diff options
author | Joonas Lahtinen <joonas.lahtinen@linux.intel.com> | 2020-04-16 14:35:16 +0300 |
---|---|---|
committer | Joonas Lahtinen <joonas.lahtinen@linux.intel.com> | 2020-04-16 14:35:16 +0300 |
commit | 2b703bbda2713fd2a7d98029ea6c44f9c3159f34 (patch) | |
tree | 00aabbb35186a4541ccccf89dead04e251d0541d /drivers/gpu/drm | |
parent | 3ffaf56e912e76cf09c560244c8804f9bebda8b1 (diff) | |
parent | 8f3d9f354286745c751374f5f1fcafee6b3f3136 (diff) | |
download | linux-2b703bbda2713fd2a7d98029ea6c44f9c3159f34.tar.bz2 |
Merge drm/drm-next into drm-intel-next-queued
Backmerging in order to pull "topic/phy-compliance".
Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Diffstat (limited to 'drivers/gpu/drm')
656 files changed, 23930 insertions, 10903 deletions
diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig index 4bffa08f610a..43594978958e 100644 --- a/drivers/gpu/drm/Kconfig +++ b/drivers/gpu/drm/Kconfig @@ -54,9 +54,6 @@ config DRM_DEBUG_MM If in doubt, say "N". -config DRM_EXPORT_FOR_TESTS - bool - config DRM_DEBUG_SELFTEST tristate "kselftests for DRM" depends on DRM @@ -470,6 +467,9 @@ config DRM_SAVAGE endif # DRM_LEGACY +config DRM_EXPORT_FOR_TESTS + bool + # Separate option because drm_panel_orientation_quirks.c is shared with fbdev config DRM_PANEL_ORIENTATION_QUIRKS tristate diff --git a/drivers/gpu/drm/Makefile b/drivers/gpu/drm/Makefile index ca0ca775d37f..7f72ef5e7811 100644 --- a/drivers/gpu/drm/Makefile +++ b/drivers/gpu/drm/Makefile @@ -39,7 +39,8 @@ obj-$(CONFIG_DRM_VRAM_HELPER) += drm_vram_helper.o drm_ttm_helper-y := drm_gem_ttm_helper.o obj-$(CONFIG_DRM_TTM_HELPER) += drm_ttm_helper.o -drm_kms_helper-y := drm_crtc_helper.o drm_dp_helper.o drm_dsc.o drm_probe_helper.o \ +drm_kms_helper-y := drm_bridge_connector.o drm_crtc_helper.o drm_dp_helper.o \ + drm_dsc.o drm_probe_helper.o \ drm_plane_helper.o drm_dp_mst_topology.o drm_atomic_helper.o \ drm_kms_helper_common.o drm_dp_dual_mode_helper.o \ drm_simple_kms_helper.o drm_modeset_helper.o \ diff --git a/drivers/gpu/drm/amd/acp/Kconfig b/drivers/gpu/drm/amd/acp/Kconfig index 13340f353ea8..216d932a7831 100644 --- a/drivers/gpu/drm/amd/acp/Kconfig +++ b/drivers/gpu/drm/amd/acp/Kconfig @@ -1,5 +1,6 @@ # SPDX-License-Identifier: MIT menu "ACP (Audio CoProcessor) Configuration" + depends on DRM_AMDGPU config DRM_AMD_ACP bool "Enable AMD Audio CoProcessor IP support" diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index da3bcff61b97..2992a49ad4a5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -579,6 +579,7 @@ struct amdgpu_asic_funcs { /* invalidate hdp read cache */ void (*invalidate_hdp)(struct amdgpu_device *adev, struct amdgpu_ring *ring); + void (*reset_hdp_ras_error_count)(struct amdgpu_device *adev); /* check if the asic needs a full reset of if soft reset will work */ bool (*need_full_reset)(struct amdgpu_device *adev); /* initialize doorbell layout for specific asic*/ @@ -969,6 +970,7 @@ struct amdgpu_device { int pstate; /* enable runtime pm on the device */ bool runpm; + bool in_runpm; bool pm_sysfs_en; bool ucode_sysfs_en; @@ -992,6 +994,8 @@ uint32_t amdgpu_mm_rreg(struct amdgpu_device *adev, uint32_t reg, uint32_t acc_flags); void amdgpu_mm_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v, uint32_t acc_flags); +void amdgpu_mm_wreg_mmio_rlc(struct amdgpu_device *adev, uint32_t reg, uint32_t v, + uint32_t acc_flags); void amdgpu_mm_wreg8(struct amdgpu_device *adev, uint32_t offset, uint8_t value); uint8_t amdgpu_mm_rreg8(struct amdgpu_device *adev, uint32_t offset); @@ -1174,9 +1178,9 @@ void amdgpu_driver_postclose_kms(struct drm_device *dev, int amdgpu_device_ip_suspend(struct amdgpu_device *adev); int amdgpu_device_suspend(struct drm_device *dev, bool fbcon); int amdgpu_device_resume(struct drm_device *dev, bool fbcon); -u32 amdgpu_get_vblank_counter_kms(struct drm_device *dev, unsigned int pipe); -int amdgpu_enable_vblank_kms(struct drm_device *dev, unsigned int pipe); -void amdgpu_disable_vblank_kms(struct drm_device *dev, unsigned int pipe); +u32 amdgpu_get_vblank_counter_kms(struct drm_crtc *crtc); +int amdgpu_enable_vblank_kms(struct drm_crtc *crtc); +void amdgpu_disable_vblank_kms(struct drm_crtc *crtc); long amdgpu_kms_compat_ioctl(struct file *filp, unsigned int cmd, unsigned long arg); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c index 8609287620ea..abfbe89e805e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c @@ -29,6 +29,7 @@ #include <linux/module.h> #include <linux/dma-buf.h> #include "amdgpu_xgmi.h" +#include <uapi/linux/kfd_ioctl.h> static const unsigned int compute_vmid_bitmap = 0xFF00; @@ -126,7 +127,7 @@ void amdgpu_amdkfd_device_init(struct amdgpu_device *adev) /* this is going to have a few of the MSBs set that we need to * clear */ - bitmap_complement(gpu_resources.queue_bitmap, + bitmap_complement(gpu_resources.cp_queue_bitmap, adev->gfx.mec.queue_bitmap, KGD_MAX_QUEUES); @@ -137,7 +138,7 @@ void amdgpu_amdkfd_device_init(struct amdgpu_device *adev) * adev->gfx.mec.num_pipe_per_mec * adev->gfx.mec.num_queue_per_pipe; for (i = last_valid_bit; i < KGD_MAX_QUEUES; ++i) - clear_bit(i, gpu_resources.queue_bitmap); + clear_bit(i, gpu_resources.cp_queue_bitmap); amdgpu_doorbell_get_kfd_info(adev, &gpu_resources.doorbell_physical_address, @@ -178,18 +179,18 @@ void amdgpu_amdkfd_interrupt(struct amdgpu_device *adev, kgd2kfd_interrupt(adev->kfd.dev, ih_ring_entry); } -void amdgpu_amdkfd_suspend(struct amdgpu_device *adev) +void amdgpu_amdkfd_suspend(struct amdgpu_device *adev, bool run_pm) { if (adev->kfd.dev) - kgd2kfd_suspend(adev->kfd.dev); + kgd2kfd_suspend(adev->kfd.dev, run_pm); } -int amdgpu_amdkfd_resume(struct amdgpu_device *adev) +int amdgpu_amdkfd_resume(struct amdgpu_device *adev, bool run_pm) { int r = 0; if (adev->kfd.dev) - r = kgd2kfd_resume(adev->kfd.dev); + r = kgd2kfd_resume(adev->kfd.dev, run_pm); return r; } @@ -224,7 +225,7 @@ void amdgpu_amdkfd_gpu_reset(struct kgd_dev *kgd) int amdgpu_amdkfd_alloc_gtt_mem(struct kgd_dev *kgd, size_t size, void **mem_obj, uint64_t *gpu_addr, - void **cpu_ptr, bool mqd_gfx9) + void **cpu_ptr, bool cp_mqd_gfx9) { struct amdgpu_device *adev = (struct amdgpu_device *)kgd; struct amdgpu_bo *bo = NULL; @@ -240,8 +241,8 @@ int amdgpu_amdkfd_alloc_gtt_mem(struct kgd_dev *kgd, size_t size, bp.type = ttm_bo_type_kernel; bp.resv = NULL; - if (mqd_gfx9) - bp.flags |= AMDGPU_GEM_CREATE_MQD_GFX9; + if (cp_mqd_gfx9) + bp.flags |= AMDGPU_GEM_CREATE_CP_MQD_GFX9; r = amdgpu_bo_create(adev, &bp, &bo); if (r) { @@ -402,7 +403,7 @@ void amdgpu_amdkfd_get_local_mem_info(struct kgd_dev *kgd, if (amdgpu_sriov_vf(adev)) mem_info->mem_clk_max = adev->clock.default_mclk / 100; - else if (adev->powerplay.pp_funcs) { + else if (adev->pm.dpm_enabled) { if (amdgpu_emu_mode == 1) mem_info->mem_clk_max = 0; else @@ -427,7 +428,7 @@ uint32_t amdgpu_amdkfd_get_max_engine_clock_in_mhz(struct kgd_dev *kgd) /* the sclk is in quantas of 10kHz */ if (amdgpu_sriov_vf(adev)) return adev->clock.default_sclk / 100; - else if (adev->powerplay.pp_funcs) + else if (adev->pm.dpm_enabled) return amdgpu_dpm_get_sclk(adev, false) / 100; else return 100; @@ -501,10 +502,11 @@ int amdgpu_amdkfd_get_dmabuf_info(struct kgd_dev *kgd, int dma_buf_fd, metadata_size, &metadata_flags); if (flags) { *flags = (bo->preferred_domains & AMDGPU_GEM_DOMAIN_VRAM) ? - ALLOC_MEM_FLAGS_VRAM : ALLOC_MEM_FLAGS_GTT; + KFD_IOC_ALLOC_MEM_FLAGS_VRAM + : KFD_IOC_ALLOC_MEM_FLAGS_GTT; if (bo->flags & AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED) - *flags |= ALLOC_MEM_FLAGS_PUBLIC; + *flags |= KFD_IOC_ALLOC_MEM_FLAGS_PUBLIC; } out_put: @@ -525,6 +527,14 @@ uint64_t amdgpu_amdkfd_get_hive_id(struct kgd_dev *kgd) return adev->gmc.xgmi.hive_id; } + +uint64_t amdgpu_amdkfd_get_unique_id(struct kgd_dev *kgd) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)kgd; + + return adev->unique_id; +} + uint8_t amdgpu_amdkfd_get_xgmi_hops_count(struct kgd_dev *dst, struct kgd_dev *src) { struct amdgpu_device *peer_adev = (struct amdgpu_device *)src; @@ -647,13 +657,9 @@ int amdgpu_amdkfd_flush_gpu_tlb_vmid(struct kgd_dev *kgd, uint16_t vmid) int amdgpu_amdkfd_flush_gpu_tlb_pasid(struct kgd_dev *kgd, uint16_t pasid) { struct amdgpu_device *adev = (struct amdgpu_device *)kgd; - uint32_t flush_type = 0; + const uint32_t flush_type = 0; bool all_hub = false; - if (adev->gmc.xgmi.num_physical_nodes && - adev->asic_type == CHIP_VEGA20) - flush_type = 2; - if (adev->family == AMDGPU_FAMILY_AI) all_hub = true; @@ -677,6 +683,11 @@ void amdgpu_amdkfd_unreserve_memory_limit(struct amdgpu_bo *bo) { } +int amdgpu_amdkfd_remove_fence_on_pt_pd_bos(struct amdgpu_bo *bo) +{ + return 0; +} + void amdgpu_amdkfd_gpuvm_destroy_cb(struct amdgpu_device *adev, struct amdgpu_vm *vm) { @@ -713,11 +724,11 @@ void kgd2kfd_exit(void) { } -void kgd2kfd_suspend(struct kfd_dev *kfd) +void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm) { } -int kgd2kfd_resume(struct kfd_dev *kfd) +int kgd2kfd_resume(struct kfd_dev *kfd, bool run_pm) { return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h index 47b0f2957d1f..13feb313e9b3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h @@ -96,6 +96,7 @@ struct amdgpu_amdkfd_fence *amdgpu_amdkfd_fence_create(u64 context, struct mm_struct *mm); bool amdkfd_fence_check_mm(struct dma_fence *f, struct mm_struct *mm); struct amdgpu_amdkfd_fence *to_amdgpu_amdkfd_fence(struct dma_fence *f); +int amdgpu_amdkfd_remove_fence_on_pt_pd_bos(struct amdgpu_bo *bo); struct amdkfd_process_info { /* List head of all VMs that belong to a KFD process */ @@ -122,8 +123,8 @@ struct amdkfd_process_info { int amdgpu_amdkfd_init(void); void amdgpu_amdkfd_fini(void); -void amdgpu_amdkfd_suspend(struct amdgpu_device *adev); -int amdgpu_amdkfd_resume(struct amdgpu_device *adev); +void amdgpu_amdkfd_suspend(struct amdgpu_device *adev, bool run_pm); +int amdgpu_amdkfd_resume(struct amdgpu_device *adev, bool run_pm); void amdgpu_amdkfd_interrupt(struct amdgpu_device *adev, const void *ih_ring_entry); void amdgpu_amdkfd_device_probe(struct amdgpu_device *adev); @@ -171,6 +172,7 @@ int amdgpu_amdkfd_get_dmabuf_info(struct kgd_dev *kgd, int dma_buf_fd, uint32_t *flags); uint64_t amdgpu_amdkfd_get_vram_usage(struct kgd_dev *kgd); uint64_t amdgpu_amdkfd_get_hive_id(struct kgd_dev *kgd); +uint64_t amdgpu_amdkfd_get_unique_id(struct kgd_dev *kgd); uint64_t amdgpu_amdkfd_get_mmio_remap_phys_addr(struct kgd_dev *kgd); uint32_t amdgpu_amdkfd_get_num_gws(struct kgd_dev *kgd); uint8_t amdgpu_amdkfd_get_xgmi_hops_count(struct kgd_dev *dst, struct kgd_dev *src); @@ -240,6 +242,9 @@ int amdgpu_amdkfd_gpuvm_import_dmabuf(struct kgd_dev *kgd, void amdgpu_amdkfd_gpuvm_init_mem_limits(void); void amdgpu_amdkfd_unreserve_memory_limit(struct amdgpu_bo *bo); +int amdgpu_amdkfd_get_tile_config(struct kgd_dev *kgd, + struct tile_config *config); + /* KGD2KFD callbacks */ int kgd2kfd_init(void); void kgd2kfd_exit(void); @@ -249,8 +254,8 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd, struct drm_device *ddev, const struct kgd2kfd_shared_resources *gpu_resources); void kgd2kfd_device_exit(struct kfd_dev *kfd); -void kgd2kfd_suspend(struct kfd_dev *kfd); -int kgd2kfd_resume(struct kfd_dev *kfd); +void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm); +int kgd2kfd_resume(struct kfd_dev *kfd, bool run_pm); int kgd2kfd_pre_reset(struct kfd_dev *kfd); int kgd2kfd_post_reset(struct kfd_dev *kfd); void kgd2kfd_interrupt(struct kfd_dev *kfd, const void *ih_ring_entry); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c index 4bcc175a149d..6529caca88fe 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c @@ -79,7 +79,7 @@ static uint32_t get_sdma_rlc_reg_offset(struct amdgpu_device *adev, dev_warn(adev->dev, "Invalid sdma engine id (%d), using engine id 0\n", engine_id); - /* fall through */ + fallthrough; case 0: sdma_engine_reg_base = SOC15_REG_OFFSET(SDMA0, 0, mmSDMA0_RLC0_RB_CNTL) - mmSDMA0_RLC0_RB_CNTL; @@ -319,7 +319,6 @@ const struct kfd2kgd_calls arcturus_kfd2kgd = { .address_watch_get_offset = kgd_gfx_v9_address_watch_get_offset, .get_atc_vmid_pasid_mapping_info = kgd_gfx_v9_get_atc_vmid_pasid_mapping_info, - .get_tile_config = kgd_gfx_v9_get_tile_config, .set_vm_context_page_table_base = kgd_set_vm_context_page_table_base, .get_hive_id = amdgpu_amdkfd_get_hive_id, }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c index a7b17c8deb00..4ec6d0c03201 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c @@ -42,38 +42,6 @@ enum hqd_dequeue_request_type { SAVE_WAVES }; -/* Because of REG_GET_FIELD() being used, we put this function in the - * asic specific file. - */ -static int amdgpu_amdkfd_get_tile_config(struct kgd_dev *kgd, - struct tile_config *config) -{ - struct amdgpu_device *adev = (struct amdgpu_device *)kgd; - - config->gb_addr_config = adev->gfx.config.gb_addr_config; -#if 0 -/* TODO - confirm REG_GET_FIELD x2, should be OK as is... but - * MC_ARB_RAMCFG register doesn't exist on Vega10 - initial amdgpu - * changes commented out related code, doing the same here for now but - * need to sync with Ken et al - */ - config->num_banks = REG_GET_FIELD(adev->gfx.config.mc_arb_ramcfg, - MC_ARB_RAMCFG, NOOFBANK); - config->num_ranks = REG_GET_FIELD(adev->gfx.config.mc_arb_ramcfg, - MC_ARB_RAMCFG, NOOFRANKS); -#endif - - config->tile_config_ptr = adev->gfx.config.tile_mode_array; - config->num_tile_configs = - ARRAY_SIZE(adev->gfx.config.tile_mode_array); - config->macro_tile_config_ptr = - adev->gfx.config.macrotile_mode_array; - config->num_macro_tile_configs = - ARRAY_SIZE(adev->gfx.config.macrotile_mode_array); - - return 0; -} - static inline struct amdgpu_device *get_amdgpu_device(struct kgd_dev *kgd) { return (struct amdgpu_device *)kgd; @@ -805,7 +773,7 @@ const struct kfd2kgd_calls gfx_v10_kfd2kgd = { .address_watch_get_offset = kgd_address_watch_get_offset, .get_atc_vmid_pasid_mapping_info = get_atc_vmid_pasid_mapping_info, - .get_tile_config = amdgpu_amdkfd_get_tile_config, .set_vm_context_page_table_base = set_vm_context_page_table_base, .get_hive_id = amdgpu_amdkfd_get_hive_id, + .get_unique_id = amdgpu_amdkfd_get_unique_id, }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c index 8f052e98a3c6..0b7e78748540 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c @@ -84,31 +84,6 @@ union TCP_WATCH_CNTL_BITS { float f32All; }; -/* Because of REG_GET_FIELD() being used, we put this function in the - * asic specific file. - */ -static int get_tile_config(struct kgd_dev *kgd, - struct tile_config *config) -{ - struct amdgpu_device *adev = (struct amdgpu_device *)kgd; - - config->gb_addr_config = adev->gfx.config.gb_addr_config; - config->num_banks = REG_GET_FIELD(adev->gfx.config.mc_arb_ramcfg, - MC_ARB_RAMCFG, NOOFBANK); - config->num_ranks = REG_GET_FIELD(adev->gfx.config.mc_arb_ramcfg, - MC_ARB_RAMCFG, NOOFRANKS); - - config->tile_config_ptr = adev->gfx.config.tile_mode_array; - config->num_tile_configs = - ARRAY_SIZE(adev->gfx.config.tile_mode_array); - config->macro_tile_config_ptr = - adev->gfx.config.macrotile_mode_array; - config->num_macro_tile_configs = - ARRAY_SIZE(adev->gfx.config.macrotile_mode_array); - - return 0; -} - static inline struct amdgpu_device *get_amdgpu_device(struct kgd_dev *kgd) { return (struct amdgpu_device *)kgd; @@ -730,7 +705,6 @@ const struct kfd2kgd_calls gfx_v7_kfd2kgd = { .address_watch_get_offset = kgd_address_watch_get_offset, .get_atc_vmid_pasid_mapping_info = get_atc_vmid_pasid_mapping_info, .set_scratch_backing_va = set_scratch_backing_va, - .get_tile_config = get_tile_config, .set_vm_context_page_table_base = set_vm_context_page_table_base, .read_vmid_from_vmfault_reg = read_vmid_from_vmfault_reg, }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c index 19a10db93d68..ccd635b812b5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c @@ -41,31 +41,6 @@ enum hqd_dequeue_request_type { RESET_WAVES }; -/* Because of REG_GET_FIELD() being used, we put this function in the - * asic specific file. - */ -static int get_tile_config(struct kgd_dev *kgd, - struct tile_config *config) -{ - struct amdgpu_device *adev = (struct amdgpu_device *)kgd; - - config->gb_addr_config = adev->gfx.config.gb_addr_config; - config->num_banks = REG_GET_FIELD(adev->gfx.config.mc_arb_ramcfg, - MC_ARB_RAMCFG, NOOFBANK); - config->num_ranks = REG_GET_FIELD(adev->gfx.config.mc_arb_ramcfg, - MC_ARB_RAMCFG, NOOFRANKS); - - config->tile_config_ptr = adev->gfx.config.tile_mode_array; - config->num_tile_configs = - ARRAY_SIZE(adev->gfx.config.tile_mode_array); - config->macro_tile_config_ptr = - adev->gfx.config.macrotile_mode_array; - config->num_macro_tile_configs = - ARRAY_SIZE(adev->gfx.config.macrotile_mode_array); - - return 0; -} - static inline struct amdgpu_device *get_amdgpu_device(struct kgd_dev *kgd) { return (struct amdgpu_device *)kgd; @@ -676,6 +651,5 @@ const struct kfd2kgd_calls gfx_v8_kfd2kgd = { .get_atc_vmid_pasid_mapping_info = get_atc_vmid_pasid_mapping_info, .set_scratch_backing_va = set_scratch_backing_va, - .get_tile_config = get_tile_config, .set_vm_context_page_table_base = set_vm_context_page_table_base, }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c index 8562afe5b761..df841c2ac5e7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c @@ -48,28 +48,6 @@ enum hqd_dequeue_request_type { RESET_WAVES }; - -/* Because of REG_GET_FIELD() being used, we put this function in the - * asic specific file. - */ -int kgd_gfx_v9_get_tile_config(struct kgd_dev *kgd, - struct tile_config *config) -{ - struct amdgpu_device *adev = (struct amdgpu_device *)kgd; - - config->gb_addr_config = adev->gfx.config.gb_addr_config; - - config->tile_config_ptr = adev->gfx.config.tile_mode_array; - config->num_tile_configs = - ARRAY_SIZE(adev->gfx.config.tile_mode_array); - config->macro_tile_config_ptr = - adev->gfx.config.macrotile_mode_array; - config->num_macro_tile_configs = - ARRAY_SIZE(adev->gfx.config.macrotile_mode_array); - - return 0; -} - static inline struct amdgpu_device *get_amdgpu_device(struct kgd_dev *kgd) { return (struct amdgpu_device *)kgd; @@ -736,7 +714,7 @@ const struct kfd2kgd_calls gfx_v9_kfd2kgd = { .address_watch_get_offset = kgd_gfx_v9_address_watch_get_offset, .get_atc_vmid_pasid_mapping_info = kgd_gfx_v9_get_atc_vmid_pasid_mapping_info, - .get_tile_config = kgd_gfx_v9_get_tile_config, .set_vm_context_page_table_base = kgd_gfx_v9_set_vm_context_page_table_base, .get_hive_id = amdgpu_amdkfd_get_hive_id, + .get_unique_id = amdgpu_amdkfd_get_unique_id, }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h index 63d3e6683dfe..aedf67d57449 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h @@ -60,5 +60,3 @@ uint32_t kgd_gfx_v9_address_watch_get_offset(struct kgd_dev *kgd, bool kgd_gfx_v9_get_atc_vmid_pasid_mapping_info(struct kgd_dev *kgd, uint8_t vmid, uint16_t *p_pasid); -int kgd_gfx_v9_get_tile_config(struct kgd_dev *kgd, - struct tile_config *config); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c index fa8ac9d19a7a..9dff792c9290 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c @@ -29,6 +29,7 @@ #include "amdgpu_vm.h" #include "amdgpu_amdkfd.h" #include "amdgpu_dma_buf.h" +#include <uapi/linux/kfd_ioctl.h> /* BO flag to indicate a KFD userptr BO */ #define AMDGPU_AMDKFD_USERPTR_BO (1ULL << 63) @@ -276,6 +277,42 @@ static int amdgpu_amdkfd_remove_eviction_fence(struct amdgpu_bo *bo, return 0; } +int amdgpu_amdkfd_remove_fence_on_pt_pd_bos(struct amdgpu_bo *bo) +{ + struct amdgpu_bo *root = bo; + struct amdgpu_vm_bo_base *vm_bo; + struct amdgpu_vm *vm; + struct amdkfd_process_info *info; + struct amdgpu_amdkfd_fence *ef; + int ret; + + /* we can always get vm_bo from root PD bo.*/ + while (root->parent) + root = root->parent; + + vm_bo = root->vm_bo; + if (!vm_bo) + return 0; + + vm = vm_bo->vm; + if (!vm) + return 0; + + info = vm->process_info; + if (!info || !info->eviction_fence) + return 0; + + ef = container_of(dma_fence_get(&info->eviction_fence->base), + struct amdgpu_amdkfd_fence, base); + + BUG_ON(!dma_resv_trylock(bo->tbo.base.resv)); + ret = amdgpu_amdkfd_remove_eviction_fence(bo, ef); + dma_resv_unlock(bo->tbo.base.resv); + + dma_fence_put(&ef->base); + return ret; +} + static int amdgpu_amdkfd_bo_validate(struct amdgpu_bo *bo, uint32_t domain, bool wait) { @@ -364,18 +401,18 @@ static int vm_update_pds(struct amdgpu_vm *vm, struct amdgpu_sync *sync) static uint64_t get_pte_flags(struct amdgpu_device *adev, struct kgd_mem *mem) { struct amdgpu_device *bo_adev = amdgpu_ttm_adev(mem->bo->tbo.bdev); - bool coherent = mem->alloc_flags & ALLOC_MEM_FLAGS_COHERENT; + bool coherent = mem->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_COHERENT; uint32_t mapping_flags; mapping_flags = AMDGPU_VM_PAGE_READABLE; - if (mem->alloc_flags & ALLOC_MEM_FLAGS_WRITABLE) + if (mem->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_WRITABLE) mapping_flags |= AMDGPU_VM_PAGE_WRITEABLE; - if (mem->alloc_flags & ALLOC_MEM_FLAGS_EXECUTABLE) + if (mem->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_EXECUTABLE) mapping_flags |= AMDGPU_VM_PAGE_EXECUTABLE; switch (adev->asic_type) { case CHIP_ARCTURUS: - if (mem->alloc_flags & ALLOC_MEM_FLAGS_VRAM) { + if (mem->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_VRAM) { if (bo_adev == adev) mapping_flags |= coherent ? AMDGPU_VM_MTYPE_CC : AMDGPU_VM_MTYPE_RW; @@ -847,9 +884,9 @@ static int process_sync_pds_resv(struct amdkfd_process_info *process_info, vm_list_node) { struct amdgpu_bo *pd = peer_vm->root.base.bo; - ret = amdgpu_sync_resv(NULL, - sync, pd->tbo.base.resv, - AMDGPU_FENCE_OWNER_KFD, false); + ret = amdgpu_sync_resv(NULL, sync, pd->tbo.base.resv, + AMDGPU_SYNC_NE_OWNER, + AMDGPU_FENCE_OWNER_KFD); if (ret) return ret; } @@ -1044,6 +1081,8 @@ void amdgpu_amdkfd_gpuvm_destroy_cb(struct amdgpu_device *adev, list_del(&vm->vm_list_node); mutex_unlock(&process_info->lock); + vm->process_info = NULL; + /* Release per-process resources when last compute VM is destroyed */ if (!process_info->n_vms) { WARN_ON(!list_empty(&process_info->kfd_bo_list)); @@ -1122,24 +1161,24 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu( /* * Check on which domain to allocate BO */ - if (flags & ALLOC_MEM_FLAGS_VRAM) { + if (flags & KFD_IOC_ALLOC_MEM_FLAGS_VRAM) { domain = alloc_domain = AMDGPU_GEM_DOMAIN_VRAM; alloc_flags = AMDGPU_GEM_CREATE_VRAM_WIPE_ON_RELEASE; - alloc_flags |= (flags & ALLOC_MEM_FLAGS_PUBLIC) ? + alloc_flags |= (flags & KFD_IOC_ALLOC_MEM_FLAGS_PUBLIC) ? AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED : AMDGPU_GEM_CREATE_NO_CPU_ACCESS; - } else if (flags & ALLOC_MEM_FLAGS_GTT) { + } else if (flags & KFD_IOC_ALLOC_MEM_FLAGS_GTT) { domain = alloc_domain = AMDGPU_GEM_DOMAIN_GTT; alloc_flags = 0; - } else if (flags & ALLOC_MEM_FLAGS_USERPTR) { + } else if (flags & KFD_IOC_ALLOC_MEM_FLAGS_USERPTR) { domain = AMDGPU_GEM_DOMAIN_GTT; alloc_domain = AMDGPU_GEM_DOMAIN_CPU; alloc_flags = 0; if (!offset || !*offset) return -EINVAL; user_addr = untagged_addr(*offset); - } else if (flags & (ALLOC_MEM_FLAGS_DOORBELL | - ALLOC_MEM_FLAGS_MMIO_REMAP)) { + } else if (flags & (KFD_IOC_ALLOC_MEM_FLAGS_DOORBELL | + KFD_IOC_ALLOC_MEM_FLAGS_MMIO_REMAP)) { domain = AMDGPU_GEM_DOMAIN_GTT; alloc_domain = AMDGPU_GEM_DOMAIN_CPU; bo_type = ttm_bo_type_sg; @@ -1160,7 +1199,7 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu( } INIT_LIST_HEAD(&(*mem)->bo_va_list); mutex_init(&(*mem)->lock); - (*mem)->aql_queue = !!(flags & ALLOC_MEM_FLAGS_AQL_QUEUE_MEM); + (*mem)->aql_queue = !!(flags & KFD_IOC_ALLOC_MEM_FLAGS_AQL_QUEUE_MEM); /* Workaround for AQL queue wraparound bug. Map the same * memory twice. That means we only actually allocate half @@ -1642,10 +1681,12 @@ int amdgpu_amdkfd_gpuvm_import_dmabuf(struct kgd_dev *kgd, INIT_LIST_HEAD(&(*mem)->bo_va_list); mutex_init(&(*mem)->lock); + (*mem)->alloc_flags = ((bo->preferred_domains & AMDGPU_GEM_DOMAIN_VRAM) ? - ALLOC_MEM_FLAGS_VRAM : ALLOC_MEM_FLAGS_GTT) | - ALLOC_MEM_FLAGS_WRITABLE | ALLOC_MEM_FLAGS_EXECUTABLE; + KFD_IOC_ALLOC_MEM_FLAGS_VRAM : KFD_IOC_ALLOC_MEM_FLAGS_GTT) + | KFD_IOC_ALLOC_MEM_FLAGS_WRITABLE + | KFD_IOC_ALLOC_MEM_FLAGS_EXECUTABLE; (*mem)->bo = amdgpu_bo_ref(bo); (*mem)->va = va; @@ -2204,3 +2245,25 @@ int amdgpu_amdkfd_remove_gws_from_process(void *info, void *mem) kfree(mem); return 0; } + +/* Returns GPU-specific tiling mode information */ +int amdgpu_amdkfd_get_tile_config(struct kgd_dev *kgd, + struct tile_config *config) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)kgd; + + config->gb_addr_config = adev->gfx.config.gb_addr_config; + config->tile_config_ptr = adev->gfx.config.tile_mode_array; + config->num_tile_configs = + ARRAY_SIZE(adev->gfx.config.tile_mode_array); + config->macro_tile_config_ptr = + adev->gfx.config.macrotile_mode_array; + config->num_macro_tile_configs = + ARRAY_SIZE(adev->gfx.config.macrotile_mode_array); + + /* Those values are not set from GFX9 onwards */ + config->num_banks = adev->gfx.config.num_banks; + config->num_ranks = adev->gfx.config.num_ranks; + + return 0; +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c index 50dff69a0f6e..b1172d93c99c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c @@ -192,30 +192,35 @@ static bool amdgpu_read_bios_from_rom(struct amdgpu_device *adev) static bool amdgpu_read_platform_bios(struct amdgpu_device *adev) { - uint8_t __iomem *bios; - size_t size; + phys_addr_t rom = adev->pdev->rom; + size_t romlen = adev->pdev->romlen; + void __iomem *bios; adev->bios = NULL; - bios = pci_platform_rom(adev->pdev, &size); - if (!bios) { + if (!rom || romlen == 0) return false; - } - adev->bios = kzalloc(size, GFP_KERNEL); - if (adev->bios == NULL) + adev->bios = kzalloc(romlen, GFP_KERNEL); + if (!adev->bios) return false; - memcpy_fromio(adev->bios, bios, size); + bios = ioremap(rom, romlen); + if (!bios) + goto free_bios; - if (!check_atom_bios(adev->bios, size)) { - kfree(adev->bios); - return false; - } + memcpy_fromio(adev->bios, bios, romlen); + iounmap(bios); - adev->bios_size = size; + if (!check_atom_bios(adev->bios, romlen)) + goto free_bios; + + adev->bios_size = romlen; return true; +free_bios: + kfree(adev->bios); + return false; } #ifdef CONFIG_ACPI diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c index a62cbc8199de..f355d9a752d2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c @@ -1461,6 +1461,20 @@ static enum drm_mode_status amdgpu_connector_dp_mode_valid(struct drm_connector return MODE_OK; } +static int +amdgpu_connector_late_register(struct drm_connector *connector) +{ + struct amdgpu_connector *amdgpu_connector = to_amdgpu_connector(connector); + int r = 0; + + if (amdgpu_connector->ddc_bus->has_aux) { + amdgpu_connector->ddc_bus->aux.dev = amdgpu_connector->base.kdev; + r = drm_dp_aux_register(&amdgpu_connector->ddc_bus->aux); + } + + return r; +} + static const struct drm_connector_helper_funcs amdgpu_connector_dp_helper_funcs = { .get_modes = amdgpu_connector_dp_get_modes, .mode_valid = amdgpu_connector_dp_mode_valid, @@ -1475,6 +1489,7 @@ static const struct drm_connector_funcs amdgpu_connector_dp_funcs = { .early_unregister = amdgpu_connector_unregister, .destroy = amdgpu_connector_destroy, .force = amdgpu_connector_dvi_force, + .late_register = amdgpu_connector_late_register, }; static const struct drm_connector_funcs amdgpu_connector_edp_funcs = { @@ -1485,6 +1500,7 @@ static const struct drm_connector_funcs amdgpu_connector_edp_funcs = { .early_unregister = amdgpu_connector_unregister, .destroy = amdgpu_connector_destroy, .force = amdgpu_connector_dvi_force, + .late_register = amdgpu_connector_late_register, }; void @@ -1931,7 +1947,6 @@ amdgpu_connector_add(struct amdgpu_device *adev, connector->polled = DRM_CONNECTOR_POLL_HPD; connector->display_info.subpixel_order = subpixel_order; - drm_connector_register(connector); if (has_aux) amdgpu_atombios_dp_aux_init(amdgpu_connector); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c index a52a084158b1..af91627b19b0 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -28,6 +28,7 @@ #include <linux/file.h> #include <linux/pagemap.h> #include <linux/sync_file.h> +#include <linux/dma-buf.h> #include <drm/amdgpu_drm.h> #include <drm/drm_syncobj.h> @@ -415,7 +416,9 @@ static int amdgpu_cs_bo_validate(struct amdgpu_cs_parser *p, /* Don't move this buffer if we have depleted our allowance * to move it. Don't move anything if the threshold is zero. */ - if (p->bytes_moved < p->bytes_moved_threshold) { + if (p->bytes_moved < p->bytes_moved_threshold && + (!bo->tbo.base.dma_buf || + list_empty(&bo->tbo.base.dma_buf->attachments))) { if (!amdgpu_gmc_vram_full_visible(&adev->gmc) && (bo->flags & AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED)) { /* And don't move a CPU_ACCESS_REQUIRED BO to limited @@ -651,16 +654,19 @@ out: static int amdgpu_cs_sync_rings(struct amdgpu_cs_parser *p) { + struct amdgpu_fpriv *fpriv = p->filp->driver_priv; struct amdgpu_bo_list_entry *e; int r; list_for_each_entry(e, &p->validated, tv.head) { struct amdgpu_bo *bo = ttm_to_amdgpu_bo(e->tv.bo); struct dma_resv *resv = bo->tbo.base.resv; + enum amdgpu_sync_mode sync_mode; - r = amdgpu_sync_resv(p->adev, &p->job->sync, resv, p->filp, - amdgpu_bo_explicit_sync(bo)); - + sync_mode = amdgpu_bo_explicit_sync(bo) ? + AMDGPU_SYNC_EXPLICIT : AMDGPU_SYNC_NE_OWNER; + r = amdgpu_sync_resv(p->adev, &p->job->sync, resv, sync_mode, + &fpriv->vm); if (r) return r; } @@ -1202,7 +1208,6 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p, struct amdgpu_fpriv *fpriv = p->filp->driver_priv; struct drm_sched_entity *entity = p->entity; enum drm_sched_priority priority; - struct amdgpu_ring *ring; struct amdgpu_bo_list_entry *e; struct amdgpu_job *job; uint64_t seq; @@ -1211,7 +1216,7 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p, job = p->job; p->job = NULL; - r = drm_sched_job_init(&job->base, entity, p->filp); + r = drm_sched_job_init(&job->base, entity, &fpriv->vm); if (r) goto error_unlock; @@ -1255,9 +1260,6 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p, priority = job->base.s_priority; drm_sched_entity_push_job(&job->base, entity); - ring = to_amdgpu_ring(entity->rq->sched); - amdgpu_ring_priority_get(ring, priority); - amdgpu_vm_move_to_lru_tail(p->adev, &fpriv->vm); ttm_eu_fence_buffer_objects(&p->ticket, &p->validated, p->fence); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c index 94a6c42f29ea..6ed36a2c5f73 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c @@ -61,12 +61,24 @@ static int amdgpu_ctx_priority_permit(struct drm_file *filp, return -EACCES; } +static enum gfx_pipe_priority amdgpu_ctx_sched_prio_to_compute_prio(enum drm_sched_priority prio) +{ + switch (prio) { + case DRM_SCHED_PRIORITY_HIGH_HW: + case DRM_SCHED_PRIORITY_KERNEL: + return AMDGPU_GFX_PIPE_PRIO_HIGH; + default: + return AMDGPU_GFX_PIPE_PRIO_NORMAL; + } +} + static int amdgpu_ctx_init_entity(struct amdgpu_ctx *ctx, const u32 hw_ip, const u32 ring) { struct amdgpu_device *adev = ctx->adev; struct amdgpu_ctx_entity *entity; struct drm_gpu_scheduler **scheds = NULL, *sched = NULL; unsigned num_scheds = 0; + enum gfx_pipe_priority hw_prio; enum drm_sched_priority priority; int r; @@ -79,46 +91,51 @@ static int amdgpu_ctx_init_entity(struct amdgpu_ctx *ctx, const u32 hw_ip, const priority = (ctx->override_priority == DRM_SCHED_PRIORITY_UNSET) ? ctx->init_priority : ctx->override_priority; switch (hw_ip) { - case AMDGPU_HW_IP_GFX: - sched = &adev->gfx.gfx_ring[0].sched; - scheds = &sched; - num_scheds = 1; - break; - case AMDGPU_HW_IP_COMPUTE: - scheds = adev->gfx.compute_sched; - num_scheds = adev->gfx.num_compute_sched; - break; - case AMDGPU_HW_IP_DMA: - scheds = adev->sdma.sdma_sched; - num_scheds = adev->sdma.num_sdma_sched; - break; - case AMDGPU_HW_IP_UVD: - sched = &adev->uvd.inst[0].ring.sched; - scheds = &sched; - num_scheds = 1; - break; - case AMDGPU_HW_IP_VCE: - sched = &adev->vce.ring[0].sched; - scheds = &sched; - num_scheds = 1; - break; - case AMDGPU_HW_IP_UVD_ENC: - sched = &adev->uvd.inst[0].ring_enc[0].sched; - scheds = &sched; - num_scheds = 1; - break; - case AMDGPU_HW_IP_VCN_DEC: - scheds = adev->vcn.vcn_dec_sched; - num_scheds = adev->vcn.num_vcn_dec_sched; - break; - case AMDGPU_HW_IP_VCN_ENC: - scheds = adev->vcn.vcn_enc_sched; - num_scheds = adev->vcn.num_vcn_enc_sched; - break; - case AMDGPU_HW_IP_VCN_JPEG: - scheds = adev->jpeg.jpeg_sched; - num_scheds = adev->jpeg.num_jpeg_sched; - break; + case AMDGPU_HW_IP_GFX: + sched = &adev->gfx.gfx_ring[0].sched; + scheds = &sched; + num_scheds = 1; + break; + case AMDGPU_HW_IP_COMPUTE: + hw_prio = amdgpu_ctx_sched_prio_to_compute_prio(priority); + scheds = adev->gfx.compute_prio_sched[hw_prio]; + num_scheds = adev->gfx.num_compute_sched[hw_prio]; + break; + case AMDGPU_HW_IP_DMA: + scheds = adev->sdma.sdma_sched; + num_scheds = adev->sdma.num_sdma_sched; + break; + case AMDGPU_HW_IP_UVD: + sched = &adev->uvd.inst[0].ring.sched; + scheds = &sched; + num_scheds = 1; + break; + case AMDGPU_HW_IP_VCE: + sched = &adev->vce.ring[0].sched; + scheds = &sched; + num_scheds = 1; + break; + case AMDGPU_HW_IP_UVD_ENC: + sched = &adev->uvd.inst[0].ring_enc[0].sched; + scheds = &sched; + num_scheds = 1; + break; + case AMDGPU_HW_IP_VCN_DEC: + sched = drm_sched_pick_best(adev->vcn.vcn_dec_sched, + adev->vcn.num_vcn_dec_sched); + scheds = &sched; + num_scheds = 1; + break; + case AMDGPU_HW_IP_VCN_ENC: + sched = drm_sched_pick_best(adev->vcn.vcn_enc_sched, + adev->vcn.num_vcn_enc_sched); + scheds = &sched; + num_scheds = 1; + break; + case AMDGPU_HW_IP_VCN_JPEG: + scheds = adev->jpeg.jpeg_sched; + num_scheds = adev->jpeg.num_jpeg_sched; + break; } r = drm_sched_entity_init(&entity->entity, priority, scheds, num_scheds, @@ -502,6 +519,29 @@ struct dma_fence *amdgpu_ctx_get_fence(struct amdgpu_ctx *ctx, return fence; } +static void amdgpu_ctx_set_entity_priority(struct amdgpu_ctx *ctx, + struct amdgpu_ctx_entity *aentity, + int hw_ip, + enum drm_sched_priority priority) +{ + struct amdgpu_device *adev = ctx->adev; + enum gfx_pipe_priority hw_prio; + struct drm_gpu_scheduler **scheds = NULL; + unsigned num_scheds; + + /* set sw priority */ + drm_sched_entity_set_priority(&aentity->entity, priority); + + /* set hw priority */ + if (hw_ip == AMDGPU_HW_IP_COMPUTE) { + hw_prio = amdgpu_ctx_sched_prio_to_compute_prio(priority); + scheds = adev->gfx.compute_prio_sched[hw_prio]; + num_scheds = adev->gfx.num_compute_sched[hw_prio]; + drm_sched_entity_modify_sched(&aentity->entity, scheds, + num_scheds); + } +} + void amdgpu_ctx_priority_override(struct amdgpu_ctx *ctx, enum drm_sched_priority priority) { @@ -514,13 +554,11 @@ void amdgpu_ctx_priority_override(struct amdgpu_ctx *ctx, ctx->init_priority : ctx->override_priority; for (i = 0; i < AMDGPU_HW_IP_NUM; ++i) { for (j = 0; j < amdgpu_ctx_num_entities[i]; ++j) { - struct drm_sched_entity *entity; - if (!ctx->entities[i][j]) continue; - entity = &ctx->entities[i][j]->entity; - drm_sched_entity_set_priority(entity, ctx_prio); + amdgpu_ctx_set_entity_priority(ctx, ctx->entities[i][j], + i, ctx_prio); } } } @@ -628,20 +666,53 @@ void amdgpu_ctx_mgr_fini(struct amdgpu_ctx_mgr *mgr) mutex_destroy(&mgr->lock); } + +static void amdgpu_ctx_init_compute_sched(struct amdgpu_device *adev) +{ + int num_compute_sched_normal = 0; + int num_compute_sched_high = AMDGPU_MAX_COMPUTE_RINGS - 1; + int i; + + /* use one drm sched array, gfx.compute_sched to store both high and + * normal priority drm compute schedulers */ + for (i = 0; i < adev->gfx.num_compute_rings; i++) { + if (!adev->gfx.compute_ring[i].has_high_prio) + adev->gfx.compute_sched[num_compute_sched_normal++] = + &adev->gfx.compute_ring[i].sched; + else + adev->gfx.compute_sched[num_compute_sched_high--] = + &adev->gfx.compute_ring[i].sched; + } + + /* compute ring only has two priority for now */ + i = AMDGPU_GFX_PIPE_PRIO_NORMAL; + adev->gfx.compute_prio_sched[i] = &adev->gfx.compute_sched[0]; + adev->gfx.num_compute_sched[i] = num_compute_sched_normal; + + i = AMDGPU_GFX_PIPE_PRIO_HIGH; + if (num_compute_sched_high == (AMDGPU_MAX_COMPUTE_RINGS - 1)) { + /* When compute has no high priority rings then use */ + /* normal priority sched array */ + adev->gfx.compute_prio_sched[i] = &adev->gfx.compute_sched[0]; + adev->gfx.num_compute_sched[i] = num_compute_sched_normal; + } else { + adev->gfx.compute_prio_sched[i] = + &adev->gfx.compute_sched[num_compute_sched_high - 1]; + adev->gfx.num_compute_sched[i] = + adev->gfx.num_compute_rings - num_compute_sched_normal; + } +} + void amdgpu_ctx_init_sched(struct amdgpu_device *adev) { int i, j; + amdgpu_ctx_init_compute_sched(adev); for (i = 0; i < adev->gfx.num_gfx_rings; i++) { adev->gfx.gfx_sched[i] = &adev->gfx.gfx_ring[i].sched; adev->gfx.num_gfx_sched++; } - for (i = 0; i < adev->gfx.num_compute_rings; i++) { - adev->gfx.compute_sched[i] = &adev->gfx.compute_ring[i].sched; - adev->gfx.num_compute_sched++; - } - for (i = 0; i < adev->sdma.num_instances; i++) { adev->sdma.sdma_sched[i] = &adev->sdma.instance[i].ring.sched; adev->sdma.num_sdma_sched++; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c index f24ed9a1a3e5..c0f9a651dc06 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c @@ -31,6 +31,9 @@ #include <drm/drm_debugfs.h> #include "amdgpu.h" +#include "amdgpu_pm.h" +#include "amdgpu_dm_debugfs.h" +#include "amdgpu_ras.h" /** * amdgpu_debugfs_add_files - Add simple debugfs entries @@ -176,7 +179,7 @@ static int amdgpu_debugfs_process_reg_op(bool read, struct file *f, } else { r = get_user(value, (uint32_t *)buf); if (!r) - WREG32(*pos >> 2, value); + amdgpu_mm_wreg_mmio_rlc(adev, *pos >> 2, value, 0); } if (r) { result = r; @@ -781,11 +784,11 @@ static ssize_t amdgpu_debugfs_gpr_read(struct file *f, char __user *buf, ssize_t result = 0; uint32_t offset, se, sh, cu, wave, simd, thread, bank, *data; - if (size & 3 || *pos & 3) + if (size > 4096 || size & 3 || *pos & 3) return -EINVAL; /* decode offset */ - offset = *pos & GENMASK_ULL(11, 0); + offset = (*pos & GENMASK_ULL(11, 0)) >> 2; se = (*pos & GENMASK_ULL(19, 12)) >> 12; sh = (*pos & GENMASK_ULL(27, 20)) >> 20; cu = (*pos & GENMASK_ULL(35, 28)) >> 28; @@ -823,7 +826,7 @@ static ssize_t amdgpu_debugfs_gpr_read(struct file *f, char __user *buf, while (size) { uint32_t value; - value = data[offset++]; + value = data[result >> 2]; r = put_user(value, (uint32_t *)buf); if (r) { result = r; @@ -840,6 +843,55 @@ err: return result; } +/** + * amdgpu_debugfs_regs_gfxoff_write - Enable/disable GFXOFF + * + * @f: open file handle + * @buf: User buffer to write data from + * @size: Number of bytes to write + * @pos: Offset to seek to + * + * Write a 32-bit zero to disable or a 32-bit non-zero to enable + */ +static ssize_t amdgpu_debugfs_gfxoff_write(struct file *f, const char __user *buf, + size_t size, loff_t *pos) +{ + struct amdgpu_device *adev = file_inode(f)->i_private; + ssize_t result = 0; + int r; + + if (size & 0x3 || *pos & 0x3) + return -EINVAL; + + r = pm_runtime_get_sync(adev->ddev->dev); + if (r < 0) + return r; + + while (size) { + uint32_t value; + + r = get_user(value, (uint32_t *)buf); + if (r) { + pm_runtime_mark_last_busy(adev->ddev->dev); + pm_runtime_put_autosuspend(adev->ddev->dev); + return r; + } + + amdgpu_gfx_off_ctrl(adev, value ? true : false); + + result += 4; + buf += 4; + *pos += 4; + size -= 4; + } + + pm_runtime_mark_last_busy(adev->ddev->dev); + pm_runtime_put_autosuspend(adev->ddev->dev); + + return result; +} + + static const struct file_operations amdgpu_debugfs_regs_fops = { .owner = THIS_MODULE, .read = amdgpu_debugfs_regs_read, @@ -888,6 +940,11 @@ static const struct file_operations amdgpu_debugfs_gpr_fops = { .llseek = default_llseek }; +static const struct file_operations amdgpu_debugfs_gfxoff_fops = { + .owner = THIS_MODULE, + .write = amdgpu_debugfs_gfxoff_write, +}; + static const struct file_operations *debugfs_regs[] = { &amdgpu_debugfs_regs_fops, &amdgpu_debugfs_regs_didt_fops, @@ -897,6 +954,7 @@ static const struct file_operations *debugfs_regs[] = { &amdgpu_debugfs_sensors_fops, &amdgpu_debugfs_wave_fops, &amdgpu_debugfs_gpr_fops, + &amdgpu_debugfs_gfxoff_fops, }; static const char *debugfs_regs_names[] = { @@ -908,6 +966,7 @@ static const char *debugfs_regs_names[] = { "amdgpu_sensors", "amdgpu_wave", "amdgpu_gpr", + "amdgpu_gfxoff", }; /** @@ -934,18 +993,6 @@ int amdgpu_debugfs_regs_init(struct amdgpu_device *adev) return 0; } -void amdgpu_debugfs_regs_cleanup(struct amdgpu_device *adev) -{ - unsigned i; - - for (i = 0; i < ARRAY_SIZE(debugfs_regs); i++) { - if (adev->debugfs_regs[i]) { - debugfs_remove(adev->debugfs_regs[i]); - adev->debugfs_regs[i] = NULL; - } - } -} - static int amdgpu_debugfs_test_ib(struct seq_file *m, void *data) { struct drm_info_node *node = (struct drm_info_node *) m->private; @@ -1211,11 +1258,47 @@ failure: return 0; } +static int amdgpu_debugfs_sclk_set(void *data, u64 val) +{ + int ret = 0; + uint32_t max_freq, min_freq; + struct amdgpu_device *adev = (struct amdgpu_device *)data; + + if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) + return -EINVAL; + + ret = pm_runtime_get_sync(adev->ddev->dev); + if (ret < 0) + return ret; + + if (is_support_sw_smu(adev)) { + ret = smu_get_dpm_freq_range(&adev->smu, SMU_SCLK, &min_freq, &max_freq, true); + if (ret || val > max_freq || val < min_freq) + return -EINVAL; + ret = smu_set_soft_freq_range(&adev->smu, SMU_SCLK, (uint32_t)val, (uint32_t)val, true); + } else { + return 0; + } + + pm_runtime_mark_last_busy(adev->ddev->dev); + pm_runtime_put_autosuspend(adev->ddev->dev); + + if (ret) + return -EINVAL; + + return 0; +} + DEFINE_SIMPLE_ATTRIBUTE(fops_ib_preempt, NULL, amdgpu_debugfs_ib_preempt, "%llu\n"); +DEFINE_SIMPLE_ATTRIBUTE(fops_sclk_set, NULL, + amdgpu_debugfs_sclk_set, "%llu\n"); + int amdgpu_debugfs_init(struct amdgpu_device *adev) { + int r, i; + adev->debugfs_preempt = debugfs_create_file("amdgpu_preempt_ib", 0600, adev->ddev->primary->debugfs_root, adev, @@ -1225,24 +1308,78 @@ int amdgpu_debugfs_init(struct amdgpu_device *adev) return -EIO; } + adev->smu.debugfs_sclk = + debugfs_create_file("amdgpu_force_sclk", 0200, + adev->ddev->primary->debugfs_root, adev, + &fops_sclk_set); + if (!(adev->smu.debugfs_sclk)) { + DRM_ERROR("unable to create amdgpu_set_sclk debugsfs file\n"); + return -EIO; + } + + /* Register debugfs entries for amdgpu_ttm */ + r = amdgpu_ttm_debugfs_init(adev); + if (r) { + DRM_ERROR("Failed to init debugfs\n"); + return r; + } + + r = amdgpu_debugfs_pm_init(adev); + if (r) { + DRM_ERROR("Failed to register debugfs file for dpm!\n"); + return r; + } + + if (amdgpu_debugfs_sa_init(adev)) { + dev_err(adev->dev, "failed to register debugfs file for SA\n"); + } + + if (amdgpu_debugfs_fence_init(adev)) + dev_err(adev->dev, "fence debugfs file creation failed\n"); + + r = amdgpu_debugfs_gem_init(adev); + if (r) + DRM_ERROR("registering gem debugfs failed (%d).\n", r); + + r = amdgpu_debugfs_regs_init(adev); + if (r) + DRM_ERROR("registering register debugfs failed (%d).\n", r); + + r = amdgpu_debugfs_firmware_init(adev); + if (r) + DRM_ERROR("registering firmware debugfs failed (%d).\n", r); + +#if defined(CONFIG_DRM_AMD_DC) + if (amdgpu_device_has_dc_support(adev)) { + if (dtn_debugfs_init(adev)) + DRM_ERROR("amdgpu: failed initialize dtn debugfs support.\n"); + } +#endif + + for (i = 0; i < AMDGPU_MAX_RINGS; ++i) { + struct amdgpu_ring *ring = adev->rings[i]; + + if (!ring) + continue; + + if (amdgpu_debugfs_ring_init(adev, ring)) { + DRM_ERROR("Failed to register debugfs file for rings !\n"); + } + } + + amdgpu_ras_debugfs_create_all(adev); + return amdgpu_debugfs_add_files(adev, amdgpu_debugfs_list, ARRAY_SIZE(amdgpu_debugfs_list)); } -void amdgpu_debugfs_preempt_cleanup(struct amdgpu_device *adev) -{ - debugfs_remove(adev->debugfs_preempt); -} - #else int amdgpu_debugfs_init(struct amdgpu_device *adev) { return 0; } -void amdgpu_debugfs_preempt_cleanup(struct amdgpu_device *adev) { } int amdgpu_debugfs_regs_init(struct amdgpu_device *adev) { return 0; } -void amdgpu_debugfs_regs_cleanup(struct amdgpu_device *adev) { } #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.h index f289d28ad6b2..de12d1101526 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.h @@ -32,9 +32,8 @@ struct amdgpu_debugfs { }; int amdgpu_debugfs_regs_init(struct amdgpu_device *adev); -void amdgpu_debugfs_regs_cleanup(struct amdgpu_device *adev); int amdgpu_debugfs_init(struct amdgpu_device *adev); -void amdgpu_debugfs_preempt_cleanup(struct amdgpu_device *adev); +void amdgpu_debugfs_fini(struct amdgpu_device *adev); int amdgpu_debugfs_add_files(struct amdgpu_device *adev, const struct drm_info_list *files, unsigned nfiles); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 53fdafd85cb8..559dc24ef436 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -183,20 +183,51 @@ bool amdgpu_device_supports_baco(struct drm_device *dev) void amdgpu_device_vram_access(struct amdgpu_device *adev, loff_t pos, uint32_t *buf, size_t size, bool write) { - uint64_t last; unsigned long flags; + uint32_t hi = ~0; + uint64_t last; + + +#ifdef CONFIG_64BIT + last = min(pos + size, adev->gmc.visible_vram_size); + if (last > pos) { + void __iomem *addr = adev->mman.aper_base_kaddr + pos; + size_t count = last - pos; + + if (write) { + memcpy_toio(addr, buf, count); + mb(); + amdgpu_asic_flush_hdp(adev, NULL); + } else { + amdgpu_asic_invalidate_hdp(adev, NULL); + mb(); + memcpy_fromio(buf, addr, count); + } + + if (count == size) + return; + + pos += count; + buf += count / 4; + size -= count; + } +#endif + + spin_lock_irqsave(&adev->mmio_idx_lock, flags); + for (last = pos + size; pos < last; pos += 4) { + uint32_t tmp = pos >> 31; - last = size - 4; - for (last += pos; pos <= last; pos += 4) { - spin_lock_irqsave(&adev->mmio_idx_lock, flags); WREG32_NO_KIQ(mmMM_INDEX, ((uint32_t)pos) | 0x80000000); - WREG32_NO_KIQ(mmMM_INDEX_HI, pos >> 31); + if (tmp != hi) { + WREG32_NO_KIQ(mmMM_INDEX_HI, tmp); + hi = tmp; + } if (write) WREG32_NO_KIQ(mmMM_DATA, *buf++); else *buf++ = RREG32_NO_KIQ(mmMM_DATA); - spin_unlock_irqrestore(&adev->mmio_idx_lock, flags); } + spin_unlock_irqrestore(&adev->mmio_idx_lock, flags); } /* @@ -275,6 +306,26 @@ void amdgpu_mm_wreg8(struct amdgpu_device *adev, uint32_t offset, uint8_t value) BUG(); } +void static inline amdgpu_mm_wreg_mmio(struct amdgpu_device *adev, uint32_t reg, uint32_t v, uint32_t acc_flags) +{ + trace_amdgpu_mm_wreg(adev->pdev->device, reg, v); + + if ((reg * 4) < adev->rmmio_size && !(acc_flags & AMDGPU_REGS_IDX)) + writel(v, ((void __iomem *)adev->rmmio) + (reg * 4)); + else { + unsigned long flags; + + spin_lock_irqsave(&adev->mmio_idx_lock, flags); + writel((reg * 4), ((void __iomem *)adev->rmmio) + (mmMM_INDEX * 4)); + writel(v, ((void __iomem *)adev->rmmio) + (mmMM_DATA * 4)); + spin_unlock_irqrestore(&adev->mmio_idx_lock, flags); + } + + if (adev->asic_type >= CHIP_VEGA10 && reg == 1 && adev->last_mm_index == 0x5702C) { + udelay(500); + } +} + /** * amdgpu_mm_wreg - write to a memory mapped IO register * @@ -288,8 +339,6 @@ void amdgpu_mm_wreg8(struct amdgpu_device *adev, uint32_t offset, uint8_t value) void amdgpu_mm_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v, uint32_t acc_flags) { - trace_amdgpu_mm_wreg(adev->pdev->device, reg, v); - if (adev->asic_type >= CHIP_VEGA10 && reg == 0) { adev->last_mm_index = v; } @@ -297,20 +346,26 @@ void amdgpu_mm_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v, if ((acc_flags & AMDGPU_REGS_KIQ) || (!(acc_flags & AMDGPU_REGS_NO_KIQ) && amdgpu_sriov_runtime(adev))) return amdgpu_kiq_wreg(adev, reg, v); - if ((reg * 4) < adev->rmmio_size && !(acc_flags & AMDGPU_REGS_IDX)) - writel(v, ((void __iomem *)adev->rmmio) + (reg * 4)); - else { - unsigned long flags; + amdgpu_mm_wreg_mmio(adev, reg, v, acc_flags); +} - spin_lock_irqsave(&adev->mmio_idx_lock, flags); - writel((reg * 4), ((void __iomem *)adev->rmmio) + (mmMM_INDEX * 4)); - writel(v, ((void __iomem *)adev->rmmio) + (mmMM_DATA * 4)); - spin_unlock_irqrestore(&adev->mmio_idx_lock, flags); - } +/* + * amdgpu_mm_wreg_mmio_rlc - write register either with mmio or with RLC path if in range + * + * this function is invoked only the debugfs register access + * */ +void amdgpu_mm_wreg_mmio_rlc(struct amdgpu_device *adev, uint32_t reg, uint32_t v, + uint32_t acc_flags) +{ + if (amdgpu_sriov_fullaccess(adev) && + adev->gfx.rlc.funcs && + adev->gfx.rlc.funcs->is_rlcg_access_range) { - if (adev->asic_type >= CHIP_VEGA10 && reg == 1 && adev->last_mm_index == 0x5702C) { - udelay(500); + if (adev->gfx.rlc.funcs->is_rlcg_access_range(adev, reg)) + return adev->gfx.rlc.funcs->rlcg_wreg(adev, reg, v); } + + amdgpu_mm_wreg_mmio(adev, reg, v, acc_flags); } /** @@ -2285,8 +2340,6 @@ static int amdgpu_device_ip_suspend_phase1(struct amdgpu_device *adev) { int i, r; - amdgpu_device_set_pg_state(adev, AMD_PG_STATE_UNGATE); - amdgpu_device_set_cg_state(adev, AMD_CG_STATE_UNGATE); for (i = adev->num_ip_blocks - 1; i >= 0; i--) { if (!adev->ip_blocks[i].status.valid) @@ -2344,15 +2397,16 @@ static int amdgpu_device_ip_suspend_phase2(struct amdgpu_device *adev) } adev->ip_blocks[i].status.hw = false; /* handle putting the SMC in the appropriate state */ - if (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_SMC) { - r = amdgpu_dpm_set_mp1_state(adev, adev->mp1_state); - if (r) { - DRM_ERROR("SMC failed to set mp1 state %d, %d\n", - adev->mp1_state, r); - return r; + if(!amdgpu_sriov_vf(adev)){ + if (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_SMC) { + r = amdgpu_dpm_set_mp1_state(adev, adev->mp1_state); + if (r) { + DRM_ERROR("SMC failed to set mp1 state %d, %d\n", + adev->mp1_state, r); + return r; + } } } - adev->ip_blocks[i].status.hw = false; } @@ -2686,6 +2740,9 @@ static void amdgpu_device_xgmi_reset_func(struct work_struct *__work) if (adev->asic_reset_res) goto fail; + + if (adev->mmhub.funcs && adev->mmhub.funcs->reset_ras_error_count) + adev->mmhub.funcs->reset_ras_error_count(adev); } else { task_barrier_full(&hive->tb); @@ -2800,7 +2857,7 @@ int amdgpu_device_init(struct amdgpu_device *adev, adev->usec_timeout = AMDGPU_MAX_USEC_TIMEOUT; if (amdgpu_emu_mode == 1) - adev->usec_timeout *= 2; + adev->usec_timeout *= 10; adev->gmc.gart_size = 512 * 1024 * 1024; adev->accel_working = false; adev->num_rings = 0; @@ -3088,22 +3145,6 @@ fence_driver_init: } else adev->ucode_sysfs_en = true; - r = amdgpu_debugfs_gem_init(adev); - if (r) - DRM_ERROR("registering gem debugfs failed (%d).\n", r); - - r = amdgpu_debugfs_regs_init(adev); - if (r) - DRM_ERROR("registering register debugfs failed (%d).\n", r); - - r = amdgpu_debugfs_firmware_init(adev); - if (r) - DRM_ERROR("registering firmware debugfs failed (%d).\n", r); - - r = amdgpu_debugfs_init(adev); - if (r) - DRM_ERROR("Creating debugfs files failed (%d).\n", r); - if ((amdgpu_testing & 1)) { if (adev->accel_working) amdgpu_test_moves(adev); @@ -3177,6 +3218,12 @@ void amdgpu_device_fini(struct amdgpu_device *adev) flush_delayed_work(&adev->delayed_init_work); adev->shutdown = true; + /* make sure IB test finished before entering exclusive mode + * to avoid preemption on IB test + * */ + if (amdgpu_sriov_vf(adev)) + amdgpu_virt_request_full_gpu(adev, false); + /* disable all interrupts */ amdgpu_irq_disable_all(adev); if (adev->mode_info.mode_config_initialized){ @@ -3219,13 +3266,11 @@ void amdgpu_device_fini(struct amdgpu_device *adev) adev->rmmio = NULL; amdgpu_device_doorbell_fini(adev); - amdgpu_debugfs_regs_cleanup(adev); device_remove_file(adev->dev, &dev_attr_pcie_replay_count); if (adev->ucode_sysfs_en) amdgpu_ucode_sysfs_fini(adev); if (IS_ENABLED(CONFIG_PERF_EVENTS)) amdgpu_pmu_fini(adev); - amdgpu_debugfs_preempt_cleanup(adev); if (amdgpu_discovery && adev->asic_type >= CHIP_NAVI10) amdgpu_discovery_fini(adev); } @@ -3309,7 +3354,10 @@ int amdgpu_device_suspend(struct drm_device *dev, bool fbcon) } } - amdgpu_amdkfd_suspend(adev); + amdgpu_device_set_pg_state(adev, AMD_PG_STATE_UNGATE); + amdgpu_device_set_cg_state(adev, AMD_CG_STATE_UNGATE); + + amdgpu_amdkfd_suspend(adev, !fbcon); amdgpu_ras_suspend(adev); @@ -3393,7 +3441,7 @@ int amdgpu_device_resume(struct drm_device *dev, bool fbcon) } } } - r = amdgpu_amdkfd_resume(adev); + r = amdgpu_amdkfd_resume(adev, !fbcon); if (r) return r; @@ -3866,8 +3914,15 @@ static int amdgpu_do_asic_reset(struct amdgpu_hive_info *hive, } } - if (!r && amdgpu_ras_intr_triggered()) + if (!r && amdgpu_ras_intr_triggered()) { + list_for_each_entry(tmp_adev, device_list_handle, gmc.xgmi.head) { + if (tmp_adev->mmhub.funcs && + tmp_adev->mmhub.funcs->reset_ras_error_count) + tmp_adev->mmhub.funcs->reset_ras_error_count(tmp_adev); + } + amdgpu_ras_intr_cleared(); + } list_for_each_entry(tmp_adev, device_list_handle, gmc.xgmi.head) { if (need_full_reset) { @@ -3913,6 +3968,8 @@ static int amdgpu_do_asic_reset(struct amdgpu_hive_info *hive, if (r) goto out; + amdgpu_fbdev_set_suspend(tmp_adev, 0); + /* must succeed. */ amdgpu_ras_resume(tmp_adev); @@ -4086,6 +4143,8 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, */ amdgpu_unregister_gpu_instance(tmp_adev); + amdgpu_fbdev_set_suspend(adev, 1); + /* disable ras on ALL IPs */ if (!(in_ras_intr && !use_baco) && amdgpu_device_ip_need_full_reset(tmp_adev)) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c index f95092741c38..27d8ae19a7a4 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c @@ -307,7 +307,7 @@ int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev) for (hw_ip = 0; hw_ip < MAX_HWIP; hw_ip++) { if (hw_id_map[hw_ip] == le16_to_cpu(ip->hw_id)) { - DRM_INFO("set register base offset for %s\n", + DRM_DEBUG("set register base offset for %s\n", hw_id_names[le16_to_cpu(ip->hw_id)]); adev->reg_offset[hw_ip][ip->number_instance] = ip->base_address; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c index 6d520a3eec40..84cee27cd7ef 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c @@ -99,7 +99,7 @@ static void amdgpu_display_flip_work_func(struct work_struct *__work) & (DRM_SCANOUTPOS_VALID | DRM_SCANOUTPOS_IN_VBLANK)) == (DRM_SCANOUTPOS_VALID | DRM_SCANOUTPOS_IN_VBLANK) && (int)(work->target_vblank - - amdgpu_get_vblank_counter_kms(adev->ddev, amdgpu_crtc->crtc_id)) > 0) { + amdgpu_get_vblank_counter_kms(crtc)) > 0) { schedule_delayed_work(&work->flip_work, usecs_to_jiffies(1000)); return; } @@ -219,7 +219,7 @@ int amdgpu_display_crtc_page_flip_target(struct drm_crtc *crtc, if (!adev->enable_virtual_display) work->base = amdgpu_bo_gpu_offset(new_abo); work->target_vblank = target - (uint32_t)drm_crtc_vblank_count(crtc) + - amdgpu_get_vblank_counter_kms(dev, work->crtc_id); + amdgpu_get_vblank_counter_kms(crtc); /* we borrow the event spin lock for protecting flip_wrok */ spin_lock_irqsave(&crtc->dev->event_lock, flags); @@ -924,3 +924,15 @@ int amdgpu_display_crtc_idx_to_irq_type(struct amdgpu_device *adev, int crtc) return AMDGPU_CRTC_IRQ_NONE; } } + +bool amdgpu_crtc_get_scanout_position(struct drm_crtc *crtc, + bool in_vblank_irq, int *vpos, + int *hpos, ktime_t *stime, ktime_t *etime, + const struct drm_display_mode *mode) +{ + struct drm_device *dev = crtc->dev; + unsigned int pipe = crtc->index; + + return amdgpu_display_get_crtc_scanoutpos(dev, pipe, 0, vpos, hpos, + stime, etime, mode); +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c index a59cd47aa6c1..ffeb20f11c07 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c @@ -223,6 +223,37 @@ static void amdgpu_dma_buf_detach(struct dma_buf *dmabuf, } /** + * amdgpu_dma_buf_pin - &dma_buf_ops.pin implementation + * + * @attach: attachment to pin down + * + * Pin the BO which is backing the DMA-buf so that it can't move any more. + */ +static int amdgpu_dma_buf_pin(struct dma_buf_attachment *attach) +{ + struct drm_gem_object *obj = attach->dmabuf->priv; + struct amdgpu_bo *bo = gem_to_amdgpu_bo(obj); + + /* pin buffer into GTT */ + return amdgpu_bo_pin(bo, AMDGPU_GEM_DOMAIN_GTT); +} + +/** + * amdgpu_dma_buf_unpin - &dma_buf_ops.unpin implementation + * + * @attach: attachment to unpin + * + * Unpin a previously pinned BO to make it movable again. + */ +static void amdgpu_dma_buf_unpin(struct dma_buf_attachment *attach) +{ + struct drm_gem_object *obj = attach->dmabuf->priv; + struct amdgpu_bo *bo = gem_to_amdgpu_bo(obj); + + amdgpu_bo_unpin(bo); +} + +/** * amdgpu_dma_buf_map - &dma_buf_ops.map_dma_buf implementation * @attach: DMA-buf attachment * @dir: DMA direction @@ -244,9 +275,19 @@ static struct sg_table *amdgpu_dma_buf_map(struct dma_buf_attachment *attach, struct sg_table *sgt; long r; - r = amdgpu_bo_pin(bo, AMDGPU_GEM_DOMAIN_GTT); - if (r) - return ERR_PTR(r); + if (!bo->pin_count) { + /* move buffer into GTT */ + struct ttm_operation_ctx ctx = { false, false }; + + amdgpu_bo_placement_from_domain(bo, AMDGPU_GEM_DOMAIN_GTT); + r = ttm_bo_validate(&bo->tbo, &bo->placement, &ctx); + if (r) + return ERR_PTR(r); + + } else if (!(amdgpu_mem_type_to_domain(bo->tbo.mem.mem_type) & + AMDGPU_GEM_DOMAIN_GTT)) { + return ERR_PTR(-EBUSY); + } sgt = drm_prime_pages_to_sg(bo->tbo.ttm->pages, bo->tbo.num_pages); if (IS_ERR(sgt)) @@ -277,13 +318,9 @@ static void amdgpu_dma_buf_unmap(struct dma_buf_attachment *attach, struct sg_table *sgt, enum dma_data_direction dir) { - struct drm_gem_object *obj = attach->dmabuf->priv; - struct amdgpu_bo *bo = gem_to_amdgpu_bo(obj); - dma_unmap_sg(attach->dev, sgt->sgl, sgt->nents, dir); sg_free_table(sgt); kfree(sgt); - amdgpu_bo_unpin(bo); } /** @@ -327,9 +364,10 @@ static int amdgpu_dma_buf_begin_cpu_access(struct dma_buf *dma_buf, } const struct dma_buf_ops amdgpu_dmabuf_ops = { - .dynamic_mapping = true, .attach = amdgpu_dma_buf_attach, .detach = amdgpu_dma_buf_detach, + .pin = amdgpu_dma_buf_pin, + .unpin = amdgpu_dma_buf_unpin, .map_dma_buf = amdgpu_dma_buf_map, .unmap_dma_buf = amdgpu_dma_buf_unmap, .release = drm_gem_dmabuf_release, @@ -413,6 +451,73 @@ error: } /** + * amdgpu_dma_buf_move_notify - &attach.move_notify implementation + * + * @attach: the DMA-buf attachment + * + * Invalidate the DMA-buf attachment, making sure that the we re-create the + * mapping before the next use. + */ +static void +amdgpu_dma_buf_move_notify(struct dma_buf_attachment *attach) +{ + struct drm_gem_object *obj = attach->importer_priv; + struct ww_acquire_ctx *ticket = dma_resv_locking_ctx(obj->resv); + struct amdgpu_bo *bo = gem_to_amdgpu_bo(obj); + struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev); + struct ttm_operation_ctx ctx = { false, false }; + struct ttm_placement placement = {}; + struct amdgpu_vm_bo_base *bo_base; + int r; + + if (bo->tbo.mem.mem_type == TTM_PL_SYSTEM) + return; + + r = ttm_bo_validate(&bo->tbo, &placement, &ctx); + if (r) { + DRM_ERROR("Failed to invalidate DMA-buf import (%d))\n", r); + return; + } + + for (bo_base = bo->vm_bo; bo_base; bo_base = bo_base->next) { + struct amdgpu_vm *vm = bo_base->vm; + struct dma_resv *resv = vm->root.base.bo->tbo.base.resv; + + if (ticket) { + /* When we get an error here it means that somebody + * else is holding the VM lock and updating page tables + * So we can just continue here. + */ + r = dma_resv_lock(resv, ticket); + if (r) + continue; + + } else { + /* TODO: This is more problematic and we actually need + * to allow page tables updates without holding the + * lock. + */ + if (!dma_resv_trylock(resv)) + continue; + } + + r = amdgpu_vm_clear_freed(adev, vm, NULL); + if (!r) + r = amdgpu_vm_handle_moved(adev, vm); + + if (r && r != -EBUSY) + DRM_ERROR("Failed to invalidate VM page tables (%d))\n", + r); + + dma_resv_unlock(resv); + } +} + +static const struct dma_buf_attach_ops amdgpu_dma_buf_attach_ops = { + .move_notify = amdgpu_dma_buf_move_notify +}; + +/** * amdgpu_gem_prime_import - &drm_driver.gem_prime_import implementation * @dev: DRM device * @dma_buf: Shared DMA buffer @@ -444,7 +549,8 @@ struct drm_gem_object *amdgpu_gem_prime_import(struct drm_device *dev, if (IS_ERR(obj)) return obj; - attach = dma_buf_dynamic_attach(dma_buf, dev->dev, true); + attach = dma_buf_dynamic_attach(dma_buf, dev->dev, + &amdgpu_dma_buf_attach_ops, obj); if (IS_ERR(attach)) { drm_gem_object_put(obj); return ERR_CAST(attach); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.c index a2e8c3dfb4f1..ba1bb95a3cf9 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.c @@ -1171,3 +1171,20 @@ int amdgpu_dpm_set_xgmi_pstate(struct amdgpu_device *adev, return ret; } + +int amdgpu_dpm_set_df_cstate(struct amdgpu_device *adev, + uint32_t cstate) +{ + int ret = 0; + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + void *pp_handle = adev->powerplay.pp_handle; + struct smu_context *smu = &adev->smu; + + if (is_support_sw_smu(adev)) + ret = smu_set_df_cstate(smu, cstate); + else if (pp_funcs && + pp_funcs->set_df_cstate) + ret = pp_funcs->set_df_cstate(pp_handle, cstate); + + return ret; +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.h index 902ca6c00cca..936d85aa0fbc 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.h @@ -448,6 +448,8 @@ struct amdgpu_pm { /* powerplay feature */ uint32_t pp_feature; + /* Used for I2C access to various EEPROMs on relevant ASICs */ + struct i2c_adapter smu_i2c; }; #define R600_SSTU_DFLT 0 @@ -533,4 +535,7 @@ int amdgpu_dpm_baco_exit(struct amdgpu_device *adev); int amdgpu_dpm_baco_enter(struct amdgpu_device *adev); +int amdgpu_dpm_set_df_cstate(struct amdgpu_device *adev, + uint32_t cstate); + #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 94e2fd758e01..8ea86ffdea0d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -1021,6 +1021,7 @@ static int amdgpu_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent) { struct drm_device *dev; + struct amdgpu_device *adev; unsigned long flags = ent->driver_data; int ret, retry = 0; bool supports_atomic = false; @@ -1090,6 +1091,8 @@ static int amdgpu_pci_probe(struct pci_dev *pdev, pci_set_drvdata(pdev, dev); + amdgpu_driver_load_kms(dev, ent->driver_data); + retry_init: ret = drm_dev_register(dev, ent->driver_data); if (ret == -EAGAIN && ++retry <= 3) { @@ -1100,6 +1103,11 @@ retry_init: } else if (ret) goto err_pci; + adev = dev->dev_private; + ret = amdgpu_debugfs_init(adev); + if (ret) + DRM_ERROR("Creating debugfs files failed (%d).\n", ret); + return 0; err_pci: @@ -1119,9 +1127,10 @@ amdgpu_pci_remove(struct pci_dev *pdev) #endif DRM_ERROR("Hotplug removal is not supported\n"); drm_dev_unplug(dev); - drm_dev_put(dev); + amdgpu_driver_unload_kms(dev); pci_disable_device(pdev); pci_set_drvdata(pdev, NULL); + drm_dev_put(dev); } static void @@ -1220,11 +1229,15 @@ static int amdgpu_pmops_runtime_suspend(struct device *dev) } } + adev->in_runpm = true; if (amdgpu_device_supports_boco(drm_dev)) drm_dev->switch_power_state = DRM_SWITCH_POWER_CHANGING; drm_kms_helper_poll_disable(drm_dev); ret = amdgpu_device_suspend(drm_dev, false); + if (ret) + return ret; + if (amdgpu_device_supports_boco(drm_dev)) { /* Only need to handle PCI state in the driver for ATPX * PCI core handles it for _PR3. @@ -1278,6 +1291,7 @@ static int amdgpu_pmops_runtime_resume(struct device *dev) drm_kms_helper_poll_enable(drm_dev); if (amdgpu_device_supports_boco(drm_dev)) drm_dev->switch_power_state = DRM_SWITCH_POWER_ON; + adev->in_runpm = false; return 0; } @@ -1285,24 +1299,55 @@ static int amdgpu_pmops_runtime_idle(struct device *dev) { struct drm_device *drm_dev = dev_get_drvdata(dev); struct amdgpu_device *adev = drm_dev->dev_private; - struct drm_crtc *crtc; + /* we don't want the main rpm_idle to call suspend - we want to autosuspend */ + int ret = 1; if (!adev->runpm) { pm_runtime_forbid(dev); return -EBUSY; } - list_for_each_entry(crtc, &drm_dev->mode_config.crtc_list, head) { - if (crtc->enabled) { - DRM_DEBUG_DRIVER("failing to power off - crtc active\n"); - return -EBUSY; + if (amdgpu_device_has_dc_support(adev)) { + struct drm_crtc *crtc; + + drm_modeset_lock_all(drm_dev); + + drm_for_each_crtc(crtc, drm_dev) { + if (crtc->state->active) { + ret = -EBUSY; + break; + } + } + + drm_modeset_unlock_all(drm_dev); + + } else { + struct drm_connector *list_connector; + struct drm_connector_list_iter iter; + + mutex_lock(&drm_dev->mode_config.mutex); + drm_modeset_lock(&drm_dev->mode_config.connection_mutex, NULL); + + drm_connector_list_iter_begin(drm_dev, &iter); + drm_for_each_connector_iter(list_connector, &iter) { + if (list_connector->dpms == DRM_MODE_DPMS_ON) { + ret = -EBUSY; + break; + } } + + drm_connector_list_iter_end(&iter); + + drm_modeset_unlock(&drm_dev->mode_config.connection_mutex); + mutex_unlock(&drm_dev->mode_config.mutex); } + if (ret == -EBUSY) + DRM_DEBUG_DRIVER("failing to power off - crtc active\n"); + pm_runtime_mark_last_busy(dev); pm_runtime_autosuspend(dev); - /* we don't want the main rpm_idle to call suspend - we want to autosuspend */ - return 1; + return ret; } long amdgpu_drm_ioctl(struct file *filp, @@ -1377,32 +1422,15 @@ int amdgpu_file_to_fpriv(struct file *filp, struct amdgpu_fpriv **fpriv) return 0; } -static bool -amdgpu_get_crtc_scanout_position(struct drm_device *dev, unsigned int pipe, - bool in_vblank_irq, int *vpos, int *hpos, - ktime_t *stime, ktime_t *etime, - const struct drm_display_mode *mode) -{ - return amdgpu_display_get_crtc_scanoutpos(dev, pipe, 0, vpos, hpos, - stime, etime, mode); -} - static struct drm_driver kms_driver = { .driver_features = - DRIVER_USE_AGP | DRIVER_ATOMIC | + DRIVER_ATOMIC | DRIVER_GEM | DRIVER_RENDER | DRIVER_MODESET | DRIVER_SYNCOBJ | DRIVER_SYNCOBJ_TIMELINE, - .load = amdgpu_driver_load_kms, .open = amdgpu_driver_open_kms, .postclose = amdgpu_driver_postclose_kms, .lastclose = amdgpu_driver_lastclose_kms, - .unload = amdgpu_driver_unload_kms, - .get_vblank_counter = amdgpu_get_vblank_counter_kms, - .enable_vblank = amdgpu_enable_vblank_kms, - .disable_vblank = amdgpu_disable_vblank_kms, - .get_vblank_timestamp = drm_calc_vbltimestamp_from_scanoutpos, - .get_scanout_position = amdgpu_get_crtc_scanout_position, .irq_handler = amdgpu_irq_handler, .ioctls = amdgpu_ioctls_kms, .gem_free_object_unlocked = amdgpu_gem_object_free, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c index 2672dc64a310..9ae7b61f696a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c @@ -336,15 +336,12 @@ int amdgpu_fbdev_init(struct amdgpu_device *adev) drm_fb_helper_prepare(adev->ddev, &rfbdev->helper, &amdgpu_fb_helper_funcs); - ret = drm_fb_helper_init(adev->ddev, &rfbdev->helper, - AMDGPUFB_CONN_LIMIT); + ret = drm_fb_helper_init(adev->ddev, &rfbdev->helper); if (ret) { kfree(rfbdev); return ret; } - drm_fb_helper_single_add_all_connectors(&rfbdev->helper); - /* disable all the possible outputs/crtcs before entering KMS mode */ if (!amdgpu_device_has_dc_support(adev)) drm_helper_disable_unused_functions(adev->ddev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c index 3c01252b1e0e..7531527067df 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c @@ -503,9 +503,6 @@ int amdgpu_fence_driver_init_ring(struct amdgpu_ring *ring, */ int amdgpu_fence_driver_init(struct amdgpu_device *adev) { - if (amdgpu_debugfs_fence_init(adev)) - dev_err(adev->dev, "fence debugfs file creation failed\n"); - return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c index 0f960b498792..6b9c9193cdfa 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c @@ -192,6 +192,14 @@ static bool amdgpu_gfx_is_multipipe_capable(struct amdgpu_device *adev) return adev->gfx.mec.num_mec > 1; } +bool amdgpu_gfx_is_high_priority_compute_queue(struct amdgpu_device *adev, + int queue) +{ + /* Policy: make queue 0 of each pipe as high priority compute queue */ + return (queue == 0); + +} + void amdgpu_gfx_compute_queue_acquire(struct amdgpu_device *adev) { int i, queue, pipe, mec; @@ -477,7 +485,7 @@ int amdgpu_gfx_disable_kcq(struct amdgpu_device *adev) kiq->pmf->kiq_unmap_queues(kiq_ring, &adev->gfx.compute_ring[i], RESET_QUEUES, 0, 0); - return amdgpu_ring_test_ring(kiq_ring); + return amdgpu_ring_test_helper(kiq_ring); } int amdgpu_gfx_enable_kcq(struct amdgpu_device *adev) @@ -565,7 +573,6 @@ int amdgpu_gfx_ras_late_init(struct amdgpu_device *adev) int r; struct ras_fs_if fs_info = { .sysfs_name = "gfx_err_count", - .debugfs_name = "gfx_err_inject", }; struct ras_ih_if ih_info = { .cb = amdgpu_gfx_process_ras_data_cb, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h index ca17ffb01301..5825692d07e4 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h @@ -41,6 +41,15 @@ #define AMDGPU_MAX_GFX_QUEUES KGD_MAX_QUEUES #define AMDGPU_MAX_COMPUTE_QUEUES KGD_MAX_QUEUES +enum gfx_pipe_priority { + AMDGPU_GFX_PIPE_PRIO_NORMAL = 1, + AMDGPU_GFX_PIPE_PRIO_HIGH, + AMDGPU_GFX_PIPE_PRIO_MAX +}; + +#define AMDGPU_GFX_QUEUE_PRIORITY_MINIMUM 0 +#define AMDGPU_GFX_QUEUE_PRIORITY_MAXIMUM 15 + struct amdgpu_mec { struct amdgpu_bo *hpd_eop_obj; u64 hpd_eop_gpu_addr; @@ -151,6 +160,8 @@ struct amdgpu_gfx_config { unsigned num_gpus; unsigned multi_gpu_tile_size; unsigned mc_arb_ramcfg; + unsigned num_banks; + unsigned num_ranks; unsigned gb_addr_config; unsigned num_rbs; unsigned gs_vgt_table_depth; @@ -204,6 +215,7 @@ struct amdgpu_gfx_funcs { u32 queue, u32 vmid); int (*ras_error_inject)(struct amdgpu_device *adev, void *inject_if); int (*query_ras_error_count) (struct amdgpu_device *adev, void *ras_error_status); + void (*reset_ras_error_count) (struct amdgpu_device *adev); }; struct sq_work { @@ -278,8 +290,9 @@ struct amdgpu_gfx { uint32_t num_gfx_sched; unsigned num_gfx_rings; struct amdgpu_ring compute_ring[AMDGPU_MAX_COMPUTE_RINGS]; + struct drm_gpu_scheduler **compute_prio_sched[AMDGPU_GFX_PIPE_PRIO_MAX]; struct drm_gpu_scheduler *compute_sched[AMDGPU_MAX_COMPUTE_RINGS]; - uint32_t num_compute_sched; + uint32_t num_compute_sched[AMDGPU_GFX_PIPE_PRIO_MAX]; unsigned num_compute_rings; struct amdgpu_irq_src eop_irq; struct amdgpu_irq_src priv_reg_irq; @@ -361,6 +374,8 @@ void amdgpu_gfx_bit_to_mec_queue(struct amdgpu_device *adev, int bit, int *mec, int *pipe, int *queue); bool amdgpu_gfx_is_mec_queue_enabled(struct amdgpu_device *adev, int mec, int pipe, int queue); +bool amdgpu_gfx_is_high_priority_compute_queue(struct amdgpu_device *adev, + int queue); int amdgpu_gfx_me_queue_to_bit(struct amdgpu_device *adev, int me, int pipe, int queue); void amdgpu_gfx_bit_to_me_queue(struct amdgpu_device *adev, int bit, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h index d3c27a3c43f6..7546da0cc70c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h @@ -195,6 +195,7 @@ struct amdgpu_gmc { uint32_t srbm_soft_reset; bool prt_warning; uint64_t stolen_size; + uint32_t sdpif_register; /* apertures */ u64 shared_aperture_start; u64 shared_aperture_end; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c index 60655834d649..ccbd7acfc4cb 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c @@ -48,7 +48,6 @@ * produce command buffers which are send to the kernel and * put in IBs for execution by the requested ring. */ -static int amdgpu_debugfs_sa_init(struct amdgpu_device *adev); /** * amdgpu_ib_get - request an IB (Indirect Buffer) @@ -295,9 +294,7 @@ int amdgpu_ib_pool_init(struct amdgpu_device *adev) } adev->ib_pool_ready = true; - if (amdgpu_debugfs_sa_init(adev)) { - dev_err(adev->dev, "failed to register debugfs file for SA\n"); - } + return 0; } @@ -421,7 +418,7 @@ static const struct drm_info_list amdgpu_debugfs_sa_list[] = { #endif -static int amdgpu_debugfs_sa_init(struct amdgpu_device *adev) +int amdgpu_debugfs_sa_init(struct amdgpu_device *adev) { #if defined(CONFIG_DEBUG_FS) return amdgpu_debugfs_add_files(adev, amdgpu_debugfs_sa_list, 1); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c index d42be880a236..4981e443a884 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c @@ -117,12 +117,10 @@ void amdgpu_job_free_resources(struct amdgpu_job *job) static void amdgpu_job_free_cb(struct drm_sched_job *s_job) { - struct amdgpu_ring *ring = to_amdgpu_ring(s_job->sched); struct amdgpu_job *job = to_amdgpu_job(s_job); drm_sched_job_cleanup(s_job); - amdgpu_ring_priority_put(ring, s_job->s_priority); dma_fence_put(job->fence); amdgpu_sync_free(&job->sync); amdgpu_sync_free(&job->sched_sync); @@ -143,7 +141,6 @@ int amdgpu_job_submit(struct amdgpu_job *job, struct drm_sched_entity *entity, void *owner, struct dma_fence **f) { enum drm_sched_priority priority; - struct amdgpu_ring *ring; int r; if (!f) @@ -158,9 +155,6 @@ int amdgpu_job_submit(struct amdgpu_job *job, struct drm_sched_entity *entity, priority = job->base.s_priority; drm_sched_entity_push_job(&job->base, entity); - ring = to_amdgpu_ring(entity->rq->sched); - amdgpu_ring_priority_get(ring, priority); - return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c index 60591dbc2097..fd1dc3236eca 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c @@ -88,9 +88,6 @@ void amdgpu_driver_unload_kms(struct drm_device *dev) if (adev->rmmio == NULL) goto done_free; - if (amdgpu_sriov_vf(adev)) - amdgpu_virt_request_full_gpu(adev, false); - if (adev->runpm) { pm_runtime_get_sync(dev->dev); pm_runtime_forbid(dev->dev); @@ -170,10 +167,17 @@ int amdgpu_driver_load_kms(struct drm_device *dev, unsigned long flags) } if (amdgpu_device_supports_boco(dev) && - (amdgpu_runtime_pm != 0)) /* enable runpm by default */ + (amdgpu_runtime_pm != 0)) /* enable runpm by default for boco */ + adev->runpm = true; + else if (amdgpu_device_supports_baco(dev) && + (amdgpu_runtime_pm != 0) && + (adev->asic_type >= CHIP_TOPAZ) && + (adev->asic_type != CHIP_VEGA10) && + (adev->asic_type != CHIP_VEGA20) && + (adev->asic_type != CHIP_ARCTURUS)) /* enable runpm on VI+ */ adev->runpm = true; else if (amdgpu_device_supports_baco(dev) && - (amdgpu_runtime_pm > 0)) /* enable runpm if runpm=1 */ + (amdgpu_runtime_pm > 0)) /* enable runpm if runpm=1 on CI */ adev->runpm = true; /* Call ACPI methods: require modeset init @@ -1110,14 +1114,15 @@ void amdgpu_driver_postclose_kms(struct drm_device *dev, /** * amdgpu_get_vblank_counter_kms - get frame count * - * @dev: drm dev pointer - * @pipe: crtc to get the frame count from + * @crtc: crtc to get the frame count from * * Gets the frame count on the requested crtc (all asics). * Returns frame count on success, -EINVAL on failure. */ -u32 amdgpu_get_vblank_counter_kms(struct drm_device *dev, unsigned int pipe) +u32 amdgpu_get_vblank_counter_kms(struct drm_crtc *crtc) { + struct drm_device *dev = crtc->dev; + unsigned int pipe = crtc->index; struct amdgpu_device *adev = dev->dev_private; int vpos, hpos, stat; u32 count; @@ -1177,14 +1182,15 @@ u32 amdgpu_get_vblank_counter_kms(struct drm_device *dev, unsigned int pipe) /** * amdgpu_enable_vblank_kms - enable vblank interrupt * - * @dev: drm dev pointer - * @pipe: crtc to enable vblank interrupt for + * @crtc: crtc to enable vblank interrupt for * * Enable the interrupt on the requested crtc (all asics). * Returns 0 on success, -EINVAL on failure. */ -int amdgpu_enable_vblank_kms(struct drm_device *dev, unsigned int pipe) +int amdgpu_enable_vblank_kms(struct drm_crtc *crtc) { + struct drm_device *dev = crtc->dev; + unsigned int pipe = crtc->index; struct amdgpu_device *adev = dev->dev_private; int idx = amdgpu_display_crtc_idx_to_irq_type(adev, pipe); @@ -1194,13 +1200,14 @@ int amdgpu_enable_vblank_kms(struct drm_device *dev, unsigned int pipe) /** * amdgpu_disable_vblank_kms - disable vblank interrupt * - * @dev: drm dev pointer - * @pipe: crtc to disable vblank interrupt for + * @crtc: crtc to disable vblank interrupt for * * Disable the interrupt on the requested crtc (all asics). */ -void amdgpu_disable_vblank_kms(struct drm_device *dev, unsigned int pipe) +void amdgpu_disable_vblank_kms(struct drm_crtc *crtc) { + struct drm_device *dev = crtc->dev; + unsigned int pipe = crtc->index; struct amdgpu_device *adev = dev->dev_private; int idx = amdgpu_display_crtc_idx_to_irq_type(adev, pipe); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.c index 676c48c02d77..ead3dc572ec5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.c @@ -32,7 +32,6 @@ int amdgpu_mmhub_ras_late_init(struct amdgpu_device *adev) }; struct ras_fs_if fs_info = { .sysfs_name = "mmhub_err_count", - .debugfs_name = "mmhub_err_inject", }; if (!adev->mmhub.ras_if) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h index 1cd78940cf82..e89fb35fec71 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h @@ -26,6 +26,7 @@ struct amdgpu_mmhub_funcs { int (*ras_late_init)(struct amdgpu_device *adev); void (*query_ras_error_count)(struct amdgpu_device *adev, void *ras_error_status); + void (*reset_ras_error_count)(struct amdgpu_device *adev); }; struct amdgpu_mmhub { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h index eb9975f4decb..37ba07e2feb5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h @@ -612,6 +612,11 @@ void amdgpu_panel_mode_fixup(struct drm_encoder *encoder, struct drm_display_mode *adjusted_mode); int amdgpu_display_crtc_idx_to_irq_type(struct amdgpu_device *adev, int crtc); +bool amdgpu_crtc_get_scanout_position(struct drm_crtc *crtc, + bool in_vblank_irq, int *vpos, + int *hpos, ktime_t *stime, ktime_t *etime, + const struct drm_display_mode *mode); + /* fbdev layer */ int amdgpu_fbdev_init(struct amdgpu_device *adev); void amdgpu_fbdev_fini(struct amdgpu_device *adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c index 7d5c3a9de9ea..6201a5f4b4fa 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c @@ -30,7 +30,6 @@ int amdgpu_nbio_ras_late_init(struct amdgpu_device *adev) }; struct ras_fs_if fs_info = { .sysfs_name = "pcie_bif_err_count", - .debugfs_name = "pcie_bif_err_inject", }; if (!adev->nbio.ras_if) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c index e3f16b49e970..c687f5415b3f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c @@ -31,6 +31,7 @@ */ #include <linux/list.h> #include <linux/slab.h> +#include <linux/dma-buf.h> #include <drm/amdgpu_drm.h> #include <drm/drm_cache.h> @@ -925,6 +926,9 @@ int amdgpu_bo_pin_restricted(struct amdgpu_bo *bo, u32 domain, return 0; } + if (bo->tbo.base.import_attach) + dma_buf_pin(bo->tbo.base.import_attach); + bo->flags |= AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS; /* force to pin into visible video ram */ if (!(bo->flags & AMDGPU_GEM_CREATE_NO_CPU_ACCESS)) @@ -1008,6 +1012,9 @@ int amdgpu_bo_unpin(struct amdgpu_bo *bo) amdgpu_bo_subtract_pin_size(bo); + if (bo->tbo.base.import_attach) + dma_buf_unpin(bo->tbo.base.import_attach); + for (i = 0; i < bo->placement.num_placement; i++) { bo->placements[i].lpfn = 0; bo->placements[i].flags &= ~TTM_PL_FLAG_NO_EVICT; @@ -1274,6 +1281,10 @@ void amdgpu_bo_move_notify(struct ttm_buffer_object *bo, amdgpu_bo_kunmap(abo); + if (abo->tbo.base.dma_buf && !abo->tbo.base.import_attach && + bo->mem.mem_type != TTM_PL_SYSTEM) + dma_buf_move_notify(abo->tbo.base.dma_buf); + /* remember the eviction */ if (evict) atomic64_inc(&adev->num_evictions); @@ -1307,6 +1318,12 @@ void amdgpu_bo_release_notify(struct ttm_buffer_object *bo) if (abo->kfd_bo) amdgpu_amdkfd_unreserve_memory_limit(abo); + /* We only remove the fence if the resv has individualized. */ + WARN_ON_ONCE(bo->type == ttm_bo_type_kernel + && bo->base.resv != &bo->base._resv); + if (bo->base.resv == &bo->base._resv) + amdgpu_amdkfd_remove_fence_on_pt_pd_bos(abo); + if (bo->mem.mem_type != TTM_PL_VRAM || !bo->mem.mm_node || !(abo->flags & AMDGPU_GEM_CREATE_VRAM_WIPE_ON_RELEASE)) return; @@ -1403,30 +1420,52 @@ void amdgpu_bo_fence(struct amdgpu_bo *bo, struct dma_fence *fence, } /** - * amdgpu_sync_wait_resv - Wait for BO reservation fences + * amdgpu_bo_sync_wait_resv - Wait for BO reservation fences * - * @bo: buffer object + * @adev: amdgpu device pointer + * @resv: reservation object to sync to + * @sync_mode: synchronization mode * @owner: fence owner * @intr: Whether the wait is interruptible * + * Extract the fences from the reservation object and waits for them to finish. + * * Returns: * 0 on success, errno otherwise. */ -int amdgpu_bo_sync_wait(struct amdgpu_bo *bo, void *owner, bool intr) +int amdgpu_bo_sync_wait_resv(struct amdgpu_device *adev, struct dma_resv *resv, + enum amdgpu_sync_mode sync_mode, void *owner, + bool intr) { - struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev); struct amdgpu_sync sync; int r; amdgpu_sync_create(&sync); - amdgpu_sync_resv(adev, &sync, bo->tbo.base.resv, owner, false); + amdgpu_sync_resv(adev, &sync, resv, sync_mode, owner); r = amdgpu_sync_wait(&sync, intr); amdgpu_sync_free(&sync); - return r; } /** + * amdgpu_bo_sync_wait - Wrapper for amdgpu_bo_sync_wait_resv + * @bo: buffer object to wait for + * @owner: fence owner + * @intr: Whether the wait is interruptible + * + * Wrapper to wait for fences in a BO. + * Returns: + * 0 on success, errno otherwise. + */ +int amdgpu_bo_sync_wait(struct amdgpu_bo *bo, void *owner, bool intr) +{ + struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev); + + return amdgpu_bo_sync_wait_resv(adev, bo->tbo.base.resv, + AMDGPU_SYNC_NE_OWNER, owner, intr); +} + +/** * amdgpu_bo_gpu_offset - return GPU offset of bo * @bo: amdgpu object for which we query the offset * diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h index 36dec51d1ef1..5e39ecd8cc28 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h @@ -277,6 +277,9 @@ void amdgpu_bo_release_notify(struct ttm_buffer_object *bo); int amdgpu_bo_fault_reserve_notify(struct ttm_buffer_object *bo); void amdgpu_bo_fence(struct amdgpu_bo *bo, struct dma_fence *fence, bool shared); +int amdgpu_bo_sync_wait_resv(struct amdgpu_device *adev, struct dma_resv *resv, + enum amdgpu_sync_mode sync_mode, void *owner, + bool intr); int amdgpu_bo_sync_wait(struct amdgpu_bo *bo, void *owner, bool intr); u64 amdgpu_bo_gpu_offset(struct amdgpu_bo *bo); int amdgpu_bo_validate(struct amdgpu_bo *bo); @@ -316,6 +319,7 @@ void amdgpu_sa_bo_free(struct amdgpu_device *adev, void amdgpu_sa_bo_dump_debug_info(struct amdgpu_sa_manager *sa_manager, struct seq_file *m); #endif +int amdgpu_debugfs_sa_init(struct amdgpu_device *adev); bool amdgpu_bo_support_uswc(u64 bo_flags); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c index b03b1eb7ba04..abe94a55ecad 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c @@ -41,8 +41,6 @@ #include "hwmgr.h" #define WIDTH_4K 3840 -static int amdgpu_debugfs_pm_init(struct amdgpu_device *adev); - static const struct cg_flag_name clocks[] = { {AMD_CG_SUPPORT_GFX_MGCG, "Graphics Medium Grain Clock Gating"}, {AMD_CG_SUPPORT_GFX_MGLS, "Graphics Medium Grain memory Light Sleep"}, @@ -91,9 +89,13 @@ void amdgpu_pm_acpi_event_handler(struct amdgpu_device *adev) adev->pm.ac_power = true; else adev->pm.ac_power = false; - if (adev->powerplay.pp_funcs->enable_bapm) + if (adev->powerplay.pp_funcs && + adev->powerplay.pp_funcs->enable_bapm) amdgpu_dpm_enable_bapm(adev, adev->pm.ac_power); mutex_unlock(&adev->pm.mutex); + + if (is_support_sw_smu(adev)) + smu_set_ac_dc(&adev->smu); } } @@ -3398,11 +3400,6 @@ int amdgpu_pm_sysfs_init(struct amdgpu_device *adev) DRM_ERROR("failed to create device file unique_id\n"); return ret; } - ret = amdgpu_debugfs_pm_init(adev); - if (ret) { - DRM_ERROR("Failed to register debugfs file for dpm!\n"); - return ret; - } if ((adev->asic_type >= CHIP_VEGA10) && !(adev->flags & AMD_IS_APU)) { @@ -3669,7 +3666,7 @@ static const struct drm_info_list amdgpu_pm_info_list[] = { }; #endif -static int amdgpu_debugfs_pm_init(struct amdgpu_device *adev) +int amdgpu_debugfs_pm_init(struct amdgpu_device *adev) { #if defined(CONFIG_DEBUG_FS) return amdgpu_debugfs_add_files(adev, amdgpu_pm_info_list, ARRAY_SIZE(amdgpu_pm_info_list)); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.h index 3da1da277805..5db0ef86e84c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.h @@ -43,4 +43,6 @@ void amdgpu_dpm_enable_uvd(struct amdgpu_device *adev, bool enable); void amdgpu_dpm_enable_vce(struct amdgpu_device *adev, bool enable); void amdgpu_dpm_enable_jpeg(struct amdgpu_device *adev, bool enable); +int amdgpu_debugfs_pm_init(struct amdgpu_device *adev); + #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c index 3a1570dafe34..deaa26808841 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c @@ -24,6 +24,7 @@ */ #include <linux/firmware.h> +#include <linux/dma-mapping.h> #include "amdgpu.h" #include "amdgpu_psp.h" @@ -38,6 +39,42 @@ static void psp_set_funcs(struct amdgpu_device *adev); +static int psp_sysfs_init(struct amdgpu_device *adev); +static void psp_sysfs_fini(struct amdgpu_device *adev); + +/* + * Due to DF Cstate management centralized to PMFW, the firmware + * loading sequence will be updated as below: + * - Load KDB + * - Load SYS_DRV + * - Load tOS + * - Load PMFW + * - Setup TMR + * - Load other non-psp fw + * - Load ASD + * - Load XGMI/RAS/HDCP/DTM TA if any + * + * This new sequence is required for + * - Arcturus + * - Navi12 and onwards + */ +static void psp_check_pmfw_centralized_cstate_management(struct psp_context *psp) +{ + struct amdgpu_device *adev = psp->adev; + + psp->pmfw_centralized_cstate_management = false; + + if (amdgpu_sriov_vf(adev)) + return; + + if (adev->flags & AMD_IS_APU) + return; + + if ((adev->asic_type == CHIP_ARCTURUS) || + (adev->asic_type >= CHIP_NAVI12)) + psp->pmfw_centralized_cstate_management = true; +} + static int psp_early_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; @@ -75,6 +112,8 @@ static int psp_early_init(void *handle) psp->adev = adev; + psp_check_pmfw_centralized_cstate_management(psp); + return 0; } @@ -101,6 +140,13 @@ static int psp_sw_init(void *handle) return ret; } + if (adev->asic_type == CHIP_NAVI10) { + ret= psp_sysfs_init(adev); + if (ret) { + return ret; + } + } + return 0; } @@ -117,6 +163,10 @@ static int psp_sw_fini(void *handle) release_firmware(adev->psp.ta_fw); adev->psp.ta_fw = NULL; } + + if (adev->asic_type == CHIP_NAVI10) + psp_sysfs_fini(adev); + return 0; } @@ -150,6 +200,7 @@ psp_cmd_submit_buf(struct psp_context *psp, int ret; int index; int timeout = 2000; + bool ras_intr = false; mutex_lock(&psp->mutex); @@ -174,7 +225,8 @@ psp_cmd_submit_buf(struct psp_context *psp, * because gpu reset thread triggered and lock resource should * be released for psp resume sequence. */ - if (amdgpu_ras_intr_triggered()) + ras_intr = amdgpu_ras_intr_triggered(); + if (ras_intr) break; msleep(1); amdgpu_asic_invalidate_hdp(psp->adev, NULL); @@ -187,7 +239,7 @@ psp_cmd_submit_buf(struct psp_context *psp, * during psp initialization to avoid breaking hw_init and it doesn't * return -EINVAL. */ - if (psp->cmd_buf_mem->resp.status || !timeout) { + if ((psp->cmd_buf_mem->resp.status || !timeout) && !ras_intr) { if (ucode) DRM_WARN("failed to load ucode id (%d) ", ucode->ucode_id); @@ -558,7 +610,7 @@ int psp_xgmi_invoke(struct psp_context *psp, uint32_t ta_cmd_id) return psp_ta_invoke(psp, ta_cmd_id, psp->xgmi_context.session_id); } -static int psp_xgmi_terminate(struct psp_context *psp) +int psp_xgmi_terminate(struct psp_context *psp) { int ret; @@ -579,7 +631,7 @@ static int psp_xgmi_terminate(struct psp_context *psp) return 0; } -static int psp_xgmi_initialize(struct psp_context *psp) +int psp_xgmi_initialize(struct psp_context *psp) { struct ta_xgmi_shared_memory *xgmi_cmd; int ret; @@ -766,7 +818,7 @@ static int psp_ras_initialize(struct psp_context *psp) if (!psp->adev->psp.ta_ras_ucode_size || !psp->adev->psp.ta_ras_start_addr) { - dev_warn(psp->adev->dev, "RAS: ras ta ucode is not available\n"); + dev_info(psp->adev->dev, "RAS: optional ras ta ucode is not available\n"); return 0; } @@ -850,7 +902,7 @@ static int psp_hdcp_initialize(struct psp_context *psp) if (!psp->adev->psp.ta_hdcp_ucode_size || !psp->adev->psp.ta_hdcp_start_addr) { - dev_warn(psp->adev->dev, "HDCP: hdcp ta ucode is not available\n"); + dev_info(psp->adev->dev, "HDCP: optional hdcp ta ucode is not available\n"); return 0; } @@ -996,7 +1048,7 @@ static int psp_dtm_initialize(struct psp_context *psp) if (!psp->adev->psp.ta_dtm_ucode_size || !psp->adev->psp.ta_dtm_start_addr) { - dev_warn(psp->adev->dev, "DTM: dtm ta ucode is not available\n"); + dev_info(psp->adev->dev, "DTM: optional dtm ta ucode is not available\n"); return 0; } @@ -1013,6 +1065,30 @@ static int psp_dtm_initialize(struct psp_context *psp) return 0; } +static int psp_dtm_unload(struct psp_context *psp) +{ + int ret; + struct psp_gfx_cmd_resp *cmd; + + /* + * TODO: bypass the unloading in sriov for now + */ + if (amdgpu_sriov_vf(psp->adev)) + return 0; + + cmd = kzalloc(sizeof(struct psp_gfx_cmd_resp), GFP_KERNEL); + if (!cmd) + return -ENOMEM; + + psp_prep_ta_unload_cmd_buf(cmd, psp->dtm_context.session_id); + + ret = psp_cmd_submit_buf(psp, NULL, cmd, psp->fence_buf_mc_addr); + + kfree(cmd); + + return ret; +} + int psp_dtm_invoke(struct psp_context *psp, uint32_t ta_cmd_id) { /* @@ -1037,7 +1113,7 @@ static int psp_dtm_terminate(struct psp_context *psp) if (!psp->dtm_context.dtm_initialized) return 0; - ret = psp_hdcp_unload(psp); + ret = psp_dtm_unload(psp); if (ret) return ret; @@ -1057,7 +1133,7 @@ static int psp_hw_start(struct psp_context *psp) struct amdgpu_device *adev = psp->adev; int ret; - if (!amdgpu_sriov_vf(adev) || !adev->in_gpu_reset) { + if (!amdgpu_sriov_vf(adev)) { if (psp->kdb_bin_size && (psp->funcs->bootloader_load_kdb != NULL)) { ret = psp_bootloader_load_kdb(psp); @@ -1092,10 +1168,17 @@ static int psp_hw_start(struct psp_context *psp) return ret; } - ret = psp_tmr_load(psp); - if (ret) { - DRM_ERROR("PSP load tmr failed!\n"); - return ret; + /* + * For those ASICs with DF Cstate management centralized + * to PMFW, TMR setup should be performed after PMFW + * loaded and before other non-psp firmware loaded. + */ + if (!psp->pmfw_centralized_cstate_management) { + ret = psp_tmr_load(psp); + if (ret) { + DRM_ERROR("PSP load tmr failed!\n"); + return ret; + } } return 0; @@ -1292,9 +1375,10 @@ static int psp_np_fw_load(struct psp_context *psp) struct amdgpu_firmware_info *ucode; struct amdgpu_device* adev = psp->adev; - if (psp->autoload_supported) { + if (psp->autoload_supported || + psp->pmfw_centralized_cstate_management) { ucode = &adev->firmware.ucode[AMDGPU_UCODE_ID_SMC]; - if (!ucode->fw) + if (!ucode->fw || amdgpu_sriov_vf(adev)) goto out; ret = psp_execute_np_fw_load(psp, ucode); @@ -1302,6 +1386,14 @@ static int psp_np_fw_load(struct psp_context *psp) return ret; } + if (psp->pmfw_centralized_cstate_management) { + ret = psp_tmr_load(psp); + if (ret) { + DRM_ERROR("PSP load tmr failed!\n"); + return ret; + } + } + out: for (i = 0; i < adev->firmware.max_ucodes; i++) { ucode = &adev->firmware.ucode[i]; @@ -1309,7 +1401,9 @@ out: continue; if (ucode->ucode_id == AMDGPU_UCODE_ID_SMC && - (psp_smu_reload_quirk(psp) || psp->autoload_supported)) + (psp_smu_reload_quirk(psp) || + psp->autoload_supported || + psp->pmfw_centralized_cstate_management)) continue; if (amdgpu_sriov_vf(adev) && @@ -1420,16 +1514,6 @@ skip_memalloc: return ret; } - if (adev->gmc.xgmi.num_physical_nodes > 1) { - ret = psp_xgmi_initialize(psp); - /* Warning the XGMI seesion initialize failure - * Instead of stop driver initialization - */ - if (ret) - dev_err(psp->adev->dev, - "XGMI: Failed to initialize XGMI session\n"); - } - if (psp->adev->psp.ta_fw) { ret = psp_ras_initialize(psp); if (ret) @@ -1494,10 +1578,6 @@ static int psp_hw_fini(void *handle) void *tmr_buf; void **pptr; - if (adev->gmc.xgmi.num_physical_nodes > 1 && - psp->xgmi_context.initialized == 1) - psp_xgmi_terminate(psp); - if (psp->adev->psp.ta_fw) { psp_ras_terminate(psp); psp_dtm_terminate(psp); @@ -1753,6 +1833,97 @@ static int psp_set_powergating_state(void *handle, return 0; } +static ssize_t psp_usbc_pd_fw_sysfs_read(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + struct drm_device *ddev = dev_get_drvdata(dev); + struct amdgpu_device *adev = ddev->dev_private; + uint32_t fw_ver; + int ret; + + if (!adev->ip_blocks[AMD_IP_BLOCK_TYPE_PSP].status.late_initialized) { + DRM_INFO("PSP block is not ready yet."); + return -EBUSY; + } + + mutex_lock(&adev->psp.mutex); + ret = psp_read_usbc_pd_fw(&adev->psp, &fw_ver); + mutex_unlock(&adev->psp.mutex); + + if (ret) { + DRM_ERROR("Failed to read USBC PD FW, err = %d", ret); + return ret; + } + + return snprintf(buf, PAGE_SIZE, "%x\n", fw_ver); +} + +static ssize_t psp_usbc_pd_fw_sysfs_write(struct device *dev, + struct device_attribute *attr, + const char *buf, + size_t count) +{ + struct drm_device *ddev = dev_get_drvdata(dev); + struct amdgpu_device *adev = ddev->dev_private; + void *cpu_addr; + dma_addr_t dma_addr; + int ret; + char fw_name[100]; + const struct firmware *usbc_pd_fw; + + if (!adev->ip_blocks[AMD_IP_BLOCK_TYPE_PSP].status.late_initialized) { + DRM_INFO("PSP block is not ready yet."); + return -EBUSY; + } + + snprintf(fw_name, sizeof(fw_name), "amdgpu/%s", buf); + ret = request_firmware(&usbc_pd_fw, fw_name, adev->dev); + if (ret) + goto fail; + + /* We need contiguous physical mem to place the FW for psp to access */ + cpu_addr = dma_alloc_coherent(adev->dev, usbc_pd_fw->size, &dma_addr, GFP_KERNEL); + + ret = dma_mapping_error(adev->dev, dma_addr); + if (ret) + goto rel_buf; + + memcpy_toio(cpu_addr, usbc_pd_fw->data, usbc_pd_fw->size); + + /* + * x86 specific workaround. + * Without it the buffer is invisible in PSP. + * + * TODO Remove once PSP starts snooping CPU cache + */ +#ifdef CONFIG_X86 + clflush_cache_range(cpu_addr, (usbc_pd_fw->size & ~(L1_CACHE_BYTES - 1))); +#endif + + mutex_lock(&adev->psp.mutex); + ret = psp_load_usbc_pd_fw(&adev->psp, dma_addr); + mutex_unlock(&adev->psp.mutex); + +rel_buf: + dma_free_coherent(adev->dev, usbc_pd_fw->size, cpu_addr, dma_addr); + release_firmware(usbc_pd_fw); + +fail: + if (ret) { + DRM_ERROR("Failed to load USBC PD FW, err = %d", ret); + return ret; + } + + return count; +} + +static DEVICE_ATTR(usbc_pd_fw, S_IRUGO | S_IWUSR, + psp_usbc_pd_fw_sysfs_read, + psp_usbc_pd_fw_sysfs_write); + + + const struct amd_ip_funcs psp_ip_funcs = { .name = "psp", .early_init = psp_early_init, @@ -1771,6 +1942,21 @@ const struct amd_ip_funcs psp_ip_funcs = { .set_powergating_state = psp_set_powergating_state, }; +static int psp_sysfs_init(struct amdgpu_device *adev) +{ + int ret = device_create_file(adev->dev, &dev_attr_usbc_pd_fw); + + if (ret) + DRM_ERROR("Failed to create USBC PD FW control file!"); + + return ret; +} + +static void psp_sysfs_fini(struct amdgpu_device *adev) +{ + device_remove_file(adev->dev, &dev_attr_usbc_pd_fw); +} + static const struct amdgpu_psp_funcs psp_funcs = { .check_fw_loading_status = psp_check_fw_loading_status, }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h index 611021514c52..297435c0c7c1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h @@ -114,6 +114,8 @@ struct psp_funcs int (*mem_training)(struct psp_context *psp, uint32_t ops); uint32_t (*ring_get_wptr)(struct psp_context *psp); void (*ring_set_wptr)(struct psp_context *psp, uint32_t value); + int (*load_usbc_pd_fw)(struct psp_context *psp, dma_addr_t dma_addr); + int (*read_usbc_pd_fw)(struct psp_context *psp, uint32_t *fw_ver); }; #define AMDGPU_XGMI_MAX_CONNECTED_NODES 64 @@ -264,6 +266,8 @@ struct psp_context atomic_t fence_value; /* flag to mark whether gfx fw autoload is supported or not */ bool autoload_supported; + /* flag to mark whether df cstate management centralized to PMFW */ + bool pmfw_centralized_cstate_management; /* xgmi ta firmware and buffer */ const struct firmware *ta_fw; @@ -349,6 +353,14 @@ struct amdgpu_psp_funcs { #define psp_ring_get_wptr(psp) (psp)->funcs->ring_get_wptr((psp)) #define psp_ring_set_wptr(psp, value) (psp)->funcs->ring_set_wptr((psp), (value)) +#define psp_load_usbc_pd_fw(psp, dma_addr) \ + ((psp)->funcs->load_usbc_pd_fw ? \ + (psp)->funcs->load_usbc_pd_fw((psp), (dma_addr)) : -EINVAL) + +#define psp_read_usbc_pd_fw(psp, fw_ver) \ + ((psp)->funcs->read_usbc_pd_fw ? \ + (psp)->funcs->read_usbc_pd_fw((psp), fw_ver) : -EINVAL) + extern const struct amd_ip_funcs psp_ip_funcs; extern const struct amdgpu_ip_block_version psp_v3_1_ip_block; @@ -362,6 +374,8 @@ int psp_gpu_reset(struct amdgpu_device *adev); int psp_update_vcn_sram(struct amdgpu_device *adev, int inst_idx, uint64_t cmd_gpu_addr, int cmd_size); +int psp_xgmi_initialize(struct psp_context *psp); +int psp_xgmi_terminate(struct psp_context *psp); int psp_xgmi_invoke(struct psp_context *psp, uint32_t ta_cmd_id); int psp_ras_invoke(struct psp_context *psp, uint32_t ta_cmd_id); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c index cef94e2169fe..ab379b44679c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c @@ -31,6 +31,7 @@ #include "amdgpu.h" #include "amdgpu_ras.h" #include "amdgpu_atomfirmware.h" +#include "amdgpu_xgmi.h" #include "ivsrcid/nbio/irqsrcs_nbif_7_4.h" const char *ras_error_string[] = { @@ -280,6 +281,11 @@ static ssize_t amdgpu_ras_debugfs_ctrl_write(struct file *f, const char __user * struct ras_debug_if data; int ret = 0; + if (amdgpu_ras_intr_triggered()) { + DRM_WARN("RAS WARN: error injection currently inaccessible\n"); + return size; + } + ret = amdgpu_ras_debugfs_ctrl_parse_data(f, buf, size, pos, &data); if (ret) return -EINVAL; @@ -393,6 +399,10 @@ static ssize_t amdgpu_ras_sysfs_read(struct device *dev, .head = obj->head, }; + if (amdgpu_ras_intr_triggered()) + return snprintf(buf, PAGE_SIZE, + "Query currently inaccessible\n"); + if (amdgpu_ras_error_query(obj->adev, &info)) return -EINVAL; @@ -720,6 +730,9 @@ int amdgpu_ras_error_query(struct amdgpu_device *adev, if (adev->nbio.funcs->query_ras_error_count) adev->nbio.funcs->query_ras_error_count(adev, &err_data); break; + case AMDGPU_RAS_BLOCK__XGMI_WAFL: + amdgpu_xgmi_query_ras_error_count(adev, &err_data); + break; default: break; } @@ -742,20 +755,6 @@ int amdgpu_ras_error_query(struct amdgpu_device *adev, return 0; } -uint64_t get_xgmi_relative_phy_addr(struct amdgpu_device *adev, uint64_t addr) -{ - uint32_t df_inst_id; - - if ((!adev->df.funcs) || - (!adev->df.funcs->get_df_inst_id) || - (!adev->df.funcs->get_dram_base_addr)) - return addr; - - df_inst_id = adev->df.funcs->get_df_inst_id(adev); - - return addr + adev->df.funcs->get_dram_base_addr(adev, df_inst_id); -} - /* wrapper of psp_ras_trigger_error */ int amdgpu_ras_error_inject(struct amdgpu_device *adev, struct ras_inject_if *info) @@ -775,8 +774,9 @@ int amdgpu_ras_error_inject(struct amdgpu_device *adev, /* Calculate XGMI relative offset */ if (adev->gmc.xgmi.num_physical_nodes > 1) { - block_info.address = get_xgmi_relative_phy_addr(adev, - block_info.address); + block_info.address = + amdgpu_xgmi_get_relative_phy_addr(adev, + block_info.address); } switch (info->head.block) { @@ -1122,6 +1122,32 @@ void amdgpu_ras_debugfs_create(struct amdgpu_device *adev, &amdgpu_ras_debugfs_ops); } +void amdgpu_ras_debugfs_create_all(struct amdgpu_device *adev) +{ + struct amdgpu_ras *con = amdgpu_ras_get_context(adev); + struct ras_manager *obj; + struct ras_fs_if fs_info; + + /* + * it won't be called in resume path, no need to check + * suspend and gpu reset status + */ + if (!con) + return; + + amdgpu_ras_debugfs_create_ctrl_node(adev); + + list_for_each_entry(obj, &con->head, node) { + if (amdgpu_ras_is_supported(adev, obj->head.block) && + (obj->attr_inuse == 1)) { + sprintf(fs_info.debugfs_name, "%s_err_inject", + ras_block_str(obj->head.block)); + fs_info.head = obj->head; + amdgpu_ras_debugfs_create(adev, &fs_info); + } + } +} + void amdgpu_ras_debugfs_remove(struct amdgpu_device *adev, struct ras_common_if *head) { @@ -1154,7 +1180,6 @@ static void amdgpu_ras_debugfs_remove_all(struct amdgpu_device *adev) static int amdgpu_ras_fs_init(struct amdgpu_device *adev) { amdgpu_ras_sysfs_create_feature_node(adev); - amdgpu_ras_debugfs_create_ctrl_node(adev); return 0; } @@ -1319,6 +1344,33 @@ static int amdgpu_ras_interrupt_remove_all(struct amdgpu_device *adev) } /* ih end */ +/* traversal all IPs except NBIO to query error counter */ +static void amdgpu_ras_log_on_err_counter(struct amdgpu_device *adev) +{ + struct amdgpu_ras *con = amdgpu_ras_get_context(adev); + struct ras_manager *obj; + + if (!con) + return; + + list_for_each_entry(obj, &con->head, node) { + struct ras_query_if info = { + .head = obj->head, + }; + + /* + * PCIE_BIF IP has one different isr by ras controller + * interrupt, the specific ras counter query will be + * done in that isr. So skip such block from common + * sync flood interrupt isr calling. + */ + if (info.head.block == AMDGPU_RAS_BLOCK__PCIE_BIF) + continue; + + amdgpu_ras_error_query(adev, &info); + } +} + /* recovery begin */ /* return 0 on success. @@ -1372,6 +1424,22 @@ static void amdgpu_ras_do_recovery(struct work_struct *work) { struct amdgpu_ras *ras = container_of(work, struct amdgpu_ras, recovery_work); + struct amdgpu_device *remote_adev = NULL; + struct amdgpu_device *adev = ras->adev; + struct list_head device_list, *device_list_handle = NULL; + struct amdgpu_hive_info *hive = amdgpu_get_xgmi_hive(adev, false); + + /* Build list of devices to query RAS related errors */ + if (hive && adev->gmc.xgmi.num_physical_nodes > 1) { + device_list_handle = &hive->device_list; + } else { + list_add_tail(&adev->gmc.xgmi.head, &device_list); + device_list_handle = &device_list; + } + + list_for_each_entry(remote_adev, device_list_handle, gmc.xgmi.head) { + amdgpu_ras_log_on_err_counter(remote_adev); + } if (amdgpu_device_should_recover_gpu(ras->adev)) amdgpu_device_gpu_recover(ras->adev, 0); @@ -1713,18 +1781,30 @@ static void amdgpu_ras_check_supported(struct amdgpu_device *adev, *hw_supported = 0; *supported = 0; - if (amdgpu_sriov_vf(adev) || + if (amdgpu_sriov_vf(adev) || !adev->is_atom_fw || (adev->asic_type != CHIP_VEGA20 && adev->asic_type != CHIP_ARCTURUS)) return; - if (adev->is_atom_fw && - (amdgpu_atomfirmware_mem_ecc_supported(adev) || - amdgpu_atomfirmware_sram_ecc_supported(adev))) - *hw_supported = AMDGPU_RAS_BLOCK_MASK; + if (amdgpu_atomfirmware_mem_ecc_supported(adev)) { + DRM_INFO("HBM ECC is active.\n"); + *hw_supported |= (1 << AMDGPU_RAS_BLOCK__UMC | + 1 << AMDGPU_RAS_BLOCK__DF); + } else + DRM_INFO("HBM ECC is not presented.\n"); + + if (amdgpu_atomfirmware_sram_ecc_supported(adev)) { + DRM_INFO("SRAM ECC is active.\n"); + *hw_supported |= ~(1 << AMDGPU_RAS_BLOCK__UMC | + 1 << AMDGPU_RAS_BLOCK__DF); + } else + DRM_INFO("SRAM ECC is not presented.\n"); + + /* hw_supported needs to be aligned with RAS block mask. */ + *hw_supported &= AMDGPU_RAS_BLOCK_MASK; *supported = amdgpu_ras_enable == 0 ? - 0 : *hw_supported & amdgpu_ras_mask; + 0 : *hw_supported & amdgpu_ras_mask; } int amdgpu_ras_init(struct amdgpu_device *adev) @@ -1825,8 +1905,6 @@ int amdgpu_ras_late_init(struct amdgpu_device *adev, goto interrupt; } - amdgpu_ras_debugfs_create(adev, fs_info); - r = amdgpu_ras_sysfs_create(adev, fs_info); if (r) goto sysfs; @@ -1835,7 +1913,6 @@ int amdgpu_ras_late_init(struct amdgpu_device *adev, cleanup: amdgpu_ras_sysfs_remove(adev, ras_block); sysfs: - amdgpu_ras_debugfs_remove(adev, ras_block); if (ih_info->cb) amdgpu_ras_interrupt_remove_handler(adev, ih_info); interrupt: @@ -1852,7 +1929,6 @@ void amdgpu_ras_late_fini(struct amdgpu_device *adev, return; amdgpu_ras_sysfs_remove(adev, ras_block); - amdgpu_ras_debugfs_remove(adev, ras_block); if (ih_info->cb) amdgpu_ras_interrupt_remove_handler(adev, ih_info); amdgpu_ras_feature_enable(adev, ras_block, 0); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h index a5fe29a9373e..55c3eceb390d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h @@ -592,6 +592,8 @@ int amdgpu_ras_sysfs_remove(struct amdgpu_device *adev, void amdgpu_ras_debugfs_create(struct amdgpu_device *adev, struct ras_fs_if *head); +void amdgpu_ras_debugfs_create_all(struct amdgpu_device *adev); + void amdgpu_ras_debugfs_remove(struct amdgpu_device *adev, struct ras_common_if *head); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c index 2a8e04895595..c0096097bbcf 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c @@ -25,10 +25,11 @@ #include "amdgpu.h" #include "amdgpu_ras.h" #include <linux/bits.h> -#include "smu_v11_0_i2c.h" +#include "atom.h" -#define EEPROM_I2C_TARGET_ADDR_ARCTURUS 0xA8 -#define EEPROM_I2C_TARGET_ADDR_VEGA20 0xA0 +#define EEPROM_I2C_TARGET_ADDR_VEGA20 0xA0 +#define EEPROM_I2C_TARGET_ADDR_ARCTURUS 0xA8 +#define EEPROM_I2C_TARGET_ADDR_ARCTURUS_D342 0xA0 /* * The 2 macros bellow represent the actual size in bytes that @@ -55,6 +56,45 @@ #define to_amdgpu_device(x) (container_of(x, struct amdgpu_ras, eeprom_control))->adev +static bool __get_eeprom_i2c_addr_arct(struct amdgpu_device *adev, + uint16_t *i2c_addr) +{ + struct atom_context *atom_ctx = adev->mode_info.atom_context; + + if (!i2c_addr || !atom_ctx) + return false; + + if (strnstr(atom_ctx->vbios_version, + "D342", + sizeof(atom_ctx->vbios_version))) + *i2c_addr = EEPROM_I2C_TARGET_ADDR_ARCTURUS_D342; + else + *i2c_addr = EEPROM_I2C_TARGET_ADDR_ARCTURUS; + + return true; +} + +static bool __get_eeprom_i2c_addr(struct amdgpu_device *adev, + uint16_t *i2c_addr) +{ + if (!i2c_addr) + return false; + + switch (adev->asic_type) { + case CHIP_VEGA20: + *i2c_addr = EEPROM_I2C_TARGET_ADDR_VEGA20; + break; + + case CHIP_ARCTURUS: + return __get_eeprom_i2c_addr_arct(adev, i2c_addr); + + default: + return false; + } + + return true; +} + static void __encode_table_header_to_buff(struct amdgpu_ras_eeprom_table_header *hdr, unsigned char *buff) { @@ -83,6 +123,7 @@ static int __update_table_header(struct amdgpu_ras_eeprom_control *control, unsigned char *buff) { int ret = 0; + struct amdgpu_device *adev = to_amdgpu_device(control); struct i2c_msg msg = { .addr = 0, .flags = 0, @@ -96,15 +137,13 @@ static int __update_table_header(struct amdgpu_ras_eeprom_control *control, msg.addr = control->i2c_address; - ret = i2c_transfer(&control->eeprom_accessor, &msg, 1); + ret = i2c_transfer(&adev->pm.smu_i2c, &msg, 1); if (ret < 1) DRM_ERROR("Failed to write EEPROM table header, ret:%d", ret); return ret; } - - static uint32_t __calc_hdr_byte_sum(struct amdgpu_ras_eeprom_control *control) { int i; @@ -212,32 +251,18 @@ int amdgpu_ras_eeprom_init(struct amdgpu_ras_eeprom_control *control) .buf = buff, }; - mutex_init(&control->tbl_mutex); - - switch (adev->asic_type) { - case CHIP_VEGA20: - control->i2c_address = EEPROM_I2C_TARGET_ADDR_VEGA20; - ret = smu_v11_0_i2c_eeprom_control_init(&control->eeprom_accessor); - break; - - case CHIP_ARCTURUS: - control->i2c_address = EEPROM_I2C_TARGET_ADDR_ARCTURUS; - ret = smu_i2c_eeprom_init(&adev->smu, &control->eeprom_accessor); - break; + /* Verify i2c adapter is initialized */ + if (!adev->pm.smu_i2c.algo) + return -ENOENT; - default: - return 0; - } + if (!__get_eeprom_i2c_addr(adev, &control->i2c_address)) + return -EINVAL; - if (ret) { - DRM_ERROR("Failed to init I2C controller, ret:%d", ret); - return ret; - } + mutex_init(&control->tbl_mutex); msg.addr = control->i2c_address; - /* Read/Create table header from EEPROM address 0 */ - ret = i2c_transfer(&control->eeprom_accessor, &msg, 1); + ret = i2c_transfer(&adev->pm.smu_i2c, &msg, 1); if (ret < 1) { DRM_ERROR("Failed to read EEPROM table header, ret:%d", ret); return ret; @@ -263,23 +288,6 @@ int amdgpu_ras_eeprom_init(struct amdgpu_ras_eeprom_control *control) return ret == 1 ? 0 : -EIO; } -void amdgpu_ras_eeprom_fini(struct amdgpu_ras_eeprom_control *control) -{ - struct amdgpu_device *adev = to_amdgpu_device(control); - - switch (adev->asic_type) { - case CHIP_VEGA20: - smu_v11_0_i2c_eeprom_control_fini(&control->eeprom_accessor); - break; - case CHIP_ARCTURUS: - smu_i2c_eeprom_fini(&adev->smu, &control->eeprom_accessor); - break; - - default: - return; - } -} - static void __encode_table_record_to_buff(struct amdgpu_ras_eeprom_control *control, struct eeprom_table_record *record, unsigned char *buff) @@ -436,7 +444,7 @@ int amdgpu_ras_eeprom_process_recods(struct amdgpu_ras_eeprom_control *control, control->next_addr += EEPROM_TABLE_RECORD_SIZE; } - ret = i2c_transfer(&control->eeprom_accessor, msgs, num); + ret = i2c_transfer(&adev->pm.smu_i2c, msgs, num); if (ret < 1) { DRM_ERROR("Failed to process EEPROM table records, ret:%d", ret); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h index ca78f812d436..7e8647a05df7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h @@ -44,7 +44,6 @@ struct amdgpu_ras_eeprom_table_header { struct amdgpu_ras_eeprom_control { struct amdgpu_ras_eeprom_table_header tbl_hdr; - struct i2c_adapter eeprom_accessor; uint32_t next_addr; unsigned int num_recs; struct mutex tbl_mutex; @@ -79,7 +78,6 @@ struct eeprom_table_record { }__attribute__((__packed__)); int amdgpu_ras_eeprom_init(struct amdgpu_ras_eeprom_control *control); -void amdgpu_ras_eeprom_fini(struct amdgpu_ras_eeprom_control *control); int amdgpu_ras_eeprom_reset_table(struct amdgpu_ras_eeprom_control *control); int amdgpu_ras_eeprom_process_recods(struct amdgpu_ras_eeprom_control *control, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c index e5c83e164d82..a7e1d0425ed0 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c @@ -48,9 +48,6 @@ * wptr. The GPU then starts fetching commands and executes * them until the pointers are equal again. */ -static int amdgpu_debugfs_ring_init(struct amdgpu_device *adev, - struct amdgpu_ring *ring); -static void amdgpu_debugfs_ring_fini(struct amdgpu_ring *ring); /** * amdgpu_ring_alloc - allocate space on the ring buffer @@ -154,76 +151,6 @@ void amdgpu_ring_undo(struct amdgpu_ring *ring) } /** - * amdgpu_ring_priority_put - restore a ring's priority - * - * @ring: amdgpu_ring structure holding the information - * @priority: target priority - * - * Release a request for executing at @priority - */ -void amdgpu_ring_priority_put(struct amdgpu_ring *ring, - enum drm_sched_priority priority) -{ - int i; - - if (!ring->funcs->set_priority) - return; - - if (atomic_dec_return(&ring->num_jobs[priority]) > 0) - return; - - /* no need to restore if the job is already at the lowest priority */ - if (priority == DRM_SCHED_PRIORITY_NORMAL) - return; - - mutex_lock(&ring->priority_mutex); - /* something higher prio is executing, no need to decay */ - if (ring->priority > priority) - goto out_unlock; - - /* decay priority to the next level with a job available */ - for (i = priority; i >= DRM_SCHED_PRIORITY_MIN; i--) { - if (i == DRM_SCHED_PRIORITY_NORMAL - || atomic_read(&ring->num_jobs[i])) { - ring->priority = i; - ring->funcs->set_priority(ring, i); - break; - } - } - -out_unlock: - mutex_unlock(&ring->priority_mutex); -} - -/** - * amdgpu_ring_priority_get - change the ring's priority - * - * @ring: amdgpu_ring structure holding the information - * @priority: target priority - * - * Request a ring's priority to be raised to @priority (refcounted). - */ -void amdgpu_ring_priority_get(struct amdgpu_ring *ring, - enum drm_sched_priority priority) -{ - if (!ring->funcs->set_priority) - return; - - if (atomic_inc_return(&ring->num_jobs[priority]) <= 0) - return; - - mutex_lock(&ring->priority_mutex); - if (priority <= ring->priority) - goto out_unlock; - - ring->priority = priority; - ring->funcs->set_priority(ring, priority); - -out_unlock: - mutex_unlock(&ring->priority_mutex); -} - -/** * amdgpu_ring_init - init driver ring struct. * * @adev: amdgpu_device pointer @@ -334,10 +261,6 @@ int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring, for (i = 0; i < DRM_SCHED_PRIORITY_MAX; ++i) atomic_set(&ring->num_jobs[i], 0); - if (amdgpu_debugfs_ring_init(adev, ring)) { - DRM_ERROR("Failed to register debugfs file for rings !\n"); - } - return 0; } @@ -351,12 +274,13 @@ int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring, */ void amdgpu_ring_fini(struct amdgpu_ring *ring) { - ring->sched.ready = false; /* Not to finish a ring which is not initialized */ if (!(ring->adev) || !(ring->adev->rings[ring->idx])) return; + ring->sched.ready = false; + amdgpu_device_wb_free(ring->adev, ring->rptr_offs); amdgpu_device_wb_free(ring->adev, ring->wptr_offs); @@ -367,8 +291,6 @@ void amdgpu_ring_fini(struct amdgpu_ring *ring) &ring->gpu_addr, (void **)&ring->ring); - amdgpu_debugfs_ring_fini(ring); - dma_fence_put(ring->vmid_wait); ring->vmid_wait = NULL; ring->me = 0; @@ -485,8 +407,8 @@ static const struct file_operations amdgpu_debugfs_ring_fops = { #endif -static int amdgpu_debugfs_ring_init(struct amdgpu_device *adev, - struct amdgpu_ring *ring) +int amdgpu_debugfs_ring_init(struct amdgpu_device *adev, + struct amdgpu_ring *ring) { #if defined(CONFIG_DEBUG_FS) struct drm_minor *minor = adev->ddev->primary; @@ -507,13 +429,6 @@ static int amdgpu_debugfs_ring_init(struct amdgpu_device *adev, return 0; } -static void amdgpu_debugfs_ring_fini(struct amdgpu_ring *ring) -{ -#if defined(CONFIG_DEBUG_FS) - debugfs_remove(ring->ent); -#endif -} - /** * amdgpu_ring_test_helper - tests ring and set sched readiness status * diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h index 930316e60155..9a443013d70d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h @@ -167,9 +167,6 @@ struct amdgpu_ring_funcs { uint32_t reg0, uint32_t reg1, uint32_t ref, uint32_t mask); void (*emit_tmz)(struct amdgpu_ring *ring, bool start); - /* priority functions */ - void (*set_priority) (struct amdgpu_ring *ring, - enum drm_sched_priority priority); /* Try to soft recover the ring to make the fence signal */ void (*soft_recovery)(struct amdgpu_ring *ring, unsigned vmid); int (*preempt_ib)(struct amdgpu_ring *ring); @@ -222,6 +219,7 @@ struct amdgpu_ring { struct mutex priority_mutex; /* protected by priority_mutex */ int priority; + bool has_high_prio; #if defined(CONFIG_DEBUG_FS) struct dentry *ent; @@ -258,10 +256,6 @@ void amdgpu_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count); void amdgpu_ring_generic_pad_ib(struct amdgpu_ring *ring, struct amdgpu_ib *ib); void amdgpu_ring_commit(struct amdgpu_ring *ring); void amdgpu_ring_undo(struct amdgpu_ring *ring); -void amdgpu_ring_priority_get(struct amdgpu_ring *ring, - enum drm_sched_priority priority); -void amdgpu_ring_priority_put(struct amdgpu_ring *ring, - enum drm_sched_priority priority); int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring, unsigned ring_size, struct amdgpu_irq_src *irq_src, unsigned irq_type); @@ -328,4 +322,8 @@ static inline void amdgpu_ring_write_multiple(struct amdgpu_ring *ring, int amdgpu_ring_test_helper(struct amdgpu_ring *ring); +int amdgpu_debugfs_ring_init(struct amdgpu_device *adev, + struct amdgpu_ring *ring); +void amdgpu_debugfs_ring_fini(struct amdgpu_ring *ring); + #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h index d3d4707f2168..60bb3e8b3118 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h @@ -126,6 +126,9 @@ struct amdgpu_rlc_funcs { void (*stop)(struct amdgpu_device *adev); void (*reset)(struct amdgpu_device *adev); void (*start)(struct amdgpu_device *adev); + void (*update_spm_vmid)(struct amdgpu_device *adev, unsigned vmid); + void (*rlcg_wreg)(struct amdgpu_device *adev, u32 offset, u32 v); + bool (*is_rlcg_access_range)(struct amdgpu_device *adev, uint32_t reg); }; struct amdgpu_rlc { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c index a2ee30b16212..250a309e4dee 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c @@ -70,7 +70,8 @@ uint64_t amdgpu_sdma_get_csa_mc_addr(struct amdgpu_ring *ring, uint32_t index = 0; int r; - if (vmid == 0 || !amdgpu_mcbp) + /* don't enable OS preemption on SDMA under SRIOV */ + if (amdgpu_sriov_vf(adev) || vmid == 0 || !amdgpu_mcbp) return 0; r = amdgpu_sdma_get_index_from_ring(ring, &index); @@ -92,7 +93,6 @@ int amdgpu_sdma_ras_late_init(struct amdgpu_device *adev, struct ras_ih_if *ih_info = (struct ras_ih_if *)ras_ih_info; struct ras_fs_if fs_info = { .sysfs_name = "sdma_err_count", - .debugfs_name = "sdma_err_inject", }; if (!ih_info) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h index 485335267d78..4b352206354b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h @@ -56,6 +56,7 @@ struct amdgpu_sdma_ras_funcs { void (*ras_fini)(struct amdgpu_device *adev); int (*query_ras_error_count)(struct amdgpu_device *adev, uint32_t instance, void *ras_error_status); + void (*reset_ras_error_count)(struct amdgpu_device *adev); }; struct amdgpu_sdma { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c index a09b6b9c27d1..b86392253696 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c @@ -202,18 +202,17 @@ int amdgpu_sync_vm_fence(struct amdgpu_sync *sync, struct dma_fence *fence) * * @sync: sync object to add fences from reservation object to * @resv: reservation object with embedded fence - * @explicit_sync: true if we should only sync to the exclusive fence + * @mode: how owner affects which fences we sync to + * @owner: owner of the planned job submission * * Sync to the fence */ -int amdgpu_sync_resv(struct amdgpu_device *adev, - struct amdgpu_sync *sync, - struct dma_resv *resv, - void *owner, bool explicit_sync) +int amdgpu_sync_resv(struct amdgpu_device *adev, struct amdgpu_sync *sync, + struct dma_resv *resv, enum amdgpu_sync_mode mode, + void *owner) { struct dma_resv_list *flist; struct dma_fence *f; - void *fence_owner; unsigned i; int r = 0; @@ -229,30 +228,46 @@ int amdgpu_sync_resv(struct amdgpu_device *adev, return r; for (i = 0; i < flist->shared_count; ++i) { + void *fence_owner; + f = rcu_dereference_protected(flist->shared[i], dma_resv_held(resv)); + + fence_owner = amdgpu_sync_get_owner(f); + + /* Always sync to moves, no matter what */ + if (fence_owner == AMDGPU_FENCE_OWNER_UNDEFINED) { + r = amdgpu_sync_fence(sync, f, false); + if (r) + break; + } + /* We only want to trigger KFD eviction fences on * evict or move jobs. Skip KFD fences otherwise. */ - fence_owner = amdgpu_sync_get_owner(f); if (fence_owner == AMDGPU_FENCE_OWNER_KFD && owner != AMDGPU_FENCE_OWNER_UNDEFINED) continue; - if (amdgpu_sync_same_dev(adev, f)) { - /* VM updates only sync with moves but not with user - * command submissions or KFD evictions fences - */ - if (owner == AMDGPU_FENCE_OWNER_VM && - fence_owner != AMDGPU_FENCE_OWNER_UNDEFINED) + /* Ignore fences depending on the sync mode */ + switch (mode) { + case AMDGPU_SYNC_ALWAYS: + break; + + case AMDGPU_SYNC_NE_OWNER: + if (amdgpu_sync_same_dev(adev, f) && + fence_owner == owner) continue; + break; - /* Ignore fence from the same owner and explicit one as - * long as it isn't undefined. - */ - if (owner != AMDGPU_FENCE_OWNER_UNDEFINED && - (fence_owner == owner || explicit_sync)) + case AMDGPU_SYNC_EQ_OWNER: + if (amdgpu_sync_same_dev(adev, f) && + fence_owner != owner) continue; + break; + + case AMDGPU_SYNC_EXPLICIT: + continue; } r = amdgpu_sync_fence(sync, f, false); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.h index d62c2b81d92b..cfbe5788b8b9 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.h @@ -31,6 +31,13 @@ struct dma_resv; struct amdgpu_device; struct amdgpu_ring; +enum amdgpu_sync_mode { + AMDGPU_SYNC_ALWAYS, + AMDGPU_SYNC_NE_OWNER, + AMDGPU_SYNC_EQ_OWNER, + AMDGPU_SYNC_EXPLICIT +}; + /* * Container for fences used to sync command submissions. */ @@ -43,11 +50,9 @@ void amdgpu_sync_create(struct amdgpu_sync *sync); int amdgpu_sync_fence(struct amdgpu_sync *sync, struct dma_fence *f, bool explicit); int amdgpu_sync_vm_fence(struct amdgpu_sync *sync, struct dma_fence *fence); -int amdgpu_sync_resv(struct amdgpu_device *adev, - struct amdgpu_sync *sync, - struct dma_resv *resv, - void *owner, - bool explicit_sync); +int amdgpu_sync_resv(struct amdgpu_device *adev, struct amdgpu_sync *sync, + struct dma_resv *resv, enum amdgpu_sync_mode mode, + void *owner); struct dma_fence *amdgpu_sync_peek_fence(struct amdgpu_sync *sync, struct amdgpu_ring *ring); struct dma_fence *amdgpu_sync_get_fence(struct amdgpu_sync *sync, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c index 3ab46d4647e4..6309ff72bd78 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c @@ -60,15 +60,14 @@ #include "amdgpu_ras.h" #include "bif/bif_4_1_d.h" +#define AMDGPU_TTM_VRAM_MAX_DW_READ (size_t)128 + static int amdgpu_map_buffer(struct ttm_buffer_object *bo, struct ttm_mem_reg *mem, unsigned num_pages, uint64_t offset, unsigned window, struct amdgpu_ring *ring, uint64_t *addr); -static int amdgpu_ttm_debugfs_init(struct amdgpu_device *adev); -static void amdgpu_ttm_debugfs_fini(struct amdgpu_device *adev); - /** * amdgpu_init_mem_type - Initialize a memory manager for a specific type of * memory request. @@ -771,7 +770,6 @@ struct amdgpu_ttm_tt { static const uint64_t hmm_range_flags[HMM_PFN_FLAG_MAX] = { (1 << 0), /* HMM_PFN_VALID */ (1 << 1), /* HMM_PFN_WRITE */ - 0 /* HMM_PFN_DEVICE_PRIVATE */ }; static const uint64_t hmm_range_values[HMM_PFN_VALUE_MAX] = { @@ -852,7 +850,7 @@ retry: range->notifier_seq = mmu_interval_read_begin(&bo->notifier); down_read(&mm->mmap_sem); - r = hmm_range_fault(range, 0); + r = hmm_range_fault(range); up_read(&mm->mmap_sem); if (unlikely(r <= 0)) { /* @@ -969,7 +967,7 @@ static int amdgpu_ttm_tt_pin_userptr(struct ttm_tt *ttm) /* Map SG to device */ r = -ENOMEM; nents = dma_map_sg(adev->dev, ttm->sg->sgl, ttm->sg->nents, direction); - if (nents != ttm->sg->nents) + if (nents == 0) goto release_sg; /* convert SG to linear array of pages and dma addresses */ @@ -1029,7 +1027,7 @@ int amdgpu_ttm_gart_bind(struct amdgpu_device *adev, struct amdgpu_ttm_tt *gtt = (void *)ttm; int r; - if (abo->flags & AMDGPU_GEM_CREATE_MQD_GFX9) { + if (abo->flags & AMDGPU_GEM_CREATE_CP_MQD_GFX9) { uint64_t page_idx = 1; r = amdgpu_gart_bind(adev, gtt->offset, page_idx, @@ -1037,7 +1035,10 @@ int amdgpu_ttm_gart_bind(struct amdgpu_device *adev, if (r) goto gart_bind_fail; - /* Patch mtype of the second part BO */ + /* The memory type of the first page defaults to UC. Now + * modify the memory type to NC from the second page of + * the BO onward. + */ flags &= ~AMDGPU_PTE_MTYPE_VG10_MASK; flags |= AMDGPU_PTE_MTYPE_VG10(AMDGPU_MTYPE_NC); @@ -1591,7 +1592,7 @@ static int amdgpu_ttm_access_memory(struct ttm_buffer_object *bo, while (len && pos < adev->gmc.mc_vram_size) { uint64_t aligned_pos = pos & ~(uint64_t)3; - uint32_t bytes = 4 - (pos & 3); + uint64_t bytes = 4 - (pos & 3); uint32_t shift = (pos & 3) * 8; uint32_t mask = 0xffffffff << shift; @@ -1600,20 +1601,28 @@ static int amdgpu_ttm_access_memory(struct ttm_buffer_object *bo, bytes = len; } - spin_lock_irqsave(&adev->mmio_idx_lock, flags); - WREG32_NO_KIQ(mmMM_INDEX, ((uint32_t)aligned_pos) | 0x80000000); - WREG32_NO_KIQ(mmMM_INDEX_HI, aligned_pos >> 31); - if (!write || mask != 0xffffffff) - value = RREG32_NO_KIQ(mmMM_DATA); - if (write) { - value &= ~mask; - value |= (*(uint32_t *)buf << shift) & mask; - WREG32_NO_KIQ(mmMM_DATA, value); - } - spin_unlock_irqrestore(&adev->mmio_idx_lock, flags); - if (!write) { - value = (value & mask) >> shift; - memcpy(buf, &value, bytes); + if (mask != 0xffffffff) { + spin_lock_irqsave(&adev->mmio_idx_lock, flags); + WREG32_NO_KIQ(mmMM_INDEX, ((uint32_t)aligned_pos) | 0x80000000); + WREG32_NO_KIQ(mmMM_INDEX_HI, aligned_pos >> 31); + if (!write || mask != 0xffffffff) + value = RREG32_NO_KIQ(mmMM_DATA); + if (write) { + value &= ~mask; + value |= (*(uint32_t *)buf << shift) & mask; + WREG32_NO_KIQ(mmMM_DATA, value); + } + spin_unlock_irqrestore(&adev->mmio_idx_lock, flags); + if (!write) { + value = (value & mask) >> shift; + memcpy(buf, &value, bytes); + } + } else { + bytes = (nodes->start + nodes->size) << PAGE_SHIFT; + bytes = min(bytes - pos, (uint64_t)len & ~0x3ull); + + amdgpu_device_vram_access(adev, pos, (uint32_t *)buf, + bytes, write); } ret += bytes; @@ -1830,9 +1839,11 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) *The reserved vram for memory training must be pinned to the specified *place on the VRAM, so reserve it early. */ - r = amdgpu_ttm_training_reserve_vram_init(adev); - if (r) - return r; + if (!amdgpu_sriov_vf(adev)) { + r = amdgpu_ttm_training_reserve_vram_init(adev); + if (r) + return r; + } /* allocate memory as required for VGA * This is used for VGA emulation and pre-OS scanout buffers to @@ -1905,12 +1916,6 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) return r; } - /* Register debugfs entries for amdgpu_ttm */ - r = amdgpu_ttm_debugfs_init(adev); - if (r) { - DRM_ERROR("Failed to init debugfs\n"); - return r; - } return 0; } @@ -1932,7 +1937,6 @@ void amdgpu_ttm_fini(struct amdgpu_device *adev) if (!adev->mman.initialized) return; - amdgpu_ttm_debugfs_fini(adev); amdgpu_ttm_training_reserve_vram_fini(adev); /* return the IP Discovery TMR memory back to VRAM */ amdgpu_bo_free_kernel(&adev->discovery_memory, NULL, NULL); @@ -2107,8 +2111,8 @@ int amdgpu_copy_buffer(struct amdgpu_ring *ring, uint64_t src_offset, } if (resv) { r = amdgpu_sync_resv(adev, &job->sync, resv, - AMDGPU_FENCE_OWNER_UNDEFINED, - false); + AMDGPU_SYNC_ALWAYS, + AMDGPU_FENCE_OWNER_UNDEFINED); if (r) { DRM_ERROR("sync failed (%d).\n", r); goto error_free; @@ -2192,7 +2196,8 @@ int amdgpu_fill_buffer(struct amdgpu_bo *bo, if (resv) { r = amdgpu_sync_resv(adev, &job->sync, resv, - AMDGPU_FENCE_OWNER_UNDEFINED, false); + AMDGPU_SYNC_ALWAYS, + AMDGPU_FENCE_OWNER_UNDEFINED); if (r) { DRM_ERROR("sync failed (%d).\n", r); goto error_free; @@ -2273,7 +2278,6 @@ static ssize_t amdgpu_ttm_vram_read(struct file *f, char __user *buf, { struct amdgpu_device *adev = file_inode(f)->i_private; ssize_t result = 0; - int r; if (size & 0x3 || *pos & 0x3) return -EINVAL; @@ -2281,27 +2285,19 @@ static ssize_t amdgpu_ttm_vram_read(struct file *f, char __user *buf, if (*pos >= adev->gmc.mc_vram_size) return -ENXIO; + size = min(size, (size_t)(adev->gmc.mc_vram_size - *pos)); while (size) { - unsigned long flags; - uint32_t value; - - if (*pos >= adev->gmc.mc_vram_size) - return result; + size_t bytes = min(size, AMDGPU_TTM_VRAM_MAX_DW_READ * 4); + uint32_t value[AMDGPU_TTM_VRAM_MAX_DW_READ]; - spin_lock_irqsave(&adev->mmio_idx_lock, flags); - WREG32_NO_KIQ(mmMM_INDEX, ((uint32_t)*pos) | 0x80000000); - WREG32_NO_KIQ(mmMM_INDEX_HI, *pos >> 31); - value = RREG32_NO_KIQ(mmMM_DATA); - spin_unlock_irqrestore(&adev->mmio_idx_lock, flags); - - r = put_user(value, (uint32_t *)buf); - if (r) - return r; + amdgpu_device_vram_access(adev, *pos, value, bytes, false); + if (copy_to_user(buf, value, bytes)) + return -EFAULT; - result += 4; - buf += 4; - *pos += 4; - size -= 4; + result += bytes; + buf += bytes; + *pos += bytes; + size -= bytes; } return result; @@ -2538,7 +2534,7 @@ static const struct { #endif -static int amdgpu_ttm_debugfs_init(struct amdgpu_device *adev) +int amdgpu_ttm_debugfs_init(struct amdgpu_device *adev) { #if defined(CONFIG_DEBUG_FS) unsigned count; @@ -2573,13 +2569,3 @@ static int amdgpu_ttm_debugfs_init(struct amdgpu_device *adev) return 0; #endif } - -static void amdgpu_ttm_debugfs_fini(struct amdgpu_device *adev) -{ -#if defined(CONFIG_DEBUG_FS) - unsigned i; - - for (i = 0; i < ARRAY_SIZE(ttm_debugfs_entries); i++) - debugfs_remove(adev->mman.debugfs_entries[i]); -#endif -} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h index 0dddedc06ae3..bd05bbb4878d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h @@ -133,4 +133,6 @@ uint64_t amdgpu_ttm_tt_pde_flags(struct ttm_tt *ttm, struct ttm_mem_reg *mem); uint64_t amdgpu_ttm_tt_pte_flags(struct amdgpu_device *adev, struct ttm_tt *ttm, struct ttm_mem_reg *mem); +int amdgpu_ttm_debugfs_init(struct amdgpu_device *adev); + #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c index f4d40855147b..9dd51f0d2c11 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c @@ -28,7 +28,6 @@ int amdgpu_umc_ras_late_init(struct amdgpu_device *adev) int r; struct ras_fs_if fs_info = { .sysfs_name = "umc_err_count", - .debugfs_name = "umc_err_inject", }; struct ras_ih_if ih_info = { .cb = amdgpu_umc_process_ras_data_cb, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c index a92f3b18e657..5fd32ad1c575 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c @@ -1099,7 +1099,8 @@ static int amdgpu_uvd_send_msg(struct amdgpu_ring *ring, struct amdgpu_bo *bo, goto err_free; } else { r = amdgpu_sync_resv(adev, &job->sync, bo->tbo.base.resv, - AMDGPU_FENCE_OWNER_UNDEFINED, false); + AMDGPU_SYNC_ALWAYS, + AMDGPU_FENCE_OWNER_UNDEFINED); if (r) goto err_free; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c index f96464e2c157..a41272fbcba2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c @@ -493,14 +493,9 @@ static int amdgpu_vcn_dec_get_destroy_msg(struct amdgpu_ring *ring, uint32_t han int amdgpu_vcn_dec_ring_test_ib(struct amdgpu_ring *ring, long timeout) { - struct amdgpu_device *adev = ring->adev; struct dma_fence *fence; long r; - /* temporarily disable ib test for sriov */ - if (amdgpu_sriov_vf(adev)) - return 0; - r = amdgpu_vcn_dec_get_create_msg(ring, 1, NULL); if (r) goto error; @@ -527,6 +522,9 @@ int amdgpu_vcn_enc_ring_test_ring(struct amdgpu_ring *ring) unsigned i; int r; + if (amdgpu_sriov_vf(adev)) + return 0; + r = amdgpu_ring_alloc(ring, 16); if (r) return r; @@ -656,15 +654,10 @@ err: int amdgpu_vcn_enc_ring_test_ib(struct amdgpu_ring *ring, long timeout) { - struct amdgpu_device *adev = ring->adev; struct dma_fence *fence = NULL; struct amdgpu_bo *bo = NULL; long r; - /* temporarily disable ib test for sriov */ - if (amdgpu_sriov_vf(adev)) - return 0; - r = amdgpu_bo_create_reserved(ring->adev, 128 * 1024, PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM, &bo, NULL, NULL); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h index daaf909d009a..f0128f745bd2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h @@ -270,6 +270,9 @@ struct amdgpu_virt { #define amdgpu_sriov_runtime(adev) \ ((adev)->virt.caps & AMDGPU_SRIOV_CAPS_RUNTIME) +#define amdgpu_sriov_fullaccess(adev) \ +(amdgpu_sriov_vf((adev)) && !amdgpu_sriov_runtime((adev))) + #define amdgpu_passthrough(adev) \ ((adev)->virt.caps & AMDGPU_PASSTHROUGH_MODE) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c index d16231d6a790..6d9252a27916 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c @@ -120,23 +120,17 @@ static inline void amdgpu_vm_eviction_unlock(struct amdgpu_vm *vm) static unsigned amdgpu_vm_level_shift(struct amdgpu_device *adev, unsigned level) { - unsigned shift = 0xff; - switch (level) { case AMDGPU_VM_PDB2: case AMDGPU_VM_PDB1: case AMDGPU_VM_PDB0: - shift = 9 * (AMDGPU_VM_PDB0 - level) + + return 9 * (AMDGPU_VM_PDB0 - level) + adev->vm_manager.block_size; - break; case AMDGPU_VM_PTB: - shift = 0; - break; + return 0; default: - dev_err(adev->dev, "the level%d isn't supported.\n", level); + return ~0; } - - return shift; } /** @@ -235,19 +229,6 @@ static void amdgpu_vm_bo_evicted(struct amdgpu_vm_bo_base *vm_bo) else list_move_tail(&vm_bo->vm_status, &vm->evicted); } - -/** - * amdgpu_vm_bo_relocated - vm_bo is reloacted - * - * @vm_bo: vm_bo which is relocated - * - * State for PDs/PTs which needs to update their parent PD. - */ -static void amdgpu_vm_bo_relocated(struct amdgpu_vm_bo_base *vm_bo) -{ - list_move(&vm_bo->vm_status, &vm_bo->vm->relocated); -} - /** * amdgpu_vm_bo_moved - vm_bo is moved * @@ -291,6 +272,22 @@ static void amdgpu_vm_bo_invalidated(struct amdgpu_vm_bo_base *vm_bo) } /** + * amdgpu_vm_bo_relocated - vm_bo is reloacted + * + * @vm_bo: vm_bo which is relocated + * + * State for PDs/PTs which needs to update their parent PD. + * For the root PD, just move to idle state. + */ +static void amdgpu_vm_bo_relocated(struct amdgpu_vm_bo_base *vm_bo) +{ + if (vm_bo->bo->parent) + list_move(&vm_bo->vm_status, &vm_bo->vm->relocated); + else + amdgpu_vm_bo_idle(vm_bo); +} + +/** * amdgpu_vm_bo_done - vm_bo is done * * @vm_bo: vm_bo which is now done @@ -588,8 +585,8 @@ void amdgpu_vm_get_pd_bo(struct amdgpu_vm *vm, { entry->priority = 0; entry->tv.bo = &vm->root.base.bo->tbo; - /* One for TTM and one for the CS job */ - entry->tv.num_shared = 2; + /* Two for VM updates, one for TTM and one for the CS job */ + entry->tv.num_shared = 4; entry->user_pages = NULL; list_add(&entry->tv.head, validated); } @@ -697,10 +694,7 @@ int amdgpu_vm_validate_pt_bos(struct amdgpu_device *adev, struct amdgpu_vm *vm, amdgpu_vm_bo_moved(bo_base); } else { vm->update_funcs->map_table(bo); - if (bo->parent) - amdgpu_vm_bo_relocated(bo_base); - else - amdgpu_vm_bo_idle(bo_base); + amdgpu_vm_bo_relocated(bo_base); } } @@ -803,7 +797,7 @@ static int amdgpu_vm_clear_bo(struct amdgpu_device *adev, params.vm = vm; params.direct = direct; - r = vm->update_funcs->prepare(¶ms, AMDGPU_FENCE_OWNER_KFD, NULL); + r = vm->update_funcs->prepare(¶ms, NULL, AMDGPU_SYNC_EXPLICIT); if (r) return r; @@ -1086,8 +1080,12 @@ int amdgpu_vm_flush(struct amdgpu_ring *ring, struct amdgpu_job *job, struct dma_fence *fence = NULL; bool pasid_mapping_needed = false; unsigned patch_offset = 0; + bool update_spm_vmid_needed = (job->vm && (job->vm->reserved_vmid[vmhub] != NULL)); int r; + if (update_spm_vmid_needed && adev->gfx.rlc.funcs->update_spm_vmid) + adev->gfx.rlc.funcs->update_spm_vmid(adev, job->vmid); + if (amdgpu_vmid_had_gpu_reset(adev, id)) { gds_switch_needed = true; vm_flush_needed = true; @@ -1299,7 +1297,7 @@ int amdgpu_vm_update_pdes(struct amdgpu_device *adev, params.vm = vm; params.direct = direct; - r = vm->update_funcs->prepare(¶ms, AMDGPU_FENCE_OWNER_VM, NULL); + r = vm->update_funcs->prepare(¶ms, NULL, AMDGPU_SYNC_EXPLICIT); if (r) return r; @@ -1448,21 +1446,14 @@ static int amdgpu_vm_update_ptes(struct amdgpu_vm_update_params *params, uint64_t incr, entry_end, pe_start; struct amdgpu_bo *pt; - /* make sure that the page tables covering the address range are - * actually allocated - */ - r = amdgpu_vm_alloc_pts(params->adev, params->vm, &cursor, - params->direct); - if (r) - return r; - - pt = cursor.entry->base.bo; - - /* The root level can't be a huge page */ - if (cursor.level == adev->vm_manager.root_level) { - if (!amdgpu_vm_pt_descendant(adev, &cursor)) - return -ENOENT; - continue; + if (flags & (AMDGPU_PTE_VALID | AMDGPU_PTE_PRT)) { + /* make sure that the page tables covering the + * address range are actually allocated + */ + r = amdgpu_vm_alloc_pts(params->adev, params->vm, + &cursor, params->direct); + if (r) + return r; } shift = amdgpu_vm_level_shift(adev, cursor.level); @@ -1480,25 +1471,38 @@ static int amdgpu_vm_update_ptes(struct amdgpu_vm_update_params *params, * smaller than the address shift. Go to the next * child entry and try again. */ - if (!amdgpu_vm_pt_descendant(adev, &cursor)) - return -ENOENT; - continue; - } else if (frag >= parent_shift && - cursor.level - 1 != adev->vm_manager.root_level) { + if (amdgpu_vm_pt_descendant(adev, &cursor)) + continue; + } else if (frag >= parent_shift) { /* If the fragment size is even larger than the parent - * shift we should go up one level and check it again - * unless one level up is the root level. + * shift we should go up one level and check it again. */ if (!amdgpu_vm_pt_ancestor(&cursor)) - return -ENOENT; + return -EINVAL; continue; } + pt = cursor.entry->base.bo; + if (!pt) { + /* We need all PDs and PTs for mapping something, */ + if (flags & AMDGPU_PTE_VALID) + return -ENOENT; + + /* but unmapping something can happen at a higher + * level. + */ + if (!amdgpu_vm_pt_ancestor(&cursor)) + return -EINVAL; + + pt = cursor.entry->base.bo; + shift = parent_shift; + } + /* Looks good so far, calculate parameters for the update */ incr = (uint64_t)AMDGPU_GPU_PAGE_SIZE << shift; mask = amdgpu_vm_entries_mask(adev, cursor.level); pe_start = ((cursor.pfn >> shift) & mask) * 8; - entry_end = (uint64_t)(mask + 1) << shift; + entry_end = ((uint64_t)mask + 1) << shift; entry_end += cursor.pfn & ~(entry_end - 1); entry_end = min(entry_end, end); @@ -1506,6 +1510,10 @@ static int amdgpu_vm_update_ptes(struct amdgpu_vm_update_params *params, uint64_t upd_end = min(entry_end, frag_end); unsigned nptes = (upd_end - frag_start) >> shift; + /* This can happen when we set higher level PDs to + * silent to stop fault floods. + */ + nptes = max(nptes, 1u); amdgpu_vm_update_flags(params, pt, cursor.level, pe_start, dst, nptes, incr, flags | AMDGPU_PTE_FRAG(frag)); @@ -1550,7 +1558,7 @@ static int amdgpu_vm_update_ptes(struct amdgpu_vm_update_params *params, * @adev: amdgpu_device pointer * @vm: requested vm * @direct: direct submission in a page fault - * @exclusive: fence we need to sync to + * @resv: fences we need to sync to * @start: start of mapped range * @last: last mapped entry * @flags: flags for the entries @@ -1565,14 +1573,14 @@ static int amdgpu_vm_update_ptes(struct amdgpu_vm_update_params *params, */ static int amdgpu_vm_bo_update_mapping(struct amdgpu_device *adev, struct amdgpu_vm *vm, bool direct, - struct dma_fence *exclusive, + struct dma_resv *resv, uint64_t start, uint64_t last, uint64_t flags, uint64_t addr, dma_addr_t *pages_addr, struct dma_fence **fence) { struct amdgpu_vm_update_params params; - void *owner = AMDGPU_FENCE_OWNER_VM; + enum amdgpu_sync_mode sync_mode; int r; memset(¶ms, 0, sizeof(params)); @@ -1581,9 +1589,13 @@ static int amdgpu_vm_bo_update_mapping(struct amdgpu_device *adev, params.direct = direct; params.pages_addr = pages_addr; - /* sync to everything except eviction fences on unmapping */ + /* Implicitly sync to command submissions in the same VM before + * unmapping. Sync to moving fences before mapping. + */ if (!(flags & AMDGPU_PTE_VALID)) - owner = AMDGPU_FENCE_OWNER_KFD; + sync_mode = AMDGPU_SYNC_EQ_OWNER; + else + sync_mode = AMDGPU_SYNC_EXPLICIT; amdgpu_vm_eviction_lock(vm); if (vm->evicting) { @@ -1591,7 +1603,14 @@ static int amdgpu_vm_bo_update_mapping(struct amdgpu_device *adev, goto error_unlock; } - r = vm->update_funcs->prepare(¶ms, owner, exclusive); + if (flags & (AMDGPU_PTE_VALID | AMDGPU_PTE_PRT)) { + struct amdgpu_bo *root = vm->root.base.bo; + + if (!dma_fence_is_signaled(vm->last_direct)) + amdgpu_bo_fence(root, vm->last_direct, true); + } + + r = vm->update_funcs->prepare(¶ms, resv, sync_mode); if (r) goto error_unlock; @@ -1610,7 +1629,7 @@ error_unlock: * amdgpu_vm_bo_split_mapping - split a mapping into smaller chunks * * @adev: amdgpu_device pointer - * @exclusive: fence we need to sync to + * @resv: fences we need to sync to * @pages_addr: DMA addresses to use for mapping * @vm: requested vm * @mapping: mapped range and flags to use for the update @@ -1626,7 +1645,7 @@ error_unlock: * 0 for success, -EINVAL for failure. */ static int amdgpu_vm_bo_split_mapping(struct amdgpu_device *adev, - struct dma_fence *exclusive, + struct dma_resv *resv, dma_addr_t *pages_addr, struct amdgpu_vm *vm, struct amdgpu_bo_va_mapping *mapping, @@ -1696,13 +1715,13 @@ static int amdgpu_vm_bo_split_mapping(struct amdgpu_device *adev, AMDGPU_GPU_PAGES_IN_CPU_PAGE; } - } else if (flags & AMDGPU_PTE_VALID) { + } else if (flags & (AMDGPU_PTE_VALID | AMDGPU_PTE_PRT)) { addr += bo_adev->vm_manager.vram_base_offset; addr += pfn << PAGE_SHIFT; } last = min((uint64_t)mapping->last, start + max_entries - 1); - r = amdgpu_vm_bo_update_mapping(adev, vm, false, exclusive, + r = amdgpu_vm_bo_update_mapping(adev, vm, false, resv, start, last, flags, addr, dma_addr, fence); if (r) @@ -1741,7 +1760,8 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va, dma_addr_t *pages_addr = NULL; struct ttm_mem_reg *mem; struct drm_mm_node *nodes; - struct dma_fence *exclusive, **last_update; + struct dma_fence **last_update; + struct dma_resv *resv; uint64_t flags; struct amdgpu_device *bo_adev = adev; int r; @@ -1749,7 +1769,7 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va, if (clear || !bo) { mem = NULL; nodes = NULL; - exclusive = NULL; + resv = vm->root.base.bo->tbo.base.resv; } else { struct ttm_dma_tt *ttm; @@ -1759,7 +1779,7 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va, ttm = container_of(bo->tbo.ttm, struct ttm_dma_tt, ttm); pages_addr = ttm->dma_address; } - exclusive = bo->tbo.moving; + resv = bo->tbo.base.resv; } if (bo) { @@ -1769,7 +1789,8 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va, flags = 0x0; } - if (clear || (bo && bo->tbo.base.resv == vm->root.base.bo->tbo.base.resv)) + if (clear || (bo && bo->tbo.base.resv == + vm->root.base.bo->tbo.base.resv)) last_update = &vm->last_update; else last_update = &bo_va->last_pt_update; @@ -1783,7 +1804,7 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va, } list_for_each_entry(mapping, &bo_va->invalids, list) { - r = amdgpu_vm_bo_split_mapping(adev, exclusive, pages_addr, vm, + r = amdgpu_vm_bo_split_mapping(adev, resv, pages_addr, vm, mapping, flags, bo_adev, nodes, last_update); if (r) @@ -1978,6 +1999,7 @@ int amdgpu_vm_clear_freed(struct amdgpu_device *adev, struct amdgpu_vm *vm, struct dma_fence **fence) { + struct dma_resv *resv = vm->root.base.bo->tbo.base.resv; struct amdgpu_bo_va_mapping *mapping; uint64_t init_pte_value = 0; struct dma_fence *f = NULL; @@ -1992,7 +2014,7 @@ int amdgpu_vm_clear_freed(struct amdgpu_device *adev, mapping->start < AMDGPU_GMC_HOLE_START) init_pte_value = AMDGPU_PTE_DEFAULT_ATC; - r = amdgpu_vm_bo_update_mapping(adev, vm, false, NULL, + r = amdgpu_vm_bo_update_mapping(adev, vm, false, resv, mapping->start, mapping->last, init_pte_value, 0, NULL, &f); amdgpu_vm_free_mapping(adev, vm, mapping, f); @@ -2563,8 +2585,7 @@ bool amdgpu_vm_evictable(struct amdgpu_bo *bo) return false; /* Don't evict VM page tables while they are updated */ - if (!dma_fence_is_signaled(bo_base->vm->last_direct) || - !dma_fence_is_signaled(bo_base->vm->last_delayed)) { + if (!dma_fence_is_signaled(bo_base->vm->last_direct)) { amdgpu_vm_eviction_unlock(bo_base->vm); return false; } @@ -2741,11 +2762,7 @@ long amdgpu_vm_wait_idle(struct amdgpu_vm *vm, long timeout) if (timeout <= 0) return timeout; - timeout = dma_fence_wait_timeout(vm->last_direct, true, timeout); - if (timeout <= 0) - return timeout; - - return dma_fence_wait_timeout(vm->last_delayed, true, timeout); + return dma_fence_wait_timeout(vm->last_direct, true, timeout); } /** @@ -2818,7 +2835,6 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm, vm->update_funcs = &amdgpu_vm_sdma_funcs; vm->last_update = NULL; vm->last_direct = dma_fence_get_stub(); - vm->last_delayed = dma_fence_get_stub(); mutex_init(&vm->eviction_lock); vm->evicting = false; @@ -2873,7 +2889,6 @@ error_free_root: error_free_delayed: dma_fence_put(vm->last_direct); - dma_fence_put(vm->last_delayed); drm_sched_entity_destroy(&vm->delayed); error_free_direct: @@ -3076,8 +3091,6 @@ void amdgpu_vm_fini(struct amdgpu_device *adev, struct amdgpu_vm *vm) dma_fence_wait(vm->last_direct, false); dma_fence_put(vm->last_direct); - dma_fence_wait(vm->last_delayed, false); - dma_fence_put(vm->last_delayed); list_for_each_entry_safe(mapping, tmp, &vm->freed, list) { if (mapping->flags & AMDGPU_PTE_PRT && prt_fini_needed) { @@ -3188,6 +3201,7 @@ int amdgpu_vm_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) union drm_amdgpu_vm *args = data; struct amdgpu_device *adev = dev->dev_private; struct amdgpu_fpriv *fpriv = filp->driver_priv; + long timeout = msecs_to_jiffies(2000); int r; switch (args->in.op) { @@ -3199,6 +3213,21 @@ int amdgpu_vm_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) return r; break; case AMDGPU_VM_OP_UNRESERVE_VMID: + if (amdgpu_sriov_runtime(adev)) + timeout = 8 * timeout; + + /* Wait vm idle to make sure the vmid set in SPM_VMID is + * not referenced anymore. + */ + r = amdgpu_bo_reserve(fpriv->vm.root.base.bo, true); + if (r) + return r; + + r = amdgpu_vm_wait_idle(&fpriv->vm, timeout); + if (r < 0) + return r; + + amdgpu_bo_unreserve(fpriv->vm.root.base.bo); amdgpu_vmid_free_reserved(adev, &fpriv->vm, AMDGPU_GFXHUB_0); break; default: diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h index b4640ab38c95..06fe30e1492d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h @@ -227,8 +227,8 @@ struct amdgpu_vm_update_params { struct amdgpu_vm_update_funcs { int (*map_table)(struct amdgpu_bo *bo); - int (*prepare)(struct amdgpu_vm_update_params *p, void * owner, - struct dma_fence *exclusive); + int (*prepare)(struct amdgpu_vm_update_params *p, struct dma_resv *resv, + enum amdgpu_sync_mode sync_mode); int (*update)(struct amdgpu_vm_update_params *p, struct amdgpu_bo *bo, uint64_t pe, uint64_t addr, unsigned count, uint32_t incr, uint64_t flags); @@ -276,7 +276,6 @@ struct amdgpu_vm { /* Last submission to the scheduler entities */ struct dma_fence *last_direct; - struct dma_fence *last_delayed; unsigned int pasid; /* dedicated to vm */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c index 73fec7a0ced5..e38516304070 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c @@ -44,26 +44,14 @@ static int amdgpu_vm_cpu_map_table(struct amdgpu_bo *table) * Returns: * Negativ errno, 0 for success. */ -static int amdgpu_vm_cpu_prepare(struct amdgpu_vm_update_params *p, void *owner, - struct dma_fence *exclusive) +static int amdgpu_vm_cpu_prepare(struct amdgpu_vm_update_params *p, + struct dma_resv *resv, + enum amdgpu_sync_mode sync_mode) { - int r; - - /* Wait for any BO move to be completed */ - if (exclusive) { - r = dma_fence_wait(exclusive, true); - if (unlikely(r)) - return r; - } - - /* Don't wait for submissions during page fault */ - if (p->direct) + if (!resv) return 0; - /* Wait for PT BOs to be idle. PTs share the same resv. object - * as the root PD BO - */ - return amdgpu_bo_sync_wait(p->vm->root.base.bo, owner, true); + return amdgpu_bo_sync_wait_resv(p->adev, resv, sync_mode, p->vm, true); } /** @@ -86,6 +74,13 @@ static int amdgpu_vm_cpu_update(struct amdgpu_vm_update_params *p, { unsigned int i; uint64_t value; + int r; + + if (bo->tbo.moving) { + r = dma_fence_wait(bo->tbo.moving, true); + if (r) + return r; + } pe += (unsigned long)amdgpu_bo_kptr(bo); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c index 19b7f80758f1..cf96c335b258 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c @@ -58,9 +58,9 @@ static int amdgpu_vm_sdma_map_table(struct amdgpu_bo *table) * Negativ errno, 0 for success. */ static int amdgpu_vm_sdma_prepare(struct amdgpu_vm_update_params *p, - void *owner, struct dma_fence *exclusive) + struct dma_resv *resv, + enum amdgpu_sync_mode sync_mode) { - struct amdgpu_bo *root = p->vm->root.base.bo; unsigned int ndw = AMDGPU_VM_SDMA_MIN_NUM_DW; int r; @@ -70,17 +70,10 @@ static int amdgpu_vm_sdma_prepare(struct amdgpu_vm_update_params *p, p->num_dw_left = ndw; - /* Wait for moves to be completed */ - r = amdgpu_sync_fence(&p->job->sync, exclusive, false); - if (r) - return r; - - /* Don't wait for any submissions during page fault handling */ - if (p->direct) + if (!resv) return 0; - return amdgpu_sync_resv(p->adev, &p->job->sync, root->tbo.base.resv, - owner, false); + return amdgpu_sync_resv(p->adev, &p->job->sync, resv, sync_mode, p->vm); } /** @@ -111,12 +104,13 @@ static int amdgpu_vm_sdma_commit(struct amdgpu_vm_update_params *p, if (r) goto error; - tmp = dma_fence_get(f); - if (p->direct) + if (p->direct) { + tmp = dma_fence_get(f); swap(p->vm->last_direct, tmp); - else - swap(p->vm->last_delayed, tmp); - dma_fence_put(tmp); + dma_fence_put(tmp); + } else { + dma_resv_add_shared_fence(p->vm->root.base.bo->tbo.base.resv, f); + } if (fence && !p->direct) swap(*fence, f); @@ -147,7 +141,7 @@ static void amdgpu_vm_sdma_copy_ptes(struct amdgpu_vm_update_params *p, src += p->num_dw_left * 4; - pe += amdgpu_bo_gpu_offset(bo); + pe += amdgpu_gmc_sign_extend(bo->tbo.offset); trace_amdgpu_vm_copy_ptes(pe, src, count, p->direct); amdgpu_vm_copy_pte(p->adev, ib, pe, src, count); @@ -174,7 +168,7 @@ static void amdgpu_vm_sdma_set_ptes(struct amdgpu_vm_update_params *p, { struct amdgpu_ib *ib = p->job->ibs; - pe += amdgpu_bo_gpu_offset(bo); + pe += amdgpu_gmc_sign_extend(bo->tbo.offset); trace_amdgpu_vm_set_ptes(pe, addr, count, incr, flags, p->direct); if (count < 3) { amdgpu_vm_write_pte(p->adev, ib, pe, addr | flags, @@ -208,6 +202,11 @@ static int amdgpu_vm_sdma_update(struct amdgpu_vm_update_params *p, uint64_t *pte; int r; + /* Wait for PD/PT moves to be completed */ + r = amdgpu_sync_fence(&p->job->sync, bo->tbo.moving, false); + if (r) + return r; + do { ndw = p->num_dw_left; ndw -= p->job->ibs->length_dw; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c index a97af422575a..95b3327168ac 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c @@ -26,7 +26,12 @@ #include "amdgpu_xgmi.h" #include "amdgpu_smu.h" #include "amdgpu_ras.h" +#include "soc15.h" #include "df/df_3_6_offset.h" +#include "xgmi/xgmi_4_0_0_smn.h" +#include "xgmi/xgmi_4_0_0_sh_mask.h" +#include "wafl/wafl2_4_0_0_smn.h" +#include "wafl/wafl2_4_0_0_sh_mask.h" static DEFINE_MUTEX(xgmi_mutex); @@ -36,6 +41,109 @@ static DEFINE_MUTEX(xgmi_mutex); static struct amdgpu_hive_info xgmi_hives[AMDGPU_MAX_XGMI_HIVE]; static unsigned hive_count = 0; +static const int xgmi_pcs_err_status_reg_vg20[] = { + smnXGMI0_PCS_GOPX16_PCS_ERROR_STATUS, + smnXGMI0_PCS_GOPX16_PCS_ERROR_STATUS + 0x100000, +}; + +static const int wafl_pcs_err_status_reg_vg20[] = { + smnPCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, + smnPCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS + 0x100000, +}; + +static const int xgmi_pcs_err_status_reg_arct[] = { + smnXGMI0_PCS_GOPX16_PCS_ERROR_STATUS, + smnXGMI0_PCS_GOPX16_PCS_ERROR_STATUS + 0x100000, + smnXGMI0_PCS_GOPX16_PCS_ERROR_STATUS + 0x500000, + smnXGMI0_PCS_GOPX16_PCS_ERROR_STATUS + 0x600000, + smnXGMI0_PCS_GOPX16_PCS_ERROR_STATUS + 0x700000, + smnXGMI0_PCS_GOPX16_PCS_ERROR_STATUS + 0x800000, +}; + +/* same as vg20*/ +static const int wafl_pcs_err_status_reg_arct[] = { + smnPCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, + smnPCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS + 0x100000, +}; + +static const struct amdgpu_pcs_ras_field xgmi_pcs_ras_fields[] = { + {"XGMI PCS DataLossErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, DataLossErr)}, + {"XGMI PCS TrainingErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, TrainingErr)}, + {"XGMI PCS CRCErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, CRCErr)}, + {"XGMI PCS BERExceededErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, BERExceededErr)}, + {"XGMI PCS TxMetaDataErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, TxMetaDataErr)}, + {"XGMI PCS ReplayBufParityErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, ReplayBufParityErr)}, + {"XGMI PCS DataParityErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, DataParityErr)}, + {"XGMI PCS ReplayFifoOverflowErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, ReplayFifoOverflowErr)}, + {"XGMI PCS ReplayFifoUnderflowErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, ReplayFifoUnderflowErr)}, + {"XGMI PCS ElasticFifoOverflowErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, ElasticFifoOverflowErr)}, + {"XGMI PCS DeskewErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, DeskewErr)}, + {"XGMI PCS DataStartupLimitErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, DataStartupLimitErr)}, + {"XGMI PCS FCInitTimeoutErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, FCInitTimeoutErr)}, + {"XGMI PCS RecoveryTimeoutErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, RecoveryTimeoutErr)}, + {"XGMI PCS ReadySerialTimeoutErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, ReadySerialTimeoutErr)}, + {"XGMI PCS ReadySerialAttemptErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, ReadySerialAttemptErr)}, + {"XGMI PCS RecoveryAttemptErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, RecoveryAttemptErr)}, + {"XGMI PCS RecoveryRelockAttemptErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, RecoveryRelockAttemptErr)}, +}; + +static const struct amdgpu_pcs_ras_field wafl_pcs_ras_fields[] = { + {"WAFL PCS DataLossErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, DataLossErr)}, + {"WAFL PCS TrainingErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, TrainingErr)}, + {"WAFL PCS CRCErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, CRCErr)}, + {"WAFL PCS BERExceededErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, BERExceededErr)}, + {"WAFL PCS TxMetaDataErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, TxMetaDataErr)}, + {"WAFL PCS ReplayBufParityErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, ReplayBufParityErr)}, + {"WAFL PCS DataParityErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, DataParityErr)}, + {"WAFL PCS ReplayFifoOverflowErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, ReplayFifoOverflowErr)}, + {"WAFL PCS ReplayFifoUnderflowErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, ReplayFifoUnderflowErr)}, + {"WAFL PCS ElasticFifoOverflowErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, ElasticFifoOverflowErr)}, + {"WAFL PCS DeskewErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, DeskewErr)}, + {"WAFL PCS DataStartupLimitErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, DataStartupLimitErr)}, + {"WAFL PCS FCInitTimeoutErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, FCInitTimeoutErr)}, + {"WAFL PCS RecoveryTimeoutErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, RecoveryTimeoutErr)}, + {"WAFL PCS ReadySerialTimeoutErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, ReadySerialTimeoutErr)}, + {"WAFL PCS ReadySerialAttemptErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, ReadySerialAttemptErr)}, + {"WAFL PCS RecoveryAttemptErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, RecoveryAttemptErr)}, + {"WAFL PCS RecoveryRelockAttemptErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, RecoveryRelockAttemptErr)}, +}; + void *amdgpu_xgmi_hive_try_lock(struct amdgpu_hive_info *hive) { return &hive->device_list; @@ -365,6 +473,13 @@ int amdgpu_xgmi_add_device(struct amdgpu_device *adev) return 0; if (amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_PSP)) { + ret = psp_xgmi_initialize(&adev->psp); + if (ret) { + dev_err(adev->dev, + "XGMI: Failed to initialize xgmi session\n"); + return ret; + } + ret = psp_xgmi_get_hive_id(&adev->psp, &adev->gmc.xgmi.hive_id); if (ret) { dev_err(adev->dev, @@ -451,16 +566,16 @@ exit: return ret; } -void amdgpu_xgmi_remove_device(struct amdgpu_device *adev) +int amdgpu_xgmi_remove_device(struct amdgpu_device *adev) { struct amdgpu_hive_info *hive; if (!adev->gmc.xgmi.supported) - return; + return -EINVAL; hive = amdgpu_get_xgmi_hive(adev, 1); if (!hive) - return; + return -EINVAL; if (!(hive->number_devices--)) { amdgpu_xgmi_sysfs_destroy(adev, hive); @@ -471,6 +586,8 @@ void amdgpu_xgmi_remove_device(struct amdgpu_device *adev) amdgpu_xgmi_sysfs_rem_dev_info(adev, hive); mutex_unlock(&hive->hive_lock); } + + return psp_xgmi_terminate(&adev->psp); } int amdgpu_xgmi_ras_late_init(struct amdgpu_device *adev) @@ -481,7 +598,6 @@ int amdgpu_xgmi_ras_late_init(struct amdgpu_device *adev) }; struct ras_fs_if fs_info = { .sysfs_name = "xgmi_wafl_err_count", - .debugfs_name = "xgmi_wafl_err_inject", }; if (!adev->gmc.xgmi.supported || @@ -521,3 +637,129 @@ void amdgpu_xgmi_ras_fini(struct amdgpu_device *adev) kfree(ras_if); } } + +uint64_t amdgpu_xgmi_get_relative_phy_addr(struct amdgpu_device *adev, + uint64_t addr) +{ + uint32_t df_inst_id; + uint64_t dram_base_addr = 0; + const struct amdgpu_df_funcs *df_funcs = adev->df.funcs; + + if ((!df_funcs) || + (!df_funcs->get_df_inst_id) || + (!df_funcs->get_dram_base_addr)) { + dev_warn(adev->dev, + "XGMI: relative phy_addr algorithm is not supported\n"); + return addr; + } + + if (amdgpu_dpm_set_df_cstate(adev, DF_CSTATE_DISALLOW)) { + dev_warn(adev->dev, + "failed to disable DF-Cstate, DF register may not be accessible\n"); + return addr; + } + + df_inst_id = df_funcs->get_df_inst_id(adev); + dram_base_addr = df_funcs->get_dram_base_addr(adev, df_inst_id); + + if (amdgpu_dpm_set_df_cstate(adev, DF_CSTATE_ALLOW)) + dev_warn(adev->dev, "failed to enable DF-Cstate\n"); + + return addr + dram_base_addr; +} + +static int amdgpu_xgmi_query_pcs_error_status(struct amdgpu_device *adev, + uint32_t value, + uint32_t *ue_count, + uint32_t *ce_count, + bool is_xgmi_pcs) +{ + int i; + int ue_cnt; + + if (is_xgmi_pcs) { + /* query xgmi pcs error status, + * only ue is supported */ + for (i = 0; i < ARRAY_SIZE(xgmi_pcs_ras_fields); i ++) { + ue_cnt = (value & + xgmi_pcs_ras_fields[i].pcs_err_mask) >> + xgmi_pcs_ras_fields[i].pcs_err_shift; + if (ue_cnt) { + dev_info(adev->dev, "%s detected\n", + xgmi_pcs_ras_fields[i].err_name); + *ue_count += ue_cnt; + } + } + } else { + /* query wafl pcs error status, + * only ue is supported */ + for (i = 0; i < ARRAY_SIZE(wafl_pcs_ras_fields); i++) { + ue_cnt = (value & + wafl_pcs_ras_fields[i].pcs_err_mask) >> + wafl_pcs_ras_fields[i].pcs_err_shift; + if (ue_cnt) { + dev_info(adev->dev, "%s detected\n", + wafl_pcs_ras_fields[i].err_name); + *ue_count += ue_cnt; + } + } + } + + return 0; +} + +int amdgpu_xgmi_query_ras_error_count(struct amdgpu_device *adev, + void *ras_error_status) +{ + struct ras_err_data *err_data = (struct ras_err_data *)ras_error_status; + int i; + uint32_t data; + uint32_t ue_cnt = 0, ce_cnt = 0; + + if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__XGMI_WAFL)) + return -EINVAL; + + err_data->ue_count = 0; + err_data->ce_count = 0; + + switch (adev->asic_type) { + case CHIP_ARCTURUS: + /* check xgmi pcs error */ + for (i = 0; i < ARRAY_SIZE(xgmi_pcs_err_status_reg_arct); i++) { + data = RREG32_PCIE(xgmi_pcs_err_status_reg_arct[i]); + if (data) + amdgpu_xgmi_query_pcs_error_status(adev, + data, &ue_cnt, &ce_cnt, true); + } + /* check wafl pcs error */ + for (i = 0; i < ARRAY_SIZE(wafl_pcs_err_status_reg_arct); i++) { + data = RREG32_PCIE(wafl_pcs_err_status_reg_arct[i]); + if (data) + amdgpu_xgmi_query_pcs_error_status(adev, + data, &ue_cnt, &ce_cnt, false); + } + break; + case CHIP_VEGA20: + default: + /* check xgmi pcs error */ + for (i = 0; i < ARRAY_SIZE(xgmi_pcs_err_status_reg_vg20); i++) { + data = RREG32_PCIE(xgmi_pcs_err_status_reg_vg20[i]); + if (data) + amdgpu_xgmi_query_pcs_error_status(adev, + data, &ue_cnt, &ce_cnt, true); + } + /* check wafl pcs error */ + for (i = 0; i < ARRAY_SIZE(wafl_pcs_err_status_reg_vg20); i++) { + data = RREG32_PCIE(wafl_pcs_err_status_reg_vg20[i]); + if (data) + amdgpu_xgmi_query_pcs_error_status(adev, + data, &ue_cnt, &ce_cnt, false); + } + break; + } + + err_data->ue_count += ue_cnt; + err_data->ce_count += ce_cnt; + + return 0; +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h index 74011fbc2251..4a92067fe595 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h @@ -37,15 +37,25 @@ struct amdgpu_hive_info { struct task_barrier tb; }; +struct amdgpu_pcs_ras_field { + const char *err_name; + uint32_t pcs_err_mask; + uint32_t pcs_err_shift; +}; + struct amdgpu_hive_info *amdgpu_get_xgmi_hive(struct amdgpu_device *adev, int lock); int amdgpu_xgmi_update_topology(struct amdgpu_hive_info *hive, struct amdgpu_device *adev); int amdgpu_xgmi_add_device(struct amdgpu_device *adev); -void amdgpu_xgmi_remove_device(struct amdgpu_device *adev); +int amdgpu_xgmi_remove_device(struct amdgpu_device *adev); int amdgpu_xgmi_set_pstate(struct amdgpu_device *adev, int pstate); int amdgpu_xgmi_get_hops_count(struct amdgpu_device *adev, struct amdgpu_device *peer_adev); int amdgpu_xgmi_ras_late_init(struct amdgpu_device *adev); void amdgpu_xgmi_ras_fini(struct amdgpu_device *adev); +uint64_t amdgpu_xgmi_get_relative_phy_addr(struct amdgpu_device *adev, + uint64_t addr); +int amdgpu_xgmi_query_ras_error_count(struct amdgpu_device *adev, + void *ras_error_status); static inline bool amdgpu_xgmi_same_hive(struct amdgpu_device *adev, struct amdgpu_device *bo_adev) diff --git a/drivers/gpu/drm/amd/amdgpu/atom.c b/drivers/gpu/drm/amd/amdgpu/atom.c index dd30f4e61a8c..cae426c7c086 100644 --- a/drivers/gpu/drm/amd/amdgpu/atom.c +++ b/drivers/gpu/drm/amd/amdgpu/atom.c @@ -744,8 +744,8 @@ static void atom_op_jump(atom_exec_context *ctx, int *ptr, int arg) cjiffies = jiffies; if (time_after(cjiffies, ctx->last_jump_jiffies)) { cjiffies -= ctx->last_jump_jiffies; - if ((jiffies_to_msecs(cjiffies) > 5000)) { - DRM_ERROR("atombios stuck in loop for more than 5secs aborting\n"); + if ((jiffies_to_msecs(cjiffies) > 10000)) { + DRM_ERROR("atombios stuck in loop for more than 10secs aborting\n"); ctx->abort = true; } } else { diff --git a/drivers/gpu/drm/amd/amdgpu/atombios_dp.c b/drivers/gpu/drm/amd/amdgpu/atombios_dp.c index ea702a64f807..9b74cfdba7b8 100644 --- a/drivers/gpu/drm/amd/amdgpu/atombios_dp.c +++ b/drivers/gpu/drm/amd/amdgpu/atombios_dp.c @@ -186,16 +186,10 @@ amdgpu_atombios_dp_aux_transfer(struct drm_dp_aux *aux, struct drm_dp_aux_msg *m void amdgpu_atombios_dp_aux_init(struct amdgpu_connector *amdgpu_connector) { - int ret; - amdgpu_connector->ddc_bus->rec.hpd = amdgpu_connector->hpd.hpd; - amdgpu_connector->ddc_bus->aux.dev = amdgpu_connector->base.kdev; amdgpu_connector->ddc_bus->aux.transfer = amdgpu_atombios_dp_aux_transfer; - ret = drm_dp_aux_register(&amdgpu_connector->ddc_bus->aux); - if (!ret) - amdgpu_connector->ddc_bus->has_aux = true; - - WARN(ret, "drm_dp_aux_register_i2c_bus() failed with error %d\n", ret); + drm_dp_aux_init(&amdgpu_connector->ddc_bus->aux); + amdgpu_connector->ddc_bus->has_aux = true; } /***** general DP utility functions *****/ diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c index 40d2ac723dd6..2512e7ebfedf 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c @@ -2494,6 +2494,10 @@ static const struct drm_crtc_funcs dce_v10_0_crtc_funcs = { .set_config = amdgpu_display_crtc_set_config, .destroy = dce_v10_0_crtc_destroy, .page_flip_target = amdgpu_display_crtc_page_flip_target, + .get_vblank_counter = amdgpu_get_vblank_counter_kms, + .enable_vblank = amdgpu_enable_vblank_kms, + .disable_vblank = amdgpu_disable_vblank_kms, + .get_vblank_timestamp = drm_crtc_vblank_helper_get_vblank_timestamp, }; static void dce_v10_0_crtc_dpms(struct drm_crtc *crtc, int mode) @@ -2685,6 +2689,7 @@ static const struct drm_crtc_helper_funcs dce_v10_0_crtc_helper_funcs = { .prepare = dce_v10_0_crtc_prepare, .commit = dce_v10_0_crtc_commit, .disable = dce_v10_0_crtc_disable, + .get_scanout_position = amdgpu_crtc_get_scanout_position, }; static int dce_v10_0_crtc_init(struct amdgpu_device *adev, int index) diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c index 898ef72d423c..0dde22db9848 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c @@ -2573,6 +2573,10 @@ static const struct drm_crtc_funcs dce_v11_0_crtc_funcs = { .set_config = amdgpu_display_crtc_set_config, .destroy = dce_v11_0_crtc_destroy, .page_flip_target = amdgpu_display_crtc_page_flip_target, + .get_vblank_counter = amdgpu_get_vblank_counter_kms, + .enable_vblank = amdgpu_enable_vblank_kms, + .disable_vblank = amdgpu_disable_vblank_kms, + .get_vblank_timestamp = drm_crtc_vblank_helper_get_vblank_timestamp, }; static void dce_v11_0_crtc_dpms(struct drm_crtc *crtc, int mode) @@ -2793,6 +2797,7 @@ static const struct drm_crtc_helper_funcs dce_v11_0_crtc_helper_funcs = { .prepare = dce_v11_0_crtc_prepare, .commit = dce_v11_0_crtc_commit, .disable = dce_v11_0_crtc_disable, + .get_scanout_position = amdgpu_crtc_get_scanout_position, }; static int dce_v11_0_crtc_init(struct amdgpu_device *adev, int index) diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c index db15a112becc..84219534bd38 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c @@ -2388,6 +2388,10 @@ static const struct drm_crtc_funcs dce_v6_0_crtc_funcs = { .set_config = amdgpu_display_crtc_set_config, .destroy = dce_v6_0_crtc_destroy, .page_flip_target = amdgpu_display_crtc_page_flip_target, + .get_vblank_counter = amdgpu_get_vblank_counter_kms, + .enable_vblank = amdgpu_enable_vblank_kms, + .disable_vblank = amdgpu_disable_vblank_kms, + .get_vblank_timestamp = drm_crtc_vblank_helper_get_vblank_timestamp, }; static void dce_v6_0_crtc_dpms(struct drm_crtc *crtc, int mode) @@ -2575,6 +2579,7 @@ static const struct drm_crtc_helper_funcs dce_v6_0_crtc_helper_funcs = { .prepare = dce_v6_0_crtc_prepare, .commit = dce_v6_0_crtc_commit, .disable = dce_v6_0_crtc_disable, + .get_scanout_position = amdgpu_crtc_get_scanout_position, }; static int dce_v6_0_crtc_init(struct amdgpu_device *adev, int index) diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c index f06c9022c1fd..3a640702d7d1 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c @@ -2395,6 +2395,10 @@ static const struct drm_crtc_funcs dce_v8_0_crtc_funcs = { .set_config = amdgpu_display_crtc_set_config, .destroy = dce_v8_0_crtc_destroy, .page_flip_target = amdgpu_display_crtc_page_flip_target, + .get_vblank_counter = amdgpu_get_vblank_counter_kms, + .enable_vblank = amdgpu_enable_vblank_kms, + .disable_vblank = amdgpu_disable_vblank_kms, + .get_vblank_timestamp = drm_crtc_vblank_helper_get_vblank_timestamp, }; static void dce_v8_0_crtc_dpms(struct drm_crtc *crtc, int mode) @@ -2593,6 +2597,7 @@ static const struct drm_crtc_helper_funcs dce_v8_0_crtc_helper_funcs = { .prepare = dce_v8_0_crtc_prepare, .commit = dce_v8_0_crtc_commit, .disable = dce_v8_0_crtc_disable, + .get_scanout_position = amdgpu_crtc_get_scanout_position, }; static int dce_v8_0_crtc_init(struct amdgpu_device *adev, int index) diff --git a/drivers/gpu/drm/amd/amdgpu/dce_virtual.c b/drivers/gpu/drm/amd/amdgpu/dce_virtual.c index e4f94863332c..13e12be667fc 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_virtual.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_virtual.c @@ -123,6 +123,10 @@ static const struct drm_crtc_funcs dce_virtual_crtc_funcs = { .set_config = amdgpu_display_crtc_set_config, .destroy = dce_virtual_crtc_destroy, .page_flip_target = amdgpu_display_crtc_page_flip_target, + .get_vblank_counter = amdgpu_get_vblank_counter_kms, + .enable_vblank = amdgpu_enable_vblank_kms, + .disable_vblank = amdgpu_disable_vblank_kms, + .get_vblank_timestamp = drm_crtc_vblank_helper_get_vblank_timestamp, }; static void dce_virtual_crtc_dpms(struct drm_crtc *crtc, int mode) @@ -218,6 +222,7 @@ static const struct drm_crtc_helper_funcs dce_virtual_crtc_helper_funcs = { .prepare = dce_virtual_crtc_prepare, .commit = dce_virtual_crtc_commit, .disable = dce_virtual_crtc_disable, + .get_scanout_position = amdgpu_crtc_get_scanout_position, }; static int dce_virtual_crtc_init(struct amdgpu_device *adev, int index) @@ -609,7 +614,6 @@ static int dce_virtual_connector_encoder_init(struct amdgpu_device *adev, connector->display_info.subpixel_order = SubPixelHorizontalRGB; connector->interlace_allowed = false; connector->doublescan_allowed = false; - drm_connector_register(connector); /* link them */ drm_connector_attach_encoder(connector, encoder); diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c index 1785fdad6ecb..d78059fd2c72 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c @@ -35,6 +35,8 @@ #include "gc/gc_10_1_0_offset.h" #include "gc/gc_10_1_0_sh_mask.h" +#include "smuio/smuio_11_0_0_offset.h" +#include "smuio/smuio_11_0_0_sh_mask.h" #include "navi10_enum.h" #include "hdp/hdp_5_0_0_offset.h" #include "ivsrcid/gfx/irqsrcs_gfx_10_1.h" @@ -52,7 +54,7 @@ * 1. Primary ring * 2. Async ring */ -#define GFX10_NUM_GFX_RINGS 2 +#define GFX10_NUM_GFX_RINGS_NV1X 1 #define GFX10_MEC_HPD_SIZE 2048 #define F32_CE_PROGRAM_RAM_SIZE 65536 @@ -222,6 +224,49 @@ static const struct soc15_reg_golden golden_settings_gc_10_1_2[] = SOC15_REG_GOLDEN_VALUE(GC, 0, mmUTCL1_CTRL, 0xffffffff, 0x00800000) }; +static void gfx_v10_rlcg_wreg(struct amdgpu_device *adev, u32 offset, u32 v) +{ + static void *scratch_reg0; + static void *scratch_reg1; + static void *scratch_reg2; + static void *scratch_reg3; + static void *spare_int; + static uint32_t grbm_cntl; + static uint32_t grbm_idx; + uint32_t i = 0; + uint32_t retries = 50000; + + scratch_reg0 = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG0_BASE_IDX] + mmSCRATCH_REG0)*4; + scratch_reg1 = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG1_BASE_IDX] + mmSCRATCH_REG1)*4; + scratch_reg2 = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG1_BASE_IDX] + mmSCRATCH_REG2)*4; + scratch_reg3 = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG1_BASE_IDX] + mmSCRATCH_REG3)*4; + spare_int = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmRLC_SPARE_INT_BASE_IDX] + mmRLC_SPARE_INT)*4; + + grbm_cntl = adev->reg_offset[GC_HWIP][0][mmGRBM_GFX_CNTL_BASE_IDX] + mmGRBM_GFX_CNTL; + grbm_idx = adev->reg_offset[GC_HWIP][0][mmGRBM_GFX_INDEX_BASE_IDX] + mmGRBM_GFX_INDEX; + + if (amdgpu_sriov_runtime(adev)) { + pr_err("shouldn't call rlcg write register during runtime\n"); + return; + } + + writel(v, scratch_reg0); + writel(offset | 0x80000000, scratch_reg1); + writel(1, spare_int); + for (i = 0; i < retries; i++) { + u32 tmp; + + tmp = readl(scratch_reg1); + if (!(tmp & 0x80000000)) + break; + + udelay(10); + } + + if (i >= retries) + pr_err("timeout: rlcg program reg:0x%05x failed !\n", offset); +} + static const struct soc15_reg_golden golden_settings_gc_10_1_nv14[] = { /* Pending on emulation bring up */ @@ -234,7 +279,7 @@ static const struct soc15_reg_golden golden_settings_gc_10_1_2_nv12[] = #define DEFAULT_SH_MEM_CONFIG \ ((SH_MEM_ADDRESS_MODE_64 << SH_MEM_CONFIG__ADDRESS_MODE__SHIFT) | \ - (SH_MEM_ALIGNMENT_MODE_UNALIGNED << SH_MEM_CONFIG__ALIGNMENT_MODE__SHIFT) | \ + (SH_MEM_ALIGNMENT_MODE_DWORD << SH_MEM_CONFIG__ALIGNMENT_MODE__SHIFT) | \ (SH_MEM_RETRY_MODE_ALL << SH_MEM_CONFIG__RETRY_MODE__SHIFT) | \ (3 << SH_MEM_CONFIG__INITIAL_INST_PREFETCH__SHIFT)) @@ -500,29 +545,28 @@ static int gfx_v10_0_ring_test_ib(struct amdgpu_ring *ring, long timeout) struct amdgpu_device *adev = ring->adev; struct amdgpu_ib ib; struct dma_fence *f = NULL; - uint32_t scratch; - uint32_t tmp = 0; + unsigned index; + uint64_t gpu_addr; + uint32_t tmp; long r; - r = amdgpu_gfx_scratch_get(adev, &scratch); - if (r) { - DRM_ERROR("amdgpu: failed to get scratch reg (%ld).\n", r); + r = amdgpu_device_wb_get(adev, &index); + if (r) return r; - } - - WREG32(scratch, 0xCAFEDEAD); + gpu_addr = adev->wb.gpu_addr + (index * 4); + adev->wb.wb[index] = cpu_to_le32(0xCAFEDEAD); memset(&ib, 0, sizeof(ib)); - r = amdgpu_ib_get(adev, NULL, 256, &ib); - if (r) { - DRM_ERROR("amdgpu: failed to get ib (%ld).\n", r); + r = amdgpu_ib_get(adev, NULL, 16, &ib); + if (r) goto err1; - } - ib.ptr[0] = PACKET3(PACKET3_SET_UCONFIG_REG, 1); - ib.ptr[1] = ((scratch - PACKET3_SET_UCONFIG_REG_START)); - ib.ptr[2] = 0xDEADBEEF; - ib.length_dw = 3; + ib.ptr[0] = PACKET3(PACKET3_WRITE_DATA, 3); + ib.ptr[1] = WRITE_DATA_DST_SEL(5) | WR_CONFIRM; + ib.ptr[2] = lower_32_bits(gpu_addr); + ib.ptr[3] = upper_32_bits(gpu_addr); + ib.ptr[4] = 0xDEADBEEF; + ib.length_dw = 5; r = amdgpu_ib_schedule(ring, 1, &ib, NULL, &f); if (r) @@ -530,15 +574,13 @@ static int gfx_v10_0_ring_test_ib(struct amdgpu_ring *ring, long timeout) r = dma_fence_wait_timeout(f, false, timeout); if (r == 0) { - DRM_ERROR("amdgpu: IB test timed out.\n"); r = -ETIMEDOUT; goto err2; } else if (r < 0) { - DRM_ERROR("amdgpu: fence wait failed (%ld).\n", r); goto err2; } - tmp = RREG32(scratch); + tmp = adev->wb.wb[index]; if (tmp == 0xDEADBEEF) r = 0; else @@ -547,8 +589,7 @@ err2: amdgpu_ib_free(adev, &ib, NULL); dma_fence_put(f); err1: - amdgpu_gfx_scratch_free(adev, scratch); - + amdgpu_device_wb_free(adev, index); return r; } @@ -1016,6 +1057,10 @@ static int gfx_v10_0_rlc_init(struct amdgpu_device *adev) return r; } + /* init spm vmid with 0xf */ + if (adev->gfx.rlc.funcs->update_spm_vmid) + adev->gfx.rlc.funcs->update_spm_vmid(adev, 0xf); + return 0; } @@ -1068,7 +1113,7 @@ static int gfx_v10_0_mec_init(struct amdgpu_device *adev) return r; } - memset(hpd, 0, adev->gfx.mec.hpd_eop_obj->tbo.mem.size); + memset(hpd, 0, mec_hpd_size); amdgpu_bo_kunmap(adev->gfx.mec.hpd_eop_obj); amdgpu_bo_unreserve(adev->gfx.mec.hpd_eop_obj); @@ -1304,7 +1349,7 @@ static int gfx_v10_0_sw_init(void *handle) case CHIP_NAVI14: case CHIP_NAVI12: adev->gfx.me.num_me = 1; - adev->gfx.me.num_pipe_per_me = 2; + adev->gfx.me.num_pipe_per_me = 1; adev->gfx.me.num_queue_per_pipe = 1; adev->gfx.mec.num_mec = 2; adev->gfx.mec.num_pipe_per_mec = 4; @@ -1783,11 +1828,11 @@ static int gfx_v10_0_init_csb(struct amdgpu_device *adev) adev->gfx.rlc.funcs->get_csb_buffer(adev, adev->gfx.rlc.cs_ptr); /* csib */ - WREG32_SOC15(GC, 0, mmRLC_CSIB_ADDR_HI, + WREG32_SOC15_RLC(GC, 0, mmRLC_CSIB_ADDR_HI, adev->gfx.rlc.clear_state_gpu_addr >> 32); - WREG32_SOC15(GC, 0, mmRLC_CSIB_ADDR_LO, + WREG32_SOC15_RLC(GC, 0, mmRLC_CSIB_ADDR_LO, adev->gfx.rlc.clear_state_gpu_addr & 0xfffffffc); - WREG32_SOC15(GC, 0, mmRLC_CSIB_LENGTH, adev->gfx.rlc.clear_state_size); + WREG32_SOC15_RLC(GC, 0, mmRLC_CSIB_LENGTH, adev->gfx.rlc.clear_state_size); return 0; } @@ -1895,6 +1940,11 @@ static int gfx_v10_0_rlc_resume(struct amdgpu_device *adev) if (!amdgpu_sriov_vf(adev)) /* enable RLC SRM */ gfx_v10_0_rlc_enable_srm(adev); } else { + if (amdgpu_sriov_vf(adev)) { + gfx_v10_0_init_csb(adev); + return 0; + } + adev->gfx.rlc.funcs->stop(adev); /* disable CG */ @@ -2395,7 +2445,7 @@ static int gfx_v10_0_cp_gfx_enable(struct amdgpu_device *adev, bool enable) for (i = 0; i < adev->gfx.num_gfx_rings; i++) adev->gfx.gfx_ring[i].sched.ready = false; } - WREG32_SOC15(GC, 0, mmCP_ME_CNTL, tmp); + WREG32_SOC15_RLC(GC, 0, mmCP_ME_CNTL, tmp); for (i = 0; i < adev->usec_timeout; i++) { if (RREG32_SOC15(GC, 0, mmCP_STAT) == 0) @@ -2710,18 +2760,20 @@ static int gfx_v10_0_cp_gfx_start(struct amdgpu_device *adev) amdgpu_ring_commit(ring); /* submit cs packet to copy state 0 to next available state */ - ring = &adev->gfx.gfx_ring[1]; - r = amdgpu_ring_alloc(ring, 2); - if (r) { - DRM_ERROR("amdgpu: cp failed to lock ring (%d).\n", r); - return r; - } - - amdgpu_ring_write(ring, PACKET3(PACKET3_CLEAR_STATE, 0)); - amdgpu_ring_write(ring, 0); + if (adev->gfx.num_gfx_rings > 1) { + /* maximum supported gfx ring is 2 */ + ring = &adev->gfx.gfx_ring[1]; + r = amdgpu_ring_alloc(ring, 2); + if (r) { + DRM_ERROR("amdgpu: cp failed to lock ring (%d).\n", r); + return r; + } - amdgpu_ring_commit(ring); + amdgpu_ring_write(ring, PACKET3(PACKET3_CLEAR_STATE, 0)); + amdgpu_ring_write(ring, 0); + amdgpu_ring_commit(ring); + } return 0; } @@ -2818,39 +2870,41 @@ static int gfx_v10_0_cp_gfx_resume(struct amdgpu_device *adev) mutex_unlock(&adev->srbm_mutex); /* Init gfx ring 1 for pipe 1 */ - mutex_lock(&adev->srbm_mutex); - gfx_v10_0_cp_gfx_switch_pipe(adev, PIPE_ID1); - ring = &adev->gfx.gfx_ring[1]; - rb_bufsz = order_base_2(ring->ring_size / 8); - tmp = REG_SET_FIELD(0, CP_RB1_CNTL, RB_BUFSZ, rb_bufsz); - tmp = REG_SET_FIELD(tmp, CP_RB1_CNTL, RB_BLKSZ, rb_bufsz - 2); - WREG32_SOC15(GC, 0, mmCP_RB1_CNTL, tmp); - /* Initialize the ring buffer's write pointers */ - ring->wptr = 0; - WREG32_SOC15(GC, 0, mmCP_RB1_WPTR, lower_32_bits(ring->wptr)); - WREG32_SOC15(GC, 0, mmCP_RB1_WPTR_HI, upper_32_bits(ring->wptr)); - /* Set the wb address wether it's enabled or not */ - rptr_addr = adev->wb.gpu_addr + (ring->rptr_offs * 4); - WREG32_SOC15(GC, 0, mmCP_RB1_RPTR_ADDR, lower_32_bits(rptr_addr)); - WREG32_SOC15(GC, 0, mmCP_RB1_RPTR_ADDR_HI, upper_32_bits(rptr_addr) & - CP_RB1_RPTR_ADDR_HI__RB_RPTR_ADDR_HI_MASK); - wptr_gpu_addr = adev->wb.gpu_addr + (ring->wptr_offs * 4); - WREG32_SOC15(GC, 0, mmCP_RB_WPTR_POLL_ADDR_LO, - lower_32_bits(wptr_gpu_addr)); - WREG32_SOC15(GC, 0, mmCP_RB_WPTR_POLL_ADDR_HI, - upper_32_bits(wptr_gpu_addr)); - - mdelay(1); - WREG32_SOC15(GC, 0, mmCP_RB1_CNTL, tmp); - - rb_addr = ring->gpu_addr >> 8; - WREG32_SOC15(GC, 0, mmCP_RB1_BASE, rb_addr); - WREG32_SOC15(GC, 0, mmCP_RB1_BASE_HI, upper_32_bits(rb_addr)); - WREG32_SOC15(GC, 0, mmCP_RB1_ACTIVE, 1); - - gfx_v10_0_cp_gfx_set_doorbell(adev, ring); - mutex_unlock(&adev->srbm_mutex); - + if (adev->gfx.num_gfx_rings > 1) { + mutex_lock(&adev->srbm_mutex); + gfx_v10_0_cp_gfx_switch_pipe(adev, PIPE_ID1); + /* maximum supported gfx ring is 2 */ + ring = &adev->gfx.gfx_ring[1]; + rb_bufsz = order_base_2(ring->ring_size / 8); + tmp = REG_SET_FIELD(0, CP_RB1_CNTL, RB_BUFSZ, rb_bufsz); + tmp = REG_SET_FIELD(tmp, CP_RB1_CNTL, RB_BLKSZ, rb_bufsz - 2); + WREG32_SOC15(GC, 0, mmCP_RB1_CNTL, tmp); + /* Initialize the ring buffer's write pointers */ + ring->wptr = 0; + WREG32_SOC15(GC, 0, mmCP_RB1_WPTR, lower_32_bits(ring->wptr)); + WREG32_SOC15(GC, 0, mmCP_RB1_WPTR_HI, upper_32_bits(ring->wptr)); + /* Set the wb address wether it's enabled or not */ + rptr_addr = adev->wb.gpu_addr + (ring->rptr_offs * 4); + WREG32_SOC15(GC, 0, mmCP_RB1_RPTR_ADDR, lower_32_bits(rptr_addr)); + WREG32_SOC15(GC, 0, mmCP_RB1_RPTR_ADDR_HI, upper_32_bits(rptr_addr) & + CP_RB1_RPTR_ADDR_HI__RB_RPTR_ADDR_HI_MASK); + wptr_gpu_addr = adev->wb.gpu_addr + (ring->wptr_offs * 4); + WREG32_SOC15(GC, 0, mmCP_RB_WPTR_POLL_ADDR_LO, + lower_32_bits(wptr_gpu_addr)); + WREG32_SOC15(GC, 0, mmCP_RB_WPTR_POLL_ADDR_HI, + upper_32_bits(wptr_gpu_addr)); + + mdelay(1); + WREG32_SOC15(GC, 0, mmCP_RB1_CNTL, tmp); + + rb_addr = ring->gpu_addr >> 8; + WREG32_SOC15(GC, 0, mmCP_RB1_BASE, rb_addr); + WREG32_SOC15(GC, 0, mmCP_RB1_BASE_HI, upper_32_bits(rb_addr)); + WREG32_SOC15(GC, 0, mmCP_RB1_ACTIVE, 1); + + gfx_v10_0_cp_gfx_set_doorbell(adev, ring); + mutex_unlock(&adev->srbm_mutex); + } /* Switch to pipe 0 */ mutex_lock(&adev->srbm_mutex); gfx_v10_0_cp_gfx_switch_pipe(adev, PIPE_ID0); @@ -3164,12 +3218,7 @@ static int gfx_v10_0_kiq_enable_kgq(struct amdgpu_device *adev) for (i = 0; i < adev->gfx.num_gfx_rings; i++) kiq->pmf->kiq_map_queues(kiq_ring, &adev->gfx.gfx_ring[i]); - r = amdgpu_ring_test_ring(kiq_ring); - if (r) { - DRM_ERROR("kfq enable failed\n"); - kiq_ring->sched.ready = false; - } - return r; + return amdgpu_ring_test_helper(kiq_ring); } #endif @@ -3212,6 +3261,22 @@ done: return r; } +static void gfx_v10_0_compute_mqd_set_priority(struct amdgpu_ring *ring, struct v10_compute_mqd *mqd) +{ + struct amdgpu_device *adev = ring->adev; + + if (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE) { + if (amdgpu_gfx_is_high_priority_compute_queue(adev, ring->queue)) { + mqd->cp_hqd_pipe_priority = AMDGPU_GFX_PIPE_PRIO_HIGH; + ring->has_high_prio = true; + mqd->cp_hqd_queue_priority = + AMDGPU_GFX_QUEUE_PRIORITY_MAXIMUM; + } else { + ring->has_high_prio = false; + } + } +} + static int gfx_v10_0_compute_mqd_init(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; @@ -3337,6 +3402,9 @@ static int gfx_v10_0_compute_mqd_init(struct amdgpu_ring *ring) tmp = REG_SET_FIELD(tmp, CP_HQD_IB_CONTROL, MIN_IB_AVAIL_SIZE, 3); mqd->cp_hqd_ib_control = tmp; + /* set static priority for a compute queue/ring */ + gfx_v10_0_compute_mqd_set_priority(ring, mqd); + /* map_queues packet doesn't need activate the queue, * so only kiq need set this field. */ @@ -3513,6 +3581,7 @@ static int gfx_v10_0_kcq_init_queue(struct amdgpu_ring *ring) /* reset ring buffer */ ring->wptr = 0; + atomic64_set((atomic64_t *)&adev->wb.wb[ring->wptr_offs], 0); amdgpu_ring_clear_ring(ring); } else { amdgpu_ring_clear_ring(ring); @@ -3785,7 +3854,7 @@ static int gfx_v10_0_kiq_disable_kgq(struct amdgpu_device *adev) kiq->pmf->kiq_unmap_queues(kiq_ring, &adev->gfx.gfx_ring[i], PREEMPT_QUEUES, 0, 0); - return amdgpu_ring_test_ring(kiq_ring); + return amdgpu_ring_test_helper(kiq_ring); } #endif @@ -3923,11 +3992,12 @@ static uint64_t gfx_v10_0_get_gpu_clock_counter(struct amdgpu_device *adev) { uint64_t clock; + amdgpu_gfx_off_ctrl(adev, false); mutex_lock(&adev->gfx.gpu_clock_mutex); - WREG32_SOC15(GC, 0, mmRLC_CAPTURE_GPU_CLOCK_COUNT, 1); - clock = (uint64_t)RREG32_SOC15(GC, 0, mmRLC_GPU_CLOCK_COUNT_LSB) | - ((uint64_t)RREG32_SOC15(GC, 0, mmRLC_GPU_CLOCK_COUNT_MSB) << 32ULL); + clock = (uint64_t)RREG32_SOC15(SMUIO, 0, mmGOLDEN_TSC_COUNT_LOWER) | + ((uint64_t)RREG32_SOC15(SMUIO, 0, mmGOLDEN_TSC_COUNT_UPPER) << 32ULL); mutex_unlock(&adev->gfx.gpu_clock_mutex); + amdgpu_gfx_off_ctrl(adev, true); return clock; } @@ -3964,7 +4034,8 @@ static int gfx_v10_0_early_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - adev->gfx.num_gfx_rings = GFX10_NUM_GFX_RINGS; + adev->gfx.num_gfx_rings = GFX10_NUM_GFX_RINGS_NV1X; + adev->gfx.num_compute_rings = AMDGPU_MAX_COMPUTE_RINGS; gfx_v10_0_set_kiq_pm4_funcs(adev); @@ -4033,6 +4104,12 @@ static void gfx_v10_0_update_medium_grain_clock_gating(struct amdgpu_device *ade /* It is disabled by HW by default */ if (enable && (adev->cg_flags & AMD_CG_SUPPORT_GFX_MGCG)) { + /* 0 - Disable some blocks' MGCG */ + WREG32_SOC15(GC, 0, mmGRBM_GFX_INDEX, 0xe0000000); + WREG32_SOC15(GC, 0, mmCGTT_WD_CLK_CTRL, 0xff000000); + WREG32_SOC15(GC, 0, mmCGTT_VGT_CLK_CTRL, 0xff000000); + WREG32_SOC15(GC, 0, mmCGTT_IA_CLK_CTRL, 0xff000000); + /* 1 - RLC_CGTT_MGCG_OVERRIDE */ def = data = RREG32_SOC15(GC, 0, mmRLC_CGTT_MGCG_OVERRIDE); data &= ~(RLC_CGTT_MGCG_OVERRIDE__GRBM_CGTT_SCLK_OVERRIDE_MASK | @@ -4072,19 +4149,20 @@ static void gfx_v10_0_update_medium_grain_clock_gating(struct amdgpu_device *ade if (def != data) WREG32_SOC15(GC, 0, mmRLC_CGTT_MGCG_OVERRIDE, data); - /* 2 - disable MGLS in RLC */ + /* 2 - disable MGLS in CP */ + data = RREG32_SOC15(GC, 0, mmCP_MEM_SLP_CNTL); + if (data & CP_MEM_SLP_CNTL__CP_MEM_LS_EN_MASK) { + data &= ~CP_MEM_SLP_CNTL__CP_MEM_LS_EN_MASK; + WREG32_SOC15(GC, 0, mmCP_MEM_SLP_CNTL, data); + } + + /* 3 - disable MGLS in RLC */ data = RREG32_SOC15(GC, 0, mmRLC_MEM_SLP_CNTL); if (data & RLC_MEM_SLP_CNTL__RLC_MEM_LS_EN_MASK) { data &= ~RLC_MEM_SLP_CNTL__RLC_MEM_LS_EN_MASK; WREG32_SOC15(GC, 0, mmRLC_MEM_SLP_CNTL, data); } - /* 3 - disable MGLS in CP */ - data = RREG32_SOC15(GC, 0, mmCP_MEM_SLP_CNTL); - if (data & CP_MEM_SLP_CNTL__CP_MEM_LS_EN_MASK) { - data &= ~CP_MEM_SLP_CNTL__CP_MEM_LS_EN_MASK; - WREG32_SOC15(GC, 0, mmCP_MEM_SLP_CNTL, data); - } } } @@ -4195,7 +4273,7 @@ static int gfx_v10_0_update_gfx_clock_gating(struct amdgpu_device *adev, /* === CGCG /CGLS for GFX 3D Only === */ gfx_v10_0_update_3d_clock_gating(adev, enable); /* === MGCG + MGLS === */ - gfx_v10_0_update_medium_grain_clock_gating(adev, enable); + /* gfx_v10_0_update_medium_grain_clock_gating(adev, enable); */ } if (adev->cg_flags & @@ -4212,6 +4290,45 @@ static int gfx_v10_0_update_gfx_clock_gating(struct amdgpu_device *adev, return 0; } +static void gfx_v10_0_update_spm_vmid(struct amdgpu_device *adev, unsigned vmid) +{ + u32 data; + + data = RREG32_SOC15(GC, 0, mmRLC_SPM_MC_CNTL); + + data &= ~RLC_SPM_MC_CNTL__RLC_SPM_VMID_MASK; + data |= (vmid & RLC_SPM_MC_CNTL__RLC_SPM_VMID_MASK) << RLC_SPM_MC_CNTL__RLC_SPM_VMID__SHIFT; + + WREG32_SOC15(GC, 0, mmRLC_SPM_MC_CNTL, data); +} + +static bool gfx_v10_0_check_rlcg_range(struct amdgpu_device *adev, + uint32_t offset, + struct soc15_reg_rlcg *entries, int arr_size) +{ + int i; + uint32_t reg; + + if (!entries) + return false; + + for (i = 0; i < arr_size; i++) { + const struct soc15_reg_rlcg *entry; + + entry = &entries[i]; + reg = adev->reg_offset[entry->hwip][entry->instance][entry->segment] + entry->reg; + if (offset == reg) + return true; + } + + return false; +} + +static bool gfx_v10_0_is_rlcg_access_range(struct amdgpu_device *adev, u32 offset) +{ + return gfx_v10_0_check_rlcg_range(adev, offset, NULL, 0); +} + static const struct amdgpu_rlc_funcs gfx_v10_0_rlc_funcs = { .is_rlc_enabled = gfx_v10_0_is_rlc_enabled, .set_safe_mode = gfx_v10_0_set_safe_mode, @@ -4222,7 +4339,10 @@ static const struct amdgpu_rlc_funcs gfx_v10_0_rlc_funcs = { .resume = gfx_v10_0_rlc_resume, .stop = gfx_v10_0_rlc_stop, .reset = gfx_v10_0_rlc_reset, - .start = gfx_v10_0_rlc_start + .start = gfx_v10_0_rlc_start, + .update_spm_vmid = gfx_v10_0_update_spm_vmid, + .rlcg_wreg = gfx_v10_rlcg_wreg, + .is_rlcg_access_range = gfx_v10_0_is_rlcg_access_range, }; static int gfx_v10_0_set_powergating_state(void *handle, @@ -4411,15 +4531,15 @@ static void gfx_v10_0_ring_emit_ib_gfx(struct amdgpu_ring *ring, control |= ib->length_dw | (vmid << 24); - if (amdgpu_mcbp && (ib->flags & AMDGPU_IB_FLAG_PREEMPT)) { + if ((amdgpu_sriov_vf(ring->adev) || amdgpu_mcbp) && (ib->flags & AMDGPU_IB_FLAG_PREEMPT)) { control |= INDIRECT_BUFFER_PRE_ENB(1); if (flags & AMDGPU_IB_PREEMPTED) control |= INDIRECT_BUFFER_PRE_RESUME(1); - if (!(ib->flags & AMDGPU_IB_FLAG_CE)) + if (!(ib->flags & AMDGPU_IB_FLAG_CE) && vmid) gfx_v10_0_ring_emit_de_meta(ring, - flags & AMDGPU_IB_PREEMPTED ? true : false); + (!amdgpu_sriov_vf(ring->adev) && flags & AMDGPU_IB_PREEMPTED) ? true : false); } amdgpu_ring_write(ring, header); @@ -4566,9 +4686,9 @@ static void gfx_v10_0_ring_emit_cntxcntl(struct amdgpu_ring *ring, uint32_t flag { uint32_t dw2 = 0; - if (amdgpu_mcbp) + if (amdgpu_mcbp || amdgpu_sriov_vf(ring->adev)) gfx_v10_0_ring_emit_ce_meta(ring, - flags & AMDGPU_IB_PREEMPTED ? true : false); + (!amdgpu_sriov_vf(ring->adev) && flags & AMDGPU_IB_PREEMPTED) ? true : false); gfx_v10_0_ring_emit_tmz(ring, true); diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c index 8f20a5dd44fe..733d398c61cc 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c @@ -3346,6 +3346,10 @@ static int gfx_v7_0_rlc_init(struct amdgpu_device *adev) return r; } + /* init spm vmid with 0xf */ + if (adev->gfx.rlc.funcs->update_spm_vmid) + adev->gfx.rlc.funcs->update_spm_vmid(adev, 0xf); + return 0; } @@ -3570,6 +3574,18 @@ static int gfx_v7_0_rlc_resume(struct amdgpu_device *adev) return 0; } +static void gfx_v7_0_update_spm_vmid(struct amdgpu_device *adev, unsigned vmid) +{ + u32 data; + + data = RREG32(mmRLC_SPM_VMID); + + data &= ~RLC_SPM_VMID__RLC_SPM_VMID_MASK; + data |= (vmid & RLC_SPM_VMID__RLC_SPM_VMID_MASK) << RLC_SPM_VMID__RLC_SPM_VMID__SHIFT; + + WREG32(mmRLC_SPM_VMID, data); +} + static void gfx_v7_0_enable_cgcg(struct amdgpu_device *adev, bool enable) { u32 data, orig, tmp, tmp2; @@ -4221,7 +4237,8 @@ static const struct amdgpu_rlc_funcs gfx_v7_0_rlc_funcs = { .resume = gfx_v7_0_rlc_resume, .stop = gfx_v7_0_rlc_stop, .reset = gfx_v7_0_rlc_reset, - .start = gfx_v7_0_rlc_start + .start = gfx_v7_0_rlc_start, + .update_spm_vmid = gfx_v7_0_update_spm_vmid }; static int gfx_v7_0_early_init(void *handle) @@ -4338,6 +4355,11 @@ static void gfx_v7_0_gpu_early_init(struct amdgpu_device *adev) adev->gfx.config.mc_arb_ramcfg = RREG32(mmMC_ARB_RAMCFG); mc_arb_ramcfg = adev->gfx.config.mc_arb_ramcfg; + adev->gfx.config.num_banks = REG_GET_FIELD(mc_arb_ramcfg, + MC_ARB_RAMCFG, NOOFBANK); + adev->gfx.config.num_ranks = REG_GET_FIELD(mc_arb_ramcfg, + MC_ARB_RAMCFG, NOOFRANKS); + adev->gfx.config.num_tile_pipes = adev->gfx.config.max_tile_pipes; adev->gfx.config.mem_max_burst_length_bytes = 256; if (adev->flags & AMD_IS_APU) { diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c index fa245973de12..fc32586ef80b 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c @@ -1318,6 +1318,10 @@ static int gfx_v8_0_rlc_init(struct amdgpu_device *adev) return r; } + /* init spm vmid with 0xf */ + if (adev->gfx.rlc.funcs->update_spm_vmid) + adev->gfx.rlc.funcs->update_spm_vmid(adev, 0xf); + return 0; } @@ -1820,6 +1824,11 @@ static int gfx_v8_0_gpu_early_init(struct amdgpu_device *adev) adev->gfx.config.mc_arb_ramcfg = RREG32(mmMC_ARB_RAMCFG); mc_arb_ramcfg = adev->gfx.config.mc_arb_ramcfg; + adev->gfx.config.num_banks = REG_GET_FIELD(mc_arb_ramcfg, + MC_ARB_RAMCFG, NOOFBANK); + adev->gfx.config.num_ranks = REG_GET_FIELD(mc_arb_ramcfg, + MC_ARB_RAMCFG, NOOFRANKS); + adev->gfx.config.num_tile_pipes = adev->gfx.config.max_tile_pipes; adev->gfx.config.mem_max_burst_length_bytes = 256; if (adev->flags & AMD_IS_APU) { @@ -4421,6 +4430,22 @@ static int gfx_v8_0_deactivate_hqd(struct amdgpu_device *adev, u32 req) return r; } +static void gfx_v8_0_mqd_set_priority(struct amdgpu_ring *ring, struct vi_mqd *mqd) +{ + struct amdgpu_device *adev = ring->adev; + + if (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE) { + if (amdgpu_gfx_is_high_priority_compute_queue(adev, ring->queue)) { + mqd->cp_hqd_pipe_priority = AMDGPU_GFX_PIPE_PRIO_HIGH; + ring->has_high_prio = true; + mqd->cp_hqd_queue_priority = + AMDGPU_GFX_QUEUE_PRIORITY_MAXIMUM; + } else { + ring->has_high_prio = false; + } + } +} + static int gfx_v8_0_mqd_init(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; @@ -4544,9 +4569,6 @@ static int gfx_v8_0_mqd_init(struct amdgpu_ring *ring) /* defaults */ mqd->cp_hqd_eop_rptr = RREG32(mmCP_HQD_EOP_RPTR); mqd->cp_hqd_eop_wptr = RREG32(mmCP_HQD_EOP_WPTR); - mqd->cp_hqd_pipe_priority = RREG32(mmCP_HQD_PIPE_PRIORITY); - mqd->cp_hqd_queue_priority = RREG32(mmCP_HQD_QUEUE_PRIORITY); - mqd->cp_hqd_quantum = RREG32(mmCP_HQD_QUANTUM); mqd->cp_hqd_ctx_save_base_addr_lo = RREG32(mmCP_HQD_CTX_SAVE_BASE_ADDR_LO); mqd->cp_hqd_ctx_save_base_addr_hi = RREG32(mmCP_HQD_CTX_SAVE_BASE_ADDR_HI); mqd->cp_hqd_cntl_stack_offset = RREG32(mmCP_HQD_CNTL_STACK_OFFSET); @@ -4558,6 +4580,10 @@ static int gfx_v8_0_mqd_init(struct amdgpu_ring *ring) mqd->cp_hqd_eop_wptr_mem = RREG32(mmCP_HQD_EOP_WPTR_MEM); mqd->cp_hqd_eop_dones = RREG32(mmCP_HQD_EOP_DONES); + /* set static priority for a queue/ring */ + gfx_v8_0_mqd_set_priority(ring, mqd); + mqd->cp_hqd_quantum = RREG32(mmCP_HQD_QUANTUM); + /* map_queues packet doesn't need activate the queue, * so only kiq need set this field. */ @@ -5589,6 +5615,18 @@ static void gfx_v8_0_unset_safe_mode(struct amdgpu_device *adev) } } +static void gfx_v8_0_update_spm_vmid(struct amdgpu_device *adev, unsigned vmid) +{ + u32 data; + + data = RREG32(mmRLC_SPM_VMID); + + data &= ~RLC_SPM_VMID__RLC_SPM_VMID_MASK; + data |= (vmid & RLC_SPM_VMID__RLC_SPM_VMID_MASK) << RLC_SPM_VMID__RLC_SPM_VMID__SHIFT; + + WREG32(mmRLC_SPM_VMID, data); +} + static const struct amdgpu_rlc_funcs iceland_rlc_funcs = { .is_rlc_enabled = gfx_v8_0_is_rlc_enabled, .set_safe_mode = gfx_v8_0_set_safe_mode, @@ -5600,7 +5638,8 @@ static const struct amdgpu_rlc_funcs iceland_rlc_funcs = { .resume = gfx_v8_0_rlc_resume, .stop = gfx_v8_0_rlc_stop, .reset = gfx_v8_0_rlc_reset, - .start = gfx_v8_0_rlc_start + .start = gfx_v8_0_rlc_start, + .update_spm_vmid = gfx_v8_0_update_spm_vmid }; static void gfx_v8_0_update_medium_grain_clock_gating(struct amdgpu_device *adev, @@ -6094,7 +6133,7 @@ static void gfx_v8_0_ring_emit_ib_gfx(struct amdgpu_ring *ring, if (amdgpu_sriov_vf(ring->adev) && (ib->flags & AMDGPU_IB_FLAG_PREEMPT)) { control |= INDIRECT_BUFFER_PRE_ENB(1); - if (!(ib->flags & AMDGPU_IB_FLAG_CE)) + if (!(ib->flags & AMDGPU_IB_FLAG_CE) && vmid) gfx_v8_0_ring_emit_de_meta(ring); } @@ -6236,104 +6275,6 @@ static void gfx_v8_0_ring_set_wptr_compute(struct amdgpu_ring *ring) WDOORBELL32(ring->doorbell_index, lower_32_bits(ring->wptr)); } -static void gfx_v8_0_ring_set_pipe_percent(struct amdgpu_ring *ring, - bool acquire) -{ - struct amdgpu_device *adev = ring->adev; - int pipe_num, tmp, reg; - int pipe_percent = acquire ? SPI_WCL_PIPE_PERCENT_GFX__VALUE_MASK : 0x1; - - pipe_num = ring->me * adev->gfx.mec.num_pipe_per_mec + ring->pipe; - - /* first me only has 2 entries, GFX and HP3D */ - if (ring->me > 0) - pipe_num -= 2; - - reg = mmSPI_WCL_PIPE_PERCENT_GFX + pipe_num; - tmp = RREG32(reg); - tmp = REG_SET_FIELD(tmp, SPI_WCL_PIPE_PERCENT_GFX, VALUE, pipe_percent); - WREG32(reg, tmp); -} - -static void gfx_v8_0_pipe_reserve_resources(struct amdgpu_device *adev, - struct amdgpu_ring *ring, - bool acquire) -{ - int i, pipe; - bool reserve; - struct amdgpu_ring *iring; - - mutex_lock(&adev->gfx.pipe_reserve_mutex); - pipe = amdgpu_gfx_mec_queue_to_bit(adev, ring->me, ring->pipe, 0); - if (acquire) - set_bit(pipe, adev->gfx.pipe_reserve_bitmap); - else - clear_bit(pipe, adev->gfx.pipe_reserve_bitmap); - - if (!bitmap_weight(adev->gfx.pipe_reserve_bitmap, AMDGPU_MAX_COMPUTE_QUEUES)) { - /* Clear all reservations - everyone reacquires all resources */ - for (i = 0; i < adev->gfx.num_gfx_rings; ++i) - gfx_v8_0_ring_set_pipe_percent(&adev->gfx.gfx_ring[i], - true); - - for (i = 0; i < adev->gfx.num_compute_rings; ++i) - gfx_v8_0_ring_set_pipe_percent(&adev->gfx.compute_ring[i], - true); - } else { - /* Lower all pipes without a current reservation */ - for (i = 0; i < adev->gfx.num_gfx_rings; ++i) { - iring = &adev->gfx.gfx_ring[i]; - pipe = amdgpu_gfx_mec_queue_to_bit(adev, - iring->me, - iring->pipe, - 0); - reserve = test_bit(pipe, adev->gfx.pipe_reserve_bitmap); - gfx_v8_0_ring_set_pipe_percent(iring, reserve); - } - - for (i = 0; i < adev->gfx.num_compute_rings; ++i) { - iring = &adev->gfx.compute_ring[i]; - pipe = amdgpu_gfx_mec_queue_to_bit(adev, - iring->me, - iring->pipe, - 0); - reserve = test_bit(pipe, adev->gfx.pipe_reserve_bitmap); - gfx_v8_0_ring_set_pipe_percent(iring, reserve); - } - } - - mutex_unlock(&adev->gfx.pipe_reserve_mutex); -} - -static void gfx_v8_0_hqd_set_priority(struct amdgpu_device *adev, - struct amdgpu_ring *ring, - bool acquire) -{ - uint32_t pipe_priority = acquire ? 0x2 : 0x0; - uint32_t queue_priority = acquire ? 0xf : 0x0; - - mutex_lock(&adev->srbm_mutex); - vi_srbm_select(adev, ring->me, ring->pipe, ring->queue, 0); - - WREG32(mmCP_HQD_PIPE_PRIORITY, pipe_priority); - WREG32(mmCP_HQD_QUEUE_PRIORITY, queue_priority); - - vi_srbm_select(adev, 0, 0, 0, 0); - mutex_unlock(&adev->srbm_mutex); -} -static void gfx_v8_0_ring_set_priority_compute(struct amdgpu_ring *ring, - enum drm_sched_priority priority) -{ - struct amdgpu_device *adev = ring->adev; - bool acquire = priority == DRM_SCHED_PRIORITY_HIGH_HW; - - if (ring->funcs->type != AMDGPU_RING_TYPE_COMPUTE) - return; - - gfx_v8_0_hqd_set_priority(adev, ring, acquire); - gfx_v8_0_pipe_reserve_resources(adev, ring, acquire); -} - static void gfx_v8_0_ring_emit_fence_compute(struct amdgpu_ring *ring, u64 addr, u64 seq, unsigned flags) @@ -6966,7 +6907,6 @@ static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_compute = { .test_ib = gfx_v8_0_ring_test_ib, .insert_nop = amdgpu_ring_insert_nop, .pad_ib = amdgpu_ring_generic_pad_ib, - .set_priority = gfx_v8_0_ring_set_priority_compute, .emit_wreg = gfx_v8_0_ring_emit_wreg, }; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c index b33a4eb39193..e6b113ed2f40 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c @@ -697,6 +697,11 @@ static const struct soc15_reg_golden golden_settings_gc_9_4_1_arct[] = SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQ_FIFO_SIZES, 0xffffffff, 0x00000f00), }; +static const struct soc15_reg_rlcg rlcg_access_gc_9_0[] = { + {SOC15_REG_ENTRY(GC, 0, mmGRBM_GFX_INDEX)}, + {SOC15_REG_ENTRY(GC, 0, mmSQ_IND_INDEX)}, +}; + static const u32 GFX_RLC_SRM_INDEX_CNTL_ADDR_OFFSETS[] = { mmRLC_SRM_INDEX_CNTL_ADDR_0 - mmRLC_SRM_INDEX_CNTL_ADDR_0, @@ -721,6 +726,59 @@ static const u32 GFX_RLC_SRM_INDEX_CNTL_DATA_OFFSETS[] = mmRLC_SRM_INDEX_CNTL_DATA_7 - mmRLC_SRM_INDEX_CNTL_DATA_0, }; +void gfx_v9_0_rlcg_wreg(struct amdgpu_device *adev, u32 offset, u32 v) +{ + static void *scratch_reg0; + static void *scratch_reg1; + static void *scratch_reg2; + static void *scratch_reg3; + static void *spare_int; + static uint32_t grbm_cntl; + static uint32_t grbm_idx; + + scratch_reg0 = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG0_BASE_IDX] + mmSCRATCH_REG0)*4; + scratch_reg1 = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG1_BASE_IDX] + mmSCRATCH_REG1)*4; + scratch_reg2 = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG1_BASE_IDX] + mmSCRATCH_REG2)*4; + scratch_reg3 = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG1_BASE_IDX] + mmSCRATCH_REG3)*4; + spare_int = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmRLC_SPARE_INT_BASE_IDX] + mmRLC_SPARE_INT)*4; + + grbm_cntl = adev->reg_offset[GC_HWIP][0][mmGRBM_GFX_CNTL_BASE_IDX] + mmGRBM_GFX_CNTL; + grbm_idx = adev->reg_offset[GC_HWIP][0][mmGRBM_GFX_INDEX_BASE_IDX] + mmGRBM_GFX_INDEX; + + if (amdgpu_sriov_runtime(adev)) { + pr_err("shouldn't call rlcg write register during runtime\n"); + return; + } + + if (offset == grbm_cntl || offset == grbm_idx) { + if (offset == grbm_cntl) + writel(v, scratch_reg2); + else if (offset == grbm_idx) + writel(v, scratch_reg3); + + writel(v, ((void __iomem *)adev->rmmio) + (offset * 4)); + } else { + uint32_t i = 0; + uint32_t retries = 50000; + + writel(v, scratch_reg0); + writel(offset | 0x80000000, scratch_reg1); + writel(1, spare_int); + for (i = 0; i < retries; i++) { + u32 tmp; + + tmp = readl(scratch_reg1); + if (!(tmp & 0x80000000)) + break; + + udelay(10); + } + if (i >= retries) + pr_err("timeout: rlcg program reg:0x%05x failed !\n", offset); + } + +} + #define VEGA10_GB_ADDR_CONFIG_GOLDEN 0x2a114042 #define VEGA12_GB_ADDR_CONFIG_GOLDEN 0x24104041 #define RAVEN_GB_ADDR_CONFIG_GOLDEN 0x24000042 @@ -738,9 +796,9 @@ static void gfx_v9_0_ring_emit_de_meta(struct amdgpu_ring *ring); static u64 gfx_v9_0_ring_get_rptr_compute(struct amdgpu_ring *ring); static int gfx_v9_0_query_ras_error_count(struct amdgpu_device *adev, void *ras_error_status); -static void gfx_v9_0_clear_ras_edc_counter(struct amdgpu_device *adev); static int gfx_v9_0_ras_error_inject(struct amdgpu_device *adev, void *inject_if); +static void gfx_v9_0_reset_ras_error_count(struct amdgpu_device *adev); static void gfx_v9_0_kiq_set_resources(struct amdgpu_ring *kiq_ring, uint64_t queue_mask) @@ -1106,10 +1164,11 @@ static void gfx_v9_0_check_fw_write_wait(struct amdgpu_device *adev) adev->gfx.me_fw_write_wait = false; adev->gfx.mec_fw_write_wait = false; - if ((adev->gfx.mec_fw_version < 0x000001a5) || + if ((adev->asic_type != CHIP_ARCTURUS) && + ((adev->gfx.mec_fw_version < 0x000001a5) || (adev->gfx.mec_feature_version < 46) || (adev->gfx.pfp_fw_version < 0x000000b7) || - (adev->gfx.pfp_feature_version < 46)) + (adev->gfx.pfp_feature_version < 46))) DRM_WARN_ONCE("CP firmware version too old, please update!"); switch (adev->asic_type) { @@ -1158,6 +1217,8 @@ static void gfx_v9_0_check_fw_write_wait(struct amdgpu_device *adev) adev->gfx.mec_fw_write_wait = true; break; default: + adev->gfx.me_fw_write_wait = true; + adev->gfx.mec_fw_write_wait = true; break; } } @@ -1193,6 +1254,14 @@ static bool gfx_v9_0_should_disable_gfxoff(struct pci_dev *pdev) return false; } +static bool is_raven_kicker(struct amdgpu_device *adev) +{ + if (adev->pm.fw_version >= 0x41e2b) + return true; + else + return false; +} + static void gfx_v9_0_check_if_need_gfxoff(struct amdgpu_device *adev) { if (gfx_v9_0_should_disable_gfxoff(adev->pdev)) @@ -1205,9 +1274,8 @@ static void gfx_v9_0_check_if_need_gfxoff(struct amdgpu_device *adev) break; case CHIP_RAVEN: if (!(adev->rev_id >= 0x8 || adev->pdev->device == 0x15d8) && - ((adev->gfx.rlc_fw_version != 106 && + ((!is_raven_kicker(adev) && adev->gfx.rlc_fw_version < 531) || - (adev->gfx.rlc_fw_version == 53815) || (adev->gfx.rlc_feature_version < 1) || !adev->gfx.rlc.is_rlc_v2_1)) adev->pm.pp_feature &= ~PP_GFXOFF_MASK; @@ -1839,6 +1907,10 @@ static int gfx_v9_0_rlc_init(struct amdgpu_device *adev) break; } + /* init spm vmid with 0xf */ + if (adev->gfx.rlc.funcs->update_spm_vmid) + adev->gfx.rlc.funcs->update_spm_vmid(adev, 0xf); + return 0; } @@ -1876,7 +1948,7 @@ static int gfx_v9_0_mec_init(struct amdgpu_device *adev) return r; } - memset(hpd, 0, adev->gfx.mec.hpd_eop_obj->tbo.mem.size); + memset(hpd, 0, mec_hpd_size); amdgpu_bo_kunmap(adev->gfx.mec.hpd_eop_obj); amdgpu_bo_unreserve(adev->gfx.mec.hpd_eop_obj); @@ -1909,7 +1981,7 @@ static int gfx_v9_0_mec_init(struct amdgpu_device *adev) static uint32_t wave_read_ind(struct amdgpu_device *adev, uint32_t simd, uint32_t wave, uint32_t address) { - WREG32_SOC15(GC, 0, mmSQ_IND_INDEX, + WREG32_SOC15_RLC(GC, 0, mmSQ_IND_INDEX, (wave << SQ_IND_INDEX__WAVE_ID__SHIFT) | (simd << SQ_IND_INDEX__SIMD_ID__SHIFT) | (address << SQ_IND_INDEX__INDEX__SHIFT) | @@ -1921,7 +1993,7 @@ static void wave_read_regs(struct amdgpu_device *adev, uint32_t simd, uint32_t wave, uint32_t thread, uint32_t regno, uint32_t num, uint32_t *out) { - WREG32_SOC15(GC, 0, mmSQ_IND_INDEX, + WREG32_SOC15_RLC(GC, 0, mmSQ_IND_INDEX, (wave << SQ_IND_INDEX__WAVE_ID__SHIFT) | (simd << SQ_IND_INDEX__SIMD_ID__SHIFT) | (regno << SQ_IND_INDEX__INDEX__SHIFT) | @@ -1985,7 +2057,8 @@ static const struct amdgpu_gfx_funcs gfx_v9_0_gfx_funcs = { .read_wave_vgprs = &gfx_v9_0_read_wave_vgprs, .select_me_pipe_q = &gfx_v9_0_select_me_pipe_q, .ras_error_inject = &gfx_v9_0_ras_error_inject, - .query_ras_error_count = &gfx_v9_0_query_ras_error_count + .query_ras_error_count = &gfx_v9_0_query_ras_error_count, + .reset_ras_error_count = &gfx_v9_0_reset_ras_error_count, }; static const struct amdgpu_gfx_funcs gfx_v9_4_gfx_funcs = { @@ -1996,7 +2069,8 @@ static const struct amdgpu_gfx_funcs gfx_v9_4_gfx_funcs = { .read_wave_vgprs = &gfx_v9_0_read_wave_vgprs, .select_me_pipe_q = &gfx_v9_0_select_me_pipe_q, .ras_error_inject = &gfx_v9_4_ras_error_inject, - .query_ras_error_count = &gfx_v9_4_query_ras_error_count + .query_ras_error_count = &gfx_v9_4_query_ras_error_count, + .reset_ras_error_count = &gfx_v9_4_reset_ras_error_count, }; static int gfx_v9_0_gpu_early_init(struct amdgpu_device *adev) @@ -3302,6 +3376,22 @@ static void gfx_v9_0_kiq_setting(struct amdgpu_ring *ring) WREG32_SOC15_RLC(GC, 0, mmRLC_CP_SCHEDULERS, tmp); } +static void gfx_v9_0_mqd_set_priority(struct amdgpu_ring *ring, struct v9_mqd *mqd) +{ + struct amdgpu_device *adev = ring->adev; + + if (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE) { + if (amdgpu_gfx_is_high_priority_compute_queue(adev, ring->queue)) { + mqd->cp_hqd_pipe_priority = AMDGPU_GFX_PIPE_PRIO_HIGH; + ring->has_high_prio = true; + mqd->cp_hqd_queue_priority = + AMDGPU_GFX_QUEUE_PRIORITY_MAXIMUM; + } else { + ring->has_high_prio = false; + } + } +} + static int gfx_v9_0_mqd_init(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; @@ -3438,6 +3528,10 @@ static int gfx_v9_0_mqd_init(struct amdgpu_ring *ring) tmp = REG_SET_FIELD(tmp, CP_HQD_IB_CONTROL, MIN_IB_AVAIL_SIZE, 3); mqd->cp_hqd_ib_control = tmp; + /* set static priority for a queue/ring */ + gfx_v9_0_mqd_set_priority(ring, mqd); + mqd->cp_hqd_quantum = RREG32(mmCP_HQD_QUANTUM); + /* map_queues packet doesn't need activate the queue, * so only kiq need set this field. */ @@ -3656,6 +3750,7 @@ static int gfx_v9_0_kcq_init_queue(struct amdgpu_ring *ring) /* reset ring buffer */ ring->wptr = 0; + atomic64_set((atomic64_t *)&adev->wb.wb[ring->wptr_offs], 0); amdgpu_ring_clear_ring(ring); } else { amdgpu_ring_clear_ring(ring); @@ -3955,28 +4050,78 @@ static int gfx_v9_0_soft_reset(void *handle) return 0; } +static uint64_t gfx_v9_0_kiq_read_clock(struct amdgpu_device *adev) +{ + signed long r, cnt = 0; + unsigned long flags; + uint32_t seq; + struct amdgpu_kiq *kiq = &adev->gfx.kiq; + struct amdgpu_ring *ring = &kiq->ring; + + BUG_ON(!ring->funcs->emit_rreg); + + spin_lock_irqsave(&kiq->ring_lock, flags); + amdgpu_ring_alloc(ring, 32); + amdgpu_ring_write(ring, PACKET3(PACKET3_COPY_DATA, 4)); + amdgpu_ring_write(ring, 9 | /* src: register*/ + (5 << 8) | /* dst: memory */ + (1 << 16) | /* count sel */ + (1 << 20)); /* write confirm */ + amdgpu_ring_write(ring, 0); + amdgpu_ring_write(ring, 0); + amdgpu_ring_write(ring, lower_32_bits(adev->wb.gpu_addr + + kiq->reg_val_offs * 4)); + amdgpu_ring_write(ring, upper_32_bits(adev->wb.gpu_addr + + kiq->reg_val_offs * 4)); + amdgpu_fence_emit_polling(ring, &seq); + amdgpu_ring_commit(ring); + spin_unlock_irqrestore(&kiq->ring_lock, flags); + + r = amdgpu_fence_wait_polling(ring, seq, MAX_KIQ_REG_WAIT); + + /* don't wait anymore for gpu reset case because this way may + * block gpu_recover() routine forever, e.g. this virt_kiq_rreg + * is triggered in TTM and ttm_bo_lock_delayed_workqueue() will + * never return if we keep waiting in virt_kiq_rreg, which cause + * gpu_recover() hang there. + * + * also don't wait anymore for IRQ context + * */ + if (r < 1 && (adev->in_gpu_reset || in_interrupt())) + goto failed_kiq_read; + + might_sleep(); + while (r < 1 && cnt++ < MAX_KIQ_REG_TRY) { + msleep(MAX_KIQ_REG_BAILOUT_INTERVAL); + r = amdgpu_fence_wait_polling(ring, seq, MAX_KIQ_REG_WAIT); + } + + if (cnt > MAX_KIQ_REG_TRY) + goto failed_kiq_read; + + return (uint64_t)adev->wb.wb[kiq->reg_val_offs] | + (uint64_t)adev->wb.wb[kiq->reg_val_offs + 1 ] << 32ULL; + +failed_kiq_read: + pr_err("failed to read gpu clock\n"); + return ~0; +} + static uint64_t gfx_v9_0_get_gpu_clock_counter(struct amdgpu_device *adev) { uint64_t clock; + amdgpu_gfx_off_ctrl(adev, false); mutex_lock(&adev->gfx.gpu_clock_mutex); if (adev->asic_type == CHIP_VEGA10 && amdgpu_sriov_runtime(adev)) { - uint32_t tmp, lsb, msb, i = 0; - do { - if (i != 0) - udelay(1); - tmp = RREG32_SOC15(GC, 0, mmRLC_REFCLOCK_TIMESTAMP_MSB); - lsb = RREG32_SOC15(GC, 0, mmRLC_REFCLOCK_TIMESTAMP_LSB); - msb = RREG32_SOC15(GC, 0, mmRLC_REFCLOCK_TIMESTAMP_MSB); - i++; - } while (unlikely(tmp != msb) && (i < adev->usec_timeout)); - clock = (uint64_t)lsb | ((uint64_t)msb << 32ULL); + clock = gfx_v9_0_kiq_read_clock(adev); } else { WREG32_SOC15(GC, 0, mmRLC_CAPTURE_GPU_CLOCK_COUNT, 1); clock = (uint64_t)RREG32_SOC15(GC, 0, mmRLC_GPU_CLOCK_COUNT_LSB) | ((uint64_t)RREG32_SOC15(GC, 0, mmRLC_GPU_CLOCK_COUNT_MSB) << 32ULL); } mutex_unlock(&adev->gfx.gpu_clock_mutex); + amdgpu_gfx_off_ctrl(adev, true); return clock; } @@ -4043,6 +4188,101 @@ static const u32 sgpr_init_compute_shader[] = 0xbe800080, 0xbf810000, }; +static const u32 vgpr_init_compute_shader_arcturus[] = { + 0xd3d94000, 0x18000080, 0xd3d94001, 0x18000080, 0xd3d94002, 0x18000080, + 0xd3d94003, 0x18000080, 0xd3d94004, 0x18000080, 0xd3d94005, 0x18000080, + 0xd3d94006, 0x18000080, 0xd3d94007, 0x18000080, 0xd3d94008, 0x18000080, + 0xd3d94009, 0x18000080, 0xd3d9400a, 0x18000080, 0xd3d9400b, 0x18000080, + 0xd3d9400c, 0x18000080, 0xd3d9400d, 0x18000080, 0xd3d9400e, 0x18000080, + 0xd3d9400f, 0x18000080, 0xd3d94010, 0x18000080, 0xd3d94011, 0x18000080, + 0xd3d94012, 0x18000080, 0xd3d94013, 0x18000080, 0xd3d94014, 0x18000080, + 0xd3d94015, 0x18000080, 0xd3d94016, 0x18000080, 0xd3d94017, 0x18000080, + 0xd3d94018, 0x18000080, 0xd3d94019, 0x18000080, 0xd3d9401a, 0x18000080, + 0xd3d9401b, 0x18000080, 0xd3d9401c, 0x18000080, 0xd3d9401d, 0x18000080, + 0xd3d9401e, 0x18000080, 0xd3d9401f, 0x18000080, 0xd3d94020, 0x18000080, + 0xd3d94021, 0x18000080, 0xd3d94022, 0x18000080, 0xd3d94023, 0x18000080, + 0xd3d94024, 0x18000080, 0xd3d94025, 0x18000080, 0xd3d94026, 0x18000080, + 0xd3d94027, 0x18000080, 0xd3d94028, 0x18000080, 0xd3d94029, 0x18000080, + 0xd3d9402a, 0x18000080, 0xd3d9402b, 0x18000080, 0xd3d9402c, 0x18000080, + 0xd3d9402d, 0x18000080, 0xd3d9402e, 0x18000080, 0xd3d9402f, 0x18000080, + 0xd3d94030, 0x18000080, 0xd3d94031, 0x18000080, 0xd3d94032, 0x18000080, + 0xd3d94033, 0x18000080, 0xd3d94034, 0x18000080, 0xd3d94035, 0x18000080, + 0xd3d94036, 0x18000080, 0xd3d94037, 0x18000080, 0xd3d94038, 0x18000080, + 0xd3d94039, 0x18000080, 0xd3d9403a, 0x18000080, 0xd3d9403b, 0x18000080, + 0xd3d9403c, 0x18000080, 0xd3d9403d, 0x18000080, 0xd3d9403e, 0x18000080, + 0xd3d9403f, 0x18000080, 0xd3d94040, 0x18000080, 0xd3d94041, 0x18000080, + 0xd3d94042, 0x18000080, 0xd3d94043, 0x18000080, 0xd3d94044, 0x18000080, + 0xd3d94045, 0x18000080, 0xd3d94046, 0x18000080, 0xd3d94047, 0x18000080, + 0xd3d94048, 0x18000080, 0xd3d94049, 0x18000080, 0xd3d9404a, 0x18000080, + 0xd3d9404b, 0x18000080, 0xd3d9404c, 0x18000080, 0xd3d9404d, 0x18000080, + 0xd3d9404e, 0x18000080, 0xd3d9404f, 0x18000080, 0xd3d94050, 0x18000080, + 0xd3d94051, 0x18000080, 0xd3d94052, 0x18000080, 0xd3d94053, 0x18000080, + 0xd3d94054, 0x18000080, 0xd3d94055, 0x18000080, 0xd3d94056, 0x18000080, + 0xd3d94057, 0x18000080, 0xd3d94058, 0x18000080, 0xd3d94059, 0x18000080, + 0xd3d9405a, 0x18000080, 0xd3d9405b, 0x18000080, 0xd3d9405c, 0x18000080, + 0xd3d9405d, 0x18000080, 0xd3d9405e, 0x18000080, 0xd3d9405f, 0x18000080, + 0xd3d94060, 0x18000080, 0xd3d94061, 0x18000080, 0xd3d94062, 0x18000080, + 0xd3d94063, 0x18000080, 0xd3d94064, 0x18000080, 0xd3d94065, 0x18000080, + 0xd3d94066, 0x18000080, 0xd3d94067, 0x18000080, 0xd3d94068, 0x18000080, + 0xd3d94069, 0x18000080, 0xd3d9406a, 0x18000080, 0xd3d9406b, 0x18000080, + 0xd3d9406c, 0x18000080, 0xd3d9406d, 0x18000080, 0xd3d9406e, 0x18000080, + 0xd3d9406f, 0x18000080, 0xd3d94070, 0x18000080, 0xd3d94071, 0x18000080, + 0xd3d94072, 0x18000080, 0xd3d94073, 0x18000080, 0xd3d94074, 0x18000080, + 0xd3d94075, 0x18000080, 0xd3d94076, 0x18000080, 0xd3d94077, 0x18000080, + 0xd3d94078, 0x18000080, 0xd3d94079, 0x18000080, 0xd3d9407a, 0x18000080, + 0xd3d9407b, 0x18000080, 0xd3d9407c, 0x18000080, 0xd3d9407d, 0x18000080, + 0xd3d9407e, 0x18000080, 0xd3d9407f, 0x18000080, 0xd3d94080, 0x18000080, + 0xd3d94081, 0x18000080, 0xd3d94082, 0x18000080, 0xd3d94083, 0x18000080, + 0xd3d94084, 0x18000080, 0xd3d94085, 0x18000080, 0xd3d94086, 0x18000080, + 0xd3d94087, 0x18000080, 0xd3d94088, 0x18000080, 0xd3d94089, 0x18000080, + 0xd3d9408a, 0x18000080, 0xd3d9408b, 0x18000080, 0xd3d9408c, 0x18000080, + 0xd3d9408d, 0x18000080, 0xd3d9408e, 0x18000080, 0xd3d9408f, 0x18000080, + 0xd3d94090, 0x18000080, 0xd3d94091, 0x18000080, 0xd3d94092, 0x18000080, + 0xd3d94093, 0x18000080, 0xd3d94094, 0x18000080, 0xd3d94095, 0x18000080, + 0xd3d94096, 0x18000080, 0xd3d94097, 0x18000080, 0xd3d94098, 0x18000080, + 0xd3d94099, 0x18000080, 0xd3d9409a, 0x18000080, 0xd3d9409b, 0x18000080, + 0xd3d9409c, 0x18000080, 0xd3d9409d, 0x18000080, 0xd3d9409e, 0x18000080, + 0xd3d9409f, 0x18000080, 0xd3d940a0, 0x18000080, 0xd3d940a1, 0x18000080, + 0xd3d940a2, 0x18000080, 0xd3d940a3, 0x18000080, 0xd3d940a4, 0x18000080, + 0xd3d940a5, 0x18000080, 0xd3d940a6, 0x18000080, 0xd3d940a7, 0x18000080, + 0xd3d940a8, 0x18000080, 0xd3d940a9, 0x18000080, 0xd3d940aa, 0x18000080, + 0xd3d940ab, 0x18000080, 0xd3d940ac, 0x18000080, 0xd3d940ad, 0x18000080, + 0xd3d940ae, 0x18000080, 0xd3d940af, 0x18000080, 0xd3d940b0, 0x18000080, + 0xd3d940b1, 0x18000080, 0xd3d940b2, 0x18000080, 0xd3d940b3, 0x18000080, + 0xd3d940b4, 0x18000080, 0xd3d940b5, 0x18000080, 0xd3d940b6, 0x18000080, + 0xd3d940b7, 0x18000080, 0xd3d940b8, 0x18000080, 0xd3d940b9, 0x18000080, + 0xd3d940ba, 0x18000080, 0xd3d940bb, 0x18000080, 0xd3d940bc, 0x18000080, + 0xd3d940bd, 0x18000080, 0xd3d940be, 0x18000080, 0xd3d940bf, 0x18000080, + 0xd3d940c0, 0x18000080, 0xd3d940c1, 0x18000080, 0xd3d940c2, 0x18000080, + 0xd3d940c3, 0x18000080, 0xd3d940c4, 0x18000080, 0xd3d940c5, 0x18000080, + 0xd3d940c6, 0x18000080, 0xd3d940c7, 0x18000080, 0xd3d940c8, 0x18000080, + 0xd3d940c9, 0x18000080, 0xd3d940ca, 0x18000080, 0xd3d940cb, 0x18000080, + 0xd3d940cc, 0x18000080, 0xd3d940cd, 0x18000080, 0xd3d940ce, 0x18000080, + 0xd3d940cf, 0x18000080, 0xd3d940d0, 0x18000080, 0xd3d940d1, 0x18000080, + 0xd3d940d2, 0x18000080, 0xd3d940d3, 0x18000080, 0xd3d940d4, 0x18000080, + 0xd3d940d5, 0x18000080, 0xd3d940d6, 0x18000080, 0xd3d940d7, 0x18000080, + 0xd3d940d8, 0x18000080, 0xd3d940d9, 0x18000080, 0xd3d940da, 0x18000080, + 0xd3d940db, 0x18000080, 0xd3d940dc, 0x18000080, 0xd3d940dd, 0x18000080, + 0xd3d940de, 0x18000080, 0xd3d940df, 0x18000080, 0xd3d940e0, 0x18000080, + 0xd3d940e1, 0x18000080, 0xd3d940e2, 0x18000080, 0xd3d940e3, 0x18000080, + 0xd3d940e4, 0x18000080, 0xd3d940e5, 0x18000080, 0xd3d940e6, 0x18000080, + 0xd3d940e7, 0x18000080, 0xd3d940e8, 0x18000080, 0xd3d940e9, 0x18000080, + 0xd3d940ea, 0x18000080, 0xd3d940eb, 0x18000080, 0xd3d940ec, 0x18000080, + 0xd3d940ed, 0x18000080, 0xd3d940ee, 0x18000080, 0xd3d940ef, 0x18000080, + 0xd3d940f0, 0x18000080, 0xd3d940f1, 0x18000080, 0xd3d940f2, 0x18000080, + 0xd3d940f3, 0x18000080, 0xd3d940f4, 0x18000080, 0xd3d940f5, 0x18000080, + 0xd3d940f6, 0x18000080, 0xd3d940f7, 0x18000080, 0xd3d940f8, 0x18000080, + 0xd3d940f9, 0x18000080, 0xd3d940fa, 0x18000080, 0xd3d940fb, 0x18000080, + 0xd3d940fc, 0x18000080, 0xd3d940fd, 0x18000080, 0xd3d940fe, 0x18000080, + 0xd3d940ff, 0x18000080, 0xb07c0000, 0xbe8a00ff, 0x000000f8, 0xbf11080a, + 0x7e000280, 0x7e020280, 0x7e040280, 0x7e060280, 0x7e080280, 0x7e0a0280, + 0x7e0c0280, 0x7e0e0280, 0x808a880a, 0xbe80320a, 0xbf84fff5, 0xbf9c0000, + 0xd28c0001, 0x0001007f, 0xd28d0001, 0x0002027e, 0x10020288, 0xb88b0904, + 0xb78b4000, 0xd1196a01, 0x00001701, 0xbe8a0087, 0xbefc00c1, 0xd89c4000, + 0x00020201, 0xd89cc080, 0x00040401, 0x320202ff, 0x00000800, 0x808a810a, + 0xbf84fff8, 0xbf810000, +}; + /* When below register arrays changed, please update gpr_reg_size, and sec_ded_counter_reg_size in function gfx_v9_0_do_edc_gpr_workarounds, to cover all gfx9 ASICs */ @@ -4063,6 +4303,23 @@ static const struct soc15_reg_entry vgpr_init_regs[] = { { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_STATIC_THREAD_MGMT_SE7), 0xffffffff }, }; +static const struct soc15_reg_entry vgpr_init_regs_arcturus[] = { + { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_RESOURCE_LIMITS), 0x0000000 }, + { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_NUM_THREAD_X), 0x40 }, + { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_NUM_THREAD_Y), 4 }, + { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_NUM_THREAD_Z), 1 }, + { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_PGM_RSRC1), 0xbf }, + { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_PGM_RSRC2), 0x400000 }, /* 64KB LDS */ + { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_STATIC_THREAD_MGMT_SE0), 0xffffffff }, + { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_STATIC_THREAD_MGMT_SE1), 0xffffffff }, + { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_STATIC_THREAD_MGMT_SE2), 0xffffffff }, + { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_STATIC_THREAD_MGMT_SE3), 0xffffffff }, + { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_STATIC_THREAD_MGMT_SE4), 0xffffffff }, + { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_STATIC_THREAD_MGMT_SE5), 0xffffffff }, + { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_STATIC_THREAD_MGMT_SE6), 0xffffffff }, + { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_STATIC_THREAD_MGMT_SE7), 0xffffffff }, +}; + static const struct soc15_reg_entry sgpr1_init_regs[] = { { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_RESOURCE_LIMITS), 0x0000000 }, { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_NUM_THREAD_X), 0x40 }, @@ -4131,7 +4388,6 @@ static const struct soc15_reg_entry gfx_v9_0_edc_counter_regs[] = { { SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT2), 0, 1, 16}, { SOC15_REG_ENTRY(GC, 0, mmTCA_EDC_CNT), 0, 1, 2}, { SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT3), 0, 4, 6}, - { SOC15_REG_ENTRY(HDP, 0, mmHDP_EDC_CNT), 0, 1, 1}, }; static int gfx_v9_0_do_edc_gds_workarounds(struct amdgpu_device *adev) @@ -4194,7 +4450,10 @@ static int gfx_v9_0_do_edc_gpr_workarounds(struct amdgpu_device *adev) adev->gfx.config.max_cu_per_sh * adev->gfx.config.max_sh_per_se; int sgpr_work_group_size = 5; - int gpr_reg_size = compute_dim_x / 16 + 6; + int gpr_reg_size = adev->gfx.config.max_shader_engines + 6; + int vgpr_init_shader_size; + const u32 *vgpr_init_shader_ptr; + const struct soc15_reg_entry *vgpr_init_regs_ptr; /* only support when RAS is enabled */ if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__GFX)) @@ -4204,6 +4463,16 @@ static int gfx_v9_0_do_edc_gpr_workarounds(struct amdgpu_device *adev) if (!ring->sched.ready) return 0; + if (adev->asic_type == CHIP_ARCTURUS) { + vgpr_init_shader_ptr = vgpr_init_compute_shader_arcturus; + vgpr_init_shader_size = sizeof(vgpr_init_compute_shader_arcturus); + vgpr_init_regs_ptr = vgpr_init_regs_arcturus; + } else { + vgpr_init_shader_ptr = vgpr_init_compute_shader; + vgpr_init_shader_size = sizeof(vgpr_init_compute_shader); + vgpr_init_regs_ptr = vgpr_init_regs; + } + total_size = (gpr_reg_size * 3 + 4 + 5 + 2) * 4; /* VGPRS */ total_size += @@ -4212,7 +4481,7 @@ static int gfx_v9_0_do_edc_gpr_workarounds(struct amdgpu_device *adev) (gpr_reg_size * 3 + 4 + 5 + 2) * 4; /* SGPRS2 */ total_size = ALIGN(total_size, 256); vgpr_offset = total_size; - total_size += ALIGN(sizeof(vgpr_init_compute_shader), 256); + total_size += ALIGN(vgpr_init_shader_size, 256); sgpr_offset = total_size; total_size += sizeof(sgpr_init_compute_shader); @@ -4225,8 +4494,8 @@ static int gfx_v9_0_do_edc_gpr_workarounds(struct amdgpu_device *adev) } /* load the compute shaders */ - for (i = 0; i < ARRAY_SIZE(vgpr_init_compute_shader); i++) - ib.ptr[i + (vgpr_offset / 4)] = vgpr_init_compute_shader[i]; + for (i = 0; i < vgpr_init_shader_size/sizeof(u32); i++) + ib.ptr[i + (vgpr_offset / 4)] = vgpr_init_shader_ptr[i]; for (i = 0; i < ARRAY_SIZE(sgpr_init_compute_shader); i++) ib.ptr[i + (sgpr_offset / 4)] = sgpr_init_compute_shader[i]; @@ -4238,9 +4507,9 @@ static int gfx_v9_0_do_edc_gpr_workarounds(struct amdgpu_device *adev) /* write the register state for the compute dispatch */ for (i = 0; i < gpr_reg_size; i++) { ib.ptr[ib.length_dw++] = PACKET3(PACKET3_SET_SH_REG, 1); - ib.ptr[ib.length_dw++] = SOC15_REG_ENTRY_OFFSET(vgpr_init_regs[i]) + ib.ptr[ib.length_dw++] = SOC15_REG_ENTRY_OFFSET(vgpr_init_regs_ptr[i]) - PACKET3_SET_SH_REG_START; - ib.ptr[ib.length_dw++] = vgpr_init_regs[i].reg_value; + ib.ptr[ib.length_dw++] = vgpr_init_regs_ptr[i].reg_value; } /* write the shader start address: mmCOMPUTE_PGM_LO, mmCOMPUTE_PGM_HI */ gpu_addr = (ib.gpu_addr + (u64)vgpr_offset) >> 8; @@ -4252,7 +4521,7 @@ static int gfx_v9_0_do_edc_gpr_workarounds(struct amdgpu_device *adev) /* write dispatch packet */ ib.ptr[ib.length_dw++] = PACKET3(PACKET3_DISPATCH_DIRECT, 3); - ib.ptr[ib.length_dw++] = compute_dim_x; /* x */ + ib.ptr[ib.length_dw++] = compute_dim_x * 2; /* x */ ib.ptr[ib.length_dw++] = 1; /* y */ ib.ptr[ib.length_dw++] = 1; /* z */ ib.ptr[ib.length_dw++] = @@ -4332,18 +4601,6 @@ static int gfx_v9_0_do_edc_gpr_workarounds(struct amdgpu_device *adev) goto fail; } - switch (adev->asic_type) - { - case CHIP_VEGA20: - gfx_v9_0_clear_ras_edc_counter(adev); - break; - case CHIP_ARCTURUS: - gfx_v9_4_clear_ras_edc_counter(adev); - break; - default: - break; - } - fail: amdgpu_ib_free(adev, &ib, NULL); dma_fence_put(f); @@ -4391,6 +4648,10 @@ static int gfx_v9_0_ecc_late_init(void *handle) if (r) return r; + if (adev->gfx.funcs && + adev->gfx.funcs->reset_ras_error_count) + adev->gfx.funcs->reset_ras_error_count(adev); + r = amdgpu_gfx_ras_late_init(adev); if (r) return r; @@ -4695,6 +4956,47 @@ static int gfx_v9_0_update_gfx_clock_gating(struct amdgpu_device *adev, return 0; } +static void gfx_v9_0_update_spm_vmid(struct amdgpu_device *adev, unsigned vmid) +{ + u32 data; + + data = RREG32_SOC15(GC, 0, mmRLC_SPM_MC_CNTL); + + data &= ~RLC_SPM_MC_CNTL__RLC_SPM_VMID_MASK; + data |= (vmid & RLC_SPM_MC_CNTL__RLC_SPM_VMID_MASK) << RLC_SPM_MC_CNTL__RLC_SPM_VMID__SHIFT; + + WREG32_SOC15(GC, 0, mmRLC_SPM_MC_CNTL, data); +} + +static bool gfx_v9_0_check_rlcg_range(struct amdgpu_device *adev, + uint32_t offset, + struct soc15_reg_rlcg *entries, int arr_size) +{ + int i; + uint32_t reg; + + if (!entries) + return false; + + for (i = 0; i < arr_size; i++) { + const struct soc15_reg_rlcg *entry; + + entry = &entries[i]; + reg = adev->reg_offset[entry->hwip][entry->instance][entry->segment] + entry->reg; + if (offset == reg) + return true; + } + + return false; +} + +static bool gfx_v9_0_is_rlcg_access_range(struct amdgpu_device *adev, u32 offset) +{ + return gfx_v9_0_check_rlcg_range(adev, offset, + (void *)rlcg_access_gc_9_0, + ARRAY_SIZE(rlcg_access_gc_9_0)); +} + static const struct amdgpu_rlc_funcs gfx_v9_0_rlc_funcs = { .is_rlc_enabled = gfx_v9_0_is_rlc_enabled, .set_safe_mode = gfx_v9_0_set_safe_mode, @@ -4706,7 +5008,10 @@ static const struct amdgpu_rlc_funcs gfx_v9_0_rlc_funcs = { .resume = gfx_v9_0_rlc_resume, .stop = gfx_v9_0_rlc_stop, .reset = gfx_v9_0_rlc_reset, - .start = gfx_v9_0_rlc_start + .start = gfx_v9_0_rlc_start, + .update_spm_vmid = gfx_v9_0_update_spm_vmid, + .rlcg_wreg = gfx_v9_0_rlcg_wreg, + .is_rlcg_access_range = gfx_v9_0_is_rlcg_access_range, }; static int gfx_v9_0_set_powergating_state(void *handle, @@ -4909,7 +5214,7 @@ static void gfx_v9_0_ring_emit_ib_gfx(struct amdgpu_ring *ring, if (amdgpu_sriov_vf(ring->adev) && (ib->flags & AMDGPU_IB_FLAG_PREEMPT)) { control |= INDIRECT_BUFFER_PRE_ENB(1); - if (!(ib->flags & AMDGPU_IB_FLAG_CE)) + if (!(ib->flags & AMDGPU_IB_FLAG_CE) && vmid) gfx_v9_0_ring_emit_de_meta(ring); } @@ -5034,105 +5339,6 @@ static u64 gfx_v9_0_ring_get_wptr_compute(struct amdgpu_ring *ring) return wptr; } -static void gfx_v9_0_ring_set_pipe_percent(struct amdgpu_ring *ring, - bool acquire) -{ - struct amdgpu_device *adev = ring->adev; - int pipe_num, tmp, reg; - int pipe_percent = acquire ? SPI_WCL_PIPE_PERCENT_GFX__VALUE_MASK : 0x1; - - pipe_num = ring->me * adev->gfx.mec.num_pipe_per_mec + ring->pipe; - - /* first me only has 2 entries, GFX and HP3D */ - if (ring->me > 0) - pipe_num -= 2; - - reg = SOC15_REG_OFFSET(GC, 0, mmSPI_WCL_PIPE_PERCENT_GFX) + pipe_num; - tmp = RREG32(reg); - tmp = REG_SET_FIELD(tmp, SPI_WCL_PIPE_PERCENT_GFX, VALUE, pipe_percent); - WREG32(reg, tmp); -} - -static void gfx_v9_0_pipe_reserve_resources(struct amdgpu_device *adev, - struct amdgpu_ring *ring, - bool acquire) -{ - int i, pipe; - bool reserve; - struct amdgpu_ring *iring; - - mutex_lock(&adev->gfx.pipe_reserve_mutex); - pipe = amdgpu_gfx_mec_queue_to_bit(adev, ring->me, ring->pipe, 0); - if (acquire) - set_bit(pipe, adev->gfx.pipe_reserve_bitmap); - else - clear_bit(pipe, adev->gfx.pipe_reserve_bitmap); - - if (!bitmap_weight(adev->gfx.pipe_reserve_bitmap, AMDGPU_MAX_COMPUTE_QUEUES)) { - /* Clear all reservations - everyone reacquires all resources */ - for (i = 0; i < adev->gfx.num_gfx_rings; ++i) - gfx_v9_0_ring_set_pipe_percent(&adev->gfx.gfx_ring[i], - true); - - for (i = 0; i < adev->gfx.num_compute_rings; ++i) - gfx_v9_0_ring_set_pipe_percent(&adev->gfx.compute_ring[i], - true); - } else { - /* Lower all pipes without a current reservation */ - for (i = 0; i < adev->gfx.num_gfx_rings; ++i) { - iring = &adev->gfx.gfx_ring[i]; - pipe = amdgpu_gfx_mec_queue_to_bit(adev, - iring->me, - iring->pipe, - 0); - reserve = test_bit(pipe, adev->gfx.pipe_reserve_bitmap); - gfx_v9_0_ring_set_pipe_percent(iring, reserve); - } - - for (i = 0; i < adev->gfx.num_compute_rings; ++i) { - iring = &adev->gfx.compute_ring[i]; - pipe = amdgpu_gfx_mec_queue_to_bit(adev, - iring->me, - iring->pipe, - 0); - reserve = test_bit(pipe, adev->gfx.pipe_reserve_bitmap); - gfx_v9_0_ring_set_pipe_percent(iring, reserve); - } - } - - mutex_unlock(&adev->gfx.pipe_reserve_mutex); -} - -static void gfx_v9_0_hqd_set_priority(struct amdgpu_device *adev, - struct amdgpu_ring *ring, - bool acquire) -{ - uint32_t pipe_priority = acquire ? 0x2 : 0x0; - uint32_t queue_priority = acquire ? 0xf : 0x0; - - mutex_lock(&adev->srbm_mutex); - soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0); - - WREG32_SOC15_RLC(GC, 0, mmCP_HQD_PIPE_PRIORITY, pipe_priority); - WREG32_SOC15_RLC(GC, 0, mmCP_HQD_QUEUE_PRIORITY, queue_priority); - - soc15_grbm_select(adev, 0, 0, 0, 0); - mutex_unlock(&adev->srbm_mutex); -} - -static void gfx_v9_0_ring_set_priority_compute(struct amdgpu_ring *ring, - enum drm_sched_priority priority) -{ - struct amdgpu_device *adev = ring->adev; - bool acquire = priority == DRM_SCHED_PRIORITY_HIGH_HW; - - if (ring->funcs->type != AMDGPU_RING_TYPE_COMPUTE) - return; - - gfx_v9_0_hqd_set_priority(adev, ring, acquire); - gfx_v9_0_pipe_reserve_resources(adev, ring, acquire); -} - static void gfx_v9_0_ring_set_wptr_compute(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; @@ -6312,10 +6518,13 @@ static int gfx_v9_0_ras_error_count(const struct soc15_reg_entry *reg, return 0; } -static void gfx_v9_0_clear_ras_edc_counter(struct amdgpu_device *adev) +static void gfx_v9_0_reset_ras_error_count(struct amdgpu_device *adev) { int i, j, k; + if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__GFX)) + return; + /* read back registers to clear the counters */ mutex_lock(&adev->grbm_idx_mutex); for (i = 0; i < ARRAY_SIZE(gfx_v9_0_edc_counter_regs); i++) { @@ -6503,7 +6712,6 @@ static const struct amdgpu_ring_funcs gfx_v9_0_ring_funcs_compute = { .test_ib = gfx_v9_0_ring_test_ib, .insert_nop = amdgpu_ring_insert_nop, .pad_ib = amdgpu_ring_generic_pad_ib, - .set_priority = gfx_v9_0_ring_set_priority_compute, .emit_wreg = gfx_v9_0_ring_emit_wreg, .emit_reg_wait = gfx_v9_0_ring_emit_reg_wait, .emit_reg_write_reg_wait = gfx_v9_0_ring_emit_reg_write_reg_wait, diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c index f099f13d7f1e..dce945ef21a5 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c @@ -710,14 +710,16 @@ static int gfx_v9_4_query_utc_edc_status(struct amdgpu_device *adev, sec_count = REG_GET_FIELD(data, VML2_MEM_ECC_CNTL, SEC_COUNT); if (sec_count) { - DRM_INFO("Instance[%d]: SubBlock %s, SEC %d\n", i, + dev_info(adev->dev, + "Instance[%d]: SubBlock %s, SEC %d\n", i, vml2_mems[i], sec_count); err_data->ce_count += sec_count; } ded_count = REG_GET_FIELD(data, VML2_MEM_ECC_CNTL, DED_COUNT); if (ded_count) { - DRM_INFO("Instance[%d]: SubBlock %s, DED %d\n", i, + dev_info(adev->dev, + "Instance[%d]: SubBlock %s, DED %d\n", i, vml2_mems[i], ded_count); err_data->ue_count += ded_count; } @@ -893,10 +895,13 @@ int gfx_v9_4_query_ras_error_count(struct amdgpu_device *adev, return 0; } -void gfx_v9_4_clear_ras_edc_counter(struct amdgpu_device *adev) +void gfx_v9_4_reset_ras_error_count(struct amdgpu_device *adev) { int i, j, k; + if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__GFX)) + return; + mutex_lock(&adev->grbm_idx_mutex); for (i = 0; i < ARRAY_SIZE(gfx_v9_4_edc_counter_regs); i++) { for (j = 0; j < gfx_v9_4_edc_counter_regs[i].se_num; j++) { diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.h b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.h index 2e3f6f755ad4..1ffecc5c0f0a 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.h +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.h @@ -32,4 +32,6 @@ int gfx_v9_4_query_ras_error_count(struct amdgpu_device *adev, int gfx_v9_4_ras_error_inject(struct amdgpu_device *adev, void *inject_if); +void gfx_v9_4_reset_ras_error_count(struct amdgpu_device *adev); + #endif /* __GFX_V9_4_H__ */ diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c index b70c7b483c24..cc866c367939 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c @@ -81,24 +81,31 @@ static void gfxhub_v2_0_init_system_aperture_regs(struct amdgpu_device *adev) { uint64_t value; - /* Disable AGP. */ - WREG32_SOC15(GC, 0, mmGCMC_VM_AGP_BASE, 0); - WREG32_SOC15(GC, 0, mmGCMC_VM_AGP_TOP, 0); - WREG32_SOC15(GC, 0, mmGCMC_VM_AGP_BOT, 0x00FFFFFF); - - /* Program the system aperture low logical page number. */ - WREG32_SOC15(GC, 0, mmGCMC_VM_SYSTEM_APERTURE_LOW_ADDR, - adev->gmc.vram_start >> 18); - WREG32_SOC15(GC, 0, mmGCMC_VM_SYSTEM_APERTURE_HIGH_ADDR, - adev->gmc.vram_end >> 18); - - /* Set default page address. */ - value = adev->vram_scratch.gpu_addr - adev->gmc.vram_start - + adev->vm_manager.vram_base_offset; - WREG32_SOC15(GC, 0, mmGCMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_LSB, - (u32)(value >> 12)); - WREG32_SOC15(GC, 0, mmGCMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB, - (u32)(value >> 44)); + if (!amdgpu_sriov_vf(adev)) { + /* + * the new L1 policy will block SRIOV guest from writing + * these regs, and they will be programed at host. + * so skip programing these regs. + */ + /* Disable AGP. */ + WREG32_SOC15(GC, 0, mmGCMC_VM_AGP_BASE, 0); + WREG32_SOC15(GC, 0, mmGCMC_VM_AGP_TOP, 0); + WREG32_SOC15(GC, 0, mmGCMC_VM_AGP_BOT, 0x00FFFFFF); + + /* Program the system aperture low logical page number. */ + WREG32_SOC15(GC, 0, mmGCMC_VM_SYSTEM_APERTURE_LOW_ADDR, + adev->gmc.vram_start >> 18); + WREG32_SOC15(GC, 0, mmGCMC_VM_SYSTEM_APERTURE_HIGH_ADDR, + adev->gmc.vram_end >> 18); + + /* Set default page address. */ + value = adev->vram_scratch.gpu_addr - adev->gmc.vram_start + + adev->vm_manager.vram_base_offset; + WREG32_SOC15(GC, 0, mmGCMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_LSB, + (u32)(value >> 12)); + WREG32_SOC15(GC, 0, mmGCMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB, + (u32)(value >> 44)); + } /* Program "protection fault". */ WREG32_SOC15(GC, 0, mmGCVM_L2_PROTECTION_FAULT_DEFAULT_ADDR_LO32, @@ -135,6 +142,10 @@ static void gfxhub_v2_0_init_cache_regs(struct amdgpu_device *adev) { uint32_t tmp; + /* These regs are not accessible for VF, PF will program these in SRIOV */ + if (amdgpu_sriov_vf(adev)) + return; + /* Setup L2 cache */ tmp = RREG32_SOC15(GC, 0, mmGCVM_L2_CNTL); tmp = REG_SET_FIELD(tmp, GCVM_L2_CNTL, ENABLE_L2_CACHE, 1); @@ -256,18 +267,6 @@ static void gfxhub_v2_0_program_invalidation(struct amdgpu_device *adev) int gfxhub_v2_0_gart_enable(struct amdgpu_device *adev) { - if (amdgpu_sriov_vf(adev)) { - /* - * GCMC_VM_FB_LOCATION_BASE/TOP is NULL for VF, becuase they are - * VF copy registers so vbios post doesn't program them, for - * SRIOV driver need to program them - */ - WREG32_SOC15(GC, 0, mmGCMC_VM_FB_LOCATION_BASE, - adev->gmc.vram_start >> 24); - WREG32_SOC15(GC, 0, mmGCMC_VM_FB_LOCATION_TOP, - adev->gmc.vram_end >> 24); - } - /* GART Enable. */ gfxhub_v2_0_init_gart_aperture_regs(adev); gfxhub_v2_0_init_system_aperture_regs(adev); @@ -298,9 +297,11 @@ void gfxhub_v2_0_gart_disable(struct amdgpu_device *adev) ENABLE_ADVANCED_DRIVER_MODEL, 0); WREG32_SOC15(GC, 0, mmGCMC_VM_MX_L1_TLB_CNTL, tmp); - /* Setup L2 cache */ - WREG32_FIELD15(GC, 0, GCVM_L2_CNTL, ENABLE_L2_CACHE, 0); - WREG32_SOC15(GC, 0, mmGCVM_L2_CNTL3, 0); + if (!amdgpu_sriov_vf(adev)) { + /* Setup L2 cache */ + WREG32_FIELD15(GC, 0, GCVM_L2_CNTL, ENABLE_L2_CACHE, 0); + WREG32_SOC15(GC, 0, mmGCVM_L2_CNTL3, 0); + } } /** diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c index 90216abf14a4..8606f877478f 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c @@ -476,13 +476,26 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, { bool use_semaphore = gmc_v9_0_use_invalidate_semaphore(adev, vmhub); const unsigned eng = 17; - u32 j, inv_req, tmp; + u32 j, inv_req, inv_req2, tmp; struct amdgpu_vmhub *hub; BUG_ON(vmhub >= adev->num_vmhubs); hub = &adev->vmhub[vmhub]; - inv_req = gmc_v9_0_get_invalidate_req(vmid, flush_type); + if (adev->gmc.xgmi.num_physical_nodes && + adev->asic_type == CHIP_VEGA20) { + /* Vega20+XGMI caches PTEs in TC and TLB. Add a + * heavy-weight TLB flush (type 2), which flushes + * both. Due to a race condition with concurrent + * memory accesses using the same TLB cache line, we + * still need a second TLB flush after this. + */ + inv_req = gmc_v9_0_get_invalidate_req(vmid, 2); + inv_req2 = gmc_v9_0_get_invalidate_req(vmid, flush_type); + } else { + inv_req = gmc_v9_0_get_invalidate_req(vmid, flush_type); + inv_req2 = 0; + } /* This is necessary for a HW workaround under SRIOV as well * as GFXOFF under bare metal @@ -521,21 +534,27 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, DRM_ERROR("Timeout waiting for sem acquire in VM flush!\n"); } - WREG32_NO_KIQ(hub->vm_inv_eng0_req + eng, inv_req); + do { + WREG32_NO_KIQ(hub->vm_inv_eng0_req + eng, inv_req); - /* - * Issue a dummy read to wait for the ACK register to be cleared - * to avoid a false ACK due to the new fast GRBM interface. - */ - if (vmhub == AMDGPU_GFXHUB_0) - RREG32_NO_KIQ(hub->vm_inv_eng0_req + eng); + /* + * Issue a dummy read to wait for the ACK register to + * be cleared to avoid a false ACK due to the new fast + * GRBM interface. + */ + if (vmhub == AMDGPU_GFXHUB_0) + RREG32_NO_KIQ(hub->vm_inv_eng0_req + eng); - for (j = 0; j < adev->usec_timeout; j++) { - tmp = RREG32_NO_KIQ(hub->vm_inv_eng0_ack + eng); - if (tmp & (1 << vmid)) - break; - udelay(1); - } + for (j = 0; j < adev->usec_timeout; j++) { + tmp = RREG32_NO_KIQ(hub->vm_inv_eng0_ack + eng); + if (tmp & (1 << vmid)) + break; + udelay(1); + } + + inv_req = inv_req2; + inv_req2 = 0; + } while (inv_req); /* TODO: It needs to continue working on debugging with semaphore for GFXHUB as well. */ if (use_semaphore) @@ -577,9 +596,26 @@ static int gmc_v9_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev, return -EIO; if (ring->sched.ready) { + /* Vega20+XGMI caches PTEs in TC and TLB. Add a + * heavy-weight TLB flush (type 2), which flushes + * both. Due to a race condition with concurrent + * memory accesses using the same TLB cache line, we + * still need a second TLB flush after this. + */ + bool vega20_xgmi_wa = (adev->gmc.xgmi.num_physical_nodes && + adev->asic_type == CHIP_VEGA20); + /* 2 dwords flush + 8 dwords fence */ + unsigned int ndw = kiq->pmf->invalidate_tlbs_size + 8; + + if (vega20_xgmi_wa) + ndw += kiq->pmf->invalidate_tlbs_size; + spin_lock(&adev->gfx.kiq.ring_lock); /* 2 dwords flush + 8 dwords fence */ - amdgpu_ring_alloc(ring, kiq->pmf->invalidate_tlbs_size + 8); + amdgpu_ring_alloc(ring, ndw); + if (vega20_xgmi_wa) + kiq->pmf->kiq_invalidate_tlbs(ring, + pasid, 2, all_hub); kiq->pmf->kiq_invalidate_tlbs(ring, pasid, flush_type, all_hub); amdgpu_fence_emit_polling(ring, &seq); @@ -886,32 +922,25 @@ static int gmc_v9_0_late_init(void *handle) if (r) return r; /* Check if ecc is available */ - if (!amdgpu_sriov_vf(adev)) { - switch (adev->asic_type) { - case CHIP_VEGA10: - case CHIP_VEGA20: - case CHIP_ARCTURUS: - r = amdgpu_atomfirmware_mem_ecc_supported(adev); - if (!r) { - DRM_INFO("ECC is not present.\n"); - if (adev->df.funcs->enable_ecc_force_par_wr_rmw) - adev->df.funcs->enable_ecc_force_par_wr_rmw(adev, false); - } else { - DRM_INFO("ECC is active.\n"); - } - - r = amdgpu_atomfirmware_sram_ecc_supported(adev); - if (!r) { - DRM_INFO("SRAM ECC is not present.\n"); - } else { - DRM_INFO("SRAM ECC is active.\n"); - } - break; - default: - break; - } + if (!amdgpu_sriov_vf(adev) && (adev->asic_type == CHIP_VEGA10)) { + r = amdgpu_atomfirmware_mem_ecc_supported(adev); + if (!r) { + DRM_INFO("ECC is not present.\n"); + if (adev->df.funcs->enable_ecc_force_par_wr_rmw) + adev->df.funcs->enable_ecc_force_par_wr_rmw(adev, false); + } else + DRM_INFO("ECC is active.\n"); + + r = amdgpu_atomfirmware_sram_ecc_supported(adev); + if (!r) + DRM_INFO("SRAM ECC is not present.\n"); + else + DRM_INFO("SRAM ECC is active.\n"); } + if (adev->mmhub.funcs && adev->mmhub.funcs->reset_ras_error_count) + adev->mmhub.funcs->reset_ras_error_count(adev); + r = amdgpu_gmc_ras_late_init(adev); if (r) return r; @@ -1272,6 +1301,19 @@ static void gmc_v9_0_init_golden_registers(struct amdgpu_device *adev) } /** + * gmc_v9_0_restore_registers - restores regs + * + * @adev: amdgpu_device pointer + * + * This restores register values, saved at suspend. + */ +static void gmc_v9_0_restore_registers(struct amdgpu_device *adev) +{ + if (adev->asic_type == CHIP_RAVEN) + WREG32(mmDCHUBBUB_SDPIF_MMIO_CNTRL_0, adev->gmc.sdpif_register); +} + +/** * gmc_v9_0_gart_enable - gart enable * * @adev: amdgpu_device pointer @@ -1377,6 +1419,20 @@ static int gmc_v9_0_hw_init(void *handle) } /** + * gmc_v9_0_save_registers - saves regs + * + * @adev: amdgpu_device pointer + * + * This saves potential register values that should be + * restored upon resume + */ +static void gmc_v9_0_save_registers(struct amdgpu_device *adev) +{ + if (adev->asic_type == CHIP_RAVEN) + adev->gmc.sdpif_register = RREG32(mmDCHUBBUB_SDPIF_MMIO_CNTRL_0); +} + +/** * gmc_v9_0_gart_disable - gart disable * * @adev: amdgpu_device pointer @@ -1412,9 +1468,16 @@ static int gmc_v9_0_hw_fini(void *handle) static int gmc_v9_0_suspend(void *handle) { + int r; struct amdgpu_device *adev = (struct amdgpu_device *)handle; - return gmc_v9_0_hw_fini(adev); + r = gmc_v9_0_hw_fini(adev); + if (r) + return r; + + gmc_v9_0_save_registers(adev); + + return 0; } static int gmc_v9_0_resume(void *handle) @@ -1422,6 +1485,7 @@ static int gmc_v9_0_resume(void *handle) int r; struct amdgpu_device *adev = (struct amdgpu_device *)handle; + gmc_v9_0_restore_registers(adev); r = gmc_v9_0_hw_init(adev); if (r) return r; diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c index ff2e6e1ccde7..6173951db7b4 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c @@ -693,7 +693,7 @@ static int jpeg_v2_0_set_clockgating_state(void *handle, bool enable = (state == AMD_CG_STATE_GATE); if (enable) { - if (jpeg_v2_0_is_idle(handle)) + if (!jpeg_v2_0_is_idle(handle)) return -EBUSY; jpeg_v2_0_enable_clock_gating(adev); } else { diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c index c6d046df4b70..c04c2078a7c1 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c @@ -477,7 +477,7 @@ static int jpeg_v2_5_set_clockgating_state(void *handle, continue; if (enable) { - if (jpeg_v2_5_is_idle(handle)) + if (!jpeg_v2_5_is_idle(handle)) return -EBUSY; jpeg_v2_5_enable_clock_gating(adev, i); } else { diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c index 49a3a56ec017..396c2a624de0 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c @@ -747,7 +747,19 @@ static void mmhub_v1_0_query_ras_error_count(struct amdgpu_device *adev, err_data->ue_count += ded_count; } +static void mmhub_v1_0_reset_ras_error_count(struct amdgpu_device *adev) +{ + uint32_t i; + + /* read back edc counter registers to reset the counters to 0 */ + if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__MMHUB)) { + for (i = 0; i < ARRAY_SIZE(mmhub_v1_0_edc_cnt_regs); i++) + RREG32(SOC15_REG_ENTRY_OFFSET(mmhub_v1_0_edc_cnt_regs[i])); + } +} + const struct amdgpu_mmhub_funcs mmhub_v1_0_funcs = { .ras_late_init = amdgpu_mmhub_ras_late_init, .query_ras_error_count = mmhub_v1_0_query_ras_error_count, + .reset_ras_error_count = mmhub_v1_0_reset_ras_error_count, }; diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c index bde189680521..fb3f228458e5 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c @@ -72,11 +72,18 @@ static void mmhub_v2_0_init_system_aperture_regs(struct amdgpu_device *adev) WREG32_SOC15(MMHUB, 0, mmMMMC_VM_AGP_TOP, 0); WREG32_SOC15(MMHUB, 0, mmMMMC_VM_AGP_BOT, 0x00FFFFFF); - /* Program the system aperture low logical page number. */ - WREG32_SOC15(MMHUB, 0, mmMMMC_VM_SYSTEM_APERTURE_LOW_ADDR, - adev->gmc.vram_start >> 18); - WREG32_SOC15(MMHUB, 0, mmMMMC_VM_SYSTEM_APERTURE_HIGH_ADDR, - adev->gmc.vram_end >> 18); + if (!amdgpu_sriov_vf(adev)) { + /* + * the new L1 policy will block SRIOV guest from writing + * these regs, and they will be programed at host. + * so skip programing these regs. + */ + /* Program the system aperture low logical page number. */ + WREG32_SOC15(MMHUB, 0, mmMMMC_VM_SYSTEM_APERTURE_LOW_ADDR, + adev->gmc.vram_start >> 18); + WREG32_SOC15(MMHUB, 0, mmMMMC_VM_SYSTEM_APERTURE_HIGH_ADDR, + adev->gmc.vram_end >> 18); + } /* Set default page address. */ value = adev->vram_scratch.gpu_addr - adev->gmc.vram_start + @@ -247,18 +254,6 @@ static void mmhub_v2_0_program_invalidation(struct amdgpu_device *adev) int mmhub_v2_0_gart_enable(struct amdgpu_device *adev) { - if (amdgpu_sriov_vf(adev)) { - /* - * MMMC_VM_FB_LOCATION_BASE/TOP is NULL for VF, becuase they are - * VF copy registers so vbios post doesn't program them, for - * SRIOV driver need to program them - */ - WREG32_SOC15(MMHUB, 0, mmMMMC_VM_FB_LOCATION_BASE, - adev->gmc.vram_start >> 24); - WREG32_SOC15(MMHUB, 0, mmMMMC_VM_FB_LOCATION_TOP, - adev->gmc.vram_end >> 24); - } - /* GART Enable. */ mmhub_v2_0_init_gart_aperture_regs(adev); mmhub_v2_0_init_system_aperture_regs(adev); diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c index a5281df8d84f..c0e3efcb09bf 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c @@ -1539,8 +1539,11 @@ static const struct soc15_reg_entry mmhub_v9_4_edc_cnt_regs[] = { { SOC15_REG_ENTRY(MMHUB, 0, mmMMEA7_EDC_CNT3), 0, 0, 0 }, }; -static int mmhub_v9_4_get_ras_error_count(const struct soc15_reg_entry *reg, - uint32_t value, uint32_t *sec_count, uint32_t *ded_count) +static int mmhub_v9_4_get_ras_error_count(struct amdgpu_device *adev, + const struct soc15_reg_entry *reg, + uint32_t value, + uint32_t *sec_count, + uint32_t *ded_count) { uint32_t i; uint32_t sec_cnt, ded_cnt; @@ -1553,7 +1556,7 @@ static int mmhub_v9_4_get_ras_error_count(const struct soc15_reg_entry *reg, mmhub_v9_4_ras_fields[i].sec_count_mask) >> mmhub_v9_4_ras_fields[i].sec_count_shift; if (sec_cnt) { - DRM_INFO("MMHUB SubBlock %s, SEC %d\n", + dev_info(adev->dev, "MMHUB SubBlock %s, SEC %d\n", mmhub_v9_4_ras_fields[i].name, sec_cnt); *sec_count += sec_cnt; @@ -1563,7 +1566,7 @@ static int mmhub_v9_4_get_ras_error_count(const struct soc15_reg_entry *reg, mmhub_v9_4_ras_fields[i].ded_count_mask) >> mmhub_v9_4_ras_fields[i].ded_count_shift; if (ded_cnt) { - DRM_INFO("MMHUB SubBlock %s, DED %d\n", + dev_info(adev->dev, "MMHUB SubBlock %s, DED %d\n", mmhub_v9_4_ras_fields[i].name, ded_cnt); *ded_count += ded_cnt; @@ -1588,7 +1591,7 @@ static void mmhub_v9_4_query_ras_error_count(struct amdgpu_device *adev, reg_value = RREG32(SOC15_REG_ENTRY_OFFSET(mmhub_v9_4_edc_cnt_regs[i])); if (reg_value) - mmhub_v9_4_get_ras_error_count(&mmhub_v9_4_edc_cnt_regs[i], + mmhub_v9_4_get_ras_error_count(adev, &mmhub_v9_4_edc_cnt_regs[i], reg_value, &sec_count, &ded_count); } @@ -1596,7 +1599,19 @@ static void mmhub_v9_4_query_ras_error_count(struct amdgpu_device *adev, err_data->ue_count += ded_count; } +static void mmhub_v9_4_reset_ras_error_count(struct amdgpu_device *adev) +{ + uint32_t i; + + /* read back edc counter registers to reset the counters to 0 */ + if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__MMHUB)) { + for (i = 0; i < ARRAY_SIZE(mmhub_v9_4_edc_cnt_regs); i++) + RREG32(SOC15_REG_ENTRY_OFFSET(mmhub_v9_4_edc_cnt_regs[i])); + } +} + const struct amdgpu_mmhub_funcs mmhub_v9_4_funcs = { .ras_late_init = amdgpu_mmhub_ras_late_init, .query_ras_error_count = mmhub_v9_4_query_ras_error_count, + .reset_ras_error_count = mmhub_v9_4_reset_ras_error_count, }; diff --git a/drivers/gpu/drm/amd/amdgpu/mmsch_v2_0.h b/drivers/gpu/drm/amd/amdgpu/mmsch_v2_0.h new file mode 100644 index 000000000000..1b5086c7d4e6 --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/mmsch_v2_0.h @@ -0,0 +1,338 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ + +#ifndef __MMSCH_V2_0_H__ +#define __MMSCH_V2_0_H__ + +// addressBlock: uvd0_mmsch_dec +// base address: 0x1e000 +#define mmMMSCH_UCODE_ADDR 0x0000 +#define mmMMSCH_UCODE_ADDR_BASE_IDX 0 +#define mmMMSCH_UCODE_DATA 0x0001 +#define mmMMSCH_UCODE_DATA_BASE_IDX 0 +#define mmMMSCH_SRAM_ADDR 0x0002 +#define mmMMSCH_SRAM_ADDR_BASE_IDX 0 +#define mmMMSCH_SRAM_DATA 0x0003 +#define mmMMSCH_SRAM_DATA_BASE_IDX 0 +#define mmMMSCH_VF_SRAM_OFFSET 0x0004 +#define mmMMSCH_VF_SRAM_OFFSET_BASE_IDX 0 +#define mmMMSCH_DB_SRAM_OFFSET 0x0005 +#define mmMMSCH_DB_SRAM_OFFSET_BASE_IDX 0 +#define mmMMSCH_CTX_SRAM_OFFSET 0x0006 +#define mmMMSCH_CTX_SRAM_OFFSET_BASE_IDX 0 +#define mmMMSCH_CTL 0x0007 +#define mmMMSCH_CTL_BASE_IDX 0 +#define mmMMSCH_INTR 0x0008 +#define mmMMSCH_INTR_BASE_IDX 0 +#define mmMMSCH_INTR_ACK 0x0009 +#define mmMMSCH_INTR_ACK_BASE_IDX 0 +#define mmMMSCH_INTR_STATUS 0x000a +#define mmMMSCH_INTR_STATUS_BASE_IDX 0 +#define mmMMSCH_VF_VMID 0x000b +#define mmMMSCH_VF_VMID_BASE_IDX 0 +#define mmMMSCH_VF_CTX_ADDR_LO 0x000c +#define mmMMSCH_VF_CTX_ADDR_LO_BASE_IDX 0 +#define mmMMSCH_VF_CTX_ADDR_HI 0x000d +#define mmMMSCH_VF_CTX_ADDR_HI_BASE_IDX 0 +#define mmMMSCH_VF_CTX_SIZE 0x000e +#define mmMMSCH_VF_CTX_SIZE_BASE_IDX 0 +#define mmMMSCH_VF_GPCOM_ADDR_LO 0x000f +#define mmMMSCH_VF_GPCOM_ADDR_LO_BASE_IDX 0 +#define mmMMSCH_VF_GPCOM_ADDR_HI 0x0010 +#define mmMMSCH_VF_GPCOM_ADDR_HI_BASE_IDX 0 +#define mmMMSCH_VF_GPCOM_SIZE 0x0011 +#define mmMMSCH_VF_GPCOM_SIZE_BASE_IDX 0 +#define mmMMSCH_VF_MAILBOX_HOST 0x0012 +#define mmMMSCH_VF_MAILBOX_HOST_BASE_IDX 0 +#define mmMMSCH_VF_MAILBOX_RESP 0x0013 +#define mmMMSCH_VF_MAILBOX_RESP_BASE_IDX 0 +#define mmMMSCH_VF_MAILBOX_0 0x0014 +#define mmMMSCH_VF_MAILBOX_0_BASE_IDX 0 +#define mmMMSCH_VF_MAILBOX_0_RESP 0x0015 +#define mmMMSCH_VF_MAILBOX_0_RESP_BASE_IDX 0 +#define mmMMSCH_VF_MAILBOX_1 0x0016 +#define mmMMSCH_VF_MAILBOX_1_BASE_IDX 0 +#define mmMMSCH_VF_MAILBOX_1_RESP 0x0017 +#define mmMMSCH_VF_MAILBOX_1_RESP_BASE_IDX 0 +#define mmMMSCH_CNTL 0x001c +#define mmMMSCH_CNTL_BASE_IDX 0 +#define mmMMSCH_NONCACHE_OFFSET0 0x001d +#define mmMMSCH_NONCACHE_OFFSET0_BASE_IDX 0 +#define mmMMSCH_NONCACHE_SIZE0 0x001e +#define mmMMSCH_NONCACHE_SIZE0_BASE_IDX 0 +#define mmMMSCH_NONCACHE_OFFSET1 0x001f +#define mmMMSCH_NONCACHE_OFFSET1_BASE_IDX 0 +#define mmMMSCH_NONCACHE_SIZE1 0x0020 +#define mmMMSCH_NONCACHE_SIZE1_BASE_IDX 0 +#define mmMMSCH_PDEBUG_STATUS 0x0021 +#define mmMMSCH_PDEBUG_STATUS_BASE_IDX 0 +#define mmMMSCH_PDEBUG_DATA_32UPPERBITS 0x0022 +#define mmMMSCH_PDEBUG_DATA_32UPPERBITS_BASE_IDX 0 +#define mmMMSCH_PDEBUG_DATA_32LOWERBITS 0x0023 +#define mmMMSCH_PDEBUG_DATA_32LOWERBITS_BASE_IDX 0 +#define mmMMSCH_PDEBUG_EPC 0x0024 +#define mmMMSCH_PDEBUG_EPC_BASE_IDX 0 +#define mmMMSCH_PDEBUG_EXCCAUSE 0x0025 +#define mmMMSCH_PDEBUG_EXCCAUSE_BASE_IDX 0 +#define mmMMSCH_PROC_STATE1 0x0026 +#define mmMMSCH_PROC_STATE1_BASE_IDX 0 +#define mmMMSCH_LAST_MC_ADDR 0x0027 +#define mmMMSCH_LAST_MC_ADDR_BASE_IDX 0 +#define mmMMSCH_LAST_MEM_ACCESS_HI 0x0028 +#define mmMMSCH_LAST_MEM_ACCESS_HI_BASE_IDX 0 +#define mmMMSCH_LAST_MEM_ACCESS_LO 0x0029 +#define mmMMSCH_LAST_MEM_ACCESS_LO_BASE_IDX 0 +#define mmMMSCH_IOV_ACTIVE_FCN_ID 0x002a +#define mmMMSCH_IOV_ACTIVE_FCN_ID_BASE_IDX 0 +#define mmMMSCH_SCRATCH_0 0x002b +#define mmMMSCH_SCRATCH_0_BASE_IDX 0 +#define mmMMSCH_SCRATCH_1 0x002c +#define mmMMSCH_SCRATCH_1_BASE_IDX 0 +#define mmMMSCH_GPUIOV_SCH_BLOCK_0 0x002d +#define mmMMSCH_GPUIOV_SCH_BLOCK_0_BASE_IDX 0 +#define mmMMSCH_GPUIOV_CMD_CONTROL_0 0x002e +#define mmMMSCH_GPUIOV_CMD_CONTROL_0_BASE_IDX 0 +#define mmMMSCH_GPUIOV_CMD_STATUS_0 0x002f +#define mmMMSCH_GPUIOV_CMD_STATUS_0_BASE_IDX 0 +#define mmMMSCH_GPUIOV_VM_BUSY_STATUS_0 0x0030 +#define mmMMSCH_GPUIOV_VM_BUSY_STATUS_0_BASE_IDX 0 +#define mmMMSCH_GPUIOV_ACTIVE_FCNS_0 0x0031 +#define mmMMSCH_GPUIOV_ACTIVE_FCNS_0_BASE_IDX 0 +#define mmMMSCH_GPUIOV_ACTIVE_FCN_ID_0 0x0032 +#define mmMMSCH_GPUIOV_ACTIVE_FCN_ID_0_BASE_IDX 0 +#define mmMMSCH_GPUIOV_DW6_0 0x0033 +#define mmMMSCH_GPUIOV_DW6_0_BASE_IDX 0 +#define mmMMSCH_GPUIOV_DW7_0 0x0034 +#define mmMMSCH_GPUIOV_DW7_0_BASE_IDX 0 +#define mmMMSCH_GPUIOV_DW8_0 0x0035 +#define mmMMSCH_GPUIOV_DW8_0_BASE_IDX 0 +#define mmMMSCH_GPUIOV_SCH_BLOCK_1 0x0036 +#define mmMMSCH_GPUIOV_SCH_BLOCK_1_BASE_IDX 0 +#define mmMMSCH_GPUIOV_CMD_CONTROL_1 0x0037 +#define mmMMSCH_GPUIOV_CMD_CONTROL_1_BASE_IDX 0 +#define mmMMSCH_GPUIOV_CMD_STATUS_1 0x0038 +#define mmMMSCH_GPUIOV_CMD_STATUS_1_BASE_IDX 0 +#define mmMMSCH_GPUIOV_VM_BUSY_STATUS_1 0x0039 +#define mmMMSCH_GPUIOV_VM_BUSY_STATUS_1_BASE_IDX 0 +#define mmMMSCH_GPUIOV_ACTIVE_FCNS_1 0x003a +#define mmMMSCH_GPUIOV_ACTIVE_FCNS_1_BASE_IDX 0 +#define mmMMSCH_GPUIOV_ACTIVE_FCN_ID_1 0x003b +#define mmMMSCH_GPUIOV_ACTIVE_FCN_ID_1_BASE_IDX 0 +#define mmMMSCH_GPUIOV_DW6_1 0x003c +#define mmMMSCH_GPUIOV_DW6_1_BASE_IDX 0 +#define mmMMSCH_GPUIOV_DW7_1 0x003d +#define mmMMSCH_GPUIOV_DW7_1_BASE_IDX 0 +#define mmMMSCH_GPUIOV_DW8_1 0x003e +#define mmMMSCH_GPUIOV_DW8_1_BASE_IDX 0 +#define mmMMSCH_GPUIOV_CNTXT 0x003f +#define mmMMSCH_GPUIOV_CNTXT_BASE_IDX 0 +#define mmMMSCH_SCRATCH_2 0x0040 +#define mmMMSCH_SCRATCH_2_BASE_IDX 0 +#define mmMMSCH_SCRATCH_3 0x0041 +#define mmMMSCH_SCRATCH_3_BASE_IDX 0 +#define mmMMSCH_SCRATCH_4 0x0042 +#define mmMMSCH_SCRATCH_4_BASE_IDX 0 +#define mmMMSCH_SCRATCH_5 0x0043 +#define mmMMSCH_SCRATCH_5_BASE_IDX 0 +#define mmMMSCH_SCRATCH_6 0x0044 +#define mmMMSCH_SCRATCH_6_BASE_IDX 0 +#define mmMMSCH_SCRATCH_7 0x0045 +#define mmMMSCH_SCRATCH_7_BASE_IDX 0 +#define mmMMSCH_VFID_FIFO_HEAD_0 0x0046 +#define mmMMSCH_VFID_FIFO_HEAD_0_BASE_IDX 0 +#define mmMMSCH_VFID_FIFO_TAIL_0 0x0047 +#define mmMMSCH_VFID_FIFO_TAIL_0_BASE_IDX 0 +#define mmMMSCH_VFID_FIFO_HEAD_1 0x0048 +#define mmMMSCH_VFID_FIFO_HEAD_1_BASE_IDX 0 +#define mmMMSCH_VFID_FIFO_TAIL_1 0x0049 +#define mmMMSCH_VFID_FIFO_TAIL_1_BASE_IDX 0 +#define mmMMSCH_NACK_STATUS 0x004a +#define mmMMSCH_NACK_STATUS_BASE_IDX 0 +#define mmMMSCH_VF_MAILBOX0_DATA 0x004b +#define mmMMSCH_VF_MAILBOX0_DATA_BASE_IDX 0 +#define mmMMSCH_VF_MAILBOX1_DATA 0x004c +#define mmMMSCH_VF_MAILBOX1_DATA_BASE_IDX 0 +#define mmMMSCH_GPUIOV_SCH_BLOCK_IP_0 0x004d +#define mmMMSCH_GPUIOV_SCH_BLOCK_IP_0_BASE_IDX 0 +#define mmMMSCH_GPUIOV_CMD_STATUS_IP_0 0x004e +#define mmMMSCH_GPUIOV_CMD_STATUS_IP_0_BASE_IDX 0 +#define mmMMSCH_GPUIOV_ACTIVE_FCN_ID_IP_0 0x004f +#define mmMMSCH_GPUIOV_ACTIVE_FCN_ID_IP_0_BASE_IDX 0 +#define mmMMSCH_GPUIOV_SCH_BLOCK_IP_1 0x0050 +#define mmMMSCH_GPUIOV_SCH_BLOCK_IP_1_BASE_IDX 0 +#define mmMMSCH_GPUIOV_CMD_STATUS_IP_1 0x0051 +#define mmMMSCH_GPUIOV_CMD_STATUS_IP_1_BASE_IDX 0 +#define mmMMSCH_GPUIOV_ACTIVE_FCN_ID_IP_1 0x0052 +#define mmMMSCH_GPUIOV_ACTIVE_FCN_ID_IP_1_BASE_IDX 0 +#define mmMMSCH_GPUIOV_CNTXT_IP 0x0053 +#define mmMMSCH_GPUIOV_CNTXT_IP_BASE_IDX 0 +#define mmMMSCH_GPUIOV_SCH_BLOCK_2 0x0054 +#define mmMMSCH_GPUIOV_SCH_BLOCK_2_BASE_IDX 0 +#define mmMMSCH_GPUIOV_CMD_CONTROL_2 0x0055 +#define mmMMSCH_GPUIOV_CMD_CONTROL_2_BASE_IDX 0 +#define mmMMSCH_GPUIOV_CMD_STATUS_2 0x0056 +#define mmMMSCH_GPUIOV_CMD_STATUS_2_BASE_IDX 0 +#define mmMMSCH_GPUIOV_VM_BUSY_STATUS_2 0x0057 +#define mmMMSCH_GPUIOV_VM_BUSY_STATUS_2_BASE_IDX 0 +#define mmMMSCH_GPUIOV_ACTIVE_FCNS_2 0x0058 +#define mmMMSCH_GPUIOV_ACTIVE_FCNS_2_BASE_IDX 0 +#define mmMMSCH_GPUIOV_ACTIVE_FCN_ID_2 0x0059 +#define mmMMSCH_GPUIOV_ACTIVE_FCN_ID_2_BASE_IDX 0 +#define mmMMSCH_GPUIOV_DW6_2 0x005a +#define mmMMSCH_GPUIOV_DW6_2_BASE_IDX 0 +#define mmMMSCH_GPUIOV_DW7_2 0x005b +#define mmMMSCH_GPUIOV_DW7_2_BASE_IDX 0 +#define mmMMSCH_GPUIOV_DW8_2 0x005c +#define mmMMSCH_GPUIOV_DW8_2_BASE_IDX 0 +#define mmMMSCH_GPUIOV_SCH_BLOCK_IP_2 0x005d +#define mmMMSCH_GPUIOV_SCH_BLOCK_IP_2_BASE_IDX 0 +#define mmMMSCH_GPUIOV_CMD_STATUS_IP_2 0x005e +#define mmMMSCH_GPUIOV_CMD_STATUS_IP_2_BASE_IDX 0 +#define mmMMSCH_GPUIOV_ACTIVE_FCN_ID_IP_2 0x005f +#define mmMMSCH_GPUIOV_ACTIVE_FCN_ID_IP_2_BASE_IDX 0 +#define mmMMSCH_VFID_FIFO_HEAD_2 0x0060 +#define mmMMSCH_VFID_FIFO_HEAD_2_BASE_IDX 0 +#define mmMMSCH_VFID_FIFO_TAIL_2 0x0061 +#define mmMMSCH_VFID_FIFO_TAIL_2_BASE_IDX 0 +#define mmMMSCH_VM_BUSY_STATUS_0 0x0062 +#define mmMMSCH_VM_BUSY_STATUS_0_BASE_IDX 0 +#define mmMMSCH_VM_BUSY_STATUS_1 0x0063 +#define mmMMSCH_VM_BUSY_STATUS_1_BASE_IDX 0 +#define mmMMSCH_VM_BUSY_STATUS_2 0x0064 +#define mmMMSCH_VM_BUSY_STATUS_2_BASE_IDX 0 + +#define MMSCH_VERSION_MAJOR 2 +#define MMSCH_VERSION_MINOR 0 +#define MMSCH_VERSION (MMSCH_VERSION_MAJOR << 16 | MMSCH_VERSION_MINOR) + +enum mmsch_v2_0_command_type { + MMSCH_COMMAND__DIRECT_REG_WRITE = 0, + MMSCH_COMMAND__DIRECT_REG_POLLING = 2, + MMSCH_COMMAND__DIRECT_REG_READ_MODIFY_WRITE = 3, + MMSCH_COMMAND__INDIRECT_REG_WRITE = 8, + MMSCH_COMMAND__END = 0xf +}; + +struct mmsch_v2_0_init_header { + uint32_t version; + uint32_t header_size; + uint32_t vcn_init_status; + uint32_t vcn_table_offset; + uint32_t vcn_table_size; +}; + +struct mmsch_v2_0_cmd_direct_reg_header { + uint32_t reg_offset : 28; + uint32_t command_type : 4; +}; + +struct mmsch_v2_0_cmd_indirect_reg_header { + uint32_t reg_offset : 20; + uint32_t reg_idx_space : 8; + uint32_t command_type : 4; +}; + +struct mmsch_v2_0_cmd_direct_write { + struct mmsch_v2_0_cmd_direct_reg_header cmd_header; + uint32_t reg_value; +}; + +struct mmsch_v2_0_cmd_direct_read_modify_write { + struct mmsch_v2_0_cmd_direct_reg_header cmd_header; + uint32_t write_data; + uint32_t mask_value; +}; + +struct mmsch_v2_0_cmd_direct_polling { + struct mmsch_v2_0_cmd_direct_reg_header cmd_header; + uint32_t mask_value; + uint32_t wait_value; +}; + +struct mmsch_v2_0_cmd_end { + struct mmsch_v2_0_cmd_direct_reg_header cmd_header; +}; + +struct mmsch_v2_0_cmd_indirect_write { + struct mmsch_v2_0_cmd_indirect_reg_header cmd_header; + uint32_t reg_value; +}; + +static inline void mmsch_v2_0_insert_direct_wt(struct mmsch_v2_0_cmd_direct_write *direct_wt, + uint32_t *init_table, + uint32_t reg_offset, + uint32_t value) +{ + direct_wt->cmd_header.reg_offset = reg_offset; + direct_wt->reg_value = value; + memcpy((void *)init_table, direct_wt, sizeof(struct mmsch_v2_0_cmd_direct_write)); +} + +static inline void mmsch_v2_0_insert_direct_rd_mod_wt(struct mmsch_v2_0_cmd_direct_read_modify_write *direct_rd_mod_wt, + uint32_t *init_table, + uint32_t reg_offset, + uint32_t mask, uint32_t data) +{ + direct_rd_mod_wt->cmd_header.reg_offset = reg_offset; + direct_rd_mod_wt->mask_value = mask; + direct_rd_mod_wt->write_data = data; + memcpy((void *)init_table, direct_rd_mod_wt, + sizeof(struct mmsch_v2_0_cmd_direct_read_modify_write)); +} + +static inline void mmsch_v2_0_insert_direct_poll(struct mmsch_v2_0_cmd_direct_polling *direct_poll, + uint32_t *init_table, + uint32_t reg_offset, + uint32_t mask, uint32_t wait) +{ + direct_poll->cmd_header.reg_offset = reg_offset; + direct_poll->mask_value = mask; + direct_poll->wait_value = wait; + memcpy((void *)init_table, direct_poll, sizeof(struct mmsch_v2_0_cmd_direct_polling)); +} + +#define MMSCH_V2_0_INSERT_DIRECT_RD_MOD_WT(reg, mask, data) { \ + mmsch_v2_0_insert_direct_rd_mod_wt(&direct_rd_mod_wt, \ + init_table, (reg), \ + (mask), (data)); \ + init_table += sizeof(struct mmsch_v2_0_cmd_direct_read_modify_write)/4; \ + table_size += sizeof(struct mmsch_v2_0_cmd_direct_read_modify_write)/4; \ +} + +#define MMSCH_V2_0_INSERT_DIRECT_WT(reg, value) { \ + mmsch_v2_0_insert_direct_wt(&direct_wt, \ + init_table, (reg), \ + (value)); \ + init_table += sizeof(struct mmsch_v2_0_cmd_direct_write)/4; \ + table_size += sizeof(struct mmsch_v2_0_cmd_direct_write)/4; \ +} + +#define MMSCH_V2_0_INSERT_DIRECT_POLL(reg, mask, wait) { \ + mmsch_v2_0_insert_direct_poll(&direct_poll, \ + init_table, (reg), \ + (mask), (wait)); \ + init_table += sizeof(struct mmsch_v2_0_cmd_direct_polling)/4; \ + table_size += sizeof(struct mmsch_v2_0_cmd_direct_polling)/4; \ +} + +#endif diff --git a/drivers/gpu/drm/amd/amdgpu/navi10_ih.c b/drivers/gpu/drm/amd/amdgpu/navi10_ih.c index cf557a428298..e08245a446fc 100644 --- a/drivers/gpu/drm/amd/amdgpu/navi10_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/navi10_ih.c @@ -32,6 +32,7 @@ #include "soc15_common.h" #include "navi10_ih.h" +#define MAX_REARM_RETRY 10 static void navi10_ih_set_interrupt_funcs(struct amdgpu_device *adev); @@ -284,6 +285,38 @@ static void navi10_ih_decode_iv(struct amdgpu_device *adev, } /** + * navi10_ih_irq_rearm - rearm IRQ if lost + * + * @adev: amdgpu_device pointer + * + */ +static void navi10_ih_irq_rearm(struct amdgpu_device *adev, + struct amdgpu_ih_ring *ih) +{ + uint32_t reg_rptr = 0; + uint32_t v = 0; + uint32_t i = 0; + + if (ih == &adev->irq.ih) + reg_rptr = SOC15_REG_OFFSET(OSSSYS, 0, mmIH_RB_RPTR); + else if (ih == &adev->irq.ih1) + reg_rptr = SOC15_REG_OFFSET(OSSSYS, 0, mmIH_RB_RPTR_RING1); + else if (ih == &adev->irq.ih2) + reg_rptr = SOC15_REG_OFFSET(OSSSYS, 0, mmIH_RB_RPTR_RING2); + else + return; + + /* Rearm IRQ / re-write doorbell if doorbell write is lost */ + for (i = 0; i < MAX_REARM_RETRY; i++) { + v = RREG32_NO_KIQ(reg_rptr); + if ((v < ih->ring_size) && (v != ih->rptr)) + WDOORBELL32(ih->doorbell_index, ih->rptr); + else + break; + } +} + +/** * navi10_ih_set_rptr - set the IH ring buffer rptr * * @adev: amdgpu_device pointer @@ -297,6 +330,9 @@ static void navi10_ih_set_rptr(struct amdgpu_device *adev, /* XXX check if swapping is necessary on BE */ *ih->rptr_cpu = ih->rptr; WDOORBELL32(ih->doorbell_index, ih->rptr); + + if (amdgpu_sriov_vf(adev)) + navi10_ih_irq_rearm(adev, ih); } else WREG32_SOC15(OSSSYS, 0, mmIH_RB_RPTR, ih->rptr); } diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c index 65eb378fa035..149d386590df 100644 --- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c +++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c @@ -318,6 +318,7 @@ static void nbio_v7_4_handle_ras_controller_intr_no_bifring(struct amdgpu_device { uint32_t bif_doorbell_intr_cntl; struct ras_manager *obj = amdgpu_ras_find_obj(adev, adev->nbio.ras_if); + struct ras_err_data err_data = {0, 0, 0, NULL}; bif_doorbell_intr_cntl = RREG32_SOC15(NBIO, 0, mmBIF_DOORBELL_INT_CNTL); if (REG_GET_FIELD(bif_doorbell_intr_cntl, @@ -332,7 +333,19 @@ static void nbio_v7_4_handle_ras_controller_intr_no_bifring(struct amdgpu_device * clear error status after ras_controller_intr according to * hw team and count ue number for query */ - nbio_v7_4_query_ras_error_count(adev, &obj->err_data); + nbio_v7_4_query_ras_error_count(adev, &err_data); + + /* logging on error counter and printing for awareness */ + obj->err_data.ue_count += err_data.ue_count; + obj->err_data.ce_count += err_data.ce_count; + + if (err_data.ce_count) + DRM_INFO("%ld correctable errors detected in %s block\n", + obj->err_data.ce_count, adev->nbio.ras_if->name); + + if (err_data.ue_count) + DRM_INFO("%ld uncorrectable errors detected in %s block\n", + obj->err_data.ue_count, adev->nbio.ras_if->name); DRM_WARN("RAS controller interrupt triggered by NBIF error\n"); diff --git a/drivers/gpu/drm/amd/amdgpu/nv.c b/drivers/gpu/drm/amd/amdgpu/nv.c index 2d1bebdf1603..033cbbca2072 100644 --- a/drivers/gpu/drm/amd/amdgpu/nv.c +++ b/drivers/gpu/drm/amd/amdgpu/nv.c @@ -516,7 +516,8 @@ int nv_set_ip_blocks(struct amdgpu_device *adev) !amdgpu_sriov_vf(adev)) amdgpu_device_ip_block_add(adev, &smu_v11_0_ip_block); amdgpu_device_ip_block_add(adev, &vcn_v2_0_ip_block); - amdgpu_device_ip_block_add(adev, &jpeg_v2_0_ip_block); + if (!amdgpu_sriov_vf(adev)) + amdgpu_device_ip_block_add(adev, &jpeg_v2_0_ip_block); break; default: return -EINVAL; diff --git a/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h b/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h index 36b65797434e..a44fd6060d5b 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h +++ b/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h @@ -31,6 +31,9 @@ #define GFX_CMD_RESERVED_MASK 0x7FF00000 #define GFX_CMD_RESPONSE_MASK 0x80000000 +/* USBC PD FW version retrieval command */ +#define C2PMSG_CMD_GFX_USB_PD_FW_VER 0x2000000 + /* TEE Gfx Command IDs for the register interface. * Command ID must be between 0x00010000 and 0x000F0000. */ diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c index 0829188c1a5c..0afd610a1263 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c @@ -26,6 +26,7 @@ #include "amdgpu.h" #include "amdgpu_psp.h" +#include "amdgpu_ras.h" #include "amdgpu_ucode.h" #include "soc15_common.h" #include "psp_v11_0.h" @@ -65,6 +66,9 @@ MODULE_FIRMWARE("amdgpu/arcturus_ta.bin"); /* memory training timeout define */ #define MEM_TRAIN_SEND_MSG_TIMEOUT_US 3000000 +/* For large FW files the time to complete can be very long */ +#define USBC_PD_POLLING_LIMIT_S 240 + static int psp_v11_0_init_microcode(struct psp_context *psp) { struct amdgpu_device *adev = psp->adev; @@ -420,7 +424,8 @@ static int psp_v11_0_ring_init(struct psp_context *psp, struct psp_ring *ring; struct amdgpu_device *adev = psp->adev; - psp_v11_0_reroute_ih(psp); + if (!amdgpu_sriov_vf(adev)) + psp_v11_0_reroute_ih(psp); ring = &psp->km_ring; @@ -864,6 +869,11 @@ static int psp_v11_0_ras_trigger_error(struct psp_context *psp, if (ret) return -EINVAL; + /* If err_event_athub occurs error inject was successful, however + return status from TA is no long reliable */ + if (amdgpu_ras_intr_triggered()) + return 0; + return ras_cmd->ras_status; } @@ -1108,6 +1118,82 @@ static void psp_v11_0_ring_set_wptr(struct psp_context *psp, uint32_t value) WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_67, value); } +static int psp_v11_0_load_usbc_pd_fw(struct psp_context *psp, dma_addr_t dma_addr) +{ + struct amdgpu_device *adev = psp->adev; + uint32_t reg_status; + int ret, i = 0; + + /* Write lower 32-bit address of the PD Controller FW */ + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_36, lower_32_bits(dma_addr)); + ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_35), + 0x80000000, 0x80000000, false); + if (ret) + return ret; + + /* Fireup interrupt so PSP can pick up the lower address */ + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_35, 0x800000); + ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_35), + 0x80000000, 0x80000000, false); + if (ret) + return ret; + + reg_status = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_35); + + if ((reg_status & 0xFFFF) != 0) { + DRM_ERROR("Lower address load failed - MP0_SMN_C2PMSG_35.Bits [15:0] = %02x...\n", + reg_status & 0xFFFF); + return -EIO; + } + + /* Write upper 32-bit address of the PD Controller FW */ + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_36, upper_32_bits(dma_addr)); + + ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_35), + 0x80000000, 0x80000000, false); + if (ret) + return ret; + + /* Fireup interrupt so PSP can pick up the upper address */ + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_35, 0x4000000); + + /* FW load takes very long time */ + do { + msleep(1000); + reg_status = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_35); + + if (reg_status & 0x80000000) + goto done; + + } while (++i < USBC_PD_POLLING_LIMIT_S); + + return -ETIME; +done: + + if ((reg_status & 0xFFFF) != 0) { + DRM_ERROR("Upper address load failed - MP0_SMN_C2PMSG_35.Bits [15:0] = x%04x\n", + reg_status & 0xFFFF); + return -EIO; + } + + return 0; +} + +static int psp_v11_0_read_usbc_pd_fw(struct psp_context *psp, uint32_t *fw_ver) +{ + struct amdgpu_device *adev = psp->adev; + int ret; + + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_35, C2PMSG_CMD_GFX_USB_PD_FW_VER); + + ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_35), + 0x80000000, 0x80000000, false); + if (!ret) + *fw_ver = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_36); + + return ret; +} + static const struct psp_funcs psp_v11_0_funcs = { .init_microcode = psp_v11_0_init_microcode, .bootloader_load_kdb = psp_v11_0_bootloader_load_kdb, @@ -1132,6 +1218,8 @@ static const struct psp_funcs psp_v11_0_funcs = { .mem_training = psp_v11_0_memory_training, .ring_get_wptr = psp_v11_0_ring_get_wptr, .ring_set_wptr = psp_v11_0_ring_set_wptr, + .load_usbc_pd_fw = psp_v11_0_load_usbc_pd_fw, + .read_usbc_pd_fw = psp_v11_0_read_usbc_pd_fw }; void psp_v11_0_set_psp_funcs(struct psp_context *psp) diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c index e55884d204bd..5f3a5ee2a3f4 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c @@ -677,7 +677,7 @@ static uint64_t sdma_v4_0_ring_get_wptr(struct amdgpu_ring *ring) } /** - * sdma_v4_0_ring_set_wptr - commit the write pointer + * sdma_v4_0_page_ring_set_wptr - commit the write pointer * * @ring: amdgpu ring pointer * @@ -977,7 +977,7 @@ static void sdma_v4_0_page_stop(struct amdgpu_device *adev) } /** - * sdma_v_0_ctx_switch_enable - stop the async dma engines context switch + * sdma_v4_0_ctx_switch_enable - stop the async dma engines context switch * * @adev: amdgpu_device pointer * @enable: enable/disable the DMA MEs context switch. @@ -1801,13 +1801,9 @@ static int sdma_v4_0_late_init(void *handle) struct ras_ih_if ih_info = { .cb = sdma_v4_0_process_ras_data_cb, }; - int i; - /* read back edc counter registers to clear the counters */ - if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__SDMA)) { - for (i = 0; i < adev->sdma.num_instances; i++) - RREG32_SDMA(i, mmSDMA0_EDC_COUNTER); - } + if (adev->sdma.funcs && adev->sdma.funcs->reset_ras_error_count) + adev->sdma.funcs->reset_ras_error_count(adev); if (adev->sdma.funcs && adev->sdma.funcs->ras_late_init) return adev->sdma.funcs->ras_late_init(adev, &ih_info); @@ -2572,10 +2568,22 @@ static int sdma_v4_0_query_ras_error_count(struct amdgpu_device *adev, return 0; }; +static void sdma_v4_0_reset_ras_error_count(struct amdgpu_device *adev) +{ + int i; + + /* read back edc counter registers to clear the counters */ + if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__SDMA)) { + for (i = 0; i < adev->sdma.num_instances; i++) + RREG32_SDMA(i, mmSDMA0_EDC_COUNTER); + } +} + static const struct amdgpu_sdma_ras_funcs sdma_v4_0_ras_funcs = { .ras_late_init = amdgpu_sdma_ras_late_init, .ras_fini = amdgpu_sdma_ras_fini, .query_ras_error_count = sdma_v4_0_query_ras_error_count, + .reset_ras_error_count = sdma_v4_0_reset_ras_error_count, }; static void sdma_v4_0_set_ras_funcs(struct amdgpu_device *adev) diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c index 67b9830b7c7e..ebfd2cdf4e65 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c @@ -746,11 +746,9 @@ static int sdma_v5_0_gfx_resume(struct amdgpu_device *adev) sdma_v5_0_enable(adev, true); } - r = amdgpu_ring_test_ring(ring); - if (r) { - ring->sched.ready = false; + r = amdgpu_ring_test_helper(ring); + if (r) return r; - } if (adev->mman.buffer_funcs_ring == ring) amdgpu_ttm_set_buffer_funcs_status(adev, true); diff --git a/drivers/gpu/drm/amd/amdgpu/si_dpm.c b/drivers/gpu/drm/amd/amdgpu/si_dpm.c index 4cb4c891120b..0860e85a2d35 100644 --- a/drivers/gpu/drm/amd/amdgpu/si_dpm.c +++ b/drivers/gpu/drm/amd/amdgpu/si_dpm.c @@ -3439,7 +3439,6 @@ static void si_apply_state_adjust_rules(struct amdgpu_device *adev, if (adev->asic_type == CHIP_HAINAN) { if ((adev->pdev->revision == 0x81) || - (adev->pdev->revision == 0x83) || (adev->pdev->revision == 0xC3) || (adev->pdev->device == 0x6664) || (adev->pdev->device == 0x6665) || diff --git a/drivers/gpu/drm/amd/amdgpu/smu_v11_0_i2c.c b/drivers/gpu/drm/amd/amdgpu/smu_v11_0_i2c.c index c902f26cf50d..9bffbab35041 100644 --- a/drivers/gpu/drm/amd/amdgpu/smu_v11_0_i2c.c +++ b/drivers/gpu/drm/amd/amdgpu/smu_v11_0_i2c.c @@ -46,8 +46,7 @@ #define I2C_NO_STOP 1 #define I2C_RESTART 2 -#define to_amdgpu_device(x) (container_of(x, struct amdgpu_ras, eeprom_control.eeprom_accessor))->adev -#define to_eeprom_control(x) container_of(x, struct amdgpu_ras_eeprom_control, eeprom_accessor) +#define to_amdgpu_device(x) (container_of(x, struct amdgpu_device, pm.smu_i2c)) static void smu_v11_0_i2c_set_clock_gating(struct i2c_adapter *control, bool en) { @@ -592,7 +591,8 @@ static uint32_t smu_v11_0_i2c_eeprom_write_data(struct i2c_adapter *control, static void lock_bus(struct i2c_adapter *i2c, unsigned int flags) { - struct amdgpu_ras_eeprom_control *control = to_eeprom_control(i2c); + struct amdgpu_device *adev = to_amdgpu_device(i2c); + struct amdgpu_ras_eeprom_control *control = &adev->psp.ras.ras->eeprom_control; if (!smu_v11_0_i2c_bus_lock(i2c)) { DRM_ERROR("Failed to lock the bus from SMU"); @@ -610,7 +610,8 @@ static int trylock_bus(struct i2c_adapter *i2c, unsigned int flags) static void unlock_bus(struct i2c_adapter *i2c, unsigned int flags) { - struct amdgpu_ras_eeprom_control *control = to_eeprom_control(i2c); + struct amdgpu_device *adev = to_amdgpu_device(i2c); + struct amdgpu_ras_eeprom_control *control = &adev->psp.ras.ras->eeprom_control; if (!smu_v11_0_i2c_bus_unlock(i2c)) { DRM_ERROR("Failed to unlock the bus from SMU"); @@ -630,7 +631,8 @@ static int smu_v11_0_i2c_eeprom_i2c_xfer(struct i2c_adapter *i2c_adap, struct i2c_msg *msgs, int num) { int i, ret; - struct amdgpu_ras_eeprom_control *control = to_eeprom_control(i2c_adap); + struct amdgpu_device *adev = to_amdgpu_device(i2c_adap); + struct amdgpu_ras_eeprom_control *control = &adev->psp.ras.ras->eeprom_control; if (!control->bus_locked) { DRM_ERROR("I2C bus unlocked, stopping transaction!"); @@ -679,7 +681,7 @@ int smu_v11_0_i2c_eeprom_control_init(struct i2c_adapter *control) control->class = I2C_CLASS_SPD; control->dev.parent = &adev->pdev->dev; control->algo = &smu_v11_0_i2c_eeprom_i2c_algo; - snprintf(control->name, sizeof(control->name), "RAS EEPROM"); + snprintf(control->name, sizeof(control->name), "AMDGPU EEPROM"); control->lock_ops = &smu_v11_0_i2c_i2c_lock_ops; res = i2c_add_adapter(control); diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c index 15f3424a1ff7..a40499d51c93 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15.c +++ b/drivers/gpu/drm/amd/amdgpu/soc15.c @@ -89,6 +89,13 @@ #define HDP_MEM_POWER_CTRL__RC_MEM_POWER_CTRL_EN_MASK 0x00010000L #define HDP_MEM_POWER_CTRL__RC_MEM_POWER_LS_EN_MASK 0x00020000L #define mmHDP_MEM_POWER_CTRL_BASE_IDX 0 + +/* for Vega20/arcturus regiter offset change */ +#define mmROM_INDEX_VG20 0x00e4 +#define mmROM_INDEX_VG20_BASE_IDX 0 +#define mmROM_DATA_VG20 0x00e5 +#define mmROM_DATA_VG20_BASE_IDX 0 + /* * Indirect registers accessor */ @@ -272,7 +279,12 @@ static u32 soc15_get_config_memsize(struct amdgpu_device *adev) static u32 soc15_get_xclk(struct amdgpu_device *adev) { - return adev->clock.spll.reference_freq; + u32 reference_clock = adev->clock.spll.reference_freq; + + if (adev->asic_type == CHIP_RAVEN) + return reference_clock / 4; + + return reference_clock; } @@ -304,6 +316,8 @@ static bool soc15_read_bios_from_rom(struct amdgpu_device *adev, { u32 *dw_ptr; u32 i, length_dw; + uint32_t rom_index_offset; + uint32_t rom_data_offset; if (bios == NULL) return false; @@ -316,11 +330,23 @@ static bool soc15_read_bios_from_rom(struct amdgpu_device *adev, dw_ptr = (u32 *)bios; length_dw = ALIGN(length_bytes, 4) / 4; + switch (adev->asic_type) { + case CHIP_VEGA20: + case CHIP_ARCTURUS: + rom_index_offset = SOC15_REG_OFFSET(SMUIO, 0, mmROM_INDEX_VG20); + rom_data_offset = SOC15_REG_OFFSET(SMUIO, 0, mmROM_DATA_VG20); + break; + default: + rom_index_offset = SOC15_REG_OFFSET(SMUIO, 0, mmROM_INDEX); + rom_data_offset = SOC15_REG_OFFSET(SMUIO, 0, mmROM_DATA); + break; + } + /* set rom index to 0 */ - WREG32(SOC15_REG_OFFSET(SMUIO, 0, mmROM_INDEX), 0); + WREG32(rom_index_offset, 0); /* read out the rom data */ for (i = 0; i < length_dw; i++) - dw_ptr[i] = RREG32(SOC15_REG_OFFSET(SMUIO, 0, mmROM_DATA)); + dw_ptr[i] = RREG32(rom_data_offset); return true; } @@ -826,6 +852,15 @@ static bool soc15_need_full_reset(struct amdgpu_device *adev) /* change this when we implement soft reset */ return true; } + +static void vega20_reset_hdp_ras_error_count(struct amdgpu_device *adev) +{ + if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__HDP)) + return; + /*read back hdp ras counter to reset it to 0 */ + RREG32_SOC15(HDP, 0, mmHDP_EDC_CNT); +} + static void soc15_get_pcie_usage(struct amdgpu_device *adev, uint64_t *count0, uint64_t *count1) { @@ -993,6 +1028,7 @@ static const struct amdgpu_asic_funcs vega20_asic_funcs = .get_config_memsize = &soc15_get_config_memsize, .flush_hdp = &soc15_flush_hdp, .invalidate_hdp = &soc15_invalidate_hdp, + .reset_hdp_ras_error_count = &vega20_reset_hdp_ras_error_count, .need_full_reset = &soc15_need_full_reset, .init_doorbell_index = &vega20_doorbell_index_init, .get_pcie_usage = &vega20_get_pcie_usage, @@ -1238,6 +1274,10 @@ static int soc15_common_late_init(void *handle) if (amdgpu_sriov_vf(adev)) xgpu_ai_mailbox_get_irq(adev); + if (adev->asic_funcs && + adev->asic_funcs->reset_hdp_ras_error_count) + adev->asic_funcs->reset_hdp_ras_error_count(adev); + if (adev->nbio.funcs->ras_late_init) r = adev->nbio.funcs->ras_late_init(adev); diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.h b/drivers/gpu/drm/amd/amdgpu/soc15.h index d0fb7a67c1a3..b03f950c486c 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15.h +++ b/drivers/gpu/drm/amd/amdgpu/soc15.h @@ -42,6 +42,13 @@ struct soc15_reg_golden { u32 or_mask; }; +struct soc15_reg_rlcg { + u32 hwip; + u32 instance; + u32 segment; + u32 reg; +}; + struct soc15_reg_entry { uint32_t hwip; uint32_t inst; diff --git a/drivers/gpu/drm/amd/amdgpu/soc15_common.h b/drivers/gpu/drm/amd/amdgpu/soc15_common.h index 19e870c79896..c893c645a4b2 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15_common.h +++ b/drivers/gpu/drm/amd/amdgpu/soc15_common.h @@ -70,10 +70,9 @@ } \ } while (0) -#define AMDGPU_VIRT_SUPPORT_RLC_PRG_REG(a) (amdgpu_sriov_vf((a)) && !amdgpu_sriov_runtime((a))) #define WREG32_RLC(reg, value) \ do { \ - if (AMDGPU_VIRT_SUPPORT_RLC_PRG_REG(adev)) { \ + if (amdgpu_sriov_fullaccess(adev)) { \ uint32_t i = 0; \ uint32_t retries = 50000; \ uint32_t r0 = adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG0_BASE_IDX] + mmSCRATCH_REG0; \ @@ -98,7 +97,7 @@ #define WREG32_SOC15_RLC_SHADOW(ip, inst, reg, value) \ do { \ uint32_t target_reg = adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg;\ - if (AMDGPU_VIRT_SUPPORT_RLC_PRG_REG(adev)) { \ + if (amdgpu_sriov_fullaccess(adev)) { \ uint32_t r2 = adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG1_BASE_IDX] + mmSCRATCH_REG2; \ uint32_t r3 = adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG1_BASE_IDX] + mmSCRATCH_REG3; \ uint32_t grbm_cntl = adev->reg_offset[GC_HWIP][0][mmGRBM_GFX_CNTL_BASE_IDX] + mmGRBM_GFX_CNTL; \ diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v6_1.c b/drivers/gpu/drm/amd/amdgpu/umc_v6_1.c index 793bf70e64b1..14d346321a5f 100644 --- a/drivers/gpu/drm/amd/amdgpu/umc_v6_1.c +++ b/drivers/gpu/drm/amd/amdgpu/umc_v6_1.c @@ -186,6 +186,10 @@ static void umc_v6_1_query_ras_error_count(struct amdgpu_device *adev, if (rsmu_umc_index_state) umc_v6_1_disable_umc_index_mode(adev); + if ((adev->asic_type == CHIP_ARCTURUS) && + amdgpu_dpm_set_df_cstate(adev, DF_CSTATE_DISALLOW)) + DRM_WARN("Fail to disable DF-Cstate.\n"); + LOOP_UMC_INST_AND_CH(umc_inst, ch_inst) { umc_reg_offset = get_umc_6_reg_offset(adev, umc_inst, @@ -199,6 +203,10 @@ static void umc_v6_1_query_ras_error_count(struct amdgpu_device *adev, &(err_data->ue_count)); } + if ((adev->asic_type == CHIP_ARCTURUS) && + amdgpu_dpm_set_df_cstate(adev, DF_CSTATE_ALLOW)) + DRM_WARN("Fail to enable DF-Cstate\n"); + if (rsmu_umc_index_state) umc_v6_1_enable_umc_index_mode(adev); } @@ -228,7 +236,11 @@ static void umc_v6_1_query_error_address(struct amdgpu_device *adev, SOC15_REG_OFFSET(UMC, 0, mmMCA_UMC_UMC0_MCUMC_ADDRT0); } - /* skip error address process if -ENOMEM */ + mc_umc_status = RREG64_PCIE((mc_umc_status_addr + umc_reg_offset) * 4); + + if (mc_umc_status == 0) + return; + if (!err_data->err_addr) { /* clear umc status */ WREG64_PCIE((mc_umc_status_addr + umc_reg_offset) * 4, 0x0ULL); @@ -236,7 +248,6 @@ static void umc_v6_1_query_error_address(struct amdgpu_device *adev, } err_rec = &err_data->err_addr[err_data->err_addr_cnt]; - mc_umc_status = RREG64_PCIE((mc_umc_status_addr + umc_reg_offset) * 4); /* calculate error address if ue/ce error is detected */ if (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1 && @@ -288,6 +299,10 @@ static void umc_v6_1_query_ras_error_address(struct amdgpu_device *adev, if (rsmu_umc_index_state) umc_v6_1_disable_umc_index_mode(adev); + if ((adev->asic_type == CHIP_ARCTURUS) && + amdgpu_dpm_set_df_cstate(adev, DF_CSTATE_DISALLOW)) + DRM_WARN("Fail to disable DF-Cstate.\n"); + LOOP_UMC_INST_AND_CH(umc_inst, ch_inst) { umc_reg_offset = get_umc_6_reg_offset(adev, umc_inst, @@ -300,6 +315,10 @@ static void umc_v6_1_query_ras_error_address(struct amdgpu_device *adev, umc_inst); } + if ((adev->asic_type == CHIP_ARCTURUS) && + amdgpu_dpm_set_df_cstate(adev, DF_CSTATE_ALLOW)) + DRM_WARN("Fail to enable DF-Cstate\n"); + if (rsmu_umc_index_state) umc_v6_1_enable_umc_index_mode(adev); } diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c index 71f61afdc655..09b0572b838d 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c @@ -1352,7 +1352,7 @@ static int vcn_v1_0_set_clockgating_state(void *handle, if (enable) { /* wait for STATUS to clear */ - if (vcn_v1_0_is_idle(handle)) + if (!vcn_v1_0_is_idle(handle)) return -EBUSY; vcn_v1_0_enable_clock_gating(adev); } else { diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c index c387c81f8695..ec8091a661df 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c @@ -29,6 +29,7 @@ #include "soc15d.h" #include "amdgpu_pm.h" #include "amdgpu_psp.h" +#include "mmsch_v2_0.h" #include "vcn/vcn_2_0_0_offset.h" #include "vcn/vcn_2_0_0_sh_mask.h" @@ -54,7 +55,7 @@ static int vcn_v2_0_set_powergating_state(void *handle, enum amd_powergating_state state); static int vcn_v2_0_pause_dpg_mode(struct amdgpu_device *adev, int inst_idx, struct dpg_pause_state *new_state); - +static int vcn_v2_0_start_sriov(struct amdgpu_device *adev); /** * vcn_v2_0_early_init - set function pointers * @@ -67,7 +68,10 @@ static int vcn_v2_0_early_init(void *handle) struct amdgpu_device *adev = (struct amdgpu_device *)handle; adev->vcn.num_vcn_inst = 1; - adev->vcn.num_enc_rings = 2; + if (amdgpu_sriov_vf(adev)) + adev->vcn.num_enc_rings = 1; + else + adev->vcn.num_enc_rings = 2; vcn_v2_0_set_dec_ring_funcs(adev); vcn_v2_0_set_enc_ring_funcs(adev); @@ -154,7 +158,10 @@ static int vcn_v2_0_sw_init(void *handle) for (i = 0; i < adev->vcn.num_enc_rings; ++i) { ring = &adev->vcn.inst->ring_enc[i]; ring->use_doorbell = true; - ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 2 + i; + if (!amdgpu_sriov_vf(adev)) + ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 2 + i; + else + ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 1 + i; sprintf(ring->name, "vcn_enc%d", i); r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst->irq, 0); if (r) @@ -163,6 +170,10 @@ static int vcn_v2_0_sw_init(void *handle) adev->vcn.pause_dpg_mode = vcn_v2_0_pause_dpg_mode; + r = amdgpu_virt_alloc_mm_table(adev); + if (r) + return r; + return 0; } @@ -178,6 +189,8 @@ static int vcn_v2_0_sw_fini(void *handle) int r; struct amdgpu_device *adev = (struct amdgpu_device *)handle; + amdgpu_virt_free_mm_table(adev); + r = amdgpu_vcn_suspend(adev); if (r) return r; @@ -203,6 +216,9 @@ static int vcn_v2_0_hw_init(void *handle) adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell, ring->doorbell_index, 0); + if (amdgpu_sriov_vf(adev)) + vcn_v2_0_start_sriov(adev); + r = amdgpu_ring_test_helper(ring); if (r) goto done; @@ -304,6 +320,9 @@ static void vcn_v2_0_mc_resume(struct amdgpu_device *adev) uint32_t size = AMDGPU_GPU_PAGE_ALIGN(adev->vcn.fw->size + 4); uint32_t offset; + if (amdgpu_sriov_vf(adev)) + return; + /* cache window 0: fw */ if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) { WREG32_SOC15(UVD, 0, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW, @@ -448,6 +467,9 @@ static void vcn_v2_0_disable_clock_gating(struct amdgpu_device *adev) { uint32_t data; + if (amdgpu_sriov_vf(adev)) + return; + /* UVD disable CGC */ data = RREG32_SOC15(VCN, 0, mmUVD_CGC_CTRL); if (adev->cg_flags & AMD_CG_SUPPORT_VCN_MGCG) @@ -606,6 +628,9 @@ static void vcn_v2_0_enable_clock_gating(struct amdgpu_device *adev) { uint32_t data = 0; + if (amdgpu_sriov_vf(adev)) + return; + /* enable UVD CGC */ data = RREG32_SOC15(VCN, 0, mmUVD_CGC_CTRL); if (adev->cg_flags & AMD_CG_SUPPORT_VCN_MGCG) @@ -658,6 +683,9 @@ static void vcn_v2_0_disable_static_power_gating(struct amdgpu_device *adev) uint32_t data = 0; int ret; + if (amdgpu_sriov_vf(adev)) + return; + if (adev->pg_flags & AMD_PG_SUPPORT_VCN) { data = (1 << UVD_PGFSM_CONFIG__UVDM_PWR_CONFIG__SHIFT | 1 << UVD_PGFSM_CONFIG__UVDU_PWR_CONFIG__SHIFT @@ -705,6 +733,9 @@ static void vcn_v2_0_enable_static_power_gating(struct amdgpu_device *adev) uint32_t data = 0; int ret; + if (amdgpu_sriov_vf(adev)) + return; + if (adev->pg_flags & AMD_PG_SUPPORT_VCN) { /* Before power off, this indicator has to be turned on */ data = RREG32_SOC15(VCN, 0, mmUVD_POWER_STATUS); @@ -1215,9 +1246,12 @@ static int vcn_v2_0_set_clockgating_state(void *handle, struct amdgpu_device *adev = (struct amdgpu_device *)handle; bool enable = (state == AMD_CG_STATE_GATE); + if (amdgpu_sriov_vf(adev)) + return 0; + if (enable) { /* wait for STATUS to clear */ - if (vcn_v2_0_is_idle(handle)) + if (!vcn_v2_0_is_idle(handle)) return -EBUSY; vcn_v2_0_enable_clock_gating(adev); } else { @@ -1631,6 +1665,9 @@ int vcn_v2_0_dec_ring_test_ring(struct amdgpu_ring *ring) unsigned i; int r; + if (amdgpu_sriov_vf(adev)) + return 0; + WREG32(adev->vcn.inst[ring->me].external.scratch9, 0xCAFEDEAD); r = amdgpu_ring_alloc(ring, 4); if (r) @@ -1667,6 +1704,11 @@ static int vcn_v2_0_set_powergating_state(void *handle, int ret; struct amdgpu_device *adev = (struct amdgpu_device *)handle; + if (amdgpu_sriov_vf(adev)) { + adev->vcn.cur_state = AMD_PG_STATE_UNGATE; + return 0; + } + if (state == adev->vcn.cur_state) return 0; @@ -1680,6 +1722,215 @@ static int vcn_v2_0_set_powergating_state(void *handle, return ret; } +static int vcn_v2_0_start_mmsch(struct amdgpu_device *adev, + struct amdgpu_mm_table *table) +{ + uint32_t data = 0, loop; + uint64_t addr = table->gpu_addr; + struct mmsch_v2_0_init_header *header; + uint32_t size; + int i; + + header = (struct mmsch_v2_0_init_header *)table->cpu_addr; + size = header->header_size + header->vcn_table_size; + + /* 1, write to vce_mmsch_vf_ctx_addr_lo/hi register with GPU mc addr + * of memory descriptor location + */ + WREG32_SOC15(UVD, 0, mmMMSCH_VF_CTX_ADDR_LO, lower_32_bits(addr)); + WREG32_SOC15(UVD, 0, mmMMSCH_VF_CTX_ADDR_HI, upper_32_bits(addr)); + + /* 2, update vmid of descriptor */ + data = RREG32_SOC15(UVD, 0, mmMMSCH_VF_VMID); + data &= ~MMSCH_VF_VMID__VF_CTX_VMID_MASK; + /* use domain0 for MM scheduler */ + data |= (0 << MMSCH_VF_VMID__VF_CTX_VMID__SHIFT); + WREG32_SOC15(UVD, 0, mmMMSCH_VF_VMID, data); + + /* 3, notify mmsch about the size of this descriptor */ + WREG32_SOC15(UVD, 0, mmMMSCH_VF_CTX_SIZE, size); + + /* 4, set resp to zero */ + WREG32_SOC15(UVD, 0, mmMMSCH_VF_MAILBOX_RESP, 0); + + adev->vcn.inst->ring_dec.wptr = 0; + adev->vcn.inst->ring_dec.wptr_old = 0; + vcn_v2_0_dec_ring_set_wptr(&adev->vcn.inst->ring_dec); + + for (i = 0; i < adev->vcn.num_enc_rings; ++i) { + adev->vcn.inst->ring_enc[i].wptr = 0; + adev->vcn.inst->ring_enc[i].wptr_old = 0; + vcn_v2_0_enc_ring_set_wptr(&adev->vcn.inst->ring_enc[i]); + } + + /* 5, kick off the initialization and wait until + * VCE_MMSCH_VF_MAILBOX_RESP becomes non-zero + */ + WREG32_SOC15(UVD, 0, mmMMSCH_VF_MAILBOX_HOST, 0x10000001); + + data = RREG32_SOC15(UVD, 0, mmMMSCH_VF_MAILBOX_RESP); + loop = 1000; + while ((data & 0x10000002) != 0x10000002) { + udelay(10); + data = RREG32_SOC15(UVD, 0, mmMMSCH_VF_MAILBOX_RESP); + loop--; + if (!loop) + break; + } + + if (!loop) { + DRM_ERROR("failed to init MMSCH, " \ + "mmMMSCH_VF_MAILBOX_RESP = 0x%08x\n", data); + return -EBUSY; + } + + return 0; +} + +static int vcn_v2_0_start_sriov(struct amdgpu_device *adev) +{ + int r; + uint32_t tmp; + struct amdgpu_ring *ring; + uint32_t offset, size; + uint32_t table_size = 0; + struct mmsch_v2_0_cmd_direct_write direct_wt = { {0} }; + struct mmsch_v2_0_cmd_direct_read_modify_write direct_rd_mod_wt = { {0} }; + struct mmsch_v2_0_cmd_direct_polling direct_poll = { {0} }; + struct mmsch_v2_0_cmd_end end = { {0} }; + struct mmsch_v2_0_init_header *header; + uint32_t *init_table = adev->virt.mm_table.cpu_addr; + uint8_t i = 0; + + header = (struct mmsch_v2_0_init_header *)init_table; + direct_wt.cmd_header.command_type = MMSCH_COMMAND__DIRECT_REG_WRITE; + direct_rd_mod_wt.cmd_header.command_type = + MMSCH_COMMAND__DIRECT_REG_READ_MODIFY_WRITE; + direct_poll.cmd_header.command_type = + MMSCH_COMMAND__DIRECT_REG_POLLING; + end.cmd_header.command_type = MMSCH_COMMAND__END; + + if (header->vcn_table_offset == 0 && header->vcn_table_size == 0) { + header->version = MMSCH_VERSION; + header->header_size = sizeof(struct mmsch_v2_0_init_header) >> 2; + + header->vcn_table_offset = header->header_size; + + init_table += header->vcn_table_offset; + + size = AMDGPU_GPU_PAGE_ALIGN(adev->vcn.fw->size + 4); + + MMSCH_V2_0_INSERT_DIRECT_RD_MOD_WT( + SOC15_REG_OFFSET(UVD, i, mmUVD_STATUS), + 0xFFFFFFFF, 0x00000004); + + /* mc resume*/ + if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) { + tmp = AMDGPU_UCODE_ID_VCN; + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, + mmUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW), + adev->firmware.ucode[tmp].tmr_mc_addr_lo); + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, + mmUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH), + adev->firmware.ucode[tmp].tmr_mc_addr_hi); + offset = 0; + } else { + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, + mmUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW), + lower_32_bits(adev->vcn.inst->gpu_addr)); + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, + mmUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH), + upper_32_bits(adev->vcn.inst->gpu_addr)); + offset = size; + } + + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, mmUVD_VCPU_CACHE_OFFSET0), + 0); + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, mmUVD_VCPU_CACHE_SIZE0), + size); + + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, + mmUVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW), + lower_32_bits(adev->vcn.inst->gpu_addr + offset)); + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, + mmUVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH), + upper_32_bits(adev->vcn.inst->gpu_addr + offset)); + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, mmUVD_VCPU_CACHE_OFFSET1), + 0); + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, mmUVD_VCPU_CACHE_SIZE1), + AMDGPU_VCN_STACK_SIZE); + + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, + mmUVD_LMI_VCPU_CACHE2_64BIT_BAR_LOW), + lower_32_bits(adev->vcn.inst->gpu_addr + offset + + AMDGPU_VCN_STACK_SIZE)); + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, + mmUVD_LMI_VCPU_CACHE2_64BIT_BAR_HIGH), + upper_32_bits(adev->vcn.inst->gpu_addr + offset + + AMDGPU_VCN_STACK_SIZE)); + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, mmUVD_VCPU_CACHE_OFFSET2), + 0); + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, mmUVD_VCPU_CACHE_SIZE2), + AMDGPU_VCN_CONTEXT_SIZE); + + for (r = 0; r < adev->vcn.num_enc_rings; ++r) { + ring = &adev->vcn.inst->ring_enc[r]; + ring->wptr = 0; + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, mmUVD_RB_BASE_LO), + lower_32_bits(ring->gpu_addr)); + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, mmUVD_RB_BASE_HI), + upper_32_bits(ring->gpu_addr)); + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, mmUVD_RB_SIZE), + ring->ring_size / 4); + } + + ring = &adev->vcn.inst->ring_dec; + ring->wptr = 0; + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, + mmUVD_LMI_RBC_RB_64BIT_BAR_LOW), + lower_32_bits(ring->gpu_addr)); + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, + mmUVD_LMI_RBC_RB_64BIT_BAR_HIGH), + upper_32_bits(ring->gpu_addr)); + /* force RBC into idle state */ + tmp = order_base_2(ring->ring_size); + tmp = REG_SET_FIELD(0, UVD_RBC_RB_CNTL, RB_BUFSZ, tmp); + tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_BLKSZ, 1); + tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_NO_FETCH, 1); + tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_NO_UPDATE, 1); + tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_RPTR_WR_EN, 1); + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, mmUVD_RBC_RB_CNTL), tmp); + + /* add end packet */ + tmp = sizeof(struct mmsch_v2_0_cmd_end); + memcpy((void *)init_table, &end, tmp); + table_size += (tmp / 4); + header->vcn_table_size = table_size; + + } + return vcn_v2_0_start_mmsch(adev, &adev->virt.mm_table); +} + static const struct amd_ip_funcs vcn_v2_0_ip_funcs = { .name = "vcn_v2_0", .early_init = vcn_v2_0_early_init, diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c index 2d64ba1adf99..c6363f5ad564 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c @@ -74,29 +74,30 @@ static int amdgpu_ih_clientid_vcns[] = { static int vcn_v2_5_early_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - if (adev->asic_type == CHIP_ARCTURUS) { - u32 harvest; - int i; - - adev->vcn.num_vcn_inst = VCN25_MAX_HW_INSTANCES_ARCTURUS; - for (i = 0; i < adev->vcn.num_vcn_inst; i++) { - harvest = RREG32_SOC15(UVD, i, mmCC_UVD_HARVESTING); - if (harvest & CC_UVD_HARVESTING__UVD_DISABLE_MASK) - adev->vcn.harvest_config |= 1 << i; - } - - if (adev->vcn.harvest_config == (AMDGPU_VCN_HARVEST_VCN0 | - AMDGPU_VCN_HARVEST_VCN1)) - /* both instances are harvested, disable the block */ - return -ENOENT; - } else - adev->vcn.num_vcn_inst = 1; if (amdgpu_sriov_vf(adev)) { adev->vcn.num_vcn_inst = 2; adev->vcn.harvest_config = 0; adev->vcn.num_enc_rings = 1; } else { + if (adev->asic_type == CHIP_ARCTURUS) { + u32 harvest; + int i; + + adev->vcn.num_vcn_inst = VCN25_MAX_HW_INSTANCES_ARCTURUS; + for (i = 0; i < adev->vcn.num_vcn_inst; i++) { + harvest = RREG32_SOC15(UVD, i, mmCC_UVD_HARVESTING); + if (harvest & CC_UVD_HARVESTING__UVD_DISABLE_MASK) + adev->vcn.harvest_config |= 1 << i; + } + + if (adev->vcn.harvest_config == (AMDGPU_VCN_HARVEST_VCN0 | + AMDGPU_VCN_HARVEST_VCN1)) + /* both instances are harvested, disable the block */ + return -ENOENT; + } else + adev->vcn.num_vcn_inst = 1; + adev->vcn.num_enc_rings = 2; } @@ -1672,7 +1673,7 @@ static int vcn_v2_5_set_clockgating_state(void *handle, return 0; if (enable) { - if (vcn_v2_5_is_idle(handle)) + if (!vcn_v2_5_is_idle(handle)) return -EBUSY; vcn_v2_5_enable_clock_gating(adev); } else { diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c index 3f0300e53727..0ec5f25adf56 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c @@ -127,6 +127,8 @@ static int kfd_open(struct inode *inode, struct file *filep) return PTR_ERR(process); if (kfd_is_locked()) { + dev_dbg(kfd_device, "kfd is locked!\n" + "process %d unreferenced", process->pasid); kfd_unref_process(process); return -EAGAIN; } @@ -1167,7 +1169,7 @@ static int kfd_ioctl_get_tile_config(struct file *filep, if (!dev) return -EINVAL; - dev->kfd2kgd->get_tile_config(dev->kgd, &config); + amdgpu_amdkfd_get_tile_config(dev->kgd, &config); args->gb_addr_config = config.gb_addr_config; args->num_banks = config.num_banks; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c index 2a9e40131735..05bc6d96ec52 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c @@ -648,6 +648,9 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd, if (kfd->kfd2kgd->get_hive_id) kfd->hive_id = kfd->kfd2kgd->get_hive_id(kfd->kgd); + if (kfd->kfd2kgd->get_unique_id) + kfd->unique_id = kfd->kfd2kgd->get_unique_id(kfd->kgd); + if (kfd_interrupt_init(kfd)) { dev_err(kfd_device, "Error initializing interrupts\n"); goto kfd_interrupt_error; @@ -710,7 +713,7 @@ out: void kgd2kfd_device_exit(struct kfd_dev *kfd) { if (kfd->init_complete) { - kgd2kfd_suspend(kfd); + kgd2kfd_suspend(kfd, false); device_queue_manager_uninit(kfd->dqm); kfd_interrupt_exit(kfd); kfd_topology_remove_device(kfd); @@ -731,7 +734,7 @@ int kgd2kfd_pre_reset(struct kfd_dev *kfd) kfd->dqm->ops.pre_reset(kfd->dqm); - kgd2kfd_suspend(kfd); + kgd2kfd_suspend(kfd, false); kfd_signal_reset_event(kfd); return 0; @@ -765,21 +768,23 @@ bool kfd_is_locked(void) return (atomic_read(&kfd_locked) > 0); } -void kgd2kfd_suspend(struct kfd_dev *kfd) +void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm) { if (!kfd->init_complete) return; - /* For first KFD device suspend all the KFD processes */ - if (atomic_inc_return(&kfd_locked) == 1) - kfd_suspend_all_processes(); + /* for runtime suspend, skip locking kfd */ + if (!run_pm) { + /* For first KFD device suspend all the KFD processes */ + if (atomic_inc_return(&kfd_locked) == 1) + kfd_suspend_all_processes(); + } kfd->dqm->ops.stop(kfd->dqm); - kfd_iommu_suspend(kfd); } -int kgd2kfd_resume(struct kfd_dev *kfd) +int kgd2kfd_resume(struct kfd_dev *kfd, bool run_pm) { int ret, count; @@ -790,10 +795,13 @@ int kgd2kfd_resume(struct kfd_dev *kfd) if (ret) return ret; - count = atomic_dec_return(&kfd_locked); - WARN_ONCE(count < 0, "KFD suspend / resume ref. error"); - if (count == 0) - ret = kfd_resume_all_processes(); + /* for runtime resume, skip unlocking kfd */ + if (!run_pm) { + count = atomic_dec_return(&kfd_locked); + WARN_ONCE(count < 0, "KFD suspend / resume ref. error"); + if (count == 0) + ret = kfd_resume_all_processes(); + } return ret; } @@ -1104,9 +1112,9 @@ kfd_gtt_out: return 0; kfd_gtt_no_free_chunk: - pr_debug("Allocation failed with mem_obj = %p\n", mem_obj); + pr_debug("Allocation failed with mem_obj = %p\n", *mem_obj); mutex_unlock(&kfd->gtt_sa_lock); - kfree(mem_obj); + kfree(*mem_obj); return -ENOMEM; } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c index 80d22bf702e8..77ea0f0cb163 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c @@ -78,14 +78,14 @@ static bool is_pipe_enabled(struct device_queue_manager *dqm, int mec, int pipe) /* queue is available for KFD usage if bit is 1 */ for (i = 0; i < dqm->dev->shared_resources.num_queue_per_pipe; ++i) if (test_bit(pipe_offset + i, - dqm->dev->shared_resources.queue_bitmap)) + dqm->dev->shared_resources.cp_queue_bitmap)) return true; return false; } -unsigned int get_queues_num(struct device_queue_manager *dqm) +unsigned int get_cp_queues_num(struct device_queue_manager *dqm) { - return bitmap_weight(dqm->dev->shared_resources.queue_bitmap, + return bitmap_weight(dqm->dev->shared_resources.cp_queue_bitmap, KGD_MAX_QUEUES); } @@ -109,6 +109,11 @@ static unsigned int get_num_xgmi_sdma_engines(struct device_queue_manager *dqm) return dqm->dev->device_info->num_xgmi_sdma_engines; } +static unsigned int get_num_all_sdma_engines(struct device_queue_manager *dqm) +{ + return get_num_sdma_engines(dqm) + get_num_xgmi_sdma_engines(dqm); +} + unsigned int get_num_sdma_queues(struct device_queue_manager *dqm) { return dqm->dev->device_info->num_sdma_engines @@ -132,6 +137,22 @@ void program_sh_mem_settings(struct device_queue_manager *dqm, qpd->sh_mem_bases); } +void increment_queue_count(struct device_queue_manager *dqm, + enum kfd_queue_type type) +{ + dqm->active_queue_count++; + if (type == KFD_QUEUE_TYPE_COMPUTE || type == KFD_QUEUE_TYPE_DIQ) + dqm->active_cp_queue_count++; +} + +void decrement_queue_count(struct device_queue_manager *dqm, + enum kfd_queue_type type) +{ + dqm->active_queue_count--; + if (type == KFD_QUEUE_TYPE_COMPUTE || type == KFD_QUEUE_TYPE_DIQ) + dqm->active_cp_queue_count--; +} + static int allocate_doorbell(struct qcm_process_device *qpd, struct queue *q) { struct kfd_dev *dev = qpd->dqm->dev; @@ -281,8 +302,6 @@ static int create_queue_nocpsch(struct device_queue_manager *dqm, struct mqd_manager *mqd_mgr; int retval; - print_queue(q); - dqm_lock(dqm); if (dqm->total_queue_count >= max_num_of_queues_per_device) { @@ -359,12 +378,7 @@ add_queue_to_list: list_add(&q->list, &qpd->queues_list); qpd->queue_count++; if (q->properties.is_active) - dqm->queue_count++; - - if (q->properties.type == KFD_QUEUE_TYPE_SDMA) - dqm->sdma_queue_count++; - else if (q->properties.type == KFD_QUEUE_TYPE_SDMA_XGMI) - dqm->xgmi_sdma_queue_count++; + increment_queue_count(dqm, q->properties.type); /* * Unconditionally increment this counter, regardless of the queue's @@ -446,15 +460,13 @@ static int destroy_queue_nocpsch_locked(struct device_queue_manager *dqm, mqd_mgr = dqm->mqd_mgrs[get_mqd_type_from_queue_type( q->properties.type)]; - if (q->properties.type == KFD_QUEUE_TYPE_COMPUTE) { + if (q->properties.type == KFD_QUEUE_TYPE_COMPUTE) deallocate_hqd(dqm, q); - } else if (q->properties.type == KFD_QUEUE_TYPE_SDMA) { - dqm->sdma_queue_count--; + else if (q->properties.type == KFD_QUEUE_TYPE_SDMA) deallocate_sdma_queue(dqm, q); - } else if (q->properties.type == KFD_QUEUE_TYPE_SDMA_XGMI) { - dqm->xgmi_sdma_queue_count--; + else if (q->properties.type == KFD_QUEUE_TYPE_SDMA_XGMI) deallocate_sdma_queue(dqm, q); - } else { + else { pr_debug("q->properties.type %d is invalid\n", q->properties.type); return -EINVAL; @@ -494,7 +506,7 @@ static int destroy_queue_nocpsch_locked(struct device_queue_manager *dqm, } qpd->queue_count--; if (q->properties.is_active) - dqm->queue_count--; + decrement_queue_count(dqm, q->properties.type); return retval; } @@ -563,13 +575,13 @@ static int update_queue(struct device_queue_manager *dqm, struct queue *q) /* * check active state vs. the previous state and modify * counter accordingly. map_queues_cpsch uses the - * dqm->queue_count to determine whether a new runlist must be + * dqm->active_queue_count to determine whether a new runlist must be * uploaded. */ if (q->properties.is_active && !prev_active) - dqm->queue_count++; + increment_queue_count(dqm, q->properties.type); else if (!q->properties.is_active && prev_active) - dqm->queue_count--; + decrement_queue_count(dqm, q->properties.type); if (dqm->sched_policy != KFD_SCHED_POLICY_NO_HWS) retval = map_queues_cpsch(dqm); @@ -618,7 +630,7 @@ static int evict_process_queues_nocpsch(struct device_queue_manager *dqm, mqd_mgr = dqm->mqd_mgrs[get_mqd_type_from_queue_type( q->properties.type)]; q->properties.is_active = false; - dqm->queue_count--; + decrement_queue_count(dqm, q->properties.type); if (WARN_ONCE(!dqm->sched_running, "Evict when stopped\n")) continue; @@ -662,7 +674,7 @@ static int evict_process_queues_cpsch(struct device_queue_manager *dqm, continue; q->properties.is_active = false; - dqm->queue_count--; + decrement_queue_count(dqm, q->properties.type); } retval = execute_queues_cpsch(dqm, qpd->is_debug ? @@ -731,7 +743,7 @@ static int restore_process_queues_nocpsch(struct device_queue_manager *dqm, mqd_mgr = dqm->mqd_mgrs[get_mqd_type_from_queue_type( q->properties.type)]; q->properties.is_active = true; - dqm->queue_count++; + increment_queue_count(dqm, q->properties.type); if (WARN_ONCE(!dqm->sched_running, "Restore when stopped\n")) continue; @@ -786,7 +798,7 @@ static int restore_process_queues_cpsch(struct device_queue_manager *dqm, continue; q->properties.is_active = true; - dqm->queue_count++; + increment_queue_count(dqm, q->properties.type); } retval = execute_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0); @@ -899,16 +911,15 @@ static int initialize_nocpsch(struct device_queue_manager *dqm) mutex_init(&dqm->lock_hidden); INIT_LIST_HEAD(&dqm->queues); - dqm->queue_count = dqm->next_pipe_to_allocate = 0; - dqm->sdma_queue_count = 0; - dqm->xgmi_sdma_queue_count = 0; + dqm->active_queue_count = dqm->next_pipe_to_allocate = 0; + dqm->active_cp_queue_count = 0; for (pipe = 0; pipe < get_pipes_per_mec(dqm); pipe++) { int pipe_offset = pipe * get_queues_per_pipe(dqm); for (queue = 0; queue < get_queues_per_pipe(dqm); queue++) if (test_bit(pipe_offset + queue, - dqm->dev->shared_resources.queue_bitmap)) + dqm->dev->shared_resources.cp_queue_bitmap)) dqm->allocated_queues[pipe] |= 1 << queue; } @@ -924,7 +935,7 @@ static void uninitialize(struct device_queue_manager *dqm) { int i; - WARN_ON(dqm->queue_count > 0 || dqm->processes_count > 0); + WARN_ON(dqm->active_queue_count > 0 || dqm->processes_count > 0); kfree(dqm->allocated_queues); for (i = 0 ; i < KFD_MQD_TYPE_MAX ; i++) @@ -966,8 +977,11 @@ static int allocate_sdma_queue(struct device_queue_manager *dqm, int bit; if (q->properties.type == KFD_QUEUE_TYPE_SDMA) { - if (dqm->sdma_bitmap == 0) + if (dqm->sdma_bitmap == 0) { + pr_err("No more SDMA queue to allocate\n"); return -ENOMEM; + } + bit = __ffs64(dqm->sdma_bitmap); dqm->sdma_bitmap &= ~(1ULL << bit); q->sdma_id = bit; @@ -976,8 +990,10 @@ static int allocate_sdma_queue(struct device_queue_manager *dqm, q->properties.sdma_queue_id = q->sdma_id / get_num_sdma_engines(dqm); } else if (q->properties.type == KFD_QUEUE_TYPE_SDMA_XGMI) { - if (dqm->xgmi_sdma_bitmap == 0) + if (dqm->xgmi_sdma_bitmap == 0) { + pr_err("No more XGMI SDMA queue to allocate\n"); return -ENOMEM; + } bit = __ffs64(dqm->xgmi_sdma_bitmap); dqm->xgmi_sdma_bitmap &= ~(1ULL << bit); q->sdma_id = bit; @@ -1029,7 +1045,7 @@ static int set_sched_resources(struct device_queue_manager *dqm) mec = (i / dqm->dev->shared_resources.num_queue_per_pipe) / dqm->dev->shared_resources.num_pipe_per_mec; - if (!test_bit(i, dqm->dev->shared_resources.queue_bitmap)) + if (!test_bit(i, dqm->dev->shared_resources.cp_queue_bitmap)) continue; /* only acquire queues from the first MEC */ @@ -1064,9 +1080,9 @@ static int initialize_cpsch(struct device_queue_manager *dqm) mutex_init(&dqm->lock_hidden); INIT_LIST_HEAD(&dqm->queues); - dqm->queue_count = dqm->processes_count = 0; - dqm->sdma_queue_count = 0; - dqm->xgmi_sdma_queue_count = 0; + dqm->active_queue_count = dqm->processes_count = 0; + dqm->active_cp_queue_count = 0; + dqm->active_runlist = false; dqm->sdma_bitmap = ~0ULL >> (64 - get_num_sdma_queues(dqm)); dqm->xgmi_sdma_bitmap = ~0ULL >> (64 - get_num_xgmi_sdma_queues(dqm)); @@ -1158,7 +1174,7 @@ static int create_kernel_queue_cpsch(struct device_queue_manager *dqm, dqm->total_queue_count); list_add(&kq->list, &qpd->priv_queue_list); - dqm->queue_count++; + increment_queue_count(dqm, kq->queue->properties.type); qpd->is_debug = true; execute_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0); dqm_unlock(dqm); @@ -1172,7 +1188,7 @@ static void destroy_kernel_queue_cpsch(struct device_queue_manager *dqm, { dqm_lock(dqm); list_del(&kq->list); - dqm->queue_count--; + decrement_queue_count(dqm, kq->queue->properties.type); qpd->is_debug = false; execute_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_ALL_QUEUES, 0); /* @@ -1238,13 +1254,9 @@ static int create_queue_cpsch(struct device_queue_manager *dqm, struct queue *q, list_add(&q->list, &qpd->queues_list); qpd->queue_count++; - if (q->properties.type == KFD_QUEUE_TYPE_SDMA) - dqm->sdma_queue_count++; - else if (q->properties.type == KFD_QUEUE_TYPE_SDMA_XGMI) - dqm->xgmi_sdma_queue_count++; - if (q->properties.is_active) { - dqm->queue_count++; + increment_queue_count(dqm, q->properties.type); + retval = execute_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0); } @@ -1298,20 +1310,6 @@ int amdkfd_fence_wait_timeout(unsigned int *fence_addr, return 0; } -static int unmap_sdma_queues(struct device_queue_manager *dqm) -{ - int i, retval = 0; - - for (i = 0; i < dqm->dev->device_info->num_sdma_engines + - dqm->dev->device_info->num_xgmi_sdma_engines; i++) { - retval = pm_send_unmap_queue(&dqm->packets, KFD_QUEUE_TYPE_SDMA, - KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0, false, i); - if (retval) - return retval; - } - return retval; -} - /* dqm->lock mutex has to be locked before calling this function */ static int map_queues_cpsch(struct device_queue_manager *dqm) { @@ -1319,7 +1317,7 @@ static int map_queues_cpsch(struct device_queue_manager *dqm) if (!dqm->sched_running) return 0; - if (dqm->queue_count <= 0 || dqm->processes_count <= 0) + if (dqm->active_queue_count <= 0 || dqm->processes_count <= 0) return 0; if (dqm->active_runlist) return 0; @@ -1349,12 +1347,6 @@ static int unmap_queues_cpsch(struct device_queue_manager *dqm, if (!dqm->active_runlist) return retval; - pr_debug("Before destroying queues, sdma queue count is : %u, xgmi sdma queue count is : %u\n", - dqm->sdma_queue_count, dqm->xgmi_sdma_queue_count); - - if (dqm->sdma_queue_count > 0 || dqm->xgmi_sdma_queue_count) - unmap_sdma_queues(dqm); - retval = pm_send_unmap_queue(&dqm->packets, KFD_QUEUE_TYPE_COMPUTE, filter, filter_param, false, 0); if (retval) @@ -1427,18 +1419,15 @@ static int destroy_queue_cpsch(struct device_queue_manager *dqm, deallocate_doorbell(qpd, q); - if (q->properties.type == KFD_QUEUE_TYPE_SDMA) { - dqm->sdma_queue_count--; + if (q->properties.type == KFD_QUEUE_TYPE_SDMA) deallocate_sdma_queue(dqm, q); - } else if (q->properties.type == KFD_QUEUE_TYPE_SDMA_XGMI) { - dqm->xgmi_sdma_queue_count--; + else if (q->properties.type == KFD_QUEUE_TYPE_SDMA_XGMI) deallocate_sdma_queue(dqm, q); - } list_del(&q->list); qpd->queue_count--; if (q->properties.is_active) { - dqm->queue_count--; + decrement_queue_count(dqm, q->properties.type); retval = execute_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0); if (retval == -ETIME) @@ -1648,7 +1637,7 @@ static int process_termination_cpsch(struct device_queue_manager *dqm, /* Clean all kernel queues */ list_for_each_entry_safe(kq, kq_next, &qpd->priv_queue_list, list) { list_del(&kq->list); - dqm->queue_count--; + decrement_queue_count(dqm, kq->queue->properties.type); qpd->is_debug = false; dqm->total_queue_count--; filter = KFD_UNMAP_QUEUES_FILTER_ALL_QUEUES; @@ -1656,16 +1645,13 @@ static int process_termination_cpsch(struct device_queue_manager *dqm, /* Clear all user mode queues */ list_for_each_entry(q, &qpd->queues_list, list) { - if (q->properties.type == KFD_QUEUE_TYPE_SDMA) { - dqm->sdma_queue_count--; + if (q->properties.type == KFD_QUEUE_TYPE_SDMA) deallocate_sdma_queue(dqm, q); - } else if (q->properties.type == KFD_QUEUE_TYPE_SDMA_XGMI) { - dqm->xgmi_sdma_queue_count--; + else if (q->properties.type == KFD_QUEUE_TYPE_SDMA_XGMI) deallocate_sdma_queue(dqm, q); - } if (q->properties.is_active) - dqm->queue_count--; + decrement_queue_count(dqm, q->properties.type); dqm->total_queue_count--; } @@ -1742,14 +1728,13 @@ static int allocate_hiq_sdma_mqd(struct device_queue_manager *dqm) struct kfd_dev *dev = dqm->dev; struct kfd_mem_obj *mem_obj = &dqm->hiq_sdma_mqd; uint32_t size = dqm->mqd_mgrs[KFD_MQD_TYPE_SDMA]->mqd_size * - (dev->device_info->num_sdma_engines + - dev->device_info->num_xgmi_sdma_engines) * + get_num_all_sdma_engines(dqm) * dev->device_info->num_sdma_queues_per_engine + dqm->mqd_mgrs[KFD_MQD_TYPE_HIQ]->mqd_size; retval = amdgpu_amdkfd_alloc_gtt_mem(dev->kgd, size, &(mem_obj->gtt_mem), &(mem_obj->gpu_addr), - (void *)&(mem_obj->cpu_ptr), true); + (void *)&(mem_obj->cpu_ptr), false); return retval; } @@ -1979,7 +1964,7 @@ int dqm_debugfs_hqds(struct seq_file *m, void *data) for (queue = 0; queue < get_queues_per_pipe(dqm); queue++) { if (!test_bit(pipe_offset + queue, - dqm->dev->shared_resources.queue_bitmap)) + dqm->dev->shared_resources.cp_queue_bitmap)) continue; r = dqm->dev->kfd2kgd->hqd_dump( @@ -1995,8 +1980,7 @@ int dqm_debugfs_hqds(struct seq_file *m, void *data) } } - for (pipe = 0; pipe < get_num_sdma_engines(dqm) + - get_num_xgmi_sdma_engines(dqm); pipe++) { + for (pipe = 0; pipe < get_num_all_sdma_engines(dqm); pipe++) { for (queue = 0; queue < dqm->dev->device_info->num_sdma_queues_per_engine; queue++) { diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h index 871d3b628d2d..50d919f814e9 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h @@ -180,9 +180,8 @@ struct device_queue_manager { struct list_head queues; unsigned int saved_flags; unsigned int processes_count; - unsigned int queue_count; - unsigned int sdma_queue_count; - unsigned int xgmi_sdma_queue_count; + unsigned int active_queue_count; + unsigned int active_cp_queue_count; unsigned int total_queue_count; unsigned int next_pipe_to_allocate; unsigned int *allocated_queues; @@ -219,7 +218,7 @@ void device_queue_manager_init_v10_navi10( struct device_queue_manager_asic_ops *asic_ops); void program_sh_mem_settings(struct device_queue_manager *dqm, struct qcm_process_device *qpd); -unsigned int get_queues_num(struct device_queue_manager *dqm); +unsigned int get_cp_queues_num(struct device_queue_manager *dqm); unsigned int get_queues_per_pipe(struct device_queue_manager *dqm); unsigned int get_pipes_per_mec(struct device_queue_manager *dqm); unsigned int get_num_sdma_queues(struct device_queue_manager *dqm); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_events.c b/drivers/gpu/drm/amd/amdkfd/kfd_events.c index 1f8365575b12..15476fca8fa6 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_events.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_events.c @@ -187,7 +187,7 @@ static int create_signal_event(struct file *devkfd, if (p->signal_mapped_size && p->signal_event_count == p->signal_mapped_size / 8) { if (!p->signal_event_limit_reached) { - pr_warn("Signal event wasn't created because limit was reached\n"); + pr_debug("Signal event wasn't created because limit was reached\n"); p->signal_event_limit_reached = true; } return -ENOSPC; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c b/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c index bb77b8890e77..78714f9a8b11 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c @@ -316,7 +316,7 @@ static void kfd_init_apertures_vi(struct kfd_process_device *pdd, uint8_t id) { /* * node id couldn't be 0 - the three MSB bits of - * aperture shoudn't be 0 + * aperture shouldn't be 0 */ pdd->lds_base = MAKE_LDS_APP_BASE_VI(); pdd->lds_limit = MAKE_LDS_APP_LIMIT(pdd->lds_base); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c index 436b7f518979..48cda3073b70 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c @@ -87,9 +87,21 @@ static struct kfd_mem_obj *allocate_mqd(struct kfd_dev *kfd, int retval; struct kfd_mem_obj *mqd_mem_obj = NULL; - /* From V9, for CWSR, the control stack is located on the next page - * boundary after the mqd, we will use the gtt allocation function - * instead of sub-allocation function. + /* For V9 only, due to a HW bug, the control stack of a user mode + * compute queue needs to be allocated just behind the page boundary + * of its regular MQD buffer. So we allocate an enlarged MQD buffer: + * the first page of the buffer serves as the regular MQD buffer + * purpose and the remaining is for control stack. Although the two + * parts are in the same buffer object, they need different memory + * types: MQD part needs UC (uncached) as usual, while control stack + * needs NC (non coherent), which is different from the UC type which + * is used when control stack is allocated in user space. + * + * Because of all those, we use the gtt allocation function instead + * of sub-allocation function for this enlarged MQD buffer. Moreover, + * in order to achieve two memory types in a single buffer object, we + * pass a special bo flag AMDGPU_GEM_CREATE_CP_MQD_GFX9 to instruct + * amdgpu memory functions to do so. */ if (kfd->cwsr_enabled && (q->type == KFD_QUEUE_TYPE_COMPUTE)) { mqd_mem_obj = kzalloc(sizeof(struct kfd_mem_obj), GFP_KERNEL); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c index dc406e6dee23..efdb75e7677b 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c @@ -47,9 +47,8 @@ static void pm_calc_rlib_size(struct packet_manager *pm, struct kfd_dev *dev = pm->dqm->dev; process_count = pm->dqm->processes_count; - queue_count = pm->dqm->queue_count; - compute_queue_count = queue_count - pm->dqm->sdma_queue_count - - pm->dqm->xgmi_sdma_queue_count; + queue_count = pm->dqm->active_queue_count; + compute_queue_count = pm->dqm->active_cp_queue_count; /* check if there is over subscription * Note: the arbitration between the number of VMIDs and @@ -62,7 +61,7 @@ static void pm_calc_rlib_size(struct packet_manager *pm, max_proc_per_quantum = dev->max_proc_per_quantum; if ((process_count > max_proc_per_quantum) || - compute_queue_count > get_queues_num(pm->dqm)) { + compute_queue_count > get_cp_queues_num(pm->dqm)) { *over_subscription = true; pr_debug("Over subscribed runlist\n"); } @@ -141,7 +140,7 @@ static int pm_create_runlist_ib(struct packet_manager *pm, pm->ib_size_bytes = alloc_size_bytes; pr_debug("Building runlist ib process count: %d queues count %d\n", - pm->dqm->processes_count, pm->dqm->queue_count); + pm->dqm->processes_count, pm->dqm->active_queue_count); /* build the run list ib packet */ list_for_each_entry(cur, queues, list) { diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h index 6af1b5881f43..4a3049841086 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h @@ -41,6 +41,7 @@ #include <drm/drm_drv.h> #include <drm/drm_device.h> #include <kgd_kfd_interface.h> +#include <linux/swap.h> #include "amd_shared.h" @@ -293,6 +294,9 @@ struct kfd_dev { /* xGMI */ uint64_t hive_id; + + /* UUID */ + uint64_t unique_id; bool pci_atomic_requested; @@ -502,6 +506,9 @@ struct queue { struct kfd_process *process; struct kfd_dev *device; void *gws; + + /* procfs */ + struct kobject kobj; }; /* @@ -646,6 +653,7 @@ struct kfd_process_device { * function. */ bool already_dequeued; + bool runtime_inuse; /* Is this process/pasid bound to this device? (amd_iommu_bind_pasid) */ enum kfd_pdd_bound bound; @@ -729,6 +737,7 @@ struct kfd_process { /* Kobj for our procfs */ struct kobject *kobj; + struct kobject *kobj_queues; struct attribute attr_pasid; }; @@ -835,6 +844,8 @@ extern struct device *kfd_device; /* KFD's procfs */ void kfd_procfs_init(void); void kfd_procfs_shutdown(void); +int kfd_procfs_add_queue(struct queue *q); +void kfd_procfs_del_queue(struct queue *q); /* Topology */ int kfd_topology_init(void); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c index 25b90f70aecd..fe0cd49d4ea7 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c @@ -31,6 +31,7 @@ #include <linux/compat.h> #include <linux/mman.h> #include <linux/file.h> +#include <linux/pm_runtime.h> #include "amdgpu_amdkfd.h" #include "amdgpu.h" @@ -132,6 +133,88 @@ void kfd_procfs_shutdown(void) } } +static ssize_t kfd_procfs_queue_show(struct kobject *kobj, + struct attribute *attr, char *buffer) +{ + struct queue *q = container_of(kobj, struct queue, kobj); + + if (!strcmp(attr->name, "size")) + return snprintf(buffer, PAGE_SIZE, "%llu", + q->properties.queue_size); + else if (!strcmp(attr->name, "type")) + return snprintf(buffer, PAGE_SIZE, "%d", q->properties.type); + else if (!strcmp(attr->name, "gpuid")) + return snprintf(buffer, PAGE_SIZE, "%u", q->device->id); + else + pr_err("Invalid attribute"); + + return 0; +} + +static struct attribute attr_queue_size = { + .name = "size", + .mode = KFD_SYSFS_FILE_MODE +}; + +static struct attribute attr_queue_type = { + .name = "type", + .mode = KFD_SYSFS_FILE_MODE +}; + +static struct attribute attr_queue_gpuid = { + .name = "gpuid", + .mode = KFD_SYSFS_FILE_MODE +}; + +static struct attribute *procfs_queue_attrs[] = { + &attr_queue_size, + &attr_queue_type, + &attr_queue_gpuid, + NULL +}; + +static const struct sysfs_ops procfs_queue_ops = { + .show = kfd_procfs_queue_show, +}; + +static struct kobj_type procfs_queue_type = { + .sysfs_ops = &procfs_queue_ops, + .default_attrs = procfs_queue_attrs, +}; + +int kfd_procfs_add_queue(struct queue *q) +{ + struct kfd_process *proc; + int ret; + + if (!q || !q->process) + return -EINVAL; + proc = q->process; + + /* Create proc/<pid>/queues/<queue id> folder */ + if (!proc->kobj_queues) + return -EFAULT; + ret = kobject_init_and_add(&q->kobj, &procfs_queue_type, + proc->kobj_queues, "%u", q->properties.queue_id); + if (ret < 0) { + pr_warn("Creating proc/<pid>/queues/%u failed", + q->properties.queue_id); + kobject_put(&q->kobj); + return ret; + } + + return 0; +} + +void kfd_procfs_del_queue(struct queue *q) +{ + if (!q) + return; + + kobject_del(&q->kobj); + kobject_put(&q->kobj); +} + int kfd_process_create_wq(void) { if (!kfd_process_wq) @@ -244,10 +327,10 @@ err_alloc_mem: static int kfd_process_device_reserve_ib_mem(struct kfd_process_device *pdd) { struct qcm_process_device *qpd = &pdd->qpd; - uint32_t flags = ALLOC_MEM_FLAGS_GTT | - ALLOC_MEM_FLAGS_NO_SUBSTITUTE | - ALLOC_MEM_FLAGS_WRITABLE | - ALLOC_MEM_FLAGS_EXECUTABLE; + uint32_t flags = KFD_IOC_ALLOC_MEM_FLAGS_GTT | + KFD_IOC_ALLOC_MEM_FLAGS_NO_SUBSTITUTE | + KFD_IOC_ALLOC_MEM_FLAGS_WRITABLE | + KFD_IOC_ALLOC_MEM_FLAGS_EXECUTABLE; void *kaddr; int ret; @@ -323,6 +406,11 @@ struct kfd_process *kfd_create_process(struct file *filep) if (ret) pr_warn("Creating pasid for pid %d failed", (int)process->lead_thread->pid); + + process->kobj_queues = kobject_create_and_add("queues", + process->kobj); + if (!process->kobj_queues) + pr_warn("Creating KFD proc/queues folder failed"); } out: if (!IS_ERR(process)) @@ -440,6 +528,16 @@ static void kfd_process_destroy_pdds(struct kfd_process *p) kfree(pdd->qpd.doorbell_bitmap); idr_destroy(&pdd->alloc_idr); + /* + * before destroying pdd, make sure to report availability + * for auto suspend + */ + if (pdd->runtime_inuse) { + pm_runtime_mark_last_busy(pdd->dev->ddev->dev); + pm_runtime_put_autosuspend(pdd->dev->ddev->dev); + pdd->runtime_inuse = false; + } + kfree(pdd); } } @@ -457,6 +555,9 @@ static void kfd_process_wq_release(struct work_struct *work) /* Remove the procfs files */ if (p->kobj) { sysfs_remove_file(p->kobj, &p->attr_pasid); + kobject_del(p->kobj_queues); + kobject_put(p->kobj_queues); + p->kobj_queues = NULL; kobject_del(p->kobj); kobject_put(p->kobj); p->kobj = NULL; @@ -540,6 +641,11 @@ static void kfd_process_notifier_release(struct mmu_notifier *mn, /* Indicate to other users that MM is no longer valid */ p->mm = NULL; + /* Signal the eviction fence after user mode queues are + * destroyed. This allows any BOs to be freed without + * triggering pointless evictions or waiting for fences. + */ + dma_fence_signal(p->ef); mutex_unlock(&p->mutex); @@ -591,8 +697,9 @@ static int kfd_process_device_init_cwsr_dgpu(struct kfd_process_device *pdd) { struct kfd_dev *dev = pdd->dev; struct qcm_process_device *qpd = &pdd->qpd; - uint32_t flags = ALLOC_MEM_FLAGS_GTT | - ALLOC_MEM_FLAGS_NO_SUBSTITUTE | ALLOC_MEM_FLAGS_EXECUTABLE; + uint32_t flags = KFD_IOC_ALLOC_MEM_FLAGS_GTT + | KFD_IOC_ALLOC_MEM_FLAGS_NO_SUBSTITUTE + | KFD_IOC_ALLOC_MEM_FLAGS_EXECUTABLE; void *kaddr; int ret; @@ -754,6 +861,7 @@ struct kfd_process_device *kfd_create_process_device_data(struct kfd_dev *dev, pdd->process = p; pdd->bound = PDD_UNBOUND; pdd->already_dequeued = false; + pdd->runtime_inuse = false; list_add(&pdd->per_device_list, &p->per_device_data); /* Init idr used for memory handle translation */ @@ -843,15 +951,41 @@ struct kfd_process_device *kfd_bind_process_to_device(struct kfd_dev *dev, return ERR_PTR(-ENOMEM); } + /* + * signal runtime-pm system to auto resume and prevent + * further runtime suspend once device pdd is created until + * pdd is destroyed. + */ + if (!pdd->runtime_inuse) { + err = pm_runtime_get_sync(dev->ddev->dev); + if (err < 0) + return ERR_PTR(err); + } + err = kfd_iommu_bind_process_to_device(pdd); if (err) - return ERR_PTR(err); + goto out; err = kfd_process_device_init_vm(pdd, NULL); if (err) - return ERR_PTR(err); + goto out; + + /* + * make sure that runtime_usage counter is incremented just once + * per pdd + */ + pdd->runtime_inuse = true; return pdd; + +out: + /* balance runpm reference count and exit with error */ + if (!pdd->runtime_inuse) { + pm_runtime_mark_last_busy(dev->ddev->dev); + pm_runtime_put_autosuspend(dev->ddev->dev); + } + + return ERR_PTR(err); } struct kfd_process_device *kfd_get_first_process_device_data( diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c index 31fcd1b51f00..084c35f55d59 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c @@ -241,23 +241,18 @@ int pqm_create_queue(struct process_queue_manager *pqm, switch (type) { case KFD_QUEUE_TYPE_SDMA: case KFD_QUEUE_TYPE_SDMA_XGMI: - if ((type == KFD_QUEUE_TYPE_SDMA && dev->dqm->sdma_queue_count - >= get_num_sdma_queues(dev->dqm)) || - (type == KFD_QUEUE_TYPE_SDMA_XGMI && - dev->dqm->xgmi_sdma_queue_count - >= get_num_xgmi_sdma_queues(dev->dqm))) { - pr_debug("Over-subscription is not allowed for SDMA.\n"); - retval = -EPERM; - goto err_create_queue; - } - + /* SDMA queues are always allocated statically no matter + * which scheduler mode is used. We also do not need to + * check whether a SDMA queue can be allocated here, because + * allocate_sdma_queue() in create_queue() has the + * corresponding check logic. + */ retval = init_user_queue(pqm, dev, &q, properties, f, *qid); if (retval != 0) goto err_create_queue; pqn->q = q; pqn->kq = NULL; retval = dev->dqm->ops.create_queue(dev->dqm, q, &pdd->qpd); - pr_debug("DQM returned %d for create_queue\n", retval); print_queue(q); break; @@ -266,7 +261,7 @@ int pqm_create_queue(struct process_queue_manager *pqm, if ((dev->dqm->sched_policy == KFD_SCHED_POLICY_HWS_NO_OVERSUBSCRIPTION) && ((dev->dqm->processes_count >= dev->vm_info.vmid_num_kfd) || - (dev->dqm->queue_count >= get_queues_num(dev->dqm)))) { + (dev->dqm->active_queue_count >= get_cp_queues_num(dev->dqm)))) { pr_debug("Over-subscription is not allowed when amdkfd.sched_policy == 1\n"); retval = -EPERM; goto err_create_queue; @@ -278,7 +273,6 @@ int pqm_create_queue(struct process_queue_manager *pqm, pqn->q = q; pqn->kq = NULL; retval = dev->dqm->ops.create_queue(dev->dqm, q, &pdd->qpd); - pr_debug("DQM returned %d for create_queue\n", retval); print_queue(q); break; case KFD_QUEUE_TYPE_DIQ: @@ -299,7 +293,7 @@ int pqm_create_queue(struct process_queue_manager *pqm, } if (retval != 0) { - pr_err("Pasid 0x%x DQM create queue %d failed. ret %d\n", + pr_err("Pasid 0x%x DQM create queue type %d failed. ret %d\n", pqm->process->pasid, type, retval); goto err_create_queue; } @@ -322,12 +316,16 @@ int pqm_create_queue(struct process_queue_manager *pqm, if (q) { pr_debug("PQM done creating queue\n"); + kfd_procfs_add_queue(q); print_queue_properties(&q->properties); } return retval; err_create_queue: + uninit_queue(q); + if (kq) + kernel_queue_uninit(kq, false); kfree(pqn); err_allocate_pqn: /* check if queues list is empty unregister process from device */ @@ -378,6 +376,7 @@ int pqm_destroy_queue(struct process_queue_manager *pqm, unsigned int qid) } if (pqn->q) { + kfd_procfs_del_queue(pqn->q); dqm = pqn->q->device->dqm; retval = dqm->ops.destroy_queue(dqm, &pdd->qpd, pqn->q); if (retval) { diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c index 203c823d65f1..aa0bfa78a667 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c @@ -490,6 +490,8 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr, dev->node_props.num_sdma_queues_per_engine); sysfs_show_32bit_prop(buffer, "num_cp_queues", dev->node_props.num_cp_queues); + sysfs_show_64bit_prop(buffer, "unique_id", + dev->node_props.unique_id); if (dev->gpu) { log_max_watch_addr = @@ -1318,7 +1320,8 @@ int kfd_topology_add_device(struct kfd_dev *gpu) dev->node_props.num_gws = (hws_gws_support && dev->gpu->dqm->sched_policy != KFD_SCHED_POLICY_NO_HWS) ? amdgpu_amdkfd_get_num_gws(dev->gpu->kgd) : 0; - dev->node_props.num_cp_queues = get_queues_num(dev->gpu->dqm); + dev->node_props.num_cp_queues = get_cp_queues_num(dev->gpu->dqm); + dev->node_props.unique_id = gpu->unique_id; kfd_fill_mem_clk_max_info(dev); kfd_fill_iolink_non_crat_info(dev); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.h b/drivers/gpu/drm/amd/amdkfd/kfd_topology.h index 74e9b1682af8..46eeecaf1b68 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.h @@ -54,6 +54,7 @@ struct kfd_node_properties { uint64_t hive_id; + uint64_t unique_id; uint32_t cpu_cores_count; uint32_t simd_count; uint32_t mem_banks_count; diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index 63e8a12a74bc..f7c5cdc10a70 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -98,6 +98,9 @@ MODULE_FIRMWARE(FIRMWARE_RENOIR_DMUB); #define FIRMWARE_RAVEN_DMCU "amdgpu/raven_dmcu.bin" MODULE_FIRMWARE(FIRMWARE_RAVEN_DMCU); +#define FIRMWARE_NAVI12_DMCU "amdgpu/navi12_dmcu.bin" +MODULE_FIRMWARE(FIRMWARE_NAVI12_DMCU); + /* Number of bytes in PSP header for firmware. */ #define PSP_HEADER_BYTES 0x100 @@ -129,9 +132,6 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev); /* removes and deallocates the drm structures, created by the above function */ static void amdgpu_dm_destroy_drm_device(struct amdgpu_display_manager *dm); -static void -amdgpu_dm_update_connector_after_detect(struct amdgpu_dm_connector *aconnector); - static int amdgpu_dm_plane_init(struct amdgpu_display_manager *dm, struct drm_plane *plane, unsigned long possible_crtcs, @@ -383,8 +383,8 @@ static void dm_pflip_high_irq(void *interrupt_params) * of pageflip completion, so last_flip_vblank is the forbidden count * for queueing new pageflips if vsync + VRR is enabled. */ - amdgpu_crtc->last_flip_vblank = amdgpu_get_vblank_counter_kms(adev->ddev, - amdgpu_crtc->crtc_id); + amdgpu_crtc->last_flip_vblank = + amdgpu_get_vblank_counter_kms(&amdgpu_crtc->base); amdgpu_crtc->pflip_status = AMDGPU_FLIP_NONE; spin_unlock_irqrestore(&adev->ddev->event_lock, flags); @@ -407,8 +407,9 @@ static void dm_vupdate_high_irq(void *interrupt_params) if (acrtc) { acrtc_state = to_dm_crtc_state(acrtc->base.state); - DRM_DEBUG_DRIVER("crtc:%d, vupdate-vrr:%d\n", acrtc->crtc_id, - amdgpu_dm_vrr_active(acrtc_state)); + DRM_DEBUG_VBL("crtc:%d, vupdate-vrr:%d\n", + acrtc->crtc_id, + amdgpu_dm_vrr_active(acrtc_state)); /* Core vblank handling is done here after end of front-porch in * vrr mode, as vblank timestamping will give valid results @@ -458,8 +459,9 @@ static void dm_crtc_high_irq(void *interrupt_params) if (acrtc) { acrtc_state = to_dm_crtc_state(acrtc->base.state); - DRM_DEBUG_DRIVER("crtc:%d, vupdate-vrr:%d\n", acrtc->crtc_id, - amdgpu_dm_vrr_active(acrtc_state)); + DRM_DEBUG_VBL("crtc:%d, vupdate-vrr:%d\n", + acrtc->crtc_id, + amdgpu_dm_vrr_active(acrtc_state)); /* Core vblank handling at start of front-porch is only possible * in non-vrr mode, as only there vblank timestamping will give @@ -522,8 +524,9 @@ static void dm_dcn_crtc_high_irq(void *interrupt_params) acrtc_state = to_dm_crtc_state(acrtc->base.state); - DRM_DEBUG_DRIVER("crtc:%d, vupdate-vrr:%d\n", acrtc->crtc_id, - amdgpu_dm_vrr_active(acrtc_state)); + DRM_DEBUG_VBL("crtc:%d, vupdate-vrr:%d, planes:%d\n", acrtc->crtc_id, + amdgpu_dm_vrr_active(acrtc_state), + acrtc_state->active_planes); amdgpu_dm_crtc_handle_crc_irq(&acrtc->base); drm_crtc_handle_vblank(&acrtc->base); @@ -543,7 +546,18 @@ static void dm_dcn_crtc_high_irq(void *interrupt_params) &acrtc_state->vrr_params.adjust); } - if (acrtc->pflip_status == AMDGPU_FLIP_SUBMITTED) { + /* + * If there aren't any active_planes then DCH HUBP may be clock-gated. + * In that case, pageflip completion interrupts won't fire and pageflip + * completion events won't get delivered. Prevent this by sending + * pending pageflip events from here if a flip is still pending. + * + * If any planes are enabled, use dm_pflip_high_irq() instead, to + * avoid race conditions between flip programming and completion, + * which could cause too early flip completion events. + */ + if (acrtc->pflip_status == AMDGPU_FLIP_SUBMITTED && + acrtc_state->active_planes == 0) { if (acrtc->event) { drm_crtc_send_vblank_event(&acrtc->base, acrtc->event); acrtc->event = NULL; @@ -801,10 +815,20 @@ static int dm_dmub_hw_init(struct amdgpu_device *adev) fw_bss_data_size = le32_to_cpu(hdr->bss_data_bytes); - memcpy(fb_info->fb[DMUB_WINDOW_0_INST_CONST].cpu_addr, fw_inst_const, - fw_inst_const_size); + /* if adev->firmware.load_type == AMDGPU_FW_LOAD_PSP, + * amdgpu_ucode_init_single_fw will load dmub firmware + * fw_inst_const part to cw0; otherwise, the firmware back door load + * will be done by dm_dmub_hw_init + */ + if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) { + memcpy(fb_info->fb[DMUB_WINDOW_0_INST_CONST].cpu_addr, fw_inst_const, + fw_inst_const_size); + } + memcpy(fb_info->fb[DMUB_WINDOW_2_BSS_DATA].cpu_addr, fw_bss_data, fw_bss_data_size); + + /* Copy firmware bios info into FB memory. */ memcpy(fb_info->fb[DMUB_WINDOW_3_VBIOS].cpu_addr, adev->bios, adev->bios_size); @@ -823,6 +847,10 @@ static int dm_dmub_hw_init(struct amdgpu_device *adev) hw_params.fb_base = adev->gmc.fb_start; hw_params.fb_offset = adev->gmc.aper_base; + /* backdoor load firmware and trigger dmub running */ + if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) + hw_params.load_inst_const = true; + if (dmcu) hw_params.psp_version = dmcu->psp_version; @@ -885,7 +913,7 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) init_data.asic_id.chip_family = adev->family; - init_data.asic_id.pci_revision_id = adev->rev_id; + init_data.asic_id.pci_revision_id = adev->pdev->revision; init_data.asic_id.hw_internal_rev = adev->external_rev_id; init_data.asic_id.vram_width = adev->gmc.vram_width; @@ -960,7 +988,7 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) #ifdef CONFIG_DRM_AMD_DC_HDCP if (adev->asic_type >= CHIP_RAVEN) { - adev->dm.hdcp_workqueue = hdcp_create_workqueue(&adev->psp, &init_params.cp_psp, adev->dm.dc); + adev->dm.hdcp_workqueue = hdcp_create_workqueue(adev, &init_params.cp_psp, adev->dm.dc); if (!adev->dm.hdcp_workqueue) DRM_ERROR("amdgpu: failed to initialize hdcp_workqueue.\n"); @@ -991,11 +1019,6 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) goto error; } -#if defined(CONFIG_DEBUG_FS) - if (dtn_debugfs_init(adev)) - DRM_ERROR("amdgpu: failed initialize dtn debugfs support.\n"); -#endif - DRM_DEBUG_DRIVER("KMS initialized.\n"); return 0; @@ -1079,9 +1102,11 @@ static int load_dmcu_fw(struct amdgpu_device *adev) case CHIP_VEGA20: case CHIP_NAVI10: case CHIP_NAVI14: - case CHIP_NAVI12: case CHIP_RENOIR: return 0; + case CHIP_NAVI12: + fw_name_dmcu = FIRMWARE_NAVI12_DMCU; + break; case CHIP_RAVEN: if (ASICREV_IS_PICASSO(adev->external_rev_id)) fw_name_dmcu = FIRMWARE_RAVEN_DMCU; @@ -1192,22 +1217,21 @@ static int dm_dmub_sw_init(struct amdgpu_device *adev) return 0; } - if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) { - DRM_WARN("Only PSP firmware loading is supported for DMUB\n"); - return 0; - } - hdr = (const struct dmcub_firmware_header_v1_0 *)adev->dm.dmub_fw->data; - adev->firmware.ucode[AMDGPU_UCODE_ID_DMCUB].ucode_id = - AMDGPU_UCODE_ID_DMCUB; - adev->firmware.ucode[AMDGPU_UCODE_ID_DMCUB].fw = adev->dm.dmub_fw; - adev->firmware.fw_size += - ALIGN(le32_to_cpu(hdr->inst_const_bytes), PAGE_SIZE); - adev->dm.dmcub_fw_version = le32_to_cpu(hdr->header.ucode_version); + if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) { + adev->firmware.ucode[AMDGPU_UCODE_ID_DMCUB].ucode_id = + AMDGPU_UCODE_ID_DMCUB; + adev->firmware.ucode[AMDGPU_UCODE_ID_DMCUB].fw = + adev->dm.dmub_fw; + adev->firmware.fw_size += + ALIGN(le32_to_cpu(hdr->inst_const_bytes), PAGE_SIZE); - DRM_INFO("Loading DMUB firmware via PSP: version=0x%08X\n", - adev->dm.dmcub_fw_version); + DRM_INFO("Loading DMUB firmware via PSP: version=0x%08X\n", + adev->dm.dmcub_fw_version); + } + + adev->dm.dmcub_fw_version = le32_to_cpu(hdr->header.ucode_version); adev->dm.dmub_srv = kzalloc(sizeof(*adev->dm.dmub_srv), GFP_KERNEL); dmub_srv = adev->dm.dmub_srv; @@ -1422,6 +1446,73 @@ static void s3_handle_mst(struct drm_device *dev, bool suspend) drm_kms_helper_hotplug_event(dev); } +static int amdgpu_dm_smu_write_watermarks_table(struct amdgpu_device *adev) +{ + struct smu_context *smu = &adev->smu; + int ret = 0; + + if (!is_support_sw_smu(adev)) + return 0; + + /* This interface is for dGPU Navi1x.Linux dc-pplib interface depends + * on window driver dc implementation. + * For Navi1x, clock settings of dcn watermarks are fixed. the settings + * should be passed to smu during boot up and resume from s3. + * boot up: dc calculate dcn watermark clock settings within dc_create, + * dcn20_resource_construct + * then call pplib functions below to pass the settings to smu: + * smu_set_watermarks_for_clock_ranges + * smu_set_watermarks_table + * navi10_set_watermarks_table + * smu_write_watermarks_table + * + * For Renoir, clock settings of dcn watermark are also fixed values. + * dc has implemented different flow for window driver: + * dc_hardware_init / dc_set_power_state + * dcn10_init_hw + * notify_wm_ranges + * set_wm_ranges + * -- Linux + * smu_set_watermarks_for_clock_ranges + * renoir_set_watermarks_table + * smu_write_watermarks_table + * + * For Linux, + * dc_hardware_init -> amdgpu_dm_init + * dc_set_power_state --> dm_resume + * + * therefore, this function apply to navi10/12/14 but not Renoir + * * + */ + switch(adev->asic_type) { + case CHIP_NAVI10: + case CHIP_NAVI14: + case CHIP_NAVI12: + break; + default: + return 0; + } + + mutex_lock(&smu->mutex); + + /* pass data to smu controller */ + if ((smu->watermarks_bitmap & WATERMARKS_EXIST) && + !(smu->watermarks_bitmap & WATERMARKS_LOADED)) { + ret = smu_write_watermarks_table(smu); + + if (ret) { + mutex_unlock(&smu->mutex); + DRM_ERROR("Failed to update WMTABLE!\n"); + return ret; + } + smu->watermarks_bitmap |= WATERMARKS_LOADED; + } + + mutex_unlock(&smu->mutex); + + return 0; +} + /** * dm_hw_init() - Initialize DC device * @handle: The base driver device containing the amdgpu_dm device. @@ -1700,6 +1791,8 @@ static int dm_resume(void *handle) amdgpu_dm_irq_resume_late(adev); + amdgpu_dm_smu_write_watermarks_table(adev); + return 0; } @@ -1758,8 +1851,63 @@ static struct drm_mode_config_helper_funcs amdgpu_dm_mode_config_helperfuncs = { .atomic_commit_tail = amdgpu_dm_atomic_commit_tail }; -static void -amdgpu_dm_update_connector_after_detect(struct amdgpu_dm_connector *aconnector) +static void update_connector_ext_caps(struct amdgpu_dm_connector *aconnector) +{ + u32 max_cll, min_cll, max, min, q, r; + struct amdgpu_dm_backlight_caps *caps; + struct amdgpu_display_manager *dm; + struct drm_connector *conn_base; + struct amdgpu_device *adev; + static const u8 pre_computed_values[] = { + 50, 51, 52, 53, 55, 56, 57, 58, 59, 61, 62, 63, 65, 66, 68, 69, + 71, 72, 74, 75, 77, 79, 81, 82, 84, 86, 88, 90, 92, 94, 96, 98}; + + if (!aconnector || !aconnector->dc_link) + return; + + conn_base = &aconnector->base; + adev = conn_base->dev->dev_private; + dm = &adev->dm; + caps = &dm->backlight_caps; + caps->ext_caps = &aconnector->dc_link->dpcd_sink_ext_caps; + caps->aux_support = false; + max_cll = conn_base->hdr_sink_metadata.hdmi_type1.max_cll; + min_cll = conn_base->hdr_sink_metadata.hdmi_type1.min_cll; + + if (caps->ext_caps->bits.oled == 1 || + caps->ext_caps->bits.sdr_aux_backlight_control == 1 || + caps->ext_caps->bits.hdr_aux_backlight_control == 1) + caps->aux_support = true; + + /* From the specification (CTA-861-G), for calculating the maximum + * luminance we need to use: + * Luminance = 50*2**(CV/32) + * Where CV is a one-byte value. + * For calculating this expression we may need float point precision; + * to avoid this complexity level, we take advantage that CV is divided + * by a constant. From the Euclids division algorithm, we know that CV + * can be written as: CV = 32*q + r. Next, we replace CV in the + * Luminance expression and get 50*(2**q)*(2**(r/32)), hence we just + * need to pre-compute the value of r/32. For pre-computing the values + * We just used the following Ruby line: + * (0...32).each {|cv| puts (50*2**(cv/32.0)).round} + * The results of the above expressions can be verified at + * pre_computed_values. + */ + q = max_cll >> 5; + r = max_cll % 32; + max = (1 << q) * pre_computed_values[r]; + + // min luminance: maxLum * (CV/255)^2 / 100 + q = DIV_ROUND_CLOSEST(min_cll, 255); + min = max * DIV_ROUND_CLOSEST((q * q), 100); + + caps->aux_max_input_signal = max; + caps->aux_min_input_signal = min; +} + +void amdgpu_dm_update_connector_after_detect( + struct amdgpu_dm_connector *aconnector) { struct drm_connector *connector = &aconnector->base; struct drm_device *dev = connector->dev; @@ -1872,7 +2020,7 @@ amdgpu_dm_update_connector_after_detect(struct amdgpu_dm_connector *aconnector) aconnector->edid); } amdgpu_dm_update_freesync_caps(connector, aconnector->edid); - + update_connector_ext_caps(aconnector); } else { drm_dp_cec_unset_edid(&aconnector->dm_dp_aux.aux); amdgpu_dm_update_freesync_caps(connector, NULL); @@ -1911,7 +2059,7 @@ static void handle_hpd_irq(void *param) mutex_lock(&aconnector->hpd_lock); #ifdef CONFIG_DRM_AMD_DC_HDCP - if (adev->asic_type >= CHIP_RAVEN) + if (adev->dm.hdcp_workqueue) hdcp_reset_display(adev->dm.hdcp_workqueue, aconnector->dc_link->link_index); #endif if (aconnector->fake_enable) @@ -2088,8 +2236,10 @@ static void handle_hpd_rx_irq(void *param) } } #ifdef CONFIG_DRM_AMD_DC_HDCP - if (hpd_irq_data.bytes.device_service_irq.bits.CP_IRQ) - hdcp_handle_cpirq(adev->dm.hdcp_workqueue, aconnector->base.index); + if (hpd_irq_data.bytes.device_service_irq.bits.CP_IRQ) { + if (adev->dm.hdcp_workqueue) + hdcp_handle_cpirq(adev->dm.hdcp_workqueue, aconnector->base.index); + } #endif if ((dc_link->cur_link_settings.lane_count != LANE_COUNT_UNKNOWN) || (dc_link->type == dc_connection_mst_branch)) @@ -2484,6 +2634,7 @@ static int amdgpu_dm_mode_config_init(struct amdgpu_device *adev) #define AMDGPU_DM_DEFAULT_MIN_BACKLIGHT 12 #define AMDGPU_DM_DEFAULT_MAX_BACKLIGHT 255 +#define AUX_BL_DEFAULT_TRANSITION_TIME_MS 50 #if defined(CONFIG_BACKLIGHT_CLASS_DEVICE) ||\ defined(CONFIG_BACKLIGHT_CLASS_DEVICE_MODULE) @@ -2498,9 +2649,11 @@ static void amdgpu_dm_update_backlight_caps(struct amdgpu_display_manager *dm) amdgpu_acpi_get_backlight_caps(dm->adev, &caps); if (caps.caps_valid) { + dm->backlight_caps.caps_valid = true; + if (caps.aux_support) + return; dm->backlight_caps.min_input_signal = caps.min_input_signal; dm->backlight_caps.max_input_signal = caps.max_input_signal; - dm->backlight_caps.caps_valid = true; } else { dm->backlight_caps.min_input_signal = AMDGPU_DM_DEFAULT_MIN_BACKLIGHT; @@ -2508,40 +2661,95 @@ static void amdgpu_dm_update_backlight_caps(struct amdgpu_display_manager *dm) AMDGPU_DM_DEFAULT_MAX_BACKLIGHT; } #else + if (dm->backlight_caps.aux_support) + return; + dm->backlight_caps.min_input_signal = AMDGPU_DM_DEFAULT_MIN_BACKLIGHT; dm->backlight_caps.max_input_signal = AMDGPU_DM_DEFAULT_MAX_BACKLIGHT; #endif } +static int set_backlight_via_aux(struct dc_link *link, uint32_t brightness) +{ + bool rc; + + if (!link) + return 1; + + rc = dc_link_set_backlight_level_nits(link, true, brightness, + AUX_BL_DEFAULT_TRANSITION_TIME_MS); + + return rc ? 0 : 1; +} + +static u32 convert_brightness(const struct amdgpu_dm_backlight_caps *caps, + const uint32_t user_brightness) +{ + u32 min, max, conversion_pace; + u32 brightness = user_brightness; + + if (!caps) + goto out; + + if (!caps->aux_support) { + max = caps->max_input_signal; + min = caps->min_input_signal; + /* + * The brightness input is in the range 0-255 + * It needs to be rescaled to be between the + * requested min and max input signal + * It also needs to be scaled up by 0x101 to + * match the DC interface which has a range of + * 0 to 0xffff + */ + conversion_pace = 0x101; + brightness = + user_brightness + * conversion_pace + * (max - min) + / AMDGPU_MAX_BL_LEVEL + + min * conversion_pace; + } else { + /* TODO + * We are doing a linear interpolation here, which is OK but + * does not provide the optimal result. We probably want + * something close to the Perceptual Quantizer (PQ) curve. + */ + max = caps->aux_max_input_signal; + min = caps->aux_min_input_signal; + + brightness = (AMDGPU_MAX_BL_LEVEL - user_brightness) * min + + user_brightness * max; + // Multiple the value by 1000 since we use millinits + brightness *= 1000; + brightness = DIV_ROUND_CLOSEST(brightness, AMDGPU_MAX_BL_LEVEL); + } + +out: + return brightness; +} + static int amdgpu_dm_backlight_update_status(struct backlight_device *bd) { struct amdgpu_display_manager *dm = bl_get_data(bd); struct amdgpu_dm_backlight_caps caps; - uint32_t brightness = bd->props.brightness; + struct dc_link *link = NULL; + u32 brightness; + bool rc; amdgpu_dm_update_backlight_caps(dm); caps = dm->backlight_caps; - /* - * The brightness input is in the range 0-255 - * It needs to be rescaled to be between the - * requested min and max input signal - * - * It also needs to be scaled up by 0x101 to - * match the DC interface which has a range of - * 0 to 0xffff - */ - brightness = - brightness - * 0x101 - * (caps.max_input_signal - caps.min_input_signal) - / AMDGPU_MAX_BL_LEVEL - + caps.min_input_signal * 0x101; - - if (dc_link_set_backlight_level(dm->backlight_link, - brightness, 0)) - return 0; - else - return 1; + + link = (struct dc_link *)dm->backlight_link; + + brightness = convert_brightness(&caps, bd->props.brightness); + // Change brightness based on AUX property + if (caps.aux_support) + return set_backlight_via_aux(link, brightness); + + rc = dc_link_set_backlight_level(dm->backlight_link, brightness, 0); + + return rc ? 0 : 1; } static int amdgpu_dm_backlight_get_brightness(struct backlight_device *bd) @@ -2826,6 +3034,9 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev) if (adev->asic_type != CHIP_CARRIZO && adev->asic_type != CHIP_STONEY) dm->dc->debug.disable_stutter = amdgpu_pp_feature_mask & PP_STUTTER_MODE ? false : true; + /* No userspace support. */ + dm->dc->debug.disable_tri_buf = true; + return 0; fail: kfree(aencoder); @@ -3428,6 +3639,9 @@ fill_dc_plane_info_and_addr(struct amdgpu_device *adev, case DRM_FORMAT_NV12: plane_info->format = SURFACE_PIXEL_FORMAT_VIDEO_420_YCrCb; break; + case DRM_FORMAT_P010: + plane_info->format = SURFACE_PIXEL_FORMAT_VIDEO_420_10bpc_YCrCb; + break; default: DRM_ERROR( "Unsupported screen format %s\n", @@ -4117,9 +4331,22 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector, struct dmcu *dmcu = core_dc->res_pool->dmcu; stream->psr_version = dmcu->dmcu_version.psr_version; - mod_build_vsc_infopacket(stream, - &stream->vsc_infopacket, - &stream->use_vsc_sdp_for_colorimetry); + + // + // should decide stream support vsc sdp colorimetry capability + // before building vsc info packet + // + stream->use_vsc_sdp_for_colorimetry = false; + if (aconnector->dc_sink->sink_signal == SIGNAL_TYPE_DISPLAY_PORT_MST) { + stream->use_vsc_sdp_for_colorimetry = + aconnector->dc_sink->is_vsc_sdp_colorimetry_supported; + } else { + if (stream->link->dpcd_caps.dpcd_rev.raw >= 0x14 && + stream->link->dpcd_caps.dprx_feature.bits.VSC_SDP_COLORIMETRY_SUPPORTED) { + stream->use_vsc_sdp_for_colorimetry = true; + } + } + mod_build_vsc_infopacket(stream, &stream->vsc_infopacket); } } finish: @@ -4269,8 +4496,10 @@ static const struct drm_crtc_funcs amdgpu_dm_crtc_funcs = { .set_crc_source = amdgpu_dm_crtc_set_crc_source, .verify_crc_source = amdgpu_dm_crtc_verify_crc_source, .get_crc_sources = amdgpu_dm_crtc_get_crc_sources, + .get_vblank_counter = amdgpu_get_vblank_counter_kms, .enable_vblank = dm_enable_vblank, .disable_vblank = dm_disable_vblank, + .get_vblank_timestamp = drm_crtc_vblank_helper_get_vblank_timestamp, }; static enum drm_connector_status @@ -4491,6 +4720,19 @@ amdgpu_dm_connector_atomic_duplicate_state(struct drm_connector *connector) return &new_state->base; } +static int +amdgpu_dm_connector_late_register(struct drm_connector *connector) +{ +#if defined(CONFIG_DEBUG_FS) + struct amdgpu_dm_connector *amdgpu_dm_connector = + to_amdgpu_dm_connector(connector); + + connector_debugfs_init(amdgpu_dm_connector); +#endif + + return 0; +} + static const struct drm_connector_funcs amdgpu_dm_connector_funcs = { .reset = amdgpu_dm_connector_funcs_reset, .detect = amdgpu_dm_connector_detect, @@ -4500,6 +4742,7 @@ static const struct drm_connector_funcs amdgpu_dm_connector_funcs = { .atomic_destroy_state = drm_atomic_helper_connector_destroy_state, .atomic_set_property = amdgpu_dm_connector_atomic_set_property, .atomic_get_property = amdgpu_dm_connector_atomic_get_property, + .late_register = amdgpu_dm_connector_late_register, .early_unregister = amdgpu_dm_connector_unregister }; @@ -4876,7 +5119,8 @@ static bool dm_crtc_helper_mode_fixup(struct drm_crtc *crtc, static const struct drm_crtc_helper_funcs amdgpu_dm_crtc_helper_funcs = { .disable = dm_crtc_helper_disable, .atomic_check = dm_crtc_helper_atomic_check, - .mode_fixup = dm_crtc_helper_mode_fixup + .mode_fixup = dm_crtc_helper_mode_fixup, + .get_scanout_position = amdgpu_crtc_get_scanout_position, }; static void dm_encoder_helper_disable(struct drm_encoder *encoder) @@ -5294,6 +5538,8 @@ static int get_plane_formats(const struct drm_plane *plane, if (plane_cap && plane_cap->pixel_format_support.nv12) formats[num_formats++] = DRM_FORMAT_NV12; + if (plane_cap && plane_cap->pixel_format_support.p010) + formats[num_formats++] = DRM_FORMAT_P010; break; case DRM_PLANE_TYPE_OVERLAY: @@ -5346,12 +5592,15 @@ static int amdgpu_dm_plane_init(struct amdgpu_display_manager *dm, } if (plane->type == DRM_PLANE_TYPE_PRIMARY && - plane_cap && plane_cap->pixel_format_support.nv12) { + plane_cap && + (plane_cap->pixel_format_support.nv12 || + plane_cap->pixel_format_support.p010)) { /* This only affects YUV formats. */ drm_plane_create_color_properties( plane, BIT(DRM_COLOR_YCBCR_BT601) | - BIT(DRM_COLOR_YCBCR_BT709), + BIT(DRM_COLOR_YCBCR_BT709) | + BIT(DRM_COLOR_YCBCR_BT2020), BIT(DRM_COLOR_YCBCR_LIMITED_RANGE) | BIT(DRM_COLOR_YCBCR_FULL_RANGE), DRM_COLOR_YCBCR_BT709, DRM_COLOR_YCBCR_LIMITED_RANGE); @@ -5680,7 +5929,8 @@ void amdgpu_dm_connector_init_helper(struct amdgpu_display_manager *dm, adev->mode_info.underscan_vborder_property, 0); - drm_connector_attach_max_bpc_property(&aconnector->base, 8, 16); + if (!aconnector->mst_port) + drm_connector_attach_max_bpc_property(&aconnector->base, 8, 16); /* This defaults to the max in the range, but we want 8bpc for non-edp. */ aconnector->base.state->max_bpc = (connector_type == DRM_MODE_CONNECTOR_eDP) ? 16 : 8; @@ -5699,10 +5949,11 @@ void amdgpu_dm_connector_init_helper(struct amdgpu_display_manager *dm, &aconnector->base.base, dm->ddev->mode_config.hdr_output_metadata_property, 0); - drm_connector_attach_vrr_capable_property( - &aconnector->base); + if (!aconnector->mst_port) + drm_connector_attach_vrr_capable_property(&aconnector->base); + #ifdef CONFIG_DRM_AMD_DC_HDCP - if (adev->asic_type >= CHIP_RAVEN) + if (adev->dm.hdcp_workqueue) drm_connector_attach_content_protection_property(&aconnector->base, true); #endif } @@ -5839,13 +6090,6 @@ static int amdgpu_dm_connector_init(struct amdgpu_display_manager *dm, drm_connector_attach_encoder( &aconnector->base, &aencoder->base); - drm_connector_register(&aconnector->base); -#if defined(CONFIG_DEBUG_FS) - connector_debugfs_init(aconnector); - aconnector->debugfs_dpcd_address = 0; - aconnector->debugfs_dpcd_size = 0; -#endif - if (connector_type == DRM_MODE_CONNECTOR_DisplayPort || connector_type == DRM_MODE_CONNECTOR_eDP) amdgpu_dm_initialize_dp_connector(dm, aconnector); @@ -6030,12 +6274,6 @@ static int get_cursor_position(struct drm_plane *plane, struct drm_crtc *crtc, y <= -amdgpu_crtc->max_cursor_height) return 0; - if (crtc->primary->state) { - /* avivo cursor are offset into the total surface */ - x += crtc->primary->state->src_x >> 16; - y += crtc->primary->state->src_y >> 16; - } - if (x < 0) { xorigin = min(-x, amdgpu_crtc->max_cursor_width - 1); x = 0; @@ -6045,6 +6283,7 @@ static int get_cursor_position(struct drm_plane *plane, struct drm_crtc *crtc, y = 0; } position->enable = true; + position->translate_by_source = true; position->x = x; position->y = y; position->x_hotspot = xorigin; @@ -6331,7 +6570,6 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, uint32_t target_vblank, last_flip_vblank; bool vrr_active = amdgpu_dm_vrr_active(acrtc_state); bool pflip_present = false; - bool swizzle = true; struct { struct dc_surface_update surface_updates[MAX_SURFACES]; struct dc_plane_info plane_infos[MAX_SURFACES]; @@ -6377,9 +6615,6 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, dc_plane = dm_new_plane_state->dc_state; - if (dc_plane && !dc_plane->tiling_info.gfx9.swizzle) - swizzle = false; - bundle->surface_updates[planes_count].surface = dc_plane; if (new_pcrtc_state->color_mgmt_changed) { bundle->surface_updates[planes_count].gamma = dc_plane->gamma_correction; @@ -6480,7 +6715,7 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, * clients using the GLX_OML_sync_control extension or * DRI3/Present extension with defined target_msc. */ - last_flip_vblank = amdgpu_get_vblank_counter_kms(dm->ddev, acrtc_attach->crtc_id); + last_flip_vblank = amdgpu_get_vblank_counter_kms(pcrtc); } else { /* For variable refresh rate mode only: @@ -6509,7 +6744,7 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, & (DRM_SCANOUTPOS_VALID | DRM_SCANOUTPOS_IN_VBLANK)) == (DRM_SCANOUTPOS_VALID | DRM_SCANOUTPOS_IN_VBLANK) && (int)(target_vblank - - amdgpu_get_vblank_counter_kms(dm->ddev, acrtc_attach->crtc_id)) > 0)) { + amdgpu_get_vblank_counter_kms(pcrtc)) > 0)) { usleep_range(1000, 1100); } @@ -6588,8 +6823,7 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, amdgpu_dm_link_setup_psr(acrtc_state->stream); else if ((acrtc_state->update_type == UPDATE_TYPE_FAST) && acrtc_state->stream->link->psr_feature_enabled && - !acrtc_state->stream->link->psr_allow_active && - swizzle) { + !acrtc_state->stream->link->psr_allow_active) { amdgpu_dm_psr_enable(acrtc_state->stream); } diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h index 7ea9acb0358d..5cab3e65d992 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h @@ -90,15 +90,41 @@ struct dm_comressor_info { }; /** - * struct amdgpu_dm_backlight_caps - Usable range of backlight values from ACPI - * @min_input_signal: minimum possible input in range 0-255 - * @max_input_signal: maximum possible input in range 0-255 - * @caps_valid: true if these values are from the ACPI interface + * struct amdgpu_dm_backlight_caps - Information about backlight + * + * Describe the backlight support for ACPI or eDP AUX. */ struct amdgpu_dm_backlight_caps { + /** + * @ext_caps: Keep the data struct with all the information about the + * display support for HDR. + */ + union dpcd_sink_ext_caps *ext_caps; + /** + * @aux_min_input_signal: Min brightness value supported by the display + */ + u32 aux_min_input_signal; + /** + * @aux_max_input_signal: Max brightness value supported by the display + * in nits. + */ + u32 aux_max_input_signal; + /** + * @min_input_signal: minimum possible input in range 0-255. + */ int min_input_signal; + /** + * @max_input_signal: maximum possible input in range 0-255. + */ int max_input_signal; + /** + * @caps_valid: true if these values are from the ACPI interface. + */ bool caps_valid; + /** + * @aux_support: Describes if the display supports AUX backlight. + */ + bool aux_support; }; /** @@ -457,6 +483,9 @@ int amdgpu_dm_update_crtc_color_mgmt(struct dm_crtc_state *crtc); int amdgpu_dm_update_plane_color_mgmt(struct dm_crtc_state *crtc, struct dc_plane_state *dc_plane_state); +void amdgpu_dm_update_connector_after_detect( + struct amdgpu_dm_connector *aconnector); + extern const struct drm_encoder_helper_funcs amdgpu_dm_encoder_helper_funcs; #endif /* __AMDGPU_DM_H__ */ diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c index f81d3439ee8c..0461fecd68db 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c @@ -32,6 +32,19 @@ #include "amdgpu_dm.h" #include "amdgpu_dm_debugfs.h" #include "dm_helpers.h" +#include "dmub/inc/dmub_srv.h" + +struct dmub_debugfs_trace_header { + uint32_t entry_count; + uint32_t reserved[3]; +}; + +struct dmub_debugfs_trace_entry { + uint32_t trace_code; + uint32_t tick_count; + uint32_t param0; + uint32_t param1; +}; /* function description * get/ set DP configuration: lane_count, link_rate, spread_spectrum @@ -675,6 +688,73 @@ static ssize_t dp_phy_test_pattern_debugfs_write(struct file *f, const char __us return bytes_from_user; } +/** + * Returns the DMCUB tracebuffer contents. + * Example usage: cat /sys/kernel/debug/dri/0/amdgpu_dm_dmub_tracebuffer + */ +static int dmub_tracebuffer_show(struct seq_file *m, void *data) +{ + struct amdgpu_device *adev = m->private; + struct dmub_srv_fb_info *fb_info = adev->dm.dmub_fb_info; + struct dmub_debugfs_trace_entry *entries; + uint8_t *tbuf_base; + uint32_t tbuf_size, max_entries, num_entries, i; + + if (!fb_info) + return 0; + + tbuf_base = (uint8_t *)fb_info->fb[DMUB_WINDOW_5_TRACEBUFF].cpu_addr; + if (!tbuf_base) + return 0; + + tbuf_size = fb_info->fb[DMUB_WINDOW_5_TRACEBUFF].size; + max_entries = (tbuf_size - sizeof(struct dmub_debugfs_trace_header)) / + sizeof(struct dmub_debugfs_trace_entry); + + num_entries = + ((struct dmub_debugfs_trace_header *)tbuf_base)->entry_count; + + num_entries = min(num_entries, max_entries); + + entries = (struct dmub_debugfs_trace_entry + *)(tbuf_base + + sizeof(struct dmub_debugfs_trace_header)); + + for (i = 0; i < num_entries; ++i) { + struct dmub_debugfs_trace_entry *entry = &entries[i]; + + seq_printf(m, + "trace_code=%u tick_count=%u param0=%u param1=%u\n", + entry->trace_code, entry->tick_count, entry->param0, + entry->param1); + } + + return 0; +} + +/** + * Returns the DMCUB firmware state contents. + * Example usage: cat /sys/kernel/debug/dri/0/amdgpu_dm_dmub_fw_state + */ +static int dmub_fw_state_show(struct seq_file *m, void *data) +{ + struct amdgpu_device *adev = m->private; + struct dmub_srv_fb_info *fb_info = adev->dm.dmub_fb_info; + uint8_t *state_base; + uint32_t state_size; + + if (!fb_info) + return 0; + + state_base = (uint8_t *)fb_info->fb[DMUB_WINDOW_6_FW_STATE].cpu_addr; + if (!state_base) + return 0; + + state_size = fb_info->fb[DMUB_WINDOW_6_FW_STATE].size; + + return seq_write(m, state_base, state_size); +} + /* * Returns the current and maximum output bpc for the connector. * Example usage: cat /sys/kernel/debug/dri/0/DP-1/output_bpc @@ -880,6 +960,8 @@ static ssize_t dp_dpcd_data_read(struct file *f, char __user *buf, return read_size - r; } +DEFINE_SHOW_ATTRIBUTE(dmub_fw_state); +DEFINE_SHOW_ATTRIBUTE(dmub_tracebuffer); DEFINE_SHOW_ATTRIBUTE(output_bpc); DEFINE_SHOW_ATTRIBUTE(vrr_range); @@ -1008,6 +1090,9 @@ void connector_debugfs_init(struct amdgpu_dm_connector *connector) debugfs_create_file_unsafe("force_yuv420_output", 0644, dir, connector, &force_yuv420_output_fops); + connector->debugfs_dpcd_address = 0; + connector->debugfs_dpcd_size = 0; + } /* @@ -1188,5 +1273,11 @@ int dtn_debugfs_init(struct amdgpu_device *adev) debugfs_create_file_unsafe("amdgpu_dm_visual_confirm", 0644, root, adev, &visual_confirm_fops); + debugfs_create_file_unsafe("amdgpu_dm_dmub_tracebuffer", 0644, root, + adev, &dmub_tracebuffer_fops); + + debugfs_create_file_unsafe("amdgpu_dm_dmub_fw_state", 0644, root, + adev, &dmub_fw_state_fops); + return 0; } diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c index 0acd3409dd6c..78e1c11d4ae5 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c @@ -28,6 +28,13 @@ #include "amdgpu_dm.h" #include "dm_helpers.h" #include <drm/drm_hdcp.h> +#include "hdcp_psp.h" + +/* + * If the SRM version being loaded is less than or equal to the + * currently loaded SRM, psp will return 0xFFFF as the version + */ +#define PSP_SRM_VERSION_MAX 0xFFFF static bool lp_write_i2c(void *handle, uint32_t address, const uint8_t *data, uint32_t size) @@ -67,6 +74,59 @@ lp_read_dpcd(void *handle, uint32_t address, uint8_t *data, uint32_t size) return dm_helpers_dp_read_dpcd(link->ctx, link, address, data, size); } +static uint8_t *psp_get_srm(struct psp_context *psp, uint32_t *srm_version, uint32_t *srm_size) +{ + + struct ta_hdcp_shared_memory *hdcp_cmd; + + if (!psp->hdcp_context.hdcp_initialized) { + DRM_WARN("Failed to get hdcp srm. HDCP TA is not initialized."); + return NULL; + } + + hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; + memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); + + hdcp_cmd->cmd_id = TA_HDCP_COMMAND__HDCP_GET_SRM; + psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); + + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) + return NULL; + + *srm_version = hdcp_cmd->out_msg.hdcp_get_srm.srm_version; + *srm_size = hdcp_cmd->out_msg.hdcp_get_srm.srm_buf_size; + + + return hdcp_cmd->out_msg.hdcp_get_srm.srm_buf; +} + +static int psp_set_srm(struct psp_context *psp, uint8_t *srm, uint32_t srm_size, uint32_t *srm_version) +{ + + struct ta_hdcp_shared_memory *hdcp_cmd; + + if (!psp->hdcp_context.hdcp_initialized) { + DRM_WARN("Failed to get hdcp srm. HDCP TA is not initialized."); + return -EINVAL; + } + + hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; + memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); + + memcpy(hdcp_cmd->in_msg.hdcp_set_srm.srm_buf, srm, srm_size); + hdcp_cmd->in_msg.hdcp_set_srm.srm_buf_size = srm_size; + hdcp_cmd->cmd_id = TA_HDCP_COMMAND__HDCP_SET_SRM; + + psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); + + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS || hdcp_cmd->out_msg.hdcp_set_srm.valid_signature != 1 || + hdcp_cmd->out_msg.hdcp_set_srm.srm_version == PSP_SRM_VERSION_MAX) + return -EINVAL; + + *srm_version = hdcp_cmd->out_msg.hdcp_set_srm.srm_version; + return 0; +} + static void process_output(struct hdcp_workqueue *hdcp_work) { struct mod_hdcp_output output = hdcp_work->output; @@ -88,6 +148,18 @@ static void process_output(struct hdcp_workqueue *hdcp_work) schedule_delayed_work(&hdcp_work->property_validate_dwork, msecs_to_jiffies(0)); } +static void link_lock(struct hdcp_workqueue *work, bool lock) +{ + + int i = 0; + + for (i = 0; i < work->max_link; i++) { + if (lock) + mutex_lock(&work[i].mutex); + else + mutex_unlock(&work[i].mutex); + } +} void hdcp_update_display(struct hdcp_workqueue *hdcp_work, unsigned int link_index, struct amdgpu_dm_connector *aconnector, @@ -112,11 +184,21 @@ void hdcp_update_display(struct hdcp_workqueue *hdcp_work, hdcp_w->link.adjust.hdcp2.force_type = MOD_HDCP_FORCE_TYPE_0; if (enable_encryption) { + /* Explicitly set the saved SRM as sysfs call will be after we already enabled hdcp + * (s3 resume case) + */ + if (hdcp_work->srm_size > 0) + psp_set_srm(hdcp_work->hdcp.config.psp.handle, hdcp_work->srm, hdcp_work->srm_size, + &hdcp_work->srm_version); + display->adjust.disable = 0; - if (content_type == DRM_MODE_HDCP_CONTENT_TYPE0) + if (content_type == DRM_MODE_HDCP_CONTENT_TYPE0) { + hdcp_w->link.adjust.hdcp1.disable = 0; hdcp_w->link.adjust.hdcp2.force_type = MOD_HDCP_FORCE_TYPE_0; - else if (content_type == DRM_MODE_HDCP_CONTENT_TYPE1) + } else if (content_type == DRM_MODE_HDCP_CONTENT_TYPE1) { + hdcp_w->link.adjust.hdcp1.disable = 1; hdcp_w->link.adjust.hdcp2.force_type = MOD_HDCP_FORCE_TYPE_1; + } schedule_delayed_work(&hdcp_w->property_validate_dwork, msecs_to_jiffies(DRM_HDCP_CHECK_PERIOD_MS)); @@ -184,7 +266,7 @@ static void event_callback(struct work_struct *work) mutex_lock(&hdcp_work->mutex); - cancel_delayed_work(&hdcp_work->watchdog_timer_dwork); + cancel_delayed_work(&hdcp_work->callback_dwork); mod_hdcp_process_event(&hdcp_work->hdcp, MOD_HDCP_EVENT_CALLBACK, &hdcp_work->output); @@ -265,6 +347,8 @@ static void event_watchdog_timer(struct work_struct *work) mutex_lock(&hdcp_work->mutex); + cancel_delayed_work(&hdcp_work->watchdog_timer_dwork); + mod_hdcp_process_event(&hdcp_work->hdcp, MOD_HDCP_EVENT_WATCHDOG_TIMEOUT, &hdcp_work->output); @@ -301,8 +385,9 @@ void hdcp_destroy(struct hdcp_workqueue *hdcp_work) cancel_delayed_work_sync(&hdcp_work[i].watchdog_timer_dwork); } + kfree(hdcp_work->srm); + kfree(hdcp_work->srm_temp); kfree(hdcp_work); - } static void update_config(void *handle, struct cp_psp_stream_config *config) @@ -332,26 +417,171 @@ static void update_config(void *handle, struct cp_psp_stream_config *config) link->dig_be = config->link_enc_inst; link->ddc_line = aconnector->dc_link->ddc_hw_inst + 1; link->dp.rev = aconnector->dc_link->dpcd_caps.dpcd_rev.raw; + link->dp.mst_supported = config->mst_supported; display->adjust.disable = 1; - link->adjust.auth_delay = 2; + link->adjust.auth_delay = 3; + link->adjust.hdcp1.disable = 0; hdcp_update_display(hdcp_work, link_index, aconnector, DRM_MODE_HDCP_CONTENT_TYPE0, false); } -struct hdcp_workqueue *hdcp_create_workqueue(void *psp_context, struct cp_psp *cp_psp, struct dc *dc) + +/* NOTE: From the usermodes prospective you only need to call write *ONCE*, the kernel + * will automatically call once or twice depending on the size + * + * call: "cat file > /sys/class/drm/card0/device/hdcp_srm" from usermode no matter what the size is + * + * The kernel can only send PAGE_SIZE at once and since MAX_SRM_FILE(5120) > PAGE_SIZE(4096), + * srm_data_write can be called multiple times. + * + * sysfs interface doesn't tell us the size we will get so we are sending partial SRMs to psp and on + * the last call we will send the full SRM. PSP will fail on every call before the last. + * + * This means we don't know if the SRM is good until the last call. And because of this limitation we + * cannot throw errors early as it will stop the kernel from writing to sysfs + * + * Example 1: + * Good SRM size = 5096 + * first call to write 4096 -> PSP fails + * Second call to write 1000 -> PSP Pass -> SRM is set + * + * Example 2: + * Bad SRM size = 4096 + * first call to write 4096 -> PSP fails (This is the same as above, but we don't know if this + * is the last call) + * + * Solution?: + * 1: Parse the SRM? -> It is signed so we don't know the EOF + * 2: We can have another sysfs that passes the size before calling set. -> simpler solution + * below + * + * Easy Solution: + * Always call get after Set to verify if set was successful. + * +----------------------+ + * | Why it works: | + * +----------------------+ + * PSP will only update its srm if its older than the one we are trying to load. + * Always do set first than get. + * -if we try to "1. SET" a older version PSP will reject it and we can "2. GET" the newer + * version and save it + * + * -if we try to "1. SET" a newer version PSP will accept it and we can "2. GET" the + * same(newer) version back and save it + * + * -if we try to "1. SET" a newer version and PSP rejects it. That means the format is + * incorrect/corrupted and we should correct our SRM by getting it from PSP + */ +static ssize_t srm_data_write(struct file *filp, struct kobject *kobj, struct bin_attribute *bin_attr, char *buffer, + loff_t pos, size_t count) +{ + struct hdcp_workqueue *work; + uint32_t srm_version = 0; + + work = container_of(bin_attr, struct hdcp_workqueue, attr); + link_lock(work, true); + + memcpy(work->srm_temp + pos, buffer, count); + + if (!psp_set_srm(work->hdcp.config.psp.handle, work->srm_temp, pos + count, &srm_version)) { + DRM_DEBUG_DRIVER("HDCP SRM SET version 0x%X", srm_version); + memcpy(work->srm, work->srm_temp, pos + count); + work->srm_size = pos + count; + work->srm_version = srm_version; + } + + + link_lock(work, false); + + return count; +} + +static ssize_t srm_data_read(struct file *filp, struct kobject *kobj, struct bin_attribute *bin_attr, char *buffer, + loff_t pos, size_t count) +{ + struct hdcp_workqueue *work; + uint8_t *srm = NULL; + uint32_t srm_version; + uint32_t srm_size; + size_t ret = count; + + work = container_of(bin_attr, struct hdcp_workqueue, attr); + + link_lock(work, true); + + srm = psp_get_srm(work->hdcp.config.psp.handle, &srm_version, &srm_size); + + if (!srm) + return -EINVAL; + + if (pos >= srm_size) + ret = 0; + + if (srm_size - pos < count) { + memcpy(buffer, srm + pos, srm_size - pos); + ret = srm_size - pos; + goto ret; + } + + memcpy(buffer, srm + pos, count); + +ret: + link_lock(work, false); + return ret; +} + +/* From the hdcp spec (5.Renewability) SRM needs to be stored in a non-volatile memory. + * + * For example, + * if Application "A" sets the SRM (ver 2) and we reboot/suspend and later when Application "B" + * needs to use HDCP, the version in PSP should be SRM(ver 2). So SRM should be persistent + * across boot/reboots/suspend/resume/shutdown + * + * Currently when the system goes down (suspend/shutdown) the SRM is cleared from PSP. For HDCP we need + * to make the SRM persistent. + * + * -PSP owns the checking of SRM but doesn't have the ability to store it in a non-volatile memory. + * -The kernel cannot write to the file systems. + * -So we need usermode to do this for us, which is why an interface for usermode is needed + * + * + * + * Usermode can read/write to/from PSP using the sysfs interface + * For example: + * to save SRM from PSP to storage : cat /sys/class/drm/card0/device/hdcp_srm > srmfile + * to load from storage to PSP: cat srmfile > /sys/class/drm/card0/device/hdcp_srm + */ +static const struct bin_attribute data_attr = { + .attr = {.name = "hdcp_srm", .mode = 0664}, + .size = PSP_HDCP_SRM_FIRST_GEN_MAX_SIZE, /* Limit SRM size */ + .write = srm_data_write, + .read = srm_data_read, +}; + + +struct hdcp_workqueue *hdcp_create_workqueue(struct amdgpu_device *adev, struct cp_psp *cp_psp, struct dc *dc) { int max_caps = dc->caps.max_links; - struct hdcp_workqueue *hdcp_work = kzalloc(max_caps*sizeof(*hdcp_work), GFP_KERNEL); + struct hdcp_workqueue *hdcp_work; int i = 0; + hdcp_work = kcalloc(max_caps, sizeof(*hdcp_work), GFP_KERNEL); if (hdcp_work == NULL) + return NULL; + + hdcp_work->srm = kcalloc(PSP_HDCP_SRM_FIRST_GEN_MAX_SIZE, sizeof(*hdcp_work->srm), GFP_KERNEL); + + if (hdcp_work->srm == NULL) + goto fail_alloc_context; + + hdcp_work->srm_temp = kcalloc(PSP_HDCP_SRM_FIRST_GEN_MAX_SIZE, sizeof(*hdcp_work->srm_temp), GFP_KERNEL); + + if (hdcp_work->srm_temp == NULL) goto fail_alloc_context; hdcp_work->max_link = max_caps; for (i = 0; i < max_caps; i++) { - mutex_init(&hdcp_work[i].mutex); INIT_WORK(&hdcp_work[i].cpirq_work, event_cpirq); @@ -360,7 +590,7 @@ struct hdcp_workqueue *hdcp_create_workqueue(void *psp_context, struct cp_psp *c INIT_DELAYED_WORK(&hdcp_work[i].watchdog_timer_dwork, event_watchdog_timer); INIT_DELAYED_WORK(&hdcp_work[i].property_validate_dwork, event_property_validate); - hdcp_work[i].hdcp.config.psp.handle = psp_context; + hdcp_work[i].hdcp.config.psp.handle = &adev->psp; hdcp_work[i].hdcp.config.ddc.handle = dc_get_link_at_index(dc, i); hdcp_work[i].hdcp.config.ddc.funcs.write_i2c = lp_write_i2c; hdcp_work[i].hdcp.config.ddc.funcs.read_i2c = lp_read_i2c; @@ -371,9 +601,17 @@ struct hdcp_workqueue *hdcp_create_workqueue(void *psp_context, struct cp_psp *c cp_psp->funcs.update_stream_config = update_config; cp_psp->handle = hdcp_work; + /* File created at /sys/class/drm/card0/device/hdcp_srm*/ + hdcp_work[0].attr = data_attr; + + if (sysfs_create_bin_file(&adev->dev->kobj, &hdcp_work[0].attr)) + DRM_WARN("Failed to create device file hdcp_srm"); + return hdcp_work; fail_alloc_context: + kfree(hdcp_work->srm); + kfree(hdcp_work->srm_temp); kfree(hdcp_work); return NULL; diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.h index 6abde86bce4a..5159b3a5e5b0 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.h @@ -30,6 +30,7 @@ #include "hdcp.h" #include "dc.h" #include "dm_cp_psp.h" +#include "amdgpu.h" struct mod_hdcp; struct mod_hdcp_link; @@ -52,6 +53,12 @@ struct hdcp_workqueue { enum mod_hdcp_encryption_status encryption_status; uint8_t max_link; + + uint8_t *srm; + uint8_t *srm_temp; + uint32_t srm_version; + uint32_t srm_size; + struct bin_attribute attr; }; void hdcp_update_display(struct hdcp_workqueue *hdcp_work, @@ -64,6 +71,6 @@ void hdcp_reset_display(struct hdcp_workqueue *work, unsigned int link_index); void hdcp_handle_cpirq(struct hdcp_workqueue *work, unsigned int link_index); void hdcp_destroy(struct hdcp_workqueue *work); -struct hdcp_workqueue *hdcp_create_workqueue(void *psp_context, struct cp_psp *cp_psp, struct dc *dc); +struct hdcp_workqueue *hdcp_create_workqueue(struct amdgpu_device *adev, struct cp_psp *cp_psp, struct dc *dc); #endif /* AMDGPU_DM_AMDGPU_DM_HDCP_H_ */ diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c index 318b474ff20e..c20fb08c450b 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c @@ -400,8 +400,8 @@ bool dm_helpers_dp_mst_start_top_mgr( struct amdgpu_dm_connector *aconnector = link->priv; if (!aconnector) { - DRM_ERROR("Failed to found connector for link!"); - return false; + DRM_ERROR("Failed to find connector for link!"); + return false; } if (boot) { @@ -423,8 +423,8 @@ void dm_helpers_dp_mst_stop_top_mgr( struct amdgpu_dm_connector *aconnector = link->priv; if (!aconnector) { - DRM_ERROR("Failed to found connector for link!"); - return; + DRM_ERROR("Failed to find connector for link!"); + return; } DRM_INFO("DM_MST: stopping TM on aconnector: %p [id: %d]\n", @@ -445,7 +445,7 @@ bool dm_helpers_dp_read_dpcd( struct amdgpu_dm_connector *aconnector = link->priv; if (!aconnector) { - DRM_ERROR("Failed to found connector for link!"); + DRM_ERROR("Failed to find connector for link!"); return false; } @@ -463,7 +463,7 @@ bool dm_helpers_dp_write_dpcd( struct amdgpu_dm_connector *aconnector = link->priv; if (!aconnector) { - DRM_ERROR("Failed to found connector for link!"); + DRM_ERROR("Failed to find connector for link!"); return false; } @@ -483,7 +483,7 @@ bool dm_helpers_submit_i2c( bool result; if (!aconnector) { - DRM_ERROR("Failed to found connector for link!"); + DRM_ERROR("Failed to find connector for link!"); return false; } @@ -538,7 +538,7 @@ bool dm_helpers_is_dp_sink_present(struct dc_link *link) struct amdgpu_dm_connector *aconnector = link->priv; if (!aconnector) { - BUG_ON("Failed to found connector for link!"); + BUG_ON("Failed to find connector for link!"); return true; } @@ -580,6 +580,20 @@ enum dc_edid_status dm_helpers_read_local_edid( /* We don't need the original edid anymore */ kfree(edid); + /* connector->display_info will be parsed from EDID and saved + * into drm_connector->display_info from edid by call stack + * below: + * drm_parse_ycbcr420_deep_color_info + * drm_parse_hdmi_forum_vsdb + * drm_parse_cea_ext + * drm_add_display_info + * drm_connector_update_edid_property + * + * drm_connector->display_info will be used by amdgpu_dm funcs, + * like fill_stream_properties_from_drm_display_mode + */ + amdgpu_dm_update_connector_after_detect(aconnector); + edid_status = dm_helpers_parse_edid_caps( ctx, &sink->dc_edid, diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c index 5672f7765919..fabbe78d5aef 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c @@ -154,15 +154,18 @@ amdgpu_dm_mst_connector_late_register(struct drm_connector *connector) { struct amdgpu_dm_connector *amdgpu_dm_connector = to_amdgpu_dm_connector(connector); - struct drm_dp_mst_port *port = amdgpu_dm_connector->port; + int r; + + amdgpu_dm_connector->dm_dp_aux.aux.dev = connector->kdev; + r = drm_dp_aux_register(&amdgpu_dm_connector->dm_dp_aux.aux); + if (r) + return r; #if defined(CONFIG_DEBUG_FS) connector_debugfs_init(amdgpu_dm_connector); - amdgpu_dm_connector->debugfs_dpcd_address = 0; - amdgpu_dm_connector->debugfs_dpcd_size = 0; #endif - return drm_dp_mst_connector_late_register(connector, port); + return r; } static void @@ -204,7 +207,7 @@ static bool validate_dsc_caps_on_connector(struct amdgpu_dm_connector *aconnecto if (!dc_dsc_parse_dsc_dpcd(aconnector->dc_link->ctx->dc, dsc_caps, NULL, - &dc_sink->sink_dsc_caps.dsc_dec_caps)) + &dc_sink->dsc_caps.dsc_dec_caps)) return false; return true; @@ -259,8 +262,8 @@ static int dm_dp_mst_get_modes(struct drm_connector *connector) #if defined(CONFIG_DRM_AMD_DC_DCN) if (!validate_dsc_caps_on_connector(aconnector)) - memset(&aconnector->dc_sink->sink_dsc_caps, - 0, sizeof(aconnector->dc_sink->sink_dsc_caps)); + memset(&aconnector->dc_sink->dsc_caps, + 0, sizeof(aconnector->dc_sink->dsc_caps)); #endif } } @@ -407,6 +410,14 @@ dm_dp_add_mst_connector(struct drm_dp_mst_topology_mgr *mgr, drm_connector_attach_encoder(&aconnector->base, &aconnector->mst_encoder->base); + connector->max_bpc_property = master->base.max_bpc_property; + if (connector->max_bpc_property) + drm_connector_attach_max_bpc_property(connector, 8, 16); + + connector->vrr_capable_property = master->base.vrr_capable_property; + if (connector->vrr_capable_property) + drm_connector_attach_vrr_capable_property(connector); + drm_object_attach_property( &connector->base, dev->mode_config.path_property, @@ -437,9 +448,6 @@ dm_dp_add_mst_connector(struct drm_dp_mst_topology_mgr *mgr, static void dm_dp_destroy_mst_connector(struct drm_dp_mst_topology_mgr *mgr, struct drm_connector *connector) { - struct amdgpu_dm_connector *master = container_of(mgr, struct amdgpu_dm_connector, mst_mgr); - struct drm_device *dev = master->base.dev; - struct amdgpu_device *adev = dev->dev_private; struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector); DRM_INFO("DM_MST: Disabling connector: %p [id: %d] [master: %p]\n", @@ -451,42 +459,26 @@ static void dm_dp_destroy_mst_connector(struct drm_dp_mst_topology_mgr *mgr, aconnector->dc_sink); dc_sink_release(aconnector->dc_sink); aconnector->dc_sink = NULL; + aconnector->dc_link->cur_link_settings.lane_count = 0; } drm_connector_unregister(connector); - if (adev->mode_info.rfbdev) - drm_fb_helper_remove_one_connector(&adev->mode_info.rfbdev->helper, connector); drm_connector_put(connector); } -static void dm_dp_mst_register_connector(struct drm_connector *connector) -{ - struct drm_device *dev = connector->dev; - struct amdgpu_device *adev = dev->dev_private; - - if (adev->mode_info.rfbdev) - drm_fb_helper_add_one_connector(&adev->mode_info.rfbdev->helper, connector); - else - DRM_ERROR("adev->mode_info.rfbdev is NULL\n"); - - drm_connector_register(connector); -} - static const struct drm_dp_mst_topology_cbs dm_mst_cbs = { .add_connector = dm_dp_add_mst_connector, .destroy_connector = dm_dp_destroy_mst_connector, - .register_connector = dm_dp_mst_register_connector }; void amdgpu_dm_initialize_dp_connector(struct amdgpu_display_manager *dm, struct amdgpu_dm_connector *aconnector) { aconnector->dm_dp_aux.aux.name = "dmdc"; - aconnector->dm_dp_aux.aux.dev = aconnector->base.kdev; aconnector->dm_dp_aux.aux.transfer = dm_dp_aux_transfer; aconnector->dm_dp_aux.ddc_service = aconnector->dc_link->ddc; - drm_dp_aux_register(&aconnector->dm_dp_aux.aux); + drm_dp_aux_init(&aconnector->dm_dp_aux.aux); drm_dp_cec_register_connector(&aconnector->dm_dp_aux.aux, &aconnector->base); @@ -547,7 +539,7 @@ static void set_dsc_configs_from_fairness_vars(struct dsc_mst_fairness_params *p memset(¶ms[i].timing->dsc_cfg, 0, sizeof(params[i].timing->dsc_cfg)); if (vars[i].dsc_enabled && dc_dsc_compute_config( params[i].sink->ctx->dc->res_pool->dscs[0], - ¶ms[i].sink->sink_dsc_caps.dsc_dec_caps, + ¶ms[i].sink->dsc_caps.dsc_dec_caps, params[i].sink->ctx->dc->debug.dsc_min_slice_height_override, 0, params[i].timing, @@ -568,7 +560,7 @@ static int bpp_x16_from_pbn(struct dsc_mst_fairness_params param, int pbn) kbps = div_u64((u64)pbn * 994 * 8 * 54, 64); dc_dsc_compute_config( param.sink->ctx->dc->res_pool->dscs[0], - ¶m.sink->sink_dsc_caps.dsc_dec_caps, + ¶m.sink->dsc_caps.dsc_dec_caps, param.sink->ctx->dc->debug.dsc_min_slice_height_override, (int) kbps, param.timing, &dsc_config); @@ -765,14 +757,14 @@ static bool compute_mst_dsc_configs_for_link(struct drm_atomic_state *state, params[count].sink = stream->sink; aconnector = (struct amdgpu_dm_connector *)stream->dm_stream_context; params[count].port = aconnector->port; - params[count].compression_possible = stream->sink->sink_dsc_caps.dsc_dec_caps.is_dsc_supported; + params[count].compression_possible = stream->sink->dsc_caps.dsc_dec_caps.is_dsc_supported; dc_dsc_get_policy_for_timing(params[count].timing, &dsc_policy); if (!dc_dsc_compute_bandwidth_range( stream->sink->ctx->dc->res_pool->dscs[0], stream->sink->ctx->dc->debug.dsc_min_slice_height_override, dsc_policy.min_target_bpp, dsc_policy.max_target_bpp, - &stream->sink->sink_dsc_caps.dsc_dec_caps, + &stream->sink->dsc_caps.dsc_dec_caps, &stream->timing, ¶ms[count].bw_range)) params[count].bw_range.stream_kbps = dc_bandwidth_in_kbps_from_timing(&stream->timing); @@ -854,7 +846,7 @@ bool compute_mst_dsc_configs_for_state(struct drm_atomic_state *state, if (!aconnector || !aconnector->dc_sink) continue; - if (!aconnector->dc_sink->sink_dsc_caps.dsc_dec_caps.is_dsc_supported) + if (!aconnector->dc_sink->dsc_caps.dsc_dec_caps.is_dsc_supported) continue; if (computed_streams[i]) diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c index 2f1c9584ac32..37fa7b48250e 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c +++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c @@ -267,7 +267,7 @@ static struct atom_display_object_path_v2 *get_bios_object( && id.enum_id == obj_id.enum_id) return &bp->object_info_tbl.v1_4->display_path[i]; } - /* fall through */ + fallthrough; case OBJECT_TYPE_CONNECTOR: case OBJECT_TYPE_GENERIC: /* Both Generic and Connector Object ID @@ -280,7 +280,7 @@ static struct atom_display_object_path_v2 *get_bios_object( && id.enum_id == obj_id.enum_id) return &bp->object_info_tbl.v1_4->display_path[i]; } - /* fall through */ + fallthrough; default: return NULL; } diff --git a/drivers/gpu/drm/amd/display/dc/bios/command_table2.c b/drivers/gpu/drm/amd/display/dc/bios/command_table2.c index c4ba6e84db65..8edc2506d49e 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/command_table2.c +++ b/drivers/gpu/drm/amd/display/dc/bios/command_table2.c @@ -221,8 +221,8 @@ static void init_transmitter_control(struct bios_parser *bp) uint8_t frev; uint8_t crev; - if (BIOS_CMD_TABLE_REVISION(dig1transmittercontrol, frev, crev) == false) - BREAK_TO_DEBUGGER(); + BIOS_CMD_TABLE_REVISION(dig1transmittercontrol, frev, crev); + switch (crev) { case 6: bp->cmd_tbl.transmitter_control = transmitter_control_v1_6; diff --git a/drivers/gpu/drm/amd/display/dc/bios/command_table_helper2.c b/drivers/gpu/drm/amd/display/dc/bios/command_table_helper2.c index 7388c987c595..204d7942a6e5 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/command_table_helper2.c +++ b/drivers/gpu/drm/amd/display/dc/bios/command_table_helper2.c @@ -53,25 +53,18 @@ bool dal_bios_parser_init_cmd_tbl_helper2( case DCE_VERSION_11_2: case DCE_VERSION_11_22: + case DCE_VERSION_12_0: + case DCE_VERSION_12_1: *h = dal_cmd_tbl_helper_dce112_get_table2(); return true; #if defined(CONFIG_DRM_AMD_DC_DCN) case DCN_VERSION_1_0: case DCN_VERSION_1_01: - *h = dal_cmd_tbl_helper_dce112_get_table2(); - return true; -#endif - case DCN_VERSION_2_0: - *h = dal_cmd_tbl_helper_dce112_get_table2(); - return true; case DCN_VERSION_2_1: *h = dal_cmd_tbl_helper_dce112_get_table2(); return true; - case DCE_VERSION_12_0: - case DCE_VERSION_12_1: - *h = dal_cmd_tbl_helper_dce112_get_table2(); - return true; +#endif default: /* Unsupported DCE */ diff --git a/drivers/gpu/drm/amd/display/dc/calcs/dce_calcs.c b/drivers/gpu/drm/amd/display/dc/calcs/dce_calcs.c index 5d081c42e81b..2c6db379afae 100644 --- a/drivers/gpu/drm/amd/display/dc/calcs/dce_calcs.c +++ b/drivers/gpu/drm/amd/display/dc/calcs/dce_calcs.c @@ -3265,33 +3265,33 @@ bool bw_calcs(struct dc_context *ctx, bw_fixed_to_int(bw_mul(data-> stutter_exit_watermark[9], bw_int_to_fixed(1000))); - calcs_output->stutter_entry_wm_ns[0].b_mark = - bw_fixed_to_int(bw_mul(data-> - stutter_entry_watermark[4], bw_int_to_fixed(1000))); - calcs_output->stutter_entry_wm_ns[1].b_mark = - bw_fixed_to_int(bw_mul(data-> - stutter_entry_watermark[5], bw_int_to_fixed(1000))); - calcs_output->stutter_entry_wm_ns[2].b_mark = - bw_fixed_to_int(bw_mul(data-> - stutter_entry_watermark[6], bw_int_to_fixed(1000))); - if (ctx->dc->caps.max_slave_planes) { - calcs_output->stutter_entry_wm_ns[3].b_mark = + calcs_output->stutter_entry_wm_ns[0].b_mark = bw_fixed_to_int(bw_mul(data-> - stutter_entry_watermark[0], bw_int_to_fixed(1000))); - calcs_output->stutter_entry_wm_ns[4].b_mark = + stutter_entry_watermark[4], bw_int_to_fixed(1000))); + calcs_output->stutter_entry_wm_ns[1].b_mark = bw_fixed_to_int(bw_mul(data-> - stutter_entry_watermark[1], bw_int_to_fixed(1000))); - } else { - calcs_output->stutter_entry_wm_ns[3].b_mark = + stutter_entry_watermark[5], bw_int_to_fixed(1000))); + calcs_output->stutter_entry_wm_ns[2].b_mark = bw_fixed_to_int(bw_mul(data-> - stutter_entry_watermark[7], bw_int_to_fixed(1000))); - calcs_output->stutter_entry_wm_ns[4].b_mark = + stutter_entry_watermark[6], bw_int_to_fixed(1000))); + if (ctx->dc->caps.max_slave_planes) { + calcs_output->stutter_entry_wm_ns[3].b_mark = + bw_fixed_to_int(bw_mul(data-> + stutter_entry_watermark[0], bw_int_to_fixed(1000))); + calcs_output->stutter_entry_wm_ns[4].b_mark = + bw_fixed_to_int(bw_mul(data-> + stutter_entry_watermark[1], bw_int_to_fixed(1000))); + } else { + calcs_output->stutter_entry_wm_ns[3].b_mark = + bw_fixed_to_int(bw_mul(data-> + stutter_entry_watermark[7], bw_int_to_fixed(1000))); + calcs_output->stutter_entry_wm_ns[4].b_mark = + bw_fixed_to_int(bw_mul(data-> + stutter_entry_watermark[8], bw_int_to_fixed(1000))); + } + calcs_output->stutter_entry_wm_ns[5].b_mark = bw_fixed_to_int(bw_mul(data-> - stutter_entry_watermark[8], bw_int_to_fixed(1000))); - } - calcs_output->stutter_entry_wm_ns[5].b_mark = - bw_fixed_to_int(bw_mul(data-> - stutter_entry_watermark[9], bw_int_to_fixed(1000))); + stutter_entry_watermark[9], bw_int_to_fixed(1000))); calcs_output->urgent_wm_ns[0].b_mark = bw_fixed_to_int(bw_mul(data-> diff --git a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c b/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c index 1a37550731de..3960a8db94cb 100644 --- a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c +++ b/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c @@ -703,11 +703,24 @@ static void hack_bounding_box(struct dcn_bw_internal_vars *v, } -unsigned int get_highest_allowed_voltage_level(uint32_t hw_internal_rev) +unsigned int get_highest_allowed_voltage_level(uint32_t hw_internal_rev, uint32_t pci_revision_id) { - /* for dali & pollock, the highest voltage level we want is 0 */ - if (ASICREV_IS_POLLOCK(hw_internal_rev) || ASICREV_IS_DALI(hw_internal_rev)) - return 0; + /* for low power RV2 variants, the highest voltage level we want is 0 */ + if (ASICREV_IS_RAVEN2(hw_internal_rev)) + switch (pci_revision_id) { + case PRID_DALI_DE: + case PRID_DALI_DF: + case PRID_DALI_E3: + case PRID_DALI_E4: + case PRID_POLLOCK_94: + case PRID_POLLOCK_95: + case PRID_POLLOCK_E9: + case PRID_POLLOCK_EA: + case PRID_POLLOCK_EB: + return 0; + default: + break; + } /* we are ok with all levels */ return 4; @@ -1277,7 +1290,9 @@ bool dcn_validate_bandwidth( PERFORMANCE_TRACE_END(); BW_VAL_TRACE_FINISH(); - if (bw_limit_pass && v->voltage_level <= get_highest_allowed_voltage_level(dc->ctx->asic_id.hw_internal_rev)) + if (bw_limit_pass && v->voltage_level <= get_highest_allowed_voltage_level( + dc->ctx->asic_id.hw_internal_rev, + dc->ctx->asic_id.pci_revision_id)) return true; else return false; diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c index a78e5c74c79c..8ec2dfe45d40 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c @@ -63,6 +63,25 @@ int clk_mgr_helper_get_active_display_cnt( return display_count; } +int clk_mgr_helper_get_active_plane_cnt( + struct dc *dc, + struct dc_state *context) +{ + int i, total_plane_count; + + total_plane_count = 0; + for (i = 0; i < context->stream_count; i++) { + const struct dc_stream_status stream_status = context->stream_status[i]; + + /* + * Sum up plane_count for all streams ( active and virtual ). + */ + total_plane_count += stream_status.plane_count; + } + + return total_plane_count; +} + void clk_mgr_exit_optimized_pwr_state(const struct dc *dc, struct clk_mgr *clk_mgr) { struct dc_link *edp_link = get_edp_link(dc); @@ -134,13 +153,6 @@ struct clk_mgr *dc_clk_mgr_create(struct dc_context *ctx, struct pp_smu_funcs *p #if defined(CONFIG_DRM_AMD_DC_DCN) case FAMILY_RV: - if (ASICREV_IS_DALI(asic_id.hw_internal_rev) || - ASICREV_IS_POLLOCK(asic_id.hw_internal_rev)) { - /* TEMP: this check has to come before ASICREV_IS_RENOIR */ - /* which also incorrectly returns true for Dali/Pollock*/ - rv2_clk_mgr_construct(ctx, clk_mgr, pp_smu); - break; - } if (ASICREV_IS_RENOIR(asic_id.hw_internal_rev)) { rn_clk_mgr_construct(ctx, clk_mgr, pp_smu, dccg); break; diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c index 49ce46b543ea..55d09adbf0d9 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c @@ -115,12 +115,11 @@ void dcn20_update_clocks_update_dpp_dto(struct clk_mgr_internal *clk_mgr, dpp_inst = i; dppclk_khz = context->res_ctx.pipe_ctx[i].plane_res.bw.dppclk_khz; - prev_dppclk_khz = clk_mgr->base.ctx->dc->current_state->res_ctx.pipe_ctx[i].plane_res.bw.dppclk_khz; + prev_dppclk_khz = clk_mgr->dccg->pipe_dppclk_khz[i]; - if ((prev_dppclk_khz > dppclk_khz && safe_to_lower) || prev_dppclk_khz < dppclk_khz) { + if (safe_to_lower || prev_dppclk_khz < dppclk_khz) clk_mgr->dccg->funcs->update_dpp_dto( clk_mgr->dccg, dpp_inst, dppclk_khz); - } } } @@ -158,6 +157,8 @@ void dcn2_update_clocks(struct clk_mgr *clk_mgr_base, bool dpp_clock_lowered = false; struct dmcu *dmcu = clk_mgr_base->ctx->dc->res_pool->dmcu; bool force_reset = false; + bool p_state_change_support; + int total_plane_count; if (dc->work_arounds.skip_clock_update) return; @@ -213,9 +214,11 @@ void dcn2_update_clocks(struct clk_mgr *clk_mgr_base, pp_smu->set_hard_min_socclk_by_freq(&pp_smu->pp_smu, clk_mgr_base->clks.socclk_khz / 1000); } - if (should_update_pstate_support(safe_to_lower, new_clocks->p_state_change_support, clk_mgr_base->clks.p_state_change_support)) { + total_plane_count = clk_mgr_helper_get_active_plane_cnt(dc, context); + p_state_change_support = new_clocks->p_state_change_support || (total_plane_count == 0); + if (should_update_pstate_support(safe_to_lower, p_state_change_support, clk_mgr_base->clks.p_state_change_support)) { clk_mgr_base->clks.prev_p_state_change_support = clk_mgr_base->clks.p_state_change_support; - clk_mgr_base->clks.p_state_change_support = new_clocks->p_state_change_support; + clk_mgr_base->clks.p_state_change_support = p_state_change_support; if (pp_smu && pp_smu->set_pstate_handshake_support) pp_smu->set_pstate_handshake_support(&pp_smu->pp_smu, clk_mgr_base->clks.p_state_change_support); } diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c index 9ef3f7b91a1d..24c5765890fa 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c @@ -46,6 +46,7 @@ /* Constants */ #define LPDDR_MEM_RETRAIN_LATENCY 4.977 /* Number obtained from LPDDR4 Training Counter Requirement doc */ +#define SMU_VER_55_51_0 0x373300 /* SMU Version that is able to set DISPCLK below 100MHz */ /* Macros */ @@ -405,7 +406,7 @@ void rn_init_clocks(struct clk_mgr *clk_mgr) clk_mgr->clks.pwr_state = DCN_PWR_STATE_UNKNOWN; } -void build_watermark_ranges(struct clk_bw_params *bw_params, struct pp_smu_wm_range_sets *ranges) +static void build_watermark_ranges(struct clk_bw_params *bw_params, struct pp_smu_wm_range_sets *ranges) { int i, num_valid_sets; @@ -465,16 +466,15 @@ void build_watermark_ranges(struct clk_bw_params *bw_params, struct pp_smu_wm_ra static void rn_notify_wm_ranges(struct clk_mgr *clk_mgr_base) { struct dc_debug_options *debug = &clk_mgr_base->ctx->dc->debug; - struct pp_smu_wm_range_sets ranges = {0}; struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base); struct pp_smu_funcs *pp_smu = clk_mgr->pp_smu; if (!debug->disable_pplib_wm_range) { - build_watermark_ranges(clk_mgr_base->bw_params, &ranges); + build_watermark_ranges(clk_mgr_base->bw_params, &clk_mgr_base->ranges); /* Notify PP Lib/SMU which Watermarks to use for which clock ranges */ if (pp_smu && pp_smu->rn_funcs.set_wm_ranges) - pp_smu->rn_funcs.set_wm_ranges(&pp_smu->rn_funcs.pp_smu, &ranges); + pp_smu->rn_funcs.set_wm_ranges(&pp_smu->rn_funcs.pp_smu, &clk_mgr_base->ranges); } } @@ -504,7 +504,7 @@ static struct clk_mgr_funcs dcn21_funcs = { .notify_wm_ranges = rn_notify_wm_ranges }; -struct clk_bw_params rn_bw_params = { +static struct clk_bw_params rn_bw_params = { .vram_type = Ddr4MemType, .num_channels = 1, .clk_table = { @@ -544,7 +544,7 @@ struct clk_bw_params rn_bw_params = { }; -struct wm_table ddr4_wm_table = { +static struct wm_table ddr4_wm_table = { .entries = { { .wm_inst = WM_A, @@ -581,7 +581,7 @@ struct wm_table ddr4_wm_table = { } }; -struct wm_table lpddr4_wm_table = { +static struct wm_table lpddr4_wm_table = { .entries = { { .wm_inst = WM_A, @@ -643,7 +643,7 @@ static void rn_clk_mgr_helper_populate_bw_params(struct clk_bw_params *bw_params /* Find lowest DPM, FCLK is filled in reverse order*/ for (i = PP_SMU_NUM_FCLK_DPM_LEVELS - 1; i >= 0; i--) { - if (clock_table->FClocks[i].Freq != 0) { + if (clock_table->FClocks[i].Freq != 0 && clock_table->FClocks[i].Vol != 0) { j = i; break; } @@ -721,6 +721,13 @@ void rn_clk_mgr_construct( } else { struct clk_log_info log_info = {0}; + clk_mgr->smu_ver = rn_vbios_smu_get_smu_version(clk_mgr); + + /* SMU Version 55.51.0 and up no longer have an issue + * that needs to limit minimum dispclk */ + if (clk_mgr->smu_ver >= SMU_VER_55_51_0) + debug->min_disp_clk_khz = 0; + /* TODO: Check we get what we expect during bringup */ clk_mgr->base.dentist_vco_freq_khz = get_vco_frequency_from_reg(clk_mgr); diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c index 04441dbcba76..8489f1e56892 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc.c @@ -283,6 +283,8 @@ bool dc_stream_adjust_vmin_vmax(struct dc *dc, int i = 0; bool ret = false; + stream->adjust = *adjust; + for (i = 0; i < MAX_PIPES; i++) { struct pipe_ctx *pipe = &dc->current_state->res_ctx.pipe_ctx[i]; @@ -701,7 +703,7 @@ static bool dc_construct(struct dc *dc, dc_ctx->created_bios = true; } - + dc->vendor_signature = init_params->vendor_signature; /* Create GPIO service */ dc_ctx->gpio_service = dal_gpio_service_create( @@ -761,6 +763,28 @@ static bool disable_all_writeback_pipes_for_stream( return true; } +void apply_ctx_interdependent_lock(struct dc *dc, struct dc_state *context, struct dc_stream_state *stream, bool lock) +{ + int i = 0; + + /* Checks if interdependent update function pointer is NULL or not, takes care of DCE110 case */ + if (dc->hwss.interdependent_update_lock) + dc->hwss.interdependent_update_lock(dc, context, lock); + else { + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; + struct pipe_ctx *old_pipe_ctx = &dc->current_state->res_ctx.pipe_ctx[i]; + + // Copied conditions that were previously in dce110_apply_ctx_for_surface + if (stream == pipe_ctx->stream) { + if (!pipe_ctx->top_pipe && + (pipe_ctx->plane_state || old_pipe_ctx->plane_state)) + dc->hwss.pipe_control_lock(dc, pipe_ctx, lock); + } + } + } +} + static void disable_dangling_plane(struct dc *dc, struct dc_state *context) { int i, j; @@ -786,11 +810,20 @@ static void disable_dangling_plane(struct dc *dc, struct dc_state *context) if (should_disable && old_stream) { dc_rem_all_planes_for_stream(dc, old_stream, dangling_context); disable_all_writeback_pipes_for_stream(dc, old_stream, dangling_context); - if (dc->hwss.apply_ctx_for_surface) + + if (dc->hwss.apply_ctx_for_surface) { + apply_ctx_interdependent_lock(dc, dc->current_state, old_stream, true); dc->hwss.apply_ctx_for_surface(dc, old_stream, 0, dangling_context); + apply_ctx_interdependent_lock(dc, dc->current_state, old_stream, false); + dc->hwss.post_unlock_program_front_end(dc, dangling_context); + } + if (dc->hwss.program_front_end_for_ctx) { + dc->hwss.interdependent_update_lock(dc, dc->current_state, true); + dc->hwss.program_front_end_for_ctx(dc, dangling_context); + dc->hwss.interdependent_update_lock(dc, dc->current_state, false); + dc->hwss.post_unlock_program_front_end(dc, dangling_context); + } } - if (dc->hwss.program_front_end_for_ctx) - dc->hwss.program_front_end_for_ctx(dc, dangling_context); } current_ctx = dc->current_state; @@ -1210,16 +1243,19 @@ static enum dc_status dc_commit_state_no_check(struct dc *dc, struct dc_state *c /* re-program planes for existing stream, in case we need to * free up plane resource for later use */ - if (dc->hwss.apply_ctx_for_surface) + if (dc->hwss.apply_ctx_for_surface) { for (i = 0; i < context->stream_count; i++) { if (context->streams[i]->mode_changed) continue; - + apply_ctx_interdependent_lock(dc, context, context->streams[i], true); dc->hwss.apply_ctx_for_surface( dc, context->streams[i], context->stream_status[i].plane_count, context); /* use new pipe config in new context */ + apply_ctx_interdependent_lock(dc, context, context->streams[i], false); + dc->hwss.post_unlock_program_front_end(dc, context); } + } /* Program hardware */ for (i = 0; i < dc->res_pool->pipe_count; i++) { @@ -1238,19 +1274,27 @@ static enum dc_status dc_commit_state_no_check(struct dc *dc, struct dc_state *c } /* Program all planes within new context*/ - if (dc->hwss.program_front_end_for_ctx) + if (dc->hwss.program_front_end_for_ctx) { + dc->hwss.interdependent_update_lock(dc, context, true); dc->hwss.program_front_end_for_ctx(dc, context); + dc->hwss.interdependent_update_lock(dc, context, false); + dc->hwss.post_unlock_program_front_end(dc, context); + } for (i = 0; i < context->stream_count; i++) { const struct dc_link *link = context->streams[i]->link; if (!context->streams[i]->mode_changed) continue; - if (dc->hwss.apply_ctx_for_surface) + if (dc->hwss.apply_ctx_for_surface) { + apply_ctx_interdependent_lock(dc, context, context->streams[i], true); dc->hwss.apply_ctx_for_surface( dc, context->streams[i], context->stream_status[i].plane_count, context); + apply_ctx_interdependent_lock(dc, context, context->streams[i], false); + dc->hwss.post_unlock_program_front_end(dc, context); + } /* * enable stereo @@ -1318,10 +1362,24 @@ bool dc_commit_state(struct dc *dc, struct dc_state *context) return (result == DC_OK); } -bool dc_is_hw_initialized(struct dc *dc) +static bool is_flip_pending_in_pipes(struct dc *dc, struct dc_state *context) { - struct dc_bios *dcb = dc->ctx->dc_bios; - return dcb->funcs->is_accelerated_mode(dcb); + int i; + struct pipe_ctx *pipe; + + for (i = 0; i < MAX_PIPES; i++) { + pipe = &context->res_ctx.pipe_ctx[i]; + + if (!pipe->plane_state) + continue; + + /* Must set to false to start with, due to OR in update function */ + pipe->plane_state->status.is_flip_pending = false; + dc->hwss.update_pending_status(pipe); + if (pipe->plane_state->status.is_flip_pending) + return true; + } + return false; } bool dc_post_update_surfaces_to_stream(struct dc *dc) @@ -1329,11 +1387,14 @@ bool dc_post_update_surfaces_to_stream(struct dc *dc) int i; struct dc_state *context = dc->current_state; - if (!dc->optimized_required || dc->optimize_seamless_boot_streams > 0) + if ((!dc->optimized_required) || dc->optimize_seamless_boot_streams > 0) return true; post_surface_trace(dc); + if (is_flip_pending_in_pipes(dc, context)) + return true; + for (i = 0; i < dc->res_pool->pipe_count; i++) if (context->res_ctx.pipe_ctx[i].stream == NULL || context->res_ctx.pipe_ctx[i].plane_state == NULL) { @@ -1341,9 +1402,11 @@ bool dc_post_update_surfaces_to_stream(struct dc *dc) dc->hwss.disable_plane(dc, &context->res_ctx.pipe_ctx[i]); } + dc->hwss.optimize_bandwidth(dc, context); + dc->optimized_required = false; + dc->wm_optimized_required = false; - dc->hwss.optimize_bandwidth(dc, context); return true; } @@ -1665,6 +1728,9 @@ static enum surface_update_type det_surface_update(const struct dc *dc, if (u->coeff_reduction_factor) update_flags->bits.coeff_reduction_change = 1; + if (u->gamut_remap_matrix) + update_flags->bits.gamut_remap_change = 1; + if (u->gamma) { enum surface_pixel_format format = SURFACE_PIXEL_FORMAT_GRPH_BEGIN; @@ -1690,7 +1756,8 @@ static enum surface_update_type det_surface_update(const struct dc *dc, if (update_flags->bits.input_csc_change || update_flags->bits.coeff_reduction_change - || update_flags->bits.gamma_change) { + || update_flags->bits.gamma_change + || update_flags->bits.gamut_remap_change) { type = UPDATE_TYPE_FULL; elevate_update_type(&overall_type, type); } @@ -1734,14 +1801,15 @@ static enum surface_update_type check_update_surfaces_for_stream( if (stream_update->wb_update) su_flags->bits.wb_update = 1; + + if (stream_update->dsc_config) + su_flags->bits.dsc_changed = 1; + if (su_flags->raw != 0) overall_type = UPDATE_TYPE_FULL; if (stream_update->output_csc_transform || stream_update->output_color_space) su_flags->bits.out_csc = 1; - - if (stream_update->dsc_config) - overall_type = UPDATE_TYPE_FULL; } for (i = 0 ; i < surface_count; i++) { @@ -1776,8 +1844,11 @@ enum surface_update_type dc_check_update_surfaces_for_stream( type = check_update_surfaces_for_stream(dc, updates, surface_count, stream_update, stream_status); if (type == UPDATE_TYPE_FULL) { - if (stream_update) + if (stream_update) { + uint32_t dsc_changed = stream_update->stream->update_flags.bits.dsc_changed; stream_update->stream->update_flags.raw = 0xFFFFFFFF; + stream_update->stream->update_flags.bits.dsc_changed = dsc_changed; + } for (i = 0; i < surface_count; i++) updates[i].surface->update_flags.raw = 0xFFFFFFFF; } @@ -1791,6 +1862,8 @@ enum surface_update_type dc_check_update_surfaces_for_stream( } else if (memcmp(&dc->current_state->bw_ctx.bw.dcn.clk, &dc->clk_mgr->clks, offsetof(struct dc_clocks, prev_p_state_change_support)) != 0) { dc->optimized_required = true; } + + dc->optimized_required |= dc->wm_optimized_required; } return type; @@ -1829,6 +1902,8 @@ static void copy_surface_update_to_plane( surface->time.index++; if (surface->time.index >= DC_PLANE_UPDATE_TIMES_MAX) surface->time.index = 0; + + surface->triplebuffer_flips = srf_update->flip_addr->triplebuffer_flips; } if (srf_update->scaling_info) { @@ -1928,6 +2003,10 @@ static void copy_surface_update_to_plane( if (srf_update->coeff_reduction_factor) surface->coeff_reduction_factor = *srf_update->coeff_reduction_factor; + + if (srf_update->gamut_remap_matrix) + surface->gamut_remap_matrix = + *srf_update->gamut_remap_matrix; } static void copy_stream_update_to_stream(struct dc *dc, @@ -2093,18 +2172,14 @@ static void commit_planes_do_stream_update(struct dc *dc, } } - if (stream_update->dsc_config && dc->hwss.pipe_control_lock_global) { - dc->hwss.pipe_control_lock_global(dc, pipe_ctx, true); - dp_update_dsc_config(pipe_ctx); - dc->hwss.pipe_control_lock_global(dc, pipe_ctx, false); - } /* Full fe update*/ if (update_type == UPDATE_TYPE_FAST) continue; - if (stream_update->dpms_off) { - dc->hwss.pipe_control_lock(dc, pipe_ctx, true); + if (stream_update->dsc_config) + dp_update_dsc_config(pipe_ctx); + if (stream_update->dpms_off) { if (*stream_update->dpms_off) { core_link_disable_stream(pipe_ctx); /* for dpms, keep acquired resources*/ @@ -2118,8 +2193,6 @@ static void commit_planes_do_stream_update(struct dc *dc, core_link_enable_stream(dc->current_state, pipe_ctx); } - - dc->hwss.pipe_control_lock(dc, pipe_ctx, false); } if (stream_update->abm_level && pipe_ctx->stream_res.abm) { @@ -2175,6 +2248,32 @@ static void commit_planes_for_stream(struct dc *dc, context_clock_trace(dc, context); } + for (j = 0; j < dc->res_pool->pipe_count; j++) { + struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[j]; + + if (!pipe_ctx->top_pipe && + !pipe_ctx->prev_odm_pipe && + pipe_ctx->stream && + pipe_ctx->stream == stream) { + top_pipe_to_program = pipe_ctx; + } + } + + if ((update_type != UPDATE_TYPE_FAST) && stream->update_flags.bits.dsc_changed) + if (top_pipe_to_program->stream_res.tg->funcs->lock_doublebuffer_enable) + top_pipe_to_program->stream_res.tg->funcs->lock_doublebuffer_enable( + top_pipe_to_program->stream_res.tg); + + if ((update_type != UPDATE_TYPE_FAST) && dc->hwss.interdependent_update_lock) + dc->hwss.interdependent_update_lock(dc, context, true); + else + /* Lock the top pipe while updating plane addrs, since freesync requires + * plane addr update event triggers to be synchronized. + * top_pipe_to_program is expected to never be NULL + */ + dc->hwss.pipe_control_lock(dc, top_pipe_to_program, true); + + // Stream updates if (stream_update) commit_planes_do_stream_update(dc, stream, stream_update, update_type, context); @@ -2189,6 +2288,12 @@ static void commit_planes_for_stream(struct dc *dc, if (dc->hwss.program_front_end_for_ctx) dc->hwss.program_front_end_for_ctx(dc, context); + if ((update_type != UPDATE_TYPE_FAST) && dc->hwss.interdependent_update_lock) + dc->hwss.interdependent_update_lock(dc, context, false); + else + dc->hwss.pipe_control_lock(dc, top_pipe_to_program, false); + + dc->hwss.post_unlock_program_front_end(dc, context); return; } @@ -2224,8 +2329,6 @@ static void commit_planes_for_stream(struct dc *dc, pipe_ctx->stream == stream) { struct dc_stream_status *stream_status = NULL; - top_pipe_to_program = pipe_ctx; - if (!pipe_ctx->plane_state) continue; @@ -2270,12 +2373,6 @@ static void commit_planes_for_stream(struct dc *dc, // Update Type FAST, Surface updates if (update_type == UPDATE_TYPE_FAST) { - /* Lock the top pipe while updating plane addrs, since freesync requires - * plane addr update event triggers to be synchronized. - * top_pipe_to_program is expected to never be NULL - */ - dc->hwss.pipe_control_lock(dc, top_pipe_to_program, true); - if (dc->hwss.set_flip_control_gsl) for (i = 0; i < surface_count; i++) { struct dc_plane_state *plane_state = srf_updates[i].surface; @@ -2317,9 +2414,30 @@ static void commit_planes_for_stream(struct dc *dc, dc->hwss.update_plane_addr(dc, pipe_ctx); } } + } + if ((update_type != UPDATE_TYPE_FAST) && dc->hwss.interdependent_update_lock) + dc->hwss.interdependent_update_lock(dc, context, false); + else dc->hwss.pipe_control_lock(dc, top_pipe_to_program, false); - } + + if ((update_type != UPDATE_TYPE_FAST) && stream->update_flags.bits.dsc_changed) + if (top_pipe_to_program->stream_res.tg->funcs->lock_doublebuffer_enable) { + top_pipe_to_program->stream_res.tg->funcs->wait_for_state( + top_pipe_to_program->stream_res.tg, + CRTC_STATE_VACTIVE); + top_pipe_to_program->stream_res.tg->funcs->wait_for_state( + top_pipe_to_program->stream_res.tg, + CRTC_STATE_VBLANK); + top_pipe_to_program->stream_res.tg->funcs->wait_for_state( + top_pipe_to_program->stream_res.tg, + CRTC_STATE_VACTIVE); + top_pipe_to_program->stream_res.tg->funcs->lock_doublebuffer_disable( + top_pipe_to_program->stream_res.tg); + } + + if (update_type != UPDATE_TYPE_FAST) + dc->hwss.post_unlock_program_front_end(dc, context); // Fire manual trigger only when bottom plane is flipped for (j = 0; j < dc->res_pool->pipe_count; j++) { @@ -2347,7 +2465,7 @@ void dc_commit_updates_for_stream(struct dc *dc, enum surface_update_type update_type; struct dc_state *context; struct dc_context *dc_ctx = dc->ctx; - int i; + int i, j; stream_status = dc_stream_get_status(stream); context = dc->current_state; @@ -2385,6 +2503,17 @@ void dc_commit_updates_for_stream(struct dc *dc, copy_surface_update_to_plane(surface, &srf_updates[i]); + if (update_type >= UPDATE_TYPE_MED) { + for (j = 0; j < dc->res_pool->pipe_count; j++) { + struct pipe_ctx *pipe_ctx = + &context->res_ctx.pipe_ctx[j]; + + if (pipe_ctx->plane_state != surface) + continue; + + resource_build_scaling_params(pipe_ctx); + } + } } copy_stream_update_to_stream(dc, context, stream, stream_update); diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link.c b/drivers/gpu/drm/amd/display/dc/core/dc_link.c index a09119c10d7c..67cfff1586e9 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link.c @@ -45,7 +45,7 @@ #include "dpcd_defs.h" #include "dmcu.h" #include "hw/clk_mgr.h" -#include "../dce/dmub_psr.h" +#include "dce/dmub_psr.h" #define DC_LOGGER_INIT(logger) @@ -585,20 +585,23 @@ static void read_current_link_settings_on_detect(struct dc_link *link) LINK_SPREAD_05_DOWNSPREAD_30KHZ : LINK_SPREAD_DISABLED; } -static bool detect_dp( - struct dc_link *link, - struct display_sink_capability *sink_caps, - bool *converter_disable_audio, - struct audio_support *audio_support, - enum dc_detect_reason reason) +static bool detect_dp(struct dc_link *link, + struct display_sink_capability *sink_caps, + bool *converter_disable_audio, + struct audio_support *audio_support, + enum dc_detect_reason reason) { bool boot = false; + sink_caps->signal = link_detect_sink(link, reason); sink_caps->transaction_type = get_ddc_transaction_type(sink_caps->signal); if (sink_caps->transaction_type == DDC_TRANSACTION_TYPE_I2C_OVER_AUX) { sink_caps->signal = SIGNAL_TYPE_DISPLAY_PORT; + + dpcd_set_source_specific_data(link); + if (!detect_dp_sink_caps(link)) return false; @@ -606,9 +609,8 @@ static bool detect_dp( sink_caps->signal = SIGNAL_TYPE_DISPLAY_PORT_MST; link->type = dc_connection_mst_branch; - dal_ddc_service_set_transaction_type( - link->ddc, - sink_caps->transaction_type); + dal_ddc_service_set_transaction_type(link->ddc, + sink_caps->transaction_type); /* * This call will initiate MST topology discovery. Which @@ -637,13 +639,10 @@ static bool detect_dp( if (reason == DETECT_REASON_BOOT) boot = true; - dm_helpers_dp_update_branch_info( - link->ctx, - link); + dm_helpers_dp_update_branch_info(link->ctx, link); - if (!dm_helpers_dp_mst_start_top_mgr( - link->ctx, - link, boot)) { + if (!dm_helpers_dp_mst_start_top_mgr(link->ctx, + link, boot)) { /* MST not supported */ link->type = dc_connection_single; sink_caps->signal = SIGNAL_TYPE_DISPLAY_PORT; @@ -651,7 +650,7 @@ static bool detect_dp( } if (link->type != dc_connection_mst_branch && - is_dp_active_dongle(link)) { + is_dp_active_dongle(link)) { /* DP active dongles */ link->type = dc_connection_active_dongle; if (!link->dpcd_caps.sink_count.bits.SINK_COUNT) { @@ -662,14 +661,15 @@ static bool detect_dp( return true; } - if (link->dpcd_caps.dongle_type != DISPLAY_DONGLE_DP_HDMI_CONVERTER) + if (link->dpcd_caps.dongle_type != + DISPLAY_DONGLE_DP_HDMI_CONVERTER) *converter_disable_audio = true; } } else { /* DP passive dongles */ sink_caps->signal = dp_passive_dongle_detection(link->ddc, - sink_caps, - audio_support); + sink_caps, + audio_support); } return true; @@ -769,8 +769,16 @@ static bool dc_link_detect_helper(struct dc_link *link, if ((link->connector_signal == SIGNAL_TYPE_LVDS || link->connector_signal == SIGNAL_TYPE_EDP) && - link->local_sink) + link->local_sink) { + + // need to re-write OUI and brightness in resume case + if (link->connector_signal == SIGNAL_TYPE_EDP) { + dpcd_set_source_specific_data(link); + dc_link_set_default_brightness_aux(link); //TODO: use cached + } + return true; + } if (false == dc_link_detect_sink(link, &new_connection_type)) { BREAK_TO_DEBUGGER(); @@ -818,6 +826,10 @@ static bool dc_link_detect_helper(struct dc_link *link, } case SIGNAL_TYPE_EDP: { + read_current_link_settings_on_detect(link); + + dpcd_set_source_specific_data(link); + detect_edp_sink_caps(link); read_current_link_settings_on_detect(link); sink_caps.transaction_type = DDC_TRANSACTION_TYPE_I2C_OVER_AUX; @@ -958,10 +970,16 @@ static bool dc_link_detect_helper(struct dc_link *link, break; } + if (link->local_sink->edid_caps.panel_patch.disable_fec) + link->ctx->dc->debug.disable_fec = true; + // Check if edid is the same if ((prev_sink != NULL) && ((edid_status == EDID_THE_SAME) || (edid_status == EDID_OK))) same_edid = is_same_edid(&prev_sink->dc_edid, &sink->dc_edid); + if (sink->edid_caps.panel_patch.skip_scdc_overwrite) + link->ctx->dc->debug.hdmi20_disable = true; + if (link->connector_signal == SIGNAL_TYPE_DISPLAY_PORT && sink_caps.transaction_type == DDC_TRANSACTION_TYPE_I2C_OVER_AUX) { /* @@ -1480,9 +1498,8 @@ static void enable_stream_features(struct pipe_ctx *pipe_ctx) } } -static enum dc_status enable_link_dp( - struct dc_state *state, - struct pipe_ctx *pipe_ctx) +static enum dc_status enable_link_dp(struct dc_state *state, + struct pipe_ctx *pipe_ctx) { struct dc_stream_state *stream = pipe_ctx->stream; enum dc_status status; @@ -1492,6 +1509,7 @@ static enum dc_status enable_link_dp( bool fec_enable; int i; bool apply_seamless_boot_optimization = false; + uint32_t bl_oled_enable_delay = 50; // in ms // check for seamless boot for (i = 0; i < state->stream_count; i++) { @@ -1513,31 +1531,45 @@ static enum dc_status enable_link_dp( pipe_ctx->stream_res.pix_clk_params.requested_sym_clk = link_settings.link_rate * LINK_RATE_REF_FREQ_IN_KHZ; if (state->clk_mgr && !apply_seamless_boot_optimization) - state->clk_mgr->funcs->update_clocks(state->clk_mgr, state, false); + state->clk_mgr->funcs->update_clocks(state->clk_mgr, + state, false); + + // during mode switch we do DP_SET_POWER off then on, and OUI is lost + dpcd_set_source_specific_data(link); skip_video_pattern = true; if (link_settings.link_rate == LINK_RATE_LOW) - skip_video_pattern = false; - - if (perform_link_training_with_retries( - &link_settings, - skip_video_pattern, - LINK_TRAINING_ATTEMPTS, - pipe_ctx, - pipe_ctx->stream->signal)) { + skip_video_pattern = false; + + if (perform_link_training_with_retries(&link_settings, + skip_video_pattern, + LINK_TRAINING_ATTEMPTS, + pipe_ctx, + pipe_ctx->stream->signal)) { link->cur_link_settings = link_settings; status = DC_OK; - } - else + } else { status = DC_FAIL_DP_LINK_TRAINING; + } - if (link->preferred_training_settings.fec_enable != NULL) + if (link->preferred_training_settings.fec_enable) fec_enable = *link->preferred_training_settings.fec_enable; else fec_enable = true; dp_set_fec_enable(link, fec_enable); + + // during mode set we do DP_SET_POWER off then on, aux writes are lost + if (link->dpcd_sink_ext_caps.bits.oled == 1 || + link->dpcd_sink_ext_caps.bits.sdr_aux_backlight_control == 1 || + link->dpcd_sink_ext_caps.bits.hdr_aux_backlight_control == 1) { + dc_link_set_default_brightness_aux(link); // TODO: use cached if known + if (link->dpcd_sink_ext_caps.bits.oled == 1) + msleep(bl_oled_enable_delay); + dc_link_backlight_enable_aux(link, true); + } + return status; } @@ -1733,8 +1765,7 @@ static void write_i2c_retimer_setting( slave_address, buffer[0], buffer[1], i2c_success?1:0); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; /* Based on DP159 specs, APPLY_RX_TX_CHANGE bit in 0x0A * needs to be set to 1 on every 0xA-0xC write. @@ -1752,8 +1783,7 @@ static void write_i2c_retimer_setting( pipe_ctx->stream->link->ddc, slave_address, &offset, 1, &value, 1); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; } buffer[0] = offset; @@ -1765,8 +1795,7 @@ static void write_i2c_retimer_setting( offset = 0x%x, reg_val = 0x%x, i2c_success = %d\n", slave_address, buffer[0], buffer[1], i2c_success?1:0); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; } } } @@ -1786,8 +1815,7 @@ static void write_i2c_retimer_setting( slave_address, buffer[0], buffer[1], i2c_success?1:0); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; /* Based on DP159 specs, APPLY_RX_TX_CHANGE bit in 0x0A * needs to be set to 1 on every 0xA-0xC write. @@ -1805,8 +1833,7 @@ static void write_i2c_retimer_setting( pipe_ctx->stream->link->ddc, slave_address, &offset, 1, &value, 1); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; } buffer[0] = offset; @@ -1818,8 +1845,7 @@ static void write_i2c_retimer_setting( offset = 0x%x, reg_val = 0x%x, i2c_success = %d\n", slave_address, buffer[0], buffer[1], i2c_success?1:0); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; } } } @@ -1837,8 +1863,7 @@ static void write_i2c_retimer_setting( offset = 0x%x, reg_val = 0x%x, i2c_success = %d\n", slave_address, buffer[0], buffer[1], i2c_success?1:0); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; /* Write offset 0x00 to 0x23 */ buffer[0] = 0x00; @@ -1849,8 +1874,7 @@ static void write_i2c_retimer_setting( offset = 0x%x, reg_val = 0x%x, i2c_success = %d\n", slave_address, buffer[0], buffer[1], i2c_success?1:0); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; /* Write offset 0xff to 0x00 */ buffer[0] = 0xff; @@ -1861,10 +1885,14 @@ static void write_i2c_retimer_setting( offset = 0x%x, reg_val = 0x%x, i2c_success = %d\n", slave_address, buffer[0], buffer[1], i2c_success?1:0); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; } + + return; + +i2c_write_fail: + DC_LOG_DEBUG("Set retimer failed"); } static void write_i2c_default_retimer_setting( @@ -1889,8 +1917,7 @@ static void write_i2c_default_retimer_setting( offset = 0x%x, reg_val = 0x%x, i2c_success = %d\n", slave_address, buffer[0], buffer[1], i2c_success?1:0); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; /* Write offset 0x0A to 0x17 */ buffer[0] = 0x0A; @@ -1901,8 +1928,7 @@ static void write_i2c_default_retimer_setting( offset = 0x%x, reg_val = 0x%x, i2c_success = %d\n", slave_address, buffer[0], buffer[1], i2c_success?1:0); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; /* Write offset 0x0B to 0xDA or 0xD8 */ buffer[0] = 0x0B; @@ -1913,8 +1939,7 @@ static void write_i2c_default_retimer_setting( offset = 0x%x, reg_val = 0x%x, i2c_success = %d\n", slave_address, buffer[0], buffer[1], i2c_success?1:0); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; /* Write offset 0x0A to 0x17 */ buffer[0] = 0x0A; @@ -1925,8 +1950,7 @@ static void write_i2c_default_retimer_setting( offset = 0x%x, reg_val= 0x%x, i2c_success = %d\n", slave_address, buffer[0], buffer[1], i2c_success?1:0); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; /* Write offset 0x0C to 0x1D or 0x91 */ buffer[0] = 0x0C; @@ -1937,8 +1961,7 @@ static void write_i2c_default_retimer_setting( offset = 0x%x, reg_val = 0x%x, i2c_success = %d\n", slave_address, buffer[0], buffer[1], i2c_success?1:0); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; /* Write offset 0x0A to 0x17 */ buffer[0] = 0x0A; @@ -1949,8 +1972,7 @@ static void write_i2c_default_retimer_setting( offset = 0x%x, reg_val = 0x%x, i2c_success = %d\n", slave_address, buffer[0], buffer[1], i2c_success?1:0); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; if (is_vga_mode) { @@ -1965,8 +1987,7 @@ static void write_i2c_default_retimer_setting( offset = 0x%x, reg_val = 0x%x, i2c_success = %d\n", slave_address, buffer[0], buffer[1], i2c_success?1:0); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; /* Write offset 0x00 to 0x23 */ buffer[0] = 0x00; @@ -1977,8 +1998,7 @@ static void write_i2c_default_retimer_setting( offset = 0x%x, reg_val= 0x%x, i2c_success = %d\n", slave_address, buffer[0], buffer[1], i2c_success?1:0); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; /* Write offset 0xff to 0x00 */ buffer[0] = 0xff; @@ -1989,9 +2009,13 @@ static void write_i2c_default_retimer_setting( offset = 0x%x, reg_val= 0x%x, i2c_success = %d end here\n", slave_address, buffer[0], buffer[1], i2c_success?1:0); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; } + + return; + +i2c_write_fail: + DC_LOG_DEBUG("Set default retimer failed"); } static void write_i2c_redriver_setting( @@ -2020,8 +2044,7 @@ static void write_i2c_redriver_setting( slave_address, buffer[3], buffer[4], buffer[5], buffer[6], i2c_success?1:0); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + DC_LOG_DEBUG("Set redriver failed"); } static void disable_link(struct dc_link *link, enum signal_type signal) @@ -2400,8 +2423,8 @@ bool dc_link_set_psr_allow_active(struct dc_link *link, bool allow_active, bool struct dmcu *dmcu = dc->res_pool->dmcu; struct dmub_psr *psr = dc->res_pool->psr; - if ((psr != NULL) && link->psr_feature_enabled) - psr->funcs->set_psr_enable(psr, allow_active); + if (psr != NULL && link->psr_feature_enabled) + psr->funcs->psr_enable(psr, allow_active); else if ((dmcu != NULL && dmcu->funcs->is_dmcu_initialized(dmcu)) && link->psr_feature_enabled) dmcu->funcs->set_psr_enable(dmcu, allow_active, wait); @@ -2417,7 +2440,7 @@ bool dc_link_get_psr_state(const struct dc_link *link, uint32_t *psr_state) struct dmub_psr *psr = dc->res_pool->psr; if (psr != NULL && link->psr_feature_enabled) - psr->funcs->get_psr_state(psr_state); + psr->funcs->psr_get_state(psr, psr_state); else if (dmcu != NULL && link->psr_feature_enabled) dmcu->funcs->get_psr_state(dmcu, psr_state); @@ -2589,7 +2612,7 @@ bool dc_link_setup_psr(struct dc_link *link, psr_context->frame_delay = 0; if (psr) - link->psr_feature_enabled = psr->funcs->setup_psr(psr, link, psr_context); + link->psr_feature_enabled = psr->funcs->psr_copy_settings(psr, link, psr_context); else link->psr_feature_enabled = dmcu->funcs->setup_psr(dmcu, link, psr_context); @@ -2922,10 +2945,13 @@ static void update_psp_stream_config(struct pipe_ctx *pipe_ctx, bool dpms_off) memset(&config, 0, sizeof(config)); config.otg_inst = (uint8_t) pipe_ctx->stream_res.tg->inst; - config.stream_enc_inst = (uint8_t) pipe_ctx->stream_res.stream_enc->id; + /*stream_enc_inst*/ + config.stream_enc_inst = (uint8_t) pipe_ctx->stream_res.stream_enc->stream_enc_inst; config.link_enc_inst = pipe_ctx->stream->link->link_enc_hw_inst; config.dpms_off = dpms_off; config.dm_stream_ctx = pipe_ctx->stream->dm_stream_context; + config.mst_supported = (pipe_ctx->stream->signal == + SIGNAL_TYPE_DISPLAY_PORT_MST); cp_psp->funcs.update_stream_config(cp_psp->handle, &config); } } @@ -3061,6 +3087,9 @@ void core_link_enable_stream( dc->hwss.unblank_stream(pipe_ctx, &pipe_ctx->stream->link->cur_link_settings); + if (stream->sink_patches.delay_ignore_msa > 0) + msleep(stream->sink_patches.delay_ignore_msa); + if (dc_is_dp_signal(pipe_ctx->stream->signal)) enable_stream_features(pipe_ctx); #if defined(CONFIG_DRM_AMD_DC_HDCP) @@ -3373,7 +3402,7 @@ uint32_t dc_link_bandwidth_kbps( link_bw_kbps *= 8; /* 8 bits per byte*/ link_bw_kbps *= link_setting->lane_count; - if (link->dpcd_caps.fec_cap.bits.FEC_CAPABLE) { + if (dc_link_is_fec_supported(link) && !link->dc->debug.disable_fec) { /* Account for FEC overhead. * We have to do it based on caps, * and not based on FEC being set ready, @@ -3417,3 +3446,11 @@ void dc_link_overwrite_extended_receiver_cap( dp_overwrite_extended_receiver_cap(link); } +bool dc_link_is_fec_supported(const struct dc_link *link) +{ + return (dc_is_dp_signal(link->connector_signal) && + link->link_enc->features.fec_supported && + link->dpcd_caps.fec_cap.bits.FEC_CAPABLE && + !IS_FPGA_MAXIMUS_DC(link->ctx->dce_environment)); +} + diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c index a49c10d5df26..256889eed93e 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c @@ -126,22 +126,16 @@ struct aux_payloads { struct vector payloads; }; -static struct i2c_payloads *dal_ddc_i2c_payloads_create(struct dc_context *ctx, uint32_t count) +static bool dal_ddc_i2c_payloads_create( + struct dc_context *ctx, + struct i2c_payloads *payloads, + uint32_t count) { - struct i2c_payloads *payloads; - - payloads = kzalloc(sizeof(struct i2c_payloads), GFP_KERNEL); - - if (!payloads) - return NULL; - if (dal_vector_construct( &payloads->payloads, ctx, count, sizeof(struct i2c_payload))) - return payloads; - - kfree(payloads); - return NULL; + return true; + return false; } static struct i2c_payload *dal_ddc_i2c_payloads_get(struct i2c_payloads *p) @@ -154,14 +148,12 @@ static uint32_t dal_ddc_i2c_payloads_get_count(struct i2c_payloads *p) return p->payloads.count; } -static void dal_ddc_i2c_payloads_destroy(struct i2c_payloads **p) +static void dal_ddc_i2c_payloads_destroy(struct i2c_payloads *p) { - if (!p || !*p) + if (!p) return; - dal_vector_destruct(&(*p)->payloads); - kfree(*p); - *p = NULL; + dal_vector_destruct(&p->payloads); } #define DDC_MIN(a, b) (((a) < (b)) ? (a) : (b)) @@ -524,9 +516,13 @@ bool dal_ddc_service_query_ddc_data( uint32_t payloads_num = write_payloads + read_payloads; + if (write_size > EDID_SEGMENT_SIZE || read_size > EDID_SEGMENT_SIZE) return false; + if (!payloads_num) + return false; + /*TODO: len of payload data for i2c and aux is uint8!!!!, * but we want to read 256 over i2c!!!!*/ if (dal_ddc_service_is_in_aux_transaction_mode(ddc)) { @@ -557,23 +553,25 @@ bool dal_ddc_service_query_ddc_data( ret = dal_ddc_submit_aux_command(ddc, &payload); } } else { - struct i2c_payloads *payloads = - dal_ddc_i2c_payloads_create(ddc->ctx, payloads_num); + struct i2c_command command = {0}; + struct i2c_payloads payloads; - struct i2c_command command = { - .payloads = dal_ddc_i2c_payloads_get(payloads), - .number_of_payloads = 0, - .engine = DDC_I2C_COMMAND_ENGINE, - .speed = ddc->ctx->dc->caps.i2c_speed_in_khz }; + if (!dal_ddc_i2c_payloads_create(ddc->ctx, &payloads, payloads_num)) + return false; + + command.payloads = dal_ddc_i2c_payloads_get(&payloads); + command.number_of_payloads = 0; + command.engine = DDC_I2C_COMMAND_ENGINE; + command.speed = ddc->ctx->dc->caps.i2c_speed_in_khz; dal_ddc_i2c_payloads_add( - payloads, address, write_size, write_buf, true); + &payloads, address, write_size, write_buf, true); dal_ddc_i2c_payloads_add( - payloads, address, read_size, read_buf, false); + &payloads, address, read_size, read_buf, false); command.number_of_payloads = - dal_ddc_i2c_payloads_get_count(payloads); + dal_ddc_i2c_payloads_get_count(&payloads); ret = dm_helpers_submit_i2c( ddc->ctx, @@ -686,6 +684,10 @@ void dal_ddc_service_write_scdc_data(struct ddc_service *ddc_service, uint8_t write_buffer[2] = {0}; /*Lower than 340 Scramble bit from SCDC caps*/ + if (ddc_service->link->local_sink && + ddc_service->link->local_sink->edid_caps.panel_patch.skip_scdc_overwrite) + return; + dal_ddc_service_query_ddc_data(ddc_service, slave_address, &offset, sizeof(offset), &sink_version, sizeof(sink_version)); if (sink_version == 1) { @@ -715,6 +717,10 @@ void dal_ddc_service_read_scdc_data(struct ddc_service *ddc_service) uint8_t offset = HDMI_SCDC_TMDS_CONFIG; uint8_t tmds_config = 0; + if (ddc_service->link->local_sink && + ddc_service->link->local_sink->edid_caps.panel_patch.skip_scdc_overwrite) + return; + dal_ddc_service_query_ddc_data(ddc_service, slave_address, &offset, sizeof(offset), &tmds_config, sizeof(tmds_config)); if (tmds_config & 0x1) { diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c index cb731c1d30b1..7cbb1efb4f68 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c @@ -945,6 +945,17 @@ static enum link_training_result perform_channel_equalization_sequence( } #define TRAINING_AUX_RD_INTERVAL 100 //us +static void start_clock_recovery_pattern_early(struct dc_link *link, + struct link_training_settings *lt_settings, + uint32_t offset) +{ + DC_LOG_HW_LINK_TRAINING("%s\n GPU sends TPS1. Wait 400us.\n", + __func__); + dp_set_hw_training_pattern(link, DP_TRAINING_PATTERN_SEQUENCE_1, offset); + dp_set_hw_lane_settings(link, lt_settings, offset); + udelay(400); +} + static enum link_training_result perform_clock_recovery_sequence( struct dc_link *link, struct link_training_settings *lt_settings, @@ -962,7 +973,8 @@ static enum link_training_result perform_clock_recovery_sequence( retries_cr = 0; retry_count = 0; - dp_set_hw_training_pattern(link, tr_pattern, offset); + if (!link->ctx->dc->work_arounds.lt_early_cr_pattern) + dp_set_hw_training_pattern(link, tr_pattern, offset); /* najeeb - The synaptics MST hub can put the LT in * infinite loop by switching the VS @@ -1434,6 +1446,13 @@ enum link_training_result dc_link_dp_perform_link_training( &link->preferred_training_settings, <_settings); + /* Configure lttpr mode */ + if (!link->is_lttpr_mode_transparent) + configure_lttpr_mode(link); + + if (link->ctx->dc->work_arounds.lt_early_cr_pattern) + start_clock_recovery_pattern_early(link, <_settings, DPRX); + /* 1. set link rate, lane count and spread. */ dpcd_set_link_settings(link, <_settings); @@ -1445,8 +1464,6 @@ enum link_training_result dc_link_dp_perform_link_training( dp_set_fec_ready(link, fec_enable); if (!link->is_lttpr_mode_transparent) { - /* Configure lttpr mode */ - configure_lttpr_mode(link); /* 2. perform link training (set link training done * to false is done as well) @@ -1654,6 +1671,8 @@ enum link_training_result dc_link_dp_sync_lt_attempt( dp_set_panel_mode(link, panel_mode); /* Attempt to train with given link training settings */ + if (link->ctx->dc->work_arounds.lt_early_cr_pattern) + start_clock_recovery_pattern_early(link, <_settings, DPRX); /* Set link rate, lane count and spread. */ dpcd_set_link_settings(link, <_settings); @@ -1892,6 +1911,16 @@ bool dp_verify_link_cap( /* disable PHY done possible by BIOS, will be done by driver itself */ dp_disable_link_phy(link, link->connector_signal); + dp_cs_id = get_clock_source_id(link); + + /* link training starts with the maximum common settings + * supported by both sink and ASIC. + */ + initial_link_settings = get_common_supported_link_settings( + *known_limit_link_setting, + max_link_cap); + cur_link_setting = initial_link_settings; + /* Temporary Renoir-specific workaround for SWDEV-215184; * PHY will sometimes be in bad state on hotplugging display from certain USB-C dongle, * so add extra cycle of enabling and disabling the PHY before first link training. @@ -1902,15 +1931,6 @@ bool dp_verify_link_cap( dp_disable_link_phy(link, link->connector_signal); } - dp_cs_id = get_clock_source_id(link); - - /* link training starts with the maximum common settings - * supported by both sink and ASIC. - */ - initial_link_settings = get_common_supported_link_settings( - *known_limit_link_setting, - max_link_cap); - cur_link_setting = initial_link_settings; do { skip_video_pattern = true; @@ -2654,9 +2674,12 @@ static void dp_test_send_link_test_pattern(struct dc_link *link) break; } - test_pattern_color_space = dpcd_test_params.bits.YCBCR_COEFS ? - DP_TEST_PATTERN_COLOR_SPACE_YCBCR709 : - DP_TEST_PATTERN_COLOR_SPACE_YCBCR601; + if (dpcd_test_params.bits.CLR_FORMAT == 0) + test_pattern_color_space = DP_TEST_PATTERN_COLOR_SPACE_RGB; + else + test_pattern_color_space = dpcd_test_params.bits.YCBCR_COEFS ? + DP_TEST_PATTERN_COLOR_SPACE_YCBCR709 : + DP_TEST_PATTERN_COLOR_SPACE_YCBCR601; dc_link_dp_set_test_pattern( link, @@ -3165,6 +3188,23 @@ static void dp_wa_power_up_0010FA(struct dc_link *link, uint8_t *dpcd_data, link->wa_flags.dp_keep_receiver_powered = false; } +/* Read additional sink caps defined in source specific DPCD area + * This function currently only reads from SinkCapability address (DP_SOURCE_SINK_CAP) + */ +static bool dpcd_read_sink_ext_caps(struct dc_link *link) +{ + uint8_t dpcd_data; + + if (!link) + return false; + + if (core_link_read_dpcd(link, DP_SOURCE_SINK_CAP, &dpcd_data, 1) != DC_OK) + return false; + + link->dpcd_sink_ext_caps.raw = dpcd_data; + return true; +} + static bool retrieve_link_cap(struct dc_link *link) { /* DP_ADAPTER_CAP - DP_DPCD_REV + 1 == 16 and also DP_DSC_BITS_PER_PIXEL_INC - DP_DSC_SUPPORT + 1 == 16, @@ -3401,6 +3441,17 @@ static bool retrieve_link_cap(struct dc_link *link) sink_id.ieee_device_id, sizeof(sink_id.ieee_device_id)); + /* Quirk Apple MBP 2017 15" Retina panel: Wrong DP_MAX_LINK_RATE */ + { + uint8_t str_mbp_2017[] = { 101, 68, 21, 101, 98, 97 }; + + if ((link->dpcd_caps.sink_dev_id == 0x0010fa) && + !memcmp(link->dpcd_caps.sink_dev_id_str, str_mbp_2017, + sizeof(str_mbp_2017))) { + link->reported_link_cap.link_rate = 0x0c; + } + } + core_link_read_dpcd( link, DP_SINK_HW_REVISION_START, @@ -3437,6 +3488,9 @@ static bool retrieve_link_cap(struct dc_link *link) sizeof(link->dpcd_caps.dsc_caps.dsc_ext_caps.raw)); } + if (!dpcd_read_sink_ext_caps(link)) + link->dpcd_sink_ext_caps.raw = 0; + /* Connectivity log: detection */ CONN_DATA_DETECT(link, dpcd_data, sizeof(dpcd_data), "Rx Caps: "); @@ -3589,6 +3643,8 @@ void detect_edp_sink_caps(struct dc_link *link) } } link->verified_link_cap = link->reported_link_cap; + + dc_link_set_default_brightness_aux(link); } void dc_link_dp_enable_hpd(const struct dc_link *link) @@ -3680,7 +3736,8 @@ static void set_crtc_test_pattern(struct dc_link *link, struct pipe_ctx *odm_pipe; enum controller_dp_color_space controller_color_space; int opp_cnt = 1; - int count; + int offset = 0; + int dpg_width = width; switch (test_pattern_color_space) { case DP_TEST_PATTERN_COLOR_SPACE_RGB: @@ -3702,33 +3759,30 @@ static void set_crtc_test_pattern(struct dc_link *link, for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) opp_cnt++; + dpg_width = width / opp_cnt; + offset = dpg_width; - width /= opp_cnt; + opp->funcs->opp_set_disp_pattern_generator(opp, + controller_test_pattern, + controller_color_space, + color_depth, + NULL, + dpg_width, + height, + 0); for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) { struct output_pixel_processor *odm_opp = odm_pipe->stream_res.opp; - odm_opp->funcs->opp_program_bit_depth_reduction(odm_opp, ¶ms); odm_opp->funcs->opp_set_disp_pattern_generator(odm_opp, controller_test_pattern, controller_color_space, color_depth, NULL, - width, - height); - } - opp->funcs->opp_set_disp_pattern_generator(opp, - controller_test_pattern, - controller_color_space, - color_depth, - NULL, - width, - height); - /* wait for dpg to blank pixel data with test pattern */ - for (count = 0; count < 1000; count++) { - if (opp->funcs->dpg_is_blanked(opp)) - break; - udelay(100); + dpg_width, + height, + offset); + offset += offset; } } } @@ -3746,11 +3800,12 @@ static void set_crtc_test_pattern(struct dc_link *link, else if (opp->funcs->opp_set_disp_pattern_generator) { struct pipe_ctx *odm_pipe; int opp_cnt = 1; + int dpg_width = width; for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) opp_cnt++; - width /= opp_cnt; + dpg_width = width / opp_cnt; for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) { struct output_pixel_processor *odm_opp = odm_pipe->stream_res.opp; @@ -3760,16 +3815,18 @@ static void set_crtc_test_pattern(struct dc_link *link, CONTROLLER_DP_COLOR_SPACE_UDEFINED, color_depth, NULL, - width, - height); + dpg_width, + height, + 0); } opp->funcs->opp_set_disp_pattern_generator(opp, CONTROLLER_DP_TEST_PATTERN_VIDEOMODE, CONTROLLER_DP_COLOR_SPACE_UDEFINED, color_depth, NULL, - width, - height); + dpg_width, + height, + 0); } } break; @@ -3947,6 +4004,11 @@ bool dc_link_dp_set_test_pattern( default: break; } + + if (pipe_ctx->stream_res.tg->funcs->lock_doublebuffer_enable) + pipe_ctx->stream_res.tg->funcs->lock_doublebuffer_enable( + pipe_ctx->stream_res.tg); + pipe_ctx->stream_res.tg->funcs->lock(pipe_ctx->stream_res.tg); /* update MSA to requested color space */ pipe_ctx->stream_res.stream_enc->funcs->dp_set_stream_attribute(pipe_ctx->stream_res.stream_enc, &pipe_ctx->stream->timing, @@ -3954,9 +4016,27 @@ bool dc_link_dp_set_test_pattern( pipe_ctx->stream->use_vsc_sdp_for_colorimetry, link->dpcd_caps.dprx_feature.bits.SST_SPLIT_SDP_CAP); + if (pipe_ctx->stream->use_vsc_sdp_for_colorimetry) { + if (test_pattern == DP_TEST_PATTERN_COLOR_SQUARES_CEA) + pipe_ctx->stream->vsc_infopacket.sb[17] |= (1 << 7); // sb17 bit 7 Dynamic Range: 0 = VESA range, 1 = CTA range + else + pipe_ctx->stream->vsc_infopacket.sb[17] &= ~(1 << 7); + resource_build_info_frame(pipe_ctx); + link->dc->hwss.update_info_frame(pipe_ctx); + } + /* CRTC Patterns */ set_crtc_test_pattern(link, pipe_ctx, test_pattern, test_pattern_color_space); - + pipe_ctx->stream_res.tg->funcs->unlock(pipe_ctx->stream_res.tg); + pipe_ctx->stream_res.tg->funcs->wait_for_state(pipe_ctx->stream_res.tg, + CRTC_STATE_VACTIVE); + pipe_ctx->stream_res.tg->funcs->wait_for_state(pipe_ctx->stream_res.tg, + CRTC_STATE_VBLANK); + pipe_ctx->stream_res.tg->funcs->wait_for_state(pipe_ctx->stream_res.tg, + CRTC_STATE_VACTIVE); + if (pipe_ctx->stream_res.tg->funcs->lock_doublebuffer_disable) + pipe_ctx->stream_res.tg->funcs->lock_doublebuffer_disable( + pipe_ctx->stream_res.tg); /* Set Test Pattern state */ link->test_pattern_enabled = true; } @@ -4086,8 +4166,7 @@ void dp_set_fec_ready(struct dc_link *link, bool ready) struct link_encoder *link_enc = link->link_enc; uint8_t fec_config = 0; - if (link->dc->debug.disable_fec || - IS_FPGA_MAXIMUS_DC(link->ctx->dce_environment)) + if (!dc_link_is_fec_supported(link) || link->dc->debug.disable_fec) return; if (link_enc->funcs->fec_set_ready && @@ -4122,8 +4201,7 @@ void dp_set_fec_enable(struct dc_link *link, bool enable) { struct link_encoder *link_enc = link->link_enc; - if (link->dc->debug.disable_fec || - IS_FPGA_MAXIMUS_DC(link->ctx->dce_environment)) + if (!dc_link_is_fec_supported(link) || link->dc->debug.disable_fec) return; if (link_enc->funcs->fec_set_enable && @@ -4146,3 +4224,148 @@ void dp_set_fec_enable(struct dc_link *link, bool enable) } } +void dpcd_set_source_specific_data(struct dc_link *link) +{ + const uint32_t post_oui_delay = 30; // 30ms + + if (!link->dc->vendor_signature.is_valid) { + struct dpcd_amd_signature amd_signature; + amd_signature.AMD_IEEE_TxSignature_byte1 = 0x0; + amd_signature.AMD_IEEE_TxSignature_byte2 = 0x0; + amd_signature.AMD_IEEE_TxSignature_byte3 = 0x1A; + amd_signature.device_id_byte1 = + (uint8_t)(link->ctx->asic_id.chip_id); + amd_signature.device_id_byte2 = + (uint8_t)(link->ctx->asic_id.chip_id >> 8); + memset(&amd_signature.zero, 0, 4); + amd_signature.dce_version = + (uint8_t)(link->ctx->dce_version); + amd_signature.dal_version_byte1 = 0x0; // needed? where to get? + amd_signature.dal_version_byte2 = 0x0; // needed? where to get? + + core_link_write_dpcd(link, DP_SOURCE_OUI, + (uint8_t *)(&amd_signature), + sizeof(amd_signature)); + + } else { + core_link_write_dpcd(link, DP_SOURCE_OUI, + link->dc->vendor_signature.data.raw, + sizeof(link->dc->vendor_signature.data.raw)); + } + + // Sink may need to configure internals based on vendor, so allow some + // time before proceeding with possibly vendor specific transactions + msleep(post_oui_delay); +} + +bool dc_link_set_backlight_level_nits(struct dc_link *link, + bool isHDR, + uint32_t backlight_millinits, + uint32_t transition_time_in_ms) +{ + struct dpcd_source_backlight_set dpcd_backlight_set; + uint8_t backlight_control = isHDR ? 1 : 0; + + if (!link || (link->connector_signal != SIGNAL_TYPE_EDP && + link->connector_signal != SIGNAL_TYPE_DISPLAY_PORT)) + return false; + + // OLEDs have no PWM, they can only use AUX + if (link->dpcd_sink_ext_caps.bits.oled == 1) + backlight_control = 1; + + *(uint32_t *)&dpcd_backlight_set.backlight_level_millinits = backlight_millinits; + *(uint16_t *)&dpcd_backlight_set.backlight_transition_time_ms = (uint16_t)transition_time_in_ms; + + + if (core_link_write_dpcd(link, DP_SOURCE_BACKLIGHT_LEVEL, + (uint8_t *)(&dpcd_backlight_set), + sizeof(dpcd_backlight_set)) != DC_OK) + return false; + + if (core_link_write_dpcd(link, DP_SOURCE_BACKLIGHT_CONTROL, + &backlight_control, 1) != DC_OK) + return false; + + return true; +} + +bool dc_link_get_backlight_level_nits(struct dc_link *link, + uint32_t *backlight_millinits_avg, + uint32_t *backlight_millinits_peak) +{ + union dpcd_source_backlight_get dpcd_backlight_get; + + memset(&dpcd_backlight_get, 0, sizeof(union dpcd_source_backlight_get)); + + if (!link || (link->connector_signal != SIGNAL_TYPE_EDP && + link->connector_signal != SIGNAL_TYPE_DISPLAY_PORT)) + return false; + + if (!core_link_read_dpcd(link, DP_SOURCE_BACKLIGHT_CURRENT_PEAK, + dpcd_backlight_get.raw, + sizeof(union dpcd_source_backlight_get))) + return false; + + *backlight_millinits_avg = + dpcd_backlight_get.bytes.backlight_millinits_avg; + *backlight_millinits_peak = + dpcd_backlight_get.bytes.backlight_millinits_peak; + + /* On non-supported panels dpcd_read usually succeeds with 0 returned */ + if (*backlight_millinits_avg == 0 || + *backlight_millinits_avg > *backlight_millinits_peak) + return false; + + return true; +} + +bool dc_link_backlight_enable_aux(struct dc_link *link, bool enable) +{ + uint8_t backlight_enable = enable ? 1 : 0; + + if (!link || (link->connector_signal != SIGNAL_TYPE_EDP && + link->connector_signal != SIGNAL_TYPE_DISPLAY_PORT)) + return false; + + if (core_link_write_dpcd(link, DP_SOURCE_BACKLIGHT_ENABLE, + &backlight_enable, 1) != DC_OK) + return false; + + return true; +} + +// we read default from 0x320 because we expect BIOS wrote it there +// regular get_backlight_nit reads from panel set at 0x326 +bool dc_link_read_default_bl_aux(struct dc_link *link, uint32_t *backlight_millinits) +{ + if (!link || (link->connector_signal != SIGNAL_TYPE_EDP && + link->connector_signal != SIGNAL_TYPE_DISPLAY_PORT)) + return false; + + if (!core_link_read_dpcd(link, DP_SOURCE_BACKLIGHT_LEVEL, + (uint8_t *) backlight_millinits, + sizeof(uint32_t))) + return false; + + return true; +} + +bool dc_link_set_default_brightness_aux(struct dc_link *link) +{ + uint32_t default_backlight; + + if (link && + (link->dpcd_sink_ext_caps.bits.hdr_aux_backlight_control == 1 || + link->dpcd_sink_ext_caps.bits.sdr_aux_backlight_control == 1)) { + if (!dc_link_read_default_bl_aux(link, &default_backlight)) + default_backlight = 150000; + // if < 5 nits or > 5000, it might be wrong readback + if (default_backlight < 5000 || default_backlight > 5000000) + default_backlight = 150000; // + + return dc_link_set_backlight_level_nits(link, true, + default_backlight, 0); + } + return false; +} diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c index ddb855045767..51e0ee6e7695 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c @@ -153,18 +153,19 @@ bool edp_receiver_ready_T9(struct dc_link *link) unsigned char edpRev = 0; enum dc_status result = DC_OK; result = core_link_read_dpcd(link, DP_EDP_DPCD_REV, &edpRev, sizeof(edpRev)); - if (edpRev < DP_EDP_12) - return true; - /* start from eDP version 1.2, SINK_STAUS indicate the sink is ready.*/ - do { - sinkstatus = 1; - result = core_link_read_dpcd(link, DP_SINK_STATUS, &sinkstatus, sizeof(sinkstatus)); - if (sinkstatus == 0) - break; - if (result != DC_OK) - break; - udelay(100); //MAx T9 - } while (++tries < 50); + + /* start from eDP version 1.2, SINK_STAUS indicate the sink is ready.*/ + if (result == DC_OK && edpRev >= DP_EDP_12) { + do { + sinkstatus = 1; + result = core_link_read_dpcd(link, DP_SINK_STATUS, &sinkstatus, sizeof(sinkstatus)); + if (sinkstatus == 0) + break; + if (result != DC_OK) + break; + udelay(100); //MAx T9 + } while (++tries < 50); + } if (link->local_sink->edid_caps.panel_patch.extra_delay_backlight_off > 0) udelay(link->local_sink->edid_caps.panel_patch.extra_delay_backlight_off * 1000); @@ -183,21 +184,22 @@ bool edp_receiver_ready_T7(struct dc_link *link) unsigned long long time_taken_in_ns = 0; result = core_link_read_dpcd(link, DP_EDP_DPCD_REV, &edpRev, sizeof(edpRev)); - if (result == DC_OK && edpRev < DP_EDP_12) - return true; - /* start from eDP version 1.2, SINK_STAUS indicate the sink is ready.*/ - enter_timestamp = dm_get_timestamp(link->ctx); - do { - sinkstatus = 0; - result = core_link_read_dpcd(link, DP_SINK_STATUS, &sinkstatus, sizeof(sinkstatus)); - if (sinkstatus == 1) - break; - if (result != DC_OK) - break; - udelay(25); - finish_timestamp = dm_get_timestamp(link->ctx); - time_taken_in_ns = dm_get_elapse_time_in_ns(link->ctx, finish_timestamp, enter_timestamp); - } while (time_taken_in_ns < 50 * 1000000); //MAx T7 is 50ms + + if (result == DC_OK && edpRev >= DP_EDP_12) { + /* start from eDP version 1.2, SINK_STAUS indicate the sink is ready.*/ + enter_timestamp = dm_get_timestamp(link->ctx); + do { + sinkstatus = 0; + result = core_link_read_dpcd(link, DP_SINK_STATUS, &sinkstatus, sizeof(sinkstatus)); + if (sinkstatus == 1) + break; + if (result != DC_OK) + break; + udelay(25); + finish_timestamp = dm_get_timestamp(link->ctx); + time_taken_in_ns = dm_get_elapse_time_in_ns(link->ctx, finish_timestamp, enter_timestamp); + } while (time_taken_in_ns < 50 * 1000000); //MAx T7 is 50ms + } if (link->local_sink->edid_caps.panel_patch.extra_t7_ms > 0) udelay(link->local_sink->edid_caps.panel_patch.extra_t7_ms * 1000); @@ -429,6 +431,7 @@ void dp_set_dsc_on_stream(struct pipe_ctx *pipe_ctx, bool enable) dsc_cfg.pic_height = stream->timing.v_addressable + stream->timing.v_border_top + stream->timing.v_border_bottom; dsc_cfg.pixel_encoding = stream->timing.pixel_encoding; dsc_cfg.color_depth = stream->timing.display_color_depth; + dsc_cfg.is_odm = pipe_ctx->next_odm_pipe ? true : false; dsc_cfg.dc_dsc_cfg = stream->timing.dsc_cfg; ASSERT(dsc_cfg.dc_dsc_cfg.num_slices_h % opp_cnt == 0); dsc_cfg.dc_dsc_cfg.num_slices_h /= opp_cnt; @@ -533,6 +536,7 @@ bool dp_set_dsc_pps_sdp(struct pipe_ctx *pipe_ctx, bool enable) dsc_cfg.pic_height = stream->timing.v_addressable + stream->timing.v_border_top + stream->timing.v_border_bottom; dsc_cfg.pixel_encoding = stream->timing.pixel_encoding; dsc_cfg.color_depth = stream->timing.display_color_depth; + dsc_cfg.is_odm = pipe_ctx->next_odm_pipe ? true : false; dsc_cfg.dc_dsc_cfg = stream->timing.dsc_cfg; DC_LOG_DSC(" "); diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c index a0eb9e533a61..f4bcc71b2920 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c @@ -46,12 +46,12 @@ #include "dce100/dce100_resource.h" #include "dce110/dce110_resource.h" #include "dce112/dce112_resource.h" +#include "dce120/dce120_resource.h" #if defined(CONFIG_DRM_AMD_DC_DCN) #include "dcn10/dcn10_resource.h" -#endif #include "dcn20/dcn20_resource.h" #include "dcn21/dcn21_resource.h" -#include "dce120/dce120_resource.h" +#endif #define DC_LOGGER_INIT(logger) @@ -532,6 +532,51 @@ static inline void get_vp_scan_direction( *flip_horz_scan_dir = !*flip_horz_scan_dir; } +int get_num_odm_splits(struct pipe_ctx *pipe) +{ + int odm_split_count = 0; + struct pipe_ctx *next_pipe = pipe->next_odm_pipe; + while (next_pipe) { + odm_split_count++; + next_pipe = next_pipe->next_odm_pipe; + } + pipe = pipe->prev_odm_pipe; + while (pipe) { + odm_split_count++; + pipe = pipe->prev_odm_pipe; + } + return odm_split_count; +} + +static void calculate_split_count_and_index(struct pipe_ctx *pipe_ctx, int *split_count, int *split_idx) +{ + *split_count = get_num_odm_splits(pipe_ctx); + *split_idx = 0; + if (*split_count == 0) { + /*Check for mpc split*/ + struct pipe_ctx *split_pipe = pipe_ctx->top_pipe; + + while (split_pipe && split_pipe->plane_state == pipe_ctx->plane_state) { + (*split_idx)++; + (*split_count)++; + split_pipe = split_pipe->top_pipe; + } + split_pipe = pipe_ctx->bottom_pipe; + while (split_pipe && split_pipe->plane_state == pipe_ctx->plane_state) { + (*split_count)++; + split_pipe = split_pipe->bottom_pipe; + } + } else { + /*Get odm split index*/ + struct pipe_ctx *split_pipe = pipe_ctx->prev_odm_pipe; + + while (split_pipe) { + (*split_idx)++; + split_pipe = split_pipe->prev_odm_pipe; + } + } +} + static void calculate_viewport(struct pipe_ctx *pipe_ctx) { const struct dc_plane_state *plane_state = pipe_ctx->plane_state; @@ -541,16 +586,16 @@ static void calculate_viewport(struct pipe_ctx *pipe_ctx) struct rect clip, dest; int vpc_div = (data->format == PIXEL_FORMAT_420BPP8 || data->format == PIXEL_FORMAT_420BPP10) ? 2 : 1; - bool pri_split = pipe_ctx->bottom_pipe && - pipe_ctx->bottom_pipe->plane_state == pipe_ctx->plane_state; - bool sec_split = pipe_ctx->top_pipe && - pipe_ctx->top_pipe->plane_state == pipe_ctx->plane_state; + int split_count = 0; + int split_idx = 0; bool orthogonal_rotation, flip_y_start, flip_x_start; + calculate_split_count_and_index(pipe_ctx, &split_count, &split_idx); + if (stream->view_format == VIEW_3D_FORMAT_SIDE_BY_SIDE || stream->view_format == VIEW_3D_FORMAT_TOP_AND_BOTTOM) { - pri_split = false; - sec_split = false; + split_count = 0; + split_idx = 0; } /* The actual clip is an intersection between stream @@ -609,23 +654,32 @@ static void calculate_viewport(struct pipe_ctx *pipe_ctx) data->viewport.height = clip.height * surf_src.height / dest.height; /* Handle split */ - if (pri_split || sec_split) { + if (split_count) { + /* extra pixels in the division remainder need to go to pipes after + * the extra pixel index minus one(epimo) defined here as: + */ + int epimo = 0; + if (orthogonal_rotation) { - if (flip_y_start != pri_split) - data->viewport.height /= 2; - else { - data->viewport.y += data->viewport.height / 2; - /* Ceil offset pipe */ - data->viewport.height = (data->viewport.height + 1) / 2; - } + if (flip_y_start) + split_idx = split_count - split_idx; + + epimo = split_count - data->viewport.height % (split_count + 1); + + data->viewport.y += (data->viewport.height / (split_count + 1)) * split_idx; + if (split_idx > epimo) + data->viewport.y += split_idx - epimo - 1; + data->viewport.height = data->viewport.height / (split_count + 1) + (split_idx > epimo ? 1 : 0); } else { - if (flip_x_start != pri_split) - data->viewport.width /= 2; - else { - data->viewport.x += data->viewport.width / 2; - /* Ceil offset pipe */ - data->viewport.width = (data->viewport.width + 1) / 2; - } + if (flip_x_start) + split_idx = split_count - split_idx; + + epimo = split_count - data->viewport.width % (split_count + 1); + + data->viewport.x += (data->viewport.width / (split_count + 1)) * split_idx; + if (split_idx > epimo) + data->viewport.x += split_idx - epimo - 1; + data->viewport.width = data->viewport.width / (split_count + 1) + (split_idx > epimo ? 1 : 0); } } @@ -644,58 +698,58 @@ static void calculate_recout(struct pipe_ctx *pipe_ctx) { const struct dc_plane_state *plane_state = pipe_ctx->plane_state; const struct dc_stream_state *stream = pipe_ctx->stream; + struct scaler_data *data = &pipe_ctx->plane_res.scl_data; struct rect surf_clip = plane_state->clip_rect; - bool pri_split = pipe_ctx->bottom_pipe && - pipe_ctx->bottom_pipe->plane_state == pipe_ctx->plane_state; - bool sec_split = pipe_ctx->top_pipe && - pipe_ctx->top_pipe->plane_state == pipe_ctx->plane_state; - bool top_bottom_split = stream->view_format == VIEW_3D_FORMAT_TOP_AND_BOTTOM; - - pipe_ctx->plane_res.scl_data.recout.x = stream->dst.x; + bool pri_split_tb = pipe_ctx->bottom_pipe && + pipe_ctx->bottom_pipe->plane_state == pipe_ctx->plane_state && + stream->view_format == VIEW_3D_FORMAT_TOP_AND_BOTTOM; + bool sec_split_tb = pipe_ctx->top_pipe && + pipe_ctx->top_pipe->plane_state == pipe_ctx->plane_state && + stream->view_format == VIEW_3D_FORMAT_TOP_AND_BOTTOM; + int split_count = 0; + int split_idx = 0; + + calculate_split_count_and_index(pipe_ctx, &split_count, &split_idx); + + data->recout.x = stream->dst.x; if (stream->src.x < surf_clip.x) - pipe_ctx->plane_res.scl_data.recout.x += (surf_clip.x - - stream->src.x) * stream->dst.width + data->recout.x += (surf_clip.x - stream->src.x) * stream->dst.width / stream->src.width; - pipe_ctx->plane_res.scl_data.recout.width = surf_clip.width * - stream->dst.width / stream->src.width; - if (pipe_ctx->plane_res.scl_data.recout.width + pipe_ctx->plane_res.scl_data.recout.x > - stream->dst.x + stream->dst.width) - pipe_ctx->plane_res.scl_data.recout.width = - stream->dst.x + stream->dst.width - - pipe_ctx->plane_res.scl_data.recout.x; + data->recout.width = surf_clip.width * stream->dst.width / stream->src.width; + if (data->recout.width + data->recout.x > stream->dst.x + stream->dst.width) + data->recout.width = stream->dst.x + stream->dst.width - data->recout.x; - pipe_ctx->plane_res.scl_data.recout.y = stream->dst.y; + data->recout.y = stream->dst.y; if (stream->src.y < surf_clip.y) - pipe_ctx->plane_res.scl_data.recout.y += (surf_clip.y - - stream->src.y) * stream->dst.height + data->recout.y += (surf_clip.y - stream->src.y) * stream->dst.height / stream->src.height; - pipe_ctx->plane_res.scl_data.recout.height = surf_clip.height * - stream->dst.height / stream->src.height; - if (pipe_ctx->plane_res.scl_data.recout.height + pipe_ctx->plane_res.scl_data.recout.y > - stream->dst.y + stream->dst.height) - pipe_ctx->plane_res.scl_data.recout.height = - stream->dst.y + stream->dst.height - - pipe_ctx->plane_res.scl_data.recout.y; + data->recout.height = surf_clip.height * stream->dst.height / stream->src.height; + if (data->recout.height + data->recout.y > stream->dst.y + stream->dst.height) + data->recout.height = stream->dst.y + stream->dst.height - data->recout.y; /* Handle h & v split, handle rotation using viewport */ - if (sec_split && top_bottom_split) { - pipe_ctx->plane_res.scl_data.recout.y += - pipe_ctx->plane_res.scl_data.recout.height / 2; + if (sec_split_tb) { + data->recout.y += data->recout.height / 2; /* Floor primary pipe, ceil 2ndary pipe */ - pipe_ctx->plane_res.scl_data.recout.height = - (pipe_ctx->plane_res.scl_data.recout.height + 1) / 2; - } else if (pri_split && top_bottom_split) - pipe_ctx->plane_res.scl_data.recout.height /= 2; - else if (sec_split) { - pipe_ctx->plane_res.scl_data.recout.x += - pipe_ctx->plane_res.scl_data.recout.width / 2; - /* Ceil offset pipe */ - pipe_ctx->plane_res.scl_data.recout.width = - (pipe_ctx->plane_res.scl_data.recout.width + 1) / 2; - } else if (pri_split) - pipe_ctx->plane_res.scl_data.recout.width /= 2; + data->recout.height = (data->recout.height + 1) / 2; + } else if (pri_split_tb) + data->recout.height /= 2; + else if (split_count) { + /* extra pixels in the division remainder need to go to pipes after + * the extra pixel index minus one(epimo) defined here as: + */ + int epimo = split_count - data->recout.width % (split_count + 1); + + /*no recout offset due to odm */ + if (!pipe_ctx->next_odm_pipe && !pipe_ctx->prev_odm_pipe) { + data->recout.x += (data->recout.width / (split_count + 1)) * split_idx; + if (split_idx > epimo) + data->recout.x += split_idx - epimo - 1; + } + data->recout.width = data->recout.width / (split_count + 1) + (split_idx > epimo ? 1 : 0); + } } static void calculate_scaling_ratios(struct pipe_ctx *pipe_ctx) @@ -832,12 +886,14 @@ static void calculate_inits_and_adj_vp(struct pipe_ctx *pipe_ctx) { const struct dc_plane_state *plane_state = pipe_ctx->plane_state; const struct dc_stream_state *stream = pipe_ctx->stream; + struct pipe_ctx *odm_pipe = pipe_ctx->prev_odm_pipe; struct scaler_data *data = &pipe_ctx->plane_res.scl_data; struct rect src = pipe_ctx->plane_state->src_rect; int recout_skip_h, recout_skip_v, surf_size_h, surf_size_v; int vpc_div = (data->format == PIXEL_FORMAT_420BPP8 || data->format == PIXEL_FORMAT_420BPP10) ? 2 : 1; bool orthogonal_rotation, flip_vert_scan_dir, flip_horz_scan_dir; + int odm_idx = 0; /* * Need to calculate the scan direction for viewport to make adjustments @@ -869,6 +925,14 @@ static void calculate_inits_and_adj_vp(struct pipe_ctx *pipe_ctx) * stream->dst.width / stream->src.width - src.x * plane_state->dst_rect.width / src.width * stream->dst.width / stream->src.width); + /*modified recout_skip_h calculation due to odm having no recout offset*/ + while (odm_pipe) { + odm_idx++; + odm_pipe = odm_pipe->prev_odm_pipe; + } + if (odm_idx) + recout_skip_h += odm_idx * data->recout.width; + recout_skip_v = data->recout.y - (stream->dst.y + (plane_state->dst_rect.y - stream->src.y) * stream->dst.height / stream->src.height - src.y * plane_state->dst_rect.height / src.height @@ -1013,6 +1077,7 @@ bool resource_build_scaling_params(struct pipe_ctx *pipe_ctx) * on certain displays, such as the Sharp 4k */ pipe_ctx->plane_res.scl_data.lb_params.depth = LB_PIXEL_DEPTH_30BPP; + pipe_ctx->plane_res.scl_data.lb_params.alpha_en = plane_state->per_pixel_alpha; pipe_ctx->plane_res.scl_data.recout.x += timing->h_border_left; pipe_ctx->plane_res.scl_data.recout.y += timing->v_border_top; @@ -1021,6 +1086,8 @@ bool resource_build_scaling_params(struct pipe_ctx *pipe_ctx) store_h_border_left + timing->h_border_right; pipe_ctx->plane_res.scl_data.v_active = timing->v_addressable + timing->v_border_top + timing->v_border_bottom; + if (pipe_ctx->next_odm_pipe || pipe_ctx->prev_odm_pipe) + pipe_ctx->plane_res.scl_data.h_active /= get_num_odm_splits(pipe_ctx) + 1; /* Taps calculations */ if (pipe_ctx->plane_res.xfm != NULL) @@ -2034,7 +2101,7 @@ enum dc_status resource_map_pool_resources( for (i = 0; i < context->stream_count; i++) if (context->streams[i] == stream) { context->stream_status[i].primary_otg_inst = pipe_ctx->stream_res.tg->inst; - context->stream_status[i].stream_enc_inst = pipe_ctx->stream_res.stream_enc->id; + context->stream_status[i].stream_enc_inst = pipe_ctx->stream_res.stream_enc->stream_enc_inst; context->stream_status[i].audio_inst = pipe_ctx->stream_res.audio ? pipe_ctx->stream_res.audio->inst : -1; @@ -2108,10 +2175,10 @@ enum dc_status dc_validate_global_state( if (pipe_ctx->stream != stream) continue; - if (dc->res_pool->funcs->get_default_swizzle_mode && + if (dc->res_pool->funcs->patch_unknown_plane_state && pipe_ctx->plane_state && pipe_ctx->plane_state->tiling_info.gfx9.swizzle == DC_SW_UNKNOWN) { - result = dc->res_pool->funcs->get_default_swizzle_mode(pipe_ctx->plane_state); + result = dc->res_pool->funcs->patch_unknown_plane_state(pipe_ctx->plane_state); if (result != DC_OK) return result; } diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_vm_helper.c b/drivers/gpu/drm/amd/display/dc/core/dc_vm_helper.c index a96d8de9380e..64cf24a9ab08 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_vm_helper.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_vm_helper.c @@ -47,6 +47,9 @@ int dc_setup_system_context(struct dc *dc, struct dc_phy_addr_space_config *pa_c */ memcpy(&dc->vm_pa_config, pa_config, sizeof(struct dc_phy_addr_space_config)); dc->vm_pa_config.valid = true; + + if (pa_config->is_hvm_enabled == 0) + dc->debug.nv12_iflip_vm_wa = false; } return num_vmids; @@ -62,7 +65,7 @@ int dc_get_vmid_use_vector(struct dc *dc) int i; int in_use = 0; - for (i = 0; i < dc->vm_helper->num_vmid; i++) + for (i = 0; i < MAX_HUBP; i++) in_use |= dc->vm_helper->hubp_vmid_usage[i].vmid_usage[0] | dc->vm_helper->hubp_vmid_usage[i].vmid_usage[1]; return in_use; diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h index 8ff25b5dd2f6..1935cf6601eb 100644 --- a/drivers/gpu/drm/amd/display/dc/dc.h +++ b/drivers/gpu/drm/amd/display/dc/dc.h @@ -39,7 +39,7 @@ #include "inc/hw/dmcu.h" #include "dml/display_mode_lib.h" -#define DC_VER "3.2.69" +#define DC_VER "3.2.76" #define MAX_SURFACES 3 #define MAX_PLANES 6 @@ -126,6 +126,7 @@ struct dc_bug_wa { bool no_connect_phy_config; bool dedcn20_305_wa; bool skip_clock_update; + bool lt_early_cr_pattern; }; struct dc_dcc_surface_param { @@ -229,6 +230,7 @@ struct dc_config { bool forced_clocks; bool disable_extended_timeout_support; // Used to disable extended timeout and lttpr feature as well bool multi_mon_pp_mclk_switch; + bool psr_on_dmub; }; enum visual_confirm { @@ -388,6 +390,7 @@ struct dc_debug_options { int always_scale; bool disable_pplib_clock_request; bool disable_clock_gate; + bool disable_mem_low_power; bool disable_dmcu; bool disable_psr; bool force_abm_enable; @@ -453,6 +456,7 @@ struct dc_phy_addr_space_config { } gart_config; bool valid; + bool is_hvm_enabled; uint64_t page_table_default_page_addr; }; @@ -518,6 +522,7 @@ struct dc { /* Require to optimize clocks and bandwidth for added/removed planes */ bool optimized_required; + bool wm_optimized_required; /* Require to maintain clocks and bandwidth for UEFI enabled HW */ int optimize_seamless_boot_streams; @@ -526,6 +531,7 @@ struct dc { struct compressor *fbc_compressor; struct dc_debug_data debug_data; + struct dpcd_vendor_signature vendor_signature; const char *build_id; struct vm_helper *vm_helper; @@ -565,12 +571,14 @@ struct dc_init_data { struct dc_reg_helper_state *dmub_offload; struct dc_config flags; - uint32_t log_mask; + uint64_t log_mask; + /** * gpu_info FW provided soc bounding box struct or 0 if not * available in FW */ const struct gpu_info_soc_bounding_box_v1_0 *soc_bounding_box; + struct dpcd_vendor_signature vendor_signature; }; struct dc_callback_init { @@ -682,7 +690,6 @@ struct dc_3dlut { struct kref refcount; struct tetrahedral_params lut_3d; struct fixed31_32 hdr_multiplier; - bool initialized; /*remove after diag fix*/ union dc_3dlut_state state; struct dc_context *ctx; }; @@ -719,6 +726,7 @@ union surface_update_flags { uint32_t output_tf_change:1; uint32_t pixel_format_change:1; uint32_t plane_size_change:1; + uint32_t gamut_remap_change:1; /* Full updates */ uint32_t new_plane:1; @@ -753,6 +761,7 @@ struct dc_plane_state { struct dc_csc_transform input_csc_color_matrix; struct fixed31_32 coeff_reduction_factor; struct fixed31_32 hdr_mult; + struct colorspace_transform gamut_remap_matrix; // TODO: No longer used, remove struct dc_hdr_static_metadata hdr_static_ctx; @@ -832,6 +841,7 @@ struct dc_surface_update { const struct dc_transfer_func *func_shaper; const struct dc_3dlut *lut3d_func; const struct dc_transfer_func *blend_tf; + const struct colorspace_transform *gamut_remap_matrix; }; /* @@ -865,6 +875,7 @@ struct dc_flip_addrs { unsigned int flip_timestamp_in_us; bool flip_immediate; /* TODO: add flip duration for FreeSync */ + bool triplebuffer_flips; }; bool dc_post_update_surfaces_to_stream( @@ -979,6 +990,20 @@ struct dpcd_caps { }; +union dpcd_sink_ext_caps { + struct { + /* 0 - Sink supports backlight adjust via PWM during SDR/HDR mode + * 1 - Sink supports backlight adjust via AUX during SDR/HDR mode. + */ + uint8_t sdr_aux_backlight_control : 1; + uint8_t hdr_aux_backlight_control : 1; + uint8_t reserved_1 : 2; + uint8_t oled : 1; + uint8_t reserved : 3; + } bits; + uint8_t raw; +}; + #include "dc_link.h" /******************************************************************************* @@ -1004,6 +1029,11 @@ struct dc_sink_dsc_caps { struct dsc_dec_dpcd_caps dsc_dec_caps; }; +struct dc_sink_fec_caps { + bool is_rx_fec_supported; + bool is_topology_fec_supported; +}; + /* * The sink structure contains EDID and other display device properties */ @@ -1017,7 +1047,10 @@ struct dc_sink { struct stereo_3d_features features_3d[TIMING_3D_FORMAT_MAX]; bool converter_disable_audio; - struct dc_sink_dsc_caps sink_dsc_caps; + struct dc_sink_dsc_caps dsc_caps; + struct dc_sink_fec_caps fec_caps; + + bool is_vsc_sdp_colorimetry_supported; /* private to DC core */ struct dc_link *link; @@ -1075,7 +1108,6 @@ unsigned int dc_get_current_backlight_pwm(struct dc *dc); unsigned int dc_get_target_backlight_pwm(struct dc *dc); bool dc_is_dmcu_initialized(struct dc *dc); -bool dc_is_hw_initialized(struct dc *dc); enum dc_status dc_set_clock(struct dc *dc, enum dc_clock_type clock_type, uint32_t clk_khz, uint32_t stepping); void dc_get_clock(struct dc *dc, enum dc_clock_type clock_type, struct dc_clock_config *clock_cfg); diff --git a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h index dfe4472c9e40..bb2730e9521e 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h @@ -432,6 +432,54 @@ struct dp_sink_hw_fw_revision { uint8_t ieee_fw_rev[2]; }; +struct dpcd_vendor_signature { + bool is_valid; + + union dpcd_ieee_vendor_signature { + struct { + uint8_t ieee_oui[3];/*24-bit IEEE OUI*/ + uint8_t ieee_device_id[6];/*usually 6-byte ASCII name*/ + uint8_t ieee_hw_rev; + uint8_t ieee_fw_rev[2]; + }; + uint8_t raw[12]; + } data; +}; + +struct dpcd_amd_signature { + uint8_t AMD_IEEE_TxSignature_byte1; + uint8_t AMD_IEEE_TxSignature_byte2; + uint8_t AMD_IEEE_TxSignature_byte3; + uint8_t device_id_byte1; + uint8_t device_id_byte2; + uint8_t zero[4]; + uint8_t dce_version; + uint8_t dal_version_byte1; + uint8_t dal_version_byte2; +}; + +struct dpcd_source_backlight_set { + struct { + uint8_t byte0; + uint8_t byte1; + uint8_t byte2; + uint8_t byte3; + } backlight_level_millinits; + + struct { + uint8_t byte0; + uint8_t byte1; + } backlight_transition_time_ms; +}; + +union dpcd_source_backlight_get { + struct { + uint32_t backlight_millinits_peak; /* 326h */ + uint32_t backlight_millinits_avg; /* 32Ah */ + } bytes; + uint8_t raw[8]; +}; + /*DPCD register of DP receiver capability field bits-*/ union edp_configuration_cap { struct { diff --git a/drivers/gpu/drm/amd/display/dc/dc_hw_types.h b/drivers/gpu/drm/amd/display/dc/dc_hw_types.h index 25c50bcab9e9..a8dc3082e3e1 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_hw_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_hw_types.h @@ -385,6 +385,8 @@ struct dc_cursor_position { */ bool enable; + /* Translate cursor x/y by the source rectangle for each plane. */ + bool translate_by_source; }; struct dc_cursor_mi_param { diff --git a/drivers/gpu/drm/amd/display/dc/dc_link.h b/drivers/gpu/drm/amd/display/dc/dc_link.h index d25603128394..00ff5e98278c 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_link.h +++ b/drivers/gpu/drm/amd/display/dc/dc_link.h @@ -26,6 +26,7 @@ #ifndef DC_LINK_H_ #define DC_LINK_H_ +#include "dc.h" #include "dc_types.h" #include "grph_object_defs.h" @@ -128,6 +129,7 @@ struct dc_link { enum edp_revision edp_revision; bool psr_feature_enabled; bool psr_allow_active; + union dpcd_sink_ext_caps dpcd_sink_ext_caps; /* MST record stream using this link */ struct link_flags { @@ -178,6 +180,21 @@ bool dc_link_set_backlight_level(const struct dc_link *dc_link, uint32_t backlight_pwm_u16_16, uint32_t frame_ramp); +/* Set/get nits-based backlight level of an embedded panel (eDP, LVDS). */ +bool dc_link_set_backlight_level_nits(struct dc_link *link, + bool isHDR, + uint32_t backlight_millinits, + uint32_t transition_time_in_ms); + +bool dc_link_get_backlight_level_nits(struct dc_link *link, + uint32_t *backlight_millinits, + uint32_t *backlight_millinits_peak); + +bool dc_link_backlight_enable_aux(struct dc_link *link, bool enable); + +bool dc_link_read_default_bl_aux(struct dc_link *link, uint32_t *backlight_millinits); +bool dc_link_set_default_brightness_aux(struct dc_link *link); + int dc_link_get_backlight_level(const struct dc_link *dc_link); bool dc_link_set_abm_disable(const struct dc_link *dc_link); @@ -316,4 +333,7 @@ bool dc_submit_i2c_oem( uint32_t dc_bandwidth_in_kbps_from_timing( const struct dc_crtc_timing *timing); + +bool dc_link_is_fec_supported(const struct dc_link *link); + #endif /* DC_LINK_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dc_stream.h b/drivers/gpu/drm/amd/display/dc/dc_stream.h index 92096de79dec..a5c7ef47b8d3 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_stream.h +++ b/drivers/gpu/drm/amd/display/dc/dc_stream.h @@ -118,6 +118,7 @@ union stream_update_flags { uint32_t dpms_off:1; uint32_t gamut_remap:1; uint32_t wb_update:1; + uint32_t dsc_changed : 1; } bits; uint32_t raw; diff --git a/drivers/gpu/drm/amd/display/dc/dc_types.h b/drivers/gpu/drm/amd/display/dc/dc_types.h index e59532d98cb4..0d210104ba0a 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_types.h @@ -229,7 +229,9 @@ struct dc_panel_patch { unsigned int extra_t12_ms; unsigned int extra_delay_backlight_off; unsigned int extra_t7_ms; - unsigned int manage_secondary_link; + unsigned int skip_scdc_overwrite; + unsigned int delay_ignore_msa; + unsigned int disable_fec; }; struct dc_edid_caps { diff --git a/drivers/gpu/drm/amd/display/dc/dce/Makefile b/drivers/gpu/drm/amd/display/dc/dce/Makefile index fdf3d8f87eee..fbfcff700971 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dce/Makefile @@ -29,7 +29,7 @@ DCE = dce_audio.o dce_stream_encoder.o dce_link_encoder.o dce_hwseq.o \ dce_mem_input.o dce_clock_source.o dce_scl_filters.o dce_transform.o \ dce_opp.o dce_dmcu.o dce_abm.o dce_ipp.o dce_aux.o \ -dce_i2c.o dce_i2c_hw.o dce_i2c_sw.o +dce_i2c.o dce_i2c_hw.o dce_i2c_sw.o dmub_psr.o AMD_DAL_DCE = $(addprefix $(AMDDALPATH)/dc/dce/,$(DCE)) diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c b/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c index 68c4049cbc2a..743042d5905a 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c @@ -645,7 +645,7 @@ bool dce_aux_transfer_with_retries(struct ddc_service *ddc, case AUX_TRANSACTION_REPLY_AUX_DEFER: case AUX_TRANSACTION_REPLY_I2C_OVER_AUX_DEFER: retry_on_defer = true; - /* fall through */ + fallthrough; case AUX_TRANSACTION_REPLY_I2C_OVER_AUX_NACK: if (++aux_defer_retries >= AUX_MAX_DEFER_RETRIES) { goto fail; diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c b/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c index 30d953acd016..f0cebe721bcc 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c @@ -378,6 +378,11 @@ static bool dcn10_dmcu_init(struct dmcu *dmcu) struct dce_dmcu *dmcu_dce = TO_DCE_DMCU(dmcu); const struct dc_config *config = &dmcu->ctx->dc->config; bool status = false; + struct dc_context *ctx = dmcu->ctx; + unsigned int i; + // 5 4 3 2 1 0 + // F E D C B A - bit 0 is A, bit 5 is F + unsigned int tx_interrupt_mask = 0; PERF_TRACE(); /* Definition of DC_DMCU_SCRATCH @@ -387,6 +392,15 @@ static bool dcn10_dmcu_init(struct dmcu *dmcu) */ dmcu->dmcu_state = REG_READ(DC_DMCU_SCRATCH); + for (i = 0; i < ctx->dc->link_count; i++) { + if (ctx->dc->links[i]->link_enc->features.flags.bits.DP_IS_USB_C) { + if (ctx->dc->links[i]->link_enc->transmitter >= TRANSMITTER_UNIPHY_A && + ctx->dc->links[i]->link_enc->transmitter <= TRANSMITTER_UNIPHY_F) { + tx_interrupt_mask |= 1 << ctx->dc->links[i]->link_enc->transmitter; + } + } + } + switch (dmcu->dmcu_state) { case DMCU_UNLOADED: status = false; @@ -401,6 +415,8 @@ static bool dcn10_dmcu_init(struct dmcu *dmcu) /* Set backlight ramping stepsize */ REG_WRITE(MASTER_COMM_DATA_REG2, abm_gain_stepsize); + REG_WRITE(MASTER_COMM_DATA_REG3, tx_interrupt_mask); + /* Set command to initialize microcontroller */ REG_UPDATE(MASTER_COMM_CMD_REG, MASTER_COMM_CMD_REG_BYTE0, MCP_INIT_DMCU); diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.c b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.c index 066188ba7949..24adec407972 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.c @@ -267,6 +267,9 @@ static void set_speed( uint32_t xtal_ref_div = 0; uint32_t prescale = 0; + if (speed == 0) + return; + REG_GET(MICROSECOND_TIME_BASE_DIV, XTAL_REF_DIV, &xtal_ref_div); if (xtal_ref_div == 0) @@ -274,17 +277,15 @@ static void set_speed( prescale = ((dce_i2c_hw->reference_frequency * 2) / xtal_ref_div) / speed; - if (speed) { - if (dce_i2c_hw->masks->DC_I2C_DDC1_START_STOP_TIMING_CNTL) - REG_UPDATE_N(SPEED, 3, - FN(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_PRESCALE), prescale, - FN(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_THRESHOLD), 2, - FN(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_START_STOP_TIMING_CNTL), speed > 50 ? 2:1); - else - REG_UPDATE_N(SPEED, 2, - FN(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_PRESCALE), prescale, - FN(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_THRESHOLD), 2); - } + if (dce_i2c_hw->masks->DC_I2C_DDC1_START_STOP_TIMING_CNTL) + REG_UPDATE_N(SPEED, 3, + FN(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_PRESCALE), prescale, + FN(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_THRESHOLD), 2, + FN(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_START_STOP_TIMING_CNTL), speed > 50 ? 2:1); + else + REG_UPDATE_N(SPEED, 2, + FN(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_PRESCALE), prescale, + FN(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_THRESHOLD), 2); } static bool setup_engine( diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_mem_input.c b/drivers/gpu/drm/amd/display/dc/dce/dce_mem_input.c index 8aa937f496c4..51481e922eb9 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_mem_input.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_mem_input.c @@ -479,7 +479,7 @@ static void program_grph_pixel_format( case SURFACE_PIXEL_FORMAT_GRPH_ABGR16161616F: sign = 1; floating = 1; - /* fall through */ + fallthrough; case SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616F: /* shouldn't this get float too? */ case SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616: grph_depth = 3; diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_scl_filters.c b/drivers/gpu/drm/amd/display/dc/dce/dce_scl_filters.c index 48862bebf29e..7311f312369f 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_scl_filters.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_scl_filters.c @@ -22,1004 +22,1330 @@ * Authors: AMD * */ - #include "transform.h" +//========================================= +// <num_taps> = 2 +// <num_phases> = 16 +// <scale_ratio> = 0.833333 (input/output) +// <sharpness> = 0 +// <CoefType> = ModifiedLanczos +// <CoefQuant> = s1.10 +// <CoefOut> = s1.12 +//========================================= static const uint16_t filter_2tap_16p[18] = { - 4096, 0, - 3840, 256, - 3584, 512, - 3328, 768, - 3072, 1024, - 2816, 1280, - 2560, 1536, - 2304, 1792, - 2048, 2048 + 0x1000, 0x0000, + 0x0FF0, 0x0010, + 0x0FB0, 0x0050, + 0x0F34, 0x00CC, + 0x0E68, 0x0198, + 0x0D44, 0x02BC, + 0x0BC4, 0x043C, + 0x09FC, 0x0604, + 0x0800, 0x0800 }; +//========================================= +// <num_taps> = 3 +// <num_phases> = 16 +// <scale_ratio> = 0.83333 (input/output) +// <sharpness> = 0 +// <CoefType> = ModifiedLanczos +// <CoefQuant> = 1.10 +// <CoefOut> = 1.12 +//========================================= static const uint16_t filter_3tap_16p_upscale[27] = { - 2048, 2048, 0, - 1708, 2424, 16348, - 1372, 2796, 16308, - 1056, 3148, 16272, - 768, 3464, 16244, - 512, 3728, 16236, - 296, 3928, 16252, - 124, 4052, 16296, - 0, 4096, 0 + 0x0804, 0x07FC, 0x0000, + 0x06AC, 0x0978, 0x3FDC, + 0x055C, 0x0AF0, 0x3FB4, + 0x0420, 0x0C50, 0x3F90, + 0x0300, 0x0D88, 0x3F78, + 0x0200, 0x0E90, 0x3F70, + 0x0128, 0x0F5C, 0x3F7C, + 0x007C, 0x0FD8, 0x3FAC, + 0x0000, 0x1000, 0x0000 }; -static const uint16_t filter_3tap_16p_117[27] = { - 2048, 2048, 0, - 1824, 2276, 16376, - 1600, 2496, 16380, - 1376, 2700, 16, - 1156, 2880, 52, - 948, 3032, 108, - 756, 3144, 192, - 580, 3212, 296, - 428, 3236, 428 +//========================================= +// <num_taps> = 3 +// <num_phases> = 16 +// <scale_ratio> = 1.16666 (input/output) +// <sharpness> = 0 +// <CoefType> = ModifiedLanczos +// <CoefQuant> = 1.10 +// <CoefOut> = 1.12 +//========================================= +static const uint16_t filter_3tap_16p_116[27] = { + 0x0804, 0x07FC, 0x0000, + 0x0700, 0x0914, 0x3FEC, + 0x0604, 0x0A1C, 0x3FE0, + 0x050C, 0x0B14, 0x3FE0, + 0x041C, 0x0BF4, 0x3FF0, + 0x0340, 0x0CB0, 0x0010, + 0x0274, 0x0D3C, 0x0050, + 0x01C0, 0x0D94, 0x00AC, + 0x0128, 0x0DB4, 0x0124 }; -static const uint16_t filter_3tap_16p_150[27] = { - 2048, 2048, 0, - 1872, 2184, 36, - 1692, 2308, 88, - 1516, 2420, 156, - 1340, 2516, 236, - 1168, 2592, 328, - 1004, 2648, 440, - 844, 2684, 560, - 696, 2696, 696 +//========================================= +// <num_taps> = 3 +// <num_phases> = 16 +// <scale_ratio> = 1.49999 (input/output) +// <sharpness> = 0 +// <CoefType> = ModifiedLanczos +// <CoefQuant> = 1.10 +// <CoefOut> = 1.12 +//========================================= +static const uint16_t filter_3tap_16p_149[27] = { + 0x0804, 0x07FC, 0x0000, + 0x0730, 0x08CC, 0x0004, + 0x0660, 0x098C, 0x0014, + 0x0590, 0x0A3C, 0x0034, + 0x04C4, 0x0AD4, 0x0068, + 0x0400, 0x0B54, 0x00AC, + 0x0348, 0x0BB0, 0x0108, + 0x029C, 0x0BEC, 0x0178, + 0x0200, 0x0C00, 0x0200 }; +//========================================= +// <num_taps> = 3 +// <num_phases> = 16 +// <scale_ratio> = 1.83332 (input/output) +// <sharpness> = 0 +// <CoefType> = ModifiedLanczos +// <CoefQuant> = 1.10 +// <CoefOut> = 1.12 +//========================================= static const uint16_t filter_3tap_16p_183[27] = { - 2048, 2048, 0, - 1892, 2104, 92, - 1744, 2152, 196, - 1592, 2196, 300, - 1448, 2232, 412, - 1304, 2256, 528, - 1168, 2276, 648, - 1032, 2288, 772, - 900, 2292, 900 + 0x0804, 0x07FC, 0x0000, + 0x0754, 0x0880, 0x002C, + 0x06A8, 0x08F0, 0x0068, + 0x05FC, 0x0954, 0x00B0, + 0x0550, 0x09AC, 0x0104, + 0x04A8, 0x09F0, 0x0168, + 0x0408, 0x0A20, 0x01D8, + 0x036C, 0x0A40, 0x0254, + 0x02DC, 0x0A48, 0x02DC }; +//========================================= +// <num_taps> = 4 +// <num_phases> = 16 +// <scale_ratio> = 0.83333 (input/output) +// <sharpness> = 0 +// <CoefType> = ModifiedLanczos +// <CoefQuant> = 1.10 +// <CoefOut> = 1.12 +//========================================= static const uint16_t filter_4tap_16p_upscale[36] = { - 0, 4096, 0, 0, - 16240, 4056, 180, 16380, - 16136, 3952, 404, 16364, - 16072, 3780, 664, 16344, - 16040, 3556, 952, 16312, - 16036, 3284, 1268, 16272, - 16052, 2980, 1604, 16224, - 16084, 2648, 1952, 16176, - 16128, 2304, 2304, 16128 + 0x0000, 0x1000, 0x0000, 0x0000, + 0x3F74, 0x0FDC, 0x00B4, 0x3FFC, + 0x3F0C, 0x0F70, 0x0194, 0x3FF0, + 0x3ECC, 0x0EC4, 0x0298, 0x3FD8, + 0x3EAC, 0x0DE4, 0x03B8, 0x3FB8, + 0x3EA4, 0x0CD8, 0x04F4, 0x3F90, + 0x3EB8, 0x0BA0, 0x0644, 0x3F64, + 0x3ED8, 0x0A54, 0x07A0, 0x3F34, + 0x3F00, 0x08FC, 0x0900, 0x3F04 }; -static const uint16_t filter_4tap_16p_117[36] = { - 428, 3236, 428, 0, - 276, 3232, 604, 16364, - 148, 3184, 800, 16340, - 44, 3104, 1016, 16312, - 16344, 2984, 1244, 16284, - 16284, 2832, 1488, 16256, - 16244, 2648, 1732, 16236, - 16220, 2440, 1976, 16220, - 16212, 2216, 2216, 16212 +//========================================= +// <num_taps> = 4 +// <num_phases> = 16 +// <scale_ratio> = 1.16666 (input/output) +// <sharpness> = 0 +// <CoefType> = ModifiedLanczos +// <CoefQuant> = 1.10 +// <CoefOut> = 1.12 +//========================================= +static const uint16_t filter_4tap_16p_116[36] = { + 0x01A8, 0x0CB4, 0x01A4, 0x0000, + 0x0110, 0x0CB0, 0x0254, 0x3FEC, + 0x0090, 0x0C80, 0x031C, 0x3FD4, + 0x0024, 0x0C2C, 0x03F4, 0x3FBC, + 0x3FD8, 0x0BAC, 0x04DC, 0x3FA0, + 0x3F9C, 0x0B14, 0x05CC, 0x3F84, + 0x3F70, 0x0A60, 0x06C4, 0x3F6C, + 0x3F5C, 0x098C, 0x07BC, 0x3F5C, + 0x3F54, 0x08AC, 0x08AC, 0x3F54 }; -static const uint16_t filter_4tap_16p_150[36] = { - 696, 2700, 696, 0, - 560, 2700, 848, 16364, - 436, 2676, 1008, 16348, - 328, 2628, 1180, 16336, - 232, 2556, 1356, 16328, - 152, 2460, 1536, 16328, - 84, 2344, 1716, 16332, - 28, 2208, 1888, 16348, - 16376, 2052, 2052, 16376 +//========================================= +// <num_taps> = 4 +// <num_phases> = 16 +// <scale_ratio> = 1.49999 (input/output) +// <sharpness> = 0 +// <CoefType> = ModifiedLanczos +// <CoefQuant> = 1.10 +// <CoefOut> = 1.12 +//========================================= +static const uint16_t filter_4tap_16p_149[36] = { + 0x02B8, 0x0A90, 0x02B8, 0x0000, + 0x0230, 0x0A90, 0x0350, 0x3FF0, + 0x01B8, 0x0A78, 0x03F0, 0x3FE0, + 0x0148, 0x0A48, 0x049C, 0x3FD4, + 0x00E8, 0x0A00, 0x054C, 0x3FCC, + 0x0098, 0x09A0, 0x0600, 0x3FC8, + 0x0054, 0x0928, 0x06B4, 0x3FD0, + 0x001C, 0x08A4, 0x0760, 0x3FE0, + 0x3FFC, 0x0804, 0x0804, 0x3FFC }; +//========================================= +// <num_taps> = 4 +// <num_phases> = 16 +// <scale_ratio> = 1.83332 (input/output) +// <sharpness> = 0 +// <CoefType> = ModifiedLanczos +// <CoefQuant> = 1.10 +// <CoefOut> = 1.12 +//========================================= static const uint16_t filter_4tap_16p_183[36] = { - 940, 2208, 940, 0, - 832, 2200, 1052, 4, - 728, 2180, 1164, 16, - 628, 2148, 1280, 36, - 536, 2100, 1392, 60, - 448, 2044, 1504, 92, - 368, 1976, 1612, 132, - 296, 1900, 1716, 176, - 232, 1812, 1812, 232 + 0x03B0, 0x08A0, 0x03B0, 0x0000, + 0x0348, 0x0898, 0x041C, 0x0004, + 0x02DC, 0x0884, 0x0490, 0x0010, + 0x0278, 0x0864, 0x0500, 0x0024, + 0x021C, 0x0838, 0x0570, 0x003C, + 0x01C8, 0x07FC, 0x05E0, 0x005C, + 0x0178, 0x07B8, 0x064C, 0x0084, + 0x0130, 0x076C, 0x06B0, 0x00B4, + 0x00F0, 0x0714, 0x0710, 0x00EC }; +//========================================= +// <num_taps> = 2 +// <num_phases> = 64 +// <scale_ratio> = 0.833333 (input/output) +// <sharpness> = 0 +// <CoefType> = ModifiedLanczos +// <CoefQuant> = s1.10 +// <CoefOut> = s1.12 +//========================================= static const uint16_t filter_2tap_64p[66] = { - 4096, 0, - 4032, 64, - 3968, 128, - 3904, 192, - 3840, 256, - 3776, 320, - 3712, 384, - 3648, 448, - 3584, 512, - 3520, 576, - 3456, 640, - 3392, 704, - 3328, 768, - 3264, 832, - 3200, 896, - 3136, 960, - 3072, 1024, - 3008, 1088, - 2944, 1152, - 2880, 1216, - 2816, 1280, - 2752, 1344, - 2688, 1408, - 2624, 1472, - 2560, 1536, - 2496, 1600, - 2432, 1664, - 2368, 1728, - 2304, 1792, - 2240, 1856, - 2176, 1920, - 2112, 1984, - 2048, 2048 }; + 0x1000, 0x0000, + 0x1000, 0x0000, + 0x0FFC, 0x0004, + 0x0FF8, 0x0008, + 0x0FF0, 0x0010, + 0x0FE4, 0x001C, + 0x0FD8, 0x0028, + 0x0FC4, 0x003C, + 0x0FB0, 0x0050, + 0x0F98, 0x0068, + 0x0F7C, 0x0084, + 0x0F58, 0x00A8, + 0x0F34, 0x00CC, + 0x0F08, 0x00F8, + 0x0ED8, 0x0128, + 0x0EA4, 0x015C, + 0x0E68, 0x0198, + 0x0E28, 0x01D8, + 0x0DE4, 0x021C, + 0x0D98, 0x0268, + 0x0D44, 0x02BC, + 0x0CEC, 0x0314, + 0x0C90, 0x0370, + 0x0C2C, 0x03D4, + 0x0BC4, 0x043C, + 0x0B58, 0x04A8, + 0x0AE8, 0x0518, + 0x0A74, 0x058C, + 0x09FC, 0x0604, + 0x0980, 0x0680, + 0x0900, 0x0700, + 0x0880, 0x0780, + 0x0800, 0x0800 +}; +//========================================= +// <num_taps> = 3 +// <num_phases> = 64 +// <scale_ratio> = 0.83333 (input/output) +// <sharpness> = 0 +// <CoefType> = ModifiedLanczos +// <CoefQuant> = 1.10 +// <CoefOut> = 1.12 +//========================================= static const uint16_t filter_3tap_64p_upscale[99] = { - 2048, 2048, 0, - 1960, 2140, 16376, - 1876, 2236, 16364, - 1792, 2328, 16356, - 1708, 2424, 16348, - 1620, 2516, 16336, - 1540, 2612, 16328, - 1456, 2704, 16316, - 1372, 2796, 16308, - 1292, 2884, 16296, - 1212, 2976, 16288, - 1136, 3060, 16280, - 1056, 3148, 16272, - 984, 3228, 16264, - 908, 3312, 16256, - 836, 3388, 16248, - 768, 3464, 16244, - 700, 3536, 16240, - 636, 3604, 16236, - 572, 3668, 16236, - 512, 3728, 16236, - 456, 3784, 16236, - 400, 3836, 16240, - 348, 3884, 16244, - 296, 3928, 16252, - 252, 3964, 16260, - 204, 4000, 16268, - 164, 4028, 16284, - 124, 4052, 16296, - 88, 4072, 16316, - 56, 4084, 16336, - 24, 4092, 16356, - 0, 4096, 0 + 0x0804, 0x07FC, 0x0000, + 0x07A8, 0x0860, 0x3FF8, + 0x0754, 0x08BC, 0x3FF0, + 0x0700, 0x0918, 0x3FE8, + 0x06AC, 0x0978, 0x3FDC, + 0x0654, 0x09D8, 0x3FD4, + 0x0604, 0x0A34, 0x3FC8, + 0x05B0, 0x0A90, 0x3FC0, + 0x055C, 0x0AF0, 0x3FB4, + 0x050C, 0x0B48, 0x3FAC, + 0x04BC, 0x0BA0, 0x3FA4, + 0x0470, 0x0BF4, 0x3F9C, + 0x0420, 0x0C50, 0x3F90, + 0x03D8, 0x0C9C, 0x3F8C, + 0x038C, 0x0CF0, 0x3F84, + 0x0344, 0x0D40, 0x3F7C, + 0x0300, 0x0D88, 0x3F78, + 0x02BC, 0x0DD0, 0x3F74, + 0x027C, 0x0E14, 0x3F70, + 0x023C, 0x0E54, 0x3F70, + 0x0200, 0x0E90, 0x3F70, + 0x01C8, 0x0EC8, 0x3F70, + 0x0190, 0x0EFC, 0x3F74, + 0x015C, 0x0F2C, 0x3F78, + 0x0128, 0x0F5C, 0x3F7C, + 0x00FC, 0x0F7C, 0x3F88, + 0x00CC, 0x0FA4, 0x3F90, + 0x00A4, 0x0FC0, 0x3F9C, + 0x007C, 0x0FD8, 0x3FAC, + 0x0058, 0x0FE8, 0x3FC0, + 0x0038, 0x0FF4, 0x3FD4, + 0x0018, 0x1000, 0x3FE8, + 0x0000, 0x1000, 0x0000 }; -static const uint16_t filter_3tap_64p_117[99] = { - 2048, 2048, 0, - 1992, 2104, 16380, - 1936, 2160, 16380, - 1880, 2220, 16376, - 1824, 2276, 16376, - 1768, 2332, 16376, - 1712, 2388, 16376, - 1656, 2444, 16376, - 1600, 2496, 16380, - 1544, 2548, 0, - 1488, 2600, 4, - 1432, 2652, 8, - 1376, 2700, 16, - 1320, 2748, 20, - 1264, 2796, 32, - 1212, 2840, 40, - 1156, 2880, 52, - 1104, 2920, 64, - 1052, 2960, 80, - 1000, 2996, 92, - 948, 3032, 108, - 900, 3060, 128, - 852, 3092, 148, - 804, 3120, 168, - 756, 3144, 192, - 712, 3164, 216, - 668, 3184, 240, - 624, 3200, 268, - 580, 3212, 296, - 540, 3220, 328, - 500, 3228, 360, - 464, 3232, 392, - 428, 3236, 428 +//========================================= +// <num_taps> = 3 +// <num_phases> = 64 +// <scale_ratio> = 1.16666 (input/output) +// <sharpness> = 0 +// <CoefType> = ModifiedLanczos +// <CoefQuant> = 1.10 +// <CoefOut> = 1.12 +//========================================= +static const uint16_t filter_3tap_64p_116[99] = { + 0x0804, 0x07FC, 0x0000, + 0x07C0, 0x0844, 0x3FFC, + 0x0780, 0x0888, 0x3FF8, + 0x0740, 0x08D0, 0x3FF0, + 0x0700, 0x0914, 0x3FEC, + 0x06C0, 0x0958, 0x3FE8, + 0x0684, 0x0998, 0x3FE4, + 0x0644, 0x09DC, 0x3FE0, + 0x0604, 0x0A1C, 0x3FE0, + 0x05C4, 0x0A5C, 0x3FE0, + 0x0588, 0x0A9C, 0x3FDC, + 0x0548, 0x0ADC, 0x3FDC, + 0x050C, 0x0B14, 0x3FE0, + 0x04CC, 0x0B54, 0x3FE0, + 0x0490, 0x0B8C, 0x3FE4, + 0x0458, 0x0BC0, 0x3FE8, + 0x041C, 0x0BF4, 0x3FF0, + 0x03E0, 0x0C28, 0x3FF8, + 0x03A8, 0x0C58, 0x0000, + 0x0374, 0x0C88, 0x0004, + 0x0340, 0x0CB0, 0x0010, + 0x0308, 0x0CD8, 0x0020, + 0x02D8, 0x0CFC, 0x002C, + 0x02A0, 0x0D20, 0x0040, + 0x0274, 0x0D3C, 0x0050, + 0x0244, 0x0D58, 0x0064, + 0x0214, 0x0D70, 0x007C, + 0x01E8, 0x0D84, 0x0094, + 0x01C0, 0x0D94, 0x00AC, + 0x0198, 0x0DA0, 0x00C8, + 0x0170, 0x0DAC, 0x00E4, + 0x014C, 0x0DB0, 0x0104, + 0x0128, 0x0DB4, 0x0124 }; -static const uint16_t filter_3tap_64p_150[99] = { - 2048, 2048, 0, - 2004, 2080, 8, - 1960, 2116, 16, - 1916, 2148, 28, - 1872, 2184, 36, - 1824, 2216, 48, - 1780, 2248, 60, - 1736, 2280, 76, - 1692, 2308, 88, - 1648, 2336, 104, - 1604, 2368, 120, - 1560, 2392, 136, - 1516, 2420, 156, - 1472, 2444, 172, - 1428, 2472, 192, - 1384, 2492, 212, - 1340, 2516, 236, - 1296, 2536, 256, - 1252, 2556, 280, - 1212, 2576, 304, - 1168, 2592, 328, - 1124, 2608, 356, - 1084, 2624, 384, - 1044, 2636, 412, - 1004, 2648, 440, - 964, 2660, 468, - 924, 2668, 500, - 884, 2676, 528, - 844, 2684, 560, - 808, 2688, 596, - 768, 2692, 628, - 732, 2696, 664, - 696, 2696, 696 +//========================================= +// <num_taps> = 3 +// <num_phases> = 64 +// <scale_ratio> = 1.49999 (input/output) +// <sharpness> = 0 +// <CoefType> = ModifiedLanczos +// <CoefQuant> = 1.10 +// <CoefOut> = 1.12 +//========================================= +static const uint16_t filter_3tap_64p_149[99] = { + 0x0804, 0x07FC, 0x0000, + 0x07CC, 0x0834, 0x0000, + 0x0798, 0x0868, 0x0000, + 0x0764, 0x089C, 0x0000, + 0x0730, 0x08CC, 0x0004, + 0x0700, 0x08FC, 0x0004, + 0x06CC, 0x092C, 0x0008, + 0x0698, 0x095C, 0x000C, + 0x0660, 0x098C, 0x0014, + 0x062C, 0x09B8, 0x001C, + 0x05FC, 0x09E4, 0x0020, + 0x05C4, 0x0A10, 0x002C, + 0x0590, 0x0A3C, 0x0034, + 0x055C, 0x0A64, 0x0040, + 0x0528, 0x0A8C, 0x004C, + 0x04F8, 0x0AB0, 0x0058, + 0x04C4, 0x0AD4, 0x0068, + 0x0490, 0x0AF8, 0x0078, + 0x0460, 0x0B18, 0x0088, + 0x0430, 0x0B38, 0x0098, + 0x0400, 0x0B54, 0x00AC, + 0x03D0, 0x0B6C, 0x00C4, + 0x03A0, 0x0B88, 0x00D8, + 0x0374, 0x0B9C, 0x00F0, + 0x0348, 0x0BB0, 0x0108, + 0x0318, 0x0BC4, 0x0124, + 0x02EC, 0x0BD4, 0x0140, + 0x02C4, 0x0BE0, 0x015C, + 0x029C, 0x0BEC, 0x0178, + 0x0274, 0x0BF4, 0x0198, + 0x024C, 0x0BFC, 0x01B8, + 0x0228, 0x0BFC, 0x01DC, + 0x0200, 0x0C00, 0x0200 }; +//========================================= +// <num_taps> = 3 +// <num_phases> = 64 +// <scale_ratio> = 1.83332 (input/output) +// <sharpness> = 0 +// <CoefType> = ModifiedLanczos +// <CoefQuant> = 1.10 +// <CoefOut> = 1.12 +//========================================= static const uint16_t filter_3tap_64p_183[99] = { - 2048, 2048, 0, - 2008, 2060, 20, - 1968, 2076, 44, - 1932, 2088, 68, - 1892, 2104, 92, - 1856, 2116, 120, - 1816, 2128, 144, - 1780, 2140, 168, - 1744, 2152, 196, - 1704, 2164, 220, - 1668, 2176, 248, - 1632, 2188, 272, - 1592, 2196, 300, - 1556, 2204, 328, - 1520, 2216, 356, - 1484, 2224, 384, - 1448, 2232, 412, - 1412, 2240, 440, - 1376, 2244, 468, - 1340, 2252, 496, - 1304, 2256, 528, - 1272, 2264, 556, - 1236, 2268, 584, - 1200, 2272, 616, - 1168, 2276, 648, - 1132, 2280, 676, - 1100, 2284, 708, - 1064, 2288, 740, - 1032, 2288, 772, - 996, 2292, 800, - 964, 2292, 832, - 932, 2292, 868, - 900, 2292, 900 + 0x0804, 0x07FC, 0x0000, + 0x07D4, 0x0824, 0x0008, + 0x07AC, 0x0840, 0x0014, + 0x0780, 0x0860, 0x0020, + 0x0754, 0x0880, 0x002C, + 0x0728, 0x089C, 0x003C, + 0x0700, 0x08B8, 0x0048, + 0x06D4, 0x08D4, 0x0058, + 0x06A8, 0x08F0, 0x0068, + 0x067C, 0x090C, 0x0078, + 0x0650, 0x0924, 0x008C, + 0x0628, 0x093C, 0x009C, + 0x05FC, 0x0954, 0x00B0, + 0x05D0, 0x096C, 0x00C4, + 0x05A8, 0x0980, 0x00D8, + 0x0578, 0x0998, 0x00F0, + 0x0550, 0x09AC, 0x0104, + 0x0528, 0x09BC, 0x011C, + 0x04FC, 0x09D0, 0x0134, + 0x04D4, 0x09E0, 0x014C, + 0x04A8, 0x09F0, 0x0168, + 0x0480, 0x09FC, 0x0184, + 0x045C, 0x0A08, 0x019C, + 0x0434, 0x0A14, 0x01B8, + 0x0408, 0x0A20, 0x01D8, + 0x03E0, 0x0A2C, 0x01F4, + 0x03B8, 0x0A34, 0x0214, + 0x0394, 0x0A38, 0x0234, + 0x036C, 0x0A40, 0x0254, + 0x0348, 0x0A44, 0x0274, + 0x0324, 0x0A48, 0x0294, + 0x0300, 0x0A48, 0x02B8, + 0x02DC, 0x0A48, 0x02DC }; +//========================================= +// <num_taps> = 4 +// <num_phases> = 64 +// <scale_ratio> = 0.83333 (input/output) +// <sharpness> = 0 +// <CoefType> = ModifiedLanczos +// <CoefQuant> = 1.10 +// <CoefOut> = 1.12 +//========================================= static const uint16_t filter_4tap_64p_upscale[132] = { - 0, 4096, 0, 0, - 16344, 4092, 40, 0, - 16308, 4084, 84, 16380, - 16272, 4072, 132, 16380, - 16240, 4056, 180, 16380, - 16212, 4036, 232, 16376, - 16184, 4012, 288, 16372, - 16160, 3984, 344, 16368, - 16136, 3952, 404, 16364, - 16116, 3916, 464, 16360, - 16100, 3872, 528, 16356, - 16084, 3828, 596, 16348, - 16072, 3780, 664, 16344, - 16060, 3728, 732, 16336, - 16052, 3676, 804, 16328, - 16044, 3616, 876, 16320, - 16040, 3556, 952, 16312, - 16036, 3492, 1028, 16300, - 16032, 3424, 1108, 16292, - 16032, 3356, 1188, 16280, - 16036, 3284, 1268, 16272, - 16036, 3212, 1352, 16260, - 16040, 3136, 1436, 16248, - 16044, 3056, 1520, 16236, - 16052, 2980, 1604, 16224, - 16060, 2896, 1688, 16212, - 16064, 2816, 1776, 16200, - 16076, 2732, 1864, 16188, - 16084, 2648, 1952, 16176, - 16092, 2564, 2040, 16164, - 16104, 2476, 2128, 16152, - 16116, 2388, 2216, 16140, - 16128, 2304, 2304, 16128 }; + 0x0000, 0x1000, 0x0000, 0x0000, + 0x3FDC, 0x0FFC, 0x0028, 0x0000, + 0x3FB4, 0x0FF8, 0x0054, 0x0000, + 0x3F94, 0x0FE8, 0x0084, 0x0000, + 0x3F74, 0x0FDC, 0x00B4, 0x3FFC, + 0x3F58, 0x0FC4, 0x00E8, 0x3FFC, + 0x3F3C, 0x0FAC, 0x0120, 0x3FF8, + 0x3F24, 0x0F90, 0x0158, 0x3FF4, + 0x3F0C, 0x0F70, 0x0194, 0x3FF0, + 0x3EF8, 0x0F4C, 0x01D0, 0x3FEC, + 0x3EE8, 0x0F20, 0x0210, 0x3FE8, + 0x3ED8, 0x0EF4, 0x0254, 0x3FE0, + 0x3ECC, 0x0EC4, 0x0298, 0x3FD8, + 0x3EC0, 0x0E90, 0x02DC, 0x3FD4, + 0x3EB8, 0x0E58, 0x0324, 0x3FCC, + 0x3EB0, 0x0E20, 0x036C, 0x3FC4, + 0x3EAC, 0x0DE4, 0x03B8, 0x3FB8, + 0x3EA8, 0x0DA4, 0x0404, 0x3FB0, + 0x3EA4, 0x0D60, 0x0454, 0x3FA8, + 0x3EA4, 0x0D1C, 0x04A4, 0x3F9C, + 0x3EA4, 0x0CD8, 0x04F4, 0x3F90, + 0x3EA8, 0x0C88, 0x0548, 0x3F88, + 0x3EAC, 0x0C3C, 0x059C, 0x3F7C, + 0x3EB0, 0x0BF0, 0x05F0, 0x3F70, + 0x3EB8, 0x0BA0, 0x0644, 0x3F64, + 0x3EBC, 0x0B54, 0x0698, 0x3F58, + 0x3EC4, 0x0B00, 0x06F0, 0x3F4C, + 0x3ECC, 0x0AAC, 0x0748, 0x3F40, + 0x3ED8, 0x0A54, 0x07A0, 0x3F34, + 0x3EE0, 0x0A04, 0x07F8, 0x3F24, + 0x3EEC, 0x09AC, 0x0850, 0x3F18, + 0x3EF8, 0x0954, 0x08A8, 0x3F0C, + 0x3F00, 0x08FC, 0x0900, 0x3F04 +}; -static const uint16_t filter_4tap_64p_117[132] = { - 420, 3248, 420, 0, - 380, 3248, 464, 16380, - 344, 3248, 508, 16372, - 308, 3248, 552, 16368, - 272, 3240, 596, 16364, - 236, 3236, 644, 16356, - 204, 3224, 692, 16352, - 172, 3212, 744, 16344, - 144, 3196, 796, 16340, - 116, 3180, 848, 16332, - 88, 3160, 900, 16324, - 60, 3136, 956, 16320, - 36, 3112, 1012, 16312, - 16, 3084, 1068, 16304, - 16380, 3056, 1124, 16296, - 16360, 3024, 1184, 16292, - 16340, 2992, 1244, 16284, - 16324, 2956, 1304, 16276, - 16308, 2920, 1364, 16268, - 16292, 2880, 1424, 16264, - 16280, 2836, 1484, 16256, - 16268, 2792, 1548, 16252, - 16256, 2748, 1608, 16244, - 16248, 2700, 1668, 16240, - 16240, 2652, 1732, 16232, - 16232, 2604, 1792, 16228, - 16228, 2552, 1856, 16224, - 16220, 2500, 1916, 16220, - 16216, 2444, 1980, 16216, - 16216, 2388, 2040, 16216, - 16212, 2332, 2100, 16212, - 16212, 2276, 2160, 16212, - 16212, 2220, 2220, 16212 }; +//========================================= +// <num_taps> = 4 +// <num_phases> = 64 +// <scale_ratio> = 1.16666 (input/output) +// <sharpness> = 0 +// <CoefType> = ModifiedLanczos +// <CoefQuant> = 1.10 +// <CoefOut> = 1.12 +//========================================= +static const uint16_t filter_4tap_64p_116[132] = { + 0x01A8, 0x0CB4, 0x01A4, 0x0000, + 0x017C, 0x0CB8, 0x01D0, 0x3FFC, + 0x0158, 0x0CB8, 0x01F8, 0x3FF8, + 0x0130, 0x0CB4, 0x0228, 0x3FF4, + 0x0110, 0x0CB0, 0x0254, 0x3FEC, + 0x00EC, 0x0CA8, 0x0284, 0x3FE8, + 0x00CC, 0x0C9C, 0x02B4, 0x3FE4, + 0x00AC, 0x0C90, 0x02E8, 0x3FDC, + 0x0090, 0x0C80, 0x031C, 0x3FD4, + 0x0070, 0x0C70, 0x0350, 0x3FD0, + 0x0058, 0x0C5C, 0x0384, 0x3FC8, + 0x003C, 0x0C48, 0x03BC, 0x3FC0, + 0x0024, 0x0C2C, 0x03F4, 0x3FBC, + 0x0010, 0x0C10, 0x042C, 0x3FB4, + 0x3FFC, 0x0BF4, 0x0464, 0x3FAC, + 0x3FE8, 0x0BD4, 0x04A0, 0x3FA4, + 0x3FD8, 0x0BAC, 0x04DC, 0x3FA0, + 0x3FC4, 0x0B8C, 0x0518, 0x3F98, + 0x3FB4, 0x0B68, 0x0554, 0x3F90, + 0x3FA8, 0x0B40, 0x0590, 0x3F88, + 0x3F9C, 0x0B14, 0x05CC, 0x3F84, + 0x3F90, 0x0AEC, 0x0608, 0x3F7C, + 0x3F84, 0x0ABC, 0x0648, 0x3F78, + 0x3F7C, 0x0A90, 0x0684, 0x3F70, + 0x3F70, 0x0A60, 0x06C4, 0x3F6C, + 0x3F6C, 0x0A2C, 0x0700, 0x3F68, + 0x3F64, 0x09F8, 0x0740, 0x3F64, + 0x3F60, 0x09C4, 0x077C, 0x3F60, + 0x3F5C, 0x098C, 0x07BC, 0x3F5C, + 0x3F58, 0x0958, 0x07F8, 0x3F58, + 0x3F58, 0x091C, 0x0834, 0x3F58, + 0x3F54, 0x08E4, 0x0870, 0x3F58, + 0x3F54, 0x08AC, 0x08AC, 0x3F54 +}; -static const uint16_t filter_4tap_64p_150[132] = { - 696, 2700, 696, 0, - 660, 2704, 732, 16380, - 628, 2704, 768, 16376, - 596, 2704, 804, 16372, - 564, 2700, 844, 16364, - 532, 2696, 884, 16360, - 500, 2692, 924, 16356, - 472, 2684, 964, 16352, - 440, 2676, 1004, 16352, - 412, 2668, 1044, 16348, - 384, 2656, 1088, 16344, - 360, 2644, 1128, 16340, - 332, 2632, 1172, 16336, - 308, 2616, 1216, 16336, - 284, 2600, 1260, 16332, - 260, 2580, 1304, 16332, - 236, 2560, 1348, 16328, - 216, 2540, 1392, 16328, - 196, 2516, 1436, 16328, - 176, 2492, 1480, 16324, - 156, 2468, 1524, 16324, - 136, 2440, 1568, 16328, - 120, 2412, 1612, 16328, - 104, 2384, 1656, 16328, - 88, 2352, 1700, 16332, - 72, 2324, 1744, 16332, - 60, 2288, 1788, 16336, - 48, 2256, 1828, 16340, - 36, 2220, 1872, 16344, - 24, 2184, 1912, 16352, - 12, 2148, 1952, 16356, - 4, 2112, 1996, 16364, - 16380, 2072, 2036, 16372 }; +//========================================= +// <num_taps> = 4 +// <num_phases> = 64 +// <scale_ratio> = 1.49999 (input/output) +// <sharpness> = 0 +// <CoefType> = ModifiedLanczos +// <CoefQuant> = 1.10 +// <CoefOut> = 1.12 +//========================================= +static const uint16_t filter_4tap_64p_149[132] = { + 0x02B8, 0x0A90, 0x02B8, 0x0000, + 0x0294, 0x0A94, 0x02DC, 0x3FFC, + 0x0274, 0x0A94, 0x0300, 0x3FF8, + 0x0250, 0x0A94, 0x0328, 0x3FF4, + 0x0230, 0x0A90, 0x0350, 0x3FF0, + 0x0214, 0x0A8C, 0x0374, 0x3FEC, + 0x01F0, 0x0A88, 0x03A0, 0x3FE8, + 0x01D4, 0x0A80, 0x03C8, 0x3FE4, + 0x01B8, 0x0A78, 0x03F0, 0x3FE0, + 0x0198, 0x0A70, 0x041C, 0x3FDC, + 0x0180, 0x0A64, 0x0444, 0x3FD8, + 0x0164, 0x0A54, 0x0470, 0x3FD8, + 0x0148, 0x0A48, 0x049C, 0x3FD4, + 0x0130, 0x0A38, 0x04C8, 0x3FD0, + 0x0118, 0x0A24, 0x04F4, 0x3FD0, + 0x0100, 0x0A14, 0x0520, 0x3FCC, + 0x00E8, 0x0A00, 0x054C, 0x3FCC, + 0x00D4, 0x09E8, 0x057C, 0x3FC8, + 0x00C0, 0x09D0, 0x05A8, 0x3FC8, + 0x00AC, 0x09B8, 0x05D4, 0x3FC8, + 0x0098, 0x09A0, 0x0600, 0x3FC8, + 0x0084, 0x0984, 0x0630, 0x3FC8, + 0x0074, 0x0964, 0x065C, 0x3FCC, + 0x0064, 0x0948, 0x0688, 0x3FCC, + 0x0054, 0x0928, 0x06B4, 0x3FD0, + 0x0044, 0x0908, 0x06E0, 0x3FD4, + 0x0038, 0x08E8, 0x070C, 0x3FD4, + 0x002C, 0x08C4, 0x0738, 0x3FD8, + 0x001C, 0x08A4, 0x0760, 0x3FE0, + 0x0014, 0x087C, 0x078C, 0x3FE4, + 0x0008, 0x0858, 0x07B4, 0x3FEC, + 0x0000, 0x0830, 0x07DC, 0x3FF4, + 0x3FFC, 0x0804, 0x0804, 0x3FFC +}; +//========================================= +// <num_taps> = 4 +// <num_phases> = 64 +// <scale_ratio> = 1.83332 (input/output) +// <sharpness> = 0 +// <CoefType> = ModifiedLanczos +// <CoefQuant> = 1.10 +// <CoefOut> = 1.12 +//========================================= static const uint16_t filter_4tap_64p_183[132] = { - 944, 2204, 944, 0, - 916, 2204, 972, 0, - 888, 2200, 996, 0, - 860, 2200, 1024, 4, - 832, 2196, 1052, 4, - 808, 2192, 1080, 8, - 780, 2188, 1108, 12, - 756, 2180, 1140, 12, - 728, 2176, 1168, 16, - 704, 2168, 1196, 20, - 680, 2160, 1224, 24, - 656, 2152, 1252, 28, - 632, 2144, 1280, 36, - 608, 2132, 1308, 40, - 584, 2120, 1336, 48, - 560, 2112, 1364, 52, - 536, 2096, 1392, 60, - 516, 2084, 1420, 68, - 492, 2072, 1448, 76, - 472, 2056, 1476, 84, - 452, 2040, 1504, 92, - 428, 2024, 1532, 100, - 408, 2008, 1560, 112, - 392, 1992, 1584, 120, - 372, 1972, 1612, 132, - 352, 1956, 1636, 144, - 336, 1936, 1664, 156, - 316, 1916, 1688, 168, - 300, 1896, 1712, 180, - 284, 1876, 1736, 192, - 268, 1852, 1760, 208, - 252, 1832, 1784, 220, - 236, 1808, 1808, 236 }; + 0x03B0, 0x08A0, 0x03B0, 0x0000, + 0x0394, 0x08A0, 0x03CC, 0x0000, + 0x037C, 0x089C, 0x03E8, 0x0000, + 0x0360, 0x089C, 0x0400, 0x0004, + 0x0348, 0x0898, 0x041C, 0x0004, + 0x032C, 0x0894, 0x0438, 0x0008, + 0x0310, 0x0890, 0x0454, 0x000C, + 0x02F8, 0x0888, 0x0474, 0x000C, + 0x02DC, 0x0884, 0x0490, 0x0010, + 0x02C4, 0x087C, 0x04AC, 0x0014, + 0x02AC, 0x0874, 0x04C8, 0x0018, + 0x0290, 0x086C, 0x04E4, 0x0020, + 0x0278, 0x0864, 0x0500, 0x0024, + 0x0264, 0x0858, 0x051C, 0x0028, + 0x024C, 0x084C, 0x0538, 0x0030, + 0x0234, 0x0844, 0x0554, 0x0034, + 0x021C, 0x0838, 0x0570, 0x003C, + 0x0208, 0x0828, 0x058C, 0x0044, + 0x01F0, 0x081C, 0x05A8, 0x004C, + 0x01DC, 0x080C, 0x05C4, 0x0054, + 0x01C8, 0x07FC, 0x05E0, 0x005C, + 0x01B4, 0x07EC, 0x05FC, 0x0064, + 0x019C, 0x07DC, 0x0618, 0x0070, + 0x018C, 0x07CC, 0x0630, 0x0078, + 0x0178, 0x07B8, 0x064C, 0x0084, + 0x0164, 0x07A8, 0x0664, 0x0090, + 0x0150, 0x0794, 0x0680, 0x009C, + 0x0140, 0x0780, 0x0698, 0x00A8, + 0x0130, 0x076C, 0x06B0, 0x00B4, + 0x0120, 0x0758, 0x06C8, 0x00C0, + 0x0110, 0x0740, 0x06E0, 0x00D0, + 0x0100, 0x072C, 0x06F8, 0x00DC, + 0x00F0, 0x0714, 0x0710, 0x00EC +}; +//========================================= +// <num_taps> = 5 +// <num_phases> = 64 +// <scale_ratio> = 0.83333 (input/output) +// <sharpness> = 0 +// <CoefType> = ModifiedLanczos +// <CoefQuant> = 1.10 +// <CoefOut> = 1.12 +//========================================= static const uint16_t filter_5tap_64p_upscale[165] = { - 15936, 2496, 2496, 15936, 0, - 15948, 2404, 2580, 15924, 0, - 15960, 2312, 2664, 15912, 4, - 15976, 2220, 2748, 15904, 8, - 15992, 2128, 2832, 15896, 12, - 16004, 2036, 2912, 15888, 16, - 16020, 1944, 2992, 15880, 20, - 16036, 1852, 3068, 15876, 20, - 16056, 1760, 3140, 15876, 24, - 16072, 1668, 3216, 15872, 28, - 16088, 1580, 3284, 15872, 32, - 16104, 1492, 3352, 15876, 32, - 16120, 1404, 3420, 15876, 36, - 16140, 1316, 3480, 15884, 40, - 16156, 1228, 3540, 15892, 40, - 16172, 1144, 3600, 15900, 40, - 16188, 1060, 3652, 15908, 44, - 16204, 980, 3704, 15924, 44, - 16220, 900, 3756, 15936, 44, - 16236, 824, 3800, 15956, 44, - 16248, 744, 3844, 15972, 44, - 16264, 672, 3884, 15996, 44, - 16276, 600, 3920, 16020, 44, - 16292, 528, 3952, 16044, 40, - 16304, 460, 3980, 16072, 40, - 16316, 396, 4008, 16104, 36, - 16328, 332, 4032, 16136, 32, - 16336, 272, 4048, 16172, 28, - 16348, 212, 4064, 16208, 24, - 16356, 156, 4080, 16248, 16, - 16368, 100, 4088, 16292, 12, - 16376, 48, 4092, 16336, 4, - 0, 0, 4096, 0, 0 }; + 0x3E40, 0x09C0, 0x09C0, 0x3E40, 0x0000, + 0x3E50, 0x0964, 0x0A18, 0x3E34, 0x0000, + 0x3E5C, 0x0908, 0x0A6C, 0x3E2C, 0x0004, + 0x3E6C, 0x08AC, 0x0AC0, 0x3E20, 0x0008, + 0x3E78, 0x0850, 0x0B14, 0x3E18, 0x000C, + 0x3E88, 0x07F4, 0x0B60, 0x3E14, 0x0010, + 0x3E98, 0x0798, 0x0BB0, 0x3E0C, 0x0014, + 0x3EA8, 0x073C, 0x0C00, 0x3E08, 0x0014, + 0x3EB8, 0x06E4, 0x0C48, 0x3E04, 0x0018, + 0x3ECC, 0x0684, 0x0C90, 0x3E04, 0x001C, + 0x3EDC, 0x062C, 0x0CD4, 0x3E04, 0x0020, + 0x3EEC, 0x05D4, 0x0D1C, 0x3E04, 0x0020, + 0x3EFC, 0x057C, 0x0D5C, 0x3E08, 0x0024, + 0x3F0C, 0x0524, 0x0D98, 0x3E10, 0x0028, + 0x3F20, 0x04CC, 0x0DD8, 0x3E14, 0x0028, + 0x3F30, 0x0478, 0x0E14, 0x3E1C, 0x0028, + 0x3F40, 0x0424, 0x0E48, 0x3E28, 0x002C, + 0x3F50, 0x03D4, 0x0E7C, 0x3E34, 0x002C, + 0x3F60, 0x0384, 0x0EAC, 0x3E44, 0x002C, + 0x3F6C, 0x0338, 0x0EDC, 0x3E54, 0x002C, + 0x3F7C, 0x02E8, 0x0F08, 0x3E68, 0x002C, + 0x3F8C, 0x02A0, 0x0F2C, 0x3E7C, 0x002C, + 0x3F98, 0x0258, 0x0F50, 0x3E94, 0x002C, + 0x3FA4, 0x0210, 0x0F74, 0x3EB0, 0x0028, + 0x3FB0, 0x01CC, 0x0F90, 0x3ECC, 0x0028, + 0x3FC0, 0x018C, 0x0FA8, 0x3EE8, 0x0024, + 0x3FC8, 0x014C, 0x0FC0, 0x3F0C, 0x0020, + 0x3FD4, 0x0110, 0x0FD4, 0x3F2C, 0x001C, + 0x3FE0, 0x00D4, 0x0FE0, 0x3F54, 0x0018, + 0x3FE8, 0x009C, 0x0FF0, 0x3F7C, 0x0010, + 0x3FF0, 0x0064, 0x0FFC, 0x3FA4, 0x000C, + 0x3FFC, 0x0030, 0x0FFC, 0x3FD4, 0x0004, + 0x0000, 0x0000, 0x1000, 0x0000, 0x0000 +}; -static const uint16_t filter_5tap_64p_117[165] = { - 16056, 2372, 2372, 16056, 0, - 16052, 2312, 2432, 16060, 0, - 16052, 2252, 2488, 16064, 0, - 16052, 2188, 2548, 16072, 0, - 16052, 2124, 2600, 16076, 0, - 16052, 2064, 2656, 16088, 0, - 16052, 2000, 2708, 16096, 0, - 16056, 1932, 2760, 16108, 0, - 16060, 1868, 2808, 16120, 0, - 16064, 1804, 2856, 16132, 0, - 16068, 1740, 2904, 16148, 16380, - 16076, 1676, 2948, 16164, 16380, - 16080, 1612, 2992, 16180, 16376, - 16088, 1544, 3032, 16200, 16372, - 16096, 1480, 3072, 16220, 16372, - 16104, 1420, 3108, 16244, 16368, - 16112, 1356, 3144, 16268, 16364, - 16120, 1292, 3180, 16292, 16360, - 16128, 1232, 3212, 16320, 16356, - 16136, 1168, 3240, 16344, 16352, - 16144, 1108, 3268, 16376, 16344, - 16156, 1048, 3292, 20, 16340, - 16164, 988, 3316, 52, 16332, - 16172, 932, 3336, 88, 16328, - 16184, 872, 3356, 124, 16320, - 16192, 816, 3372, 160, 16316, - 16204, 760, 3388, 196, 16308, - 16212, 708, 3400, 236, 16300, - 16220, 656, 3412, 276, 16292, - 16232, 604, 3420, 320, 16284, - 16240, 552, 3424, 364, 16276, - 16248, 504, 3428, 408, 16268, - 16256, 456, 3428, 456, 16256 }; +//========================================= +// <num_taps> = 5 +// <num_phases> = 64 +// <scale_ratio> = 1.16666 (input/output) +// <sharpness> = 0 +// <CoefType> = ModifiedLanczos +// <CoefQuant> = 1.10 +// <CoefOut> = 1.12 +//========================================= +static const uint16_t filter_5tap_64p_116[165] = { + 0x3EDC, 0x0924, 0x0924, 0x3EDC, 0x0000, + 0x3ED8, 0x08EC, 0x095C, 0x3EE0, 0x0000, + 0x3ED4, 0x08B0, 0x0994, 0x3EE8, 0x0000, + 0x3ED0, 0x0878, 0x09C8, 0x3EF0, 0x0000, + 0x3ED0, 0x083C, 0x09FC, 0x3EF8, 0x0000, + 0x3ED0, 0x0800, 0x0A2C, 0x3F04, 0x0000, + 0x3ED0, 0x07C4, 0x0A5C, 0x3F10, 0x0000, + 0x3ED0, 0x0788, 0x0A8C, 0x3F1C, 0x0000, + 0x3ED0, 0x074C, 0x0AC0, 0x3F28, 0x3FFC, + 0x3ED4, 0x0710, 0x0AE8, 0x3F38, 0x3FFC, + 0x3ED8, 0x06D0, 0x0B18, 0x3F48, 0x3FF8, + 0x3EDC, 0x0694, 0x0B3C, 0x3F5C, 0x3FF8, + 0x3EE0, 0x0658, 0x0B68, 0x3F6C, 0x3FF4, + 0x3EE4, 0x061C, 0x0B90, 0x3F80, 0x3FF0, + 0x3EEC, 0x05DC, 0x0BB4, 0x3F98, 0x3FEC, + 0x3EF0, 0x05A0, 0x0BD8, 0x3FB0, 0x3FE8, + 0x3EF8, 0x0564, 0x0BF8, 0x3FC8, 0x3FE4, + 0x3EFC, 0x0528, 0x0C1C, 0x3FE0, 0x3FE0, + 0x3F04, 0x04EC, 0x0C38, 0x3FFC, 0x3FDC, + 0x3F0C, 0x04B4, 0x0C54, 0x0014, 0x3FD8, + 0x3F14, 0x047C, 0x0C70, 0x0030, 0x3FD0, + 0x3F1C, 0x0440, 0x0C88, 0x0050, 0x3FCC, + 0x3F24, 0x0408, 0x0CA0, 0x0070, 0x3FC4, + 0x3F2C, 0x03D0, 0x0CB0, 0x0094, 0x3FC0, + 0x3F34, 0x0398, 0x0CC4, 0x00B8, 0x3FB8, + 0x3F3C, 0x0364, 0x0CD4, 0x00DC, 0x3FB0, + 0x3F48, 0x032C, 0x0CE0, 0x0100, 0x3FAC, + 0x3F50, 0x02F8, 0x0CEC, 0x0128, 0x3FA4, + 0x3F58, 0x02C4, 0x0CF8, 0x0150, 0x3F9C, + 0x3F60, 0x0290, 0x0D00, 0x017C, 0x3F94, + 0x3F68, 0x0260, 0x0D04, 0x01A8, 0x3F8C, + 0x3F74, 0x0230, 0x0D04, 0x01D4, 0x3F84, + 0x3F7C, 0x0200, 0x0D08, 0x0200, 0x3F7C +}; -static const uint16_t filter_5tap_64p_150[165] = { - 16368, 2064, 2064, 16368, 0, - 16352, 2028, 2100, 16380, 16380, - 16340, 1996, 2132, 12, 16376, - 16328, 1960, 2168, 24, 16376, - 16316, 1924, 2204, 44, 16372, - 16308, 1888, 2236, 60, 16368, - 16296, 1848, 2268, 76, 16364, - 16288, 1812, 2300, 96, 16360, - 16280, 1772, 2328, 116, 16356, - 16272, 1736, 2360, 136, 16352, - 16268, 1696, 2388, 160, 16348, - 16260, 1656, 2416, 180, 16344, - 16256, 1616, 2440, 204, 16340, - 16248, 1576, 2464, 228, 16336, - 16244, 1536, 2492, 252, 16332, - 16240, 1496, 2512, 276, 16324, - 16240, 1456, 2536, 304, 16320, - 16236, 1416, 2556, 332, 16316, - 16232, 1376, 2576, 360, 16312, - 16232, 1336, 2592, 388, 16308, - 16232, 1296, 2612, 416, 16300, - 16232, 1256, 2628, 448, 16296, - 16232, 1216, 2640, 480, 16292, - 16232, 1172, 2652, 512, 16288, - 16232, 1132, 2664, 544, 16284, - 16232, 1092, 2676, 576, 16280, - 16236, 1056, 2684, 608, 16272, - 16236, 1016, 2692, 644, 16268, - 16240, 976, 2700, 680, 16264, - 16240, 936, 2704, 712, 16260, - 16244, 900, 2708, 748, 16256, - 16248, 860, 2708, 788, 16252, - 16248, 824, 2708, 824, 16248 }; +//========================================= +// <num_taps> = 5 +// <num_phases> = 64 +// <scale_ratio> = 1.49999 (input/output) +// <sharpness> = 0 +// <CoefType> = ModifiedLanczos +// <CoefQuant> = 1.10 +// <CoefOut> = 1.12 +//========================================= +static const uint16_t filter_5tap_64p_149[165] = { + 0x3FF4, 0x080C, 0x080C, 0x3FF4, 0x0000, + 0x3FE8, 0x07E8, 0x0830, 0x0000, 0x0000, + 0x3FDC, 0x07C8, 0x0850, 0x0010, 0x3FFC, + 0x3FD0, 0x07A4, 0x0878, 0x001C, 0x3FF8, + 0x3FC4, 0x0780, 0x0898, 0x0030, 0x3FF4, + 0x3FB8, 0x075C, 0x08B8, 0x0040, 0x3FF4, + 0x3FB0, 0x0738, 0x08D8, 0x0050, 0x3FF0, + 0x3FA8, 0x0710, 0x08F8, 0x0064, 0x3FEC, + 0x3FA0, 0x06EC, 0x0914, 0x0078, 0x3FE8, + 0x3F98, 0x06C4, 0x0934, 0x008C, 0x3FE4, + 0x3F90, 0x06A0, 0x094C, 0x00A4, 0x3FE0, + 0x3F8C, 0x0678, 0x0968, 0x00B8, 0x3FDC, + 0x3F84, 0x0650, 0x0984, 0x00D0, 0x3FD8, + 0x3F80, 0x0628, 0x099C, 0x00E8, 0x3FD4, + 0x3F7C, 0x0600, 0x09B8, 0x0100, 0x3FCC, + 0x3F78, 0x05D8, 0x09D0, 0x0118, 0x3FC8, + 0x3F74, 0x05B0, 0x09E4, 0x0134, 0x3FC4, + 0x3F70, 0x0588, 0x09F8, 0x0150, 0x3FC0, + 0x3F70, 0x0560, 0x0A08, 0x016C, 0x3FBC, + 0x3F6C, 0x0538, 0x0A20, 0x0188, 0x3FB4, + 0x3F6C, 0x0510, 0x0A30, 0x01A4, 0x3FB0, + 0x3F6C, 0x04E8, 0x0A3C, 0x01C4, 0x3FAC, + 0x3F6C, 0x04C0, 0x0A48, 0x01E4, 0x3FA8, + 0x3F6C, 0x0498, 0x0A58, 0x0200, 0x3FA4, + 0x3F6C, 0x0470, 0x0A60, 0x0224, 0x3FA0, + 0x3F6C, 0x0448, 0x0A70, 0x0244, 0x3F98, + 0x3F70, 0x0420, 0x0A78, 0x0264, 0x3F94, + 0x3F70, 0x03F8, 0x0A80, 0x0288, 0x3F90, + 0x3F74, 0x03D4, 0x0A84, 0x02A8, 0x3F8C, + 0x3F74, 0x03AC, 0x0A8C, 0x02CC, 0x3F88, + 0x3F78, 0x0384, 0x0A90, 0x02F0, 0x3F84, + 0x3F7C, 0x0360, 0x0A90, 0x0314, 0x3F80, + 0x3F7C, 0x033C, 0x0A90, 0x033C, 0x3F7C +}; +//========================================= +// <num_taps> = 5 +// <num_phases> = 64 +// <scale_ratio> = 1.83332 (input/output) +// <sharpness> = 0 +// <CoefType> = ModifiedLanczos +// <CoefQuant> = 1.10 +// <CoefOut> = 1.12 +//========================================= static const uint16_t filter_5tap_64p_183[165] = { - 228, 1816, 1816, 228, 0, - 216, 1792, 1836, 248, 16380, - 200, 1772, 1860, 264, 16376, - 184, 1748, 1884, 280, 16376, - 168, 1728, 1904, 300, 16372, - 156, 1704, 1928, 316, 16368, - 144, 1680, 1948, 336, 16364, - 128, 1656, 1968, 356, 16364, - 116, 1632, 1988, 376, 16360, - 104, 1604, 2008, 396, 16356, - 96, 1580, 2024, 416, 16356, - 84, 1556, 2044, 440, 16352, - 72, 1528, 2060, 460, 16348, - 64, 1504, 2076, 484, 16348, - 52, 1476, 2092, 504, 16344, - 44, 1448, 2104, 528, 16344, - 36, 1424, 2120, 552, 16340, - 28, 1396, 2132, 576, 16340, - 20, 1368, 2144, 600, 16340, - 12, 1340, 2156, 624, 16336, - 4, 1312, 2168, 652, 16336, - 0, 1284, 2180, 676, 16336, - 16376, 1256, 2188, 700, 16332, - 16372, 1228, 2196, 728, 16332, - 16368, 1200, 2204, 752, 16332, - 16364, 1172, 2212, 780, 16332, - 16356, 1144, 2216, 808, 16332, - 16352, 1116, 2220, 836, 16332, - 16352, 1084, 2224, 860, 16332, - 16348, 1056, 2228, 888, 16336, - 16344, 1028, 2232, 916, 16336, - 16340, 1000, 2232, 944, 16336, - 16340, 972, 2232, 972, 16340 }; + 0x0168, 0x069C, 0x0698, 0x0164, 0x0000, + 0x0154, 0x068C, 0x06AC, 0x0174, 0x0000, + 0x0144, 0x0674, 0x06C0, 0x0188, 0x0000, + 0x0138, 0x0664, 0x06D0, 0x0198, 0x3FFC, + 0x0128, 0x0654, 0x06E0, 0x01A8, 0x3FFC, + 0x0118, 0x0640, 0x06F0, 0x01BC, 0x3FFC, + 0x010C, 0x0630, 0x0700, 0x01CC, 0x3FF8, + 0x00FC, 0x061C, 0x0710, 0x01E0, 0x3FF8, + 0x00F0, 0x060C, 0x071C, 0x01F0, 0x3FF8, + 0x00E4, 0x05F4, 0x072C, 0x0204, 0x3FF8, + 0x00D8, 0x05E4, 0x0738, 0x0218, 0x3FF4, + 0x00CC, 0x05D0, 0x0744, 0x022C, 0x3FF4, + 0x00C0, 0x05B8, 0x0754, 0x0240, 0x3FF4, + 0x00B4, 0x05A4, 0x0760, 0x0254, 0x3FF4, + 0x00A8, 0x0590, 0x076C, 0x0268, 0x3FF4, + 0x009C, 0x057C, 0x0778, 0x027C, 0x3FF4, + 0x0094, 0x0564, 0x0780, 0x0294, 0x3FF4, + 0x0088, 0x0550, 0x0788, 0x02A8, 0x3FF8, + 0x0080, 0x0538, 0x0794, 0x02BC, 0x3FF8, + 0x0074, 0x0524, 0x079C, 0x02D4, 0x3FF8, + 0x006C, 0x0510, 0x07A4, 0x02E8, 0x3FF8, + 0x0064, 0x04F4, 0x07AC, 0x0300, 0x3FFC, + 0x005C, 0x04E4, 0x07B0, 0x0314, 0x3FFC, + 0x0054, 0x04C8, 0x07B8, 0x032C, 0x0000, + 0x004C, 0x04B4, 0x07C0, 0x0340, 0x0000, + 0x0044, 0x04A0, 0x07C4, 0x0358, 0x0000, + 0x003C, 0x0488, 0x07C8, 0x0370, 0x0004, + 0x0038, 0x0470, 0x07CC, 0x0384, 0x0008, + 0x0030, 0x045C, 0x07D0, 0x039C, 0x0008, + 0x002C, 0x0444, 0x07D0, 0x03B4, 0x000C, + 0x0024, 0x042C, 0x07D4, 0x03CC, 0x0010, + 0x0020, 0x0414, 0x07D4, 0x03E0, 0x0018, + 0x001C, 0x03FC, 0x07D4, 0x03F8, 0x001C +}; +//========================================= +// <num_taps> = 6 +// <num_phases> = 64 +// <scale_ratio> = 0.83333 (input/output) +// <sharpness> = 0 +// <CoefType> = ModifiedLanczos +// <CoefQuant> = 1.10 +// <CoefOut> = 1.12 +//========================================= static const uint16_t filter_6tap_64p_upscale[198] = { - 0, 0, 4092, 0, 0, 0, - 12, 16332, 4092, 52, 16368, 0, - 24, 16280, 4088, 108, 16356, 0, - 36, 16236, 4080, 168, 16340, 0, - 44, 16188, 4064, 228, 16324, 0, - 56, 16148, 4052, 292, 16308, 0, - 64, 16108, 4032, 356, 16292, 4, - 72, 16072, 4008, 424, 16276, 4, - 80, 16036, 3980, 492, 16256, 4, - 88, 16004, 3952, 564, 16240, 8, - 96, 15972, 3920, 636, 16220, 8, - 100, 15944, 3884, 712, 16204, 12, - 108, 15916, 3844, 788, 16184, 16, - 112, 15896, 3800, 864, 16164, 20, - 116, 15872, 3756, 944, 16144, 20, - 120, 15852, 3708, 1024, 16124, 24, - 120, 15836, 3656, 1108, 16104, 28, - 124, 15824, 3600, 1192, 16084, 32, - 124, 15808, 3544, 1276, 16064, 36, - 124, 15800, 3484, 1360, 16044, 40, - 128, 15792, 3420, 1448, 16024, 44, - 128, 15784, 3352, 1536, 16004, 48, - 124, 15780, 3288, 1624, 15988, 52, - 124, 15776, 3216, 1712, 15968, 56, - 124, 15776, 3144, 1800, 15948, 64, - 120, 15776, 3068, 1888, 15932, 68, - 120, 15780, 2992, 1976, 15912, 72, - 116, 15784, 2916, 2064, 15896, 76, - 112, 15792, 2836, 2152, 15880, 80, - 108, 15796, 2752, 2244, 15868, 84, - 104, 15804, 2672, 2328, 15852, 88, - 104, 15816, 2588, 2416, 15840, 92, - 100, 15828, 2504, 2504, 15828, 100 }; + 0x0000, 0x0000, 0x1000, 0x0000, 0x0000, 0x0000, + 0x000C, 0x3FD0, 0x0FFC, 0x0034, 0x3FF4, 0x0000, + 0x0018, 0x3F9C, 0x0FF8, 0x006C, 0x3FE8, 0x0000, + 0x0024, 0x3F6C, 0x0FF0, 0x00A8, 0x3FD8, 0x0000, + 0x002C, 0x3F44, 0x0FE4, 0x00E4, 0x3FC8, 0x0000, + 0x0038, 0x3F18, 0x0FD4, 0x0124, 0x3FB8, 0x0000, + 0x0040, 0x3EF0, 0x0FC0, 0x0164, 0x3FA8, 0x0004, + 0x0048, 0x3EC8, 0x0FAC, 0x01A8, 0x3F98, 0x0004, + 0x0050, 0x3EA8, 0x0F94, 0x01EC, 0x3F84, 0x0004, + 0x0058, 0x3E84, 0x0F74, 0x0234, 0x3F74, 0x0008, + 0x0060, 0x3E68, 0x0F54, 0x027C, 0x3F60, 0x0008, + 0x0064, 0x3E4C, 0x0F30, 0x02C8, 0x3F4C, 0x000C, + 0x006C, 0x3E30, 0x0F04, 0x0314, 0x3F3C, 0x0010, + 0x0070, 0x3E18, 0x0EDC, 0x0360, 0x3F28, 0x0014, + 0x0074, 0x3E04, 0x0EB0, 0x03B0, 0x3F14, 0x0014, + 0x0078, 0x3DF0, 0x0E80, 0x0400, 0x3F00, 0x0018, + 0x0078, 0x3DE0, 0x0E4C, 0x0454, 0x3EEC, 0x001C, + 0x007C, 0x3DD0, 0x0E14, 0x04A8, 0x3ED8, 0x0020, + 0x007C, 0x3DC4, 0x0DDC, 0x04FC, 0x3EC4, 0x0024, + 0x007C, 0x3DBC, 0x0DA0, 0x0550, 0x3EB0, 0x0028, + 0x0080, 0x3DB4, 0x0D5C, 0x05A8, 0x3E9C, 0x002C, + 0x0080, 0x3DAC, 0x0D1C, 0x0600, 0x3E88, 0x0030, + 0x007C, 0x3DA8, 0x0CDC, 0x0658, 0x3E74, 0x0034, + 0x007C, 0x3DA4, 0x0C94, 0x06B0, 0x3E64, 0x0038, + 0x007C, 0x3DA4, 0x0C48, 0x0708, 0x3E50, 0x0040, + 0x0078, 0x3DA4, 0x0C00, 0x0760, 0x3E40, 0x0044, + 0x0078, 0x3DA8, 0x0BB4, 0x07B8, 0x3E2C, 0x0048, + 0x0074, 0x3DAC, 0x0B68, 0x0810, 0x3E1C, 0x004C, + 0x0070, 0x3DB4, 0x0B18, 0x0868, 0x3E0C, 0x0050, + 0x006C, 0x3DBC, 0x0AC4, 0x08C4, 0x3DFC, 0x0054, + 0x0068, 0x3DC4, 0x0A74, 0x0918, 0x3DF0, 0x0058, + 0x0068, 0x3DCC, 0x0A20, 0x0970, 0x3DE0, 0x005C, + 0x0064, 0x3DD4, 0x09C8, 0x09C8, 0x3DD4, 0x0064 +}; -static const uint16_t filter_6tap_64p_117[198] = { - 16168, 476, 3568, 476, 16168, 0, - 16180, 428, 3564, 528, 16156, 0, - 16192, 376, 3556, 584, 16144, 4, - 16204, 328, 3548, 636, 16128, 4, - 16216, 280, 3540, 692, 16116, 8, - 16228, 232, 3524, 748, 16104, 12, - 16240, 188, 3512, 808, 16092, 12, - 16252, 148, 3492, 864, 16080, 16, - 16264, 104, 3472, 924, 16068, 16, - 16276, 64, 3452, 984, 16056, 20, - 16284, 28, 3428, 1044, 16048, 24, - 16296, 16376, 3400, 1108, 16036, 24, - 16304, 16340, 3372, 1168, 16024, 28, - 16316, 16304, 3340, 1232, 16016, 32, - 16324, 16272, 3308, 1296, 16004, 32, - 16332, 16244, 3272, 1360, 15996, 36, - 16344, 16212, 3236, 1424, 15988, 36, - 16352, 16188, 3200, 1488, 15980, 40, - 16360, 16160, 3160, 1552, 15972, 40, - 16368, 16136, 3116, 1616, 15964, 40, - 16372, 16112, 3072, 1680, 15956, 44, - 16380, 16092, 3028, 1744, 15952, 44, - 0, 16072, 2980, 1808, 15948, 44, - 8, 16052, 2932, 1872, 15944, 48, - 12, 16036, 2880, 1936, 15940, 48, - 16, 16020, 2828, 2000, 15936, 48, - 20, 16008, 2776, 2064, 15936, 48, - 24, 15996, 2724, 2128, 15936, 48, - 28, 15984, 2668, 2192, 15936, 48, - 32, 15972, 2612, 2252, 15940, 44, - 36, 15964, 2552, 2316, 15940, 44, - 40, 15956, 2496, 2376, 15944, 44, - 40, 15952, 2436, 2436, 15952, 40 }; +//========================================= +// <num_taps> = 6 +// <num_phases> = 64 +// <scale_ratio> = 1.16666 (input/output) +// <sharpness> = 0 +// <CoefType> = ModifiedLanczos +// <CoefQuant> = 1.10 +// <CoefOut> = 1.12 +//========================================= +static const uint16_t filter_6tap_64p_116[198] = { + 0x3F0C, 0x0240, 0x0D68, 0x0240, 0x3F0C, 0x0000, + 0x3F18, 0x0210, 0x0D64, 0x0274, 0x3F00, 0x0000, + 0x3F24, 0x01E0, 0x0D58, 0x02A8, 0x3EF8, 0x0004, + 0x3F2C, 0x01B0, 0x0D58, 0x02DC, 0x3EEC, 0x0004, + 0x3F38, 0x0180, 0x0D50, 0x0310, 0x3EE0, 0x0008, + 0x3F44, 0x0154, 0x0D40, 0x0348, 0x3ED8, 0x0008, + 0x3F50, 0x0128, 0x0D34, 0x037C, 0x3ECC, 0x000C, + 0x3F5C, 0x00FC, 0x0D20, 0x03B4, 0x3EC4, 0x0010, + 0x3F64, 0x00D4, 0x0D14, 0x03EC, 0x3EB8, 0x0010, + 0x3F70, 0x00AC, 0x0CFC, 0x0424, 0x3EB0, 0x0014, + 0x3F78, 0x0084, 0x0CE8, 0x0460, 0x3EA8, 0x0014, + 0x3F84, 0x0060, 0x0CCC, 0x0498, 0x3EA0, 0x0018, + 0x3F90, 0x003C, 0x0CB4, 0x04D0, 0x3E98, 0x0018, + 0x3F98, 0x0018, 0x0C9C, 0x050C, 0x3E90, 0x0018, + 0x3FA0, 0x3FFC, 0x0C78, 0x0548, 0x3E88, 0x001C, + 0x3FAC, 0x3FDC, 0x0C54, 0x0584, 0x3E84, 0x001C, + 0x3FB4, 0x3FBC, 0x0C3C, 0x05BC, 0x3E7C, 0x001C, + 0x3FBC, 0x3FA0, 0x0C14, 0x05F8, 0x3E78, 0x0020, + 0x3FC4, 0x3F84, 0x0BF0, 0x0634, 0x3E74, 0x0020, + 0x3FCC, 0x3F68, 0x0BCC, 0x0670, 0x3E70, 0x0020, + 0x3FD4, 0x3F50, 0x0BA4, 0x06AC, 0x3E6C, 0x0020, + 0x3FDC, 0x3F38, 0x0B78, 0x06E8, 0x3E6C, 0x0020, + 0x3FE0, 0x3F24, 0x0B50, 0x0724, 0x3E68, 0x0020, + 0x3FE8, 0x3F0C, 0x0B24, 0x0760, 0x3E68, 0x0020, + 0x3FF0, 0x3EFC, 0x0AF4, 0x0798, 0x3E68, 0x0020, + 0x3FF4, 0x3EE8, 0x0AC8, 0x07D4, 0x3E68, 0x0020, + 0x3FFC, 0x3ED8, 0x0A94, 0x0810, 0x3E6C, 0x001C, + 0x0000, 0x3EC8, 0x0A64, 0x0848, 0x3E70, 0x001C, + 0x0000, 0x3EB8, 0x0A38, 0x0880, 0x3E74, 0x001C, + 0x0004, 0x3EAC, 0x0A04, 0x08BC, 0x3E78, 0x0018, + 0x0008, 0x3EA4, 0x09D0, 0x08F4, 0x3E7C, 0x0014, + 0x000C, 0x3E98, 0x0998, 0x092C, 0x3E84, 0x0014, + 0x0010, 0x3E90, 0x0964, 0x0960, 0x3E8C, 0x0010 +}; -static const uint16_t filter_6tap_64p_150[198] = { - 16148, 920, 2724, 920, 16148, 0, - 16152, 880, 2724, 956, 16148, 0, - 16152, 844, 2720, 996, 16144, 0, - 16156, 804, 2716, 1032, 16144, 0, - 16156, 768, 2712, 1072, 16144, 0, - 16160, 732, 2708, 1112, 16144, 16380, - 16164, 696, 2700, 1152, 16144, 16380, - 16168, 660, 2692, 1192, 16148, 16380, - 16172, 628, 2684, 1232, 16148, 16380, - 16176, 592, 2672, 1272, 16152, 16376, - 16180, 560, 2660, 1312, 16152, 16376, - 16184, 524, 2648, 1348, 16156, 16376, - 16192, 492, 2632, 1388, 16160, 16372, - 16196, 460, 2616, 1428, 16164, 16372, - 16200, 432, 2600, 1468, 16168, 16368, - 16204, 400, 2584, 1508, 16176, 16364, - 16212, 368, 2564, 1548, 16180, 16364, - 16216, 340, 2544, 1588, 16188, 16360, - 16220, 312, 2524, 1628, 16196, 16356, - 16228, 284, 2504, 1668, 16204, 16356, - 16232, 256, 2480, 1704, 16212, 16352, - 16240, 232, 2456, 1744, 16224, 16348, - 16244, 204, 2432, 1780, 16232, 16344, - 16248, 180, 2408, 1820, 16244, 16340, - 16256, 156, 2380, 1856, 16256, 16336, - 16260, 132, 2352, 1896, 16268, 16332, - 16268, 108, 2324, 1932, 16280, 16328, - 16272, 88, 2296, 1968, 16292, 16324, - 16276, 64, 2268, 2004, 16308, 16320, - 16284, 44, 2236, 2036, 16324, 16312, - 16288, 24, 2204, 2072, 16340, 16308, - 16292, 8, 2172, 2108, 16356, 16304, - 16300, 16372, 2140, 2140, 16372, 16300 }; +//========================================= +// <num_taps> = 6 +// <num_phases> = 64 +// <scale_ratio> = 1.49999 (input/output) +// <sharpness> = 0 +// <CoefType> = ModifiedLanczos +// <CoefQuant> = 1.10 +// <CoefOut> = 1.12 +//========================================= +static const uint16_t filter_6tap_64p_149[198] = { + 0x3F14, 0x0394, 0x0AB0, 0x0394, 0x3F14, 0x0000, + 0x3F18, 0x036C, 0x0AB0, 0x03B8, 0x3F14, 0x0000, + 0x3F18, 0x0348, 0x0AAC, 0x03E0, 0x3F14, 0x0000, + 0x3F1C, 0x0320, 0x0AAC, 0x0408, 0x3F10, 0x0000, + 0x3F20, 0x02FC, 0x0AA8, 0x042C, 0x3F10, 0x0000, + 0x3F24, 0x02D8, 0x0AA0, 0x0454, 0x3F10, 0x0000, + 0x3F28, 0x02B4, 0x0A98, 0x047C, 0x3F10, 0x0000, + 0x3F28, 0x0290, 0x0A90, 0x04A4, 0x3F14, 0x0000, + 0x3F30, 0x026C, 0x0A84, 0x04CC, 0x3F14, 0x0000, + 0x3F34, 0x024C, 0x0A7C, 0x04F4, 0x3F14, 0x3FFC, + 0x3F38, 0x0228, 0x0A70, 0x051C, 0x3F18, 0x3FFC, + 0x3F3C, 0x0208, 0x0A64, 0x0544, 0x3F1C, 0x3FF8, + 0x3F40, 0x01E8, 0x0A54, 0x056C, 0x3F20, 0x3FF8, + 0x3F44, 0x01C8, 0x0A48, 0x0594, 0x3F24, 0x3FF4, + 0x3F4C, 0x01A8, 0x0A34, 0x05BC, 0x3F28, 0x3FF4, + 0x3F50, 0x0188, 0x0A28, 0x05E4, 0x3F2C, 0x3FF0, + 0x3F54, 0x016C, 0x0A10, 0x060C, 0x3F34, 0x3FF0, + 0x3F5C, 0x014C, 0x09FC, 0x0634, 0x3F3C, 0x3FEC, + 0x3F60, 0x0130, 0x09EC, 0x065C, 0x3F40, 0x3FE8, + 0x3F68, 0x0114, 0x09D0, 0x0684, 0x3F48, 0x3FE8, + 0x3F6C, 0x00F8, 0x09B8, 0x06AC, 0x3F54, 0x3FE4, + 0x3F74, 0x00E0, 0x09A0, 0x06D0, 0x3F5C, 0x3FE0, + 0x3F78, 0x00C4, 0x098C, 0x06F8, 0x3F64, 0x3FDC, + 0x3F7C, 0x00AC, 0x0970, 0x0720, 0x3F70, 0x3FD8, + 0x3F84, 0x0094, 0x0954, 0x0744, 0x3F7C, 0x3FD4, + 0x3F88, 0x007C, 0x093C, 0x0768, 0x3F88, 0x3FD0, + 0x3F90, 0x0064, 0x091C, 0x0790, 0x3F94, 0x3FCC, + 0x3F94, 0x0050, 0x08FC, 0x07B4, 0x3FA4, 0x3FC8, + 0x3F98, 0x003C, 0x08E0, 0x07D8, 0x3FB0, 0x3FC4, + 0x3FA0, 0x0024, 0x08C0, 0x07FC, 0x3FC0, 0x3FC0, + 0x3FA4, 0x0014, 0x08A4, 0x081C, 0x3FD0, 0x3FB8, + 0x3FAC, 0x0000, 0x0880, 0x0840, 0x3FE0, 0x3FB4, + 0x3FB0, 0x3FF0, 0x0860, 0x0860, 0x3FF0, 0x3FB0 +}; +//========================================= +// <num_taps> = 6 +// <num_phases> = 64 +// <scale_ratio> = 1.83332 (input/output) +// <sharpness> = 0 +// <CoefType> = ModifiedLanczos +// <CoefQuant> = 1.10 +// <CoefOut> = 1.12 +//========================================= static const uint16_t filter_6tap_64p_183[198] = { - 16296, 1032, 2196, 1032, 16296, 0, - 16292, 1004, 2200, 1060, 16304, 16380, - 16288, 976, 2200, 1088, 16308, 16380, - 16284, 952, 2196, 1116, 16312, 16376, - 16284, 924, 2196, 1144, 16320, 16376, - 16280, 900, 2192, 1172, 16324, 16372, - 16276, 872, 2192, 1200, 16332, 16368, - 16276, 848, 2188, 1228, 16340, 16368, - 16272, 820, 2180, 1256, 16348, 16364, - 16272, 796, 2176, 1280, 16356, 16360, - 16268, 768, 2168, 1308, 16364, 16360, - 16268, 744, 2164, 1336, 16372, 16356, - 16268, 716, 2156, 1364, 16380, 16352, - 16264, 692, 2148, 1392, 4, 16352, - 16264, 668, 2136, 1420, 16, 16348, - 16264, 644, 2128, 1448, 28, 16344, - 16264, 620, 2116, 1472, 36, 16340, - 16264, 596, 2108, 1500, 48, 16340, - 16268, 572, 2096, 1524, 60, 16336, - 16268, 548, 2080, 1552, 72, 16332, - 16268, 524, 2068, 1576, 88, 16328, - 16268, 504, 2056, 1604, 100, 16324, - 16272, 480, 2040, 1628, 112, 16324, - 16272, 456, 2024, 1652, 128, 16320, - 16272, 436, 2008, 1680, 144, 16316, - 16276, 416, 1992, 1704, 156, 16312, - 16276, 392, 1976, 1724, 172, 16308, - 16280, 372, 1956, 1748, 188, 16308, - 16280, 352, 1940, 1772, 204, 16304, - 16284, 332, 1920, 1796, 224, 16300, - 16288, 312, 1900, 1816, 240, 16296, - 16288, 296, 1880, 1840, 256, 16296, - 16292, 276, 1860, 1860, 276, 16292 }; + 0x002C, 0x0420, 0x076C, 0x041C, 0x002C, 0x0000, + 0x0028, 0x040C, 0x0768, 0x0430, 0x0034, 0x0000, + 0x0020, 0x03F8, 0x0768, 0x0448, 0x003C, 0x3FFC, + 0x0018, 0x03E4, 0x0768, 0x045C, 0x0044, 0x3FFC, + 0x0014, 0x03D0, 0x0768, 0x0470, 0x004C, 0x3FF8, + 0x000C, 0x03BC, 0x0764, 0x0484, 0x0058, 0x3FF8, + 0x0008, 0x03A4, 0x0764, 0x049C, 0x0060, 0x3FF4, + 0x0004, 0x0390, 0x0760, 0x04B0, 0x0068, 0x3FF4, + 0x0000, 0x037C, 0x0760, 0x04C4, 0x0070, 0x3FF0, + 0x3FFC, 0x0364, 0x075C, 0x04D8, 0x007C, 0x3FF0, + 0x3FF8, 0x0350, 0x0758, 0x04F0, 0x0084, 0x3FEC, + 0x3FF4, 0x033C, 0x0750, 0x0504, 0x0090, 0x3FEC, + 0x3FF0, 0x0328, 0x074C, 0x0518, 0x009C, 0x3FE8, + 0x3FEC, 0x0314, 0x0744, 0x052C, 0x00A8, 0x3FE8, + 0x3FE8, 0x0304, 0x0740, 0x0540, 0x00B0, 0x3FE4, + 0x3FE4, 0x02EC, 0x073C, 0x0554, 0x00BC, 0x3FE4, + 0x3FE0, 0x02DC, 0x0734, 0x0568, 0x00C8, 0x3FE0, + 0x3FE0, 0x02C4, 0x072C, 0x057C, 0x00D4, 0x3FE0, + 0x3FDC, 0x02B4, 0x0724, 0x058C, 0x00E4, 0x3FDC, + 0x3FDC, 0x02A0, 0x0718, 0x05A0, 0x00F0, 0x3FDC, + 0x3FD8, 0x028C, 0x0714, 0x05B4, 0x00FC, 0x3FD8, + 0x3FD8, 0x0278, 0x0704, 0x05C8, 0x010C, 0x3FD8, + 0x3FD4, 0x0264, 0x0700, 0x05D8, 0x0118, 0x3FD8, + 0x3FD4, 0x0254, 0x06F0, 0x05EC, 0x0128, 0x3FD4, + 0x3FD0, 0x0244, 0x06E8, 0x05FC, 0x0134, 0x3FD4, + 0x3FD0, 0x0230, 0x06DC, 0x060C, 0x0144, 0x3FD4, + 0x3FD0, 0x021C, 0x06D0, 0x0620, 0x0154, 0x3FD0, + 0x3FD0, 0x0208, 0x06C4, 0x0630, 0x0164, 0x3FD0, + 0x3FD0, 0x01F8, 0x06B8, 0x0640, 0x0170, 0x3FD0, + 0x3FCC, 0x01E8, 0x06AC, 0x0650, 0x0180, 0x3FD0, + 0x3FCC, 0x01D8, 0x069C, 0x0660, 0x0190, 0x3FD0, + 0x3FCC, 0x01C4, 0x068C, 0x0670, 0x01A4, 0x3FD0, + 0x3FCC, 0x01B8, 0x0680, 0x067C, 0x01B4, 0x3FCC +}; +//========================================= +// <num_taps> = 7 +// <num_phases> = 64 +// <scale_ratio> = 0.83333 (input/output) +// <sharpness> = 0 +// <CoefType> = ModifiedLanczos +// <CoefQuant> = 1.10 +// <CoefOut> = 1.12 +//========================================= static const uint16_t filter_7tap_64p_upscale[231] = { - 176, 15760, 2488, 2488, 15760, 176, 0, - 172, 15772, 2404, 2572, 15752, 180, 16380, - 168, 15784, 2324, 2656, 15740, 184, 16380, - 164, 15800, 2240, 2736, 15732, 188, 16376, - 160, 15812, 2152, 2816, 15728, 192, 16376, - 152, 15828, 2068, 2896, 15724, 192, 16376, - 148, 15848, 1984, 2972, 15720, 196, 16372, - 140, 15864, 1896, 3048, 15720, 196, 16372, - 136, 15884, 1812, 3124, 15720, 196, 16368, - 128, 15900, 1724, 3196, 15720, 196, 16368, - 120, 15920, 1640, 3268, 15724, 196, 16368, - 116, 15940, 1552, 3336, 15732, 196, 16364, - 108, 15964, 1468, 3400, 15740, 196, 16364, - 104, 15984, 1384, 3464, 15748, 192, 16364, - 96, 16004, 1300, 3524, 15760, 188, 16364, - 88, 16028, 1216, 3584, 15776, 184, 16364, - 84, 16048, 1132, 3640, 15792, 180, 16360, - 76, 16072, 1048, 3692, 15812, 176, 16360, - 68, 16092, 968, 3744, 15832, 168, 16360, - 64, 16116, 888, 3788, 15856, 160, 16360, - 56, 16140, 812, 3832, 15884, 152, 16360, - 52, 16160, 732, 3876, 15912, 144, 16360, - 44, 16184, 656, 3912, 15944, 136, 16364, - 40, 16204, 584, 3944, 15976, 124, 16364, - 32, 16228, 512, 3976, 16012, 116, 16364, - 28, 16248, 440, 4004, 16048, 104, 16364, - 24, 16268, 372, 4028, 16092, 88, 16368, - 20, 16288, 304, 4048, 16132, 76, 16368, - 12, 16308, 240, 4064, 16180, 60, 16372, - 8, 16328, 176, 4076, 16228, 48, 16372, - 4, 16348, 112, 4088, 16276, 32, 16376, - 0, 16364, 56, 4092, 16328, 16, 16380, - 0, 0, 0, 4096, 0, 0, 0 }; + 0x00B0, 0x3D98, 0x09BC, 0x09B8, 0x3D94, 0x00B0, 0x0000, + 0x00AC, 0x3DA0, 0x0968, 0x0A10, 0x3D88, 0x00B4, 0x0000, + 0x00A8, 0x3DAC, 0x0914, 0x0A60, 0x3D80, 0x00B8, 0x0000, + 0x00A4, 0x3DB8, 0x08C0, 0x0AB4, 0x3D78, 0x00BC, 0x3FFC, + 0x00A0, 0x3DC8, 0x0868, 0x0B00, 0x3D74, 0x00C0, 0x3FFC, + 0x0098, 0x3DD8, 0x0818, 0x0B54, 0x3D6C, 0x00C0, 0x3FF8, + 0x0094, 0x3DE8, 0x07C0, 0x0B9C, 0x3D6C, 0x00C4, 0x3FF8, + 0x008C, 0x3DFC, 0x0768, 0x0BEC, 0x3D68, 0x00C4, 0x3FF8, + 0x0088, 0x3E0C, 0x0714, 0x0C38, 0x3D68, 0x00C4, 0x3FF4, + 0x0080, 0x3E20, 0x06BC, 0x0C80, 0x3D6C, 0x00C4, 0x3FF4, + 0x0078, 0x3E34, 0x0668, 0x0CC4, 0x3D70, 0x00C4, 0x3FF4, + 0x0074, 0x3E48, 0x0610, 0x0D08, 0x3D78, 0x00C4, 0x3FF0, + 0x006C, 0x3E5C, 0x05BC, 0x0D48, 0x3D80, 0x00C4, 0x3FF0, + 0x0068, 0x3E74, 0x0568, 0x0D84, 0x3D88, 0x00C0, 0x3FF0, + 0x0060, 0x3E88, 0x0514, 0x0DC8, 0x3D94, 0x00BC, 0x3FEC, + 0x0058, 0x3E9C, 0x04C0, 0x0E04, 0x3DA4, 0x00B8, 0x3FEC, + 0x0054, 0x3EB4, 0x046C, 0x0E38, 0x3DB4, 0x00B4, 0x3FEC, + 0x004C, 0x3ECC, 0x0418, 0x0E6C, 0x3DC8, 0x00B0, 0x3FEC, + 0x0044, 0x3EE0, 0x03C8, 0x0EA4, 0x3DDC, 0x00A8, 0x3FEC, + 0x0040, 0x3EF8, 0x0378, 0x0ED0, 0x3DF4, 0x00A0, 0x3FEC, + 0x0038, 0x3F0C, 0x032C, 0x0EFC, 0x3E10, 0x0098, 0x3FEC, + 0x0034, 0x3F24, 0x02DC, 0x0F24, 0x3E2C, 0x0090, 0x3FEC, + 0x002C, 0x3F38, 0x0294, 0x0F4C, 0x3E48, 0x0088, 0x3FEC, + 0x0028, 0x3F50, 0x0248, 0x0F68, 0x3E6C, 0x007C, 0x3FF0, + 0x0020, 0x3F64, 0x0200, 0x0F88, 0x3E90, 0x0074, 0x3FF0, + 0x001C, 0x3F7C, 0x01B8, 0x0FA4, 0x3EB4, 0x0068, 0x3FF0, + 0x0018, 0x3F90, 0x0174, 0x0FBC, 0x3EDC, 0x0058, 0x3FF4, + 0x0014, 0x3FA4, 0x0130, 0x0FD0, 0x3F08, 0x004C, 0x3FF4, + 0x000C, 0x3FB8, 0x00F0, 0x0FE4, 0x3F34, 0x003C, 0x3FF8, + 0x0008, 0x3FCC, 0x00B0, 0x0FF0, 0x3F64, 0x0030, 0x3FF8, + 0x0004, 0x3FDC, 0x0070, 0x0FFC, 0x3F98, 0x0020, 0x3FFC, + 0x0000, 0x3FF0, 0x0038, 0x0FFC, 0x3FCC, 0x0010, 0x0000, + 0x0000, 0x0000, 0x0000, 0x1000, 0x0000, 0x0000, 0x0000 +}; -static const uint16_t filter_7tap_64p_117[231] = { - 92, 15868, 2464, 2464, 15868, 92, 0, - 96, 15864, 2404, 2528, 15876, 88, 0, - 100, 15860, 2344, 2584, 15884, 84, 0, - 104, 15856, 2280, 2644, 15892, 76, 0, - 108, 15852, 2216, 2700, 15904, 72, 0, - 108, 15852, 2152, 2756, 15916, 64, 0, - 112, 15852, 2088, 2812, 15932, 60, 0, - 112, 15852, 2024, 2864, 15948, 52, 0, - 112, 15856, 1960, 2916, 15964, 44, 0, - 116, 15860, 1892, 2964, 15984, 36, 0, - 116, 15864, 1828, 3016, 16004, 24, 4, - 116, 15868, 1760, 3060, 16024, 16, 4, - 116, 15876, 1696, 3108, 16048, 8, 8, - 116, 15884, 1628, 3152, 16072, 16380, 8, - 112, 15892, 1564, 3192, 16100, 16372, 8, - 112, 15900, 1496, 3232, 16124, 16360, 12, - 112, 15908, 1428, 3268, 16156, 16348, 12, - 108, 15920, 1364, 3304, 16188, 16336, 16, - 108, 15928, 1300, 3340, 16220, 16324, 20, - 104, 15940, 1232, 3372, 16252, 16312, 20, - 104, 15952, 1168, 3400, 16288, 16300, 24, - 100, 15964, 1104, 3428, 16328, 16284, 28, - 96, 15980, 1040, 3452, 16364, 16272, 28, - 96, 15992, 976, 3476, 20, 16256, 32, - 92, 16004, 916, 3496, 64, 16244, 36, - 88, 16020, 856, 3516, 108, 16228, 40, - 84, 16032, 792, 3532, 152, 16216, 44, - 80, 16048, 732, 3544, 200, 16200, 48, - 80, 16064, 676, 3556, 248, 16184, 48, - 76, 16080, 616, 3564, 296, 16168, 52, - 72, 16092, 560, 3568, 344, 16156, 56, - 68, 16108, 504, 3572, 396, 16140, 60, - 64, 16124, 452, 3576, 452, 16124, 64 }; +//========================================= +// <num_taps> = 7 +// <num_phases> = 64 +// <scale_ratio> = 1.16666 (input/output) +// <sharpness> = 0 +// <CoefType> = ModifiedLanczos +// <CoefQuant> = 1.10 +// <CoefOut> = 1.12 +//========================================= +static const uint16_t filter_7tap_64p_116[231] = { + 0x0020, 0x3E58, 0x0988, 0x0988, 0x3E58, 0x0020, 0x0000, + 0x0024, 0x3E4C, 0x0954, 0x09C0, 0x3E64, 0x0018, 0x0000, + 0x002C, 0x3E44, 0x091C, 0x09F4, 0x3E70, 0x0010, 0x0000, + 0x0030, 0x3E3C, 0x08E8, 0x0A24, 0x3E80, 0x0008, 0x0000, + 0x0034, 0x3E34, 0x08AC, 0x0A5C, 0x3E90, 0x0000, 0x0000, + 0x003C, 0x3E30, 0x0870, 0x0A84, 0x3EA0, 0x3FFC, 0x0004, + 0x0040, 0x3E28, 0x0838, 0x0AB4, 0x3EB4, 0x3FF4, 0x0004, + 0x0044, 0x3E24, 0x07FC, 0x0AE4, 0x3EC8, 0x3FEC, 0x0004, + 0x0048, 0x3E24, 0x07C4, 0x0B08, 0x3EDC, 0x3FE4, 0x0008, + 0x0048, 0x3E20, 0x0788, 0x0B3C, 0x3EF4, 0x3FD8, 0x0008, + 0x004C, 0x3E20, 0x074C, 0x0B60, 0x3F0C, 0x3FD0, 0x000C, + 0x0050, 0x3E20, 0x0710, 0x0B8C, 0x3F24, 0x3FC4, 0x000C, + 0x0050, 0x3E20, 0x06D4, 0x0BB0, 0x3F40, 0x3FBC, 0x0010, + 0x0054, 0x3E24, 0x0698, 0x0BD4, 0x3F5C, 0x3FB0, 0x0010, + 0x0054, 0x3E24, 0x065C, 0x0BFC, 0x3F78, 0x3FA4, 0x0014, + 0x0054, 0x3E28, 0x0624, 0x0C1C, 0x3F98, 0x3F98, 0x0014, + 0x0058, 0x3E2C, 0x05E4, 0x0C3C, 0x3FB8, 0x3F8C, 0x0018, + 0x0058, 0x3E34, 0x05A8, 0x0C58, 0x3FD8, 0x3F80, 0x001C, + 0x0058, 0x3E38, 0x0570, 0x0C78, 0x3FF8, 0x3F74, 0x001C, + 0x0058, 0x3E40, 0x0534, 0x0C94, 0x0018, 0x3F68, 0x0020, + 0x0058, 0x3E48, 0x04F4, 0x0CAC, 0x0040, 0x3F5C, 0x0024, + 0x0058, 0x3E50, 0x04BC, 0x0CC4, 0x0064, 0x3F50, 0x0024, + 0x0054, 0x3E58, 0x0484, 0x0CD8, 0x008C, 0x3F44, 0x0028, + 0x0054, 0x3E60, 0x0448, 0x0CEC, 0x00B4, 0x3F38, 0x002C, + 0x0054, 0x3E68, 0x0410, 0x0CFC, 0x00E0, 0x3F28, 0x0030, + 0x0054, 0x3E74, 0x03D4, 0x0D0C, 0x010C, 0x3F1C, 0x0030, + 0x0050, 0x3E7C, 0x03A0, 0x0D18, 0x0138, 0x3F10, 0x0034, + 0x0050, 0x3E88, 0x0364, 0x0D24, 0x0164, 0x3F04, 0x0038, + 0x004C, 0x3E94, 0x0330, 0x0D30, 0x0194, 0x3EF4, 0x0038, + 0x004C, 0x3EA0, 0x02F8, 0x0D34, 0x01C4, 0x3EE8, 0x003C, + 0x0048, 0x3EAC, 0x02C0, 0x0D3C, 0x01F4, 0x3EDC, 0x0040, + 0x0048, 0x3EB8, 0x0290, 0x0D3C, 0x0224, 0x3ED0, 0x0040, + 0x0044, 0x3EC4, 0x0258, 0x0D40, 0x0258, 0x3EC4, 0x0044 +}; -static const uint16_t filter_7tap_64p_150[231] = { - 16224, 16380, 2208, 2208, 16380, 16224, 0, - 16232, 16360, 2172, 2236, 16, 16216, 0, - 16236, 16340, 2140, 2268, 40, 16212, 0, - 16244, 16324, 2104, 2296, 60, 16204, 4, - 16252, 16304, 2072, 2324, 84, 16196, 4, - 16256, 16288, 2036, 2352, 108, 16192, 4, - 16264, 16268, 2000, 2380, 132, 16184, 8, - 16272, 16252, 1960, 2408, 160, 16176, 8, - 16276, 16240, 1924, 2432, 184, 16172, 8, - 16284, 16224, 1888, 2456, 212, 16164, 8, - 16288, 16212, 1848, 2480, 240, 16160, 12, - 16296, 16196, 1812, 2500, 268, 16152, 12, - 16300, 16184, 1772, 2524, 296, 16144, 12, - 16308, 16172, 1736, 2544, 324, 16140, 12, - 16312, 16164, 1696, 2564, 356, 16136, 12, - 16320, 16152, 1656, 2584, 388, 16128, 12, - 16324, 16144, 1616, 2600, 416, 16124, 12, - 16328, 16136, 1576, 2616, 448, 16116, 12, - 16332, 16128, 1536, 2632, 480, 16112, 12, - 16340, 16120, 1496, 2648, 516, 16108, 12, - 16344, 16112, 1456, 2660, 548, 16104, 12, - 16348, 16104, 1416, 2672, 580, 16100, 12, - 16352, 16100, 1376, 2684, 616, 16096, 12, - 16356, 16096, 1336, 2696, 652, 16092, 12, - 16360, 16092, 1296, 2704, 688, 16088, 12, - 16364, 16088, 1256, 2712, 720, 16084, 12, - 16368, 16084, 1220, 2720, 760, 16084, 8, - 16368, 16080, 1180, 2724, 796, 16080, 8, - 16372, 16080, 1140, 2732, 832, 16080, 8, - 16376, 16076, 1100, 2732, 868, 16076, 4, - 16380, 16076, 1060, 2736, 908, 16076, 4, - 16380, 16076, 1020, 2740, 944, 16076, 0, - 0, 16076, 984, 2740, 984, 16076, 0 }; +//========================================= +// <num_taps> = 7 +// <num_phases> = 64 +// <scale_ratio> = 1.49999 (input/output) +// <sharpness> = 0 +// <CoefType> = ModifiedLanczos +// <CoefQuant> = 1.10 +// <CoefOut> = 1.12 +//========================================= +static const uint16_t filter_7tap_64p_149[231] = { + 0x3F68, 0x3FEC, 0x08A8, 0x08AC, 0x3FF0, 0x3F68, 0x0000, + 0x3F70, 0x3FDC, 0x0888, 0x08CC, 0x0000, 0x3F60, 0x0000, + 0x3F74, 0x3FC8, 0x0868, 0x08F0, 0x0014, 0x3F58, 0x0000, + 0x3F7C, 0x3FB4, 0x0844, 0x0908, 0x002C, 0x3F54, 0x0004, + 0x3F84, 0x3FA4, 0x0820, 0x0924, 0x0044, 0x3F4C, 0x0004, + 0x3F88, 0x3F90, 0x0800, 0x0944, 0x005C, 0x3F44, 0x0004, + 0x3F90, 0x3F80, 0x07D8, 0x095C, 0x0074, 0x3F40, 0x0008, + 0x3F98, 0x3F70, 0x07B0, 0x097C, 0x008C, 0x3F38, 0x0008, + 0x3F9C, 0x3F60, 0x0790, 0x0994, 0x00A8, 0x3F30, 0x0008, + 0x3FA4, 0x3F54, 0x0764, 0x09B0, 0x00C4, 0x3F28, 0x0008, + 0x3FA8, 0x3F48, 0x0740, 0x09C4, 0x00DC, 0x3F24, 0x000C, + 0x3FB0, 0x3F38, 0x0718, 0x09DC, 0x00FC, 0x3F1C, 0x000C, + 0x3FB4, 0x3F2C, 0x06F0, 0x09F4, 0x0118, 0x3F18, 0x000C, + 0x3FBC, 0x3F24, 0x06C8, 0x0A08, 0x0134, 0x3F10, 0x000C, + 0x3FC0, 0x3F18, 0x06A0, 0x0A1C, 0x0154, 0x3F08, 0x0010, + 0x3FC8, 0x3F10, 0x0678, 0x0A2C, 0x0170, 0x3F04, 0x0010, + 0x3FCC, 0x3F04, 0x0650, 0x0A40, 0x0190, 0x3F00, 0x0010, + 0x3FD0, 0x3EFC, 0x0628, 0x0A54, 0x01B0, 0x3EF8, 0x0010, + 0x3FD4, 0x3EF4, 0x0600, 0x0A64, 0x01D0, 0x3EF4, 0x0010, + 0x3FDC, 0x3EEC, 0x05D8, 0x0A6C, 0x01F4, 0x3EF0, 0x0010, + 0x3FE0, 0x3EE8, 0x05B0, 0x0A7C, 0x0214, 0x3EE8, 0x0010, + 0x3FE4, 0x3EE0, 0x0588, 0x0A88, 0x0238, 0x3EE4, 0x0010, + 0x3FE8, 0x3EDC, 0x055C, 0x0A98, 0x0258, 0x3EE0, 0x0010, + 0x3FEC, 0x3ED8, 0x0534, 0x0AA0, 0x027C, 0x3EDC, 0x0010, + 0x3FF0, 0x3ED4, 0x050C, 0x0AAC, 0x02A0, 0x3ED8, 0x000C, + 0x3FF4, 0x3ED0, 0x04E4, 0x0AB4, 0x02C4, 0x3ED4, 0x000C, + 0x3FF4, 0x3ECC, 0x04C0, 0x0ABC, 0x02E8, 0x3ED0, 0x000C, + 0x3FF8, 0x3ECC, 0x0494, 0x0AC0, 0x030C, 0x3ED0, 0x000C, + 0x3FFC, 0x3EC8, 0x046C, 0x0AC8, 0x0334, 0x3ECC, 0x0008, + 0x0000, 0x3EC8, 0x0444, 0x0AC8, 0x0358, 0x3ECC, 0x0008, + 0x0000, 0x3EC8, 0x041C, 0x0ACC, 0x0380, 0x3EC8, 0x0008, + 0x0000, 0x3EC8, 0x03F4, 0x0AD0, 0x03A8, 0x3EC8, 0x0004, + 0x0004, 0x3EC8, 0x03CC, 0x0AD0, 0x03CC, 0x3EC8, 0x0004 +}; +//========================================= +// <num_taps> = 7 +// <num_phases> = 64 +// <scale_ratio> = 1.83332 (input/output) +// <sharpness> = 0 +// <CoefType> = ModifiedLanczos +// <CoefQuant> = 1.10 +// <CoefOut> = 1.12 +//========================================= static const uint16_t filter_7tap_64p_183[231] = { - 16216, 324, 1884, 1884, 324, 16216, 0, - 16220, 304, 1864, 1904, 344, 16216, 0, - 16224, 284, 1844, 1924, 364, 16216, 0, - 16224, 264, 1824, 1944, 384, 16212, 16380, - 16228, 248, 1804, 1960, 408, 16212, 16380, - 16228, 228, 1784, 1976, 428, 16208, 16380, - 16232, 212, 1760, 1996, 452, 16208, 16380, - 16236, 192, 1740, 2012, 472, 16208, 16376, - 16240, 176, 1716, 2028, 496, 16208, 16376, - 16240, 160, 1696, 2040, 516, 16208, 16376, - 16244, 144, 1672, 2056, 540, 16208, 16376, - 16248, 128, 1648, 2068, 564, 16208, 16372, - 16252, 112, 1624, 2084, 588, 16208, 16372, - 16256, 96, 1600, 2096, 612, 16208, 16368, - 16256, 84, 1576, 2108, 636, 16208, 16368, - 16260, 68, 1552, 2120, 660, 16208, 16368, - 16264, 56, 1524, 2132, 684, 16212, 16364, - 16268, 40, 1500, 2140, 712, 16212, 16364, - 16272, 28, 1476, 2152, 736, 16216, 16360, - 16276, 16, 1448, 2160, 760, 16216, 16356, - 16280, 4, 1424, 2168, 788, 16220, 16356, - 16284, 16376, 1396, 2176, 812, 16224, 16352, - 16288, 16368, 1372, 2184, 840, 16224, 16352, - 16292, 16356, 1344, 2188, 864, 16228, 16348, - 16292, 16344, 1320, 2196, 892, 16232, 16344, - 16296, 16336, 1292, 2200, 916, 16236, 16344, - 16300, 16324, 1264, 2204, 944, 16240, 16340, - 16304, 16316, 1240, 2208, 972, 16248, 16336, - 16308, 16308, 1212, 2212, 996, 16252, 16332, - 16312, 16300, 1184, 2216, 1024, 16256, 16332, - 16316, 16292, 1160, 2216, 1052, 16264, 16328, - 16316, 16284, 1132, 2216, 1076, 16268, 16324, - 16320, 16276, 1104, 2216, 1104, 16276, 16320 }; + 0x3FA4, 0x01E8, 0x0674, 0x0674, 0x01E8, 0x3FA4, 0x0000, + 0x3FA4, 0x01D4, 0x0668, 0x0684, 0x01F8, 0x3FA4, 0x0000, + 0x3FA4, 0x01C4, 0x0658, 0x0690, 0x0208, 0x3FA8, 0x0000, + 0x3FA0, 0x01B4, 0x064C, 0x06A0, 0x021C, 0x3FA8, 0x3FFC, + 0x3FA0, 0x01A4, 0x063C, 0x06AC, 0x022C, 0x3FAC, 0x3FFC, + 0x3FA0, 0x0194, 0x0630, 0x06B4, 0x0240, 0x3FAC, 0x3FFC, + 0x3FA0, 0x0184, 0x0620, 0x06C4, 0x0250, 0x3FB0, 0x3FF8, + 0x3FA0, 0x0174, 0x0614, 0x06CC, 0x0264, 0x3FB0, 0x3FF8, + 0x3FA0, 0x0164, 0x0604, 0x06D8, 0x0278, 0x3FB4, 0x3FF4, + 0x3FA0, 0x0154, 0x05F4, 0x06E4, 0x0288, 0x3FB8, 0x3FF4, + 0x3FA0, 0x0148, 0x05E4, 0x06EC, 0x029C, 0x3FBC, 0x3FF0, + 0x3FA0, 0x0138, 0x05D4, 0x06F4, 0x02B0, 0x3FC0, 0x3FF0, + 0x3FA0, 0x0128, 0x05C4, 0x0704, 0x02C4, 0x3FC0, 0x3FEC, + 0x3FA0, 0x011C, 0x05B4, 0x0708, 0x02D8, 0x3FC4, 0x3FEC, + 0x3FA4, 0x010C, 0x05A4, 0x0714, 0x02E8, 0x3FC8, 0x3FE8, + 0x3FA4, 0x0100, 0x0590, 0x0718, 0x02FC, 0x3FD0, 0x3FE8, + 0x3FA4, 0x00F0, 0x0580, 0x0724, 0x0310, 0x3FD4, 0x3FE4, + 0x3FA4, 0x00E4, 0x056C, 0x072C, 0x0324, 0x3FD8, 0x3FE4, + 0x3FA8, 0x00D8, 0x055C, 0x0730, 0x0338, 0x3FDC, 0x3FE0, + 0x3FA8, 0x00CC, 0x0548, 0x0738, 0x034C, 0x3FE4, 0x3FDC, + 0x3FA8, 0x00BC, 0x0538, 0x0740, 0x0360, 0x3FE8, 0x3FDC, + 0x3FAC, 0x00B0, 0x0528, 0x0744, 0x0374, 0x3FEC, 0x3FD8, + 0x3FAC, 0x00A4, 0x0514, 0x0748, 0x0388, 0x3FF4, 0x3FD8, + 0x3FB0, 0x0098, 0x0500, 0x074C, 0x039C, 0x3FFC, 0x3FD4, + 0x3FB0, 0x0090, 0x04EC, 0x0750, 0x03B0, 0x0000, 0x3FD4, + 0x3FB0, 0x0084, 0x04DC, 0x0758, 0x03C4, 0x0004, 0x3FD0, + 0x3FB4, 0x0078, 0x04CC, 0x0758, 0x03D8, 0x000C, 0x3FCC, + 0x3FB4, 0x006C, 0x04B8, 0x075C, 0x03EC, 0x0014, 0x3FCC, + 0x3FB8, 0x0064, 0x04A0, 0x0760, 0x0400, 0x001C, 0x3FC8, + 0x3FB8, 0x0058, 0x0490, 0x0760, 0x0414, 0x0024, 0x3FC8, + 0x3FBC, 0x0050, 0x047C, 0x0760, 0x0428, 0x002C, 0x3FC4, + 0x3FBC, 0x0048, 0x0464, 0x0764, 0x043C, 0x0034, 0x3FC4, + 0x3FC0, 0x003C, 0x0454, 0x0764, 0x0450, 0x003C, 0x3FC0 +}; +//========================================= +// <num_taps> = 8 +// <num_phases> = 64 +// <scale_ratio> = 0.83333 (input/output) +// <sharpness> = 0 +// <CoefType> = ModifiedLanczos +// <CoefQuant> = 1.10 +// <CoefOut> = 1.12 +//========================================= static const uint16_t filter_8tap_64p_upscale[264] = { - 0, 0, 0, 4096, 0, 0, 0, 0, - 16376, 20, 16328, 4092, 56, 16364, 4, 0, - 16372, 36, 16272, 4088, 116, 16340, 12, 0, - 16364, 56, 16220, 4080, 180, 16320, 20, 0, - 16360, 76, 16172, 4064, 244, 16296, 24, 16380, - 16356, 92, 16124, 4048, 312, 16276, 32, 16380, - 16352, 108, 16080, 4032, 380, 16252, 40, 16380, - 16344, 124, 16036, 4008, 452, 16228, 48, 16380, - 16340, 136, 15996, 3980, 524, 16204, 56, 16380, - 16340, 152, 15956, 3952, 600, 16180, 64, 16376, - 16336, 164, 15920, 3920, 672, 16156, 76, 16376, - 16332, 176, 15888, 3884, 752, 16132, 84, 16376, - 16328, 188, 15860, 3844, 828, 16104, 92, 16372, - 16328, 200, 15828, 3800, 908, 16080, 100, 16372, - 16324, 208, 15804, 3756, 992, 16056, 108, 16372, - 16324, 216, 15780, 3708, 1072, 16032, 120, 16368, - 16320, 224, 15760, 3656, 1156, 16008, 128, 16368, - 16320, 232, 15740, 3604, 1240, 15984, 136, 16364, - 16320, 240, 15724, 3548, 1324, 15960, 144, 16364, - 16320, 244, 15708, 3488, 1412, 15936, 152, 16360, - 16320, 248, 15696, 3428, 1496, 15912, 160, 16360, - 16320, 252, 15688, 3364, 1584, 15892, 172, 16356, - 16320, 256, 15680, 3296, 1672, 15868, 180, 16352, - 16320, 256, 15672, 3228, 1756, 15848, 188, 16352, - 16320, 256, 15668, 3156, 1844, 15828, 192, 16348, - 16320, 260, 15668, 3084, 1932, 15808, 200, 16348, - 16320, 256, 15668, 3012, 2020, 15792, 208, 16344, - 16324, 256, 15668, 2936, 2108, 15772, 216, 16344, - 16324, 256, 15672, 2856, 2192, 15756, 220, 16340, - 16324, 252, 15676, 2776, 2280, 15740, 228, 16336, - 16328, 252, 15684, 2696, 2364, 15728, 232, 16336, - 16328, 248, 15692, 2616, 2448, 15716, 240, 16332, - 16332, 244, 15704, 2532, 2532, 15704, 244, 16332 }; + 0x0000, 0x0000, 0x0000, 0x1000, 0x0000, 0x0000, 0x0000, 0x0000, + 0x3FFC, 0x0014, 0x3FC8, 0x1000, 0x0038, 0x3FEC, 0x0004, 0x0000, + 0x3FF4, 0x0024, 0x3F94, 0x0FFC, 0x0074, 0x3FD8, 0x000C, 0x0000, + 0x3FF0, 0x0038, 0x3F60, 0x0FEC, 0x00B4, 0x3FC4, 0x0014, 0x0000, + 0x3FEC, 0x004C, 0x3F2C, 0x0FE4, 0x00F4, 0x3FAC, 0x0018, 0x0000, + 0x3FE4, 0x005C, 0x3F00, 0x0FD4, 0x0138, 0x3F94, 0x0020, 0x0000, + 0x3FE0, 0x006C, 0x3ED0, 0x0FC4, 0x017C, 0x3F7C, 0x0028, 0x0000, + 0x3FDC, 0x007C, 0x3EA8, 0x0FA4, 0x01C4, 0x3F68, 0x0030, 0x0000, + 0x3FD8, 0x0088, 0x3E80, 0x0F90, 0x020C, 0x3F50, 0x0038, 0x3FFC, + 0x3FD4, 0x0098, 0x3E58, 0x0F70, 0x0258, 0x3F38, 0x0040, 0x3FFC, + 0x3FD0, 0x00A4, 0x3E34, 0x0F54, 0x02A0, 0x3F1C, 0x004C, 0x3FFC, + 0x3FD0, 0x00B0, 0x3E14, 0x0F28, 0x02F0, 0x3F04, 0x0054, 0x3FFC, + 0x3FCC, 0x00BC, 0x3DF4, 0x0F08, 0x033C, 0x3EEC, 0x005C, 0x3FF8, + 0x3FC8, 0x00C8, 0x3DD8, 0x0EDC, 0x038C, 0x3ED4, 0x0064, 0x3FF8, + 0x3FC8, 0x00D0, 0x3DC0, 0x0EAC, 0x03E0, 0x3EBC, 0x006C, 0x3FF4, + 0x3FC4, 0x00D8, 0x3DA8, 0x0E7C, 0x0430, 0x3EA4, 0x0078, 0x3FF4, + 0x3FC4, 0x00E0, 0x3D94, 0x0E48, 0x0484, 0x3E8C, 0x0080, 0x3FF0, + 0x3FC4, 0x00E8, 0x3D80, 0x0E10, 0x04D8, 0x3E74, 0x0088, 0x3FF0, + 0x3FC4, 0x00F0, 0x3D70, 0x0DD8, 0x052C, 0x3E5C, 0x0090, 0x3FEC, + 0x3FC0, 0x00F4, 0x3D60, 0x0DA0, 0x0584, 0x3E44, 0x0098, 0x3FEC, + 0x3FC0, 0x00F8, 0x3D54, 0x0D68, 0x05D8, 0x3E2C, 0x00A0, 0x3FE8, + 0x3FC0, 0x00FC, 0x3D48, 0x0D20, 0x0630, 0x3E18, 0x00AC, 0x3FE8, + 0x3FC0, 0x0100, 0x3D40, 0x0CE0, 0x0688, 0x3E00, 0x00B4, 0x3FE4, + 0x3FC4, 0x0100, 0x3D3C, 0x0C98, 0x06DC, 0x3DEC, 0x00BC, 0x3FE4, + 0x3FC4, 0x0100, 0x3D38, 0x0C58, 0x0734, 0x3DD8, 0x00C0, 0x3FE0, + 0x3FC4, 0x0104, 0x3D38, 0x0C0C, 0x078C, 0x3DC4, 0x00C8, 0x3FDC, + 0x3FC4, 0x0100, 0x3D38, 0x0BC4, 0x07E4, 0x3DB0, 0x00D0, 0x3FDC, + 0x3FC4, 0x0100, 0x3D38, 0x0B78, 0x083C, 0x3DA0, 0x00D8, 0x3FD8, + 0x3FC8, 0x0100, 0x3D3C, 0x0B28, 0x0890, 0x3D90, 0x00DC, 0x3FD8, + 0x3FC8, 0x00FC, 0x3D40, 0x0ADC, 0x08E8, 0x3D80, 0x00E4, 0x3FD4, + 0x3FCC, 0x00FC, 0x3D48, 0x0A84, 0x093C, 0x3D74, 0x00E8, 0x3FD4, + 0x3FCC, 0x00F8, 0x3D50, 0x0A38, 0x0990, 0x3D64, 0x00F0, 0x3FD0, + 0x3FD0, 0x00F4, 0x3D58, 0x09E0, 0x09E4, 0x3D5C, 0x00F4, 0x3FD0 +}; -static const uint16_t filter_8tap_64p_117[264] = { - 116, 16100, 428, 3564, 428, 16100, 116, 0, - 112, 16116, 376, 3564, 484, 16084, 120, 16380, - 104, 16136, 324, 3560, 540, 16064, 124, 16380, - 100, 16152, 272, 3556, 600, 16048, 128, 16380, - 96, 16168, 220, 3548, 656, 16032, 136, 16376, - 88, 16188, 172, 3540, 716, 16016, 140, 16376, - 84, 16204, 124, 3528, 780, 16000, 144, 16376, - 80, 16220, 76, 3512, 840, 15984, 148, 16372, - 76, 16236, 32, 3496, 904, 15968, 152, 16372, - 68, 16252, 16376, 3480, 968, 15952, 156, 16372, - 64, 16268, 16332, 3456, 1032, 15936, 160, 16372, - 60, 16284, 16292, 3432, 1096, 15920, 164, 16368, - 56, 16300, 16252, 3408, 1164, 15908, 164, 16368, - 48, 16316, 16216, 3380, 1228, 15892, 168, 16368, - 44, 16332, 16180, 3348, 1296, 15880, 168, 16368, - 40, 16348, 16148, 3316, 1364, 15868, 172, 16364, - 36, 16360, 16116, 3284, 1428, 15856, 172, 16364, - 32, 16376, 16084, 3248, 1496, 15848, 176, 16364, - 28, 4, 16052, 3208, 1564, 15836, 176, 16364, - 24, 16, 16028, 3168, 1632, 15828, 176, 16364, - 20, 28, 16000, 3124, 1700, 15820, 176, 16364, - 16, 40, 15976, 3080, 1768, 15812, 176, 16364, - 12, 52, 15952, 3036, 1836, 15808, 176, 16364, - 8, 64, 15932, 2988, 1904, 15800, 176, 16364, - 4, 76, 15912, 2940, 1972, 15800, 172, 16364, - 4, 84, 15892, 2888, 2040, 15796, 172, 16364, - 0, 96, 15876, 2836, 2104, 15792, 168, 16364, - 16380, 104, 15864, 2780, 2172, 15792, 164, 16364, - 16380, 112, 15848, 2724, 2236, 15792, 160, 16364, - 16376, 120, 15836, 2668, 2300, 15796, 156, 16368, - 16376, 128, 15828, 2608, 2364, 15800, 152, 16368, - 16372, 136, 15816, 2548, 2428, 15804, 148, 16368, - 16372, 140, 15812, 2488, 2488, 15812, 140, 16372 }; +//========================================= +// <num_taps> = 8 +// <num_phases> = 64 +// <scale_ratio> = 1.16666 (input/output) +// <sharpness> = 0 +// <CoefType> = ModifiedLanczos +// <CoefQuant> = 1.10 +// <CoefOut> = 1.12 +//========================================= +static const uint16_t filter_8tap_64p_116[264] = { + 0x0080, 0x3E90, 0x0268, 0x0D14, 0x0264, 0x3E90, 0x0080, 0x0000, + 0x007C, 0x3E9C, 0x0238, 0x0D14, 0x0298, 0x3E84, 0x0080, 0x0000, + 0x0078, 0x3EAC, 0x0200, 0x0D10, 0x02D0, 0x3E78, 0x0084, 0x0000, + 0x0078, 0x3EB8, 0x01D0, 0x0D0C, 0x0304, 0x3E6C, 0x0084, 0x0000, + 0x0074, 0x3EC8, 0x01A0, 0x0D00, 0x033C, 0x3E60, 0x0088, 0x0000, + 0x0070, 0x3ED4, 0x0170, 0x0D00, 0x0374, 0x3E54, 0x0088, 0x3FFC, + 0x006C, 0x3EE4, 0x0140, 0x0CF8, 0x03AC, 0x3E48, 0x0088, 0x3FFC, + 0x006C, 0x3EF0, 0x0114, 0x0CE8, 0x03E4, 0x3E3C, 0x008C, 0x3FFC, + 0x0068, 0x3F00, 0x00E8, 0x0CD8, 0x041C, 0x3E34, 0x008C, 0x3FFC, + 0x0064, 0x3F10, 0x00BC, 0x0CCC, 0x0454, 0x3E28, 0x008C, 0x3FFC, + 0x0060, 0x3F1C, 0x0090, 0x0CBC, 0x0490, 0x3E20, 0x008C, 0x3FFC, + 0x005C, 0x3F2C, 0x0068, 0x0CA4, 0x04CC, 0x3E18, 0x008C, 0x3FFC, + 0x0058, 0x3F38, 0x0040, 0x0C94, 0x0504, 0x3E10, 0x008C, 0x3FFC, + 0x0054, 0x3F48, 0x001C, 0x0C7C, 0x0540, 0x3E08, 0x0088, 0x3FFC, + 0x0050, 0x3F54, 0x3FF8, 0x0C60, 0x057C, 0x3E04, 0x0088, 0x3FFC, + 0x004C, 0x3F64, 0x3FD4, 0x0C44, 0x05B8, 0x3DFC, 0x0088, 0x3FFC, + 0x0048, 0x3F70, 0x3FB4, 0x0C28, 0x05F4, 0x3DF8, 0x0084, 0x3FFC, + 0x0044, 0x3F80, 0x3F90, 0x0C0C, 0x0630, 0x3DF4, 0x0080, 0x3FFC, + 0x0040, 0x3F8C, 0x3F70, 0x0BE8, 0x066C, 0x3DF4, 0x0080, 0x3FFC, + 0x003C, 0x3F9C, 0x3F50, 0x0BC8, 0x06A8, 0x3DF0, 0x007C, 0x3FFC, + 0x0038, 0x3FA8, 0x3F34, 0x0BA0, 0x06E4, 0x3DF0, 0x0078, 0x0000, + 0x0034, 0x3FB4, 0x3F18, 0x0B80, 0x071C, 0x3DF0, 0x0074, 0x0000, + 0x0030, 0x3FC0, 0x3EFC, 0x0B5C, 0x0758, 0x3DF0, 0x0070, 0x0000, + 0x002C, 0x3FCC, 0x3EE4, 0x0B34, 0x0794, 0x3DF4, 0x0068, 0x0000, + 0x002C, 0x3FDC, 0x3ECC, 0x0B08, 0x07CC, 0x3DF4, 0x0064, 0x0000, + 0x0028, 0x3FE4, 0x3EB4, 0x0AE0, 0x0808, 0x3DF8, 0x0060, 0x0000, + 0x0024, 0x3FF0, 0x3EA0, 0x0AB0, 0x0840, 0x3E00, 0x0058, 0x0004, + 0x0020, 0x3FFC, 0x3E90, 0x0A84, 0x0878, 0x3E04, 0x0050, 0x0004, + 0x001C, 0x0004, 0x3E7C, 0x0A54, 0x08B0, 0x3E0C, 0x004C, 0x0008, + 0x0018, 0x000C, 0x3E68, 0x0A28, 0x08E8, 0x3E18, 0x0044, 0x0008, + 0x0018, 0x0018, 0x3E54, 0x09F4, 0x0920, 0x3E20, 0x003C, 0x000C, + 0x0014, 0x0020, 0x3E48, 0x09C0, 0x0954, 0x3E2C, 0x0034, 0x0010, + 0x0010, 0x002C, 0x3E3C, 0x098C, 0x0988, 0x3E38, 0x002C, 0x0010 +}; -static const uint16_t filter_8tap_64p_150[264] = { - 16380, 16020, 1032, 2756, 1032, 16020, 16380, 0, - 0, 16020, 992, 2756, 1068, 16024, 16376, 0, - 4, 16020, 952, 2752, 1108, 16024, 16372, 0, - 8, 16020, 916, 2748, 1148, 16028, 16368, 0, - 12, 16020, 876, 2744, 1184, 16032, 16364, 4, - 16, 16020, 840, 2740, 1224, 16036, 16356, 4, - 20, 16024, 800, 2732, 1264, 16040, 16352, 4, - 20, 16024, 764, 2724, 1304, 16044, 16348, 8, - 24, 16028, 728, 2716, 1344, 16052, 16340, 8, - 28, 16028, 692, 2704, 1380, 16056, 16336, 12, - 28, 16032, 656, 2696, 1420, 16064, 16328, 12, - 32, 16036, 620, 2684, 1460, 16072, 16324, 12, - 36, 16040, 584, 2668, 1500, 16080, 16316, 16, - 36, 16044, 548, 2656, 1536, 16088, 16308, 16, - 36, 16048, 516, 2640, 1576, 16096, 16304, 20, - 40, 16052, 480, 2624, 1612, 16108, 16296, 20, - 40, 16060, 448, 2608, 1652, 16120, 16288, 20, - 44, 16064, 416, 2588, 1692, 16132, 16280, 24, - 44, 16068, 384, 2568, 1728, 16144, 16276, 24, - 44, 16076, 352, 2548, 1764, 16156, 16268, 28, - 44, 16080, 320, 2528, 1804, 16168, 16260, 28, - 44, 16088, 292, 2508, 1840, 16184, 16252, 28, - 44, 16096, 264, 2484, 1876, 16200, 16244, 32, - 48, 16100, 232, 2460, 1912, 16216, 16236, 32, - 48, 16108, 204, 2436, 1948, 16232, 16228, 32, - 48, 16116, 176, 2412, 1980, 16248, 16220, 36, - 48, 16124, 152, 2384, 2016, 16264, 16216, 36, - 44, 16128, 124, 2356, 2052, 16284, 16208, 36, - 44, 16136, 100, 2328, 2084, 16304, 16200, 40, - 44, 16144, 72, 2300, 2116, 16324, 16192, 40, - 44, 16152, 48, 2272, 2148, 16344, 16184, 40, - 44, 16160, 24, 2244, 2180, 16364, 16176, 40, - 44, 16168, 4, 2212, 2212, 4, 16168, 44 }; +//========================================= +// <num_taps> = 8 +// <num_phases> = 64 +// <scale_ratio> = 1.49999 (input/output) +// <sharpness> = 0 +// <CoefType> = ModifiedLanczos +// <CoefQuant> = 1.10 +// <CoefOut> = 1.12 +//========================================= +static const uint16_t filter_8tap_64p_149[264] = { + 0x0008, 0x3E8C, 0x03F8, 0x0AE8, 0x03F8, 0x3E8C, 0x0008, 0x0000, + 0x000C, 0x3E8C, 0x03D0, 0x0AE8, 0x0420, 0x3E90, 0x0000, 0x0000, + 0x000C, 0x3E8C, 0x03AC, 0x0AE8, 0x0444, 0x3E90, 0x0000, 0x0000, + 0x0010, 0x3E90, 0x0384, 0x0AE0, 0x046C, 0x3E94, 0x3FFC, 0x0000, + 0x0014, 0x3E90, 0x035C, 0x0ADC, 0x0494, 0x3E94, 0x3FF8, 0x0004, + 0x0018, 0x3E90, 0x0334, 0x0AD8, 0x04BC, 0x3E98, 0x3FF4, 0x0004, + 0x001C, 0x3E94, 0x0310, 0x0AD0, 0x04E4, 0x3E9C, 0x3FEC, 0x0004, + 0x0020, 0x3E98, 0x02E8, 0x0AC4, 0x050C, 0x3EA0, 0x3FE8, 0x0008, + 0x0020, 0x3E98, 0x02C4, 0x0AC0, 0x0534, 0x3EA4, 0x3FE4, 0x0008, + 0x0024, 0x3E9C, 0x02A0, 0x0AB4, 0x055C, 0x3EAC, 0x3FDC, 0x0008, + 0x0024, 0x3EA0, 0x027C, 0x0AA8, 0x0584, 0x3EB0, 0x3FD8, 0x000C, + 0x0028, 0x3EA4, 0x0258, 0x0A9C, 0x05AC, 0x3EB8, 0x3FD0, 0x000C, + 0x0028, 0x3EA8, 0x0234, 0x0A90, 0x05D4, 0x3EC0, 0x3FC8, 0x0010, + 0x002C, 0x3EAC, 0x0210, 0x0A80, 0x05FC, 0x3EC8, 0x3FC4, 0x0010, + 0x002C, 0x3EB4, 0x01F0, 0x0A70, 0x0624, 0x3ED0, 0x3FBC, 0x0010, + 0x002C, 0x3EB8, 0x01CC, 0x0A60, 0x064C, 0x3EDC, 0x3FB4, 0x0014, + 0x0030, 0x3EBC, 0x01A8, 0x0A50, 0x0674, 0x3EE4, 0x3FB0, 0x0014, + 0x0030, 0x3EC4, 0x0188, 0x0A38, 0x069C, 0x3EF0, 0x3FA8, 0x0018, + 0x0030, 0x3ECC, 0x0168, 0x0A28, 0x06C0, 0x3EFC, 0x3FA0, 0x0018, + 0x0030, 0x3ED0, 0x0148, 0x0A14, 0x06E8, 0x3F08, 0x3F98, 0x001C, + 0x0030, 0x3ED8, 0x012C, 0x0A00, 0x070C, 0x3F14, 0x3F90, 0x001C, + 0x0034, 0x3EE0, 0x0108, 0x09E4, 0x0734, 0x3F24, 0x3F8C, 0x001C, + 0x0034, 0x3EE4, 0x00EC, 0x09CC, 0x0758, 0x3F34, 0x3F84, 0x0020, + 0x0034, 0x3EEC, 0x00D0, 0x09B8, 0x077C, 0x3F40, 0x3F7C, 0x0020, + 0x0034, 0x3EF4, 0x00B4, 0x0998, 0x07A4, 0x3F50, 0x3F74, 0x0024, + 0x0030, 0x3EFC, 0x0098, 0x0980, 0x07C8, 0x3F64, 0x3F6C, 0x0024, + 0x0030, 0x3F04, 0x0080, 0x0968, 0x07E8, 0x3F74, 0x3F64, 0x0024, + 0x0030, 0x3F0C, 0x0060, 0x094C, 0x080C, 0x3F88, 0x3F5C, 0x0028, + 0x0030, 0x3F14, 0x0048, 0x0930, 0x0830, 0x3F98, 0x3F54, 0x0028, + 0x0030, 0x3F1C, 0x0030, 0x0914, 0x0850, 0x3FAC, 0x3F4C, 0x0028, + 0x0030, 0x3F24, 0x0018, 0x08F0, 0x0874, 0x3FC0, 0x3F44, 0x002C, + 0x002C, 0x3F2C, 0x0000, 0x08D4, 0x0894, 0x3FD8, 0x3F3C, 0x002C, + 0x002C, 0x3F34, 0x3FEC, 0x08B4, 0x08B4, 0x3FEC, 0x3F34, 0x002C +}; +//========================================= +// <num_taps> = 8 +// <num_phases> = 64 +// <scale_ratio> = 1.83332 (input/output) +// <sharpness> = 0 +// <CoefType> = ModifiedLanczos +// <CoefQuant> = 1.10 +// <CoefOut> = 1.12 +//========================================= static const uint16_t filter_8tap_64p_183[264] = { - 16264, 16264, 1164, 2244, 1164, 16264, 16264, 0, - 16268, 16256, 1136, 2240, 1188, 16272, 16260, 0, - 16272, 16248, 1108, 2240, 1216, 16280, 16256, 0, - 16276, 16240, 1080, 2236, 1240, 16292, 16252, 0, - 16280, 16232, 1056, 2236, 1268, 16300, 16248, 0, - 16284, 16224, 1028, 2232, 1292, 16312, 16244, 0, - 16288, 16216, 1000, 2228, 1320, 16324, 16240, 0, - 16292, 16212, 976, 2224, 1344, 16336, 16236, 0, - 16296, 16204, 948, 2220, 1372, 16348, 16232, 0, - 16300, 16200, 920, 2212, 1396, 16360, 16228, 4, - 16304, 16196, 896, 2204, 1424, 16372, 16224, 4, - 16308, 16188, 868, 2200, 1448, 0, 16220, 4, - 16312, 16184, 844, 2192, 1472, 12, 16216, 4, - 16316, 16180, 816, 2184, 1500, 28, 16212, 4, - 16320, 16176, 792, 2172, 1524, 40, 16208, 4, - 16324, 16172, 764, 2164, 1548, 56, 16204, 0, - 16328, 16172, 740, 2156, 1572, 72, 16200, 0, - 16328, 16168, 712, 2144, 1596, 88, 16196, 0, - 16332, 16164, 688, 2132, 1620, 100, 16192, 0, - 16336, 16164, 664, 2120, 1644, 120, 16192, 0, - 16340, 16160, 640, 2108, 1668, 136, 16188, 0, - 16344, 16160, 616, 2096, 1688, 152, 16184, 0, - 16344, 16160, 592, 2080, 1712, 168, 16180, 0, - 16348, 16156, 568, 2068, 1736, 188, 16176, 16380, - 16352, 16156, 544, 2052, 1756, 204, 16176, 16380, - 16352, 16156, 520, 2036, 1780, 224, 16172, 16380, - 16356, 16156, 496, 2024, 1800, 244, 16172, 16380, - 16360, 16156, 472, 2008, 1820, 260, 16168, 16376, - 16360, 16156, 452, 1988, 1840, 280, 16164, 16376, - 16364, 16156, 428, 1972, 1860, 300, 16164, 16376, - 16364, 16156, 408, 1956, 1880, 320, 16164, 16372, - 16368, 16160, 384, 1936, 1900, 344, 16160, 16372, - 16368, 16160, 364, 1920, 1920, 364, 16160, 16368 }; + 0x3F88, 0x0048, 0x047C, 0x0768, 0x047C, 0x0048, 0x3F88, 0x0000, + 0x3F88, 0x003C, 0x0468, 0x076C, 0x0490, 0x0054, 0x3F84, 0x0000, + 0x3F8C, 0x0034, 0x0454, 0x0768, 0x04A4, 0x005C, 0x3F84, 0x0000, + 0x3F8C, 0x0028, 0x0444, 0x076C, 0x04B4, 0x0068, 0x3F80, 0x0000, + 0x3F90, 0x0020, 0x042C, 0x0768, 0x04C8, 0x0074, 0x3F80, 0x0000, + 0x3F90, 0x0018, 0x041C, 0x0764, 0x04DC, 0x0080, 0x3F7C, 0x0000, + 0x3F94, 0x0010, 0x0408, 0x075C, 0x04F0, 0x008C, 0x3F7C, 0x0000, + 0x3F94, 0x0004, 0x03F8, 0x0760, 0x0500, 0x0098, 0x3F7C, 0x3FFC, + 0x3F98, 0x0000, 0x03E0, 0x075C, 0x0514, 0x00A4, 0x3F78, 0x3FFC, + 0x3F9C, 0x3FF8, 0x03CC, 0x0754, 0x0528, 0x00B0, 0x3F78, 0x3FFC, + 0x3F9C, 0x3FF0, 0x03B8, 0x0754, 0x0538, 0x00BC, 0x3F78, 0x3FFC, + 0x3FA0, 0x3FE8, 0x03A4, 0x0750, 0x054C, 0x00CC, 0x3F74, 0x3FF8, + 0x3FA4, 0x3FE0, 0x0390, 0x074C, 0x055C, 0x00D8, 0x3F74, 0x3FF8, + 0x3FA4, 0x3FDC, 0x037C, 0x0744, 0x0570, 0x00E4, 0x3F74, 0x3FF8, + 0x3FA8, 0x3FD4, 0x0368, 0x0740, 0x0580, 0x00F4, 0x3F74, 0x3FF4, + 0x3FA8, 0x3FCC, 0x0354, 0x073C, 0x0590, 0x0104, 0x3F74, 0x3FF4, + 0x3FAC, 0x3FC8, 0x0340, 0x0730, 0x05A4, 0x0110, 0x3F74, 0x3FF4, + 0x3FB0, 0x3FC0, 0x0330, 0x0728, 0x05B4, 0x0120, 0x3F74, 0x3FF0, + 0x3FB0, 0x3FBC, 0x031C, 0x0724, 0x05C4, 0x0130, 0x3F70, 0x3FF0, + 0x3FB4, 0x3FB4, 0x0308, 0x0720, 0x05D4, 0x013C, 0x3F70, 0x3FF0, + 0x3FB8, 0x3FB0, 0x02F4, 0x0714, 0x05E4, 0x014C, 0x3F74, 0x3FEC, + 0x3FB8, 0x3FAC, 0x02E0, 0x0708, 0x05F8, 0x015C, 0x3F74, 0x3FEC, + 0x3FBC, 0x3FA8, 0x02CC, 0x0704, 0x0604, 0x016C, 0x3F74, 0x3FE8, + 0x3FC0, 0x3FA0, 0x02BC, 0x06F8, 0x0614, 0x017C, 0x3F74, 0x3FE8, + 0x3FC0, 0x3F9C, 0x02A8, 0x06F4, 0x0624, 0x018C, 0x3F74, 0x3FE4, + 0x3FC4, 0x3F98, 0x0294, 0x06E8, 0x0634, 0x019C, 0x3F74, 0x3FE4, + 0x3FC8, 0x3F94, 0x0284, 0x06D8, 0x0644, 0x01AC, 0x3F78, 0x3FE0, + 0x3FC8, 0x3F90, 0x0270, 0x06D4, 0x0650, 0x01BC, 0x3F78, 0x3FE0, + 0x3FCC, 0x3F8C, 0x025C, 0x06C8, 0x0660, 0x01D0, 0x3F78, 0x3FDC, + 0x3FCC, 0x3F8C, 0x024C, 0x06B8, 0x066C, 0x01E0, 0x3F7C, 0x3FDC, + 0x3FD0, 0x3F88, 0x0238, 0x06B0, 0x067C, 0x01F0, 0x3F7C, 0x3FD8, + 0x3FD4, 0x3F84, 0x0228, 0x069C, 0x0688, 0x0204, 0x3F80, 0x3FD8, + 0x3FD4, 0x3F84, 0x0214, 0x0694, 0x0694, 0x0214, 0x3F84, 0x3FD4 +}; const uint16_t *get_filter_3tap_16p(struct fixed31_32 ratio) { if (ratio.value < dc_fixpt_one.value) return filter_3tap_16p_upscale; else if (ratio.value < dc_fixpt_from_fraction(4, 3).value) - return filter_3tap_16p_117; + return filter_3tap_16p_116; else if (ratio.value < dc_fixpt_from_fraction(5, 3).value) - return filter_3tap_16p_150; + return filter_3tap_16p_149; else return filter_3tap_16p_183; } @@ -1029,9 +1355,9 @@ const uint16_t *get_filter_3tap_64p(struct fixed31_32 ratio) if (ratio.value < dc_fixpt_one.value) return filter_3tap_64p_upscale; else if (ratio.value < dc_fixpt_from_fraction(4, 3).value) - return filter_3tap_64p_117; + return filter_3tap_64p_116; else if (ratio.value < dc_fixpt_from_fraction(5, 3).value) - return filter_3tap_64p_150; + return filter_3tap_64p_149; else return filter_3tap_64p_183; } @@ -1041,9 +1367,9 @@ const uint16_t *get_filter_4tap_16p(struct fixed31_32 ratio) if (ratio.value < dc_fixpt_one.value) return filter_4tap_16p_upscale; else if (ratio.value < dc_fixpt_from_fraction(4, 3).value) - return filter_4tap_16p_117; + return filter_4tap_16p_116; else if (ratio.value < dc_fixpt_from_fraction(5, 3).value) - return filter_4tap_16p_150; + return filter_4tap_16p_149; else return filter_4tap_16p_183; } @@ -1053,9 +1379,9 @@ const uint16_t *get_filter_4tap_64p(struct fixed31_32 ratio) if (ratio.value < dc_fixpt_one.value) return filter_4tap_64p_upscale; else if (ratio.value < dc_fixpt_from_fraction(4, 3).value) - return filter_4tap_64p_117; + return filter_4tap_64p_116; else if (ratio.value < dc_fixpt_from_fraction(5, 3).value) - return filter_4tap_64p_150; + return filter_4tap_64p_149; else return filter_4tap_64p_183; } @@ -1065,9 +1391,9 @@ const uint16_t *get_filter_5tap_64p(struct fixed31_32 ratio) if (ratio.value < dc_fixpt_one.value) return filter_5tap_64p_upscale; else if (ratio.value < dc_fixpt_from_fraction(4, 3).value) - return filter_5tap_64p_117; + return filter_5tap_64p_116; else if (ratio.value < dc_fixpt_from_fraction(5, 3).value) - return filter_5tap_64p_150; + return filter_5tap_64p_149; else return filter_5tap_64p_183; } @@ -1077,9 +1403,9 @@ const uint16_t *get_filter_6tap_64p(struct fixed31_32 ratio) if (ratio.value < dc_fixpt_one.value) return filter_6tap_64p_upscale; else if (ratio.value < dc_fixpt_from_fraction(4, 3).value) - return filter_6tap_64p_117; + return filter_6tap_64p_116; else if (ratio.value < dc_fixpt_from_fraction(5, 3).value) - return filter_6tap_64p_150; + return filter_6tap_64p_149; else return filter_6tap_64p_183; } @@ -1089,9 +1415,9 @@ const uint16_t *get_filter_7tap_64p(struct fixed31_32 ratio) if (ratio.value < dc_fixpt_one.value) return filter_7tap_64p_upscale; else if (ratio.value < dc_fixpt_from_fraction(4, 3).value) - return filter_7tap_64p_117; + return filter_7tap_64p_116; else if (ratio.value < dc_fixpt_from_fraction(5, 3).value) - return filter_7tap_64p_150; + return filter_7tap_64p_149; else return filter_7tap_64p_183; } @@ -1101,9 +1427,9 @@ const uint16_t *get_filter_8tap_64p(struct fixed31_32 ratio) if (ratio.value < dc_fixpt_one.value) return filter_8tap_64p_upscale; else if (ratio.value < dc_fixpt_from_fraction(4, 3).value) - return filter_8tap_64p_117; + return filter_8tap_64p_116; else if (ratio.value < dc_fixpt_from_fraction(5, 3).value) - return filter_8tap_64p_150; + return filter_8tap_64p_149; else return filter_8tap_64p_183; } diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_scl_filters_old.c b/drivers/gpu/drm/amd/display/dc/dce/dce_scl_filters_old.c new file mode 100644 index 000000000000..bb0e1b80ec3c --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_scl_filters_old.c @@ -0,0 +1,25 @@ +/* + * Copyright 2012-16 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c index 225955ec6d39..bc109d4fc6e6 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c @@ -27,25 +27,55 @@ #include "dc.h" #include "dc_dmub_srv.h" #include "../../dmub/inc/dmub_srv.h" -#include "dmub_fw_state.h" +#include "../../dmub/inc/dmub_gpint_cmd.h" #include "core_types.h" -#include "ipp.h" #define MAX_PIPES 6 /** * Get PSR state from firmware. */ -static void dmub_get_psr_state(uint32_t *psr_state) +static void dmub_psr_get_state(struct dmub_psr *dmub, uint32_t *psr_state) { - // Not yet implemented - // Trigger GPINT interrupt from firmware + struct dmub_srv *srv = dmub->ctx->dmub_srv->dmub; + + // Send gpint command and wait for ack + dmub_srv_send_gpint_command(srv, DMUB_GPINT__GET_PSR_STATE, 0, 30); + + dmub_srv_get_gpint_response(srv, psr_state); +} + +/** + * Set PSR version. + */ +static bool dmub_psr_set_version(struct dmub_psr *dmub, struct dc_stream_state *stream) +{ + union dmub_rb_cmd cmd; + struct dc_context *dc = dmub->ctx; + + cmd.psr_set_version.header.type = DMUB_CMD__PSR; + cmd.psr_set_version.header.sub_type = DMUB_CMD__PSR_SET_VERSION; + + if (stream->psr_version == 0x0) // Unsupported + return false; + else if (stream->psr_version == 0x1) + cmd.psr_set_version.psr_set_version_data.version = PSR_VERSION_1; + else if (stream->psr_version == 0x2) + cmd.psr_set_version.psr_set_version_data.version = PSR_VERSION_2; + + cmd.psr_enable.header.payload_bytes = sizeof(struct dmub_cmd_psr_set_version_data); + + dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd.psr_enable.header); + dc_dmub_srv_cmd_execute(dc->dmub_srv); + dc_dmub_srv_wait_idle(dc->dmub_srv); + + return true; } /** * Enable/Disable PSR. */ -static void dmub_set_psr_enable(struct dmub_psr *dmub, bool enable) +static void dmub_psr_enable(struct dmub_psr *dmub, bool enable) { union dmub_rb_cmd cmd; struct dc_context *dc = dmub->ctx; @@ -67,13 +97,13 @@ static void dmub_set_psr_enable(struct dmub_psr *dmub, bool enable) /** * Set PSR level. */ -static void dmub_set_psr_level(struct dmub_psr *dmub, uint16_t psr_level) +static void dmub_psr_set_level(struct dmub_psr *dmub, uint16_t psr_level) { union dmub_rb_cmd cmd; uint32_t psr_state = 0; struct dc_context *dc = dmub->ctx; - dmub_get_psr_state(&psr_state); + dmub_psr_get_state(dmub, &psr_state); if (psr_state == 0) return; @@ -91,7 +121,7 @@ static void dmub_set_psr_level(struct dmub_psr *dmub, uint16_t psr_level) /** * Setup PSR by programming phy registers and sending psr hw context values to firmware. */ -static bool dmub_setup_psr(struct dmub_psr *dmub, +static bool dmub_psr_copy_settings(struct dmub_psr *dmub, struct dc_link *link, struct psr_context *psr_context) { @@ -101,21 +131,22 @@ static bool dmub_setup_psr(struct dmub_psr *dmub, = &cmd.psr_copy_settings.psr_copy_settings_data; struct pipe_ctx *pipe_ctx = NULL; struct resource_context *res_ctx = &link->ctx->dc->current_state->res_ctx; + int i = 0; - for (int i = 0; i < MAX_PIPES; i++) { - if (res_ctx && - res_ctx->pipe_ctx[i].stream && - res_ctx->pipe_ctx[i].stream->link && - res_ctx->pipe_ctx[i].stream->link == link && - res_ctx->pipe_ctx[i].stream->link->connector_signal == SIGNAL_TYPE_EDP) { + for (i = 0; i < MAX_PIPES; i++) { + if (res_ctx->pipe_ctx[i].stream && + res_ctx->pipe_ctx[i].stream->link == link && + res_ctx->pipe_ctx[i].stream->link->connector_signal == SIGNAL_TYPE_EDP) { pipe_ctx = &res_ctx->pipe_ctx[i]; break; } } - if (!pipe_ctx || - !&pipe_ctx->plane_res || - !&pipe_ctx->stream_res) + if (!pipe_ctx) + return false; + + // First, set the psr version + if (!dmub_psr_set_version(dmub, pipe_ctx->stream)) return false; // Program DP DPHY fast training registers @@ -138,10 +169,6 @@ static bool dmub_setup_psr(struct dmub_psr *dmub, copy_settings_data->mpcc_inst = pipe_ctx->plane_res.mpcc_inst; - if (pipe_ctx->plane_res.hubp) - copy_settings_data->hubp_inst = pipe_ctx->plane_res.hubp->inst; - else - copy_settings_data->hubp_inst = 0; if (pipe_ctx->plane_res.dpp) copy_settings_data->dpp_inst = pipe_ctx->plane_res.dpp->inst; else @@ -157,18 +184,9 @@ static bool dmub_setup_psr(struct dmub_psr *dmub, // Misc copy_settings_data->psr_level = psr_context->psr_level.u32all; - copy_settings_data->hyst_frames = psr_context->timehyst_frames; - copy_settings_data->hyst_lines = psr_context->hyst_lines; - copy_settings_data->phy_type = psr_context->phyType; - copy_settings_data->aux_repeat = psr_context->aux_repeats; - copy_settings_data->smu_optimizations_en = psr_context->allow_smu_optimizations; - copy_settings_data->skip_wait_for_pll_lock = psr_context->skipPsrWaitForPllLock; + copy_settings_data->smu_optimizations_en = psr_context->allow_smu_optimizations; copy_settings_data->frame_delay = psr_context->frame_delay; - copy_settings_data->smu_phy_id = psr_context->smuPhyId; - copy_settings_data->num_of_controllers = psr_context->numberOfControllers; copy_settings_data->frame_cap_ind = psr_context->psrFrameCaptureIndicationReq; - copy_settings_data->phy_num = psr_context->frame_delay & 0x7; - copy_settings_data->link_rate = psr_context->frame_delay & 0xF; dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd.psr_copy_settings.header); dc_dmub_srv_cmd_execute(dc->dmub_srv); @@ -178,10 +196,10 @@ static bool dmub_setup_psr(struct dmub_psr *dmub, } static const struct dmub_psr_funcs psr_funcs = { - .set_psr_enable = dmub_set_psr_enable, - .setup_psr = dmub_setup_psr, - .get_psr_state = dmub_get_psr_state, - .set_psr_level = dmub_set_psr_level, + .psr_copy_settings = dmub_psr_copy_settings, + .psr_enable = dmub_psr_enable, + .psr_get_state = dmub_psr_get_state, + .psr_set_level = dmub_psr_set_level, }; /** @@ -215,6 +233,6 @@ struct dmub_psr *dmub_psr_create(struct dc_context *ctx) */ void dmub_psr_destroy(struct dmub_psr **dmub) { - kfree(dmub); + kfree(*dmub); *dmub = NULL; } diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.h b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.h index 229958de3035..f404fecd6410 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.h +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.h @@ -27,6 +27,7 @@ #define _DMUB_PSR_H_ #include "os_types.h" +#include "dc_link.h" struct dmub_psr { struct dc_context *ctx; @@ -34,14 +35,14 @@ struct dmub_psr { }; struct dmub_psr_funcs { - void (*set_psr_enable)(struct dmub_psr *dmub, bool enable); - bool (*setup_psr)(struct dmub_psr *dmub, struct dc_link *link, struct psr_context *psr_context); - void (*get_psr_state)(uint32_t *psr_state); - void (*set_psr_level)(struct dmub_psr *dmub, uint16_t psr_level); + bool (*psr_copy_settings)(struct dmub_psr *dmub, struct dc_link *link, struct psr_context *psr_context); + void (*psr_enable)(struct dmub_psr *dmub, bool enable); + void (*psr_get_state)(struct dmub_psr *dmub, uint32_t *psr_state); + void (*psr_set_level)(struct dmub_psr *dmub, uint16_t psr_level); }; struct dmub_psr *dmub_psr_create(struct dc_context *ctx); void dmub_psr_destroy(struct dmub_psr **dmub); -#endif /* _DCE_DMUB_H_ */ +#endif /* _DMUB_PSR_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c index 5b689273ff44..c279982947e1 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c @@ -71,6 +71,8 @@ #define PANEL_POWER_UP_TIMEOUT 300 #define PANEL_POWER_DOWN_TIMEOUT 500 #define HPD_CHECK_INTERVAL 10 +#define OLED_POST_T7_DELAY 100 +#define OLED_PRE_T11_DELAY 150 #define CTX \ hws->ctx @@ -696,8 +698,10 @@ void dce110_enable_stream(struct pipe_ctx *pipe_ctx) } /*todo: cloned in stream enc, fix*/ -static bool is_panel_backlight_on(struct dce_hwseq *hws) +bool dce110_is_panel_backlight_on(struct dc_link *link) { + struct dc_context *ctx = link->ctx; + struct dce_hwseq *hws = ctx->dc->hwseq; uint32_t value; REG_GET(LVTMA_PWRSEQ_CNTL, LVTMA_BLON, &value); @@ -705,11 +709,12 @@ static bool is_panel_backlight_on(struct dce_hwseq *hws) return value; } -static bool is_panel_powered_on(struct dce_hwseq *hws) +bool dce110_is_panel_powered_on(struct dc_link *link) { + struct dc_context *ctx = link->ctx; + struct dce_hwseq *hws = ctx->dc->hwseq; uint32_t pwr_seq_state, dig_on, dig_on_ovrd; - REG_GET(LVTMA_PWRSEQ_STATE, LVTMA_PWRSEQ_TARGET_STATE_R, &pwr_seq_state); REG_GET_2(LVTMA_PWRSEQ_CNTL, LVTMA_DIGON, &dig_on, LVTMA_DIGON_OVRD, &dig_on_ovrd); @@ -816,7 +821,7 @@ void dce110_edp_power_control( return; } - if (power_up != is_panel_powered_on(hwseq)) { + if (power_up != hwseq->funcs.is_panel_powered_on(link)) { /* Send VBIOS command to prompt eDP panel power */ if (power_up) { unsigned long long current_ts = dm_get_timestamp(ctx); @@ -896,7 +901,7 @@ void dce110_edp_backlight_control( return; } - if (enable && is_panel_backlight_on(hws)) { + if (enable && hws->funcs.is_panel_backlight_on(link)) { DC_LOG_HW_RESUME_S3( "%s: panel already powered up. Do nothing.\n", __func__); @@ -936,9 +941,21 @@ void dce110_edp_backlight_control( if (cntl.action == TRANSMITTER_CONTROL_BACKLIGHT_ON) edp_receiver_ready_T7(link); link_transmitter_control(ctx->dc_bios, &cntl); + + if (enable && link->dpcd_sink_ext_caps.bits.oled) + msleep(OLED_POST_T7_DELAY); + + if (link->dpcd_sink_ext_caps.bits.oled || + link->dpcd_sink_ext_caps.bits.hdr_aux_backlight_control == 1 || + link->dpcd_sink_ext_caps.bits.sdr_aux_backlight_control == 1) + dc_link_backlight_enable_aux(link, enable); + /*edp 1.2*/ if (cntl.action == TRANSMITTER_CONTROL_BACKLIGHT_OFF) edp_receiver_ready_T9(link); + + if (!enable && link->dpcd_sink_ext_caps.bits.oled) + msleep(OLED_PRE_T11_DELAY); } void dce110_enable_audio_stream(struct pipe_ctx *pipe_ctx) @@ -2576,17 +2593,6 @@ static void dce110_apply_ctx_for_surface( for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; - struct pipe_ctx *old_pipe_ctx = &dc->current_state->res_ctx.pipe_ctx[i]; - - if (stream == pipe_ctx->stream) { - if (!pipe_ctx->top_pipe && - (pipe_ctx->plane_state || old_pipe_ctx->plane_state)) - dc->hwss.pipe_control_lock(dc, pipe_ctx, true); - } - } - - for (i = 0; i < dc->res_pool->pipe_count; i++) { - struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; if (pipe_ctx->stream != stream) continue; @@ -2607,20 +2613,16 @@ static void dce110_apply_ctx_for_surface( } - for (i = 0; i < dc->res_pool->pipe_count; i++) { - struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; - struct pipe_ctx *old_pipe_ctx = &dc->current_state->res_ctx.pipe_ctx[i]; - - if ((stream == pipe_ctx->stream) && - (!pipe_ctx->top_pipe) && - (pipe_ctx->plane_state || old_pipe_ctx->plane_state)) - dc->hwss.pipe_control_lock(dc, pipe_ctx, false); - } - if (dc->fbc_compressor) enable_fbc(dc, context); } +static void dce110_post_unlock_program_front_end( + struct dc *dc, + struct dc_state *context) +{ +} + static void dce110_power_down_fe(struct dc *dc, struct pipe_ctx *pipe_ctx) { struct dce_hwseq *hws = dc->hwseq; @@ -2683,6 +2685,23 @@ void dce110_set_cursor_position(struct pipe_ctx *pipe_ctx) .mirror = pipe_ctx->plane_state->horizontal_mirror }; + /** + * If the cursor's source viewport is clipped then we need to + * translate the cursor to appear in the correct position on + * the screen. + * + * This translation isn't affected by scaling so it needs to be + * done *after* we adjust the position for the scale factor. + * + * This is only done by opt-in for now since there are still + * some usecases like tiled display that might enable the + * cursor on both streams while expecting dc to clip it. + */ + if (pos_cpy.translate_by_source) { + pos_cpy.x += pipe_ctx->plane_state->src_rect.x; + pos_cpy.y += pipe_ctx->plane_state->src_rect.y; + } + if (pipe_ctx->plane_state->address.type == PLN_ADDR_TYPE_VIDEO_PROGRESSIVE) pos_cpy.enable = false; @@ -2722,6 +2741,7 @@ static const struct hw_sequencer_funcs dce110_funcs = { .init_hw = init_hw, .apply_ctx_to_hw = dce110_apply_ctx_to_hw, .apply_ctx_for_surface = dce110_apply_ctx_for_surface, + .post_unlock_program_front_end = dce110_post_unlock_program_front_end, .update_plane_addr = update_plane_addr, .update_pending_status = dce110_update_pending_status, .enable_accelerated_mode = dce110_enable_accelerated_mode, @@ -2736,6 +2756,7 @@ static const struct hw_sequencer_funcs dce110_funcs = { .disable_audio_stream = dce110_disable_audio_stream, .disable_plane = dce110_power_down_fe, .pipe_control_lock = dce_pipe_control_lock, + .interdependent_update_lock = NULL, .prepare_bandwidth = dce110_prepare_bandwidth, .optimize_bandwidth = dce110_optimize_bandwidth, .set_drr = set_drr, @@ -2763,6 +2784,8 @@ static const struct hwseq_private_funcs dce110_private_funcs = { .disable_stream_gating = NULL, .enable_stream_gating = NULL, .edp_backlight_control = dce110_edp_backlight_control, + .is_panel_backlight_on = dce110_is_panel_backlight_on, + .is_panel_powered_on = dce110_is_panel_powered_on, }; void dce110_hw_sequencer_construct(struct dc *dc) diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.h index 26a9c14a58b1..34be166e8ff0 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.h +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.h @@ -85,5 +85,9 @@ void dce110_edp_wait_for_hpd_ready( struct dc_link *link, bool power_up); +bool dce110_is_panel_backlight_on(struct dc_link *link); + +bool dce110_is_panel_powered_on(struct dc_link *link); + #endif /* __DC_HWSS_DCE110_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c index bbd6e01b3eca..47a39eb9400b 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c @@ -316,6 +316,7 @@ bool cm_helper_translate_curve_to_hw_format( struct pwl_result_data *rgb_resulted; struct pwl_result_data *rgb; struct pwl_result_data *rgb_plus_1; + struct pwl_result_data *rgb_minus_1; int32_t region_start, region_end; int32_t i; @@ -465,9 +466,20 @@ bool cm_helper_translate_curve_to_hw_format( rgb = rgb_resulted; rgb_plus_1 = rgb_resulted + 1; + rgb_minus_1 = rgb; i = 1; while (i != hw_points + 1) { + + if (i >= hw_points - 1) { + if (dc_fixpt_lt(rgb_plus_1->red, rgb->red)) + rgb_plus_1->red = dc_fixpt_add(rgb->red, rgb_minus_1->delta_red); + if (dc_fixpt_lt(rgb_plus_1->green, rgb->green)) + rgb_plus_1->green = dc_fixpt_add(rgb->green, rgb_minus_1->delta_green); + if (dc_fixpt_lt(rgb_plus_1->blue, rgb->blue)) + rgb_plus_1->blue = dc_fixpt_add(rgb->blue, rgb_minus_1->delta_blue); + } + rgb->delta_red = dc_fixpt_sub(rgb_plus_1->red, rgb->red); rgb->delta_green = dc_fixpt_sub(rgb_plus_1->green, rgb->green); rgb->delta_blue = dc_fixpt_sub(rgb_plus_1->blue, rgb->blue); @@ -482,6 +494,7 @@ bool cm_helper_translate_curve_to_hw_format( } ++rgb_plus_1; + rgb_minus_1 = rgb; ++rgb; ++i; } diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c index f36a0d8cedfe..deccab0228d2 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c @@ -128,8 +128,8 @@ bool hubbub1_verify_allow_pstate_change_high( * pstate takes around ~100us on linux. Unknown currently as to * why it takes that long on linux */ - static unsigned int pstate_wait_timeout_us = 200; - static unsigned int pstate_wait_expected_timeout_us = 40; + const unsigned int pstate_wait_timeout_us = 200; + const unsigned int pstate_wait_expected_timeout_us = 40; static unsigned int max_sampled_pstate_wait_us; /* data collection */ static bool forced_pstate_allow; /* help with revert wa */ @@ -147,8 +147,9 @@ bool hubbub1_verify_allow_pstate_change_high( forced_pstate_allow = false; } - /* RV2: - * dchubbubdebugind, at: 0xB + /* The following table only applies to DCN1 and DCN2, + * for newer DCNs, need to consult with HW IP folks to read RTL + * HUBBUB:DCHUBBUB_TEST_ARB_DEBUG10 DCHUBBUBDEBUGIND:0xB * description * 0: Pipe0 Plane0 Allow Pstate Change * 1: Pipe0 Plane1 Allow Pstate Change @@ -181,64 +182,6 @@ bool hubbub1_verify_allow_pstate_change_high( * 28: WB0 Allow Pstate Change * 29: WB1 Allow Pstate Change * 30: Arbiter's allow_pstate_change - * 31: SOC pstate change request" - */ - /*DCN2.x: - HUBBUB:DCHUBBUB_TEST_ARB_DEBUG10 DCHUBBUBDEBUGIND:0xB - 0: Pipe0 Plane0 Allow P-state Change - 1: Pipe0 Plane1 Allow P-state Change - 2: Pipe0 Cursor0 Allow P-state Change - 3: Pipe0 Cursor1 Allow P-state Change - 4: Pipe1 Plane0 Allow P-state Change - 5: Pipe1 Plane1 Allow P-state Change - 6: Pipe1 Cursor0 Allow P-state Change - 7: Pipe1 Cursor1 Allow P-state Change - 8: Pipe2 Plane0 Allow P-state Change - 9: Pipe2 Plane1 Allow P-state Change - 10: Pipe2 Cursor0 Allow P-state Change - 11: Pipe2 Cursor1 Allow P-state Change - 12: Pipe3 Plane0 Allow P-state Change - 13: Pipe3 Plane1 Allow P-state Change - 14: Pipe3 Cursor0 Allow P-state Change - 15: Pipe3 Cursor1 Allow P-state Change - 16: Pipe4 Plane0 Allow P-state Change - 17: Pipe4 Plane1 Allow P-state Change - 18: Pipe4 Cursor0 Allow P-state Change - 19: Pipe4 Cursor1 Allow P-state Change - 20: Pipe5 Plane0 Allow P-state Change - 21: Pipe5 Plane1 Allow P-state Change - 22: Pipe5 Cursor0 Allow P-state Change - 23: Pipe5 Cursor1 Allow P-state Change - 24: Pipe6 Plane0 Allow P-state Change - 25: Pipe6 Plane1 Allow P-state Change - 26: Pipe6 Cursor0 Allow P-state Change - 27: Pipe6 Cursor1 Allow P-state Change - 28: WB0 Allow P-state Change - 29: WB1 Allow P-state Change - 30: Arbiter`s Allow P-state Change - 31: SOC P-state Change request - */ - /* RV1: - * dchubbubdebugind, at: 0x7 - * description "3-0: Pipe0 cursor0 QOS - * 7-4: Pipe1 cursor0 QOS - * 11-8: Pipe2 cursor0 QOS - * 15-12: Pipe3 cursor0 QOS - * 16: Pipe0 Plane0 Allow Pstate Change - * 17: Pipe1 Plane0 Allow Pstate Change - * 18: Pipe2 Plane0 Allow Pstate Change - * 19: Pipe3 Plane0 Allow Pstate Change - * 20: Pipe0 Plane1 Allow Pstate Change - * 21: Pipe1 Plane1 Allow Pstate Change - * 22: Pipe2 Plane1 Allow Pstate Change - * 23: Pipe3 Plane1 Allow Pstate Change - * 24: Pipe0 cursor0 Allow Pstate Change - * 25: Pipe1 cursor0 Allow Pstate Change - * 26: Pipe2 cursor0 Allow Pstate Change - * 27: Pipe3 cursor0 Allow Pstate Change - * 28: WB0 Allow Pstate Change - * 29: WB1 Allow Pstate Change - * 30: Arbiter's allow_pstate_change * 31: SOC pstate change request */ @@ -300,7 +243,7 @@ void hubbub1_wm_change_req_wa(struct hubbub *hubbub) DCHUBBUB_ARB_WATERMARK_CHANGE_REQUEST, 1); } -void hubbub1_program_urgent_watermarks( +bool hubbub1_program_urgent_watermarks( struct hubbub *hubbub, struct dcn_watermark_set *watermarks, unsigned int refclk_mhz, @@ -308,6 +251,7 @@ void hubbub1_program_urgent_watermarks( { struct dcn10_hubbub *hubbub1 = TO_DCN10_HUBBUB(hubbub); uint32_t prog_wm_value; + bool wm_pending = false; /* Repeat for water mark set A, B, C and D. */ /* clock state A */ @@ -321,7 +265,8 @@ void hubbub1_program_urgent_watermarks( DC_LOG_BANDWIDTH_CALCS("URGENCY_WATERMARK_A calculated =%d\n" "HW register value = 0x%x\n", watermarks->a.urgent_ns, prog_wm_value); - } + } else if (watermarks->a.urgent_ns < hubbub1->watermarks.a.urgent_ns) + wm_pending = true; if (safe_to_lower || watermarks->a.pte_meta_urgent_ns > hubbub1->watermarks.a.pte_meta_urgent_ns) { hubbub1->watermarks.a.pte_meta_urgent_ns = watermarks->a.pte_meta_urgent_ns; @@ -331,7 +276,8 @@ void hubbub1_program_urgent_watermarks( DC_LOG_BANDWIDTH_CALCS("PTE_META_URGENCY_WATERMARK_A calculated =%d\n" "HW register value = 0x%x\n", watermarks->a.pte_meta_urgent_ns, prog_wm_value); - } + } else if (watermarks->a.pte_meta_urgent_ns < hubbub1->watermarks.a.pte_meta_urgent_ns) + wm_pending = true; /* clock state B */ if (safe_to_lower || watermarks->b.urgent_ns > hubbub1->watermarks.b.urgent_ns) { @@ -344,7 +290,8 @@ void hubbub1_program_urgent_watermarks( DC_LOG_BANDWIDTH_CALCS("URGENCY_WATERMARK_B calculated =%d\n" "HW register value = 0x%x\n", watermarks->b.urgent_ns, prog_wm_value); - } + } else if (watermarks->b.urgent_ns < hubbub1->watermarks.b.urgent_ns) + wm_pending = true; if (safe_to_lower || watermarks->b.pte_meta_urgent_ns > hubbub1->watermarks.b.pte_meta_urgent_ns) { hubbub1->watermarks.b.pte_meta_urgent_ns = watermarks->b.pte_meta_urgent_ns; @@ -354,7 +301,8 @@ void hubbub1_program_urgent_watermarks( DC_LOG_BANDWIDTH_CALCS("PTE_META_URGENCY_WATERMARK_B calculated =%d\n" "HW register value = 0x%x\n", watermarks->b.pte_meta_urgent_ns, prog_wm_value); - } + } else if (watermarks->b.pte_meta_urgent_ns < hubbub1->watermarks.b.pte_meta_urgent_ns) + wm_pending = true; /* clock state C */ if (safe_to_lower || watermarks->c.urgent_ns > hubbub1->watermarks.c.urgent_ns) { @@ -367,7 +315,8 @@ void hubbub1_program_urgent_watermarks( DC_LOG_BANDWIDTH_CALCS("URGENCY_WATERMARK_C calculated =%d\n" "HW register value = 0x%x\n", watermarks->c.urgent_ns, prog_wm_value); - } + } else if (watermarks->c.urgent_ns < hubbub1->watermarks.c.urgent_ns) + wm_pending = true; if (safe_to_lower || watermarks->c.pte_meta_urgent_ns > hubbub1->watermarks.c.pte_meta_urgent_ns) { hubbub1->watermarks.c.pte_meta_urgent_ns = watermarks->c.pte_meta_urgent_ns; @@ -377,7 +326,8 @@ void hubbub1_program_urgent_watermarks( DC_LOG_BANDWIDTH_CALCS("PTE_META_URGENCY_WATERMARK_C calculated =%d\n" "HW register value = 0x%x\n", watermarks->c.pte_meta_urgent_ns, prog_wm_value); - } + } else if (watermarks->c.pte_meta_urgent_ns < hubbub1->watermarks.c.pte_meta_urgent_ns) + wm_pending = true; /* clock state D */ if (safe_to_lower || watermarks->d.urgent_ns > hubbub1->watermarks.d.urgent_ns) { @@ -390,7 +340,8 @@ void hubbub1_program_urgent_watermarks( DC_LOG_BANDWIDTH_CALCS("URGENCY_WATERMARK_D calculated =%d\n" "HW register value = 0x%x\n", watermarks->d.urgent_ns, prog_wm_value); - } + } else if (watermarks->d.urgent_ns < hubbub1->watermarks.d.urgent_ns) + wm_pending = true; if (safe_to_lower || watermarks->d.pte_meta_urgent_ns > hubbub1->watermarks.d.pte_meta_urgent_ns) { hubbub1->watermarks.d.pte_meta_urgent_ns = watermarks->d.pte_meta_urgent_ns; @@ -400,10 +351,13 @@ void hubbub1_program_urgent_watermarks( DC_LOG_BANDWIDTH_CALCS("PTE_META_URGENCY_WATERMARK_D calculated =%d\n" "HW register value = 0x%x\n", watermarks->d.pte_meta_urgent_ns, prog_wm_value); - } + } else if (watermarks->d.pte_meta_urgent_ns < hubbub1->watermarks.d.pte_meta_urgent_ns) + wm_pending = true; + + return wm_pending; } -void hubbub1_program_stutter_watermarks( +bool hubbub1_program_stutter_watermarks( struct hubbub *hubbub, struct dcn_watermark_set *watermarks, unsigned int refclk_mhz, @@ -411,6 +365,7 @@ void hubbub1_program_stutter_watermarks( { struct dcn10_hubbub *hubbub1 = TO_DCN10_HUBBUB(hubbub); uint32_t prog_wm_value; + bool wm_pending = false; /* clock state A */ if (safe_to_lower || watermarks->a.cstate_pstate.cstate_enter_plus_exit_ns @@ -425,7 +380,9 @@ void hubbub1_program_stutter_watermarks( DC_LOG_BANDWIDTH_CALCS("SR_ENTER_EXIT_WATERMARK_A calculated =%d\n" "HW register value = 0x%x\n", watermarks->a.cstate_pstate.cstate_enter_plus_exit_ns, prog_wm_value); - } + } else if (watermarks->a.cstate_pstate.cstate_enter_plus_exit_ns + < hubbub1->watermarks.a.cstate_pstate.cstate_enter_plus_exit_ns) + wm_pending = true; if (safe_to_lower || watermarks->a.cstate_pstate.cstate_exit_ns > hubbub1->watermarks.a.cstate_pstate.cstate_exit_ns) { @@ -439,7 +396,9 @@ void hubbub1_program_stutter_watermarks( DC_LOG_BANDWIDTH_CALCS("SR_EXIT_WATERMARK_A calculated =%d\n" "HW register value = 0x%x\n", watermarks->a.cstate_pstate.cstate_exit_ns, prog_wm_value); - } + } else if (watermarks->a.cstate_pstate.cstate_exit_ns + < hubbub1->watermarks.a.cstate_pstate.cstate_exit_ns) + wm_pending = true; /* clock state B */ if (safe_to_lower || watermarks->b.cstate_pstate.cstate_enter_plus_exit_ns @@ -454,7 +413,9 @@ void hubbub1_program_stutter_watermarks( DC_LOG_BANDWIDTH_CALCS("SR_ENTER_EXIT_WATERMARK_B calculated =%d\n" "HW register value = 0x%x\n", watermarks->b.cstate_pstate.cstate_enter_plus_exit_ns, prog_wm_value); - } + } else if (watermarks->b.cstate_pstate.cstate_enter_plus_exit_ns + < hubbub1->watermarks.b.cstate_pstate.cstate_enter_plus_exit_ns) + wm_pending = true; if (safe_to_lower || watermarks->b.cstate_pstate.cstate_exit_ns > hubbub1->watermarks.b.cstate_pstate.cstate_exit_ns) { @@ -468,7 +429,9 @@ void hubbub1_program_stutter_watermarks( DC_LOG_BANDWIDTH_CALCS("SR_EXIT_WATERMARK_B calculated =%d\n" "HW register value = 0x%x\n", watermarks->b.cstate_pstate.cstate_exit_ns, prog_wm_value); - } + } else if (watermarks->b.cstate_pstate.cstate_exit_ns + < hubbub1->watermarks.b.cstate_pstate.cstate_exit_ns) + wm_pending = true; /* clock state C */ if (safe_to_lower || watermarks->c.cstate_pstate.cstate_enter_plus_exit_ns @@ -483,7 +446,9 @@ void hubbub1_program_stutter_watermarks( DC_LOG_BANDWIDTH_CALCS("SR_ENTER_EXIT_WATERMARK_C calculated =%d\n" "HW register value = 0x%x\n", watermarks->c.cstate_pstate.cstate_enter_plus_exit_ns, prog_wm_value); - } + } else if (watermarks->c.cstate_pstate.cstate_enter_plus_exit_ns + < hubbub1->watermarks.c.cstate_pstate.cstate_enter_plus_exit_ns) + wm_pending = true; if (safe_to_lower || watermarks->c.cstate_pstate.cstate_exit_ns > hubbub1->watermarks.c.cstate_pstate.cstate_exit_ns) { @@ -497,7 +462,9 @@ void hubbub1_program_stutter_watermarks( DC_LOG_BANDWIDTH_CALCS("SR_EXIT_WATERMARK_C calculated =%d\n" "HW register value = 0x%x\n", watermarks->c.cstate_pstate.cstate_exit_ns, prog_wm_value); - } + } else if (watermarks->c.cstate_pstate.cstate_exit_ns + < hubbub1->watermarks.c.cstate_pstate.cstate_exit_ns) + wm_pending = true; /* clock state D */ if (safe_to_lower || watermarks->d.cstate_pstate.cstate_enter_plus_exit_ns @@ -512,7 +479,9 @@ void hubbub1_program_stutter_watermarks( DC_LOG_BANDWIDTH_CALCS("SR_ENTER_EXIT_WATERMARK_D calculated =%d\n" "HW register value = 0x%x\n", watermarks->d.cstate_pstate.cstate_enter_plus_exit_ns, prog_wm_value); - } + } else if (watermarks->d.cstate_pstate.cstate_enter_plus_exit_ns + < hubbub1->watermarks.d.cstate_pstate.cstate_enter_plus_exit_ns) + wm_pending = true; if (safe_to_lower || watermarks->d.cstate_pstate.cstate_exit_ns > hubbub1->watermarks.d.cstate_pstate.cstate_exit_ns) { @@ -526,11 +495,14 @@ void hubbub1_program_stutter_watermarks( DC_LOG_BANDWIDTH_CALCS("SR_EXIT_WATERMARK_D calculated =%d\n" "HW register value = 0x%x\n", watermarks->d.cstate_pstate.cstate_exit_ns, prog_wm_value); - } + } else if (watermarks->d.cstate_pstate.cstate_exit_ns + < hubbub1->watermarks.d.cstate_pstate.cstate_exit_ns) + wm_pending = true; + return wm_pending; } -void hubbub1_program_pstate_watermarks( +bool hubbub1_program_pstate_watermarks( struct hubbub *hubbub, struct dcn_watermark_set *watermarks, unsigned int refclk_mhz, @@ -538,6 +510,7 @@ void hubbub1_program_pstate_watermarks( { struct dcn10_hubbub *hubbub1 = TO_DCN10_HUBBUB(hubbub); uint32_t prog_wm_value; + bool wm_pending = false; /* clock state A */ if (safe_to_lower || watermarks->a.cstate_pstate.pstate_change_ns @@ -552,7 +525,9 @@ void hubbub1_program_pstate_watermarks( DC_LOG_BANDWIDTH_CALCS("DRAM_CLK_CHANGE_WATERMARK_A calculated =%d\n" "HW register value = 0x%x\n\n", watermarks->a.cstate_pstate.pstate_change_ns, prog_wm_value); - } + } else if (watermarks->a.cstate_pstate.pstate_change_ns + < hubbub1->watermarks.a.cstate_pstate.pstate_change_ns) + wm_pending = true; /* clock state B */ if (safe_to_lower || watermarks->b.cstate_pstate.pstate_change_ns @@ -567,7 +542,9 @@ void hubbub1_program_pstate_watermarks( DC_LOG_BANDWIDTH_CALCS("DRAM_CLK_CHANGE_WATERMARK_B calculated =%d\n" "HW register value = 0x%x\n\n", watermarks->b.cstate_pstate.pstate_change_ns, prog_wm_value); - } + } else if (watermarks->b.cstate_pstate.pstate_change_ns + < hubbub1->watermarks.b.cstate_pstate.pstate_change_ns) + wm_pending = true; /* clock state C */ if (safe_to_lower || watermarks->c.cstate_pstate.pstate_change_ns @@ -582,7 +559,9 @@ void hubbub1_program_pstate_watermarks( DC_LOG_BANDWIDTH_CALCS("DRAM_CLK_CHANGE_WATERMARK_C calculated =%d\n" "HW register value = 0x%x\n\n", watermarks->c.cstate_pstate.pstate_change_ns, prog_wm_value); - } + } else if (watermarks->c.cstate_pstate.pstate_change_ns + < hubbub1->watermarks.c.cstate_pstate.pstate_change_ns) + wm_pending = true; /* clock state D */ if (safe_to_lower || watermarks->d.cstate_pstate.pstate_change_ns @@ -597,23 +576,33 @@ void hubbub1_program_pstate_watermarks( DC_LOG_BANDWIDTH_CALCS("DRAM_CLK_CHANGE_WATERMARK_D calculated =%d\n" "HW register value = 0x%x\n\n", watermarks->d.cstate_pstate.pstate_change_ns, prog_wm_value); - } + } else if (watermarks->d.cstate_pstate.pstate_change_ns + < hubbub1->watermarks.d.cstate_pstate.pstate_change_ns) + wm_pending = true; + + return wm_pending; } -void hubbub1_program_watermarks( +bool hubbub1_program_watermarks( struct hubbub *hubbub, struct dcn_watermark_set *watermarks, unsigned int refclk_mhz, bool safe_to_lower) { struct dcn10_hubbub *hubbub1 = TO_DCN10_HUBBUB(hubbub); + bool wm_pending = false; /* * Need to clamp to max of the register values (i.e. no wrap) * for dcn1, all wm registers are 21-bit wide */ - hubbub1_program_urgent_watermarks(hubbub, watermarks, refclk_mhz, safe_to_lower); - hubbub1_program_stutter_watermarks(hubbub, watermarks, refclk_mhz, safe_to_lower); - hubbub1_program_pstate_watermarks(hubbub, watermarks, refclk_mhz, safe_to_lower); + if (hubbub1_program_urgent_watermarks(hubbub, watermarks, refclk_mhz, safe_to_lower)) + wm_pending = true; + + if (hubbub1_program_stutter_watermarks(hubbub, watermarks, refclk_mhz, safe_to_lower)) + wm_pending = true; + + if (hubbub1_program_pstate_watermarks(hubbub, watermarks, refclk_mhz, safe_to_lower)) + wm_pending = true; REG_UPDATE(DCHUBBUB_ARB_SAT_LEVEL, DCHUBBUB_ARB_SAT_LEVEL, 60 * refclk_mhz); @@ -627,6 +616,7 @@ void hubbub1_program_watermarks( DCHUBBUB_ARB_WATERMARK_CHANGE_DONE_INTERRUPT_DISABLE, 1, DCHUBBUB_ARB_WATERMARK_CHANGE_REQUEST, 1); #endif + return wm_pending; } void hubbub1_update_dchub( @@ -840,8 +830,8 @@ static void hubbub1_det_request_size( hubbub1_get_blk256_size(&blk256_width, &blk256_height, bpe); - swath_bytes_horz_wc = height * blk256_height * bpe; - swath_bytes_vert_wc = width * blk256_width * bpe; + swath_bytes_horz_wc = width * blk256_height * bpe; + swath_bytes_vert_wc = height * blk256_width * bpe; *req128_horz_wc = (2 * swath_bytes_horz_wc <= detile_buf_size) ? false : /* full 256B request */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.h index af57751253de..343a537172c7 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.h @@ -308,7 +308,7 @@ bool hubbub1_verify_allow_pstate_change_high( void hubbub1_wm_change_req_wa(struct hubbub *hubbub); -void hubbub1_program_watermarks( +bool hubbub1_program_watermarks( struct hubbub *hubbub, struct dcn_watermark_set *watermarks, unsigned int refclk_mhz, @@ -331,17 +331,17 @@ void hubbub1_construct(struct hubbub *hubbub, const struct dcn_hubbub_shift *hubbub_shift, const struct dcn_hubbub_mask *hubbub_mask); -void hubbub1_program_urgent_watermarks( +bool hubbub1_program_urgent_watermarks( struct hubbub *hubbub, struct dcn_watermark_set *watermarks, unsigned int refclk_mhz, bool safe_to_lower); -void hubbub1_program_stutter_watermarks( +bool hubbub1_program_stutter_watermarks( struct hubbub *hubbub, struct dcn_watermark_set *watermarks, unsigned int refclk_mhz, bool safe_to_lower); -void hubbub1_program_pstate_watermarks( +bool hubbub1_program_pstate_watermarks( struct hubbub *hubbub, struct dcn_watermark_set *watermarks, unsigned int refclk_mhz, diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c index 1008ac8a0f2a..b0357546471b 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c @@ -48,8 +48,8 @@ #include "dc_link_dp.h" #include "dccg.h" #include "clk_mgr.h" - - +#include "link_hwss.h" +#include "dpcd_defs.h" #include "dsc.h" #define DC_LOGGER_INIT(logger) @@ -82,7 +82,7 @@ void print_microsec(struct dc_context *dc_ctx, us_x10 % frac); } -static void dcn10_lock_all_pipes(struct dc *dc, +void dcn10_lock_all_pipes(struct dc *dc, struct dc_state *context, bool lock) { @@ -93,6 +93,7 @@ static void dcn10_lock_all_pipes(struct dc *dc, for (i = 0; i < dc->res_pool->pipe_count; i++) { pipe_ctx = &context->res_ctx.pipe_ctx[i]; tg = pipe_ctx->stream_res.tg; + /* * Only lock the top pipe's tg to prevent redundant * (un)locking. Also skip if pipe is disabled. @@ -103,9 +104,9 @@ static void dcn10_lock_all_pipes(struct dc *dc, continue; if (lock) - tg->funcs->lock(tg); + dc->hwss.pipe_control_lock(dc, pipe_ctx, true); else - tg->funcs->unlock(tg); + dc->hwss.pipe_control_lock(dc, pipe_ctx, false); } } @@ -900,6 +901,10 @@ static void dcn10_reset_back_end_for_pipe( * parent pipe. */ if (pipe_ctx->top_pipe == NULL) { + + if (pipe_ctx->stream_res.abm) + pipe_ctx->stream_res.abm->funcs->set_abm_immediate_disable(pipe_ctx->stream_res.abm); + pipe_ctx->stream_res.tg->funcs->disable_crtc(pipe_ctx->stream_res.tg); pipe_ctx->stream_res.tg->funcs->enable_optc_clock(pipe_ctx->stream_res.tg, false); @@ -1263,7 +1268,8 @@ void dcn10_init_hw(struct dc *dc) } //Enable ability to power gate / don't force power on permanently - hws->funcs.enable_power_gating_plane(hws, true); + if (hws->funcs.enable_power_gating_plane) + hws->funcs.enable_power_gating_plane(hws, true); return; } @@ -1317,6 +1323,31 @@ void dcn10_init_hw(struct dc *dc) if (hws->funcs.dsc_pg_control != NULL) hws->funcs.dsc_pg_control(hws, res_pool->dscs[i]->inst, false); + /* we want to turn off all dp displays before doing detection */ + if (dc->config.power_down_display_on_boot) { + uint8_t dpcd_power_state = '\0'; + enum dc_status status = DC_ERROR_UNEXPECTED; + + for (i = 0; i < dc->link_count; i++) { + if (dc->links[i]->connector_signal != SIGNAL_TYPE_DISPLAY_PORT) + continue; + + /* + * core_link_read_dpcd() will invoke dm_helpers_dp_read_dpcd(), + * which needs to read dpcd info with the help of aconnector. + * If aconnector (dc->links[i]->prev) is NULL, then dpcd status + * cannot be read. + */ + if (dc->links[i]->priv) { + /* if any of the displays are lit up turn them off */ + status = core_link_read_dpcd(dc->links[i], DP_SET_POWER, + &dpcd_power_state, sizeof(dpcd_power_state)); + if (status == DC_OK && dpcd_power_state == DP_POWER_STATE_D0) + dp_receiver_power_ctrl(dc->links[i], false); + } + } + } + /* If taking control over from VBIOS, we may want to optimize our first * mode set, so we need to skip powering down pipes until we know which * pipes we want to use. @@ -1325,6 +1356,9 @@ void dcn10_init_hw(struct dc *dc) */ if (dcb->funcs->is_accelerated_mode(dcb) || dc->config.power_down_display_on_boot) { hws->funcs.init_pipes(dc, dc->current_state); + if (dc->res_pool->hubbub->funcs->allow_self_refresh_control) + dc->res_pool->hubbub->funcs->allow_self_refresh_control(dc->res_pool->hubbub, + !dc->res_pool->hubbub->ctx->dc->debug.disable_stutter); } for (i = 0; i < res_pool->audio_count; i++) { @@ -1355,8 +1389,8 @@ void dcn10_init_hw(struct dc *dc) REG_UPDATE(DCFCLK_CNTL, DCFCLK_GATE_DIS, 0); } - - hws->funcs.enable_power_gating_plane(dc->hwseq, true); + if (hws->funcs.enable_power_gating_plane) + hws->funcs.enable_power_gating_plane(dc->hwseq, true); if (dc->clk_mgr->funcs->notify_wm_ranges) dc->clk_mgr->funcs->notify_wm_ranges(dc->clk_mgr); @@ -1576,7 +1610,7 @@ void dcn10_pipe_control_lock( /* use TG master update lock to lock everything on the TG * therefore only top pipe need to lock */ - if (pipe->top_pipe) + if (!pipe || pipe->top_pipe) return; if (dc->debug.sanity_checks) @@ -1970,6 +2004,12 @@ void dcn10_program_gamut_remap(struct pipe_ctx *pipe_ctx) for (i = 0; i < CSC_TEMPERATURE_MATRIX_SIZE; i++) adjust.temperature_matrix[i] = pipe_ctx->stream->gamut_remap_matrix.matrix[i]; + } else if (pipe_ctx->plane_state && + pipe_ctx->plane_state->gamut_remap_matrix.enable_remap == true) { + adjust.gamut_adjust_type = GRAPHICS_GAMUT_ADJUST_TYPE_SW; + for (i = 0; i < CSC_TEMPERATURE_MATRIX_SIZE; i++) + adjust.temperature_matrix[i] = + pipe_ctx->plane_state->gamut_remap_matrix.matrix[i]; } pipe_ctx->plane_res.dpp->funcs->dpp_set_gamut_remap(pipe_ctx->plane_res.dpp, &adjust); @@ -2090,6 +2130,10 @@ void dcn10_get_hdr_visual_confirm_color( if (top_pipe_ctx->stream->out_transfer_func->tf == TRANSFER_FUNCTION_PQ) { /* HDR10, ARGB2101010 - set boarder color to red */ color->color_r_cr = color_value; + } else if (top_pipe_ctx->stream->out_transfer_func->tf == TRANSFER_FUNCTION_GAMMA22) { + /* FreeSync 2 ARGB2101010 - set boarder color to pink */ + color->color_r_cr = color_value; + color->color_b_cb = color_value; } break; case PIXEL_FORMAT_FP16: @@ -2512,12 +2556,17 @@ void dcn10_apply_ctx_for_surface( int i; struct timing_generator *tg; uint32_t underflow_check_delay_us; - bool removed_pipe[4] = { false }; bool interdependent_update = false; struct pipe_ctx *top_pipe_to_program = dcn10_find_top_pipe_for_stream(dc, context, stream); DC_LOGGER_INIT(dc->ctx->logger); + // Clear pipe_ctx flag + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; + pipe_ctx->update_flags.raw = 0; + } + if (!top_pipe_to_program) return; @@ -2531,11 +2580,6 @@ void dcn10_apply_ctx_for_surface( if (underflow_check_delay_us != 0xFFFFFFFF && hws->funcs.did_underflow_occur) ASSERT(hws->funcs.did_underflow_occur(dc, top_pipe_to_program)); - if (interdependent_update) - dcn10_lock_all_pipes(dc, context, true); - else - dcn10_pipe_control_lock(dc, top_pipe_to_program, true); - if (underflow_check_delay_us != 0xFFFFFFFF) udelay(underflow_check_delay_us); @@ -2552,18 +2596,6 @@ void dcn10_apply_ctx_for_surface( struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; struct pipe_ctx *old_pipe_ctx = &dc->current_state->res_ctx.pipe_ctx[i]; - /* - * Powergate reused pipes that are not powergated - * fairly hacky right now, using opp_id as indicator - * TODO: After move dc_post to dc_update, this will - * be removed. - */ - if (pipe_ctx->plane_state && !old_pipe_ctx->plane_state) { - if (old_pipe_ctx->stream_res.tg == tg && - old_pipe_ctx->plane_res.hubp && - old_pipe_ctx->plane_res.hubp->opp_id != OPP_ID_INVALID) - dc->hwss.disable_plane(dc, old_pipe_ctx); - } if ((!pipe_ctx->plane_state || pipe_ctx->stream_res.tg != old_pipe_ctx->stream_res.tg) && @@ -2571,7 +2603,7 @@ void dcn10_apply_ctx_for_surface( old_pipe_ctx->stream_res.tg == tg) { hws->funcs.plane_atomic_disconnect(dc, old_pipe_ctx); - removed_pipe[i] = true; + pipe_ctx->update_flags.bits.disable = 1; DC_LOG_DC("Reset mpcc for pipe %d\n", old_pipe_ctx->pipe_idx); @@ -2597,21 +2629,35 @@ void dcn10_apply_ctx_for_surface( &pipe_ctx->dlg_regs, &pipe_ctx->ttu_regs); } +} - if (interdependent_update) - dcn10_lock_all_pipes(dc, context, false); - else - dcn10_pipe_control_lock(dc, top_pipe_to_program, false); +void dcn10_post_unlock_program_front_end( + struct dc *dc, + struct dc_state *context) +{ + int i; + + DC_LOGGER_INIT(dc->ctx->logger); + + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; - if (num_planes == 0) - false_optc_underflow_wa(dc, stream, tg); + if (!pipe_ctx->top_pipe && + !pipe_ctx->prev_odm_pipe && + pipe_ctx->stream) { + struct timing_generator *tg = pipe_ctx->stream_res.tg; + + if (context->stream_status[i].plane_count == 0) + false_optc_underflow_wa(dc, pipe_ctx->stream, tg); + } + } for (i = 0; i < dc->res_pool->pipe_count; i++) - if (removed_pipe[i]) + if (context->res_ctx.pipe_ctx[i].update_flags.bits.disable) dc->hwss.disable_plane(dc, &dc->current_state->res_ctx.pipe_ctx[i]); for (i = 0; i < dc->res_pool->pipe_count; i++) - if (removed_pipe[i]) { + if (context->res_ctx.pipe_ctx[i].update_flags.bits.disable) { dc->hwss.optimize_bandwidth(dc, context); break; } @@ -2656,7 +2702,7 @@ void dcn10_prepare_bandwidth( false); } - hubbub->funcs->program_watermarks(hubbub, + dc->wm_optimized_required = hubbub->funcs->program_watermarks(hubbub, &context->bw_ctx.bw.dcn.watermarks, dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000, true); @@ -2693,6 +2739,7 @@ void dcn10_optimize_bandwidth( &context->bw_ctx.bw.dcn.watermarks, dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000, true); + dcn10_stereo_hw_frame_pack_wa(dc, context); if (dc->debug.pplib_wm_report_mode == WM_REPORT_OVERRIDE) @@ -2884,6 +2931,7 @@ void dcn10_update_pending_status(struct pipe_ctx *pipe_ctx) struct dc_plane_state *plane_state = pipe_ctx->plane_state; struct timing_generator *tg = pipe_ctx->stream_res.tg; bool flip_pending; + struct dc *dc = plane_state->ctx->dc; if (plane_state == NULL) return; @@ -2901,6 +2949,19 @@ void dcn10_update_pending_status(struct pipe_ctx *pipe_ctx) plane_state->status.is_right_eye = !tg->funcs->is_stereo_left_eye(pipe_ctx->stream_res.tg); } + + if (dc->hwseq->wa_state.disallow_self_refresh_during_multi_plane_transition_applied) { + struct dce_hwseq *hwseq = dc->hwseq; + struct timing_generator *tg = dc->res_pool->timing_generators[0]; + unsigned int cur_frame = tg->funcs->get_frame_count(tg); + + if (cur_frame != hwseq->wa_state.disallow_self_refresh_during_multi_plane_transition_applied_on_frame) { + struct hubbub *hubbub = dc->res_pool->hubbub; + + hubbub->funcs->allow_self_refresh_control(hubbub, !dc->debug.disable_stutter); + hwseq->wa_state.disallow_self_refresh_during_multi_plane_transition_applied = false; + } + } } void dcn10_update_dchub(struct dce_hwseq *hws, struct dchub_init_data *dh_data) @@ -2960,12 +3021,50 @@ void dcn10_set_cursor_position(struct pipe_ctx *pipe_ctx) int x_pos = pos_cpy.x; int y_pos = pos_cpy.y; - // translate cursor from stream space to plane space + /** + * DC cursor is stream space, HW cursor is plane space and drawn + * as part of the framebuffer. + * + * Cursor position can't be negative, but hotspot can be used to + * shift cursor out of the plane bounds. Hotspot must be smaller + * than the cursor size. + */ + + /** + * Translate cursor from stream space to plane space. + * + * If the cursor is scaled then we need to scale the position + * to be in the approximately correct place. We can't do anything + * about the actual size being incorrect, that's a limitation of + * the hardware. + */ x_pos = (x_pos - x_plane) * pipe_ctx->plane_state->src_rect.width / pipe_ctx->plane_state->dst_rect.width; y_pos = (y_pos - y_plane) * pipe_ctx->plane_state->src_rect.height / pipe_ctx->plane_state->dst_rect.height; + /** + * If the cursor's source viewport is clipped then we need to + * translate the cursor to appear in the correct position on + * the screen. + * + * This translation isn't affected by scaling so it needs to be + * done *after* we adjust the position for the scale factor. + * + * This is only done by opt-in for now since there are still + * some usecases like tiled display that might enable the + * cursor on both streams while expecting dc to clip it. + */ + if (pos_cpy.translate_by_source) { + x_pos += pipe_ctx->plane_state->src_rect.x; + y_pos += pipe_ctx->plane_state->src_rect.y; + } + + /** + * If the position is negative then we need to add to the hotspot + * to shift the cursor outside the plane. + */ + if (x_pos < 0) { pos_cpy.x_hotspot -= x_pos; x_pos = 0; diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h index 4d20f6586bb5..16a50e05ffbf 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h @@ -70,11 +70,18 @@ void dcn10_reset_hw_ctx_wrap( struct dc *dc, struct dc_state *context); void dcn10_disable_plane(struct dc *dc, struct pipe_ctx *pipe_ctx); +void dcn10_lock_all_pipes( + struct dc *dc, + struct dc_state *context, + bool lock); void dcn10_apply_ctx_for_surface( struct dc *dc, const struct dc_stream_state *stream, int num_planes, struct dc_state *context); +void dcn10_post_unlock_program_front_end( + struct dc *dc, + struct dc_state *context); void dcn10_hubp_pg_control( struct dce_hwseq *hws, unsigned int hubp_inst, diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_init.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_init.c index e7e5352ec424..dd02d3983695 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_init.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_init.c @@ -32,6 +32,7 @@ static const struct hw_sequencer_funcs dcn10_funcs = { .init_hw = dcn10_init_hw, .apply_ctx_to_hw = dce110_apply_ctx_to_hw, .apply_ctx_for_surface = dcn10_apply_ctx_for_surface, + .post_unlock_program_front_end = dcn10_post_unlock_program_front_end, .update_plane_addr = dcn10_update_plane_addr, .update_dchub = dcn10_update_dchub, .update_pending_status = dcn10_update_pending_status, @@ -49,6 +50,7 @@ static const struct hw_sequencer_funcs dcn10_funcs = { .disable_audio_stream = dce110_disable_audio_stream, .disable_plane = dcn10_disable_plane, .pipe_control_lock = dcn10_pipe_control_lock, + .interdependent_update_lock = dcn10_lock_all_pipes, .prepare_bandwidth = dcn10_prepare_bandwidth, .optimize_bandwidth = dcn10_optimize_bandwidth, .set_drr = dcn10_set_drr, @@ -85,6 +87,8 @@ static const struct hwseq_private_funcs dcn10_private_funcs = { .reset_hw_ctx_wrap = dcn10_reset_hw_ctx_wrap, .enable_stream_timing = dcn10_enable_stream_timing, .edp_backlight_control = dce110_edp_backlight_control, + .is_panel_backlight_on = dce110_is_panel_backlight_on, + .is_panel_powered_on = dce110_is_panel_powered_on, .disable_stream_gating = NULL, .enable_stream_gating = NULL, .setup_vupdate_interrupt = dcn10_setup_vupdate_interrupt, diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c index 1a37c90e9d43..d3617d6785a7 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c @@ -782,6 +782,11 @@ bool dcn10_link_encoder_validate_output_with_stream( struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); bool is_valid; + //if SCDC (340-600MHz) is disabled, set to HDMI 1.4 timing limit + if (stream->sink->edid_caps.panel_patch.skip_scdc_overwrite && + enc10->base.features.max_hdmi_pixel_clock > 300000) + enc10->base.features.max_hdmi_pixel_clock = 300000; + switch (stream->signal) { case SIGNAL_TYPE_DVI_SINGLE_LINK: case SIGNAL_TYPE_DVI_DUAL_LINK: diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.h index eb13589b9a81..762109174fb8 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.h @@ -62,11 +62,11 @@ SRI(DP_DPHY_FAST_TRAINING, DP, id), \ SRI(DP_SEC_CNTL1, DP, id), \ SRI(DP_DPHY_BS_SR_SWAP_CNTL, DP, id), \ - SRI(DP_DPHY_INTERNAL_CTRL, DP, id), \ SRI(DP_DPHY_HBR2_PATTERN_CONTROL, DP, id) #define LE_DCN10_REG_LIST(id)\ + SRI(DP_DPHY_INTERNAL_CTRL, DP, id), \ LE_DCN_COMMON_REG_LIST(id) struct dcn10_link_enc_aux_registers { diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c index a9a43b397db9..17d96ec6acd8 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c @@ -299,7 +299,6 @@ void optc1_set_vtg_params(struct timing_generator *optc, uint32_t asic_blank_end; uint32_t v_init; uint32_t v_fp2 = 0; - int32_t vertical_line_start; struct optc *optc1 = DCN10TG_FROM_TG(optc); @@ -316,9 +315,8 @@ void optc1_set_vtg_params(struct timing_generator *optc, patched_crtc_timing.v_border_top; /* if VSTARTUP is before VSYNC, FP2 is the offset, otherwise 0 */ - vertical_line_start = asic_blank_end - optc1->vstartup_start + 1; - if (vertical_line_start < 0) - v_fp2 = -vertical_line_start; + if (optc1->vstartup_start > asic_blank_end) + v_fp2 = optc1->vstartup_start - asic_blank_end; /* Interlace */ if (REG(OTG_INTERLACE_CONTROL)) { @@ -345,6 +343,23 @@ void optc1_set_blank_data_double_buffer(struct timing_generator *optc, bool enab } /** + * optc1_set_timing_double_buffer() - DRR double buffering control + * + * Sets double buffer point for V_TOTAL, H_TOTAL, VTOTAL_MIN, + * VTOTAL_MAX, VTOTAL_MIN_SEL and VTOTAL_MAX_SEL registers. + * + * Options: any time, start of frame, dp start of frame (range timing) + */ +void optc1_set_timing_double_buffer(struct timing_generator *optc, bool enable) +{ + struct optc *optc1 = DCN10TG_FROM_TG(optc); + uint32_t mode = enable ? 2 : 0; + + REG_UPDATE(OTG_DOUBLE_BUFFER_CONTROL, + OTG_RANGE_TIMING_DBUF_UPDATE_MODE, mode); +} + +/** * unblank_crtc * Call ASIC Control Object to UnBlank CRTC. */ @@ -1195,7 +1210,7 @@ static void optc1_enable_stereo(struct timing_generator *optc, REG_UPDATE_3(OTG_STEREO_CONTROL, OTG_STEREO_EN, stereo_en, OTG_STEREO_SYNC_OUTPUT_LINE_NUM, 0, - OTG_STEREO_SYNC_OUTPUT_POLARITY, 0); + OTG_STEREO_SYNC_OUTPUT_POLARITY, flags->RIGHT_EYE_POLARITY == 0 ? 0 : 1); if (flags->PROGRAM_POLARITY) REG_UPDATE(OTG_STEREO_CONTROL, @@ -1355,6 +1370,7 @@ void optc1_clear_optc_underflow(struct timing_generator *optc) void optc1_tg_init(struct timing_generator *optc) { optc1_set_blank_data_double_buffer(optc, true); + optc1_set_timing_double_buffer(optc, true); optc1_clear_optc_underflow(optc); } diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h index f277656d5464..9a459a8fe8a0 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h @@ -185,6 +185,7 @@ struct dcn_optc_registers { SF(OTG0_OTG_GLOBAL_CONTROL0, OTG_MASTER_UPDATE_LOCK_SEL, mask_sh),\ SF(OTG0_OTG_DOUBLE_BUFFER_CONTROL, OTG_UPDATE_PENDING, mask_sh),\ SF(OTG0_OTG_DOUBLE_BUFFER_CONTROL, OTG_BLANK_DATA_DOUBLE_BUFFER_EN, mask_sh),\ + SF(OTG0_OTG_DOUBLE_BUFFER_CONTROL, OTG_RANGE_TIMING_DBUF_UPDATE_MODE, mask_sh),\ SF(OTG0_OTG_H_TOTAL, OTG_H_TOTAL, mask_sh),\ SF(OTG0_OTG_H_BLANK_START_END, OTG_H_BLANK_START, mask_sh),\ SF(OTG0_OTG_H_BLANK_START_END, OTG_H_BLANK_END, mask_sh),\ @@ -643,6 +644,8 @@ bool optc1_is_optc_underflow_occurred(struct timing_generator *optc); void optc1_set_blank_data_double_buffer(struct timing_generator *optc, bool enable); +void optc1_set_timing_double_buffer(struct timing_generator *optc, bool enable); + bool optc1_get_otg_active_size(struct timing_generator *optc, uint32_t *otg_active_width, uint32_t *otg_active_height); diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c index 3b71898e859e..07265ca7d28c 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c @@ -552,7 +552,8 @@ static const struct dc_plane_cap plane_cap = { .pixel_format_support = { .argb8888 = true, .nv12 = true, - .fp16 = true + .fp16 = true, + .p010 = true }, .max_upscale_factor = { @@ -570,7 +571,7 @@ static const struct dc_plane_cap plane_cap = { static const struct dc_debug_options debug_defaults_drv = { .sanity_checks = true, - .disable_dmcu = true, + .disable_dmcu = false, .force_abm_enable = false, .timing_trace = false, .clock_trace = true, @@ -584,7 +585,7 @@ static const struct dc_debug_options debug_defaults_drv = { .disable_pplib_clock_request = false, .disable_pplib_wm_range = false, .pplib_wm_report_mode = WM_REPORT_DEFAULT, - .pipe_split_policy = MPC_SPLIT_AVOID_MULT_DISP, + .pipe_split_policy = MPC_SPLIT_DYNAMIC, .force_single_disp_pipe_split = true, .disable_dcc = DCC_ENABLE, .voltage_align_fclk = true, @@ -598,7 +599,7 @@ static const struct dc_debug_options debug_defaults_drv = { }; static const struct dc_debug_options debug_defaults_diags = { - .disable_dmcu = true, + .disable_dmcu = false, .force_abm_enable = false, .timing_trace = true, .clock_trace = true, @@ -1233,7 +1234,7 @@ static enum dc_status dcn10_validate_global(struct dc *dc, struct dc_state *cont return DC_OK; } -static enum dc_status dcn10_get_default_swizzle_mode(struct dc_plane_state *plane_state) +static enum dc_status dcn10_patch_unknown_plane_state(struct dc_plane_state *plane_state) { enum dc_status result = DC_OK; @@ -1295,7 +1296,7 @@ static const struct resource_funcs dcn10_res_pool_funcs = { .validate_plane = dcn10_validate_plane, .validate_global = dcn10_validate_global, .add_stream_to_ctx = dcn10_add_stream_to_ctx, - .get_default_swizzle_mode = dcn10_get_default_swizzle_mode, + .patch_unknown_plane_state = dcn10_patch_unknown_plane_state, .find_first_free_match_stream_enc_for_link = dcn10_find_first_free_match_stream_enc_for_link }; diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.c index 376c4264d295..7eba9333c328 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.c @@ -1667,5 +1667,6 @@ void dcn10_stream_encoder_construct( enc1->regs = regs; enc1->se_shift = se_shift; enc1->se_mask = se_mask; + enc1->base.stream_enc_inst = eng_id - ENGINE_ID_DIGA; } diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.c index 50bffbfdd394..62cc2651e00c 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.c @@ -70,6 +70,8 @@ void dccg2_update_dpp_dto(struct dccg *dccg, int dpp_inst, int req_dppclk) REG_UPDATE(DPPCLK_DTO_CTRL, DPPCLK_DTO_ENABLE[dpp_inst], 0); } + + dccg->pipe_dppclk_khz[dpp_inst] = req_dppclk; } void dccg2_get_dccg_ref_freq(struct dccg *dccg, diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.c index 13e057d7ee93..42bba7c9548b 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.c @@ -369,84 +369,6 @@ void dpp2_set_cursor_attributes( } } -#define IDENTITY_RATIO(ratio) (dc_fixpt_u3d19(ratio) == (1 << 19)) - -bool dpp2_get_optimal_number_of_taps( - struct dpp *dpp, - struct scaler_data *scl_data, - const struct scaling_taps *in_taps) -{ - /* Some ASICs does not support FP16 scaling, so we reject modes require this*/ - if (scl_data->viewport.width != scl_data->h_active && - scl_data->viewport.height != scl_data->v_active && - dpp->caps->dscl_data_proc_format == DSCL_DATA_PRCESSING_FIXED_FORMAT && - scl_data->format == PIXEL_FORMAT_FP16) - return false; - - if (scl_data->viewport.width > scl_data->h_active && - dpp->ctx->dc->debug.max_downscale_src_width != 0 && - scl_data->viewport.width > dpp->ctx->dc->debug.max_downscale_src_width) - return false; - - /* TODO: add lb check */ - - /* No support for programming ratio of 8, drop to 7.99999.. */ - if (scl_data->ratios.horz.value == (8ll << 32)) - scl_data->ratios.horz.value--; - if (scl_data->ratios.vert.value == (8ll << 32)) - scl_data->ratios.vert.value--; - if (scl_data->ratios.horz_c.value == (8ll << 32)) - scl_data->ratios.horz_c.value--; - if (scl_data->ratios.vert_c.value == (8ll << 32)) - scl_data->ratios.vert_c.value--; - - /* Set default taps if none are provided */ - if (in_taps->h_taps == 0) { - if (dc_fixpt_ceil(scl_data->ratios.horz) > 4) - scl_data->taps.h_taps = 8; - else - scl_data->taps.h_taps = 4; - } else - scl_data->taps.h_taps = in_taps->h_taps; - if (in_taps->v_taps == 0) { - if (dc_fixpt_ceil(scl_data->ratios.vert) > 4) - scl_data->taps.v_taps = 8; - else - scl_data->taps.v_taps = 4; - } else - scl_data->taps.v_taps = in_taps->v_taps; - if (in_taps->v_taps_c == 0) { - if (dc_fixpt_ceil(scl_data->ratios.vert_c) > 4) - scl_data->taps.v_taps_c = 4; - else - scl_data->taps.v_taps_c = 2; - } else - scl_data->taps.v_taps_c = in_taps->v_taps_c; - if (in_taps->h_taps_c == 0) { - if (dc_fixpt_ceil(scl_data->ratios.horz_c) > 4) - scl_data->taps.h_taps_c = 4; - else - scl_data->taps.h_taps_c = 2; - } else if ((in_taps->h_taps_c % 2) != 0 && in_taps->h_taps_c != 1) - /* Only 1 and even h_taps_c are supported by hw */ - scl_data->taps.h_taps_c = in_taps->h_taps_c - 1; - else - scl_data->taps.h_taps_c = in_taps->h_taps_c; - - if (!dpp->ctx->dc->debug.always_scale) { - if (IDENTITY_RATIO(scl_data->ratios.horz)) - scl_data->taps.h_taps = 1; - if (IDENTITY_RATIO(scl_data->ratios.vert)) - scl_data->taps.v_taps = 1; - if (IDENTITY_RATIO(scl_data->ratios.horz_c)) - scl_data->taps.h_taps_c = 1; - if (IDENTITY_RATIO(scl_data->ratios.vert_c)) - scl_data->taps.v_taps_c = 1; - } - - return true; -} - void oppn20_dummy_program_regamma_pwl( struct dpp *dpp, const struct pwl_params *params, diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.c index 6bdfee20b6a7..1b1ae9ce2799 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.c @@ -369,6 +369,7 @@ static bool dsc_prepare_config(const struct dsc_config *dsc_cfg, struct dsc_reg_ dsc_reg_vals->pps.block_pred_enable = dsc_cfg->dc_dsc_cfg.block_pred_enable; dsc_reg_vals->pps.line_buf_depth = dsc_cfg->dc_dsc_cfg.linebuf_depth; dsc_reg_vals->alternate_ich_encoding_en = dsc_reg_vals->pps.dsc_version_minor == 1 ? 0 : 1; + dsc_reg_vals->ich_reset_at_eol = (dsc_cfg->is_odm || dsc_reg_vals->num_slices_h > 1) ? 0xF : 0; // TODO: in addition to validating slice height (pic height must be divisible by slice height), // see what happens when the same condition doesn't apply for slice_width/pic_width. @@ -531,7 +532,6 @@ static void dsc_update_from_dsc_parameters(struct dsc_reg_values *reg_vals, cons reg_vals->pps.rc_buf_thresh[i] = reg_vals->pps.rc_buf_thresh[i] >> 6; reg_vals->rc_buffer_model_size = dsc_params->rc_buffer_model_size; - reg_vals->ich_reset_at_eol = reg_vals->num_slices_h == 1 ? 0 : 0xf; } static void dsc_write_to_registers(struct display_stream_compressor *dsc, const struct dsc_reg_values *reg_vals) diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.c index 9235f7d29454..c0b21d7450d4 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.c @@ -562,19 +562,23 @@ void hubbub2_get_dchub_ref_freq(struct hubbub *hubbub, } } -static void hubbub2_program_watermarks( +static bool hubbub2_program_watermarks( struct hubbub *hubbub, struct dcn_watermark_set *watermarks, unsigned int refclk_mhz, bool safe_to_lower) { struct dcn20_hubbub *hubbub1 = TO_DCN20_HUBBUB(hubbub); + bool wm_pending = false; /* * Need to clamp to max of the register values (i.e. no wrap) * for dcn1, all wm registers are 21-bit wide */ - hubbub1_program_urgent_watermarks(hubbub, watermarks, refclk_mhz, safe_to_lower); - hubbub1_program_stutter_watermarks(hubbub, watermarks, refclk_mhz, safe_to_lower); + if (hubbub1_program_urgent_watermarks(hubbub, watermarks, refclk_mhz, safe_to_lower)) + wm_pending = true; + + if (hubbub1_program_stutter_watermarks(hubbub, watermarks, refclk_mhz, safe_to_lower)) + wm_pending = true; /* * There's a special case when going from p-state support to p-state unsupported @@ -592,6 +596,7 @@ static void hubbub2_program_watermarks( REG_UPDATE(DCHUBBUB_ARB_DF_REQ_OUTSTAND, DCHUBBUB_ARB_MIN_REQ_OUTSTAND, 180); hubbub->funcs->allow_self_refresh_control(hubbub, !hubbub->ctx->dc->debug.disable_stutter); + return wm_pending; } static const struct hubbub_funcs hubbub2_funcs = { diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c index a444fed94184..22f421e82733 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c @@ -307,7 +307,8 @@ void dcn20_init_blank( COLOR_DEPTH_UNDEFINED, &black_color, otg_active_width, - otg_active_height); + otg_active_height, + 0); if (num_opps == 2) { bottom_opp->funcs->opp_set_disp_pattern_generator( @@ -317,7 +318,8 @@ void dcn20_init_blank( COLOR_DEPTH_UNDEFINED, &black_color, otg_active_width, - otg_active_height); + otg_active_height, + 0); } hws->funcs.wait_for_blank_complete(opp); @@ -645,6 +647,9 @@ enum dc_status dcn20_enable_stream_timing( return DC_ERROR_UNEXPECTED; } + if (dc->hwseq->funcs.PLAT_58856_wa && (!dc_is_dp_signal(stream->signal))) + dc->hwseq->funcs.PLAT_58856_wa(context, pipe_ctx); + pipe_ctx->stream_res.tg->funcs->program_timing( pipe_ctx->stream_res.tg, &stream->timing, @@ -974,7 +979,8 @@ void dcn20_blank_pixel_data( stream->timing.display_color_depth, &black_color, width, - height); + height, + 0); for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) { odm_pipe->stream_res.opp->funcs->opp_set_disp_pattern_generator( @@ -985,7 +991,8 @@ void dcn20_blank_pixel_data( stream->timing.display_color_depth, &black_color, width, - height); + height, + 0); } if (!blank) @@ -1088,29 +1095,6 @@ void dcn20_enable_plane( // } } - -void dcn20_pipe_control_lock_global( - struct dc *dc, - struct pipe_ctx *pipe, - bool lock) -{ - if (lock) { - pipe->stream_res.tg->funcs->lock_doublebuffer_enable( - pipe->stream_res.tg); - pipe->stream_res.tg->funcs->lock(pipe->stream_res.tg); - } else { - pipe->stream_res.tg->funcs->unlock(pipe->stream_res.tg); - pipe->stream_res.tg->funcs->wait_for_state(pipe->stream_res.tg, - CRTC_STATE_VACTIVE); - pipe->stream_res.tg->funcs->wait_for_state(pipe->stream_res.tg, - CRTC_STATE_VBLANK); - pipe->stream_res.tg->funcs->wait_for_state(pipe->stream_res.tg, - CRTC_STATE_VACTIVE); - pipe->stream_res.tg->funcs->lock_doublebuffer_disable( - pipe->stream_res.tg); - } -} - void dcn20_pipe_control_lock( struct dc *dc, struct pipe_ctx *pipe, @@ -1121,7 +1105,7 @@ void dcn20_pipe_control_lock( /* use TG master update lock to lock everything on the TG * therefore only top pipe need to lock */ - if (pipe->top_pipe) + if (!pipe || pipe->top_pipe) return; if (pipe->plane_state != NULL) @@ -1389,6 +1373,7 @@ static void dcn20_update_dchubp_dpp( } if (pipe_ctx->update_flags.bits.viewport || + (context == dc->current_state && plane_state->update_flags.bits.position_change) || (context == dc->current_state && plane_state->update_flags.bits.scaling_change) || (context == dc->current_state && pipe_ctx->stream->update_flags.bits.scaling)) { @@ -1536,48 +1521,32 @@ static void dcn20_program_pipe( } } -static bool does_pipe_need_lock(struct pipe_ctx *pipe) -{ - if ((pipe->plane_state && pipe->plane_state->update_flags.raw) - || pipe->update_flags.raw) - return true; - if (pipe->bottom_pipe) - return does_pipe_need_lock(pipe->bottom_pipe); - - return false; -} - void dcn20_program_front_end_for_ctx( struct dc *dc, struct dc_state *context) { - const unsigned int TIMEOUT_FOR_PIPE_ENABLE_MS = 100; int i; struct dce_hwseq *hws = dc->hwseq; - bool pipe_locked[MAX_PIPES] = {false}; DC_LOGGER_INIT(dc->ctx->logger); - /* Carry over GSL groups in case the context is changing. */ - for (i = 0; i < dc->res_pool->pipe_count; i++) - if (context->res_ctx.pipe_ctx[i].stream == dc->current_state->res_ctx.pipe_ctx[i].stream) - context->res_ctx.pipe_ctx[i].stream_res.gsl_group = - dc->current_state->res_ctx.pipe_ctx[i].stream_res.gsl_group; + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; + + if (!pipe_ctx->top_pipe && !pipe_ctx->prev_odm_pipe && pipe_ctx->plane_state) { + ASSERT(!pipe_ctx->plane_state->triplebuffer_flips); + if (dc->hwss.program_triplebuffer != NULL && + !dc->debug.disable_tri_buf) { + /*turn off triple buffer for full update*/ + dc->hwss.program_triplebuffer( + dc, pipe_ctx, pipe_ctx->plane_state->triplebuffer_flips); + } + } + } /* Set pipe update flags and lock pipes */ for (i = 0; i < dc->res_pool->pipe_count; i++) dcn20_detect_pipe_changes(&dc->current_state->res_ctx.pipe_ctx[i], &context->res_ctx.pipe_ctx[i]); - for (i = 0; i < dc->res_pool->pipe_count; i++) - if (!context->res_ctx.pipe_ctx[i].top_pipe && - does_pipe_need_lock(&context->res_ctx.pipe_ctx[i])) { - struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; - - if (pipe_ctx->update_flags.bits.tg_changed || pipe_ctx->update_flags.bits.enable) - dc->hwss.pipe_control_lock(dc, pipe_ctx, true); - if (!pipe_ctx->update_flags.bits.enable) - dc->hwss.pipe_control_lock(dc, &dc->current_state->res_ctx.pipe_ctx[i], true); - pipe_locked[i] = true; - } /* OTG blank before disabling all front ends */ for (i = 0; i < dc->res_pool->pipe_count; i++) @@ -1615,17 +1584,17 @@ void dcn20_program_front_end_for_ctx( hws->funcs.program_all_writeback_pipes_in_tree(dc, pipe->stream, context); } } +} - /* Unlock all locked pipes */ - for (i = 0; i < dc->res_pool->pipe_count; i++) - if (pipe_locked[i]) { - struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; +void dcn20_post_unlock_program_front_end( + struct dc *dc, + struct dc_state *context) +{ + int i; + const unsigned int TIMEOUT_FOR_PIPE_ENABLE_MS = 100; + struct dce_hwseq *hwseq = dc->hwseq; - if (pipe_ctx->update_flags.bits.tg_changed || pipe_ctx->update_flags.bits.enable) - dc->hwss.pipe_control_lock(dc, pipe_ctx, false); - if (!pipe_ctx->update_flags.bits.enable) - dc->hwss.pipe_control_lock(dc, &dc->current_state->res_ctx.pipe_ctx[i], false); - } + DC_LOGGER_INIT(dc->ctx->logger); for (i = 0; i < dc->res_pool->pipe_count; i++) if (context->res_ctx.pipe_ctx[i].update_flags.bits.disable) @@ -1651,11 +1620,26 @@ void dcn20_program_front_end_for_ctx( } /* WA to apply WM setting*/ - if (dc->hwseq->wa.DEGVIDCN21) + if (hwseq->wa.DEGVIDCN21) dc->res_pool->hubbub->funcs->apply_DEDCN21_147_wa(dc->res_pool->hubbub); -} + /* WA for stutter underflow during MPO transitions when adding 2nd plane */ + if (hwseq->wa.disallow_self_refresh_during_multi_plane_transition) { + + if (dc->current_state->stream_status[0].plane_count == 1 && + context->stream_status[0].plane_count > 1) { + + struct timing_generator *tg = dc->res_pool->timing_generators[0]; + + dc->res_pool->hubbub->funcs->allow_self_refresh_control(dc->res_pool->hubbub, false); + + hwseq->wa_state.disallow_self_refresh_during_multi_plane_transition_applied = true; + hwseq->wa_state.disallow_self_refresh_during_multi_plane_transition_applied_on_frame = tg->funcs->get_frame_count(tg); + } + } +} + void dcn20_prepare_bandwidth( struct dc *dc, struct dc_state *context) @@ -1668,7 +1652,7 @@ void dcn20_prepare_bandwidth( false); /* program dchubbub watermarks */ - hubbub->funcs->program_watermarks(hubbub, + dc->wm_optimized_required = hubbub->funcs->program_watermarks(hubbub, &context->bw_ctx.bw.dcn.watermarks, dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000, false); @@ -2052,6 +2036,10 @@ static void dcn20_reset_back_end_for_pipe( * parent pipe. */ if (pipe_ctx->top_pipe == NULL) { + + if (pipe_ctx->stream_res.abm) + pipe_ctx->stream_res.abm->funcs->set_abm_immediate_disable(pipe_ctx->stream_res.abm); + pipe_ctx->stream_res.tg->funcs->disable_crtc(pipe_ctx->stream_res.tg); pipe_ctx->stream_res.tg->funcs->enable_optc_clock(pipe_ctx->stream_res.tg, false); diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.h index 02c9be5ebd47..63ce763f148e 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.h @@ -35,6 +35,9 @@ bool dcn20_set_shaper_3dlut( void dcn20_program_front_end_for_ctx( struct dc *dc, struct dc_state *context); +void dcn20_post_unlock_program_front_end( + struct dc *dc, + struct dc_state *context); void dcn20_update_plane_addr(const struct dc *dc, struct pipe_ctx *pipe_ctx); void dcn20_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx); bool dcn20_set_input_transfer_func(struct dc *dc, struct pipe_ctx *pipe_ctx, @@ -58,10 +61,6 @@ void dcn20_pipe_control_lock( struct dc *dc, struct pipe_ctx *pipe, bool lock); -void dcn20_pipe_control_lock_global( - struct dc *dc, - struct pipe_ctx *pipe, - bool lock); void dcn20_prepare_bandwidth( struct dc *dc, struct dc_state *context); diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c index d51e02fdab4d..1e73357eda34 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c @@ -33,6 +33,7 @@ static const struct hw_sequencer_funcs dcn20_funcs = { .apply_ctx_to_hw = dce110_apply_ctx_to_hw, .apply_ctx_for_surface = NULL, .program_front_end_for_ctx = dcn20_program_front_end_for_ctx, + .post_unlock_program_front_end = dcn20_post_unlock_program_front_end, .update_plane_addr = dcn20_update_plane_addr, .update_dchub = dcn10_update_dchub, .update_pending_status = dcn10_update_pending_status, @@ -50,7 +51,7 @@ static const struct hw_sequencer_funcs dcn20_funcs = { .disable_audio_stream = dce110_disable_audio_stream, .disable_plane = dcn20_disable_plane, .pipe_control_lock = dcn20_pipe_control_lock, - .pipe_control_lock_global = dcn20_pipe_control_lock_global, + .interdependent_update_lock = dcn10_lock_all_pipes, .prepare_bandwidth = dcn20_prepare_bandwidth, .optimize_bandwidth = dcn20_optimize_bandwidth, .update_bandwidth = dcn20_update_bandwidth, @@ -96,6 +97,8 @@ static const struct hwseq_private_funcs dcn20_private_funcs = { .reset_hw_ctx_wrap = dcn20_reset_hw_ctx_wrap, .enable_stream_timing = dcn20_enable_stream_timing, .edp_backlight_control = dce110_edp_backlight_control, + .is_panel_backlight_on = dce110_is_panel_backlight_on, + .is_panel_powered_on = dce110_is_panel_powered_on, .disable_stream_gating = dcn20_disable_stream_gating, .enable_stream_gating = dcn20_enable_stream_gating, .setup_vupdate_interrupt = dcn20_setup_vupdate_interrupt, @@ -108,7 +111,6 @@ static const struct hwseq_private_funcs dcn20_private_funcs = { .enable_power_gating_plane = dcn20_enable_power_gating_plane, .dpp_pg_control = dcn20_dpp_pg_control, .hubp_pg_control = dcn20_hubp_pg_control, - .dsc_pg_control = NULL, .update_odm = dcn20_update_odm, .dsc_pg_control = dcn20_dsc_pg_control, .get_surface_visual_confirm_color = dcn10_get_surface_visual_confirm_color, diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mmhubbub.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mmhubbub.h index 3fccd5eeecbb..7bcee5894d2e 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mmhubbub.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mmhubbub.h @@ -36,26 +36,6 @@ #define BASE(seg) \ BASE_INNER(seg) -#define SR(reg_name)\ - .reg_name = BASE(mm ## reg_name ## _BASE_IDX) + \ - mm ## reg_name - -#define SRI(reg_name, block, id)\ - .reg_name = BASE(mm ## block ## id ## _ ## reg_name ## _BASE_IDX) + \ - mm ## block ## id ## _ ## reg_name - -#define SRI2(reg_name, block, id)\ - .reg_name = BASE(mm ## reg_name ## _BASE_IDX) + \ - mm ## reg_name - -#define SRII(reg_name, block, id)\ - .reg_name[id] = BASE(mm ## block ## id ## _ ## reg_name ## _BASE_IDX) + \ - mm ## block ## id ## _ ## reg_name - -#define SF(reg_name, field_name, post_fix)\ - .field_name = reg_name ## __ ## field_name ## post_fix - - #define MCIF_WB_COMMON_REG_LIST_DCN2_0(inst) \ SRI(MCIF_WB_BUFMGR_SW_CONTROL, MCIF_WB, inst),\ SRI(MCIF_WB_BUFMGR_CUR_LINE_R, MCIF_WB, inst),\ diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.c index 023cc71fad0f..138321e151eb 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.c @@ -45,7 +45,8 @@ void opp2_set_disp_pattern_generator( enum dc_color_depth color_depth, const struct tg_color *solid_color, int width, - int height) + int height, + int offset) { struct dcn20_opp *oppn20 = TO_DCN20_OPP(opp); enum test_pattern_color_format bit_depth; @@ -92,6 +93,11 @@ void opp2_set_disp_pattern_generator( DPG_ACTIVE_WIDTH, width, DPG_ACTIVE_HEIGHT, height); + /* set DPG offset */ + REG_SET_2(DPG_OFFSET_SEGMENT, 0, + DPG_X_OFFSET, offset, + DPG_SEGMENT_WIDTH, 0); + switch (test_pattern) { case CONTROLLER_DP_TEST_PATTERN_COLORSQUARES: case CONTROLLER_DP_TEST_PATTERN_COLORSQUARES_CEA: diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.h index 4093bec172c1..64c5b429c79a 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.h @@ -36,6 +36,7 @@ #define OPP_DPG_REG_LIST(id) \ SRI(DPG_CONTROL, DPG, id), \ SRI(DPG_DIMENSIONS, DPG, id), \ + SRI(DPG_OFFSET_SEGMENT, DPG, id), \ SRI(DPG_COLOUR_B_CB, DPG, id), \ SRI(DPG_COLOUR_G_Y, DPG, id), \ SRI(DPG_COLOUR_R_CR, DPG, id), \ @@ -53,6 +54,7 @@ uint32_t FMT_422_CONTROL; \ uint32_t DPG_CONTROL; \ uint32_t DPG_DIMENSIONS; \ + uint32_t DPG_OFFSET_SEGMENT; \ uint32_t DPG_COLOUR_B_CB; \ uint32_t DPG_COLOUR_G_Y; \ uint32_t DPG_COLOUR_R_CR; \ @@ -68,6 +70,8 @@ OPP_SF(DPG0_DPG_CONTROL, DPG_HRES, mask_sh), \ OPP_SF(DPG0_DPG_DIMENSIONS, DPG_ACTIVE_WIDTH, mask_sh), \ OPP_SF(DPG0_DPG_DIMENSIONS, DPG_ACTIVE_HEIGHT, mask_sh), \ + OPP_SF(DPG0_DPG_OFFSET_SEGMENT, DPG_X_OFFSET, mask_sh), \ + OPP_SF(DPG0_DPG_OFFSET_SEGMENT, DPG_SEGMENT_WIDTH, mask_sh), \ OPP_SF(DPG0_DPG_COLOUR_R_CR, DPG_COLOUR0_R_CR, mask_sh), \ OPP_SF(DPG0_DPG_COLOUR_R_CR, DPG_COLOUR1_R_CR, mask_sh), \ OPP_SF(DPG0_DPG_COLOUR_B_CB, DPG_COLOUR0_B_CB, mask_sh), \ @@ -97,6 +101,8 @@ type DPG_HRES; \ type DPG_ACTIVE_WIDTH; \ type DPG_ACTIVE_HEIGHT; \ + type DPG_X_OFFSET; \ + type DPG_SEGMENT_WIDTH; \ type DPG_COLOUR0_R_CR; \ type DPG_COLOUR1_R_CR; \ type DPG_COLOUR0_B_CB; \ @@ -144,7 +150,8 @@ void opp2_set_disp_pattern_generator( enum dc_color_depth color_depth, const struct tg_color *solid_color, int width, - int height); + int height, + int offset); bool opp2_dpg_is_blanked(struct output_pixel_processor *opp); diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c index 85f90f3e24cb..5cdbba0cd873 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c @@ -153,6 +153,7 @@ struct _vcs_dpi_ip_params_st dcn2_0_ip = { .xfc_supported = true, .xfc_fill_bw_overhead_percent = 10.0, .xfc_fill_constant_bytes = 0, + .number_of_cursors = 1, }; struct _vcs_dpi_ip_params_st dcn2_0_nv14_ip = { @@ -220,7 +221,8 @@ struct _vcs_dpi_ip_params_st dcn2_0_nv14_ip = { .xfc_supported = true, .xfc_fill_bw_overhead_percent = 10.0, .xfc_fill_constant_bytes = 0, - .ptoi_supported = 0 + .ptoi_supported = 0, + .number_of_cursors = 1, }; struct _vcs_dpi_soc_bounding_box_st dcn2_0_soc = { @@ -335,6 +337,117 @@ struct _vcs_dpi_soc_bounding_box_st dcn2_0_soc = { .use_urgent_burst_bw = 0 }; +struct _vcs_dpi_soc_bounding_box_st dcn2_0_nv14_soc = { + .clock_limits = { + { + .state = 0, + .dcfclk_mhz = 560.0, + .fabricclk_mhz = 560.0, + .dispclk_mhz = 513.0, + .dppclk_mhz = 513.0, + .phyclk_mhz = 540.0, + .socclk_mhz = 560.0, + .dscclk_mhz = 171.0, + .dram_speed_mts = 8960.0, + }, + { + .state = 1, + .dcfclk_mhz = 694.0, + .fabricclk_mhz = 694.0, + .dispclk_mhz = 642.0, + .dppclk_mhz = 642.0, + .phyclk_mhz = 600.0, + .socclk_mhz = 694.0, + .dscclk_mhz = 214.0, + .dram_speed_mts = 11104.0, + }, + { + .state = 2, + .dcfclk_mhz = 875.0, + .fabricclk_mhz = 875.0, + .dispclk_mhz = 734.0, + .dppclk_mhz = 734.0, + .phyclk_mhz = 810.0, + .socclk_mhz = 875.0, + .dscclk_mhz = 245.0, + .dram_speed_mts = 14000.0, + }, + { + .state = 3, + .dcfclk_mhz = 1000.0, + .fabricclk_mhz = 1000.0, + .dispclk_mhz = 1100.0, + .dppclk_mhz = 1100.0, + .phyclk_mhz = 810.0, + .socclk_mhz = 1000.0, + .dscclk_mhz = 367.0, + .dram_speed_mts = 16000.0, + }, + { + .state = 4, + .dcfclk_mhz = 1200.0, + .fabricclk_mhz = 1200.0, + .dispclk_mhz = 1284.0, + .dppclk_mhz = 1284.0, + .phyclk_mhz = 810.0, + .socclk_mhz = 1200.0, + .dscclk_mhz = 428.0, + .dram_speed_mts = 16000.0, + }, + /*Extra state, no dispclk ramping*/ + { + .state = 5, + .dcfclk_mhz = 1200.0, + .fabricclk_mhz = 1200.0, + .dispclk_mhz = 1284.0, + .dppclk_mhz = 1284.0, + .phyclk_mhz = 810.0, + .socclk_mhz = 1200.0, + .dscclk_mhz = 428.0, + .dram_speed_mts = 16000.0, + }, + }, + .num_states = 5, + .sr_exit_time_us = 8.6, + .sr_enter_plus_exit_time_us = 10.9, + .urgent_latency_us = 4.0, + .urgent_latency_pixel_data_only_us = 4.0, + .urgent_latency_pixel_mixed_with_vm_data_us = 4.0, + .urgent_latency_vm_data_only_us = 4.0, + .urgent_out_of_order_return_per_channel_pixel_only_bytes = 4096, + .urgent_out_of_order_return_per_channel_pixel_and_vm_bytes = 4096, + .urgent_out_of_order_return_per_channel_vm_only_bytes = 4096, + .pct_ideal_dram_sdp_bw_after_urgent_pixel_only = 40.0, + .pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = 40.0, + .pct_ideal_dram_sdp_bw_after_urgent_vm_only = 40.0, + .max_avg_sdp_bw_use_normal_percent = 40.0, + .max_avg_dram_bw_use_normal_percent = 40.0, + .writeback_latency_us = 12.0, + .ideal_dram_bw_after_urgent_percent = 40.0, + .max_request_size_bytes = 256, + .dram_channel_width_bytes = 2, + .fabric_datapath_to_dcn_data_return_bytes = 64, + .dcn_downspread_percent = 0.5, + .downspread_percent = 0.38, + .dram_page_open_time_ns = 50.0, + .dram_rw_turnaround_time_ns = 17.5, + .dram_return_buffer_per_channel_bytes = 8192, + .round_trip_ping_latency_dcfclk_cycles = 131, + .urgent_out_of_order_return_per_channel_bytes = 256, + .channel_interleave_bytes = 256, + .num_banks = 8, + .num_chans = 8, + .vmm_page_size_bytes = 4096, + .dram_clock_change_latency_us = 404.0, + .dummy_pstate_latency_us = 5.0, + .writeback_dram_clock_change_latency_us = 23.0, + .return_bus_width_bytes = 64, + .dispclk_dppclk_vco_speed_mhz = 3850, + .xfc_bus_transport_time_us = 20, + .xfc_xbuf_latency_tolerance_us = 4, + .use_urgent_burst_bw = 0 +}; + struct _vcs_dpi_soc_bounding_box_st dcn2_0_nv12_soc = { 0 }; #ifndef mmDP0_DP_DPHY_INTERNAL_CTRL @@ -899,7 +1012,8 @@ static const struct dc_plane_cap plane_cap = { .pixel_format_support = { .argb8888 = true, .nv12 = true, - .fp16 = true + .fp16 = true, + .p010 = true }, .max_upscale_factor = { @@ -928,7 +1042,7 @@ static const struct resource_caps res_cap_nv14 = { }; static const struct dc_debug_options debug_defaults_drv = { - .disable_dmcu = true, + .disable_dmcu = false, .force_abm_enable = false, .timing_trace = false, .clock_trace = true, @@ -947,7 +1061,7 @@ static const struct dc_debug_options debug_defaults_drv = { }; static const struct dc_debug_options debug_defaults_diags = { - .disable_dmcu = true, + .disable_dmcu = false, .force_abm_enable = false, .timing_trace = true, .clock_trace = true, @@ -1143,6 +1257,7 @@ static const struct encoder_feature_support link_enc_feature = { .max_hdmi_pixel_clock = 600000, .hdmi_ycbcr420_supported = true, .dp_ycbcr420_supported = true, + .fec_supported = true, .flags.bits.IS_HBR2_CAPABLE = true, .flags.bits.IS_HBR3_CAPABLE = true, .flags.bits.IS_TPS3_CAPABLE = true, @@ -1557,7 +1672,7 @@ static void acquire_dsc(struct resource_context *res_ctx, } } -static void release_dsc(struct resource_context *res_ctx, +void dcn20_release_dsc(struct resource_context *res_ctx, const struct resource_pool *pool, struct display_stream_compressor **dsc) { @@ -1617,7 +1732,7 @@ static enum dc_status remove_dsc_from_stream_resource(struct dc *dc, pipe_ctx = &new_ctx->res_ctx.pipe_ctx[i]; if (pipe_ctx->stream_res.dsc) - release_dsc(&new_ctx->res_ctx, dc->res_pool, &pipe_ctx->stream_res.dsc); + dcn20_release_dsc(&new_ctx->res_ctx, dc->res_pool, &pipe_ctx->stream_res.dsc); } } @@ -1861,22 +1976,6 @@ void dcn20_populate_dml_writeback_from_context( } -static int get_num_odm_heads(struct pipe_ctx *pipe) -{ - int odm_head_count = 0; - struct pipe_ctx *next_pipe = pipe->next_odm_pipe; - while (next_pipe) { - odm_head_count++; - next_pipe = next_pipe->next_odm_pipe; - } - pipe = pipe->prev_odm_pipe; - while (pipe) { - odm_head_count++; - pipe = pipe->prev_odm_pipe; - } - return odm_head_count ? odm_head_count + 1 : 0; -} - int dcn20_populate_dml_pipes_from_context( struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes) { @@ -1956,8 +2055,8 @@ int dcn20_populate_dml_pipes_from_context( pipes[pipe_cnt].dout.dp_lanes = 4; pipes[pipe_cnt].pipe.dest.vtotal_min = res_ctx->pipe_ctx[i].stream->adjust.v_total_min; pipes[pipe_cnt].pipe.dest.vtotal_max = res_ctx->pipe_ctx[i].stream->adjust.v_total_max; - switch (get_num_odm_heads(&res_ctx->pipe_ctx[i])) { - case 2: + switch (get_num_odm_splits(&res_ctx->pipe_ctx[i])) { + case 1: pipes[pipe_cnt].pipe.dest.odm_combine = dm_odm_combine_mode_2to1; break; default: @@ -1965,9 +2064,14 @@ int dcn20_populate_dml_pipes_from_context( } pipes[pipe_cnt].pipe.src.hsplit_grp = res_ctx->pipe_ctx[i].pipe_idx; if (res_ctx->pipe_ctx[i].top_pipe && res_ctx->pipe_ctx[i].top_pipe->plane_state - == res_ctx->pipe_ctx[i].plane_state) - pipes[pipe_cnt].pipe.src.hsplit_grp = res_ctx->pipe_ctx[i].top_pipe->pipe_idx; - else if (res_ctx->pipe_ctx[i].prev_odm_pipe) { + == res_ctx->pipe_ctx[i].plane_state) { + struct pipe_ctx *first_pipe = res_ctx->pipe_ctx[i].top_pipe; + + while (first_pipe->top_pipe && first_pipe->top_pipe->plane_state + == res_ctx->pipe_ctx[i].plane_state) + first_pipe = first_pipe->top_pipe; + pipes[pipe_cnt].pipe.src.hsplit_grp = first_pipe->pipe_idx; + } else if (res_ctx->pipe_ctx[i].prev_odm_pipe) { struct pipe_ctx *first_pipe = res_ctx->pipe_ctx[i].prev_odm_pipe; while (first_pipe->prev_odm_pipe) @@ -2052,16 +2156,20 @@ int dcn20_populate_dml_pipes_from_context( /* todo: default max for now, until there is logic reflecting this in dc*/ pipes[pipe_cnt].dout.output_bpc = 12; /* - * Use max cursor settings for calculations to minimize + * For graphic plane, cursor number is 1, nv12 is 0 * bw calculations due to cursor on/off */ - pipes[pipe_cnt].pipe.src.num_cursors = 2; + if (res_ctx->pipe_ctx[i].plane_state && + res_ctx->pipe_ctx[i].plane_state->address.type == PLN_ADDR_TYPE_VIDEO_PROGRESSIVE) + pipes[pipe_cnt].pipe.src.num_cursors = 0; + else + pipes[pipe_cnt].pipe.src.num_cursors = dc->dml.ip.number_of_cursors; + pipes[pipe_cnt].pipe.src.cur0_src_width = 256; pipes[pipe_cnt].pipe.src.cur0_bpp = dm_cur_32bit; - pipes[pipe_cnt].pipe.src.cur1_src_width = 256; - pipes[pipe_cnt].pipe.src.cur1_bpp = dm_cur_32bit; if (!res_ctx->pipe_ctx[i].plane_state) { + pipes[pipe_cnt].pipe.src.is_hsplit = pipes[pipe_cnt].pipe.dest.odm_combine != dm_odm_combine_mode_disabled; pipes[pipe_cnt].pipe.src.source_scan = dm_horz; pipes[pipe_cnt].pipe.src.sw_mode = dm_sw_linear; pipes[pipe_cnt].pipe.src.macro_tile_size = dm_64k_tile; @@ -2087,19 +2195,21 @@ int dcn20_populate_dml_pipes_from_context( pipes[pipe_cnt].pipe.scale_ratio_depth.scl_enable = 0; /*Lb only or Full scl*/ pipes[pipe_cnt].pipe.scale_taps.htaps = 1; pipes[pipe_cnt].pipe.scale_taps.vtaps = 1; - pipes[pipe_cnt].pipe.src.is_hsplit = 0; - pipes[pipe_cnt].pipe.dest.odm_combine = 0; pipes[pipe_cnt].pipe.dest.vtotal_min = v_total; pipes[pipe_cnt].pipe.dest.vtotal_max = v_total; + + if (pipes[pipe_cnt].pipe.dest.odm_combine == dm_odm_combine_mode_2to1) { + pipes[pipe_cnt].pipe.src.viewport_width /= 2; + pipes[pipe_cnt].pipe.dest.recout_width /= 2; + } } else { struct dc_plane_state *pln = res_ctx->pipe_ctx[i].plane_state; struct scaler_data *scl = &res_ctx->pipe_ctx[i].plane_res.scl_data; pipes[pipe_cnt].pipe.src.immediate_flip = pln->flip_immediate; - pipes[pipe_cnt].pipe.src.is_hsplit = (res_ctx->pipe_ctx[i].bottom_pipe - && res_ctx->pipe_ctx[i].bottom_pipe->plane_state == pln) - || (res_ctx->pipe_ctx[i].top_pipe - && res_ctx->pipe_ctx[i].top_pipe->plane_state == pln); + pipes[pipe_cnt].pipe.src.is_hsplit = (res_ctx->pipe_ctx[i].bottom_pipe && res_ctx->pipe_ctx[i].bottom_pipe->plane_state == pln) + || (res_ctx->pipe_ctx[i].top_pipe && res_ctx->pipe_ctx[i].top_pipe->plane_state == pln) + || pipes[pipe_cnt].pipe.dest.odm_combine != dm_odm_combine_mode_disabled; pipes[pipe_cnt].pipe.src.source_scan = pln->rotation == ROTATION_ANGLE_90 || pln->rotation == ROTATION_ANGLE_270 ? dm_vert : dm_horz; pipes[pipe_cnt].pipe.src.viewport_y_y = scl->viewport.y; @@ -2124,18 +2234,22 @@ int dcn20_populate_dml_pipes_from_context( pipes[pipe_cnt].pipe.src.dcc = pln->dcc.enable; pipes[pipe_cnt].pipe.dest.recout_width = scl->recout.width; pipes[pipe_cnt].pipe.dest.recout_height = scl->recout.height; - pipes[pipe_cnt].pipe.dest.full_recout_width = scl->recout.width; pipes[pipe_cnt].pipe.dest.full_recout_height = scl->recout.height; - if (res_ctx->pipe_ctx[i].bottom_pipe && res_ctx->pipe_ctx[i].bottom_pipe->plane_state == pln) { - pipes[pipe_cnt].pipe.dest.full_recout_width += - res_ctx->pipe_ctx[i].bottom_pipe->plane_res.scl_data.recout.width; - pipes[pipe_cnt].pipe.dest.full_recout_height += - res_ctx->pipe_ctx[i].bottom_pipe->plane_res.scl_data.recout.height; - } else if (res_ctx->pipe_ctx[i].top_pipe && res_ctx->pipe_ctx[i].top_pipe->plane_state == pln) { - pipes[pipe_cnt].pipe.dest.full_recout_width += - res_ctx->pipe_ctx[i].top_pipe->plane_res.scl_data.recout.width; - pipes[pipe_cnt].pipe.dest.full_recout_height += - res_ctx->pipe_ctx[i].top_pipe->plane_res.scl_data.recout.height; + pipes[pipe_cnt].pipe.dest.full_recout_width = scl->recout.width; + if (pipes[pipe_cnt].pipe.dest.odm_combine == dm_odm_combine_mode_2to1) + pipes[pipe_cnt].pipe.dest.full_recout_width *= 2; + else { + struct pipe_ctx *split_pipe = res_ctx->pipe_ctx[i].bottom_pipe; + + while (split_pipe && split_pipe->plane_state == pln) { + pipes[pipe_cnt].pipe.dest.full_recout_width += split_pipe->plane_res.scl_data.recout.width; + split_pipe = split_pipe->bottom_pipe; + } + split_pipe = res_ctx->pipe_ctx[i].top_pipe; + while (split_pipe && split_pipe->plane_state == pln) { + pipes[pipe_cnt].pipe.dest.full_recout_width += split_pipe->plane_res.scl_data.recout.width; + split_pipe = split_pipe->top_pipe; + } } pipes[pipe_cnt].pipe.scale_ratio_depth.lb_depth = dm_lb_16; @@ -2302,6 +2416,7 @@ bool dcn20_validate_dsc(struct dc *dc, struct dc_state *new_ctx) + stream->timing.v_border_bottom; dsc_cfg.pixel_encoding = stream->timing.pixel_encoding; dsc_cfg.color_depth = stream->timing.display_color_depth; + dsc_cfg.is_odm = pipe_ctx->next_odm_pipe ? true : false; dsc_cfg.dc_dsc_cfg = stream->timing.dsc_cfg; dsc_cfg.dc_dsc_cfg.num_slices_h /= opp_cnt; @@ -2388,7 +2503,7 @@ struct pipe_ctx *dcn20_find_secondary_pipe(struct dc *dc, return secondary_pipe; } -void dcn20_merge_pipes_for_validate( +static void dcn20_merge_pipes_for_validate( struct dc *dc, struct dc_state *context) { @@ -2413,7 +2528,7 @@ void dcn20_merge_pipes_for_validate( odm_pipe->prev_odm_pipe = NULL; odm_pipe->next_odm_pipe = NULL; if (odm_pipe->stream_res.dsc) - release_dsc(&context->res_ctx, dc->res_pool, &odm_pipe->stream_res.dsc); + dcn20_release_dsc(&context->res_ctx, dc->res_pool, &odm_pipe->stream_res.dsc); /* Clear plane_res and stream_res */ memset(&odm_pipe->plane_res, 0, sizeof(odm_pipe->plane_res)); memset(&odm_pipe->stream_res, 0, sizeof(odm_pipe->stream_res)); @@ -2451,41 +2566,29 @@ int dcn20_validate_apply_pipe_split_flags( struct dc *dc, struct dc_state *context, int vlevel, - bool *split) + bool *split, + bool *merge) { int i, pipe_idx, vlevel_split; + int plane_count = 0; bool force_split = false; - bool avoid_split = dc->debug.pipe_split_policy != MPC_SPLIT_DYNAMIC; + bool avoid_split = dc->debug.pipe_split_policy == MPC_SPLIT_AVOID; - /* Single display loop, exits if there is more than one display */ + if (context->stream_count > 1) { + if (dc->debug.pipe_split_policy == MPC_SPLIT_AVOID_MULT_DISP) + avoid_split = true; + } else if (dc->debug.force_single_disp_pipe_split) + force_split = true; + + /* TODO: fix dc bugs and remove this split threshold thing */ for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; - bool exit_loop = false; - - if (!pipe->stream || pipe->top_pipe) - continue; - if (dc->debug.force_single_disp_pipe_split) { - if (!force_split) - force_split = true; - else { - force_split = false; - exit_loop = true; - } - } - if (dc->debug.pipe_split_policy == MPC_SPLIT_AVOID_MULT_DISP) { - if (avoid_split) - avoid_split = false; - else { - avoid_split = true; - exit_loop = true; - } - } - if (exit_loop) - break; + if (pipe->stream && !pipe->prev_odm_pipe && + (!pipe->top_pipe || pipe->top_pipe->plane_state != pipe->plane_state)) + ++plane_count; } - /* TODO: fix dc bugs and remove this split threshold thing */ - if (context->stream_count > dc->res_pool->pipe_count / 2) + if (plane_count > dc->res_pool->pipe_count / 2) avoid_split = true; /* Avoid split loop looks for lowest voltage level that allows most unsplit pipes possible */ @@ -2508,11 +2611,12 @@ int dcn20_validate_apply_pipe_split_flags( /* Split loop sets which pipe should be split based on dml outputs and dc flags */ for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; + int pipe_plane = context->bw_ctx.dml.vba.pipe_plane[pipe_idx]; if (!context->res_ctx.pipe_ctx[i].stream) continue; - if (force_split || context->bw_ctx.dml.vba.NoOfDPP[vlevel][context->bw_ctx.dml.vba.maxMpcComb][pipe_idx] > 1) + if (force_split || context->bw_ctx.dml.vba.NoOfDPP[vlevel][context->bw_ctx.dml.vba.maxMpcComb][pipe_plane] > 1) split[i] = true; if ((pipe->stream->view_format == VIEW_3D_FORMAT_SIDE_BY_SIDE || @@ -2525,10 +2629,44 @@ int dcn20_validate_apply_pipe_split_flags( split[i] = true; if (dc->debug.force_odm_combine & (1 << pipe->stream_res.tg->inst)) { split[i] = true; - context->bw_ctx.dml.vba.ODMCombineEnablePerState[vlevel][pipe_idx] = dm_odm_combine_mode_2to1; + context->bw_ctx.dml.vba.ODMCombineEnablePerState[vlevel][pipe_plane] = dm_odm_combine_mode_2to1; + } + context->bw_ctx.dml.vba.ODMCombineEnabled[pipe_plane] = + context->bw_ctx.dml.vba.ODMCombineEnablePerState[vlevel][pipe_plane]; + + if (pipe->prev_odm_pipe && context->bw_ctx.dml.vba.ODMCombineEnabled[pipe_plane] != dm_odm_combine_mode_disabled) { + /*Already split odm pipe tree, don't try to split again*/ + split[i] = false; + split[pipe->prev_odm_pipe->pipe_idx] = false; + } else if (pipe->top_pipe && pipe->plane_state == pipe->top_pipe->plane_state + && context->bw_ctx.dml.vba.ODMCombineEnabled[pipe_plane] == dm_odm_combine_mode_disabled) { + /*Already split mpc tree, don't try to split again, assumes only 2x mpc combine*/ + split[i] = false; + split[pipe->top_pipe->pipe_idx] = false; + } else if (pipe->prev_odm_pipe || (pipe->top_pipe && pipe->plane_state == pipe->top_pipe->plane_state)) { + if (split[i] == false) { + /*Exiting mpc/odm combine*/ + merge[i] = true; + if (pipe->prev_odm_pipe) { + ASSERT(0); /*should not actually happen yet*/ + merge[pipe->prev_odm_pipe->pipe_idx] = true; + } else + merge[pipe->top_pipe->pipe_idx] = true; + } else { + /*Transition from mpc combine to odm combine or vice versa*/ + ASSERT(0); /*should not actually happen yet*/ + split[i] = true; + merge[i] = true; + if (pipe->prev_odm_pipe) { + split[pipe->prev_odm_pipe->pipe_idx] = true; + merge[pipe->prev_odm_pipe->pipe_idx] = true; + } else { + split[pipe->top_pipe->pipe_idx] = true; + merge[pipe->top_pipe->pipe_idx] = true; + } + } } - context->bw_ctx.dml.vba.ODMCombineEnabled[pipe_idx] = - context->bw_ctx.dml.vba.ODMCombineEnablePerState[vlevel][pipe_idx]; + /* Adjust dppclk when split is forced, do not bother with dispclk */ if (split[i] && context->bw_ctx.dml.vba.NoOfDPP[vlevel][context->bw_ctx.dml.vba.maxMpcComb][pipe_idx] == 1) context->bw_ctx.dml.vba.RequiredDPPCLK[vlevel][context->bw_ctx.dml.vba.maxMpcComb][pipe_idx] /= 2; @@ -2570,7 +2708,7 @@ bool dcn20_fast_validate_bw( if (vlevel > context->bw_ctx.dml.soc.num_states) goto validate_fail; - vlevel = dcn20_validate_apply_pipe_split_flags(dc, context, vlevel, split); + vlevel = dcn20_validate_apply_pipe_split_flags(dc, context, vlevel, split, NULL); /*initialize pipe_just_split_from to invalid idx*/ for (i = 0; i < MAX_PIPES; i++) @@ -2790,6 +2928,9 @@ void dcn20_calculate_dlg_params( != dm_dram_clock_change_unsupported; context->bw_ctx.bw.dcn.clk.dppclk_khz = 0; + if (context->bw_ctx.bw.dcn.clk.dispclk_khz < dc->debug.min_disp_clk_khz) + context->bw_ctx.bw.dcn.clk.dispclk_khz = dc->debug.min_disp_clk_khz; + /* * An artifact of dml pipe split/odm is that pipes get merged back together for * calculation. Therefore we need to only extract for first pipe in ascending index order @@ -3027,7 +3168,7 @@ static struct dc_cap_funcs cap_funcs = { }; -enum dc_status dcn20_get_default_swizzle_mode(struct dc_plane_state *plane_state) +enum dc_status dcn20_patch_unknown_plane_state(struct dc_plane_state *plane_state) { enum dc_status result = DC_OK; @@ -3053,7 +3194,7 @@ static struct resource_funcs dcn20_res_pool_funcs = { .add_stream_to_ctx = dcn20_add_stream_to_ctx, .remove_stream_from_ctx = dcn20_remove_stream_from_ctx, .populate_dml_writeback_from_context = dcn20_populate_dml_writeback_from_context, - .get_default_swizzle_mode = dcn20_get_default_swizzle_mode, + .patch_unknown_plane_state = dcn20_patch_unknown_plane_state, .set_mcif_arb_params = dcn20_set_mcif_arb_params, .populate_dml_pipes = dcn20_populate_dml_pipes_from_context, .find_first_free_match_stream_enc_for_link = dcn10_find_first_free_match_stream_enc_for_link @@ -3202,7 +3343,7 @@ void dcn20_cap_soc_clocks( void dcn20_update_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_st *bb, struct pp_smu_nv_clock_table *max_clocks, unsigned int *uclk_states, unsigned int num_states) { - struct _vcs_dpi_voltage_scaling_st calculated_states[MAX_CLOCK_LIMIT_STATES]; + struct _vcs_dpi_voltage_scaling_st calculated_states[DC__VOLTAGE_STATES]; int i; int num_calculated_states = 0; int min_dcfclk = 0; @@ -3291,6 +3432,9 @@ void dcn20_patch_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_st static struct _vcs_dpi_soc_bounding_box_st *get_asic_rev_soc_bb( uint32_t hw_internal_rev) { + if (ASICREV_IS_NAVI14_M(hw_internal_rev)) + return &dcn2_0_nv14_soc; + if (ASICREV_IS_NAVI12_P(hw_internal_rev)) return &dcn2_0_nv12_soc; @@ -3772,6 +3916,15 @@ static bool dcn20_resource_construct( dcn20_hw_sequencer_construct(dc); + // IF NV12, set PG function pointer to NULL. It's not that + // PG isn't supported for NV12, it's that we don't want to + // program the registers because that will cause more power + // to be consumed. We could have created dcn20_init_hw to get + // the same effect by checking ASIC rev, but there was a + // request at some point to not check ASIC rev on hw sequencer. + if (ASICREV_IS_NAVI12_P(dc->ctx->asic_id.hw_internal_rev)) + dc->hwseq->funcs.enable_power_gating_plane = NULL; + dc->caps.max_planes = pool->base.pipe_count; for (i = 0; i < dc->caps.max_planes; ++i) diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h index f5893840b79b..9d5bff9455fd 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h @@ -119,14 +119,15 @@ void dcn20_set_mcif_arb_params( display_e2e_pipe_params_st *pipes, int pipe_cnt); bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, bool fast_validate); -void dcn20_merge_pipes_for_validate( - struct dc *dc, - struct dc_state *context); int dcn20_validate_apply_pipe_split_flags( struct dc *dc, struct dc_state *context, int vlevel, - bool *split); + bool *split, + bool *merge); +void dcn20_release_dsc(struct resource_context *res_ctx, + const struct resource_pool *pool, + struct display_stream_compressor **dsc); bool dcn20_validate_dsc(struct dc *dc, struct dc_state *new_ctx); void dcn20_split_stream_for_mpc( struct resource_context *res_ctx, @@ -159,7 +160,7 @@ enum dc_status dcn20_build_mapped_resource(const struct dc *dc, struct dc_state enum dc_status dcn20_add_stream_to_ctx(struct dc *dc, struct dc_state *new_ctx, struct dc_stream_state *dc_stream); enum dc_status dcn20_add_dsc_to_stream_resource(struct dc *dc, struct dc_state *dc_ctx, struct dc_stream_state *dc_stream); enum dc_status dcn20_remove_stream_from_ctx(struct dc *dc, struct dc_state *new_ctx, struct dc_stream_state *dc_stream); -enum dc_status dcn20_get_default_swizzle_mode(struct dc_plane_state *plane_state); +enum dc_status dcn20_patch_unknown_plane_state(struct dc_plane_state *plane_state); void dcn20_patch_bounding_box( struct dc *dc, diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_stream_encoder.c index 9b70a1e7b962..99a7ef6ab878 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_stream_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_stream_encoder.c @@ -616,5 +616,6 @@ void dcn20_stream_encoder_construct( enc1->regs = regs; enc1->se_shift = se_shift; enc1->se_mask = se_mask; + enc1->base.stream_enc_inst = eng_id - ENGINE_ID_DIGA; } diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_vmid.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_vmid.h index 02fafb013fc6..f1ef46e8da5b 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_vmid.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_vmid.h @@ -34,13 +34,6 @@ #define BASE(seg) \ BASE_INNER(seg) -#define SRI(reg_name, block, id)\ - .reg_name = BASE(mm ## block ## id ## _ ## reg_name ## _BASE_IDX) + \ - mm ## block ## id ## _ ## reg_name - -#define SF(reg_name, field_name, post_fix)\ - .field_name = reg_name ## __ ## field_name ## post_fix - #define DCN20_VMID_REG_LIST(id)\ SRI(CNTL, DCN_VM_CONTEXT, id),\ SRI(PAGE_TABLE_BASE_ADDR_HI32, DCN_VM_CONTEXT, id),\ diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubbub.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubbub.c index f546260c15b7..5e2d14b897af 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubbub.c @@ -141,7 +141,7 @@ int hubbub21_init_dchub(struct hubbub *hubbub, return NUM_VMID; } -void hubbub21_program_urgent_watermarks( +bool hubbub21_program_urgent_watermarks( struct hubbub *hubbub, struct dcn_watermark_set *watermarks, unsigned int refclk_mhz, @@ -149,6 +149,7 @@ void hubbub21_program_urgent_watermarks( { struct dcn20_hubbub *hubbub1 = TO_DCN20_HUBBUB(hubbub); uint32_t prog_wm_value; + bool wm_pending = false; /* Repeat for water mark set A, B, C and D. */ /* clock state A */ @@ -163,7 +164,8 @@ void hubbub21_program_urgent_watermarks( DC_LOG_BANDWIDTH_CALCS("URGENCY_WATERMARK_A calculated =%d\n" "HW register value = 0x%x\n", watermarks->a.urgent_ns, prog_wm_value); - } + } else if (watermarks->a.urgent_ns < hubbub1->watermarks.a.urgent_ns) + wm_pending = true; /* determine the transfer time for a quantity of data for a particular requestor.*/ if (safe_to_lower || watermarks->a.frac_urg_bw_flip @@ -172,7 +174,9 @@ void hubbub21_program_urgent_watermarks( REG_SET(DCHUBBUB_ARB_FRAC_URG_BW_FLIP_A, 0, DCHUBBUB_ARB_FRAC_URG_BW_FLIP_A, watermarks->a.frac_urg_bw_flip); - } + } else if (watermarks->a.frac_urg_bw_flip + < hubbub1->watermarks.a.frac_urg_bw_flip) + wm_pending = true; if (safe_to_lower || watermarks->a.frac_urg_bw_nom > hubbub1->watermarks.a.frac_urg_bw_nom) { @@ -180,14 +184,18 @@ void hubbub21_program_urgent_watermarks( REG_SET(DCHUBBUB_ARB_FRAC_URG_BW_NOM_A, 0, DCHUBBUB_ARB_FRAC_URG_BW_NOM_A, watermarks->a.frac_urg_bw_nom); - } + } else if (watermarks->a.frac_urg_bw_nom + < hubbub1->watermarks.a.frac_urg_bw_nom) + wm_pending = true; + if (safe_to_lower || watermarks->a.urgent_latency_ns > hubbub1->watermarks.a.urgent_latency_ns) { hubbub1->watermarks.a.urgent_latency_ns = watermarks->a.urgent_latency_ns; prog_wm_value = convert_and_clamp(watermarks->a.urgent_latency_ns, refclk_mhz, 0x1fffff); REG_SET(DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_A, 0, DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_A, prog_wm_value); - } + } else if (watermarks->a.urgent_latency_ns < hubbub1->watermarks.a.urgent_latency_ns) + wm_pending = true; /* clock state B */ if (safe_to_lower || watermarks->b.urgent_ns > hubbub1->watermarks.b.urgent_ns) { @@ -201,7 +209,8 @@ void hubbub21_program_urgent_watermarks( DC_LOG_BANDWIDTH_CALCS("URGENCY_WATERMARK_B calculated =%d\n" "HW register value = 0x%x\n", watermarks->b.urgent_ns, prog_wm_value); - } + } else if (watermarks->b.urgent_ns < hubbub1->watermarks.b.urgent_ns) + wm_pending = true; /* determine the transfer time for a quantity of data for a particular requestor.*/ if (safe_to_lower || watermarks->a.frac_urg_bw_flip @@ -210,7 +219,9 @@ void hubbub21_program_urgent_watermarks( REG_SET(DCHUBBUB_ARB_FRAC_URG_BW_FLIP_B, 0, DCHUBBUB_ARB_FRAC_URG_BW_FLIP_B, watermarks->a.frac_urg_bw_flip); - } + } else if (watermarks->a.frac_urg_bw_flip + < hubbub1->watermarks.a.frac_urg_bw_flip) + wm_pending = true; if (safe_to_lower || watermarks->a.frac_urg_bw_nom > hubbub1->watermarks.a.frac_urg_bw_nom) { @@ -218,7 +229,9 @@ void hubbub21_program_urgent_watermarks( REG_SET(DCHUBBUB_ARB_FRAC_URG_BW_NOM_B, 0, DCHUBBUB_ARB_FRAC_URG_BW_NOM_B, watermarks->a.frac_urg_bw_nom); - } + } else if (watermarks->a.frac_urg_bw_nom + < hubbub1->watermarks.a.frac_urg_bw_nom) + wm_pending = true; if (safe_to_lower || watermarks->b.urgent_latency_ns > hubbub1->watermarks.b.urgent_latency_ns) { hubbub1->watermarks.b.urgent_latency_ns = watermarks->b.urgent_latency_ns; @@ -226,7 +239,8 @@ void hubbub21_program_urgent_watermarks( refclk_mhz, 0x1fffff); REG_SET(DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_B, 0, DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_B, prog_wm_value); - } + } else if (watermarks->b.urgent_latency_ns < hubbub1->watermarks.b.urgent_latency_ns) + wm_pending = true; /* clock state C */ if (safe_to_lower || watermarks->c.urgent_ns > hubbub1->watermarks.c.urgent_ns) { @@ -240,7 +254,8 @@ void hubbub21_program_urgent_watermarks( DC_LOG_BANDWIDTH_CALCS("URGENCY_WATERMARK_C calculated =%d\n" "HW register value = 0x%x\n", watermarks->c.urgent_ns, prog_wm_value); - } + } else if (watermarks->c.urgent_ns < hubbub1->watermarks.c.urgent_ns) + wm_pending = true; /* determine the transfer time for a quantity of data for a particular requestor.*/ if (safe_to_lower || watermarks->a.frac_urg_bw_flip @@ -249,7 +264,9 @@ void hubbub21_program_urgent_watermarks( REG_SET(DCHUBBUB_ARB_FRAC_URG_BW_FLIP_C, 0, DCHUBBUB_ARB_FRAC_URG_BW_FLIP_C, watermarks->a.frac_urg_bw_flip); - } + } else if (watermarks->a.frac_urg_bw_flip + < hubbub1->watermarks.a.frac_urg_bw_flip) + wm_pending = true; if (safe_to_lower || watermarks->a.frac_urg_bw_nom > hubbub1->watermarks.a.frac_urg_bw_nom) { @@ -257,7 +274,9 @@ void hubbub21_program_urgent_watermarks( REG_SET(DCHUBBUB_ARB_FRAC_URG_BW_NOM_C, 0, DCHUBBUB_ARB_FRAC_URG_BW_NOM_C, watermarks->a.frac_urg_bw_nom); - } + } else if (watermarks->a.frac_urg_bw_nom + < hubbub1->watermarks.a.frac_urg_bw_nom) + wm_pending = true; if (safe_to_lower || watermarks->c.urgent_latency_ns > hubbub1->watermarks.c.urgent_latency_ns) { hubbub1->watermarks.c.urgent_latency_ns = watermarks->c.urgent_latency_ns; @@ -265,7 +284,8 @@ void hubbub21_program_urgent_watermarks( refclk_mhz, 0x1fffff); REG_SET(DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_C, 0, DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_C, prog_wm_value); - } + } else if (watermarks->c.urgent_latency_ns < hubbub1->watermarks.c.urgent_latency_ns) + wm_pending = true; /* clock state D */ if (safe_to_lower || watermarks->d.urgent_ns > hubbub1->watermarks.d.urgent_ns) { @@ -279,7 +299,8 @@ void hubbub21_program_urgent_watermarks( DC_LOG_BANDWIDTH_CALCS("URGENCY_WATERMARK_D calculated =%d\n" "HW register value = 0x%x\n", watermarks->d.urgent_ns, prog_wm_value); - } + } else if (watermarks->d.urgent_ns < hubbub1->watermarks.d.urgent_ns) + wm_pending = true; /* determine the transfer time for a quantity of data for a particular requestor.*/ if (safe_to_lower || watermarks->a.frac_urg_bw_flip @@ -288,7 +309,9 @@ void hubbub21_program_urgent_watermarks( REG_SET(DCHUBBUB_ARB_FRAC_URG_BW_FLIP_D, 0, DCHUBBUB_ARB_FRAC_URG_BW_FLIP_D, watermarks->a.frac_urg_bw_flip); - } + } else if (watermarks->a.frac_urg_bw_flip + < hubbub1->watermarks.a.frac_urg_bw_flip) + wm_pending = true; if (safe_to_lower || watermarks->a.frac_urg_bw_nom > hubbub1->watermarks.a.frac_urg_bw_nom) { @@ -296,7 +319,9 @@ void hubbub21_program_urgent_watermarks( REG_SET(DCHUBBUB_ARB_FRAC_URG_BW_NOM_D, 0, DCHUBBUB_ARB_FRAC_URG_BW_NOM_D, watermarks->a.frac_urg_bw_nom); - } + } else if (watermarks->a.frac_urg_bw_nom + < hubbub1->watermarks.a.frac_urg_bw_nom) + wm_pending = true; if (safe_to_lower || watermarks->d.urgent_latency_ns > hubbub1->watermarks.d.urgent_latency_ns) { hubbub1->watermarks.d.urgent_latency_ns = watermarks->d.urgent_latency_ns; @@ -304,10 +329,13 @@ void hubbub21_program_urgent_watermarks( refclk_mhz, 0x1fffff); REG_SET(DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_D, 0, DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_D, prog_wm_value); - } + } else if (watermarks->d.urgent_latency_ns < hubbub1->watermarks.d.urgent_latency_ns) + wm_pending = true; + + return wm_pending; } -void hubbub21_program_stutter_watermarks( +bool hubbub21_program_stutter_watermarks( struct hubbub *hubbub, struct dcn_watermark_set *watermarks, unsigned int refclk_mhz, @@ -315,6 +343,7 @@ void hubbub21_program_stutter_watermarks( { struct dcn20_hubbub *hubbub1 = TO_DCN20_HUBBUB(hubbub); uint32_t prog_wm_value; + bool wm_pending = false; /* clock state A */ if (safe_to_lower || watermarks->a.cstate_pstate.cstate_enter_plus_exit_ns @@ -330,7 +359,9 @@ void hubbub21_program_stutter_watermarks( DC_LOG_BANDWIDTH_CALCS("SR_ENTER_EXIT_WATERMARK_A calculated =%d\n" "HW register value = 0x%x\n", watermarks->a.cstate_pstate.cstate_enter_plus_exit_ns, prog_wm_value); - } + } else if (watermarks->a.cstate_pstate.cstate_enter_plus_exit_ns + < hubbub1->watermarks.a.cstate_pstate.cstate_enter_plus_exit_ns) + wm_pending = true; if (safe_to_lower || watermarks->a.cstate_pstate.cstate_exit_ns > hubbub1->watermarks.a.cstate_pstate.cstate_exit_ns) { @@ -345,7 +376,9 @@ void hubbub21_program_stutter_watermarks( DC_LOG_BANDWIDTH_CALCS("SR_EXIT_WATERMARK_A calculated =%d\n" "HW register value = 0x%x\n", watermarks->a.cstate_pstate.cstate_exit_ns, prog_wm_value); - } + } else if (watermarks->a.cstate_pstate.cstate_exit_ns + < hubbub1->watermarks.a.cstate_pstate.cstate_exit_ns) + wm_pending = true; /* clock state B */ if (safe_to_lower || watermarks->b.cstate_pstate.cstate_enter_plus_exit_ns @@ -361,7 +394,9 @@ void hubbub21_program_stutter_watermarks( DC_LOG_BANDWIDTH_CALCS("SR_ENTER_EXIT_WATERMARK_B calculated =%d\n" "HW register value = 0x%x\n", watermarks->b.cstate_pstate.cstate_enter_plus_exit_ns, prog_wm_value); - } + } else if (watermarks->b.cstate_pstate.cstate_enter_plus_exit_ns + < hubbub1->watermarks.b.cstate_pstate.cstate_enter_plus_exit_ns) + wm_pending = true; if (safe_to_lower || watermarks->b.cstate_pstate.cstate_exit_ns > hubbub1->watermarks.b.cstate_pstate.cstate_exit_ns) { @@ -376,7 +411,9 @@ void hubbub21_program_stutter_watermarks( DC_LOG_BANDWIDTH_CALCS("SR_EXIT_WATERMARK_B calculated =%d\n" "HW register value = 0x%x\n", watermarks->b.cstate_pstate.cstate_exit_ns, prog_wm_value); - } + } else if (watermarks->b.cstate_pstate.cstate_exit_ns + < hubbub1->watermarks.b.cstate_pstate.cstate_exit_ns) + wm_pending = true; /* clock state C */ if (safe_to_lower || watermarks->c.cstate_pstate.cstate_enter_plus_exit_ns @@ -392,7 +429,9 @@ void hubbub21_program_stutter_watermarks( DC_LOG_BANDWIDTH_CALCS("SR_ENTER_EXIT_WATERMARK_C calculated =%d\n" "HW register value = 0x%x\n", watermarks->c.cstate_pstate.cstate_enter_plus_exit_ns, prog_wm_value); - } + } else if (watermarks->c.cstate_pstate.cstate_enter_plus_exit_ns + < hubbub1->watermarks.c.cstate_pstate.cstate_enter_plus_exit_ns) + wm_pending = true; if (safe_to_lower || watermarks->c.cstate_pstate.cstate_exit_ns > hubbub1->watermarks.c.cstate_pstate.cstate_exit_ns) { @@ -407,7 +446,9 @@ void hubbub21_program_stutter_watermarks( DC_LOG_BANDWIDTH_CALCS("SR_EXIT_WATERMARK_C calculated =%d\n" "HW register value = 0x%x\n", watermarks->c.cstate_pstate.cstate_exit_ns, prog_wm_value); - } + } else if (watermarks->c.cstate_pstate.cstate_exit_ns + < hubbub1->watermarks.c.cstate_pstate.cstate_exit_ns) + wm_pending = true; /* clock state D */ if (safe_to_lower || watermarks->d.cstate_pstate.cstate_enter_plus_exit_ns @@ -423,7 +464,9 @@ void hubbub21_program_stutter_watermarks( DC_LOG_BANDWIDTH_CALCS("SR_ENTER_EXIT_WATERMARK_D calculated =%d\n" "HW register value = 0x%x\n", watermarks->d.cstate_pstate.cstate_enter_plus_exit_ns, prog_wm_value); - } + } else if (watermarks->d.cstate_pstate.cstate_enter_plus_exit_ns + < hubbub1->watermarks.d.cstate_pstate.cstate_enter_plus_exit_ns) + wm_pending = true; if (safe_to_lower || watermarks->d.cstate_pstate.cstate_exit_ns > hubbub1->watermarks.d.cstate_pstate.cstate_exit_ns) { @@ -438,10 +481,14 @@ void hubbub21_program_stutter_watermarks( DC_LOG_BANDWIDTH_CALCS("SR_EXIT_WATERMARK_D calculated =%d\n" "HW register value = 0x%x\n", watermarks->d.cstate_pstate.cstate_exit_ns, prog_wm_value); - } + } else if (watermarks->d.cstate_pstate.cstate_exit_ns + < hubbub1->watermarks.d.cstate_pstate.cstate_exit_ns) + wm_pending = true; + + return wm_pending; } -void hubbub21_program_pstate_watermarks( +bool hubbub21_program_pstate_watermarks( struct hubbub *hubbub, struct dcn_watermark_set *watermarks, unsigned int refclk_mhz, @@ -450,6 +497,8 @@ void hubbub21_program_pstate_watermarks( struct dcn20_hubbub *hubbub1 = TO_DCN20_HUBBUB(hubbub); uint32_t prog_wm_value; + bool wm_pending = false; + /* clock state A */ if (safe_to_lower || watermarks->a.cstate_pstate.pstate_change_ns > hubbub1->watermarks.a.cstate_pstate.pstate_change_ns) { @@ -464,7 +513,9 @@ void hubbub21_program_pstate_watermarks( DC_LOG_BANDWIDTH_CALCS("DRAM_CLK_CHANGE_WATERMARK_A calculated =%d\n" "HW register value = 0x%x\n\n", watermarks->a.cstate_pstate.pstate_change_ns, prog_wm_value); - } + } else if (watermarks->a.cstate_pstate.pstate_change_ns + < hubbub1->watermarks.a.cstate_pstate.pstate_change_ns) + wm_pending = true; /* clock state B */ if (safe_to_lower || watermarks->b.cstate_pstate.pstate_change_ns @@ -480,7 +531,9 @@ void hubbub21_program_pstate_watermarks( DC_LOG_BANDWIDTH_CALCS("DRAM_CLK_CHANGE_WATERMARK_B calculated =%d\n" "HW register value = 0x%x\n\n", watermarks->b.cstate_pstate.pstate_change_ns, prog_wm_value); - } + } else if (watermarks->b.cstate_pstate.pstate_change_ns + < hubbub1->watermarks.b.cstate_pstate.pstate_change_ns) + wm_pending = false; /* clock state C */ if (safe_to_lower || watermarks->c.cstate_pstate.pstate_change_ns @@ -496,7 +549,9 @@ void hubbub21_program_pstate_watermarks( DC_LOG_BANDWIDTH_CALCS("DRAM_CLK_CHANGE_WATERMARK_C calculated =%d\n" "HW register value = 0x%x\n\n", watermarks->c.cstate_pstate.pstate_change_ns, prog_wm_value); - } + } else if (watermarks->c.cstate_pstate.pstate_change_ns + < hubbub1->watermarks.c.cstate_pstate.pstate_change_ns) + wm_pending = true; /* clock state D */ if (safe_to_lower || watermarks->d.cstate_pstate.pstate_change_ns @@ -512,20 +567,30 @@ void hubbub21_program_pstate_watermarks( DC_LOG_BANDWIDTH_CALCS("DRAM_CLK_CHANGE_WATERMARK_D calculated =%d\n" "HW register value = 0x%x\n\n", watermarks->d.cstate_pstate.pstate_change_ns, prog_wm_value); - } + } else if (watermarks->d.cstate_pstate.pstate_change_ns + < hubbub1->watermarks.d.cstate_pstate.pstate_change_ns) + wm_pending = true; + + return wm_pending; } -void hubbub21_program_watermarks( +bool hubbub21_program_watermarks( struct hubbub *hubbub, struct dcn_watermark_set *watermarks, unsigned int refclk_mhz, bool safe_to_lower) { struct dcn20_hubbub *hubbub1 = TO_DCN20_HUBBUB(hubbub); + bool wm_pending = false; + + if (hubbub21_program_urgent_watermarks(hubbub, watermarks, refclk_mhz, safe_to_lower)) + wm_pending = true; - hubbub21_program_urgent_watermarks(hubbub, watermarks, refclk_mhz, safe_to_lower); - hubbub21_program_stutter_watermarks(hubbub, watermarks, refclk_mhz, safe_to_lower); - hubbub21_program_pstate_watermarks(hubbub, watermarks, refclk_mhz, safe_to_lower); + if (hubbub21_program_stutter_watermarks(hubbub, watermarks, refclk_mhz, safe_to_lower)) + wm_pending = true; + + if (hubbub21_program_pstate_watermarks(hubbub, watermarks, refclk_mhz, safe_to_lower)) + wm_pending = true; /* * The DCHub arbiter has a mechanism to dynamically rate limit the DCHub request stream to the fabric. @@ -549,6 +614,8 @@ void hubbub21_program_watermarks( DCHUBBUB_ARB_MAX_QOS_COMMIT_THRESHOLD, 0xF); hubbub1_allow_self_refresh_control(hubbub, !hubbub->ctx->dc->debug.disable_stutter); + + return wm_pending; } void hubbub21_wm_read_state(struct hubbub *hubbub, @@ -635,6 +702,7 @@ static const struct hubbub_funcs hubbub21_funcs = { .wm_read_state = hubbub21_wm_read_state, .get_dchub_ref_freq = hubbub2_get_dchub_ref_freq, .program_watermarks = hubbub21_program_watermarks, + .allow_self_refresh_control = hubbub1_allow_self_refresh_control, .apply_DEDCN21_147_wa = hubbub21_apply_DEDCN21_147_wa, }; diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubbub.h b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubbub.h index c4840dfb1fa5..ef3ef28509ed 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubbub.h +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubbub.h @@ -113,22 +113,22 @@ void dcn21_dchvm_init(struct hubbub *hubbub); int hubbub21_init_dchub(struct hubbub *hubbub, struct dcn_hubbub_phys_addr_config *pa_config); -void hubbub21_program_watermarks( +bool hubbub21_program_watermarks( struct hubbub *hubbub, struct dcn_watermark_set *watermarks, unsigned int refclk_mhz, bool safe_to_lower); -void hubbub21_program_urgent_watermarks( +bool hubbub21_program_urgent_watermarks( struct hubbub *hubbub, struct dcn_watermark_set *watermarks, unsigned int refclk_mhz, bool safe_to_lower); -void hubbub21_program_stutter_watermarks( +bool hubbub21_program_stutter_watermarks( struct hubbub *hubbub, struct dcn_watermark_set *watermarks, unsigned int refclk_mhz, bool safe_to_lower); -void hubbub21_program_pstate_watermarks( +bool hubbub21_program_pstate_watermarks( struct hubbub *hubbub, struct dcn_watermark_set *watermarks, unsigned int refclk_mhz, diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.c index cf09b9335728..d285ba622d61 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.c @@ -79,32 +79,47 @@ void apply_DEDCN21_142_wa_for_hostvm_deadline( struct _vcs_dpi_display_dlg_regs_st *dlg_attr) { struct dcn21_hubp *hubp21 = TO_DCN21_HUBP(hubp); - uint32_t cur_value; + uint32_t refcyc_per_vm_group_vblank; + uint32_t refcyc_per_vm_req_vblank; + uint32_t refcyc_per_vm_group_flip; + uint32_t refcyc_per_vm_req_flip; + const uint32_t uninitialized_hw_default = 0; - REG_GET(VBLANK_PARAMETERS_5, REFCYC_PER_VM_GROUP_VBLANK, &cur_value); - if (cur_value > dlg_attr->refcyc_per_vm_group_vblank) + REG_GET(VBLANK_PARAMETERS_5, + REFCYC_PER_VM_GROUP_VBLANK, &refcyc_per_vm_group_vblank); + + if (refcyc_per_vm_group_vblank == uninitialized_hw_default || + refcyc_per_vm_group_vblank > dlg_attr->refcyc_per_vm_group_vblank) REG_SET(VBLANK_PARAMETERS_5, 0, REFCYC_PER_VM_GROUP_VBLANK, dlg_attr->refcyc_per_vm_group_vblank); REG_GET(VBLANK_PARAMETERS_6, - REFCYC_PER_VM_REQ_VBLANK, - &cur_value); - if (cur_value > dlg_attr->refcyc_per_vm_req_vblank) + REFCYC_PER_VM_REQ_VBLANK, &refcyc_per_vm_req_vblank); + + if (refcyc_per_vm_req_vblank == uninitialized_hw_default || + refcyc_per_vm_req_vblank > dlg_attr->refcyc_per_vm_req_vblank) REG_SET(VBLANK_PARAMETERS_6, 0, REFCYC_PER_VM_REQ_VBLANK, dlg_attr->refcyc_per_vm_req_vblank); - REG_GET(FLIP_PARAMETERS_3, REFCYC_PER_VM_GROUP_FLIP, &cur_value); - if (cur_value > dlg_attr->refcyc_per_vm_group_flip) + REG_GET(FLIP_PARAMETERS_3, + REFCYC_PER_VM_GROUP_FLIP, &refcyc_per_vm_group_flip); + + if (refcyc_per_vm_group_flip == uninitialized_hw_default || + refcyc_per_vm_group_flip > dlg_attr->refcyc_per_vm_group_flip) REG_SET(FLIP_PARAMETERS_3, 0, REFCYC_PER_VM_GROUP_FLIP, dlg_attr->refcyc_per_vm_group_flip); - REG_GET(FLIP_PARAMETERS_4, REFCYC_PER_VM_REQ_FLIP, &cur_value); - if (cur_value > dlg_attr->refcyc_per_vm_req_flip) + REG_GET(FLIP_PARAMETERS_4, + REFCYC_PER_VM_REQ_FLIP, &refcyc_per_vm_req_flip); + + if (refcyc_per_vm_req_flip == uninitialized_hw_default || + refcyc_per_vm_req_flip > dlg_attr->refcyc_per_vm_req_flip) REG_SET(FLIP_PARAMETERS_4, 0, REFCYC_PER_VM_REQ_FLIP, dlg_attr->refcyc_per_vm_req_flip); REG_SET(FLIP_PARAMETERS_5, 0, REFCYC_PER_PTE_GROUP_FLIP_C, dlg_attr->refcyc_per_pte_group_flip_c); + REG_SET(FLIP_PARAMETERS_6, 0, REFCYC_PER_META_CHUNK_FLIP_C, dlg_attr->refcyc_per_meta_chunk_flip_c); } @@ -325,13 +340,9 @@ void hubp21_set_vm_system_aperture_settings(struct hubp *hubp, { struct dcn21_hubp *hubp21 = TO_DCN21_HUBP(hubp); - PHYSICAL_ADDRESS_LOC mc_vm_apt_default; PHYSICAL_ADDRESS_LOC mc_vm_apt_low; PHYSICAL_ADDRESS_LOC mc_vm_apt_high; - // The format of default addr is 48:12 of the 48 bit addr - mc_vm_apt_default.quad_part = apt->sys_default.quad_part >> 12; - // The format of high/low are 48:18 of the 48 bit addr mc_vm_apt_low.quad_part = apt->sys_low.quad_part >> 18; mc_vm_apt_high.quad_part = apt->sys_high.quad_part >> 18; diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.c index 081ad8e43d58..ada65b1a7eb1 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.c @@ -112,3 +112,25 @@ void dcn21_optimize_pwr_state( true); } +/* If user hotplug a HDMI monitor while in monitor off, + * OS will do a mode set (with output timing) but keep output off. + * In this case DAL will ask vbios to power up the pll in the PHY. + * If user unplug the monitor (while we are on monitor off) or + * system attempt to enter modern standby (which we will disable PLL), + * PHY will hang on the next mode set attempt. + * if enable PLL follow by disable PLL (without executing lane enable/disable), + * RDPCS_PHY_DP_MPLLB_STATE remains 1, + * which indicate that PLL disable attempt actually didn’t go through. + * As a workaround, insert PHY lane enable/disable before PLL disable. + */ +void dcn21_PLAT_58856_wa(struct dc_state *context, struct pipe_ctx *pipe_ctx) +{ + if (!pipe_ctx->stream->dpms_off) + return; + + pipe_ctx->stream->dpms_off = false; + core_link_enable_stream(context, pipe_ctx); + core_link_disable_stream(pipe_ctx); + pipe_ctx->stream->dpms_off = true; +} + diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.h b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.h index 182736096123..26bf24d3b59f 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.h @@ -44,4 +44,7 @@ void dcn21_optimize_pwr_state( const struct dc *dc, struct dc_state *context); +void dcn21_PLAT_58856_wa(struct dc_state *context, + struct pipe_ctx *pipe_ctx); + #endif /* __DC_HWSS_DCN21_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c index 4861aa5c59ae..b9ff9767e08f 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c @@ -34,6 +34,7 @@ static const struct hw_sequencer_funcs dcn21_funcs = { .apply_ctx_to_hw = dce110_apply_ctx_to_hw, .apply_ctx_for_surface = NULL, .program_front_end_for_ctx = dcn20_program_front_end_for_ctx, + .post_unlock_program_front_end = dcn20_post_unlock_program_front_end, .update_plane_addr = dcn20_update_plane_addr, .update_dchub = dcn10_update_dchub, .update_pending_status = dcn10_update_pending_status, @@ -51,7 +52,7 @@ static const struct hw_sequencer_funcs dcn21_funcs = { .disable_audio_stream = dce110_disable_audio_stream, .disable_plane = dcn20_disable_plane, .pipe_control_lock = dcn20_pipe_control_lock, - .pipe_control_lock_global = dcn20_pipe_control_lock_global, + .interdependent_update_lock = dcn10_lock_all_pipes, .prepare_bandwidth = dcn20_prepare_bandwidth, .optimize_bandwidth = dcn20_optimize_bandwidth, .update_bandwidth = dcn20_update_bandwidth, @@ -104,6 +105,8 @@ static const struct hwseq_private_funcs dcn21_private_funcs = { .reset_hw_ctx_wrap = dcn20_reset_hw_ctx_wrap, .enable_stream_timing = dcn20_enable_stream_timing, .edp_backlight_control = dce110_edp_backlight_control, + .is_panel_backlight_on = dce110_is_panel_backlight_on, + .is_panel_powered_on = dce110_is_panel_powered_on, .disable_stream_gating = dcn20_disable_stream_gating, .enable_stream_gating = dcn20_enable_stream_gating, .setup_vupdate_interrupt = dcn20_setup_vupdate_interrupt, @@ -116,7 +119,6 @@ static const struct hwseq_private_funcs dcn21_private_funcs = { .enable_power_gating_plane = dcn20_enable_power_gating_plane, .dpp_pg_control = dcn20_dpp_pg_control, .hubp_pg_control = dcn20_hubp_pg_control, - .dsc_pg_control = NULL, .update_odm = dcn20_update_odm, .dsc_pg_control = dcn20_dsc_pg_control, .get_surface_visual_confirm_color = dcn10_get_surface_visual_confirm_color, @@ -128,6 +130,7 @@ static const struct hwseq_private_funcs dcn21_private_funcs = { .dccg_init = dcn20_dccg_init, .set_blend_lut = dcn20_set_blend_lut, .set_shaper_3dlut = dcn20_set_shaper_3dlut, + .PLAT_58856_wa = dcn21_PLAT_58856_wa, }; void dcn21_hw_sequencer_construct(struct dc *dc) diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c index 33d0a176841a..b25484aa8222 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c @@ -84,7 +84,7 @@ #include "dcn21_resource.h" #include "vm_helper.h" #include "dcn20/dcn20_vmid.h" -#include "../dce/dmub_psr.h" +#include "dce/dmub_psr.h" #define SOC_BOUNDING_BOX_VALID false #define DC_LOGGER_INIT(logger) @@ -156,17 +156,18 @@ struct _vcs_dpi_ip_params_st dcn2_1_ip = { .xfc_supported = false, .xfc_fill_bw_overhead_percent = 10.0, .xfc_fill_constant_bytes = 0, - .ptoi_supported = 0 + .ptoi_supported = 0, + .number_of_cursors = 1, }; struct _vcs_dpi_soc_bounding_box_st dcn2_1_soc = { .clock_limits = { { .state = 0, - .dcfclk_mhz = 304.0, - .fabricclk_mhz = 600.0, - .dispclk_mhz = 618.0, - .dppclk_mhz = 440.0, + .dcfclk_mhz = 400.0, + .fabricclk_mhz = 400.0, + .dispclk_mhz = 600.0, + .dppclk_mhz = 400.00, .phyclk_mhz = 600.0, .socclk_mhz = 278.0, .dscclk_mhz = 205.67, @@ -174,10 +175,10 @@ struct _vcs_dpi_soc_bounding_box_st dcn2_1_soc = { }, { .state = 1, - .dcfclk_mhz = 304.0, - .fabricclk_mhz = 600.0, - .dispclk_mhz = 618.0, - .dppclk_mhz = 618.0, + .dcfclk_mhz = 464.52, + .fabricclk_mhz = 800.0, + .dispclk_mhz = 654.55, + .dppclk_mhz = 626.09, .phyclk_mhz = 600.0, .socclk_mhz = 278.0, .dscclk_mhz = 205.67, @@ -185,32 +186,65 @@ struct _vcs_dpi_soc_bounding_box_st dcn2_1_soc = { }, { .state = 2, - .dcfclk_mhz = 608.0, - .fabricclk_mhz = 1066.0, - .dispclk_mhz = 888.0, - .dppclk_mhz = 888.0, - .phyclk_mhz = 810.0, + .dcfclk_mhz = 514.29, + .fabricclk_mhz = 933.0, + .dispclk_mhz = 757.89, + .dppclk_mhz = 685.71, + .phyclk_mhz = 600.0, .socclk_mhz = 278.0, .dscclk_mhz = 287.67, - .dram_speed_mts = 2133.0, + .dram_speed_mts = 1866.0, }, { .state = 3, - .dcfclk_mhz = 676.0, - .fabricclk_mhz = 1600.0, - .dispclk_mhz = 1015.0, - .dppclk_mhz = 1015.0, - .phyclk_mhz = 810.0, + .dcfclk_mhz = 576.00, + .fabricclk_mhz = 1067.0, + .dispclk_mhz = 847.06, + .dppclk_mhz = 757.89, + .phyclk_mhz = 600.0, .socclk_mhz = 715.0, .dscclk_mhz = 318.334, - .dram_speed_mts = 4266.0, + .dram_speed_mts = 2134.0, }, { .state = 4, - .dcfclk_mhz = 810.0, + .dcfclk_mhz = 626.09, + .fabricclk_mhz = 1200.0, + .dispclk_mhz = 900.00, + .dppclk_mhz = 847.06, + .phyclk_mhz = 810.0, + .socclk_mhz = 953.0, + .dscclk_mhz = 489.0, + .dram_speed_mts = 2400.0, + }, + { + .state = 5, + .dcfclk_mhz = 685.71, + .fabricclk_mhz = 1333.0, + .dispclk_mhz = 1028.57, + .dppclk_mhz = 960.00, + .phyclk_mhz = 810.0, + .socclk_mhz = 278.0, + .dscclk_mhz = 287.67, + .dram_speed_mts = 2666.0, + }, + { + .state = 6, + .dcfclk_mhz = 757.89, + .fabricclk_mhz = 1467.0, + .dispclk_mhz = 1107.69, + .dppclk_mhz = 1028.57, + .phyclk_mhz = 810.0, + .socclk_mhz = 715.0, + .dscclk_mhz = 318.334, + .dram_speed_mts = 3200.0, + }, + { + .state = 7, + .dcfclk_mhz = 847.06, .fabricclk_mhz = 1600.0, .dispclk_mhz = 1395.0, - .dppclk_mhz = 1285.0, + .dppclk_mhz = 1285.00, .phyclk_mhz = 1325.0, .socclk_mhz = 953.0, .dscclk_mhz = 489.0, @@ -218,8 +252,8 @@ struct _vcs_dpi_soc_bounding_box_st dcn2_1_soc = { }, /*Extra state, no dispclk ramping*/ { - .state = 5, - .dcfclk_mhz = 810.0, + .state = 8, + .dcfclk_mhz = 847.06, .fabricclk_mhz = 1600.0, .dispclk_mhz = 1395.0, .dppclk_mhz = 1285.0, @@ -266,7 +300,7 @@ struct _vcs_dpi_soc_bounding_box_st dcn2_1_soc = { .xfc_bus_transport_time_us = 4, .xfc_xbuf_latency_tolerance_us = 4, .use_urgent_burst_bw = 1, - .num_states = 5 + .num_states = 8 }; #ifndef MAX @@ -804,7 +838,8 @@ static const struct dc_plane_cap plane_cap = { .pixel_format_support = { .argb8888 = true, .nv12 = true, - .fp16 = true + .fp16 = true, + .p010 = true }, .max_upscale_factor = { @@ -821,11 +856,12 @@ static const struct dc_plane_cap plane_cap = { }; static const struct dc_debug_options debug_defaults_drv = { - .disable_dmcu = true, + .disable_dmcu = false, .force_abm_enable = false, .timing_trace = false, .clock_trace = true, .disable_pplib_clock_request = true, + .min_disp_clk_khz = 100000, .pipe_split_policy = MPC_SPLIT_AVOID_MULT_DISP, .force_single_disp_pipe_split = false, .disable_dcc = DCC_ENABLE, @@ -841,7 +877,7 @@ static const struct dc_debug_options debug_defaults_drv = { }; static const struct dc_debug_options debug_defaults_diags = { - .disable_dmcu = true, + .disable_dmcu = false, .force_abm_enable = false, .timing_trace = true, .clock_trace = true, @@ -962,6 +998,9 @@ static void dcn21_resource_destruct(struct dcn21_resource_pool *pool) if (pool->base.dmcu != NULL) dce_dmcu_destroy(&pool->base.dmcu); + if (pool->base.psr != NULL) + dmub_psr_destroy(&pool->base.psr); + if (pool->base.dccg != NULL) dcn_dccg_destroy(&pool->base.dccg); @@ -1335,26 +1374,65 @@ static void update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_param { struct dcn21_resource_pool *pool = TO_DCN21_RES_POOL(dc->res_pool); struct clk_limit_table *clk_table = &bw_params->clk_table; - int i; - - dcn2_1_ip.max_num_otg = pool->base.res_cap->num_timing_generator; - dcn2_1_ip.max_num_dpp = pool->base.pipe_count; - dcn2_1_soc.num_chans = bw_params->num_channels; + unsigned int i, j, k; + int closest_clk_lvl; + + // Default clock levels are used for diags, which may lead to overclocking. + if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) && !IS_DIAG_DC(dc->ctx->dce_environment)) { + dcn2_1_ip.max_num_otg = pool->base.res_cap->num_timing_generator; + dcn2_1_ip.max_num_dpp = pool->base.pipe_count; + dcn2_1_soc.num_chans = bw_params->num_channels; + + /* Vmin: leave lowest DCN clocks, override with dcfclk, fclk, memclk from fuse */ + dcn2_1_soc.clock_limits[0].state = 0; + dcn2_1_soc.clock_limits[0].dcfclk_mhz = clk_table->entries[0].dcfclk_mhz; + dcn2_1_soc.clock_limits[0].fabricclk_mhz = clk_table->entries[0].fclk_mhz; + dcn2_1_soc.clock_limits[0].socclk_mhz = clk_table->entries[0].socclk_mhz; + dcn2_1_soc.clock_limits[0].dram_speed_mts = clk_table->entries[0].memclk_mhz * 2; + + /* + * Other levels: find closest DCN clocks that fit the given clock limit using dcfclk + * as indicator + */ - for (i = 0; i < clk_table->num_entries; i++) { + closest_clk_lvl = -1; + /* index currently being filled */ + k = 1; + for (i = 1; i < clk_table->num_entries; i++) { + /* loop backwards, skip duplicate state*/ + for (j = dcn2_1_soc.num_states - 1; j >= k; j--) { + if ((unsigned int) dcn2_1_soc.clock_limits[j].dcfclk_mhz <= clk_table->entries[i].dcfclk_mhz) { + closest_clk_lvl = j; + break; + } + } - dcn2_1_soc.clock_limits[i].state = i; - dcn2_1_soc.clock_limits[i].dcfclk_mhz = clk_table->entries[i].dcfclk_mhz; - dcn2_1_soc.clock_limits[i].fabricclk_mhz = clk_table->entries[i].fclk_mhz; - dcn2_1_soc.clock_limits[i].socclk_mhz = clk_table->entries[i].socclk_mhz; - dcn2_1_soc.clock_limits[i].dram_speed_mts = clk_table->entries[i].memclk_mhz * 2; + /* if found a lvl that fits, use the DCN clks from it, if not, go to next clk limit*/ + if (closest_clk_lvl != -1) { + dcn2_1_soc.clock_limits[k].state = i; + dcn2_1_soc.clock_limits[k].dcfclk_mhz = clk_table->entries[i].dcfclk_mhz; + dcn2_1_soc.clock_limits[k].fabricclk_mhz = clk_table->entries[i].fclk_mhz; + dcn2_1_soc.clock_limits[k].socclk_mhz = clk_table->entries[i].socclk_mhz; + dcn2_1_soc.clock_limits[k].dram_speed_mts = clk_table->entries[i].memclk_mhz * 2; + + dcn2_1_soc.clock_limits[k].dispclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].dispclk_mhz; + dcn2_1_soc.clock_limits[k].dppclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].dppclk_mhz; + dcn2_1_soc.clock_limits[k].dram_bw_per_chan_gbps = dcn2_1_soc.clock_limits[closest_clk_lvl].dram_bw_per_chan_gbps; + dcn2_1_soc.clock_limits[k].dscclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].dscclk_mhz; + dcn2_1_soc.clock_limits[k].dtbclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].dtbclk_mhz; + dcn2_1_soc.clock_limits[k].phyclk_d18_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].phyclk_d18_mhz; + dcn2_1_soc.clock_limits[k].phyclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].phyclk_mhz; + k++; + } + } + dcn2_1_soc.num_states = k; } - dcn2_1_soc.clock_limits[i] = dcn2_1_soc.clock_limits[i - 1]; - dcn2_1_soc.num_states = i; - // diags does not retrieve proper values from SMU, do not update DML instance for diags - if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) && !IS_DIAG_DC(dc->ctx->dce_environment)) - dml_init_instance(&dc->dml, &dcn2_1_soc, &dcn2_1_ip, DML_PROJECT_DCN21); + /* duplicate last level */ + dcn2_1_soc.clock_limits[dcn2_1_soc.num_states] = dcn2_1_soc.clock_limits[dcn2_1_soc.num_states - 1]; + dcn2_1_soc.clock_limits[dcn2_1_soc.num_states].state = dcn2_1_soc.num_states; + + dml_init_instance(&dc->dml, &dcn2_1_soc, &dcn2_1_ip, DML_PROJECT_DCN21); } /* Temporary Place holder until we can get them from fuse */ @@ -1476,6 +1554,7 @@ static struct dce_hwseq *dcn21_hwseq_create( hws->shifts = &hwseq_shift; hws->masks = &hwseq_mask; hws->wa.DEGVIDCN21 = true; + hws->wa.disallow_self_refresh_during_multi_plane_transition = true; } return hws; } @@ -1499,6 +1578,7 @@ static const struct encoder_feature_support link_enc_feature = { .max_hdmi_pixel_clock = 600000, .hdmi_ycbcr420_supported = true, .dp_ycbcr420_supported = true, + .fec_supported = true, .flags.bits.IS_HBR2_CAPABLE = true, .flags.bits.IS_HBR3_CAPABLE = true, .flags.bits.IS_TPS3_CAPABLE = true, @@ -1639,6 +1719,19 @@ static int dcn21_populate_dml_pipes_from_context( return pipe_cnt; } +enum dc_status dcn21_patch_unknown_plane_state(struct dc_plane_state *plane_state) +{ + enum dc_status result = DC_OK; + + if (plane_state->ctx->dc->debug.disable_dcc == DCC_ENABLE) { + plane_state->dcc.enable = 1; + /* align to our worst case block width */ + plane_state->dcc.meta_pitch = ((plane_state->src_rect.width + 1023) / 1024) * 1024; + } + result = dcn20_patch_unknown_plane_state(plane_state); + return result; +} + static struct resource_funcs dcn21_res_pool_funcs = { .destroy = dcn21_destroy_resource_pool, .link_enc_create = dcn21_link_encoder_create, @@ -1648,7 +1741,7 @@ static struct resource_funcs dcn21_res_pool_funcs = { .remove_stream_from_ctx = dcn20_remove_stream_from_ctx, .acquire_idle_pipe_for_layer = dcn20_acquire_idle_pipe_for_layer, .populate_dml_writeback_from_context = dcn20_populate_dml_writeback_from_context, - .get_default_swizzle_mode = dcn20_get_default_swizzle_mode, + .patch_unknown_plane_state = dcn21_patch_unknown_plane_state, .set_mcif_arb_params = dcn20_set_mcif_arb_params, .find_first_free_match_stream_enc_for_link = dcn10_find_first_free_match_stream_enc_for_link, .update_bw_bounding_box = update_bw_bounding_box @@ -1695,6 +1788,7 @@ static bool dcn21_resource_construct( dc->caps.force_dp_tps4_for_cp2520 = true; dc->caps.extended_aux_timeout_support = true; dc->caps.dmcub_support = true; + dc->caps.is_apu = true; if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV) dc->debug = debug_defaults_drv; @@ -1758,9 +1852,15 @@ static bool dcn21_resource_construct( goto create_fail; } - // Leave as NULL to not affect current dmcu psr programming sequence - // Will be uncommented when functionality is confirmed to be working - pool->base.psr = NULL; + if (dc->debug.disable_dmcu) { + pool->base.psr = dmub_psr_create(ctx); + + if (pool->base.psr == NULL) { + dm_error("DC: failed to create psr obj!\n"); + BREAK_TO_DEBUGGER(); + goto create_fail; + } + } pool->base.abm = dce_abm_create(ctx, &abm_regs, diff --git a/drivers/gpu/drm/amd/display/dc/dm_cp_psp.h b/drivers/gpu/drm/amd/display/dc/dm_cp_psp.h index 626d22d437f4..968c46dfb506 100644 --- a/drivers/gpu/drm/amd/display/dc/dm_cp_psp.h +++ b/drivers/gpu/drm/amd/display/dc/dm_cp_psp.h @@ -32,6 +32,7 @@ struct cp_psp_stream_config { uint8_t otg_inst; uint8_t link_enc_inst; uint8_t stream_enc_inst; + uint8_t mst_supported; void *dm_stream_ctx; bool dpms_off; }; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dc_features.h b/drivers/gpu/drm/amd/display/dc/dml/dc_features.h index ea4cde952f4f..2a1983324629 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dc_features.h +++ b/drivers/gpu/drm/amd/display/dc/dml/dc_features.h @@ -29,7 +29,7 @@ #define DC__PRESENT 1 #define DC__PRESENT__1 1 #define DC__NUM_DPP 4 -#define DC__VOLTAGE_STATES 7 +#define DC__VOLTAGE_STATES 9 #define DC__NUM_DPP__4 1 #define DC__NUM_DPP__0_PRESENT 1 #define DC__NUM_DPP__1_PRESENT 1 diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20v2.c b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20v2.c index 485a9c62ec58..5bbbafacc720 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20v2.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20v2.c @@ -2614,6 +2614,14 @@ static void dml20v2_DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndP if (mode_lib->vba.DRAMClockChangeSupportsVActive && mode_lib->vba.MinActiveDRAMClockChangeMargin > 60) { + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.PrefetchMode[mode_lib->vba.VoltageLevel][mode_lib->vba.maxMpcComb] == 0) { + if (mode_lib->vba.DRAMClockChangeWatermark > + dml_max(mode_lib->vba.StutterEnterPlusExitWatermark, mode_lib->vba.UrgentWatermark)) + mode_lib->vba.MinTTUVBlank[k] += 25; + } + } mode_lib->vba.DRAMClockChangeWatermark += 25; mode_lib->vba.DRAMClockChangeSupport[0][0] = dm_dram_clock_change_vactive; } else if (mode_lib->vba.DummyPStateCheck && diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h index 658f81e757e9..687010c17324 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h +++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h @@ -22,11 +22,12 @@ * Authors: AMD * */ + +#include "dc_features.h" + #ifndef __DISPLAY_MODE_STRUCTS_H__ #define __DISPLAY_MODE_STRUCTS_H__ -#define MAX_CLOCK_LIMIT_STATES 8 - typedef struct _vcs_dpi_voltage_scaling_st voltage_scaling_st; typedef struct _vcs_dpi_soc_bounding_box_st soc_bounding_box_st; typedef struct _vcs_dpi_ip_params_st ip_params_st; @@ -61,12 +62,15 @@ struct _vcs_dpi_voltage_scaling_st { double dram_speed_mts; double fabricclk_mhz; double dispclk_mhz; + double dram_bw_per_chan_gbps; double phyclk_mhz; double dppclk_mhz; double dtbclk_mhz; }; struct _vcs_dpi_soc_bounding_box_st { + struct _vcs_dpi_voltage_scaling_st clock_limits[DC__VOLTAGE_STATES]; + unsigned int num_states; double sr_exit_time_us; double sr_enter_plus_exit_time_us; double urgent_latency_us; @@ -109,8 +113,7 @@ struct _vcs_dpi_soc_bounding_box_st { double xfc_bus_transport_time_us; double xfc_xbuf_latency_tolerance_us; int use_urgent_burst_bw; - unsigned int num_states; - struct _vcs_dpi_voltage_scaling_st clock_limits[MAX_CLOCK_LIMIT_STATES]; + double min_dcfclk; bool do_urgent_latency_adjustment; double urgent_latency_adjustment_fabric_clock_component_us; double urgent_latency_adjustment_fabric_clock_reference_mhz; @@ -189,7 +192,7 @@ struct _vcs_dpi_ip_params_st { unsigned int min_vblank_lines; unsigned int dppclk_delay_subtotal; unsigned int dispclk_delay_subtotal; - unsigned int dcfclk_cstate_latency; + double dcfclk_cstate_latency; unsigned int dppclk_delay_scl; unsigned int dppclk_delay_scl_lb_only; unsigned int dppclk_delay_cnvc_formatter; @@ -202,6 +205,7 @@ struct _vcs_dpi_ip_params_st { unsigned int LineBufferFixedBpp; unsigned int can_vstartup_lines_exceed_vsync_plus_back_porch_lines_minus_one; unsigned int bug_forcing_LC_req_same_size_fixed; + unsigned int number_of_cursors; }; struct _vcs_dpi_display_xfc_params_st { @@ -325,7 +329,6 @@ struct _vcs_dpi_display_pipe_dest_params_st { unsigned int vupdate_width; unsigned int vready_offset; unsigned char interlaced; - unsigned char embedded; double pixel_rate_mhz; unsigned char synchronized_vblank_all_planes; unsigned char otg_inst; diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c index b3c96d9b472f..6b525c52124c 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c +++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c @@ -266,8 +266,6 @@ static void fetch_socbb_params(struct display_mode_lib *mode_lib) mode_lib->vba.MaxDispclk[i] = soc->clock_limits[i].dispclk_mhz; mode_lib->vba.DTBCLKPerState[i] = soc->clock_limits[i].dtbclk_mhz; } - mode_lib->vba.MinVoltageLevel = 0; - mode_lib->vba.MaxVoltageLevel = mode_lib->vba.soc.num_states; mode_lib->vba.DoUrgentLatencyAdjustment = soc->do_urgent_latency_adjustment; @@ -379,7 +377,6 @@ static void fetch_pipe_params(struct display_mode_lib *mode_lib) mode_lib->vba.pipe_plane[j] = mode_lib->vba.NumberOfActivePlanes; - mode_lib->vba.EmbeddedPanel[mode_lib->vba.NumberOfActivePlanes] = dst->embedded; mode_lib->vba.DPPPerPlane[mode_lib->vba.NumberOfActivePlanes] = 1; mode_lib->vba.SourceScan[mode_lib->vba.NumberOfActivePlanes] = (enum scan_direction_class) (src->source_scan); @@ -396,11 +393,11 @@ static void fetch_pipe_params(struct display_mode_lib *mode_lib) mode_lib->vba.ViewportYStartC[mode_lib->vba.NumberOfActivePlanes] = src->viewport_y_c; mode_lib->vba.PitchY[mode_lib->vba.NumberOfActivePlanes] = src->data_pitch; - mode_lib->vba.SurfaceHeightY[mode_lib->vba.NumberOfActivePlanes] = src->viewport_height; - mode_lib->vba.SurfaceWidthY[mode_lib->vba.NumberOfActivePlanes] = src->viewport_width; + mode_lib->vba.SurfaceWidthY[mode_lib->vba.NumberOfActivePlanes] = src->surface_width_y; + mode_lib->vba.SurfaceHeightY[mode_lib->vba.NumberOfActivePlanes] = src->surface_height_y; mode_lib->vba.PitchC[mode_lib->vba.NumberOfActivePlanes] = src->data_pitch_c; - mode_lib->vba.SurfaceHeightC[mode_lib->vba.NumberOfActivePlanes] = src->viewport_height_c; - mode_lib->vba.SurfaceWidthC[mode_lib->vba.NumberOfActivePlanes] = src->viewport_width_c; + mode_lib->vba.SurfaceHeightC[mode_lib->vba.NumberOfActivePlanes] = src->surface_height_c; + mode_lib->vba.SurfaceWidthC[mode_lib->vba.NumberOfActivePlanes] = src->surface_width_c; mode_lib->vba.DCCMetaPitchY[mode_lib->vba.NumberOfActivePlanes] = src->meta_pitch; mode_lib->vba.DCCMetaPitchC[mode_lib->vba.NumberOfActivePlanes] = src->meta_pitch_c; mode_lib->vba.HRatio[mode_lib->vba.NumberOfActivePlanes] = scl->hscl_ratio; diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.h index 2875efd85467..5d82fc5a7ed7 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.h +++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.h @@ -389,7 +389,6 @@ struct vba_vars_st { /* vba mode support */ /*inputs*/ - bool EmbeddedPanel[DC__NUM_DPP__MAX]; bool SupportGFX7CompatibleTilingIn32bppAnd64bpp; double MaxHSCLRatio; double MaxVSCLRatio; @@ -842,8 +841,6 @@ struct vba_vars_st { double DCCRateChroma[DC__NUM_DPP__MAX]; double PHYCLKD18PerState[DC__VOLTAGE_STATES + 1]; - int MinVoltageLevel; - int MaxVoltageLevel; bool WritebackSupportInterleaveAndUsingWholeBufferForASingleStream; bool NumberOfHDMIFRLSupport; @@ -880,7 +877,6 @@ struct vba_vars_st { double TotalMetaRowBandwidth[DC__VOLTAGE_STATES + 1][2]; double TotalVActiveCursorBandwidth[DC__VOLTAGE_STATES + 1][2]; double TotalVActivePixelBandwidth[DC__VOLTAGE_STATES + 1][2]; - bool UseMinimumRequiredDCFCLK; double WritebackDelayTime[DC__NUM_DPP__MAX]; unsigned int DCCYIndependentBlock[DC__NUM_DPP__MAX]; unsigned int DCCCIndependentBlock[DC__NUM_DPP__MAX]; diff --git a/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.c b/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.c index d2d36d48caaa..f252af1947c3 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.c @@ -47,9 +47,9 @@ #include "dce120/hw_factory_dce120.h" #if defined(CONFIG_DRM_AMD_DC_DCN) #include "dcn10/hw_factory_dcn10.h" -#endif #include "dcn20/hw_factory_dcn20.h" #include "dcn21/hw_factory_dcn21.h" +#endif #include "diagnostics/hw_factory_diag.h" diff --git a/drivers/gpu/drm/amd/display/dc/gpio/hw_translate.c b/drivers/gpu/drm/amd/display/dc/gpio/hw_translate.c index 5d396657a1ee..04e2c0f74cb0 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/hw_translate.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/hw_translate.c @@ -45,9 +45,9 @@ #include "dce120/hw_translate_dce120.h" #if defined(CONFIG_DRM_AMD_DC_DCN) #include "dcn10/hw_translate_dcn10.h" -#endif #include "dcn20/hw_translate_dcn20.h" #include "dcn21/hw_translate_dcn21.h" +#endif #include "diagnostics/hw_translate_diag.h" diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_types.h b/drivers/gpu/drm/amd/display/dc/inc/core_types.h index f285b76888fb..d523fc9547e7 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/core_types.h +++ b/drivers/gpu/drm/amd/display/dc/inc/core_types.h @@ -124,7 +124,7 @@ struct resource_funcs { struct dc *dc, struct dc_state *new_ctx, struct dc_stream_state *stream); - enum dc_status (*get_default_swizzle_mode)( + enum dc_status (*patch_unknown_plane_state)( struct dc_plane_state *plane_state); struct stream_encoder *(*find_first_free_match_stream_enc_for_link)( diff --git a/drivers/gpu/drm/amd/display/dc/inc/dc_link_dp.h b/drivers/gpu/drm/amd/display/dc/inc/dc_link_dp.h index 8b1f0ce6c2a7..e94e5fbf2aa2 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/dc_link_dp.h +++ b/drivers/gpu/drm/amd/display/dc/inc/dc_link_dp.h @@ -78,6 +78,8 @@ void dp_set_panel_mode(struct dc_link *link, enum dp_panel_mode panel_mode); bool dp_overwrite_extended_receiver_cap(struct dc_link *link); +void dpcd_set_source_specific_data(struct dc_link *link); + void dp_set_fec_ready(struct dc_link *link, bool ready); void dp_set_fec_enable(struct dc_link *link, bool enable); bool dp_set_dsc_enable(struct pipe_ctx *pipe_ctx, bool enable); diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h index ac530c057ddd..ce65678c03b2 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h @@ -27,6 +27,7 @@ #define __DAL_CLK_MGR_H__ #include "dc.h" +#include "dm_pp_smu.h" #define DCN_MINIMUM_DISPCLK_Khz 100000 #define DCN_MINIMUM_DPPCLK_Khz 100000 @@ -193,6 +194,7 @@ struct clk_mgr { int dentist_vco_freq_khz; struct clk_state_registers_and_bypass boot_snapshot; struct clk_bw_params *bw_params; + struct pp_smu_wm_range_sets ranges; }; /* forward declarations */ diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h index 862952c0286a..9311d0de377f 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h @@ -296,6 +296,10 @@ int clk_mgr_helper_get_active_display_cnt( struct dc *dc, struct dc_state *context); +int clk_mgr_helper_get_active_plane_cnt( + struct dc *dc, + struct dc_state *context); + #endif //__DAL_CLK_MGR_INTERNAL_H__ diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h index 05ee5295d2c1..336c80a18175 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h @@ -27,11 +27,12 @@ #define __DAL_DCCG_H__ #include "dc_types.h" +#include "hw_shared.h" struct dccg { struct dc_context *ctx; const struct dccg_funcs *funcs; - + int pipe_dppclk_khz[MAX_PIPES]; int ref_dppclk; }; diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h index c0dc1d0f5cae..f5dd0cc73c63 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h @@ -134,7 +134,7 @@ struct hubbub_funcs { unsigned int dccg_ref_freq_inKhz, unsigned int *dchub_ref_freq_inKhz); - void (*program_watermarks)( + bool (*program_watermarks)( struct hubbub *hubbub, struct dcn_watermark_set *watermarks, unsigned int refclk_mhz, diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dsc.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dsc.h index c59740084ebc..7c2a3328b208 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/dsc.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dsc.h @@ -39,6 +39,7 @@ struct dsc_config { uint32_t pic_height; enum dc_pixel_encoding pixel_encoding; enum dc_color_depth color_depth; /* Bits per component */ + bool is_odm; struct dc_dsc_config dc_dsc_cfg; }; diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dwb.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dwb.h index 459f95f52486..f30ab4916242 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/dwb.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dwb.h @@ -25,16 +25,15 @@ #ifndef __DC_DWBC_H__ #define __DC_DWBC_H__ +#include "dal_types.h" #include "dc_hw_types.h" - #define DWB_SW_V2 1 #define DWB_MCIF_BUF_COUNT 4 /* forward declaration of mcif_wb struct */ struct mcif_wb; -enum dce_version; enum dwb_sw_version { dwb_ver_1_0 = 1, diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/link_encoder.h b/drivers/gpu/drm/amd/display/dc/inc/hw/link_encoder.h index fb748f082c56..c2b392a533b1 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/link_encoder.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/link_encoder.h @@ -68,6 +68,7 @@ struct encoder_feature_support { unsigned int max_hdmi_pixel_clock; bool hdmi_ycbcr420_supported; bool dp_ycbcr420_supported; + bool fec_supported; }; union dpcd_psr_configuration { diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/opp.h b/drivers/gpu/drm/amd/display/dc/inc/hw/opp.h index 7575564b2265..2717352eb697 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/opp.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/opp.h @@ -310,7 +310,8 @@ struct opp_funcs { enum dc_color_depth color_depth, const struct tg_color *solid_color, int width, - int height); + int height, + int offset); bool (*dpg_is_blanked)( struct output_pixel_processor *opp); diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h b/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h index 351b387ad606..ac6523c0828e 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h @@ -103,6 +103,7 @@ struct stream_encoder { struct dc_context *ctx; struct dc_bios *bp; enum engine_id id; + uint32_t stream_enc_inst; }; struct enc_state { diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h index 209118f9f193..d4c1fb242c63 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h @@ -66,6 +66,8 @@ struct hw_sequencer_funcs { int num_planes, struct dc_state *context); void (*program_front_end_for_ctx)(struct dc *dc, struct dc_state *context); + void (*post_unlock_program_front_end)(struct dc *dc, + struct dc_state *context); void (*update_plane_addr)(const struct dc *dc, struct pipe_ctx *pipe_ctx); void (*update_dchub)(struct dce_hwseq *hws, @@ -78,10 +80,10 @@ struct hw_sequencer_funcs { void (*update_pending_status)(struct pipe_ctx *pipe_ctx); /* Pipe Lock Related */ - void (*pipe_control_lock_global)(struct dc *dc, - struct pipe_ctx *pipe, bool lock); void (*pipe_control_lock)(struct dc *dc, struct pipe_ctx *pipe, bool lock); + void (*interdependent_update_lock)(struct dc *dc, + struct dc_state *context, bool lock); void (*set_flip_control_gsl)(struct pipe_ctx *pipe_ctx, bool flip_immediate); diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h index ecf566378ccd..52a26e6be066 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h @@ -40,10 +40,13 @@ struct dce_hwseq_wa { bool false_optc_underflow; bool DEGVIDCN10_254; bool DEGVIDCN21; + bool disallow_self_refresh_during_multi_plane_transition; }; struct hwseq_wa_state { bool DEGVIDCN10_253_applied; + bool disallow_self_refresh_during_multi_plane_transition_applied; + unsigned int disallow_self_refresh_during_multi_plane_transition_applied_on_frame; }; struct pipe_ctx; @@ -97,6 +100,8 @@ struct hwseq_private_funcs { struct dc *dc); void (*edp_backlight_control)(struct dc_link *link, bool enable); + bool (*is_panel_backlight_on)(struct dc_link *link); + bool (*is_panel_powered_on)(struct dc_link *link); void (*setup_vupdate_interrupt)(struct dc *dc, struct pipe_ctx *pipe_ctx); bool (*did_underflow_occur)(struct dc *dc, struct pipe_ctx *pipe_ctx); @@ -140,6 +145,8 @@ struct hwseq_private_funcs { const struct dc_plane_state *plane_state); bool (*set_shaper_3dlut)(struct pipe_ctx *pipe_ctx, const struct dc_plane_state *plane_state); + void (*PLAT_58856_wa)(struct dc_state *context, + struct pipe_ctx *pipe_ctx); }; struct dce_hwseq { diff --git a/drivers/gpu/drm/amd/display/dc/inc/resource.h b/drivers/gpu/drm/amd/display/dc/inc/resource.h index 5ae8ada154ef..ca4c36c0c9bc 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/resource.h +++ b/drivers/gpu/drm/amd/display/dc/inc/resource.h @@ -179,4 +179,7 @@ unsigned int resource_pixel_format_to_bpp(enum surface_pixel_format format); void get_audio_check(struct audio_info *aud_modes, struct audio_check *aud_chk); + +int get_num_odm_splits(struct pipe_ctx *pipe); + #endif /* DRIVERS_GPU_DRM_AMD_DC_DEV_DC_INC_RESOURCE_H_ */ diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h index cd9532b4f14d..10b5fa9d2588 100644 --- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h +++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h @@ -50,6 +50,7 @@ enum dmub_cmd_type { DMUB_CMD__REG_REG_WAIT = 4, DMUB_CMD__PLAT_54186_WA = 5, DMUB_CMD__PSR = 64, + DMUB_CMD__ABM = 66, DMUB_CMD__VBIOS = 128, }; @@ -216,7 +217,6 @@ struct dmub_rb_cmd_dpphy_init { struct dmub_cmd_psr_copy_settings_data { uint16_t psr_level; - uint8_t hubp_inst; uint8_t dpp_inst; uint8_t mpcc_inst; uint8_t opp_inst; @@ -225,17 +225,8 @@ struct dmub_cmd_psr_copy_settings_data { uint8_t digbe_inst; uint8_t dpphy_inst; uint8_t aux_inst; - uint8_t hyst_frames; - uint8_t hyst_lines; - uint8_t phy_num; - uint8_t phy_type; - uint8_t aux_repeat; uint8_t smu_optimizations_en; - uint8_t skip_wait_for_pll_lock; uint8_t frame_delay; - uint8_t smu_phy_id; - uint8_t num_of_controllers; - uint8_t link_rate; uint8_t frame_cap_ind; }; @@ -257,13 +248,59 @@ struct dmub_rb_cmd_psr_enable { struct dmub_cmd_header header; }; -struct dmub_cmd_psr_setup_data { +struct dmub_cmd_psr_set_version_data { enum psr_version version; // PSR version 1 or 2 }; -struct dmub_rb_cmd_psr_setup { +struct dmub_rb_cmd_psr_set_version { struct dmub_cmd_header header; - struct dmub_cmd_psr_setup_data psr_setup_data; + struct dmub_cmd_psr_set_version_data psr_set_version_data; +}; + +struct dmub_cmd_abm_set_pipe_data { + uint32_t ramping_boundary; + uint32_t otg_inst; +}; + +struct dmub_rb_cmd_abm_set_pipe { + struct dmub_cmd_header header; + struct dmub_cmd_abm_set_pipe_data abm_set_pipe_data; +}; + +struct dmub_cmd_abm_set_backlight_data { + uint32_t frame_ramp; +}; + +struct dmub_rb_cmd_abm_set_backlight { + struct dmub_cmd_header header; + struct dmub_cmd_abm_set_backlight_data abm_set_backlight_data; +}; + +struct dmub_cmd_abm_set_level_data { + uint32_t level; +}; + +struct dmub_rb_cmd_abm_set_level { + struct dmub_cmd_header header; + struct dmub_cmd_abm_set_level_data abm_set_level_data; +}; + +struct dmub_cmd_abm_set_ambient_level_data { + uint32_t ambient_lux; +}; + +struct dmub_rb_cmd_abm_set_ambient_level { + struct dmub_cmd_header header; + struct dmub_cmd_abm_set_ambient_level_data abm_set_ambient_level_data; +}; + +struct dmub_cmd_abm_set_pwm_frac_data { + uint32_t fractional_pwm; +}; + +struct dmub_rb_cmd_abm_set_pwm_frac { + struct dmub_cmd_header header; + struct dmub_cmd_abm_set_pwm_frac_data abm_set_pwm_frac_data; }; union dmub_rb_cmd { @@ -277,11 +314,16 @@ union dmub_rb_cmd { struct dmub_rb_cmd_enable_disp_power_gating enable_disp_power_gating; struct dmub_rb_cmd_dpphy_init dpphy_init; struct dmub_rb_cmd_dig1_transmitter_control dig1_transmitter_control; - struct dmub_rb_cmd_psr_enable psr_enable; + struct dmub_rb_cmd_psr_set_version psr_set_version; struct dmub_rb_cmd_psr_copy_settings psr_copy_settings; + struct dmub_rb_cmd_psr_enable psr_enable; struct dmub_rb_cmd_psr_set_level psr_set_level; struct dmub_rb_cmd_PLAT_54186_wa PLAT_54186_wa; - struct dmub_rb_cmd_psr_setup psr_setup; + struct dmub_rb_cmd_abm_set_pipe abm_set_pipe; + struct dmub_rb_cmd_abm_set_backlight abm_set_backlight; + struct dmub_rb_cmd_abm_set_level abm_set_level; + struct dmub_rb_cmd_abm_set_ambient_level abm_set_ambient_level; + struct dmub_rb_cmd_abm_set_pwm_frac abm_set_pwm_frac; }; #pragma pack(pop) diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd_dal.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd_dal.h index 7b69eb37f762..d37535d21928 100644 --- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd_dal.h +++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd_dal.h @@ -32,7 +32,7 @@ */ enum dmub_cmd_psr_type { - DMUB_CMD__PSR_SETUP = 0, + DMUB_CMD__PSR_SET_VERSION = 0, DMUB_CMD__PSR_COPY_SETTINGS = 1, DMUB_CMD__PSR_ENABLE = 2, DMUB_CMD__PSR_DISABLE = 3, @@ -42,7 +42,16 @@ enum dmub_cmd_psr_type { enum psr_version { PSR_VERSION_1 = 0x10, // PSR Version 1 PSR_VERSION_2 = 0x20, // PSR Version 2, includes selective update - PSR_VERSION_2_Y_COORD = 0x21, // PSR Version 2, includes Y-coordinate support for SU + PSR_VERSION_2_1 = 0x21, // PSR Version 2, includes Y-coordinate support for SU +}; + +enum dmub_cmd_abm_type { + DMUB_CMD__ABM_INIT_CONFIG = 0, + DMUB_CMD__ABM_SET_PIPE = 1, + DMUB_CMD__ABM_SET_BACKLIGHT = 2, + DMUB_CMD__ABM_SET_LEVEL = 3, + DMUB_CMD__ABM_SET_AMBIENT_LEVEL = 4, + DMUB_CMD__ABM_SET_PWM_FRAC = 5, }; #endif /* _DMUB_CMD_DAL_H_ */ diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_gpint_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_gpint_cmd.h new file mode 100644 index 000000000000..652d6fc061b6 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_gpint_cmd.h @@ -0,0 +1,75 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef _DMUB_GPINT_CMD_H_ +#define _DMUB_GPINT_CMD_H_ + +#include "dmub_types.h" + +/** + * The register format for sending a command via the GPINT. + */ +union dmub_gpint_data_register { + struct { + uint32_t param : 16; + uint32_t command_code : 12; + uint32_t status : 4; + } bits; + uint32_t all; +}; + +/** + * The shifts and masks below may alternatively be used to format and read + * the command register bits. + */ + +#define DMUB_GPINT_DATA_PARAM_MASK 0xFFFF +#define DMUB_GPINT_DATA_PARAM_SHIFT 0 + +#define DMUB_GPINT_DATA_COMMAND_CODE_MASK 0xFFF +#define DMUB_GPINT_DATA_COMMAND_CODE_SHIFT 16 + +#define DMUB_GPINT_DATA_STATUS_MASK 0xF +#define DMUB_GPINT_DATA_STATUS_SHIFT 28 + +/* + * Command IDs should be treated as stable ABI. + * Do not reuse or modify IDs. + */ + +enum dmub_gpint_command { + DMUB_GPINT__INVALID_COMMAND = 0, + DMUB_GPINT__GET_FW_VERSION = 1, + DMUB_GPINT__STOP_FW = 2, + DMUB_GPINT__GET_PSR_STATE = 7, +}; + +/** + * Command responses. + */ + +#define DMUB_GPINT__STOP_FW_RESPONSE 0xDEADDEAD + +#endif /* _DMUB_GPINT_CMD_H_ */ diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_srv.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_srv.h index f8917594036a..c2671f2616c8 100644 --- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_srv.h +++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_srv.h @@ -66,6 +66,7 @@ #include "dmub_types.h" #include "dmub_cmd.h" +#include "dmub_gpint_cmd.h" #include "dmub_rb.h" #if defined(__cplusplus) @@ -103,7 +104,7 @@ enum dmub_window_id { DMUB_WINDOW_4_MAILBOX, DMUB_WINDOW_5_TRACEBUFF, DMUB_WINDOW_6_FW_STATE, - DMUB_WINDOW_7_RESERVED, + DMUB_WINDOW_7_SCRATCH_MEM, DMUB_WINDOW_TOTAL, }; @@ -262,6 +263,14 @@ struct dmub_srv_hw_funcs { bool (*is_phy_init)(struct dmub_srv *dmub); bool (*is_auto_load_done)(struct dmub_srv *dmub); + + void (*set_gpint)(struct dmub_srv *dmub, + union dmub_gpint_data_register reg); + + bool (*is_gpint_acked)(struct dmub_srv *dmub, + union dmub_gpint_data_register reg); + + uint32_t (*get_gpint_response)(struct dmub_srv *dmub); }; /** @@ -307,6 +316,7 @@ struct dmub_srv { enum dmub_asic asic; void *user_ctx; bool is_virtual; + struct dmub_fb scratch_mem_fb; volatile const struct dmub_fw_state *fw_state; /* private: internal use only */ @@ -516,6 +526,45 @@ enum dmub_status dmub_srv_wait_for_phy_init(struct dmub_srv *dmub, enum dmub_status dmub_srv_wait_for_idle(struct dmub_srv *dmub, uint32_t timeout_us); +/** + * dmub_srv_send_gpint_command() - Sends a GPINT based command. + * @dmub: the dmub service + * @command_code: the command code to send + * @param: the command parameter to send + * @timeout_us: the maximum number of microseconds to wait + * + * Sends a command via the general purpose interrupt (GPINT). + * Waits for the number of microseconds specified by timeout_us + * for the command ACK before returning. + * + * Can be called after software initialization. + * + * Return: + * DMUB_STATUS_OK - success + * DMUB_STATUS_TIMEOUT - wait for ACK timed out + * DMUB_STATUS_INVALID - unspecified error + */ +enum dmub_status +dmub_srv_send_gpint_command(struct dmub_srv *dmub, + enum dmub_gpint_command command_code, + uint16_t param, uint32_t timeout_us); + +/** + * dmub_srv_get_gpint_response() - Queries the GPINT response. + * @dmub: the dmub service + * @response: the response for the last GPINT + * + * Returns the response code for the last GPINT interrupt. + * + * Can be called after software initialization. + * + * Return: + * DMUB_STATUS_OK - success + * DMUB_STATUS_INVALID - unspecified error + */ +enum dmub_status dmub_srv_get_gpint_response(struct dmub_srv *dmub, + uint32_t *response); + #if defined(__cplusplus) } #endif diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn20.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn20.c index b2ca8e0dbac9..63bb9e2c81de 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn20.c +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn20.c @@ -60,6 +60,12 @@ static void dmub_dcn20_get_fb_base_offset(struct dmub_srv *dmub, { uint32_t tmp; + if (dmub->fb_base || dmub->fb_offset) { + *fb_base = dmub->fb_base; + *fb_offset = dmub->fb_offset; + return; + } + REG_GET(DCN_VM_FB_LOCATION_BASE, FB_BASE, &tmp); *fb_base = (uint64_t)tmp << 24; @@ -77,11 +83,52 @@ static inline void dmub_dcn20_translate_addr(const union dmub_addr *addr_in, void dmub_dcn20_reset(struct dmub_srv *dmub) { + union dmub_gpint_data_register cmd; + const uint32_t timeout = 30; + uint32_t in_reset, scratch, i; + + REG_GET(DMCUB_CNTL, DMCUB_SOFT_RESET, &in_reset); + + if (in_reset == 0) { + cmd.bits.status = 1; + cmd.bits.command_code = DMUB_GPINT__STOP_FW; + cmd.bits.param = 0; + + dmub->hw_funcs.set_gpint(dmub, cmd); + + /** + * Timeout covers both the ACK and the wait + * for remaining work to finish. + * + * This is mostly bound by the PHY disable sequence. + * Each register check will be greater than 1us, so + * don't bother using udelay. + */ + + for (i = 0; i < timeout; ++i) { + if (dmub->hw_funcs.is_gpint_acked(dmub, cmd)) + break; + } + + for (i = 0; i < timeout; ++i) { + scratch = dmub->hw_funcs.get_gpint_response(dmub); + if (scratch == DMUB_GPINT__STOP_FW_RESPONSE) + break; + } + + /* Clear the GPINT command manually so we don't reset again. */ + cmd.all = 0; + dmub->hw_funcs.set_gpint(dmub, cmd); + + /* Force reset in case we timed out, DMCUB is likely hung. */ + } + REG_UPDATE(DMCUB_CNTL, DMCUB_SOFT_RESET, 1); REG_UPDATE(DMCUB_CNTL, DMCUB_ENABLE, 0); REG_UPDATE(MMHUBBUB_SOFT_RESET, DMUIF_SOFT_RESET, 1); REG_WRITE(DMCUB_INBOX1_RPTR, 0); REG_WRITE(DMCUB_INBOX1_WPTR, 0); + REG_WRITE(DMCUB_SCRATCH0, 0); } void dmub_dcn20_reset_release(struct dmub_srv *dmub) @@ -217,3 +264,25 @@ bool dmub_dcn20_is_supported(struct dmub_srv *dmub) return supported; } + +void dmub_dcn20_set_gpint(struct dmub_srv *dmub, + union dmub_gpint_data_register reg) +{ + REG_WRITE(DMCUB_GPINT_DATAIN1, reg.all); +} + +bool dmub_dcn20_is_gpint_acked(struct dmub_srv *dmub, + union dmub_gpint_data_register reg) +{ + union dmub_gpint_data_register test; + + reg.bits.status = 0; + test.all = REG_READ(DMCUB_GPINT_DATAIN1); + + return test.all == reg.all; +} + +uint32_t dmub_dcn20_get_gpint_response(struct dmub_srv *dmub) +{ + return REG_READ(DMCUB_SCRATCH7); +} diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn20.h b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn20.h index 04b0fa13153d..7f046c73927e 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn20.h +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn20.h @@ -91,6 +91,7 @@ struct dmub_srv; DMUB_SR(DMCUB_SCRATCH13) \ DMUB_SR(DMCUB_SCRATCH14) \ DMUB_SR(DMCUB_SCRATCH15) \ + DMUB_SR(DMCUB_GPINT_DATAIN1) \ DMUB_SR(CC_DC_PIPE_DIS) \ DMUB_SR(MMHUBBUB_SOFT_RESET) \ DMUB_SR(DCN_VM_FB_LOCATION_BASE) \ @@ -183,4 +184,12 @@ bool dmub_dcn20_is_hw_init(struct dmub_srv *dmub); bool dmub_dcn20_is_supported(struct dmub_srv *dmub); +void dmub_dcn20_set_gpint(struct dmub_srv *dmub, + union dmub_gpint_data_register reg); + +bool dmub_dcn20_is_gpint_acked(struct dmub_srv *dmub, + union dmub_gpint_data_register reg); + +uint32_t dmub_dcn20_get_gpint_response(struct dmub_srv *dmub); + #endif /* _DMUB_DCN20_H_ */ diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c index 85a518bf8a76..ce32cc7933c4 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c @@ -52,8 +52,11 @@ /* Default tracebuffer size if meta is absent. */ #define DMUB_TRACE_BUFFER_SIZE (1024) +/* Default scratch mem size. */ +#define DMUB_SCRATCH_MEM_SIZE (256) + /* Number of windows in use. */ -#define DMUB_NUM_WINDOWS (DMUB_WINDOW_6_FW_STATE + 1) +#define DMUB_NUM_WINDOWS (DMUB_WINDOW_TOTAL) /* Base addresses. */ #define DMUB_CW0_BASE (0x60000000) @@ -126,6 +129,9 @@ static bool dmub_srv_hw_setup(struct dmub_srv *dmub, enum dmub_asic asic) funcs->set_inbox1_wptr = dmub_dcn20_set_inbox1_wptr; funcs->is_supported = dmub_dcn20_is_supported; funcs->is_hw_init = dmub_dcn20_is_hw_init; + funcs->set_gpint = dmub_dcn20_set_gpint; + funcs->is_gpint_acked = dmub_dcn20_is_gpint_acked; + funcs->get_gpint_response = dmub_dcn20_get_gpint_response; if (asic == DMUB_ASIC_DCN21) { dmub->regs = &dmub_srv_dcn21_regs; @@ -208,9 +214,11 @@ dmub_srv_calc_region_info(struct dmub_srv *dmub, struct dmub_region *mail = &out->regions[DMUB_WINDOW_4_MAILBOX]; struct dmub_region *trace_buff = &out->regions[DMUB_WINDOW_5_TRACEBUFF]; struct dmub_region *fw_state = &out->regions[DMUB_WINDOW_6_FW_STATE]; + struct dmub_region *scratch_mem = &out->regions[DMUB_WINDOW_7_SCRATCH_MEM]; const struct dmub_fw_meta_info *fw_info; uint32_t fw_state_size = DMUB_FW_STATE_SIZE; uint32_t trace_buffer_size = DMUB_TRACE_BUFFER_SIZE; + uint32_t scratch_mem_size = DMUB_SCRATCH_MEM_SIZE; if (!dmub->sw_init) return DMUB_STATUS_INVALID; @@ -253,7 +261,10 @@ dmub_srv_calc_region_info(struct dmub_srv *dmub, fw_state->base = dmub_align(trace_buff->top, 256); fw_state->top = fw_state->base + dmub_align(fw_state_size, 64); - out->fb_size = dmub_align(fw_state->top, 4096); + scratch_mem->base = dmub_align(fw_state->top, 256); + scratch_mem->top = scratch_mem->base + dmub_align(scratch_mem_size, 64); + + out->fb_size = dmub_align(scratch_mem->top, 4096); return DMUB_STATUS_OK; } @@ -331,6 +342,7 @@ enum dmub_status dmub_srv_hw_init(struct dmub_srv *dmub, struct dmub_fb *mail_fb = params->fb[DMUB_WINDOW_4_MAILBOX]; struct dmub_fb *tracebuff_fb = params->fb[DMUB_WINDOW_5_TRACEBUFF]; struct dmub_fb *fw_state_fb = params->fb[DMUB_WINDOW_6_FW_STATE]; + struct dmub_fb *scratch_mem_fb = params->fb[DMUB_WINDOW_7_SCRATCH_MEM]; struct dmub_rb_init_params rb_params; struct dmub_window cw0, cw1, cw2, cw3, cw4, cw5, cw6; @@ -367,7 +379,7 @@ enum dmub_status dmub_srv_hw_init(struct dmub_srv *dmub, dmub->hw_funcs.reset(dmub); if (inst_fb && data_fb && bios_fb && mail_fb && tracebuff_fb && - fw_state_fb) { + fw_state_fb && scratch_mem_fb) { cw2.offset.quad_part = data_fb->gpu_addr; cw2.region.base = DMUB_CW0_BASE + inst_fb->size; cw2.region.top = cw2.region.base + data_fb->size; @@ -393,6 +405,8 @@ enum dmub_status dmub_srv_hw_init(struct dmub_srv *dmub, dmub->fw_state = fw_state_fb->cpu_addr; + dmub->scratch_mem_fb = *scratch_mem_fb; + if (dmub->hw_funcs.setup_windows) dmub->hw_funcs.setup_windows(dmub, &cw2, &cw3, &cw4, &cw5, &cw6); @@ -522,3 +536,50 @@ enum dmub_status dmub_srv_wait_for_idle(struct dmub_srv *dmub, return DMUB_STATUS_TIMEOUT; } + +enum dmub_status +dmub_srv_send_gpint_command(struct dmub_srv *dmub, + enum dmub_gpint_command command_code, + uint16_t param, uint32_t timeout_us) +{ + union dmub_gpint_data_register reg; + uint32_t i; + + if (!dmub->sw_init) + return DMUB_STATUS_INVALID; + + if (!dmub->hw_funcs.set_gpint) + return DMUB_STATUS_INVALID; + + if (!dmub->hw_funcs.is_gpint_acked) + return DMUB_STATUS_INVALID; + + reg.bits.status = 1; + reg.bits.command_code = command_code; + reg.bits.param = param; + + dmub->hw_funcs.set_gpint(dmub, reg); + + for (i = 0; i < timeout_us; ++i) { + if (dmub->hw_funcs.is_gpint_acked(dmub, reg)) + return DMUB_STATUS_OK; + } + + return DMUB_STATUS_TIMEOUT; +} + +enum dmub_status dmub_srv_get_gpint_response(struct dmub_srv *dmub, + uint32_t *response) +{ + *response = 0; + + if (!dmub->sw_init) + return DMUB_STATUS_INVALID; + + if (!dmub->hw_funcs.get_gpint_response) + return DMUB_STATUS_INVALID; + + *response = dmub->hw_funcs.get_gpint_response(dmub); + + return DMUB_STATUS_OK; +} diff --git a/drivers/gpu/drm/amd/display/include/dal_asic_id.h b/drivers/gpu/drm/amd/display/include/dal_asic_id.h index a2903985b9e8..2359e88d6029 100644 --- a/drivers/gpu/drm/amd/display/include/dal_asic_id.h +++ b/drivers/gpu/drm/amd/display/include/dal_asic_id.h @@ -134,31 +134,28 @@ #define PICASSO_A0 0x41 /* DCN1_01 */ #define RAVEN2_A0 0x81 -#define RAVEN2_15D8_REV_94 0x94 -#define RAVEN2_15D8_REV_95 0x95 -#define RAVEN2_15D8_REV_E3 0xE3 -#define RAVEN2_15D8_REV_E4 0xE4 -#define RAVEN2_15D8_REV_E9 0xE9 -#define RAVEN2_15D8_REV_EA 0xEA -#define RAVEN2_15D8_REV_EB 0xEB #define RAVEN1_F0 0xF0 #define RAVEN_UNKNOWN 0xFF +#define RENOIR_A0 0x91 #ifndef ASICREV_IS_RAVEN #define ASICREV_IS_RAVEN(eChipRev) ((eChipRev >= RAVEN_A0) && eChipRev < RAVEN_UNKNOWN) #endif +#define PRID_DALI_DE 0xDE +#define PRID_DALI_DF 0xDF +#define PRID_DALI_E3 0xE3 +#define PRID_DALI_E4 0xE4 + +#define PRID_POLLOCK_94 0x94 +#define PRID_POLLOCK_95 0x95 +#define PRID_POLLOCK_E9 0xE9 +#define PRID_POLLOCK_EA 0xEA +#define PRID_POLLOCK_EB 0xEB #define ASICREV_IS_PICASSO(eChipRev) ((eChipRev >= PICASSO_A0) && (eChipRev < RAVEN2_A0)) #ifndef ASICREV_IS_RAVEN2 -#define ASICREV_IS_RAVEN2(eChipRev) ((eChipRev >= RAVEN2_A0) && (eChipRev < RAVEN1_F0)) +#define ASICREV_IS_RAVEN2(eChipRev) ((eChipRev >= RAVEN2_A0) && (eChipRev < RENOIR_A0)) #endif #define ASICREV_IS_RV1_F0(eChipRev) ((eChipRev >= RAVEN1_F0) && (eChipRev < RAVEN_UNKNOWN)) -#define ASICREV_IS_DALI(eChipRev) ((eChipRev == RAVEN2_15D8_REV_E3) \ - || (eChipRev == RAVEN2_15D8_REV_E4)) -#define ASICREV_IS_POLLOCK(eChipRev) (eChipRev == RAVEN2_15D8_REV_94 \ - || eChipRev == RAVEN2_15D8_REV_95 \ - || eChipRev == RAVEN2_15D8_REV_E9 \ - || eChipRev == RAVEN2_15D8_REV_EA \ - || eChipRev == RAVEN2_15D8_REV_EB) #define FAMILY_RV 142 /* DCN 1*/ @@ -175,9 +172,7 @@ enum { #define ASICREV_IS_NAVI10_P(eChipRev) (eChipRev < NV_NAVI12_P_A0) #define ASICREV_IS_NAVI12_P(eChipRev) ((eChipRev >= NV_NAVI12_P_A0) && (eChipRev < NV_NAVI14_M_A0)) #define ASICREV_IS_NAVI14_M(eChipRev) ((eChipRev >= NV_NAVI14_M_A0) && (eChipRev < NV_UNKNOWN)) -#define RENOIR_A0 0x91 -#define DEVICE_ID_RENOIR_1636 0x1636 // Renoir -#define ASICREV_IS_RENOIR(eChipRev) ((eChipRev >= RENOIR_A0) && (eChipRev < 0xFF)) +#define ASICREV_IS_RENOIR(eChipRev) ((eChipRev >= RENOIR_A0) && (eChipRev < RAVEN1_F0)) /* * ASIC chip ID @@ -187,6 +182,9 @@ enum { #define DEVICE_ID_TEMASH_9839 0x9839 #define DEVICE_ID_TEMASH_983D 0x983D +/* RENOIR */ +#define DEVICE_ID_RENOIR_1636 0x1636 + /* Asic Family IDs for different asic family. */ #define FAMILY_CI 120 /* Sea Islands: Hawaii (P), Bonaire (M) */ #define FAMILY_KV 125 /* Fusion => Kaveri: Spectre, Spooky; Kabini: Kalindi */ diff --git a/drivers/gpu/drm/amd/display/include/dpcd_defs.h b/drivers/gpu/drm/amd/display/include/dpcd_defs.h index 2c90d1b46c8b..3d29646c7cb4 100644 --- a/drivers/gpu/drm/amd/display/include/dpcd_defs.h +++ b/drivers/gpu/drm/amd/display/include/dpcd_defs.h @@ -149,4 +149,12 @@ enum dpcd_psr_sink_states { PSR_SINK_STATE_SINK_INTERNAL_ERROR = 7, }; +#define DP_SOURCE_TABLE_REVISION 0x310 +#define DP_SOURCE_PAYLOAD_SIZE 0x311 +#define DP_SOURCE_SINK_CAP 0x317 +#define DP_SOURCE_BACKLIGHT_LEVEL 0x320 +#define DP_SOURCE_BACKLIGHT_CURRENT_PEAK 0x326 +#define DP_SOURCE_BACKLIGHT_CONTROL 0x32E +#define DP_SOURCE_BACKLIGHT_ENABLE 0x32F + #endif /* __DAL_DPCD_DEFS_H__ */ diff --git a/drivers/gpu/drm/amd/display/include/logger_types.h b/drivers/gpu/drm/amd/display/include/logger_types.h index 89a709267019..d66f9d8eefb4 100644 --- a/drivers/gpu/drm/amd/display/include/logger_types.h +++ b/drivers/gpu/drm/amd/display/include/logger_types.h @@ -124,36 +124,37 @@ enum dc_log_type { #define DC_MIN_LOG_MASK ((1 << LOG_ERROR) | \ (1 << LOG_DETECTION_EDID_PARSER)) -#define DC_DEFAULT_LOG_MASK ((1 << LOG_ERROR) | \ - (1 << LOG_WARNING) | \ - (1 << LOG_EVENT_MODE_SET) | \ - (1 << LOG_EVENT_DETECTION) | \ - (1 << LOG_EVENT_LINK_TRAINING) | \ - (1 << LOG_EVENT_LINK_LOSS) | \ - (1 << LOG_EVENT_UNDERFLOW) | \ - (1 << LOG_RESOURCE) | \ - (1 << LOG_FEATURE_OVERRIDE) | \ - (1 << LOG_DETECTION_EDID_PARSER) | \ - (1 << LOG_DC) | \ - (1 << LOG_HW_HOTPLUG) | \ - (1 << LOG_HW_SET_MODE) | \ - (1 << LOG_HW_RESUME_S3) | \ - (1 << LOG_HW_HPD_IRQ) | \ - (1 << LOG_SYNC) | \ - (1 << LOG_BANDWIDTH_VALIDATION) | \ - (1 << LOG_MST) | \ - (1 << LOG_DETECTION_DP_CAPS) | \ - (1 << LOG_BACKLIGHT)) | \ - (1 << LOG_I2C_AUX) | \ - (1 << LOG_IF_TRACE) | \ - (1 << LOG_DTN) /* | \ - (1 << LOG_DEBUG) | \ - (1 << LOG_BIOS) | \ - (1 << LOG_SURFACE) | \ - (1 << LOG_SCALER) | \ - (1 << LOG_DML) | \ - (1 << LOG_HW_LINK_TRAINING) | \ - (1 << LOG_HW_AUDIO)| \ - (1 << LOG_BANDWIDTH_CALCS)*/ +#define DC_DEFAULT_LOG_MASK ((1ULL << LOG_ERROR) | \ + (1ULL << LOG_WARNING) | \ + (1ULL << LOG_EVENT_MODE_SET) | \ + (1ULL << LOG_EVENT_DETECTION) | \ + (1ULL << LOG_EVENT_LINK_TRAINING) | \ + (1ULL << LOG_EVENT_LINK_LOSS) | \ + (1ULL << LOG_EVENT_UNDERFLOW) | \ + (1ULL << LOG_RESOURCE) | \ + (1ULL << LOG_FEATURE_OVERRIDE) | \ + (1ULL << LOG_DETECTION_EDID_PARSER) | \ + (1ULL << LOG_DC) | \ + (1ULL << LOG_HW_HOTPLUG) | \ + (1ULL << LOG_HW_SET_MODE) | \ + (1ULL << LOG_HW_RESUME_S3) | \ + (1ULL << LOG_HW_HPD_IRQ) | \ + (1ULL << LOG_SYNC) | \ + (1ULL << LOG_BANDWIDTH_VALIDATION) | \ + (1ULL << LOG_MST) | \ + (1ULL << LOG_DETECTION_DP_CAPS) | \ + (1ULL << LOG_BACKLIGHT)) | \ + (1ULL << LOG_I2C_AUX) | \ + (1ULL << LOG_IF_TRACE) | \ + (1ULL << LOG_HDMI_FRL) | \ + (1ULL << LOG_DTN) /* | \ + (1ULL << LOG_DEBUG) | \ + (1ULL << LOG_BIOS) | \ + (1ULL << LOG_SURFACE) | \ + (1ULL << LOG_SCALER) | \ + (1ULL << LOG_DML) | \ + (1ULL << LOG_HW_LINK_TRAINING) | \ + (1ULL << LOG_HW_AUDIO)| \ + (1ULL << LOG_BANDWIDTH_CALCS)*/ #endif /* __DAL_LOGGER_TYPES_H__ */ diff --git a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c index b9992ebf77a6..c33454a9e0b4 100644 --- a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c +++ b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c @@ -524,12 +524,12 @@ static void build_vrr_infopacket_data(const struct mod_vrr_params *vrr, infopacket->sb[6] |= 0x04; /* PB7 = FreeSync Minimum refresh rate (Hz) */ - infopacket->sb[7] = (unsigned char)(vrr->min_refresh_in_uhz / 1000000); + infopacket->sb[7] = (unsigned char)((vrr->min_refresh_in_uhz + 500000) / 1000000); /* PB8 = FreeSync Maximum refresh rate (Hz) * Note: We should never go above the field rate of the mode timing set. */ - infopacket->sb[8] = (unsigned char)(vrr->max_refresh_in_uhz / 1000000); + infopacket->sb[8] = (unsigned char)((vrr->max_refresh_in_uhz + 500000) / 1000000); //FreeSync HDR @@ -734,6 +734,7 @@ void mod_freesync_build_vrr_params(struct mod_freesync *mod_freesync, { struct core_freesync *core_freesync = NULL; unsigned long long nominal_field_rate_in_uhz = 0; + unsigned long long rounded_nominal_in_uhz = 0; unsigned int refresh_range = 0; unsigned long long min_refresh_in_uhz = 0; unsigned long long max_refresh_in_uhz = 0; @@ -747,24 +748,23 @@ void mod_freesync_build_vrr_params(struct mod_freesync *mod_freesync, nominal_field_rate_in_uhz = mod_freesync_calc_nominal_field_rate(stream); - /* Rounded to the nearest Hz */ - nominal_field_rate_in_uhz = 1000000ULL * - div_u64(nominal_field_rate_in_uhz + 500000, 1000000); - min_refresh_in_uhz = in_config->min_refresh_in_uhz; max_refresh_in_uhz = in_config->max_refresh_in_uhz; - // Don't allow min > max - if (min_refresh_in_uhz > max_refresh_in_uhz) - min_refresh_in_uhz = max_refresh_in_uhz; - // Full range may be larger than current video timing, so cap at nominal if (max_refresh_in_uhz > nominal_field_rate_in_uhz) max_refresh_in_uhz = nominal_field_rate_in_uhz; // Full range may be larger than current video timing, so cap at nominal - if (min_refresh_in_uhz > nominal_field_rate_in_uhz) - min_refresh_in_uhz = nominal_field_rate_in_uhz; + if (min_refresh_in_uhz > max_refresh_in_uhz) + min_refresh_in_uhz = max_refresh_in_uhz; + + // If a monitor reports exactly max refresh of 2x of min, enforce it on nominal + rounded_nominal_in_uhz = + div_u64(nominal_field_rate_in_uhz + 50000, 100000) * 100000; + if (in_config->max_refresh_in_uhz == (2 * in_config->min_refresh_in_uhz) && + in_config->max_refresh_in_uhz == rounded_nominal_in_uhz) + min_refresh_in_uhz = div_u64(nominal_field_rate_in_uhz, 2); if (!vrr_settings_require_update(core_freesync, in_config, (unsigned int)min_refresh_in_uhz, (unsigned int)max_refresh_in_uhz, @@ -796,11 +796,6 @@ void mod_freesync_build_vrr_params(struct mod_freesync *mod_freesync, refresh_range = in_out_vrr->max_refresh_in_uhz - in_out_vrr->min_refresh_in_uhz; - in_out_vrr->btr.margin_in_us = in_out_vrr->max_duration_in_us - - 2 * in_out_vrr->min_duration_in_us; - if (in_out_vrr->btr.margin_in_us > BTR_MAX_MARGIN) - in_out_vrr->btr.margin_in_us = BTR_MAX_MARGIN; - in_out_vrr->supported = true; } @@ -808,9 +803,14 @@ void mod_freesync_build_vrr_params(struct mod_freesync *mod_freesync, in_out_vrr->btr.btr_enabled = in_config->btr; - if (in_out_vrr->max_refresh_in_uhz < - 2 * in_out_vrr->min_refresh_in_uhz) + if (in_out_vrr->max_refresh_in_uhz < (2 * in_out_vrr->min_refresh_in_uhz)) in_out_vrr->btr.btr_enabled = false; + else { + in_out_vrr->btr.margin_in_us = in_out_vrr->max_duration_in_us - + 2 * in_out_vrr->min_duration_in_us; + if (in_out_vrr->btr.margin_in_us > BTR_MAX_MARGIN) + in_out_vrr->btr.margin_in_us = BTR_MAX_MARGIN; + } in_out_vrr->btr.btr_active = false; in_out_vrr->btr.inserted_duration_in_us = 0; @@ -1012,8 +1012,8 @@ unsigned long long mod_freesync_calc_nominal_field_rate( unsigned int total = stream->timing.h_total * stream->timing.v_total; /* Calculate nominal field rate for stream, rounded up to nearest integer */ - nominal_field_rate_in_uhz = stream->timing.pix_clk_100hz / 10; - nominal_field_rate_in_uhz *= 1000ULL * 1000ULL * 1000ULL; + nominal_field_rate_in_uhz = stream->timing.pix_clk_100hz; + nominal_field_rate_in_uhz *= 100000000ULL; nominal_field_rate_in_uhz = div_u64(nominal_field_rate_in_uhz, total); diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.c index 8aa528e874c4..cc1d3f470b99 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.c +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.c @@ -52,8 +52,8 @@ static uint8_t is_cp_desired_hdcp1(struct mod_hdcp *hdcp) * hdcp is not desired */ for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) { - if (hdcp->connection.displays[i].state != MOD_HDCP_DISPLAY_INACTIVE && - !hdcp->connection.displays[i].adjust.disable) { + if (hdcp->displays[i].state != MOD_HDCP_DISPLAY_INACTIVE && + !hdcp->displays[i].adjust.disable) { is_auth_needed = 1; break; } @@ -61,7 +61,8 @@ static uint8_t is_cp_desired_hdcp1(struct mod_hdcp *hdcp) return (hdcp->connection.hdcp1_retry_count < MAX_NUM_OF_ATTEMPTS) && is_auth_needed && - !hdcp->connection.link.adjust.hdcp1.disable; + !hdcp->connection.link.adjust.hdcp1.disable && + !hdcp->connection.is_hdcp1_revoked; } static uint8_t is_cp_desired_hdcp2(struct mod_hdcp *hdcp) @@ -72,8 +73,8 @@ static uint8_t is_cp_desired_hdcp2(struct mod_hdcp *hdcp) * hdcp is not desired */ for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) { - if (hdcp->connection.displays[i].state != MOD_HDCP_DISPLAY_INACTIVE && - !hdcp->connection.displays[i].adjust.disable) { + if (hdcp->displays[i].state != MOD_HDCP_DISPLAY_INACTIVE && + !hdcp->displays[i].adjust.disable) { is_auth_needed = 1; break; } @@ -103,8 +104,6 @@ static enum mod_hdcp_status execution(struct mod_hdcp *hdcp, event_ctx->unexpected_event = 1; goto out; } - /* update topology event if hdcp is not desired */ - status = mod_hdcp_add_display_topology(hdcp); } else if (is_in_hdcp1_states(hdcp)) { status = mod_hdcp_hdcp1_execution(hdcp, event_ctx, &input->hdcp1); } else if (is_in_hdcp1_dp_states(hdcp)) { @@ -115,6 +114,9 @@ static enum mod_hdcp_status execution(struct mod_hdcp *hdcp, } else if (is_in_hdcp2_dp_states(hdcp)) { status = mod_hdcp_hdcp2_dp_execution(hdcp, event_ctx, &input->hdcp2); + } else { + event_ctx->unexpected_event = 1; + goto out; } out: return status; @@ -191,14 +193,7 @@ static enum mod_hdcp_status reset_authentication(struct mod_hdcp *hdcp, mod_hdcp_hdcp1_destroy_session(hdcp); } - if (hdcp->auth.trans_input.hdcp1.add_topology == PASS) { - status = mod_hdcp_remove_display_topology(hdcp); - if (status != MOD_HDCP_STATUS_SUCCESS) { - output->callback_needed = 0; - output->watchdog_timer_needed = 0; - goto out; - } - } + HDCP_TOP_RESET_AUTH_TRACE(hdcp); memset(&hdcp->auth, 0, sizeof(struct mod_hdcp_authentication)); memset(&hdcp->state, 0, sizeof(struct mod_hdcp_state)); @@ -212,25 +207,12 @@ static enum mod_hdcp_status reset_authentication(struct mod_hdcp *hdcp, goto out; } } - if (hdcp->auth.trans_input.hdcp2.add_topology == PASS) { - status = mod_hdcp_remove_display_topology(hdcp); - if (status != MOD_HDCP_STATUS_SUCCESS) { - output->callback_needed = 0; - output->watchdog_timer_needed = 0; - goto out; - } - } + HDCP_TOP_RESET_AUTH_TRACE(hdcp); memset(&hdcp->auth, 0, sizeof(struct mod_hdcp_authentication)); memset(&hdcp->state, 0, sizeof(struct mod_hdcp_state)); set_state_id(hdcp, output, HDCP_INITIALIZED); } else if (is_in_cp_not_desired_state(hdcp)) { - status = mod_hdcp_remove_display_topology(hdcp); - if (status != MOD_HDCP_STATUS_SUCCESS) { - output->callback_needed = 0; - output->watchdog_timer_needed = 0; - goto out; - } HDCP_TOP_RESET_AUTH_TRACE(hdcp); memset(&hdcp->auth, 0, sizeof(struct mod_hdcp_authentication)); memset(&hdcp->state, 0, sizeof(struct mod_hdcp_state)); @@ -337,16 +319,19 @@ enum mod_hdcp_status mod_hdcp_add_display(struct mod_hdcp *hdcp, if (status != MOD_HDCP_STATUS_SUCCESS) goto out; - /* add display to connection */ - hdcp->connection.link = *link; - *display_container = *display; - /* reset retry counters */ reset_retry_counts(hdcp); /* reset error trace */ memset(&hdcp->connection.trace, 0, sizeof(hdcp->connection.trace)); + /* add display to connection */ + hdcp->connection.link = *link; + *display_container = *display; + status = mod_hdcp_add_display_to_topology(hdcp, display->index); + if (status != MOD_HDCP_STATUS_SUCCESS) + goto out; + /* request authentication */ if (current_state(hdcp) != HDCP_INITIALIZED) set_state_id(hdcp, output, HDCP_INITIALIZED); @@ -379,17 +364,20 @@ enum mod_hdcp_status mod_hdcp_remove_display(struct mod_hdcp *hdcp, if (status != MOD_HDCP_STATUS_SUCCESS) goto out; - /* remove display */ - display->state = MOD_HDCP_DISPLAY_INACTIVE; - /* clear retry counters */ reset_retry_counts(hdcp); /* reset error trace */ memset(&hdcp->connection.trace, 0, sizeof(hdcp->connection.trace)); - /* request authentication for remaining displays*/ - if (get_active_display_count(hdcp) > 0) + /* remove display */ + status = mod_hdcp_remove_display_from_topology(hdcp, index); + if (status != MOD_HDCP_STATUS_SUCCESS) + goto out; + display->state = MOD_HDCP_DISPLAY_INACTIVE; + + /* request authentication when connection is not reset */ + if (current_state(hdcp) != HDCP_UNINITIALIZED) callback_in_ms(hdcp->connection.link.adjust.auth_delay * 1000, output); out: @@ -496,10 +484,8 @@ enum mod_hdcp_operation_mode mod_hdcp_signal_type_to_operation_mode( break; case SIGNAL_TYPE_EDP: case SIGNAL_TYPE_DISPLAY_PORT: - mode = MOD_HDCP_MODE_DP; - break; case SIGNAL_TYPE_DISPLAY_PORT_MST: - mode = MOD_HDCP_MODE_DP_MST; + mode = MOD_HDCP_MODE_DP; break; default: break; diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.h b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.h index af78e4f1be68..5cb4546be0ef 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.h +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.h @@ -41,7 +41,6 @@ enum mod_hdcp_trans_input_result { struct mod_hdcp_transition_input_hdcp1 { uint8_t bksv_read; uint8_t bksv_validation; - uint8_t add_topology; uint8_t create_session; uint8_t an_write; uint8_t aksv_write; @@ -71,7 +70,6 @@ struct mod_hdcp_transition_input_hdcp1 { struct mod_hdcp_transition_input_hdcp2 { uint8_t hdcp2version_read; uint8_t hdcp2_capable_check; - uint8_t add_topology; uint8_t create_session; uint8_t ake_init_prepare; uint8_t ake_init_write; @@ -167,9 +165,9 @@ struct mod_hdcp_auth_counters { /* contains values per connection */ struct mod_hdcp_connection { struct mod_hdcp_link link; - struct mod_hdcp_display displays[MAX_NUM_OF_DISPLAYS]; uint8_t is_repeater; uint8_t is_km_stored; + uint8_t is_hdcp1_revoked; uint8_t is_hdcp2_revoked; struct mod_hdcp_trace trace; uint8_t hdcp1_retry_count; @@ -202,6 +200,8 @@ struct mod_hdcp { struct mod_hdcp_config config; /* per connection */ struct mod_hdcp_connection connection; + /* per displays */ + struct mod_hdcp_display displays[MAX_NUM_OF_DISPLAYS]; /* per authentication attempt */ struct mod_hdcp_authentication auth; /* per state in an authentication */ @@ -327,10 +327,10 @@ void mod_hdcp_dump_binary_message(uint8_t *msg, uint32_t msg_size, /* TODO: add adjustment log */ /* psp functions */ -enum mod_hdcp_status mod_hdcp_add_display_topology( - struct mod_hdcp *hdcp); -enum mod_hdcp_status mod_hdcp_remove_display_topology( - struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_add_display_to_topology( + struct mod_hdcp *hdcp, uint8_t index); +enum mod_hdcp_status mod_hdcp_remove_display_from_topology( + struct mod_hdcp *hdcp, uint8_t index); enum mod_hdcp_status mod_hdcp_hdcp1_create_session(struct mod_hdcp *hdcp); enum mod_hdcp_status mod_hdcp_hdcp1_destroy_session(struct mod_hdcp *hdcp); enum mod_hdcp_status mod_hdcp_hdcp1_validate_rx(struct mod_hdcp *hdcp); @@ -392,13 +392,13 @@ enum mod_hdcp_status mod_hdcp_write_content_type(struct mod_hdcp *hdcp); /* hdcp version helpers */ static inline uint8_t is_dp_hdcp(struct mod_hdcp *hdcp) { - return (hdcp->connection.link.mode == MOD_HDCP_MODE_DP || - hdcp->connection.link.mode == MOD_HDCP_MODE_DP_MST); + return (hdcp->connection.link.mode == MOD_HDCP_MODE_DP); } static inline uint8_t is_dp_mst_hdcp(struct mod_hdcp *hdcp) { - return (hdcp->connection.link.mode == MOD_HDCP_MODE_DP_MST); + return (hdcp->connection.link.mode == MOD_HDCP_MODE_DP && + hdcp->connection.link.dp.mst_supported); } static inline uint8_t is_hdmi_dvi_sl_hdcp(struct mod_hdcp *hdcp) @@ -519,7 +519,7 @@ static inline uint8_t get_active_display_count(struct mod_hdcp *hdcp) uint8_t i; for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) - if (is_display_active(&hdcp->connection.displays[i])) + if (is_display_active(&hdcp->displays[i])) added_count++; return added_count; } @@ -530,7 +530,7 @@ static inline uint8_t get_added_display_count(struct mod_hdcp *hdcp) uint8_t i; for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) - if (is_display_added(&hdcp->connection.displays[i])) + if (is_display_added(&hdcp->displays[i])) added_count++; return added_count; } @@ -542,8 +542,8 @@ static inline struct mod_hdcp_display *get_first_added_display( struct mod_hdcp_display *display = NULL; for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) - if (is_display_added(&hdcp->connection.displays[i])) { - display = &hdcp->connection.displays[i]; + if (is_display_added(&hdcp->displays[i])) { + display = &hdcp->displays[i]; break; } return display; @@ -556,9 +556,9 @@ static inline struct mod_hdcp_display *get_active_display_at_index( struct mod_hdcp_display *display = NULL; for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) - if (hdcp->connection.displays[i].index == index && - is_display_active(&hdcp->connection.displays[i])) { - display = &hdcp->connection.displays[i]; + if (hdcp->displays[i].index == index && + is_display_active(&hdcp->displays[i])) { + display = &hdcp->displays[i]; break; } return display; @@ -571,8 +571,8 @@ static inline struct mod_hdcp_display *get_empty_display_container( struct mod_hdcp_display *display = NULL; for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) - if (!is_display_active(&hdcp->connection.displays[i])) { - display = &hdcp->connection.displays[i]; + if (!is_display_active(&hdcp->displays[i])) { + display = &hdcp->displays[i]; break; } return display; diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_execution.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_execution.c index 37670db64855..37c8c05497d6 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_execution.c +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_execution.c @@ -168,10 +168,6 @@ static enum mod_hdcp_status exchange_ksvs(struct mod_hdcp *hdcp, goto out; } - if (!mod_hdcp_execute_and_set(mod_hdcp_add_display_topology, - &input->add_topology, &status, - hdcp, "add_topology")) - goto out; if (!mod_hdcp_execute_and_set(mod_hdcp_hdcp1_create_session, &input->create_session, &status, hdcp, "create_session")) diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_transition.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_transition.c index 76edcbe51f71..f3711914364e 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_transition.c +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_transition.c @@ -46,8 +46,7 @@ enum mod_hdcp_status mod_hdcp_hdcp1_transition(struct mod_hdcp *hdcp, set_state_id(hdcp, output, H1_A1_EXCHANGE_KSVS); break; case H1_A1_EXCHANGE_KSVS: - if (input->add_topology != PASS || - input->create_session != PASS) { + if (input->create_session != PASS) { /* out of sync with psp state */ adjust->hdcp1.disable = 1; fail_and_restart_in_ms(0, &status, output); @@ -173,8 +172,7 @@ enum mod_hdcp_status mod_hdcp_hdcp1_dp_transition(struct mod_hdcp *hdcp, set_state_id(hdcp, output, D1_A1_EXCHANGE_KSVS); break; case D1_A1_EXCHANGE_KSVS: - if (input->add_topology != PASS || - input->create_session != PASS) { + if (input->create_session != PASS) { /* out of sync with psp state */ adjust->hdcp1.disable = 1; fail_and_restart_in_ms(0, &status, output); @@ -210,7 +208,8 @@ enum mod_hdcp_status mod_hdcp_hdcp1_dp_transition(struct mod_hdcp *hdcp, fail_and_restart_in_ms(0, &status, output); break; } else if (input->rx_validation != PASS) { - if (hdcp->state.stay_count < 2) { + if (hdcp->state.stay_count < 2 && + !hdcp->connection.is_hdcp1_revoked) { /* allow 2 additional retries */ callback_in_ms(0, output); increment_stay_counter(hdcp); @@ -231,6 +230,9 @@ enum mod_hdcp_status mod_hdcp_hdcp1_dp_transition(struct mod_hdcp *hdcp, (!conn->is_repeater && is_dp_mst_hdcp(hdcp) && input->stream_encryption_dp != PASS)) { fail_and_restart_in_ms(0, &status, output); break; + } else if (conn->hdcp1_retry_count < conn->link.adjust.hdcp1.min_auth_retries_wa) { + fail_and_restart_in_ms(0, &status, output); + break; } if (conn->is_repeater) { set_watchdog_in_ms(hdcp, 5000, output); @@ -290,7 +292,8 @@ enum mod_hdcp_status mod_hdcp_hdcp1_dp_transition(struct mod_hdcp *hdcp, fail_and_restart_in_ms(0, &status, output); break; } else if (input->ksvlist_vp_validation != PASS) { - if (hdcp->state.stay_count < 2) { + if (hdcp->state.stay_count < 2 && + !hdcp->connection.is_hdcp1_revoked) { /* allow 2 additional retries */ callback_in_ms(0, output); increment_stay_counter(hdcp); diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_execution.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_execution.c index f730b94ac3c0..491c00f48026 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_execution.c +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_execution.c @@ -34,7 +34,7 @@ static inline enum mod_hdcp_status check_receiver_id_list_ready(struct mod_hdcp if (is_dp_hdcp(hdcp)) is_ready = HDCP_2_2_DP_RXSTATUS_READY(hdcp->auth.msg.hdcp2.rxstatus_dp) ? 1 : 0; else - is_ready = (HDCP_2_2_HDMI_RXSTATUS_READY(hdcp->auth.msg.hdcp2.rxstatus[0]) && + is_ready = (HDCP_2_2_HDMI_RXSTATUS_READY(hdcp->auth.msg.hdcp2.rxstatus[1]) && (HDCP_2_2_HDMI_RXSTATUS_MSG_SZ_HI(hdcp->auth.msg.hdcp2.rxstatus[1]) << 8 | hdcp->auth.msg.hdcp2.rxstatus[0])) ? 1 : 0; return is_ready ? MOD_HDCP_STATUS_SUCCESS : @@ -46,8 +46,8 @@ static inline enum mod_hdcp_status check_hdcp2_capable(struct mod_hdcp *hdcp) enum mod_hdcp_status status; if (is_dp_hdcp(hdcp)) - status = (hdcp->auth.msg.hdcp2.rxcaps_dp[2] & HDCP_2_2_RX_CAPS_VERSION_VAL) && - HDCP_2_2_DP_HDCP_CAPABLE(hdcp->auth.msg.hdcp2.rxcaps_dp[0]) ? + status = (hdcp->auth.msg.hdcp2.rxcaps_dp[0] == HDCP_2_2_RX_CAPS_VERSION_VAL) && + HDCP_2_2_DP_HDCP_CAPABLE(hdcp->auth.msg.hdcp2.rxcaps_dp[2]) ? MOD_HDCP_STATUS_SUCCESS : MOD_HDCP_STATUS_HDCP2_NOT_CAPABLE; else @@ -67,7 +67,7 @@ static inline enum mod_hdcp_status check_reauthentication_request( MOD_HDCP_STATUS_HDCP2_REAUTH_REQUEST : MOD_HDCP_STATUS_SUCCESS; else - ret = HDCP_2_2_HDMI_RXSTATUS_REAUTH_REQ(hdcp->auth.msg.hdcp2.rxstatus[0]) ? + ret = HDCP_2_2_HDMI_RXSTATUS_REAUTH_REQ(hdcp->auth.msg.hdcp2.rxstatus[1]) ? MOD_HDCP_STATUS_HDCP2_REAUTH_REQUEST : MOD_HDCP_STATUS_SUCCESS; return ret; @@ -259,6 +259,7 @@ static enum mod_hdcp_status known_hdcp2_capable_rx(struct mod_hdcp *hdcp, event_ctx->unexpected_event = 1; goto out; } + if (!mod_hdcp_execute_and_set(mod_hdcp_read_hdcp2version, &input->hdcp2version_read, &status, hdcp, "hdcp2version_read")) @@ -281,10 +282,7 @@ static enum mod_hdcp_status send_ake_init(struct mod_hdcp *hdcp, event_ctx->unexpected_event = 1; goto out; } - if (!mod_hdcp_execute_and_set(mod_hdcp_add_display_topology, - &input->add_topology, &status, - hdcp, "add_topology")) - goto out; + if (!mod_hdcp_execute_and_set(mod_hdcp_hdcp2_create_session, &input->create_session, &status, hdcp, "create_session")) diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_transition.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_transition.c index 8cae3e3aacd5..e738c7ae66ec 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_transition.c +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_transition.c @@ -47,8 +47,7 @@ enum mod_hdcp_status mod_hdcp_hdcp2_transition(struct mod_hdcp *hdcp, } break; case H2_A1_SEND_AKE_INIT: - if (input->add_topology != PASS || - input->create_session != PASS || + if (input->create_session != PASS || input->ake_init_prepare != PASS) { /* out of sync with psp state */ adjust->hdcp2.disable = 1; @@ -389,8 +388,7 @@ enum mod_hdcp_status mod_hdcp_hdcp2_dp_transition(struct mod_hdcp *hdcp, } break; case D2_A1_SEND_AKE_INIT: - if (input->add_topology != PASS || - input->create_session != PASS || + if (input->create_session != PASS || input->ake_init_prepare != PASS) { /* out of sync with psp state */ adjust->hdcp2.disable = 1; diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_ddc.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_ddc.c index ff9d54812e62..bb5130f4228d 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_ddc.c +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_ddc.c @@ -65,6 +65,7 @@ enum mod_hdcp_ddc_message_id { MOD_HDCP_MESSAGE_ID_READ_LC_SEND_L_PRIME, MOD_HDCP_MESSAGE_ID_WRITE_SKE_SEND_EKS, MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_SEND_RECEIVERID_LIST, + MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_SEND_RECEIVERID_LIST_PART2, MOD_HDCP_MESSAGE_ID_WRITE_REPEATER_AUTH_SEND_ACK, MOD_HDCP_MESSAGE_ID_WRITE_REPEATER_AUTH_STREAM_MANAGE, MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_STREAM_READY, @@ -101,6 +102,7 @@ static const uint8_t hdcp_i2c_offsets[] = { [MOD_HDCP_MESSAGE_ID_READ_LC_SEND_L_PRIME] = 0x80, [MOD_HDCP_MESSAGE_ID_WRITE_SKE_SEND_EKS] = 0x60, [MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_SEND_RECEIVERID_LIST] = 0x80, + [MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_SEND_RECEIVERID_LIST_PART2] = 0x80, [MOD_HDCP_MESSAGE_ID_WRITE_REPEATER_AUTH_SEND_ACK] = 0x60, [MOD_HDCP_MESSAGE_ID_WRITE_REPEATER_AUTH_STREAM_MANAGE] = 0x60, [MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_STREAM_READY] = 0x80, @@ -135,6 +137,7 @@ static const uint32_t hdcp_dpcd_addrs[] = { [MOD_HDCP_MESSAGE_ID_READ_LC_SEND_L_PRIME] = 0x692f8, [MOD_HDCP_MESSAGE_ID_WRITE_SKE_SEND_EKS] = 0x69318, [MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_SEND_RECEIVERID_LIST] = 0x69330, + [MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_SEND_RECEIVERID_LIST_PART2] = 0x69340, [MOD_HDCP_MESSAGE_ID_WRITE_REPEATER_AUTH_SEND_ACK] = 0x693e0, [MOD_HDCP_MESSAGE_ID_WRITE_REPEATER_AUTH_STREAM_MANAGE] = 0x693f0, [MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_STREAM_READY] = 0x69473, @@ -405,7 +408,7 @@ enum mod_hdcp_status mod_hdcp_read_ake_cert(struct mod_hdcp *hdcp) enum mod_hdcp_status status; if (is_dp_hdcp(hdcp)) { - hdcp->auth.msg.hdcp2.ake_cert[0] = 3; + hdcp->auth.msg.hdcp2.ake_cert[0] = HDCP_2_2_AKE_SEND_CERT; status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_AKE_SEND_CERT, hdcp->auth.msg.hdcp2.ake_cert+1, sizeof(hdcp->auth.msg.hdcp2.ake_cert)-1); @@ -423,7 +426,7 @@ enum mod_hdcp_status mod_hdcp_read_h_prime(struct mod_hdcp *hdcp) enum mod_hdcp_status status; if (is_dp_hdcp(hdcp)) { - hdcp->auth.msg.hdcp2.ake_h_prime[0] = 7; + hdcp->auth.msg.hdcp2.ake_h_prime[0] = HDCP_2_2_AKE_SEND_HPRIME; status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_AKE_SEND_H_PRIME, hdcp->auth.msg.hdcp2.ake_h_prime+1, sizeof(hdcp->auth.msg.hdcp2.ake_h_prime)-1); @@ -441,7 +444,7 @@ enum mod_hdcp_status mod_hdcp_read_pairing_info(struct mod_hdcp *hdcp) enum mod_hdcp_status status; if (is_dp_hdcp(hdcp)) { - hdcp->auth.msg.hdcp2.ake_pairing_info[0] = 8; + hdcp->auth.msg.hdcp2.ake_pairing_info[0] = HDCP_2_2_AKE_SEND_PAIRING_INFO; status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_AKE_SEND_PAIRING_INFO, hdcp->auth.msg.hdcp2.ake_pairing_info+1, sizeof(hdcp->auth.msg.hdcp2.ake_pairing_info)-1); @@ -459,7 +462,7 @@ enum mod_hdcp_status mod_hdcp_read_l_prime(struct mod_hdcp *hdcp) enum mod_hdcp_status status; if (is_dp_hdcp(hdcp)) { - hdcp->auth.msg.hdcp2.lc_l_prime[0] = 10; + hdcp->auth.msg.hdcp2.lc_l_prime[0] = HDCP_2_2_LC_SEND_LPRIME; status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_LC_SEND_L_PRIME, hdcp->auth.msg.hdcp2.lc_l_prime+1, sizeof(hdcp->auth.msg.hdcp2.lc_l_prime)-1); @@ -474,14 +477,27 @@ enum mod_hdcp_status mod_hdcp_read_l_prime(struct mod_hdcp *hdcp) enum mod_hdcp_status mod_hdcp_read_rx_id_list(struct mod_hdcp *hdcp) { - enum mod_hdcp_status status; + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; if (is_dp_hdcp(hdcp)) { - hdcp->auth.msg.hdcp2.rx_id_list[0] = 12; - status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_SEND_RECEIVERID_LIST, - hdcp->auth.msg.hdcp2.rx_id_list+1, - sizeof(hdcp->auth.msg.hdcp2.rx_id_list)-1); + uint32_t device_count = 0; + uint32_t rx_id_list_size = 0; + uint32_t bytes_read = 0; + hdcp->auth.msg.hdcp2.rx_id_list[0] = HDCP_2_2_REP_SEND_RECVID_LIST; + status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_SEND_RECEIVERID_LIST, + hdcp->auth.msg.hdcp2.rx_id_list+1, + HDCP_MAX_AUX_TRANSACTION_SIZE); + if (status == MOD_HDCP_STATUS_SUCCESS) { + bytes_read = HDCP_MAX_AUX_TRANSACTION_SIZE; + device_count = HDCP_2_2_DEV_COUNT_LO(hdcp->auth.msg.hdcp2.rx_id_list[2]) + + (HDCP_2_2_DEV_COUNT_HI(hdcp->auth.msg.hdcp2.rx_id_list[1]) << 4); + rx_id_list_size = MIN((21 + 5 * device_count), + (sizeof(hdcp->auth.msg.hdcp2.rx_id_list) - 1)); + status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_SEND_RECEIVERID_LIST_PART2, + hdcp->auth.msg.hdcp2.rx_id_list + 1 + bytes_read, + (rx_id_list_size - 1) / HDCP_MAX_AUX_TRANSACTION_SIZE * HDCP_MAX_AUX_TRANSACTION_SIZE); + } } else { status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_SEND_RECEIVERID_LIST, hdcp->auth.msg.hdcp2.rx_id_list, @@ -495,7 +511,7 @@ enum mod_hdcp_status mod_hdcp_read_stream_ready(struct mod_hdcp *hdcp) enum mod_hdcp_status status; if (is_dp_hdcp(hdcp)) { - hdcp->auth.msg.hdcp2.repeater_auth_stream_ready[0] = 17; + hdcp->auth.msg.hdcp2.repeater_auth_stream_ready[0] = HDCP_2_2_REP_STREAM_READY; status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_STREAM_READY, hdcp->auth.msg.hdcp2.repeater_auth_stream_ready+1, sizeof(hdcp->auth.msg.hdcp2.repeater_auth_stream_ready)-1); diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_log.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_log.c index 724ebcee9a19..44956f9ba178 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_log.c +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_log.c @@ -90,10 +90,14 @@ char *mod_hdcp_status_to_str(int32_t status) return "MOD_HDCP_STATUS_HDCP1_R0_PRIME_PENDING"; case MOD_HDCP_STATUS_HDCP1_VALIDATE_RX_FAILURE: return "MOD_HDCP_STATUS_HDCP1_VALIDATE_RX_FAILURE"; + case MOD_HDCP_STATUS_HDCP1_BKSV_REVOKED: + return "MOD_HDCP_STATUS_HDCP1_BKSV_REVOKED"; case MOD_HDCP_STATUS_HDCP1_KSV_LIST_NOT_READY: return "MOD_HDCP_STATUS_HDCP1_KSV_LIST_NOT_READY"; case MOD_HDCP_STATUS_HDCP1_VALIDATE_KSV_LIST_FAILURE: return "MOD_HDCP_STATUS_HDCP1_VALIDATE_KSV_LIST_FAILURE"; + case MOD_HDCP_STATUS_HDCP1_KSV_LIST_REVOKED: + return "MOD_HDCP_STATUS_HDCP1_KSV_LIST_REVOKED"; case MOD_HDCP_STATUS_HDCP1_ENABLE_ENCRYPTION: return "MOD_HDCP_STATUS_HDCP1_ENABLE_ENCRYPTION"; case MOD_HDCP_STATUS_HDCP1_ENABLE_STREAM_ENCRYPTION_FAILURE: diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_log.h b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_log.h index ff91373ebada..d3192b9d0c3d 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_log.h +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_log.h @@ -37,10 +37,11 @@ /* default logs */ #define HDCP_ERROR_TRACE(hdcp, status) \ HDCP_LOG_ERR(hdcp, \ - "[Link %d] WARNING %s IN STATE %s", \ + "[Link %d] WARNING %s IN STATE %s STAY COUNT %d", \ hdcp->config.index, \ mod_hdcp_status_to_str(status), \ - mod_hdcp_state_id_to_str(hdcp->state.id)) + mod_hdcp_state_id_to_str(hdcp->state.id), \ + hdcp->state.stay_count) #define HDCP_HDCP1_ENABLED_TRACE(hdcp, displayIndex) \ HDCP_LOG_VER(hdcp, \ "[Link %d] HDCP 1.4 enabled on display %d", \ @@ -49,6 +50,15 @@ HDCP_LOG_VER(hdcp, \ "[Link %d] HDCP 2.2 enabled on display %d", \ hdcp->config.index, displayIndex) +#define HDCP_HDCP1_DISABLED_TRACE(hdcp, displayIndex) \ + HDCP_LOG_VER(hdcp, \ + "[Link %d] HDCP 1.4 disabled on display %d", \ + hdcp->config.index, displayIndex) +#define HDCP_HDCP2_DISABLED_TRACE(hdcp, displayIndex) \ + HDCP_LOG_VER(hdcp, \ + "[Link %d] HDCP 2.2 disabled on display %d", \ + hdcp->config.index, displayIndex) + /* state machine logs */ #define HDCP_REMOVE_DISPLAY_TRACE(hdcp, displayIndex) \ HDCP_LOG_FSM(hdcp, \ @@ -102,6 +112,9 @@ sizeof(hdcp->auth.msg.hdcp1.bksv)); \ HDCP_DDC_READ_TRACE(hdcp, "BCAPS", &hdcp->auth.msg.hdcp1.bcaps, \ sizeof(hdcp->auth.msg.hdcp1.bcaps)); \ + HDCP_DDC_READ_TRACE(hdcp, "BSTATUS", \ + (uint8_t *)&hdcp->auth.msg.hdcp1.bstatus, \ + sizeof(hdcp->auth.msg.hdcp1.bstatus)); \ HDCP_DDC_WRITE_TRACE(hdcp, "AN", hdcp->auth.msg.hdcp1.an, \ sizeof(hdcp->auth.msg.hdcp1.an)); \ HDCP_DDC_WRITE_TRACE(hdcp, "AKSV", hdcp->auth.msg.hdcp1.aksv, \ diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.c index 7911dc157d5a..c2929815c3ee 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.c +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.c @@ -44,83 +44,80 @@ static void hdcp2_message_init(struct mod_hdcp *hdcp, in->process.msg3_desc.msg_id = TA_HDCP_HDCP2_MSG_ID__NULL_MESSAGE; in->process.msg3_desc.msg_size = 0; } -enum mod_hdcp_status mod_hdcp_remove_display_topology(struct mod_hdcp *hdcp) -{ - - struct psp_context *psp = hdcp->config.psp.handle; - struct ta_dtm_shared_memory *dtm_cmd; - struct mod_hdcp_display *display = NULL; - uint8_t i; +enum mod_hdcp_status mod_hdcp_remove_display_from_topology( + struct mod_hdcp *hdcp, uint8_t index) + { + struct psp_context *psp = hdcp->config.psp.handle; + struct ta_dtm_shared_memory *dtm_cmd; + struct mod_hdcp_display *display = + get_active_display_at_index(hdcp, index); dtm_cmd = (struct ta_dtm_shared_memory *)psp->dtm_context.dtm_shared_buf; - for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) { - if (is_display_added(&(hdcp->connection.displays[i]))) { - - memset(dtm_cmd, 0, sizeof(struct ta_dtm_shared_memory)); - - display = &hdcp->connection.displays[i]; + if (!display || !is_display_added(display)) + return MOD_HDCP_STATUS_DISPLAY_NOT_FOUND; - dtm_cmd->cmd_id = TA_DTM_COMMAND__TOPOLOGY_UPDATE_V2; - dtm_cmd->dtm_in_message.topology_update_v2.display_handle = display->index; - dtm_cmd->dtm_in_message.topology_update_v2.is_active = 0; - dtm_cmd->dtm_status = TA_DTM_STATUS__GENERIC_FAILURE; + memset(dtm_cmd, 0, sizeof(struct ta_dtm_shared_memory)); - psp_dtm_invoke(psp, dtm_cmd->cmd_id); + dtm_cmd->cmd_id = TA_DTM_COMMAND__TOPOLOGY_UPDATE_V2; + dtm_cmd->dtm_in_message.topology_update_v2.display_handle = display->index; + dtm_cmd->dtm_in_message.topology_update_v2.is_active = 0; + dtm_cmd->dtm_status = TA_DTM_STATUS__GENERIC_FAILURE; - if (dtm_cmd->dtm_status != TA_DTM_STATUS__SUCCESS) - return MOD_HDCP_STATUS_UPDATE_TOPOLOGY_FAILURE; + psp_dtm_invoke(psp, dtm_cmd->cmd_id); - display->state = MOD_HDCP_DISPLAY_ACTIVE; - HDCP_TOP_REMOVE_DISPLAY_TRACE(hdcp, display->index); - } - } + if (dtm_cmd->dtm_status != TA_DTM_STATUS__SUCCESS) + return MOD_HDCP_STATUS_UPDATE_TOPOLOGY_FAILURE; - return MOD_HDCP_STATUS_SUCCESS; + display->state = MOD_HDCP_DISPLAY_ACTIVE; + HDCP_TOP_REMOVE_DISPLAY_TRACE(hdcp, display->index); + + return MOD_HDCP_STATUS_SUCCESS; + } - -enum mod_hdcp_status mod_hdcp_add_display_topology(struct mod_hdcp *hdcp) +enum mod_hdcp_status mod_hdcp_add_display_to_topology(struct mod_hdcp *hdcp, + uint8_t index) { struct psp_context *psp = hdcp->config.psp.handle; struct ta_dtm_shared_memory *dtm_cmd; - struct mod_hdcp_display *display = NULL; + struct mod_hdcp_display *display = + get_active_display_at_index(hdcp, index); struct mod_hdcp_link *link = &hdcp->connection.link; - uint8_t i; if (!psp->dtm_context.dtm_initialized) { DRM_ERROR("Failed to add display topology, DTM TA is not initialized."); return MOD_HDCP_STATUS_FAILURE; } + if (!display || is_display_added(display)) + return MOD_HDCP_STATUS_UPDATE_TOPOLOGY_FAILURE; + dtm_cmd = (struct ta_dtm_shared_memory *)psp->dtm_context.dtm_shared_buf; - for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) { - if (hdcp->connection.displays[i].state == MOD_HDCP_DISPLAY_ACTIVE) { - display = &hdcp->connection.displays[i]; - - memset(dtm_cmd, 0, sizeof(struct ta_dtm_shared_memory)); - - dtm_cmd->cmd_id = TA_DTM_COMMAND__TOPOLOGY_UPDATE_V2; - dtm_cmd->dtm_in_message.topology_update_v2.display_handle = display->index; - dtm_cmd->dtm_in_message.topology_update_v2.is_active = 1; - dtm_cmd->dtm_in_message.topology_update_v2.controller = display->controller; - dtm_cmd->dtm_in_message.topology_update_v2.ddc_line = link->ddc_line; - dtm_cmd->dtm_in_message.topology_update_v2.dig_be = link->dig_be; - dtm_cmd->dtm_in_message.topology_update_v2.dig_fe = display->dig_fe; - dtm_cmd->dtm_in_message.topology_update_v2.dp_mst_vcid = display->vc_id; - dtm_cmd->dtm_in_message.topology_update_v2.max_hdcp_supported_version = - TA_DTM_HDCP_VERSION_MAX_SUPPORTED__2_2; - dtm_cmd->dtm_status = TA_DTM_STATUS__GENERIC_FAILURE; - - psp_dtm_invoke(psp, dtm_cmd->cmd_id); - - if (dtm_cmd->dtm_status != TA_DTM_STATUS__SUCCESS) - return MOD_HDCP_STATUS_UPDATE_TOPOLOGY_FAILURE; - - display->state = MOD_HDCP_DISPLAY_ACTIVE_AND_ADDED; - HDCP_TOP_ADD_DISPLAY_TRACE(hdcp, display->index); - } - } + memset(dtm_cmd, 0, sizeof(struct ta_dtm_shared_memory)); + + dtm_cmd->cmd_id = TA_DTM_COMMAND__TOPOLOGY_UPDATE_V2; + dtm_cmd->dtm_in_message.topology_update_v2.display_handle = display->index; + dtm_cmd->dtm_in_message.topology_update_v2.is_active = 1; + dtm_cmd->dtm_in_message.topology_update_v2.controller = display->controller; + dtm_cmd->dtm_in_message.topology_update_v2.ddc_line = link->ddc_line; + dtm_cmd->dtm_in_message.topology_update_v2.dig_be = link->dig_be; + dtm_cmd->dtm_in_message.topology_update_v2.dig_fe = display->dig_fe; + if (is_dp_hdcp(hdcp)) + dtm_cmd->dtm_in_message.topology_update_v2.is_assr = link->dp.assr_supported; + + dtm_cmd->dtm_in_message.topology_update_v2.dp_mst_vcid = display->vc_id; + dtm_cmd->dtm_in_message.topology_update_v2.max_hdcp_supported_version = + TA_DTM_HDCP_VERSION_MAX_SUPPORTED__2_2; + dtm_cmd->dtm_status = TA_DTM_STATUS__GENERIC_FAILURE; + + psp_dtm_invoke(psp, dtm_cmd->cmd_id); + + if (dtm_cmd->dtm_status != TA_DTM_STATUS__SUCCESS) + return MOD_HDCP_STATUS_UPDATE_TOPOLOGY_FAILURE; + + display->state = MOD_HDCP_DISPLAY_ACTIVE_AND_ADDED; + HDCP_TOP_ADD_DISPLAY_TRACE(hdcp, display->index); return MOD_HDCP_STATUS_SUCCESS; } @@ -164,6 +161,7 @@ enum mod_hdcp_status mod_hdcp_hdcp1_destroy_session(struct mod_hdcp *hdcp) struct psp_context *psp = hdcp->config.psp.handle; struct ta_hdcp_shared_memory *hdcp_cmd; + uint8_t i = 0; hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); @@ -177,6 +175,14 @@ enum mod_hdcp_status mod_hdcp_hdcp1_destroy_session(struct mod_hdcp *hdcp) return MOD_HDCP_STATUS_HDCP1_DESTROY_SESSION_FAILURE; HDCP_TOP_HDCP1_DESTROY_SESSION_TRACE(hdcp); + for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) + if (is_display_encryption_enabled( + &hdcp->displays[i])) { + hdcp->displays[i].state = + MOD_HDCP_DISPLAY_ACTIVE_AND_ADDED; + HDCP_HDCP1_DISABLED_TRACE(hdcp, + hdcp->displays[i].index); + } return MOD_HDCP_STATUS_SUCCESS; } @@ -210,6 +216,10 @@ enum mod_hdcp_status mod_hdcp_hdcp1_validate_rx(struct mod_hdcp *hdcp) } else if (hdcp_cmd->out_msg.hdcp1_first_part_authentication.authentication_status == TA_HDCP_AUTHENTICATION_STATUS__HDCP1_AUTHENTICATED) { hdcp->connection.is_repeater = 0; + } else if (hdcp_cmd->out_msg.hdcp1_first_part_authentication.authentication_status == + TA_HDCP_AUTHENTICATION_STATUS__HDCP1_KSV_REVOKED) { + hdcp->connection.is_hdcp1_revoked = 1; + return MOD_HDCP_STATUS_HDCP1_BKSV_REVOKED; } else return MOD_HDCP_STATUS_HDCP1_VALIDATE_RX_FAILURE; @@ -245,6 +255,7 @@ enum mod_hdcp_status mod_hdcp_hdcp1_validate_ksvlist_vp(struct mod_hdcp *hdcp) { struct psp_context *psp = hdcp->config.psp.handle; struct ta_hdcp_shared_memory *hdcp_cmd; + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); @@ -264,10 +275,19 @@ enum mod_hdcp_status mod_hdcp_hdcp1_validate_ksvlist_vp(struct mod_hdcp *hdcp) psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); - if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) - return MOD_HDCP_STATUS_HDCP1_VALIDATE_KSV_LIST_FAILURE; + if (hdcp_cmd->hdcp_status == TA_HDCP_STATUS__SUCCESS && + hdcp_cmd->out_msg.hdcp1_second_part_authentication.authentication_status == + TA_HDCP_AUTHENTICATION_STATUS__HDCP1_AUTHENTICATED) { + status = MOD_HDCP_STATUS_SUCCESS; + } else if (hdcp_cmd->out_msg.hdcp1_second_part_authentication.authentication_status == + TA_HDCP_AUTHENTICATION_STATUS__HDCP1_KSV_REVOKED) { + hdcp->connection.is_hdcp1_revoked = 1; + status = MOD_HDCP_STATUS_HDCP1_KSV_LIST_REVOKED; + } else { + status = MOD_HDCP_STATUS_HDCP1_VALIDATE_KSV_LIST_FAILURE; + } - return MOD_HDCP_STATUS_SUCCESS; + return status; } enum mod_hdcp_status mod_hdcp_hdcp1_enable_dp_stream_encryption(struct mod_hdcp *hdcp) @@ -281,14 +301,14 @@ enum mod_hdcp_status mod_hdcp_hdcp1_enable_dp_stream_encryption(struct mod_hdcp for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) { - if (hdcp->connection.displays[i].state != MOD_HDCP_DISPLAY_ACTIVE_AND_ADDED || - hdcp->connection.displays[i].adjust.disable) + if (hdcp->displays[i].state != MOD_HDCP_DISPLAY_ACTIVE_AND_ADDED || + hdcp->displays[i].adjust.disable) continue; memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); hdcp_cmd->in_msg.hdcp1_enable_dp_stream_encryption.session_handle = hdcp->auth.id; - hdcp_cmd->in_msg.hdcp1_enable_dp_stream_encryption.display_handle = hdcp->connection.displays[i].index; + hdcp_cmd->in_msg.hdcp1_enable_dp_stream_encryption.display_handle = hdcp->displays[i].index; hdcp_cmd->cmd_id = TA_HDCP_COMMAND__HDCP1_ENABLE_DP_STREAM_ENCRYPTION; psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); @@ -296,8 +316,8 @@ enum mod_hdcp_status mod_hdcp_hdcp1_enable_dp_stream_encryption(struct mod_hdcp if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) return MOD_HDCP_STATUS_HDCP1_ENABLE_STREAM_ENCRYPTION_FAILURE; - hdcp->connection.displays[i].state = MOD_HDCP_DISPLAY_ENCRYPTION_ENABLED; - HDCP_HDCP1_ENABLED_TRACE(hdcp, hdcp->connection.displays[i].index); + hdcp->displays[i].state = MOD_HDCP_DISPLAY_ENCRYPTION_ENABLED; + HDCP_HDCP1_ENABLED_TRACE(hdcp, hdcp->displays[i].index); } return MOD_HDCP_STATUS_SUCCESS; @@ -385,6 +405,7 @@ enum mod_hdcp_status mod_hdcp_hdcp2_destroy_session(struct mod_hdcp *hdcp) { struct psp_context *psp = hdcp->config.psp.handle; struct ta_hdcp_shared_memory *hdcp_cmd; + uint8_t i = 0; hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); @@ -398,6 +419,14 @@ enum mod_hdcp_status mod_hdcp_hdcp2_destroy_session(struct mod_hdcp *hdcp) return MOD_HDCP_STATUS_HDCP2_DESTROY_SESSION_FAILURE; HDCP_TOP_HDCP2_DESTROY_SESSION_TRACE(hdcp); + for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) + if (is_display_encryption_enabled( + &hdcp->displays[i])) { + hdcp->displays[i].state = + MOD_HDCP_DISPLAY_ACTIVE_AND_ADDED; + HDCP_HDCP2_DISABLED_TRACE(hdcp, + hdcp->displays[i].index); + } return MOD_HDCP_STATUS_SUCCESS; } @@ -473,9 +502,12 @@ enum mod_hdcp_status mod_hdcp_hdcp2_validate_ake_cert(struct mod_hdcp *hdcp) hdcp->connection.is_km_stored = msg_out->process.is_km_stored ? 1 : 0; hdcp->connection.is_repeater = msg_out->process.is_repeater ? 1 : 0; return MOD_HDCP_STATUS_SUCCESS; + } else if (msg_out->process.msg1_status == TA_HDCP2_MSG_AUTHENTICATION_STATUS__RECEIVERID_REVOKED) { + hdcp->connection.is_hdcp2_revoked = 1; + return MOD_HDCP_STATUS_HDCP2_AKE_CERT_REVOKED; } - return MOD_HDCP_STATUS_FAILURE; + return MOD_HDCP_STATUS_HDCP2_VALIDATE_AKE_CERT_FAILURE; } enum mod_hdcp_status mod_hdcp_hdcp2_validate_h_prime(struct mod_hdcp *hdcp) @@ -630,20 +662,15 @@ enum mod_hdcp_status mod_hdcp_hdcp2_enable_encryption(struct mod_hdcp *hdcp) { struct psp_context *psp = hdcp->config.psp.handle; struct ta_hdcp_shared_memory *hdcp_cmd; - struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_input_v2 *msg_in; struct mod_hdcp_display *display = get_first_added_display(hdcp); hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); - msg_in = &hdcp_cmd->in_msg.hdcp2_prepare_process_authentication_message_v2; - - hdcp2_message_init(hdcp, msg_in); - if (!display) return MOD_HDCP_STATUS_DISPLAY_NOT_FOUND; - hdcp_cmd->in_msg.hdcp1_enable_encryption.session_handle = hdcp->auth.id; + hdcp_cmd->in_msg.hdcp2_set_encryption.session_handle = hdcp->auth.id; hdcp_cmd->cmd_id = TA_HDCP_COMMAND__HDCP2_SET_ENCRYPTION; psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); @@ -695,6 +722,9 @@ enum mod_hdcp_status mod_hdcp_hdcp2_validate_rx_id_list(struct mod_hdcp *hdcp) hdcp->connection.is_km_stored = msg_out->process.is_km_stored ? 1 : 0; hdcp->connection.is_repeater = msg_out->process.is_repeater ? 1 : 0; return MOD_HDCP_STATUS_SUCCESS; + } else if (msg_out->process.msg1_status == TA_HDCP2_MSG_AUTHENTICATION_STATUS__RECEIVERID_REVOKED) { + hdcp->connection.is_hdcp2_revoked = 1; + return MOD_HDCP_STATUS_HDCP2_RX_ID_LIST_REVOKED; } @@ -717,10 +747,10 @@ enum mod_hdcp_status mod_hdcp_hdcp2_enable_dp_stream_encryption(struct mod_hdcp for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) { - if (hdcp->connection.displays[i].state != MOD_HDCP_DISPLAY_ACTIVE_AND_ADDED || - hdcp->connection.displays[i].adjust.disable) + if (hdcp->displays[i].state != MOD_HDCP_DISPLAY_ACTIVE_AND_ADDED || + hdcp->displays[i].adjust.disable) continue; - hdcp_cmd->in_msg.hdcp2_enable_dp_stream_encryption.display_handle = hdcp->connection.displays[i].index; + hdcp_cmd->in_msg.hdcp2_enable_dp_stream_encryption.display_handle = hdcp->displays[i].index; hdcp_cmd->in_msg.hdcp2_enable_dp_stream_encryption.session_handle = hdcp->auth.id; hdcp_cmd->cmd_id = TA_HDCP_COMMAND__HDCP2_ENABLE_DP_STREAM_ENCRYPTION; @@ -729,8 +759,8 @@ enum mod_hdcp_status mod_hdcp_hdcp2_enable_dp_stream_encryption(struct mod_hdcp if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) break; - hdcp->connection.displays[i].state = MOD_HDCP_DISPLAY_ENCRYPTION_ENABLED; - HDCP_HDCP2_ENABLED_TRACE(hdcp, hdcp->connection.displays[i].index); + hdcp->displays[i].state = MOD_HDCP_DISPLAY_ENCRYPTION_ENABLED; + HDCP_HDCP2_ENABLED_TRACE(hdcp, hdcp->displays[i].index); } return (hdcp_cmd->hdcp_status == TA_HDCP_STATUS__SUCCESS) ? MOD_HDCP_STATUS_SUCCESS diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.h b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.h index 82a5e997d573..1a663dbbf810 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.h +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.h @@ -117,6 +117,8 @@ struct ta_dtm_shared_memory { int psp_cmd_submit_buf(struct psp_context *psp, struct amdgpu_firmware_info *ucode, struct psp_gfx_cmd_resp *cmd, uint64_t fence_mc_addr); +enum { PSP_HDCP_SRM_FIRST_GEN_MAX_SIZE = 5120 }; + enum ta_hdcp_command { TA_HDCP_COMMAND__INITIALIZE, TA_HDCP_COMMAND__HDCP1_CREATE_SESSION, @@ -134,7 +136,10 @@ enum ta_hdcp_command { TA_HDCP_COMMAND__UNUSED_3, TA_HDCP_COMMAND__HDCP2_CREATE_SESSION_V2, TA_HDCP_COMMAND__HDCP2_PREPARE_PROCESS_AUTHENTICATION_MSG_V2, - TA_HDCP_COMMAND__HDCP2_ENABLE_DP_STREAM_ENCRYPTION + TA_HDCP_COMMAND__HDCP2_ENABLE_DP_STREAM_ENCRYPTION, + TA_HDCP_COMMAND__HDCP_DESTROY_ALL_SESSIONS, + TA_HDCP_COMMAND__HDCP_SET_SRM, + TA_HDCP_COMMAND__HDCP_GET_SRM }; enum ta_hdcp2_msg_id { @@ -235,7 +240,8 @@ enum ta_hdcp_authentication_status { TA_HDCP_AUTHENTICATION_STATUS__HDCP22_AUTHENTICATION_PENDING = 0x06, TA_HDCP_AUTHENTICATION_STATUS__HDCP22_AUTHENTICATION_FAILED = 0x07, TA_HDCP_AUTHENTICATION_STATUS__HDCP22_AUTHENTICATED = 0x08, - TA_HDCP_AUTHENTICATION_STATUS__HDCP1_KSV_VALIDATION_FAILED = 0x09 + TA_HDCP_AUTHENTICATION_STATUS__HDCP1_KSV_VALIDATION_FAILED = 0x09, + TA_HDCP_AUTHENTICATION_STATUS__HDCP1_KSV_REVOKED = 0x0A }; enum ta_hdcp2_msg_authentication_status { @@ -253,7 +259,8 @@ enum ta_hdcp2_msg_authentication_status { TA_HDCP2_MSG_AUTHENTICATION_STATUS__INVALID_SEQ_NUM, TA_HDCP2_MSG_AUTHENTICATION_STATUS__INVALID_SIZE, TA_HDCP2_MSG_AUTHENTICATION_STATUS__INVALID_LENGTH, - TA_HDCP2_MSG_AUTHENTICATION_STATUS__REAUTH_REQUEST + TA_HDCP2_MSG_AUTHENTICATION_STATUS__REAUTH_REQUEST, + TA_HDCP2_MSG_AUTHENTICATION_STATUS__RECEIVERID_REVOKED }; enum ta_hdcp_content_type { @@ -415,6 +422,22 @@ struct ta_hdcp_cmd_hdcp2_enable_dp_stream_encryption_input { uint32_t display_handle; }; +struct ta_hdcp_cmd_set_srm_input { + uint32_t srm_buf_size; + uint8_t srm_buf[PSP_HDCP_SRM_FIRST_GEN_MAX_SIZE]; +}; + +struct ta_hdcp_cmd_set_srm_output { + uint8_t valid_signature; + uint32_t srm_version; +}; + +struct ta_hdcp_cmd_get_srm_output { + uint32_t srm_version; + uint32_t srm_buf_size; + uint8_t srm_buf[PSP_HDCP_SRM_FIRST_GEN_MAX_SIZE]; +}; + /**********************************************************/ /* Common input structure for HDCP callbacks */ union ta_hdcp_cmd_input { @@ -432,6 +455,7 @@ union ta_hdcp_cmd_input { struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_input_v2 hdcp2_prepare_process_authentication_message_v2; struct ta_hdcp_cmd_hdcp2_enable_dp_stream_encryption_input hdcp2_enable_dp_stream_encryption; + struct ta_hdcp_cmd_set_srm_input hdcp_set_srm; }; /* Common output structure for HDCP callbacks */ @@ -444,6 +468,8 @@ union ta_hdcp_cmd_output { struct ta_hdcp_cmd_hdcp2_create_session_output_v2 hdcp2_create_session_v2; struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_output_v2 hdcp2_prepare_process_authentication_message_v2; + struct ta_hdcp_cmd_set_srm_output hdcp_set_srm; + struct ta_hdcp_cmd_get_srm_output hdcp_get_srm; }; /**********************************************************/ diff --git a/drivers/gpu/drm/amd/display/modules/inc/mod_hdcp.h b/drivers/gpu/drm/amd/display/modules/inc/mod_hdcp.h index f2a0e1a064da..c088602bc1a0 100644 --- a/drivers/gpu/drm/amd/display/modules/inc/mod_hdcp.h +++ b/drivers/gpu/drm/amd/display/modules/inc/mod_hdcp.h @@ -56,8 +56,10 @@ enum mod_hdcp_status { MOD_HDCP_STATUS_HDCP1_NOT_CAPABLE, MOD_HDCP_STATUS_HDCP1_R0_PRIME_PENDING, MOD_HDCP_STATUS_HDCP1_VALIDATE_RX_FAILURE, + MOD_HDCP_STATUS_HDCP1_BKSV_REVOKED, MOD_HDCP_STATUS_HDCP1_KSV_LIST_NOT_READY, MOD_HDCP_STATUS_HDCP1_VALIDATE_KSV_LIST_FAILURE, + MOD_HDCP_STATUS_HDCP1_KSV_LIST_REVOKED, MOD_HDCP_STATUS_HDCP1_ENABLE_ENCRYPTION, MOD_HDCP_STATUS_HDCP1_ENABLE_STREAM_ENCRYPTION_FAILURE, MOD_HDCP_STATUS_HDCP1_MAX_CASCADE_EXCEEDED_FAILURE, @@ -100,6 +102,7 @@ enum mod_hdcp_status { struct mod_hdcp_displayport { uint8_t rev; uint8_t assr_supported; + uint8_t mst_supported; }; struct mod_hdcp_hdmi { @@ -108,8 +111,7 @@ struct mod_hdcp_hdmi { enum mod_hdcp_operation_mode { MOD_HDCP_MODE_OFF, MOD_HDCP_MODE_DEFAULT, - MOD_HDCP_MODE_DP, - MOD_HDCP_MODE_DP_MST + MOD_HDCP_MODE_DP }; enum mod_hdcp_display_state { @@ -155,7 +157,8 @@ struct mod_hdcp_display_adjustment { struct mod_hdcp_link_adjustment_hdcp1 { uint8_t disable : 1; uint8_t postpone_encryption : 1; - uint8_t reserved : 6; + uint8_t min_auth_retries_wa : 1; + uint8_t reserved : 5; }; enum mod_hdcp_force_hdcp_type { diff --git a/drivers/gpu/drm/amd/display/modules/inc/mod_info_packet.h b/drivers/gpu/drm/amd/display/modules/inc/mod_info_packet.h index 42cbeffac640..13c57ff2abdc 100644 --- a/drivers/gpu/drm/amd/display/modules/inc/mod_info_packet.h +++ b/drivers/gpu/drm/amd/display/modules/inc/mod_info_packet.h @@ -34,8 +34,7 @@ struct dc_info_packet; struct mod_vrr_params; void mod_build_vsc_infopacket(const struct dc_stream_state *stream, - struct dc_info_packet *info_packet, - bool *use_vsc_sdp_for_colorimetry); + struct dc_info_packet *info_packet); void mod_build_hf_vsif_infopacket(const struct dc_stream_state *stream, struct dc_info_packet *info_packet, int ALLMEnabled, int ALLMValue); diff --git a/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c b/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c index 6a8a056424b8..cff3ab15fc0c 100644 --- a/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c +++ b/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c @@ -130,8 +130,7 @@ enum ColorimetryYCCDP { }; void mod_build_vsc_infopacket(const struct dc_stream_state *stream, - struct dc_info_packet *info_packet, - bool *use_vsc_sdp_for_colorimetry) + struct dc_info_packet *info_packet) { unsigned int vsc_packet_revision = vsc_packet_undefined; unsigned int i; @@ -139,11 +138,6 @@ void mod_build_vsc_infopacket(const struct dc_stream_state *stream, unsigned int colorimetryFormat = 0; bool stereo3dSupport = false; - /* Initialize first, later if infopacket is valid determine if VSC SDP - * should be used to signal colorimetry format and pixel encoding. - */ - *use_vsc_sdp_for_colorimetry = false; - if (stream->timing.timing_3d_format != TIMING_3D_FORMAT_NONE && stream->view_format != VIEW_3D_FORMAT_NONE) { vsc_packet_revision = vsc_packet_rev1; stereo3dSupport = true; @@ -153,9 +147,8 @@ void mod_build_vsc_infopacket(const struct dc_stream_state *stream, if (stream->psr_version != 0) vsc_packet_revision = vsc_packet_rev2; - /* Update to revision 5 for extended colorimetry support for DPCD 1.4+ */ - if (stream->link->dpcd_caps.dpcd_rev.raw >= 0x14 && - stream->link->dpcd_caps.dprx_feature.bits.VSC_SDP_COLORIMETRY_SUPPORTED) + /* Update to revision 5 for extended colorimetry support */ + if (stream->use_vsc_sdp_for_colorimetry) vsc_packet_revision = vsc_packet_rev5; /* VSC packet not needed based on the features @@ -269,13 +262,6 @@ void mod_build_vsc_infopacket(const struct dc_stream_state *stream, info_packet->valid = true; - /* If we are using VSC SDP revision 05h, use this to signal for - * colorimetry format and pixel encoding. HW should later be - * programmed to set MSA MISC1 bit 6 to indicate ignore - * colorimetry format and pixel encoding in the MSA. - */ - *use_vsc_sdp_for_colorimetry = true; - /* Set VSC SDP fields for pixel encoding and colorimetry format from DP 1.3 specs * Data Bytes DB 18~16 * Bits 3:0 (Colorimetry Format) | Bits 7:4 (Pixel Encoding) diff --git a/drivers/gpu/drm/amd/display/modules/vmid/vmid.c b/drivers/gpu/drm/amd/display/modules/vmid/vmid.c index f0a153704f6e..00f132f8ad55 100644 --- a/drivers/gpu/drm/amd/display/modules/vmid/vmid.c +++ b/drivers/gpu/drm/amd/display/modules/vmid/vmid.c @@ -40,14 +40,18 @@ struct core_vmid { static void add_ptb_to_table(struct core_vmid *core_vmid, unsigned int vmid, uint64_t ptb) { - core_vmid->ptb_assigned_to_vmid[vmid] = ptb; - core_vmid->num_vmids_available--; + if (vmid < MAX_VMID) { + core_vmid->ptb_assigned_to_vmid[vmid] = ptb; + core_vmid->num_vmids_available--; + } } static void clear_entry_from_vmid_table(struct core_vmid *core_vmid, unsigned int vmid) { - core_vmid->ptb_assigned_to_vmid[vmid] = 0; - core_vmid->num_vmids_available++; + if (vmid < MAX_VMID) { + core_vmid->ptb_assigned_to_vmid[vmid] = 0; + core_vmid->num_vmids_available++; + } } static void evict_vmids(struct core_vmid *core_vmid) @@ -57,7 +61,7 @@ static void evict_vmids(struct core_vmid *core_vmid) // At this point any positions with value 0 are unused vmids, evict them for (i = 1; i < core_vmid->num_vmid; i++) { - if (ord & (1u << i)) + if (!(ord & (1u << i))) clear_entry_from_vmid_table(core_vmid, i); } } @@ -91,7 +95,7 @@ static int get_next_available_vmid(struct core_vmid *core_vmid) uint8_t mod_vmid_get_for_ptb(struct mod_vmid *mod_vmid, uint64_t ptb) { struct core_vmid *core_vmid = MOD_VMID_TO_CORE(mod_vmid); - unsigned int vmid = 0; + int vmid = 0; // Physical address gets vmid 0 if (ptb == 0) diff --git a/drivers/gpu/drm/amd/include/asic_reg/dce/dce_12_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/dce/dce_12_0_offset.h index b6f74bf4af02..27bb8c1ab858 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/dce/dce_12_0_offset.h +++ b/drivers/gpu/drm/amd/include/asic_reg/dce/dce_12_0_offset.h @@ -7376,6 +7376,8 @@ #define mmCRTC4_CRTC_DRR_CONTROL 0x0f3e #define mmCRTC4_CRTC_DRR_CONTROL_BASE_IDX 2 +#define mmDCHUBBUB_SDPIF_MMIO_CNTRL_0 0x395d +#define mmDCHUBBUB_SDPIF_MMIO_CNTRL_0_BASE_IDX 2 // addressBlock: dce_dc_fmt4_dispdec // base address: 0x2000 diff --git a/drivers/gpu/drm/amd/include/asic_reg/wafl/wafl2_4_0_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/wafl/wafl2_4_0_0_sh_mask.h new file mode 100644 index 000000000000..82b6cc25205e --- /dev/null +++ b/drivers/gpu/drm/amd/include/asic_reg/wafl/wafl2_4_0_0_sh_mask.h @@ -0,0 +1,69 @@ +/* + * Copyright 2020 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ + +#ifndef _wafl2_4_0_0_SH_MASK_HEADER +#define _wafl2_4_0_0_SH_MASK_HEADER + +//PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__DataLossErr__SHIFT 0x0 +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__TrainingErr__SHIFT 0x1 +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__CRCErr__SHIFT 0x5 +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__BERExceededErr__SHIFT 0x6 +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__TxMetaDataErr__SHIFT 0x7 +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__ReplayBufParityErr__SHIFT 0x8 +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__DataParityErr__SHIFT 0x9 +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__ReplayFifoOverflowErr__SHIFT 0xa +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__ReplayFifoUnderflowErr__SHIFT 0xb +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__ElasticFifoOverflowErr__SHIFT 0xc +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__DeskewErr__SHIFT 0xd +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__DataStartupLimitErr__SHIFT 0xf +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__FCInitTimeoutErr__SHIFT 0x10 +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__RecoveryTimeoutErr__SHIFT 0x11 +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__ReadySerialTimeoutErr__SHIFT 0x12 +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__ReadySerialAttemptErr__SHIFT 0x13 +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__RecoveryAttemptErr__SHIFT 0x14 +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__RecoveryRelockAttemptErr__SHIFT 0x15 +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__ClearBERAccum__SHIFT 0x17 +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__BERAccumulator__SHIFT 0x18 +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__DataLossErr_MASK 0x00000001L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__TrainingErr_MASK 0x00000002L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__CRCErr_MASK 0x00000020L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__BERExceededErr_MASK 0x00000040L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__TxMetaDataErr_MASK 0x00000080L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__ReplayBufParityErr_MASK 0x00000100L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__DataParityErr_MASK 0x00000200L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__ReplayFifoOverflowErr_MASK 0x00000400L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__ReplayFifoUnderflowErr_MASK 0x00000800L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__ElasticFifoOverflowErr_MASK 0x00001000L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__DeskewErr_MASK 0x00002000L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__DataStartupLimitErr_MASK 0x00008000L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__FCInitTimeoutErr_MASK 0x00010000L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__RecoveryTimeoutErr_MASK 0x00020000L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__ReadySerialTimeoutErr_MASK 0x00040000L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__ReadySerialAttemptErr_MASK 0x00080000L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__RecoveryAttemptErr_MASK 0x00100000L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__RecoveryRelockAttemptErr_MASK 0x00200000L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__ClearBERAccum_MASK 0x00800000L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__BERAccumulator_MASK 0xFF000000L + +#endif diff --git a/drivers/gpu/drm/amd/include/asic_reg/wafl/wafl2_4_0_0_smn.h b/drivers/gpu/drm/amd/include/asic_reg/wafl/wafl2_4_0_0_smn.h new file mode 100644 index 000000000000..4a51a90c611a --- /dev/null +++ b/drivers/gpu/drm/amd/include/asic_reg/wafl/wafl2_4_0_0_smn.h @@ -0,0 +1,29 @@ +/* + * Copyright 2020 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ + +#ifndef _wafl2_4_0_0_SMN_HEADER +#define _wafl2_4_0_0_SMN_HEADER + +#define smnPCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS 0x11cf0210 + +#endif diff --git a/drivers/gpu/drm/amd/include/asic_reg/xgmi/xgmi_4_0_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/xgmi/xgmi_4_0_0_sh_mask.h new file mode 100644 index 000000000000..f37712f05b03 --- /dev/null +++ b/drivers/gpu/drm/amd/include/asic_reg/xgmi/xgmi_4_0_0_sh_mask.h @@ -0,0 +1,69 @@ +/* + * Copyright 2020 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ + +#ifndef _xgmi_4_0_0_SH_MASK_HEADER +#define _xgmi_4_0_0_SH_MASK_HEADER + +//PCS_GOPX16_PCS_ERROR_STATUS +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__DataLossErr__SHIFT 0x0 +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__TrainingErr__SHIFT 0x1 +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__CRCErr__SHIFT 0x5 +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__BERExceededErr__SHIFT 0x6 +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__TxMetaDataErr__SHIFT 0x7 +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__ReplayBufParityErr__SHIFT 0x8 +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__DataParityErr__SHIFT 0x9 +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__ReplayFifoOverflowErr__SHIFT 0xa +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__ReplayFifoUnderflowErr__SHIFT 0xb +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__ElasticFifoOverflowErr__SHIFT 0xc +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__DeskewErr__SHIFT 0xd +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__DataStartupLimitErr__SHIFT 0xf +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__FCInitTimeoutErr__SHIFT 0x10 +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__RecoveryTimeoutErr__SHIFT 0x11 +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__ReadySerialTimeoutErr__SHIFT 0x12 +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__ReadySerialAttemptErr__SHIFT 0x13 +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__RecoveryAttemptErr__SHIFT 0x14 +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__RecoveryRelockAttemptErr__SHIFT 0x15 +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__ClearBERAccum__SHIFT 0x17 +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__BERAccumulator__SHIFT 0x18 +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__DataLossErr_MASK 0x00000001L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__TrainingErr_MASK 0x00000002L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__CRCErr_MASK 0x00000020L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__BERExceededErr_MASK 0x00000040L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__TxMetaDataErr_MASK 0x00000080L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__ReplayBufParityErr_MASK 0x00000100L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__DataParityErr_MASK 0x00000200L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__ReplayFifoOverflowErr_MASK 0x00000400L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__ReplayFifoUnderflowErr_MASK 0x00000800L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__ElasticFifoOverflowErr_MASK 0x00001000L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__DeskewErr_MASK 0x00002000L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__DataStartupLimitErr_MASK 0x00008000L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__FCInitTimeoutErr_MASK 0x00010000L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__RecoveryTimeoutErr_MASK 0x00020000L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__ReadySerialTimeoutErr_MASK 0x00040000L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__ReadySerialAttemptErr_MASK 0x00080000L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__RecoveryAttemptErr_MASK 0x00100000L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__RecoveryRelockAttemptErr_MASK 0x00200000L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__ClearBERAccum_MASK 0x00800000L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__BERAccumulator_MASK 0xFF000000L + +#endif diff --git a/drivers/gpu/drm/amd/include/asic_reg/xgmi/xgmi_4_0_0_smn.h b/drivers/gpu/drm/amd/include/asic_reg/xgmi/xgmi_4_0_0_smn.h new file mode 100644 index 000000000000..6ccbac4ce87e --- /dev/null +++ b/drivers/gpu/drm/amd/include/asic_reg/xgmi/xgmi_4_0_0_smn.h @@ -0,0 +1,29 @@ +/* + * Copyright 2020 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ + +#ifndef _xgmi_4_0_0_SMN_HEADER +#define _xgmi_4_0_0_SMN_HEADER + +#define smnXGMI0_PCS_GOPX16_PCS_ERROR_STATUS 0x11af0210 + +#endif diff --git a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h index a607b1034962..a3c238c39ef5 100644 --- a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h +++ b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h @@ -123,7 +123,7 @@ struct kgd2kfd_shared_resources { uint32_t num_queue_per_pipe; /* Bit n == 1 means Queue n is available for KFD */ - DECLARE_BITMAP(queue_bitmap, KGD_MAX_QUEUES); + DECLARE_BITMAP(cp_queue_bitmap, KGD_MAX_QUEUES); /* SDMA doorbell assignments (SOC15 and later chips only). Only * specific doorbells are routed to each SDMA engine. Others @@ -151,6 +151,7 @@ struct kgd2kfd_shared_resources { /* Minor device number of the render node */ int drm_render_minor; + }; struct tile_config { @@ -166,27 +167,6 @@ struct tile_config { #define KFD_MAX_NUM_OF_QUEUES_PER_DEVICE_DEFAULT 4096 -/* - * Allocation flag domains - * NOTE: This must match the corresponding definitions in kfd_ioctl.h. - */ -#define ALLOC_MEM_FLAGS_VRAM (1 << 0) -#define ALLOC_MEM_FLAGS_GTT (1 << 1) -#define ALLOC_MEM_FLAGS_USERPTR (1 << 2) -#define ALLOC_MEM_FLAGS_DOORBELL (1 << 3) -#define ALLOC_MEM_FLAGS_MMIO_REMAP (1 << 4) - -/* - * Allocation flags attributes/access options. - * NOTE: This must match the corresponding definitions in kfd_ioctl.h. - */ -#define ALLOC_MEM_FLAGS_WRITABLE (1 << 31) -#define ALLOC_MEM_FLAGS_EXECUTABLE (1 << 30) -#define ALLOC_MEM_FLAGS_PUBLIC (1 << 29) -#define ALLOC_MEM_FLAGS_NO_SUBSTITUTE (1 << 28) /* TODO */ -#define ALLOC_MEM_FLAGS_AQL_QUEUE_MEM (1 << 27) -#define ALLOC_MEM_FLAGS_COHERENT (1 << 26) /* For GFXv9 or later */ - /** * struct kfd2kgd_calls * @@ -222,8 +202,6 @@ struct tile_config { * @set_scratch_backing_va: Sets VA for scratch backing memory of a VMID. * Only used for no cp scheduling mode * - * @get_tile_config: Returns GPU-specific tiling mode information - * * @set_vm_context_page_table_base: Program page table base for a VMID * * @invalidate_tlbs: Invalidate TLBs for a specific PASID @@ -236,6 +214,8 @@ struct tile_config { * * @get_hive_id: Returns hive id of current device, 0 if xgmi is not enabled * + * @get_unique_id: Returns uuid id of current device + * * This structure contains function pointers to services that the kgd driver * provides to amdkfd driver. * @@ -307,12 +287,11 @@ struct kfd2kgd_calls { void (*set_scratch_backing_va)(struct kgd_dev *kgd, uint64_t va, uint32_t vmid); - int (*get_tile_config)(struct kgd_dev *kgd, struct tile_config *config); - void (*set_vm_context_page_table_base)(struct kgd_dev *kgd, uint32_t vmid, uint64_t page_table_base); uint32_t (*read_vmid_from_vmfault_reg)(struct kgd_dev *kgd); uint64_t (*get_hive_id)(struct kgd_dev *kgd); + uint64_t (*get_unique_id)(struct kgd_dev *kgd); }; diff --git a/drivers/gpu/drm/amd/powerplay/amd_powerplay.c b/drivers/gpu/drm/amd/powerplay/amd_powerplay.c index c195575366a3..2a12614a12c2 100644 --- a/drivers/gpu/drm/amd/powerplay/amd_powerplay.c +++ b/drivers/gpu/drm/amd/powerplay/amd_powerplay.c @@ -1452,7 +1452,8 @@ static int pp_get_asic_baco_state(void *handle, int *state) if (!hwmgr) return -EINVAL; - if (!hwmgr->pm_en || !hwmgr->hwmgr_func->get_asic_baco_state) + if (!(hwmgr->not_vf && amdgpu_dpm) || + !hwmgr->hwmgr_func->get_asic_baco_state) return 0; mutex_lock(&hwmgr->smu_lock); diff --git a/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c b/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c index 99ad4ddbe12f..e8b27fab6aa1 100644 --- a/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c +++ b/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c @@ -23,15 +23,12 @@ #include <linux/firmware.h> #include <linux/pci.h> -#include "pp_debug.h" #include "amdgpu.h" #include "amdgpu_smu.h" #include "smu_internal.h" -#include "soc15_common.h" #include "smu_v11_0.h" #include "smu_v12_0.h" #include "atom.h" -#include "amd_pcie.h" #include "vega20_ppt.h" #include "arcturus_ppt.h" #include "navi10_ppt.h" @@ -121,20 +118,20 @@ static int smu_feature_update_enable_state(struct smu_context *smu, if (enabled) { ret = smu_send_smc_msg_with_param(smu, SMU_MSG_EnableSmuFeaturesLow, - feature_low); + feature_low, NULL); if (ret) return ret; ret = smu_send_smc_msg_with_param(smu, SMU_MSG_EnableSmuFeaturesHigh, - feature_high); + feature_high, NULL); if (ret) return ret; } else { ret = smu_send_smc_msg_with_param(smu, SMU_MSG_DisableSmuFeaturesLow, - feature_low); + feature_low, NULL); if (ret) return ret; ret = smu_send_smc_msg_with_param(smu, SMU_MSG_DisableSmuFeaturesHigh, - feature_high); + feature_high, NULL); if (ret) return ret; } @@ -195,21 +192,13 @@ int smu_get_smc_version(struct smu_context *smu, uint32_t *if_version, uint32_t return -EINVAL; if (if_version) { - ret = smu_send_smc_msg(smu, SMU_MSG_GetDriverIfVersion); - if (ret) - return ret; - - ret = smu_read_smc_arg(smu, if_version); + ret = smu_send_smc_msg(smu, SMU_MSG_GetDriverIfVersion, if_version); if (ret) return ret; } if (smu_version) { - ret = smu_send_smc_msg(smu, SMU_MSG_GetSmuVersion); - if (ret) - return ret; - - ret = smu_read_smc_arg(smu, smu_version); + ret = smu_send_smc_msg(smu, SMU_MSG_GetSmuVersion, smu_version); if (ret) return ret; } @@ -218,17 +207,19 @@ int smu_get_smc_version(struct smu_context *smu, uint32_t *if_version, uint32_t } int smu_set_soft_freq_range(struct smu_context *smu, enum smu_clk_type clk_type, - uint32_t min, uint32_t max) + uint32_t min, uint32_t max, bool lock_needed) { int ret = 0; - if (min <= 0 && max <= 0) - return -EINVAL; - if (!smu_clk_dpm_is_enabled(smu, clk_type)) return 0; + if (lock_needed) + mutex_lock(&smu->mutex); ret = smu_set_soft_freq_limited_range(smu, clk_type, min, max); + if (lock_needed) + mutex_unlock(&smu->mutex); + return ret; } @@ -251,7 +242,7 @@ int smu_set_hard_freq_range(struct smu_context *smu, enum smu_clk_type clk_type, if (max > 0) { param = (uint32_t)((clk_id << 16) | (max & 0xffff)); ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMaxByFreq, - param); + param, NULL); if (ret) return ret; } @@ -259,7 +250,7 @@ int smu_set_hard_freq_range(struct smu_context *smu, enum smu_clk_type clk_type, if (min > 0) { param = (uint32_t)((clk_id << 16) | (min & 0xffff)); ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinByFreq, - param); + param, NULL); if (ret) return ret; } @@ -335,12 +326,8 @@ int smu_get_dpm_freq_by_index(struct smu_context *smu, enum smu_clk_type clk_typ param = (uint32_t)(((clk_id & 0xffff) << 16) | (level & 0xffff)); - ret = smu_send_smc_msg_with_param(smu,SMU_MSG_GetDpmFreqByIndex, - param); - if (ret) - return ret; - - ret = smu_read_smc_arg(smu, ¶m); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetDpmFreqByIndex, + param, ¶m); if (ret) return ret; @@ -542,7 +529,8 @@ int smu_update_table(struct smu_context *smu, enum smu_table_id table_index, int ret = smu_send_smc_msg_with_param(smu, drv2smu ? SMU_MSG_TransferTableDram2Smu : SMU_MSG_TransferTableSmu2Dram, - table_id | ((argument & 0xFFFF) << 16)); + table_id | ((argument & 0xFFFF) << 16), + NULL); if (ret) return ret; @@ -900,6 +888,7 @@ static int smu_sw_init(void *handle) mutex_init(&smu->sensor_lock); mutex_init(&smu->metrics_lock); + mutex_init(&smu->message_lock); smu->watermarks_bitmap = 0; smu->power_profile_mode = PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT; @@ -943,6 +932,13 @@ static int smu_sw_init(void *handle) return ret; } + if (adev->smu.ppt_funcs->i2c_eeprom_init) { + ret = smu_i2c_eeprom_init(smu, &adev->pm.smu_i2c); + + if (ret) + return ret; + } + return 0; } @@ -952,6 +948,9 @@ static int smu_sw_fini(void *handle) struct smu_context *smu = &adev->smu; int ret; + if (adev->smu.ppt_funcs->i2c_eeprom_fini) + smu_i2c_eeprom_fini(smu, &adev->pm.smu_i2c); + kfree(smu->irq_source); smu->irq_source = NULL; @@ -1113,12 +1112,12 @@ static int smu_smc_table_hw_init(struct smu_context *smu, return ret; } + ret = smu_set_driver_table_location(smu); + if (ret) + return ret; + /* smu_dump_pptable(smu); */ if (!amdgpu_sriov_vf(adev)) { - ret = smu_set_driver_table_location(smu); - if (ret) - return ret; - /* * Copy pptable bo in the vram to smc with SMU MSGs such as * SetDriverDramAddr and TransferTableDram2Smu. @@ -1155,6 +1154,21 @@ static int smu_smc_table_hw_init(struct smu_context *smu, } } } + + if (smu->ppt_funcs->set_power_source) { + /* + * For Navi1X, manually switch it to AC mode as PMFW + * may boot it with DC mode. + */ + if (adev->pm.ac_power) + ret = smu_set_power_source(smu, SMU_POWER_SOURCE_AC); + else + ret = smu_set_power_source(smu, SMU_POWER_SOURCE_DC); + if (ret) { + pr_err("Failed to switch to %s mode!\n", adev->pm.ac_power ? "AC" : "DC"); + return ret; + } + } } if (adev->asic_type != CHIP_ARCTURUS) { ret = smu_notify_display_change(smu); @@ -1454,29 +1468,84 @@ int smu_reset(struct smu_context *smu) return ret; } +static int smu_disable_dpm(struct smu_context *smu) +{ + struct amdgpu_device *adev = smu->adev; + uint32_t smu_version; + int ret = 0; + bool use_baco = !smu->is_apu && + ((adev->in_gpu_reset && + (amdgpu_asic_reset_method(adev) == AMD_RESET_METHOD_BACO)) || + (adev->in_runpm && amdgpu_asic_supports_baco(adev))); + + ret = smu_get_smc_version(smu, NULL, &smu_version); + if (ret) { + pr_err("Failed to get smu version.\n"); + return ret; + } + + /* + * Disable all enabled SMU features. + * This should be handled in SMU FW, as a backup + * driver can issue call to SMU FW until sequence + * in SMU FW is operational. + */ + ret = smu_system_features_control(smu, false); + if (ret) { + pr_err("Failed to disable smu features.\n"); + return ret; + } + + /* + * Arcturus does not have BACO bit in disable feature mask. + * Enablement of BACO bit on Arcturus should be skipped. + */ + if (adev->asic_type == CHIP_ARCTURUS) { + if (use_baco && (smu_version > 0x360e00)) + return 0; + } + + /* For baco, need to leave BACO feature enabled */ + if (use_baco) { + /* + * Correct the way for checking whether SMU_FEATURE_BACO_BIT + * is supported. + * + * Since 'smu_feature_is_enabled(smu, SMU_FEATURE_BACO_BIT)' will + * always return false as the 'smu_system_features_control(smu, false)' + * was just issued above which disabled all SMU features. + * + * Thus 'smu_feature_get_index(smu, SMU_FEATURE_BACO_BIT)' is used + * now for the checking. + */ + if (smu_feature_get_index(smu, SMU_FEATURE_BACO_BIT) >= 0) { + ret = smu_feature_set_enabled(smu, SMU_FEATURE_BACO_BIT, true); + if (ret) { + pr_warn("set BACO feature enabled failed, return %d\n", ret); + return ret; + } + } + } + + return ret; +} + static int smu_suspend(void *handle) { - int ret; struct amdgpu_device *adev = (struct amdgpu_device *)handle; struct smu_context *smu = &adev->smu; - bool baco_feature_is_enabled = false; + int ret; - if (!smu->pm_enabled) + if (amdgpu_sriov_vf(adev)&& !amdgpu_sriov_is_pp_one_vf(adev)) return 0; - if(!smu->is_apu) - baco_feature_is_enabled = smu_feature_is_enabled(smu, SMU_FEATURE_BACO_BIT); - - ret = smu_system_features_control(smu, false); - if (ret) - return ret; + if (!smu->pm_enabled) + return 0; - if (baco_feature_is_enabled) { - ret = smu_feature_set_enabled(smu, SMU_FEATURE_BACO_BIT, true); - if (ret) { - pr_warn("set BACO feature enabled failed, return %d\n", ret); + if(!amdgpu_sriov_vf(adev)) { + ret = smu_disable_dpm(smu); + if (ret) return ret; - } } smu->watermarks_bitmap &= ~(WATERMARKS_LOADED); @@ -1942,7 +2011,7 @@ int smu_set_mp1_state(struct smu_context *smu, return 0; } - ret = smu_send_smc_msg(smu, msg); + ret = smu_send_smc_msg(smu, msg, NULL); if (ret) pr_err("[PrepareMp1] Failed!\n"); @@ -2006,8 +2075,11 @@ int smu_set_watermarks_for_clock_ranges(struct smu_context *smu, smu_feature_is_enabled(smu, SMU_FEATURE_DPM_DCEFCLK_BIT) && smu_feature_is_enabled(smu, SMU_FEATURE_DPM_SOCCLK_BIT)) { smu_set_watermarks_table(smu, table, clock_ranges); - smu->watermarks_bitmap |= WATERMARKS_EXIST; - smu->watermarks_bitmap &= ~WATERMARKS_LOADED; + + if (!(smu->watermarks_bitmap & WATERMARKS_EXIST)) { + smu->watermarks_bitmap |= WATERMARKS_EXIST; + smu->watermarks_bitmap &= ~WATERMARKS_LOADED; + } } mutex_unlock(&smu->mutex); @@ -2015,6 +2087,29 @@ int smu_set_watermarks_for_clock_ranges(struct smu_context *smu, return 0; } +int smu_set_ac_dc(struct smu_context *smu) +{ + int ret = 0; + + /* controlled by firmware */ + if (smu->dc_controlled_by_gpio) + return 0; + + mutex_lock(&smu->mutex); + if (smu->ppt_funcs->set_power_source) { + if (smu->adev->pm.ac_power) + ret = smu_set_power_source(smu, SMU_POWER_SOURCE_AC); + else + ret = smu_set_power_source(smu, SMU_POWER_SOURCE_DC); + if (ret) + pr_err("Failed to switch to %s mode!\n", + smu->adev->pm.ac_power ? "AC" : "DC"); + } + mutex_unlock(&smu->mutex); + + return ret; +} + const struct amd_ip_funcs smu_ip_funcs = { .name = "smu", .early_init = smu_early_init, @@ -2617,12 +2712,3 @@ uint32_t smu_get_pptable_power_limit(struct smu_context *smu) return ret; } - -int smu_send_smc_msg(struct smu_context *smu, - enum smu_message_type msg) -{ - int ret; - - ret = smu_send_smc_msg_with_param(smu, msg, 0); - return ret; -} diff --git a/drivers/gpu/drm/amd/powerplay/arcturus_ppt.c b/drivers/gpu/drm/amd/powerplay/arcturus_ppt.c index 14ba6aa876e2..1ef0923f7190 100644 --- a/drivers/gpu/drm/amd/powerplay/arcturus_ppt.c +++ b/drivers/gpu/drm/amd/powerplay/arcturus_ppt.c @@ -21,7 +21,6 @@ * */ -#include "pp_debug.h" #include <linux/firmware.h> #include "amdgpu.h" #include "amdgpu_smu.h" @@ -36,13 +35,14 @@ #include "arcturus_ppt.h" #include "smu_v11_0_pptable.h" #include "arcturus_ppsmc.h" +#include "nbio/nbio_7_4_offset.h" #include "nbio/nbio_7_4_sh_mask.h" #include "amdgpu_xgmi.h" #include <linux/i2c.h> #include <linux/pci.h> #include "amdgpu_ras.h" -#define to_amdgpu_device(x) (container_of(x, struct amdgpu_ras, eeprom_control.eeprom_accessor))->adev +#define to_amdgpu_device(x) (container_of(x, struct amdgpu_device, pm.smu_i2c)) #define CTF_OFFSET_EDGE 5 #define CTF_OFFSET_HOTSPOT 5 @@ -127,6 +127,7 @@ static struct smu_11_0_cmn2aisc_mapping arcturus_message_map[SMU_MSG_MAX_COUNT] MSG_MAP(WaflTest, PPSMC_MSG_WaflTest), MSG_MAP(SetXgmiMode, PPSMC_MSG_SetXgmiMode), MSG_MAP(SetMemoryChannelEnable, PPSMC_MSG_SetMemoryChannelEnable), + MSG_MAP(DFCstateControl, PPSMC_MSG_DFCstateControl), }; static struct smu_11_0_cmn2aisc_mapping arcturus_clk_map[SMU_CLK_COUNT] = { @@ -373,13 +374,13 @@ arcturus_set_single_dpm_table(struct smu_context *smu, ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetDpmFreqByIndex, - (clk_id << 16 | 0xFF)); + (clk_id << 16 | 0xFF), + &num_of_levels); if (ret) { pr_err("[%s] failed to get dpm levels!\n", __func__); return ret; } - smu_read_smc_arg(smu, &num_of_levels); if (!num_of_levels) { pr_err("[%s] number of clk levels is invalid!\n", __func__); return -EINVAL; @@ -389,12 +390,12 @@ arcturus_set_single_dpm_table(struct smu_context *smu, for (i = 0; i < num_of_levels; i++) { ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetDpmFreqByIndex, - (clk_id << 16 | i)); + (clk_id << 16 | i), + &clk); if (ret) { pr_err("[%s] failed to get dpm freq by index!\n", __func__); return ret; } - smu_read_smc_arg(smu, &clk); if (!clk) { pr_err("[%s] clk value is invalid!\n", __func__); return -EINVAL; @@ -552,13 +553,13 @@ static int arcturus_run_btc(struct smu_context *smu) { int ret = 0; - ret = smu_send_smc_msg(smu, SMU_MSG_RunAfllBtc); + ret = smu_send_smc_msg(smu, SMU_MSG_RunAfllBtc, NULL); if (ret) { pr_err("RunAfllBtc failed!\n"); return ret; } - return smu_send_smc_msg(smu, SMU_MSG_RunDcBtc); + return smu_send_smc_msg(smu, SMU_MSG_RunDcBtc, NULL); } static int arcturus_populate_umd_state_clk(struct smu_context *smu) @@ -743,7 +744,8 @@ static int arcturus_upload_dpm_level(struct smu_context *smu, bool max, single_dpm_table->dpm_state.soft_min_level; ret = smu_send_smc_msg_with_param(smu, (max ? SMU_MSG_SetSoftMaxByFreq : SMU_MSG_SetSoftMinByFreq), - (PPCLK_GFXCLK << 16) | (freq & 0xffff)); + (PPCLK_GFXCLK << 16) | (freq & 0xffff), + NULL); if (ret) { pr_err("Failed to set soft %s gfxclk !\n", max ? "max" : "min"); @@ -758,7 +760,8 @@ static int arcturus_upload_dpm_level(struct smu_context *smu, bool max, single_dpm_table->dpm_state.soft_min_level; ret = smu_send_smc_msg_with_param(smu, (max ? SMU_MSG_SetSoftMaxByFreq : SMU_MSG_SetSoftMinByFreq), - (PPCLK_UCLK << 16) | (freq & 0xffff)); + (PPCLK_UCLK << 16) | (freq & 0xffff), + NULL); if (ret) { pr_err("Failed to set soft %s memclk !\n", max ? "max" : "min"); @@ -773,7 +776,8 @@ static int arcturus_upload_dpm_level(struct smu_context *smu, bool max, single_dpm_table->dpm_state.soft_min_level; ret = smu_send_smc_msg_with_param(smu, (max ? SMU_MSG_SetSoftMaxByFreq : SMU_MSG_SetSoftMinByFreq), - (PPCLK_SOCCLK << 16) | (freq & 0xffff)); + (PPCLK_SOCCLK << 16) | (freq & 0xffff), + NULL); if (ret) { pr_err("Failed to set soft %s socclk !\n", max ? "max" : "min"); @@ -790,8 +794,21 @@ static int arcturus_force_clk_levels(struct smu_context *smu, struct arcturus_dpm_table *dpm_table; struct arcturus_single_dpm_table *single_dpm_table; uint32_t soft_min_level, soft_max_level; + uint32_t smu_version; int ret = 0; + ret = smu_get_smc_version(smu, NULL, &smu_version); + if (ret) { + pr_err("Failed to get smu version!\n"); + return ret; + } + + if (smu_version >= 0x361200) { + pr_err("Forcing clock level is not supported with " + "54.18 and onwards SMU firmwares\n"); + return -EOPNOTSUPP; + } + soft_min_level = mask ? (ffs(mask) - 1) : 0; soft_max_level = mask ? (fls(mask) - 1) : 0; @@ -1288,12 +1305,11 @@ static int arcturus_get_power_limit(struct smu_context *smu, return -EINVAL; ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetPptLimit, - power_src << 16); + power_src << 16, &asic_default_power_limit); if (ret) { pr_err("[%s] get PPT limit failed!", __func__); return ret; } - smu_read_smc_arg(smu, &asic_default_power_limit); } else { /* the last hope to figure out the ppt limit */ if (!pptable) { @@ -1497,7 +1513,8 @@ static int arcturus_set_power_profile_mode(struct smu_context *smu, ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetWorkloadMask, - 1 << workload_type); + 1 << workload_type, + NULL); if (ret) { pr_err("Fail to set workload type %d\n", workload_type); return ret; @@ -1508,6 +1525,38 @@ static int arcturus_set_power_profile_mode(struct smu_context *smu, return 0; } +static int arcturus_set_performance_level(struct smu_context *smu, + enum amd_dpm_forced_level level) +{ + uint32_t smu_version; + int ret; + + ret = smu_get_smc_version(smu, NULL, &smu_version); + if (ret) { + pr_err("Failed to get smu version!\n"); + return ret; + } + + switch (level) { + case AMD_DPM_FORCED_LEVEL_HIGH: + case AMD_DPM_FORCED_LEVEL_LOW: + case AMD_DPM_FORCED_LEVEL_PROFILE_STANDARD: + case AMD_DPM_FORCED_LEVEL_PROFILE_MIN_SCLK: + case AMD_DPM_FORCED_LEVEL_PROFILE_MIN_MCLK: + case AMD_DPM_FORCED_LEVEL_PROFILE_PEAK: + if (smu_version >= 0x361200) { + pr_err("Forcing clock level is not supported with " + "54.18 and onwards SMU firmwares\n"); + return -EOPNOTSUPP; + } + break; + default: + break; + } + + return smu_v11_0_set_performance_level(smu, level); +} + static void arcturus_dump_pptable(struct smu_context *smu) { struct smu_table_context *table_context = &smu->smu_table; @@ -2187,7 +2236,7 @@ static int arcturus_i2c_eeprom_control_init(struct i2c_adapter *control) control->class = I2C_CLASS_SPD; control->dev.parent = &adev->pdev->dev; control->algo = &arcturus_i2c_eeprom_i2c_algo; - snprintf(control->name, sizeof(control->name), "RAS EEPROM"); + snprintf(control->name, sizeof(control->name), "AMDGPU EEPROM"); res = i2c_add_adapter(control); if (res) @@ -2207,6 +2256,18 @@ static void arcturus_i2c_eeprom_control_fini(struct i2c_adapter *control) i2c_del_adapter(control); } +static bool arcturus_is_baco_supported(struct smu_context *smu) +{ + struct amdgpu_device *adev = smu->adev; + uint32_t val; + + if (!smu_v11_0_baco_is_support(smu)) + return false; + + val = RREG32_SOC15(NBIO, 0, mmRCC_BIF_STRAP0); + return (val & RCC_BIF_STRAP0__STRAP_PX_CAPABLE_MASK) ? true : false; +} + static uint32_t arcturus_get_pptable_power_limit(struct smu_context *smu) { PPTable_t *pptable = smu->smu_table.driver_pptable; @@ -2214,6 +2275,27 @@ static uint32_t arcturus_get_pptable_power_limit(struct smu_context *smu) return pptable->SocketPowerLimitAc[PPT_THROTTLER_PPT0]; } +static int arcturus_set_df_cstate(struct smu_context *smu, + enum pp_df_cstate state) +{ + uint32_t smu_version; + int ret; + + ret = smu_get_smc_version(smu, NULL, &smu_version); + if (ret) { + pr_err("Failed to get smu version!\n"); + return ret; + } + + /* PPSMC_MSG_DFCstateControl is supported by 54.15.0 and onwards */ + if (smu_version < 0x360F00) { + pr_err("DFCstateControl is only supported by PMFW 54.15.0 and onwards\n"); + return -EINVAL; + } + + return smu_send_smc_msg_with_param(smu, SMU_MSG_DFCstateControl, state, NULL); +} + static const struct pptable_funcs arcturus_ppt_funcs = { /* translate smu index into arcturus specific index */ .get_smu_msg_index = arcturus_get_smu_msg_index, @@ -2248,7 +2330,7 @@ static const struct pptable_funcs arcturus_ppt_funcs = { .get_profiling_clk_mask = arcturus_get_profiling_clk_mask, .get_power_profile_mode = arcturus_get_power_profile_mode, .set_power_profile_mode = arcturus_set_power_profile_mode, - .set_performance_level = smu_v11_0_set_performance_level, + .set_performance_level = arcturus_set_performance_level, /* debug (internal used) */ .dump_pptable = arcturus_dump_pptable, .get_power_limit = arcturus_get_power_limit, @@ -2277,7 +2359,6 @@ static const struct pptable_funcs arcturus_ppt_funcs = { .notify_memory_pool_location = smu_v11_0_notify_memory_pool_location, .system_features_control = smu_v11_0_system_features_control, .send_smc_msg_with_param = smu_v11_0_send_msg_with_param, - .read_smc_arg = smu_v11_0_read_arg, .init_display_count = smu_v11_0_init_display_count, .set_allowed_mask = smu_v11_0_set_allowed_mask, .get_enabled_mask = smu_v11_0_get_enabled_mask, @@ -2298,7 +2379,7 @@ static const struct pptable_funcs arcturus_ppt_funcs = { .register_irq_handler = smu_v11_0_register_irq_handler, .set_azalia_d3_pme = smu_v11_0_set_azalia_d3_pme, .get_max_sustainable_clocks_by_dc = smu_v11_0_get_max_sustainable_clocks_by_dc, - .baco_is_support= smu_v11_0_baco_is_support, + .baco_is_support= arcturus_is_baco_supported, .baco_get_state = smu_v11_0_baco_get_state, .baco_set_state = smu_v11_0_baco_set_state, .baco_enter = smu_v11_0_baco_enter, @@ -2307,6 +2388,7 @@ static const struct pptable_funcs arcturus_ppt_funcs = { .set_soft_freq_limited_range = smu_v11_0_set_soft_freq_limited_range, .override_pcie_parameters = smu_v11_0_override_pcie_parameters, .get_pptable_power_limit = arcturus_get_pptable_power_limit, + .set_df_cstate = arcturus_set_df_cstate, }; void arcturus_set_ppt_funcs(struct smu_context *smu) diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c index bf04cfefb283..7740488999df 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c @@ -1250,7 +1250,7 @@ static void smu7_set_dpm_event_sources(struct pp_hwmgr *hwmgr, uint32_t sources) switch (sources) { default: pr_err("Unknown throttling event sources."); - /* fall through */ + fallthrough; case 0: protection = false; /* src is unused */ @@ -3698,12 +3698,12 @@ static int smu7_request_link_speed_change_before_state_change( data->force_pcie_gen = PP_PCIEGen2; if (current_link_speed == PP_PCIEGen2) break; - /* fall through */ + fallthrough; case PP_PCIEGen2: if (0 == amdgpu_acpi_pcie_performance_request(hwmgr->adev, PCIE_PERF_REQ_GEN2, false)) break; + fallthrough; #endif - /* fall through */ default: data->force_pcie_gen = smu7_get_current_pcie_speed(hwmgr); break; diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c index 92a65e3daff4..f29f95be1e56 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c @@ -3382,7 +3382,7 @@ static int vega10_populate_and_upload_sclk_mclk_dpm_levels( } if (data->need_update_dpm_table & - (DPMTABLE_OD_UPDATE_SCLK + DPMTABLE_UPDATE_SCLK + DPMTABLE_UPDATE_SOCCLK)) { + (DPMTABLE_OD_UPDATE_SCLK | DPMTABLE_UPDATE_SCLK | DPMTABLE_UPDATE_SOCCLK)) { result = vega10_populate_all_graphic_levels(hwmgr); PP_ASSERT_WITH_CODE((0 == result), "Failed to populate SCLK during PopulateNewDPMClocksStates Function!", @@ -3390,7 +3390,7 @@ static int vega10_populate_and_upload_sclk_mclk_dpm_levels( } if (data->need_update_dpm_table & - (DPMTABLE_OD_UPDATE_MCLK + DPMTABLE_UPDATE_MCLK)) { + (DPMTABLE_OD_UPDATE_MCLK | DPMTABLE_UPDATE_MCLK)) { result = vega10_populate_all_memory_levels(hwmgr); PP_ASSERT_WITH_CODE((0 == result), "Failed to populate MCLK during PopulateNewDPMClocksStates Function!", diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c index 3b3ec5666051..08b6ba39a6d7 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c @@ -487,15 +487,16 @@ static int vega20_setup_asic_task(struct pp_hwmgr *hwmgr) { struct amdgpu_device *adev = (struct amdgpu_device *)(hwmgr->adev); int ret = 0; + bool use_baco = (adev->in_gpu_reset && + (amdgpu_asic_reset_method(adev) == AMD_RESET_METHOD_BACO)) || + (adev->in_runpm && amdgpu_asic_supports_baco(adev)); ret = vega20_init_sclk_threshold(hwmgr); PP_ASSERT_WITH_CODE(!ret, "Failed to init sclk threshold!", return ret); - if (adev->in_gpu_reset && - (amdgpu_asic_reset_method(adev) == AMD_RESET_METHOD_BACO)) { - + if (use_baco) { ret = vega20_baco_apply_vdci_flush_workaround(hwmgr); if (ret) pr_err("Failed to apply vega20 baco workaround!\n"); diff --git a/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h b/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h index 97b6714e83e6..ae2c318dd6fa 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h +++ b/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h @@ -362,6 +362,7 @@ struct smu_context struct mutex mutex; struct mutex sensor_lock; struct mutex metrics_lock; + struct mutex message_lock; uint64_t pool_size; struct smu_table_context smu_table; @@ -371,6 +372,9 @@ struct smu_context struct amd_pp_display_configuration *display_config; struct smu_baco_context smu_baco; void *od_settings; +#if defined(CONFIG_DEBUG_FS) + struct dentry *debugfs_sclk; +#endif uint32_t pstate_sclk; uint32_t pstate_mclk; @@ -404,6 +408,7 @@ struct smu_context uint32_t smc_if_version; bool uploading_custom_pp_table; + bool dc_controlled_by_gpio; }; struct i2c_adapter; @@ -514,8 +519,7 @@ struct pptable_funcs { int (*set_last_dcef_min_deep_sleep_clk)(struct smu_context *smu); int (*system_features_control)(struct smu_context *smu, bool en); int (*send_smc_msg_with_param)(struct smu_context *smu, - enum smu_message_type msg, uint32_t param); - int (*read_smc_arg)(struct smu_context *smu, uint32_t *arg); + enum smu_message_type msg, uint32_t param, uint32_t *read_arg); int (*init_display_count)(struct smu_context *smu, uint32_t count); int (*set_allowed_mask)(struct smu_context *smu); int (*get_enabled_mask)(struct smu_context *smu, uint32_t *feature_mask, uint32_t num); @@ -567,6 +571,7 @@ struct pptable_funcs { int (*override_pcie_parameters)(struct smu_context *smu); uint32_t (*get_pptable_power_limit)(struct smu_context *smu); int (*disable_umc_cdr_12gbps_workaround)(struct smu_context *smu); + int (*set_power_source)(struct smu_context *smu, enum smu_power_src_type power_src); }; int smu_load_microcode(struct smu_context *smu); @@ -707,7 +712,7 @@ int smu_get_dpm_level_count(struct smu_context *smu, enum smu_clk_type clk_type, int smu_get_dpm_freq_range(struct smu_context *smu, enum smu_clk_type clk_type, uint32_t *min, uint32_t *max, bool lock_needed); int smu_set_soft_freq_range(struct smu_context *smu, enum smu_clk_type clk_type, - uint32_t min, uint32_t max); + uint32_t min, uint32_t max, bool lock_needed); int smu_set_hard_freq_range(struct smu_context *smu, enum smu_clk_type clk_type, uint32_t min, uint32_t max); int smu_get_dpm_level_range(struct smu_context *smu, enum smu_clk_type clk_type, @@ -715,6 +720,7 @@ int smu_get_dpm_level_range(struct smu_context *smu, enum smu_clk_type clk_type, enum amd_dpm_forced_level smu_get_performance_level(struct smu_context *smu); int smu_force_performance_level(struct smu_context *smu, enum amd_dpm_forced_level level); int smu_set_display_count(struct smu_context *smu, uint32_t count); +int smu_set_ac_dc(struct smu_context *smu); bool smu_clk_dpm_is_enabled(struct smu_context *smu, enum smu_clk_type clk_type); const char *smu_get_message_name(struct smu_context *smu, enum smu_message_type type); const char *smu_get_feature_name(struct smu_context *smu, enum smu_feature_mask feature); diff --git a/drivers/gpu/drm/amd/powerplay/inc/arcturus_ppsmc.h b/drivers/gpu/drm/amd/powerplay/inc/arcturus_ppsmc.h index e3291259b249..f736d773f9d6 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/arcturus_ppsmc.h +++ b/drivers/gpu/drm/amd/powerplay/inc/arcturus_ppsmc.h @@ -110,7 +110,11 @@ //Others #define PPSMC_MSG_SetMemoryChannelEnable 0x39 -#define PPSMC_Message_Count 0x3A +//OOB +#define PPSMC_MSG_SetNumBadHbmPagesRetired 0x3A + +#define PPSMC_MSG_DFCstateControl 0x3B +#define PPSMC_Message_Count 0x3C typedef uint32_t PPSMC_Result; typedef uint32_t PPSMC_Msg; diff --git a/drivers/gpu/drm/amd/powerplay/inc/pp_debug.h b/drivers/gpu/drm/amd/powerplay/inc/pp_debug.h index 822cd8b5bf90..cea65093b6ad 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/pp_debug.h +++ b/drivers/gpu/drm/amd/powerplay/inc/pp_debug.h @@ -37,7 +37,7 @@ #define PP_ASSERT_WITH_CODE(cond, msg, code) \ do { \ if (!(cond)) { \ - pr_warn("%s\n", msg); \ + pr_warn_ratelimited("%s\n", msg); \ code; \ } \ } while (0) @@ -45,7 +45,7 @@ #define PP_ASSERT(cond, msg) \ do { \ if (!(cond)) { \ - pr_warn("%s\n", msg); \ + pr_warn_ratelimited("%s\n", msg); \ } \ } while (0) diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu11_driver_if_navi10.h b/drivers/gpu/drm/amd/powerplay/inc/smu11_driver_if_navi10.h index ac0120e384be..4b2da98afcd2 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/smu11_driver_if_navi10.h +++ b/drivers/gpu/drm/amd/powerplay/inc/smu11_driver_if_navi10.h @@ -701,7 +701,8 @@ typedef struct { // APCC Settings uint16_t PccThresholdLow; uint16_t PccThresholdHigh; - uint32_t PaddingAPCC[6]; //FIXME pending SPEC + uint32_t MGpuFanBoostLimitRpm; + uint32_t PaddingAPCC[5]; // Temperature Dependent Vmin uint16_t VDDGFX_TVmin; //Celcius diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0.h b/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0.h index d5314d12628a..674e426ed59b 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0.h +++ b/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0.h @@ -28,8 +28,9 @@ #define SMU11_DRIVER_IF_VERSION_INV 0xFFFFFFFF #define SMU11_DRIVER_IF_VERSION_VG20 0x13 #define SMU11_DRIVER_IF_VERSION_ARCT 0x12 -#define SMU11_DRIVER_IF_VERSION_NV10 0x33 -#define SMU11_DRIVER_IF_VERSION_NV14 0x34 +#define SMU11_DRIVER_IF_VERSION_NV10 0x35 +#define SMU11_DRIVER_IF_VERSION_NV12 0x33 +#define SMU11_DRIVER_IF_VERSION_NV14 0x36 /* MP Apertures */ #define MP0_Public 0x03800000 @@ -182,9 +183,8 @@ int smu_v11_0_system_features_control(struct smu_context *smu, int smu_v11_0_send_msg_with_param(struct smu_context *smu, enum smu_message_type msg, - uint32_t param); - -int smu_v11_0_read_arg(struct smu_context *smu, uint32_t *arg); + uint32_t param, + uint32_t *read_arg); int smu_v11_0_init_display_count(struct smu_context *smu, uint32_t count); @@ -267,4 +267,7 @@ uint32_t smu_v11_0_get_max_power_limit(struct smu_context *smu); int smu_v11_0_set_performance_level(struct smu_context *smu, enum amd_dpm_forced_level level); +int smu_v11_0_set_power_source(struct smu_context *smu, + enum smu_power_src_type power_src); + #endif diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu_v12_0.h b/drivers/gpu/drm/amd/powerplay/inc/smu_v12_0.h index d79e54b5ebf6..7fbebc1979cf 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/smu_v12_0.h +++ b/drivers/gpu/drm/amd/powerplay/inc/smu_v12_0.h @@ -40,14 +40,13 @@ struct smu_12_0_cmn2aisc_mapping { int smu_v12_0_send_msg_without_waiting(struct smu_context *smu, uint16_t msg); -int smu_v12_0_read_arg(struct smu_context *smu, uint32_t *arg); - int smu_v12_0_wait_for_response(struct smu_context *smu); int smu_v12_0_send_msg_with_param(struct smu_context *smu, enum smu_message_type msg, - uint32_t param); + uint32_t param, + uint32_t *read_arg); int smu_v12_0_check_fw_status(struct smu_context *smu); diff --git a/drivers/gpu/drm/amd/powerplay/navi10_ppt.c b/drivers/gpu/drm/amd/powerplay/navi10_ppt.c index 0d73a49166af..15030284b444 100644 --- a/drivers/gpu/drm/amd/powerplay/navi10_ppt.c +++ b/drivers/gpu/drm/amd/powerplay/navi10_ppt.c @@ -21,7 +21,6 @@ * */ -#include "pp_debug.h" #include <linux/firmware.h> #include <linux/pci.h> #include "amdgpu.h" @@ -29,14 +28,15 @@ #include "smu_internal.h" #include "atomfirmware.h" #include "amdgpu_atomfirmware.h" +#include "soc15_common.h" #include "smu_v11_0.h" #include "smu11_driver_if_navi10.h" -#include "soc15_common.h" #include "atom.h" #include "navi10_ppt.h" #include "smu_v11_0_pptable.h" #include "smu_v11_0_ppsmc.h" -#include "nbio/nbio_7_4_sh_mask.h" +#include "nbio/nbio_2_3_offset.h" +#include "nbio/nbio_2_3_sh_mask.h" #include "asic_reg/mp/mp_11_0_sh_mask.h" @@ -349,7 +349,6 @@ navi10_get_allowed_feature_mask(struct smu_context *smu, | FEATURE_MASK(FEATURE_DS_DCEFCLK_BIT) | FEATURE_MASK(FEATURE_FW_DSTATE_BIT) | FEATURE_MASK(FEATURE_BACO_BIT) - | FEATURE_MASK(FEATURE_ACDC_BIT) | FEATURE_MASK(FEATURE_GFX_SS_BIT) | FEATURE_MASK(FEATURE_APCC_DFLL_BIT) | FEATURE_MASK(FEATURE_FW_CTF_BIT) @@ -393,6 +392,9 @@ navi10_get_allowed_feature_mask(struct smu_context *smu, if (smu->adev->pg_flags & AMD_PG_SUPPORT_JPEG) *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_JPEG_PG_BIT); + if (smu->dc_controlled_by_gpio) + *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_ACDC_BIT); + /* disable DPM UCLK and DS SOCCLK on navi10 A0 secure board */ if (is_asic_secure(smu)) { /* only for navi10 A0 */ @@ -527,6 +529,9 @@ static int navi10_store_powerplay_table(struct smu_context *smu) table_context->thermal_controller_type = powerplay_table->thermal_controller_type; + if (powerplay_table->platform_caps & SMU_11_0_PP_PLATFORM_CAP_HARDWAREDC) + smu->dc_controlled_by_gpio = true; + mutex_lock(&smu_baco->mutex); if (powerplay_table->platform_caps & SMU_11_0_PP_PLATFORM_CAP_BACO || powerplay_table->platform_caps & SMU_11_0_PP_PLATFORM_CAP_MACO) @@ -661,14 +666,14 @@ static int navi10_dpm_set_uvd_enable(struct smu_context *smu, bool enable) if (enable) { /* vcn dpm on is a prerequisite for vcn power gate messages */ if (smu_feature_is_enabled(smu, SMU_FEATURE_VCN_PG_BIT)) { - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_PowerUpVcn, 1); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_PowerUpVcn, 1, NULL); if (ret) return ret; } power_gate->vcn_gated = false; } else { if (smu_feature_is_enabled(smu, SMU_FEATURE_VCN_PG_BIT)) { - ret = smu_send_smc_msg(smu, SMU_MSG_PowerDownVcn); + ret = smu_send_smc_msg(smu, SMU_MSG_PowerDownVcn, NULL); if (ret) return ret; } @@ -686,14 +691,14 @@ static int navi10_dpm_set_jpeg_enable(struct smu_context *smu, bool enable) if (enable) { if (smu_feature_is_enabled(smu, SMU_FEATURE_JPEG_PG_BIT)) { - ret = smu_send_smc_msg(smu, SMU_MSG_PowerUpJpeg); + ret = smu_send_smc_msg(smu, SMU_MSG_PowerUpJpeg, NULL); if (ret) return ret; } power_gate->jpeg_gated = false; } else { if (smu_feature_is_enabled(smu, SMU_FEATURE_JPEG_PG_BIT)) { - ret = smu_send_smc_msg(smu, SMU_MSG_PowerDownJpeg); + ret = smu_send_smc_msg(smu, SMU_MSG_PowerDownJpeg, NULL); if (ret) return ret; } @@ -970,7 +975,7 @@ static int navi10_force_clk_levels(struct smu_context *smu, if (ret) return size; - ret = smu_set_soft_freq_range(smu, clk_type, min_freq, max_freq); + ret = smu_set_soft_freq_range(smu, clk_type, min_freq, max_freq, false); if (ret) return size; break; @@ -1042,7 +1047,7 @@ static int navi10_pre_display_config_changed(struct smu_context *smu) int ret = 0; uint32_t max_freq = 0; - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_NumOfDisplays, 0); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_NumOfDisplays, 0, NULL); if (ret) return ret; @@ -1063,19 +1068,11 @@ static int navi10_display_config_changed(struct smu_context *smu) int ret = 0; if ((smu->watermarks_bitmap & WATERMARKS_EXIST) && - !(smu->watermarks_bitmap & WATERMARKS_LOADED)) { - ret = smu_write_watermarks_table(smu); - if (ret) - return ret; - - smu->watermarks_bitmap |= WATERMARKS_LOADED; - } - - if ((smu->watermarks_bitmap & WATERMARKS_EXIST) && smu_feature_is_supported(smu, SMU_FEATURE_DPM_DCEFCLK_BIT) && smu_feature_is_supported(smu, SMU_FEATURE_DPM_SOCCLK_BIT)) { ret = smu_send_smc_msg_with_param(smu, SMU_MSG_NumOfDisplays, - smu->display_config->num_display); + smu->display_config->num_display, + NULL); if (ret) return ret; } @@ -1102,7 +1099,7 @@ static int navi10_force_dpm_limit_value(struct smu_context *smu, bool highest) return ret; force_freq = highest ? max_freq : min_freq; - ret = smu_set_soft_freq_range(smu, clk_type, force_freq, force_freq); + ret = smu_set_soft_freq_range(smu, clk_type, force_freq, force_freq, false); if (ret) return ret; } @@ -1128,7 +1125,7 @@ static int navi10_unforce_dpm_levels(struct smu_context *smu) if (ret) return ret; - ret = smu_set_soft_freq_range(smu, clk_type, min_freq, max_freq); + ret = smu_set_soft_freq_range(smu, clk_type, min_freq, max_freq, false); if (ret) return ret; } @@ -1400,7 +1397,7 @@ static int navi10_set_power_profile_mode(struct smu_context *smu, long *input, u if (workload_type < 0) return -EINVAL; smu_send_smc_msg_with_param(smu, SMU_MSG_SetWorkloadMask, - 1 << workload_type); + 1 << workload_type, NULL); return ret; } @@ -1465,7 +1462,8 @@ static int navi10_notify_smc_display_config(struct smu_context *smu) if (smu_feature_is_supported(smu, SMU_FEATURE_DS_DCEFCLK_BIT)) { ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetMinDeepSleepDcefclk, - min_clocks.dcef_clock_in_sr/100); + min_clocks.dcef_clock_in_sr/100, + NULL); if (ret) { pr_err("Attempt to set divider for DCEFCLK Failed!"); return ret; @@ -1493,6 +1491,7 @@ static int navi10_set_watermarks_table(struct smu_context *smu, *clock_ranges) { int i; + int ret = 0; Watermarks_t *table = watermarks; if (!table || !clock_ranges) @@ -1544,6 +1543,18 @@ static int navi10_set_watermarks_table(struct smu_context *smu, clock_ranges->wm_mcif_clocks_ranges[i].wm_set_id; } + smu->watermarks_bitmap |= WATERMARKS_EXIST; + + /* pass data to smu controller */ + if (!(smu->watermarks_bitmap & WATERMARKS_LOADED)) { + ret = smu_write_watermarks_table(smu); + if (ret) { + pr_err("Failed to update WMTABLE!"); + return ret; + } + smu->watermarks_bitmap |= WATERMARKS_LOADED; + } + return 0; } @@ -1674,10 +1685,10 @@ static int navi10_set_standard_performance_level(struct smu_context *smu) return navi10_set_performance_level(smu, AMD_DPM_FORCED_LEVEL_AUTO); } - ret = smu_set_soft_freq_range(smu, SMU_SCLK, sclk_freq, sclk_freq); + ret = smu_set_soft_freq_range(smu, SMU_SCLK, sclk_freq, sclk_freq, false); if (ret) return ret; - ret = smu_set_soft_freq_range(smu, SMU_UCLK, uclk_freq, uclk_freq); + ret = smu_set_soft_freq_range(smu, SMU_UCLK, uclk_freq, uclk_freq, false); if (ret) return ret; @@ -1742,10 +1753,10 @@ static int navi10_set_peak_performance_level(struct smu_context *smu) if (ret) return ret; - ret = smu_set_soft_freq_range(smu, SMU_SCLK, sclk_freq, sclk_freq); + ret = smu_set_soft_freq_range(smu, SMU_SCLK, sclk_freq, sclk_freq, false); if (ret) return ret; - ret = smu_set_soft_freq_range(smu, SMU_UCLK, uclk_freq, uclk_freq); + ret = smu_set_soft_freq_range(smu, SMU_UCLK, uclk_freq, uclk_freq, false); if (ret) return ret; @@ -1855,12 +1866,11 @@ static int navi10_get_power_limit(struct smu_context *smu, return -EINVAL; ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetPptLimit, - power_src << 16); + power_src << 16, &asic_default_power_limit); if (ret) { pr_err("[%s] get PPT limit failed!", __func__); return ret; } - smu_read_smc_arg(smu, &asic_default_power_limit); } else { /* the last hope to figure out the ppt limit */ if (!pptable) { @@ -1900,7 +1910,8 @@ static int navi10_update_pcie_parameters(struct smu_context *smu, pptable->PcieLaneCount[i] : pcie_width_cap); ret = smu_send_smc_msg_with_param(smu, SMU_MSG_OverridePcieParameters, - smu_pcie_arg); + smu_pcie_arg, + NULL); if (ret) return ret; @@ -1946,13 +1957,13 @@ static int navi10_overdrive_get_gfx_clk_base_voltage(struct smu_context *smu, ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetVoltageByDpm, - param); + param, + &value); if (ret) { pr_err("[GetBaseVoltage] failed to get GFXCLK AVFS voltage from SMU!"); return ret; } - smu_read_smc_arg(smu, &value); *voltage = (uint16_t)value; return 0; @@ -1976,6 +1987,18 @@ static int navi10_setup_od_limits(struct smu_context *smu) { return 0; } +static bool navi10_is_baco_supported(struct smu_context *smu) +{ + struct amdgpu_device *adev = smu->adev; + uint32_t val; + + if (!smu_v11_0_baco_is_support(smu)) + return false; + + val = RREG32_SOC15(NBIO, 0, mmRCC_BIF_STRAP0); + return (val & RCC_BIF_STRAP0__STRAP_PX_CAPABLE_MASK) ? true : false; +} + static int navi10_set_default_od_settings(struct smu_context *smu, bool initialize) { OverDriveTable_t *od_table, *boot_od_table; int ret = 0; @@ -2209,7 +2232,7 @@ static int navi10_run_btc(struct smu_context *smu) { int ret = 0; - ret = smu_send_smc_msg(smu, SMU_MSG_RunBtc); + ret = smu_send_smc_msg(smu, SMU_MSG_RunBtc, NULL); if (ret) pr_err("RunBtc failed!\n"); @@ -2221,9 +2244,9 @@ static int navi10_dummy_pstate_control(struct smu_context *smu, bool enable) int result = 0; if (!enable) - result = smu_send_smc_msg(smu, SMU_MSG_DAL_DISABLE_DUMMY_PSTATE_CHANGE); + result = smu_send_smc_msg(smu, SMU_MSG_DAL_DISABLE_DUMMY_PSTATE_CHANGE, NULL); else - result = smu_send_smc_msg(smu, SMU_MSG_DAL_ENABLE_DUMMY_PSTATE_CHANGE); + result = smu_send_smc_msg(smu, SMU_MSG_DAL_ENABLE_DUMMY_PSTATE_CHANGE, NULL); return result; } @@ -2332,7 +2355,6 @@ static const struct pptable_funcs navi10_ppt_funcs = { .notify_memory_pool_location = smu_v11_0_notify_memory_pool_location, .system_features_control = smu_v11_0_system_features_control, .send_smc_msg_with_param = smu_v11_0_send_msg_with_param, - .read_smc_arg = smu_v11_0_read_arg, .init_display_count = smu_v11_0_init_display_count, .set_allowed_mask = smu_v11_0_set_allowed_mask, .get_enabled_mask = smu_v11_0_get_enabled_mask, @@ -2353,7 +2375,7 @@ static const struct pptable_funcs navi10_ppt_funcs = { .register_irq_handler = smu_v11_0_register_irq_handler, .set_azalia_d3_pme = smu_v11_0_set_azalia_d3_pme, .get_max_sustainable_clocks_by_dc = smu_v11_0_get_max_sustainable_clocks_by_dc, - .baco_is_support= smu_v11_0_baco_is_support, + .baco_is_support= navi10_is_baco_supported, .baco_get_state = smu_v11_0_baco_get_state, .baco_set_state = smu_v11_0_baco_set_state, .baco_enter = smu_v11_0_baco_enter, @@ -2366,6 +2388,7 @@ static const struct pptable_funcs navi10_ppt_funcs = { .get_pptable_power_limit = navi10_get_pptable_power_limit, .run_btc = navi10_run_btc, .disable_umc_cdr_12gbps_workaround = navi10_disable_umc_cdr_12gbps_workaround, + .set_power_source = smu_v11_0_set_power_source, }; void navi10_set_ppt_funcs(struct smu_context *smu) diff --git a/drivers/gpu/drm/amd/powerplay/renoir_ppt.c b/drivers/gpu/drm/amd/powerplay/renoir_ppt.c index 861e6410363b..ff73a735b888 100644 --- a/drivers/gpu/drm/amd/powerplay/renoir_ppt.c +++ b/drivers/gpu/drm/amd/powerplay/renoir_ppt.c @@ -24,7 +24,6 @@ #include "amdgpu.h" #include "amdgpu_smu.h" #include "smu_internal.h" -#include "soc15_common.h" #include "smu_v12_0_ppsmc.h" #include "smu12_driver_if.h" #include "smu_v12_0.h" @@ -111,8 +110,8 @@ static struct smu_12_0_cmn2aisc_mapping renoir_clk_map[SMU_CLK_COUNT] = { CLK_MAP(GFXCLK, CLOCK_GFXCLK), CLK_MAP(SCLK, CLOCK_GFXCLK), CLK_MAP(SOCCLK, CLOCK_SOCCLK), - CLK_MAP(UCLK, CLOCK_UMCCLK), - CLK_MAP(MCLK, CLOCK_UMCCLK), + CLK_MAP(UCLK, CLOCK_FCLK), + CLK_MAP(MCLK, CLOCK_FCLK), }; static struct smu_12_0_cmn2aisc_mapping renoir_table_map[SMU_TABLE_COUNT] = { @@ -240,6 +239,7 @@ static int renoir_print_clk_levels(struct smu_context *smu, uint32_t cur_value = 0, value = 0, count = 0, min = 0, max = 0; DpmClocks_t *clk_table = smu->smu_table.clocks_table; SmuMetrics_t metrics; + bool cur_value_match_level = false; if (!clk_table || clk_type >= SMU_CLK_COUNT) return -EINVAL; @@ -280,7 +280,7 @@ static int renoir_print_clk_levels(struct smu_context *smu, break; case SMU_MCLK: count = NUM_MEMCLK_DPM_LEVELS; - cur_value = metrics.ClockFrequency[CLOCK_UMCCLK]; + cur_value = metrics.ClockFrequency[CLOCK_FCLK]; break; case SMU_DCEFCLK: count = NUM_DCFCLK_DPM_LEVELS; @@ -298,8 +298,13 @@ static int renoir_print_clk_levels(struct smu_context *smu, GET_DPM_CUR_FREQ(clk_table, clk_type, i, value); size += sprintf(buf + size, "%d: %uMhz %s\n", i, value, cur_value == value ? "*" : ""); + if (cur_value == value) + cur_value_match_level = true; } + if (!cur_value_match_level) + size += sprintf(buf + size, " %uMhz *\n", cur_value); + return size; } @@ -342,14 +347,14 @@ static int renoir_dpm_set_uvd_enable(struct smu_context *smu, bool enable) if (enable) { /* vcn dpm on is a prerequisite for vcn power gate messages */ if (smu_feature_is_enabled(smu, SMU_FEATURE_VCN_PG_BIT)) { - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_PowerUpVcn, 0); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_PowerUpVcn, 0, NULL); if (ret) return ret; } power_gate->vcn_gated = false; } else { if (smu_feature_is_enabled(smu, SMU_FEATURE_VCN_PG_BIT)) { - ret = smu_send_smc_msg(smu, SMU_MSG_PowerDownVcn); + ret = smu_send_smc_msg(smu, SMU_MSG_PowerDownVcn, NULL); if (ret) return ret; } @@ -367,14 +372,14 @@ static int renoir_dpm_set_jpeg_enable(struct smu_context *smu, bool enable) if (enable) { if (smu_feature_is_enabled(smu, SMU_FEATURE_JPEG_PG_BIT)) { - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_PowerUpJpeg, 0); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_PowerUpJpeg, 0, NULL); if (ret) return ret; } power_gate->jpeg_gated = false; } else { if (smu_feature_is_enabled(smu, SMU_FEATURE_JPEG_PG_BIT)) { - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_PowerDownJpeg, 0); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_PowerDownJpeg, 0, NULL); if (ret) return ret; } @@ -423,7 +428,7 @@ static int renoir_force_dpm_limit_value(struct smu_context *smu, bool highest) return ret; force_freq = highest ? max_freq : min_freq; - ret = smu_set_soft_freq_range(smu, clk_type, force_freq, force_freq); + ret = smu_set_soft_freq_range(smu, clk_type, force_freq, force_freq, false); if (ret) return ret; } @@ -456,7 +461,7 @@ static int renoir_unforce_dpm_levels(struct smu_context *smu) { if (ret) return ret; - ret = smu_set_soft_freq_range(smu, clk_type, min_freq, max_freq); + ret = smu_set_soft_freq_range(smu, clk_type, min_freq, max_freq, false); if (ret) return ret; } @@ -622,22 +627,24 @@ static int renoir_force_clk_levels(struct smu_context *smu, return ret; ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxGfxClk, soft_max_level == 0 ? min_freq : - soft_max_level == 1 ? RENOIR_UMD_PSTATE_GFXCLK : max_freq); + soft_max_level == 1 ? RENOIR_UMD_PSTATE_GFXCLK : max_freq, + NULL); if (ret) return ret; ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinGfxClk, soft_min_level == 2 ? max_freq : - soft_min_level == 1 ? RENOIR_UMD_PSTATE_GFXCLK : min_freq); + soft_min_level == 1 ? RENOIR_UMD_PSTATE_GFXCLK : min_freq, + NULL); if (ret) return ret; break; case SMU_SOCCLK: GET_DPM_CUR_FREQ(clk_table, clk_type, soft_min_level, min_freq); GET_DPM_CUR_FREQ(clk_table, clk_type, soft_max_level, max_freq); - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxSocclkByFreq, max_freq); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxSocclkByFreq, max_freq, NULL); if (ret) return ret; - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinSocclkByFreq, min_freq); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinSocclkByFreq, min_freq, NULL); if (ret) return ret; break; @@ -645,10 +652,10 @@ static int renoir_force_clk_levels(struct smu_context *smu, case SMU_FCLK: GET_DPM_CUR_FREQ(clk_table, clk_type, soft_min_level, min_freq); GET_DPM_CUR_FREQ(clk_table, clk_type, soft_max_level, max_freq); - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxFclkByFreq, max_freq); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxFclkByFreq, max_freq, NULL); if (ret) return ret; - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinFclkByFreq, min_freq); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinFclkByFreq, min_freq, NULL); if (ret) return ret; break; @@ -672,14 +679,19 @@ static int renoir_set_power_profile_mode(struct smu_context *smu, long *input, u /* conv PP_SMC_POWER_PROFILE* to WORKLOAD_PPLIB_*_BIT */ workload_type = smu_workload_get_type(smu, smu->power_profile_mode); if (workload_type < 0) { - pr_err("Unsupported power profile mode %d on RENOIR\n",smu->power_profile_mode); + /* + * TODO: If some case need switch to powersave/default power mode + * then can consider enter WORKLOAD_COMPUTE/WORKLOAD_CUSTOM for power saving. + */ + pr_err_once("Unsupported power profile mode %d on RENOIR\n",smu->power_profile_mode); return -EINVAL; } ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetWorkloadMask, - 1 << workload_type); + 1 << workload_type, + NULL); if (ret) { - pr_err("Fail to set workload type %d\n", workload_type); + pr_err_once("Fail to set workload type %d\n", workload_type); return ret; } @@ -697,7 +709,7 @@ static int renoir_set_peak_clock_by_device(struct smu_context *smu) if (ret) return ret; - ret = smu_set_soft_freq_range(smu, SMU_SCLK, sclk_freq, sclk_freq); + ret = smu_set_soft_freq_range(smu, SMU_SCLK, sclk_freq, sclk_freq, false); if (ret) return ret; @@ -705,7 +717,7 @@ static int renoir_set_peak_clock_by_device(struct smu_context *smu) if (ret) return ret; - ret = smu_set_soft_freq_range(smu, SMU_UCLK, uclk_freq, uclk_freq); + ret = smu_set_soft_freq_range(smu, SMU_UCLK, uclk_freq, uclk_freq, false); if (ret) return ret; @@ -806,9 +818,10 @@ static int renoir_set_watermarks_table( clock_ranges->wm_mcif_clocks_ranges[i].wm_set_id; } + smu->watermarks_bitmap |= WATERMARKS_EXIST; + /* pass data to smu controller */ - if ((smu->watermarks_bitmap & WATERMARKS_EXIST) && - !(smu->watermarks_bitmap & WATERMARKS_LOADED)) { + if (!(smu->watermarks_bitmap & WATERMARKS_LOADED)) { ret = smu_write_watermarks_table(smu); if (ret) { pr_err("Failed to update WMTABLE!"); @@ -880,6 +893,17 @@ static int renoir_read_sensor(struct smu_context *smu, return ret; } +static bool renoir_is_dpm_running(struct smu_context *smu) +{ + /* + * Util now, the pmfw hasn't exported the interface of SMU + * feature mask to APU SKU so just force on all the feature + * at early initial stage. + */ + return true; + +} + static const struct pptable_funcs renoir_ppt_funcs = { .get_smu_msg_index = renoir_get_smu_msg_index, .get_smu_clk_index = renoir_get_smu_clk_index, @@ -909,7 +933,6 @@ static const struct pptable_funcs renoir_ppt_funcs = { .powergate_vcn = smu_v12_0_powergate_vcn, .powergate_jpeg = smu_v12_0_powergate_jpeg, .send_smc_msg_with_param = smu_v12_0_send_msg_with_param, - .read_smc_arg = smu_v12_0_read_arg, .set_gfx_cgpg = smu_v12_0_set_gfx_cgpg, .gfx_off_control = smu_v12_0_gfx_off_control, .init_smc_tables = smu_v12_0_init_smc_tables, @@ -921,6 +944,7 @@ static const struct pptable_funcs renoir_ppt_funcs = { .mode2_reset = smu_v12_0_mode2_reset, .set_soft_freq_limited_range = smu_v12_0_set_soft_freq_limited_range, .set_driver_table_location = smu_v12_0_set_driver_table_location, + .is_dpm_running = renoir_is_dpm_running, }; void renoir_set_ppt_funcs(struct smu_context *smu) diff --git a/drivers/gpu/drm/amd/powerplay/renoir_ppt.h b/drivers/gpu/drm/amd/powerplay/renoir_ppt.h index 2a390ddd37dd..89cd6da118a3 100644 --- a/drivers/gpu/drm/amd/powerplay/renoir_ppt.h +++ b/drivers/gpu/drm/amd/powerplay/renoir_ppt.h @@ -37,7 +37,7 @@ extern void renoir_set_ppt_funcs(struct smu_context *smu); freq = table->SocClocks[dpm_level].Freq; \ break; \ case SMU_MCLK: \ - freq = table->MemClocks[dpm_level].Freq; \ + freq = table->FClocks[dpm_level].Freq; \ break; \ case SMU_DCEFCLK: \ freq = table->DcfClocks[dpm_level].Freq; \ diff --git a/drivers/gpu/drm/amd/powerplay/smu_internal.h b/drivers/gpu/drm/amd/powerplay/smu_internal.h index 7bd200ffcda8..40c35bcc5a0a 100644 --- a/drivers/gpu/drm/amd/powerplay/smu_internal.h +++ b/drivers/gpu/drm/amd/powerplay/smu_internal.h @@ -79,12 +79,13 @@ #define smu_set_default_od_settings(smu, initialize) \ ((smu)->ppt_funcs->set_default_od_settings ? (smu)->ppt_funcs->set_default_od_settings((smu), (initialize)) : 0) -int smu_send_smc_msg(struct smu_context *smu, enum smu_message_type msg); +#define smu_send_smc_msg_with_param(smu, msg, param, read_arg) \ + ((smu)->ppt_funcs->send_smc_msg_with_param? (smu)->ppt_funcs->send_smc_msg_with_param((smu), (msg), (param), (read_arg)) : 0) + +static inline int smu_send_smc_msg(struct smu_context *smu, enum smu_message_type msg, uint32_t *read_arg) { + return smu_send_smc_msg_with_param(smu, msg, 0, read_arg); +} -#define smu_send_smc_msg_with_param(smu, msg, param) \ - ((smu)->ppt_funcs->send_smc_msg_with_param? (smu)->ppt_funcs->send_smc_msg_with_param((smu), (msg), (param)) : 0) -#define smu_read_smc_arg(smu, arg) \ - ((smu)->ppt_funcs->read_smc_arg? (smu)->ppt_funcs->read_smc_arg((smu), (arg)) : 0) #define smu_alloc_dpm_context(smu) \ ((smu)->ppt_funcs->alloc_dpm_context ? (smu)->ppt_funcs->alloc_dpm_context((smu)) : 0) #define smu_init_display_count(smu, count) \ @@ -210,4 +211,7 @@ int smu_send_smc_msg(struct smu_context *smu, enum smu_message_type msg); #define smu_disable_umc_cdr_12gbps_workaround(smu) \ ((smu)->ppt_funcs->disable_umc_cdr_12gbps_workaround ? (smu)->ppt_funcs->disable_umc_cdr_12gbps_workaround((smu)) : 0) +#define smu_set_power_source(smu, power_src) \ + ((smu)->ppt_funcs->set_power_source ? (smu)->ppt_funcs->set_power_source((smu), (power_src)) : 0) + #endif diff --git a/drivers/gpu/drm/amd/powerplay/smu_v11_0.c b/drivers/gpu/drm/amd/powerplay/smu_v11_0.c index 0dc49479a7eb..541c932a6005 100644 --- a/drivers/gpu/drm/amd/powerplay/smu_v11_0.c +++ b/drivers/gpu/drm/amd/powerplay/smu_v11_0.c @@ -26,7 +26,6 @@ #define SMU_11_0_PARTIAL_PPTABLE -#include "pp_debug.h" #include "amdgpu.h" #include "amdgpu_smu.h" #include "smu_internal.h" @@ -43,8 +42,6 @@ #include "asic_reg/thm/thm_11_0_2_sh_mask.h" #include "asic_reg/mp/mp_11_0_offset.h" #include "asic_reg/mp/mp_11_0_sh_mask.h" -#include "asic_reg/nbio/nbio_7_4_offset.h" -#include "asic_reg/nbio/nbio_7_4_sh_mask.h" #include "asic_reg/smuio/smuio_11_0_0_offset.h" #include "asic_reg/smuio/smuio_11_0_0_sh_mask.h" @@ -64,7 +61,7 @@ static int smu_v11_0_send_msg_without_waiting(struct smu_context *smu, return 0; } -int smu_v11_0_read_arg(struct smu_context *smu, uint32_t *arg) +static int smu_v11_0_read_arg(struct smu_context *smu, uint32_t *arg) { struct amdgpu_device *adev = smu->adev; @@ -92,7 +89,8 @@ static int smu_v11_0_wait_for_response(struct smu_context *smu) int smu_v11_0_send_msg_with_param(struct smu_context *smu, enum smu_message_type msg, - uint32_t param) + uint32_t param, + uint32_t *read_arg) { struct amdgpu_device *adev = smu->adev; int ret = 0, index = 0; @@ -101,11 +99,12 @@ smu_v11_0_send_msg_with_param(struct smu_context *smu, if (index < 0) return index; + mutex_lock(&smu->message_lock); ret = smu_v11_0_wait_for_response(smu); if (ret) { pr_err("Msg issuing pre-check failed and " "SMU may be not in the right state!\n"); - return ret; + goto out; } WREG32_SOC15(MP1, 0, mmMP1_SMN_C2PMSG_90, 0); @@ -115,10 +114,21 @@ smu_v11_0_send_msg_with_param(struct smu_context *smu, smu_v11_0_send_msg_without_waiting(smu, (uint16_t)index); ret = smu_v11_0_wait_for_response(smu); - if (ret) + if (ret) { pr_err("failed send message: %10s (%d) \tparam: 0x%08x response %#x\n", smu_get_message_name(smu, msg), index, param, ret); - + goto out; + } + if (read_arg) { + ret = smu_v11_0_read_arg(smu, read_arg); + if (ret) { + pr_err("failed to read message arg: %10s (%d) \tparam: 0x%08x response %#x\n", + smu_get_message_name(smu, msg), index, param, ret); + goto out; + } + } +out: + mutex_unlock(&smu->message_lock); return ret; } @@ -262,6 +272,9 @@ int smu_v11_0_check_fw_version(struct smu_context *smu) case CHIP_NAVI10: smu->smc_if_version = SMU11_DRIVER_IF_VERSION_NV10; break; + case CHIP_NAVI12: + smu->smc_if_version = SMU11_DRIVER_IF_VERSION_NV12; + break; case CHIP_NAVI14: smu->smc_if_version = SMU11_DRIVER_IF_VERSION_NV14; break; @@ -671,12 +684,14 @@ int smu_v11_0_notify_memory_pool_location(struct smu_context *smu) ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSystemVirtualDramAddrHigh, - address_high); + address_high, + NULL); if (ret) return ret; ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSystemVirtualDramAddrLow, - address_low); + address_low, + NULL); if (ret) return ret; @@ -685,15 +700,15 @@ int smu_v11_0_notify_memory_pool_location(struct smu_context *smu) address_low = (uint32_t)lower_32_bits(address); ret = smu_send_smc_msg_with_param(smu, SMU_MSG_DramLogSetDramAddrHigh, - address_high); + address_high, NULL); if (ret) return ret; ret = smu_send_smc_msg_with_param(smu, SMU_MSG_DramLogSetDramAddrLow, - address_low); + address_low, NULL); if (ret) return ret; ret = smu_send_smc_msg_with_param(smu, SMU_MSG_DramLogSetDramSize, - (uint32_t)memory_pool->size); + (uint32_t)memory_pool->size, NULL); if (ret) return ret; @@ -757,7 +772,7 @@ int smu_v11_0_set_deep_sleep_dcefclk(struct smu_context *smu, uint32_t clk) int ret; ret = smu_send_smc_msg_with_param(smu, - SMU_MSG_SetMinDeepSleepDcefclk, clk); + SMU_MSG_SetMinDeepSleepDcefclk, clk, NULL); if (ret) pr_err("SMU11 attempt to set divider for DCEFCLK Failed!"); @@ -784,11 +799,13 @@ int smu_v11_0_set_driver_table_location(struct smu_context *smu) if (driver_table->mc_address) { ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetDriverDramAddrHigh, - upper_32_bits(driver_table->mc_address)); + upper_32_bits(driver_table->mc_address), + NULL); if (!ret) ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetDriverDramAddrLow, - lower_32_bits(driver_table->mc_address)); + lower_32_bits(driver_table->mc_address), + NULL); } return ret; @@ -802,11 +819,13 @@ int smu_v11_0_set_tool_table_location(struct smu_context *smu) if (tool_table->mc_address) { ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetToolsDramAddrHigh, - upper_32_bits(tool_table->mc_address)); + upper_32_bits(tool_table->mc_address), + NULL); if (!ret) ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetToolsDramAddrLow, - lower_32_bits(tool_table->mc_address)); + lower_32_bits(tool_table->mc_address), + NULL); } return ret; @@ -819,7 +838,7 @@ int smu_v11_0_init_display_count(struct smu_context *smu, uint32_t count) if (!smu->pm_enabled) return ret; - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_NumOfDisplays, count); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_NumOfDisplays, count, NULL); return ret; } @@ -837,12 +856,12 @@ int smu_v11_0_set_allowed_mask(struct smu_context *smu) bitmap_copy((unsigned long *)feature_mask, feature->allowed, 64); ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetAllowedFeaturesMaskHigh, - feature_mask[1]); + feature_mask[1], NULL); if (ret) goto failed; ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetAllowedFeaturesMaskLow, - feature_mask[0]); + feature_mask[0], NULL); if (ret) goto failed; @@ -862,17 +881,11 @@ int smu_v11_0_get_enabled_mask(struct smu_context *smu, return -EINVAL; if (bitmap_empty(feature->enabled, feature->feature_num)) { - ret = smu_send_smc_msg(smu, SMU_MSG_GetEnabledSmuFeaturesHigh); - if (ret) - return ret; - ret = smu_read_smc_arg(smu, &feature_mask_high); + ret = smu_send_smc_msg(smu, SMU_MSG_GetEnabledSmuFeaturesHigh, &feature_mask_high); if (ret) return ret; - ret = smu_send_smc_msg(smu, SMU_MSG_GetEnabledSmuFeaturesLow); - if (ret) - return ret; - ret = smu_read_smc_arg(smu, &feature_mask_low); + ret = smu_send_smc_msg(smu, SMU_MSG_GetEnabledSmuFeaturesLow, &feature_mask_low); if (ret) return ret; @@ -894,10 +907,13 @@ int smu_v11_0_system_features_control(struct smu_context *smu, int ret = 0; ret = smu_send_smc_msg(smu, (en ? SMU_MSG_EnableAllSmuFeatures : - SMU_MSG_DisableAllSmuFeatures)); + SMU_MSG_DisableAllSmuFeatures), NULL); if (ret) return ret; + bitmap_zero(feature->enabled, feature->feature_num); + bitmap_zero(feature->supported, feature->feature_num); + if (en) { ret = smu_feature_get_enabled_mask(smu, feature_mask, 2); if (ret) @@ -907,9 +923,6 @@ int smu_v11_0_system_features_control(struct smu_context *smu, feature->feature_num); bitmap_copy(feature->supported, (unsigned long *)&feature_mask, feature->feature_num); - } else { - bitmap_zero(feature->enabled, feature->feature_num); - bitmap_zero(feature->supported, feature->feature_num); } return ret; @@ -923,7 +936,7 @@ int smu_v11_0_notify_display_change(struct smu_context *smu) return ret; if (smu_feature_is_enabled(smu, SMU_FEATURE_DPM_UCLK_BIT) && smu->adev->gmc.vram_type == AMDGPU_VRAM_TYPE_HBM) - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetUclkFastSwitch, 1); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetUclkFastSwitch, 1, NULL); return ret; } @@ -947,30 +960,24 @@ smu_v11_0_get_max_sustainable_clock(struct smu_context *smu, uint32_t *clock, return -EINVAL; ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetDcModeMaxDpmFreq, - clk_id << 16); + clk_id << 16, clock); if (ret) { pr_err("[GetMaxSustainableClock] Failed to get max DC clock from SMC!"); return ret; } - ret = smu_read_smc_arg(smu, clock); - if (ret) - return ret; - if (*clock != 0) return 0; /* if DC limit is zero, return AC limit */ ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetMaxDpmFreq, - clk_id << 16); + clk_id << 16, clock); if (ret) { pr_err("[GetMaxSustainableClock] failed to get max AC clock from SMC!"); return ret; } - ret = smu_read_smc_arg(smu, clock); - - return ret; + return 0; } int smu_v11_0_init_max_sustainable_clocks(struct smu_context *smu) @@ -978,8 +985,12 @@ int smu_v11_0_init_max_sustainable_clocks(struct smu_context *smu) struct smu_11_0_max_sustainable_clocks *max_sustainable_clocks; int ret = 0; - max_sustainable_clocks = kzalloc(sizeof(struct smu_11_0_max_sustainable_clocks), + if (!smu->smu_table.max_sustainable_clocks) + max_sustainable_clocks = kzalloc(sizeof(struct smu_11_0_max_sustainable_clocks), GFP_KERNEL); + else + max_sustainable_clocks = smu->smu_table.max_sustainable_clocks; + smu->smu_table.max_sustainable_clocks = (void *)max_sustainable_clocks; max_sustainable_clocks->uclock = smu->smu_table.boot_values.uclk / 100; @@ -1102,7 +1113,7 @@ int smu_v11_0_set_power_limit(struct smu_context *smu, uint32_t n) return -EOPNOTSUPP; } - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetPptLimit, n); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetPptLimit, n, NULL); if (ret) { pr_err("[%s] Set power limit Failed!\n", __func__); return ret; @@ -1132,11 +1143,7 @@ int smu_v11_0_get_current_clk_freq(struct smu_context *smu, ret = smu_get_current_clk_freq_by_table(smu, clk_id, &freq); else { ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetDpmClockFreq, - (asic_clk_id << 16)); - if (ret) - return ret; - - ret = smu_read_smc_arg(smu, &freq); + (asic_clk_id << 16), &freq); if (ret) return ret; } @@ -1371,9 +1378,9 @@ int smu_v11_0_gfx_off_control(struct smu_context *smu, bool enable) if (!(adev->pm.pp_feature & PP_GFXOFF_MASK)) return 0; if (enable) - ret = smu_send_smc_msg(smu, SMU_MSG_AllowGfxOff); + ret = smu_send_smc_msg(smu, SMU_MSG_AllowGfxOff, NULL); else - ret = smu_send_smc_msg(smu, SMU_MSG_DisallowGfxOff); + ret = smu_send_smc_msg(smu, SMU_MSG_DisallowGfxOff, NULL); break; default: break; @@ -1511,10 +1518,18 @@ int smu_v11_0_set_xgmi_pstate(struct smu_context *smu, int ret = 0; ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetXgmiMode, - pstate ? XGMI_MODE_PSTATE_D0 : XGMI_MODE_PSTATE_D3); + pstate ? XGMI_MODE_PSTATE_D0 : XGMI_MODE_PSTATE_D3, + NULL); return ret; } +static int smu_v11_0_ack_ac_dc_interrupt(struct smu_context *smu) +{ + return smu_send_smc_msg(smu, + SMU_MSG_ReenableAcDcInterrupt, + NULL); +} + #define THM_11_0__SRCID__THM_DIG_THERM_L2H 0 /* ASIC_TEMP > CG_THERMAL_INT.DIG_THERM_INTH */ #define THM_11_0__SRCID__THM_DIG_THERM_H2L 1 /* ASIC_TEMP < CG_THERMAL_INT.DIG_THERM_INTL */ @@ -1548,6 +1563,9 @@ static int smu_v11_0_irq_process(struct amdgpu_device *adev, break; } + } else if (client_id == SOC15_IH_CLIENTID_MP1) { + if (src_id == 0xfe) + smu_v11_0_ack_ac_dc_interrupt(&adev->smu); } return 0; @@ -1587,6 +1605,12 @@ int smu_v11_0_register_irq_handler(struct smu_context *smu) if (ret) return ret; + ret = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_MP1, + 0xfe, + irq_src); + if (ret) + return ret; + return ret; } @@ -1624,21 +1648,19 @@ int smu_v11_0_set_azalia_d3_pme(struct smu_context *smu) { int ret = 0; - ret = smu_send_smc_msg(smu, SMU_MSG_BacoAudioD3PME); + ret = smu_send_smc_msg(smu, SMU_MSG_BacoAudioD3PME, NULL); return ret; } static int smu_v11_0_baco_set_armd3_sequence(struct smu_context *smu, enum smu_v11_0_baco_seq baco_seq) { - return smu_send_smc_msg_with_param(smu, SMU_MSG_ArmD3, baco_seq); + return smu_send_smc_msg_with_param(smu, SMU_MSG_ArmD3, baco_seq, NULL); } bool smu_v11_0_baco_is_support(struct smu_context *smu) { - struct amdgpu_device *adev = smu->adev; struct smu_baco_context *smu_baco = &smu->smu_baco; - uint32_t val; bool baco_support; mutex_lock(&smu_baco->mutex); @@ -1653,11 +1675,7 @@ bool smu_v11_0_baco_is_support(struct smu_context *smu) !smu_feature_is_enabled(smu, SMU_FEATURE_BACO_BIT)) return false; - val = RREG32_SOC15(NBIO, 0, mmRCC_BIF_STRAP0); - if (val & RCC_BIF_STRAP0__STRAP_PX_CAPABLE_MASK) - return true; - - return false; + return true; } enum smu_baco_state smu_v11_0_baco_get_state(struct smu_context *smu) @@ -1674,11 +1692,9 @@ enum smu_baco_state smu_v11_0_baco_get_state(struct smu_context *smu) int smu_v11_0_baco_set_state(struct smu_context *smu, enum smu_baco_state state) { - struct smu_baco_context *smu_baco = &smu->smu_baco; struct amdgpu_device *adev = smu->adev; struct amdgpu_ras *ras = amdgpu_ras_get_context(adev); - uint32_t bif_doorbell_intr_cntl; uint32_t data; int ret = 0; @@ -1687,33 +1703,21 @@ int smu_v11_0_baco_set_state(struct smu_context *smu, enum smu_baco_state state) mutex_lock(&smu_baco->mutex); - bif_doorbell_intr_cntl = RREG32_SOC15(NBIO, 0, mmBIF_DOORBELL_INT_CNTL); - if (state == SMU_BACO_STATE_ENTER) { - bif_doorbell_intr_cntl = REG_SET_FIELD(bif_doorbell_intr_cntl, - BIF_DOORBELL_INT_CNTL, - DOORBELL_INTERRUPT_DISABLE, 1); - WREG32_SOC15(NBIO, 0, mmBIF_DOORBELL_INT_CNTL, bif_doorbell_intr_cntl); - if (!ras || !ras->supported) { data = RREG32_SOC15(THM, 0, mmTHM_BACO_CNTL); data |= 0x80000000; WREG32_SOC15(THM, 0, mmTHM_BACO_CNTL, data); - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_EnterBaco, 0); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_EnterBaco, 0, NULL); } else { - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_EnterBaco, 1); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_EnterBaco, 1, NULL); } } else { - ret = smu_send_smc_msg(smu, SMU_MSG_ExitBaco); + ret = smu_send_smc_msg(smu, SMU_MSG_ExitBaco, NULL); if (ret) goto out; - bif_doorbell_intr_cntl = REG_SET_FIELD(bif_doorbell_intr_cntl, - BIF_DOORBELL_INT_CNTL, - DOORBELL_INTERRUPT_DISABLE, 0); - WREG32_SOC15(NBIO, 0, mmBIF_DOORBELL_INT_CNTL, bif_doorbell_intr_cntl); - /* clear vbios scratch 6 and 7 for coming asic reinit */ WREG32(adev->bios_scratch_reg_offset + 6, 0); WREG32(adev->bios_scratch_reg_offset + 7, 0); @@ -1773,19 +1777,13 @@ int smu_v11_0_get_dpm_ultimate_freq(struct smu_context *smu, enum smu_clk_type c param = (clk_id & 0xffff) << 16; if (max) { - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetMaxDpmFreq, param); - if (ret) - goto failed; - ret = smu_read_smc_arg(smu, max); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetMaxDpmFreq, param, max); if (ret) goto failed; } if (min) { - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetMinDpmFreq, param); - if (ret) - goto failed; - ret = smu_read_smc_arg(smu, min); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetMinDpmFreq, param, min); if (ret) goto failed; } @@ -1807,7 +1805,7 @@ int smu_v11_0_set_soft_freq_limited_range(struct smu_context *smu, enum smu_clk_ if (max > 0) { param = (uint32_t)((clk_id << 16) | (max & 0xffff)); ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxByFreq, - param); + param, NULL); if (ret) return ret; } @@ -1815,7 +1813,7 @@ int smu_v11_0_set_soft_freq_limited_range(struct smu_context *smu, enum smu_clk_ if (min > 0) { param = (uint32_t)((clk_id << 16) | (min & 0xffff)); ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMinByFreq, - param); + param, NULL); if (ret) return ret; } @@ -1935,3 +1933,18 @@ int smu_v11_0_set_performance_level(struct smu_context *smu, return ret; } +int smu_v11_0_set_power_source(struct smu_context *smu, + enum smu_power_src_type power_src) +{ + int pwr_source; + + pwr_source = smu_power_get_index(smu, (uint32_t)power_src); + if (pwr_source < 0) + return -EINVAL; + + return smu_send_smc_msg_with_param(smu, + SMU_MSG_NotifyPowerSource, + pwr_source, + NULL); +} + diff --git a/drivers/gpu/drm/amd/powerplay/smu_v12_0.c b/drivers/gpu/drm/amd/powerplay/smu_v12_0.c index 870e6db2907e..169ebdad87b8 100644 --- a/drivers/gpu/drm/amd/powerplay/smu_v12_0.c +++ b/drivers/gpu/drm/amd/powerplay/smu_v12_0.c @@ -20,7 +20,6 @@ * OTHER DEALINGS IN THE SOFTWARE. */ -#include "pp_debug.h" #include <linux/firmware.h> #include "amdgpu.h" #include "amdgpu_smu.h" @@ -50,7 +49,7 @@ int smu_v12_0_send_msg_without_waiting(struct smu_context *smu, return 0; } -int smu_v12_0_read_arg(struct smu_context *smu, uint32_t *arg) +static int smu_v12_0_read_arg(struct smu_context *smu, uint32_t *arg) { struct amdgpu_device *adev = smu->adev; @@ -78,7 +77,8 @@ int smu_v12_0_wait_for_response(struct smu_context *smu) int smu_v12_0_send_msg_with_param(struct smu_context *smu, enum smu_message_type msg, - uint32_t param) + uint32_t param, + uint32_t *read_arg) { struct amdgpu_device *adev = smu->adev; int ret = 0, index = 0; @@ -87,11 +87,12 @@ smu_v12_0_send_msg_with_param(struct smu_context *smu, if (index < 0) return index; + mutex_lock(&smu->message_lock); ret = smu_v12_0_wait_for_response(smu); if (ret) { pr_err("Msg issuing pre-check failed and " "SMU may be not in the right state!\n"); - return ret; + goto out; } WREG32_SOC15(MP1, 0, mmMP1_SMN_C2PMSG_90, 0); @@ -101,10 +102,21 @@ smu_v12_0_send_msg_with_param(struct smu_context *smu, smu_v12_0_send_msg_without_waiting(smu, (uint16_t)index); ret = smu_v12_0_wait_for_response(smu); - if (ret) + if (ret) { pr_err("Failed to send message 0x%x, response 0x%x param 0x%x\n", index, ret, param); - + goto out; + } + if (read_arg) { + ret = smu_v12_0_read_arg(smu, read_arg); + if (ret) { + pr_err("Failed to read message arg 0x%x, response 0x%x param 0x%x\n", + index, ret, param); + goto out; + } + } +out: + mutex_unlock(&smu->message_lock); return ret; } @@ -163,9 +175,9 @@ int smu_v12_0_powergate_sdma(struct smu_context *smu, bool gate) return 0; if (gate) - return smu_send_smc_msg(smu, SMU_MSG_PowerDownSdma); + return smu_send_smc_msg(smu, SMU_MSG_PowerDownSdma, NULL); else - return smu_send_smc_msg(smu, SMU_MSG_PowerUpSdma); + return smu_send_smc_msg(smu, SMU_MSG_PowerUpSdma, NULL); } int smu_v12_0_powergate_vcn(struct smu_context *smu, bool gate) @@ -174,9 +186,9 @@ int smu_v12_0_powergate_vcn(struct smu_context *smu, bool gate) return 0; if (gate) - return smu_send_smc_msg(smu, SMU_MSG_PowerDownVcn); + return smu_send_smc_msg(smu, SMU_MSG_PowerDownVcn, NULL); else - return smu_send_smc_msg(smu, SMU_MSG_PowerUpVcn); + return smu_send_smc_msg(smu, SMU_MSG_PowerUpVcn, NULL); } int smu_v12_0_powergate_jpeg(struct smu_context *smu, bool gate) @@ -185,9 +197,9 @@ int smu_v12_0_powergate_jpeg(struct smu_context *smu, bool gate) return 0; if (gate) - return smu_send_smc_msg_with_param(smu, SMU_MSG_PowerDownJpeg, 0); + return smu_send_smc_msg_with_param(smu, SMU_MSG_PowerDownJpeg, 0, NULL); else - return smu_send_smc_msg_with_param(smu, SMU_MSG_PowerUpJpeg, 0); + return smu_send_smc_msg_with_param(smu, SMU_MSG_PowerUpJpeg, 0, NULL); } int smu_v12_0_set_gfx_cgpg(struct smu_context *smu, bool enable) @@ -196,7 +208,9 @@ int smu_v12_0_set_gfx_cgpg(struct smu_context *smu, bool enable) return 0; return smu_v12_0_send_msg_with_param(smu, - SMU_MSG_SetGfxCGPG, enable ? 1 : 0); + SMU_MSG_SetGfxCGPG, + enable ? 1 : 0, + NULL); } int smu_v12_0_read_sensor(struct smu_context *smu, @@ -262,10 +276,10 @@ int smu_v12_0_gfx_off_control(struct smu_context *smu, bool enable) int ret = 0, timeout = 500; if (enable) { - ret = smu_send_smc_msg(smu, SMU_MSG_AllowGfxOff); + ret = smu_send_smc_msg(smu, SMU_MSG_AllowGfxOff, NULL); } else { - ret = smu_send_smc_msg(smu, SMU_MSG_DisallowGfxOff); + ret = smu_send_smc_msg(smu, SMU_MSG_DisallowGfxOff, NULL); /* confirm gfx is back to "on" state, timeout is 0.5 second */ while (!(smu_v12_0_get_gfxoff_status(smu) == 2)) { @@ -331,17 +345,11 @@ int smu_v12_0_get_enabled_mask(struct smu_context *smu, if (!feature_mask || num < 2) return -EINVAL; - ret = smu_send_smc_msg(smu, SMU_MSG_GetEnabledSmuFeaturesHigh); - if (ret) - return ret; - ret = smu_read_smc_arg(smu, &feature_mask_high); + ret = smu_send_smc_msg(smu, SMU_MSG_GetEnabledSmuFeaturesHigh, &feature_mask_high); if (ret) return ret; - ret = smu_send_smc_msg(smu, SMU_MSG_GetEnabledSmuFeaturesLow); - if (ret) - return ret; - ret = smu_read_smc_arg(smu, &feature_mask_low); + ret = smu_send_smc_msg(smu, SMU_MSG_GetEnabledSmuFeaturesLow, &feature_mask_low); if (ret) return ret; @@ -388,14 +396,11 @@ int smu_v12_0_get_dpm_ultimate_freq(struct smu_context *smu, enum smu_clk_type c switch (clk_type) { case SMU_GFXCLK: case SMU_SCLK: - ret = smu_send_smc_msg(smu, SMU_MSG_GetMaxGfxclkFrequency); + ret = smu_send_smc_msg(smu, SMU_MSG_GetMaxGfxclkFrequency, max); if (ret) { pr_err("Attempt to get max GX frequency from SMC Failed !\n"); goto failed; } - ret = smu_read_smc_arg(smu, max); - if (ret) - goto failed; break; case SMU_UCLK: case SMU_FCLK: @@ -419,14 +424,11 @@ int smu_v12_0_get_dpm_ultimate_freq(struct smu_context *smu, enum smu_clk_type c switch (clk_type) { case SMU_GFXCLK: case SMU_SCLK: - ret = smu_send_smc_msg(smu, SMU_MSG_GetMinGfxclkFrequency); + ret = smu_send_smc_msg(smu, SMU_MSG_GetMinGfxclkFrequency, min); if (ret) { pr_err("Attempt to get min GX frequency from SMC Failed !\n"); goto failed; } - ret = smu_read_smc_arg(smu, min); - if (ret) - goto failed; break; case SMU_UCLK: case SMU_FCLK: @@ -450,7 +452,7 @@ failed: } int smu_v12_0_mode2_reset(struct smu_context *smu){ - return smu_v12_0_send_msg_with_param(smu, SMU_MSG_GfxDeviceDriverReset, SMU_RESET_MODE_2); + return smu_v12_0_send_msg_with_param(smu, SMU_MSG_GfxDeviceDriverReset, SMU_RESET_MODE_2, NULL); } int smu_v12_0_set_soft_freq_limited_range(struct smu_context *smu, enum smu_clk_type clk_type, @@ -458,45 +460,42 @@ int smu_v12_0_set_soft_freq_limited_range(struct smu_context *smu, enum smu_clk_ { int ret = 0; - if (max < min) - return -EINVAL; - switch (clk_type) { case SMU_GFXCLK: case SMU_SCLK: - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinGfxClk, min); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinGfxClk, min, NULL); if (ret) return ret; - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxGfxClk, max); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxGfxClk, max, NULL); if (ret) return ret; break; case SMU_FCLK: case SMU_MCLK: - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinFclkByFreq, min); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinFclkByFreq, min, NULL); if (ret) return ret; - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxFclkByFreq, max); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxFclkByFreq, max, NULL); if (ret) return ret; break; case SMU_SOCCLK: - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinSocclkByFreq, min); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinSocclkByFreq, min, NULL); if (ret) return ret; - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxSocclkByFreq, max); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxSocclkByFreq, max, NULL); if (ret) return ret; break; case SMU_VCLK: - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinVcn, min); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinVcn, min, NULL); if (ret) return ret; - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxVcn, max); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxVcn, max, NULL); if (ret) return ret; break; @@ -515,11 +514,13 @@ int smu_v12_0_set_driver_table_location(struct smu_context *smu) if (driver_table->mc_address) { ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetDriverDramAddrHigh, - upper_32_bits(driver_table->mc_address)); + upper_32_bits(driver_table->mc_address), + NULL); if (!ret) ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetDriverDramAddrLow, - lower_32_bits(driver_table->mc_address)); + lower_32_bits(driver_table->mc_address), + NULL); } return ret; diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/vega20_smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/vega20_smumgr.c index 49e5ef3e3876..16aa171971d3 100644 --- a/drivers/gpu/drm/amd/powerplay/smumgr/vega20_smumgr.c +++ b/drivers/gpu/drm/amd/powerplay/smumgr/vega20_smumgr.c @@ -33,6 +33,8 @@ #include "smu7_smumgr.h" #include "vega20_hwmgr.h" +#include "smu_v11_0_i2c.h" + /* MP Apertures */ #define MP0_Public 0x03800000 #define MP0_SRAM 0x03900000 @@ -406,6 +408,7 @@ static int vega20_smu_init(struct pp_hwmgr *hwmgr) struct vega20_smumgr *priv; unsigned long tools_size = 0x19000; int ret = 0; + struct amdgpu_device *adev = hwmgr->adev; struct cgs_firmware_info info = {0}; @@ -505,6 +508,10 @@ static int vega20_smu_init(struct pp_hwmgr *hwmgr) priv->smu_tables.entry[TABLE_ACTIVITY_MONITOR_COEFF].version = 0x01; priv->smu_tables.entry[TABLE_ACTIVITY_MONITOR_COEFF].size = sizeof(DpmActivityMonitorCoeffInt_t); + ret = smu_v11_0_i2c_eeprom_control_init(&adev->pm.smu_i2c); + if (ret) + goto err4; + return 0; err4: @@ -537,6 +544,9 @@ static int vega20_smu_fini(struct pp_hwmgr *hwmgr) { struct vega20_smumgr *priv = (struct vega20_smumgr *)(hwmgr->smu_backend); + struct amdgpu_device *adev = hwmgr->adev; + + smu_v11_0_i2c_eeprom_control_fini(&adev->pm.smu_i2c); if (priv) { amdgpu_bo_free_kernel(&priv->smu_tables.entry[TABLE_PPTABLE].handle, @@ -560,6 +570,7 @@ static int vega20_smu_fini(struct pp_hwmgr *hwmgr) kfree(hwmgr->smu_backend); hwmgr->smu_backend = NULL; } + return 0; } diff --git a/drivers/gpu/drm/amd/powerplay/vega20_ppt.c b/drivers/gpu/drm/amd/powerplay/vega20_ppt.c index 4ad8d6c14ee5..3f1044326dcb 100644 --- a/drivers/gpu/drm/amd/powerplay/vega20_ppt.c +++ b/drivers/gpu/drm/amd/powerplay/vega20_ppt.c @@ -21,7 +21,6 @@ * */ -#include "pp_debug.h" #include <linux/firmware.h> #include "amdgpu.h" #include "amdgpu_smu.h" @@ -36,6 +35,7 @@ #include "vega20_ppt.h" #include "vega20_pptable.h" #include "vega20_ppsmc.h" +#include "nbio/nbio_7_4_offset.h" #include "nbio/nbio_7_4_sh_mask.h" #include "asic_reg/thm/thm_11_0_2_offset.h" #include "asic_reg/thm/thm_11_0_2_sh_mask.h" @@ -587,7 +587,7 @@ static int vega20_check_powerplay_table(struct smu_context *smu) static int vega20_run_btc_afll(struct smu_context *smu) { - return smu_send_smc_msg(smu, SMU_MSG_RunAfllBtc); + return smu_send_smc_msg(smu, SMU_MSG_RunAfllBtc, NULL); } #define FEATURE_MASK(feature) (1ULL << feature) @@ -670,13 +670,13 @@ vega20_set_single_dpm_table(struct smu_context *smu, ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetDpmFreqByIndex, - (clk_id << 16 | 0xFF)); + (clk_id << 16 | 0xFF), + &num_of_levels); if (ret) { pr_err("[GetNumOfDpmLevel] failed to get dpm levels!"); return ret; } - smu_read_smc_arg(smu, &num_of_levels); if (!num_of_levels) { pr_err("[GetNumOfDpmLevel] number of clk levels is invalid!"); return -EINVAL; @@ -687,12 +687,12 @@ vega20_set_single_dpm_table(struct smu_context *smu, for (i = 0; i < num_of_levels; i++) { ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetDpmFreqByIndex, - (clk_id << 16 | i)); + (clk_id << 16 | i), + &clk); if (ret) { pr_err("[GetDpmFreqByIndex] failed to get dpm freq by index!"); return ret; } - smu_read_smc_arg(smu, &clk); if (!clk) { pr_err("[GetDpmFreqByIndex] clk value is invalid!"); return -EINVAL; @@ -1200,7 +1200,8 @@ static int vega20_upload_dpm_level(struct smu_context *smu, bool max, single_dpm_table->dpm_state.soft_min_level; ret = smu_send_smc_msg_with_param(smu, (max ? SMU_MSG_SetSoftMaxByFreq : SMU_MSG_SetSoftMinByFreq), - (PPCLK_GFXCLK << 16) | (freq & 0xffff)); + (PPCLK_GFXCLK << 16) | (freq & 0xffff), + NULL); if (ret) { pr_err("Failed to set soft %s gfxclk !\n", max ? "max" : "min"); @@ -1215,7 +1216,8 @@ static int vega20_upload_dpm_level(struct smu_context *smu, bool max, single_dpm_table->dpm_state.soft_min_level; ret = smu_send_smc_msg_with_param(smu, (max ? SMU_MSG_SetSoftMaxByFreq : SMU_MSG_SetSoftMinByFreq), - (PPCLK_UCLK << 16) | (freq & 0xffff)); + (PPCLK_UCLK << 16) | (freq & 0xffff), + NULL); if (ret) { pr_err("Failed to set soft %s memclk !\n", max ? "max" : "min"); @@ -1230,7 +1232,8 @@ static int vega20_upload_dpm_level(struct smu_context *smu, bool max, single_dpm_table->dpm_state.soft_min_level; ret = smu_send_smc_msg_with_param(smu, (max ? SMU_MSG_SetSoftMaxByFreq : SMU_MSG_SetSoftMinByFreq), - (PPCLK_SOCCLK << 16) | (freq & 0xffff)); + (PPCLK_SOCCLK << 16) | (freq & 0xffff), + NULL); if (ret) { pr_err("Failed to set soft %s socclk !\n", max ? "max" : "min"); @@ -1245,7 +1248,8 @@ static int vega20_upload_dpm_level(struct smu_context *smu, bool max, single_dpm_table->dpm_state.soft_min_level; ret = smu_send_smc_msg_with_param(smu, (max ? SMU_MSG_SetSoftMaxByFreq : SMU_MSG_SetSoftMinByFreq), - (PPCLK_FCLK << 16) | (freq & 0xffff)); + (PPCLK_FCLK << 16) | (freq & 0xffff), + NULL); if (ret) { pr_err("Failed to set soft %s fclk !\n", max ? "max" : "min"); @@ -1260,7 +1264,8 @@ static int vega20_upload_dpm_level(struct smu_context *smu, bool max, if (!max) { ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinByFreq, - (PPCLK_DCEFCLK << 16) | (freq & 0xffff)); + (PPCLK_DCEFCLK << 16) | (freq & 0xffff), + NULL); if (ret) { pr_err("Failed to set hard min dcefclk !\n"); return ret; @@ -1421,7 +1426,9 @@ static int vega20_force_clk_levels(struct smu_context *smu, } ret = smu_send_smc_msg_with_param(smu, - SMU_MSG_SetMinLinkDpmByIndex, soft_min_level); + SMU_MSG_SetMinLinkDpmByIndex, + soft_min_level, + NULL); if (ret) pr_err("Failed to set min link dpm level!\n"); @@ -1477,13 +1484,13 @@ static int vega20_overdrive_get_gfx_clk_base_voltage(struct smu_context *smu, ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetAVFSVoltageByDpm, - ((AVFS_CURVE << 24) | (OD8_HOTCURVE_TEMPERATURE << 16) | freq)); + ((AVFS_CURVE << 24) | (OD8_HOTCURVE_TEMPERATURE << 16) | freq), + voltage); if (ret) { pr_err("[GetBaseVoltage] failed to get GFXCLK AVFS voltage from SMU!"); return ret; } - smu_read_smc_arg(smu, voltage); *voltage = *voltage / VOLTAGE_SCALE; return 0; @@ -1956,8 +1963,10 @@ static int vega20_set_power_profile_mode(struct smu_context *smu, long *input, u workload_type = smu_workload_get_type(smu, smu->power_profile_mode); if (workload_type < 0) return -EINVAL; - smu_send_smc_msg_with_param(smu, SMU_MSG_SetWorkloadMask, - 1 << workload_type); + smu_send_smc_msg_with_param(smu, + SMU_MSG_SetWorkloadMask, + 1 << workload_type, + NULL); return ret; } @@ -2029,7 +2038,8 @@ vega20_set_uclk_to_highest_dpm_level(struct smu_context *smu, dpm_table->dpm_state.hard_min_level = dpm_table->dpm_levels[dpm_table->count - 1].value; ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinByFreq, - (PPCLK_UCLK << 16) | dpm_table->dpm_state.hard_min_level); + (PPCLK_UCLK << 16) | dpm_table->dpm_state.hard_min_level, + NULL); if (ret) { pr_err("[%s] Set hard min uclk failed!", __func__); return ret; @@ -2047,7 +2057,7 @@ static int vega20_pre_display_config_changed(struct smu_context *smu) if (!smu->smu_dpm.dpm_context) return -EINVAL; - smu_send_smc_msg_with_param(smu, SMU_MSG_NumOfDisplays, 0); + smu_send_smc_msg_with_param(smu, SMU_MSG_NumOfDisplays, 0, NULL); ret = vega20_set_uclk_to_highest_dpm_level(smu, &dpm_table->mem_table); if (ret) @@ -2074,7 +2084,8 @@ static int vega20_display_config_changed(struct smu_context *smu) smu_feature_is_supported(smu, SMU_FEATURE_DPM_SOCCLK_BIT)) { smu_send_smc_msg_with_param(smu, SMU_MSG_NumOfDisplays, - smu->display_config->num_display); + smu->display_config->num_display, + NULL); } return ret; @@ -2247,7 +2258,8 @@ vega20_notify_smc_display_config(struct smu_context *smu) if (smu_feature_is_supported(smu, SMU_FEATURE_DS_DCEFCLK_BIT)) { ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetMinDeepSleepDcefclk, - min_clocks.dcef_clock_in_sr/100); + min_clocks.dcef_clock_in_sr/100, + NULL); if (ret) { pr_err("Attempt to set divider for DCEFCLK Failed!"); return ret; @@ -2262,7 +2274,8 @@ vega20_notify_smc_display_config(struct smu_context *smu) memtable->dpm_state.hard_min_level = min_clocks.memory_clock/100; ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinByFreq, - (PPCLK_UCLK << 16) | memtable->dpm_state.hard_min_level); + (PPCLK_UCLK << 16) | memtable->dpm_state.hard_min_level, + NULL); if (ret) { pr_err("[%s] Set hard min uclk failed!", __func__); return ret; @@ -2853,8 +2866,10 @@ static int vega20_set_thermal_fan_table(struct smu_context *smu) struct smu_table_context *table_context = &smu->smu_table; PPTable_t *pptable = table_context->driver_pptable; - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetFanTemperatureTarget, - (uint32_t)pptable->FanTargetTemperature); + ret = smu_send_smc_msg_with_param(smu, + SMU_MSG_SetFanTemperatureTarget, + (uint32_t)pptable->FanTargetTemperature, + NULL); return ret; } @@ -2864,15 +2879,13 @@ static int vega20_get_fan_speed_rpm(struct smu_context *smu, { int ret; - ret = smu_send_smc_msg(smu, SMU_MSG_GetCurrentRpm); + ret = smu_send_smc_msg(smu, SMU_MSG_GetCurrentRpm, speed); if (ret) { pr_err("Attempt to get current RPM from SMC Failed!\n"); return ret; } - smu_read_smc_arg(smu, speed); - return 0; } @@ -3137,7 +3150,7 @@ static int vega20_set_df_cstate(struct smu_context *smu, return -EINVAL; } - return smu_send_smc_msg_with_param(smu, SMU_MSG_DFCstateControl, state); + return smu_send_smc_msg_with_param(smu, SMU_MSG_DFCstateControl, state, NULL); } static int vega20_update_pcie_parameters(struct smu_context *smu, @@ -3155,12 +3168,24 @@ static int vega20_update_pcie_parameters(struct smu_context *smu, pptable->PcieLaneCount[i] : pcie_width_cap); ret = smu_send_smc_msg_with_param(smu, SMU_MSG_OverridePcieParameters, - smu_pcie_arg); + smu_pcie_arg, + NULL); } return ret; } +static bool vega20_is_baco_supported(struct smu_context *smu) +{ + struct amdgpu_device *adev = smu->adev; + uint32_t val; + + if (!smu_v11_0_baco_is_support(smu)) + return false; + + val = RREG32_SOC15(NBIO, 0, mmRCC_BIF_STRAP0); + return (val & RCC_BIF_STRAP0__STRAP_PX_CAPABLE_MASK) ? true : false; +} static const struct pptable_funcs vega20_ppt_funcs = { .tables_init = vega20_tables_init, @@ -3229,7 +3254,6 @@ static const struct pptable_funcs vega20_ppt_funcs = { .notify_memory_pool_location = smu_v11_0_notify_memory_pool_location, .system_features_control = smu_v11_0_system_features_control, .send_smc_msg_with_param = smu_v11_0_send_msg_with_param, - .read_smc_arg = smu_v11_0_read_arg, .init_display_count = smu_v11_0_init_display_count, .set_allowed_mask = smu_v11_0_set_allowed_mask, .get_enabled_mask = smu_v11_0_get_enabled_mask, @@ -3250,7 +3274,7 @@ static const struct pptable_funcs vega20_ppt_funcs = { .register_irq_handler = smu_v11_0_register_irq_handler, .set_azalia_d3_pme = smu_v11_0_set_azalia_d3_pme, .get_max_sustainable_clocks_by_dc = smu_v11_0_get_max_sustainable_clocks_by_dc, - .baco_is_support= smu_v11_0_baco_is_support, + .baco_is_support= vega20_is_baco_supported, .baco_get_state = smu_v11_0_baco_get_state, .baco_set_state = smu_v11_0_baco_set_state, .baco_enter = smu_v11_0_baco_enter, diff --git a/drivers/gpu/drm/arc/arcpgu_hdmi.c b/drivers/gpu/drm/arc/arcpgu_hdmi.c index 8fd7094beece..52839934f2fb 100644 --- a/drivers/gpu/drm/arc/arcpgu_hdmi.c +++ b/drivers/gpu/drm/arc/arcpgu_hdmi.c @@ -40,7 +40,7 @@ int arcpgu_drm_hdmi_init(struct drm_device *drm, struct device_node *np) return ret; /* Link drm_bridge to encoder */ - ret = drm_bridge_attach(encoder, bridge, NULL); + ret = drm_bridge_attach(encoder, bridge, NULL, 0); if (ret) drm_encoder_cleanup(encoder); diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_drv.c b/drivers/gpu/drm/arm/display/komeda/komeda_drv.c index ea5cd1e17304..e7933930a657 100644 --- a/drivers/gpu/drm/arm/display/komeda/komeda_drv.c +++ b/drivers/gpu/drm/arm/display/komeda/komeda_drv.c @@ -146,14 +146,14 @@ static const struct of_device_id komeda_of_match[] = { MODULE_DEVICE_TABLE(of, komeda_of_match); -static int komeda_rt_pm_suspend(struct device *dev) +static int __maybe_unused komeda_rt_pm_suspend(struct device *dev) { struct komeda_drv *mdrv = dev_get_drvdata(dev); return komeda_dev_suspend(mdrv->mdev); } -static int komeda_rt_pm_resume(struct device *dev) +static int __maybe_unused komeda_rt_pm_resume(struct device *dev) { struct komeda_drv *mdrv = dev_get_drvdata(dev); diff --git a/drivers/gpu/drm/armada/armada_fbdev.c b/drivers/gpu/drm/armada/armada_fbdev.c index ac8a78bfda03..f2dc371bd8e5 100644 --- a/drivers/gpu/drm/armada/armada_fbdev.c +++ b/drivers/gpu/drm/armada/armada_fbdev.c @@ -129,18 +129,12 @@ int armada_fbdev_init(struct drm_device *dev) drm_fb_helper_prepare(dev, fbh, &armada_fb_helper_funcs); - ret = drm_fb_helper_init(dev, fbh, 1); + ret = drm_fb_helper_init(dev, fbh); if (ret) { DRM_ERROR("failed to initialize drm fb helper\n"); goto err_fb_helper; } - ret = drm_fb_helper_single_add_all_connectors(fbh); - if (ret) { - DRM_ERROR("failed to add fb connectors\n"); - goto err_fb_setup; - } - ret = drm_fb_helper_initial_config(fbh, 32); if (ret) { DRM_ERROR("failed to set initial config\n"); diff --git a/drivers/gpu/drm/ast/ast_drv.h b/drivers/gpu/drm/ast/ast_drv.h index f5d8780776ae..656d591b154b 100644 --- a/drivers/gpu/drm/ast/ast_drv.h +++ b/drivers/gpu/drm/ast/ast_drv.h @@ -121,6 +121,7 @@ struct ast_private { unsigned int next_index; } cursor; + struct drm_encoder encoder; struct drm_plane primary_plane; struct drm_plane cursor_plane; @@ -238,13 +239,8 @@ struct ast_crtc { u8 offset_x, offset_y; }; -struct ast_encoder { - struct drm_encoder base; -}; - #define to_ast_crtc(x) container_of(x, struct ast_crtc, base) #define to_ast_connector(x) container_of(x, struct ast_connector, base) -#define to_ast_encoder(x) container_of(x, struct ast_encoder, base) struct ast_vbios_stdtable { u8 misc; diff --git a/drivers/gpu/drm/ast/ast_mode.c b/drivers/gpu/drm/ast/ast_mode.c index 7810a84e7e9e..cdd6c46d6557 100644 --- a/drivers/gpu/drm/ast/ast_mode.c +++ b/drivers/gpu/drm/ast/ast_mode.c @@ -40,6 +40,7 @@ #include <drm/drm_gem_vram_helper.h> #include <drm/drm_plane_helper.h> #include <drm/drm_probe_helper.h> +#include <drm/drm_simple_kms_helper.h> #include "ast_drv.h" #include "ast_tables.h" @@ -957,28 +958,18 @@ err_kfree: * Encoder */ -static void ast_encoder_destroy(struct drm_encoder *encoder) -{ - drm_encoder_cleanup(encoder); - kfree(encoder); -} - -static const struct drm_encoder_funcs ast_enc_funcs = { - .destroy = ast_encoder_destroy, -}; - static int ast_encoder_init(struct drm_device *dev) { - struct ast_encoder *ast_encoder; + struct ast_private *ast = dev->dev_private; + struct drm_encoder *encoder = &ast->encoder; + int ret; - ast_encoder = kzalloc(sizeof(struct ast_encoder), GFP_KERNEL); - if (!ast_encoder) - return -ENOMEM; + ret = drm_simple_encoder_init(dev, encoder, DRM_MODE_ENCODER_DAC); + if (ret) + return ret; - drm_encoder_init(dev, &ast_encoder->base, &ast_enc_funcs, - DRM_MODE_ENCODER_DAC, NULL); + encoder->possible_crtcs = 1; - ast_encoder->base.possible_crtcs = 1; return 0; } diff --git a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_output.c b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_output.c index 121b62682d80..e2019fe97fff 100644 --- a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_output.c +++ b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_output.c @@ -114,7 +114,7 @@ static int atmel_hlcdc_attach_endpoint(struct drm_device *dev, int endpoint) } if (bridge) { - ret = drm_bridge_attach(&output->encoder, bridge, NULL); + ret = drm_bridge_attach(&output->encoder, bridge, NULL, 0); if (!ret) return 0; diff --git a/drivers/gpu/drm/bochs/bochs_drv.c b/drivers/gpu/drm/bochs/bochs_drv.c index 10460878414e..addb0568c1af 100644 --- a/drivers/gpu/drm/bochs/bochs_drv.c +++ b/drivers/gpu/drm/bochs/bochs_drv.c @@ -23,7 +23,6 @@ static void bochs_unload(struct drm_device *dev) bochs_kms_fini(bochs); bochs_mm_fini(bochs); - bochs_hw_fini(dev); kfree(bochs); dev->dev_private = NULL; } @@ -69,6 +68,7 @@ static struct drm_driver bochs_driver = { .major = 1, .minor = 0, DRM_GEM_VRAM_DRIVER, + .release = bochs_unload, }; /* ---------------------------------------------------------------------- */ @@ -148,9 +148,9 @@ static void bochs_pci_remove(struct pci_dev *pdev) { struct drm_device *dev = pci_get_drvdata(pdev); + drm_dev_unplug(dev); drm_atomic_helper_shutdown(dev); - drm_dev_unregister(dev); - bochs_unload(dev); + bochs_hw_fini(dev); drm_dev_put(dev); } diff --git a/drivers/gpu/drm/bochs/bochs_hw.c b/drivers/gpu/drm/bochs/bochs_hw.c index b615b7dfdd9d..dce4672e3fc8 100644 --- a/drivers/gpu/drm/bochs/bochs_hw.c +++ b/drivers/gpu/drm/bochs/bochs_hw.c @@ -4,6 +4,7 @@ #include <linux/pci.h> +#include <drm/drm_drv.h> #include <drm/drm_fourcc.h> #include "bochs.h" @@ -156,10 +157,8 @@ int bochs_hw_init(struct drm_device *dev) size = min(size, mem); } - if (pci_request_region(pdev, 0, "bochs-drm") != 0) { - DRM_ERROR("Cannot request framebuffer\n"); - return -EBUSY; - } + if (pci_request_region(pdev, 0, "bochs-drm") != 0) + DRM_WARN("Cannot request framebuffer, boot fb still active?\n"); bochs->fb_map = ioremap(addr, size); if (bochs->fb_map == NULL) { @@ -194,6 +193,8 @@ void bochs_hw_fini(struct drm_device *dev) { struct bochs_device *bochs = dev->dev_private; + /* TODO: shot down existing vram mappings */ + if (bochs->mmio) iounmap(bochs->mmio); if (bochs->ioports) @@ -207,6 +208,11 @@ void bochs_hw_fini(struct drm_device *dev) void bochs_hw_setmode(struct bochs_device *bochs, struct drm_display_mode *mode) { + int idx; + + if (!drm_dev_enter(bochs->dev, &idx)) + return; + bochs->xres = mode->hdisplay; bochs->yres = mode->vdisplay; bochs->bpp = 32; @@ -232,11 +238,18 @@ void bochs_hw_setmode(struct bochs_device *bochs, bochs_dispi_write(bochs, VBE_DISPI_INDEX_ENABLE, VBE_DISPI_ENABLED | VBE_DISPI_LFB_ENABLED); + + drm_dev_exit(idx); } void bochs_hw_setformat(struct bochs_device *bochs, const struct drm_format_info *format) { + int idx; + + if (!drm_dev_enter(bochs->dev, &idx)) + return; + DRM_DEBUG_DRIVER("format %c%c%c%c\n", (format->format >> 0) & 0xff, (format->format >> 8) & 0xff, @@ -256,13 +269,18 @@ void bochs_hw_setformat(struct bochs_device *bochs, __func__, format->format); break; } + + drm_dev_exit(idx); } void bochs_hw_setbase(struct bochs_device *bochs, int x, int y, int stride, u64 addr) { unsigned long offset; - unsigned int vx, vy, vwidth; + unsigned int vx, vy, vwidth, idx; + + if (!drm_dev_enter(bochs->dev, &idx)) + return; bochs->stride = stride; offset = (unsigned long)addr + @@ -277,4 +295,6 @@ void bochs_hw_setbase(struct bochs_device *bochs, bochs_dispi_write(bochs, VBE_DISPI_INDEX_VIRT_WIDTH, vwidth); bochs_dispi_write(bochs, VBE_DISPI_INDEX_X_OFFSET, vx); bochs_dispi_write(bochs, VBE_DISPI_INDEX_Y_OFFSET, vy); + + drm_dev_exit(idx); } diff --git a/drivers/gpu/drm/bochs/bochs_kms.c b/drivers/gpu/drm/bochs/bochs_kms.c index 7410b0ea4623..8066d7d370d5 100644 --- a/drivers/gpu/drm/bochs/bochs_kms.c +++ b/drivers/gpu/drm/bochs/bochs_kms.c @@ -163,6 +163,9 @@ int bochs_kms_init(struct bochs_device *bochs) void bochs_kms_fini(struct bochs_device *bochs) { + if (!bochs->dev->mode_config.num_connector) + return; + drm_atomic_helper_shutdown(bochs->dev); drm_mode_config_cleanup(bochs->dev); } diff --git a/drivers/gpu/drm/bridge/Kconfig b/drivers/gpu/drm/bridge/Kconfig index 8397bf72d2f3..aaed2347ace9 100644 --- a/drivers/gpu/drm/bridge/Kconfig +++ b/drivers/gpu/drm/bridge/Kconfig @@ -27,13 +27,16 @@ config DRM_CDNS_DSI Support Cadence DPI to DSI bridge. This is an internal bridge and is meant to be directly embedded in a SoC. -config DRM_DUMB_VGA_DAC - tristate "Dumb VGA DAC Bridge support" +config DRM_DISPLAY_CONNECTOR + tristate "Display connector support" depends on OF - select DRM_KMS_HELPER help - Support for non-programmable RGB to VGA DAC bridges, such as ADI - ADV7123, TI THS8134 and THS8135 or passive resistor ladder DACs. + Driver for display connectors with support for DDC and hot-plug + detection. Most display controller handle display connectors + internally and don't need this driver, but the DRM subsystem is + moving towards separating connector handling from display controllers + on ARM-based platforms. Saying Y here when this driver is not needed + will not cause any issue. config DRM_LVDS_CODEC tristate "Transparent LVDS encoders and decoders support" @@ -98,6 +101,7 @@ config DRM_SII902X select DRM_KMS_HELPER select REGMAP_I2C select I2C_MUX + select SND_SOC_HDMI_CODEC if SND_SOC ---help--- Silicon Image sii902x bridge chip driver. @@ -109,6 +113,14 @@ config DRM_SII9234 It is an I2C driver, that detects connection of MHL bridge and starts encapsulation of HDMI signal. +config DRM_SIMPLE_BRIDGE + tristate "Simple DRM bridge support" + depends on OF + select DRM_KMS_HELPER + help + Support for non-programmable DRM bridges, such as ADI ADV7123, TI + THS8134 and THS8135 or passive resistor ladder DACs. + config DRM_THINE_THC63LVD1024 tristate "Thine THC63LVD1024 LVDS decoder bridge" depends on OF @@ -133,6 +145,16 @@ config DRM_TOSHIBA_TC358767 ---help--- Toshiba TC358767 eDP bridge chip driver. +config DRM_TOSHIBA_TC358768 + tristate "Toshiba TC358768 MIPI DSI bridge" + depends on OF + select DRM_KMS_HELPER + select REGMAP_I2C + select DRM_PANEL + select DRM_MIPI_DSI + help + Toshiba TC358768AXBG/TC358778XBG DSI bridge chip driver. + config DRM_TI_TFP410 tristate "TI TFP410 DVI/HDMI bridge" depends on OF @@ -150,6 +172,14 @@ config DRM_TI_SN65DSI86 help Texas Instruments SN65DSI86 DSI to eDP Bridge driver +config DRM_TI_TPD12S015 + tristate "TI TPD12S015 HDMI level shifter and ESD protection" + depends on OF + select DRM_KMS_HELPER + help + Texas Instruments TPD12S015 HDMI level shifter and ESD protection + driver. + source "drivers/gpu/drm/bridge/analogix/Kconfig" source "drivers/gpu/drm/bridge/adv7511/Kconfig" diff --git a/drivers/gpu/drm/bridge/Makefile b/drivers/gpu/drm/bridge/Makefile index 1eb5376c5d68..6fb062b5b0f0 100644 --- a/drivers/gpu/drm/bridge/Makefile +++ b/drivers/gpu/drm/bridge/Makefile @@ -1,6 +1,6 @@ # SPDX-License-Identifier: GPL-2.0 obj-$(CONFIG_DRM_CDNS_DSI) += cdns-dsi.o -obj-$(CONFIG_DRM_DUMB_VGA_DAC) += dumb-vga-dac.o +obj-$(CONFIG_DRM_DISPLAY_CONNECTOR) += display-connector.o obj-$(CONFIG_DRM_LVDS_CODEC) += lvds-codec.o obj-$(CONFIG_DRM_MEGACHIPS_STDPXXXX_GE_B850V3_FW) += megachips-stdpxxxx-ge-b850v3-fw.o obj-$(CONFIG_DRM_NXP_PTN3460) += nxp-ptn3460.o @@ -9,12 +9,15 @@ obj-$(CONFIG_DRM_PARADE_PS8640) += parade-ps8640.o obj-$(CONFIG_DRM_SIL_SII8620) += sil-sii8620.o obj-$(CONFIG_DRM_SII902X) += sii902x.o obj-$(CONFIG_DRM_SII9234) += sii9234.o +obj-$(CONFIG_DRM_SIMPLE_BRIDGE) += simple-bridge.o obj-$(CONFIG_DRM_THINE_THC63LVD1024) += thc63lvd1024.o obj-$(CONFIG_DRM_TOSHIBA_TC358764) += tc358764.o obj-$(CONFIG_DRM_TOSHIBA_TC358767) += tc358767.o +obj-$(CONFIG_DRM_TOSHIBA_TC358768) += tc358768.o obj-$(CONFIG_DRM_I2C_ADV7511) += adv7511/ obj-$(CONFIG_DRM_TI_SN65DSI86) += ti-sn65dsi86.o obj-$(CONFIG_DRM_TI_TFP410) += ti-tfp410.o +obj-$(CONFIG_DRM_TI_TPD12S015) += ti-tpd12s015.o obj-y += analogix/ obj-y += synopsys/ diff --git a/drivers/gpu/drm/bridge/adv7511/Kconfig b/drivers/gpu/drm/bridge/adv7511/Kconfig index 8a56ff81f4fb..47d4eb9e845d 100644 --- a/drivers/gpu/drm/bridge/adv7511/Kconfig +++ b/drivers/gpu/drm/bridge/adv7511/Kconfig @@ -4,8 +4,9 @@ config DRM_I2C_ADV7511 depends on OF select DRM_KMS_HELPER select REGMAP_I2C + select DRM_MIPI_DSI help - Support for the Analog Device ADV7511(W) and ADV7513 HDMI encoders. + Support for the Analog Device ADV7511(W)/13/33/35 HDMI encoders. config DRM_I2C_ADV7511_AUDIO bool "ADV7511 HDMI Audio driver" @@ -15,16 +16,8 @@ config DRM_I2C_ADV7511_AUDIO Support the ADV7511 HDMI Audio interface. This is used in conjunction with the AV7511 HDMI driver. -config DRM_I2C_ADV7533 - bool "ADV7533 encoder" - depends on DRM_I2C_ADV7511 - select DRM_MIPI_DSI - default y - help - Support for the Analog Devices ADV7533 DSI to HDMI encoder. - config DRM_I2C_ADV7511_CEC - bool "ADV7511/33 HDMI CEC driver" + bool "ADV7511/33/35 HDMI CEC driver" depends on DRM_I2C_ADV7511 select CEC_CORE default y diff --git a/drivers/gpu/drm/bridge/adv7511/Makefile b/drivers/gpu/drm/bridge/adv7511/Makefile index b46ebeb35fd4..d8ceb534b51f 100644 --- a/drivers/gpu/drm/bridge/adv7511/Makefile +++ b/drivers/gpu/drm/bridge/adv7511/Makefile @@ -1,6 +1,5 @@ # SPDX-License-Identifier: GPL-2.0-only -adv7511-y := adv7511_drv.o +adv7511-y := adv7511_drv.o adv7533.o adv7511-$(CONFIG_DRM_I2C_ADV7511_AUDIO) += adv7511_audio.o adv7511-$(CONFIG_DRM_I2C_ADV7511_CEC) += adv7511_cec.o -adv7511-$(CONFIG_DRM_I2C_ADV7533) += adv7533.o obj-$(CONFIG_DRM_I2C_ADV7511) += adv7511.o diff --git a/drivers/gpu/drm/bridge/adv7511/adv7511.h b/drivers/gpu/drm/bridge/adv7511/adv7511.h index 52b2adfdc877..a9bb734366ae 100644 --- a/drivers/gpu/drm/bridge/adv7511/adv7511.h +++ b/drivers/gpu/drm/bridge/adv7511/adv7511.h @@ -320,6 +320,7 @@ struct adv7511_video_config { enum adv7511_type { ADV7511, ADV7533, + ADV7535, }; #define ADV7511_MAX_ADDRS 3 @@ -393,7 +394,6 @@ static inline int adv7511_cec_init(struct device *dev, struct adv7511 *adv7511) } #endif -#ifdef CONFIG_DRM_I2C_ADV7533 void adv7533_dsi_power_on(struct adv7511 *adv); void adv7533_dsi_power_off(struct adv7511 *adv); void adv7533_mode_set(struct adv7511 *adv, const struct drm_display_mode *mode); @@ -402,44 +402,6 @@ int adv7533_patch_cec_registers(struct adv7511 *adv); int adv7533_attach_dsi(struct adv7511 *adv); void adv7533_detach_dsi(struct adv7511 *adv); int adv7533_parse_dt(struct device_node *np, struct adv7511 *adv); -#else -static inline void adv7533_dsi_power_on(struct adv7511 *adv) -{ -} - -static inline void adv7533_dsi_power_off(struct adv7511 *adv) -{ -} - -static inline void adv7533_mode_set(struct adv7511 *adv, - const struct drm_display_mode *mode) -{ -} - -static inline int adv7533_patch_registers(struct adv7511 *adv) -{ - return -ENODEV; -} - -static inline int adv7533_patch_cec_registers(struct adv7511 *adv) -{ - return -ENODEV; -} - -static inline int adv7533_attach_dsi(struct adv7511 *adv) -{ - return -ENODEV; -} - -static inline void adv7533_detach_dsi(struct adv7511 *adv) -{ -} - -static inline int adv7533_parse_dt(struct device_node *np, struct adv7511 *adv) -{ - return -ENODEV; -} -#endif #ifdef CONFIG_DRM_I2C_ADV7511_AUDIO int adv7511_audio_init(struct device *dev, struct adv7511 *adv7511); diff --git a/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c b/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c index 9e13e466e72c..87b58c1acff4 100644 --- a/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c +++ b/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c @@ -367,7 +367,7 @@ static void adv7511_power_on(struct adv7511 *adv7511) */ regcache_sync(adv7511->regmap); - if (adv7511->type == ADV7533) + if (adv7511->type == ADV7533 || adv7511->type == ADV7535) adv7533_dsi_power_on(adv7511); adv7511->powered = true; } @@ -387,7 +387,7 @@ static void __adv7511_power_off(struct adv7511 *adv7511) static void adv7511_power_off(struct adv7511 *adv7511) { __adv7511_power_off(adv7511); - if (adv7511->type == ADV7533) + if (adv7511->type == ADV7533 || adv7511->type == ADV7535) adv7533_dsi_power_off(adv7511); adv7511->powered = false; } @@ -761,7 +761,7 @@ static void adv7511_mode_set(struct adv7511 *adv7511, regmap_update_bits(adv7511->regmap, 0x17, 0x60, (vsync_polarity << 6) | (hsync_polarity << 5)); - if (adv7511->type == ADV7533) + if (adv7511->type == ADV7533 || adv7511->type == ADV7535) adv7533_mode_set(adv7511, adj_mode); drm_mode_copy(&adv7511->curr_mode, adj_mode); @@ -847,11 +847,17 @@ static void adv7511_bridge_mode_set(struct drm_bridge *bridge, adv7511_mode_set(adv, mode, adj_mode); } -static int adv7511_bridge_attach(struct drm_bridge *bridge) +static int adv7511_bridge_attach(struct drm_bridge *bridge, + enum drm_bridge_attach_flags flags) { struct adv7511 *adv = bridge_to_adv7511(bridge); int ret; + if (flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR) { + DRM_ERROR("Fix bridge driver to make connector optional!"); + return -EINVAL; + } + if (!bridge->encoder) { DRM_ERROR("Parent encoder object not found"); return -ENODEV; @@ -874,7 +880,7 @@ static int adv7511_bridge_attach(struct drm_bridge *bridge) &adv7511_connector_helper_funcs); drm_connector_attach_encoder(&adv->connector, bridge->encoder); - if (adv->type == ADV7533) + if (adv->type == ADV7533 || adv->type == ADV7535) ret = adv7533_attach_dsi(adv); if (adv->i2c_main->irq) @@ -952,7 +958,7 @@ static bool adv7511_cec_register_volatile(struct device *dev, unsigned int reg) struct i2c_client *i2c = to_i2c_client(dev); struct adv7511 *adv7511 = i2c_get_clientdata(i2c); - if (adv7511->type == ADV7533) + if (adv7511->type == ADV7533 || adv7511->type == ADV7535) reg -= ADV7533_REG_CEC_OFFSET; switch (reg) { @@ -994,7 +1000,7 @@ static int adv7511_init_cec_regmap(struct adv7511 *adv) goto err; } - if (adv->type == ADV7533) { + if (adv->type == ADV7533 || adv->type == ADV7535) { ret = adv7533_patch_cec_registers(adv); if (ret) goto err; @@ -1242,7 +1248,7 @@ static int adv7511_remove(struct i2c_client *i2c) { struct adv7511 *adv7511 = i2c_get_clientdata(i2c); - if (adv7511->type == ADV7533) + if (adv7511->type == ADV7533 || adv7511->type == ADV7535) adv7533_detach_dsi(adv7511); i2c_unregister_device(adv7511->i2c_cec); if (adv7511->cec_clk) @@ -1266,9 +1272,8 @@ static const struct i2c_device_id adv7511_i2c_ids[] = { { "adv7511", ADV7511 }, { "adv7511w", ADV7511 }, { "adv7513", ADV7511 }, -#ifdef CONFIG_DRM_I2C_ADV7533 { "adv7533", ADV7533 }, -#endif + { "adv7535", ADV7535 }, { } }; MODULE_DEVICE_TABLE(i2c, adv7511_i2c_ids); @@ -1277,9 +1282,8 @@ static const struct of_device_id adv7511_of_ids[] = { { .compatible = "adi,adv7511", .data = (void *)ADV7511 }, { .compatible = "adi,adv7511w", .data = (void *)ADV7511 }, { .compatible = "adi,adv7513", .data = (void *)ADV7511 }, -#ifdef CONFIG_DRM_I2C_ADV7533 { .compatible = "adi,adv7533", .data = (void *)ADV7533 }, -#endif + { .compatible = "adi,adv7535", .data = (void *)ADV7535 }, { } }; MODULE_DEVICE_TABLE(of, adv7511_of_ids); diff --git a/drivers/gpu/drm/bridge/analogix/analogix-anx6345.c b/drivers/gpu/drm/bridge/analogix/analogix-anx6345.c index 56f55c53abfd..2bc6e4f85171 100644 --- a/drivers/gpu/drm/bridge/analogix/analogix-anx6345.c +++ b/drivers/gpu/drm/bridge/analogix/analogix-anx6345.c @@ -210,8 +210,7 @@ static int anx6345_dp_link_training(struct anx6345 *anx6345) if (err) return err; - dpcd[0] = drm_dp_max_link_rate(anx6345->dpcd); - dpcd[0] = drm_dp_link_rate_to_bw_code(dpcd[0]); + dpcd[0] = dp_bw; err = regmap_write(anx6345->map[I2C_IDX_DPTX], SP_DP_MAIN_LINK_BW_SET_REG, dpcd[0]); if (err) @@ -520,11 +519,17 @@ static const struct drm_connector_funcs anx6345_connector_funcs = { .atomic_destroy_state = drm_atomic_helper_connector_destroy_state, }; -static int anx6345_bridge_attach(struct drm_bridge *bridge) +static int anx6345_bridge_attach(struct drm_bridge *bridge, + enum drm_bridge_attach_flags flags) { struct anx6345 *anx6345 = bridge_to_anx6345(bridge); int err; + if (flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR) { + DRM_ERROR("Fix bridge driver to make connector optional!"); + return -EINVAL; + } + if (!bridge->encoder) { DRM_ERROR("Parent encoder object not found"); return -ENODEV; @@ -712,16 +717,20 @@ static int anx6345_i2c_probe(struct i2c_client *client, DRM_DEBUG("No panel found\n"); /* 1.2V digital core power regulator */ - anx6345->dvdd12 = devm_regulator_get(dev, "dvdd12-supply"); + anx6345->dvdd12 = devm_regulator_get(dev, "dvdd12"); if (IS_ERR(anx6345->dvdd12)) { - DRM_ERROR("dvdd12-supply not found\n"); + if (PTR_ERR(anx6345->dvdd12) != -EPROBE_DEFER) + DRM_ERROR("Failed to get dvdd12 supply (%ld)\n", + PTR_ERR(anx6345->dvdd12)); return PTR_ERR(anx6345->dvdd12); } /* 2.5V digital core power regulator */ - anx6345->dvdd25 = devm_regulator_get(dev, "dvdd25-supply"); + anx6345->dvdd25 = devm_regulator_get(dev, "dvdd25"); if (IS_ERR(anx6345->dvdd25)) { - DRM_ERROR("dvdd25-supply not found\n"); + if (PTR_ERR(anx6345->dvdd25) != -EPROBE_DEFER) + DRM_ERROR("Failed to get dvdd25 supply (%ld)\n", + PTR_ERR(anx6345->dvdd25)); return PTR_ERR(anx6345->dvdd25); } diff --git a/drivers/gpu/drm/bridge/analogix/analogix-anx78xx.c b/drivers/gpu/drm/bridge/analogix/analogix-anx78xx.c index 41867be03751..0d5a5ad0c9ee 100644 --- a/drivers/gpu/drm/bridge/analogix/analogix-anx78xx.c +++ b/drivers/gpu/drm/bridge/analogix/analogix-anx78xx.c @@ -722,10 +722,9 @@ static int anx78xx_dp_link_training(struct anx78xx *anx78xx) if (err) return err; - dpcd[0] = drm_dp_max_link_rate(anx78xx->dpcd); - dpcd[0] = drm_dp_link_rate_to_bw_code(dpcd[0]); err = regmap_write(anx78xx->map[I2C_IDX_TX_P0], - SP_DP_MAIN_LINK_BW_SET_REG, dpcd[0]); + SP_DP_MAIN_LINK_BW_SET_REG, + anx78xx->dpcd[DP_MAX_LINK_RATE]); if (err) return err; @@ -887,11 +886,17 @@ static const struct drm_connector_funcs anx78xx_connector_funcs = { .atomic_destroy_state = drm_atomic_helper_connector_destroy_state, }; -static int anx78xx_bridge_attach(struct drm_bridge *bridge) +static int anx78xx_bridge_attach(struct drm_bridge *bridge, + enum drm_bridge_attach_flags flags) { struct anx78xx *anx78xx = bridge_to_anx78xx(bridge); int err; + if (flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR) { + DRM_ERROR("Fix bridge driver to make connector optional!"); + return -EINVAL; + } + if (!bridge->encoder) { DRM_ERROR("Parent encoder object not found"); return -ENODEV; diff --git a/drivers/gpu/drm/bridge/analogix/analogix_dp_core.c b/drivers/gpu/drm/bridge/analogix/analogix_dp_core.c index dfb59a5fefea..76736fb8ed94 100644 --- a/drivers/gpu/drm/bridge/analogix/analogix_dp_core.c +++ b/drivers/gpu/drm/bridge/analogix/analogix_dp_core.c @@ -1216,13 +1216,19 @@ static const struct drm_connector_funcs analogix_dp_connector_funcs = { .atomic_destroy_state = drm_atomic_helper_connector_destroy_state, }; -static int analogix_dp_bridge_attach(struct drm_bridge *bridge) +static int analogix_dp_bridge_attach(struct drm_bridge *bridge, + enum drm_bridge_attach_flags flags) { struct analogix_dp_device *dp = bridge->driver_private; struct drm_encoder *encoder = dp->encoder; struct drm_connector *connector = NULL; int ret = 0; + if (flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR) { + DRM_ERROR("Fix bridge driver to make connector optional!"); + return -EINVAL; + } + if (!bridge->encoder) { DRM_ERROR("Parent encoder object not found"); return -ENODEV; @@ -1598,7 +1604,7 @@ static int analogix_dp_create_bridge(struct drm_device *drm_dev, bridge->driver_private = dp; bridge->funcs = &analogix_dp_bridge_funcs; - ret = drm_bridge_attach(dp->encoder, bridge, NULL); + ret = drm_bridge_attach(dp->encoder, bridge, NULL, 0); if (ret) { DRM_ERROR("failed to attach drm bridge\n"); return -EINVAL; @@ -1646,8 +1652,7 @@ static ssize_t analogix_dpaux_transfer(struct drm_dp_aux *aux, } struct analogix_dp_device * -analogix_dp_bind(struct device *dev, struct drm_device *drm_dev, - struct analogix_dp_plat_data *plat_data) +analogix_dp_probe(struct device *dev, struct analogix_dp_plat_data *plat_data) { struct platform_device *pdev = to_platform_device(dev); struct analogix_dp_device *dp; @@ -1750,22 +1755,30 @@ analogix_dp_bind(struct device *dev, struct drm_device *drm_dev, irq_flags, "analogix-dp", dp); if (ret) { dev_err(&pdev->dev, "failed to request irq\n"); - goto err_disable_pm_runtime; + return ERR_PTR(ret); } disable_irq(dp->irq); + return dp; +} +EXPORT_SYMBOL_GPL(analogix_dp_probe); + +int analogix_dp_bind(struct analogix_dp_device *dp, struct drm_device *drm_dev) +{ + int ret; + dp->drm_dev = drm_dev; dp->encoder = dp->plat_data->encoder; dp->aux.name = "DP-AUX"; dp->aux.transfer = analogix_dpaux_transfer; - dp->aux.dev = &pdev->dev; + dp->aux.dev = dp->dev; ret = drm_dp_aux_register(&dp->aux); if (ret) - return ERR_PTR(ret); + return ret; - pm_runtime_enable(dev); + pm_runtime_enable(dp->dev); ret = analogix_dp_create_bridge(drm_dev, dp); if (ret) { @@ -1773,13 +1786,12 @@ analogix_dp_bind(struct device *dev, struct drm_device *drm_dev, goto err_disable_pm_runtime; } - return dp; + return 0; err_disable_pm_runtime: + pm_runtime_disable(dp->dev); - pm_runtime_disable(dev); - - return ERR_PTR(ret); + return ret; } EXPORT_SYMBOL_GPL(analogix_dp_bind); @@ -1796,10 +1808,15 @@ void analogix_dp_unbind(struct analogix_dp_device *dp) drm_dp_aux_unregister(&dp->aux); pm_runtime_disable(dp->dev); - clk_disable_unprepare(dp->clock); } EXPORT_SYMBOL_GPL(analogix_dp_unbind); +void analogix_dp_remove(struct analogix_dp_device *dp) +{ + clk_disable_unprepare(dp->clock); +} +EXPORT_SYMBOL_GPL(analogix_dp_remove); + #ifdef CONFIG_PM int analogix_dp_suspend(struct analogix_dp_device *dp) { diff --git a/drivers/gpu/drm/bridge/cdns-dsi.c b/drivers/gpu/drm/bridge/cdns-dsi.c index b7c97f060241..69c3892caee5 100644 --- a/drivers/gpu/drm/bridge/cdns-dsi.c +++ b/drivers/gpu/drm/bridge/cdns-dsi.c @@ -644,7 +644,8 @@ static int cdns_dsi_check_conf(struct cdns_dsi *dsi, return 0; } -static int cdns_dsi_bridge_attach(struct drm_bridge *bridge) +static int cdns_dsi_bridge_attach(struct drm_bridge *bridge, + enum drm_bridge_attach_flags flags) { struct cdns_dsi_input *input = bridge_to_cdns_dsi_input(bridge); struct cdns_dsi *dsi = input_to_dsi(input); @@ -656,7 +657,8 @@ static int cdns_dsi_bridge_attach(struct drm_bridge *bridge) return -ENOTSUPP; } - return drm_bridge_attach(bridge->encoder, output->bridge, bridge); + return drm_bridge_attach(bridge->encoder, output->bridge, bridge, + flags); } static enum drm_mode_status diff --git a/drivers/gpu/drm/bridge/display-connector.c b/drivers/gpu/drm/bridge/display-connector.c new file mode 100644 index 000000000000..4d278573cdb9 --- /dev/null +++ b/drivers/gpu/drm/bridge/display-connector.c @@ -0,0 +1,295 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (C) 2019 Laurent Pinchart <laurent.pinchart@ideasonboard.com> + */ + +#include <linux/gpio/consumer.h> +#include <linux/i2c.h> +#include <linux/interrupt.h> +#include <linux/module.h> +#include <linux/mutex.h> +#include <linux/of.h> +#include <linux/of_device.h> +#include <linux/platform_device.h> + +#include <drm/drm_bridge.h> +#include <drm/drm_edid.h> + +struct display_connector { + struct drm_bridge bridge; + + struct gpio_desc *hpd_gpio; + int hpd_irq; +}; + +static inline struct display_connector * +to_display_connector(struct drm_bridge *bridge) +{ + return container_of(bridge, struct display_connector, bridge); +} + +static int display_connector_attach(struct drm_bridge *bridge, + enum drm_bridge_attach_flags flags) +{ + return flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR ? 0 : -EINVAL; +} + +static enum drm_connector_status +display_connector_detect(struct drm_bridge *bridge) +{ + struct display_connector *conn = to_display_connector(bridge); + + if (conn->hpd_gpio) { + if (gpiod_get_value_cansleep(conn->hpd_gpio)) + return connector_status_connected; + else + return connector_status_disconnected; + } + + if (conn->bridge.ddc && drm_probe_ddc(conn->bridge.ddc)) + return connector_status_connected; + + switch (conn->bridge.type) { + case DRM_MODE_CONNECTOR_DVIA: + case DRM_MODE_CONNECTOR_DVID: + case DRM_MODE_CONNECTOR_DVII: + case DRM_MODE_CONNECTOR_HDMIA: + case DRM_MODE_CONNECTOR_HDMIB: + /* + * For DVI and HDMI connectors a DDC probe failure indicates + * that no cable is connected. + */ + return connector_status_disconnected; + + case DRM_MODE_CONNECTOR_Composite: + case DRM_MODE_CONNECTOR_SVIDEO: + case DRM_MODE_CONNECTOR_VGA: + default: + /* + * Composite and S-Video connectors have no other detection + * mean than the HPD GPIO. For VGA connectors, even if we have + * an I2C bus, we can't assume that the cable is disconnected + * if drm_probe_ddc fails, as some cables don't wire the DDC + * pins. + */ + return connector_status_unknown; + } +} + +static struct edid *display_connector_get_edid(struct drm_bridge *bridge, + struct drm_connector *connector) +{ + struct display_connector *conn = to_display_connector(bridge); + + return drm_get_edid(connector, conn->bridge.ddc); +} + +static const struct drm_bridge_funcs display_connector_bridge_funcs = { + .attach = display_connector_attach, + .detect = display_connector_detect, + .get_edid = display_connector_get_edid, +}; + +static irqreturn_t display_connector_hpd_irq(int irq, void *arg) +{ + struct display_connector *conn = arg; + struct drm_bridge *bridge = &conn->bridge; + + drm_bridge_hpd_notify(bridge, display_connector_detect(bridge)); + + return IRQ_HANDLED; +} + +static int display_connector_probe(struct platform_device *pdev) +{ + struct display_connector *conn; + unsigned int type; + const char *label; + int ret; + + conn = devm_kzalloc(&pdev->dev, sizeof(*conn), GFP_KERNEL); + if (!conn) + return -ENOMEM; + + platform_set_drvdata(pdev, conn); + + type = (uintptr_t)of_device_get_match_data(&pdev->dev); + + /* Get the exact connector type. */ + switch (type) { + case DRM_MODE_CONNECTOR_DVII: { + bool analog, digital; + + analog = of_property_read_bool(pdev->dev.of_node, "analog"); + digital = of_property_read_bool(pdev->dev.of_node, "digital"); + if (analog && !digital) { + conn->bridge.type = DRM_MODE_CONNECTOR_DVIA; + } else if (!analog && digital) { + conn->bridge.type = DRM_MODE_CONNECTOR_DVID; + } else if (analog && digital) { + conn->bridge.type = DRM_MODE_CONNECTOR_DVII; + } else { + dev_err(&pdev->dev, "DVI connector with no type\n"); + return -EINVAL; + } + break; + } + + case DRM_MODE_CONNECTOR_HDMIA: { + const char *hdmi_type; + + ret = of_property_read_string(pdev->dev.of_node, "type", + &hdmi_type); + if (ret < 0) { + dev_err(&pdev->dev, "HDMI connector with no type\n"); + return -EINVAL; + } + + if (!strcmp(hdmi_type, "a") || !strcmp(hdmi_type, "c") || + !strcmp(hdmi_type, "d") || !strcmp(hdmi_type, "e")) { + conn->bridge.type = DRM_MODE_CONNECTOR_HDMIA; + } else if (!strcmp(hdmi_type, "b")) { + conn->bridge.type = DRM_MODE_CONNECTOR_HDMIB; + } else { + dev_err(&pdev->dev, + "Unsupported HDMI connector type '%s'\n", + hdmi_type); + return -EINVAL; + } + + break; + } + + default: + conn->bridge.type = type; + break; + } + + /* All the supported connector types support interlaced modes. */ + conn->bridge.interlace_allowed = true; + + /* Get the optional connector label. */ + of_property_read_string(pdev->dev.of_node, "label", &label); + + /* + * Get the HPD GPIO for DVI and HDMI connectors. If the GPIO can provide + * edge interrupts, register an interrupt handler. + */ + if (type == DRM_MODE_CONNECTOR_DVII || + type == DRM_MODE_CONNECTOR_HDMIA) { + conn->hpd_gpio = devm_gpiod_get_optional(&pdev->dev, "hpd", + GPIOD_IN); + if (IS_ERR(conn->hpd_gpio)) { + if (PTR_ERR(conn->hpd_gpio) != -EPROBE_DEFER) + dev_err(&pdev->dev, + "Unable to retrieve HPD GPIO\n"); + return PTR_ERR(conn->hpd_gpio); + } + + conn->hpd_irq = gpiod_to_irq(conn->hpd_gpio); + } else { + conn->hpd_irq = -EINVAL; + } + + if (conn->hpd_irq >= 0) { + ret = devm_request_threaded_irq(&pdev->dev, conn->hpd_irq, + NULL, display_connector_hpd_irq, + IRQF_TRIGGER_RISING | + IRQF_TRIGGER_FALLING | + IRQF_ONESHOT, + "HPD", conn); + if (ret) { + dev_info(&pdev->dev, + "Failed to request HPD edge interrupt, falling back to polling\n"); + conn->hpd_irq = -EINVAL; + } + } + + /* Retrieve the DDC I2C adapter for DVI, HDMI and VGA connectors. */ + if (type == DRM_MODE_CONNECTOR_DVII || + type == DRM_MODE_CONNECTOR_HDMIA || + type == DRM_MODE_CONNECTOR_VGA) { + struct device_node *phandle; + + phandle = of_parse_phandle(pdev->dev.of_node, "ddc-i2c-bus", 0); + if (phandle) { + conn->bridge.ddc = of_get_i2c_adapter_by_node(phandle); + of_node_put(phandle); + if (!conn->bridge.ddc) + return -EPROBE_DEFER; + } else { + dev_dbg(&pdev->dev, + "No I2C bus specified, disabling EDID readout\n"); + } + } + + conn->bridge.funcs = &display_connector_bridge_funcs; + conn->bridge.of_node = pdev->dev.of_node; + + if (conn->bridge.ddc) + conn->bridge.ops |= DRM_BRIDGE_OP_EDID + | DRM_BRIDGE_OP_DETECT; + if (conn->hpd_gpio) + conn->bridge.ops |= DRM_BRIDGE_OP_DETECT; + if (conn->hpd_irq >= 0) + conn->bridge.ops |= DRM_BRIDGE_OP_HPD; + + dev_dbg(&pdev->dev, + "Found %s display connector '%s' %s DDC bus and %s HPD GPIO (ops 0x%x)\n", + drm_get_connector_type_name(conn->bridge.type), + label ? label : "<unlabelled>", + conn->bridge.ddc ? "with" : "without", + conn->hpd_gpio ? "with" : "without", + conn->bridge.ops); + + drm_bridge_add(&conn->bridge); + + return 0; +} + +static int display_connector_remove(struct platform_device *pdev) +{ + struct display_connector *conn = platform_get_drvdata(pdev); + + drm_bridge_remove(&conn->bridge); + + if (!IS_ERR(conn->bridge.ddc)) + i2c_put_adapter(conn->bridge.ddc); + + return 0; +} + +static const struct of_device_id display_connector_match[] = { + { + .compatible = "composite-video-connector", + .data = (void *)DRM_MODE_CONNECTOR_Composite, + }, { + .compatible = "dvi-connector", + .data = (void *)DRM_MODE_CONNECTOR_DVII, + }, { + .compatible = "hdmi-connector", + .data = (void *)DRM_MODE_CONNECTOR_HDMIA, + }, { + .compatible = "svideo-connector", + .data = (void *)DRM_MODE_CONNECTOR_SVIDEO, + }, { + .compatible = "vga-connector", + .data = (void *)DRM_MODE_CONNECTOR_VGA, + }, + {}, +}; +MODULE_DEVICE_TABLE(of, display_connector_match); + +static struct platform_driver display_connector_driver = { + .probe = display_connector_probe, + .remove = display_connector_remove, + .driver = { + .name = "display-connector", + .of_match_table = display_connector_match, + }, +}; +module_platform_driver(display_connector_driver); + +MODULE_AUTHOR("Laurent Pinchart <laurent.pinchart@ideasonboard.com>"); +MODULE_DESCRIPTION("Display connector driver"); +MODULE_LICENSE("GPL"); diff --git a/drivers/gpu/drm/bridge/dumb-vga-dac.c b/drivers/gpu/drm/bridge/dumb-vga-dac.c deleted file mode 100644 index cc33dc411b9e..000000000000 --- a/drivers/gpu/drm/bridge/dumb-vga-dac.c +++ /dev/null @@ -1,300 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-or-later -/* - * Copyright (C) 2015-2016 Free Electrons - * Copyright (C) 2015-2016 NextThing Co - * - * Maxime Ripard <maxime.ripard@free-electrons.com> - */ - -#include <linux/module.h> -#include <linux/of_device.h> -#include <linux/of_graph.h> -#include <linux/regulator/consumer.h> - -#include <drm/drm_atomic_helper.h> -#include <drm/drm_bridge.h> -#include <drm/drm_crtc.h> -#include <drm/drm_print.h> -#include <drm/drm_probe_helper.h> - -struct dumb_vga { - struct drm_bridge bridge; - struct drm_connector connector; - - struct i2c_adapter *ddc; - struct regulator *vdd; -}; - -static inline struct dumb_vga * -drm_bridge_to_dumb_vga(struct drm_bridge *bridge) -{ - return container_of(bridge, struct dumb_vga, bridge); -} - -static inline struct dumb_vga * -drm_connector_to_dumb_vga(struct drm_connector *connector) -{ - return container_of(connector, struct dumb_vga, connector); -} - -static int dumb_vga_get_modes(struct drm_connector *connector) -{ - struct dumb_vga *vga = drm_connector_to_dumb_vga(connector); - struct edid *edid; - int ret; - - if (!vga->ddc) - goto fallback; - - edid = drm_get_edid(connector, vga->ddc); - if (!edid) { - DRM_INFO("EDID readout failed, falling back to standard modes\n"); - goto fallback; - } - - drm_connector_update_edid_property(connector, edid); - ret = drm_add_edid_modes(connector, edid); - kfree(edid); - return ret; - -fallback: - /* - * In case we cannot retrieve the EDIDs (broken or missing i2c - * bus), fallback on the XGA standards - */ - ret = drm_add_modes_noedid(connector, 1920, 1200); - - /* And prefer a mode pretty much anyone can handle */ - drm_set_preferred_mode(connector, 1024, 768); - - return ret; -} - -static const struct drm_connector_helper_funcs dumb_vga_con_helper_funcs = { - .get_modes = dumb_vga_get_modes, -}; - -static enum drm_connector_status -dumb_vga_connector_detect(struct drm_connector *connector, bool force) -{ - struct dumb_vga *vga = drm_connector_to_dumb_vga(connector); - - /* - * Even if we have an I2C bus, we can't assume that the cable - * is disconnected if drm_probe_ddc fails. Some cables don't - * wire the DDC pins, or the I2C bus might not be working at - * all. - */ - if (vga->ddc && drm_probe_ddc(vga->ddc)) - return connector_status_connected; - - return connector_status_unknown; -} - -static const struct drm_connector_funcs dumb_vga_con_funcs = { - .detect = dumb_vga_connector_detect, - .fill_modes = drm_helper_probe_single_connector_modes, - .destroy = drm_connector_cleanup, - .reset = drm_atomic_helper_connector_reset, - .atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state, - .atomic_destroy_state = drm_atomic_helper_connector_destroy_state, -}; - -static int dumb_vga_attach(struct drm_bridge *bridge) -{ - struct dumb_vga *vga = drm_bridge_to_dumb_vga(bridge); - int ret; - - if (!bridge->encoder) { - DRM_ERROR("Missing encoder\n"); - return -ENODEV; - } - - drm_connector_helper_add(&vga->connector, - &dumb_vga_con_helper_funcs); - ret = drm_connector_init_with_ddc(bridge->dev, &vga->connector, - &dumb_vga_con_funcs, - DRM_MODE_CONNECTOR_VGA, - vga->ddc); - if (ret) { - DRM_ERROR("Failed to initialize connector\n"); - return ret; - } - - drm_connector_attach_encoder(&vga->connector, - bridge->encoder); - - return 0; -} - -static void dumb_vga_enable(struct drm_bridge *bridge) -{ - struct dumb_vga *vga = drm_bridge_to_dumb_vga(bridge); - int ret = 0; - - if (vga->vdd) - ret = regulator_enable(vga->vdd); - - if (ret) - DRM_ERROR("Failed to enable vdd regulator: %d\n", ret); -} - -static void dumb_vga_disable(struct drm_bridge *bridge) -{ - struct dumb_vga *vga = drm_bridge_to_dumb_vga(bridge); - - if (vga->vdd) - regulator_disable(vga->vdd); -} - -static const struct drm_bridge_funcs dumb_vga_bridge_funcs = { - .attach = dumb_vga_attach, - .enable = dumb_vga_enable, - .disable = dumb_vga_disable, -}; - -static struct i2c_adapter *dumb_vga_retrieve_ddc(struct device *dev) -{ - struct device_node *phandle, *remote; - struct i2c_adapter *ddc; - - remote = of_graph_get_remote_node(dev->of_node, 1, -1); - if (!remote) - return ERR_PTR(-EINVAL); - - phandle = of_parse_phandle(remote, "ddc-i2c-bus", 0); - of_node_put(remote); - if (!phandle) - return ERR_PTR(-ENODEV); - - ddc = of_get_i2c_adapter_by_node(phandle); - of_node_put(phandle); - if (!ddc) - return ERR_PTR(-EPROBE_DEFER); - - return ddc; -} - -static int dumb_vga_probe(struct platform_device *pdev) -{ - struct dumb_vga *vga; - - vga = devm_kzalloc(&pdev->dev, sizeof(*vga), GFP_KERNEL); - if (!vga) - return -ENOMEM; - platform_set_drvdata(pdev, vga); - - vga->vdd = devm_regulator_get_optional(&pdev->dev, "vdd"); - if (IS_ERR(vga->vdd)) { - int ret = PTR_ERR(vga->vdd); - if (ret == -EPROBE_DEFER) - return -EPROBE_DEFER; - vga->vdd = NULL; - dev_dbg(&pdev->dev, "No vdd regulator found: %d\n", ret); - } - - vga->ddc = dumb_vga_retrieve_ddc(&pdev->dev); - if (IS_ERR(vga->ddc)) { - if (PTR_ERR(vga->ddc) == -ENODEV) { - dev_dbg(&pdev->dev, - "No i2c bus specified. Disabling EDID readout\n"); - vga->ddc = NULL; - } else { - dev_err(&pdev->dev, "Couldn't retrieve i2c bus\n"); - return PTR_ERR(vga->ddc); - } - } - - vga->bridge.funcs = &dumb_vga_bridge_funcs; - vga->bridge.of_node = pdev->dev.of_node; - vga->bridge.timings = of_device_get_match_data(&pdev->dev); - - drm_bridge_add(&vga->bridge); - - return 0; -} - -static int dumb_vga_remove(struct platform_device *pdev) -{ - struct dumb_vga *vga = platform_get_drvdata(pdev); - - drm_bridge_remove(&vga->bridge); - - if (vga->ddc) - i2c_put_adapter(vga->ddc); - - return 0; -} - -/* - * We assume the ADV7123 DAC is the "default" for historical reasons - * Information taken from the ADV7123 datasheet, revision D. - * NOTE: the ADV7123EP seems to have other timings and need a new timings - * set if used. - */ -static const struct drm_bridge_timings default_dac_timings = { - /* Timing specifications, datasheet page 7 */ - .input_bus_flags = DRM_BUS_FLAG_PIXDATA_SAMPLE_POSEDGE, - .setup_time_ps = 500, - .hold_time_ps = 1500, -}; - -/* - * Information taken from the THS8134, THS8134A, THS8134B datasheet named - * "SLVS205D", dated May 1990, revised March 2000. - */ -static const struct drm_bridge_timings ti_ths8134_dac_timings = { - /* From timing diagram, datasheet page 9 */ - .input_bus_flags = DRM_BUS_FLAG_PIXDATA_SAMPLE_POSEDGE, - /* From datasheet, page 12 */ - .setup_time_ps = 3000, - /* I guess this means latched input */ - .hold_time_ps = 0, -}; - -/* - * Information taken from the THS8135 datasheet named "SLAS343B", dated - * May 2001, revised April 2013. - */ -static const struct drm_bridge_timings ti_ths8135_dac_timings = { - /* From timing diagram, datasheet page 14 */ - .input_bus_flags = DRM_BUS_FLAG_PIXDATA_SAMPLE_POSEDGE, - /* From datasheet, page 16 */ - .setup_time_ps = 2000, - .hold_time_ps = 500, -}; - -static const struct of_device_id dumb_vga_match[] = { - { - .compatible = "dumb-vga-dac", - .data = NULL, - }, - { - .compatible = "adi,adv7123", - .data = &default_dac_timings, - }, - { - .compatible = "ti,ths8135", - .data = &ti_ths8135_dac_timings, - }, - { - .compatible = "ti,ths8134", - .data = &ti_ths8134_dac_timings, - }, - {}, -}; -MODULE_DEVICE_TABLE(of, dumb_vga_match); - -static struct platform_driver dumb_vga_driver = { - .probe = dumb_vga_probe, - .remove = dumb_vga_remove, - .driver = { - .name = "dumb-vga-dac", - .of_match_table = dumb_vga_match, - }, -}; -module_platform_driver(dumb_vga_driver); - -MODULE_AUTHOR("Maxime Ripard <maxime.ripard@free-electrons.com>"); -MODULE_DESCRIPTION("Dumb VGA DAC bridge driver"); -MODULE_LICENSE("GPL"); diff --git a/drivers/gpu/drm/bridge/lvds-codec.c b/drivers/gpu/drm/bridge/lvds-codec.c index 5f04cc11227e..24fb1befdfa2 100644 --- a/drivers/gpu/drm/bridge/lvds-codec.c +++ b/drivers/gpu/drm/bridge/lvds-codec.c @@ -21,19 +21,23 @@ struct lvds_codec { u32 connector_type; }; -static int lvds_codec_attach(struct drm_bridge *bridge) +static inline struct lvds_codec *to_lvds_codec(struct drm_bridge *bridge) { - struct lvds_codec *lvds_codec = container_of(bridge, - struct lvds_codec, bridge); + return container_of(bridge, struct lvds_codec, bridge); +} + +static int lvds_codec_attach(struct drm_bridge *bridge, + enum drm_bridge_attach_flags flags) +{ + struct lvds_codec *lvds_codec = to_lvds_codec(bridge); return drm_bridge_attach(bridge->encoder, lvds_codec->panel_bridge, - bridge); + bridge, flags); } static void lvds_codec_enable(struct drm_bridge *bridge) { - struct lvds_codec *lvds_codec = container_of(bridge, - struct lvds_codec, bridge); + struct lvds_codec *lvds_codec = to_lvds_codec(bridge); if (lvds_codec->powerdown_gpio) gpiod_set_value_cansleep(lvds_codec->powerdown_gpio, 0); @@ -41,14 +45,13 @@ static void lvds_codec_enable(struct drm_bridge *bridge) static void lvds_codec_disable(struct drm_bridge *bridge) { - struct lvds_codec *lvds_codec = container_of(bridge, - struct lvds_codec, bridge); + struct lvds_codec *lvds_codec = to_lvds_codec(bridge); if (lvds_codec->powerdown_gpio) gpiod_set_value_cansleep(lvds_codec->powerdown_gpio, 1); } -static struct drm_bridge_funcs funcs = { +static const struct drm_bridge_funcs funcs = { .attach = lvds_codec_attach, .enable = lvds_codec_enable, .disable = lvds_codec_disable, diff --git a/drivers/gpu/drm/bridge/megachips-stdpxxxx-ge-b850v3-fw.c b/drivers/gpu/drm/bridge/megachips-stdpxxxx-ge-b850v3-fw.c index e8a49f6146c6..6200f12a37e6 100644 --- a/drivers/gpu/drm/bridge/megachips-stdpxxxx-ge-b850v3-fw.c +++ b/drivers/gpu/drm/bridge/megachips-stdpxxxx-ge-b850v3-fw.c @@ -206,13 +206,19 @@ static irqreturn_t ge_b850v3_lvds_irq_handler(int irq, void *dev_id) return IRQ_HANDLED; } -static int ge_b850v3_lvds_attach(struct drm_bridge *bridge) +static int ge_b850v3_lvds_attach(struct drm_bridge *bridge, + enum drm_bridge_attach_flags flags) { struct drm_connector *connector = &ge_b850v3_lvds_ptr->connector; struct i2c_client *stdp4028_i2c = ge_b850v3_lvds_ptr->stdp4028_i2c; int ret; + if (flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR) { + DRM_ERROR("Fix bridge driver to make connector optional!"); + return -EINVAL; + } + if (!bridge->encoder) { DRM_ERROR("Parent encoder object not found"); return -ENODEV; diff --git a/drivers/gpu/drm/bridge/nxp-ptn3460.c b/drivers/gpu/drm/bridge/nxp-ptn3460.c index 57ff01339559..438e566ce0a4 100644 --- a/drivers/gpu/drm/bridge/nxp-ptn3460.c +++ b/drivers/gpu/drm/bridge/nxp-ptn3460.c @@ -236,11 +236,17 @@ static const struct drm_connector_funcs ptn3460_connector_funcs = { .atomic_destroy_state = drm_atomic_helper_connector_destroy_state, }; -static int ptn3460_bridge_attach(struct drm_bridge *bridge) +static int ptn3460_bridge_attach(struct drm_bridge *bridge, + enum drm_bridge_attach_flags flags) { struct ptn3460_bridge *ptn_bridge = bridge_to_ptn3460(bridge); int ret; + if (flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR) { + DRM_ERROR("Fix bridge driver to make connector optional!"); + return -EINVAL; + } + if (!bridge->encoder) { DRM_ERROR("Parent encoder object not found"); return -ENODEV; diff --git a/drivers/gpu/drm/bridge/panel.c b/drivers/gpu/drm/bridge/panel.c index f66777e24968..8461ee8304ba 100644 --- a/drivers/gpu/drm/bridge/panel.c +++ b/drivers/gpu/drm/bridge/panel.c @@ -53,12 +53,16 @@ static const struct drm_connector_funcs panel_bridge_connector_funcs = { .atomic_destroy_state = drm_atomic_helper_connector_destroy_state, }; -static int panel_bridge_attach(struct drm_bridge *bridge) +static int panel_bridge_attach(struct drm_bridge *bridge, + enum drm_bridge_attach_flags flags) { struct panel_bridge *panel_bridge = drm_bridge_to_panel_bridge(bridge); struct drm_connector *connector = &panel_bridge->connector; int ret; + if (flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR) + return 0; + if (!bridge->encoder) { DRM_ERROR("Missing encoder\n"); return -ENODEV; @@ -120,6 +124,14 @@ static void panel_bridge_post_disable(struct drm_bridge *bridge) drm_panel_unprepare(panel_bridge->panel); } +static int panel_bridge_get_modes(struct drm_bridge *bridge, + struct drm_connector *connector) +{ + struct panel_bridge *panel_bridge = drm_bridge_to_panel_bridge(bridge); + + return drm_panel_get_modes(panel_bridge->panel, connector); +} + static const struct drm_bridge_funcs panel_bridge_bridge_funcs = { .attach = panel_bridge_attach, .detach = panel_bridge_detach, @@ -127,6 +139,11 @@ static const struct drm_bridge_funcs panel_bridge_bridge_funcs = { .enable = panel_bridge_enable, .disable = panel_bridge_disable, .post_disable = panel_bridge_post_disable, + .get_modes = panel_bridge_get_modes, + .atomic_reset = drm_atomic_helper_bridge_reset, + .atomic_duplicate_state = drm_atomic_helper_bridge_duplicate_state, + .atomic_destroy_state = drm_atomic_helper_bridge_destroy_state, + .atomic_get_input_bus_fmts = drm_atomic_helper_bridge_propagate_bus_fmt, }; /** @@ -151,7 +168,7 @@ static const struct drm_bridge_funcs panel_bridge_bridge_funcs = { * known type. Calling this function with a panel whose connector type is * DRM_MODE_CONNECTOR_Unknown will return NULL. * - * See devm_drm_panel_bridge_add() for an automatically manged version of this + * See devm_drm_panel_bridge_add() for an automatically managed version of this * function. */ struct drm_bridge *drm_panel_bridge_add(struct drm_panel *panel) @@ -196,6 +213,8 @@ struct drm_bridge *drm_panel_bridge_add_typed(struct drm_panel *panel, #ifdef CONFIG_OF panel_bridge->bridge.of_node = panel->dev->of_node; #endif + panel_bridge->bridge.ops = DRM_BRIDGE_OP_MODES; + panel_bridge->bridge.type = connector_type; drm_bridge_add(&panel_bridge->bridge); diff --git a/drivers/gpu/drm/bridge/parade-ps8622.c b/drivers/gpu/drm/bridge/parade-ps8622.c index 10c47c008b40..d789ea2a7fb9 100644 --- a/drivers/gpu/drm/bridge/parade-ps8622.c +++ b/drivers/gpu/drm/bridge/parade-ps8622.c @@ -476,11 +476,17 @@ static const struct drm_connector_funcs ps8622_connector_funcs = { .atomic_destroy_state = drm_atomic_helper_connector_destroy_state, }; -static int ps8622_attach(struct drm_bridge *bridge) +static int ps8622_attach(struct drm_bridge *bridge, + enum drm_bridge_attach_flags flags) { struct ps8622_bridge *ps8622 = bridge_to_ps8622(bridge); int ret; + if (flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR) { + DRM_ERROR("Fix bridge driver to make connector optional!"); + return -EINVAL; + } + if (!bridge->encoder) { DRM_ERROR("Parent encoder object not found"); return -ENODEV; diff --git a/drivers/gpu/drm/bridge/parade-ps8640.c b/drivers/gpu/drm/bridge/parade-ps8640.c index c6c06688aff2..d3a53442d449 100644 --- a/drivers/gpu/drm/bridge/parade-ps8640.c +++ b/drivers/gpu/drm/bridge/parade-ps8640.c @@ -187,7 +187,8 @@ static void ps8640_post_disable(struct drm_bridge *bridge) DRM_ERROR("cannot disable regulators %d\n", ret); } -static int ps8640_bridge_attach(struct drm_bridge *bridge) +static int ps8640_bridge_attach(struct drm_bridge *bridge, + enum drm_bridge_attach_flags flags) { struct ps8640 *ps_bridge = bridge_to_ps8640(bridge); struct device *dev = &ps_bridge->page[0]->dev; @@ -234,7 +235,7 @@ static int ps8640_bridge_attach(struct drm_bridge *bridge) /* Attach the panel-bridge to the dsi bridge */ return drm_bridge_attach(bridge->encoder, ps_bridge->panel_bridge, - &ps_bridge->bridge); + &ps_bridge->bridge, flags); err_dsi_attach: mipi_dsi_device_unregister(dsi); diff --git a/drivers/gpu/drm/bridge/sii902x.c b/drivers/gpu/drm/bridge/sii902x.c index b70e8c5cf2e1..6dad025f8da7 100644 --- a/drivers/gpu/drm/bridge/sii902x.c +++ b/drivers/gpu/drm/bridge/sii902x.c @@ -399,12 +399,18 @@ out: mutex_unlock(&sii902x->mutex); } -static int sii902x_bridge_attach(struct drm_bridge *bridge) +static int sii902x_bridge_attach(struct drm_bridge *bridge, + enum drm_bridge_attach_flags flags) { struct sii902x *sii902x = bridge_to_sii902x(bridge); struct drm_device *drm = bridge->dev; int ret; + if (flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR) { + DRM_ERROR("Fix bridge driver to make connector optional!"); + return -EINVAL; + } + drm_connector_helper_add(&sii902x->connector, &sii902x_connector_helper_funcs); diff --git a/drivers/gpu/drm/bridge/sil-sii8620.c b/drivers/gpu/drm/bridge/sil-sii8620.c index 4c0eef406eb1..92acd336aa89 100644 --- a/drivers/gpu/drm/bridge/sil-sii8620.c +++ b/drivers/gpu/drm/bridge/sil-sii8620.c @@ -2202,7 +2202,8 @@ static inline struct sii8620 *bridge_to_sii8620(struct drm_bridge *bridge) return container_of(bridge, struct sii8620, bridge); } -static int sii8620_attach(struct drm_bridge *bridge) +static int sii8620_attach(struct drm_bridge *bridge, + enum drm_bridge_attach_flags flags) { struct sii8620 *ctx = bridge_to_sii8620(bridge); diff --git a/drivers/gpu/drm/bridge/simple-bridge.c b/drivers/gpu/drm/bridge/simple-bridge.c new file mode 100644 index 000000000000..a2dca7a3ef03 --- /dev/null +++ b/drivers/gpu/drm/bridge/simple-bridge.c @@ -0,0 +1,342 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * Copyright (C) 2015-2016 Free Electrons + * Copyright (C) 2015-2016 NextThing Co + * + * Maxime Ripard <maxime.ripard@free-electrons.com> + */ + +#include <linux/gpio/consumer.h> +#include <linux/module.h> +#include <linux/of_device.h> +#include <linux/of_graph.h> +#include <linux/regulator/consumer.h> + +#include <drm/drm_atomic_helper.h> +#include <drm/drm_bridge.h> +#include <drm/drm_crtc.h> +#include <drm/drm_print.h> +#include <drm/drm_probe_helper.h> + +struct simple_bridge_info { + const struct drm_bridge_timings *timings; + unsigned int connector_type; +}; + +struct simple_bridge { + struct drm_bridge bridge; + struct drm_connector connector; + + const struct simple_bridge_info *info; + + struct i2c_adapter *ddc; + struct regulator *vdd; + struct gpio_desc *enable; +}; + +static inline struct simple_bridge * +drm_bridge_to_simple_bridge(struct drm_bridge *bridge) +{ + return container_of(bridge, struct simple_bridge, bridge); +} + +static inline struct simple_bridge * +drm_connector_to_simple_bridge(struct drm_connector *connector) +{ + return container_of(connector, struct simple_bridge, connector); +} + +static int simple_bridge_get_modes(struct drm_connector *connector) +{ + struct simple_bridge *sbridge = drm_connector_to_simple_bridge(connector); + struct edid *edid; + int ret; + + if (!sbridge->ddc) + goto fallback; + + edid = drm_get_edid(connector, sbridge->ddc); + if (!edid) { + DRM_INFO("EDID readout failed, falling back to standard modes\n"); + goto fallback; + } + + drm_connector_update_edid_property(connector, edid); + ret = drm_add_edid_modes(connector, edid); + kfree(edid); + return ret; + +fallback: + /* + * In case we cannot retrieve the EDIDs (broken or missing i2c + * bus), fallback on the XGA standards + */ + ret = drm_add_modes_noedid(connector, 1920, 1200); + + /* And prefer a mode pretty much anyone can handle */ + drm_set_preferred_mode(connector, 1024, 768); + + return ret; +} + +static const struct drm_connector_helper_funcs simple_bridge_con_helper_funcs = { + .get_modes = simple_bridge_get_modes, +}; + +static enum drm_connector_status +simple_bridge_connector_detect(struct drm_connector *connector, bool force) +{ + struct simple_bridge *sbridge = drm_connector_to_simple_bridge(connector); + + /* + * Even if we have an I2C bus, we can't assume that the cable + * is disconnected if drm_probe_ddc fails. Some cables don't + * wire the DDC pins, or the I2C bus might not be working at + * all. + */ + if (sbridge->ddc && drm_probe_ddc(sbridge->ddc)) + return connector_status_connected; + + return connector_status_unknown; +} + +static const struct drm_connector_funcs simple_bridge_con_funcs = { + .detect = simple_bridge_connector_detect, + .fill_modes = drm_helper_probe_single_connector_modes, + .destroy = drm_connector_cleanup, + .reset = drm_atomic_helper_connector_reset, + .atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state, + .atomic_destroy_state = drm_atomic_helper_connector_destroy_state, +}; + +static int simple_bridge_attach(struct drm_bridge *bridge, + enum drm_bridge_attach_flags flags) +{ + struct simple_bridge *sbridge = drm_bridge_to_simple_bridge(bridge); + int ret; + + if (flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR) { + DRM_ERROR("Fix bridge driver to make connector optional!"); + return -EINVAL; + } + + if (!bridge->encoder) { + DRM_ERROR("Missing encoder\n"); + return -ENODEV; + } + + drm_connector_helper_add(&sbridge->connector, + &simple_bridge_con_helper_funcs); + ret = drm_connector_init_with_ddc(bridge->dev, &sbridge->connector, + &simple_bridge_con_funcs, + sbridge->info->connector_type, + sbridge->ddc); + if (ret) { + DRM_ERROR("Failed to initialize connector\n"); + return ret; + } + + drm_connector_attach_encoder(&sbridge->connector, + bridge->encoder); + + return 0; +} + +static void simple_bridge_enable(struct drm_bridge *bridge) +{ + struct simple_bridge *sbridge = drm_bridge_to_simple_bridge(bridge); + int ret; + + if (sbridge->vdd) { + ret = regulator_enable(sbridge->vdd); + if (ret) + DRM_ERROR("Failed to enable vdd regulator: %d\n", ret); + } + + gpiod_set_value_cansleep(sbridge->enable, 1); +} + +static void simple_bridge_disable(struct drm_bridge *bridge) +{ + struct simple_bridge *sbridge = drm_bridge_to_simple_bridge(bridge); + + gpiod_set_value_cansleep(sbridge->enable, 0); + + if (sbridge->vdd) + regulator_disable(sbridge->vdd); +} + +static const struct drm_bridge_funcs simple_bridge_bridge_funcs = { + .attach = simple_bridge_attach, + .enable = simple_bridge_enable, + .disable = simple_bridge_disable, +}; + +static struct i2c_adapter *simple_bridge_retrieve_ddc(struct device *dev) +{ + struct device_node *phandle, *remote; + struct i2c_adapter *ddc; + + remote = of_graph_get_remote_node(dev->of_node, 1, -1); + if (!remote) + return ERR_PTR(-EINVAL); + + phandle = of_parse_phandle(remote, "ddc-i2c-bus", 0); + of_node_put(remote); + if (!phandle) + return ERR_PTR(-ENODEV); + + ddc = of_get_i2c_adapter_by_node(phandle); + of_node_put(phandle); + if (!ddc) + return ERR_PTR(-EPROBE_DEFER); + + return ddc; +} + +static int simple_bridge_probe(struct platform_device *pdev) +{ + struct simple_bridge *sbridge; + + sbridge = devm_kzalloc(&pdev->dev, sizeof(*sbridge), GFP_KERNEL); + if (!sbridge) + return -ENOMEM; + platform_set_drvdata(pdev, sbridge); + + sbridge->info = of_device_get_match_data(&pdev->dev); + + sbridge->vdd = devm_regulator_get_optional(&pdev->dev, "vdd"); + if (IS_ERR(sbridge->vdd)) { + int ret = PTR_ERR(sbridge->vdd); + if (ret == -EPROBE_DEFER) + return -EPROBE_DEFER; + sbridge->vdd = NULL; + dev_dbg(&pdev->dev, "No vdd regulator found: %d\n", ret); + } + + sbridge->enable = devm_gpiod_get_optional(&pdev->dev, "enable", + GPIOD_OUT_LOW); + if (IS_ERR(sbridge->enable)) { + if (PTR_ERR(sbridge->enable) != -EPROBE_DEFER) + dev_err(&pdev->dev, "Unable to retrieve enable GPIO\n"); + return PTR_ERR(sbridge->enable); + } + + sbridge->ddc = simple_bridge_retrieve_ddc(&pdev->dev); + if (IS_ERR(sbridge->ddc)) { + if (PTR_ERR(sbridge->ddc) == -ENODEV) { + dev_dbg(&pdev->dev, + "No i2c bus specified. Disabling EDID readout\n"); + sbridge->ddc = NULL; + } else { + dev_err(&pdev->dev, "Couldn't retrieve i2c bus\n"); + return PTR_ERR(sbridge->ddc); + } + } + + sbridge->bridge.funcs = &simple_bridge_bridge_funcs; + sbridge->bridge.of_node = pdev->dev.of_node; + sbridge->bridge.timings = sbridge->info->timings; + + drm_bridge_add(&sbridge->bridge); + + return 0; +} + +static int simple_bridge_remove(struct platform_device *pdev) +{ + struct simple_bridge *sbridge = platform_get_drvdata(pdev); + + drm_bridge_remove(&sbridge->bridge); + + if (sbridge->ddc) + i2c_put_adapter(sbridge->ddc); + + return 0; +} + +/* + * We assume the ADV7123 DAC is the "default" for historical reasons + * Information taken from the ADV7123 datasheet, revision D. + * NOTE: the ADV7123EP seems to have other timings and need a new timings + * set if used. + */ +static const struct drm_bridge_timings default_bridge_timings = { + /* Timing specifications, datasheet page 7 */ + .input_bus_flags = DRM_BUS_FLAG_PIXDATA_SAMPLE_POSEDGE, + .setup_time_ps = 500, + .hold_time_ps = 1500, +}; + +/* + * Information taken from the THS8134, THS8134A, THS8134B datasheet named + * "SLVS205D", dated May 1990, revised March 2000. + */ +static const struct drm_bridge_timings ti_ths8134_bridge_timings = { + /* From timing diagram, datasheet page 9 */ + .input_bus_flags = DRM_BUS_FLAG_PIXDATA_SAMPLE_POSEDGE, + /* From datasheet, page 12 */ + .setup_time_ps = 3000, + /* I guess this means latched input */ + .hold_time_ps = 0, +}; + +/* + * Information taken from the THS8135 datasheet named "SLAS343B", dated + * May 2001, revised April 2013. + */ +static const struct drm_bridge_timings ti_ths8135_bridge_timings = { + /* From timing diagram, datasheet page 14 */ + .input_bus_flags = DRM_BUS_FLAG_PIXDATA_SAMPLE_POSEDGE, + /* From datasheet, page 16 */ + .setup_time_ps = 2000, + .hold_time_ps = 500, +}; + +static const struct of_device_id simple_bridge_match[] = { + { + .compatible = "dumb-vga-dac", + .data = &(const struct simple_bridge_info) { + .connector_type = DRM_MODE_CONNECTOR_VGA, + }, + }, { + .compatible = "adi,adv7123", + .data = &(const struct simple_bridge_info) { + .timings = &default_bridge_timings, + .connector_type = DRM_MODE_CONNECTOR_VGA, + }, + }, { + .compatible = "ti,opa362", + .data = &(const struct simple_bridge_info) { + .connector_type = DRM_MODE_CONNECTOR_Composite, + }, + }, { + .compatible = "ti,ths8135", + .data = &(const struct simple_bridge_info) { + .timings = &ti_ths8135_bridge_timings, + .connector_type = DRM_MODE_CONNECTOR_VGA, + }, + }, { + .compatible = "ti,ths8134", + .data = &(const struct simple_bridge_info) { + .timings = &ti_ths8134_bridge_timings, + .connector_type = DRM_MODE_CONNECTOR_VGA, + }, + }, + {}, +}; +MODULE_DEVICE_TABLE(of, simple_bridge_match); + +static struct platform_driver simple_bridge_driver = { + .probe = simple_bridge_probe, + .remove = simple_bridge_remove, + .driver = { + .name = "simple-bridge", + .of_match_table = simple_bridge_match, + }, +}; +module_platform_driver(simple_bridge_driver); + +MODULE_AUTHOR("Maxime Ripard <maxime.ripard@free-electrons.com>"); +MODULE_DESCRIPTION("Simple DRM bridge driver"); +MODULE_LICENSE("GPL"); diff --git a/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c b/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c index 67fca439bbfb..383b1073d7de 100644 --- a/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c +++ b/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c @@ -1624,28 +1624,34 @@ static void hdmi_config_AVI(struct dw_hdmi *hdmi, struct drm_display_mode *mode) frame.colorspace = HDMI_COLORSPACE_RGB; /* Set up colorimetry */ - switch (hdmi->hdmi_data.enc_out_encoding) { - case V4L2_YCBCR_ENC_601: - if (hdmi->hdmi_data.enc_in_encoding == V4L2_YCBCR_ENC_XV601) - frame.colorimetry = HDMI_COLORIMETRY_EXTENDED; - else + if (!hdmi_bus_fmt_is_rgb(hdmi->hdmi_data.enc_out_bus_format)) { + switch (hdmi->hdmi_data.enc_out_encoding) { + case V4L2_YCBCR_ENC_601: + if (hdmi->hdmi_data.enc_in_encoding == V4L2_YCBCR_ENC_XV601) + frame.colorimetry = HDMI_COLORIMETRY_EXTENDED; + else + frame.colorimetry = HDMI_COLORIMETRY_ITU_601; + frame.extended_colorimetry = + HDMI_EXTENDED_COLORIMETRY_XV_YCC_601; + break; + case V4L2_YCBCR_ENC_709: + if (hdmi->hdmi_data.enc_in_encoding == V4L2_YCBCR_ENC_XV709) + frame.colorimetry = HDMI_COLORIMETRY_EXTENDED; + else + frame.colorimetry = HDMI_COLORIMETRY_ITU_709; + frame.extended_colorimetry = + HDMI_EXTENDED_COLORIMETRY_XV_YCC_709; + break; + default: /* Carries no data */ frame.colorimetry = HDMI_COLORIMETRY_ITU_601; + frame.extended_colorimetry = + HDMI_EXTENDED_COLORIMETRY_XV_YCC_601; + break; + } + } else { + frame.colorimetry = HDMI_COLORIMETRY_NONE; frame.extended_colorimetry = - HDMI_EXTENDED_COLORIMETRY_XV_YCC_601; - break; - case V4L2_YCBCR_ENC_709: - if (hdmi->hdmi_data.enc_in_encoding == V4L2_YCBCR_ENC_XV709) - frame.colorimetry = HDMI_COLORIMETRY_EXTENDED; - else - frame.colorimetry = HDMI_COLORIMETRY_ITU_709; - frame.extended_colorimetry = - HDMI_EXTENDED_COLORIMETRY_XV_YCC_709; - break; - default: /* Carries no data */ - frame.colorimetry = HDMI_COLORIMETRY_ITU_601; - frame.extended_colorimetry = - HDMI_EXTENDED_COLORIMETRY_XV_YCC_601; - break; + HDMI_EXTENDED_COLORIMETRY_XV_YCC_601; } frame.scan_mode = HDMI_SCAN_MODE_NONE; @@ -1814,13 +1820,32 @@ static void hdmi_av_composer(struct dw_hdmi *hdmi, int hblank, vblank, h_de_hs, v_de_vs, hsync_len, vsync_len; unsigned int vdisplay, hdisplay; - vmode->mtmdsclock = vmode->mpixelclock = mode->clock * 1000; + vmode->mpixelclock = mode->clock * 1000; dev_dbg(hdmi->dev, "final pixclk = %d\n", vmode->mpixelclock); + vmode->mtmdsclock = vmode->mpixelclock; + + if (!hdmi_bus_fmt_is_yuv422(hdmi->hdmi_data.enc_out_bus_format)) { + switch (hdmi_bus_fmt_color_depth( + hdmi->hdmi_data.enc_out_bus_format)) { + case 16: + vmode->mtmdsclock = vmode->mpixelclock * 2; + break; + case 12: + vmode->mtmdsclock = vmode->mpixelclock * 3 / 2; + break; + case 10: + vmode->mtmdsclock = vmode->mpixelclock * 5 / 4; + break; + } + } + if (hdmi_bus_fmt_is_yuv420(hdmi->hdmi_data.enc_out_bus_format)) vmode->mtmdsclock /= 2; + dev_dbg(hdmi->dev, "final tmdsclock = %d\n", vmode->mtmdsclock); + /* Set up HDMI_FC_INVIDCONF */ inv_val = (hdmi->hdmi_data.hdcp_enable || (dw_hdmi_support_scdc(hdmi) && @@ -2078,11 +2103,10 @@ static int dw_hdmi_setup(struct dw_hdmi *hdmi, struct drm_display_mode *mode) hdmi->hdmi_data.video_mode.mpixelrepetitionoutput = 0; hdmi->hdmi_data.video_mode.mpixelrepetitioninput = 0; - /* TOFIX: Get input format from plat data or fallback to RGB888 */ if (hdmi->plat_data->input_bus_format) hdmi->hdmi_data.enc_in_bus_format = hdmi->plat_data->input_bus_format; - else + else if (hdmi->hdmi_data.enc_in_bus_format == MEDIA_BUS_FMT_FIXED) hdmi->hdmi_data.enc_in_bus_format = MEDIA_BUS_FMT_RGB888_1X24; /* TOFIX: Get input encoding from plat data or fallback to none */ @@ -2092,8 +2116,8 @@ static int dw_hdmi_setup(struct dw_hdmi *hdmi, struct drm_display_mode *mode) else hdmi->hdmi_data.enc_in_encoding = V4L2_YCBCR_ENC_DEFAULT; - /* TOFIX: Default to RGB888 output format */ - hdmi->hdmi_data.enc_out_bus_format = MEDIA_BUS_FMT_RGB888_1X24; + if (hdmi->hdmi_data.enc_out_bus_format == MEDIA_BUS_FMT_FIXED) + hdmi->hdmi_data.enc_out_bus_format = MEDIA_BUS_FMT_RGB888_1X24; hdmi->hdmi_data.pix_repet_factor = 0; hdmi->hdmi_data.hdcp_enable = 0; @@ -2371,7 +2395,279 @@ static const struct drm_connector_helper_funcs dw_hdmi_connector_helper_funcs = .atomic_check = dw_hdmi_connector_atomic_check, }; -static int dw_hdmi_bridge_attach(struct drm_bridge *bridge) +/* + * Possible output formats : + * - MEDIA_BUS_FMT_UYYVYY16_0_5X48, + * - MEDIA_BUS_FMT_UYYVYY12_0_5X36, + * - MEDIA_BUS_FMT_UYYVYY10_0_5X30, + * - MEDIA_BUS_FMT_UYYVYY8_0_5X24, + * - MEDIA_BUS_FMT_YUV16_1X48, + * - MEDIA_BUS_FMT_RGB161616_1X48, + * - MEDIA_BUS_FMT_UYVY12_1X24, + * - MEDIA_BUS_FMT_YUV12_1X36, + * - MEDIA_BUS_FMT_RGB121212_1X36, + * - MEDIA_BUS_FMT_UYVY10_1X20, + * - MEDIA_BUS_FMT_YUV10_1X30, + * - MEDIA_BUS_FMT_RGB101010_1X30, + * - MEDIA_BUS_FMT_UYVY8_1X16, + * - MEDIA_BUS_FMT_YUV8_1X24, + * - MEDIA_BUS_FMT_RGB888_1X24, + */ + +/* Can return a maximum of 11 possible output formats for a mode/connector */ +#define MAX_OUTPUT_SEL_FORMATS 11 + +static u32 *dw_hdmi_bridge_atomic_get_output_bus_fmts(struct drm_bridge *bridge, + struct drm_bridge_state *bridge_state, + struct drm_crtc_state *crtc_state, + struct drm_connector_state *conn_state, + unsigned int *num_output_fmts) +{ + struct drm_connector *conn = conn_state->connector; + struct drm_display_info *info = &conn->display_info; + struct drm_display_mode *mode = &crtc_state->mode; + u8 max_bpc = conn_state->max_requested_bpc; + bool is_hdmi2_sink = info->hdmi.scdc.supported || + (info->color_formats & DRM_COLOR_FORMAT_YCRCB420); + u32 *output_fmts; + unsigned int i = 0; + + *num_output_fmts = 0; + + output_fmts = kcalloc(MAX_OUTPUT_SEL_FORMATS, sizeof(*output_fmts), + GFP_KERNEL); + if (!output_fmts) + return NULL; + + /* If dw-hdmi is the only bridge, avoid negociating with ourselves */ + if (list_is_singular(&bridge->encoder->bridge_chain)) { + *num_output_fmts = 1; + output_fmts[0] = MEDIA_BUS_FMT_FIXED; + + return output_fmts; + } + + /* + * If the current mode enforces 4:2:0, force the output but format + * to 4:2:0 and do not add the YUV422/444/RGB formats + */ + if (conn->ycbcr_420_allowed && + (drm_mode_is_420_only(info, mode) || + (is_hdmi2_sink && drm_mode_is_420_also(info, mode)))) { + + /* Order bus formats from 16bit to 8bit if supported */ + if (max_bpc >= 16 && info->bpc == 16 && + (info->hdmi.y420_dc_modes & DRM_EDID_YCBCR420_DC_48)) + output_fmts[i++] = MEDIA_BUS_FMT_UYYVYY16_0_5X48; + + if (max_bpc >= 12 && info->bpc >= 12 && + (info->hdmi.y420_dc_modes & DRM_EDID_YCBCR420_DC_36)) + output_fmts[i++] = MEDIA_BUS_FMT_UYYVYY12_0_5X36; + + if (max_bpc >= 10 && info->bpc >= 10 && + (info->hdmi.y420_dc_modes & DRM_EDID_YCBCR420_DC_30)) + output_fmts[i++] = MEDIA_BUS_FMT_UYYVYY10_0_5X30; + + /* Default 8bit fallback */ + output_fmts[i++] = MEDIA_BUS_FMT_UYYVYY8_0_5X24; + + *num_output_fmts = i; + + return output_fmts; + } + + /* + * Order bus formats from 16bit to 8bit and from YUV422 to RGB + * if supported. In any case the default RGB888 format is added + */ + + if (max_bpc >= 16 && info->bpc == 16) { + if (info->color_formats & DRM_COLOR_FORMAT_YCRCB444) + output_fmts[i++] = MEDIA_BUS_FMT_YUV16_1X48; + + output_fmts[i++] = MEDIA_BUS_FMT_RGB161616_1X48; + } + + if (max_bpc >= 12 && info->bpc >= 12) { + if (info->color_formats & DRM_COLOR_FORMAT_YCRCB422) + output_fmts[i++] = MEDIA_BUS_FMT_UYVY12_1X24; + + if (info->color_formats & DRM_COLOR_FORMAT_YCRCB444) + output_fmts[i++] = MEDIA_BUS_FMT_YUV12_1X36; + + output_fmts[i++] = MEDIA_BUS_FMT_RGB121212_1X36; + } + + if (max_bpc >= 10 && info->bpc >= 10) { + if (info->color_formats & DRM_COLOR_FORMAT_YCRCB422) + output_fmts[i++] = MEDIA_BUS_FMT_UYVY10_1X20; + + if (info->color_formats & DRM_COLOR_FORMAT_YCRCB444) + output_fmts[i++] = MEDIA_BUS_FMT_YUV10_1X30; + + output_fmts[i++] = MEDIA_BUS_FMT_RGB101010_1X30; + } + + if (info->color_formats & DRM_COLOR_FORMAT_YCRCB422) + output_fmts[i++] = MEDIA_BUS_FMT_UYVY8_1X16; + + if (info->color_formats & DRM_COLOR_FORMAT_YCRCB444) + output_fmts[i++] = MEDIA_BUS_FMT_YUV8_1X24; + + /* Default 8bit RGB fallback */ + output_fmts[i++] = MEDIA_BUS_FMT_RGB888_1X24; + + *num_output_fmts = i; + + return output_fmts; +} + +/* + * Possible input formats : + * - MEDIA_BUS_FMT_RGB888_1X24 + * - MEDIA_BUS_FMT_YUV8_1X24 + * - MEDIA_BUS_FMT_UYVY8_1X16 + * - MEDIA_BUS_FMT_UYYVYY8_0_5X24 + * - MEDIA_BUS_FMT_RGB101010_1X30 + * - MEDIA_BUS_FMT_YUV10_1X30 + * - MEDIA_BUS_FMT_UYVY10_1X20 + * - MEDIA_BUS_FMT_UYYVYY10_0_5X30 + * - MEDIA_BUS_FMT_RGB121212_1X36 + * - MEDIA_BUS_FMT_YUV12_1X36 + * - MEDIA_BUS_FMT_UYVY12_1X24 + * - MEDIA_BUS_FMT_UYYVYY12_0_5X36 + * - MEDIA_BUS_FMT_RGB161616_1X48 + * - MEDIA_BUS_FMT_YUV16_1X48 + * - MEDIA_BUS_FMT_UYYVYY16_0_5X48 + */ + +/* Can return a maximum of 3 possible input formats for an output format */ +#define MAX_INPUT_SEL_FORMATS 3 + +static u32 *dw_hdmi_bridge_atomic_get_input_bus_fmts(struct drm_bridge *bridge, + struct drm_bridge_state *bridge_state, + struct drm_crtc_state *crtc_state, + struct drm_connector_state *conn_state, + u32 output_fmt, + unsigned int *num_input_fmts) +{ + u32 *input_fmts; + unsigned int i = 0; + + *num_input_fmts = 0; + + input_fmts = kcalloc(MAX_INPUT_SEL_FORMATS, sizeof(*input_fmts), + GFP_KERNEL); + if (!input_fmts) + return NULL; + + switch (output_fmt) { + /* If MEDIA_BUS_FMT_FIXED is tested, return default bus format */ + case MEDIA_BUS_FMT_FIXED: + input_fmts[i++] = MEDIA_BUS_FMT_RGB888_1X24; + break; + /* 8bit */ + case MEDIA_BUS_FMT_RGB888_1X24: + input_fmts[i++] = MEDIA_BUS_FMT_RGB888_1X24; + input_fmts[i++] = MEDIA_BUS_FMT_YUV8_1X24; + input_fmts[i++] = MEDIA_BUS_FMT_UYVY8_1X16; + break; + case MEDIA_BUS_FMT_YUV8_1X24: + input_fmts[i++] = MEDIA_BUS_FMT_YUV8_1X24; + input_fmts[i++] = MEDIA_BUS_FMT_UYVY8_1X16; + input_fmts[i++] = MEDIA_BUS_FMT_RGB888_1X24; + break; + case MEDIA_BUS_FMT_UYVY8_1X16: + input_fmts[i++] = MEDIA_BUS_FMT_UYVY8_1X16; + input_fmts[i++] = MEDIA_BUS_FMT_YUV8_1X24; + input_fmts[i++] = MEDIA_BUS_FMT_RGB888_1X24; + break; + + /* 10bit */ + case MEDIA_BUS_FMT_RGB101010_1X30: + input_fmts[i++] = MEDIA_BUS_FMT_RGB101010_1X30; + input_fmts[i++] = MEDIA_BUS_FMT_YUV10_1X30; + input_fmts[i++] = MEDIA_BUS_FMT_UYVY10_1X20; + break; + case MEDIA_BUS_FMT_YUV10_1X30: + input_fmts[i++] = MEDIA_BUS_FMT_YUV10_1X30; + input_fmts[i++] = MEDIA_BUS_FMT_UYVY10_1X20; + input_fmts[i++] = MEDIA_BUS_FMT_RGB101010_1X30; + break; + case MEDIA_BUS_FMT_UYVY10_1X20: + input_fmts[i++] = MEDIA_BUS_FMT_UYVY10_1X20; + input_fmts[i++] = MEDIA_BUS_FMT_YUV10_1X30; + input_fmts[i++] = MEDIA_BUS_FMT_RGB101010_1X30; + break; + + /* 12bit */ + case MEDIA_BUS_FMT_RGB121212_1X36: + input_fmts[i++] = MEDIA_BUS_FMT_RGB121212_1X36; + input_fmts[i++] = MEDIA_BUS_FMT_YUV12_1X36; + input_fmts[i++] = MEDIA_BUS_FMT_UYVY12_1X24; + break; + case MEDIA_BUS_FMT_YUV12_1X36: + input_fmts[i++] = MEDIA_BUS_FMT_YUV12_1X36; + input_fmts[i++] = MEDIA_BUS_FMT_UYVY12_1X24; + input_fmts[i++] = MEDIA_BUS_FMT_RGB121212_1X36; + break; + case MEDIA_BUS_FMT_UYVY12_1X24: + input_fmts[i++] = MEDIA_BUS_FMT_UYVY12_1X24; + input_fmts[i++] = MEDIA_BUS_FMT_YUV12_1X36; + input_fmts[i++] = MEDIA_BUS_FMT_RGB121212_1X36; + break; + + /* 16bit */ + case MEDIA_BUS_FMT_RGB161616_1X48: + input_fmts[i++] = MEDIA_BUS_FMT_RGB161616_1X48; + input_fmts[i++] = MEDIA_BUS_FMT_YUV16_1X48; + break; + case MEDIA_BUS_FMT_YUV16_1X48: + input_fmts[i++] = MEDIA_BUS_FMT_YUV16_1X48; + input_fmts[i++] = MEDIA_BUS_FMT_RGB161616_1X48; + break; + + /*YUV 4:2:0 */ + case MEDIA_BUS_FMT_UYYVYY8_0_5X24: + case MEDIA_BUS_FMT_UYYVYY10_0_5X30: + case MEDIA_BUS_FMT_UYYVYY12_0_5X36: + case MEDIA_BUS_FMT_UYYVYY16_0_5X48: + input_fmts[i++] = output_fmt; + break; + } + + *num_input_fmts = i; + + if (*num_input_fmts == 0) { + kfree(input_fmts); + input_fmts = NULL; + } + + return input_fmts; +} + +static int dw_hdmi_bridge_atomic_check(struct drm_bridge *bridge, + struct drm_bridge_state *bridge_state, + struct drm_crtc_state *crtc_state, + struct drm_connector_state *conn_state) +{ + struct dw_hdmi *hdmi = bridge->driver_private; + + hdmi->hdmi_data.enc_out_bus_format = + bridge_state->output_bus_cfg.format; + + hdmi->hdmi_data.enc_in_bus_format = + bridge_state->input_bus_cfg.format; + + dev_dbg(hdmi->dev, "input format 0x%04x, output format 0x%04x\n", + bridge_state->input_bus_cfg.format, + bridge_state->output_bus_cfg.format); + + return 0; +} + +static int dw_hdmi_bridge_attach(struct drm_bridge *bridge, + enum drm_bridge_attach_flags flags) { struct dw_hdmi *hdmi = bridge->driver_private; struct drm_encoder *encoder = bridge->encoder; @@ -2379,6 +2675,11 @@ static int dw_hdmi_bridge_attach(struct drm_bridge *bridge) struct cec_connector_info conn_info; struct cec_notifier *notifier; + if (flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR) { + DRM_ERROR("Fix bridge driver to make connector optional!"); + return -EINVAL; + } + connector->interlace_allowed = 1; connector->polled = DRM_CONNECTOR_POLL_HPD; @@ -2389,6 +2690,14 @@ static int dw_hdmi_bridge_attach(struct drm_bridge *bridge) DRM_MODE_CONNECTOR_HDMIA, hdmi->ddc); + /* + * drm_connector_attach_max_bpc_property() requires the + * connector to have a state. + */ + drm_atomic_helper_connector_reset(connector); + + drm_connector_attach_max_bpc_property(connector, 8, 16); + if (hdmi->version >= 0x200a && hdmi->plat_data->use_drm_infoframe) drm_object_attach_property(&connector->base, connector->dev->mode_config.hdr_output_metadata_property, 0); @@ -2473,8 +2782,14 @@ static void dw_hdmi_bridge_enable(struct drm_bridge *bridge) } static const struct drm_bridge_funcs dw_hdmi_bridge_funcs = { + .atomic_duplicate_state = drm_atomic_helper_bridge_duplicate_state, + .atomic_destroy_state = drm_atomic_helper_bridge_destroy_state, + .atomic_reset = drm_atomic_helper_bridge_reset, .attach = dw_hdmi_bridge_attach, .detach = dw_hdmi_bridge_detach, + .atomic_check = dw_hdmi_bridge_atomic_check, + .atomic_get_output_bus_fmts = dw_hdmi_bridge_atomic_get_output_bus_fmts, + .atomic_get_input_bus_fmts = dw_hdmi_bridge_atomic_get_input_bus_fmts, .enable = dw_hdmi_bridge_enable, .disable = dw_hdmi_bridge_disable, .mode_set = dw_hdmi_bridge_mode_set, @@ -2943,6 +3258,12 @@ __dw_hdmi_probe(struct platform_device *pdev, hdmi->bridge.of_node = pdev->dev.of_node; #endif + if (hdmi->version >= 0x200a) + hdmi->connector.ycbcr_420_allowed = + hdmi->plat_data->ycbcr_420_allowed; + else + hdmi->connector.ycbcr_420_allowed = false; + memset(&pdevinfo, 0, sizeof(pdevinfo)); pdevinfo.parent = dev; pdevinfo.id = PLATFORM_DEVID_AUTO; @@ -3076,7 +3397,7 @@ struct dw_hdmi *dw_hdmi_bind(struct platform_device *pdev, if (IS_ERR(hdmi)) return hdmi; - ret = drm_bridge_attach(encoder, &hdmi->bridge, NULL); + ret = drm_bridge_attach(encoder, &hdmi->bridge, NULL, 0); if (ret) { dw_hdmi_remove(hdmi); DRM_ERROR("Failed to initialize bridge with drm\n"); diff --git a/drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi.c b/drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi.c index b18351b6760a..5ef0f154aa7b 100644 --- a/drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi.c +++ b/drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi.c @@ -824,7 +824,8 @@ static void dw_mipi_dsi_bridge_post_disable(struct drm_bridge *bridge) * This needs to be fixed in the drm_bridge framework and the API * needs to be updated to manage our own call chains... */ - dsi->panel_bridge->funcs->post_disable(dsi->panel_bridge); + if (dsi->panel_bridge->funcs->post_disable) + dsi->panel_bridge->funcs->post_disable(dsi->panel_bridge); if (phy_ops->power_off) phy_ops->power_off(dsi->plat_data->priv_data); @@ -935,7 +936,8 @@ dw_mipi_dsi_bridge_mode_valid(struct drm_bridge *bridge, return mode_status; } -static int dw_mipi_dsi_bridge_attach(struct drm_bridge *bridge) +static int dw_mipi_dsi_bridge_attach(struct drm_bridge *bridge, + enum drm_bridge_attach_flags flags) { struct dw_mipi_dsi *dsi = bridge_to_dsi(bridge); @@ -948,7 +950,8 @@ static int dw_mipi_dsi_bridge_attach(struct drm_bridge *bridge) bridge->encoder->encoder_type = DRM_MODE_ENCODER_DSI; /* Attach the panel-bridge to the dsi bridge */ - return drm_bridge_attach(bridge->encoder, dsi->panel_bridge, bridge); + return drm_bridge_attach(bridge->encoder, dsi->panel_bridge, bridge, + flags); } static const struct drm_bridge_funcs dw_mipi_dsi_bridge_funcs = { @@ -1119,7 +1122,7 @@ int dw_mipi_dsi_bind(struct dw_mipi_dsi *dsi, struct drm_encoder *encoder) { int ret; - ret = drm_bridge_attach(encoder, &dsi->bridge, NULL); + ret = drm_bridge_attach(encoder, &dsi->bridge, NULL, 0); if (ret) { DRM_ERROR("Failed to initialize bridge with drm\n"); return ret; diff --git a/drivers/gpu/drm/bridge/tc358764.c b/drivers/gpu/drm/bridge/tc358764.c index 96207fcfde19..5ac1430fab04 100644 --- a/drivers/gpu/drm/bridge/tc358764.c +++ b/drivers/gpu/drm/bridge/tc358764.c @@ -349,12 +349,18 @@ static void tc358764_enable(struct drm_bridge *bridge) dev_err(ctx->dev, "error enabling panel (%d)\n", ret); } -static int tc358764_attach(struct drm_bridge *bridge) +static int tc358764_attach(struct drm_bridge *bridge, + enum drm_bridge_attach_flags flags) { struct tc358764 *ctx = bridge_to_tc358764(bridge); struct drm_device *drm = bridge->dev; int ret; + if (flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR) { + DRM_ERROR("Fix bridge driver to make connector optional!"); + return -EINVAL; + } + ctx->connector.polled = DRM_CONNECTOR_POLL_HPD; ret = drm_connector_init(drm, &ctx->connector, &tc358764_connector_funcs, @@ -369,7 +375,6 @@ static int tc358764_attach(struct drm_bridge *bridge) drm_connector_attach_encoder(&ctx->connector, bridge->encoder); drm_panel_attach(ctx->panel, &ctx->connector); ctx->connector.funcs->reset(&ctx->connector); - drm_fb_helper_add_one_connector(drm->fb_helper, &ctx->connector); drm_connector_register(&ctx->connector); return 0; @@ -378,10 +383,8 @@ static int tc358764_attach(struct drm_bridge *bridge) static void tc358764_detach(struct drm_bridge *bridge) { struct tc358764 *ctx = bridge_to_tc358764(bridge); - struct drm_device *drm = bridge->dev; drm_connector_unregister(&ctx->connector); - drm_fb_helper_remove_one_connector(drm->fb_helper, &ctx->connector); drm_panel_detach(ctx->panel); ctx->panel = NULL; drm_connector_put(&ctx->connector); diff --git a/drivers/gpu/drm/bridge/tc358767.c b/drivers/gpu/drm/bridge/tc358767.c index 3709e5ace724..e4c0ea03ae3a 100644 --- a/drivers/gpu/drm/bridge/tc358767.c +++ b/drivers/gpu/drm/bridge/tc358767.c @@ -31,6 +31,7 @@ #include <drm/drm_edid.h> #include <drm/drm_of.h> #include <drm/drm_panel.h> +#include <drm/drm_print.h> #include <drm/drm_probe_helper.h> /* Registers */ @@ -297,7 +298,7 @@ static inline int tc_poll_timeout(struct tc_data *tc, unsigned int addr, static int tc_aux_wait_busy(struct tc_data *tc) { - return tc_poll_timeout(tc, DP0_AUXSTATUS, AUX_BUSY, 0, 1000, 100000); + return tc_poll_timeout(tc, DP0_AUXSTATUS, AUX_BUSY, 0, 100, 100000); } static int tc_aux_write_data(struct tc_data *tc, const void *data, @@ -640,7 +641,7 @@ static int tc_aux_link_setup(struct tc_data *tc) if (ret) goto err; - ret = tc_poll_timeout(tc, DP_PHY_CTRL, PHY_RDY, PHY_RDY, 1, 1000); + ret = tc_poll_timeout(tc, DP_PHY_CTRL, PHY_RDY, PHY_RDY, 100, 100000); if (ret == -ETIMEDOUT) { dev_err(tc->dev, "Timeout waiting for PHY to become ready"); return ret; @@ -876,7 +877,7 @@ static int tc_wait_link_training(struct tc_data *tc) int ret; ret = tc_poll_timeout(tc, DP0_LTSTAT, LT_LOOPDONE, - LT_LOOPDONE, 1, 1000); + LT_LOOPDONE, 500, 100000); if (ret) { dev_err(tc->dev, "Link training timeout waiting for LT_LOOPDONE!\n"); return ret; @@ -949,7 +950,7 @@ static int tc_main_link_enable(struct tc_data *tc) dp_phy_ctrl &= ~(DP_PHY_RST | PHY_M1_RST | PHY_M0_RST); ret = regmap_write(tc->regmap, DP_PHY_CTRL, dp_phy_ctrl); - ret = tc_poll_timeout(tc, DP_PHY_CTRL, PHY_RDY, PHY_RDY, 1, 1000); + ret = tc_poll_timeout(tc, DP_PHY_CTRL, PHY_RDY, PHY_RDY, 500, 100000); if (ret) { dev_err(dev, "timeout waiting for phy become ready"); return ret; @@ -1403,13 +1404,19 @@ static const struct drm_connector_funcs tc_connector_funcs = { .atomic_destroy_state = drm_atomic_helper_connector_destroy_state, }; -static int tc_bridge_attach(struct drm_bridge *bridge) +static int tc_bridge_attach(struct drm_bridge *bridge, + enum drm_bridge_attach_flags flags) { u32 bus_format = MEDIA_BUS_FMT_RGB888_1X24; struct tc_data *tc = bridge_to_tc(bridge); struct drm_device *drm = bridge->dev; int ret; + if (flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR) { + DRM_ERROR("Fix bridge driver to make connector optional!"); + return -EINVAL; + } + /* Create DP/eDP connector */ drm_connector_helper_add(&tc->connector, &tc_connector_helper_funcs); ret = drm_connector_init(drm, &tc->connector, &tc_connector_funcs, diff --git a/drivers/gpu/drm/bridge/tc358768.c b/drivers/gpu/drm/bridge/tc358768.c new file mode 100644 index 000000000000..1b39e8d37834 --- /dev/null +++ b/drivers/gpu/drm/bridge/tc358768.c @@ -0,0 +1,1046 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (C) 2020 Texas Instruments Incorporated - http://www.ti.com + * Author: Peter Ujfalusi <peter.ujfalusi@ti.com> + */ + +#include <linux/clk.h> +#include <linux/device.h> +#include <linux/gpio/consumer.h> +#include <linux/i2c.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/regmap.h> +#include <linux/regulator/consumer.h> +#include <linux/slab.h> + +#include <drm/drm_atomic_helper.h> +#include <drm/drm_crtc_helper.h> +#include <drm/drm_drv.h> +#include <drm/drm_mipi_dsi.h> +#include <drm/drm_of.h> +#include <drm/drm_panel.h> +#include <video/mipi_display.h> +#include <video/videomode.h> + +/* Global (16-bit addressable) */ +#define TC358768_CHIPID 0x0000 +#define TC358768_SYSCTL 0x0002 +#define TC358768_CONFCTL 0x0004 +#define TC358768_VSDLY 0x0006 +#define TC358768_DATAFMT 0x0008 +#define TC358768_GPIOEN 0x000E +#define TC358768_GPIODIR 0x0010 +#define TC358768_GPIOIN 0x0012 +#define TC358768_GPIOOUT 0x0014 +#define TC358768_PLLCTL0 0x0016 +#define TC358768_PLLCTL1 0x0018 +#define TC358768_CMDBYTE 0x0022 +#define TC358768_PP_MISC 0x0032 +#define TC358768_DSITX_DT 0x0050 +#define TC358768_FIFOSTATUS 0x00F8 + +/* Debug (16-bit addressable) */ +#define TC358768_VBUFCTRL 0x00E0 +#define TC358768_DBG_WIDTH 0x00E2 +#define TC358768_DBG_VBLANK 0x00E4 +#define TC358768_DBG_DATA 0x00E8 + +/* TX PHY (32-bit addressable) */ +#define TC358768_CLW_DPHYCONTTX 0x0100 +#define TC358768_D0W_DPHYCONTTX 0x0104 +#define TC358768_D1W_DPHYCONTTX 0x0108 +#define TC358768_D2W_DPHYCONTTX 0x010C +#define TC358768_D3W_DPHYCONTTX 0x0110 +#define TC358768_CLW_CNTRL 0x0140 +#define TC358768_D0W_CNTRL 0x0144 +#define TC358768_D1W_CNTRL 0x0148 +#define TC358768_D2W_CNTRL 0x014C +#define TC358768_D3W_CNTRL 0x0150 + +/* TX PPI (32-bit addressable) */ +#define TC358768_STARTCNTRL 0x0204 +#define TC358768_DSITXSTATUS 0x0208 +#define TC358768_LINEINITCNT 0x0210 +#define TC358768_LPTXTIMECNT 0x0214 +#define TC358768_TCLK_HEADERCNT 0x0218 +#define TC358768_TCLK_TRAILCNT 0x021C +#define TC358768_THS_HEADERCNT 0x0220 +#define TC358768_TWAKEUP 0x0224 +#define TC358768_TCLK_POSTCNT 0x0228 +#define TC358768_THS_TRAILCNT 0x022C +#define TC358768_HSTXVREGCNT 0x0230 +#define TC358768_HSTXVREGEN 0x0234 +#define TC358768_TXOPTIONCNTRL 0x0238 +#define TC358768_BTACNTRL1 0x023C + +/* TX CTRL (32-bit addressable) */ +#define TC358768_DSI_CONTROL 0x040C +#define TC358768_DSI_STATUS 0x0410 +#define TC358768_DSI_INT 0x0414 +#define TC358768_DSI_INT_ENA 0x0418 +#define TC358768_DSICMD_RDFIFO 0x0430 +#define TC358768_DSI_ACKERR 0x0434 +#define TC358768_DSI_ACKERR_INTENA 0x0438 +#define TC358768_DSI_ACKERR_HALT 0x043c +#define TC358768_DSI_RXERR 0x0440 +#define TC358768_DSI_RXERR_INTENA 0x0444 +#define TC358768_DSI_RXERR_HALT 0x0448 +#define TC358768_DSI_ERR 0x044C +#define TC358768_DSI_ERR_INTENA 0x0450 +#define TC358768_DSI_ERR_HALT 0x0454 +#define TC358768_DSI_CONFW 0x0500 +#define TC358768_DSI_LPCMD 0x0500 +#define TC358768_DSI_RESET 0x0504 +#define TC358768_DSI_INT_CLR 0x050C +#define TC358768_DSI_START 0x0518 + +/* DSITX CTRL (16-bit addressable) */ +#define TC358768_DSICMD_TX 0x0600 +#define TC358768_DSICMD_TYPE 0x0602 +#define TC358768_DSICMD_WC 0x0604 +#define TC358768_DSICMD_WD0 0x0610 +#define TC358768_DSICMD_WD1 0x0612 +#define TC358768_DSICMD_WD2 0x0614 +#define TC358768_DSICMD_WD3 0x0616 +#define TC358768_DSI_EVENT 0x0620 +#define TC358768_DSI_VSW 0x0622 +#define TC358768_DSI_VBPR 0x0624 +#define TC358768_DSI_VACT 0x0626 +#define TC358768_DSI_HSW 0x0628 +#define TC358768_DSI_HBPR 0x062A +#define TC358768_DSI_HACT 0x062C + +/* TC358768_DSI_CONTROL (0x040C) register */ +#define TC358768_DSI_CONTROL_DIS_MODE BIT(15) +#define TC358768_DSI_CONTROL_TXMD BIT(7) +#define TC358768_DSI_CONTROL_HSCKMD BIT(5) +#define TC358768_DSI_CONTROL_EOTDIS BIT(0) + +/* TC358768_DSI_CONFW (0x0500) register */ +#define TC358768_DSI_CONFW_MODE_SET (5 << 29) +#define TC358768_DSI_CONFW_MODE_CLR (6 << 29) +#define TC358768_DSI_CONFW_ADDR_DSI_CONTROL (0x3 << 24) + +static const char * const tc358768_supplies[] = { + "vddc", "vddmipi", "vddio" +}; + +struct tc358768_dsi_output { + struct mipi_dsi_device *dev; + struct drm_panel *panel; + struct drm_bridge *bridge; +}; + +struct tc358768_priv { + struct device *dev; + struct regmap *regmap; + struct gpio_desc *reset_gpio; + struct regulator_bulk_data supplies[ARRAY_SIZE(tc358768_supplies)]; + struct clk *refclk; + int enabled; + int error; + + struct mipi_dsi_host dsi_host; + struct drm_bridge bridge; + struct tc358768_dsi_output output; + + u32 pd_lines; /* number of Parallel Port Input Data Lines */ + u32 dsi_lanes; /* number of DSI Lanes */ + + /* Parameters for PLL programming */ + u32 fbd; /* PLL feedback divider */ + u32 prd; /* PLL input divider */ + u32 frs; /* PLL Freqency range for HSCK (post divider) */ + + u32 dsiclk; /* pll_clk / 2 */ +}; + +static inline struct tc358768_priv *dsi_host_to_tc358768(struct mipi_dsi_host + *host) +{ + return container_of(host, struct tc358768_priv, dsi_host); +} + +static inline struct tc358768_priv *bridge_to_tc358768(struct drm_bridge + *bridge) +{ + return container_of(bridge, struct tc358768_priv, bridge); +} + +static int tc358768_clear_error(struct tc358768_priv *priv) +{ + int ret = priv->error; + + priv->error = 0; + return ret; +} + +static void tc358768_write(struct tc358768_priv *priv, u32 reg, u32 val) +{ + size_t count = 2; + + if (priv->error) + return; + + /* 16-bit register? */ + if (reg < 0x100 || reg >= 0x600) + count = 1; + + priv->error = regmap_bulk_write(priv->regmap, reg, &val, count); +} + +static void tc358768_read(struct tc358768_priv *priv, u32 reg, u32 *val) +{ + size_t count = 2; + + if (priv->error) + return; + + /* 16-bit register? */ + if (reg < 0x100 || reg >= 0x600) { + *val = 0; + count = 1; + } + + priv->error = regmap_bulk_read(priv->regmap, reg, val, count); +} + +static void tc358768_update_bits(struct tc358768_priv *priv, u32 reg, u32 mask, + u32 val) +{ + u32 tmp, orig; + + tc358768_read(priv, reg, &orig); + tmp = orig & ~mask; + tmp |= val & mask; + if (tmp != orig) + tc358768_write(priv, reg, tmp); +} + +static int tc358768_sw_reset(struct tc358768_priv *priv) +{ + /* Assert Reset */ + tc358768_write(priv, TC358768_SYSCTL, 1); + /* Release Reset, Exit Sleep */ + tc358768_write(priv, TC358768_SYSCTL, 0); + + return tc358768_clear_error(priv); +} + +static void tc358768_hw_enable(struct tc358768_priv *priv) +{ + int ret; + + if (priv->enabled) + return; + + ret = regulator_bulk_enable(ARRAY_SIZE(priv->supplies), priv->supplies); + if (ret < 0) + dev_err(priv->dev, "error enabling regulators (%d)\n", ret); + + if (priv->reset_gpio) + usleep_range(200, 300); + + /* + * The RESX is active low (GPIO_ACTIVE_LOW). + * DEASSERT (value = 0) the reset_gpio to enable the chip + */ + gpiod_set_value_cansleep(priv->reset_gpio, 0); + + /* wait for encoder clocks to stabilize */ + usleep_range(1000, 2000); + + priv->enabled = true; +} + +static void tc358768_hw_disable(struct tc358768_priv *priv) +{ + int ret; + + if (!priv->enabled) + return; + + /* + * The RESX is active low (GPIO_ACTIVE_LOW). + * ASSERT (value = 1) the reset_gpio to disable the chip + */ + gpiod_set_value_cansleep(priv->reset_gpio, 1); + + ret = regulator_bulk_disable(ARRAY_SIZE(priv->supplies), + priv->supplies); + if (ret < 0) + dev_err(priv->dev, "error disabling regulators (%d)\n", ret); + + priv->enabled = false; +} + +static u32 tc358768_pll_to_pclk(struct tc358768_priv *priv, u32 pll_clk) +{ + return (u32)div_u64((u64)pll_clk * priv->dsi_lanes, priv->pd_lines); +} + +static u32 tc358768_pclk_to_pll(struct tc358768_priv *priv, u32 pclk) +{ + return (u32)div_u64((u64)pclk * priv->pd_lines, priv->dsi_lanes); +} + +static int tc358768_calc_pll(struct tc358768_priv *priv, + const struct drm_display_mode *mode, + bool verify_only) +{ + const u32 frs_limits[] = { + 1000000000, + 500000000, + 250000000, + 125000000, + 62500000 + }; + unsigned long refclk; + u32 prd, target_pll, i, max_pll, min_pll; + u32 frs, best_diff, best_pll, best_prd, best_fbd; + + target_pll = tc358768_pclk_to_pll(priv, mode->clock * 1000); + + /* pll_clk = RefClk * [(FBD + 1)/ (PRD + 1)] * [1 / (2^FRS)] */ + + for (i = 0; i < ARRAY_SIZE(frs_limits); i++) + if (target_pll >= frs_limits[i]) + break; + + if (i == ARRAY_SIZE(frs_limits) || i == 0) + return -EINVAL; + + frs = i - 1; + max_pll = frs_limits[i - 1]; + min_pll = frs_limits[i]; + + refclk = clk_get_rate(priv->refclk); + + best_diff = UINT_MAX; + best_pll = 0; + best_prd = 0; + best_fbd = 0; + + for (prd = 0; prd < 16; ++prd) { + u32 divisor = (prd + 1) * (1 << frs); + u32 fbd; + + for (fbd = 0; fbd < 512; ++fbd) { + u32 pll, diff; + + pll = (u32)div_u64((u64)refclk * (fbd + 1), divisor); + + if (pll >= max_pll || pll < min_pll) + continue; + + diff = max(pll, target_pll) - min(pll, target_pll); + + if (diff < best_diff) { + best_diff = diff; + best_pll = pll; + best_prd = prd; + best_fbd = fbd; + + if (best_diff == 0) + goto found; + } + } + } + + if (best_diff == UINT_MAX) { + dev_err(priv->dev, "could not find suitable PLL setup\n"); + return -EINVAL; + } + +found: + if (verify_only) + return 0; + + priv->fbd = best_fbd; + priv->prd = best_prd; + priv->frs = frs; + priv->dsiclk = best_pll / 2; + + return 0; +} + +static int tc358768_dsi_host_attach(struct mipi_dsi_host *host, + struct mipi_dsi_device *dev) +{ + struct tc358768_priv *priv = dsi_host_to_tc358768(host); + struct drm_bridge *bridge; + struct drm_panel *panel; + struct device_node *ep; + int ret; + + if (dev->lanes > 4) { + dev_err(priv->dev, "unsupported number of data lanes(%u)\n", + dev->lanes); + return -EINVAL; + } + + /* + * tc358768 supports both Video and Pulse mode, but the driver only + * implements Video (event) mode currently + */ + if (!(dev->mode_flags & MIPI_DSI_MODE_VIDEO)) { + dev_err(priv->dev, "Only MIPI_DSI_MODE_VIDEO is supported\n"); + return -ENOTSUPP; + } + + /* + * tc358768 supports RGB888, RGB666, RGB666_PACKED and RGB565, but only + * RGB888 is verified. + */ + if (dev->format != MIPI_DSI_FMT_RGB888) { + dev_warn(priv->dev, "Only MIPI_DSI_FMT_RGB888 tested!\n"); + return -ENOTSUPP; + } + + ret = drm_of_find_panel_or_bridge(host->dev->of_node, 1, 0, &panel, + &bridge); + if (ret) + return ret; + + if (panel) { + bridge = drm_panel_bridge_add_typed(panel, + DRM_MODE_CONNECTOR_DSI); + if (IS_ERR(bridge)) + return PTR_ERR(bridge); + } + + priv->output.dev = dev; + priv->output.bridge = bridge; + priv->output.panel = panel; + + priv->dsi_lanes = dev->lanes; + + /* get input ep (port0/endpoint0) */ + ret = -EINVAL; + ep = of_graph_get_endpoint_by_regs(host->dev->of_node, 0, 0); + if (ep) { + ret = of_property_read_u32(ep, "data-lines", &priv->pd_lines); + + of_node_put(ep); + } + + if (ret) + priv->pd_lines = mipi_dsi_pixel_format_to_bpp(dev->format); + + drm_bridge_add(&priv->bridge); + + return 0; +} + +static int tc358768_dsi_host_detach(struct mipi_dsi_host *host, + struct mipi_dsi_device *dev) +{ + struct tc358768_priv *priv = dsi_host_to_tc358768(host); + + drm_bridge_remove(&priv->bridge); + if (priv->output.panel) + drm_panel_bridge_remove(priv->output.bridge); + + return 0; +} + +static ssize_t tc358768_dsi_host_transfer(struct mipi_dsi_host *host, + const struct mipi_dsi_msg *msg) +{ + struct tc358768_priv *priv = dsi_host_to_tc358768(host); + struct mipi_dsi_packet packet; + int ret; + + if (!priv->enabled) { + dev_err(priv->dev, "Bridge is not enabled\n"); + return -ENODEV; + } + + if (msg->rx_len) { + dev_warn(priv->dev, "MIPI rx is not supported\n"); + return -ENOTSUPP; + } + + if (msg->tx_len > 8) { + dev_warn(priv->dev, "Maximum 8 byte MIPI tx is supported\n"); + return -ENOTSUPP; + } + + ret = mipi_dsi_create_packet(&packet, msg); + if (ret) + return ret; + + if (mipi_dsi_packet_format_is_short(msg->type)) { + tc358768_write(priv, TC358768_DSICMD_TYPE, + (0x10 << 8) | (packet.header[0] & 0x3f)); + tc358768_write(priv, TC358768_DSICMD_WC, 0); + tc358768_write(priv, TC358768_DSICMD_WD0, + (packet.header[2] << 8) | packet.header[1]); + } else { + int i; + + tc358768_write(priv, TC358768_DSICMD_TYPE, + (0x40 << 8) | (packet.header[0] & 0x3f)); + tc358768_write(priv, TC358768_DSICMD_WC, packet.payload_length); + for (i = 0; i < packet.payload_length; i += 2) { + u16 val = packet.payload[i]; + + if (i + 1 < packet.payload_length) + val |= packet.payload[i + 1] << 8; + + tc358768_write(priv, TC358768_DSICMD_WD0 + i, val); + } + } + + /* start transfer */ + tc358768_write(priv, TC358768_DSICMD_TX, 1); + + ret = tc358768_clear_error(priv); + if (ret) + dev_warn(priv->dev, "Software disable failed: %d\n", ret); + else + ret = packet.size; + + return ret; +} + +static const struct mipi_dsi_host_ops tc358768_dsi_host_ops = { + .attach = tc358768_dsi_host_attach, + .detach = tc358768_dsi_host_detach, + .transfer = tc358768_dsi_host_transfer, +}; + +static int tc358768_bridge_attach(struct drm_bridge *bridge, + enum drm_bridge_attach_flags flags) +{ + struct tc358768_priv *priv = bridge_to_tc358768(bridge); + + if (!drm_core_check_feature(bridge->dev, DRIVER_ATOMIC)) { + dev_err(priv->dev, "needs atomic updates support\n"); + return -ENOTSUPP; + } + + return drm_bridge_attach(bridge->encoder, priv->output.bridge, bridge, + flags); +} + +static enum drm_mode_status +tc358768_bridge_mode_valid(struct drm_bridge *bridge, + const struct drm_display_mode *mode) +{ + struct tc358768_priv *priv = bridge_to_tc358768(bridge); + + if (tc358768_calc_pll(priv, mode, true)) + return MODE_CLOCK_RANGE; + + return MODE_OK; +} + +static void tc358768_bridge_disable(struct drm_bridge *bridge) +{ + struct tc358768_priv *priv = bridge_to_tc358768(bridge); + int ret; + + /* set FrmStop */ + tc358768_update_bits(priv, TC358768_PP_MISC, BIT(15), BIT(15)); + + /* wait at least for one frame */ + msleep(50); + + /* clear PP_en */ + tc358768_update_bits(priv, TC358768_CONFCTL, BIT(6), 0); + + /* set RstPtr */ + tc358768_update_bits(priv, TC358768_PP_MISC, BIT(14), BIT(14)); + + ret = tc358768_clear_error(priv); + if (ret) + dev_warn(priv->dev, "Software disable failed: %d\n", ret); +} + +static void tc358768_bridge_post_disable(struct drm_bridge *bridge) +{ + struct tc358768_priv *priv = bridge_to_tc358768(bridge); + + tc358768_hw_disable(priv); +} + +static int tc358768_setup_pll(struct tc358768_priv *priv, + const struct drm_display_mode *mode) +{ + u32 fbd, prd, frs; + int ret; + + ret = tc358768_calc_pll(priv, mode, false); + if (ret) { + dev_err(priv->dev, "PLL calculation failed: %d\n", ret); + return ret; + } + + fbd = priv->fbd; + prd = priv->prd; + frs = priv->frs; + + dev_dbg(priv->dev, "PLL: refclk %lu, fbd %u, prd %u, frs %u\n", + clk_get_rate(priv->refclk), fbd, prd, frs); + dev_dbg(priv->dev, "PLL: pll_clk: %u, DSIClk %u, DSIByteClk %u\n", + priv->dsiclk * 2, priv->dsiclk, priv->dsiclk / 4); + dev_dbg(priv->dev, "PLL: pclk %u (panel: %u)\n", + tc358768_pll_to_pclk(priv, priv->dsiclk * 2), + mode->clock * 1000); + + /* PRD[15:12] FBD[8:0] */ + tc358768_write(priv, TC358768_PLLCTL0, (prd << 12) | fbd); + + /* FRS[11:10] LBWS[9:8] CKEN[4] RESETB[1] EN[0] */ + tc358768_write(priv, TC358768_PLLCTL1, + (frs << 10) | (0x2 << 8) | BIT(1) | BIT(0)); + + /* wait for lock */ + usleep_range(1000, 2000); + + /* FRS[11:10] LBWS[9:8] CKEN[4] PLL_CKEN[4] RESETB[1] EN[0] */ + tc358768_write(priv, TC358768_PLLCTL1, + (frs << 10) | (0x2 << 8) | BIT(4) | BIT(1) | BIT(0)); + + return tc358768_clear_error(priv); +} + +#define TC358768_PRECISION 1000 +static u32 tc358768_ns_to_cnt(u32 ns, u32 period_nsk) +{ + return (ns * TC358768_PRECISION + period_nsk) / period_nsk; +} + +static u32 tc358768_to_ns(u32 nsk) +{ + return (nsk / TC358768_PRECISION); +} + +static void tc358768_bridge_pre_enable(struct drm_bridge *bridge) +{ + struct tc358768_priv *priv = bridge_to_tc358768(bridge); + struct mipi_dsi_device *dsi_dev = priv->output.dev; + u32 val, val2, lptxcnt, hact, data_type; + const struct drm_display_mode *mode; + u32 dsibclk_nsk, dsiclk_nsk, ui_nsk, phy_delay_nsk; + u32 dsiclk, dsibclk; + int ret, i; + + tc358768_hw_enable(priv); + + ret = tc358768_sw_reset(priv); + if (ret) { + dev_err(priv->dev, "Software reset failed: %d\n", ret); + tc358768_hw_disable(priv); + return; + } + + mode = &bridge->encoder->crtc->state->adjusted_mode; + ret = tc358768_setup_pll(priv, mode); + if (ret) { + dev_err(priv->dev, "PLL setup failed: %d\n", ret); + tc358768_hw_disable(priv); + return; + } + + dsiclk = priv->dsiclk; + dsibclk = dsiclk / 4; + + /* Data Format Control Register */ + val = BIT(2) | BIT(1) | BIT(0); /* rdswap_en | dsitx_en | txdt_en */ + switch (dsi_dev->format) { + case MIPI_DSI_FMT_RGB888: + val |= (0x3 << 4); + hact = mode->hdisplay * 3; + data_type = MIPI_DSI_PACKED_PIXEL_STREAM_24; + break; + case MIPI_DSI_FMT_RGB666: + val |= (0x4 << 4); + hact = mode->hdisplay * 3; + data_type = MIPI_DSI_PACKED_PIXEL_STREAM_18; + break; + + case MIPI_DSI_FMT_RGB666_PACKED: + val |= (0x4 << 4) | BIT(3); + hact = mode->hdisplay * 18 / 8; + data_type = MIPI_DSI_PIXEL_STREAM_3BYTE_18; + break; + + case MIPI_DSI_FMT_RGB565: + val |= (0x5 << 4); + hact = mode->hdisplay * 2; + data_type = MIPI_DSI_PACKED_PIXEL_STREAM_16; + break; + default: + dev_err(priv->dev, "Invalid data format (%u)\n", + dsi_dev->format); + tc358768_hw_disable(priv); + return; + } + + /* VSDly[9:0] */ + tc358768_write(priv, TC358768_VSDLY, 1); + + tc358768_write(priv, TC358768_DATAFMT, val); + tc358768_write(priv, TC358768_DSITX_DT, data_type); + + /* Enable D-PHY (HiZ->LP11) */ + tc358768_write(priv, TC358768_CLW_CNTRL, 0x0000); + /* Enable lanes */ + for (i = 0; i < dsi_dev->lanes; i++) + tc358768_write(priv, TC358768_D0W_CNTRL + i * 4, 0x0000); + + /* DSI Timings */ + dsibclk_nsk = (u32)div_u64((u64)1000000000 * TC358768_PRECISION, + dsibclk); + dsiclk_nsk = (u32)div_u64((u64)1000000000 * TC358768_PRECISION, dsiclk); + ui_nsk = dsiclk_nsk / 2; + phy_delay_nsk = dsibclk_nsk + 2 * dsiclk_nsk; + dev_dbg(priv->dev, "dsiclk_nsk: %u\n", dsiclk_nsk); + dev_dbg(priv->dev, "ui_nsk: %u\n", ui_nsk); + dev_dbg(priv->dev, "dsibclk_nsk: %u\n", dsibclk_nsk); + dev_dbg(priv->dev, "phy_delay_nsk: %u\n", phy_delay_nsk); + + /* LP11 > 100us for D-PHY Rx Init */ + val = tc358768_ns_to_cnt(100 * 1000, dsibclk_nsk) - 1; + dev_dbg(priv->dev, "LINEINITCNT: 0x%x\n", val); + tc358768_write(priv, TC358768_LINEINITCNT, val); + + /* LPTimeCnt > 50ns */ + val = tc358768_ns_to_cnt(50, dsibclk_nsk) - 1; + lptxcnt = val; + dev_dbg(priv->dev, "LPTXTIMECNT: 0x%x\n", val); + tc358768_write(priv, TC358768_LPTXTIMECNT, val); + + /* 38ns < TCLK_PREPARE < 95ns */ + val = tc358768_ns_to_cnt(65, dsibclk_nsk) - 1; + /* TCLK_PREPARE > 300ns */ + val2 = tc358768_ns_to_cnt(300 + tc358768_to_ns(3 * ui_nsk), + dsibclk_nsk); + val |= (val2 - tc358768_to_ns(phy_delay_nsk - dsibclk_nsk)) << 8; + dev_dbg(priv->dev, "TCLK_HEADERCNT: 0x%x\n", val); + tc358768_write(priv, TC358768_TCLK_HEADERCNT, val); + + /* TCLK_TRAIL > 60ns + 3*UI */ + val = 60 + tc358768_to_ns(3 * ui_nsk); + val = tc358768_ns_to_cnt(val, dsibclk_nsk) - 5; + dev_dbg(priv->dev, "TCLK_TRAILCNT: 0x%x\n", val); + tc358768_write(priv, TC358768_TCLK_TRAILCNT, val); + + /* 40ns + 4*UI < THS_PREPARE < 85ns + 6*UI */ + val = 50 + tc358768_to_ns(4 * ui_nsk); + val = tc358768_ns_to_cnt(val, dsibclk_nsk) - 1; + /* THS_ZERO > 145ns + 10*UI */ + val2 = tc358768_ns_to_cnt(145 - tc358768_to_ns(ui_nsk), dsibclk_nsk); + val |= (val2 - tc358768_to_ns(phy_delay_nsk)) << 8; + dev_dbg(priv->dev, "THS_HEADERCNT: 0x%x\n", val); + tc358768_write(priv, TC358768_THS_HEADERCNT, val); + + /* TWAKEUP > 1ms in lptxcnt steps */ + val = tc358768_ns_to_cnt(1020000, dsibclk_nsk); + val = val / (lptxcnt + 1) - 1; + dev_dbg(priv->dev, "TWAKEUP: 0x%x\n", val); + tc358768_write(priv, TC358768_TWAKEUP, val); + + /* TCLK_POSTCNT > 60ns + 52*UI */ + val = tc358768_ns_to_cnt(60 + tc358768_to_ns(52 * ui_nsk), + dsibclk_nsk) - 3; + dev_dbg(priv->dev, "TCLK_POSTCNT: 0x%x\n", val); + tc358768_write(priv, TC358768_TCLK_POSTCNT, val); + + /* 60ns + 4*UI < THS_PREPARE < 105ns + 12*UI */ + val = tc358768_ns_to_cnt(60 + tc358768_to_ns(15 * ui_nsk), + dsibclk_nsk) - 5; + dev_dbg(priv->dev, "THS_TRAILCNT: 0x%x\n", val); + tc358768_write(priv, TC358768_THS_TRAILCNT, val); + + val = BIT(0); + for (i = 0; i < dsi_dev->lanes; i++) + val |= BIT(i + 1); + tc358768_write(priv, TC358768_HSTXVREGEN, val); + + if (!(dsi_dev->mode_flags & MIPI_DSI_CLOCK_NON_CONTINUOUS)) + tc358768_write(priv, TC358768_TXOPTIONCNTRL, 0x1); + + /* TXTAGOCNT[26:16] RXTASURECNT[10:0] */ + val = tc358768_to_ns((lptxcnt + 1) * dsibclk_nsk * 4); + val = tc358768_ns_to_cnt(val, dsibclk_nsk) - 1; + val2 = tc358768_ns_to_cnt(tc358768_to_ns((lptxcnt + 1) * dsibclk_nsk), + dsibclk_nsk) - 2; + val |= val2 << 16; + dev_dbg(priv->dev, "BTACNTRL1: 0x%x\n", val); + tc358768_write(priv, TC358768_BTACNTRL1, val); + + /* START[0] */ + tc358768_write(priv, TC358768_STARTCNTRL, 1); + + /* Set event mode */ + tc358768_write(priv, TC358768_DSI_EVENT, 1); + + /* vsw (+ vbp) */ + tc358768_write(priv, TC358768_DSI_VSW, + mode->vtotal - mode->vsync_start); + /* vbp (not used in event mode) */ + tc358768_write(priv, TC358768_DSI_VBPR, 0); + /* vact */ + tc358768_write(priv, TC358768_DSI_VACT, mode->vdisplay); + + /* (hsw + hbp) * byteclk * ndl / pclk */ + val = (u32)div_u64((mode->htotal - mode->hsync_start) * + ((u64)priv->dsiclk / 4) * priv->dsi_lanes, + mode->clock * 1000); + tc358768_write(priv, TC358768_DSI_HSW, val); + /* hbp (not used in event mode) */ + tc358768_write(priv, TC358768_DSI_HBPR, 0); + /* hact (bytes) */ + tc358768_write(priv, TC358768_DSI_HACT, hact); + + /* VSYNC polarity */ + if (!(mode->flags & DRM_MODE_FLAG_NVSYNC)) + tc358768_update_bits(priv, TC358768_CONFCTL, BIT(5), BIT(5)); + /* HSYNC polarity */ + if (mode->flags & DRM_MODE_FLAG_PHSYNC) + tc358768_update_bits(priv, TC358768_PP_MISC, BIT(0), BIT(0)); + + /* Start DSI Tx */ + tc358768_write(priv, TC358768_DSI_START, 0x1); + + /* Configure DSI_Control register */ + val = TC358768_DSI_CONFW_MODE_CLR | TC358768_DSI_CONFW_ADDR_DSI_CONTROL; + val |= TC358768_DSI_CONTROL_TXMD | TC358768_DSI_CONTROL_HSCKMD | + 0x3 << 1 | TC358768_DSI_CONTROL_EOTDIS; + tc358768_write(priv, TC358768_DSI_CONFW, val); + + val = TC358768_DSI_CONFW_MODE_SET | TC358768_DSI_CONFW_ADDR_DSI_CONTROL; + val |= (dsi_dev->lanes - 1) << 1; + + if (!(dsi_dev->mode_flags & MIPI_DSI_MODE_LPM)) + val |= TC358768_DSI_CONTROL_TXMD; + + if (!(dsi_dev->mode_flags & MIPI_DSI_CLOCK_NON_CONTINUOUS)) + val |= TC358768_DSI_CONTROL_HSCKMD; + + if (dsi_dev->mode_flags & MIPI_DSI_MODE_EOT_PACKET) + val |= TC358768_DSI_CONTROL_EOTDIS; + + tc358768_write(priv, TC358768_DSI_CONFW, val); + + val = TC358768_DSI_CONFW_MODE_CLR | TC358768_DSI_CONFW_ADDR_DSI_CONTROL; + val |= TC358768_DSI_CONTROL_DIS_MODE; /* DSI mode */ + tc358768_write(priv, TC358768_DSI_CONFW, val); + + ret = tc358768_clear_error(priv); + if (ret) { + dev_err(priv->dev, "Bridge pre_enable failed: %d\n", ret); + tc358768_bridge_disable(bridge); + tc358768_bridge_post_disable(bridge); + } +} + +static void tc358768_bridge_enable(struct drm_bridge *bridge) +{ + struct tc358768_priv *priv = bridge_to_tc358768(bridge); + int ret; + + if (!priv->enabled) { + dev_err(priv->dev, "Bridge is not enabled\n"); + return; + } + + /* clear FrmStop and RstPtr */ + tc358768_update_bits(priv, TC358768_PP_MISC, 0x3 << 14, 0); + + /* set PP_en */ + tc358768_update_bits(priv, TC358768_CONFCTL, BIT(6), BIT(6)); + + ret = tc358768_clear_error(priv); + if (ret) { + dev_err(priv->dev, "Bridge enable failed: %d\n", ret); + tc358768_bridge_disable(bridge); + tc358768_bridge_post_disable(bridge); + } +} + +static const struct drm_bridge_funcs tc358768_bridge_funcs = { + .attach = tc358768_bridge_attach, + .mode_valid = tc358768_bridge_mode_valid, + .pre_enable = tc358768_bridge_pre_enable, + .enable = tc358768_bridge_enable, + .disable = tc358768_bridge_disable, + .post_disable = tc358768_bridge_post_disable, +}; + +static const struct drm_bridge_timings default_tc358768_timings = { + .input_bus_flags = DRM_BUS_FLAG_PIXDATA_SAMPLE_POSEDGE + | DRM_BUS_FLAG_SYNC_SAMPLE_NEGEDGE + | DRM_BUS_FLAG_DE_HIGH, +}; + +static bool tc358768_is_reserved_reg(unsigned int reg) +{ + switch (reg) { + case 0x114 ... 0x13f: + case 0x200: + case 0x20c: + case 0x400 ... 0x408: + case 0x41c ... 0x42f: + return true; + default: + return false; + } +} + +static bool tc358768_writeable_reg(struct device *dev, unsigned int reg) +{ + if (tc358768_is_reserved_reg(reg)) + return false; + + switch (reg) { + case TC358768_CHIPID: + case TC358768_FIFOSTATUS: + case TC358768_DSITXSTATUS ... (TC358768_DSITXSTATUS + 2): + case TC358768_DSI_CONTROL ... (TC358768_DSI_INT_ENA + 2): + case TC358768_DSICMD_RDFIFO ... (TC358768_DSI_ERR_HALT + 2): + return false; + default: + return true; + } +} + +static bool tc358768_readable_reg(struct device *dev, unsigned int reg) +{ + if (tc358768_is_reserved_reg(reg)) + return false; + + switch (reg) { + case TC358768_STARTCNTRL: + case TC358768_DSI_CONFW ... (TC358768_DSI_CONFW + 2): + case TC358768_DSI_INT_CLR ... (TC358768_DSI_INT_CLR + 2): + case TC358768_DSI_START ... (TC358768_DSI_START + 2): + case TC358768_DBG_DATA: + return false; + default: + return true; + } +} + +static const struct regmap_config tc358768_regmap_config = { + .name = "tc358768", + .reg_bits = 16, + .val_bits = 16, + .max_register = TC358768_DSI_HACT, + .cache_type = REGCACHE_NONE, + .writeable_reg = tc358768_writeable_reg, + .readable_reg = tc358768_readable_reg, + .reg_format_endian = REGMAP_ENDIAN_BIG, + .val_format_endian = REGMAP_ENDIAN_BIG, +}; + +static const struct i2c_device_id tc358768_i2c_ids[] = { + { "tc358768", 0 }, + { "tc358778", 0 }, + { } +}; +MODULE_DEVICE_TABLE(i2c, tc358768_i2c_ids); + +static const struct of_device_id tc358768_of_ids[] = { + { .compatible = "toshiba,tc358768", }, + { .compatible = "toshiba,tc358778", }, + { } +}; +MODULE_DEVICE_TABLE(of, tc358768_of_ids); + +static int tc358768_get_regulators(struct tc358768_priv *priv) +{ + int i, ret; + + for (i = 0; i < ARRAY_SIZE(priv->supplies); ++i) + priv->supplies[i].supply = tc358768_supplies[i]; + + ret = devm_regulator_bulk_get(priv->dev, ARRAY_SIZE(priv->supplies), + priv->supplies); + if (ret < 0) + dev_err(priv->dev, "failed to get regulators: %d\n", ret); + + return ret; +} + +static int tc358768_i2c_probe(struct i2c_client *client, + const struct i2c_device_id *id) +{ + struct tc358768_priv *priv; + struct device *dev = &client->dev; + struct device_node *np = dev->of_node; + int ret; + + if (!np) + return -ENODEV; + + priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL); + if (!priv) + return -ENOMEM; + + dev_set_drvdata(dev, priv); + priv->dev = dev; + + ret = tc358768_get_regulators(priv); + if (ret) + return ret; + + priv->refclk = devm_clk_get(dev, "refclk"); + if (IS_ERR(priv->refclk)) + return PTR_ERR(priv->refclk); + + /* + * RESX is low active, to disable tc358768 initially (keep in reset) + * the gpio line must be LOW. This is the ASSERTED state of + * GPIO_ACTIVE_LOW (GPIOD_OUT_HIGH == ASSERTED). + */ + priv->reset_gpio = devm_gpiod_get_optional(dev, "reset", + GPIOD_OUT_HIGH); + if (IS_ERR(priv->reset_gpio)) + return PTR_ERR(priv->reset_gpio); + + priv->regmap = devm_regmap_init_i2c(client, &tc358768_regmap_config); + if (IS_ERR(priv->regmap)) { + dev_err(dev, "Failed to init regmap\n"); + return PTR_ERR(priv->regmap); + } + + priv->dsi_host.dev = dev; + priv->dsi_host.ops = &tc358768_dsi_host_ops; + + priv->bridge.funcs = &tc358768_bridge_funcs; + priv->bridge.timings = &default_tc358768_timings; + priv->bridge.of_node = np; + + i2c_set_clientdata(client, priv); + + return mipi_dsi_host_register(&priv->dsi_host); +} + +static int tc358768_i2c_remove(struct i2c_client *client) +{ + struct tc358768_priv *priv = i2c_get_clientdata(client); + + mipi_dsi_host_unregister(&priv->dsi_host); + + return 0; +} + +static struct i2c_driver tc358768_driver = { + .driver = { + .name = "tc358768", + .of_match_table = tc358768_of_ids, + }, + .id_table = tc358768_i2c_ids, + .probe = tc358768_i2c_probe, + .remove = tc358768_i2c_remove, +}; +module_i2c_driver(tc358768_driver); + +MODULE_AUTHOR("Peter Ujfalusi <peter.ujfalusi@ti.com>"); +MODULE_DESCRIPTION("TC358768AXBG/TC358778XBG DSI bridge"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/gpu/drm/bridge/thc63lvd1024.c b/drivers/gpu/drm/bridge/thc63lvd1024.c index 3d74129b2995..97d8129760e9 100644 --- a/drivers/gpu/drm/bridge/thc63lvd1024.c +++ b/drivers/gpu/drm/bridge/thc63lvd1024.c @@ -42,11 +42,12 @@ static inline struct thc63_dev *to_thc63(struct drm_bridge *bridge) return container_of(bridge, struct thc63_dev, bridge); } -static int thc63_attach(struct drm_bridge *bridge) +static int thc63_attach(struct drm_bridge *bridge, + enum drm_bridge_attach_flags flags) { struct thc63_dev *thc63 = to_thc63(bridge); - return drm_bridge_attach(bridge->encoder, thc63->next, bridge); + return drm_bridge_attach(bridge->encoder, thc63->next, bridge, flags); } static enum drm_mode_status thc63_mode_valid(struct drm_bridge *bridge, diff --git a/drivers/gpu/drm/bridge/ti-sn65dsi86.c b/drivers/gpu/drm/bridge/ti-sn65dsi86.c index 9a2dd986afa5..6ad688b320ae 100644 --- a/drivers/gpu/drm/bridge/ti-sn65dsi86.c +++ b/drivers/gpu/drm/bridge/ti-sn65dsi86.c @@ -51,6 +51,7 @@ #define SN_ENH_FRAME_REG 0x5A #define VSTREAM_ENABLE BIT(3) #define SN_DATA_FORMAT_REG 0x5B +#define BPP_18_RGB BIT(0) #define SN_HPD_DISABLE_REG 0x5C #define HPD_DISABLE BIT(0) #define SN_AUX_WDATA_REG(x) (0x64 + (x)) @@ -100,6 +101,7 @@ struct ti_sn_bridge { struct drm_panel *panel; struct gpio_desc *enable_gpio; struct regulator_bulk_data supplies[SN_REGULATOR_SUPPLY_NUM]; + int dp_lanes; }; static const struct regmap_range ti_sn_bridge_volatile_ranges[] = { @@ -264,7 +266,8 @@ static int ti_sn_bridge_parse_regulators(struct ti_sn_bridge *pdata) pdata->supplies); } -static int ti_sn_bridge_attach(struct drm_bridge *bridge) +static int ti_sn_bridge_attach(struct drm_bridge *bridge, + enum drm_bridge_attach_flags flags) { int ret, val; struct ti_sn_bridge *pdata = bridge_to_ti_sn_bridge(bridge); @@ -275,6 +278,11 @@ static int ti_sn_bridge_attach(struct drm_bridge *bridge) .node = NULL, }; + if (flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR) { + DRM_ERROR("Fix bridge driver to make connector optional!"); + return -EINVAL; + } + ret = drm_connector_init(bridge->dev, &pdata->connector, &ti_sn_bridge_connector_funcs, DRM_MODE_CONNECTOR_eDP); @@ -312,7 +320,7 @@ static int ti_sn_bridge_attach(struct drm_bridge *bridge) goto err_dsi_host; } - /* TODO: setting to 4 lanes always for now */ + /* TODO: setting to 4 MIPI lanes always for now */ dsi->lanes = 4; dsi->format = MIPI_DSI_FMT_RGB888; dsi->mode_flags = MIPI_DSI_MODE_VIDEO; @@ -417,6 +425,32 @@ static void ti_sn_bridge_set_refclk_freq(struct ti_sn_bridge *pdata) REFCLK_FREQ(i)); } +static void ti_sn_bridge_set_dsi_rate(struct ti_sn_bridge *pdata) +{ + unsigned int bit_rate_mhz, clk_freq_mhz; + unsigned int val; + struct drm_display_mode *mode = + &pdata->bridge.encoder->crtc->state->adjusted_mode; + + /* set DSIA clk frequency */ + bit_rate_mhz = (mode->clock / 1000) * + mipi_dsi_pixel_format_to_bpp(pdata->dsi->format); + clk_freq_mhz = bit_rate_mhz / (pdata->dsi->lanes * 2); + + /* for each increment in val, frequency increases by 5MHz */ + val = (MIN_DSI_CLK_FREQ_MHZ / 5) + + (((clk_freq_mhz - MIN_DSI_CLK_FREQ_MHZ) / 5) & 0xFF); + regmap_write(pdata->regmap, SN_DSIA_CLK_FREQ_REG, val); +} + +static unsigned int ti_sn_bridge_get_bpp(struct ti_sn_bridge *pdata) +{ + if (pdata->connector.display_info.bpc <= 6) + return 18; + else + return 24; +} + /** * LUT index corresponds to register value and * LUT values corresponds to dp data rate supported @@ -426,32 +460,106 @@ static const unsigned int ti_sn_bridge_dp_rate_lut[] = { 0, 1620, 2160, 2430, 2700, 3240, 4320, 5400 }; -static void ti_sn_bridge_set_dsi_dp_rate(struct ti_sn_bridge *pdata) +static int ti_sn_bridge_calc_min_dp_rate_idx(struct ti_sn_bridge *pdata) { - unsigned int bit_rate_mhz, clk_freq_mhz, dp_rate_mhz; - unsigned int val, i; + unsigned int bit_rate_khz, dp_rate_mhz; + unsigned int i; struct drm_display_mode *mode = &pdata->bridge.encoder->crtc->state->adjusted_mode; - /* set DSIA clk frequency */ - bit_rate_mhz = (mode->clock / 1000) * - mipi_dsi_pixel_format_to_bpp(pdata->dsi->format); - clk_freq_mhz = bit_rate_mhz / (pdata->dsi->lanes * 2); + /* Calculate minimum bit rate based on our pixel clock. */ + bit_rate_khz = mode->clock * ti_sn_bridge_get_bpp(pdata); - /* for each increment in val, frequency increases by 5MHz */ - val = (MIN_DSI_CLK_FREQ_MHZ / 5) + - (((clk_freq_mhz - MIN_DSI_CLK_FREQ_MHZ) / 5) & 0xFF); - regmap_write(pdata->regmap, SN_DSIA_CLK_FREQ_REG, val); + /* Calculate minimum DP data rate, taking 80% as per DP spec */ + dp_rate_mhz = DIV_ROUND_UP(bit_rate_khz * DP_CLK_FUDGE_NUM, + 1000 * pdata->dp_lanes * DP_CLK_FUDGE_DEN); - /* set DP data rate */ - dp_rate_mhz = ((bit_rate_mhz / pdata->dsi->lanes) * DP_CLK_FUDGE_NUM) / - DP_CLK_FUDGE_DEN; - for (i = 0; i < ARRAY_SIZE(ti_sn_bridge_dp_rate_lut) - 1; i++) + for (i = 1; i < ARRAY_SIZE(ti_sn_bridge_dp_rate_lut) - 1; i++) if (ti_sn_bridge_dp_rate_lut[i] > dp_rate_mhz) break; - regmap_update_bits(pdata->regmap, SN_DATARATE_CONFIG_REG, - DP_DATARATE_MASK, DP_DATARATE(i)); + return i; +} + +static void ti_sn_bridge_read_valid_rates(struct ti_sn_bridge *pdata, + bool rate_valid[]) +{ + unsigned int rate_per_200khz; + unsigned int rate_mhz; + u8 dpcd_val; + int ret; + int i, j; + + ret = drm_dp_dpcd_readb(&pdata->aux, DP_EDP_DPCD_REV, &dpcd_val); + if (ret != 1) { + DRM_DEV_ERROR(pdata->dev, + "Can't read eDP rev (%d), assuming 1.1\n", ret); + dpcd_val = DP_EDP_11; + } + + if (dpcd_val >= DP_EDP_14) { + /* eDP 1.4 devices must provide a custom table */ + __le16 sink_rates[DP_MAX_SUPPORTED_RATES]; + + ret = drm_dp_dpcd_read(&pdata->aux, DP_SUPPORTED_LINK_RATES, + sink_rates, sizeof(sink_rates)); + + if (ret != sizeof(sink_rates)) { + DRM_DEV_ERROR(pdata->dev, + "Can't read supported rate table (%d)\n", ret); + + /* By zeroing we'll fall back to DP_MAX_LINK_RATE. */ + memset(sink_rates, 0, sizeof(sink_rates)); + } + + for (i = 0; i < ARRAY_SIZE(sink_rates); i++) { + rate_per_200khz = le16_to_cpu(sink_rates[i]); + + if (!rate_per_200khz) + break; + + rate_mhz = rate_per_200khz * 200 / 1000; + for (j = 0; + j < ARRAY_SIZE(ti_sn_bridge_dp_rate_lut); + j++) { + if (ti_sn_bridge_dp_rate_lut[j] == rate_mhz) + rate_valid[j] = true; + } + } + + for (i = 0; i < ARRAY_SIZE(ti_sn_bridge_dp_rate_lut); i++) { + if (rate_valid[i]) + return; + } + DRM_DEV_ERROR(pdata->dev, + "No matching eDP rates in table; falling back\n"); + } + + /* On older versions best we can do is use DP_MAX_LINK_RATE */ + ret = drm_dp_dpcd_readb(&pdata->aux, DP_MAX_LINK_RATE, &dpcd_val); + if (ret != 1) { + DRM_DEV_ERROR(pdata->dev, + "Can't read max rate (%d); assuming 5.4 GHz\n", + ret); + dpcd_val = DP_LINK_BW_5_4; + } + + switch (dpcd_val) { + default: + DRM_DEV_ERROR(pdata->dev, + "Unexpected max rate (%#x); assuming 5.4 GHz\n", + (int)dpcd_val); + /* fall through */ + case DP_LINK_BW_5_4: + rate_valid[7] = 1; + /* fall through */ + case DP_LINK_BW_2_7: + rate_valid[4] = 1; + /* fall through */ + case DP_LINK_BW_1_62: + rate_valid[1] = 1; + break; + } } static void ti_sn_bridge_set_video_timings(struct ti_sn_bridge *pdata) @@ -493,24 +601,30 @@ static void ti_sn_bridge_set_video_timings(struct ti_sn_bridge *pdata) usleep_range(10000, 10500); /* 10ms delay recommended by spec */ } -static void ti_sn_bridge_enable(struct drm_bridge *bridge) +static unsigned int ti_sn_get_max_lanes(struct ti_sn_bridge *pdata) { - struct ti_sn_bridge *pdata = bridge_to_ti_sn_bridge(bridge); - unsigned int val; + u8 data; int ret; - /* DSI_A lane config */ - val = CHA_DSI_LANES(4 - pdata->dsi->lanes); - regmap_update_bits(pdata->regmap, SN_DSI_LANES_REG, - CHA_DSI_LANES_MASK, val); + ret = drm_dp_dpcd_readb(&pdata->aux, DP_MAX_LANE_COUNT, &data); + if (ret != 1) { + DRM_DEV_ERROR(pdata->dev, + "Can't read lane count (%d); assuming 4\n", ret); + return 4; + } - /* DP lane config */ - val = DP_NUM_LANES(pdata->dsi->lanes - 1); - regmap_update_bits(pdata->regmap, SN_SSC_CONFIG_REG, DP_NUM_LANES_MASK, - val); + return data & DP_LANE_COUNT_MASK; +} - /* set dsi/dp clk frequency value */ - ti_sn_bridge_set_dsi_dp_rate(pdata); +static int ti_sn_link_training(struct ti_sn_bridge *pdata, int dp_rate_idx, + const char **last_err_str) +{ + unsigned int val; + int ret; + + /* set dp clk frequency value */ + regmap_update_bits(pdata->regmap, SN_DATARATE_CONFIG_REG, + DP_DATARATE_MASK, DP_DATARATE(dp_rate_idx)); /* enable DP PLL */ regmap_write(pdata->regmap, SN_PLL_ENABLE_REG, 1); @@ -519,10 +633,62 @@ static void ti_sn_bridge_enable(struct drm_bridge *bridge) val & DPPLL_SRC_DP_PLL_LOCK, 1000, 50 * 1000); if (ret) { - DRM_ERROR("DP_PLL_LOCK polling failed (%d)\n", ret); - return; + *last_err_str = "DP_PLL_LOCK polling failed"; + goto exit; + } + + /* Semi auto link training mode */ + regmap_write(pdata->regmap, SN_ML_TX_MODE_REG, 0x0A); + ret = regmap_read_poll_timeout(pdata->regmap, SN_ML_TX_MODE_REG, val, + val == ML_TX_MAIN_LINK_OFF || + val == ML_TX_NORMAL_MODE, 1000, + 500 * 1000); + if (ret) { + *last_err_str = "Training complete polling failed"; + } else if (val == ML_TX_MAIN_LINK_OFF) { + *last_err_str = "Link training failed, link is off"; + ret = -EIO; } +exit: + /* Disable the PLL if we failed */ + if (ret) + regmap_write(pdata->regmap, SN_PLL_ENABLE_REG, 0); + + return ret; +} + +static void ti_sn_bridge_enable(struct drm_bridge *bridge) +{ + struct ti_sn_bridge *pdata = bridge_to_ti_sn_bridge(bridge); + bool rate_valid[ARRAY_SIZE(ti_sn_bridge_dp_rate_lut)] = { }; + const char *last_err_str = "No supported DP rate"; + int dp_rate_idx; + unsigned int val; + int ret = -EINVAL; + + /* + * Run with the maximum number of lanes that the DP sink supports. + * + * Depending use cases, we might want to revisit this later because: + * - It's plausible that someone may have run fewer lines to the + * sink than the sink actually supports, assuming that the lines + * will just be driven at a higher rate. + * - The DP spec seems to indicate that it's more important to minimize + * the number of lanes than the link rate. + * + * If we do revisit, it would be important to measure the power impact. + */ + pdata->dp_lanes = ti_sn_get_max_lanes(pdata); + + /* DSI_A lane config */ + val = CHA_DSI_LANES(4 - pdata->dsi->lanes); + regmap_update_bits(pdata->regmap, SN_DSI_LANES_REG, + CHA_DSI_LANES_MASK, val); + + /* set dsi clk frequency value */ + ti_sn_bridge_set_dsi_rate(pdata); + /** * The SN65DSI86 only supports ASSR Display Authentication method and * this method is enabled by default. An eDP panel must support this @@ -532,17 +698,30 @@ static void ti_sn_bridge_enable(struct drm_bridge *bridge) drm_dp_dpcd_writeb(&pdata->aux, DP_EDP_CONFIGURATION_SET, DP_ALTERNATE_SCRAMBLER_RESET_ENABLE); - /* Semi auto link training mode */ - regmap_write(pdata->regmap, SN_ML_TX_MODE_REG, 0x0A); - ret = regmap_read_poll_timeout(pdata->regmap, SN_ML_TX_MODE_REG, val, - val == ML_TX_MAIN_LINK_OFF || - val == ML_TX_NORMAL_MODE, 1000, - 500 * 1000); + /* Set the DP output format (18 bpp or 24 bpp) */ + val = (ti_sn_bridge_get_bpp(pdata) == 18) ? BPP_18_RGB : 0; + regmap_update_bits(pdata->regmap, SN_DATA_FORMAT_REG, BPP_18_RGB, val); + + /* DP lane config */ + val = DP_NUM_LANES(min(pdata->dp_lanes, 3)); + regmap_update_bits(pdata->regmap, SN_SSC_CONFIG_REG, DP_NUM_LANES_MASK, + val); + + ti_sn_bridge_read_valid_rates(pdata, rate_valid); + + /* Train until we run out of rates */ + for (dp_rate_idx = ti_sn_bridge_calc_min_dp_rate_idx(pdata); + dp_rate_idx < ARRAY_SIZE(ti_sn_bridge_dp_rate_lut); + dp_rate_idx++) { + if (!rate_valid[dp_rate_idx]) + continue; + + ret = ti_sn_link_training(pdata, dp_rate_idx, &last_err_str); + if (!ret) + break; + } if (ret) { - DRM_ERROR("Training complete polling failed (%d)\n", ret); - return; - } else if (val == ML_TX_MAIN_LINK_OFF) { - DRM_ERROR("Link training failed, link is off\n"); + DRM_DEV_ERROR(pdata->dev, "%s (%d)\n", last_err_str, ret); return; } diff --git a/drivers/gpu/drm/bridge/ti-tfp410.c b/drivers/gpu/drm/bridge/ti-tfp410.c index 6f6d6d1e60ae..e3eb6364c0f7 100644 --- a/drivers/gpu/drm/bridge/ti-tfp410.c +++ b/drivers/gpu/drm/bridge/ti-tfp410.c @@ -4,14 +4,12 @@ * Author: Jyri Sarha <jsarha@ti.com> */ -#include <linux/delay.h> -#include <linux/fwnode.h> #include <linux/gpio/consumer.h> #include <linux/i2c.h> -#include <linux/irq.h> #include <linux/module.h> #include <linux/of_graph.h> #include <linux/platform_device.h> +#include <linux/workqueue.h> #include <drm/drm_atomic_helper.h> #include <drm/drm_bridge.h> @@ -24,16 +22,13 @@ struct tfp410 { struct drm_bridge bridge; struct drm_connector connector; - unsigned int connector_type; u32 bus_format; - struct i2c_adapter *ddc; - struct gpio_desc *hpd; - int hpd_irq; struct delayed_work hpd_work; struct gpio_desc *powerdown; struct drm_bridge_timings timings; + struct drm_bridge *next_bridge; struct device *dev; }; @@ -56,13 +51,18 @@ static int tfp410_get_modes(struct drm_connector *connector) struct edid *edid; int ret; - if (!dvi->ddc) - goto fallback; + edid = drm_bridge_get_edid(dvi->next_bridge, connector); + if (IS_ERR_OR_NULL(edid)) { + if (edid != ERR_PTR(-ENOTSUPP)) + DRM_INFO("EDID read failed. Fallback to standard modes\n"); - edid = drm_get_edid(connector, dvi->ddc); - if (!edid) { - DRM_INFO("EDID read failed. Fallback to standard modes\n"); - goto fallback; + /* + * No EDID, fallback on the XGA standard modes and prefer a mode + * pretty much anything can handle. + */ + ret = drm_add_modes_noedid(connector, 1920, 1200); + drm_set_preferred_mode(connector, 1024, 768); + return ret; } drm_connector_update_edid_property(connector, edid); @@ -72,15 +72,6 @@ static int tfp410_get_modes(struct drm_connector *connector) kfree(edid); return ret; - -fallback: - /* No EDID, fallback on the XGA standard modes */ - ret = drm_add_modes_noedid(connector, 1920, 1200); - - /* And prefer a mode pretty much anything can handle */ - drm_set_preferred_mode(connector, 1024, 768); - - return ret; } static const struct drm_connector_helper_funcs tfp410_con_helper_funcs = { @@ -92,21 +83,7 @@ tfp410_connector_detect(struct drm_connector *connector, bool force) { struct tfp410 *dvi = drm_connector_to_tfp410(connector); - if (dvi->hpd) { - if (gpiod_get_value_cansleep(dvi->hpd)) - return connector_status_connected; - else - return connector_status_disconnected; - } - - if (dvi->ddc) { - if (drm_probe_ddc(dvi->ddc)) - return connector_status_connected; - else - return connector_status_disconnected; - } - - return connector_status_unknown; + return drm_bridge_detect(dvi->next_bridge); } static const struct drm_connector_funcs tfp410_con_funcs = { @@ -118,41 +95,84 @@ static const struct drm_connector_funcs tfp410_con_funcs = { .atomic_destroy_state = drm_atomic_helper_connector_destroy_state, }; -static int tfp410_attach(struct drm_bridge *bridge) +static void tfp410_hpd_work_func(struct work_struct *work) +{ + struct tfp410 *dvi; + + dvi = container_of(work, struct tfp410, hpd_work.work); + + if (dvi->bridge.dev) + drm_helper_hpd_irq_event(dvi->bridge.dev); +} + +static void tfp410_hpd_callback(void *arg, enum drm_connector_status status) +{ + struct tfp410 *dvi = arg; + + mod_delayed_work(system_wq, &dvi->hpd_work, + msecs_to_jiffies(HOTPLUG_DEBOUNCE_MS)); +} + +static int tfp410_attach(struct drm_bridge *bridge, + enum drm_bridge_attach_flags flags) { struct tfp410 *dvi = drm_bridge_to_tfp410(bridge); int ret; + ret = drm_bridge_attach(bridge->encoder, dvi->next_bridge, bridge, + DRM_BRIDGE_ATTACH_NO_CONNECTOR); + if (ret < 0) + return ret; + + if (flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR) + return 0; + if (!bridge->encoder) { dev_err(dvi->dev, "Missing encoder\n"); return -ENODEV; } - if (dvi->hpd_irq >= 0) + if (dvi->next_bridge->ops & DRM_BRIDGE_OP_DETECT) dvi->connector.polled = DRM_CONNECTOR_POLL_HPD; else dvi->connector.polled = DRM_CONNECTOR_POLL_CONNECT | DRM_CONNECTOR_POLL_DISCONNECT; + if (dvi->next_bridge->ops & DRM_BRIDGE_OP_HPD) { + INIT_DELAYED_WORK(&dvi->hpd_work, tfp410_hpd_work_func); + drm_bridge_hpd_enable(dvi->next_bridge, tfp410_hpd_callback, + dvi); + } + drm_connector_helper_add(&dvi->connector, &tfp410_con_helper_funcs); ret = drm_connector_init_with_ddc(bridge->dev, &dvi->connector, &tfp410_con_funcs, - dvi->connector_type, - dvi->ddc); + dvi->next_bridge->type, + dvi->next_bridge->ddc); if (ret) { - dev_err(dvi->dev, "drm_connector_init() failed: %d\n", ret); + dev_err(dvi->dev, "drm_connector_init_with_ddc() failed: %d\n", + ret); return ret; } drm_display_info_set_bus_formats(&dvi->connector.display_info, &dvi->bus_format, 1); - drm_connector_attach_encoder(&dvi->connector, - bridge->encoder); + drm_connector_attach_encoder(&dvi->connector, bridge->encoder); return 0; } +static void tfp410_detach(struct drm_bridge *bridge) +{ + struct tfp410 *dvi = drm_bridge_to_tfp410(bridge); + + if (dvi->connector.dev && dvi->next_bridge->ops & DRM_BRIDGE_OP_HPD) { + drm_bridge_hpd_disable(dvi->next_bridge); + cancel_delayed_work_sync(&dvi->hpd_work); + } +} + static void tfp410_enable(struct drm_bridge *bridge) { struct tfp410 *dvi = drm_bridge_to_tfp410(bridge); @@ -167,31 +187,25 @@ static void tfp410_disable(struct drm_bridge *bridge) gpiod_set_value_cansleep(dvi->powerdown, 1); } -static const struct drm_bridge_funcs tfp410_bridge_funcs = { - .attach = tfp410_attach, - .enable = tfp410_enable, - .disable = tfp410_disable, -}; - -static void tfp410_hpd_work_func(struct work_struct *work) +static enum drm_mode_status tfp410_mode_valid(struct drm_bridge *bridge, + const struct drm_display_mode *mode) { - struct tfp410 *dvi; + if (mode->clock < 25000) + return MODE_CLOCK_LOW; - dvi = container_of(work, struct tfp410, hpd_work.work); + if (mode->clock > 165000) + return MODE_CLOCK_HIGH; - if (dvi->bridge.dev) - drm_helper_hpd_irq_event(dvi->bridge.dev); + return MODE_OK; } -static irqreturn_t tfp410_hpd_irq_thread(int irq, void *arg) -{ - struct tfp410 *dvi = arg; - - mod_delayed_work(system_wq, &dvi->hpd_work, - msecs_to_jiffies(HOTPLUG_DEBOUNCE_MS)); - - return IRQ_HANDLED; -} +static const struct drm_bridge_funcs tfp410_bridge_funcs = { + .attach = tfp410_attach, + .detach = tfp410_detach, + .enable = tfp410_enable, + .disable = tfp410_disable, + .mode_valid = tfp410_mode_valid, +}; static const struct drm_bridge_timings tfp410_default_timings = { .input_bus_flags = DRM_BUS_FLAG_PIXDATA_SAMPLE_POSEDGE @@ -270,51 +284,9 @@ static int tfp410_parse_timings(struct tfp410 *dvi, bool i2c) return 0; } -static int tfp410_get_connector_properties(struct tfp410 *dvi) -{ - struct device_node *connector_node, *ddc_phandle; - int ret = 0; - - /* port@1 is the connector node */ - connector_node = of_graph_get_remote_node(dvi->dev->of_node, 1, -1); - if (!connector_node) - return -ENODEV; - - if (of_device_is_compatible(connector_node, "hdmi-connector")) - dvi->connector_type = DRM_MODE_CONNECTOR_HDMIA; - else - dvi->connector_type = DRM_MODE_CONNECTOR_DVID; - - dvi->hpd = fwnode_gpiod_get_index(&connector_node->fwnode, - "hpd", 0, GPIOD_IN, "hpd"); - if (IS_ERR(dvi->hpd)) { - ret = PTR_ERR(dvi->hpd); - dvi->hpd = NULL; - if (ret == -ENOENT) - ret = 0; - else - goto fail; - } - - ddc_phandle = of_parse_phandle(connector_node, "ddc-i2c-bus", 0); - if (!ddc_phandle) - goto fail; - - dvi->ddc = of_get_i2c_adapter_by_node(ddc_phandle); - if (dvi->ddc) - dev_info(dvi->dev, "Connector's ddc i2c bus found\n"); - else - ret = -EPROBE_DEFER; - - of_node_put(ddc_phandle); - -fail: - of_node_put(connector_node); - return ret; -} - static int tfp410_init(struct device *dev, bool i2c) { + struct device_node *node; struct tfp410 *dvi; int ret; @@ -326,21 +298,31 @@ static int tfp410_init(struct device *dev, bool i2c) dvi = devm_kzalloc(dev, sizeof(*dvi), GFP_KERNEL); if (!dvi) return -ENOMEM; + + dvi->dev = dev; dev_set_drvdata(dev, dvi); dvi->bridge.funcs = &tfp410_bridge_funcs; dvi->bridge.of_node = dev->of_node; dvi->bridge.timings = &dvi->timings; - dvi->dev = dev; + dvi->bridge.type = DRM_MODE_CONNECTOR_DVID; ret = tfp410_parse_timings(dvi, i2c); if (ret) - goto fail; + return ret; - ret = tfp410_get_connector_properties(dvi); - if (ret) - goto fail; + /* Get the next bridge, connected to port@1. */ + node = of_graph_get_remote_node(dev->of_node, 1, -1); + if (!node) + return -ENODEV; + + dvi->next_bridge = of_drm_find_bridge(node); + of_node_put(node); + if (!dvi->next_bridge) + return -EPROBE_DEFER; + + /* Get the powerdown GPIO. */ dvi->powerdown = devm_gpiod_get_optional(dev, "powerdown", GPIOD_OUT_HIGH); if (IS_ERR(dvi->powerdown)) { @@ -348,48 +330,18 @@ static int tfp410_init(struct device *dev, bool i2c) return PTR_ERR(dvi->powerdown); } - if (dvi->hpd) - dvi->hpd_irq = gpiod_to_irq(dvi->hpd); - else - dvi->hpd_irq = -ENXIO; - - if (dvi->hpd_irq >= 0) { - INIT_DELAYED_WORK(&dvi->hpd_work, tfp410_hpd_work_func); - - ret = devm_request_threaded_irq(dev, dvi->hpd_irq, - NULL, tfp410_hpd_irq_thread, IRQF_TRIGGER_RISING | - IRQF_TRIGGER_FALLING | IRQF_ONESHOT, - "hdmi-hpd", dvi); - if (ret) { - DRM_ERROR("failed to register hpd interrupt\n"); - goto fail; - } - } - + /* Register the DRM bridge. */ drm_bridge_add(&dvi->bridge); return 0; -fail: - i2c_put_adapter(dvi->ddc); - if (dvi->hpd) - gpiod_put(dvi->hpd); - return ret; } static int tfp410_fini(struct device *dev) { struct tfp410 *dvi = dev_get_drvdata(dev); - if (dvi->hpd_irq >= 0) - cancel_delayed_work_sync(&dvi->hpd_work); - drm_bridge_remove(&dvi->bridge); - if (dvi->ddc) - i2c_put_adapter(dvi->ddc); - if (dvi->hpd) - gpiod_put(dvi->hpd); - return 0; } diff --git a/drivers/gpu/drm/bridge/ti-tpd12s015.c b/drivers/gpu/drm/bridge/ti-tpd12s015.c new file mode 100644 index 000000000000..514cbf0eac75 --- /dev/null +++ b/drivers/gpu/drm/bridge/ti-tpd12s015.c @@ -0,0 +1,211 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * TPD12S015 HDMI ESD protection & level shifter chip driver + * + * Copyright (C) 2019 Texas Instruments Incorporated + * + * Based on the omapdrm-specific encoder-opa362 driver + * + * Copyright (C) 2013 Texas Instruments Incorporated + * Author: Tomi Valkeinen <tomi.valkeinen@ti.com> + */ + +#include <linux/delay.h> +#include <linux/gpio/consumer.h> +#include <linux/interrupt.h> +#include <linux/module.h> +#include <linux/mutex.h> +#include <linux/of.h> +#include <linux/of_graph.h> +#include <linux/platform_device.h> + +#include <drm/drm_bridge.h> + +struct tpd12s015_device { + struct drm_bridge bridge; + + struct gpio_desc *ct_cp_hpd_gpio; + struct gpio_desc *ls_oe_gpio; + struct gpio_desc *hpd_gpio; + int hpd_irq; + + struct drm_bridge *next_bridge; +}; + +static inline struct tpd12s015_device *to_tpd12s015(struct drm_bridge *bridge) +{ + return container_of(bridge, struct tpd12s015_device, bridge); +} + +static int tpd12s015_attach(struct drm_bridge *bridge, + enum drm_bridge_attach_flags flags) +{ + struct tpd12s015_device *tpd = to_tpd12s015(bridge); + int ret; + + if (!(flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR)) + return -EINVAL; + + ret = drm_bridge_attach(bridge->encoder, tpd->next_bridge, + bridge, flags); + if (ret < 0) + return ret; + + gpiod_set_value_cansleep(tpd->ls_oe_gpio, 1); + + /* DC-DC converter needs at max 300us to get to 90% of 5V. */ + usleep_range(300, 1000); + + return 0; +} + +static void tpd12s015_detach(struct drm_bridge *bridge) +{ + struct tpd12s015_device *tpd = to_tpd12s015(bridge); + + gpiod_set_value_cansleep(tpd->ls_oe_gpio, 0); +} + +static enum drm_connector_status tpd12s015_detect(struct drm_bridge *bridge) +{ + struct tpd12s015_device *tpd = to_tpd12s015(bridge); + + if (gpiod_get_value_cansleep(tpd->hpd_gpio)) + return connector_status_connected; + else + return connector_status_disconnected; +} + +static void tpd12s015_hpd_enable(struct drm_bridge *bridge) +{ + struct tpd12s015_device *tpd = to_tpd12s015(bridge); + + gpiod_set_value_cansleep(tpd->ct_cp_hpd_gpio, 1); +} + +static void tpd12s015_hpd_disable(struct drm_bridge *bridge) +{ + struct tpd12s015_device *tpd = to_tpd12s015(bridge); + + gpiod_set_value_cansleep(tpd->ct_cp_hpd_gpio, 0); +} + +static const struct drm_bridge_funcs tpd12s015_bridge_funcs = { + .attach = tpd12s015_attach, + .detach = tpd12s015_detach, + .detect = tpd12s015_detect, + .hpd_enable = tpd12s015_hpd_enable, + .hpd_disable = tpd12s015_hpd_disable, +}; + +static irqreturn_t tpd12s015_hpd_isr(int irq, void *data) +{ + struct tpd12s015_device *tpd = data; + struct drm_bridge *bridge = &tpd->bridge; + + drm_bridge_hpd_notify(bridge, tpd12s015_detect(bridge)); + + return IRQ_HANDLED; +} + +static int tpd12s015_probe(struct platform_device *pdev) +{ + struct tpd12s015_device *tpd; + struct device_node *node; + struct gpio_desc *gpio; + int ret; + + tpd = devm_kzalloc(&pdev->dev, sizeof(*tpd), GFP_KERNEL); + if (!tpd) + return -ENOMEM; + + platform_set_drvdata(pdev, tpd); + + tpd->bridge.funcs = &tpd12s015_bridge_funcs; + tpd->bridge.of_node = pdev->dev.of_node; + tpd->bridge.type = DRM_MODE_CONNECTOR_HDMIA; + tpd->bridge.ops = DRM_BRIDGE_OP_DETECT; + + /* Get the next bridge, connected to port@1. */ + node = of_graph_get_remote_node(pdev->dev.of_node, 1, -1); + if (!node) + return -ENODEV; + + tpd->next_bridge = of_drm_find_bridge(node); + of_node_put(node); + + if (!tpd->next_bridge) + return -EPROBE_DEFER; + + /* Get the control and HPD GPIOs. */ + gpio = devm_gpiod_get_index_optional(&pdev->dev, NULL, 0, + GPIOD_OUT_LOW); + if (IS_ERR(gpio)) + return PTR_ERR(gpio); + + tpd->ct_cp_hpd_gpio = gpio; + + gpio = devm_gpiod_get_index_optional(&pdev->dev, NULL, 1, + GPIOD_OUT_LOW); + if (IS_ERR(gpio)) + return PTR_ERR(gpio); + + tpd->ls_oe_gpio = gpio; + + gpio = devm_gpiod_get_index(&pdev->dev, NULL, 2, GPIOD_IN); + if (IS_ERR(gpio)) + return PTR_ERR(gpio); + + tpd->hpd_gpio = gpio; + + /* Register the IRQ if the HPD GPIO is IRQ-capable. */ + tpd->hpd_irq = gpiod_to_irq(tpd->hpd_gpio); + if (tpd->hpd_irq) { + ret = devm_request_threaded_irq(&pdev->dev, tpd->hpd_irq, NULL, + tpd12s015_hpd_isr, + IRQF_TRIGGER_RISING | + IRQF_TRIGGER_FALLING | + IRQF_ONESHOT, + "tpd12s015 hpd", tpd); + if (ret) + return ret; + + tpd->bridge.ops |= DRM_BRIDGE_OP_HPD; + } + + /* Register the DRM bridge. */ + drm_bridge_add(&tpd->bridge); + + return 0; +} + +static int __exit tpd12s015_remove(struct platform_device *pdev) +{ + struct tpd12s015_device *tpd = platform_get_drvdata(pdev); + + drm_bridge_remove(&tpd->bridge); + + return 0; +} + +static const struct of_device_id tpd12s015_of_match[] = { + { .compatible = "ti,tpd12s015", }, + {}, +}; + +MODULE_DEVICE_TABLE(of, tpd12s015_of_match); + +static struct platform_driver tpd12s015_driver = { + .probe = tpd12s015_probe, + .remove = __exit_p(tpd12s015_remove), + .driver = { + .name = "tpd12s015", + .of_match_table = tpd12s015_of_match, + }, +}; + +module_platform_driver(tpd12s015_driver); + +MODULE_AUTHOR("Tomi Valkeinen <tomi.valkeinen@ti.com>"); +MODULE_DESCRIPTION("TPD12S015 HDMI level shifter and ESD protection driver"); +MODULE_LICENSE("GPL"); diff --git a/drivers/gpu/drm/cirrus/cirrus.c b/drivers/gpu/drm/cirrus/cirrus.c index a91fb0d7282c..d2ff63ce8eaf 100644 --- a/drivers/gpu/drm/cirrus/cirrus.c +++ b/drivers/gpu/drm/cirrus/cirrus.c @@ -151,9 +151,13 @@ static int cirrus_pitch(struct drm_framebuffer *fb) static void cirrus_set_start_address(struct cirrus_device *cirrus, u32 offset) { + int idx; u32 addr; u8 tmp; + if (!drm_dev_enter(&cirrus->dev, &idx)) + return; + addr = offset >> 2; wreg_crt(cirrus, 0x0c, (u8)((addr >> 8) & 0xff)); wreg_crt(cirrus, 0x0d, (u8)(addr & 0xff)); @@ -168,6 +172,8 @@ static void cirrus_set_start_address(struct cirrus_device *cirrus, u32 offset) tmp &= 0x7f; tmp |= (addr >> 12) & 0x80; wreg_crt(cirrus, 0x1d, tmp); + + drm_dev_exit(idx); } static int cirrus_mode_set(struct cirrus_device *cirrus, @@ -176,9 +182,12 @@ static int cirrus_mode_set(struct cirrus_device *cirrus, { int hsyncstart, hsyncend, htotal, hdispend; int vtotal, vdispend; - int tmp; + int tmp, idx; int sr07 = 0, hdr = 0; + if (!drm_dev_enter(&cirrus->dev, &idx)) + return -1; + htotal = mode->htotal / 8; hsyncend = mode->hsync_end / 8; hsyncstart = mode->hsync_start / 8; @@ -264,6 +273,7 @@ static int cirrus_mode_set(struct cirrus_device *cirrus, hdr = 0xc5; break; default: + drm_dev_exit(idx); return -1; } @@ -292,6 +302,8 @@ static int cirrus_mode_set(struct cirrus_device *cirrus, /* Unblank (needed on S3 resume, vgabios doesn't do it then) */ outb(0x20, 0x3c0); + + drm_dev_exit(idx); return 0; } @@ -300,10 +312,16 @@ static int cirrus_fb_blit_rect(struct drm_framebuffer *fb, { struct cirrus_device *cirrus = fb->dev->dev_private; void *vmap; + int idx, ret; + ret = -ENODEV; + if (!drm_dev_enter(&cirrus->dev, &idx)) + goto out; + + ret = -ENOMEM; vmap = drm_gem_shmem_vmap(fb->obj[0]); if (!vmap) - return -ENOMEM; + goto out_dev_exit; if (cirrus->cpp == fb->format->cpp[0]) drm_fb_memcpy_dstclip(cirrus->vram, @@ -323,7 +341,12 @@ static int cirrus_fb_blit_rect(struct drm_framebuffer *fb, WARN_ON_ONCE("cpp mismatch"); drm_gem_shmem_vunmap(fb->obj[0], vmap); - return 0; + ret = 0; + +out_dev_exit: + drm_dev_exit(idx); +out: + return ret; } static int cirrus_fb_blit_fullscreen(struct drm_framebuffer *fb) @@ -502,6 +525,14 @@ static void cirrus_mode_config_init(struct cirrus_device *cirrus) /* ------------------------------------------------------------------ */ +static void cirrus_release(struct drm_device *dev) +{ + struct cirrus_device *cirrus = dev->dev_private; + + drm_mode_config_cleanup(dev); + kfree(cirrus); +} + DEFINE_DRM_GEM_FOPS(cirrus_fops); static struct drm_driver cirrus_driver = { @@ -515,6 +546,7 @@ static struct drm_driver cirrus_driver = { .fops = &cirrus_fops, DRM_GEM_SHMEM_DRIVER_OPS, + .release = cirrus_release, }; static int cirrus_pci_probe(struct pci_dev *pdev, @@ -598,12 +630,11 @@ static void cirrus_pci_remove(struct pci_dev *pdev) struct drm_device *dev = pci_get_drvdata(pdev); struct cirrus_device *cirrus = dev->dev_private; - drm_dev_unregister(dev); - drm_mode_config_cleanup(dev); + drm_dev_unplug(dev); + drm_atomic_helper_shutdown(dev); iounmap(cirrus->mmio); iounmap(cirrus->vram); drm_dev_put(dev); - kfree(cirrus); pci_release_regions(pdev); } diff --git a/drivers/gpu/drm/drm_atomic.c b/drivers/gpu/drm/drm_atomic.c index 65c46ed049c5..9ccfbf213d72 100644 --- a/drivers/gpu/drm/drm_atomic.c +++ b/drivers/gpu/drm/drm_atomic.c @@ -1098,8 +1098,9 @@ EXPORT_SYMBOL(drm_atomic_get_new_bridge_state); * * This function adds all bridges attached to @encoder. This is needed to add * bridge states to @state and make them available when - * &bridge_funcs.atomic_{check,pre_enable,enable,disable_post_disable}() are - * called + * &drm_bridge_funcs.atomic_check(), &drm_bridge_funcs.atomic_pre_enable(), + * &drm_bridge_funcs.atomic_enable(), + * &drm_bridge_funcs.atomic_disable_post_disable() are called. * * Returns: * 0 on success or can fail with -EDEADLK or -ENOMEM. When the error is EDEADLK diff --git a/drivers/gpu/drm/drm_atomic_state_helper.c b/drivers/gpu/drm/drm_atomic_state_helper.c index 33141d2cdad4..8fce6a115dfe 100644 --- a/drivers/gpu/drm/drm_atomic_state_helper.c +++ b/drivers/gpu/drm/drm_atomic_state_helper.c @@ -635,7 +635,6 @@ EXPORT_SYMBOL(__drm_atomic_helper_bridge_reset); * drm_atomic_helper_bridge_reset() - Allocate and initialize a bridge state * to its default * @bridge: the bridge this state refers to - * @state: bridge state to initialize * * Allocates the bridge state and initializes it to default values. This helper * is meant to be used as a bridge &drm_bridge_funcs.atomic_reset hook for diff --git a/drivers/gpu/drm/drm_bridge.c b/drivers/gpu/drm/drm_bridge.c index 68ab933ee430..afdec8e5fc68 100644 --- a/drivers/gpu/drm/drm_bridge.c +++ b/drivers/gpu/drm/drm_bridge.c @@ -39,26 +39,56 @@ * encoder chain. * * A bridge is always attached to a single &drm_encoder at a time, but can be - * either connected to it directly, or through an intermediate bridge:: + * either connected to it directly, or through a chain of bridges:: * - * encoder ---> bridge B ---> bridge A + * [ CRTC ---> ] Encoder ---> Bridge A ---> Bridge B * - * Here, the output of the encoder feeds to bridge B, and that furthers feeds to - * bridge A. + * Here, the output of the encoder feeds to bridge A, and that furthers feeds to + * bridge B. Bridge chains can be arbitrarily long, and shall be fully linear: + * Chaining multiple bridges to the output of a bridge, or the same bridge to + * the output of different bridges, is not supported. * - * The driver using the bridge is responsible to make the associations between - * the encoder and bridges. Once these links are made, the bridges will - * participate along with encoder functions to perform mode_set/enable/disable - * through the ops provided in &drm_bridge_funcs. + * Display drivers are responsible for linking encoders with the first bridge + * in the chains. This is done by acquiring the appropriate bridge with + * of_drm_find_bridge() or drm_of_find_panel_or_bridge(), or creating it for a + * panel with drm_panel_bridge_add_typed() (or the managed version + * devm_drm_panel_bridge_add_typed()). Once acquired, the bridge shall be + * attached to the encoder with a call to drm_bridge_attach(). * - * drm_bridge, like drm_panel, aren't drm_mode_object entities like planes, + * Bridges are responsible for linking themselves with the next bridge in the + * chain, if any. This is done the same way as for encoders, with the call to + * drm_bridge_attach() occurring in the &drm_bridge_funcs.attach operation. + * + * Once these links are created, the bridges can participate along with encoder + * functions to perform mode validation and fixup (through + * drm_bridge_chain_mode_valid() and drm_atomic_bridge_chain_check()), mode + * setting (through drm_bridge_chain_mode_set()), enable (through + * drm_atomic_bridge_chain_pre_enable() and drm_atomic_bridge_chain_enable()) + * and disable (through drm_atomic_bridge_chain_disable() and + * drm_atomic_bridge_chain_post_disable()). Those functions call the + * corresponding operations provided in &drm_bridge_funcs in sequence for all + * bridges in the chain. + * + * For display drivers that use the atomic helpers + * drm_atomic_helper_check_modeset(), + * drm_atomic_helper_commit_modeset_enables() and + * drm_atomic_helper_commit_modeset_disables() (either directly in hand-rolled + * commit check and commit tail handlers, or through the higher-level + * drm_atomic_helper_check() and drm_atomic_helper_commit_tail() or + * drm_atomic_helper_commit_tail_rpm() helpers), this is done transparently and + * requires no intervention from the driver. For other drivers, the relevant + * DRM bridge chain functions shall be called manually. + * + * Bridges also participate in implementing the &drm_connector at the end of + * the bridge chain. Display drivers may use the drm_bridge_connector_init() + * helper to create the &drm_connector, or implement it manually on top of the + * connector-related operations exposed by the bridge (see the overview + * documentation of bridge operations for more details). + * + * &drm_bridge, like &drm_panel, aren't &drm_mode_object entities like planes, * CRTCs, encoders or connectors and hence are not visible to userspace. They * just provide additional hooks to get the desired output at the end of the * encoder chain. - * - * Bridges can also be chained up using the &drm_bridge.chain_node field. - * - * Both legacy CRTC helpers and the new atomic modeset helpers support bridges. */ static DEFINE_MUTEX(bridge_lock); @@ -71,6 +101,8 @@ static LIST_HEAD(bridge_list); */ void drm_bridge_add(struct drm_bridge *bridge) { + mutex_init(&bridge->hpd_mutex); + mutex_lock(&bridge_lock); list_add_tail(&bridge->list, &bridge_list); mutex_unlock(&bridge_lock); @@ -87,6 +119,8 @@ void drm_bridge_remove(struct drm_bridge *bridge) mutex_lock(&bridge_lock); list_del_init(&bridge->list); mutex_unlock(&bridge_lock); + + mutex_destroy(&bridge->hpd_mutex); } EXPORT_SYMBOL(drm_bridge_remove); @@ -121,6 +155,7 @@ static const struct drm_private_state_funcs drm_bridge_priv_state_funcs = { * @encoder: DRM encoder * @bridge: bridge to attach * @previous: previous bridge in the chain (optional) + * @flags: DRM_BRIDGE_ATTACH_* flags * * Called by a kms driver to link the bridge to an encoder's chain. The previous * argument specifies the previous bridge in the chain. If NULL, the bridge is @@ -138,7 +173,8 @@ static const struct drm_private_state_funcs drm_bridge_priv_state_funcs = { * Zero on success, error code on failure */ int drm_bridge_attach(struct drm_encoder *encoder, struct drm_bridge *bridge, - struct drm_bridge *previous) + struct drm_bridge *previous, + enum drm_bridge_attach_flags flags) { int ret; @@ -160,7 +196,7 @@ int drm_bridge_attach(struct drm_encoder *encoder, struct drm_bridge *bridge, list_add(&bridge->chain_node, &encoder->bridge_chain); if (bridge->funcs->attach) { - ret = bridge->funcs->attach(bridge); + ret = bridge->funcs->attach(bridge, flags); if (ret < 0) goto err_reset_bridge; } @@ -212,14 +248,92 @@ void drm_bridge_detach(struct drm_bridge *bridge) } /** - * DOC: bridge callbacks + * DOC: bridge operations + * + * Bridge drivers expose operations through the &drm_bridge_funcs structure. + * The DRM internals (atomic and CRTC helpers) use the helpers defined in + * drm_bridge.c to call bridge operations. Those operations are divided in + * three big categories to support different parts of the bridge usage. + * + * - The encoder-related operations support control of the bridges in the + * chain, and are roughly counterparts to the &drm_encoder_helper_funcs + * operations. They are used by the legacy CRTC and the atomic modeset + * helpers to perform mode validation, fixup and setting, and enable and + * disable the bridge automatically. * - * The &drm_bridge_funcs ops are populated by the bridge driver. The DRM - * internals (atomic and CRTC helpers) use the helpers defined in drm_bridge.c - * These helpers call a specific &drm_bridge_funcs op for all the bridges - * during encoder configuration. + * The enable and disable operations are split in + * &drm_bridge_funcs.pre_enable, &drm_bridge_funcs.enable, + * &drm_bridge_funcs.disable and &drm_bridge_funcs.post_disable to provide + * finer-grained control. * - * For detailed specification of the bridge callbacks see &drm_bridge_funcs. + * Bridge drivers may implement the legacy version of those operations, or + * the atomic version (prefixed with atomic\_), in which case they shall also + * implement the atomic state bookkeeping operations + * (&drm_bridge_funcs.atomic_duplicate_state, + * &drm_bridge_funcs.atomic_destroy_state and &drm_bridge_funcs.reset). + * Mixing atomic and non-atomic versions of the operations is not supported. + * + * - The bus format negotiation operations + * &drm_bridge_funcs.atomic_get_output_bus_fmts and + * &drm_bridge_funcs.atomic_get_input_bus_fmts allow bridge drivers to + * negotiate the formats transmitted between bridges in the chain when + * multiple formats are supported. Negotiation for formats is performed + * transparently for display drivers by the atomic modeset helpers. Only + * atomic versions of those operations exist, bridge drivers that need to + * implement them shall thus also implement the atomic version of the + * encoder-related operations. This feature is not supported by the legacy + * CRTC helpers. + * + * - The connector-related operations support implementing a &drm_connector + * based on a chain of bridges. DRM bridges traditionally create a + * &drm_connector for bridges meant to be used at the end of the chain. This + * puts additional burden on bridge drivers, especially for bridges that may + * be used in the middle of a chain or at the end of it. Furthermore, it + * requires all operations of the &drm_connector to be handled by a single + * bridge, which doesn't always match the hardware architecture. + * + * To simplify bridge drivers and make the connector implementation more + * flexible, a new model allows bridges to unconditionally skip creation of + * &drm_connector and instead expose &drm_bridge_funcs operations to support + * an externally-implemented &drm_connector. Those operations are + * &drm_bridge_funcs.detect, &drm_bridge_funcs.get_modes, + * &drm_bridge_funcs.get_edid, &drm_bridge_funcs.hpd_notify, + * &drm_bridge_funcs.hpd_enable and &drm_bridge_funcs.hpd_disable. When + * implemented, display drivers shall create a &drm_connector instance for + * each chain of bridges, and implement those connector instances based on + * the bridge connector operations. + * + * Bridge drivers shall implement the connector-related operations for all + * the features that the bridge hardware support. For instance, if a bridge + * supports reading EDID, the &drm_bridge_funcs.get_edid shall be + * implemented. This however doesn't mean that the DDC lines are wired to the + * bridge on a particular platform, as they could also be connected to an I2C + * controller of the SoC. Support for the connector-related operations on the + * running platform is reported through the &drm_bridge.ops flags. Bridge + * drivers shall detect which operations they can support on the platform + * (usually this information is provided by ACPI or DT), and set the + * &drm_bridge.ops flags for all supported operations. A flag shall only be + * set if the corresponding &drm_bridge_funcs operation is implemented, but + * an implemented operation doesn't necessarily imply that the corresponding + * flag will be set. Display drivers shall use the &drm_bridge.ops flags to + * decide which bridge to delegate a connector operation to. This mechanism + * allows providing a single static const &drm_bridge_funcs instance in + * bridge drivers, improving security by storing function pointers in + * read-only memory. + * + * In order to ease transition, bridge drivers may support both the old and + * new models by making connector creation optional and implementing the + * connected-related bridge operations. Connector creation is then controlled + * by the flags argument to the drm_bridge_attach() function. Display drivers + * that support the new model and create connectors themselves shall set the + * %DRM_BRIDGE_ATTACH_NO_CONNECTOR flag, and bridge drivers shall then skip + * connector creation. For intermediate bridges in the chain, the flag shall + * be passed to the drm_bridge_attach() call for the downstream bridge. + * Bridge drivers that implement the new model only shall return an error + * from their &drm_bridge_funcs.attach handler when the + * %DRM_BRIDGE_ATTACH_NO_CONNECTOR flag is not set. New display drivers + * should use the new model, and convert the bridge drivers they use if + * needed, in order to gradually transition to the new model. */ /** @@ -919,6 +1033,164 @@ int drm_atomic_bridge_chain_check(struct drm_bridge *bridge, } EXPORT_SYMBOL(drm_atomic_bridge_chain_check); +/** + * drm_bridge_detect - check if anything is attached to the bridge output + * @bridge: bridge control structure + * + * If the bridge supports output detection, as reported by the + * DRM_BRIDGE_OP_DETECT bridge ops flag, call &drm_bridge_funcs.detect for the + * bridge and return the connection status. Otherwise return + * connector_status_unknown. + * + * RETURNS: + * The detection status on success, or connector_status_unknown if the bridge + * doesn't support output detection. + */ +enum drm_connector_status drm_bridge_detect(struct drm_bridge *bridge) +{ + if (!(bridge->ops & DRM_BRIDGE_OP_DETECT)) + return connector_status_unknown; + + return bridge->funcs->detect(bridge); +} +EXPORT_SYMBOL_GPL(drm_bridge_detect); + +/** + * drm_bridge_get_modes - fill all modes currently valid for the sink into the + * @connector + * @bridge: bridge control structure + * @connector: the connector to fill with modes + * + * If the bridge supports output modes retrieval, as reported by the + * DRM_BRIDGE_OP_MODES bridge ops flag, call &drm_bridge_funcs.get_modes to + * fill the connector with all valid modes and return the number of modes + * added. Otherwise return 0. + * + * RETURNS: + * The number of modes added to the connector. + */ +int drm_bridge_get_modes(struct drm_bridge *bridge, + struct drm_connector *connector) +{ + if (!(bridge->ops & DRM_BRIDGE_OP_MODES)) + return 0; + + return bridge->funcs->get_modes(bridge, connector); +} +EXPORT_SYMBOL_GPL(drm_bridge_get_modes); + +/** + * drm_bridge_get_edid - get the EDID data of the connected display + * @bridge: bridge control structure + * @connector: the connector to read EDID for + * + * If the bridge supports output EDID retrieval, as reported by the + * DRM_BRIDGE_OP_EDID bridge ops flag, call &drm_bridge_funcs.get_edid to + * get the EDID and return it. Otherwise return ERR_PTR(-ENOTSUPP). + * + * RETURNS: + * The retrieved EDID on success, or an error pointer otherwise. + */ +struct edid *drm_bridge_get_edid(struct drm_bridge *bridge, + struct drm_connector *connector) +{ + if (!(bridge->ops & DRM_BRIDGE_OP_EDID)) + return ERR_PTR(-ENOTSUPP); + + return bridge->funcs->get_edid(bridge, connector); +} +EXPORT_SYMBOL_GPL(drm_bridge_get_edid); + +/** + * drm_bridge_hpd_enable - enable hot plug detection for the bridge + * @bridge: bridge control structure + * @cb: hot-plug detection callback + * @data: data to be passed to the hot-plug detection callback + * + * Call &drm_bridge_funcs.hpd_enable if implemented and register the given @cb + * and @data as hot plug notification callback. From now on the @cb will be + * called with @data when an output status change is detected by the bridge, + * until hot plug notification gets disabled with drm_bridge_hpd_disable(). + * + * Hot plug detection is supported only if the DRM_BRIDGE_OP_HPD flag is set in + * bridge->ops. This function shall not be called when the flag is not set. + * + * Only one hot plug detection callback can be registered at a time, it is an + * error to call this function when hot plug detection is already enabled for + * the bridge. + */ +void drm_bridge_hpd_enable(struct drm_bridge *bridge, + void (*cb)(void *data, + enum drm_connector_status status), + void *data) +{ + if (!(bridge->ops & DRM_BRIDGE_OP_HPD)) + return; + + mutex_lock(&bridge->hpd_mutex); + + if (WARN(bridge->hpd_cb, "Hot plug detection already enabled\n")) + goto unlock; + + bridge->hpd_cb = cb; + bridge->hpd_data = data; + + if (bridge->funcs->hpd_enable) + bridge->funcs->hpd_enable(bridge); + +unlock: + mutex_unlock(&bridge->hpd_mutex); +} +EXPORT_SYMBOL_GPL(drm_bridge_hpd_enable); + +/** + * drm_bridge_hpd_disable - disable hot plug detection for the bridge + * @bridge: bridge control structure + * + * Call &drm_bridge_funcs.hpd_disable if implemented and unregister the hot + * plug detection callback previously registered with drm_bridge_hpd_enable(). + * Once this function returns the callback will not be called by the bridge + * when an output status change occurs. + * + * Hot plug detection is supported only if the DRM_BRIDGE_OP_HPD flag is set in + * bridge->ops. This function shall not be called when the flag is not set. + */ +void drm_bridge_hpd_disable(struct drm_bridge *bridge) +{ + if (!(bridge->ops & DRM_BRIDGE_OP_HPD)) + return; + + mutex_lock(&bridge->hpd_mutex); + if (bridge->funcs->hpd_disable) + bridge->funcs->hpd_disable(bridge); + + bridge->hpd_cb = NULL; + bridge->hpd_data = NULL; + mutex_unlock(&bridge->hpd_mutex); +} +EXPORT_SYMBOL_GPL(drm_bridge_hpd_disable); + +/** + * drm_bridge_hpd_notify - notify hot plug detection events + * @bridge: bridge control structure + * @status: output connection status + * + * Bridge drivers shall call this function to report hot plug events when they + * detect a change in the output status, when hot plug detection has been + * enabled by drm_bridge_hpd_enable(). + * + * This function shall be called in a context that can sleep. + */ +void drm_bridge_hpd_notify(struct drm_bridge *bridge, + enum drm_connector_status status) +{ + mutex_lock(&bridge->hpd_mutex); + if (bridge->hpd_cb) + bridge->hpd_cb(bridge->hpd_data, status); + mutex_unlock(&bridge->hpd_mutex); +} +EXPORT_SYMBOL_GPL(drm_bridge_hpd_notify); + #ifdef CONFIG_OF /** * of_drm_find_bridge - find the bridge corresponding to the device node in diff --git a/drivers/gpu/drm/drm_bridge_connector.c b/drivers/gpu/drm/drm_bridge_connector.c new file mode 100644 index 000000000000..c6994fe673f3 --- /dev/null +++ b/drivers/gpu/drm/drm_bridge_connector.c @@ -0,0 +1,379 @@ +// SPDX-License-Identifier: GPL-2.0+ +/* + * Copyright (C) 2019 Laurent Pinchart <laurent.pinchart@ideasonboard.com> + */ + +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/slab.h> + +#include <drm/drm_atomic_state_helper.h> +#include <drm/drm_bridge.h> +#include <drm/drm_bridge_connector.h> +#include <drm/drm_connector.h> +#include <drm/drm_device.h> +#include <drm/drm_edid.h> +#include <drm/drm_modeset_helper_vtables.h> +#include <drm/drm_probe_helper.h> + +/** + * DOC: overview + * + * The DRM bridge connector helper object provides a DRM connector + * implementation that wraps a chain of &struct drm_bridge. The connector + * operations are fully implemented based on the operations of the bridges in + * the chain, and don't require any intervention from the display controller + * driver at runtime. + * + * To use the helper, display controller drivers create a bridge connector with + * a call to drm_bridge_connector_init(). This associates the newly created + * connector with the chain of bridges passed to the function and registers it + * with the DRM device. At that point the connector becomes fully usable, no + * further operation is needed. + * + * The DRM bridge connector operations are implemented based on the operations + * provided by the bridges in the chain. Each connector operation is delegated + * to the bridge closest to the connector (at the end of the chain) that + * provides the relevant functionality. + * + * To make use of this helper, all bridges in the chain shall report bridge + * operation flags (&drm_bridge->ops) and bridge output type + * (&drm_bridge->type), as well as the DRM_BRIDGE_ATTACH_NO_CONNECTOR attach + * flag (none of the bridges shall create a DRM connector directly). + */ + +/** + * struct drm_bridge_connector - A connector backed by a chain of bridges + */ +struct drm_bridge_connector { + /** + * @base: The base DRM connector + */ + struct drm_connector base; + /** + * @encoder: + * + * The encoder at the start of the bridges chain. + */ + struct drm_encoder *encoder; + /** + * @bridge_edid: + * + * The last bridge in the chain (closest to the connector) that provides + * EDID read support, if any (see &DRM_BRIDGE_OP_EDID). + */ + struct drm_bridge *bridge_edid; + /** + * @bridge_hpd: + * + * The last bridge in the chain (closest to the connector) that provides + * hot-plug detection notification, if any (see &DRM_BRIDGE_OP_HPD). + */ + struct drm_bridge *bridge_hpd; + /** + * @bridge_detect: + * + * The last bridge in the chain (closest to the connector) that provides + * connector detection, if any (see &DRM_BRIDGE_OP_DETECT). + */ + struct drm_bridge *bridge_detect; + /** + * @bridge_modes: + * + * The last bridge in the chain (closest to the connector) that provides + * connector modes detection, if any (see &DRM_BRIDGE_OP_MODES). + */ + struct drm_bridge *bridge_modes; +}; + +#define to_drm_bridge_connector(x) \ + container_of(x, struct drm_bridge_connector, base) + +/* ----------------------------------------------------------------------------- + * Bridge Connector Hot-Plug Handling + */ + +static void drm_bridge_connector_hpd_notify(struct drm_connector *connector, + enum drm_connector_status status) +{ + struct drm_bridge_connector *bridge_connector = + to_drm_bridge_connector(connector); + struct drm_bridge *bridge; + + /* Notify all bridges in the pipeline of hotplug events. */ + drm_for_each_bridge_in_chain(bridge_connector->encoder, bridge) { + if (bridge->funcs->hpd_notify) + bridge->funcs->hpd_notify(bridge, status); + } +} + +static void drm_bridge_connector_hpd_cb(void *cb_data, + enum drm_connector_status status) +{ + struct drm_bridge_connector *drm_bridge_connector = cb_data; + struct drm_connector *connector = &drm_bridge_connector->base; + struct drm_device *dev = connector->dev; + enum drm_connector_status old_status; + + mutex_lock(&dev->mode_config.mutex); + old_status = connector->status; + connector->status = status; + mutex_unlock(&dev->mode_config.mutex); + + if (old_status == status) + return; + + drm_bridge_connector_hpd_notify(connector, status); + + drm_kms_helper_hotplug_event(dev); +} + +/** + * drm_bridge_connector_enable_hpd - Enable hot-plug detection for the connector + * @connector: The DRM bridge connector + * + * This function enables hot-plug detection for the given bridge connector. + * This is typically used by display drivers in their resume handler. + */ +void drm_bridge_connector_enable_hpd(struct drm_connector *connector) +{ + struct drm_bridge_connector *bridge_connector = + to_drm_bridge_connector(connector); + struct drm_bridge *hpd = bridge_connector->bridge_hpd; + + if (hpd) + drm_bridge_hpd_enable(hpd, drm_bridge_connector_hpd_cb, + bridge_connector); +} +EXPORT_SYMBOL_GPL(drm_bridge_connector_enable_hpd); + +/** + * drm_bridge_connector_disable_hpd - Disable hot-plug detection for the + * connector + * @connector: The DRM bridge connector + * + * This function disables hot-plug detection for the given bridge connector. + * This is typically used by display drivers in their suspend handler. + */ +void drm_bridge_connector_disable_hpd(struct drm_connector *connector) +{ + struct drm_bridge_connector *bridge_connector = + to_drm_bridge_connector(connector); + struct drm_bridge *hpd = bridge_connector->bridge_hpd; + + if (hpd) + drm_bridge_hpd_disable(hpd); +} +EXPORT_SYMBOL_GPL(drm_bridge_connector_disable_hpd); + +/* ----------------------------------------------------------------------------- + * Bridge Connector Functions + */ + +static enum drm_connector_status +drm_bridge_connector_detect(struct drm_connector *connector, bool force) +{ + struct drm_bridge_connector *bridge_connector = + to_drm_bridge_connector(connector); + struct drm_bridge *detect = bridge_connector->bridge_detect; + enum drm_connector_status status; + + if (detect) { + status = detect->funcs->detect(detect); + + drm_bridge_connector_hpd_notify(connector, status); + } else { + switch (connector->connector_type) { + case DRM_MODE_CONNECTOR_DPI: + case DRM_MODE_CONNECTOR_LVDS: + case DRM_MODE_CONNECTOR_DSI: + status = connector_status_connected; + break; + default: + status = connector_status_unknown; + break; + } + } + + return status; +} + +static void drm_bridge_connector_destroy(struct drm_connector *connector) +{ + struct drm_bridge_connector *bridge_connector = + to_drm_bridge_connector(connector); + + if (bridge_connector->bridge_hpd) { + struct drm_bridge *hpd = bridge_connector->bridge_hpd; + + drm_bridge_hpd_disable(hpd); + } + + drm_connector_unregister(connector); + drm_connector_cleanup(connector); + + kfree(bridge_connector); +} + +static const struct drm_connector_funcs drm_bridge_connector_funcs = { + .reset = drm_atomic_helper_connector_reset, + .detect = drm_bridge_connector_detect, + .fill_modes = drm_helper_probe_single_connector_modes, + .destroy = drm_bridge_connector_destroy, + .atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state, + .atomic_destroy_state = drm_atomic_helper_connector_destroy_state, +}; + +/* ----------------------------------------------------------------------------- + * Bridge Connector Helper Functions + */ + +static int drm_bridge_connector_get_modes_edid(struct drm_connector *connector, + struct drm_bridge *bridge) +{ + enum drm_connector_status status; + struct edid *edid; + int n; + + status = drm_bridge_connector_detect(connector, false); + if (status != connector_status_connected) + goto no_edid; + + edid = bridge->funcs->get_edid(bridge, connector); + if (!edid || !drm_edid_is_valid(edid)) { + kfree(edid); + goto no_edid; + } + + drm_connector_update_edid_property(connector, edid); + n = drm_add_edid_modes(connector, edid); + + kfree(edid); + return n; + +no_edid: + drm_connector_update_edid_property(connector, NULL); + return 0; +} + +static int drm_bridge_connector_get_modes(struct drm_connector *connector) +{ + struct drm_bridge_connector *bridge_connector = + to_drm_bridge_connector(connector); + struct drm_bridge *bridge; + + /* + * If display exposes EDID, then we parse that in the normal way to + * build table of supported modes. + */ + bridge = bridge_connector->bridge_edid; + if (bridge) + return drm_bridge_connector_get_modes_edid(connector, bridge); + + /* + * Otherwise if the display pipeline reports modes (e.g. with a fixed + * resolution panel or an analog TV output), query it. + */ + bridge = bridge_connector->bridge_modes; + if (bridge) + return bridge->funcs->get_modes(bridge, connector); + + /* + * We can't retrieve modes, which can happen for instance for a DVI or + * VGA output with the DDC bus unconnected. The KMS core will add the + * default modes. + */ + return 0; +} + +static const struct drm_connector_helper_funcs drm_bridge_connector_helper_funcs = { + .get_modes = drm_bridge_connector_get_modes, + /* No need for .mode_valid(), the bridges are checked by the core. */ +}; + +/* ----------------------------------------------------------------------------- + * Bridge Connector Initialisation + */ + +/** + * drm_bridge_connector_init - Initialise a connector for a chain of bridges + * @drm: the DRM device + * @encoder: the encoder where the bridge chain starts + * + * Allocate, initialise and register a &drm_bridge_connector with the @drm + * device. The connector is associated with a chain of bridges that starts at + * the @encoder. All bridges in the chain shall report bridge operation flags + * (&drm_bridge->ops) and bridge output type (&drm_bridge->type), and none of + * them may create a DRM connector directly. + * + * Returns a pointer to the new connector on success, or a negative error + * pointer otherwise. + */ +struct drm_connector *drm_bridge_connector_init(struct drm_device *drm, + struct drm_encoder *encoder) +{ + struct drm_bridge_connector *bridge_connector; + struct drm_connector *connector; + struct i2c_adapter *ddc = NULL; + struct drm_bridge *bridge; + int connector_type; + + bridge_connector = kzalloc(sizeof(*bridge_connector), GFP_KERNEL); + if (!bridge_connector) + return ERR_PTR(-ENOMEM); + + bridge_connector->encoder = encoder; + + /* + * TODO: Handle doublescan_allowed, stereo_allowed and + * ycbcr_420_allowed. + */ + connector = &bridge_connector->base; + connector->interlace_allowed = true; + + /* + * Initialise connector status handling. First locate the furthest + * bridges in the pipeline that support HPD and output detection. Then + * initialise the connector polling mode, using HPD if available and + * falling back to polling if supported. If neither HPD nor output + * detection are available, we don't support hotplug detection at all. + */ + connector_type = DRM_MODE_CONNECTOR_Unknown; + drm_for_each_bridge_in_chain(encoder, bridge) { + if (!bridge->interlace_allowed) + connector->interlace_allowed = false; + + if (bridge->ops & DRM_BRIDGE_OP_EDID) + bridge_connector->bridge_edid = bridge; + if (bridge->ops & DRM_BRIDGE_OP_HPD) + bridge_connector->bridge_hpd = bridge; + if (bridge->ops & DRM_BRIDGE_OP_DETECT) + bridge_connector->bridge_detect = bridge; + if (bridge->ops & DRM_BRIDGE_OP_MODES) + bridge_connector->bridge_modes = bridge; + + if (!drm_bridge_get_next_bridge(bridge)) + connector_type = bridge->type; + + if (bridge->ddc) + ddc = bridge->ddc; + } + + if (connector_type == DRM_MODE_CONNECTOR_Unknown) { + kfree(bridge_connector); + return ERR_PTR(-EINVAL); + } + + drm_connector_init_with_ddc(drm, connector, &drm_bridge_connector_funcs, + connector_type, ddc); + drm_connector_helper_add(connector, &drm_bridge_connector_helper_funcs); + + if (bridge_connector->bridge_hpd) + connector->polled = DRM_CONNECTOR_POLL_HPD; + else if (bridge_connector->bridge_detect) + connector->polled = DRM_CONNECTOR_POLL_CONNECT + | DRM_CONNECTOR_POLL_DISCONNECT; + + return connector; +} +EXPORT_SYMBOL_GPL(drm_bridge_connector_init); diff --git a/drivers/gpu/drm/drm_bufs.c b/drivers/gpu/drm/drm_bufs.c index 19297e58b232..dcabf5698333 100644 --- a/drivers/gpu/drm/drm_bufs.c +++ b/drivers/gpu/drm/drm_bufs.c @@ -134,7 +134,7 @@ static int drm_map_handle(struct drm_device *dev, struct drm_hash_item *hash, shift, add); } -/** +/* * Core function to create a range of memory available for mapping by a * non-root process. * @@ -398,7 +398,7 @@ struct drm_local_map *drm_legacy_findmap(struct drm_device *dev, } EXPORT_SYMBOL(drm_legacy_findmap); -/** +/* * Ioctl to specify a range of memory that is available for mapping by a * non-root process. * @@ -499,7 +499,7 @@ int drm_legacy_getmap_ioctl(struct drm_device *dev, void *data, return 0; } -/** +/* * Remove a map private from list and deallocate resources if the mapping * isn't in use. * @@ -659,7 +659,7 @@ int drm_legacy_rmmap_ioctl(struct drm_device *dev, void *data, return ret; } -/** +/* * Cleanup after an error on one of the addbufs() functions. * * \param dev DRM device. @@ -694,7 +694,7 @@ static void drm_cleanup_buf_error(struct drm_device *dev, } #if IS_ENABLED(CONFIG_AGP) -/** +/* * Add AGP buffers for DMA transfers. * * \param dev struct drm_device to which the buffers are to be added. @@ -1230,7 +1230,7 @@ static int drm_legacy_addbufs_sg(struct drm_device *dev, return 0; } -/** +/* * Add buffers for DMA transfers (ioctl). * * \param inode device inode. @@ -1271,7 +1271,7 @@ int drm_legacy_addbufs(struct drm_device *dev, void *data, return ret; } -/** +/* * Get information about the buffer mappings. * * This was originally mean for debugging purposes, or by a sophisticated @@ -1362,7 +1362,7 @@ int drm_legacy_infobufs(struct drm_device *dev, void *data, return __drm_legacy_infobufs(dev, data, &request->count, copy_one_buf); } -/** +/* * Specifies a low and high water mark for buffer allocation * * \param inode device inode. @@ -1411,7 +1411,7 @@ int drm_legacy_markbufs(struct drm_device *dev, void *data, return 0; } -/** +/* * Unreserve the buffers in list, previously reserved using drmDMA. * * \param inode device inode. @@ -1463,7 +1463,7 @@ int drm_legacy_freebufs(struct drm_device *dev, void *data, return 0; } -/** +/* * Maps all of the DMA buffers into client-virtual space (ioctl). * * \param inode device inode. diff --git a/drivers/gpu/drm/drm_client.c b/drivers/gpu/drm/drm_client.c index b031b45aa8ef..6b0c6ef8b9b3 100644 --- a/drivers/gpu/drm/drm_client.c +++ b/drivers/gpu/drm/drm_client.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: GPL-2.0 +// SPDX-License-Identifier: GPL-2.0 or MIT /* * Copyright 2018 Noralf Trønnes */ diff --git a/drivers/gpu/drm/drm_client_modeset.c b/drivers/gpu/drm/drm_client_modeset.c index 6d4a29e99ae2..7443114bd713 100644 --- a/drivers/gpu/drm/drm_client_modeset.c +++ b/drivers/gpu/drm/drm_client_modeset.c @@ -951,7 +951,8 @@ bool drm_client_rotation(struct drm_mode_set *modeset, unsigned int *rotation) * depending on the hardware this may require the framebuffer * to be in a specific tiling format. */ - if ((*rotation & DRM_MODE_ROTATE_MASK) != DRM_MODE_ROTATE_180 || + if (((*rotation & DRM_MODE_ROTATE_MASK) != DRM_MODE_ROTATE_0 && + (*rotation & DRM_MODE_ROTATE_MASK) != DRM_MODE_ROTATE_180) || !plane->rotation_property) return false; @@ -1094,15 +1095,17 @@ out: } /** - * drm_client_modeset_commit_force() - Force commit CRTC configuration + * drm_client_modeset_commit_locked() - Force commit CRTC configuration * @client: DRM client * - * Commit modeset configuration to crtcs without checking if there is a DRM master. + * Commit modeset configuration to crtcs without checking if there is a DRM + * master. The assumption is that the caller already holds an internal DRM + * master reference acquired with drm_master_internal_acquire(). * * Returns: * Zero on success or negative error code on failure. */ -int drm_client_modeset_commit_force(struct drm_client_dev *client) +int drm_client_modeset_commit_locked(struct drm_client_dev *client) { struct drm_device *dev = client->dev; int ret; @@ -1116,7 +1119,7 @@ int drm_client_modeset_commit_force(struct drm_client_dev *client) return ret; } -EXPORT_SYMBOL(drm_client_modeset_commit_force); +EXPORT_SYMBOL(drm_client_modeset_commit_locked); /** * drm_client_modeset_commit() - Commit CRTC configuration @@ -1135,7 +1138,7 @@ int drm_client_modeset_commit(struct drm_client_dev *client) if (!drm_master_internal_acquire(dev)) return -EBUSY; - ret = drm_client_modeset_commit_force(client); + ret = drm_client_modeset_commit_locked(client); drm_master_internal_release(dev); diff --git a/drivers/gpu/drm/drm_connector.c b/drivers/gpu/drm/drm_connector.c index f632ca05960e..644f0ad10671 100644 --- a/drivers/gpu/drm/drm_connector.c +++ b/drivers/gpu/drm/drm_connector.c @@ -112,6 +112,21 @@ void drm_connector_ida_destroy(void) } /** + * drm_get_connector_type_name - return a string for connector type + * @type: The connector type (DRM_MODE_CONNECTOR_*) + * + * Returns: the name of the connector type, or NULL if the type is not valid. + */ +const char *drm_get_connector_type_name(unsigned int type) +{ + if (type < ARRAY_SIZE(drm_connector_enum_list)) + return drm_connector_enum_list[type].name; + + return NULL; +} +EXPORT_SYMBOL(drm_get_connector_type_name); + +/** * drm_connector_get_cmdline_mode - reads the user's cmdline mode * @connector: connector to quwery * diff --git a/drivers/gpu/drm/drm_context.c b/drivers/gpu/drm/drm_context.c index 1f802d8e5681..c99be950bf17 100644 --- a/drivers/gpu/drm/drm_context.c +++ b/drivers/gpu/drm/drm_context.c @@ -47,7 +47,7 @@ struct drm_ctx_list { /** \name Context bitmap support */ /*@{*/ -/** +/* * Free a handle from the context bitmap. * * \param dev DRM device. @@ -68,7 +68,7 @@ void drm_legacy_ctxbitmap_free(struct drm_device * dev, int ctx_handle) mutex_unlock(&dev->struct_mutex); } -/** +/* * Context bitmap allocation. * * \param dev DRM device. @@ -88,7 +88,7 @@ static int drm_legacy_ctxbitmap_next(struct drm_device * dev) return ret; } -/** +/* * Context bitmap initialization. * * \param dev DRM device. @@ -104,7 +104,7 @@ void drm_legacy_ctxbitmap_init(struct drm_device * dev) idr_init(&dev->ctx_idr); } -/** +/* * Context bitmap cleanup. * * \param dev DRM device. @@ -163,7 +163,7 @@ void drm_legacy_ctxbitmap_flush(struct drm_device *dev, struct drm_file *file) /** \name Per Context SAREA Support */ /*@{*/ -/** +/* * Get per-context SAREA. * * \param inode device inode. @@ -211,7 +211,7 @@ int drm_legacy_getsareactx(struct drm_device *dev, void *data, return 0; } -/** +/* * Set per-context SAREA. * * \param inode device inode. @@ -263,7 +263,7 @@ int drm_legacy_setsareactx(struct drm_device *dev, void *data, /** \name The actual DRM context handling routines */ /*@{*/ -/** +/* * Switch context. * * \param dev DRM device. @@ -290,7 +290,7 @@ static int drm_context_switch(struct drm_device * dev, int old, int new) return 0; } -/** +/* * Complete context switch. * * \param dev DRM device. @@ -318,7 +318,7 @@ static int drm_context_switch_complete(struct drm_device *dev, return 0; } -/** +/* * Reserve contexts. * * \param inode device inode. @@ -351,7 +351,7 @@ int drm_legacy_resctx(struct drm_device *dev, void *data, return 0; } -/** +/* * Add context. * * \param inode device inode. @@ -404,7 +404,7 @@ int drm_legacy_addctx(struct drm_device *dev, void *data, return 0; } -/** +/* * Get context. * * \param inode device inode. @@ -428,7 +428,7 @@ int drm_legacy_getctx(struct drm_device *dev, void *data, return 0; } -/** +/* * Switch context. * * \param inode device inode. @@ -452,7 +452,7 @@ int drm_legacy_switchctx(struct drm_device *dev, void *data, return drm_context_switch(dev, dev->last_context, ctx->handle); } -/** +/* * New context. * * \param inode device inode. @@ -478,7 +478,7 @@ int drm_legacy_newctx(struct drm_device *dev, void *data, return 0; } -/** +/* * Remove context. * * \param inode device inode. diff --git a/drivers/gpu/drm/drm_crtc_helper.c b/drivers/gpu/drm/drm_crtc_helper.c index 93a4eec429e8..a4d36aca45ea 100644 --- a/drivers/gpu/drm/drm_crtc_helper.c +++ b/drivers/gpu/drm/drm_crtc_helper.c @@ -244,10 +244,6 @@ drm_crtc_prepare_encoders(struct drm_device *dev) /* Disable unused encoders */ if (encoder->crtc == NULL) drm_encoder_disable(encoder); - /* Disable encoders whose CRTC is about to change */ - if (encoder_funcs->get_crtc && - encoder->crtc != (*encoder_funcs->get_crtc)(encoder)) - drm_encoder_disable(encoder); } } diff --git a/drivers/gpu/drm/drm_dp_helper.c b/drivers/gpu/drm/drm_dp_helper.c index 688c475d52b5..8ba4531e808d 100644 --- a/drivers/gpu/drm/drm_dp_helper.c +++ b/drivers/gpu/drm/drm_dp_helper.c @@ -362,6 +362,65 @@ int drm_dp_dpcd_read_link_status(struct drm_dp_aux *aux, EXPORT_SYMBOL(drm_dp_dpcd_read_link_status); /** + * drm_dp_send_real_edid_checksum() - send back real edid checksum value + * @aux: DisplayPort AUX channel + * @real_edid_checksum: real edid checksum for the last block + * + * Returns: + * True on success + */ +bool drm_dp_send_real_edid_checksum(struct drm_dp_aux *aux, + u8 real_edid_checksum) +{ + u8 link_edid_read = 0, auto_test_req = 0, test_resp = 0; + + if (drm_dp_dpcd_read(aux, DP_DEVICE_SERVICE_IRQ_VECTOR, + &auto_test_req, 1) < 1) { + DRM_ERROR("DPCD failed read at register 0x%x\n", + DP_DEVICE_SERVICE_IRQ_VECTOR); + return false; + } + auto_test_req &= DP_AUTOMATED_TEST_REQUEST; + + if (drm_dp_dpcd_read(aux, DP_TEST_REQUEST, &link_edid_read, 1) < 1) { + DRM_ERROR("DPCD failed read at register 0x%x\n", + DP_TEST_REQUEST); + return false; + } + link_edid_read &= DP_TEST_LINK_EDID_READ; + + if (!auto_test_req || !link_edid_read) { + DRM_DEBUG_KMS("Source DUT does not support TEST_EDID_READ\n"); + return false; + } + + if (drm_dp_dpcd_write(aux, DP_DEVICE_SERVICE_IRQ_VECTOR, + &auto_test_req, 1) < 1) { + DRM_ERROR("DPCD failed write at register 0x%x\n", + DP_DEVICE_SERVICE_IRQ_VECTOR); + return false; + } + + /* send back checksum for the last edid extension block data */ + if (drm_dp_dpcd_write(aux, DP_TEST_EDID_CHECKSUM, + &real_edid_checksum, 1) < 1) { + DRM_ERROR("DPCD failed write at register 0x%x\n", + DP_TEST_EDID_CHECKSUM); + return false; + } + + test_resp |= DP_TEST_EDID_CHECKSUM_WRITE; + if (drm_dp_dpcd_write(aux, DP_TEST_RESPONSE, &test_resp, 1) < 1) { + DRM_ERROR("DPCD failed write at register 0x%x\n", + DP_TEST_RESPONSE); + return false; + } + + return true; +} +EXPORT_SYMBOL(drm_dp_send_real_edid_checksum); + +/** * drm_dp_downstream_max_clock() - extract branch device max * pixel rate for legacy VGA * converter or max TMDS clock diff --git a/drivers/gpu/drm/drm_dp_mst_topology.c b/drivers/gpu/drm/drm_dp_mst_topology.c index 68247670427a..70c4b7afed12 100644 --- a/drivers/gpu/drm/drm_dp_mst_topology.c +++ b/drivers/gpu/drm/drm_dp_mst_topology.c @@ -736,6 +736,10 @@ static bool drm_dp_sideband_msg_build(struct drm_dp_sideband_msg_rx *msg, if (msg->curchunk_idx >= msg->curchunk_len) { /* do CRC */ crc4 = drm_dp_msg_data_crc4(msg->chunk, msg->curchunk_len - 1); + if (crc4 != msg->chunk[msg->curchunk_len - 1]) + print_hex_dump(KERN_DEBUG, "wrong crc", + DUMP_PREFIX_NONE, 16, 1, + msg->chunk, msg->curchunk_len, false); /* copy chunk into bigger msg */ memcpy(&msg->msg[msg->curlen], msg->chunk, msg->curchunk_len - 1); msg->curlen += msg->curchunk_len - 1; @@ -1035,7 +1039,8 @@ static bool drm_dp_sideband_parse_req(struct drm_dp_sideband_msg_rx *raw, } } -static int build_dpcd_write(struct drm_dp_sideband_msg_tx *msg, u8 port_num, u32 offset, u8 num_bytes, u8 *bytes) +static void build_dpcd_write(struct drm_dp_sideband_msg_tx *msg, + u8 port_num, u32 offset, u8 num_bytes, u8 *bytes) { struct drm_dp_sideband_msg_req_body req; @@ -1045,17 +1050,14 @@ static int build_dpcd_write(struct drm_dp_sideband_msg_tx *msg, u8 port_num, u32 req.u.dpcd_write.num_bytes = num_bytes; req.u.dpcd_write.bytes = bytes; drm_dp_encode_sideband_req(&req, msg); - - return 0; } -static int build_link_address(struct drm_dp_sideband_msg_tx *msg) +static void build_link_address(struct drm_dp_sideband_msg_tx *msg) { struct drm_dp_sideband_msg_req_body req; req.req_type = DP_LINK_ADDRESS; drm_dp_encode_sideband_req(&req, msg); - return 0; } static int build_clear_payload_id_table(struct drm_dp_sideband_msg_tx *msg) @@ -1067,7 +1069,8 @@ static int build_clear_payload_id_table(struct drm_dp_sideband_msg_tx *msg) return 0; } -static int build_enum_path_resources(struct drm_dp_sideband_msg_tx *msg, int port_num) +static int build_enum_path_resources(struct drm_dp_sideband_msg_tx *msg, + int port_num) { struct drm_dp_sideband_msg_req_body req; @@ -1078,10 +1081,11 @@ static int build_enum_path_resources(struct drm_dp_sideband_msg_tx *msg, int por return 0; } -static int build_allocate_payload(struct drm_dp_sideband_msg_tx *msg, int port_num, - u8 vcpi, uint16_t pbn, - u8 number_sdp_streams, - u8 *sdp_stream_sink) +static void build_allocate_payload(struct drm_dp_sideband_msg_tx *msg, + int port_num, + u8 vcpi, uint16_t pbn, + u8 number_sdp_streams, + u8 *sdp_stream_sink) { struct drm_dp_sideband_msg_req_body req; memset(&req, 0, sizeof(req)); @@ -1094,11 +1098,10 @@ static int build_allocate_payload(struct drm_dp_sideband_msg_tx *msg, int port_n number_sdp_streams); drm_dp_encode_sideband_req(&req, msg); msg->path_msg = true; - return 0; } -static int build_power_updown_phy(struct drm_dp_sideband_msg_tx *msg, - int port_num, bool power_up) +static void build_power_updown_phy(struct drm_dp_sideband_msg_tx *msg, + int port_num, bool power_up) { struct drm_dp_sideband_msg_req_body req; @@ -1110,7 +1113,6 @@ static int build_power_updown_phy(struct drm_dp_sideband_msg_tx *msg, req.u.port_num.port_number = port_num; drm_dp_encode_sideband_req(&req, msg); msg->path_msg = true; - return 0; } static int drm_dp_mst_assign_payload_id(struct drm_dp_mst_topology_mgr *mgr, @@ -1935,7 +1937,7 @@ static u8 drm_dp_calculate_rad(struct drm_dp_mst_port *port, return parent_lct + 1; } -static bool drm_dp_mst_is_dp_mst_end_device(u8 pdt, bool mcs) +static bool drm_dp_mst_is_end_device(u8 pdt, bool mcs) { switch (pdt) { case DP_PEER_DEVICE_DP_LEGACY_CONV: @@ -1965,13 +1967,13 @@ drm_dp_port_set_pdt(struct drm_dp_mst_port *port, u8 new_pdt, /* Teardown the old pdt, if there is one */ if (port->pdt != DP_PEER_DEVICE_NONE) { - if (drm_dp_mst_is_dp_mst_end_device(port->pdt, port->mcs)) { + if (drm_dp_mst_is_end_device(port->pdt, port->mcs)) { /* * If the new PDT would also have an i2c bus, * don't bother with reregistering it */ if (new_pdt != DP_PEER_DEVICE_NONE && - drm_dp_mst_is_dp_mst_end_device(new_pdt, new_mcs)) { + drm_dp_mst_is_end_device(new_pdt, new_mcs)) { port->pdt = new_pdt; port->mcs = new_mcs; return 0; @@ -1991,7 +1993,7 @@ drm_dp_port_set_pdt(struct drm_dp_mst_port *port, u8 new_pdt, port->mcs = new_mcs; if (port->pdt != DP_PEER_DEVICE_NONE) { - if (drm_dp_mst_is_dp_mst_end_device(port->pdt, port->mcs)) { + if (drm_dp_mst_is_end_device(port->pdt, port->mcs)) { /* add i2c over sideband */ ret = drm_dp_mst_register_i2c_bus(&port->aux); } else { @@ -2061,7 +2063,7 @@ ssize_t drm_dp_mst_dpcd_read(struct drm_dp_aux *aux, * sideband messaging as drm_dp_dpcd_write() does for local * devices via actual AUX CH. * - * Return: 0 on success, negative error code on failure. + * Return: number of bytes written on success, negative error code on failure. */ ssize_t drm_dp_mst_dpcd_write(struct drm_dp_aux *aux, unsigned int offset, void *buffer, size_t size) @@ -2073,29 +2075,27 @@ ssize_t drm_dp_mst_dpcd_write(struct drm_dp_aux *aux, offset, size, buffer); } -static void drm_dp_check_mstb_guid(struct drm_dp_mst_branch *mstb, u8 *guid) +static int drm_dp_check_mstb_guid(struct drm_dp_mst_branch *mstb, u8 *guid) { - int ret; + int ret = 0; memcpy(mstb->guid, guid, 16); if (!drm_dp_validate_guid(mstb->mgr, mstb->guid)) { if (mstb->port_parent) { - ret = drm_dp_send_dpcd_write( - mstb->mgr, - mstb->port_parent, - DP_GUID, - 16, - mstb->guid); + ret = drm_dp_send_dpcd_write(mstb->mgr, + mstb->port_parent, + DP_GUID, 16, mstb->guid); } else { - - ret = drm_dp_dpcd_write( - mstb->mgr->aux, - DP_GUID, - mstb->guid, - 16); + ret = drm_dp_dpcd_write(mstb->mgr->aux, + DP_GUID, mstb->guid, 16); } } + + if (ret < 16 && ret > 0) + return -EPROTO; + + return ret == 16 ? 0 : ret; } static void build_mst_prop_path(const struct drm_dp_mst_branch *mstb, @@ -2172,13 +2172,13 @@ drm_dp_mst_port_add_connector(struct drm_dp_mst_branch *mstb, } if (port->pdt != DP_PEER_DEVICE_NONE && - drm_dp_mst_is_dp_mst_end_device(port->pdt, port->mcs)) { + drm_dp_mst_is_end_device(port->pdt, port->mcs)) { port->cached_edid = drm_get_edid(port->connector, &port->aux.ddc); drm_connector_set_tile_property(port->connector); } - mgr->cbs->register_connector(port->connector); + drm_connector_register(port->connector); return; error: @@ -2302,14 +2302,18 @@ drm_dp_mst_handle_link_address_port(struct drm_dp_mst_branch *mstb, mutex_unlock(&mgr->lock); } - if (old_ddps != port->ddps) { - if (port->ddps) { - if (!port->input) { - drm_dp_send_enum_path_resources(mgr, mstb, - port); - } + /* + * Reprobe PBN caps on both hotplug, and when re-probing the link + * for our parent mstb + */ + if (old_ddps != port->ddps || !created) { + if (port->ddps && !port->input) { + ret = drm_dp_send_enum_path_resources(mgr, mstb, + port); + if (ret == 1) + changed = true; } else { - port->available_pbn = 0; + port->full_pbn = 0; } } @@ -2401,11 +2405,10 @@ drm_dp_mst_handle_conn_stat(struct drm_dp_mst_branch *mstb, port->ddps = conn_stat->displayport_device_plug_status; if (old_ddps != port->ddps) { - if (port->ddps) { - dowork = true; - } else { - port->available_pbn = 0; - } + if (port->ddps && !port->input) + drm_dp_send_enum_path_resources(mgr, mstb, port); + else + port->full_pbn = 0; } new_pdt = port->input ? DP_PEER_DEVICE_NONE : conn_stat->peer_device_type; @@ -2556,13 +2559,6 @@ static int drm_dp_check_and_send_link_address(struct drm_dp_mst_topology_mgr *mg if (port->input || !port->ddps) continue; - if (!port->available_pbn) { - drm_modeset_lock(&mgr->base.lock, NULL); - drm_dp_send_enum_path_resources(mgr, mstb, port); - drm_modeset_unlock(&mgr->base.lock); - changed = true; - } - if (port->mstb) mstb_child = drm_dp_mst_topology_get_mstb_validated( mgr, port->mstb); @@ -2645,7 +2641,8 @@ static bool drm_dp_validate_guid(struct drm_dp_mst_topology_mgr *mgr, return false; } -static int build_dpcd_read(struct drm_dp_sideband_msg_tx *msg, u8 port_num, u32 offset, u8 num_bytes) +static void build_dpcd_read(struct drm_dp_sideband_msg_tx *msg, + u8 port_num, u32 offset, u8 num_bytes) { struct drm_dp_sideband_msg_req_body req; @@ -2654,8 +2651,6 @@ static int build_dpcd_read(struct drm_dp_sideband_msg_tx *msg, u8 port_num, u32 req.u.dpcd_read.dpcd_address = offset; req.u.dpcd_read.num_bytes = num_bytes; drm_dp_encode_sideband_req(&req, msg); - - return 0; } static int drm_dp_send_sideband_msg(struct drm_dp_mst_topology_mgr *mgr, @@ -2881,7 +2876,7 @@ static int drm_dp_send_link_address(struct drm_dp_mst_topology_mgr *mgr, struct drm_dp_sideband_msg_tx *txmsg; struct drm_dp_link_address_ack_reply *reply; struct drm_dp_mst_port *port, *tmp; - int i, len, ret, port_mask = 0; + int i, ret, port_mask = 0; bool changed = false; txmsg = kzalloc(sizeof(*txmsg), GFP_KERNEL); @@ -2889,7 +2884,7 @@ static int drm_dp_send_link_address(struct drm_dp_mst_topology_mgr *mgr, return -ENOMEM; txmsg->dst = mstb; - len = build_link_address(txmsg); + build_link_address(txmsg); mstb->link_address_sent = true; drm_dp_queue_down_tx(mgr, txmsg); @@ -2910,7 +2905,15 @@ static int drm_dp_send_link_address(struct drm_dp_mst_topology_mgr *mgr, DRM_DEBUG_KMS("link address reply: %d\n", reply->nports); drm_dp_dump_link_address(reply); - drm_dp_check_mstb_guid(mstb, reply->guid); + ret = drm_dp_check_mstb_guid(mstb, reply->guid); + if (ret) { + char buf[64]; + + drm_dp_mst_rad_to_str(mstb->rad, mstb->lct, buf, sizeof(buf)); + DRM_ERROR("GUID check on %s failed: %d\n", + buf, ret); + goto out; + } for (i = 0; i < reply->nports; i++) { port_mask |= BIT(reply->ports[i].port_number); @@ -2951,14 +2954,14 @@ void drm_dp_send_clear_payload_id_table(struct drm_dp_mst_topology_mgr *mgr, struct drm_dp_mst_branch *mstb) { struct drm_dp_sideband_msg_tx *txmsg; - int len, ret; + int ret; txmsg = kzalloc(sizeof(*txmsg), GFP_KERNEL); if (!txmsg) return; txmsg->dst = mstb; - len = build_clear_payload_id_table(txmsg); + build_clear_payload_id_table(txmsg); drm_dp_queue_down_tx(mgr, txmsg); @@ -2976,7 +2979,6 @@ drm_dp_send_enum_path_resources(struct drm_dp_mst_topology_mgr *mgr, { struct drm_dp_enum_path_resources_ack_reply *path_res; struct drm_dp_sideband_msg_tx *txmsg; - int len; int ret; txmsg = kzalloc(sizeof(*txmsg), GFP_KERNEL); @@ -2984,12 +2986,13 @@ drm_dp_send_enum_path_resources(struct drm_dp_mst_topology_mgr *mgr, return -ENOMEM; txmsg->dst = mstb; - len = build_enum_path_resources(txmsg, port->port_num); + build_enum_path_resources(txmsg, port->port_num); drm_dp_queue_down_tx(mgr, txmsg); ret = drm_dp_mst_wait_tx_reply(mstb, txmsg); if (ret > 0) { + ret = 0; path_res = &txmsg->reply.u.path_resources; if (txmsg->reply.reply_type == DP_SIDEBAND_REPLY_NAK) { @@ -3002,14 +3005,22 @@ drm_dp_send_enum_path_resources(struct drm_dp_mst_topology_mgr *mgr, path_res->port_number, path_res->full_payload_bw_number, path_res->avail_payload_bw_number); - port->available_pbn = - path_res->avail_payload_bw_number; + + /* + * If something changed, make sure we send a + * hotplug + */ + if (port->full_pbn != path_res->full_payload_bw_number || + port->fec_capable != path_res->fec_capable) + ret = 1; + + port->full_pbn = path_res->full_payload_bw_number; port->fec_capable = path_res->fec_capable; } } kfree(txmsg); - return 0; + return ret; } static struct drm_dp_mst_port *drm_dp_get_last_connected_port_to_mstb(struct drm_dp_mst_branch *mstb) @@ -3068,7 +3079,7 @@ static int drm_dp_payload_send_msg(struct drm_dp_mst_topology_mgr *mgr, { struct drm_dp_sideband_msg_tx *txmsg; struct drm_dp_mst_branch *mstb; - int len, ret, port_num; + int ret, port_num; u8 sinks[DRM_DP_MAX_SDP_STREAMS]; int i; @@ -3093,9 +3104,9 @@ static int drm_dp_payload_send_msg(struct drm_dp_mst_topology_mgr *mgr, sinks[i] = i; txmsg->dst = mstb; - len = build_allocate_payload(txmsg, port_num, - id, - pbn, port->num_sdp_streams, sinks); + build_allocate_payload(txmsg, port_num, + id, + pbn, port->num_sdp_streams, sinks); drm_dp_queue_down_tx(mgr, txmsg); @@ -3124,7 +3135,7 @@ int drm_dp_send_power_updown_phy(struct drm_dp_mst_topology_mgr *mgr, struct drm_dp_mst_port *port, bool power_up) { struct drm_dp_sideband_msg_tx *txmsg; - int len, ret; + int ret; port = drm_dp_mst_topology_get_port_validated(mgr, port); if (!port) @@ -3137,7 +3148,7 @@ int drm_dp_send_power_updown_phy(struct drm_dp_mst_topology_mgr *mgr, } txmsg->dst = port->parent; - len = build_power_updown_phy(txmsg, port->port_num, power_up); + build_power_updown_phy(txmsg, port->port_num, power_up); drm_dp_queue_down_tx(mgr, txmsg); ret = drm_dp_mst_wait_tx_reply(port->parent, txmsg); @@ -3359,7 +3370,6 @@ static int drm_dp_send_dpcd_read(struct drm_dp_mst_topology_mgr *mgr, struct drm_dp_mst_port *port, int offset, int size, u8 *bytes) { - int len; int ret = 0; struct drm_dp_sideband_msg_tx *txmsg; struct drm_dp_mst_branch *mstb; @@ -3374,7 +3384,7 @@ static int drm_dp_send_dpcd_read(struct drm_dp_mst_topology_mgr *mgr, goto fail_put; } - len = build_dpcd_read(txmsg, port->port_num, offset, size); + build_dpcd_read(txmsg, port->port_num, offset, size); txmsg->dst = port->parent; drm_dp_queue_down_tx(mgr, txmsg); @@ -3412,7 +3422,6 @@ static int drm_dp_send_dpcd_write(struct drm_dp_mst_topology_mgr *mgr, struct drm_dp_mst_port *port, int offset, int size, u8 *bytes) { - int len; int ret; struct drm_dp_sideband_msg_tx *txmsg; struct drm_dp_mst_branch *mstb; @@ -3427,18 +3436,15 @@ static int drm_dp_send_dpcd_write(struct drm_dp_mst_topology_mgr *mgr, goto fail_put; } - len = build_dpcd_write(txmsg, port->port_num, offset, size, bytes); + build_dpcd_write(txmsg, port->port_num, offset, size, bytes); txmsg->dst = mstb; drm_dp_queue_down_tx(mgr, txmsg); ret = drm_dp_mst_wait_tx_reply(mstb, txmsg); - if (ret > 0) { - if (txmsg->reply.reply_type == DP_SIDEBAND_REPLY_NAK) - ret = -EIO; - else - ret = 0; - } + if (ret > 0 && txmsg->reply.reply_type == DP_SIDEBAND_REPLY_NAK) + ret = -EIO; + kfree(txmsg); fail_put: drm_dp_mst_topology_put_mstb(mstb); @@ -3588,13 +3594,9 @@ drm_dp_mst_topology_mgr_invalidate_mstb(struct drm_dp_mst_branch *mstb) /* The link address will need to be re-sent on resume */ mstb->link_address_sent = false; - list_for_each_entry(port, &mstb->ports, next) { - /* The PBN for each port will also need to be re-probed */ - port->available_pbn = 0; - + list_for_each_entry(port, &mstb->ports, next) if (port->mstb) drm_dp_mst_topology_mgr_invalidate_mstb(port->mstb); - } } /** @@ -3673,7 +3675,12 @@ int drm_dp_mst_topology_mgr_resume(struct drm_dp_mst_topology_mgr *mgr, DRM_DEBUG_KMS("dpcd read failed - undocked during suspend?\n"); goto out_fail; } - drm_dp_check_mstb_guid(mgr->mst_primary, guid); + + ret = drm_dp_check_mstb_guid(mgr->mst_primary, guid); + if (ret) { + DRM_DEBUG_KMS("check mstb failed - undocked during suspend?\n"); + goto out_fail; + } /* * For the final step of resuming the topology, we need to bring the @@ -4615,15 +4622,34 @@ void drm_dp_mst_dump_topology(struct seq_file *m, int ret; ret = drm_dp_dpcd_read(mgr->aux, DP_DPCD_REV, buf, DP_RECEIVER_CAP_SIZE); + if (ret) { + seq_printf(m, "dpcd read failed\n"); + goto out; + } seq_printf(m, "dpcd: %*ph\n", DP_RECEIVER_CAP_SIZE, buf); + ret = drm_dp_dpcd_read(mgr->aux, DP_FAUX_CAP, buf, 2); + if (ret) { + seq_printf(m, "faux/mst read failed\n"); + goto out; + } seq_printf(m, "faux/mst: %*ph\n", 2, buf); + ret = drm_dp_dpcd_read(mgr->aux, DP_MSTM_CTRL, buf, 1); + if (ret) { + seq_printf(m, "mst ctrl read failed\n"); + goto out; + } seq_printf(m, "mst ctrl: %*ph\n", 1, buf); /* dump the standard OUI branch header */ ret = drm_dp_dpcd_read(mgr->aux, DP_BRANCH_OUI, buf, DP_BRANCH_OUI_HEADER_SIZE); + if (ret) { + seq_printf(m, "branch oui read failed\n"); + goto out; + } seq_printf(m, "branch oui: %*phN devid: ", 3, buf); + for (i = 0x3; i < 0x8 && buf[i]; i++) seq_printf(m, "%c", buf[i]); seq_printf(m, " revision: hw: %x.%x sw: %x.%x\n", @@ -4632,6 +4658,7 @@ void drm_dp_mst_dump_topology(struct seq_file *m, seq_printf(m, "payload table: %*ph\n", DP_PAYLOAD_TABLE_SIZE, buf); } +out: mutex_unlock(&mgr->lock); } @@ -4647,11 +4674,23 @@ static void drm_dp_tx_work(struct work_struct *work) mutex_unlock(&mgr->qlock); } +static inline void drm_dp_destroy_connector(struct drm_dp_mst_port *port) +{ + if (!port->connector) + return; + + if (port->mgr->cbs->destroy_connector) { + port->mgr->cbs->destroy_connector(port->mgr, port->connector); + } else { + drm_connector_unregister(port->connector); + drm_connector_put(port->connector); + } +} + static inline void drm_dp_delayed_destroy_port(struct drm_dp_mst_port *port) { - if (port->connector) - port->mgr->cbs->destroy_connector(port->mgr, port->connector); + drm_dp_destroy_connector(port); drm_dp_port_set_pdt(port, DP_PEER_DEVICE_NONE, port->mcs); drm_dp_mst_put_port_malloc(port); @@ -4820,41 +4859,102 @@ static bool drm_dp_mst_port_downstream_of_branch(struct drm_dp_mst_port *port, return false; } -static inline -int drm_dp_mst_atomic_check_bw_limit(struct drm_dp_mst_branch *branch, - struct drm_dp_mst_topology_state *mst_state) +static int +drm_dp_mst_atomic_check_port_bw_limit(struct drm_dp_mst_port *port, + struct drm_dp_mst_topology_state *state); + +static int +drm_dp_mst_atomic_check_mstb_bw_limit(struct drm_dp_mst_branch *mstb, + struct drm_dp_mst_topology_state *state) { - struct drm_dp_mst_port *port; struct drm_dp_vcpi_allocation *vcpi; - int pbn_limit = 0, pbn_used = 0; + struct drm_dp_mst_port *port; + int pbn_used = 0, ret; + bool found = false; - list_for_each_entry(port, &branch->ports, next) { - if (port->mstb) - if (drm_dp_mst_atomic_check_bw_limit(port->mstb, mst_state)) - return -ENOSPC; + /* Check that we have at least one port in our state that's downstream + * of this branch, otherwise we can skip this branch + */ + list_for_each_entry(vcpi, &state->vcpis, next) { + if (!vcpi->pbn || + !drm_dp_mst_port_downstream_of_branch(vcpi->port, mstb)) + continue; - if (port->available_pbn > 0) - pbn_limit = port->available_pbn; + found = true; + break; } - DRM_DEBUG_ATOMIC("[MST BRANCH:%p] branch has %d PBN available\n", - branch, pbn_limit); + if (!found) + return 0; - list_for_each_entry(vcpi, &mst_state->vcpis, next) { - if (!vcpi->pbn) - continue; + if (mstb->port_parent) + DRM_DEBUG_ATOMIC("[MSTB:%p] [MST PORT:%p] Checking bandwidth limits on [MSTB:%p]\n", + mstb->port_parent->parent, mstb->port_parent, + mstb); + else + DRM_DEBUG_ATOMIC("[MSTB:%p] Checking bandwidth limits\n", + mstb); + + list_for_each_entry(port, &mstb->ports, next) { + ret = drm_dp_mst_atomic_check_port_bw_limit(port, state); + if (ret < 0) + return ret; + + pbn_used += ret; + } + + return pbn_used; +} + +static int +drm_dp_mst_atomic_check_port_bw_limit(struct drm_dp_mst_port *port, + struct drm_dp_mst_topology_state *state) +{ + struct drm_dp_vcpi_allocation *vcpi; + int pbn_used = 0; + + if (port->pdt == DP_PEER_DEVICE_NONE) + return 0; + + if (drm_dp_mst_is_end_device(port->pdt, port->mcs)) { + bool found = false; + + list_for_each_entry(vcpi, &state->vcpis, next) { + if (vcpi->port != port) + continue; + if (!vcpi->pbn) + return 0; + + found = true; + break; + } + if (!found) + return 0; - if (drm_dp_mst_port_downstream_of_branch(vcpi->port, branch)) - pbn_used += vcpi->pbn; + /* This should never happen, as it means we tried to + * set a mode before querying the full_pbn + */ + if (WARN_ON(!port->full_pbn)) + return -EINVAL; + + pbn_used = vcpi->pbn; + } else { + pbn_used = drm_dp_mst_atomic_check_mstb_bw_limit(port->mstb, + state); + if (pbn_used <= 0) + return pbn_used; } - DRM_DEBUG_ATOMIC("[MST BRANCH:%p] branch used %d PBN\n", - branch, pbn_used); - if (pbn_used > pbn_limit) { - DRM_DEBUG_ATOMIC("[MST BRANCH:%p] No available bandwidth\n", - branch); + if (pbn_used > port->full_pbn) { + DRM_DEBUG_ATOMIC("[MSTB:%p] [MST PORT:%p] required PBN of %d exceeds port limit of %d\n", + port->parent, port, pbn_used, + port->full_pbn); return -ENOSPC; } - return 0; + + DRM_DEBUG_ATOMIC("[MSTB:%p] [MST PORT:%p] uses %d out of %d PBN\n", + port->parent, port, pbn_used, port->full_pbn); + + return pbn_used; } static inline int @@ -5052,9 +5152,15 @@ int drm_dp_mst_atomic_check(struct drm_atomic_state *state) ret = drm_dp_mst_atomic_check_vcpi_alloc_limit(mgr, mst_state); if (ret) break; - ret = drm_dp_mst_atomic_check_bw_limit(mgr->mst_primary, mst_state); - if (ret) + + mutex_lock(&mgr->lock); + ret = drm_dp_mst_atomic_check_mstb_bw_limit(mgr->mst_primary, + mst_state); + mutex_unlock(&mgr->lock); + if (ret < 0) break; + else + ret = 0; } return ret; diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c index 7c18a980cd4b..7b1a628d1f6e 100644 --- a/drivers/gpu/drm/drm_drv.c +++ b/drivers/gpu/drm/drm_drv.c @@ -946,7 +946,8 @@ int drm_dev_register(struct drm_device *dev, unsigned long flags) struct drm_driver *driver = dev->driver; int ret; - mutex_lock(&drm_global_mutex); + if (drm_dev_needs_global_mutex(dev)) + mutex_lock(&drm_global_mutex); ret = drm_minor_register(dev, DRM_MINOR_RENDER); if (ret) @@ -986,7 +987,8 @@ err_minors: drm_minor_unregister(dev, DRM_MINOR_PRIMARY); drm_minor_unregister(dev, DRM_MINOR_RENDER); out_unlock: - mutex_unlock(&drm_global_mutex); + if (drm_dev_needs_global_mutex(dev)) + mutex_unlock(&drm_global_mutex); return ret; } EXPORT_SYMBOL(drm_dev_register); @@ -1079,17 +1081,14 @@ static int drm_stub_open(struct inode *inode, struct file *filp) DRM_DEBUG("\n"); - mutex_lock(&drm_global_mutex); minor = drm_minor_acquire(iminor(inode)); - if (IS_ERR(minor)) { - err = PTR_ERR(minor); - goto out_unlock; - } + if (IS_ERR(minor)) + return PTR_ERR(minor); new_fops = fops_get(minor->dev->driver->fops); if (!new_fops) { err = -ENODEV; - goto out_release; + goto out; } replace_fops(filp, new_fops); @@ -1098,10 +1097,9 @@ static int drm_stub_open(struct inode *inode, struct file *filp) else err = 0; -out_release: +out: drm_minor_release(minor); -out_unlock: - mutex_unlock(&drm_global_mutex); + return err; } diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c index 805fb004c8eb..116451101426 100644 --- a/drivers/gpu/drm/drm_edid.c +++ b/drivers/gpu/drm/drm_edid.c @@ -1590,11 +1590,22 @@ static int validate_displayid(u8 *displayid, int length, int idx); static int drm_edid_block_checksum(const u8 *raw_edid) { int i; - u8 csum = 0; - for (i = 0; i < EDID_LENGTH; i++) + u8 csum = 0, crc = 0; + + for (i = 0; i < EDID_LENGTH - 1; i++) csum += raw_edid[i]; - return csum; + crc = 0x100 - csum; + + return crc; +} + +static bool drm_edid_block_checksum_diff(const u8 *raw_edid, u8 real_checksum) +{ + if (raw_edid[EDID_LENGTH - 1] != real_checksum) + return true; + else + return false; } static bool drm_edid_is_zero(const u8 *in_edid, int length) @@ -1652,7 +1663,7 @@ bool drm_edid_block_valid(u8 *raw_edid, int block, bool print_bad_edid, } csum = drm_edid_block_checksum(raw_edid); - if (csum) { + if (drm_edid_block_checksum_diff(raw_edid, csum)) { if (edid_corrupt) *edid_corrupt = true; @@ -1793,6 +1804,11 @@ static void connector_bad_edid(struct drm_connector *connector, u8 *edid, int num_blocks) { int i; + u8 num_of_ext = edid[0x7e]; + + /* Calculate real checksum for the last edid extension block data */ + connector->real_edid_checksum = + drm_edid_block_checksum(edid + num_of_ext * EDID_LENGTH); if (connector->bad_edid_counter++ && !drm_debug_enabled(DRM_UT_KMS)) return; @@ -2196,15 +2212,29 @@ struct drm_display_mode *drm_mode_find_dmt(struct drm_device *dev, } EXPORT_SYMBOL(drm_mode_find_dmt); +static bool is_display_descriptor(const u8 d[18], u8 tag) +{ + return d[0] == 0x00 && d[1] == 0x00 && + d[2] == 0x00 && d[3] == tag; +} + +static bool is_detailed_timing_descriptor(const u8 d[18]) +{ + return d[0] != 0x00 || d[1] != 0x00; +} + typedef void detailed_cb(struct detailed_timing *timing, void *closure); static void cea_for_each_detailed_block(u8 *ext, detailed_cb *cb, void *closure) { - int i, n = 0; + int i, n; u8 d = ext[0x02]; u8 *det_base = ext + d; + if (d < 4 || d > 127) + return; + n = (127 - d) / 18; for (i = 0; i < n; i++) cb((struct detailed_timing *)(det_base + 18 * i), closure); @@ -2254,9 +2284,12 @@ static void is_rb(struct detailed_timing *t, void *data) { u8 *r = (u8 *)t; - if (r[3] == EDID_DETAIL_MONITOR_RANGE) - if (r[15] & 0x10) - *(bool *)data = true; + + if (!is_display_descriptor(r, EDID_DETAIL_MONITOR_RANGE)) + return; + + if (r[15] & 0x10) + *(bool *)data = true; } /* EDID 1.4 defines this explicitly. For EDID 1.3, we guess, badly. */ @@ -2276,7 +2309,11 @@ static void find_gtf2(struct detailed_timing *t, void *data) { u8 *r = (u8 *)t; - if (r[3] == EDID_DETAIL_MONITOR_RANGE && r[10] == 0x02) + + if (!is_display_descriptor(r, EDID_DETAIL_MONITOR_RANGE)) + return; + + if (r[10] == 0x02) *(u8 **)data = r; } @@ -2815,13 +2852,13 @@ do_inferred_modes(struct detailed_timing *timing, void *c) struct detailed_non_pixel *data = &timing->data.other_data; struct detailed_data_monitor_range *range = &data->data.range; - if (data->type != EDID_DETAIL_MONITOR_RANGE) + if (!is_display_descriptor((const u8 *)timing, EDID_DETAIL_MONITOR_RANGE)) return; closure->modes += drm_dmt_modes_for_range(closure->connector, closure->edid, timing); - + if (!version_greater(closure->edid, 1, 1)) return; /* GTF not defined yet */ @@ -2894,10 +2931,11 @@ static void do_established_modes(struct detailed_timing *timing, void *c) { struct detailed_mode_closure *closure = c; - struct detailed_non_pixel *data = &timing->data.other_data; - if (data->type == EDID_DETAIL_EST_TIMINGS) - closure->modes += drm_est3_modes(closure->connector, timing); + if (!is_display_descriptor((const u8 *)timing, EDID_DETAIL_EST_TIMINGS)) + return; + + closure->modes += drm_est3_modes(closure->connector, timing); } /** @@ -2946,19 +2984,19 @@ do_standard_modes(struct detailed_timing *timing, void *c) struct detailed_non_pixel *data = &timing->data.other_data; struct drm_connector *connector = closure->connector; struct edid *edid = closure->edid; + int i; - if (data->type == EDID_DETAIL_STD_MODES) { - int i; - for (i = 0; i < 6; i++) { - struct std_timing *std; - struct drm_display_mode *newmode; + if (!is_display_descriptor((const u8 *)timing, EDID_DETAIL_STD_MODES)) + return; - std = &data->data.timings[i]; - newmode = drm_mode_std(connector, edid, std); - if (newmode) { - drm_mode_probed_add(connector, newmode); - closure->modes++; - } + for (i = 0; i < 6; i++) { + struct std_timing *std = &data->data.timings[i]; + struct drm_display_mode *newmode; + + newmode = drm_mode_std(connector, edid, std); + if (newmode) { + drm_mode_probed_add(connector, newmode); + closure->modes++; } } } @@ -3053,15 +3091,16 @@ static void do_cvt_mode(struct detailed_timing *timing, void *c) { struct detailed_mode_closure *closure = c; - struct detailed_non_pixel *data = &timing->data.other_data; - if (data->type == EDID_DETAIL_CVT_3BYTE) - closure->modes += drm_cvt_modes(closure->connector, timing); + if (!is_display_descriptor((const u8 *)timing, EDID_DETAIL_CVT_3BYTE)) + return; + + closure->modes += drm_cvt_modes(closure->connector, timing); } static int add_cvt_modes(struct drm_connector *connector, struct edid *edid) -{ +{ struct detailed_mode_closure closure = { .connector = connector, .edid = edid, @@ -3083,27 +3122,28 @@ do_detailed_mode(struct detailed_timing *timing, void *c) struct detailed_mode_closure *closure = c; struct drm_display_mode *newmode; - if (timing->pixel_clock) { - newmode = drm_mode_detailed(closure->connector->dev, - closure->edid, timing, - closure->quirks); - if (!newmode) - return; + if (!is_detailed_timing_descriptor((const u8 *)timing)) + return; - if (closure->preferred) - newmode->type |= DRM_MODE_TYPE_PREFERRED; + newmode = drm_mode_detailed(closure->connector->dev, + closure->edid, timing, + closure->quirks); + if (!newmode) + return; - /* - * Detailed modes are limited to 10kHz pixel clock resolution, - * so fix up anything that looks like CEA/HDMI mode, but the clock - * is just slightly off. - */ - fixup_detailed_cea_mode_clock(newmode); + if (closure->preferred) + newmode->type |= DRM_MODE_TYPE_PREFERRED; - drm_mode_probed_add(closure->connector, newmode); - closure->modes++; - closure->preferred = false; - } + /* + * Detailed modes are limited to 10kHz pixel clock resolution, + * so fix up anything that looks like CEA/HDMI mode, but the clock + * is just slightly off. + */ + fixup_detailed_cea_mode_clock(newmode); + + drm_mode_probed_add(closure->connector, newmode); + closure->modes++; + closure->preferred = false; } /* @@ -3953,6 +3993,13 @@ cea_db_tag(const u8 *db) static int cea_revision(const u8 *cea) { + /* + * FIXME is this correct for the DispID variant? + * The DispID spec doesn't really specify whether + * this is the revision of the CEA extension or + * the DispID CEA data block. And the only value + * given as an example is 0. + */ return cea[1]; } @@ -3977,6 +4024,10 @@ cea_db_offsets(const u8 *cea, int *start, int *end) * no non-DTD data. */ if (cea[0] == DATA_BLOCK_CTA) { + /* + * for_each_displayid_db() has already verified + * that these stay within expected bounds. + */ *start = 3; *end = *start + cea[2]; } else if (cea[0] == CEA_EXT) { @@ -4282,8 +4333,10 @@ drm_parse_hdmi_vsdb_audio(struct drm_connector *connector, const u8 *db) static void monitor_name(struct detailed_timing *t, void *data) { - if (t->data.other_data.type == EDID_DETAIL_MONITOR_NAME) - *(u8 **)data = t->data.other_data.data.str.str; + if (!is_display_descriptor((const u8 *)t, EDID_DETAIL_MONITOR_NAME)) + return; + + *(u8 **)data = t->data.other_data.data.str.str; } static int get_monitor_name(struct edid *edid, char name[13]) @@ -4316,7 +4369,7 @@ void drm_edid_get_monitor_name(struct edid *edid, char *name, int bufsize) { int name_length; char buf[13]; - + if (bufsize <= 0) return; @@ -4381,6 +4434,7 @@ static void drm_edid_to_eld(struct drm_connector *connector, struct edid *edid) if (cea_revision(cea) >= 3) { int i, start, end; + int sad_count; if (cea_db_offsets(cea, &start, &end)) { start = 0; @@ -4392,8 +4446,6 @@ static void drm_edid_to_eld(struct drm_connector *connector, struct edid *edid) dbl = cea_db_payload_len(db); switch (cea_db_tag(db)) { - int sad_count; - case AUDIO_BLOCK: /* Audio Data Block, contains SADs */ sad_count = min(dbl / 3, 15 - total_sad_count); @@ -4594,6 +4646,9 @@ EXPORT_SYMBOL(drm_av_sync_delay); * * Parse the CEA extension according to CEA-861-B. * + * Drivers that have added the modes parsed from EDID to drm_display_info + * should use &drm_display_info.is_hdmi instead of calling this function. + * * Return: True if the monitor is HDMI, false if not or unknown. */ bool drm_detect_hdmi_monitor(struct edid *edid) @@ -4828,6 +4883,8 @@ drm_parse_hdmi_vsdb_video(struct drm_connector *connector, const u8 *db) struct drm_display_info *info = &connector->display_info; u8 len = cea_db_payload_len(db); + info->is_hdmi = true; + if (len >= 6) info->dvi_dual = db[6] & 1; if (len >= 7) @@ -4880,6 +4937,47 @@ static void drm_parse_cea_ext(struct drm_connector *connector, } } +static +void get_monitor_range(struct detailed_timing *timing, + void *info_monitor_range) +{ + struct drm_monitor_range_info *monitor_range = info_monitor_range; + const struct detailed_non_pixel *data = &timing->data.other_data; + const struct detailed_data_monitor_range *range = &data->data.range; + + if (!is_display_descriptor((const u8 *)timing, EDID_DETAIL_MONITOR_RANGE)) + return; + + /* + * Check for flag range limits only. If flag == 1 then + * no additional timing information provided. + * Default GTF, GTF Secondary curve and CVT are not + * supported + */ + if (range->flags != DRM_EDID_RANGE_LIMITS_ONLY_FLAG) + return; + + monitor_range->min_vfreq = range->min_vfreq; + monitor_range->max_vfreq = range->max_vfreq; +} + +static +void drm_get_monitor_range(struct drm_connector *connector, + const struct edid *edid) +{ + struct drm_display_info *info = &connector->display_info; + + if (!version_greater(edid, 1, 1)) + return; + + drm_for_each_detailed_block((u8 *)edid, get_monitor_range, + &info->monitor_range); + + DRM_DEBUG_KMS("Supported Monitor Refresh rate range is %d Hz - %d Hz\n", + info->monitor_range.min_vfreq, + info->monitor_range.max_vfreq); +} + /* A connector has no EDID information, so we've got no EDID to compute quirks from. Reset * all of the values which would have been set from EDID */ @@ -4896,11 +4994,13 @@ drm_reset_display_info(struct drm_connector *connector) info->cea_rev = 0; info->max_tmds_clock = 0; info->dvi_dual = false; + info->is_hdmi = false; info->has_hdmi_infoframe = false; info->rgb_quant_range_selectable = false; memset(&info->hdmi, 0, sizeof(info->hdmi)); info->non_desktop = 0; + memset(&info->monitor_range, 0, sizeof(info->monitor_range)); } u32 drm_add_display_info(struct drm_connector *connector, const struct edid *edid) @@ -4916,6 +5016,8 @@ u32 drm_add_display_info(struct drm_connector *connector, const struct edid *edi info->non_desktop = !!(quirks & EDID_QUIRK_NON_DESKTOP); + drm_get_monitor_range(connector, edid); + DRM_DEBUG_KMS("non_desktop set to %d\n", info->non_desktop); if (edid->revision < 3) @@ -5396,14 +5498,11 @@ drm_hdmi_avi_infoframe_from_display_mode(struct hdmi_avi_infoframe *frame, { enum hdmi_picture_aspect picture_aspect; u8 vic, hdmi_vic; - int err; if (!frame || !mode) return -EINVAL; - err = hdmi_avi_infoframe_init(frame); - if (err < 0) - return err; + hdmi_avi_infoframe_init(frame); if (mode->flags & DRM_MODE_FLAG_DBLCLK) frame->pixel_repeat = 1; diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c index 4c7cbce7bae7..a9771de4d17e 100644 --- a/drivers/gpu/drm/drm_fb_helper.c +++ b/drivers/gpu/drm/drm_fb_helper.c @@ -250,17 +250,7 @@ int drm_fb_helper_restore_fbdev_mode_unlocked(struct drm_fb_helper *fb_helper) return 0; mutex_lock(&fb_helper->lock); - /* - * TODO: - * We should bail out here if there is a master by dropping _force. - * Currently these igt tests fail if we do that: - * - kms_fbcon_fbt@psr - * - kms_fbcon_fbt@psr-suspend - * - * So first these tests need to be fixed so they drop master or don't - * have an fd open. - */ - ret = drm_client_modeset_commit_force(&fb_helper->client); + ret = drm_client_modeset_commit(&fb_helper->client); do_delayed = fb_helper->delayed_hotplug; if (do_delayed) @@ -294,7 +284,7 @@ static bool drm_fb_helper_force_kernel_mode(void) continue; mutex_lock(&helper->lock); - ret = drm_client_modeset_commit_force(&helper->client); + ret = drm_client_modeset_commit_locked(&helper->client); if (ret) error = true; mutex_unlock(&helper->lock); @@ -460,7 +450,6 @@ EXPORT_SYMBOL(drm_fb_helper_prepare); * drm_fb_helper_init - initialize a &struct drm_fb_helper * @dev: drm device * @fb_helper: driver-allocated fbdev helper structure to initialize - * @max_conn_count: max connector count (not used) * * This allocates the structures for the fbdev helper with the given limits. * Note that this won't yet touch the hardware (through the driver interfaces) @@ -473,8 +462,7 @@ EXPORT_SYMBOL(drm_fb_helper_prepare); * Zero if everything went ok, nonzero otherwise. */ int drm_fb_helper_init(struct drm_device *dev, - struct drm_fb_helper *fb_helper, - int max_conn_count) + struct drm_fb_helper *fb_helper) { int ret; @@ -1357,7 +1345,7 @@ static int pan_display_atomic(struct fb_var_screeninfo *var, pan_set(fb_helper, var->xoffset, var->yoffset); - ret = drm_client_modeset_commit_force(&fb_helper->client); + ret = drm_client_modeset_commit_locked(&fb_helper->client); if (!ret) { info->var.xoffset = var->xoffset; info->var.yoffset = var->yoffset; @@ -2135,7 +2123,7 @@ static int drm_fbdev_client_hotplug(struct drm_client_dev *client) drm_fb_helper_prepare(dev, fb_helper, &drm_fb_helper_generic_funcs); - ret = drm_fb_helper_init(dev, fb_helper, 0); + ret = drm_fb_helper_init(dev, fb_helper); if (ret) goto err; diff --git a/drivers/gpu/drm/drm_file.c b/drivers/gpu/drm/drm_file.c index 1075b3a8b5b1..eb009d3ab48f 100644 --- a/drivers/gpu/drm/drm_file.c +++ b/drivers/gpu/drm/drm_file.c @@ -48,9 +48,45 @@ #include "drm_internal.h" #include "drm_legacy.h" +#if defined(CONFIG_MMU) && defined(CONFIG_TRANSPARENT_HUGEPAGE) +#include <uapi/asm/mman.h> +#include <drm/drm_vma_manager.h> +#endif + /* from BKL pushdown */ DEFINE_MUTEX(drm_global_mutex); +bool drm_dev_needs_global_mutex(struct drm_device *dev) +{ + /* + * Legacy drivers rely on all kinds of BKL locking semantics, don't + * bother. They also still need BKL locking for their ioctls, so better + * safe than sorry. + */ + if (drm_core_check_feature(dev, DRIVER_LEGACY)) + return true; + + /* + * The deprecated ->load callback must be called after the driver is + * already registered. This means such drivers rely on the BKL to make + * sure an open can't proceed until the driver is actually fully set up. + * Similar hilarity holds for the unload callback. + */ + if (dev->driver->load || dev->driver->unload) + return true; + + /* + * Drivers with the lastclose callback assume that it's synchronized + * against concurrent opens, which again needs the BKL. The proper fix + * is to use the drm_client infrastructure with proper locking for each + * client. + */ + if (dev->driver->lastclose) + return true; + + return false; +} + /** * DOC: file operations * @@ -379,6 +415,9 @@ int drm_open(struct inode *inode, struct file *filp) return PTR_ERR(minor); dev = minor->dev; + if (drm_dev_needs_global_mutex(dev)) + mutex_lock(&drm_global_mutex); + if (!atomic_fetch_inc(&dev->open_count)) need_setup = 1; @@ -395,10 +434,16 @@ int drm_open(struct inode *inode, struct file *filp) goto err_undo; } } + + if (drm_dev_needs_global_mutex(dev)) + mutex_unlock(&drm_global_mutex); + return 0; err_undo: atomic_dec(&dev->open_count); + if (drm_dev_needs_global_mutex(dev)) + mutex_unlock(&drm_global_mutex); drm_minor_release(minor); return retcode; } @@ -438,7 +483,8 @@ int drm_release(struct inode *inode, struct file *filp) struct drm_minor *minor = file_priv->minor; struct drm_device *dev = minor->dev; - mutex_lock(&drm_global_mutex); + if (drm_dev_needs_global_mutex(dev)) + mutex_lock(&drm_global_mutex); DRM_DEBUG("open_count = %d\n", atomic_read(&dev->open_count)); @@ -447,7 +493,8 @@ int drm_release(struct inode *inode, struct file *filp) if (atomic_dec_and_test(&dev->open_count)) drm_lastclose(dev); - mutex_unlock(&drm_global_mutex); + if (drm_dev_needs_global_mutex(dev)) + mutex_unlock(&drm_global_mutex); drm_minor_release(minor); @@ -830,3 +877,139 @@ struct file *mock_drm_getfile(struct drm_minor *minor, unsigned int flags) return file; } EXPORT_SYMBOL_FOR_TESTS_ONLY(mock_drm_getfile); + +#ifdef CONFIG_MMU +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +/* + * drm_addr_inflate() attempts to construct an aligned area by inflating + * the area size and skipping the unaligned start of the area. + * adapted from shmem_get_unmapped_area() + */ +static unsigned long drm_addr_inflate(unsigned long addr, + unsigned long len, + unsigned long pgoff, + unsigned long flags, + unsigned long huge_size) +{ + unsigned long offset, inflated_len; + unsigned long inflated_addr; + unsigned long inflated_offset; + + offset = (pgoff << PAGE_SHIFT) & (huge_size - 1); + if (offset && offset + len < 2 * huge_size) + return addr; + if ((addr & (huge_size - 1)) == offset) + return addr; + + inflated_len = len + huge_size - PAGE_SIZE; + if (inflated_len > TASK_SIZE) + return addr; + if (inflated_len < len) + return addr; + + inflated_addr = current->mm->get_unmapped_area(NULL, 0, inflated_len, + 0, flags); + if (IS_ERR_VALUE(inflated_addr)) + return addr; + if (inflated_addr & ~PAGE_MASK) + return addr; + + inflated_offset = inflated_addr & (huge_size - 1); + inflated_addr += offset - inflated_offset; + if (inflated_offset > offset) + inflated_addr += huge_size; + + if (inflated_addr > TASK_SIZE - len) + return addr; + + return inflated_addr; +} + +/** + * drm_get_unmapped_area() - Get an unused user-space virtual memory area + * suitable for huge page table entries. + * @file: The struct file representing the address space being mmap()'d. + * @uaddr: Start address suggested by user-space. + * @len: Length of the area. + * @pgoff: The page offset into the address space. + * @flags: mmap flags + * @mgr: The address space manager used by the drm driver. This argument can + * probably be removed at some point when all drivers use the same + * address space manager. + * + * This function attempts to find an unused user-space virtual memory area + * that can accommodate the size we want to map, and that is properly + * aligned to facilitate huge page table entries matching actual + * huge pages or huge page aligned memory in buffer objects. Buffer objects + * are assumed to start at huge page boundary pfns (io memory) or be + * populated by huge pages aligned to the start of the buffer object + * (system- or coherent memory). Adapted from shmem_get_unmapped_area. + * + * Return: aligned user-space address. + */ +unsigned long drm_get_unmapped_area(struct file *file, + unsigned long uaddr, unsigned long len, + unsigned long pgoff, unsigned long flags, + struct drm_vma_offset_manager *mgr) +{ + unsigned long addr; + unsigned long inflated_addr; + struct drm_vma_offset_node *node; + + if (len > TASK_SIZE) + return -ENOMEM; + + /* + * @pgoff is the file page-offset the huge page boundaries of + * which typically aligns to physical address huge page boundaries. + * That's not true for DRM, however, where physical address huge + * page boundaries instead are aligned with the offset from + * buffer object start. So adjust @pgoff to be the offset from + * buffer object start. + */ + drm_vma_offset_lock_lookup(mgr); + node = drm_vma_offset_lookup_locked(mgr, pgoff, 1); + if (node) + pgoff -= node->vm_node.start; + drm_vma_offset_unlock_lookup(mgr); + + addr = current->mm->get_unmapped_area(file, uaddr, len, pgoff, flags); + if (IS_ERR_VALUE(addr)) + return addr; + if (addr & ~PAGE_MASK) + return addr; + if (addr > TASK_SIZE - len) + return addr; + + if (len < HPAGE_PMD_SIZE) + return addr; + if (flags & MAP_FIXED) + return addr; + /* + * Our priority is to support MAP_SHARED mapped hugely; + * and support MAP_PRIVATE mapped hugely too, until it is COWed. + * But if caller specified an address hint, respect that as before. + */ + if (uaddr) + return addr; + + inflated_addr = drm_addr_inflate(addr, len, pgoff, flags, + HPAGE_PMD_SIZE); + + if (IS_ENABLED(CONFIG_HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD) && + len >= HPAGE_PUD_SIZE) + inflated_addr = drm_addr_inflate(inflated_addr, len, pgoff, + flags, HPAGE_PUD_SIZE); + return inflated_addr; +} +#else /* CONFIG_TRANSPARENT_HUGEPAGE */ +unsigned long drm_get_unmapped_area(struct file *file, + unsigned long uaddr, unsigned long len, + unsigned long pgoff, unsigned long flags, + struct drm_vma_offset_manager *mgr) +{ + return current->mm->get_unmapped_area(file, uaddr, len, pgoff, flags); +} +#endif /* CONFIG_TRANSPARENT_HUGEPAGE */ +EXPORT_SYMBOL_GPL(drm_get_unmapped_area); +#endif /* CONFIG_MMU */ diff --git a/drivers/gpu/drm/drm_format_helper.c b/drivers/gpu/drm/drm_format_helper.c index 0897cb9aeaff..3b818f2b2392 100644 --- a/drivers/gpu/drm/drm_format_helper.c +++ b/drivers/gpu/drm/drm_format_helper.c @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: GPL-2.0 */ +// SPDX-License-Identifier: GPL-2.0 or MIT /* * Copyright (C) 2016 Noralf Trønnes * diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c index a9e4a610445a..37627d06fb06 100644 --- a/drivers/gpu/drm/drm_gem.c +++ b/drivers/gpu/drm/drm_gem.c @@ -218,7 +218,7 @@ drm_gem_object_handle_put_unlocked(struct drm_gem_object *obj) struct drm_device *dev = obj->dev; bool final = false; - if (WARN_ON(obj->handle_count == 0)) + if (WARN_ON(READ_ONCE(obj->handle_count) == 0)) return; /* diff --git a/drivers/gpu/drm/drm_gem_shmem_helper.c b/drivers/gpu/drm/drm_gem_shmem_helper.c index a421a2eed48a..df31e5782eed 100644 --- a/drivers/gpu/drm/drm_gem_shmem_helper.c +++ b/drivers/gpu/drm/drm_gem_shmem_helper.c @@ -254,11 +254,16 @@ static void *drm_gem_shmem_vmap_locked(struct drm_gem_shmem_object *shmem) if (ret) goto err_zero_use; - if (obj->import_attach) + if (obj->import_attach) { shmem->vaddr = dma_buf_vmap(obj->import_attach->dmabuf); - else + } else { + pgprot_t prot = PAGE_KERNEL; + + if (!shmem->map_cached) + prot = pgprot_writecombine(prot); shmem->vaddr = vmap(shmem->pages, obj->size >> PAGE_SHIFT, - VM_MAP, pgprot_writecombine(PAGE_KERNEL)); + VM_MAP, prot); + } if (!shmem->vaddr) { DRM_DEBUG_KMS("Failed to vmap pages\n"); @@ -540,8 +545,9 @@ int drm_gem_shmem_mmap(struct drm_gem_object *obj, struct vm_area_struct *vma) } vma->vm_flags |= VM_MIXEDMAP | VM_DONTEXPAND; - vma->vm_page_prot = pgprot_writecombine(vm_get_page_prot(vma->vm_flags)); - vma->vm_page_prot = pgprot_decrypted(vma->vm_page_prot); + vma->vm_page_prot = vm_get_page_prot(vma->vm_flags); + if (!shmem->map_cached) + vma->vm_page_prot = pgprot_writecombine(vma->vm_page_prot); vma->vm_ops = &drm_gem_shmem_vm_ops; return 0; diff --git a/drivers/gpu/drm/drm_hdcp.c b/drivers/gpu/drm/drm_hdcp.c index 9191633a3c43..7f386adcf872 100644 --- a/drivers/gpu/drm/drm_hdcp.c +++ b/drivers/gpu/drm/drm_hdcp.c @@ -23,14 +23,6 @@ #include "drm_internal.h" -static struct hdcp_srm { - u32 revoked_ksv_cnt; - u8 *revoked_ksv_list; - - /* Mutex to protect above struct member */ - struct mutex mutex; -} *srm_data; - static inline void drm_hdcp_print_ksv(const u8 *ksv) { DRM_DEBUG("\t%#02x, %#02x, %#02x, %#02x, %#02x\n", @@ -60,11 +52,11 @@ static u32 drm_hdcp_get_revoked_ksv_count(const u8 *buf, u32 vrls_length) return ksv_count; } -static u32 drm_hdcp_get_revoked_ksvs(const u8 *buf, u8 *revoked_ksv_list, +static u32 drm_hdcp_get_revoked_ksvs(const u8 *buf, u8 **revoked_ksv_list, u32 vrls_length) { - u32 parsed_bytes = 0, ksv_count = 0; u32 vrl_ksv_cnt, vrl_ksv_sz, vrl_idx = 0; + u32 parsed_bytes = 0, ksv_count = 0; do { vrl_ksv_cnt = *buf; @@ -74,10 +66,10 @@ static u32 drm_hdcp_get_revoked_ksvs(const u8 *buf, u8 *revoked_ksv_list, DRM_DEBUG("vrl: %d, Revoked KSVs: %d\n", vrl_idx++, vrl_ksv_cnt); - memcpy(revoked_ksv_list, buf, vrl_ksv_sz); + memcpy((*revoked_ksv_list) + (ksv_count * DRM_HDCP_KSV_LEN), + buf, vrl_ksv_sz); ksv_count += vrl_ksv_cnt; - revoked_ksv_list += vrl_ksv_sz; buf += vrl_ksv_sz; parsed_bytes += (vrl_ksv_sz + 1); @@ -91,7 +83,8 @@ static inline u32 get_vrl_length(const u8 *buf) return drm_hdcp_be24_to_cpu(buf); } -static int drm_hdcp_parse_hdcp1_srm(const u8 *buf, size_t count) +static int drm_hdcp_parse_hdcp1_srm(const u8 *buf, size_t count, + u8 **revoked_ksv_list, u32 *revoked_ksv_cnt) { struct hdcp_srm_header *header; u32 vrl_length, ksv_count; @@ -131,29 +124,28 @@ static int drm_hdcp_parse_hdcp1_srm(const u8 *buf, size_t count) ksv_count = drm_hdcp_get_revoked_ksv_count(buf, vrl_length); if (!ksv_count) { DRM_DEBUG("Revoked KSV count is 0\n"); - return count; + return 0; } - kfree(srm_data->revoked_ksv_list); - srm_data->revoked_ksv_list = kcalloc(ksv_count, DRM_HDCP_KSV_LEN, - GFP_KERNEL); - if (!srm_data->revoked_ksv_list) { + *revoked_ksv_list = kcalloc(ksv_count, DRM_HDCP_KSV_LEN, GFP_KERNEL); + if (!*revoked_ksv_list) { DRM_ERROR("Out of Memory\n"); return -ENOMEM; } - if (drm_hdcp_get_revoked_ksvs(buf, srm_data->revoked_ksv_list, + if (drm_hdcp_get_revoked_ksvs(buf, revoked_ksv_list, vrl_length) != ksv_count) { - srm_data->revoked_ksv_cnt = 0; - kfree(srm_data->revoked_ksv_list); + *revoked_ksv_cnt = 0; + kfree(*revoked_ksv_list); return -EINVAL; } - srm_data->revoked_ksv_cnt = ksv_count; - return count; + *revoked_ksv_cnt = ksv_count; + return 0; } -static int drm_hdcp_parse_hdcp2_srm(const u8 *buf, size_t count) +static int drm_hdcp_parse_hdcp2_srm(const u8 *buf, size_t count, + u8 **revoked_ksv_list, u32 *revoked_ksv_cnt) { struct hdcp_srm_header *header; u32 vrl_length, ksv_count, ksv_sz; @@ -195,13 +187,11 @@ static int drm_hdcp_parse_hdcp2_srm(const u8 *buf, size_t count) ksv_count = (*buf << 2) | DRM_HDCP_2_KSV_COUNT_2_LSBITS(*(buf + 1)); if (!ksv_count) { DRM_DEBUG("Revoked KSV count is 0\n"); - return count; + return 0; } - kfree(srm_data->revoked_ksv_list); - srm_data->revoked_ksv_list = kcalloc(ksv_count, DRM_HDCP_KSV_LEN, - GFP_KERNEL); - if (!srm_data->revoked_ksv_list) { + *revoked_ksv_list = kcalloc(ksv_count, DRM_HDCP_KSV_LEN, GFP_KERNEL); + if (!*revoked_ksv_list) { DRM_ERROR("Out of Memory\n"); return -ENOMEM; } @@ -210,10 +200,10 @@ static int drm_hdcp_parse_hdcp2_srm(const u8 *buf, size_t count) buf += DRM_HDCP_2_NO_OF_DEV_PLUS_RESERVED_SZ; DRM_DEBUG("Revoked KSVs: %d\n", ksv_count); - memcpy(srm_data->revoked_ksv_list, buf, ksv_sz); + memcpy(*revoked_ksv_list, buf, ksv_sz); - srm_data->revoked_ksv_cnt = ksv_count; - return count; + *revoked_ksv_cnt = ksv_count; + return 0; } static inline bool is_srm_version_hdcp1(const u8 *buf) @@ -226,22 +216,27 @@ static inline bool is_srm_version_hdcp2(const u8 *buf) return *buf == (u8)(DRM_HDCP_2_SRM_ID << 4 | DRM_HDCP_2_INDICATOR); } -static void drm_hdcp_srm_update(const u8 *buf, size_t count) +static int drm_hdcp_srm_update(const u8 *buf, size_t count, + u8 **revoked_ksv_list, u32 *revoked_ksv_cnt) { if (count < sizeof(struct hdcp_srm_header)) - return; + return -EINVAL; if (is_srm_version_hdcp1(buf)) - drm_hdcp_parse_hdcp1_srm(buf, count); + return drm_hdcp_parse_hdcp1_srm(buf, count, revoked_ksv_list, + revoked_ksv_cnt); else if (is_srm_version_hdcp2(buf)) - drm_hdcp_parse_hdcp2_srm(buf, count); + return drm_hdcp_parse_hdcp2_srm(buf, count, revoked_ksv_list, + revoked_ksv_cnt); + else + return -EINVAL; } -static void drm_hdcp_request_srm(struct drm_device *drm_dev) +static int drm_hdcp_request_srm(struct drm_device *drm_dev, + u8 **revoked_ksv_list, u32 *revoked_ksv_cnt) { char fw_name[36] = "display_hdcp_srm.bin"; const struct firmware *fw; - int ret; ret = request_firmware_direct(&fw, (const char *)fw_name, @@ -250,10 +245,12 @@ static void drm_hdcp_request_srm(struct drm_device *drm_dev) goto exit; if (fw->size && fw->data) - drm_hdcp_srm_update(fw->data, fw->size); + ret = drm_hdcp_srm_update(fw->data, fw->size, revoked_ksv_list, + revoked_ksv_cnt); exit: release_firmware(fw); + return ret; } /** @@ -279,71 +276,34 @@ exit: * https://www.digital-cp.com/sites/default/files/specifications/HDCP%20on%20HDMI%20Specification%20Rev2_2_Final1.pdf * * Returns: - * TRUE on any of the KSV is revoked, else FALSE. + * Count of the revoked KSVs or -ve error number incase of the failure. */ -bool drm_hdcp_check_ksvs_revoked(struct drm_device *drm_dev, u8 *ksvs, - u32 ksv_count) +int drm_hdcp_check_ksvs_revoked(struct drm_device *drm_dev, u8 *ksvs, + u32 ksv_count) { - u32 rev_ksv_cnt, cnt, i, j; - u8 *rev_ksv_list; - - if (!srm_data) - return false; - - mutex_lock(&srm_data->mutex); - drm_hdcp_request_srm(drm_dev); - - rev_ksv_cnt = srm_data->revoked_ksv_cnt; - rev_ksv_list = srm_data->revoked_ksv_list; - - /* If the Revoked ksv list is empty */ - if (!rev_ksv_cnt || !rev_ksv_list) { - mutex_unlock(&srm_data->mutex); - return false; - } - - for (cnt = 0; cnt < ksv_count; cnt++) { - rev_ksv_list = srm_data->revoked_ksv_list; - for (i = 0; i < rev_ksv_cnt; i++) { - for (j = 0; j < DRM_HDCP_KSV_LEN; j++) - if (ksvs[j] != rev_ksv_list[j]) { - break; - } else if (j == (DRM_HDCP_KSV_LEN - 1)) { - DRM_DEBUG("Revoked KSV is "); - drm_hdcp_print_ksv(ksvs); - mutex_unlock(&srm_data->mutex); - return true; - } - /* Move the offset to next KSV in the revoked list */ - rev_ksv_list += DRM_HDCP_KSV_LEN; - } - - /* Iterate to next ksv_offset */ - ksvs += DRM_HDCP_KSV_LEN; - } - mutex_unlock(&srm_data->mutex); - return false; + u32 revoked_ksv_cnt = 0, i, j; + u8 *revoked_ksv_list = NULL; + int ret = 0; + + ret = drm_hdcp_request_srm(drm_dev, &revoked_ksv_list, + &revoked_ksv_cnt); + + /* revoked_ksv_cnt will be zero when above function failed */ + for (i = 0; i < revoked_ksv_cnt; i++) + for (j = 0; j < ksv_count; j++) + if (!memcmp(&ksvs[j * DRM_HDCP_KSV_LEN], + &revoked_ksv_list[i * DRM_HDCP_KSV_LEN], + DRM_HDCP_KSV_LEN)) { + DRM_DEBUG("Revoked KSV is "); + drm_hdcp_print_ksv(&ksvs[j * DRM_HDCP_KSV_LEN]); + ret++; + } + + kfree(revoked_ksv_list); + return ret; } EXPORT_SYMBOL_GPL(drm_hdcp_check_ksvs_revoked); -int drm_setup_hdcp_srm(struct class *drm_class) -{ - srm_data = kzalloc(sizeof(*srm_data), GFP_KERNEL); - if (!srm_data) - return -ENOMEM; - mutex_init(&srm_data->mutex); - - return 0; -} - -void drm_teardown_hdcp_srm(struct class *drm_class) -{ - if (srm_data) { - kfree(srm_data->revoked_ksv_list); - kfree(srm_data); - } -} - static struct drm_prop_enum_list drm_cp_enum_list[] = { { DRM_MODE_CONTENT_PROTECTION_UNDESIRED, "Undesired" }, { DRM_MODE_CONTENT_PROTECTION_DESIRED, "Desired" }, diff --git a/drivers/gpu/drm/drm_internal.h b/drivers/gpu/drm/drm_internal.h index 6937bf923f05..5714a78365ac 100644 --- a/drivers/gpu/drm/drm_internal.h +++ b/drivers/gpu/drm/drm_internal.h @@ -41,6 +41,7 @@ struct drm_printer; /* drm_file.c */ extern struct mutex drm_global_mutex; +bool drm_dev_needs_global_mutex(struct drm_device *dev); struct drm_file *drm_file_alloc(struct drm_minor *minor); void drm_file_free(struct drm_file *file); void drm_lastclose(struct drm_device *dev); @@ -235,7 +236,3 @@ int drm_syncobj_query_ioctl(struct drm_device *dev, void *data, void drm_framebuffer_print_info(struct drm_printer *p, unsigned int indent, const struct drm_framebuffer *fb); int drm_framebuffer_debugfs_init(struct drm_minor *minor); - -/* drm_hdcp.c */ -int drm_setup_hdcp_srm(struct class *drm_class); -void drm_teardown_hdcp_srm(struct class *drm_class); diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c index 03bce566a8c3..588be45abd7a 100644 --- a/drivers/gpu/drm/drm_irq.c +++ b/drivers/gpu/drm/drm_irq.c @@ -111,10 +111,6 @@ int drm_irq_install(struct drm_device *dev, int irq) if (irq == 0) return -EINVAL; - /* Driver must have been initialized */ - if (!dev->dev_private) - return -EINVAL; - if (dev->irq_enabled) return -EBUSY; dev->irq_enabled = true; diff --git a/drivers/gpu/drm/drm_lease.c b/drivers/gpu/drm/drm_lease.c index b481cafdde28..825abe38201a 100644 --- a/drivers/gpu/drm/drm_lease.c +++ b/drivers/gpu/drm/drm_lease.c @@ -542,10 +542,12 @@ int drm_mode_create_lease_ioctl(struct drm_device *dev, } DRM_DEBUG_LEASE("Creating lease\n"); + /* lessee will take the ownership of leases */ lessee = drm_lease_create(lessor, &leases); if (IS_ERR(lessee)) { ret = PTR_ERR(lessee); + idr_destroy(&leases); goto out_leases; } @@ -580,7 +582,6 @@ out_lessee: out_leases: put_unused_fd(fd); - idr_destroy(&leases); DRM_DEBUG_LEASE("drm_mode_create_lease_ioctl failed: %d\n", ret); return ret; diff --git a/drivers/gpu/drm/drm_lock.c b/drivers/gpu/drm/drm_lock.c index 2c79e8199e3c..f16eefbf2829 100644 --- a/drivers/gpu/drm/drm_lock.c +++ b/drivers/gpu/drm/drm_lock.c @@ -46,7 +46,7 @@ static int drm_lock_take(struct drm_lock_data *lock_data, unsigned int context); -/** +/* * Take the heavyweight lock. * * \param lock lock pointer. @@ -93,7 +93,7 @@ int drm_lock_take(struct drm_lock_data *lock_data, return 0; } -/** +/* * This takes a lock forcibly and hands it to context. Should ONLY be used * inside *_unlock to give lock to kernel before calling *_dma_schedule. * @@ -150,7 +150,7 @@ static int drm_legacy_lock_free(struct drm_lock_data *lock_data, return 0; } -/** +/* * Lock ioctl. * * \param inode device inode. @@ -243,7 +243,7 @@ int drm_legacy_lock(struct drm_device *dev, void *data, return 0; } -/** +/* * Unlock ioctl. * * \param inode device inode. @@ -275,7 +275,7 @@ int drm_legacy_unlock(struct drm_device *dev, void *data, struct drm_file *file_ return 0; } -/** +/* * This function returns immediately and takes the hw lock * with the kernel context if it is free, otherwise it gets the highest priority when and if * it is eventually released. @@ -287,7 +287,6 @@ int drm_legacy_unlock(struct drm_device *dev, void *data, struct drm_file *file_ * This should be sufficient to wait for GPU idle without * having to worry about starvation. */ - void drm_legacy_idlelock_take(struct drm_lock_data *lock_data) { int ret; diff --git a/drivers/gpu/drm/drm_mm.c b/drivers/gpu/drm/drm_mm.c index 2a6e34663146..8981abe8b7c9 100644 --- a/drivers/gpu/drm/drm_mm.c +++ b/drivers/gpu/drm/drm_mm.c @@ -399,10 +399,10 @@ next_hole(struct drm_mm *mm, */ int drm_mm_reserve_node(struct drm_mm *mm, struct drm_mm_node *node) { - u64 end = node->start + node->size; struct drm_mm_node *hole; u64 hole_start, hole_end; u64 adj_start, adj_end; + u64 end; end = node->start + node->size; if (unlikely(end <= node->start)) diff --git a/drivers/gpu/drm/drm_modes.c b/drivers/gpu/drm/drm_modes.c index 10336b144c72..d4d64518e11b 100644 --- a/drivers/gpu/drm/drm_modes.c +++ b/drivers/gpu/drm/drm_modes.c @@ -1698,6 +1698,13 @@ static int drm_mode_parse_cmdline_options(const char *str, if (rotation && freestanding) return -EINVAL; + if (!(rotation & DRM_MODE_ROTATE_MASK)) + rotation |= DRM_MODE_ROTATE_0; + + /* Make sure there is exactly one rotation defined */ + if (!is_power_of_2(rotation & DRM_MODE_ROTATE_MASK)) + return -EINVAL; + mode->rotation_reflection = rotation; return 0; diff --git a/drivers/gpu/drm/drm_pci.c b/drivers/gpu/drm/drm_pci.c index c6bb98729a26..81aa21561982 100644 --- a/drivers/gpu/drm/drm_pci.c +++ b/drivers/gpu/drm/drm_pci.c @@ -75,7 +75,6 @@ drm_dma_handle_t *drm_pci_alloc(struct drm_device * dev, size_t size, size_t ali return dmah; } - EXPORT_SYMBOL(drm_pci_alloc); /** @@ -167,6 +166,18 @@ int drm_irq_by_busid(struct drm_device *dev, void *data, return drm_pci_irq_by_busid(dev, p); } +void drm_pci_agp_destroy(struct drm_device *dev) +{ + if (dev->agp) { + arch_phys_wc_del(dev->agp->agp_mtrr); + drm_legacy_agp_clear(dev); + kfree(dev->agp); + dev->agp = NULL; + } +} + +#ifdef CONFIG_DRM_LEGACY + static void drm_pci_agp_init(struct drm_device *dev) { if (drm_core_check_feature(dev, DRIVER_USE_AGP)) { @@ -181,33 +192,9 @@ static void drm_pci_agp_init(struct drm_device *dev) } } -void drm_pci_agp_destroy(struct drm_device *dev) -{ - if (dev->agp) { - arch_phys_wc_del(dev->agp->agp_mtrr); - drm_legacy_agp_clear(dev); - kfree(dev->agp); - dev->agp = NULL; - } -} - -/** - * drm_get_pci_dev - Register a PCI device with the DRM subsystem - * @pdev: PCI device - * @ent: entry from the PCI ID table that matches @pdev - * @driver: DRM device driver - * - * Attempt to gets inter module "drm" information. If we are first - * then register the character device and inter module information. - * Try and register, if we fail to register, backout previous work. - * - * NOTE: This function is deprecated, please use drm_dev_alloc() and - * drm_dev_register() instead and remove your &drm_driver.load callback. - * - * Return: 0 on success or a negative error code on failure. - */ -int drm_get_pci_dev(struct pci_dev *pdev, const struct pci_device_id *ent, - struct drm_driver *driver) +static int drm_get_pci_dev(struct pci_dev *pdev, + const struct pci_device_id *ent, + struct drm_driver *driver) { struct drm_device *dev; int ret; @@ -250,9 +237,6 @@ err_free: drm_dev_put(dev); return ret; } -EXPORT_SYMBOL(drm_get_pci_dev); - -#ifdef CONFIG_DRM_LEGACY /** * drm_legacy_pci_init - shadow-attach a legacy DRM PCI driver diff --git a/drivers/gpu/drm/drm_prime.c b/drivers/gpu/drm/drm_prime.c index 86d9b0e45c8c..282774e469ac 100644 --- a/drivers/gpu/drm/drm_prime.c +++ b/drivers/gpu/drm/drm_prime.c @@ -962,27 +962,40 @@ int drm_prime_sg_to_page_addr_arrays(struct sg_table *sgt, struct page **pages, unsigned count; struct scatterlist *sg; struct page *page; - u32 len, index; + u32 page_len, page_index; dma_addr_t addr; + u32 dma_len, dma_index; - index = 0; + /* + * Scatterlist elements contains both pages and DMA addresses, but + * one shoud not assume 1:1 relation between them. The sg->length is + * the size of the physical memory chunk described by the sg->page, + * while sg_dma_len(sg) is the size of the DMA (IO virtual) chunk + * described by the sg_dma_address(sg). + */ + page_index = 0; + dma_index = 0; for_each_sg(sgt->sgl, sg, sgt->nents, count) { - len = sg->length; + page_len = sg->length; page = sg_page(sg); + dma_len = sg_dma_len(sg); addr = sg_dma_address(sg); - while (len > 0) { - if (WARN_ON(index >= max_entries)) + while (pages && page_len > 0) { + if (WARN_ON(page_index >= max_entries)) return -1; - if (pages) - pages[index] = page; - if (addrs) - addrs[index] = addr; - + pages[page_index] = page; page++; + page_len -= PAGE_SIZE; + page_index++; + } + while (addrs && dma_len > 0) { + if (WARN_ON(dma_index >= max_entries)) + return -1; + addrs[dma_index] = addr; addr += PAGE_SIZE; - len -= PAGE_SIZE; - index++; + dma_len -= PAGE_SIZE; + dma_index++; } } return 0; diff --git a/drivers/gpu/drm/drm_scatter.c b/drivers/gpu/drm/drm_scatter.c index d5c386154246..ca520028b2cb 100644 --- a/drivers/gpu/drm/drm_scatter.c +++ b/drivers/gpu/drm/drm_scatter.c @@ -99,6 +99,9 @@ int drm_legacy_sg_alloc(struct drm_device *dev, void *data, if (!drm_core_check_feature(dev, DRIVER_SG)) return -EOPNOTSUPP; + if (request->size > SIZE_MAX - PAGE_SIZE) + return -EINVAL; + if (dev->sg) return -EINVAL; diff --git a/drivers/gpu/drm/drm_simple_kms_helper.c b/drivers/gpu/drm/drm_simple_kms_helper.c index 15fb516ae2d8..74946690aba4 100644 --- a/drivers/gpu/drm/drm_simple_kms_helper.c +++ b/drivers/gpu/drm/drm_simple_kms_helper.c @@ -26,12 +26,51 @@ * entity. Some flexibility for code reuse is provided through a separately * allocated &drm_connector object and supporting optional &drm_bridge * encoder drivers. + * + * Many drivers require only a very simple encoder that fulfills the minimum + * requirements of the display pipeline and does not add additional + * functionality. The function drm_simple_encoder_init() provides an + * implementation of such an encoder. */ -static const struct drm_encoder_funcs drm_simple_kms_encoder_funcs = { +static const struct drm_encoder_funcs drm_simple_encoder_funcs_cleanup = { .destroy = drm_encoder_cleanup, }; +/** + * drm_simple_encoder_init - Initialize a preallocated encoder with + * basic functionality. + * @dev: drm device + * @encoder: the encoder to initialize + * @encoder_type: user visible type of the encoder + * + * Initialises a preallocated encoder that has no further functionality. + * Settings for possible CRTC and clones are left to their initial values. + * The encoder will be cleaned up automatically as part of the mode-setting + * cleanup. + * + * The caller of drm_simple_encoder_init() is responsible for freeing + * the encoder's memory after the encoder has been cleaned up. At the + * moment this only works reliably if the encoder data structure is + * stored in the device structure. Free the encoder's memory as part of + * the device release function. + * + * FIXME: Later improvements to DRM's resource management may allow for + * an automated kfree() of the encoder's memory. + * + * Returns: + * Zero on success, error code on failure. + */ +int drm_simple_encoder_init(struct drm_device *dev, + struct drm_encoder *encoder, + int encoder_type) +{ + return drm_encoder_init(dev, encoder, + &drm_simple_encoder_funcs_cleanup, + encoder_type, NULL); +} +EXPORT_SYMBOL(drm_simple_encoder_init); + static enum drm_mode_status drm_simple_kms_crtc_mode_valid(struct drm_crtc *crtc, const struct drm_display_mode *mode) @@ -229,7 +268,7 @@ static const struct drm_plane_funcs drm_simple_kms_plane_funcs = { int drm_simple_display_pipe_attach_bridge(struct drm_simple_display_pipe *pipe, struct drm_bridge *bridge) { - return drm_bridge_attach(&pipe->encoder, bridge, NULL); + return drm_bridge_attach(&pipe->encoder, bridge, NULL, 0); } EXPORT_SYMBOL(drm_simple_display_pipe_attach_bridge); @@ -288,8 +327,7 @@ int drm_simple_display_pipe_init(struct drm_device *dev, return ret; encoder->possible_crtcs = drm_crtc_mask(crtc); - ret = drm_encoder_init(dev, encoder, &drm_simple_kms_encoder_funcs, - DRM_MODE_ENCODER_NONE, NULL); + ret = drm_simple_encoder_init(dev, encoder, DRM_MODE_ENCODER_NONE); if (ret || !connector) return ret; diff --git a/drivers/gpu/drm/drm_sysfs.c b/drivers/gpu/drm/drm_sysfs.c index dd2bc85f43cc..939f0032aab1 100644 --- a/drivers/gpu/drm/drm_sysfs.c +++ b/drivers/gpu/drm/drm_sysfs.c @@ -85,7 +85,6 @@ int drm_sysfs_init(void) } drm_class->devnode = drm_devnode; - drm_setup_hdcp_srm(drm_class); return 0; } @@ -98,7 +97,6 @@ void drm_sysfs_destroy(void) { if (IS_ERR_OR_NULL(drm_class)) return; - drm_teardown_hdcp_srm(drm_class); class_remove_file(drm_class, &class_attr_version.attr); class_destroy(drm_class); drm_class = NULL; @@ -230,7 +228,7 @@ static ssize_t modes_show(struct device *device, mutex_lock(&connector->dev->mode_config.mutex); list_for_each_entry(mode, &connector->modes, head) { - written += snprintf(buf + written, PAGE_SIZE - written, "%s\n", + written += scnprintf(buf + written, PAGE_SIZE - written, "%s\n", mode->name); } mutex_unlock(&connector->dev->mode_config.mutex); diff --git a/drivers/gpu/drm/drm_vblank.c b/drivers/gpu/drm/drm_vblank.c index 27a99c486f02..da7b0b0c1090 100644 --- a/drivers/gpu/drm/drm_vblank.c +++ b/drivers/gpu/drm/drm_vblank.c @@ -30,6 +30,7 @@ #include <drm/drm_crtc.h> #include <drm/drm_drv.h> #include <drm/drm_framebuffer.h> +#include <drm/drm_modeset_helper_vtables.h> #include <drm/drm_print.h> #include <drm/drm_vblank.h> @@ -143,10 +144,9 @@ static u32 __get_vblank_counter(struct drm_device *dev, unsigned int pipe) if (crtc->funcs->get_vblank_counter) return crtc->funcs->get_vblank_counter(crtc); - } - - if (dev->driver->get_vblank_counter) + } else if (dev->driver->get_vblank_counter) { return dev->driver->get_vblank_counter(dev, pipe); + } return drm_vblank_no_hw_counter(dev, pipe); } @@ -338,7 +338,8 @@ u64 drm_crtc_accurate_vblank_count(struct drm_crtc *crtc) u64 vblank; unsigned long flags; - WARN_ONCE(drm_debug_enabled(DRM_UT_VBL) && !dev->driver->get_vblank_timestamp, + WARN_ONCE(drm_debug_enabled(DRM_UT_VBL) && + !crtc->funcs->get_vblank_timestamp, "This function requires support for accurate vblank timestamps."); spin_lock_irqsave(&dev->vblank_time_lock, flags); @@ -360,13 +361,11 @@ static void __disable_vblank(struct drm_device *dev, unsigned int pipe) if (WARN_ON(!crtc)) return; - if (crtc->funcs->disable_vblank) { + if (crtc->funcs->disable_vblank) crtc->funcs->disable_vblank(crtc); - return; - } + } else { + dev->driver->disable_vblank(dev, pipe); } - - dev->driver->disable_vblank(dev, pipe); } /* @@ -486,19 +485,6 @@ int drm_vblank_init(struct drm_device *dev, unsigned int num_crtcs) DRM_INFO("Supports vblank timestamp caching Rev 2 (21.10.2013).\n"); - /* Driver specific high-precision vblank timestamping supported? */ - if (dev->driver->get_vblank_timestamp) - DRM_INFO("Driver supports precise vblank timestamp query.\n"); - else - DRM_INFO("No driver support for vblank timestamp query.\n"); - - /* Must have precise timestamping for reliable vblank instant disable */ - if (dev->vblank_disable_immediate && !dev->driver->get_vblank_timestamp) { - dev->vblank_disable_immediate = false; - DRM_INFO("Setting vblank_disable_immediate to false because " - "get_vblank_timestamp == NULL\n"); - } - return 0; err: @@ -551,9 +537,9 @@ EXPORT_SYMBOL(drm_crtc_vblank_waitqueue); * * Calculate and store various constants which are later needed by vblank and * swap-completion timestamping, e.g, by - * drm_calc_vbltimestamp_from_scanoutpos(). They are derived from CRTC's true - * scanout timing, so they take things like panel scaling or other adjustments - * into account. + * drm_crtc_vblank_helper_get_vblank_timestamp(). They are derived from + * CRTC's true scanout timing, so they take things like panel scaling or + * other adjustments into account. */ void drm_calc_timestamping_constants(struct drm_crtc *crtc, const struct drm_display_mode *mode) @@ -604,9 +590,9 @@ void drm_calc_timestamping_constants(struct drm_crtc *crtc, EXPORT_SYMBOL(drm_calc_timestamping_constants); /** - * drm_calc_vbltimestamp_from_scanoutpos - precise vblank timestamp helper - * @dev: DRM device - * @pipe: index of CRTC whose vblank timestamp to retrieve + * drm_crtc_vblank_helper_get_vblank_timestamp_internal - precise vblank + * timestamp helper + * @crtc: CRTC whose vblank timestamp to retrieve * @max_error: Desired maximum allowable error in timestamps (nanosecs) * On return contains true maximum error of timestamp * @vblank_time: Pointer to time which should receive the timestamp @@ -614,11 +600,12 @@ EXPORT_SYMBOL(drm_calc_timestamping_constants); * True when called from drm_crtc_handle_vblank(). Some drivers * need to apply some workarounds for gpu-specific vblank irq quirks * if flag is set. + * @get_scanout_position: + * Callback function to retrieve the scanout position. See + * @struct drm_crtc_helper_funcs.get_scanout_position. * * Implements calculation of exact vblank timestamps from given drm_display_mode - * timings and current video scanout position of a CRTC. This can be directly - * used as the &drm_driver.get_vblank_timestamp implementation of a kms driver - * if &drm_driver.get_scanout_position is implemented. + * timings and current video scanout position of a CRTC. * * The current implementation only handles standard video modes. For double scan * and interlaced modes the driver is supposed to adjust the hardware mode @@ -634,34 +621,30 @@ EXPORT_SYMBOL(drm_calc_timestamping_constants); * Returns true on success, and false on failure, i.e. when no accurate * timestamp could be acquired. */ -bool drm_calc_vbltimestamp_from_scanoutpos(struct drm_device *dev, - unsigned int pipe, - int *max_error, - ktime_t *vblank_time, - bool in_vblank_irq) +bool +drm_crtc_vblank_helper_get_vblank_timestamp_internal( + struct drm_crtc *crtc, int *max_error, ktime_t *vblank_time, + bool in_vblank_irq, + drm_vblank_get_scanout_position_func get_scanout_position) { + struct drm_device *dev = crtc->dev; + unsigned int pipe = crtc->index; + struct drm_vblank_crtc *vblank = &dev->vblank[pipe]; struct timespec64 ts_etime, ts_vblank_time; ktime_t stime, etime; bool vbl_status; - struct drm_crtc *crtc; const struct drm_display_mode *mode; - struct drm_vblank_crtc *vblank = &dev->vblank[pipe]; int vpos, hpos, i; int delta_ns, duration_ns; - if (!drm_core_check_feature(dev, DRIVER_MODESET)) - return false; - - crtc = drm_crtc_from_index(dev, pipe); - - if (pipe >= dev->num_crtcs || !crtc) { + if (pipe >= dev->num_crtcs) { DRM_ERROR("Invalid crtc %u\n", pipe); return false; } /* Scanout position query not supported? Should not happen. */ - if (!dev->driver->get_scanout_position) { - DRM_ERROR("Called from driver w/o get_scanout_position()!?\n"); + if (!get_scanout_position) { + DRM_ERROR("Called from CRTC w/o get_scanout_position()!?\n"); return false; } @@ -676,7 +659,6 @@ bool drm_calc_vbltimestamp_from_scanoutpos(struct drm_device *dev, if (mode->crtc_clock == 0) { DRM_DEBUG("crtc %u: Noop due to uninitialized mode.\n", pipe); WARN_ON_ONCE(drm_drv_uses_atomic_modeset(dev)); - return false; } @@ -692,11 +674,10 @@ bool drm_calc_vbltimestamp_from_scanoutpos(struct drm_device *dev, * Get vertical and horizontal scanout position vpos, hpos, * and bounding timestamps stime, etime, pre/post query. */ - vbl_status = dev->driver->get_scanout_position(dev, pipe, - in_vblank_irq, - &vpos, &hpos, - &stime, &etime, - mode); + vbl_status = get_scanout_position(crtc, in_vblank_irq, + &vpos, &hpos, + &stime, &etime, + mode); /* Return as no-op if scanout query unsupported or failed. */ if (!vbl_status) { @@ -748,7 +729,49 @@ bool drm_calc_vbltimestamp_from_scanoutpos(struct drm_device *dev, return true; } -EXPORT_SYMBOL(drm_calc_vbltimestamp_from_scanoutpos); +EXPORT_SYMBOL(drm_crtc_vblank_helper_get_vblank_timestamp_internal); + +/** + * drm_crtc_vblank_helper_get_vblank_timestamp - precise vblank timestamp + * helper + * @crtc: CRTC whose vblank timestamp to retrieve + * @max_error: Desired maximum allowable error in timestamps (nanosecs) + * On return contains true maximum error of timestamp + * @vblank_time: Pointer to time which should receive the timestamp + * @in_vblank_irq: + * True when called from drm_crtc_handle_vblank(). Some drivers + * need to apply some workarounds for gpu-specific vblank irq quirks + * if flag is set. + * + * Implements calculation of exact vblank timestamps from given drm_display_mode + * timings and current video scanout position of a CRTC. This can be directly + * used as the &drm_crtc_funcs.get_vblank_timestamp implementation of a kms + * driver if &drm_crtc_helper_funcs.get_scanout_position is implemented. + * + * The current implementation only handles standard video modes. For double scan + * and interlaced modes the driver is supposed to adjust the hardware mode + * (taken from &drm_crtc_state.adjusted mode for atomic modeset drivers) to + * match the scanout position reported. + * + * Note that atomic drivers must call drm_calc_timestamping_constants() before + * enabling a CRTC. The atomic helpers already take care of that in + * drm_atomic_helper_update_legacy_modeset_state(). + * + * Returns: + * + * Returns true on success, and false on failure, i.e. when no accurate + * timestamp could be acquired. + */ +bool drm_crtc_vblank_helper_get_vblank_timestamp(struct drm_crtc *crtc, + int *max_error, + ktime_t *vblank_time, + bool in_vblank_irq) +{ + return drm_crtc_vblank_helper_get_vblank_timestamp_internal( + crtc, max_error, vblank_time, in_vblank_irq, + crtc->helper_private->get_scanout_position); +} +EXPORT_SYMBOL(drm_crtc_vblank_helper_get_vblank_timestamp); /** * drm_get_last_vbltimestamp - retrieve raw timestamp for the most recent @@ -775,15 +798,19 @@ static bool drm_get_last_vbltimestamp(struct drm_device *dev, unsigned int pipe, ktime_t *tvblank, bool in_vblank_irq) { + struct drm_crtc *crtc = drm_crtc_from_index(dev, pipe); bool ret = false; /* Define requested maximum error on timestamps (nanoseconds). */ int max_error = (int) drm_timestamp_precision * 1000; /* Query driver if possible and precision timestamping enabled. */ - if (dev->driver->get_vblank_timestamp && (max_error > 0)) - ret = dev->driver->get_vblank_timestamp(dev, pipe, &max_error, + if (crtc && crtc->funcs->get_vblank_timestamp && max_error > 0) { + struct drm_crtc *crtc = drm_crtc_from_index(dev, pipe); + + ret = crtc->funcs->get_vblank_timestamp(crtc, &max_error, tvblank, in_vblank_irq); + } /* GPU high precision timestamp query unsupported or failed. * Return current monotonic/gettimeofday timestamp as best estimate. @@ -1005,9 +1032,11 @@ static int __enable_vblank(struct drm_device *dev, unsigned int pipe) if (crtc->funcs->enable_vblank) return crtc->funcs->enable_vblank(crtc); + } else if (dev->driver->enable_vblank) { + return dev->driver->enable_vblank(dev, pipe); } - return dev->driver->enable_vblank(dev, pipe); + return -EINVAL; } static int drm_vblank_enable(struct drm_device *dev, unsigned int pipe) @@ -1766,6 +1795,8 @@ done: static void drm_handle_vblank_events(struct drm_device *dev, unsigned int pipe) { + struct drm_crtc *crtc = drm_crtc_from_index(dev, pipe); + bool high_prec = false; struct drm_pending_vblank_event *e, *t; ktime_t now; u64 seq; @@ -1788,8 +1819,10 @@ static void drm_handle_vblank_events(struct drm_device *dev, unsigned int pipe) send_vblank_event(dev, e, seq, now); } - trace_drm_vblank_event(pipe, seq, now, - dev->driver->get_vblank_timestamp != NULL); + if (crtc && crtc->funcs->get_vblank_timestamp) + high_prec = true; + + trace_drm_vblank_event(pipe, seq, now, high_prec); } /** diff --git a/drivers/gpu/drm/drm_vm.c b/drivers/gpu/drm/drm_vm.c index 64619fe90046..aa88911bbc06 100644 --- a/drivers/gpu/drm/drm_vm.c +++ b/drivers/gpu/drm/drm_vm.c @@ -102,7 +102,7 @@ static pgprot_t drm_dma_prot(uint32_t map_type, struct vm_area_struct *vma) return tmp; } -/** +/* * \c fault method for AGP virtual memory. * * \param vma virtual memory area. @@ -192,7 +192,7 @@ static vm_fault_t drm_vm_fault(struct vm_fault *vmf) } #endif -/** +/* * \c nopage method for shared virtual memory. * * \param vma virtual memory area. @@ -225,7 +225,7 @@ static vm_fault_t drm_vm_shm_fault(struct vm_fault *vmf) return 0; } -/** +/* * \c close method for shared virtual memory. * * \param vma virtual memory area. @@ -294,7 +294,7 @@ static void drm_vm_shm_close(struct vm_area_struct *vma) mutex_unlock(&dev->struct_mutex); } -/** +/* * \c fault method for DMA virtual memory. * * \param address access address. @@ -329,7 +329,7 @@ static vm_fault_t drm_vm_dma_fault(struct vm_fault *vmf) return 0; } -/** +/* * \c fault method for scatter-gather virtual memory. * * \param address access address. @@ -435,7 +435,7 @@ static void drm_vm_close_locked(struct drm_device *dev, } } -/** +/* * \c close method for all virtual memory types. * * \param vma virtual memory area. @@ -453,7 +453,7 @@ static void drm_vm_close(struct vm_area_struct *vma) mutex_unlock(&dev->struct_mutex); } -/** +/* * mmap DMA memory. * * \param file_priv DRM file private. @@ -513,7 +513,7 @@ static resource_size_t drm_core_get_reg_ofs(struct drm_device *dev) #endif } -/** +/* * mmap DMA memory. * * \param file_priv DRM file private. diff --git a/drivers/gpu/drm/etnaviv/etnaviv_buffer.c b/drivers/gpu/drm/etnaviv/etnaviv_buffer.c index 32d9fac587f9..76d38561c910 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_buffer.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_buffer.c @@ -12,6 +12,7 @@ #include "common.xml.h" #include "state.xml.h" +#include "state_blt.xml.h" #include "state_hi.xml.h" #include "state_3d.xml.h" #include "cmdstream.xml.h" @@ -233,6 +234,8 @@ void etnaviv_buffer_end(struct etnaviv_gpu *gpu) struct etnaviv_cmdbuf *buffer = &gpu->buffer; unsigned int waitlink_offset = buffer->user_size - 16; u32 link_target, flush = 0; + bool has_blt = !!(gpu->identity.minor_features5 & + chipMinorFeatures5_BLT_ENGINE); lockdep_assert_held(&gpu->lock); @@ -248,16 +251,38 @@ void etnaviv_buffer_end(struct etnaviv_gpu *gpu) if (flush) { unsigned int dwords = 7; + if (has_blt) + dwords += 10; + link_target = etnaviv_buffer_reserve(gpu, buffer, dwords); CMD_SEM(buffer, SYNC_RECIPIENT_FE, SYNC_RECIPIENT_PE); CMD_STALL(buffer, SYNC_RECIPIENT_FE, SYNC_RECIPIENT_PE); + if (has_blt) { + CMD_LOAD_STATE(buffer, VIVS_BLT_ENABLE, 0x1); + CMD_SEM(buffer, SYNC_RECIPIENT_FE, SYNC_RECIPIENT_BLT); + CMD_STALL(buffer, SYNC_RECIPIENT_FE, SYNC_RECIPIENT_BLT); + CMD_LOAD_STATE(buffer, VIVS_BLT_ENABLE, 0x0); + } CMD_LOAD_STATE(buffer, VIVS_GL_FLUSH_CACHE, flush); - if (gpu->exec_state == ETNA_PIPE_3D) - CMD_LOAD_STATE(buffer, VIVS_TS_FLUSH_CACHE, - VIVS_TS_FLUSH_CACHE_FLUSH); + if (gpu->exec_state == ETNA_PIPE_3D) { + if (has_blt) { + CMD_LOAD_STATE(buffer, VIVS_BLT_ENABLE, 0x1); + CMD_LOAD_STATE(buffer, VIVS_BLT_SET_COMMAND, 0x1); + CMD_LOAD_STATE(buffer, VIVS_BLT_ENABLE, 0x0); + } else { + CMD_LOAD_STATE(buffer, VIVS_TS_FLUSH_CACHE, + VIVS_TS_FLUSH_CACHE_FLUSH); + } + } CMD_SEM(buffer, SYNC_RECIPIENT_FE, SYNC_RECIPIENT_PE); CMD_STALL(buffer, SYNC_RECIPIENT_FE, SYNC_RECIPIENT_PE); + if (has_blt) { + CMD_LOAD_STATE(buffer, VIVS_BLT_ENABLE, 0x1); + CMD_SEM(buffer, SYNC_RECIPIENT_FE, SYNC_RECIPIENT_BLT); + CMD_STALL(buffer, SYNC_RECIPIENT_FE, SYNC_RECIPIENT_BLT); + CMD_LOAD_STATE(buffer, VIVS_BLT_ENABLE, 0x0); + } CMD_END(buffer); etnaviv_buffer_replace_wait(buffer, waitlink_offset, @@ -323,6 +348,8 @@ void etnaviv_buffer_queue(struct etnaviv_gpu *gpu, u32 exec_state, bool switch_mmu_context = gpu->mmu_context != mmu_context; unsigned int new_flush_seq = READ_ONCE(gpu->mmu_context->flush_seq); bool need_flush = switch_mmu_context || gpu->flush_seq != new_flush_seq; + bool has_blt = !!(gpu->identity.minor_features5 & + chipMinorFeatures5_BLT_ENGINE); lockdep_assert_held(&gpu->lock); @@ -433,6 +460,15 @@ void etnaviv_buffer_queue(struct etnaviv_gpu *gpu, u32 exec_state, * 2 semaphore stall + 1 event + 1 wait + 1 link. */ return_dwords = 7; + + /* + * When the BLT engine is present we need 6 more dwords in the return + * target: 3 enable/flush/disable + 4 enable/semaphore stall/disable, + * but we don't need the normal TS flush state. + */ + if (has_blt) + return_dwords += 6; + return_target = etnaviv_buffer_reserve(gpu, buffer, return_dwords); CMD_LINK(cmdbuf, return_dwords, return_target); @@ -447,11 +483,25 @@ void etnaviv_buffer_queue(struct etnaviv_gpu *gpu, u32 exec_state, CMD_LOAD_STATE(buffer, VIVS_GL_FLUSH_CACHE, VIVS_GL_FLUSH_CACHE_DEPTH | VIVS_GL_FLUSH_CACHE_COLOR); - CMD_LOAD_STATE(buffer, VIVS_TS_FLUSH_CACHE, - VIVS_TS_FLUSH_CACHE_FLUSH); + if (has_blt) { + CMD_LOAD_STATE(buffer, VIVS_BLT_ENABLE, 0x1); + CMD_LOAD_STATE(buffer, VIVS_BLT_SET_COMMAND, 0x1); + CMD_LOAD_STATE(buffer, VIVS_BLT_ENABLE, 0x0); + } else { + CMD_LOAD_STATE(buffer, VIVS_TS_FLUSH_CACHE, + VIVS_TS_FLUSH_CACHE_FLUSH); + } } CMD_SEM(buffer, SYNC_RECIPIENT_FE, SYNC_RECIPIENT_PE); CMD_STALL(buffer, SYNC_RECIPIENT_FE, SYNC_RECIPIENT_PE); + + if (has_blt) { + CMD_LOAD_STATE(buffer, VIVS_BLT_ENABLE, 0x1); + CMD_SEM(buffer, SYNC_RECIPIENT_FE, SYNC_RECIPIENT_BLT); + CMD_STALL(buffer, SYNC_RECIPIENT_FE, SYNC_RECIPIENT_BLT); + CMD_LOAD_STATE(buffer, VIVS_BLT_ENABLE, 0x0); + } + CMD_LOAD_STATE(buffer, VIVS_GL_EVENT, VIVS_GL_EVENT_EVENT_ID(event) | VIVS_GL_EVENT_FROM_PE); CMD_WAIT(buffer); diff --git a/drivers/gpu/drm/etnaviv/etnaviv_drv.c b/drivers/gpu/drm/etnaviv/etnaviv_drv.c index 6b43c1c94e8f..a8685b2e1803 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_drv.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_drv.c @@ -551,6 +551,7 @@ static int etnaviv_bind(struct device *dev) mutex_init(&priv->gem_lock); INIT_LIST_HEAD(&priv->gem_list); priv->num_gpus = 0; + priv->shm_gfp_mask = GFP_HIGHUSER | __GFP_RETRY_MAYFAIL | __GFP_NOWARN; priv->cmdbuf_suballoc = etnaviv_cmdbuf_suballoc_new(drm->dev); if (IS_ERR(priv->cmdbuf_suballoc)) { diff --git a/drivers/gpu/drm/etnaviv/etnaviv_drv.h b/drivers/gpu/drm/etnaviv/etnaviv_drv.h index efc656efeb0f..4d8dc9236e5f 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_drv.h +++ b/drivers/gpu/drm/etnaviv/etnaviv_drv.h @@ -35,6 +35,7 @@ struct etnaviv_drm_private { int num_gpus; struct device_dma_parameters dma_parms; struct etnaviv_gpu *gpu[ETNA_MAX_PIPES]; + gfp_t shm_gfp_mask; struct etnaviv_cmdbuf_suballoc *cmdbuf_suballoc; struct etnaviv_iommu_global *mmu_global; diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem.c b/drivers/gpu/drm/etnaviv/etnaviv_gem.c index 6adea180d629..dc9ef302f517 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_gem.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_gem.c @@ -602,6 +602,7 @@ static int etnaviv_gem_new_impl(struct drm_device *dev, u32 size, u32 flags, int etnaviv_gem_new_handle(struct drm_device *dev, struct drm_file *file, u32 size, u32 flags, u32 *handle) { + struct etnaviv_drm_private *priv = dev->dev_private; struct drm_gem_object *obj = NULL; int ret; @@ -624,8 +625,7 @@ int etnaviv_gem_new_handle(struct drm_device *dev, struct drm_file *file, * above new_inode() why this is required _and_ expected if you're * going to pin these pages. */ - mapping_set_gfp_mask(obj->filp->f_mapping, GFP_HIGHUSER | - __GFP_RETRY_MAYFAIL | __GFP_NOWARN); + mapping_set_gfp_mask(obj->filp->f_mapping, priv->shm_gfp_mask); etnaviv_gem_obj_add(dev, obj); diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem.h b/drivers/gpu/drm/etnaviv/etnaviv_gem.h index 6b68fe16041b..98e60df882b6 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_gem.h +++ b/drivers/gpu/drm/etnaviv/etnaviv_gem.h @@ -105,7 +105,7 @@ struct etnaviv_gem_submit { unsigned int nr_pmrs; struct etnaviv_perfmon_request *pmrs; unsigned int nr_bos; - struct etnaviv_gem_submit_bo bos[0]; + struct etnaviv_gem_submit_bo bos[]; /* No new members here, the previous one is variable-length! */ }; diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gpu.c b/drivers/gpu/drm/etnaviv/etnaviv_gpu.c index 799ec20b267d..a31eeff2b297 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_gpu.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_gpu.c @@ -333,9 +333,13 @@ static void etnaviv_hw_identify(struct etnaviv_gpu *gpu) gpu->identity.revision = etnaviv_field(chipIdentity, VIVS_HI_CHIP_IDENTITY_REVISION); } else { + u32 chipDate = gpu_read(gpu, VIVS_HI_CHIP_DATE); gpu->identity.model = gpu_read(gpu, VIVS_HI_CHIP_MODEL); gpu->identity.revision = gpu_read(gpu, VIVS_HI_CHIP_REV); + gpu->identity.product_id = gpu_read(gpu, VIVS_HI_CHIP_PRODUCT_ID); + gpu->identity.customer_id = gpu_read(gpu, VIVS_HI_CHIP_CUSTOMER_ID); + gpu->identity.eco_id = gpu_read(gpu, VIVS_HI_CHIP_ECO_ID); /* * !!!! HACK ALERT !!!! @@ -350,7 +354,6 @@ static void etnaviv_hw_identify(struct etnaviv_gpu *gpu) /* Another special case */ if (etnaviv_is_model_rev(gpu, GC300, 0x2201)) { - u32 chipDate = gpu_read(gpu, VIVS_HI_CHIP_DATE); u32 chipTime = gpu_read(gpu, VIVS_HI_CHIP_TIME); if (chipDate == 0x20080814 && chipTime == 0x12051100) { @@ -373,6 +376,12 @@ static void etnaviv_hw_identify(struct etnaviv_gpu *gpu) gpu->identity.model = chipModel_GC3000; gpu->identity.revision &= 0xffff; } + + if (etnaviv_is_model_rev(gpu, GC1000, 0x5037) && (chipDate == 0x20120617)) + gpu->identity.eco_id = 1; + + if (etnaviv_is_model_rev(gpu, GC320, 0x5303) && (chipDate == 0x20140511)) + gpu->identity.eco_id = 1; } dev_info(gpu->dev, "model: GC%x, revision: %x\n", @@ -506,7 +515,7 @@ static int etnaviv_hw_reset(struct etnaviv_gpu *gpu) /* read idle register. */ idle = gpu_read(gpu, VIVS_HI_IDLE_STATE); - /* try reseting again if FE it not idle */ + /* try resetting again if FE is not idle */ if ((idle & VIVS_HI_IDLE_STATE_FE) == 0) { dev_dbg(gpu->dev, "FE is not idle\n"); continue; @@ -772,6 +781,14 @@ int etnaviv_gpu_init(struct etnaviv_gpu *gpu) gpu->identity.features &= ~chipFeatures_FAST_CLEAR; } + /* + * If the GPU is part of a system with DMA addressing limitations, + * request pages for our SHM backend buffers from the DMA32 zone to + * hopefully avoid performance killing SWIOTLB bounce buffering. + */ + if (dma_addressing_limited(gpu->dev)) + priv->shm_gfp_mask |= GFP_DMA32; + /* Create buffer: */ ret = etnaviv_cmdbuf_init(priv->cmdbuf_suballoc, &gpu->buffer, PAGE_SIZE); @@ -851,6 +868,13 @@ int etnaviv_gpu_debugfs(struct etnaviv_gpu *gpu, struct seq_file *m) verify_dma(gpu, &debug); + seq_puts(m, "\tidentity\n"); + seq_printf(m, "\t model: 0x%x\n", gpu->identity.model); + seq_printf(m, "\t revision: 0x%x\n", gpu->identity.revision); + seq_printf(m, "\t product_id: 0x%x\n", gpu->identity.product_id); + seq_printf(m, "\t customer_id: 0x%x\n", gpu->identity.customer_id); + seq_printf(m, "\t eco_id: 0x%x\n", gpu->identity.eco_id); + seq_puts(m, "\tfeatures\n"); seq_printf(m, "\t major_features: 0x%08x\n", gpu->identity.features); @@ -930,6 +954,20 @@ int etnaviv_gpu_debugfs(struct etnaviv_gpu *gpu, struct seq_file *m) seq_puts(m, "\t FP is not idle\n"); if ((idle & VIVS_HI_IDLE_STATE_TS) == 0) seq_puts(m, "\t TS is not idle\n"); + if ((idle & VIVS_HI_IDLE_STATE_BL) == 0) + seq_puts(m, "\t BL is not idle\n"); + if ((idle & VIVS_HI_IDLE_STATE_ASYNCFE) == 0) + seq_puts(m, "\t ASYNCFE is not idle\n"); + if ((idle & VIVS_HI_IDLE_STATE_MC) == 0) + seq_puts(m, "\t MC is not idle\n"); + if ((idle & VIVS_HI_IDLE_STATE_PPA) == 0) + seq_puts(m, "\t PPA is not idle\n"); + if ((idle & VIVS_HI_IDLE_STATE_WD) == 0) + seq_puts(m, "\t WD is not idle\n"); + if ((idle & VIVS_HI_IDLE_STATE_NN) == 0) + seq_puts(m, "\t NN is not idle\n"); + if ((idle & VIVS_HI_IDLE_STATE_TP) == 0) + seq_puts(m, "\t TP is not idle\n"); if (idle & VIVS_HI_IDLE_STATE_AXI_LP) seq_puts(m, "\t AXI low power mode\n"); @@ -1805,11 +1843,15 @@ static int etnaviv_gpu_rpm_suspend(struct device *dev) if (atomic_read(&gpu->sched.hw_rq_count)) return -EBUSY; - /* Check whether the hardware (except FE) is idle */ - mask = gpu->idle_mask & ~VIVS_HI_IDLE_STATE_FE; + /* Check whether the hardware (except FE and MC) is idle */ + mask = gpu->idle_mask & ~(VIVS_HI_IDLE_STATE_FE | + VIVS_HI_IDLE_STATE_MC); idle = gpu_read(gpu, VIVS_HI_IDLE_STATE) & mask; - if (idle != mask) + if (idle != mask) { + dev_warn_ratelimited(dev, "GPU not yet idle, mask: 0x%08x\n", + idle); return -EBUSY; + } return etnaviv_gpu_hw_suspend(gpu); } diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gpu.h b/drivers/gpu/drm/etnaviv/etnaviv_gpu.h index 97bb48042b4d..8ea48697d132 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_gpu.h +++ b/drivers/gpu/drm/etnaviv/etnaviv_gpu.h @@ -15,11 +15,11 @@ struct etnaviv_gem_submit; struct etnaviv_vram_mapping; struct etnaviv_chip_identity { - /* Chip model. */ u32 model; - - /* Revision value.*/ u32 revision; + u32 product_id; + u32 customer_id; + u32 eco_id; /* Supported feature fields. */ u32 features; diff --git a/drivers/gpu/drm/etnaviv/etnaviv_hwdb.c b/drivers/gpu/drm/etnaviv/etnaviv_hwdb.c index 39b463db76c9..167971a09be7 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_hwdb.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_hwdb.c @@ -7,8 +7,42 @@ static const struct etnaviv_chip_identity etnaviv_chip_identities[] = { { + .model = 0x400, + .revision = 0x4652, + .product_id = 0x70001, + .customer_id = 0x100, + .eco_id = 0, + .stream_count = 4, + .register_max = 64, + .thread_count = 128, + .shader_core_count = 1, + .vertex_cache_size = 8, + .vertex_output_buffer_size = 1024, + .pixel_pipes = 1, + .instruction_count = 256, + .num_constants = 320, + .buffer_size = 0, + .varyings_count = 8, + .features = 0xa0e9e004, + .minor_features0 = 0xe1299fff, + .minor_features1 = 0xbe13b219, + .minor_features2 = 0xce110010, + .minor_features3 = 0x8000001, + .minor_features4 = 0x20102, + .minor_features5 = 0x120000, + .minor_features6 = 0x0, + .minor_features7 = 0x0, + .minor_features8 = 0x0, + .minor_features9 = 0x0, + .minor_features10 = 0x0, + .minor_features11 = 0x0, + }, + { .model = 0x7000, .revision = 0x6214, + .product_id = ~0U, + .customer_id = ~0U, + .eco_id = ~0U, .stream_count = 16, .register_max = 64, .thread_count = 1024, @@ -43,7 +77,13 @@ bool etnaviv_fill_identity_from_hwdb(struct etnaviv_gpu *gpu) for (i = 0; i < ARRAY_SIZE(etnaviv_chip_identities); i++) { if (etnaviv_chip_identities[i].model == ident->model && - etnaviv_chip_identities[i].revision == ident->revision) { + etnaviv_chip_identities[i].revision == ident->revision && + (etnaviv_chip_identities[i].product_id == ident->product_id || + etnaviv_chip_identities[i].product_id == ~0U) && + (etnaviv_chip_identities[i].customer_id == ident->customer_id || + etnaviv_chip_identities[i].customer_id == ~0U) && + (etnaviv_chip_identities[i].eco_id == ident->eco_id || + etnaviv_chip_identities[i].eco_id == ~0U)) { memcpy(ident, &etnaviv_chip_identities[i], sizeof(*ident)); return true; diff --git a/drivers/gpu/drm/etnaviv/etnaviv_perfmon.c b/drivers/gpu/drm/etnaviv/etnaviv_perfmon.c index 8adbf2861bff..e6795bafcbb9 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_perfmon.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_perfmon.c @@ -32,6 +32,7 @@ struct etnaviv_pm_domain { }; struct etnaviv_pm_domain_meta { + unsigned int feature; const struct etnaviv_pm_domain *domains; u32 nr_domains; }; @@ -410,36 +411,78 @@ static const struct etnaviv_pm_domain doms_vg[] = { static const struct etnaviv_pm_domain_meta doms_meta[] = { { + .feature = chipFeatures_PIPE_3D, .nr_domains = ARRAY_SIZE(doms_3d), .domains = &doms_3d[0] }, { + .feature = chipFeatures_PIPE_2D, .nr_domains = ARRAY_SIZE(doms_2d), .domains = &doms_2d[0] }, { + .feature = chipFeatures_PIPE_VG, .nr_domains = ARRAY_SIZE(doms_vg), .domains = &doms_vg[0] } }; +static unsigned int num_pm_domains(const struct etnaviv_gpu *gpu) +{ + unsigned int num = 0, i; + + for (i = 0; i < ARRAY_SIZE(doms_meta); i++) { + const struct etnaviv_pm_domain_meta *meta = &doms_meta[i]; + + if (gpu->identity.features & meta->feature) + num += meta->nr_domains; + } + + return num; +} + +static const struct etnaviv_pm_domain *pm_domain(const struct etnaviv_gpu *gpu, + unsigned int index) +{ + const struct etnaviv_pm_domain *domain = NULL; + unsigned int offset = 0, i; + + for (i = 0; i < ARRAY_SIZE(doms_meta); i++) { + const struct etnaviv_pm_domain_meta *meta = &doms_meta[i]; + + if (!(gpu->identity.features & meta->feature)) + continue; + + if (meta->nr_domains < (index - offset)) { + offset += meta->nr_domains; + continue; + } + + domain = meta->domains + (index - offset); + } + + return domain; +} + int etnaviv_pm_query_dom(struct etnaviv_gpu *gpu, struct drm_etnaviv_pm_domain *domain) { - const struct etnaviv_pm_domain_meta *meta = &doms_meta[domain->pipe]; + const unsigned int nr_domains = num_pm_domains(gpu); const struct etnaviv_pm_domain *dom; - if (domain->iter >= meta->nr_domains) + if (domain->iter >= nr_domains) return -EINVAL; - dom = meta->domains + domain->iter; + dom = pm_domain(gpu, domain->iter); + if (!dom) + return -EINVAL; domain->id = domain->iter; domain->nr_signals = dom->nr_signals; strncpy(domain->name, dom->name, sizeof(domain->name)); domain->iter++; - if (domain->iter == meta->nr_domains) + if (domain->iter == nr_domains) domain->iter = 0xff; return 0; @@ -448,14 +491,16 @@ int etnaviv_pm_query_dom(struct etnaviv_gpu *gpu, int etnaviv_pm_query_sig(struct etnaviv_gpu *gpu, struct drm_etnaviv_pm_signal *signal) { - const struct etnaviv_pm_domain_meta *meta = &doms_meta[signal->pipe]; + const unsigned int nr_domains = num_pm_domains(gpu); const struct etnaviv_pm_domain *dom; const struct etnaviv_pm_signal *sig; - if (signal->domain >= meta->nr_domains) + if (signal->domain >= nr_domains) return -EINVAL; - dom = meta->domains + signal->domain; + dom = pm_domain(gpu, signal->domain); + if (!dom) + return -EINVAL; if (signal->iter >= dom->nr_signals) return -EINVAL; diff --git a/drivers/gpu/drm/etnaviv/state_blt.xml.h b/drivers/gpu/drm/etnaviv/state_blt.xml.h index daae55995def..0e8bcf9dcc93 100644 --- a/drivers/gpu/drm/etnaviv/state_blt.xml.h +++ b/drivers/gpu/drm/etnaviv/state_blt.xml.h @@ -46,6 +46,8 @@ DEALINGS IN THE SOFTWARE. /* This is a cut-down version of the state_blt.xml.h file */ +#define VIVS_BLT_SET_COMMAND 0x000140ac + #define VIVS_BLT_ENABLE 0x000140b8 #define VIVS_BLT_ENABLE_ENABLE 0x00000001 diff --git a/drivers/gpu/drm/etnaviv/state_hi.xml.h b/drivers/gpu/drm/etnaviv/state_hi.xml.h index 41d8da2b6f4f..deaaa99fa654 100644 --- a/drivers/gpu/drm/etnaviv/state_hi.xml.h +++ b/drivers/gpu/drm/etnaviv/state_hi.xml.h @@ -8,17 +8,17 @@ http://0x04.net/cgit/index.cgi/rules-ng-ng git clone git://0x04.net/rules-ng-ng The rules-ng-ng source files this header was generated from are: -- state.xml ( 26087 bytes, from 2017-12-18 16:51:59) -- common.xml ( 35468 bytes, from 2018-01-22 13:48:54) -- common_3d.xml ( 14615 bytes, from 2017-12-18 16:51:59) -- state_hi.xml ( 30232 bytes, from 2018-02-15 15:48:01) -- copyright.xml ( 1597 bytes, from 2016-12-08 16:37:56) -- state_2d.xml ( 51552 bytes, from 2016-12-08 16:37:56) -- state_3d.xml ( 79992 bytes, from 2017-12-18 16:51:59) -- state_blt.xml ( 13405 bytes, from 2017-12-18 16:51:59) -- state_vg.xml ( 5975 bytes, from 2016-12-08 16:37:56) - -Copyright (C) 2012-2018 by the following authors: +- state.xml ( 26666 bytes, from 2019-12-20 21:20:35) +- common.xml ( 35468 bytes, from 2018-02-10 13:09:26) +- common_3d.xml ( 15058 bytes, from 2019-12-28 20:02:03) +- state_hi.xml ( 30552 bytes, from 2019-12-28 20:02:48) +- copyright.xml ( 1597 bytes, from 2018-02-10 13:09:26) +- state_2d.xml ( 51552 bytes, from 2018-02-10 13:09:26) +- state_3d.xml ( 83098 bytes, from 2019-12-28 20:02:03) +- state_blt.xml ( 14252 bytes, from 2019-10-20 19:59:15) +- state_vg.xml ( 5975 bytes, from 2018-02-10 13:09:26) + +Copyright (C) 2012-2019 by the following authors: - Wladimir J. van der Laan <laanwj@gmail.com> - Christian Gmeiner <christian.gmeiner@gmail.com> - Lucas Stach <l.stach@pengutronix.de> @@ -48,6 +48,9 @@ DEALINGS IN THE SOFTWARE. #define MMU_EXCEPTION_SLAVE_NOT_PRESENT 0x00000001 #define MMU_EXCEPTION_PAGE_NOT_PRESENT 0x00000002 #define MMU_EXCEPTION_WRITE_VIOLATION 0x00000003 +#define MMU_EXCEPTION_OUT_OF_BOUND 0x00000004 +#define MMU_EXCEPTION_READ_SECURITY_VIOLATION 0x00000005 +#define MMU_EXCEPTION_WRITE_SECURITY_VIOLATION 0x00000006 #define VIVS_HI 0x00000000 #define VIVS_HI_CLOCK_CONTROL 0x00000000 @@ -81,6 +84,13 @@ DEALINGS IN THE SOFTWARE. #define VIVS_HI_IDLE_STATE_IM 0x00000200 #define VIVS_HI_IDLE_STATE_FP 0x00000400 #define VIVS_HI_IDLE_STATE_TS 0x00000800 +#define VIVS_HI_IDLE_STATE_BL 0x00001000 +#define VIVS_HI_IDLE_STATE_ASYNCFE 0x00002000 +#define VIVS_HI_IDLE_STATE_MC 0x00004000 +#define VIVS_HI_IDLE_STATE_PPA 0x00008000 +#define VIVS_HI_IDLE_STATE_WD 0x00010000 +#define VIVS_HI_IDLE_STATE_NN 0x00020000 +#define VIVS_HI_IDLE_STATE_TP 0x00040000 #define VIVS_HI_IDLE_STATE_AXI_LP 0x80000000 #define VIVS_HI_AXI_CONFIG 0x00000008 @@ -140,6 +150,8 @@ DEALINGS IN THE SOFTWARE. #define VIVS_HI_CHIP_TIME 0x0000002c +#define VIVS_HI_CHIP_CUSTOMER_ID 0x00000030 + #define VIVS_HI_CHIP_MINOR_FEATURE_0 0x00000034 #define VIVS_HI_CACHE_CONTROL 0x00000038 @@ -237,6 +249,8 @@ DEALINGS IN THE SOFTWARE. #define VIVS_HI_BLT_INTR 0x000000d4 +#define VIVS_HI_CHIP_ECO_ID 0x000000e8 + #define VIVS_HI_AUXBIT 0x000000ec #define VIVS_PM 0x00000000 diff --git a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c index 8428ae12dfa5..1f79bc2a881e 100644 --- a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c +++ b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c @@ -55,6 +55,7 @@ static const char * const decon_clks_name[] = { struct decon_context { struct device *dev; struct drm_device *drm_dev; + void *dma_priv; struct exynos_drm_crtc *crtc; struct exynos_drm_plane planes[WINDOWS_NR]; struct exynos_drm_plane_config configs[WINDOWS_NR]; @@ -644,7 +645,7 @@ static int decon_bind(struct device *dev, struct device *master, void *data) decon_clear_channels(ctx->crtc); - return exynos_drm_register_dma(drm_dev, dev); + return exynos_drm_register_dma(drm_dev, dev, &ctx->dma_priv); } static void decon_unbind(struct device *dev, struct device *master, void *data) @@ -654,7 +655,7 @@ static void decon_unbind(struct device *dev, struct device *master, void *data) decon_atomic_disable(ctx->crtc); /* detach this sub driver from iommu mapping if supported. */ - exynos_drm_unregister_dma(ctx->drm_dev, ctx->dev); + exynos_drm_unregister_dma(ctx->drm_dev, ctx->dev, &ctx->dma_priv); } static const struct component_ops decon_component_ops = { diff --git a/drivers/gpu/drm/exynos/exynos7_drm_decon.c b/drivers/gpu/drm/exynos/exynos7_drm_decon.c index ff59c641fa80..f2d87a7445c7 100644 --- a/drivers/gpu/drm/exynos/exynos7_drm_decon.c +++ b/drivers/gpu/drm/exynos/exynos7_drm_decon.c @@ -40,6 +40,7 @@ struct decon_context { struct device *dev; struct drm_device *drm_dev; + void *dma_priv; struct exynos_drm_crtc *crtc; struct exynos_drm_plane planes[WINDOWS_NR]; struct exynos_drm_plane_config configs[WINDOWS_NR]; @@ -127,19 +128,19 @@ static int decon_ctx_initialize(struct decon_context *ctx, decon_clear_channels(ctx->crtc); - return exynos_drm_register_dma(drm_dev, ctx->dev); + return exynos_drm_register_dma(drm_dev, ctx->dev, &ctx->dma_priv); } static void decon_ctx_remove(struct decon_context *ctx) { /* detach this sub driver from iommu mapping if supported. */ - exynos_drm_unregister_dma(ctx->drm_dev, ctx->dev); + exynos_drm_unregister_dma(ctx->drm_dev, ctx->dev, &ctx->dma_priv); } static u32 decon_calc_clkdiv(struct decon_context *ctx, const struct drm_display_mode *mode) { - unsigned long ideal_clk = mode->htotal * mode->vtotal * mode->vrefresh; + unsigned long ideal_clk = mode->clock; u32 clkdiv; /* Find the clock divider value that gets us closest to ideal_clk */ diff --git a/drivers/gpu/drm/exynos/exynos_dp.c b/drivers/gpu/drm/exynos/exynos_dp.c index 4785885c0f4f..5ee090691390 100644 --- a/drivers/gpu/drm/exynos/exynos_dp.c +++ b/drivers/gpu/drm/exynos/exynos_dp.c @@ -106,7 +106,8 @@ static int exynos_dp_bridge_attach(struct analogix_dp_plat_data *plat_data, /* Pre-empt DP connector creation if there's a bridge */ if (dp->ptn_bridge) { - ret = drm_bridge_attach(&dp->encoder, dp->ptn_bridge, bridge); + ret = drm_bridge_attach(&dp->encoder, dp->ptn_bridge, bridge, + 0); if (ret) { DRM_DEV_ERROR(dp->dev, "Failed to attach bridge to drm\n"); @@ -158,15 +159,8 @@ static int exynos_dp_bind(struct device *dev, struct device *master, void *data) struct drm_device *drm_dev = data; int ret; - dp->dev = dev; dp->drm_dev = drm_dev; - dp->plat_data.dev_type = EXYNOS_DP; - dp->plat_data.power_on_start = exynos_dp_poweron; - dp->plat_data.power_off = exynos_dp_poweroff; - dp->plat_data.attach = exynos_dp_bridge_attach; - dp->plat_data.get_modes = exynos_dp_get_modes; - if (!dp->plat_data.panel && !dp->ptn_bridge) { ret = exynos_dp_dt_parse_panel(dp); if (ret) @@ -184,13 +178,11 @@ static int exynos_dp_bind(struct device *dev, struct device *master, void *data) dp->plat_data.encoder = encoder; - dp->adp = analogix_dp_bind(dev, dp->drm_dev, &dp->plat_data); - if (IS_ERR(dp->adp)) { + ret = analogix_dp_bind(dp->adp, dp->drm_dev); + if (ret) dp->encoder.funcs->destroy(&dp->encoder); - return PTR_ERR(dp->adp); - } - return 0; + return ret; } static void exynos_dp_unbind(struct device *dev, struct device *master, @@ -221,6 +213,7 @@ static int exynos_dp_probe(struct platform_device *pdev) if (!dp) return -ENOMEM; + dp->dev = dev; /* * We just use the drvdata until driver run into component * add function, and then we would set drvdata to null, so @@ -246,16 +239,29 @@ static int exynos_dp_probe(struct platform_device *pdev) /* The remote port can be either a panel or a bridge */ dp->plat_data.panel = panel; + dp->plat_data.dev_type = EXYNOS_DP; + dp->plat_data.power_on_start = exynos_dp_poweron; + dp->plat_data.power_off = exynos_dp_poweroff; + dp->plat_data.attach = exynos_dp_bridge_attach; + dp->plat_data.get_modes = exynos_dp_get_modes; dp->plat_data.skip_connector = !!bridge; + dp->ptn_bridge = bridge; out: + dp->adp = analogix_dp_probe(dev, &dp->plat_data); + if (IS_ERR(dp->adp)) + return PTR_ERR(dp->adp); + return component_add(&pdev->dev, &exynos_dp_ops); } static int exynos_dp_remove(struct platform_device *pdev) { + struct exynos_dp_device *dp = platform_get_drvdata(pdev); + component_del(&pdev->dev, &exynos_dp_ops); + analogix_dp_remove(dp->adp); return 0; } diff --git a/drivers/gpu/drm/exynos/exynos_drm_dma.c b/drivers/gpu/drm/exynos/exynos_drm_dma.c index 9ebc02768847..619f81435c1b 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_dma.c +++ b/drivers/gpu/drm/exynos/exynos_drm_dma.c @@ -58,7 +58,7 @@ static inline void clear_dma_max_seg_size(struct device *dev) * mapping. */ static int drm_iommu_attach_device(struct drm_device *drm_dev, - struct device *subdrv_dev) + struct device *subdrv_dev, void **dma_priv) { struct exynos_drm_private *priv = drm_dev->dev_private; int ret; @@ -74,7 +74,14 @@ static int drm_iommu_attach_device(struct drm_device *drm_dev, return ret; if (IS_ENABLED(CONFIG_ARM_DMA_USE_IOMMU)) { - if (to_dma_iommu_mapping(subdrv_dev)) + /* + * Keep the original DMA mapping of the sub-device and + * restore it on Exynos DRM detach, otherwise the DMA + * framework considers it as IOMMU-less during the next + * probe (in case of deferred probe or modular build) + */ + *dma_priv = to_dma_iommu_mapping(subdrv_dev); + if (*dma_priv) arm_iommu_detach_device(subdrv_dev); ret = arm_iommu_attach_device(subdrv_dev, priv->mapping); @@ -98,19 +105,21 @@ static int drm_iommu_attach_device(struct drm_device *drm_dev, * mapping */ static void drm_iommu_detach_device(struct drm_device *drm_dev, - struct device *subdrv_dev) + struct device *subdrv_dev, void **dma_priv) { struct exynos_drm_private *priv = drm_dev->dev_private; - if (IS_ENABLED(CONFIG_ARM_DMA_USE_IOMMU)) + if (IS_ENABLED(CONFIG_ARM_DMA_USE_IOMMU)) { arm_iommu_detach_device(subdrv_dev); - else if (IS_ENABLED(CONFIG_IOMMU_DMA)) + arm_iommu_attach_device(subdrv_dev, *dma_priv); + } else if (IS_ENABLED(CONFIG_IOMMU_DMA)) iommu_detach_device(priv->mapping, subdrv_dev); clear_dma_max_seg_size(subdrv_dev); } -int exynos_drm_register_dma(struct drm_device *drm, struct device *dev) +int exynos_drm_register_dma(struct drm_device *drm, struct device *dev, + void **dma_priv) { struct exynos_drm_private *priv = drm->dev_private; @@ -137,13 +146,14 @@ int exynos_drm_register_dma(struct drm_device *drm, struct device *dev) priv->mapping = mapping; } - return drm_iommu_attach_device(drm, dev); + return drm_iommu_attach_device(drm, dev, dma_priv); } -void exynos_drm_unregister_dma(struct drm_device *drm, struct device *dev) +void exynos_drm_unregister_dma(struct drm_device *drm, struct device *dev, + void **dma_priv) { if (IS_ENABLED(CONFIG_EXYNOS_IOMMU)) - drm_iommu_detach_device(drm, dev); + drm_iommu_detach_device(drm, dev, dma_priv); } void exynos_drm_cleanup_dma(struct drm_device *drm) diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.c b/drivers/gpu/drm/exynos/exynos_drm_drv.c index ba0f868b2477..57defeb44522 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_drv.c +++ b/drivers/gpu/drm/exynos/exynos_drm_drv.c @@ -270,7 +270,7 @@ static int exynos_drm_bind(struct device *dev) struct drm_encoder *encoder; struct drm_device *drm; unsigned int clone_mask; - int cnt, ret; + int ret; drm = drm_dev_alloc(&exynos_drm_driver, dev); if (IS_ERR(drm)) @@ -293,10 +293,9 @@ static int exynos_drm_bind(struct device *dev) exynos_drm_mode_config_init(drm); /* setup possible_clones. */ - cnt = 0; clone_mask = 0; list_for_each_entry(encoder, &drm->mode_config.encoder_list, head) - clone_mask |= (1 << (cnt++)); + clone_mask |= drm_encoder_mask(encoder); list_for_each_entry(encoder, &drm->mode_config.encoder_list, head) encoder->possible_clones = clone_mask; diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.h b/drivers/gpu/drm/exynos/exynos_drm_drv.h index d4d21d8cfb90..6ae9056e7a18 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_drv.h +++ b/drivers/gpu/drm/exynos/exynos_drm_drv.h @@ -223,8 +223,10 @@ static inline bool is_drm_iommu_supported(struct drm_device *drm_dev) return priv->mapping ? true : false; } -int exynos_drm_register_dma(struct drm_device *drm, struct device *dev); -void exynos_drm_unregister_dma(struct drm_device *drm, struct device *dev); +int exynos_drm_register_dma(struct drm_device *drm, struct device *dev, + void **dma_priv); +void exynos_drm_unregister_dma(struct drm_device *drm, struct device *dev, + void **dma_priv); void exynos_drm_cleanup_dma(struct drm_device *drm); #ifdef CONFIG_DRM_EXYNOS_DPI diff --git a/drivers/gpu/drm/exynos/exynos_drm_dsi.c b/drivers/gpu/drm/exynos/exynos_drm_dsi.c index 33628d85edad..e080aa92338c 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_dsi.c +++ b/drivers/gpu/drm/exynos/exynos_drm_dsi.c @@ -1514,7 +1514,6 @@ static int exynos_dsi_create_connector(struct drm_encoder *encoder) return 0; connector->funcs->reset(connector); - drm_fb_helper_add_one_connector(drm->fb_helper, connector); drm_connector_register(connector); return 0; } @@ -1540,7 +1539,7 @@ static int exynos_dsi_host_attach(struct mipi_dsi_host *host, out_bridge = of_drm_find_bridge(device->dev.of_node); if (out_bridge) { - drm_bridge_attach(encoder, out_bridge, NULL); + drm_bridge_attach(encoder, out_bridge, NULL, 0); dsi->out_bridge = out_bridge; list_splice_init(&encoder->bridge_chain, &dsi->bridge_chain); } else { @@ -1717,7 +1716,7 @@ static int exynos_dsi_bind(struct device *dev, struct device *master, if (dsi->in_bridge_node) { in_bridge = of_drm_find_bridge(dsi->in_bridge_node); if (in_bridge) - drm_bridge_attach(encoder, in_bridge, NULL); + drm_bridge_attach(encoder, in_bridge, NULL, 0); } return mipi_dsi_host_register(&dsi->dsi_host); @@ -1773,8 +1772,9 @@ static int exynos_dsi_probe(struct platform_device *pdev) ret = devm_regulator_bulk_get(dev, ARRAY_SIZE(dsi->supplies), dsi->supplies); if (ret) { - dev_info(dev, "failed to get regulators: %d\n", ret); - return -EPROBE_DEFER; + if (ret != -EPROBE_DEFER) + dev_info(dev, "failed to get regulators: %d\n", ret); + return ret; } dsi->clks = devm_kcalloc(dev, @@ -1787,9 +1787,10 @@ static int exynos_dsi_probe(struct platform_device *pdev) dsi->clks[i] = devm_clk_get(dev, clk_names[i]); if (IS_ERR(dsi->clks[i])) { if (strcmp(clk_names[i], "sclk_mipi") == 0) { - strcpy(clk_names[i], OLD_SCLK_MIPI_CLK_NAME); - i--; - continue; + dsi->clks[i] = devm_clk_get(dev, + OLD_SCLK_MIPI_CLK_NAME); + if (!IS_ERR(dsi->clks[i])) + continue; } dev_info(dev, "failed to get the clock: %s\n", diff --git a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c index 647a1fd1d815..e6ceaf36fb04 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c @@ -200,21 +200,13 @@ int exynos_drm_fbdev_init(struct drm_device *dev) drm_fb_helper_prepare(dev, helper, &exynos_drm_fb_helper_funcs); - ret = drm_fb_helper_init(dev, helper, MAX_CONNECTOR); + ret = drm_fb_helper_init(dev, helper); if (ret < 0) { DRM_DEV_ERROR(dev->dev, "failed to initialize drm fb helper.\n"); goto err_init; } - ret = drm_fb_helper_single_add_all_connectors(helper); - if (ret < 0) { - DRM_DEV_ERROR(dev->dev, - "failed to register drm_fb_helper_connector.\n"); - goto err_setup; - - } - ret = drm_fb_helper_initial_config(helper, PREFERRED_BPP); if (ret < 0) { DRM_DEV_ERROR(dev->dev, diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimc.c b/drivers/gpu/drm/exynos/exynos_drm_fimc.c index 8ea2e1d77802..29ab8be8604c 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fimc.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fimc.c @@ -97,6 +97,7 @@ struct fimc_scaler { struct fimc_context { struct exynos_drm_ipp ipp; struct drm_device *drm_dev; + void *dma_priv; struct device *dev; struct exynos_drm_ipp_task *task; struct exynos_drm_ipp_formats *formats; @@ -1133,7 +1134,7 @@ static int fimc_bind(struct device *dev, struct device *master, void *data) ctx->drm_dev = drm_dev; ipp->drm_dev = drm_dev; - exynos_drm_register_dma(drm_dev, dev); + exynos_drm_register_dma(drm_dev, dev, &ctx->dma_priv); exynos_drm_ipp_register(dev, ipp, &ipp_funcs, DRM_EXYNOS_IPP_CAP_CROP | DRM_EXYNOS_IPP_CAP_ROTATE | @@ -1153,7 +1154,7 @@ static void fimc_unbind(struct device *dev, struct device *master, struct exynos_drm_ipp *ipp = &ctx->ipp; exynos_drm_ipp_unregister(dev, ipp); - exynos_drm_unregister_dma(drm_dev, dev); + exynos_drm_unregister_dma(drm_dev, dev, &ctx->dma_priv); } static const struct component_ops fimc_component_ops = { diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimd.c b/drivers/gpu/drm/exynos/exynos_drm_fimd.c index 21aec38702fc..bb67cad8371f 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fimd.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fimd.c @@ -167,6 +167,7 @@ static struct fimd_driver_data exynos5420_fimd_driver_data = { struct fimd_context { struct device *dev; struct drm_device *drm_dev; + void *dma_priv; struct exynos_drm_crtc *crtc; struct exynos_drm_plane planes[WINDOWS_NR]; struct exynos_drm_plane_config configs[WINDOWS_NR]; @@ -1090,7 +1091,7 @@ static int fimd_bind(struct device *dev, struct device *master, void *data) if (is_drm_iommu_supported(drm_dev)) fimd_clear_channels(ctx->crtc); - return exynos_drm_register_dma(drm_dev, dev); + return exynos_drm_register_dma(drm_dev, dev, &ctx->dma_priv); } static void fimd_unbind(struct device *dev, struct device *master, @@ -1100,7 +1101,7 @@ static void fimd_unbind(struct device *dev, struct device *master, fimd_atomic_disable(ctx->crtc); - exynos_drm_unregister_dma(ctx->drm_dev, ctx->dev); + exynos_drm_unregister_dma(ctx->drm_dev, ctx->dev, &ctx->dma_priv); if (ctx->encoder) exynos_dpi_remove(ctx->encoder); diff --git a/drivers/gpu/drm/exynos/exynos_drm_g2d.c b/drivers/gpu/drm/exynos/exynos_drm_g2d.c index 2a3382d43bc9..fcee33a43aca 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_g2d.c +++ b/drivers/gpu/drm/exynos/exynos_drm_g2d.c @@ -232,6 +232,7 @@ struct g2d_runqueue_node { struct g2d_data { struct device *dev; + void *dma_priv; struct clk *gate_clk; void __iomem *regs; int irq; @@ -1409,7 +1410,7 @@ static int g2d_bind(struct device *dev, struct device *master, void *data) return ret; } - ret = exynos_drm_register_dma(drm_dev, dev); + ret = exynos_drm_register_dma(drm_dev, dev, &g2d->dma_priv); if (ret < 0) { dev_err(dev, "failed to enable iommu.\n"); g2d_fini_cmdlist(g2d); @@ -1434,7 +1435,7 @@ static void g2d_unbind(struct device *dev, struct device *master, void *data) priv->g2d_dev = NULL; cancel_work_sync(&g2d->runqueue_work); - exynos_drm_unregister_dma(g2d->drm_dev, dev); + exynos_drm_unregister_dma(g2d->drm_dev, dev, &g2d->dma_priv); } static const struct component_ops g2d_component_ops = { diff --git a/drivers/gpu/drm/exynos/exynos_drm_gsc.c b/drivers/gpu/drm/exynos/exynos_drm_gsc.c index 88b6fcaa20be..45e9aee8366a 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gsc.c +++ b/drivers/gpu/drm/exynos/exynos_drm_gsc.c @@ -97,6 +97,7 @@ struct gsc_scaler { struct gsc_context { struct exynos_drm_ipp ipp; struct drm_device *drm_dev; + void *dma_priv; struct device *dev; struct exynos_drm_ipp_task *task; struct exynos_drm_ipp_formats *formats; @@ -1169,7 +1170,7 @@ static int gsc_bind(struct device *dev, struct device *master, void *data) ctx->drm_dev = drm_dev; ctx->drm_dev = drm_dev; - exynos_drm_register_dma(drm_dev, dev); + exynos_drm_register_dma(drm_dev, dev, &ctx->dma_priv); exynos_drm_ipp_register(dev, ipp, &ipp_funcs, DRM_EXYNOS_IPP_CAP_CROP | DRM_EXYNOS_IPP_CAP_ROTATE | @@ -1189,7 +1190,7 @@ static void gsc_unbind(struct device *dev, struct device *master, struct exynos_drm_ipp *ipp = &ctx->ipp; exynos_drm_ipp_unregister(dev, ipp); - exynos_drm_unregister_dma(drm_dev, dev); + exynos_drm_unregister_dma(drm_dev, dev, &ctx->dma_priv); } static const struct component_ops gsc_component_ops = { diff --git a/drivers/gpu/drm/exynos/exynos_drm_rotator.c b/drivers/gpu/drm/exynos/exynos_drm_rotator.c index b98482990d1a..dafa87b82052 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_rotator.c +++ b/drivers/gpu/drm/exynos/exynos_drm_rotator.c @@ -56,6 +56,7 @@ struct rot_variant { struct rot_context { struct exynos_drm_ipp ipp; struct drm_device *drm_dev; + void *dma_priv; struct device *dev; void __iomem *regs; struct clk *clock; @@ -243,7 +244,7 @@ static int rotator_bind(struct device *dev, struct device *master, void *data) rot->drm_dev = drm_dev; ipp->drm_dev = drm_dev; - exynos_drm_register_dma(drm_dev, dev); + exynos_drm_register_dma(drm_dev, dev, &rot->dma_priv); exynos_drm_ipp_register(dev, ipp, &ipp_funcs, DRM_EXYNOS_IPP_CAP_CROP | DRM_EXYNOS_IPP_CAP_ROTATE, @@ -261,7 +262,7 @@ static void rotator_unbind(struct device *dev, struct device *master, struct exynos_drm_ipp *ipp = &rot->ipp; exynos_drm_ipp_unregister(dev, ipp); - exynos_drm_unregister_dma(rot->drm_dev, rot->dev); + exynos_drm_unregister_dma(rot->drm_dev, rot->dev, &rot->dma_priv); } static const struct component_ops rotator_component_ops = { diff --git a/drivers/gpu/drm/exynos/exynos_drm_scaler.c b/drivers/gpu/drm/exynos/exynos_drm_scaler.c index 497973e9b2c5..93c43c8d914e 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_scaler.c +++ b/drivers/gpu/drm/exynos/exynos_drm_scaler.c @@ -39,6 +39,7 @@ struct scaler_data { struct scaler_context { struct exynos_drm_ipp ipp; struct drm_device *drm_dev; + void *dma_priv; struct device *dev; void __iomem *regs; struct clk *clock[SCALER_MAX_CLK]; @@ -450,7 +451,7 @@ static int scaler_bind(struct device *dev, struct device *master, void *data) scaler->drm_dev = drm_dev; ipp->drm_dev = drm_dev; - exynos_drm_register_dma(drm_dev, dev); + exynos_drm_register_dma(drm_dev, dev, &scaler->dma_priv); exynos_drm_ipp_register(dev, ipp, &ipp_funcs, DRM_EXYNOS_IPP_CAP_CROP | DRM_EXYNOS_IPP_CAP_ROTATE | @@ -470,7 +471,8 @@ static void scaler_unbind(struct device *dev, struct device *master, struct exynos_drm_ipp *ipp = &scaler->ipp; exynos_drm_ipp_unregister(dev, ipp); - exynos_drm_unregister_dma(scaler->drm_dev, scaler->dev); + exynos_drm_unregister_dma(scaler->drm_dev, scaler->dev, + &scaler->dma_priv); } static const struct component_ops scaler_component_ops = { diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c index 9ff921f43a93..1a7c828fc41d 100644 --- a/drivers/gpu/drm/exynos/exynos_hdmi.c +++ b/drivers/gpu/drm/exynos/exynos_hdmi.c @@ -960,7 +960,7 @@ static int hdmi_create_connector(struct drm_encoder *encoder) drm_connector_attach_encoder(connector, encoder); if (hdata->bridge) { - ret = drm_bridge_attach(encoder, hdata->bridge, NULL); + ret = drm_bridge_attach(encoder, hdata->bridge, NULL, 0); if (ret) DRM_DEV_ERROR(hdata->dev, "Failed to attach bridge\n"); } @@ -1805,18 +1805,10 @@ static int hdmi_resources_init(struct hdmi_context *hdata) hdata->reg_hdmi_en = devm_regulator_get_optional(dev, "hdmi-en"); - if (PTR_ERR(hdata->reg_hdmi_en) != -ENODEV) { + if (PTR_ERR(hdata->reg_hdmi_en) != -ENODEV) if (IS_ERR(hdata->reg_hdmi_en)) return PTR_ERR(hdata->reg_hdmi_en); - ret = regulator_enable(hdata->reg_hdmi_en); - if (ret) { - DRM_DEV_ERROR(dev, - "failed to enable hdmi-en regulator\n"); - return ret; - } - } - return hdmi_bridge_init(hdata); } @@ -2023,6 +2015,15 @@ static int hdmi_probe(struct platform_device *pdev) } } + if (!IS_ERR(hdata->reg_hdmi_en)) { + ret = regulator_enable(hdata->reg_hdmi_en); + if (ret) { + DRM_DEV_ERROR(dev, + "failed to enable hdmi-en regulator\n"); + goto err_hdmiphy; + } + } + pm_runtime_enable(dev); audio_infoframe = &hdata->audio.infoframe; @@ -2047,7 +2048,8 @@ err_unregister_audio: err_rpm_disable: pm_runtime_disable(dev); - + if (!IS_ERR(hdata->reg_hdmi_en)) + regulator_disable(hdata->reg_hdmi_en); err_hdmiphy: if (hdata->hdmiphy_port) put_device(&hdata->hdmiphy_port->dev); diff --git a/drivers/gpu/drm/exynos/exynos_mixer.c b/drivers/gpu/drm/exynos/exynos_mixer.c index 38ae9c32feef..21b726baedea 100644 --- a/drivers/gpu/drm/exynos/exynos_mixer.c +++ b/drivers/gpu/drm/exynos/exynos_mixer.c @@ -94,6 +94,7 @@ struct mixer_context { struct platform_device *pdev; struct device *dev; struct drm_device *drm_dev; + void *dma_priv; struct exynos_drm_crtc *crtc; struct exynos_drm_plane planes[MIXER_WIN_NR]; unsigned long flags; @@ -894,12 +895,14 @@ static int mixer_initialize(struct mixer_context *mixer_ctx, } } - return exynos_drm_register_dma(drm_dev, mixer_ctx->dev); + return exynos_drm_register_dma(drm_dev, mixer_ctx->dev, + &mixer_ctx->dma_priv); } static void mixer_ctx_remove(struct mixer_context *mixer_ctx) { - exynos_drm_unregister_dma(mixer_ctx->drm_dev, mixer_ctx->dev); + exynos_drm_unregister_dma(mixer_ctx->drm_dev, mixer_ctx->dev, + &mixer_ctx->dma_priv); } static int mixer_enable_vblank(struct exynos_drm_crtc *crtc) diff --git a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_rgb.c b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_rgb.c index 9598ee3cc4d2..cff344367f81 100644 --- a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_rgb.c +++ b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_rgb.c @@ -151,5 +151,5 @@ int fsl_dcu_create_outputs(struct fsl_dcu_drm_device *fsl_dev) return fsl_dcu_attach_panel(fsl_dev, panel); } - return drm_bridge_attach(&fsl_dev->encoder, bridge, NULL); + return drm_bridge_attach(&fsl_dev->encoder, bridge, NULL, 0); } diff --git a/drivers/gpu/drm/gma500/cdv_intel_display.c b/drivers/gpu/drm/gma500/cdv_intel_display.c index 1ed854f498b7..686385a66167 100644 --- a/drivers/gpu/drm/gma500/cdv_intel_display.c +++ b/drivers/gpu/drm/gma500/cdv_intel_display.c @@ -977,6 +977,9 @@ const struct drm_crtc_funcs cdv_intel_crtc_funcs = { .set_config = gma_crtc_set_config, .destroy = gma_crtc_destroy, .page_flip = gma_crtc_page_flip, + .enable_vblank = psb_enable_vblank, + .disable_vblank = psb_disable_vblank, + .get_vblank_counter = psb_get_vblank_counter, }; const struct gma_clock_funcs cdv_clock_funcs = { diff --git a/drivers/gpu/drm/gma500/framebuffer.c b/drivers/gpu/drm/gma500/framebuffer.c index 1459076d1980..1d8f67e4795a 100644 --- a/drivers/gpu/drm/gma500/framebuffer.c +++ b/drivers/gpu/drm/gma500/framebuffer.c @@ -513,14 +513,10 @@ int psb_fbdev_init(struct drm_device *dev) drm_fb_helper_prepare(dev, fb_helper, &psb_fb_helper_funcs); - ret = drm_fb_helper_init(dev, fb_helper, INTELFB_CONN_LIMIT); + ret = drm_fb_helper_init(dev, fb_helper); if (ret) goto free; - ret = drm_fb_helper_single_add_all_connectors(fb_helper); - if (ret) - goto fini; - /* disable all the possible outputs/crtcs before entering KMS mode */ drm_helper_disable_unused_functions(dev); diff --git a/drivers/gpu/drm/gma500/intel_bios.h b/drivers/gpu/drm/gma500/intel_bios.h index a1f9ce9465a5..0e6facf21e33 100644 --- a/drivers/gpu/drm/gma500/intel_bios.h +++ b/drivers/gpu/drm/gma500/intel_bios.h @@ -227,7 +227,7 @@ struct bdb_general_definitions { * number = (block_size - sizeof(bdb_general_definitions))/ * sizeof(child_device_config); */ - struct child_device_config devices[0]; + struct child_device_config devices[]; }; struct bdb_lvds_options { diff --git a/drivers/gpu/drm/gma500/psb_drv.c b/drivers/gpu/drm/gma500/psb_drv.c index 6956c8e7501c..2411eb9827b8 100644 --- a/drivers/gpu/drm/gma500/psb_drv.c +++ b/drivers/gpu/drm/gma500/psb_drv.c @@ -363,7 +363,6 @@ static int psb_driver_load(struct drm_device *dev, unsigned long flags) drm_irq_install(dev, dev->pdev->irq); dev->max_vblank_count = 0xffffff; /* only 24 bits of frame count */ - dev->driver->get_vblank_counter = psb_get_vblank_counter; psb_modeset_init(dev); psb_fbdev_init(dev); @@ -507,9 +506,6 @@ static struct drm_driver driver = { .irq_postinstall = psb_irq_postinstall, .irq_uninstall = psb_irq_uninstall, .irq_handler = psb_irq_handler, - .enable_vblank = psb_enable_vblank, - .disable_vblank = psb_disable_vblank, - .get_vblank_counter = psb_get_vblank_counter, .gem_free_object = psb_gem_free_object, .gem_vm_ops = &psb_gem_vm_ops, diff --git a/drivers/gpu/drm/gma500/psb_drv.h b/drivers/gpu/drm/gma500/psb_drv.h index 3d4ef3071d45..956926341316 100644 --- a/drivers/gpu/drm/gma500/psb_drv.h +++ b/drivers/gpu/drm/gma500/psb_drv.h @@ -681,15 +681,15 @@ extern void psb_irq_turn_off_dpst(struct drm_device *dev); extern void psb_irq_uninstall_islands(struct drm_device *dev, int hw_islands); extern int psb_vblank_wait2(struct drm_device *dev, unsigned int *sequence); extern int psb_vblank_wait(struct drm_device *dev, unsigned int *sequence); -extern int psb_enable_vblank(struct drm_device *dev, unsigned int pipe); -extern void psb_disable_vblank(struct drm_device *dev, unsigned int pipe); +extern int psb_enable_vblank(struct drm_crtc *crtc); +extern void psb_disable_vblank(struct drm_crtc *crtc); void psb_enable_pipestat(struct drm_psb_private *dev_priv, int pipe, u32 mask); void psb_disable_pipestat(struct drm_psb_private *dev_priv, int pipe, u32 mask); -extern u32 psb_get_vblank_counter(struct drm_device *dev, unsigned int pipe); +extern u32 psb_get_vblank_counter(struct drm_crtc *crtc); /* framebuffer.c */ extern int psbfb_probed(struct drm_device *dev); diff --git a/drivers/gpu/drm/gma500/psb_intel_display.c b/drivers/gpu/drm/gma500/psb_intel_display.c index fed3b563e62e..531c5485be17 100644 --- a/drivers/gpu/drm/gma500/psb_intel_display.c +++ b/drivers/gpu/drm/gma500/psb_intel_display.c @@ -433,6 +433,9 @@ const struct drm_crtc_funcs psb_intel_crtc_funcs = { .set_config = gma_crtc_set_config, .destroy = gma_crtc_destroy, .page_flip = gma_crtc_page_flip, + .enable_vblank = psb_enable_vblank, + .disable_vblank = psb_disable_vblank, + .get_vblank_counter = psb_get_vblank_counter, }; const struct gma_clock_funcs psb_clock_funcs = { diff --git a/drivers/gpu/drm/gma500/psb_irq.c b/drivers/gpu/drm/gma500/psb_irq.c index 91f90016dba9..15eb3770d817 100644 --- a/drivers/gpu/drm/gma500/psb_irq.c +++ b/drivers/gpu/drm/gma500/psb_irq.c @@ -506,8 +506,10 @@ int psb_irq_disable_dpst(struct drm_device *dev) /* * It is used to enable VBLANK interrupt */ -int psb_enable_vblank(struct drm_device *dev, unsigned int pipe) +int psb_enable_vblank(struct drm_crtc *crtc) { + struct drm_device *dev = crtc->dev; + unsigned int pipe = crtc->index; struct drm_psb_private *dev_priv = dev->dev_private; unsigned long irqflags; uint32_t reg_val = 0; @@ -545,8 +547,10 @@ int psb_enable_vblank(struct drm_device *dev, unsigned int pipe) /* * It is used to disable VBLANK interrupt */ -void psb_disable_vblank(struct drm_device *dev, unsigned int pipe) +void psb_disable_vblank(struct drm_crtc *crtc) { + struct drm_device *dev = crtc->dev; + unsigned int pipe = crtc->index; struct drm_psb_private *dev_priv = dev->dev_private; unsigned long irqflags; @@ -618,8 +622,10 @@ void mdfld_disable_te(struct drm_device *dev, int pipe) /* Called from drm generic code, passed a 'crtc', which * we use as a pipe index */ -u32 psb_get_vblank_counter(struct drm_device *dev, unsigned int pipe) +u32 psb_get_vblank_counter(struct drm_crtc *crtc) { + struct drm_device *dev = crtc->dev; + unsigned int pipe = crtc->index; uint32_t high_frame = PIPEAFRAMEHIGH; uint32_t low_frame = PIPEAFRAMEPIXEL; uint32_t pipeconf_reg = PIPEACONF; diff --git a/drivers/gpu/drm/gma500/psb_irq.h b/drivers/gpu/drm/gma500/psb_irq.h index 58fd502e3b9d..4f73998848d1 100644 --- a/drivers/gpu/drm/gma500/psb_irq.h +++ b/drivers/gpu/drm/gma500/psb_irq.h @@ -12,6 +12,7 @@ #ifndef _PSB_IRQ_H_ #define _PSB_IRQ_H_ +struct drm_crtc; struct drm_device; bool sysirq_init(struct drm_device *dev); @@ -26,9 +27,9 @@ int psb_irq_enable_dpst(struct drm_device *dev); int psb_irq_disable_dpst(struct drm_device *dev); void psb_irq_turn_on_dpst(struct drm_device *dev); void psb_irq_turn_off_dpst(struct drm_device *dev); -int psb_enable_vblank(struct drm_device *dev, unsigned int pipe); -void psb_disable_vblank(struct drm_device *dev, unsigned int pipe); -u32 psb_get_vblank_counter(struct drm_device *dev, unsigned int pipe); +int psb_enable_vblank(struct drm_crtc *crtc); +void psb_disable_vblank(struct drm_crtc *crtc); +u32 psb_get_vblank_counter(struct drm_crtc *crtc); int mdfld_enable_te(struct drm_device *dev, int pipe); void mdfld_disable_te(struct drm_device *dev, int pipe); diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_de.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_de.c index 7fa7d4933f60..55b46a7150a5 100644 --- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_de.c +++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_de.c @@ -40,6 +40,7 @@ struct hibmc_dislay_pll_config { }; static const struct hibmc_dislay_pll_config hibmc_pll_table[] = { + {640, 480, CRT_PLL1_HS_25MHZ, CRT_PLL2_HS_25MHZ}, {800, 600, CRT_PLL1_HS_40MHZ, CRT_PLL2_HS_40MHZ}, {1024, 768, CRT_PLL1_HS_65MHZ, CRT_PLL2_HS_65MHZ}, {1152, 864, CRT_PLL1_HS_80MHZ_1152, CRT_PLL2_HS_80MHZ}, @@ -47,6 +48,8 @@ static const struct hibmc_dislay_pll_config hibmc_pll_table[] = { {1280, 720, CRT_PLL1_HS_74MHZ, CRT_PLL2_HS_74MHZ}, {1280, 960, CRT_PLL1_HS_108MHZ, CRT_PLL2_HS_108MHZ}, {1280, 1024, CRT_PLL1_HS_108MHZ, CRT_PLL2_HS_108MHZ}, + {1440, 900, CRT_PLL1_HS_106MHZ, CRT_PLL2_HS_106MHZ}, + {1600, 900, CRT_PLL1_HS_108MHZ, CRT_PLL2_HS_108MHZ}, {1600, 1200, CRT_PLL1_HS_162MHZ, CRT_PLL2_HS_162MHZ}, {1920, 1080, CRT_PLL1_HS_148MHZ, CRT_PLL2_HS_148MHZ}, {1920, 1200, CRT_PLL1_HS_193MHZ, CRT_PLL2_HS_193MHZ}, @@ -80,6 +83,9 @@ static int hibmc_plane_atomic_check(struct drm_plane *plane, return -EINVAL; } + if (!crtc_state->enable) + return 0; + if (state->crtc_x + state->crtc_w > crtc_state->adjusted_mode.hdisplay || state->crtc_y + state->crtc_h > @@ -184,6 +190,20 @@ static struct drm_plane *hibmc_plane_init(struct hibmc_drm_private *priv) return plane; } +static void hibmc_crtc_dpms(struct drm_crtc *crtc, int dpms) +{ + struct hibmc_drm_private *priv = crtc->dev->dev_private; + unsigned int reg; + + reg = readl(priv->mmio + HIBMC_CRT_DISP_CTL); + reg &= ~HIBMC_CRT_DISP_CTL_DPMS_MASK; + reg |= HIBMC_FIELD(HIBMC_CRT_DISP_CTL_DPMS, dpms); + reg &= ~HIBMC_CRT_DISP_CTL_TIMING_MASK; + if (dpms == HIBMC_CRT_DPMS_ON) + reg |= HIBMC_CRT_DISP_CTL_TIMING(1); + writel(reg, priv->mmio + HIBMC_CRT_DISP_CTL); +} + static void hibmc_crtc_atomic_enable(struct drm_crtc *crtc, struct drm_crtc_state *old_state) { @@ -200,6 +220,7 @@ static void hibmc_crtc_atomic_enable(struct drm_crtc *crtc, reg |= HIBMC_CURR_GATE_DISPLAY(1); hibmc_set_current_gate(priv, reg); drm_crtc_vblank_on(crtc); + hibmc_crtc_dpms(crtc, HIBMC_CRT_DPMS_ON); } static void hibmc_crtc_atomic_disable(struct drm_crtc *crtc, @@ -208,6 +229,7 @@ static void hibmc_crtc_atomic_disable(struct drm_crtc *crtc, unsigned int reg; struct hibmc_drm_private *priv = crtc->dev->dev_private; + hibmc_crtc_dpms(crtc, HIBMC_CRT_DPMS_OFF); drm_crtc_vblank_off(crtc); hibmc_set_power_mode(priv, HIBMC_PW_MODE_CTL_MODE_SLEEP); @@ -221,6 +243,25 @@ static void hibmc_crtc_atomic_disable(struct drm_crtc *crtc, hibmc_set_current_gate(priv, reg); } +static enum drm_mode_status +hibmc_crtc_mode_valid(struct drm_crtc *crtc, + const struct drm_display_mode *mode) +{ + int i = 0; + int vrefresh = drm_mode_vrefresh(mode); + + if (vrefresh < 59 || vrefresh > 61) + return MODE_NOCLOCK; + + for (i = 0; i < ARRAY_SIZE(hibmc_pll_table); i++) { + if (hibmc_pll_table[i].hdisplay == mode->hdisplay && + hibmc_pll_table[i].vdisplay == mode->vdisplay) + return MODE_OK; + } + + return MODE_BAD; +} + static unsigned int format_pll_reg(void) { unsigned int pllreg = 0; @@ -435,6 +476,42 @@ static void hibmc_crtc_disable_vblank(struct drm_crtc *crtc) priv->mmio + HIBMC_RAW_INTERRUPT_EN); } +static void hibmc_crtc_load_lut(struct drm_crtc *crtc) +{ + struct hibmc_drm_private *priv = crtc->dev->dev_private; + void __iomem *mmio = priv->mmio; + u16 *r, *g, *b; + unsigned int reg; + int i; + + r = crtc->gamma_store; + g = r + crtc->gamma_size; + b = g + crtc->gamma_size; + + for (i = 0; i < crtc->gamma_size; i++) { + unsigned int offset = i << 2; + u8 red = *r++ >> 8; + u8 green = *g++ >> 8; + u8 blue = *b++ >> 8; + u32 rgb = (red << 16) | (green << 8) | blue; + + writel(rgb, mmio + HIBMC_CRT_PALETTE + offset); + } + + reg = readl(priv->mmio + HIBMC_CRT_DISP_CTL); + reg |= HIBMC_FIELD(HIBMC_CTL_DISP_CTL_GAMMA, 1); + writel(reg, priv->mmio + HIBMC_CRT_DISP_CTL); +} + +static int hibmc_crtc_gamma_set(struct drm_crtc *crtc, u16 *red, u16 *green, + u16 *blue, uint32_t size, + struct drm_modeset_acquire_ctx *ctx) +{ + hibmc_crtc_load_lut(crtc); + + return 0; +} + static const struct drm_crtc_funcs hibmc_crtc_funcs = { .page_flip = drm_atomic_helper_page_flip, .set_config = drm_atomic_helper_set_config, @@ -444,6 +521,7 @@ static const struct drm_crtc_funcs hibmc_crtc_funcs = { .atomic_destroy_state = drm_atomic_helper_crtc_destroy_state, .enable_vblank = hibmc_crtc_enable_vblank, .disable_vblank = hibmc_crtc_disable_vblank, + .gamma_set = hibmc_crtc_gamma_set, }; static const struct drm_crtc_helper_funcs hibmc_crtc_helper_funcs = { @@ -452,6 +530,7 @@ static const struct drm_crtc_helper_funcs hibmc_crtc_helper_funcs = { .atomic_flush = hibmc_crtc_atomic_flush, .atomic_enable = hibmc_crtc_atomic_enable, .atomic_disable = hibmc_crtc_atomic_disable, + .mode_valid = hibmc_crtc_mode_valid, }; int hibmc_de_init(struct hibmc_drm_private *priv) diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c index 4a8a4cfb4b75..222356a4f9a8 100644 --- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c +++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c @@ -91,11 +91,11 @@ static int hibmc_kms_init(struct hibmc_drm_private *priv) priv->dev->mode_config.min_width = 0; priv->dev->mode_config.min_height = 0; priv->dev->mode_config.max_width = 1920; - priv->dev->mode_config.max_height = 1440; + priv->dev->mode_config.max_height = 1200; priv->dev->mode_config.fb_base = priv->fb_base; priv->dev->mode_config.preferred_depth = 24; - priv->dev->mode_config.prefer_shadow = 0; + priv->dev->mode_config.prefer_shadow = 1; priv->dev->mode_config.funcs = (void *)&hibmc_mode_funcs; @@ -327,6 +327,11 @@ static int hibmc_pci_probe(struct pci_dev *pdev, struct drm_device *dev; int ret; + ret = drm_fb_helper_remove_conflicting_pci_framebuffers(pdev, + "hibmcdrmfb"); + if (ret) + return ret; + dev = drm_dev_alloc(&hibmc_driver, &pdev->dev); if (IS_ERR(dev)) { DRM_ERROR("failed to allocate drm_device\n"); diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_regs.h b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_regs.h index b63a1ee15ceb..17b30c393b10 100644 --- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_regs.h +++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_regs.h @@ -68,6 +68,12 @@ #define HIBMC_CRT_DISP_CTL 0x80200 +#define HIBMC_CRT_DISP_CTL_DPMS(x) ((x) << 30) +#define HIBMC_CRT_DISP_CTL_DPMS_MASK 0xc0000000 + +#define HIBMC_CRT_DPMS_ON 0 +#define HIBMC_CRT_DPMS_OFF 3 + #define HIBMC_CRT_DISP_CTL_CRTSELECT(x) ((x) << 25) #define HIBMC_CRT_DISP_CTL_CRTSELECT_MASK 0x2000000 @@ -85,6 +91,9 @@ #define HIBMC_CRT_DISP_CTL_TIMING(x) ((x) << 8) #define HIBMC_CRT_DISP_CTL_TIMING_MASK 0x100 +#define HIBMC_CTL_DISP_CTL_GAMMA(x) ((x) << 3) +#define HIBMC_CTL_DISP_CTL_GAMMA_MASK 0x08 + #define HIBMC_CRT_DISP_CTL_PLANE(x) ((x) << 2) #define HIBMC_CRT_DISP_CTL_PLANE_MASK 4 @@ -170,6 +179,7 @@ #define CRT_PLL1_HS_74MHZ 0x23941dc2 #define CRT_PLL1_HS_80MHZ 0x23941001 #define CRT_PLL1_HS_80MHZ_1152 0x23540fc2 +#define CRT_PLL1_HS_106MHZ 0x237C1641 #define CRT_PLL1_HS_108MHZ 0x23b41b01 #define CRT_PLL1_HS_162MHZ 0x23480681 #define CRT_PLL1_HS_148MHZ 0x23541dc2 @@ -182,10 +192,13 @@ #define CRT_PLL2_HS_78MHZ 0x50E147AE #define CRT_PLL2_HS_74MHZ 0x602B6AE7 #define CRT_PLL2_HS_80MHZ 0x70000000 +#define CRT_PLL2_HS_106MHZ 0x0075c28f #define CRT_PLL2_HS_108MHZ 0x80000000 #define CRT_PLL2_HS_162MHZ 0xA0000000 #define CRT_PLL2_HS_148MHZ 0xB0CCCCCD #define CRT_PLL2_HS_193MHZ 0xC0872B02 +#define HIBMC_CRT_PALETTE 0x80C00 + #define HIBMC_FIELD(field, value) (field(value) & field##_MASK) #endif diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c index 6d98fdc06f6c..678ac2ef2a93 100644 --- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c +++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c @@ -11,8 +11,10 @@ * Jianhua Li <lijianhua@huawei.com> */ +#include <drm/drm_gem_vram_helper.h> #include <drm/drm_atomic_helper.h> #include <drm/drm_probe_helper.h> +#include <drm/drm_crtc_helper.h> #include <drm/drm_print.h> #include "hibmc_drm_drv.h" @@ -20,7 +22,14 @@ static int hibmc_connector_get_modes(struct drm_connector *connector) { - return drm_add_modes_noedid(connector, 800, 600); + int count; + + count = drm_add_modes_noedid(connector, + connector->dev->mode_config.max_width, + connector->dev->mode_config.max_height); + drm_set_preferred_mode(connector, 1024, 768); + + return count; } static enum drm_mode_status hibmc_connector_mode_valid(struct drm_connector *connector, diff --git a/drivers/gpu/drm/hisilicon/kirin/dw_drm_dsi.c b/drivers/gpu/drm/hisilicon/kirin/dw_drm_dsi.c index bdcf9c6ae9e9..f31068d74b18 100644 --- a/drivers/gpu/drm/hisilicon/kirin/dw_drm_dsi.c +++ b/drivers/gpu/drm/hisilicon/kirin/dw_drm_dsi.c @@ -777,7 +777,7 @@ static int dsi_bridge_init(struct drm_device *dev, struct dw_dsi *dsi) int ret; /* associate the bridge to dsi encoder */ - ret = drm_bridge_attach(encoder, bridge, NULL); + ret = drm_bridge_attach(encoder, bridge, NULL, 0); if (ret) { DRM_ERROR("failed to attach external bridge\n"); return ret; diff --git a/drivers/gpu/drm/hisilicon/kirin/kirin_ade_reg.h b/drivers/gpu/drm/hisilicon/kirin/kirin_ade_reg.h index 0da860200410..e2ac09894a6d 100644 --- a/drivers/gpu/drm/hisilicon/kirin/kirin_ade_reg.h +++ b/drivers/gpu/drm/hisilicon/kirin/kirin_ade_reg.h @@ -83,7 +83,6 @@ #define VSIZE_OFST 20 #define LDI_INT_EN 0x741C #define FRAME_END_INT_EN_OFST 1 -#define UNDERFLOW_INT_EN_OFST 2 #define LDI_CTRL 0x7420 #define BPP_OFST 3 #define DATA_GATE_EN BIT(2) diff --git a/drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c b/drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c index 73cd28a6ea07..86000127d4ee 100644 --- a/drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c +++ b/drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c @@ -46,7 +46,6 @@ struct ade_hw_ctx { struct clk *media_noc_clk; struct clk *ade_pix_clk; struct reset_control *reset; - struct work_struct display_reset_wq; bool power_on; int irq; @@ -136,7 +135,6 @@ static void ade_init(struct ade_hw_ctx *ctx) */ ade_update_bits(base + ADE_CTRL, FRM_END_START_OFST, FRM_END_START_MASK, REG_EFFECTIVE_IN_ADEEN_FRMEND); - ade_update_bits(base + LDI_INT_EN, UNDERFLOW_INT_EN_OFST, MASK(1), 1); } static bool ade_crtc_mode_fixup(struct drm_crtc *crtc, @@ -304,17 +302,6 @@ static void ade_crtc_disable_vblank(struct drm_crtc *crtc) MASK(1), 0); } -static void drm_underflow_wq(struct work_struct *work) -{ - struct ade_hw_ctx *ctx = container_of(work, struct ade_hw_ctx, - display_reset_wq); - struct drm_device *drm_dev = ctx->crtc->dev; - struct drm_atomic_state *state; - - state = drm_atomic_helper_suspend(drm_dev); - drm_atomic_helper_resume(drm_dev, state); -} - static irqreturn_t ade_irq_handler(int irq, void *data) { struct ade_hw_ctx *ctx = data; @@ -331,12 +318,6 @@ static irqreturn_t ade_irq_handler(int irq, void *data) MASK(1), 1); drm_crtc_handle_vblank(crtc); } - if (status & BIT(UNDERFLOW_INT_EN_OFST)) { - ade_update_bits(base + LDI_INT_CLR, UNDERFLOW_INT_EN_OFST, - MASK(1), 1); - DRM_ERROR("LDI underflow!"); - schedule_work(&ctx->display_reset_wq); - } return IRQ_HANDLED; } @@ -919,7 +900,6 @@ static void *ade_hw_ctx_alloc(struct platform_device *pdev, if (ret) return ERR_PTR(-EIO); - INIT_WORK(&ctx->display_reset_wq, drm_underflow_wq); ctx->crtc = crtc; return ctx; diff --git a/drivers/gpu/drm/i2c/tda998x_drv.c b/drivers/gpu/drm/i2c/tda998x_drv.c index a63790d32d75..c3332209f27a 100644 --- a/drivers/gpu/drm/i2c/tda998x_drv.c +++ b/drivers/gpu/drm/i2c/tda998x_drv.c @@ -1356,10 +1356,16 @@ static int tda998x_connector_init(struct tda998x_priv *priv, /* DRM bridge functions */ -static int tda998x_bridge_attach(struct drm_bridge *bridge) +static int tda998x_bridge_attach(struct drm_bridge *bridge, + enum drm_bridge_attach_flags flags) { struct tda998x_priv *priv = bridge_to_tda998x_priv(bridge); + if (flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR) { + DRM_ERROR("Fix bridge driver to make connector optional!"); + return -EINVAL; + } + return tda998x_connector_init(priv, bridge->dev); } @@ -2022,7 +2028,7 @@ static int tda998x_encoder_init(struct device *dev, struct drm_device *drm) if (ret) goto err_encoder; - ret = drm_bridge_attach(&priv->encoder, &priv->bridge, NULL); + ret = drm_bridge_attach(&priv->encoder, &priv->bridge, NULL, 0); if (ret) goto err_bridge; diff --git a/drivers/gpu/drm/i915/.gitignore b/drivers/gpu/drm/i915/.gitignore index d9a77f3b59b2..81972dce1aff 100644 --- a/drivers/gpu/drm/i915/.gitignore +++ b/drivers/gpu/drm/i915/.gitignore @@ -1 +1,2 @@ +# SPDX-License-Identifier: GPL-2.0-only *.hdrtest diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile index 2fce8b0040f3..44c506b7e117 100644 --- a/drivers/gpu/drm/i915/Makefile +++ b/drivers/gpu/drm/i915/Makefile @@ -28,9 +28,6 @@ subdir-ccflags-$(CONFIG_DRM_I915_WERROR) += -Werror CFLAGS_i915_pci.o = $(call cc-disable-warning, override-init) CFLAGS_display/intel_fbdev.o = $(call cc-disable-warning, override-init) -subdir-ccflags-y += \ - $(call as-instr,movntdqa (%eax)$(comma)%xmm0,-DCONFIG_AS_MOVNTDQA) - subdir-ccflags-y += -I$(srctree)/$(src) # Please keep these build lists sorted! diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c index 131f279c1437..af5b4055b38a 100644 --- a/drivers/gpu/drm/i915/display/intel_display.c +++ b/drivers/gpu/drm/i915/display/intel_display.c @@ -16347,6 +16347,7 @@ static const struct drm_crtc_funcs bdw_crtc_funcs = { .get_vblank_counter = g4x_get_vblank_counter, .enable_vblank = bdw_enable_vblank, .disable_vblank = bdw_disable_vblank, + .get_vblank_timestamp = intel_crtc_get_vblank_timestamp, }; static const struct drm_crtc_funcs ilk_crtc_funcs = { @@ -16355,6 +16356,7 @@ static const struct drm_crtc_funcs ilk_crtc_funcs = { .get_vblank_counter = g4x_get_vblank_counter, .enable_vblank = ilk_enable_vblank, .disable_vblank = ilk_disable_vblank, + .get_vblank_timestamp = intel_crtc_get_vblank_timestamp, }; static const struct drm_crtc_funcs g4x_crtc_funcs = { @@ -16363,6 +16365,7 @@ static const struct drm_crtc_funcs g4x_crtc_funcs = { .get_vblank_counter = g4x_get_vblank_counter, .enable_vblank = i965_enable_vblank, .disable_vblank = i965_disable_vblank, + .get_vblank_timestamp = intel_crtc_get_vblank_timestamp, }; static const struct drm_crtc_funcs i965_crtc_funcs = { @@ -16371,6 +16374,7 @@ static const struct drm_crtc_funcs i965_crtc_funcs = { .get_vblank_counter = i915_get_vblank_counter, .enable_vblank = i965_enable_vblank, .disable_vblank = i965_disable_vblank, + .get_vblank_timestamp = intel_crtc_get_vblank_timestamp, }; static const struct drm_crtc_funcs i915gm_crtc_funcs = { @@ -16379,6 +16383,7 @@ static const struct drm_crtc_funcs i915gm_crtc_funcs = { .get_vblank_counter = i915_get_vblank_counter, .enable_vblank = i915gm_enable_vblank, .disable_vblank = i915gm_disable_vblank, + .get_vblank_timestamp = intel_crtc_get_vblank_timestamp, }; static const struct drm_crtc_funcs i915_crtc_funcs = { @@ -16387,6 +16392,7 @@ static const struct drm_crtc_funcs i915_crtc_funcs = { .get_vblank_counter = i915_get_vblank_counter, .enable_vblank = i8xx_enable_vblank, .disable_vblank = i8xx_disable_vblank, + .get_vblank_timestamp = intel_crtc_get_vblank_timestamp, }; static const struct drm_crtc_funcs i8xx_crtc_funcs = { @@ -16395,6 +16401,7 @@ static const struct drm_crtc_funcs i8xx_crtc_funcs = { /* no hw vblank counter */ .enable_vblank = i8xx_enable_vblank, .disable_vblank = i8xx_disable_vblank, + .get_vblank_timestamp = intel_crtc_get_vblank_timestamp, }; static struct intel_crtc *intel_crtc_alloc(void) diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c index e1694638ea93..0688be42666d 100644 --- a/drivers/gpu/drm/i915/display/intel_dp.c +++ b/drivers/gpu/drm/i915/display/intel_dp.c @@ -1391,7 +1391,7 @@ intel_dp_aux_xfer(struct intel_dp *intel_dp, * lowest possible wakeup latency and so prevent the cpu from going into * deep sleep states. */ - pm_qos_update_request(&i915->pm_qos, 0); + cpu_latency_qos_update_request(&i915->pm_qos, 0); intel_dp_check_edp(intel_dp); @@ -1524,7 +1524,7 @@ done: ret = recv_bytes; out: - pm_qos_update_request(&i915->pm_qos, PM_QOS_DEFAULT_VALUE); + cpu_latency_qos_update_request(&i915->pm_qos, PM_QOS_DEFAULT_VALUE); if (vdd) edp_panel_vdd_off(intel_dp, false); diff --git a/drivers/gpu/drm/i915/display/intel_dp_mst.c b/drivers/gpu/drm/i915/display/intel_dp_mst.c index 76ec2632f8de..61605eb8c2af 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_mst.c +++ b/drivers/gpu/drm/i915/display/intel_dp_mst.c @@ -764,36 +764,8 @@ err: return NULL; } -static void intel_dp_register_mst_connector(struct drm_connector *connector) -{ - struct drm_i915_private *dev_priv = to_i915(connector->dev); - - if (dev_priv->fbdev) - drm_fb_helper_add_one_connector(&dev_priv->fbdev->helper, - connector); - - drm_connector_register(connector); -} - -static void intel_dp_destroy_mst_connector(struct drm_dp_mst_topology_mgr *mgr, - struct drm_connector *connector) -{ - struct drm_i915_private *dev_priv = to_i915(connector->dev); - - DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n", connector->base.id, connector->name); - drm_connector_unregister(connector); - - if (dev_priv->fbdev) - drm_fb_helper_remove_one_connector(&dev_priv->fbdev->helper, - connector); - - drm_connector_put(connector); -} - static const struct drm_dp_mst_topology_cbs mst_cbs = { .add_connector = intel_dp_add_mst_connector, - .register_connector = intel_dp_register_mst_connector, - .destroy_connector = intel_dp_destroy_mst_connector, }; static struct intel_dp_mst_encoder * diff --git a/drivers/gpu/drm/i915/display/intel_fbdev.c b/drivers/gpu/drm/i915/display/intel_fbdev.c index a5ddd2067b31..bd39eb6a21b8 100644 --- a/drivers/gpu/drm/i915/display/intel_fbdev.c +++ b/drivers/gpu/drm/i915/display/intel_fbdev.c @@ -465,7 +465,7 @@ int intel_fbdev_init(struct drm_device *dev) if (!intel_fbdev_init_bios(dev, ifbdev)) ifbdev->preferred_bpp = 32; - ret = drm_fb_helper_init(dev, &ifbdev->helper, 4); + ret = drm_fb_helper_init(dev, &ifbdev->helper); if (ret) { kfree(ifbdev); return ret; @@ -474,8 +474,6 @@ int intel_fbdev_init(struct drm_device *dev) dev_priv->fbdev = ifbdev; INIT_WORK(&dev_priv->fbdev_suspend_work, intel_fbdev_suspend_worker); - drm_fb_helper_single_add_all_connectors(&ifbdev->helper); - return 0; } diff --git a/drivers/gpu/drm/i915/gt/intel_context.c b/drivers/gpu/drm/i915/gt/intel_context.c index e4aece20bc80..74ddb49b2941 100644 --- a/drivers/gpu/drm/i915/gt/intel_context.c +++ b/drivers/gpu/drm/i915/gt/intel_context.c @@ -97,6 +97,8 @@ int __intel_context_do_pin(struct intel_context *ce) { int err; + GEM_BUG_ON(intel_context_is_closed(ce)); + if (unlikely(!test_bit(CONTEXT_ALLOC_BIT, &ce->flags))) { err = intel_context_alloc_state(ce); if (err) diff --git a/drivers/gpu/drm/i915/gt/selftest_rc6.c b/drivers/gpu/drm/i915/gt/selftest_rc6.c index 2f3316c924a3..08c3dbd41b12 100644 --- a/drivers/gpu/drm/i915/gt/selftest_rc6.c +++ b/drivers/gpu/drm/i915/gt/selftest_rc6.c @@ -72,6 +72,7 @@ int live_rc6_manual(void *arg) msleep(1); /* wakeup is not immediate, takes about 100us on icl */ res[0] = rc6_residency(rc6); + dt = ktime_get(); rc0_power = energy_uJ(rc6); msleep(250); diff --git a/drivers/gpu/drm/i915/gvt/cmd_parser.c b/drivers/gpu/drm/i915/gvt/cmd_parser.c index 9e065ad0658f..a3cc080a46c6 100644 --- a/drivers/gpu/drm/i915/gvt/cmd_parser.c +++ b/drivers/gpu/drm/i915/gvt/cmd_parser.c @@ -164,6 +164,7 @@ struct decode_info { #define OP_STATE_BASE_ADDRESS OP_3D_MEDIA(0x0, 0x1, 0x01) #define OP_STATE_SIP OP_3D_MEDIA(0x0, 0x1, 0x02) #define OP_3D_MEDIA_0_1_4 OP_3D_MEDIA(0x0, 0x1, 0x04) +#define OP_SWTESS_BASE_ADDRESS OP_3D_MEDIA(0x0, 0x1, 0x03) #define OP_3DSTATE_VF_STATISTICS_GM45 OP_3D_MEDIA(0x1, 0x0, 0x0B) @@ -967,18 +968,6 @@ static int cmd_handler_lri(struct parser_exec_state *s) { int i, ret = 0; int cmd_len = cmd_length(s); - u32 valid_len = CMD_LEN(1); - - /* - * Official intel docs are somewhat sloppy , check the definition of - * MI_LOAD_REGISTER_IMM. - */ - #define MAX_VALID_LEN 127 - if ((cmd_len < valid_len) || (cmd_len > MAX_VALID_LEN)) { - gvt_err("len is not valid: len=%u valid_len=%u\n", - cmd_len, valid_len); - return -EFAULT; - } for (i = 1; i < cmd_len; i += 2) { if (IS_BROADWELL(s->engine->i915) && s->engine->id != RCS0) { @@ -2485,6 +2474,9 @@ static const struct cmd_info cmd_info[] = { {"OP_3D_MEDIA_0_1_4", OP_3D_MEDIA_0_1_4, F_LEN_VAR, R_RCS, D_ALL, ADDR_FIX_1(1), 8, NULL}, + {"OP_SWTESS_BASE_ADDRESS", OP_SWTESS_BASE_ADDRESS, + F_LEN_VAR, R_RCS, D_ALL, ADDR_FIX_2(1, 2), 3, NULL}, + {"3DSTATE_VS", OP_3DSTATE_VS, F_LEN_VAR, R_RCS, D_ALL, 0, 8, NULL}, {"3DSTATE_SF", OP_3DSTATE_SF, F_LEN_VAR, R_RCS, D_ALL, 0, 8, NULL}, diff --git a/drivers/gpu/drm/i915/gvt/display.c b/drivers/gpu/drm/i915/gvt/display.c index 6e5c9885d9fe..a83df2f84eb9 100644 --- a/drivers/gpu/drm/i915/gvt/display.c +++ b/drivers/gpu/drm/i915/gvt/display.c @@ -221,7 +221,7 @@ static void emulate_monitor_status_change(struct intel_vgpu *vgpu) ~(TRANS_DDI_BPC_MASK | TRANS_DDI_MODE_SELECT_MASK | TRANS_DDI_PORT_MASK); vgpu_vreg_t(vgpu, TRANS_DDI_FUNC_CTL(TRANSCODER_A)) |= - (TRANS_DDI_BPC_8 | TRANS_DDI_MODE_SELECT_DVI | + (TRANS_DDI_BPC_8 | TRANS_DDI_MODE_SELECT_DP_SST | (PORT_B << TRANS_DDI_PORT_SHIFT) | TRANS_DDI_FUNC_ENABLE); if (IS_BROADWELL(dev_priv)) { @@ -241,7 +241,7 @@ static void emulate_monitor_status_change(struct intel_vgpu *vgpu) ~(TRANS_DDI_BPC_MASK | TRANS_DDI_MODE_SELECT_MASK | TRANS_DDI_PORT_MASK); vgpu_vreg_t(vgpu, TRANS_DDI_FUNC_CTL(TRANSCODER_A)) |= - (TRANS_DDI_BPC_8 | TRANS_DDI_MODE_SELECT_DVI | + (TRANS_DDI_BPC_8 | TRANS_DDI_MODE_SELECT_DP_SST | (PORT_C << TRANS_DDI_PORT_SHIFT) | TRANS_DDI_FUNC_ENABLE); if (IS_BROADWELL(dev_priv)) { @@ -261,7 +261,7 @@ static void emulate_monitor_status_change(struct intel_vgpu *vgpu) ~(TRANS_DDI_BPC_MASK | TRANS_DDI_MODE_SELECT_MASK | TRANS_DDI_PORT_MASK); vgpu_vreg_t(vgpu, TRANS_DDI_FUNC_CTL(TRANSCODER_A)) |= - (TRANS_DDI_BPC_8 | TRANS_DDI_MODE_SELECT_DVI | + (TRANS_DDI_BPC_8 | TRANS_DDI_MODE_SELECT_DP_SST | (PORT_D << TRANS_DDI_PORT_SHIFT) | TRANS_DDI_FUNC_ENABLE); if (IS_BROADWELL(dev_priv)) { diff --git a/drivers/gpu/drm/i915/gvt/dmabuf.c b/drivers/gpu/drm/i915/gvt/dmabuf.c index 4ff60c793a21..37fc460414a8 100644 --- a/drivers/gpu/drm/i915/gvt/dmabuf.c +++ b/drivers/gpu/drm/i915/gvt/dmabuf.c @@ -151,12 +151,12 @@ static void dmabuf_gem_object_free(struct kref *kref) dmabuf_obj = container_of(pos, struct intel_vgpu_dmabuf_obj, list); if (dmabuf_obj == obj) { + list_del(pos); intel_gvt_hypervisor_put_vfio_device(vgpu); idr_remove(&vgpu->object_idr, dmabuf_obj->dmabuf_id); kfree(dmabuf_obj->info); kfree(dmabuf_obj); - list_del(pos); break; } } diff --git a/drivers/gpu/drm/i915/gvt/handlers.c b/drivers/gpu/drm/i915/gvt/handlers.c index 0182e2a5acff..2faf50e1b051 100644 --- a/drivers/gpu/drm/i915/gvt/handlers.c +++ b/drivers/gpu/drm/i915/gvt/handlers.c @@ -462,11 +462,14 @@ static int pipeconf_mmio_write(struct intel_vgpu *vgpu, unsigned int offset, return 0; } -/* ascendingly sorted */ +/* sorted in ascending order */ static i915_reg_t force_nonpriv_white_list[] = { + _MMIO(0xd80), GEN9_CS_DEBUG_MODE1, //_MMIO(0x20ec) GEN9_CTX_PREEMPT_REG,//_MMIO(0x2248) - PS_INVOCATION_COUNT,//_MMIO(0x2348) + CL_PRIMITIVES_COUNT, //_MMIO(0x2340) + PS_INVOCATION_COUNT, //_MMIO(0x2348) + PS_DEPTH_COUNT, //_MMIO(0x2350) GEN8_CS_CHICKEN1,//_MMIO(0x2580) _MMIO(0x2690), _MMIO(0x2694), @@ -491,6 +494,7 @@ static i915_reg_t force_nonpriv_white_list[] = { _MMIO(0xe18c), _MMIO(0xe48c), _MMIO(0xe5f4), + _MMIO(0x64844), }; /* a simple bsearch */ diff --git a/drivers/gpu/drm/i915/gvt/opregion.c b/drivers/gpu/drm/i915/gvt/opregion.c index 867e7629025b..33569b910ed5 100644 --- a/drivers/gpu/drm/i915/gvt/opregion.c +++ b/drivers/gpu/drm/i915/gvt/opregion.c @@ -147,15 +147,14 @@ static void virt_vbt_generation(struct vbt *v) /* there's features depending on version! */ v->header.version = 155; v->header.header_size = sizeof(v->header); - v->header.vbt_size = sizeof(struct vbt) - sizeof(v->header); + v->header.vbt_size = sizeof(struct vbt); v->header.bdb_offset = offsetof(struct vbt, bdb_header); strcpy(&v->bdb_header.signature[0], "BIOS_DATA_BLOCK"); v->bdb_header.version = 186; /* child_dev_size = 33 */ v->bdb_header.header_size = sizeof(v->bdb_header); - v->bdb_header.bdb_size = sizeof(struct vbt) - sizeof(struct vbt_header) - - sizeof(struct bdb_header); + v->bdb_header.bdb_size = sizeof(struct vbt) - sizeof(struct vbt_header); /* general features */ v->general_features_header.id = BDB_GENERAL_FEATURES; diff --git a/drivers/gpu/drm/i915/gvt/scheduler.c b/drivers/gpu/drm/i915/gvt/scheduler.c index 1c95bf8cbed0..cb11c3184085 100644 --- a/drivers/gpu/drm/i915/gvt/scheduler.c +++ b/drivers/gpu/drm/i915/gvt/scheduler.c @@ -296,8 +296,8 @@ shadow_context_descriptor_update(struct intel_context *ce, * Update bits 0-11 of the context descriptor which includes flags * like GEN8_CTX_* cached in desc_template */ - desc &= ~(0x3 << GEN8_CTX_ADDRESSING_MODE_SHIFT); - desc |= workload->ctx_desc.addressing_mode << + desc &= ~(0x3ull << GEN8_CTX_ADDRESSING_MODE_SHIFT); + desc |= (u64)workload->ctx_desc.addressing_mode << GEN8_CTX_ADDRESSING_MODE_SHIFT; ce->lrc_desc = desc; diff --git a/drivers/gpu/drm/i915/gvt/vgpu.c b/drivers/gpu/drm/i915/gvt/vgpu.c index abcde8ce1a9a..1d5ff88078bd 100644 --- a/drivers/gpu/drm/i915/gvt/vgpu.c +++ b/drivers/gpu/drm/i915/gvt/vgpu.c @@ -274,10 +274,17 @@ void intel_gvt_destroy_vgpu(struct intel_vgpu *vgpu) struct intel_gvt *gvt = vgpu->gvt; struct drm_i915_private *i915 = gvt->gt->i915; - mutex_lock(&vgpu->vgpu_lock); - drm_WARN(&i915->drm, vgpu->active, "vGPU is still active!\n"); + /* + * remove idr first so later clean can judge if need to stop + * service if no active vgpu. + */ + mutex_lock(&gvt->lock); + idr_remove(&gvt->vgpu_idr, vgpu->id); + mutex_unlock(&gvt->lock); + + mutex_lock(&vgpu->vgpu_lock); intel_gvt_debugfs_remove_vgpu(vgpu); intel_vgpu_clean_sched_policy(vgpu); intel_vgpu_clean_submission(vgpu); @@ -292,7 +299,6 @@ void intel_gvt_destroy_vgpu(struct intel_vgpu *vgpu) mutex_unlock(&vgpu->vgpu_lock); mutex_lock(&gvt->lock); - idr_remove(&gvt->vgpu_idr, vgpu->id); if (idr_is_empty(&gvt->vgpu_idr)) intel_gvt_clean_irq(gvt); intel_gvt_update_vgpu_types(gvt); @@ -560,9 +566,9 @@ void intel_gvt_reset_vgpu_locked(struct intel_vgpu *vgpu, bool dmlr, intel_vgpu_reset_mmio(vgpu, dmlr); populate_pvinfo_page(vgpu); - intel_vgpu_reset_display(vgpu); if (dmlr) { + intel_vgpu_reset_display(vgpu); intel_vgpu_reset_cfg_space(vgpu); /* only reset the failsafe mode when dmlr reset */ vgpu->failsafe = false; diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c index 265a9df153c8..9ab4ad7ccac9 100644 --- a/drivers/gpu/drm/i915/i915_drv.c +++ b/drivers/gpu/drm/i915/i915_drv.c @@ -431,8 +431,7 @@ static int i915_driver_early_probe(struct drm_i915_private *dev_priv) mutex_init(&dev_priv->backlight_lock); mutex_init(&dev_priv->sb_lock); - pm_qos_add_request(&dev_priv->sb_qos, - PM_QOS_CPU_DMA_LATENCY, PM_QOS_DEFAULT_VALUE); + cpu_latency_qos_add_request(&dev_priv->sb_qos, PM_QOS_DEFAULT_VALUE); mutex_init(&dev_priv->av_mutex); mutex_init(&dev_priv->wm.wm_mutex); @@ -496,7 +495,7 @@ static void i915_driver_late_release(struct drm_i915_private *dev_priv) vlv_suspend_cleanup(dev_priv); i915_workqueues_cleanup(dev_priv); - pm_qos_remove_request(&dev_priv->sb_qos); + cpu_latency_qos_remove_request(&dev_priv->sb_qos); mutex_destroy(&dev_priv->sb_lock); } @@ -673,8 +672,7 @@ static int i915_driver_hw_probe(struct drm_i915_private *dev_priv) } } - pm_qos_add_request(&dev_priv->pm_qos, PM_QOS_CPU_DMA_LATENCY, - PM_QOS_DEFAULT_VALUE); + cpu_latency_qos_add_request(&dev_priv->pm_qos, PM_QOS_DEFAULT_VALUE); intel_gt_init_workarounds(dev_priv); @@ -720,7 +718,7 @@ static int i915_driver_hw_probe(struct drm_i915_private *dev_priv) err_msi: if (pdev->msi_enabled) pci_disable_msi(pdev); - pm_qos_remove_request(&dev_priv->pm_qos); + cpu_latency_qos_remove_request(&dev_priv->pm_qos); err_mem_regions: intel_memory_regions_driver_release(dev_priv); err_ggtt: @@ -743,7 +741,7 @@ static void i915_driver_hw_remove(struct drm_i915_private *dev_priv) if (pdev->msi_enabled) pci_disable_msi(pdev); - pm_qos_remove_request(&dev_priv->pm_qos); + cpu_latency_qos_remove_request(&dev_priv->pm_qos); } /** @@ -1846,9 +1844,6 @@ static struct drm_driver driver = { .gem_prime_export = i915_gem_prime_export, .gem_prime_import = i915_gem_prime_import, - .get_vblank_timestamp = drm_calc_vbltimestamp_from_scanoutpos, - .get_scanout_position = i915_get_crtc_scanoutpos, - .dumb_create = i915_gem_dumb_create, .dumb_map_offset = i915_gem_dumb_mmap_offset, diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index adb21112df78..1502ab44f1a5 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -779,13 +779,15 @@ static int __intel_get_crtc_scanline(struct intel_crtc *crtc) return (position + crtc->scanline_offset) % vtotal; } -bool i915_get_crtc_scanoutpos(struct drm_device *dev, unsigned int index, - bool in_vblank_irq, int *vpos, int *hpos, - ktime_t *stime, ktime_t *etime, - const struct drm_display_mode *mode) +static bool i915_get_crtc_scanoutpos(struct drm_crtc *_crtc, + bool in_vblank_irq, + int *vpos, int *hpos, + ktime_t *stime, ktime_t *etime, + const struct drm_display_mode *mode) { + struct drm_device *dev = _crtc->dev; struct drm_i915_private *dev_priv = to_i915(dev); - struct intel_crtc *crtc = to_intel_crtc(drm_crtc_from_index(dev, index)); + struct intel_crtc *crtc = to_intel_crtc(_crtc); enum pipe pipe = crtc->pipe; int position; int vbl_start, vbl_end, hsync_start, htotal, vtotal; @@ -897,6 +899,14 @@ bool i915_get_crtc_scanoutpos(struct drm_device *dev, unsigned int index, return true; } +bool intel_crtc_get_vblank_timestamp(struct drm_crtc *crtc, int *max_error, + ktime_t *vblank_time, bool in_vblank_irq) +{ + return drm_crtc_vblank_helper_get_vblank_timestamp_internal( + crtc, max_error, vblank_time, in_vblank_irq, + i915_get_crtc_scanoutpos); +} + int intel_get_crtc_scanline(struct intel_crtc *crtc) { struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); diff --git a/drivers/gpu/drm/i915/i915_irq.h b/drivers/gpu/drm/i915/i915_irq.h index 812c47a9c2d6..25f25cd95818 100644 --- a/drivers/gpu/drm/i915/i915_irq.h +++ b/drivers/gpu/drm/i915/i915_irq.h @@ -101,10 +101,8 @@ void gen8_irq_power_well_post_enable(struct drm_i915_private *dev_priv, void gen8_irq_power_well_pre_disable(struct drm_i915_private *dev_priv, u8 pipe_mask); -bool i915_get_crtc_scanoutpos(struct drm_device *dev, unsigned int pipe, - bool in_vblank_irq, int *vpos, int *hpos, - ktime_t *stime, ktime_t *etime, - const struct drm_display_mode *mode); +bool intel_crtc_get_vblank_timestamp(struct drm_crtc *crtc, int *max_error, + ktime_t *vblank_time, bool in_vblank_irq); u32 i915_get_vblank_counter(struct drm_crtc *crtc); u32 g4x_get_vblank_counter(struct drm_crtc *crtc); diff --git a/drivers/gpu/drm/i915/i915_memcpy.c b/drivers/gpu/drm/i915/i915_memcpy.c index fdd550405fd3..7b3b83bd5ab8 100644 --- a/drivers/gpu/drm/i915/i915_memcpy.c +++ b/drivers/gpu/drm/i915/i915_memcpy.c @@ -35,7 +35,6 @@ static DEFINE_STATIC_KEY_FALSE(has_movntdqa); -#ifdef CONFIG_AS_MOVNTDQA static void __memcpy_ntdqa(void *dst, const void *src, unsigned long len) { kernel_fpu_begin(); @@ -93,10 +92,6 @@ static void __memcpy_ntdqu(void *dst, const void *src, unsigned long len) kernel_fpu_end(); } -#else -static void __memcpy_ntdqa(void *dst, const void *src, unsigned long len) {} -static void __memcpy_ntdqu(void *dst, const void *src, unsigned long len) {} -#endif /** * i915_memcpy_from_wc: perform an accelerated *aligned* read from WC diff --git a/drivers/gpu/drm/i915/intel_sideband.c b/drivers/gpu/drm/i915/intel_sideband.c index 1447e7516cb7..3f13baaef058 100644 --- a/drivers/gpu/drm/i915/intel_sideband.c +++ b/drivers/gpu/drm/i915/intel_sideband.c @@ -60,7 +60,7 @@ static void __vlv_punit_get(struct drm_i915_private *i915) * to the Valleyview P-unit and not all sideband communications. */ if (IS_VALLEYVIEW(i915)) { - pm_qos_update_request(&i915->sb_qos, 0); + cpu_latency_qos_update_request(&i915->sb_qos, 0); on_each_cpu(ping, NULL, 1); } } @@ -68,7 +68,8 @@ static void __vlv_punit_get(struct drm_i915_private *i915) static void __vlv_punit_put(struct drm_i915_private *i915) { if (IS_VALLEYVIEW(i915)) - pm_qos_update_request(&i915->sb_qos, PM_QOS_DEFAULT_VALUE); + cpu_latency_qos_update_request(&i915->sb_qos, + PM_QOS_DEFAULT_VALUE); iosf_mbi_punit_release(); } diff --git a/drivers/gpu/drm/imx/imx-ldb.c b/drivers/gpu/drm/imx/imx-ldb.c index 8cb2665b2c74..4da22a94790c 100644 --- a/drivers/gpu/drm/imx/imx-ldb.c +++ b/drivers/gpu/drm/imx/imx-ldb.c @@ -446,7 +446,7 @@ static int imx_ldb_register(struct drm_device *drm, if (imx_ldb_ch->bridge) { ret = drm_bridge_attach(&imx_ldb_ch->encoder, - imx_ldb_ch->bridge, NULL); + imx_ldb_ch->bridge, NULL, 0); if (ret) { DRM_ERROR("Failed to initialize bridge with drm\n"); return ret; diff --git a/drivers/gpu/drm/imx/parallel-display.c b/drivers/gpu/drm/imx/parallel-display.c index 142acff5ab37..08fafa4bf8c2 100644 --- a/drivers/gpu/drm/imx/parallel-display.c +++ b/drivers/gpu/drm/imx/parallel-display.c @@ -292,7 +292,7 @@ static int imx_pd_register(struct drm_device *drm, DRM_MODE_ENCODER_NONE, NULL); imxpd->bridge.funcs = &imx_pd_bridge_funcs; - drm_bridge_attach(encoder, &imxpd->bridge, NULL); + drm_bridge_attach(encoder, &imxpd->bridge, NULL, 0); if (!imxpd->next_bridge) { drm_connector_helper_add(&imxpd->connector, @@ -307,7 +307,7 @@ static int imx_pd_register(struct drm_device *drm, if (imxpd->next_bridge) { ret = drm_bridge_attach(encoder, imxpd->next_bridge, - &imxpd->bridge); + &imxpd->bridge, 0); if (ret < 0) { dev_err(imxpd->dev, "failed to attach bridge: %d\n", ret); diff --git a/drivers/gpu/drm/ingenic/ingenic-drm.c b/drivers/gpu/drm/ingenic/ingenic-drm.c index 6d47ef7b148c..9dfe7cb530e1 100644 --- a/drivers/gpu/drm/ingenic/ingenic-drm.c +++ b/drivers/gpu/drm/ingenic/ingenic-drm.c @@ -737,7 +737,7 @@ static int ingenic_drm_probe(struct platform_device *pdev) return ret; } - ret = drm_bridge_attach(&priv->encoder, bridge, NULL); + ret = drm_bridge_attach(&priv->encoder, bridge, NULL, 0); if (ret) { dev_err(dev, "Unable to attach bridge"); return ret; diff --git a/drivers/gpu/drm/lima/lima_gp.c b/drivers/gpu/drm/lima/lima_gp.c index 52b210f9a605..d8841c870d90 100644 --- a/drivers/gpu/drm/lima/lima_gp.c +++ b/drivers/gpu/drm/lima/lima_gp.c @@ -231,8 +231,13 @@ static int lima_gp_task_recover(struct lima_sched_pipe *pipe) } gp_write(LIMA_GP_INT_MASK, LIMA_GP_IRQ_MASK_USED); + /* Resume from where we stopped, i.e. new start is old end */ + gp_write(LIMA_GP_PLBU_ALLOC_START_ADDR, + f[LIMA_GP_PLBU_ALLOC_END_ADDR >> 2]); + f[LIMA_GP_PLBU_ALLOC_END_ADDR >> 2] = + f[LIMA_GP_PLBU_ALLOC_START_ADDR >> 2] + task->heap->heap_size; gp_write(LIMA_GP_PLBU_ALLOC_END_ADDR, - f[LIMA_GP_PLBU_ALLOC_START_ADDR >> 2] + task->heap->heap_size); + f[LIMA_GP_PLBU_ALLOC_END_ADDR >> 2]); gp_write(LIMA_GP_CMD, LIMA_GP_CMD_UPDATE_PLBU_ALLOC); return 0; } diff --git a/drivers/gpu/drm/mcde/mcde_drv.c b/drivers/gpu/drm/mcde/mcde_drv.c index 9008ddcfc528..f28cb7a576ba 100644 --- a/drivers/gpu/drm/mcde/mcde_drv.c +++ b/drivers/gpu/drm/mcde/mcde_drv.c @@ -20,11 +20,11 @@ * input formats including most variants of RGB and YUV. * * The hardware has four display pipes, and the layout is a little - * bit like this: + * bit like this:: * - * Memory -> Overlay -> Channel -> FIFO -> 5 formatters -> DSI/DPI - * External 0..5 0..3 A,B, 3 x DSI bridge - * source 0..9 C0,C1 2 x DPI + * Memory -> Overlay -> Channel -> FIFO -> 5 formatters -> DSI/DPI + * External 0..5 0..3 A,B, 3 x DSI bridge + * source 0..9 C0,C1 2 x DPI * * FIFOs A and B are for LCD and HDMI while FIFO CO/C1 are for * panels with embedded buffer. @@ -43,6 +43,7 @@ * to change as we exploit more of the hardware capabilities. * * TODO: + * * - Enabled damaged rectangles using drm_plane_enable_fb_damage_clips() * so we can selectively just transmit the damaged area to a * command-only display. diff --git a/drivers/gpu/drm/mcde/mcde_dsi.c b/drivers/gpu/drm/mcde/mcde_dsi.c index bb6528b01cd0..7af5ebb0c436 100644 --- a/drivers/gpu/drm/mcde/mcde_dsi.c +++ b/drivers/gpu/drm/mcde/mcde_dsi.c @@ -986,7 +986,8 @@ static void mcde_dsi_bridge_disable(struct drm_bridge *bridge) clk_disable_unprepare(d->lp_clk); } -static int mcde_dsi_bridge_attach(struct drm_bridge *bridge) +static int mcde_dsi_bridge_attach(struct drm_bridge *bridge, + enum drm_bridge_attach_flags flags) { struct mcde_dsi *d = bridge_to_mcde_dsi(bridge); struct drm_device *drm = bridge->dev; @@ -998,7 +999,7 @@ static int mcde_dsi_bridge_attach(struct drm_bridge *bridge) } /* Attach the DSI bridge to the output (panel etc) bridge */ - ret = drm_bridge_attach(bridge->encoder, d->bridge_out, bridge); + ret = drm_bridge_attach(bridge->encoder, d->bridge_out, bridge, flags); if (ret) { dev_err(d->dev, "failed to attach the DSI bridge\n"); return ret; diff --git a/drivers/gpu/drm/mediatek/mtk_dpi.c b/drivers/gpu/drm/mediatek/mtk_dpi.c index 01fa8b8d763d..4f0ce4cd5b8c 100644 --- a/drivers/gpu/drm/mediatek/mtk_dpi.c +++ b/drivers/gpu/drm/mediatek/mtk_dpi.c @@ -607,7 +607,7 @@ static int mtk_dpi_bind(struct device *dev, struct device *master, void *data) /* Currently DPI0 is fixed to be driven by OVL1 */ dpi->encoder.possible_crtcs = BIT(1); - ret = drm_bridge_attach(&dpi->encoder, dpi->bridge, NULL); + ret = drm_bridge_attach(&dpi->encoder, dpi->bridge, NULL, 0); if (ret) { dev_err(dev, "Failed to attach bridge: %d\n", ret); goto err_cleanup; @@ -664,6 +664,16 @@ static unsigned int mt2701_calculate_factor(int clock) return 1; } +static unsigned int mt8183_calculate_factor(int clock) +{ + if (clock <= 27000) + return 8; + else if (clock <= 167000) + return 4; + else + return 2; +} + static const struct mtk_dpi_conf mt8173_conf = { .cal_factor = mt8173_calculate_factor, .reg_h_fre_con = 0xe0, @@ -675,6 +685,11 @@ static const struct mtk_dpi_conf mt2701_conf = { .edge_sel_en = true, }; +static const struct mtk_dpi_conf mt8183_conf = { + .cal_factor = mt8183_calculate_factor, + .reg_h_fre_con = 0xe0, +}; + static int mtk_dpi_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; @@ -770,6 +785,9 @@ static const struct of_device_id mtk_dpi_of_ids[] = { { .compatible = "mediatek,mt8173-dpi", .data = &mt8173_conf, }, + { .compatible = "mediatek,mt8183-dpi", + .data = &mt8183_conf, + }, { }, }; diff --git a/drivers/gpu/drm/mediatek/mtk_drm_crtc.c b/drivers/gpu/drm/mediatek/mtk_drm_crtc.c index 0dfcd1787e65..fe85e487e477 100644 --- a/drivers/gpu/drm/mediatek/mtk_drm_crtc.c +++ b/drivers/gpu/drm/mediatek/mtk_drm_crtc.c @@ -486,6 +486,7 @@ static void mtk_drm_crtc_hw_config(struct mtk_drm_crtc *mtk_crtc) } #if IS_REACHABLE(CONFIG_MTK_CMDQ) if (mtk_crtc->cmdq_client) { + mbox_flush(mtk_crtc->cmdq_client->chan, 2000); cmdq_handle = cmdq_pkt_create(mtk_crtc->cmdq_client, PAGE_SIZE); cmdq_pkt_clear_event(cmdq_handle, mtk_crtc->cmdq_event); cmdq_pkt_wfe(cmdq_handle, mtk_crtc->cmdq_event); @@ -636,10 +637,18 @@ static const struct drm_crtc_helper_funcs mtk_crtc_helper_funcs = { static int mtk_drm_crtc_init(struct drm_device *drm, struct mtk_drm_crtc *mtk_crtc, - struct drm_plane *primary, - struct drm_plane *cursor, unsigned int pipe) + unsigned int pipe) { - int ret; + struct drm_plane *primary = NULL; + struct drm_plane *cursor = NULL; + int i, ret; + + for (i = 0; i < mtk_crtc->layer_nr; i++) { + if (mtk_crtc->planes[i].type == DRM_PLANE_TYPE_PRIMARY) + primary = &mtk_crtc->planes[i]; + else if (mtk_crtc->planes[i].type == DRM_PLANE_TYPE_CURSOR) + cursor = &mtk_crtc->planes[i]; + } ret = drm_crtc_init_with_planes(drm, &mtk_crtc->base, primary, cursor, &mtk_crtc_funcs, NULL); @@ -689,11 +698,12 @@ static int mtk_drm_crtc_num_comp_planes(struct mtk_drm_crtc *mtk_crtc, } static inline -enum drm_plane_type mtk_drm_crtc_plane_type(unsigned int plane_idx) +enum drm_plane_type mtk_drm_crtc_plane_type(unsigned int plane_idx, + unsigned int num_planes) { if (plane_idx == 0) return DRM_PLANE_TYPE_PRIMARY; - else if (plane_idx == 1) + else if (plane_idx == (num_planes - 1)) return DRM_PLANE_TYPE_CURSOR; else return DRM_PLANE_TYPE_OVERLAY; @@ -712,7 +722,8 @@ static int mtk_drm_crtc_init_comp_planes(struct drm_device *drm_dev, ret = mtk_plane_init(drm_dev, &mtk_crtc->planes[mtk_crtc->layer_nr], BIT(pipe), - mtk_drm_crtc_plane_type(mtk_crtc->layer_nr), + mtk_drm_crtc_plane_type(mtk_crtc->layer_nr, + num_planes), mtk_ddp_comp_supported_rotations(comp)); if (ret) return ret; @@ -807,9 +818,7 @@ int mtk_drm_crtc_create(struct drm_device *drm_dev, return ret; } - ret = mtk_drm_crtc_init(drm_dev, mtk_crtc, &mtk_crtc->planes[0], - mtk_crtc->layer_nr > 1 ? &mtk_crtc->planes[1] : - NULL, pipe); + ret = mtk_drm_crtc_init(drm_dev, mtk_crtc, pipe); if (ret < 0) return ret; @@ -828,7 +837,8 @@ int mtk_drm_crtc_create(struct drm_device *drm_dev, drm_crtc_index(&mtk_crtc->base)); mtk_crtc->cmdq_client = NULL; } - ret = of_property_read_u32_index(dev->of_node, "mediatek,gce-events", + ret = of_property_read_u32_index(priv->mutex_node, + "mediatek,gce-events", drm_crtc_index(&mtk_crtc->base), &mtk_crtc->cmdq_event); if (ret) diff --git a/drivers/gpu/drm/mediatek/mtk_drm_ddp_comp.c b/drivers/gpu/drm/mediatek/mtk_drm_ddp_comp.c index 1f5a112bb034..57c88de9a329 100644 --- a/drivers/gpu/drm/mediatek/mtk_drm_ddp_comp.c +++ b/drivers/gpu/drm/mediatek/mtk_drm_ddp_comp.c @@ -471,6 +471,7 @@ int mtk_ddp_comp_init(struct device *dev, struct device_node *node, /* Only DMA capable components need the LARB property */ comp->larb_dev = NULL; if (type != MTK_DISP_OVL && + type != MTK_DISP_OVL_2L && type != MTK_DISP_RDMA && type != MTK_DISP_WDMA) return 0; diff --git a/drivers/gpu/drm/mediatek/mtk_drm_plane.c b/drivers/gpu/drm/mediatek/mtk_drm_plane.c index 914cc7619cd7..c2bd683a87c8 100644 --- a/drivers/gpu/drm/mediatek/mtk_drm_plane.c +++ b/drivers/gpu/drm/mediatek/mtk_drm_plane.c @@ -80,6 +80,7 @@ static int mtk_plane_atomic_async_check(struct drm_plane *plane, struct drm_plane_state *state) { struct drm_crtc_state *crtc_state; + int ret; if (plane != state->crtc->cursor) return -EINVAL; @@ -90,6 +91,11 @@ static int mtk_plane_atomic_async_check(struct drm_plane *plane, if (!plane->state->fb) return -EINVAL; + ret = mtk_drm_crtc_plane_check(state->crtc, plane, + to_mtk_plane_state(state)); + if (ret) + return ret; + if (state->state) crtc_state = drm_atomic_get_existing_crtc_state(state->state, state->crtc); @@ -115,6 +121,7 @@ static void mtk_plane_atomic_async_update(struct drm_plane *plane, plane->state->src_y = new_state->src_y; plane->state->src_h = new_state->src_h; plane->state->src_w = new_state->src_w; + swap(plane->state->fb, new_state->fb); state->pending.async_dirty = true; mtk_drm_crtc_async_update(new_state->crtc, plane, new_state); diff --git a/drivers/gpu/drm/mediatek/mtk_dsi.c b/drivers/gpu/drm/mediatek/mtk_dsi.c index 5fa1073cf26b..0ede69830a9d 100644 --- a/drivers/gpu/drm/mediatek/mtk_dsi.c +++ b/drivers/gpu/drm/mediatek/mtk_dsi.c @@ -904,7 +904,7 @@ static int mtk_dsi_create_conn_enc(struct drm_device *drm, struct mtk_dsi *dsi) /* If there's a bridge, attach to it and let it create the connector */ if (dsi->bridge) { - ret = drm_bridge_attach(&dsi->encoder, dsi->bridge, NULL); + ret = drm_bridge_attach(&dsi->encoder, dsi->bridge, NULL, 0); if (ret) { DRM_ERROR("Failed to attach bridge to drm\n"); goto err_encoder_cleanup; diff --git a/drivers/gpu/drm/mediatek/mtk_hdmi.c b/drivers/gpu/drm/mediatek/mtk_hdmi.c index 5e4a4dbda443..ff43a3d80410 100644 --- a/drivers/gpu/drm/mediatek/mtk_hdmi.c +++ b/drivers/gpu/drm/mediatek/mtk_hdmi.c @@ -12,6 +12,7 @@ #include <linux/io.h> #include <linux/kernel.h> #include <linux/mfd/syscon.h> +#include <linux/mutex.h> #include <linux/of_platform.h> #include <linux/of.h> #include <linux/of_gpio.h> @@ -169,6 +170,9 @@ struct mtk_hdmi { bool audio_enable; bool powered; bool enabled; + hdmi_codec_plugged_cb plugged_cb; + struct device *codec_dev; + struct mutex update_plugged_status_lock; }; static inline struct mtk_hdmi *hdmi_ctx_from_bridge(struct drm_bridge *b) @@ -1194,13 +1198,26 @@ static void mtk_hdmi_clk_disable_audio(struct mtk_hdmi *hdmi) clk_disable_unprepare(hdmi->clk[MTK_HDMI_CLK_AUD_SPDIF]); } +static enum drm_connector_status +mtk_hdmi_update_plugged_status(struct mtk_hdmi *hdmi) +{ + bool connected; + + mutex_lock(&hdmi->update_plugged_status_lock); + connected = mtk_cec_hpd_high(hdmi->cec_dev); + if (hdmi->plugged_cb && hdmi->codec_dev) + hdmi->plugged_cb(hdmi->codec_dev, connected); + mutex_unlock(&hdmi->update_plugged_status_lock); + + return connected ? + connector_status_connected : connector_status_disconnected; +} + static enum drm_connector_status hdmi_conn_detect(struct drm_connector *conn, bool force) { struct mtk_hdmi *hdmi = hdmi_ctx_from_conn(conn); - - return mtk_cec_hpd_high(hdmi->cec_dev) ? - connector_status_connected : connector_status_disconnected; + return mtk_hdmi_update_plugged_status(hdmi); } static void hdmi_conn_destroy(struct drm_connector *conn) @@ -1297,11 +1314,17 @@ static void mtk_hdmi_hpd_event(bool hpd, struct device *dev) * Bridge callbacks */ -static int mtk_hdmi_bridge_attach(struct drm_bridge *bridge) +static int mtk_hdmi_bridge_attach(struct drm_bridge *bridge, + enum drm_bridge_attach_flags flags) { struct mtk_hdmi *hdmi = hdmi_ctx_from_bridge(bridge); int ret; + if (flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR) { + DRM_ERROR("Fix bridge driver to make connector optional!"); + return -EINVAL; + } + ret = drm_connector_init_with_ddc(bridge->encoder->dev, &hdmi->conn, &mtk_hdmi_connector_funcs, DRM_MODE_CONNECTOR_HDMIA, @@ -1326,7 +1349,7 @@ static int mtk_hdmi_bridge_attach(struct drm_bridge *bridge) if (hdmi->next_bridge) { ret = drm_bridge_attach(bridge->encoder, hdmi->next_bridge, - bridge); + bridge, flags); if (ret) { dev_err(hdmi->dev, "Failed to attach external bridge: %d\n", ret); @@ -1651,20 +1674,39 @@ static int mtk_hdmi_audio_get_eld(struct device *dev, void *data, uint8_t *buf, return 0; } +static int mtk_hdmi_audio_hook_plugged_cb(struct device *dev, void *data, + hdmi_codec_plugged_cb fn, + struct device *codec_dev) +{ + struct mtk_hdmi *hdmi = data; + + mutex_lock(&hdmi->update_plugged_status_lock); + hdmi->plugged_cb = fn; + hdmi->codec_dev = codec_dev; + mutex_unlock(&hdmi->update_plugged_status_lock); + + mtk_hdmi_update_plugged_status(hdmi); + + return 0; +} + static const struct hdmi_codec_ops mtk_hdmi_audio_codec_ops = { .hw_params = mtk_hdmi_audio_hw_params, .audio_startup = mtk_hdmi_audio_startup, .audio_shutdown = mtk_hdmi_audio_shutdown, .digital_mute = mtk_hdmi_audio_digital_mute, .get_eld = mtk_hdmi_audio_get_eld, + .hook_plugged_cb = mtk_hdmi_audio_hook_plugged_cb, }; -static void mtk_hdmi_register_audio_driver(struct device *dev) +static int mtk_hdmi_register_audio_driver(struct device *dev) { + struct mtk_hdmi *hdmi = dev_get_drvdata(dev); struct hdmi_codec_pdata codec_data = { .ops = &mtk_hdmi_audio_codec_ops, .max_i2s_channels = 2, .i2s = 1, + .data = hdmi, }; struct platform_device *pdev; @@ -1672,9 +1714,10 @@ static void mtk_hdmi_register_audio_driver(struct device *dev) PLATFORM_DEVID_AUTO, &codec_data, sizeof(codec_data)); if (IS_ERR(pdev)) - return; + return PTR_ERR(pdev); DRM_INFO("%s driver bound to HDMI\n", HDMI_CODEC_DRV_NAME); + return 0; } static int mtk_drm_hdmi_probe(struct platform_device *pdev) @@ -1700,6 +1743,7 @@ static int mtk_drm_hdmi_probe(struct platform_device *pdev) return ret; } + mutex_init(&hdmi->update_plugged_status_lock); platform_set_drvdata(pdev, hdmi); ret = mtk_hdmi_output_init(hdmi); @@ -1708,7 +1752,11 @@ static int mtk_drm_hdmi_probe(struct platform_device *pdev) return ret; } - mtk_hdmi_register_audio_driver(dev); + ret = mtk_hdmi_register_audio_driver(dev); + if (ret) { + dev_err(dev, "Failed to register audio driver: %d\n", ret); + return ret; + } hdmi->bridge.funcs = &mtk_hdmi_bridge_funcs; hdmi->bridge.of_node = pdev->dev.of_node; diff --git a/drivers/gpu/drm/meson/meson_dw_hdmi.c b/drivers/gpu/drm/meson/meson_dw_hdmi.c index 3bb7ffe5fc39..e8c94915a4fc 100644 --- a/drivers/gpu/drm/meson/meson_dw_hdmi.c +++ b/drivers/gpu/drm/meson/meson_dw_hdmi.c @@ -16,6 +16,7 @@ #include <drm/bridge/dw_hdmi.h> #include <drm/drm_atomic_helper.h> +#include <drm/drm_bridge.h> #include <drm/drm_device.h> #include <drm/drm_edid.h> #include <drm/drm_probe_helper.h> @@ -135,6 +136,7 @@ struct meson_dw_hdmi_data { struct meson_dw_hdmi { struct drm_encoder encoder; + struct drm_bridge bridge; struct dw_hdmi_plat_data dw_plat_data; struct meson_drm *priv; struct device *dev; @@ -148,9 +150,12 @@ struct meson_dw_hdmi { struct regulator *hdmi_supply; u32 irq_stat; struct dw_hdmi *hdmi; + unsigned long output_bus_fmt; }; #define encoder_to_meson_dw_hdmi(x) \ container_of(x, struct meson_dw_hdmi, encoder) +#define bridge_to_meson_dw_hdmi(x) \ + container_of(x, struct meson_dw_hdmi, bridge) static inline int dw_hdmi_is_compatible(struct meson_dw_hdmi *dw_hdmi, const char *compat) @@ -297,6 +302,10 @@ static void meson_hdmi_phy_setup_mode(struct meson_dw_hdmi *dw_hdmi, struct meson_drm *priv = dw_hdmi->priv; unsigned int pixel_clock = mode->clock; + /* For 420, pixel clock is half unlike venc clock */ + if (dw_hdmi->output_bus_fmt == MEDIA_BUS_FMT_UYYVYY8_0_5X24) + pixel_clock /= 2; + if (dw_hdmi_is_compatible(dw_hdmi, "amlogic,meson-gxl-dw-hdmi") || dw_hdmi_is_compatible(dw_hdmi, "amlogic,meson-gxm-dw-hdmi")) { if (pixel_clock >= 371250) { @@ -368,29 +377,40 @@ static inline void meson_dw_hdmi_phy_reset(struct meson_dw_hdmi *dw_hdmi) } static void dw_hdmi_set_vclk(struct meson_dw_hdmi *dw_hdmi, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct meson_drm *priv = dw_hdmi->priv; int vic = drm_match_cea_mode(mode); + unsigned int phy_freq; unsigned int vclk_freq; unsigned int venc_freq; unsigned int hdmi_freq; vclk_freq = mode->clock; + /* For 420, pixel clock is half unlike venc clock */ + if (dw_hdmi->output_bus_fmt == MEDIA_BUS_FMT_UYYVYY8_0_5X24) + vclk_freq /= 2; + + /* TMDS clock is pixel_clock * 10 */ + phy_freq = vclk_freq * 10; + if (!vic) { - meson_vclk_setup(priv, MESON_VCLK_TARGET_DMT, vclk_freq, - vclk_freq, vclk_freq, false); + meson_vclk_setup(priv, MESON_VCLK_TARGET_DMT, phy_freq, + vclk_freq, vclk_freq, vclk_freq, false); return; } + /* 480i/576i needs global pixel doubling */ if (mode->flags & DRM_MODE_FLAG_DBLCLK) vclk_freq *= 2; venc_freq = vclk_freq; hdmi_freq = vclk_freq; - if (meson_venc_hdmi_venc_repeat(vic)) + /* VENC double pixels for 1080i, 720p and YUV420 modes */ + if (meson_venc_hdmi_venc_repeat(vic) || + dw_hdmi->output_bus_fmt == MEDIA_BUS_FMT_UYYVYY8_0_5X24) venc_freq *= 2; vclk_freq = max(venc_freq, hdmi_freq); @@ -398,11 +418,11 @@ static void dw_hdmi_set_vclk(struct meson_dw_hdmi *dw_hdmi, if (mode->flags & DRM_MODE_FLAG_DBLCLK) venc_freq /= 2; - DRM_DEBUG_DRIVER("vclk:%d venc=%d hdmi=%d enci=%d\n", - vclk_freq, venc_freq, hdmi_freq, + DRM_DEBUG_DRIVER("vclk:%d phy=%d venc=%d hdmi=%d enci=%d\n", + phy_freq, vclk_freq, venc_freq, hdmi_freq, priv->venc.hdmi_use_enci); - meson_vclk_setup(priv, MESON_VCLK_TARGET_HDMI, vclk_freq, + meson_vclk_setup(priv, MESON_VCLK_TARGET_HDMI, phy_freq, vclk_freq, venc_freq, hdmi_freq, priv->venc.hdmi_use_enci); } @@ -437,8 +457,9 @@ static int dw_hdmi_phy_init(struct dw_hdmi *hdmi, void *data, /* Enable normal output to PHY */ dw_hdmi->data->top_write(dw_hdmi, HDMITX_TOP_BIST_CNTL, BIT(12)); - /* TMDS pattern setup (TOFIX Handle the YUV420 case) */ - if (mode->clock > 340000) { + /* TMDS pattern setup */ + if (mode->clock > 340000 && + dw_hdmi->output_bus_fmt == MEDIA_BUS_FMT_YUV8_1X24) { dw_hdmi->data->top_write(dw_hdmi, HDMITX_TOP_TMDS_CLK_PTTN_01, 0); dw_hdmi->data->top_write(dw_hdmi, HDMITX_TOP_TMDS_CLK_PTTN_23, @@ -613,6 +634,8 @@ dw_hdmi_mode_valid(struct drm_connector *connector, const struct drm_display_mode *mode) { struct meson_drm *priv = connector->dev->dev_private; + bool is_hdmi2_sink = connector->display_info.hdmi.scdc.supported; + unsigned int phy_freq; unsigned int vclk_freq; unsigned int venc_freq; unsigned int hdmi_freq; @@ -621,9 +644,11 @@ dw_hdmi_mode_valid(struct drm_connector *connector, DRM_DEBUG_DRIVER("Modeline " DRM_MODE_FMT "\n", DRM_MODE_ARG(mode)); - /* If sink max TMDS clock, we reject the mode */ + /* If sink does not support 540MHz, reject the non-420 HDMI2 modes */ if (connector->display_info.max_tmds_clock && - mode->clock > connector->display_info.max_tmds_clock) + mode->clock > connector->display_info.max_tmds_clock && + !drm_mode_is_420_only(&connector->display_info, mode) && + !drm_mode_is_420_also(&connector->display_info, mode)) return MODE_BAD; /* Check against non-VIC supported modes */ @@ -639,6 +664,15 @@ dw_hdmi_mode_valid(struct drm_connector *connector, vclk_freq = mode->clock; + /* For 420, pixel clock is half unlike venc clock */ + if (drm_mode_is_420_only(&connector->display_info, mode) || + (!is_hdmi2_sink && + drm_mode_is_420_also(&connector->display_info, mode))) + vclk_freq /= 2; + + /* TMDS clock is pixel_clock * 10 */ + phy_freq = vclk_freq * 10; + /* 480i/576i needs global pixel doubling */ if (mode->flags & DRM_MODE_FLAG_DBLCLK) vclk_freq *= 2; @@ -646,8 +680,11 @@ dw_hdmi_mode_valid(struct drm_connector *connector, venc_freq = vclk_freq; hdmi_freq = vclk_freq; - /* VENC double pixels for 1080i and 720p modes */ - if (meson_venc_hdmi_venc_repeat(vic)) + /* VENC double pixels for 1080i, 720p and YUV420 modes */ + if (meson_venc_hdmi_venc_repeat(vic) || + drm_mode_is_420_only(&connector->display_info, mode) || + (!is_hdmi2_sink && + drm_mode_is_420_also(&connector->display_info, mode))) venc_freq *= 2; vclk_freq = max(venc_freq, hdmi_freq); @@ -655,14 +692,19 @@ dw_hdmi_mode_valid(struct drm_connector *connector, if (mode->flags & DRM_MODE_FLAG_DBLCLK) venc_freq /= 2; - dev_dbg(connector->dev->dev, "%s: vclk:%d venc=%d hdmi=%d\n", __func__, - vclk_freq, venc_freq, hdmi_freq); + dev_dbg(connector->dev->dev, "%s: vclk:%d phy=%d venc=%d hdmi=%d\n", + __func__, phy_freq, vclk_freq, venc_freq, hdmi_freq); - return meson_vclk_vic_supported_freq(vclk_freq); + return meson_vclk_vic_supported_freq(phy_freq, vclk_freq); } /* Encoder */ +static const u32 meson_dw_hdmi_out_bus_fmts[] = { + MEDIA_BUS_FMT_YUV8_1X24, + MEDIA_BUS_FMT_UYYVYY8_0_5X24, +}; + static void meson_venc_hdmi_encoder_destroy(struct drm_encoder *encoder) { drm_encoder_cleanup(encoder); @@ -672,16 +714,54 @@ static const struct drm_encoder_funcs meson_venc_hdmi_encoder_funcs = { .destroy = meson_venc_hdmi_encoder_destroy, }; -static int meson_venc_hdmi_encoder_atomic_check(struct drm_encoder *encoder, +static u32 * +meson_venc_hdmi_encoder_get_inp_bus_fmts(struct drm_bridge *bridge, + struct drm_bridge_state *bridge_state, + struct drm_crtc_state *crtc_state, + struct drm_connector_state *conn_state, + u32 output_fmt, + unsigned int *num_input_fmts) +{ + u32 *input_fmts = NULL; + int i; + + *num_input_fmts = 0; + + for (i = 0 ; i < ARRAY_SIZE(meson_dw_hdmi_out_bus_fmts) ; ++i) { + if (output_fmt == meson_dw_hdmi_out_bus_fmts[i]) { + *num_input_fmts = 1; + input_fmts = kcalloc(*num_input_fmts, + sizeof(*input_fmts), + GFP_KERNEL); + if (!input_fmts) + return NULL; + + input_fmts[0] = output_fmt; + + break; + } + } + + return input_fmts; +} + +static int meson_venc_hdmi_encoder_atomic_check(struct drm_bridge *bridge, + struct drm_bridge_state *bridge_state, struct drm_crtc_state *crtc_state, struct drm_connector_state *conn_state) { + struct meson_dw_hdmi *dw_hdmi = bridge_to_meson_dw_hdmi(bridge); + + dw_hdmi->output_bus_fmt = bridge_state->output_bus_cfg.format; + + DRM_DEBUG_DRIVER("output_bus_fmt %lx\n", dw_hdmi->output_bus_fmt); + return 0; } -static void meson_venc_hdmi_encoder_disable(struct drm_encoder *encoder) +static void meson_venc_hdmi_encoder_disable(struct drm_bridge *bridge) { - struct meson_dw_hdmi *dw_hdmi = encoder_to_meson_dw_hdmi(encoder); + struct meson_dw_hdmi *dw_hdmi = bridge_to_meson_dw_hdmi(bridge); struct meson_drm *priv = dw_hdmi->priv; DRM_DEBUG_DRIVER("\n"); @@ -693,9 +773,9 @@ static void meson_venc_hdmi_encoder_disable(struct drm_encoder *encoder) writel_relaxed(0, priv->io_base + _REG(ENCP_VIDEO_EN)); } -static void meson_venc_hdmi_encoder_enable(struct drm_encoder *encoder) +static void meson_venc_hdmi_encoder_enable(struct drm_bridge *bridge) { - struct meson_dw_hdmi *dw_hdmi = encoder_to_meson_dw_hdmi(encoder); + struct meson_dw_hdmi *dw_hdmi = bridge_to_meson_dw_hdmi(bridge); struct meson_drm *priv = dw_hdmi->priv; DRM_DEBUG_DRIVER("%s\n", priv->venc.hdmi_use_enci ? "VENCI" : "VENCP"); @@ -706,32 +786,47 @@ static void meson_venc_hdmi_encoder_enable(struct drm_encoder *encoder) writel_relaxed(1, priv->io_base + _REG(ENCP_VIDEO_EN)); } -static void meson_venc_hdmi_encoder_mode_set(struct drm_encoder *encoder, - struct drm_display_mode *mode, - struct drm_display_mode *adjusted_mode) +static void meson_venc_hdmi_encoder_mode_set(struct drm_bridge *bridge, + const struct drm_display_mode *mode, + const struct drm_display_mode *adjusted_mode) { - struct meson_dw_hdmi *dw_hdmi = encoder_to_meson_dw_hdmi(encoder); + struct meson_dw_hdmi *dw_hdmi = bridge_to_meson_dw_hdmi(bridge); struct meson_drm *priv = dw_hdmi->priv; int vic = drm_match_cea_mode(mode); + unsigned int ycrcb_map = VPU_HDMI_OUTPUT_CBYCR; + bool yuv420_mode = false; DRM_DEBUG_DRIVER("\"%s\" vic %d\n", mode->name, vic); + if (dw_hdmi->output_bus_fmt == MEDIA_BUS_FMT_UYYVYY8_0_5X24) { + ycrcb_map = VPU_HDMI_OUTPUT_CRYCB; + yuv420_mode = true; + } + /* VENC + VENC-DVI Mode setup */ - meson_venc_hdmi_mode_set(priv, vic, mode); + meson_venc_hdmi_mode_set(priv, vic, ycrcb_map, yuv420_mode, mode); /* VCLK Set clock */ dw_hdmi_set_vclk(dw_hdmi, mode); - /* Setup YUV444 to HDMI-TX, no 10bit diphering */ - writel_relaxed(0, priv->io_base + _REG(VPU_HDMI_FMT_CTRL)); + if (dw_hdmi->output_bus_fmt == MEDIA_BUS_FMT_UYYVYY8_0_5X24) + /* Setup YUV420 to HDMI-TX, no 10bit diphering */ + writel_relaxed(2 | (2 << 2), + priv->io_base + _REG(VPU_HDMI_FMT_CTRL)); + else + /* Setup YUV444 to HDMI-TX, no 10bit diphering */ + writel_relaxed(0, priv->io_base + _REG(VPU_HDMI_FMT_CTRL)); } -static const struct drm_encoder_helper_funcs - meson_venc_hdmi_encoder_helper_funcs = { - .atomic_check = meson_venc_hdmi_encoder_atomic_check, - .disable = meson_venc_hdmi_encoder_disable, - .enable = meson_venc_hdmi_encoder_enable, - .mode_set = meson_venc_hdmi_encoder_mode_set, +static const struct drm_bridge_funcs meson_venc_hdmi_encoder_bridge_funcs = { + .atomic_duplicate_state = drm_atomic_helper_bridge_duplicate_state, + .atomic_destroy_state = drm_atomic_helper_bridge_destroy_state, + .atomic_get_input_bus_fmts = meson_venc_hdmi_encoder_get_inp_bus_fmts, + .atomic_reset = drm_atomic_helper_bridge_reset, + .atomic_check = meson_venc_hdmi_encoder_atomic_check, + .enable = meson_venc_hdmi_encoder_enable, + .disable = meson_venc_hdmi_encoder_disable, + .mode_set = meson_venc_hdmi_encoder_mode_set, }; /* DW HDMI Regmap */ @@ -852,6 +947,7 @@ static int meson_dw_hdmi_bind(struct device *dev, struct device *master, struct drm_device *drm = data; struct meson_drm *priv = drm->dev_private; struct dw_hdmi_plat_data *dw_plat_data; + struct drm_bridge *next_bridge; struct drm_encoder *encoder; struct resource *res; int irq; @@ -953,8 +1049,6 @@ static int meson_dw_hdmi_bind(struct device *dev, struct device *master, /* Encoder */ - drm_encoder_helper_add(encoder, &meson_venc_hdmi_encoder_helper_funcs); - ret = drm_encoder_init(drm, encoder, &meson_venc_hdmi_encoder_funcs, DRM_MODE_ENCODER_TMDS, "meson_hdmi"); if (ret) { @@ -962,6 +1056,9 @@ static int meson_dw_hdmi_bind(struct device *dev, struct device *master, return ret; } + meson_dw_hdmi->bridge.funcs = &meson_venc_hdmi_encoder_bridge_funcs; + drm_bridge_attach(encoder, &meson_dw_hdmi->bridge, NULL, 0); + encoder->possible_crtcs = BIT(0); DRM_DEBUG_DRIVER("encoder initialized\n"); @@ -974,8 +1071,8 @@ static int meson_dw_hdmi_bind(struct device *dev, struct device *master, dw_plat_data->phy_ops = &meson_dw_hdmi_phy_ops; dw_plat_data->phy_name = "meson_dw_hdmi_phy"; dw_plat_data->phy_data = meson_dw_hdmi; - dw_plat_data->input_bus_format = MEDIA_BUS_FMT_YUV8_1X24; dw_plat_data->input_bus_encoding = V4L2_YCBCR_ENC_709; + dw_plat_data->ycbcr_420_allowed = true; if (dw_hdmi_is_compatible(meson_dw_hdmi, "amlogic,meson-gxl-dw-hdmi") || dw_hdmi_is_compatible(meson_dw_hdmi, "amlogic,meson-gxm-dw-hdmi") || @@ -984,11 +1081,16 @@ static int meson_dw_hdmi_bind(struct device *dev, struct device *master, platform_set_drvdata(pdev, meson_dw_hdmi); - meson_dw_hdmi->hdmi = dw_hdmi_bind(pdev, encoder, - &meson_dw_hdmi->dw_plat_data); + meson_dw_hdmi->hdmi = dw_hdmi_probe(pdev, + &meson_dw_hdmi->dw_plat_data); if (IS_ERR(meson_dw_hdmi->hdmi)) return PTR_ERR(meson_dw_hdmi->hdmi); + next_bridge = of_drm_find_bridge(pdev->dev.of_node); + if (next_bridge) + drm_bridge_attach(encoder, next_bridge, + &meson_dw_hdmi->bridge, 0); + DRM_DEBUG_DRIVER("HDMI controller initialized\n"); return 0; diff --git a/drivers/gpu/drm/meson/meson_vclk.c b/drivers/gpu/drm/meson/meson_vclk.c index f690793ae2d5..fdf26dac9fa8 100644 --- a/drivers/gpu/drm/meson/meson_vclk.c +++ b/drivers/gpu/drm/meson/meson_vclk.c @@ -354,12 +354,17 @@ enum { /* 2970 /1 /1 /1 /5 /2 => /1 /1 */ MESON_VCLK_HDMI_297000, /* 5940 /1 /1 /2 /5 /1 => /1 /1 */ - MESON_VCLK_HDMI_594000 + MESON_VCLK_HDMI_594000, +/* 2970 /1 /1 /1 /5 /1 => /1 /2 */ + MESON_VCLK_HDMI_594000_YUV420, }; struct meson_vclk_params { + unsigned int pll_freq; + unsigned int phy_freq; + unsigned int vclk_freq; + unsigned int venc_freq; unsigned int pixel_freq; - unsigned int pll_base_freq; unsigned int pll_od1; unsigned int pll_od2; unsigned int pll_od3; @@ -367,8 +372,11 @@ struct meson_vclk_params { unsigned int vclk_div; } params[] = { [MESON_VCLK_HDMI_ENCI_54000] = { + .pll_freq = 4320000, + .phy_freq = 270000, + .vclk_freq = 54000, + .venc_freq = 54000, .pixel_freq = 54000, - .pll_base_freq = 4320000, .pll_od1 = 4, .pll_od2 = 4, .pll_od3 = 1, @@ -376,8 +384,11 @@ struct meson_vclk_params { .vclk_div = 1, }, [MESON_VCLK_HDMI_DDR_54000] = { - .pixel_freq = 54000, - .pll_base_freq = 4320000, + .pll_freq = 4320000, + .phy_freq = 270000, + .vclk_freq = 54000, + .venc_freq = 54000, + .pixel_freq = 27000, .pll_od1 = 4, .pll_od2 = 4, .pll_od3 = 1, @@ -385,8 +396,11 @@ struct meson_vclk_params { .vclk_div = 1, }, [MESON_VCLK_HDMI_DDR_148500] = { - .pixel_freq = 148500, - .pll_base_freq = 2970000, + .pll_freq = 2970000, + .phy_freq = 742500, + .vclk_freq = 148500, + .venc_freq = 148500, + .pixel_freq = 74250, .pll_od1 = 4, .pll_od2 = 1, .pll_od3 = 1, @@ -394,8 +408,11 @@ struct meson_vclk_params { .vclk_div = 1, }, [MESON_VCLK_HDMI_74250] = { + .pll_freq = 2970000, + .phy_freq = 742500, + .vclk_freq = 74250, + .venc_freq = 74250, .pixel_freq = 74250, - .pll_base_freq = 2970000, .pll_od1 = 2, .pll_od2 = 2, .pll_od3 = 2, @@ -403,8 +420,11 @@ struct meson_vclk_params { .vclk_div = 1, }, [MESON_VCLK_HDMI_148500] = { + .pll_freq = 2970000, + .phy_freq = 1485000, + .vclk_freq = 148500, + .venc_freq = 148500, .pixel_freq = 148500, - .pll_base_freq = 2970000, .pll_od1 = 1, .pll_od2 = 2, .pll_od3 = 2, @@ -412,8 +432,11 @@ struct meson_vclk_params { .vclk_div = 1, }, [MESON_VCLK_HDMI_297000] = { + .pll_freq = 5940000, + .phy_freq = 2970000, + .venc_freq = 297000, + .vclk_freq = 297000, .pixel_freq = 297000, - .pll_base_freq = 5940000, .pll_od1 = 2, .pll_od2 = 1, .pll_od3 = 1, @@ -421,14 +444,29 @@ struct meson_vclk_params { .vclk_div = 2, }, [MESON_VCLK_HDMI_594000] = { + .pll_freq = 5940000, + .phy_freq = 5940000, + .venc_freq = 594000, + .vclk_freq = 594000, .pixel_freq = 594000, - .pll_base_freq = 5940000, .pll_od1 = 1, .pll_od2 = 1, .pll_od3 = 2, .vid_pll_div = VID_PLL_DIV_5, .vclk_div = 1, }, + [MESON_VCLK_HDMI_594000_YUV420] = { + .pll_freq = 5940000, + .phy_freq = 2970000, + .venc_freq = 594000, + .vclk_freq = 594000, + .pixel_freq = 297000, + .pll_od1 = 2, + .pll_od2 = 1, + .pll_od3 = 1, + .vid_pll_div = VID_PLL_DIV_5, + .vclk_div = 1, + }, { /* sentinel */ }, }; @@ -701,6 +739,7 @@ static void meson_hdmi_pll_generic_set(struct meson_drm *priv, unsigned int od, m, frac, od1, od2, od3; if (meson_hdmi_pll_find_params(priv, pll_freq, &m, &frac, &od)) { + /* OD2 goes to the PHY, and needs to be *10, so keep OD3=1 */ od3 = 1; if (od < 4) { od1 = 2; @@ -723,21 +762,28 @@ static void meson_hdmi_pll_generic_set(struct meson_drm *priv, } enum drm_mode_status -meson_vclk_vic_supported_freq(unsigned int freq) +meson_vclk_vic_supported_freq(unsigned int phy_freq, + unsigned int vclk_freq) { int i; - DRM_DEBUG_DRIVER("freq = %d\n", freq); + DRM_DEBUG_DRIVER("phy_freq = %d vclk_freq = %d\n", + phy_freq, vclk_freq); for (i = 0 ; params[i].pixel_freq ; ++i) { DRM_DEBUG_DRIVER("i = %d pixel_freq = %d alt = %d\n", i, params[i].pixel_freq, FREQ_1000_1001(params[i].pixel_freq)); + DRM_DEBUG_DRIVER("i = %d phy_freq = %d alt = %d\n", + i, params[i].phy_freq, + FREQ_1000_1001(params[i].phy_freq/10)*10); /* Match strict frequency */ - if (freq == params[i].pixel_freq) + if (phy_freq == params[i].phy_freq && + vclk_freq == params[i].vclk_freq) return MODE_OK; /* Match 1000/1001 variant */ - if (freq == FREQ_1000_1001(params[i].pixel_freq)) + if (phy_freq == (FREQ_1000_1001(params[i].phy_freq/10)*10) && + vclk_freq == FREQ_1000_1001(params[i].vclk_freq)) return MODE_OK; } @@ -965,8 +1011,9 @@ static void meson_vclk_set(struct meson_drm *priv, unsigned int pll_base_freq, } void meson_vclk_setup(struct meson_drm *priv, unsigned int target, - unsigned int vclk_freq, unsigned int venc_freq, - unsigned int dac_freq, bool hdmi_use_enci) + unsigned int phy_freq, unsigned int vclk_freq, + unsigned int venc_freq, unsigned int dac_freq, + bool hdmi_use_enci) { bool vic_alternate_clock = false; unsigned int freq; @@ -986,7 +1033,7 @@ void meson_vclk_setup(struct meson_drm *priv, unsigned int target, * - venc_div = 1 * - encp encoder */ - meson_vclk_set(priv, vclk_freq * 10, 0, 0, 0, + meson_vclk_set(priv, phy_freq, 0, 0, 0, VID_PLL_DIV_5, 2, 1, 1, false, false); return; } @@ -1008,9 +1055,11 @@ void meson_vclk_setup(struct meson_drm *priv, unsigned int target, } for (freq = 0 ; params[freq].pixel_freq ; ++freq) { - if (vclk_freq == params[freq].pixel_freq || - vclk_freq == FREQ_1000_1001(params[freq].pixel_freq)) { - if (vclk_freq != params[freq].pixel_freq) + if ((phy_freq == params[freq].phy_freq || + phy_freq == FREQ_1000_1001(params[freq].phy_freq/10)*10) && + (vclk_freq == params[freq].vclk_freq || + vclk_freq == FREQ_1000_1001(params[freq].vclk_freq))) { + if (vclk_freq != params[freq].vclk_freq) vic_alternate_clock = true; else vic_alternate_clock = false; @@ -1039,7 +1088,7 @@ void meson_vclk_setup(struct meson_drm *priv, unsigned int target, return; } - meson_vclk_set(priv, params[freq].pll_base_freq, + meson_vclk_set(priv, params[freq].pll_freq, params[freq].pll_od1, params[freq].pll_od2, params[freq].pll_od3, params[freq].vid_pll_div, params[freq].vclk_div, hdmi_tx_div, venc_div, diff --git a/drivers/gpu/drm/meson/meson_vclk.h b/drivers/gpu/drm/meson/meson_vclk.h index b62125540aef..aed0ab2efa71 100644 --- a/drivers/gpu/drm/meson/meson_vclk.h +++ b/drivers/gpu/drm/meson/meson_vclk.h @@ -25,10 +25,11 @@ enum { enum drm_mode_status meson_vclk_dmt_supported_freq(struct meson_drm *priv, unsigned int freq); enum drm_mode_status -meson_vclk_vic_supported_freq(unsigned int freq); +meson_vclk_vic_supported_freq(unsigned int phy_freq, unsigned int vclk_freq); void meson_vclk_setup(struct meson_drm *priv, unsigned int target, - unsigned int vclk_freq, unsigned int venc_freq, - unsigned int dac_freq, bool hdmi_use_enci); + unsigned int phy_freq, unsigned int vclk_freq, + unsigned int venc_freq, unsigned int dac_freq, + bool hdmi_use_enci); #endif /* __MESON_VCLK_H */ diff --git a/drivers/gpu/drm/meson/meson_venc.c b/drivers/gpu/drm/meson/meson_venc.c index 4efd7864d5bf..f93c725b6f02 100644 --- a/drivers/gpu/drm/meson/meson_venc.c +++ b/drivers/gpu/drm/meson/meson_venc.c @@ -946,7 +946,9 @@ bool meson_venc_hdmi_venc_repeat(int vic) EXPORT_SYMBOL_GPL(meson_venc_hdmi_venc_repeat); void meson_venc_hdmi_mode_set(struct meson_drm *priv, int vic, - struct drm_display_mode *mode) + unsigned int ycrcb_map, + bool yuv420_mode, + const struct drm_display_mode *mode) { union meson_hdmi_venc_mode *vmode = NULL; union meson_hdmi_venc_mode vmode_dmt; @@ -1528,14 +1530,14 @@ void meson_venc_hdmi_mode_set(struct meson_drm *priv, int vic, if (mode->flags & DRM_MODE_FLAG_PVSYNC) reg |= VPU_HDMI_INV_VSYNC; - /* Output data format: CbYCr */ - reg |= VPU_HDMI_OUTPUT_CBYCR; + /* Output data format */ + reg |= ycrcb_map; /* * Write rate to the async FIFO between VENC and HDMI. * One write every 2 wr_clk. */ - if (venc_repeat) + if (venc_repeat || yuv420_mode) reg |= VPU_HDMI_WR_RATE(2); /* diff --git a/drivers/gpu/drm/meson/meson_venc.h b/drivers/gpu/drm/meson/meson_venc.h index 576768bdd08d..9138255ffc9e 100644 --- a/drivers/gpu/drm/meson/meson_venc.h +++ b/drivers/gpu/drm/meson/meson_venc.h @@ -60,7 +60,9 @@ extern struct meson_cvbs_enci_mode meson_cvbs_enci_ntsc; void meson_venci_cvbs_mode_set(struct meson_drm *priv, struct meson_cvbs_enci_mode *mode); void meson_venc_hdmi_mode_set(struct meson_drm *priv, int vic, - struct drm_display_mode *mode); + unsigned int ycrcb_map, + bool yuv420_mode, + const struct drm_display_mode *mode); unsigned int meson_venci_get_field(struct meson_drm *priv); void meson_venc_enable_vsync(struct meson_drm *priv); diff --git a/drivers/gpu/drm/meson/meson_venc_cvbs.c b/drivers/gpu/drm/meson/meson_venc_cvbs.c index 1bd6b6d15ffb..541f9eb2a135 100644 --- a/drivers/gpu/drm/meson/meson_venc_cvbs.c +++ b/drivers/gpu/drm/meson/meson_venc_cvbs.c @@ -213,8 +213,10 @@ static void meson_venc_cvbs_encoder_mode_set(struct drm_encoder *encoder, meson_venci_cvbs_mode_set(priv, meson_mode->enci); /* Setup 27MHz vclk2 for ENCI and VDAC */ - meson_vclk_setup(priv, MESON_VCLK_TARGET_CVBS, MESON_VCLK_CVBS, - MESON_VCLK_CVBS, MESON_VCLK_CVBS, true); + meson_vclk_setup(priv, MESON_VCLK_TARGET_CVBS, + MESON_VCLK_CVBS, MESON_VCLK_CVBS, + MESON_VCLK_CVBS, MESON_VCLK_CVBS, + true); } } diff --git a/drivers/gpu/drm/mgag200/mgag200_drv.h b/drivers/gpu/drm/mgag200/mgag200_drv.h index aa32aad222c2..9691252d6233 100644 --- a/drivers/gpu/drm/mgag200/mgag200_drv.h +++ b/drivers/gpu/drm/mgag200/mgag200_drv.h @@ -95,7 +95,6 @@ #define MATROX_DPMS_CLEARED (-1) #define to_mga_crtc(x) container_of(x, struct mga_crtc, base) -#define to_mga_encoder(x) container_of(x, struct mga_encoder, base) #define to_mga_connector(x) container_of(x, struct mga_connector, base) struct mga_crtc { @@ -110,12 +109,6 @@ struct mga_mode_info { struct mga_crtc *crtc; }; -struct mga_encoder { - struct drm_encoder base; - int last_dpms; -}; - - struct mga_i2c_chan { struct i2c_adapter adapter; struct drm_device *dev; @@ -185,6 +178,8 @@ struct mga_device { /* SE model number stored in reg 0x1e24 */ u32 unique_rev_id; + + struct drm_encoder encoder; }; static inline enum mga_type diff --git a/drivers/gpu/drm/mgag200/mgag200_mode.c b/drivers/gpu/drm/mgag200/mgag200_mode.c index 62a8e9ccb16d..d90e83959fca 100644 --- a/drivers/gpu/drm/mgag200/mgag200_mode.c +++ b/drivers/gpu/drm/mgag200/mgag200_mode.c @@ -15,6 +15,7 @@ #include <drm/drm_fourcc.h> #include <drm/drm_plane_helper.h> #include <drm/drm_probe_helper.h> +#include <drm/drm_simple_kms_helper.h> #include "mgag200_drv.h" @@ -1449,76 +1450,6 @@ static void mga_crtc_init(struct mga_device *mdev) drm_crtc_helper_add(&mga_crtc->base, &mga_helper_funcs); } -/* - * The encoder comes after the CRTC in the output pipeline, but before - * the connector. It's responsible for ensuring that the digital - * stream is appropriately converted into the output format. Setup is - * very simple in this case - all we have to do is inform qemu of the - * colour depth in order to ensure that it displays appropriately - */ - -/* - * These functions are analagous to those in the CRTC code, but are intended - * to handle any encoder-specific limitations - */ -static void mga_encoder_mode_set(struct drm_encoder *encoder, - struct drm_display_mode *mode, - struct drm_display_mode *adjusted_mode) -{ - -} - -static void mga_encoder_dpms(struct drm_encoder *encoder, int state) -{ - return; -} - -static void mga_encoder_prepare(struct drm_encoder *encoder) -{ -} - -static void mga_encoder_commit(struct drm_encoder *encoder) -{ -} - -static void mga_encoder_destroy(struct drm_encoder *encoder) -{ - struct mga_encoder *mga_encoder = to_mga_encoder(encoder); - drm_encoder_cleanup(encoder); - kfree(mga_encoder); -} - -static const struct drm_encoder_helper_funcs mga_encoder_helper_funcs = { - .dpms = mga_encoder_dpms, - .mode_set = mga_encoder_mode_set, - .prepare = mga_encoder_prepare, - .commit = mga_encoder_commit, -}; - -static const struct drm_encoder_funcs mga_encoder_encoder_funcs = { - .destroy = mga_encoder_destroy, -}; - -static struct drm_encoder *mga_encoder_init(struct drm_device *dev) -{ - struct drm_encoder *encoder; - struct mga_encoder *mga_encoder; - - mga_encoder = kzalloc(sizeof(struct mga_encoder), GFP_KERNEL); - if (!mga_encoder) - return NULL; - - encoder = &mga_encoder->base; - encoder->possible_crtcs = 0x1; - - drm_encoder_init(dev, encoder, &mga_encoder_encoder_funcs, - DRM_MODE_ENCODER_DAC, NULL); - drm_encoder_helper_add(encoder, &mga_encoder_helper_funcs); - - return encoder; -} - - static int mga_vga_get_modes(struct drm_connector *connector) { struct mga_connector *mga_connector = to_mga_connector(connector); @@ -1686,8 +1617,9 @@ static struct drm_connector *mga_vga_init(struct drm_device *dev) int mgag200_modeset_init(struct mga_device *mdev) { - struct drm_encoder *encoder; + struct drm_encoder *encoder = &mdev->encoder; struct drm_connector *connector; + int ret; mdev->mode_info.mode_config_initialized = true; @@ -1698,11 +1630,15 @@ int mgag200_modeset_init(struct mga_device *mdev) mga_crtc_init(mdev); - encoder = mga_encoder_init(mdev->dev); - if (!encoder) { - DRM_ERROR("mga_encoder_init failed\n"); - return -1; + ret = drm_simple_encoder_init(mdev->dev, encoder, + DRM_MODE_ENCODER_DAC); + if (ret) { + drm_err(mdev->dev, + "drm_simple_encoder_init() failed, error %d\n", + ret); + return ret; } + encoder->possible_crtcs = 0x1; connector = mga_vga_init(mdev->dev); if (!connector) { diff --git a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c index 7d9e63e20ded..724024a2243a 100644 --- a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c @@ -1446,18 +1446,31 @@ static const struct adreno_gpu_funcs funcs = { static void check_speed_bin(struct device *dev) { struct nvmem_cell *cell; - u32 bin, val; + u32 val; + + /* + * If the OPP table specifies a opp-supported-hw property then we have + * to set something with dev_pm_opp_set_supported_hw() or the table + * doesn't get populated so pick an arbitrary value that should + * ensure the default frequencies are selected but not conflict with any + * actual bins + */ + val = 0x80; cell = nvmem_cell_get(dev, "speed_bin"); - /* If a nvmem cell isn't defined, nothing to do */ - if (IS_ERR(cell)) - return; + if (!IS_ERR(cell)) { + void *buf = nvmem_cell_read(cell, NULL); + + if (!IS_ERR(buf)) { + u8 bin = *((u8 *) buf); - bin = *((u32 *) nvmem_cell_read(cell, NULL)); - nvmem_cell_put(cell); + val = (1 << bin); + kfree(buf); + } - val = (1 << bin); + nvmem_cell_put(cell); + } dev_pm_opp_set_supported_hw(dev, &val, 1); } diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c index 983afeaee737..c4e71abbdd53 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c @@ -2,6 +2,7 @@ /* Copyright (c) 2017-2019 The Linux Foundation. All rights reserved. */ #include <linux/clk.h> +#include <linux/dma-mapping.h> #include <linux/interconnect.h> #include <linux/pm_domain.h> #include <linux/pm_opp.h> @@ -796,12 +797,41 @@ bool a6xx_gmu_isidle(struct a6xx_gmu *gmu) return true; } +#define GBIF_CLIENT_HALT_MASK BIT(0) +#define GBIF_ARB_HALT_MASK BIT(1) + +static void a6xx_bus_clear_pending_transactions(struct adreno_gpu *adreno_gpu) +{ + struct msm_gpu *gpu = &adreno_gpu->base; + + if (!a6xx_has_gbif(adreno_gpu)) { + gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, 0xf); + spin_until((gpu_read(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL1) & + 0xf) == 0xf); + gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, 0); + + return; + } + + /* Halt new client requests on GBIF */ + gpu_write(gpu, REG_A6XX_GBIF_HALT, GBIF_CLIENT_HALT_MASK); + spin_until((gpu_read(gpu, REG_A6XX_GBIF_HALT_ACK) & + (GBIF_CLIENT_HALT_MASK)) == GBIF_CLIENT_HALT_MASK); + + /* Halt all AXI requests on GBIF */ + gpu_write(gpu, REG_A6XX_GBIF_HALT, GBIF_ARB_HALT_MASK); + spin_until((gpu_read(gpu, REG_A6XX_GBIF_HALT_ACK) & + (GBIF_ARB_HALT_MASK)) == GBIF_ARB_HALT_MASK); + + /* The GBIF halt needs to be explicitly cleared */ + gpu_write(gpu, REG_A6XX_GBIF_HALT, 0x0); +} + /* Gracefully try to shut down the GMU and by extension the GPU */ static void a6xx_gmu_shutdown(struct a6xx_gmu *gmu) { struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu); struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; - struct msm_gpu *gpu = &adreno_gpu->base; u32 val; /* @@ -819,11 +849,7 @@ static void a6xx_gmu_shutdown(struct a6xx_gmu *gmu) return; } - /* Clear the VBIF pipe before shutting down */ - gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, 0xf); - spin_until((gpu_read(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL1) & 0xf) - == 0xf); - gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, 0); + a6xx_bus_clear_pending_transactions(adreno_gpu); /* tell the GMU we want to slumber */ a6xx_gmu_notify_slumber(gmu); @@ -895,21 +921,10 @@ int a6xx_gmu_stop(struct a6xx_gpu *a6xx_gpu) static void a6xx_gmu_memory_free(struct a6xx_gmu *gmu, struct a6xx_gmu_bo *bo) { - int count, i; - u64 iova; - if (IS_ERR_OR_NULL(bo)) return; - count = bo->size >> PAGE_SHIFT; - iova = bo->iova; - - for (i = 0; i < count; i++, iova += PAGE_SIZE) { - iommu_unmap(gmu->domain, iova, PAGE_SIZE); - __free_pages(bo->pages[i], 0); - } - - kfree(bo->pages); + dma_free_wc(gmu->dev, bo->size, bo->virt, bo->iova); kfree(bo); } @@ -917,7 +932,6 @@ static struct a6xx_gmu_bo *a6xx_gmu_memory_alloc(struct a6xx_gmu *gmu, size_t size) { struct a6xx_gmu_bo *bo; - int ret, count, i; bo = kzalloc(sizeof(*bo), GFP_KERNEL); if (!bo) @@ -925,86 +939,14 @@ static struct a6xx_gmu_bo *a6xx_gmu_memory_alloc(struct a6xx_gmu *gmu, bo->size = PAGE_ALIGN(size); - count = bo->size >> PAGE_SHIFT; + bo->virt = dma_alloc_wc(gmu->dev, bo->size, &bo->iova, GFP_KERNEL); - bo->pages = kcalloc(count, sizeof(struct page *), GFP_KERNEL); - if (!bo->pages) { + if (!bo->virt) { kfree(bo); return ERR_PTR(-ENOMEM); } - for (i = 0; i < count; i++) { - bo->pages[i] = alloc_page(GFP_KERNEL); - if (!bo->pages[i]) - goto err; - } - - bo->iova = gmu->uncached_iova_base; - - for (i = 0; i < count; i++) { - ret = iommu_map(gmu->domain, - bo->iova + (PAGE_SIZE * i), - page_to_phys(bo->pages[i]), PAGE_SIZE, - IOMMU_READ | IOMMU_WRITE); - - if (ret) { - DRM_DEV_ERROR(gmu->dev, "Unable to map GMU buffer object\n"); - - for (i = i - 1 ; i >= 0; i--) - iommu_unmap(gmu->domain, - bo->iova + (PAGE_SIZE * i), - PAGE_SIZE); - - goto err; - } - } - - bo->virt = vmap(bo->pages, count, VM_IOREMAP, - pgprot_writecombine(PAGE_KERNEL)); - if (!bo->virt) - goto err; - - /* Align future IOVA addresses on 1MB boundaries */ - gmu->uncached_iova_base += ALIGN(size, SZ_1M); - return bo; - -err: - for (i = 0; i < count; i++) { - if (bo->pages[i]) - __free_pages(bo->pages[i], 0); - } - - kfree(bo->pages); - kfree(bo); - - return ERR_PTR(-ENOMEM); -} - -static int a6xx_gmu_memory_probe(struct a6xx_gmu *gmu) -{ - int ret; - - /* - * The GMU address space is hardcoded to treat the range - * 0x60000000 - 0x80000000 as un-cached memory. All buffers shared - * between the GMU and the CPU will live in this space - */ - gmu->uncached_iova_base = 0x60000000; - - - gmu->domain = iommu_domain_alloc(&platform_bus_type); - if (!gmu->domain) - return -ENODEV; - - ret = iommu_attach_device(gmu->domain, gmu->dev); - - if (ret) { - iommu_domain_free(gmu->domain); - gmu->domain = NULL; - } - - return ret; } /* Return the 'arc-level' for the given frequency */ @@ -1264,10 +1206,6 @@ void a6xx_gmu_remove(struct a6xx_gpu *a6xx_gpu) a6xx_gmu_memory_free(gmu, gmu->hfi); - iommu_detach_device(gmu->domain, gmu->dev); - - iommu_domain_free(gmu->domain); - free_irq(gmu->gmu_irq, gmu); free_irq(gmu->hfi_irq, gmu); @@ -1288,7 +1226,15 @@ int a6xx_gmu_init(struct a6xx_gpu *a6xx_gpu, struct device_node *node) gmu->dev = &pdev->dev; - of_dma_configure(gmu->dev, node, true); + /* Pass force_dma false to require the DT to set the dma region */ + ret = of_dma_configure(gmu->dev, node, false); + if (ret) + return ret; + + /* Set the mask after the of_dma_configure() */ + ret = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(31)); + if (ret) + return ret; /* Fow now, don't do anything fancy until we get our feet under us */ gmu->idle_level = GMU_IDLE_STATE_ACTIVE; @@ -1300,11 +1246,6 @@ int a6xx_gmu_init(struct a6xx_gpu *a6xx_gpu, struct device_node *node) if (ret) goto err_put_device; - /* Set up the IOMMU context bank */ - ret = a6xx_gmu_memory_probe(gmu); - if (ret) - goto err_put_device; - /* Allocate memory for for the HFI queues */ gmu->hfi = a6xx_gmu_memory_alloc(gmu, SZ_16K); if (IS_ERR(gmu->hfi)) @@ -1350,11 +1291,6 @@ err_mmio: err_memory: a6xx_gmu_memory_free(gmu, gmu->hfi); - if (gmu->domain) { - iommu_detach_device(gmu->domain, gmu->dev); - - iommu_domain_free(gmu->domain); - } ret = -ENODEV; err_put_device: diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.h b/drivers/gpu/drm/msm/adreno/a6xx_gmu.h index 2af91ed7ed0c..4af65a36d5ca 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.h +++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.h @@ -12,8 +12,7 @@ struct a6xx_gmu_bo { void *virt; size_t size; - u64 iova; - struct page **pages; + dma_addr_t iova; }; /* @@ -49,9 +48,6 @@ struct a6xx_gmu { int hfi_irq; int gmu_irq; - struct iommu_domain *domain; - u64 uncached_iova_base; - struct device *gxpd; int idle_level; diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c index daf07800cde0..68af24150de5 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c @@ -378,18 +378,6 @@ static int a6xx_hw_init(struct msm_gpu *gpu) struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); int ret; - /* - * During a previous slumber, GBIF halt is asserted to ensure - * no further transaction can go through GPU before GPU - * headswitch is turned off. - * - * This halt is deasserted once headswitch goes off but - * incase headswitch doesn't goes off clear GBIF halt - * here to ensure GPU wake-up doesn't fail because of - * halted GPU transactions. - */ - gpu_write(gpu, REG_A6XX_GBIF_HALT, 0x0); - /* Make sure the GMU keeps the GPU on while we set it up */ a6xx_gmu_set_oob(&a6xx_gpu->gmu, GMU_OOB_GPU_SET); @@ -470,10 +458,12 @@ static int a6xx_hw_init(struct msm_gpu *gpu) /* Select CP0 to always count cycles */ gpu_write(gpu, REG_A6XX_CP_PERFCTR_CP_SEL_0, PERF_CP_ALWAYS_COUNT); - gpu_write(gpu, REG_A6XX_RB_NC_MODE_CNTL, 2 << 1); - gpu_write(gpu, REG_A6XX_TPL1_NC_MODE_CNTL, 2 << 1); - gpu_write(gpu, REG_A6XX_SP_NC_MODE_CNTL, 2 << 1); - gpu_write(gpu, REG_A6XX_UCHE_MODE_CNTL, 2 << 21); + if (adreno_is_a630(adreno_gpu)) { + gpu_write(gpu, REG_A6XX_RB_NC_MODE_CNTL, 2 << 1); + gpu_write(gpu, REG_A6XX_TPL1_NC_MODE_CNTL, 2 << 1); + gpu_write(gpu, REG_A6XX_SP_NC_MODE_CNTL, 2 << 1); + gpu_write(gpu, REG_A6XX_UCHE_MODE_CNTL, 2 << 21); + } /* Enable fault detection */ gpu_write(gpu, REG_A6XX_RBBM_INTERFACE_HANG_INT_CNTL, @@ -748,39 +738,6 @@ static const u32 a6xx_register_offsets[REG_ADRENO_REGISTER_MAX] = { REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_CNTL, REG_A6XX_CP_RB_CNTL), }; -#define GBIF_CLIENT_HALT_MASK BIT(0) -#define GBIF_ARB_HALT_MASK BIT(1) - -static void a6xx_bus_clear_pending_transactions(struct adreno_gpu *adreno_gpu) -{ - struct msm_gpu *gpu = &adreno_gpu->base; - - if(!a6xx_has_gbif(adreno_gpu)){ - gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, 0xf); - spin_until((gpu_read(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL1) & - 0xf) == 0xf); - gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, 0); - - return; - } - - /* Halt new client requests on GBIF */ - gpu_write(gpu, REG_A6XX_GBIF_HALT, GBIF_CLIENT_HALT_MASK); - spin_until((gpu_read(gpu, REG_A6XX_GBIF_HALT_ACK) & - (GBIF_CLIENT_HALT_MASK)) == GBIF_CLIENT_HALT_MASK); - - /* Halt all AXI requests on GBIF */ - gpu_write(gpu, REG_A6XX_GBIF_HALT, GBIF_ARB_HALT_MASK); - spin_until((gpu_read(gpu, REG_A6XX_GBIF_HALT_ACK) & - (GBIF_ARB_HALT_MASK)) == GBIF_ARB_HALT_MASK); - - /* - * GMU needs DDR access in slumber path. Deassert GBIF halt now - * to allow for GMU to access system memory. - */ - gpu_write(gpu, REG_A6XX_GBIF_HALT, 0x0); -} - static int a6xx_pm_resume(struct msm_gpu *gpu) { struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); @@ -805,16 +762,6 @@ static int a6xx_pm_suspend(struct msm_gpu *gpu) devfreq_suspend_device(gpu->devfreq.devfreq); - /* - * Make sure the GMU is idle before continuing (because some transitions - * may use VBIF - */ - a6xx_gmu_wait_for_idle(&a6xx_gpu->gmu); - - /* Clear the VBIF pipe before shutting down */ - /* FIXME: This accesses the GPU - do we need to make sure it is on? */ - a6xx_bus_clear_pending_transactions(adreno_gpu); - return a6xx_gmu_stop(a6xx_gpu); } diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.h b/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.h index e67c20c415af..24c974c293e5 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.h +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.h @@ -379,7 +379,7 @@ static const struct a6xx_indexed_registers { }; static const struct a6xx_indexed_registers a6xx_cp_mempool_indexed = { - "CP_MEMPOOOL", REG_A6XX_CP_MEM_POOL_DBG_ADDR, + "CP_MEMPOOL", REG_A6XX_CP_MEM_POOL_DBG_ADDR, REG_A6XX_CP_MEM_POOL_DBG_DATA, 0x2060, }; diff --git a/drivers/gpu/drm/msm/adreno/a6xx_hfi.c b/drivers/gpu/drm/msm/adreno/a6xx_hfi.c index eda11abc5f01..e450e0b97211 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_hfi.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_hfi.c @@ -7,6 +7,7 @@ #include "a6xx_gmu.h" #include "a6xx_gmu.xml.h" +#include "a6xx_gpu.h" #define HFI_MSG_ID(val) [val] = #val @@ -216,48 +217,82 @@ static int a6xx_hfi_send_perf_table(struct a6xx_gmu *gmu) NULL, 0); } -static int a6xx_hfi_send_bw_table(struct a6xx_gmu *gmu) +static void a618_build_bw_table(struct a6xx_hfi_msg_bw_table *msg) { - struct a6xx_hfi_msg_bw_table msg = { 0 }; + /* Send a single "off" entry since the 618 GMU doesn't do bus scaling */ + msg->bw_level_num = 1; + + msg->ddr_cmds_num = 3; + msg->ddr_wait_bitmask = 0x01; + + msg->ddr_cmds_addrs[0] = 0x50000; + msg->ddr_cmds_addrs[1] = 0x5003c; + msg->ddr_cmds_addrs[2] = 0x5000c; + + msg->ddr_cmds_data[0][0] = 0x40000000; + msg->ddr_cmds_data[0][1] = 0x40000000; + msg->ddr_cmds_data[0][2] = 0x40000000; /* - * The sdm845 GMU doesn't do bus frequency scaling on its own but it - * does need at least one entry in the list because it might be accessed - * when the GMU is shutting down. Send a single "off" entry. + * These are the CX (CNOC) votes - these are used by the GMU but the + * votes are known and fixed for the target */ + msg->cnoc_cmds_num = 1; + msg->cnoc_wait_bitmask = 0x01; + + msg->cnoc_cmds_addrs[0] = 0x5007c; + msg->cnoc_cmds_data[0][0] = 0x40000000; + msg->cnoc_cmds_data[1][0] = 0x60000001; +} - msg.bw_level_num = 1; +static void a6xx_build_bw_table(struct a6xx_hfi_msg_bw_table *msg) +{ + /* Send a single "off" entry since the 630 GMU doesn't do bus scaling */ + msg->bw_level_num = 1; - msg.ddr_cmds_num = 3; - msg.ddr_wait_bitmask = 0x07; + msg->ddr_cmds_num = 3; + msg->ddr_wait_bitmask = 0x07; - msg.ddr_cmds_addrs[0] = 0x50000; - msg.ddr_cmds_addrs[1] = 0x5005c; - msg.ddr_cmds_addrs[2] = 0x5000c; + msg->ddr_cmds_addrs[0] = 0x50000; + msg->ddr_cmds_addrs[1] = 0x5005c; + msg->ddr_cmds_addrs[2] = 0x5000c; - msg.ddr_cmds_data[0][0] = 0x40000000; - msg.ddr_cmds_data[0][1] = 0x40000000; - msg.ddr_cmds_data[0][2] = 0x40000000; + msg->ddr_cmds_data[0][0] = 0x40000000; + msg->ddr_cmds_data[0][1] = 0x40000000; + msg->ddr_cmds_data[0][2] = 0x40000000; /* * These are the CX (CNOC) votes. This is used but the values for the * sdm845 GMU are known and fixed so we can hard code them. */ - msg.cnoc_cmds_num = 3; - msg.cnoc_wait_bitmask = 0x05; + msg->cnoc_cmds_num = 3; + msg->cnoc_wait_bitmask = 0x05; - msg.cnoc_cmds_addrs[0] = 0x50034; - msg.cnoc_cmds_addrs[1] = 0x5007c; - msg.cnoc_cmds_addrs[2] = 0x5004c; + msg->cnoc_cmds_addrs[0] = 0x50034; + msg->cnoc_cmds_addrs[1] = 0x5007c; + msg->cnoc_cmds_addrs[2] = 0x5004c; - msg.cnoc_cmds_data[0][0] = 0x40000000; - msg.cnoc_cmds_data[0][1] = 0x00000000; - msg.cnoc_cmds_data[0][2] = 0x40000000; + msg->cnoc_cmds_data[0][0] = 0x40000000; + msg->cnoc_cmds_data[0][1] = 0x00000000; + msg->cnoc_cmds_data[0][2] = 0x40000000; + + msg->cnoc_cmds_data[1][0] = 0x60000001; + msg->cnoc_cmds_data[1][1] = 0x20000001; + msg->cnoc_cmds_data[1][2] = 0x60000001; +} + + +static int a6xx_hfi_send_bw_table(struct a6xx_gmu *gmu) +{ + struct a6xx_hfi_msg_bw_table msg = { 0 }; + struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu); + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; - msg.cnoc_cmds_data[1][0] = 0x60000001; - msg.cnoc_cmds_data[1][1] = 0x20000001; - msg.cnoc_cmds_data[1][2] = 0x60000001; + if (adreno_is_a618(adreno_gpu)) + a618_build_bw_table(&msg); + else + a6xx_build_bw_table(&msg); return a6xx_hfi_send_msg(gmu, HFI_H2F_MSG_BW_TABLE, &msg, sizeof(msg), NULL, 0); diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.c b/drivers/gpu/drm/msm/adreno/adreno_gpu.c index 7fd29829b2fa..1d5c43c22269 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_gpu.c +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.c @@ -673,7 +673,7 @@ static char *adreno_gpu_ascii85_encode(u32 *src, size_t len) return NULL; for (i = 0; i < l; i++) - buf_itr += snprintf(buf + buf_itr, buffer_size - buf_itr, "%s", + buf_itr += scnprintf(buf + buf_itr, buffer_size - buf_itr, "%s", ascii85_encode(src[i], out)); return buf; diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c index bf513411b243..17448505a9b5 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c @@ -1272,6 +1272,8 @@ static const struct drm_crtc_funcs dpu_crtc_funcs = { .atomic_destroy_state = dpu_crtc_destroy_state, .late_register = dpu_crtc_late_register, .early_unregister = dpu_crtc_early_unregister, + .enable_vblank = msm_crtc_enable_vblank, + .disable_vblank = msm_crtc_disable_vblank, }; static const struct drm_crtc_helper_funcs dpu_crtc_helper_funcs = { diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c index f8ac3bf60fd6..a1b79ee2bd9d 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c @@ -164,7 +164,6 @@ enum dpu_enc_rc_states { * clks and resources after IDLE_TIMEOUT time. * @vsync_event_work: worker to handle vsync event for autorefresh * @topology: topology of the display - * @mode_set_complete: flag to indicate modeset completion * @idle_timeout: idle timeout duration in milliseconds */ struct dpu_encoder_virt { @@ -202,7 +201,6 @@ struct dpu_encoder_virt { struct delayed_work delayed_off_work; struct kthread_work vsync_event_work; struct msm_display_topology topology; - bool mode_set_complete; u32 idle_timeout; }; @@ -461,7 +459,7 @@ void dpu_encoder_helper_split_config( struct msm_display_info *disp_info; if (!phys_enc->hw_mdptop || !phys_enc->parent) { - DPU_ERROR("invalid arg(s), encoder %d\n", phys_enc != 0); + DPU_ERROR("invalid arg(s), encoder %d\n", phys_enc != NULL); return; } @@ -512,7 +510,6 @@ static void _dpu_encoder_adjust_mode(struct drm_connector *connector, if (cur_mode->vdisplay == adj_mode->vdisplay && cur_mode->hdisplay == adj_mode->hdisplay && drm_mode_vrefresh(cur_mode) == drm_mode_vrefresh(adj_mode)) { - adj_mode->private = cur_mode->private; adj_mode->private_flags |= cur_mode->private_flags; } } @@ -563,12 +560,13 @@ static int dpu_encoder_virt_atomic_check( const struct drm_display_mode *mode; struct drm_display_mode *adj_mode; struct msm_display_topology topology; + struct dpu_global_state *global_state; int i = 0; int ret = 0; if (!drm_enc || !crtc_state || !conn_state) { DPU_ERROR("invalid arg(s), drm_enc %d, crtc/conn state %d/%d\n", - drm_enc != 0, crtc_state != 0, conn_state != 0); + drm_enc != NULL, crtc_state != NULL, conn_state != NULL); return -EINVAL; } @@ -579,6 +577,7 @@ static int dpu_encoder_virt_atomic_check( dpu_kms = to_dpu_kms(priv->kms); mode = &crtc_state->mode; adj_mode = &crtc_state->adjusted_mode; + global_state = dpu_kms_get_existing_global_state(dpu_kms); trace_dpu_enc_atomic_check(DRMID(drm_enc)); /* @@ -610,17 +609,15 @@ static int dpu_encoder_virt_atomic_check( topology = dpu_encoder_get_topology(dpu_enc, dpu_kms, adj_mode); - /* Reserve dynamic resources now. Indicating AtomicTest phase */ + /* Reserve dynamic resources now. */ if (!ret) { /* * Avoid reserving resources when mode set is pending. Topology * info may not be available to complete reservation. */ - if (drm_atomic_crtc_needs_modeset(crtc_state) - && dpu_enc->mode_set_complete) { - ret = dpu_rm_reserve(&dpu_kms->rm, drm_enc, crtc_state, - topology, true); - dpu_enc->mode_set_complete = false; + if (drm_atomic_crtc_needs_modeset(crtc_state)) { + ret = dpu_rm_reserve(&dpu_kms->rm, global_state, + drm_enc, crtc_state, topology); } } @@ -957,12 +954,13 @@ static void dpu_encoder_virt_mode_set(struct drm_encoder *drm_enc, struct drm_connector *conn = NULL, *conn_iter; struct drm_crtc *drm_crtc; struct dpu_crtc_state *cstate; - struct dpu_rm_hw_iter hw_iter; + struct dpu_global_state *global_state; struct msm_display_topology topology; - struct dpu_hw_ctl *hw_ctl[MAX_CHANNELS_PER_ENC] = { NULL }; - struct dpu_hw_mixer *hw_lm[MAX_CHANNELS_PER_ENC] = { NULL }; - int num_lm = 0, num_ctl = 0; - int i, j, ret; + struct dpu_hw_blk *hw_pp[MAX_CHANNELS_PER_ENC]; + struct dpu_hw_blk *hw_ctl[MAX_CHANNELS_PER_ENC]; + struct dpu_hw_blk *hw_lm[MAX_CHANNELS_PER_ENC]; + int num_lm, num_ctl, num_pp; + int i, j; if (!drm_enc) { DPU_ERROR("invalid encoder\n"); @@ -976,6 +974,12 @@ static void dpu_encoder_virt_mode_set(struct drm_encoder *drm_enc, dpu_kms = to_dpu_kms(priv->kms); connector_list = &dpu_kms->dev->mode_config.connector_list; + global_state = dpu_kms_get_existing_global_state(dpu_kms); + if (IS_ERR_OR_NULL(global_state)) { + DPU_ERROR("Failed to get global state"); + return; + } + trace_dpu_enc_mode_set(DRMID(drm_enc)); list_for_each_entry(conn_iter, connector_list, head) @@ -996,77 +1000,57 @@ static void dpu_encoder_virt_mode_set(struct drm_encoder *drm_enc, topology = dpu_encoder_get_topology(dpu_enc, dpu_kms, adj_mode); - /* Reserve dynamic resources now. Indicating non-AtomicTest phase */ - ret = dpu_rm_reserve(&dpu_kms->rm, drm_enc, drm_crtc->state, - topology, false); - if (ret) { - DPU_ERROR_ENC(dpu_enc, - "failed to reserve hw resources, %d\n", ret); - return; - } - - dpu_rm_init_hw_iter(&hw_iter, drm_enc->base.id, DPU_HW_BLK_PINGPONG); - for (i = 0; i < MAX_CHANNELS_PER_ENC; i++) { - dpu_enc->hw_pp[i] = NULL; - if (!dpu_rm_get_hw(&dpu_kms->rm, &hw_iter)) - break; - dpu_enc->hw_pp[i] = (struct dpu_hw_pingpong *) hw_iter.hw; - } - - dpu_rm_init_hw_iter(&hw_iter, drm_enc->base.id, DPU_HW_BLK_CTL); - for (i = 0; i < MAX_CHANNELS_PER_ENC; i++) { - if (!dpu_rm_get_hw(&dpu_kms->rm, &hw_iter)) - break; - hw_ctl[i] = (struct dpu_hw_ctl *)hw_iter.hw; - num_ctl++; - } + /* Query resource that have been reserved in atomic check step. */ + num_pp = dpu_rm_get_assigned_resources(&dpu_kms->rm, global_state, + drm_enc->base.id, DPU_HW_BLK_PINGPONG, hw_pp, + ARRAY_SIZE(hw_pp)); + num_ctl = dpu_rm_get_assigned_resources(&dpu_kms->rm, global_state, + drm_enc->base.id, DPU_HW_BLK_CTL, hw_ctl, ARRAY_SIZE(hw_ctl)); + num_lm = dpu_rm_get_assigned_resources(&dpu_kms->rm, global_state, + drm_enc->base.id, DPU_HW_BLK_LM, hw_lm, ARRAY_SIZE(hw_lm)); - dpu_rm_init_hw_iter(&hw_iter, drm_enc->base.id, DPU_HW_BLK_LM); - for (i = 0; i < MAX_CHANNELS_PER_ENC; i++) { - if (!dpu_rm_get_hw(&dpu_kms->rm, &hw_iter)) - break; - hw_lm[i] = (struct dpu_hw_mixer *)hw_iter.hw; - num_lm++; - } + for (i = 0; i < MAX_CHANNELS_PER_ENC; i++) + dpu_enc->hw_pp[i] = i < num_pp ? to_dpu_hw_pingpong(hw_pp[i]) + : NULL; cstate = to_dpu_crtc_state(drm_crtc->state); for (i = 0; i < num_lm; i++) { int ctl_idx = (i < num_ctl) ? i : (num_ctl-1); - cstate->mixers[i].hw_lm = hw_lm[i]; - cstate->mixers[i].lm_ctl = hw_ctl[ctl_idx]; + cstate->mixers[i].hw_lm = to_dpu_hw_mixer(hw_lm[i]); + cstate->mixers[i].lm_ctl = to_dpu_hw_ctl(hw_ctl[ctl_idx]); } cstate->num_mixers = num_lm; for (i = 0; i < dpu_enc->num_phys_encs; i++) { + int num_blk; + struct dpu_hw_blk *hw_blk[MAX_CHANNELS_PER_ENC]; struct dpu_encoder_phys *phys = dpu_enc->phys_encs[i]; if (!dpu_enc->hw_pp[i]) { DPU_ERROR_ENC(dpu_enc, "no pp block assigned at idx: %d\n", i); - goto error; + return; } if (!hw_ctl[i]) { DPU_ERROR_ENC(dpu_enc, "no ctl block assigned at idx: %d\n", i); - goto error; + return; } phys->hw_pp = dpu_enc->hw_pp[i]; - phys->hw_ctl = hw_ctl[i]; + phys->hw_ctl = to_dpu_hw_ctl(hw_ctl[i]); - dpu_rm_init_hw_iter(&hw_iter, drm_enc->base.id, - DPU_HW_BLK_INTF); - for (j = 0; j < MAX_CHANNELS_PER_ENC; j++) { + num_blk = dpu_rm_get_assigned_resources(&dpu_kms->rm, + global_state, drm_enc->base.id, DPU_HW_BLK_INTF, + hw_blk, ARRAY_SIZE(hw_blk)); + for (j = 0; j < num_blk; j++) { struct dpu_hw_intf *hw_intf; - if (!dpu_rm_get_hw(&dpu_kms->rm, &hw_iter)) - break; - - hw_intf = (struct dpu_hw_intf *)hw_iter.hw; + hw_intf = to_dpu_hw_intf(hw_blk[i]); if (hw_intf->idx == phys->intf_idx) phys->hw_intf = hw_intf; } @@ -1074,18 +1058,13 @@ static void dpu_encoder_virt_mode_set(struct drm_encoder *drm_enc, if (!phys->hw_intf) { DPU_ERROR_ENC(dpu_enc, "no intf block assigned at idx: %d\n", i); - goto error; + return; } phys->connector = conn->state->connector; if (phys->ops.mode_set) phys->ops.mode_set(phys, mode, adj_mode); } - - dpu_enc->mode_set_complete = true; - -error: - dpu_rm_release(&dpu_kms->rm, drm_enc); } static void _dpu_encoder_virt_enable_helper(struct drm_encoder *drm_enc) @@ -1182,6 +1161,7 @@ static void dpu_encoder_virt_disable(struct drm_encoder *drm_enc) struct dpu_encoder_virt *dpu_enc = NULL; struct msm_drm_private *priv; struct dpu_kms *dpu_kms; + struct dpu_global_state *global_state; int i = 0; if (!drm_enc) { @@ -1200,6 +1180,7 @@ static void dpu_encoder_virt_disable(struct drm_encoder *drm_enc) priv = drm_enc->dev->dev_private; dpu_kms = to_dpu_kms(priv->kms); + global_state = dpu_kms_get_existing_global_state(dpu_kms); trace_dpu_enc_disable(DRMID(drm_enc)); @@ -1229,7 +1210,7 @@ static void dpu_encoder_virt_disable(struct drm_encoder *drm_enc) DPU_DEBUG_ENC(dpu_enc, "encoder disabled\n"); - dpu_rm_release(&dpu_kms->rm, drm_enc); + dpu_rm_release(global_state, drm_enc); mutex_unlock(&dpu_enc->enc_lock); } @@ -1965,7 +1946,7 @@ static int dpu_encoder_virt_add_phys_encs( if (IS_ERR_OR_NULL(enc)) { DPU_ERROR_ENC(dpu_enc, "failed to init vid enc: %ld\n", PTR_ERR(enc)); - return enc == 0 ? -EINVAL : PTR_ERR(enc); + return enc == NULL ? -EINVAL : PTR_ERR(enc); } dpu_enc->phys_encs[dpu_enc->num_phys_encs] = enc; @@ -1978,7 +1959,7 @@ static int dpu_encoder_virt_add_phys_encs( if (IS_ERR_OR_NULL(enc)) { DPU_ERROR_ENC(dpu_enc, "failed to init cmd enc: %ld\n", PTR_ERR(enc)); - return enc == 0 ? -EINVAL : PTR_ERR(enc); + return enc == NULL ? -EINVAL : PTR_ERR(enc); } dpu_enc->phys_encs[dpu_enc->num_phys_encs] = enc; @@ -2009,7 +1990,7 @@ static int dpu_encoder_setup_display(struct dpu_encoder_virt *dpu_enc, struct dpu_enc_phys_init_params phys_params; if (!dpu_enc) { - DPU_ERROR("invalid arg(s), enc %d\n", dpu_enc != 0); + DPU_ERROR("invalid arg(s), enc %d\n", dpu_enc != NULL); return -EINVAL; } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c index 39e1e280ba44..8493d68ad841 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c @@ -411,7 +411,7 @@ static void _dpu_encoder_phys_cmd_pingpong_config( to_dpu_encoder_phys_cmd(phys_enc); if (!phys_enc->hw_pp || !phys_enc->hw_ctl->ops.setup_intf_cfg) { - DPU_ERROR("invalid arg(s), enc %d\n", phys_enc != 0); + DPU_ERROR("invalid arg(s), enc %d\n", phys_enc != NULL); return; } @@ -440,7 +440,7 @@ static void dpu_encoder_phys_cmd_enable_helper( u32 flush_mask = 0; if (!phys_enc->hw_pp) { - DPU_ERROR("invalid arg(s), encoder %d\n", phys_enc != 0); + DPU_ERROR("invalid arg(s), encoder %d\n", phys_enc != NULL); return; } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c index c71c18de5966..b5a49050d131 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c @@ -239,7 +239,7 @@ static void dpu_encoder_phys_vid_setup_timing_engine( struct dpu_hw_intf_cfg intf_cfg = { 0 }; if (!phys_enc->hw_ctl->ops.setup_intf_cfg) { - DPU_ERROR("invalid encoder %d\n", phys_enc != 0); + DPU_ERROR("invalid encoder %d\n", phys_enc != NULL); return; } @@ -559,7 +559,7 @@ static void dpu_encoder_phys_vid_disable(struct dpu_encoder_phys *phys_enc) if (!phys_enc->hw_intf) { DPU_ERROR("invalid hw_intf %d hw_ctl %d\n", - phys_enc->hw_intf != 0, phys_enc->hw_ctl != 0); + phys_enc->hw_intf != NULL, phys_enc->hw_ctl != NULL); return; } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c index 528632690f1e..a05282dede91 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c @@ -255,13 +255,13 @@ static const struct dpu_format dpu_format_map[] = { INTERLEAVED_RGB_FMT(RGB565, 0, COLOR_5BIT, COLOR_6BIT, COLOR_5BIT, - C2_R_Cr, C0_G_Y, C1_B_Cb, 0, 3, + C1_B_Cb, C0_G_Y, C2_R_Cr, 0, 3, false, 2, 0, DPU_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(BGR565, 0, COLOR_5BIT, COLOR_6BIT, COLOR_5BIT, - C1_B_Cb, C0_G_Y, C2_R_Cr, 0, 3, + C2_R_Cr, C0_G_Y, C1_B_Cb, 0, 3, false, 2, 0, DPU_FETCH_LINEAR, 1), diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h index 85468981632d..0ead64d3f63d 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h @@ -90,6 +90,16 @@ struct dpu_hw_intf { }; /** + * to_dpu_hw_intf - convert base object dpu_hw_base to container + * @hw: Pointer to base hardware block + * return: Pointer to hardware block container + */ +static inline struct dpu_hw_intf *to_dpu_hw_intf(struct dpu_hw_blk *hw) +{ + return container_of(hw, struct dpu_hw_intf, base); +} + +/** * dpu_hw_intf_init(): Initializes the intf driver for the passed * interface idx. * @idx: interface index for which driver object is required diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h index 3d6f46b1db30..d73cb73e938b 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h @@ -97,6 +97,16 @@ struct dpu_hw_pingpong { }; /** + * to_dpu_hw_pingpong - convert base object dpu_hw_base to container + * @hw: Pointer to base hardware block + * return: Pointer to hardware block container + */ +static inline struct dpu_hw_pingpong *to_dpu_hw_pingpong(struct dpu_hw_blk *hw) +{ + return container_of(hw, struct dpu_hw_pingpong, base); +} + +/** * dpu_hw_pingpong_init - initializes the pingpong driver for the passed * pingpong idx. * @idx: Pingpong index for which driver object is required diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c index cb08fafb1dc1..ce19f1d39367 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c @@ -138,16 +138,12 @@ static int _dpu_debugfs_show_regset32(struct seq_file *s, void *data) { struct dpu_debugfs_regset32 *regset = s->private; struct dpu_kms *dpu_kms = regset->dpu_kms; - struct drm_device *dev; - struct msm_drm_private *priv; void __iomem *base; uint32_t i, addr; if (!dpu_kms->mmio) return 0; - dev = dpu_kms->dev; - priv = dev->dev_private; base = dpu_kms->mmio + regset->offset; /* insert padding spaces, if needed */ @@ -228,6 +224,85 @@ static int dpu_kms_debugfs_init(struct msm_kms *kms, struct drm_minor *minor) } #endif +/* Global/shared object state funcs */ + +/* + * This is a helper that returns the private state currently in operation. + * Note that this would return the "old_state" if called in the atomic check + * path, and the "new_state" after the atomic swap has been done. + */ +struct dpu_global_state * +dpu_kms_get_existing_global_state(struct dpu_kms *dpu_kms) +{ + return to_dpu_global_state(dpu_kms->global_state.state); +} + +/* + * This acquires the modeset lock set aside for global state, creates + * a new duplicated private object state. + */ +struct dpu_global_state *dpu_kms_get_global_state(struct drm_atomic_state *s) +{ + struct msm_drm_private *priv = s->dev->dev_private; + struct dpu_kms *dpu_kms = to_dpu_kms(priv->kms); + struct drm_private_state *priv_state; + int ret; + + ret = drm_modeset_lock(&dpu_kms->global_state_lock, s->acquire_ctx); + if (ret) + return ERR_PTR(ret); + + priv_state = drm_atomic_get_private_obj_state(s, + &dpu_kms->global_state); + if (IS_ERR(priv_state)) + return ERR_CAST(priv_state); + + return to_dpu_global_state(priv_state); +} + +static struct drm_private_state * +dpu_kms_global_duplicate_state(struct drm_private_obj *obj) +{ + struct dpu_global_state *state; + + state = kmemdup(obj->state, sizeof(*state), GFP_KERNEL); + if (!state) + return NULL; + + __drm_atomic_helper_private_obj_duplicate_state(obj, &state->base); + + return &state->base; +} + +static void dpu_kms_global_destroy_state(struct drm_private_obj *obj, + struct drm_private_state *state) +{ + struct dpu_global_state *dpu_state = to_dpu_global_state(state); + + kfree(dpu_state); +} + +static const struct drm_private_state_funcs dpu_kms_global_state_funcs = { + .atomic_duplicate_state = dpu_kms_global_duplicate_state, + .atomic_destroy_state = dpu_kms_global_destroy_state, +}; + +static int dpu_kms_global_obj_init(struct dpu_kms *dpu_kms) +{ + struct dpu_global_state *state; + + drm_modeset_lock_init(&dpu_kms->global_state_lock); + + state = kzalloc(sizeof(*state), GFP_KERNEL); + if (!state) + return -ENOMEM; + + drm_atomic_private_obj_init(dpu_kms->dev, &dpu_kms->global_state, + &state->base, + &dpu_kms_global_state_funcs); + return 0; +} + static int dpu_kms_enable_vblank(struct msm_kms *kms, struct drm_crtc *crtc) { return dpu_crtc_vblank(crtc, true); @@ -267,8 +342,6 @@ static ktime_t dpu_kms_vsync_time(struct msm_kms *kms, struct drm_crtc *crtc) static void dpu_kms_prepare_commit(struct msm_kms *kms, struct drm_atomic_state *state) { - struct dpu_kms *dpu_kms; - struct drm_device *dev; struct drm_crtc *crtc; struct drm_crtc_state *crtc_state; struct drm_encoder *encoder; @@ -276,8 +349,6 @@ static void dpu_kms_prepare_commit(struct msm_kms *kms, if (!kms) return; - dpu_kms = to_dpu_kms(kms); - dev = dpu_kms->dev; /* Call prepare_commit for all affected encoders */ for_each_new_crtc_in_state(state, crtc, crtc_state, i) { @@ -552,11 +623,8 @@ static long dpu_kms_round_pixclk(struct msm_kms *kms, unsigned long rate, static void _dpu_kms_hw_destroy(struct dpu_kms *dpu_kms) { - struct drm_device *dev; int i; - dev = dpu_kms->dev; - if (dpu_kms->hw_intr) dpu_hw_intr_destroy(dpu_kms->hw_intr); dpu_kms->hw_intr = NULL; @@ -760,7 +828,6 @@ static int dpu_kms_hw_init(struct msm_kms *kms) { struct dpu_kms *dpu_kms; struct drm_device *dev; - struct msm_drm_private *priv; int i, rc = -EINVAL; if (!kms) { @@ -770,7 +837,10 @@ static int dpu_kms_hw_init(struct msm_kms *kms) dpu_kms = to_dpu_kms(kms); dev = dpu_kms->dev; - priv = dev->dev_private; + + rc = dpu_kms_global_obj_init(dpu_kms); + if (rc) + return rc; atomic_set(&dpu_kms->bandwidth_ref, 0); @@ -1018,10 +1088,8 @@ static int __maybe_unused dpu_runtime_suspend(struct device *dev) int rc = -1; struct platform_device *pdev = to_platform_device(dev); struct dpu_kms *dpu_kms = platform_get_drvdata(pdev); - struct drm_device *ddev; struct dss_module_power *mp = &dpu_kms->mp; - ddev = dpu_kms->dev; rc = msm_dss_enable_clk(mp->clk_config, mp->num_clk, false); if (rc) DPU_ERROR("clock disable failed rc:%d\n", rc); diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h index c6169e7df19d..211f5de99a44 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h @@ -111,6 +111,13 @@ struct dpu_kms { struct dpu_core_perf perf; + /* + * Global private object state, Do not access directly, use + * dpu_kms_global_get_state() + */ + struct drm_modeset_lock global_state_lock; + struct drm_private_obj global_state; + struct dpu_rm rm; bool rm_init; @@ -139,6 +146,25 @@ struct vsync_info { #define to_dpu_kms(x) container_of(x, struct dpu_kms, base) +#define to_dpu_global_state(x) container_of(x, struct dpu_global_state, base) + +/* Global private object state for tracking resources that are shared across + * multiple kms objects (planes/crtcs/etc). + */ +struct dpu_global_state { + struct drm_private_state base; + + uint32_t pingpong_to_enc_id[PINGPONG_MAX - PINGPONG_0]; + uint32_t mixer_to_enc_id[LM_MAX - LM_0]; + uint32_t ctl_to_enc_id[CTL_MAX - CTL_0]; + uint32_t intf_to_enc_id[INTF_MAX - INTF_0]; +}; + +struct dpu_global_state + *dpu_kms_get_existing_global_state(struct dpu_kms *dpu_kms); +struct dpu_global_state + *__must_check dpu_kms_get_global_state(struct drm_atomic_state *s); + /** * Debugfs functions - extra helper functions for debugfs support * diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_mdss.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_mdss.c index 29705e773a4b..80d3cfc14007 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_mdss.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_mdss.c @@ -12,6 +12,7 @@ #define to_dpu_mdss(x) container_of(x, struct dpu_mdss, base) +#define HW_REV 0x0 #define HW_INTR_STATUS 0x0010 /* Max BW defined in KBps */ @@ -22,6 +23,17 @@ struct dpu_irq_controller { struct irq_domain *domain; }; +struct dpu_hw_cfg { + u32 val; + u32 offset; +}; + +struct dpu_mdss_hw_init_handler { + u32 hw_rev; + u32 hw_reg_count; + struct dpu_hw_cfg* hw_cfg; +}; + struct dpu_mdss { struct msm_mdss base; void __iomem *mmio; @@ -32,6 +44,44 @@ struct dpu_mdss { u32 num_paths; }; +static struct dpu_hw_cfg hw_cfg[] = { + { + /* UBWC global settings */ + .val = 0x1E, + .offset = 0x144, + } +}; + +static struct dpu_mdss_hw_init_handler cfg_handler[] = { + { .hw_rev = DPU_HW_VER_620, + .hw_reg_count = ARRAY_SIZE(hw_cfg), + .hw_cfg = hw_cfg + }, +}; + +static void dpu_mdss_hw_init(struct dpu_mdss *dpu_mdss, u32 hw_rev) +{ + int i; + u32 count = 0; + struct dpu_hw_cfg *hw_cfg = NULL; + + for (i = 0; i < ARRAY_SIZE(cfg_handler); i++) { + if (cfg_handler[i].hw_rev == hw_rev) { + hw_cfg = cfg_handler[i].hw_cfg; + count = cfg_handler[i].hw_reg_count; + break; + } + } + + for (i = 0; i < count; i++ ) { + writel_relaxed(hw_cfg->val, + dpu_mdss->mmio + hw_cfg->offset); + hw_cfg++; + } + + return; +} + static int dpu_mdss_parse_data_bus_icc_path(struct drm_device *dev, struct dpu_mdss *dpu_mdss) { @@ -174,12 +224,18 @@ static int dpu_mdss_enable(struct msm_mdss *mdss) struct dpu_mdss *dpu_mdss = to_dpu_mdss(mdss); struct dss_module_power *mp = &dpu_mdss->mp; int ret; + u32 mdss_rev; dpu_mdss_icc_request_bw(mdss); ret = msm_dss_enable_clk(mp->clk_config, mp->num_clk, true); - if (ret) + if (ret) { DPU_ERROR("clock enable failed, ret:%d\n", ret); + return ret; + } + + mdss_rev = readl_relaxed(dpu_mdss->mmio + HW_REV); + dpu_mdss_hw_init(dpu_mdss, mdss_rev); return ret; } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c index 23f5b1433b35..9b62451b01ee 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c @@ -12,8 +12,12 @@ #include "dpu_encoder.h" #include "dpu_trace.h" -#define RESERVED_BY_OTHER(h, r) \ - ((h)->enc_id && (h)->enc_id != r) + +static inline bool reserved_by_other(uint32_t *res_map, int idx, + uint32_t enc_id) +{ + return res_map[idx] && res_map[idx] != enc_id; +} /** * struct dpu_rm_requirements - Reservation requirements parameter bundle @@ -25,171 +29,43 @@ struct dpu_rm_requirements { struct dpu_encoder_hw_resources hw_res; }; - -/** - * struct dpu_rm_hw_blk - hardware block tracking list member - * @list: List head for list of all hardware blocks tracking items - * @id: Hardware ID number, within it's own space, ie. LM_X - * @enc_id: Encoder id to which this blk is binded - * @hw: Pointer to the hardware register access object for this block - */ -struct dpu_rm_hw_blk { - struct list_head list; - uint32_t id; - uint32_t enc_id; - struct dpu_hw_blk *hw; -}; - -void dpu_rm_init_hw_iter( - struct dpu_rm_hw_iter *iter, - uint32_t enc_id, - enum dpu_hw_blk_type type) -{ - memset(iter, 0, sizeof(*iter)); - iter->enc_id = enc_id; - iter->type = type; -} - -static bool _dpu_rm_get_hw_locked(struct dpu_rm *rm, struct dpu_rm_hw_iter *i) +int dpu_rm_destroy(struct dpu_rm *rm) { - struct list_head *blk_list; - - if (!rm || !i || i->type >= DPU_HW_BLK_MAX) { - DPU_ERROR("invalid rm\n"); - return false; - } + int i; - i->hw = NULL; - blk_list = &rm->hw_blks[i->type]; + for (i = 0; i < ARRAY_SIZE(rm->pingpong_blks); i++) { + struct dpu_hw_pingpong *hw; - if (i->blk && (&i->blk->list == blk_list)) { - DPU_DEBUG("attempt resume iteration past last\n"); - return false; - } - - i->blk = list_prepare_entry(i->blk, blk_list, list); - - list_for_each_entry_continue(i->blk, blk_list, list) { - if (i->enc_id == i->blk->enc_id) { - i->hw = i->blk->hw; - DPU_DEBUG("found type %d id %d for enc %d\n", - i->type, i->blk->id, i->enc_id); - return true; + if (rm->pingpong_blks[i]) { + hw = to_dpu_hw_pingpong(rm->pingpong_blks[i]); + dpu_hw_pingpong_destroy(hw); } } + for (i = 0; i < ARRAY_SIZE(rm->mixer_blks); i++) { + struct dpu_hw_mixer *hw; - DPU_DEBUG("no match, type %d for enc %d\n", i->type, i->enc_id); - - return false; -} - -bool dpu_rm_get_hw(struct dpu_rm *rm, struct dpu_rm_hw_iter *i) -{ - bool ret; - - mutex_lock(&rm->rm_lock); - ret = _dpu_rm_get_hw_locked(rm, i); - mutex_unlock(&rm->rm_lock); - - return ret; -} - -static void _dpu_rm_hw_destroy(enum dpu_hw_blk_type type, void *hw) -{ - switch (type) { - case DPU_HW_BLK_LM: - dpu_hw_lm_destroy(hw); - break; - case DPU_HW_BLK_CTL: - dpu_hw_ctl_destroy(hw); - break; - case DPU_HW_BLK_PINGPONG: - dpu_hw_pingpong_destroy(hw); - break; - case DPU_HW_BLK_INTF: - dpu_hw_intf_destroy(hw); - break; - case DPU_HW_BLK_SSPP: - /* SSPPs are not managed by the resource manager */ - case DPU_HW_BLK_TOP: - /* Top is a singleton, not managed in hw_blks list */ - case DPU_HW_BLK_MAX: - default: - DPU_ERROR("unsupported block type %d\n", type); - break; - } -} - -int dpu_rm_destroy(struct dpu_rm *rm) -{ - struct dpu_rm_hw_blk *hw_cur, *hw_nxt; - enum dpu_hw_blk_type type; - - for (type = 0; type < DPU_HW_BLK_MAX; type++) { - list_for_each_entry_safe(hw_cur, hw_nxt, &rm->hw_blks[type], - list) { - list_del(&hw_cur->list); - _dpu_rm_hw_destroy(type, hw_cur->hw); - kfree(hw_cur); + if (rm->mixer_blks[i]) { + hw = to_dpu_hw_mixer(rm->mixer_blks[i]); + dpu_hw_lm_destroy(hw); } } + for (i = 0; i < ARRAY_SIZE(rm->ctl_blks); i++) { + struct dpu_hw_ctl *hw; - mutex_destroy(&rm->rm_lock); - - return 0; -} - -static int _dpu_rm_hw_blk_create( - struct dpu_rm *rm, - const struct dpu_mdss_cfg *cat, - void __iomem *mmio, - enum dpu_hw_blk_type type, - uint32_t id, - const void *hw_catalog_info) -{ - struct dpu_rm_hw_blk *blk; - void *hw; - - switch (type) { - case DPU_HW_BLK_LM: - hw = dpu_hw_lm_init(id, mmio, cat); - break; - case DPU_HW_BLK_CTL: - hw = dpu_hw_ctl_init(id, mmio, cat); - break; - case DPU_HW_BLK_PINGPONG: - hw = dpu_hw_pingpong_init(id, mmio, cat); - break; - case DPU_HW_BLK_INTF: - hw = dpu_hw_intf_init(id, mmio, cat); - break; - case DPU_HW_BLK_SSPP: - /* SSPPs are not managed by the resource manager */ - case DPU_HW_BLK_TOP: - /* Top is a singleton, not managed in hw_blks list */ - case DPU_HW_BLK_MAX: - default: - DPU_ERROR("unsupported block type %d\n", type); - return -EINVAL; - } - - if (IS_ERR_OR_NULL(hw)) { - DPU_ERROR("failed hw object creation: type %d, err %ld\n", - type, PTR_ERR(hw)); - return -EFAULT; + if (rm->ctl_blks[i]) { + hw = to_dpu_hw_ctl(rm->ctl_blks[i]); + dpu_hw_ctl_destroy(hw); + } } + for (i = 0; i < ARRAY_SIZE(rm->intf_blks); i++) { + struct dpu_hw_intf *hw; - blk = kzalloc(sizeof(*blk), GFP_KERNEL); - if (!blk) { - _dpu_rm_hw_destroy(type, hw); - return -ENOMEM; + if (rm->intf_blks[i]) { + hw = to_dpu_hw_intf(rm->intf_blks[i]); + dpu_hw_intf_destroy(hw); + } } - blk->id = id; - blk->hw = hw; - blk->enc_id = 0; - list_add_tail(&blk->list, &rm->hw_blks[type]); - return 0; } @@ -198,7 +74,6 @@ int dpu_rm_init(struct dpu_rm *rm, void __iomem *mmio) { int rc, i; - enum dpu_hw_blk_type type; if (!rm || !cat || !mmio) { DPU_ERROR("invalid kms\n"); @@ -208,13 +83,9 @@ int dpu_rm_init(struct dpu_rm *rm, /* Clear, setup lists */ memset(rm, 0, sizeof(*rm)); - mutex_init(&rm->rm_lock); - - for (type = 0; type < DPU_HW_BLK_MAX; type++) - INIT_LIST_HEAD(&rm->hw_blks[type]); - /* Interrogate HW catalog and create tracking items for hw blocks */ for (i = 0; i < cat->mixer_count; i++) { + struct dpu_hw_mixer *hw; const struct dpu_lm_cfg *lm = &cat->mixer[i]; if (lm->pingpong == PINGPONG_MAX) { @@ -222,12 +93,17 @@ int dpu_rm_init(struct dpu_rm *rm, continue; } - rc = _dpu_rm_hw_blk_create(rm, cat, mmio, DPU_HW_BLK_LM, - cat->mixer[i].id, &cat->mixer[i]); - if (rc) { - DPU_ERROR("failed: lm hw not available\n"); + if (lm->id < LM_0 || lm->id >= LM_MAX) { + DPU_ERROR("skip mixer %d with invalid id\n", lm->id); + continue; + } + hw = dpu_hw_lm_init(lm->id, mmio, cat); + if (IS_ERR_OR_NULL(hw)) { + rc = PTR_ERR(hw); + DPU_ERROR("failed lm object creation: err %d\n", rc); goto fail; } + rm->mixer_blks[lm->id - LM_0] = &hw->base; if (!rm->lm_max_width) { rm->lm_max_width = lm->sblk->maxwidth; @@ -243,35 +119,59 @@ int dpu_rm_init(struct dpu_rm *rm, } for (i = 0; i < cat->pingpong_count; i++) { - rc = _dpu_rm_hw_blk_create(rm, cat, mmio, DPU_HW_BLK_PINGPONG, - cat->pingpong[i].id, &cat->pingpong[i]); - if (rc) { - DPU_ERROR("failed: pp hw not available\n"); + struct dpu_hw_pingpong *hw; + const struct dpu_pingpong_cfg *pp = &cat->pingpong[i]; + + if (pp->id < PINGPONG_0 || pp->id >= PINGPONG_MAX) { + DPU_ERROR("skip pingpong %d with invalid id\n", pp->id); + continue; + } + hw = dpu_hw_pingpong_init(pp->id, mmio, cat); + if (IS_ERR_OR_NULL(hw)) { + rc = PTR_ERR(hw); + DPU_ERROR("failed pingpong object creation: err %d\n", + rc); goto fail; } + rm->pingpong_blks[pp->id - PINGPONG_0] = &hw->base; } for (i = 0; i < cat->intf_count; i++) { - if (cat->intf[i].type == INTF_NONE) { + struct dpu_hw_intf *hw; + const struct dpu_intf_cfg *intf = &cat->intf[i]; + + if (intf->type == INTF_NONE) { DPU_DEBUG("skip intf %d with type none\n", i); continue; } - - rc = _dpu_rm_hw_blk_create(rm, cat, mmio, DPU_HW_BLK_INTF, - cat->intf[i].id, &cat->intf[i]); - if (rc) { - DPU_ERROR("failed: intf hw not available\n"); + if (intf->id < INTF_0 || intf->id >= INTF_MAX) { + DPU_ERROR("skip intf %d with invalid id\n", intf->id); + continue; + } + hw = dpu_hw_intf_init(intf->id, mmio, cat); + if (IS_ERR_OR_NULL(hw)) { + rc = PTR_ERR(hw); + DPU_ERROR("failed intf object creation: err %d\n", rc); goto fail; } + rm->intf_blks[intf->id - INTF_0] = &hw->base; } for (i = 0; i < cat->ctl_count; i++) { - rc = _dpu_rm_hw_blk_create(rm, cat, mmio, DPU_HW_BLK_CTL, - cat->ctl[i].id, &cat->ctl[i]); - if (rc) { - DPU_ERROR("failed: ctl hw not available\n"); + struct dpu_hw_ctl *hw; + const struct dpu_ctl_cfg *ctl = &cat->ctl[i]; + + if (ctl->id < CTL_0 || ctl->id >= CTL_MAX) { + DPU_ERROR("skip ctl %d with invalid id\n", ctl->id); + continue; + } + hw = dpu_hw_ctl_init(ctl->id, mmio, cat); + if (IS_ERR_OR_NULL(hw)) { + rc = PTR_ERR(hw); + DPU_ERROR("failed ctl object creation: err %d\n", rc); goto fail; } + rm->ctl_blks[ctl->id - CTL_0] = &hw->base; } return 0; @@ -279,7 +179,7 @@ int dpu_rm_init(struct dpu_rm *rm, fail: dpu_rm_destroy(rm); - return rc; + return rc ? rc : -EFAULT; } static bool _dpu_rm_needs_split_display(const struct msm_display_topology *top) @@ -288,85 +188,81 @@ static bool _dpu_rm_needs_split_display(const struct msm_display_topology *top) } /** + * _dpu_rm_check_lm_peer - check if a mixer is a peer of the primary + * @rm: dpu resource manager handle + * @primary_idx: index of primary mixer in rm->mixer_blks[] + * @peer_idx: index of other mixer in rm->mixer_blks[] + * @Return: true if rm->mixer_blks[peer_idx] is a peer of + * rm->mixer_blks[primary_idx] + */ +static bool _dpu_rm_check_lm_peer(struct dpu_rm *rm, int primary_idx, + int peer_idx) +{ + const struct dpu_lm_cfg *prim_lm_cfg; + const struct dpu_lm_cfg *peer_cfg; + + prim_lm_cfg = to_dpu_hw_mixer(rm->mixer_blks[primary_idx])->cap; + peer_cfg = to_dpu_hw_mixer(rm->mixer_blks[peer_idx])->cap; + + if (!test_bit(peer_cfg->id, &prim_lm_cfg->lm_pair_mask)) { + DPU_DEBUG("lm %d not peer of lm %d\n", peer_cfg->id, + peer_cfg->id); + return false; + } + return true; +} + +/** * _dpu_rm_check_lm_and_get_connected_blks - check if proposed layer mixer meets * proposed use case requirements, incl. hardwired dependent blocks like * pingpong * @rm: dpu resource manager handle * @enc_id: encoder id requesting for allocation - * @reqs: proposed use case requirements - * @lm: proposed layer mixer, function checks if lm, and all other hardwired - * blocks connected to the lm (pp) is available and appropriate - * @pp: output parameter, pingpong block attached to the layer mixer. - * NULL if pp was not available, or not matching requirements. - * @primary_lm: if non-null, this function check if lm is compatible primary_lm - * as well as satisfying all other requirements + * @lm_idx: index of proposed layer mixer in rm->mixer_blks[], function checks + * if lm, and all other hardwired blocks connected to the lm (pp) is + * available and appropriate + * @pp_idx: output parameter, index of pingpong block attached to the layer + * mixer in rm->pongpong_blks[]. * @Return: true if lm matches all requirements, false otherwise */ -static bool _dpu_rm_check_lm_and_get_connected_blks( - struct dpu_rm *rm, - uint32_t enc_id, - struct dpu_rm_requirements *reqs, - struct dpu_rm_hw_blk *lm, - struct dpu_rm_hw_blk **pp, - struct dpu_rm_hw_blk *primary_lm) +static bool _dpu_rm_check_lm_and_get_connected_blks(struct dpu_rm *rm, + struct dpu_global_state *global_state, + uint32_t enc_id, int lm_idx, int *pp_idx) { - const struct dpu_lm_cfg *lm_cfg = to_dpu_hw_mixer(lm->hw)->cap; - struct dpu_rm_hw_iter iter; - - *pp = NULL; - - DPU_DEBUG("check lm %d pp %d\n", - lm_cfg->id, lm_cfg->pingpong); - - /* Check if this layer mixer is a peer of the proposed primary LM */ - if (primary_lm) { - const struct dpu_lm_cfg *prim_lm_cfg = - to_dpu_hw_mixer(primary_lm->hw)->cap; - - if (!test_bit(lm_cfg->id, &prim_lm_cfg->lm_pair_mask)) { - DPU_DEBUG("lm %d not peer of lm %d\n", lm_cfg->id, - prim_lm_cfg->id); - return false; - } - } + const struct dpu_lm_cfg *lm_cfg; + int idx; /* Already reserved? */ - if (RESERVED_BY_OTHER(lm, enc_id)) { - DPU_DEBUG("lm %d already reserved\n", lm_cfg->id); + if (reserved_by_other(global_state->mixer_to_enc_id, lm_idx, enc_id)) { + DPU_DEBUG("lm %d already reserved\n", lm_idx + LM_0); return false; } - dpu_rm_init_hw_iter(&iter, 0, DPU_HW_BLK_PINGPONG); - while (_dpu_rm_get_hw_locked(rm, &iter)) { - if (iter.blk->id == lm_cfg->pingpong) { - *pp = iter.blk; - break; - } - } - - if (!*pp) { + lm_cfg = to_dpu_hw_mixer(rm->mixer_blks[lm_idx])->cap; + idx = lm_cfg->pingpong - PINGPONG_0; + if (idx < 0 || idx >= ARRAY_SIZE(rm->pingpong_blks)) { DPU_ERROR("failed to get pp on lm %d\n", lm_cfg->pingpong); return false; } - if (RESERVED_BY_OTHER(*pp, enc_id)) { - DPU_DEBUG("lm %d pp %d already reserved\n", lm->id, - (*pp)->id); + if (reserved_by_other(global_state->pingpong_to_enc_id, idx, enc_id)) { + DPU_DEBUG("lm %d pp %d already reserved\n", lm_cfg->id, + lm_cfg->pingpong); return false; } - + *pp_idx = idx; return true; } -static int _dpu_rm_reserve_lms(struct dpu_rm *rm, uint32_t enc_id, +static int _dpu_rm_reserve_lms(struct dpu_rm *rm, + struct dpu_global_state *global_state, + uint32_t enc_id, struct dpu_rm_requirements *reqs) { - struct dpu_rm_hw_blk *lm[MAX_BLOCKS]; - struct dpu_rm_hw_blk *pp[MAX_BLOCKS]; - struct dpu_rm_hw_iter iter_i, iter_j; - int lm_count = 0; - int i, rc = 0; + int lm_idx[MAX_BLOCKS]; + int pp_idx[MAX_BLOCKS]; + int i, j, lm_count = 0; if (!reqs->topology.num_lm) { DPU_ERROR("invalid number of lm: %d\n", reqs->topology.num_lm); @@ -374,36 +270,40 @@ static int _dpu_rm_reserve_lms(struct dpu_rm *rm, uint32_t enc_id, } /* Find a primary mixer */ - dpu_rm_init_hw_iter(&iter_i, 0, DPU_HW_BLK_LM); - while (lm_count != reqs->topology.num_lm && - _dpu_rm_get_hw_locked(rm, &iter_i)) { - memset(&lm, 0, sizeof(lm)); - memset(&pp, 0, sizeof(pp)); + for (i = 0; i < ARRAY_SIZE(rm->mixer_blks) && + lm_count < reqs->topology.num_lm; i++) { + if (!rm->mixer_blks[i]) + continue; lm_count = 0; - lm[lm_count] = iter_i.blk; + lm_idx[lm_count] = i; - if (!_dpu_rm_check_lm_and_get_connected_blks( - rm, enc_id, reqs, lm[lm_count], - &pp[lm_count], NULL)) + if (!_dpu_rm_check_lm_and_get_connected_blks(rm, global_state, + enc_id, i, &pp_idx[lm_count])) { continue; + } ++lm_count; /* Valid primary mixer found, find matching peers */ - dpu_rm_init_hw_iter(&iter_j, 0, DPU_HW_BLK_LM); + for (j = i + 1; j < ARRAY_SIZE(rm->mixer_blks) && + lm_count < reqs->topology.num_lm; j++) { + if (!rm->mixer_blks[j]) + continue; - while (lm_count != reqs->topology.num_lm && - _dpu_rm_get_hw_locked(rm, &iter_j)) { - if (iter_i.blk == iter_j.blk) + if (!_dpu_rm_check_lm_peer(rm, i, j)) { + DPU_DEBUG("lm %d not peer of lm %d\n", LM_0 + j, + LM_0 + i); continue; + } - if (!_dpu_rm_check_lm_and_get_connected_blks( - rm, enc_id, reqs, iter_j.blk, - &pp[lm_count], iter_i.blk)) + if (!_dpu_rm_check_lm_and_get_connected_blks(rm, + global_state, enc_id, j, + &pp_idx[lm_count])) { continue; + } - lm[lm_count] = iter_j.blk; + lm_idx[lm_count] = j; ++lm_count; } } @@ -413,65 +313,65 @@ static int _dpu_rm_reserve_lms(struct dpu_rm *rm, uint32_t enc_id, return -ENAVAIL; } - for (i = 0; i < ARRAY_SIZE(lm); i++) { - if (!lm[i]) - break; + for (i = 0; i < lm_count; i++) { + global_state->mixer_to_enc_id[lm_idx[i]] = enc_id; + global_state->pingpong_to_enc_id[pp_idx[i]] = enc_id; - lm[i]->enc_id = enc_id; - pp[i]->enc_id = enc_id; - - trace_dpu_rm_reserve_lms(lm[i]->id, enc_id, pp[i]->id); + trace_dpu_rm_reserve_lms(lm_idx[i] + LM_0, enc_id, + pp_idx[i] + PINGPONG_0); } - return rc; + return 0; } static int _dpu_rm_reserve_ctls( struct dpu_rm *rm, + struct dpu_global_state *global_state, uint32_t enc_id, const struct msm_display_topology *top) { - struct dpu_rm_hw_blk *ctls[MAX_BLOCKS]; - struct dpu_rm_hw_iter iter; - int i = 0, num_ctls = 0; - bool needs_split_display = false; - - memset(&ctls, 0, sizeof(ctls)); + int ctl_idx[MAX_BLOCKS]; + int i = 0, j, num_ctls; + bool needs_split_display; /* each hw_intf needs its own hw_ctrl to program its control path */ num_ctls = top->num_intf; needs_split_display = _dpu_rm_needs_split_display(top); - dpu_rm_init_hw_iter(&iter, 0, DPU_HW_BLK_CTL); - while (_dpu_rm_get_hw_locked(rm, &iter)) { - const struct dpu_hw_ctl *ctl = to_dpu_hw_ctl(iter.blk->hw); - unsigned long features = ctl->caps->features; + for (j = 0; j < ARRAY_SIZE(rm->ctl_blks); j++) { + const struct dpu_hw_ctl *ctl; + unsigned long features; bool has_split_display; - if (RESERVED_BY_OTHER(iter.blk, enc_id)) + if (!rm->ctl_blks[j]) + continue; + if (reserved_by_other(global_state->ctl_to_enc_id, j, enc_id)) continue; + ctl = to_dpu_hw_ctl(rm->ctl_blks[j]); + features = ctl->caps->features; has_split_display = BIT(DPU_CTL_SPLIT_DISPLAY) & features; - DPU_DEBUG("ctl %d caps 0x%lX\n", iter.blk->id, features); + DPU_DEBUG("ctl %d caps 0x%lX\n", rm->ctl_blks[j]->id, features); if (needs_split_display != has_split_display) continue; - ctls[i] = iter.blk; - DPU_DEBUG("ctl %d match\n", iter.blk->id); + ctl_idx[i] = j; + DPU_DEBUG("ctl %d match\n", j + CTL_0); if (++i == num_ctls) break; + } if (i != num_ctls) return -ENAVAIL; - for (i = 0; i < ARRAY_SIZE(ctls) && i < num_ctls; i++) { - ctls[i]->enc_id = enc_id; - trace_dpu_rm_reserve_ctls(ctls[i]->id, enc_id); + for (i = 0; i < ARRAY_SIZE(ctl_idx) && i < num_ctls; i++) { + global_state->ctl_to_enc_id[ctl_idx[i]] = enc_id; + trace_dpu_rm_reserve_ctls(i + CTL_0, enc_id); } return 0; @@ -479,40 +379,34 @@ static int _dpu_rm_reserve_ctls( static int _dpu_rm_reserve_intf( struct dpu_rm *rm, + struct dpu_global_state *global_state, uint32_t enc_id, - uint32_t id, - enum dpu_hw_blk_type type) + uint32_t id) { - struct dpu_rm_hw_iter iter; - int ret = 0; - - /* Find the block entry in the rm, and note the reservation */ - dpu_rm_init_hw_iter(&iter, 0, type); - while (_dpu_rm_get_hw_locked(rm, &iter)) { - if (iter.blk->id != id) - continue; + int idx = id - INTF_0; - if (RESERVED_BY_OTHER(iter.blk, enc_id)) { - DPU_ERROR("type %d id %d already reserved\n", type, id); - return -ENAVAIL; - } - - iter.blk->enc_id = enc_id; - trace_dpu_rm_reserve_intf(iter.blk->id, enc_id); - break; + if (idx < 0 || idx >= ARRAY_SIZE(rm->intf_blks)) { + DPU_ERROR("invalid intf id: %d", id); + return -EINVAL; } - /* Shouldn't happen since intfs are fixed at probe */ - if (!iter.hw) { - DPU_ERROR("couldn't find type %d id %d\n", type, id); + if (!rm->intf_blks[idx]) { + DPU_ERROR("couldn't find intf id %d\n", id); return -EINVAL; } - return ret; + if (reserved_by_other(global_state->intf_to_enc_id, idx, enc_id)) { + DPU_ERROR("intf id %d already reserved\n", id); + return -ENAVAIL; + } + + global_state->intf_to_enc_id[idx] = enc_id; + return 0; } static int _dpu_rm_reserve_intf_related_hw( struct dpu_rm *rm, + struct dpu_global_state *global_state, uint32_t enc_id, struct dpu_encoder_hw_resources *hw_res) { @@ -523,8 +417,7 @@ static int _dpu_rm_reserve_intf_related_hw( if (hw_res->intfs[i] == INTF_MODE_NONE) continue; id = i + INTF_0; - ret = _dpu_rm_reserve_intf(rm, enc_id, id, - DPU_HW_BLK_INTF); + ret = _dpu_rm_reserve_intf(rm, global_state, enc_id, id); if (ret) return ret; } @@ -534,25 +427,27 @@ static int _dpu_rm_reserve_intf_related_hw( static int _dpu_rm_make_reservation( struct dpu_rm *rm, + struct dpu_global_state *global_state, struct drm_encoder *enc, - struct drm_crtc_state *crtc_state, struct dpu_rm_requirements *reqs) { int ret; - ret = _dpu_rm_reserve_lms(rm, enc->base.id, reqs); + ret = _dpu_rm_reserve_lms(rm, global_state, enc->base.id, reqs); if (ret) { DPU_ERROR("unable to find appropriate mixers\n"); return ret; } - ret = _dpu_rm_reserve_ctls(rm, enc->base.id, &reqs->topology); + ret = _dpu_rm_reserve_ctls(rm, global_state, enc->base.id, + &reqs->topology); if (ret) { DPU_ERROR("unable to find appropriate CTL\n"); return ret; } - ret = _dpu_rm_reserve_intf_related_hw(rm, enc->base.id, &reqs->hw_res); + ret = _dpu_rm_reserve_intf_related_hw(rm, global_state, enc->base.id, + &reqs->hw_res); if (ret) return ret; @@ -560,9 +455,7 @@ static int _dpu_rm_make_reservation( } static int _dpu_rm_populate_requirements( - struct dpu_rm *rm, struct drm_encoder *enc, - struct drm_crtc_state *crtc_state, struct dpu_rm_requirements *reqs, struct msm_display_topology req_topology) { @@ -577,37 +470,36 @@ static int _dpu_rm_populate_requirements( return 0; } -static void _dpu_rm_release_reservation(struct dpu_rm *rm, uint32_t enc_id) +static void _dpu_rm_clear_mapping(uint32_t *res_mapping, int cnt, + uint32_t enc_id) { - struct dpu_rm_hw_blk *blk; - enum dpu_hw_blk_type type; - - for (type = 0; type < DPU_HW_BLK_MAX; type++) { - list_for_each_entry(blk, &rm->hw_blks[type], list) { - if (blk->enc_id == enc_id) { - blk->enc_id = 0; - DPU_DEBUG("rel enc %d %d %d\n", enc_id, - type, blk->id); - } - } + int i; + + for (i = 0; i < cnt; i++) { + if (res_mapping[i] == enc_id) + res_mapping[i] = 0; } } -void dpu_rm_release(struct dpu_rm *rm, struct drm_encoder *enc) +void dpu_rm_release(struct dpu_global_state *global_state, + struct drm_encoder *enc) { - mutex_lock(&rm->rm_lock); - - _dpu_rm_release_reservation(rm, enc->base.id); - - mutex_unlock(&rm->rm_lock); + _dpu_rm_clear_mapping(global_state->pingpong_to_enc_id, + ARRAY_SIZE(global_state->pingpong_to_enc_id), enc->base.id); + _dpu_rm_clear_mapping(global_state->mixer_to_enc_id, + ARRAY_SIZE(global_state->mixer_to_enc_id), enc->base.id); + _dpu_rm_clear_mapping(global_state->ctl_to_enc_id, + ARRAY_SIZE(global_state->ctl_to_enc_id), enc->base.id); + _dpu_rm_clear_mapping(global_state->intf_to_enc_id, + ARRAY_SIZE(global_state->intf_to_enc_id), enc->base.id); } int dpu_rm_reserve( struct dpu_rm *rm, + struct dpu_global_state *global_state, struct drm_encoder *enc, struct drm_crtc_state *crtc_state, - struct msm_display_topology topology, - bool test_only) + struct msm_display_topology topology) { struct dpu_rm_requirements reqs; int ret; @@ -616,31 +508,75 @@ int dpu_rm_reserve( if (!drm_atomic_crtc_needs_modeset(crtc_state)) return 0; - DRM_DEBUG_KMS("reserving hw for enc %d crtc %d test_only %d\n", - enc->base.id, crtc_state->crtc->base.id, test_only); + if (IS_ERR(global_state)) { + DPU_ERROR("failed to global state\n"); + return PTR_ERR(global_state); + } - mutex_lock(&rm->rm_lock); + DRM_DEBUG_KMS("reserving hw for enc %d crtc %d\n", + enc->base.id, crtc_state->crtc->base.id); - ret = _dpu_rm_populate_requirements(rm, enc, crtc_state, &reqs, - topology); + ret = _dpu_rm_populate_requirements(enc, &reqs, topology); if (ret) { DPU_ERROR("failed to populate hw requirements\n"); - goto end; + return ret; } - ret = _dpu_rm_make_reservation(rm, enc, crtc_state, &reqs); - if (ret) { + ret = _dpu_rm_make_reservation(rm, global_state, enc, &reqs); + if (ret) DPU_ERROR("failed to reserve hw resources: %d\n", ret); - _dpu_rm_release_reservation(rm, enc->base.id); - } else if (test_only) { - /* test_only: test the reservation and then undo */ - DPU_DEBUG("test_only: discard test [enc: %d]\n", - enc->base.id); - _dpu_rm_release_reservation(rm, enc->base.id); - } -end: - mutex_unlock(&rm->rm_lock); + return ret; } + +int dpu_rm_get_assigned_resources(struct dpu_rm *rm, + struct dpu_global_state *global_state, uint32_t enc_id, + enum dpu_hw_blk_type type, struct dpu_hw_blk **blks, int blks_size) +{ + struct dpu_hw_blk **hw_blks; + uint32_t *hw_to_enc_id; + int i, num_blks, max_blks; + + switch (type) { + case DPU_HW_BLK_PINGPONG: + hw_blks = rm->pingpong_blks; + hw_to_enc_id = global_state->pingpong_to_enc_id; + max_blks = ARRAY_SIZE(rm->pingpong_blks); + break; + case DPU_HW_BLK_LM: + hw_blks = rm->mixer_blks; + hw_to_enc_id = global_state->mixer_to_enc_id; + max_blks = ARRAY_SIZE(rm->mixer_blks); + break; + case DPU_HW_BLK_CTL: + hw_blks = rm->ctl_blks; + hw_to_enc_id = global_state->ctl_to_enc_id; + max_blks = ARRAY_SIZE(rm->ctl_blks); + break; + case DPU_HW_BLK_INTF: + hw_blks = rm->intf_blks; + hw_to_enc_id = global_state->intf_to_enc_id; + max_blks = ARRAY_SIZE(rm->intf_blks); + break; + default: + DPU_ERROR("blk type %d not managed by rm\n", type); + return 0; + } + + num_blks = 0; + for (i = 0; i < max_blks; i++) { + if (hw_to_enc_id[i] != enc_id) + continue; + + if (num_blks == blks_size) { + DPU_ERROR("More than %d resources assigned to enc %d\n", + blks_size, enc_id); + break; + } + blks[num_blks++] = hw_blks[i]; + } + + return num_blks; +} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.h index 9c580a017094..6d2b04f306f0 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.h @@ -11,37 +11,24 @@ #include "msm_kms.h" #include "dpu_hw_top.h" +struct dpu_global_state; + /** * struct dpu_rm - DPU dynamic hardware resource manager - * @hw_blks: array of lists of hardware resources present in the system, one - * list per type of hardware block + * @pingpong_blks: array of pingpong hardware resources + * @mixer_blks: array of layer mixer hardware resources + * @ctl_blks: array of ctl hardware resources + * @intf_blks: array of intf hardware resources * @lm_max_width: cached layer mixer maximum width * @rm_lock: resource manager mutex */ struct dpu_rm { - struct list_head hw_blks[DPU_HW_BLK_MAX]; - uint32_t lm_max_width; - struct mutex rm_lock; -}; + struct dpu_hw_blk *pingpong_blks[PINGPONG_MAX - PINGPONG_0]; + struct dpu_hw_blk *mixer_blks[LM_MAX - LM_0]; + struct dpu_hw_blk *ctl_blks[CTL_MAX - CTL_0]; + struct dpu_hw_blk *intf_blks[INTF_MAX - INTF_0]; -/** - * struct dpu_rm_hw_blk - resource manager internal structure - * forward declaration for single iterator definition without void pointer - */ -struct dpu_rm_hw_blk; - -/** - * struct dpu_rm_hw_iter - iterator for use with dpu_rm - * @hw: dpu_hw object requested, or NULL on failure - * @blk: dpu_rm internal block representation. Clients ignore. Used as iterator. - * @enc_id: DRM ID of Encoder client wishes to search for, or 0 for Any Encoder - * @type: Hardware Block Type client wishes to search for. - */ -struct dpu_rm_hw_iter { - void *hw; - struct dpu_rm_hw_blk *blk; - uint32_t enc_id; - enum dpu_hw_blk_type type; + uint32_t lm_max_width; }; /** @@ -74,14 +61,13 @@ int dpu_rm_destroy(struct dpu_rm *rm); * @drm_enc: DRM Encoder handle * @crtc_state: Proposed Atomic DRM CRTC State handle * @topology: Pointer to topology info for the display - * @test_only: Atomic-Test phase, discard results (unless property overrides) * @Return: 0 on Success otherwise -ERROR */ int dpu_rm_reserve(struct dpu_rm *rm, + struct dpu_global_state *global_state, struct drm_encoder *drm_enc, struct drm_crtc_state *crtc_state, - struct msm_display_topology topology, - bool test_only); + struct msm_display_topology topology); /** * dpu_rm_reserve - Given the encoder for the display chain, release any @@ -90,31 +76,14 @@ int dpu_rm_reserve(struct dpu_rm *rm, * @enc: DRM Encoder handle * @Return: 0 on Success otherwise -ERROR */ -void dpu_rm_release(struct dpu_rm *rm, struct drm_encoder *enc); +void dpu_rm_release(struct dpu_global_state *global_state, + struct drm_encoder *enc); /** - * dpu_rm_init_hw_iter - setup given iterator for new iteration over hw list - * using dpu_rm_get_hw - * @iter: iter object to initialize - * @enc_id: DRM ID of Encoder client wishes to search for, or 0 for Any Encoder - * @type: Hardware Block Type client wishes to search for. - */ -void dpu_rm_init_hw_iter( - struct dpu_rm_hw_iter *iter, - uint32_t enc_id, - enum dpu_hw_blk_type type); -/** - * dpu_rm_get_hw - retrieve reserved hw object given encoder and hw type - * Meant to do a single pass through the hardware list to iteratively - * retrieve hardware blocks of a given type for a given encoder. - * Initialize an iterator object. - * Set hw block type of interest. Set encoder id of interest, 0 for any. - * Function returns first hw of type for that encoder. - * Subsequent calls will return the next reserved hw of that type in-order. - * Iterator HW pointer will be null on failure to find hw. - * @rm: DPU Resource Manager handle - * @iter: iterator object - * @Return: true on match found, false on no match found + * Get hw resources of the given type that are assigned to this encoder. */ -bool dpu_rm_get_hw(struct dpu_rm *rm, struct dpu_rm_hw_iter *iter); +int dpu_rm_get_assigned_resources(struct dpu_rm *rm, + struct dpu_global_state *global_state, uint32_t enc_id, + enum dpu_hw_blk_type type, struct dpu_hw_blk **blks, int blks_size); #endif /* __DPU_RM_H__ */ + diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_vbif.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_vbif.c index 93ab36bd8df3..5e8c3f3e6625 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_vbif.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_vbif.c @@ -24,7 +24,7 @@ static int _dpu_vbif_wait_for_xin_halt(struct dpu_hw_vbif *vbif, u32 xin_id) int rc; if (!vbif || !vbif->cap || !vbif->ops.get_halt_ctrl) { - DPU_ERROR("invalid arguments vbif %d\n", vbif != 0); + DPU_ERROR("invalid arguments vbif %d\n", vbif != NULL); return -EINVAL; } @@ -106,7 +106,7 @@ static u32 _dpu_vbif_get_ot_limit(struct dpu_hw_vbif *vbif, u32 val; if (!vbif || !vbif->cap) { - DPU_ERROR("invalid arguments vbif %d\n", vbif != 0); + DPU_ERROR("invalid arguments vbif %d\n", vbif != NULL); return -EINVAL; } @@ -164,7 +164,7 @@ void dpu_vbif_set_ot_limit(struct dpu_kms *dpu_kms, if (!vbif || !mdp) { DPU_DEBUG("invalid arguments vbif %d mdp %d\n", - vbif != 0, mdp != 0); + vbif != NULL, mdp != NULL); return; } diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_crtc.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_crtc.c index f34dca5d4532..c9239b07fe4f 100644 --- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_crtc.c +++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_crtc.c @@ -481,6 +481,8 @@ static const struct drm_crtc_funcs mdp4_crtc_funcs = { .reset = drm_atomic_helper_crtc_reset, .atomic_duplicate_state = drm_atomic_helper_crtc_duplicate_state, .atomic_destroy_state = drm_atomic_helper_crtc_destroy_state, + .enable_vblank = msm_crtc_enable_vblank, + .disable_vblank = msm_crtc_disable_vblank, }; static const struct drm_crtc_helper_funcs mdp4_crtc_helper_funcs = { diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_crtc.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_crtc.c index 05cc04f729d6..998bef1190a3 100644 --- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_crtc.c +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_crtc.c @@ -405,6 +405,83 @@ static void mdp5_crtc_mode_set_nofb(struct drm_crtc *crtc) spin_unlock_irqrestore(&mdp5_crtc->lm_lock, flags); } +static struct drm_encoder *get_encoder_from_crtc(struct drm_crtc *crtc) +{ + struct drm_device *dev = crtc->dev; + struct drm_encoder *encoder; + + drm_for_each_encoder(encoder, dev) + if (encoder->crtc == crtc) + return encoder; + + return NULL; +} + +static bool mdp5_crtc_get_scanout_position(struct drm_crtc *crtc, + bool in_vblank_irq, + int *vpos, int *hpos, + ktime_t *stime, ktime_t *etime, + const struct drm_display_mode *mode) +{ + unsigned int pipe = crtc->index; + struct drm_encoder *encoder; + int line, vsw, vbp, vactive_start, vactive_end, vfp_end; + + + encoder = get_encoder_from_crtc(crtc); + if (!encoder) { + DRM_ERROR("no encoder found for crtc %d\n", pipe); + return false; + } + + vsw = mode->crtc_vsync_end - mode->crtc_vsync_start; + vbp = mode->crtc_vtotal - mode->crtc_vsync_end; + + /* + * the line counter is 1 at the start of the VSYNC pulse and VTOTAL at + * the end of VFP. Translate the porch values relative to the line + * counter positions. + */ + + vactive_start = vsw + vbp + 1; + + vactive_end = vactive_start + mode->crtc_vdisplay; + + /* last scan line before VSYNC */ + vfp_end = mode->crtc_vtotal; + + if (stime) + *stime = ktime_get(); + + line = mdp5_encoder_get_linecount(encoder); + + if (line < vactive_start) + line -= vactive_start; + else if (line > vactive_end) + line = line - vfp_end - vactive_start; + else + line -= vactive_start; + + *vpos = line; + *hpos = 0; + + if (etime) + *etime = ktime_get(); + + return true; +} + +static u32 mdp5_crtc_get_vblank_counter(struct drm_crtc *crtc) +{ + struct drm_encoder *encoder; + + encoder = get_encoder_from_crtc(crtc); + if (!encoder) + return 0; + + return mdp5_encoder_get_framecount(encoder); +} + static void mdp5_crtc_atomic_disable(struct drm_crtc *crtc, struct drm_crtc_state *old_state) { @@ -1054,6 +1131,10 @@ static const struct drm_crtc_funcs mdp5_crtc_funcs = { .cursor_set = mdp5_crtc_cursor_set, .cursor_move = mdp5_crtc_cursor_move, .atomic_print_state = mdp5_crtc_atomic_print_state, + .get_vblank_counter = mdp5_crtc_get_vblank_counter, + .enable_vblank = msm_crtc_enable_vblank, + .disable_vblank = msm_crtc_disable_vblank, + .get_vblank_timestamp = drm_crtc_vblank_helper_get_vblank_timestamp, }; static const struct drm_crtc_helper_funcs mdp5_crtc_helper_funcs = { @@ -1063,6 +1144,7 @@ static const struct drm_crtc_helper_funcs mdp5_crtc_helper_funcs = { .atomic_flush = mdp5_crtc_atomic_flush, .atomic_enable = mdp5_crtc_atomic_enable, .atomic_disable = mdp5_crtc_atomic_disable, + .get_scanout_position = mdp5_crtc_get_scanout_position, }; static void mdp5_crtc_vblank_irq(struct mdp_irq *irq, uint32_t irqstatus) @@ -1109,8 +1191,8 @@ static void mdp5_crtc_wait_for_pp_done(struct drm_crtc *crtc) ret = wait_for_completion_timeout(&mdp5_crtc->pp_completion, msecs_to_jiffies(50)); if (ret == 0) - dev_warn(dev->dev, "pp done time out, lm=%d\n", - mdp5_cstate->pipeline.mixer->lm); + dev_warn_ratelimited(dev->dev, "pp done time out, lm=%d\n", + mdp5_cstate->pipeline.mixer->lm); } static void mdp5_crtc_wait_for_flush_done(struct drm_crtc *crtc) diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c index e43ecd4be10a..47b989834af1 100644 --- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c @@ -583,98 +583,6 @@ static int get_clk(struct platform_device *pdev, struct clk **clkp, return 0; } -static struct drm_encoder *get_encoder_from_crtc(struct drm_crtc *crtc) -{ - struct drm_device *dev = crtc->dev; - struct drm_encoder *encoder; - - drm_for_each_encoder(encoder, dev) - if (encoder->crtc == crtc) - return encoder; - - return NULL; -} - -static bool mdp5_get_scanoutpos(struct drm_device *dev, unsigned int pipe, - bool in_vblank_irq, int *vpos, int *hpos, - ktime_t *stime, ktime_t *etime, - const struct drm_display_mode *mode) -{ - struct msm_drm_private *priv = dev->dev_private; - struct drm_crtc *crtc; - struct drm_encoder *encoder; - int line, vsw, vbp, vactive_start, vactive_end, vfp_end; - - crtc = priv->crtcs[pipe]; - if (!crtc) { - DRM_ERROR("Invalid crtc %d\n", pipe); - return false; - } - - encoder = get_encoder_from_crtc(crtc); - if (!encoder) { - DRM_ERROR("no encoder found for crtc %d\n", pipe); - return false; - } - - vsw = mode->crtc_vsync_end - mode->crtc_vsync_start; - vbp = mode->crtc_vtotal - mode->crtc_vsync_end; - - /* - * the line counter is 1 at the start of the VSYNC pulse and VTOTAL at - * the end of VFP. Translate the porch values relative to the line - * counter positions. - */ - - vactive_start = vsw + vbp + 1; - - vactive_end = vactive_start + mode->crtc_vdisplay; - - /* last scan line before VSYNC */ - vfp_end = mode->crtc_vtotal; - - if (stime) - *stime = ktime_get(); - - line = mdp5_encoder_get_linecount(encoder); - - if (line < vactive_start) { - line -= vactive_start; - } else if (line > vactive_end) { - line = line - vfp_end - vactive_start; - } else { - line -= vactive_start; - } - - *vpos = line; - *hpos = 0; - - if (etime) - *etime = ktime_get(); - - return true; -} - -static u32 mdp5_get_vblank_counter(struct drm_device *dev, unsigned int pipe) -{ - struct msm_drm_private *priv = dev->dev_private; - struct drm_crtc *crtc; - struct drm_encoder *encoder; - - if (pipe >= priv->num_crtcs) - return 0; - - crtc = priv->crtcs[pipe]; - if (!crtc) - return 0; - - encoder = get_encoder_from_crtc(crtc); - if (!encoder) - return 0; - - return mdp5_encoder_get_framecount(encoder); -} - struct msm_kms *mdp5_kms_init(struct drm_device *dev) { struct msm_drm_private *priv = dev->dev_private; @@ -725,7 +633,7 @@ struct msm_kms *mdp5_kms_init(struct drm_device *dev) if (config->platform.iommu) { iommu_dev = &pdev->dev; - if (!iommu_dev->iommu_fwspec) + if (!dev_iommu_fwspec_get(iommu_dev)) iommu_dev = iommu_dev->parent; aspace = msm_gem_address_space_create(iommu_dev, @@ -762,9 +670,6 @@ struct msm_kms *mdp5_kms_init(struct drm_device *dev) dev->mode_config.max_width = 0xffff; dev->mode_config.max_height = 0xffff; - dev->driver->get_vblank_timestamp = drm_calc_vbltimestamp_from_scanoutpos; - dev->driver->get_scanout_position = mdp5_get_scanoutpos; - dev->driver->get_vblank_counter = mdp5_get_vblank_counter; dev->max_vblank_count = 0; /* max_vblank_count is set on each CRTC */ dev->vblank_disable_immediate = true; diff --git a/drivers/gpu/drm/msm/dsi/dsi_manager.c b/drivers/gpu/drm/msm/dsi/dsi_manager.c index 104115d112eb..4b363bd7ddff 100644 --- a/drivers/gpu/drm/msm/dsi/dsi_manager.c +++ b/drivers/gpu/drm/msm/dsi/dsi_manager.c @@ -336,7 +336,7 @@ static int dsi_mgr_connector_get_modes(struct drm_connector *connector) return num; } -static int dsi_mgr_connector_mode_valid(struct drm_connector *connector, +static enum drm_mode_status dsi_mgr_connector_mode_valid(struct drm_connector *connector, struct drm_display_mode *mode) { int id = dsi_mgr_connector_get_id(connector); @@ -506,6 +506,7 @@ static void dsi_mgr_bridge_post_disable(struct drm_bridge *bridge) struct msm_dsi *msm_dsi1 = dsi_mgr_get_dsi(DSI_1); struct mipi_dsi_host *host = msm_dsi->host; struct drm_panel *panel = msm_dsi->panel; + struct msm_dsi_pll *src_pll; bool is_dual_dsi = IS_DUAL_DSI(); int ret; @@ -539,6 +540,10 @@ static void dsi_mgr_bridge_post_disable(struct drm_bridge *bridge) id, ret); } + /* Save PLL status if it is a clock source */ + src_pll = msm_dsi_phy_get_pll(msm_dsi->phy); + msm_dsi_pll_save_state(src_pll); + ret = msm_dsi_host_power_off(host); if (ret) pr_err("%s: host %d power off failed,%d\n", __func__, id, ret); @@ -684,7 +689,7 @@ struct drm_bridge *msm_dsi_manager_bridge_init(u8 id) bridge = &dsi_bridge->base; bridge->funcs = &dsi_mgr_bridge_funcs; - ret = drm_bridge_attach(encoder, bridge, NULL); + ret = drm_bridge_attach(encoder, bridge, NULL, 0); if (ret) goto fail; @@ -713,7 +718,7 @@ struct drm_connector *msm_dsi_manager_ext_bridge_init(u8 id) encoder = msm_dsi->encoder; /* link the internal dsi bridge to the external bridge */ - drm_bridge_attach(encoder, ext_bridge, int_bridge); + drm_bridge_attach(encoder, ext_bridge, int_bridge, 0); /* * we need the drm_connector created by the external bridge diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c index b0cfa67d2a57..f509ebd77500 100644 --- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c +++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c @@ -724,10 +724,6 @@ void msm_dsi_phy_disable(struct msm_dsi_phy *phy) if (!phy || !phy->cfg->ops.disable) return; - /* Save PLL status if it is a clock source */ - if (phy->usecase != MSM_DSI_PHY_SLAVE) - msm_dsi_pll_save_state(phy->pll); - phy->cfg->ops.disable(phy); dsi_phy_regulator_disable(phy); diff --git a/drivers/gpu/drm/msm/dsi/pll/dsi_pll_10nm.c b/drivers/gpu/drm/msm/dsi/pll/dsi_pll_10nm.c index 1c894548dd72..6ac04fc303f5 100644 --- a/drivers/gpu/drm/msm/dsi/pll/dsi_pll_10nm.c +++ b/drivers/gpu/drm/msm/dsi/pll/dsi_pll_10nm.c @@ -411,6 +411,12 @@ static int dsi_pll_10nm_vco_prepare(struct clk_hw *hw) if (pll_10nm->slave) dsi_pll_enable_pll_bias(pll_10nm->slave); + rc = dsi_pll_10nm_vco_set_rate(hw,pll_10nm->vco_current_rate, 0); + if (rc) { + pr_err("vco_set_rate failed, rc=%d\n", rc); + return rc; + } + /* Start PLL */ pll_write(pll_10nm->phy_cmn_mmio + REG_DSI_10nm_PHY_CMN_PLL_CNTRL, 0x01); diff --git a/drivers/gpu/drm/msm/edp/edp.c b/drivers/gpu/drm/msm/edp/edp.c index ad4e963ccd9b..106a67473af5 100644 --- a/drivers/gpu/drm/msm/edp/edp.c +++ b/drivers/gpu/drm/msm/edp/edp.c @@ -178,10 +178,6 @@ int msm_edp_modeset_init(struct msm_edp *edp, struct drm_device *dev, goto fail; } - ret = drm_bridge_attach(encoder, edp->bridge, NULL); - if (ret) - goto fail; - priv->bridges[priv->num_bridges++] = edp->bridge; priv->connectors[priv->num_connectors++] = edp->connector; diff --git a/drivers/gpu/drm/msm/edp/edp_bridge.c b/drivers/gpu/drm/msm/edp/edp_bridge.c index b65b5cc2dba2..c69a37e0c708 100644 --- a/drivers/gpu/drm/msm/edp/edp_bridge.c +++ b/drivers/gpu/drm/msm/edp/edp_bridge.c @@ -97,7 +97,7 @@ struct drm_bridge *msm_edp_bridge_init(struct msm_edp *edp) bridge = &edp_bridge->base; bridge->funcs = &edp_bridge_funcs; - ret = drm_bridge_attach(edp->encoder, bridge, NULL); + ret = drm_bridge_attach(edp->encoder, bridge, NULL, 0); if (ret) goto fail; diff --git a/drivers/gpu/drm/msm/hdmi/hdmi.c b/drivers/gpu/drm/msm/hdmi/hdmi.c index 1a9b6289637d..737453b6e596 100644 --- a/drivers/gpu/drm/msm/hdmi/hdmi.c +++ b/drivers/gpu/drm/msm/hdmi/hdmi.c @@ -327,10 +327,6 @@ int msm_hdmi_modeset_init(struct hdmi *hdmi, goto fail; } - ret = drm_bridge_attach(encoder, hdmi->bridge, NULL); - if (ret) - goto fail; - priv->bridges[priv->num_bridges++] = hdmi->bridge; priv->connectors[priv->num_connectors++] = hdmi->connector; diff --git a/drivers/gpu/drm/msm/hdmi/hdmi_bridge.c b/drivers/gpu/drm/msm/hdmi/hdmi_bridge.c index ba81338a9bf8..6e380db9287b 100644 --- a/drivers/gpu/drm/msm/hdmi/hdmi_bridge.c +++ b/drivers/gpu/drm/msm/hdmi/hdmi_bridge.c @@ -287,7 +287,7 @@ struct drm_bridge *msm_hdmi_bridge_init(struct hdmi *hdmi) bridge = &hdmi_bridge->base; bridge->funcs = &msm_hdmi_bridge_funcs; - ret = drm_bridge_attach(hdmi->encoder, bridge, NULL); + ret = drm_bridge_attach(hdmi->encoder, bridge, NULL, 0); if (ret) goto fail; diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c index e4b750b0c2d3..29295dee2a2e 100644 --- a/drivers/gpu/drm/msm/msm_drv.c +++ b/drivers/gpu/drm/msm/msm_drv.c @@ -444,8 +444,10 @@ static int msm_drm_init(struct device *dev, struct drm_driver *drv) if (!dev->dma_parms) { dev->dma_parms = devm_kzalloc(dev, sizeof(*dev->dma_parms), GFP_KERNEL); - if (!dev->dma_parms) - return -ENOMEM; + if (!dev->dma_parms) { + ret = -ENOMEM; + goto err_msm_uninit; + } } dma_set_max_seg_size(dev, DMA_BIT_MASK(32)); @@ -668,8 +670,10 @@ static void msm_irq_uninstall(struct drm_device *dev) kms->funcs->irq_uninstall(kms); } -static int msm_enable_vblank(struct drm_device *dev, unsigned int pipe) +int msm_crtc_enable_vblank(struct drm_crtc *crtc) { + struct drm_device *dev = crtc->dev; + unsigned int pipe = crtc->index; struct msm_drm_private *priv = dev->dev_private; struct msm_kms *kms = priv->kms; if (!kms) @@ -678,8 +682,10 @@ static int msm_enable_vblank(struct drm_device *dev, unsigned int pipe) return vblank_ctrl_queue_work(priv, pipe, true); } -static void msm_disable_vblank(struct drm_device *dev, unsigned int pipe) +void msm_crtc_disable_vblank(struct drm_crtc *crtc) { + struct drm_device *dev = crtc->dev; + unsigned int pipe = crtc->index; struct msm_drm_private *priv = dev->dev_private; struct msm_kms *kms = priv->kms; if (!kms) @@ -1004,8 +1010,6 @@ static struct drm_driver msm_driver = { .irq_preinstall = msm_irq_preinstall, .irq_postinstall = msm_irq_postinstall, .irq_uninstall = msm_irq_uninstall, - .enable_vblank = msm_enable_vblank, - .disable_vblank = msm_disable_vblank, .gem_free_object_unlocked = msm_gem_free_object, .gem_vm_ops = &vm_ops, .dumb_create = msm_gem_dumb_create, diff --git a/drivers/gpu/drm/msm/msm_drv.h b/drivers/gpu/drm/msm/msm_drv.h index 740bf7c70d8f..194d900a460e 100644 --- a/drivers/gpu/drm/msm/msm_drv.h +++ b/drivers/gpu/drm/msm/msm_drv.h @@ -232,6 +232,9 @@ struct drm_atomic_state *msm_atomic_state_alloc(struct drm_device *dev); void msm_atomic_state_clear(struct drm_atomic_state *state); void msm_atomic_state_free(struct drm_atomic_state *state); +int msm_crtc_enable_vblank(struct drm_crtc *crtc); +void msm_crtc_disable_vblank(struct drm_crtc *crtc); + int msm_gem_init_vma(struct msm_gem_address_space *aspace, struct msm_gem_vma *vma, int npages); void msm_gem_purge_vma(struct msm_gem_address_space *aspace, diff --git a/drivers/gpu/drm/msm/msm_fbdev.c b/drivers/gpu/drm/msm/msm_fbdev.c index db48867df47d..47235f8c5922 100644 --- a/drivers/gpu/drm/msm/msm_fbdev.c +++ b/drivers/gpu/drm/msm/msm_fbdev.c @@ -160,16 +160,12 @@ struct drm_fb_helper *msm_fbdev_init(struct drm_device *dev) drm_fb_helper_prepare(dev, helper, &msm_fb_helper_funcs); - ret = drm_fb_helper_init(dev, helper, priv->num_connectors); + ret = drm_fb_helper_init(dev, helper); if (ret) { DRM_DEV_ERROR(dev->dev, "could not init fbdev: ret=%d\n", ret); goto fail; } - ret = drm_fb_helper_single_add_all_connectors(helper); - if (ret) - goto fini; - /* the fw fb could be anywhere in memory */ drm_fb_helper_remove_conflicting_framebuffers(NULL, "msm", false); diff --git a/drivers/gpu/drm/msm/msm_gem.h b/drivers/gpu/drm/msm/msm_gem.h index 9e0953c2b7ce..30584eaf8cc8 100644 --- a/drivers/gpu/drm/msm/msm_gem.h +++ b/drivers/gpu/drm/msm/msm_gem.h @@ -157,7 +157,17 @@ struct msm_gem_submit { uint32_t handle; }; uint64_t iova; - } bos[0]; + } bos[]; }; +/* helper to determine of a buffer in submit should be dumped, used for both + * devcoredump and debugfs cmdstream dumping: + */ +static inline bool +should_dump(struct msm_gem_submit *submit, int idx) +{ + extern bool rd_full; + return rd_full || (submit->bos[idx].flags & MSM_SUBMIT_BO_DUMP); +} + #endif /* __MSM_GEM_H__ */ diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c index 18f3a5c53ffb..615c5cda5389 100644 --- a/drivers/gpu/drm/msm/msm_gpu.c +++ b/drivers/gpu/drm/msm/msm_gpu.c @@ -355,16 +355,34 @@ static void msm_gpu_crashstate_capture(struct msm_gpu *gpu, state->cmd = kstrdup(cmd, GFP_KERNEL); if (submit) { - int i; - - state->bos = kcalloc(submit->nr_cmds, + int i, nr = 0; + + /* count # of buffers to dump: */ + for (i = 0; i < submit->nr_bos; i++) + if (should_dump(submit, i)) + nr++; + /* always dump cmd bo's, but don't double count them: */ + for (i = 0; i < submit->nr_cmds; i++) + if (!should_dump(submit, submit->cmd[i].idx)) + nr++; + + state->bos = kcalloc(nr, sizeof(struct msm_gpu_state_bo), GFP_KERNEL); + for (i = 0; i < submit->nr_bos; i++) { + if (should_dump(submit, i)) { + msm_gpu_crashstate_get_bo(state, submit->bos[i].obj, + submit->bos[i].iova, submit->bos[i].flags); + } + } + for (i = 0; state->bos && i < submit->nr_cmds; i++) { int idx = submit->cmd[i].idx; - msm_gpu_crashstate_get_bo(state, submit->bos[idx].obj, - submit->bos[idx].iova, submit->bos[idx].flags); + if (!should_dump(submit, submit->cmd[i].idx)) { + msm_gpu_crashstate_get_bo(state, submit->bos[idx].obj, + submit->bos[idx].iova, submit->bos[idx].flags); + } } } diff --git a/drivers/gpu/drm/msm/msm_rd.c b/drivers/gpu/drm/msm/msm_rd.c index af7ceb246c7c..732f65df5c4f 100644 --- a/drivers/gpu/drm/msm/msm_rd.c +++ b/drivers/gpu/drm/msm/msm_rd.c @@ -43,7 +43,7 @@ #include "msm_gpu.h" #include "msm_gem.h" -static bool rd_full = false; +bool rd_full = false; MODULE_PARM_DESC(rd_full, "If true, $debugfs/.../rd will snapshot all buffer contents"); module_param_named(rd_full, rd_full, bool, 0600); @@ -336,12 +336,6 @@ static void snapshot_buf(struct msm_rd_state *rd, msm_gem_put_vaddr(&obj->base); } -static bool -should_dump(struct msm_gem_submit *submit, int idx) -{ - return rd_full || (submit->bos[idx].flags & MSM_SUBMIT_BO_DUMP); -} - /* called under struct_mutex */ void msm_rd_dump_submit(struct msm_rd_state *rd, struct msm_gem_submit *submit, const char *fmt, ...) diff --git a/drivers/gpu/drm/nouveau/dispnv04/crtc.c b/drivers/gpu/drm/nouveau/dispnv04/crtc.c index 37c50ea8f847..1f08de4241e0 100644 --- a/drivers/gpu/drm/nouveau/dispnv04/crtc.c +++ b/drivers/gpu/drm/nouveau/dispnv04/crtc.c @@ -1248,6 +1248,9 @@ static const struct drm_crtc_funcs nv04_crtc_funcs = { .set_config = drm_crtc_helper_set_config, .page_flip = nv04_crtc_page_flip, .destroy = nv_crtc_destroy, + .enable_vblank = nouveau_display_vblank_enable, + .disable_vblank = nouveau_display_vblank_disable, + .get_vblank_timestamp = drm_crtc_vblank_helper_get_vblank_timestamp, }; static const struct drm_crtc_helper_funcs nv04_crtc_helper_funcs = { @@ -1258,6 +1261,7 @@ static const struct drm_crtc_helper_funcs nv04_crtc_helper_funcs = { .mode_set_base = nv04_crtc_mode_set_base, .mode_set_base_atomic = nv04_crtc_mode_set_base_atomic, .disable = nv_crtc_disable, + .get_scanout_position = nouveau_display_scanoutpos, }; static const uint32_t modeset_formats[] = { diff --git a/drivers/gpu/drm/nouveau/dispnv04/dac.c b/drivers/gpu/drm/nouveau/dispnv04/dac.c index e8eef88a8382..ffdd447d8706 100644 --- a/drivers/gpu/drm/nouveau/dispnv04/dac.c +++ b/drivers/gpu/drm/nouveau/dispnv04/dac.c @@ -35,7 +35,8 @@ #include <subdev/bios/gpio.h> #include <subdev/gpio.h> -#include <subdev/timer.h> + +#include <nvif/timer.h> int nv04_dac_output_offset(struct drm_encoder *encoder) { diff --git a/drivers/gpu/drm/nouveau/dispnv04/hw.c b/drivers/gpu/drm/nouveau/dispnv04/hw.c index 3fdfafa8b0ad..b674d68ef28a 100644 --- a/drivers/gpu/drm/nouveau/dispnv04/hw.c +++ b/drivers/gpu/drm/nouveau/dispnv04/hw.c @@ -26,6 +26,7 @@ #include "hw.h" #include <subdev/bios/pll.h> +#include <nvif/timer.h> #define CHIPSET_NFORCE 0x01a0 #define CHIPSET_NFORCE2 0x01f0 diff --git a/drivers/gpu/drm/nouveau/dispnv50/base507c.c b/drivers/gpu/drm/nouveau/dispnv50/base507c.c index 00a85f1e1a4a..ee782151d332 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/base507c.c +++ b/drivers/gpu/drm/nouveau/dispnv50/base507c.c @@ -23,6 +23,7 @@ #include <nvif/cl507c.h> #include <nvif/event.h> +#include <nvif/timer.h> #include <drm/drm_atomic_helper.h> #include <drm/drm_fourcc.h> diff --git a/drivers/gpu/drm/nouveau/dispnv50/core507d.c b/drivers/gpu/drm/nouveau/dispnv50/core507d.c index e7fcfa6e6467..c5152c39c684 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/core507d.c +++ b/drivers/gpu/drm/nouveau/dispnv50/core507d.c @@ -23,6 +23,7 @@ #include "head.h" #include <nvif/cl507d.h> +#include <nvif/timer.h> #include "nouveau_bo.h" diff --git a/drivers/gpu/drm/nouveau/dispnv50/corec37d.c b/drivers/gpu/drm/nouveau/dispnv50/corec37d.c index 3b36dc8d36b2..c03cb987856b 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/corec37d.c +++ b/drivers/gpu/drm/nouveau/dispnv50/corec37d.c @@ -24,6 +24,8 @@ #include <nouveau_bo.h> +#include <nvif/timer.h> + void corec37d_wndw_owner(struct nv50_core *core) { diff --git a/drivers/gpu/drm/nouveau/dispnv50/curs507a.c b/drivers/gpu/drm/nouveau/dispnv50/curs507a.c index 397143b639c6..8c5cf096f69b 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/curs507a.c +++ b/drivers/gpu/drm/nouveau/dispnv50/curs507a.c @@ -24,21 +24,36 @@ #include "head.h" #include <nvif/cl507a.h> +#include <nvif/timer.h> #include <drm/drm_atomic_helper.h> #include <drm/drm_plane_helper.h> +bool +curs507a_space(struct nv50_wndw *wndw) +{ + nvif_msec(&nouveau_drm(wndw->plane.dev)->client.device, 2, + if (nvif_rd32(&wndw->wimm.base.user, 0x0008) >= 4) + return true; + ); + WARN_ON(1); + return false; +} + static void curs507a_update(struct nv50_wndw *wndw, u32 *interlock) { - nvif_wr32(&wndw->wimm.base.user, 0x0080, 0x00000000); + if (curs507a_space(wndw)) + nvif_wr32(&wndw->wimm.base.user, 0x0080, 0x00000000); } static void curs507a_point(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) { - nvif_wr32(&wndw->wimm.base.user, 0x0084, asyw->point.y << 16 | - asyw->point.x); + if (curs507a_space(wndw)) { + nvif_wr32(&wndw->wimm.base.user, 0x0084, asyw->point.y << 16 | + asyw->point.x); + } } const struct nv50_wimm_func diff --git a/drivers/gpu/drm/nouveau/dispnv50/cursc37a.c b/drivers/gpu/drm/nouveau/dispnv50/cursc37a.c index 23fb29d41efe..96dff4f09f57 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/cursc37a.c +++ b/drivers/gpu/drm/nouveau/dispnv50/cursc37a.c @@ -25,14 +25,17 @@ static void cursc37a_update(struct nv50_wndw *wndw, u32 *interlock) { - nvif_wr32(&wndw->wimm.base.user, 0x0200, 0x00000001); + if (curs507a_space(wndw)) + nvif_wr32(&wndw->wimm.base.user, 0x0200, 0x00000001); } static void cursc37a_point(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) { - nvif_wr32(&wndw->wimm.base.user, 0x0208, asyw->point.y << 16 | - asyw->point.x); + if (curs507a_space(wndw)) { + nvif_wr32(&wndw->wimm.base.user, 0x0208, asyw->point.y << 16 | + asyw->point.x); + } } static const struct nv50_wimm_func diff --git a/drivers/gpu/drm/nouveau/dispnv50/disp.c b/drivers/gpu/drm/nouveau/dispnv50/disp.c index a3dc2ba19fb2..6be9df1820c5 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/disp.c +++ b/drivers/gpu/drm/nouveau/dispnv50/disp.c @@ -45,6 +45,7 @@ #include <nvif/cl5070.h> #include <nvif/cl507d.h> #include <nvif/event.h> +#include <nvif/timer.h> #include "nouveau_drv.h" #include "nouveau_dma.h" @@ -1256,30 +1257,6 @@ nv50_mstm_prepare(struct nv50_mstm *mstm) } } -static void -nv50_mstm_destroy_connector(struct drm_dp_mst_topology_mgr *mgr, - struct drm_connector *connector) -{ - struct nouveau_drm *drm = nouveau_drm(connector->dev); - struct nv50_mstc *mstc = nv50_mstc(connector); - - drm_connector_unregister(&mstc->connector); - - drm_fb_helper_remove_one_connector(&drm->fbcon->helper, &mstc->connector); - - drm_connector_put(&mstc->connector); -} - -static void -nv50_mstm_register_connector(struct drm_connector *connector) -{ - struct nouveau_drm *drm = nouveau_drm(connector->dev); - - drm_fb_helper_add_one_connector(&drm->fbcon->helper, connector); - - drm_connector_register(connector); -} - static struct drm_connector * nv50_mstm_add_connector(struct drm_dp_mst_topology_mgr *mgr, struct drm_dp_mst_port *port, const char *path) @@ -1298,8 +1275,6 @@ nv50_mstm_add_connector(struct drm_dp_mst_topology_mgr *mgr, static const struct drm_dp_mst_topology_cbs nv50_mstm = { .add_connector = nv50_mstm_add_connector, - .register_connector = nv50_mstm_register_connector, - .destroy_connector = nv50_mstm_destroy_connector, }; void diff --git a/drivers/gpu/drm/nouveau/dispnv50/head.c b/drivers/gpu/drm/nouveau/dispnv50/head.c index d9d64602947d..8f6455697ba7 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/head.c +++ b/drivers/gpu/drm/nouveau/dispnv50/head.c @@ -29,6 +29,7 @@ #include <drm/drm_atomic_helper.h> #include <drm/drm_crtc_helper.h> +#include <drm/drm_vblank.h> #include "nouveau_connector.h" void nv50_head_flush_clr(struct nv50_head *head, @@ -413,6 +414,7 @@ nv50_head_atomic_check(struct drm_crtc *crtc, struct drm_crtc_state *state) static const struct drm_crtc_helper_funcs nv50_head_help = { .atomic_check = nv50_head_atomic_check, + .get_scanout_position = nouveau_display_scanoutpos, }; static void @@ -481,6 +483,9 @@ nv50_head_func = { .page_flip = drm_atomic_helper_page_flip, .atomic_duplicate_state = nv50_head_atomic_duplicate_state, .atomic_destroy_state = nv50_head_atomic_destroy_state, + .enable_vblank = nouveau_display_vblank_enable, + .disable_vblank = nouveau_display_vblank_disable, + .get_vblank_timestamp = drm_crtc_vblank_helper_get_vblank_timestamp, }; struct nv50_head * diff --git a/drivers/gpu/drm/nouveau/dispnv50/ovly827e.c b/drivers/gpu/drm/nouveau/dispnv50/ovly827e.c index 2e68fc736fe1..4f7ce57f2036 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/ovly827e.c +++ b/drivers/gpu/drm/nouveau/dispnv50/ovly827e.c @@ -24,6 +24,8 @@ #include <nouveau_bo.h> +#include <nvif/timer.h> + static void ovly827e_image_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) { diff --git a/drivers/gpu/drm/nouveau/dispnv50/wndw.c b/drivers/gpu/drm/nouveau/dispnv50/wndw.c index 890315291b01..bb737f9281e6 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/wndw.c +++ b/drivers/gpu/drm/nouveau/dispnv50/wndw.c @@ -458,6 +458,8 @@ nv50_wndw_atomic_check(struct drm_plane *plane, struct drm_plane_state *state) asyw->clr.ntfy = armw->ntfy.handle != 0; asyw->clr.sema = armw->sema.handle != 0; asyw->clr.xlut = armw->xlut.handle != 0; + if (asyw->clr.xlut && asyw->visible) + asyw->set.xlut = asyw->xlut.handle != 0; asyw->clr.csc = armw->csc.valid; if (wndw->func->image_clr) asyw->clr.image = armw->image.handle[0] != 0; diff --git a/drivers/gpu/drm/nouveau/dispnv50/wndw.h b/drivers/gpu/drm/nouveau/dispnv50/wndw.h index caf397475918..a7412b9d3a98 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/wndw.h +++ b/drivers/gpu/drm/nouveau/dispnv50/wndw.h @@ -97,6 +97,7 @@ struct nv50_wimm_func { }; extern const struct nv50_wimm_func curs507a; +bool curs507a_space(struct nv50_wndw *); int wndwc37e_new(struct nouveau_drm *, enum drm_plane_type, int, s32, struct nv50_wndw **); diff --git a/drivers/gpu/drm/nouveau/include/nvif/device.h b/drivers/gpu/drm/nouveau/include/nvif/device.h index 25d969dcf67d..c2a572c67a76 100644 --- a/drivers/gpu/drm/nouveau/include/nvif/device.h +++ b/drivers/gpu/drm/nouveau/include/nvif/device.h @@ -23,27 +23,6 @@ int nvif_device_init(struct nvif_object *, u32 handle, s32 oclass, void *, u32, void nvif_device_fini(struct nvif_device *); u64 nvif_device_time(struct nvif_device *); -/* Delay based on GPU time (ie. PTIMER). - * - * Will return -ETIMEDOUT unless the loop was terminated with 'break', - * where it will return the number of nanoseconds taken instead. - */ -#define nvif_nsec(d,n,cond...) ({ \ - struct nvif_device *_device = (d); \ - u64 _nsecs = (n), _time0 = nvif_device_time(_device); \ - s64 _taken = 0; \ - \ - do { \ - cond \ - } while (_taken = nvif_device_time(_device) - _time0, _taken < _nsecs);\ - \ - if (_taken >= _nsecs) \ - _taken = -ETIMEDOUT; \ - _taken; \ -}) -#define nvif_usec(d,u,cond...) nvif_nsec((d), (u) * 1000, ##cond) -#define nvif_msec(d,m,cond...) nvif_usec((d), (m) * 1000, ##cond) - /*XXX*/ #include <subdev/bios.h> #include <subdev/fb.h> diff --git a/drivers/gpu/drm/nouveau/include/nvif/timer.h b/drivers/gpu/drm/nouveau/include/nvif/timer.h new file mode 100644 index 000000000000..57587a985c4b --- /dev/null +++ b/drivers/gpu/drm/nouveau/include/nvif/timer.h @@ -0,0 +1,35 @@ +#ifndef __NVIF_TIMER_H__ +#define __NVIF_TIMER_H__ +#include <nvif/os.h> + +struct nvif_timer_wait { + struct nvif_device *device; + u64 limit; + u64 time0; + u64 time1; + int reads; +}; + +void nvif_timer_wait_init(struct nvif_device *, u64 nsec, + struct nvif_timer_wait *); +s64 nvif_timer_wait_test(struct nvif_timer_wait *); + +/* Delay based on GPU time (ie. PTIMER). + * + * Will return -ETIMEDOUT unless the loop was terminated with 'break', + * where it will return the number of nanoseconds taken instead. + */ +#define nvif_nsec(d,n,cond...) ({ \ + struct nvif_timer_wait _wait; \ + s64 _taken = 0; \ + \ + nvif_timer_wait_init((d), (n), &_wait); \ + do { \ + cond \ + } while ((_taken = nvif_timer_wait_test(&_wait)) >= 0); \ + \ + _taken; \ +}) +#define nvif_usec(d,u,cond...) nvif_nsec((d), (u) * 1000, ##cond) +#define nvif_msec(d,m,cond...) nvif_usec((d), (m) * 1000, ##cond) +#endif diff --git a/drivers/gpu/drm/nouveau/include/nvif/user.h b/drivers/gpu/drm/nouveau/include/nvif/user.h index 03c11826b693..6825574d93c2 100644 --- a/drivers/gpu/drm/nouveau/include/nvif/user.h +++ b/drivers/gpu/drm/nouveau/include/nvif/user.h @@ -10,6 +10,7 @@ struct nvif_user { struct nvif_user_func { void (*doorbell)(struct nvif_user *, u32 token); + u64 (*time)(struct nvif_user *); }; int nvif_user_init(struct nvif_device *); diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau/nouveau_bo.c index 2b4b21b02e40..c40f127de3d0 100644 --- a/drivers/gpu/drm/nouveau/nouveau_bo.c +++ b/drivers/gpu/drm/nouveau/nouveau_bo.c @@ -1494,8 +1494,13 @@ nouveau_ttm_io_mem_reserve(struct ttm_bo_device *bdev, struct ttm_mem_reg *reg) ret = nvif_object_map_handle(&mem->mem.object, &args, argc, &handle, &length); - if (ret != 1) - return ret ? ret : -EINVAL; + if (ret != 1) { + if (WARN_ON(ret == 0)) + return -EINVAL; + if (ret == -ENOSPC) + return -EAGAIN; + return ret; + } reg->bus.base = 0; reg->bus.offset = handle; diff --git a/drivers/gpu/drm/nouveau/nouveau_debugfs.c b/drivers/gpu/drm/nouveau/nouveau_debugfs.c index 7dfbbbc1beea..15a3d40edf02 100644 --- a/drivers/gpu/drm/nouveau/nouveau_debugfs.c +++ b/drivers/gpu/drm/nouveau/nouveau_debugfs.c @@ -222,22 +222,18 @@ nouveau_drm_debugfs_init(struct drm_minor *minor) { struct nouveau_drm *drm = nouveau_drm(minor->dev); struct dentry *dentry; - int i, ret; + int i; for (i = 0; i < ARRAY_SIZE(nouveau_debugfs_files); i++) { - dentry = debugfs_create_file(nouveau_debugfs_files[i].name, - S_IRUGO | S_IWUSR, - minor->debugfs_root, minor->dev, - nouveau_debugfs_files[i].fops); - if (!dentry) - return -ENOMEM; + debugfs_create_file(nouveau_debugfs_files[i].name, + S_IRUGO | S_IWUSR, + minor->debugfs_root, minor->dev, + nouveau_debugfs_files[i].fops); } - ret = drm_debugfs_create_files(nouveau_debugfs_list, - NOUVEAU_DEBUGFS_ENTRIES, - minor->debugfs_root, minor); - if (ret) - return ret; + drm_debugfs_create_files(nouveau_debugfs_list, + NOUVEAU_DEBUGFS_ENTRIES, + minor->debugfs_root, minor); /* Set the size of the vbios since we know it, and it's confusing to * userspace if it wants to seek() but the file has a length of 0 diff --git a/drivers/gpu/drm/nouveau/nouveau_display.c b/drivers/gpu/drm/nouveau/nouveau_display.c index 53f9bceaf17a..700817dc4fa0 100644 --- a/drivers/gpu/drm/nouveau/nouveau_display.c +++ b/drivers/gpu/drm/nouveau/nouveau_display.c @@ -54,15 +54,10 @@ nouveau_display_vblank_handler(struct nvif_notify *notify) } int -nouveau_display_vblank_enable(struct drm_device *dev, unsigned int pipe) +nouveau_display_vblank_enable(struct drm_crtc *crtc) { - struct drm_crtc *crtc; struct nouveau_crtc *nv_crtc; - crtc = drm_crtc_from_index(dev, pipe); - if (!crtc) - return -EINVAL; - nv_crtc = nouveau_crtc(crtc); nvif_notify_get(&nv_crtc->vblank); @@ -70,15 +65,10 @@ nouveau_display_vblank_enable(struct drm_device *dev, unsigned int pipe) } void -nouveau_display_vblank_disable(struct drm_device *dev, unsigned int pipe) +nouveau_display_vblank_disable(struct drm_crtc *crtc) { - struct drm_crtc *crtc; struct nouveau_crtc *nv_crtc; - crtc = drm_crtc_from_index(dev, pipe); - if (!crtc) - return; - nv_crtc = nouveau_crtc(crtc); nvif_notify_put(&nv_crtc->vblank); } @@ -136,21 +126,13 @@ nouveau_display_scanoutpos_head(struct drm_crtc *crtc, int *vpos, int *hpos, } bool -nouveau_display_scanoutpos(struct drm_device *dev, unsigned int pipe, +nouveau_display_scanoutpos(struct drm_crtc *crtc, bool in_vblank_irq, int *vpos, int *hpos, ktime_t *stime, ktime_t *etime, const struct drm_display_mode *mode) { - struct drm_crtc *crtc; - - list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) { - if (nouveau_crtc(crtc)->index == pipe) { - return nouveau_display_scanoutpos_head(crtc, vpos, hpos, - stime, etime); - } - } - - return false; + return nouveau_display_scanoutpos_head(crtc, vpos, hpos, + stime, etime); } static void diff --git a/drivers/gpu/drm/nouveau/nouveau_display.h b/drivers/gpu/drm/nouveau/nouveau_display.h index 6e8e66882e45..de004018ab5c 100644 --- a/drivers/gpu/drm/nouveau/nouveau_display.h +++ b/drivers/gpu/drm/nouveau/nouveau_display.h @@ -61,11 +61,12 @@ int nouveau_display_init(struct drm_device *dev, bool resume, bool runtime); void nouveau_display_fini(struct drm_device *dev, bool suspend, bool runtime); int nouveau_display_suspend(struct drm_device *dev, bool runtime); void nouveau_display_resume(struct drm_device *dev, bool runtime); -int nouveau_display_vblank_enable(struct drm_device *, unsigned int); -void nouveau_display_vblank_disable(struct drm_device *, unsigned int); -bool nouveau_display_scanoutpos(struct drm_device *, unsigned int, - bool, int *, int *, ktime_t *, - ktime_t *, const struct drm_display_mode *); +int nouveau_display_vblank_enable(struct drm_crtc *crtc); +void nouveau_display_vblank_disable(struct drm_crtc *crtc); +bool nouveau_display_scanoutpos(struct drm_crtc *crtc, + bool in_vblank_irq, int *vpos, int *hpos, + ktime_t *stime, ktime_t *etime, + const struct drm_display_mode *mode); int nouveau_display_dumb_create(struct drm_file *, struct drm_device *, struct drm_mode_create_dumb *args); diff --git a/drivers/gpu/drm/nouveau/nouveau_dmem.c b/drivers/gpu/drm/nouveau/nouveau_dmem.c index 0ad5d87b5a8e..ad89e09a0be3 100644 --- a/drivers/gpu/drm/nouveau/nouveau_dmem.c +++ b/drivers/gpu/drm/nouveau/nouveau_dmem.c @@ -28,6 +28,7 @@ #include <nvif/class.h> #include <nvif/object.h> +#include <nvif/if000c.h> #include <nvif/if500b.h> #include <nvif/if900b.h> @@ -176,6 +177,7 @@ static vm_fault_t nouveau_dmem_migrate_to_ram(struct vm_fault *vmf) .end = vmf->address + PAGE_SIZE, .src = &src, .dst = &dst, + .src_owner = drm->dev, }; /* @@ -526,6 +528,7 @@ nouveau_dmem_init(struct nouveau_drm *drm) drm->dmem->pagemap.type = MEMORY_DEVICE_PRIVATE; drm->dmem->pagemap.res = *res; drm->dmem->pagemap.ops = &nouveau_dmem_pagemap_ops; + drm->dmem->pagemap.owner = drm->dev; if (IS_ERR(devm_memremap_pages(device, &drm->dmem->pagemap))) goto out_free; @@ -669,12 +672,6 @@ out: return ret; } -static inline bool -nouveau_dmem_page(struct nouveau_drm *drm, struct page *page) -{ - return is_device_private_page(page) && drm->dmem == page_to_dmem(page); -} - void nouveau_dmem_convert_pfn(struct nouveau_drm *drm, struct hmm_range *range) @@ -690,18 +687,12 @@ nouveau_dmem_convert_pfn(struct nouveau_drm *drm, if (page == NULL) continue; - if (!(range->pfns[i] & range->flags[HMM_PFN_DEVICE_PRIVATE])) { + if (!is_device_private_page(page)) continue; - } - - if (!nouveau_dmem_page(drm, page)) { - WARN(1, "Some unknown device memory !\n"); - range->pfns[i] = 0; - continue; - } addr = nouveau_dmem_page_addr(page); range->pfns[i] &= ((1UL << range->pfn_shift) - 1); range->pfns[i] |= (addr >> PAGE_SHIFT) << range->pfn_shift; + range->pfns[i] |= NVIF_VMM_PFNMAP_V0_VRAM; } } diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c index b65ae817eabf..ca4087f5a15b 100644 --- a/drivers/gpu/drm/nouveau/nouveau_drm.c +++ b/drivers/gpu/drm/nouveau/nouveau_drm.c @@ -618,6 +618,64 @@ nouveau_drm_device_fini(struct drm_device *dev) kfree(drm); } +/* + * On some Intel PCIe bridge controllers doing a + * D0 -> D3hot -> D3cold -> D0 sequence causes Nvidia GPUs to not reappear. + * Skipping the intermediate D3hot step seems to make it work again. This is + * probably caused by not meeting the expectation the involved AML code has + * when the GPU is put into D3hot state before invoking it. + * + * This leads to various manifestations of this issue: + * - AML code execution to power on the GPU hits an infinite loop (as the + * code waits on device memory to change). + * - kernel crashes, as all PCI reads return -1, which most code isn't able + * to handle well enough. + * + * In all cases dmesg will contain at least one line like this: + * 'nouveau 0000:01:00.0: Refused to change power state, currently in D3' + * followed by a lot of nouveau timeouts. + * + * In the \_SB.PCI0.PEG0.PG00._OFF code deeper down writes bit 0x80 to the not + * documented PCI config space register 0x248 of the Intel PCIe bridge + * controller (0x1901) in order to change the state of the PCIe link between + * the PCIe port and the GPU. There are alternative code paths using other + * registers, which seem to work fine (executed pre Windows 8): + * - 0xbc bit 0x20 (publicly available documentation claims 'reserved') + * - 0xb0 bit 0x10 (link disable) + * Changing the conditions inside the firmware by poking into the relevant + * addresses does resolve the issue, but it seemed to be ACPI private memory + * and not any device accessible memory at all, so there is no portable way of + * changing the conditions. + * On a XPS 9560 that means bits [0,3] on \CPEX need to be cleared. + * + * The only systems where this behavior can be seen are hybrid graphics laptops + * with a secondary Nvidia Maxwell, Pascal or Turing GPU. It's unclear whether + * this issue only occurs in combination with listed Intel PCIe bridge + * controllers and the mentioned GPUs or other devices as well. + * + * documentation on the PCIe bridge controller can be found in the + * "7th Generation Intel® Processor Families for H Platforms Datasheet Volume 2" + * Section "12 PCI Express* Controller (x16) Registers" + */ + +static void quirk_broken_nv_runpm(struct pci_dev *pdev) +{ + struct drm_device *dev = pci_get_drvdata(pdev); + struct nouveau_drm *drm = nouveau_drm(dev); + struct pci_dev *bridge = pci_upstream_bridge(pdev); + + if (!bridge || bridge->vendor != PCI_VENDOR_ID_INTEL) + return; + + switch (bridge->device) { + case 0x1901: + drm->old_pm_cap = pdev->pm_cap; + pdev->pm_cap = 0; + NV_INFO(drm, "Disabling PCI power management to avoid bug\n"); + break; + } +} + static int nouveau_drm_probe(struct pci_dev *pdev, const struct pci_device_id *pent) { @@ -699,6 +757,7 @@ static int nouveau_drm_probe(struct pci_dev *pdev, if (ret) goto fail_drm_dev_init; + quirk_broken_nv_runpm(pdev); return 0; fail_drm_dev_init: @@ -734,7 +793,11 @@ static void nouveau_drm_remove(struct pci_dev *pdev) { struct drm_device *dev = pci_get_drvdata(pdev); + struct nouveau_drm *drm = nouveau_drm(dev); + /* revert our workaround */ + if (drm->old_pm_cap) + pdev->pm_cap = drm->old_pm_cap; nouveau_drm_device_remove(dev); pci_disable_device(pdev); } @@ -1120,11 +1183,6 @@ driver_stub = { .debugfs_init = nouveau_drm_debugfs_init, #endif - .enable_vblank = nouveau_display_vblank_enable, - .disable_vblank = nouveau_display_vblank_disable, - .get_scanout_position = nouveau_display_scanoutpos, - .get_vblank_timestamp = drm_calc_vbltimestamp_from_scanoutpos, - .ioctls = nouveau_ioctls, .num_ioctls = ARRAY_SIZE(nouveau_ioctls), .fops = &nouveau_driver_fops, diff --git a/drivers/gpu/drm/nouveau/nouveau_drv.h b/drivers/gpu/drm/nouveau/nouveau_drv.h index c2c332fbde97..2a6519737800 100644 --- a/drivers/gpu/drm/nouveau/nouveau_drv.h +++ b/drivers/gpu/drm/nouveau/nouveau_drv.h @@ -140,6 +140,8 @@ struct nouveau_drm { struct list_head clients; + u8 old_pm_cap; + struct { struct agp_bridge_data *bridge; u32 base; diff --git a/drivers/gpu/drm/nouveau/nouveau_fbcon.c b/drivers/gpu/drm/nouveau/nouveau_fbcon.c index 0c5cdda3c336..24d543a01f43 100644 --- a/drivers/gpu/drm/nouveau/nouveau_fbcon.c +++ b/drivers/gpu/drm/nouveau/nouveau_fbcon.c @@ -558,14 +558,10 @@ nouveau_fbcon_init(struct drm_device *dev) drm_fb_helper_prepare(dev, &fbcon->helper, &nouveau_fbcon_helper_funcs); - ret = drm_fb_helper_init(dev, &fbcon->helper, 4); + ret = drm_fb_helper_init(dev, &fbcon->helper); if (ret) goto free; - ret = drm_fb_helper_single_add_all_connectors(&fbcon->helper); - if (ret) - goto fini; - if (preferred_bpp != 8 && preferred_bpp != 16 && preferred_bpp != 32) { if (drm->client.device.info.ram_size <= 32 * 1024 * 1024) preferred_bpp = 8; diff --git a/drivers/gpu/drm/nouveau/nouveau_svm.c b/drivers/gpu/drm/nouveau/nouveau_svm.c index df9bf1fd1bc0..645fedd77e21 100644 --- a/drivers/gpu/drm/nouveau/nouveau_svm.c +++ b/drivers/gpu/drm/nouveau/nouveau_svm.c @@ -171,6 +171,11 @@ nouveau_svmm_bind(struct drm_device *dev, void *data, mm = get_task_mm(current); down_read(&mm->mmap_sem); + if (!cli->svm.svmm) { + up_read(&mm->mmap_sem); + return -EINVAL; + } + for (addr = args->va_start, end = args->va_start + size; addr < end;) { struct vm_area_struct *vma; unsigned long next; @@ -179,6 +184,7 @@ nouveau_svmm_bind(struct drm_device *dev, void *data, if (!vma) break; + addr = max(addr, vma->vm_start); next = min(vma->vm_end, end); /* This is a best effort so we ignore errors */ nouveau_dmem_migrate_vma(cli->drm, vma, addr, next); @@ -367,7 +373,6 @@ static const u64 nouveau_svm_pfn_flags[HMM_PFN_FLAG_MAX] = { [HMM_PFN_VALID ] = NVIF_VMM_PFNMAP_V0_V, [HMM_PFN_WRITE ] = NVIF_VMM_PFNMAP_V0_W, - [HMM_PFN_DEVICE_PRIVATE] = NVIF_VMM_PFNMAP_V0_VRAM, }; static const u64 @@ -541,7 +546,7 @@ static int nouveau_range_fault(struct nouveau_svmm *svmm, range.default_flags = 0; range.pfn_flags_mask = -1UL; down_read(&mm->mmap_sem); - ret = hmm_range_fault(&range, 0); + ret = hmm_range_fault(&range); up_read(&mm->mmap_sem); if (ret <= 0) { if (ret == 0 || ret == -EBUSY) @@ -657,9 +662,6 @@ nouveau_svm_fault(struct nvif_notify *notify) limit = start + (ARRAY_SIZE(args.phys) << PAGE_SHIFT); if (start < svmm->unmanaged.limit) limit = min_t(u64, limit, svmm->unmanaged.start); - else - if (limit > svmm->unmanaged.start) - start = max_t(u64, start, svmm->unmanaged.limit); SVMM_DBG(svmm, "wndw %016llx-%016llx", start, limit); mm = svmm->notifier.mm; diff --git a/drivers/gpu/drm/nouveau/nvif/Kbuild b/drivers/gpu/drm/nouveau/nvif/Kbuild index 50d583d63807..f194d354c1f5 100644 --- a/drivers/gpu/drm/nouveau/nvif/Kbuild +++ b/drivers/gpu/drm/nouveau/nvif/Kbuild @@ -8,6 +8,7 @@ nvif-y += nvif/fifo.o nvif-y += nvif/mem.o nvif-y += nvif/mmu.o nvif-y += nvif/notify.o +nvif-y += nvif/timer.o nvif-y += nvif/vmm.o # Usermode classes diff --git a/drivers/gpu/drm/nouveau/nvif/device.c b/drivers/gpu/drm/nouveau/nvif/device.c index 1ec101ba3b42..0e92db44bbc8 100644 --- a/drivers/gpu/drm/nouveau/nvif/device.c +++ b/drivers/gpu/drm/nouveau/nvif/device.c @@ -27,11 +27,15 @@ u64 nvif_device_time(struct nvif_device *device) { - struct nv_device_time_v0 args = {}; - int ret = nvif_object_mthd(&device->object, NV_DEVICE_V0_TIME, - &args, sizeof(args)); - WARN_ON_ONCE(ret != 0); - return args.time; + if (!device->user.func) { + struct nv_device_time_v0 args = {}; + int ret = nvif_object_mthd(&device->object, NV_DEVICE_V0_TIME, + &args, sizeof(args)); + WARN_ON_ONCE(ret != 0); + return args.time; + } + + return device->user.func->time(&device->user); } void diff --git a/drivers/gpu/drm/nouveau/nvif/timer.c b/drivers/gpu/drm/nouveau/nvif/timer.c new file mode 100644 index 000000000000..602c1a258d10 --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvif/timer.c @@ -0,0 +1,56 @@ +/* + * Copyright 2020 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include <nvif/timer.h> +#include <nvif/device.h> + +s64 +nvif_timer_wait_test(struct nvif_timer_wait *wait) +{ + u64 time = nvif_device_time(wait->device); + + if (wait->reads == 0) { + wait->time0 = time; + wait->time1 = time; + } + + if (wait->time1 == time) { + if (WARN_ON(wait->reads++ == 16)) + return -ETIMEDOUT; + } else { + wait->time1 = time; + wait->reads = 1; + } + + if (wait->time1 - wait->time0 > wait->limit) + return -ETIMEDOUT; + + return wait->time1 - wait->time0; +} + +void +nvif_timer_wait_init(struct nvif_device *device, u64 nsec, + struct nvif_timer_wait *wait) +{ + wait->device = device; + wait->limit = nsec; + wait->reads = 0; +} diff --git a/drivers/gpu/drm/nouveau/nvif/userc361.c b/drivers/gpu/drm/nouveau/nvif/userc361.c index 19f9958e7e01..1116f871b272 100644 --- a/drivers/gpu/drm/nouveau/nvif/userc361.c +++ b/drivers/gpu/drm/nouveau/nvif/userc361.c @@ -21,6 +21,19 @@ */ #include <nvif/user.h> +static u64 +nvif_userc361_time(struct nvif_user *user) +{ + u32 hi, lo; + + do { + hi = nvif_rd32(&user->object, 0x084); + lo = nvif_rd32(&user->object, 0x080); + } while (hi != nvif_rd32(&user->object, 0x084)); + + return ((u64)hi << 32 | lo); +} + static void nvif_userc361_doorbell(struct nvif_user *user, u32 token) { @@ -30,4 +43,5 @@ nvif_userc361_doorbell(struct nvif_user *user, u32 token) const struct nvif_user_func nvif_userc361 = { .doorbell = nvif_userc361_doorbell, + .time = nvif_userc361_time, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c index c7d700916eae..8ebbe1656008 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c @@ -2579,6 +2579,7 @@ nv166_chipset = { static const struct nvkm_device_chip nv167_chipset = { .name = "TU117", + .acr = tu102_acr_new, .bar = tu102_bar_new, .bios = nvkm_bios_new, .bus = gf100_bus_new, @@ -2607,6 +2608,7 @@ nv167_chipset = { .disp = tu102_disp_new, .dma = gv100_dma_new, .fifo = tu102_fifo_new, + .gr = tu102_gr_new, .nvdec[0] = gm107_nvdec_new, .nvenc[0] = gm107_nvenc_new, .sec2 = tu102_sec2_new, @@ -2615,6 +2617,7 @@ nv167_chipset = { static const struct nvkm_device_chip nv168_chipset = { .name = "TU116", + .acr = tu102_acr_new, .bar = tu102_bar_new, .bios = nvkm_bios_new, .bus = gf100_bus_new, @@ -2643,6 +2646,7 @@ nv168_chipset = { .disp = tu102_disp_new, .dma = gv100_dma_new, .fifo = tu102_fifo_new, + .gr = tu102_gr_new, .nvdec[0] = gm107_nvdec_new, .nvenc[0] = gm107_nvenc_new, .sec2 = tu102_sec2_new, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c index dd8f85b8b3a7..f2f5636efac4 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c @@ -1981,8 +1981,34 @@ gf100_gr_init_(struct nvkm_gr *base) { struct gf100_gr *gr = gf100_gr(base); struct nvkm_subdev *subdev = &base->engine.subdev; + struct nvkm_device *device = subdev->device; + bool reset = device->chipset == 0x137 || device->chipset == 0x138; u32 ret; + /* On certain GP107/GP108 boards, we trigger a weird issue where + * GR will stop responding to PRI accesses after we've asked the + * SEC2 RTOS to boot the GR falcons. This happens with far more + * frequency when cold-booting a board (ie. returning from D3). + * + * The root cause for this is not known and has proven difficult + * to isolate, with many avenues being dead-ends. + * + * A workaround was discovered by Karol, whereby putting GR into + * reset for an extended period right before initialisation + * prevents the problem from occuring. + * + * XXX: As RM does not require any such workaround, this is more + * of a hack than a true fix. + */ + reset = nvkm_boolopt(device->cfgopt, "NvGrResetWar", reset); + if (reset) { + nvkm_mask(device, 0x000200, 0x00001000, 0x00000000); + nvkm_rd32(device, 0x000200); + msleep(50); + nvkm_mask(device, 0x000200, 0x00001000, 0x00001000); + nvkm_rd32(device, 0x000200); + } + nvkm_pmu_pgob(gr->base.engine.subdev.device->pmu, false); ret = nvkm_falcon_get(&gr->fecs.falcon, subdev); diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/tu102.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/tu102.c index 454668b1cf54..a9efa4d78be9 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/tu102.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/tu102.c @@ -164,6 +164,32 @@ MODULE_FIRMWARE("nvidia/tu106/gr/sw_nonctx.bin"); MODULE_FIRMWARE("nvidia/tu106/gr/sw_bundle_init.bin"); MODULE_FIRMWARE("nvidia/tu106/gr/sw_method_init.bin"); +MODULE_FIRMWARE("nvidia/tu117/gr/fecs_bl.bin"); +MODULE_FIRMWARE("nvidia/tu117/gr/fecs_inst.bin"); +MODULE_FIRMWARE("nvidia/tu117/gr/fecs_data.bin"); +MODULE_FIRMWARE("nvidia/tu117/gr/fecs_sig.bin"); +MODULE_FIRMWARE("nvidia/tu117/gr/gpccs_bl.bin"); +MODULE_FIRMWARE("nvidia/tu117/gr/gpccs_inst.bin"); +MODULE_FIRMWARE("nvidia/tu117/gr/gpccs_data.bin"); +MODULE_FIRMWARE("nvidia/tu117/gr/gpccs_sig.bin"); +MODULE_FIRMWARE("nvidia/tu117/gr/sw_ctx.bin"); +MODULE_FIRMWARE("nvidia/tu117/gr/sw_nonctx.bin"); +MODULE_FIRMWARE("nvidia/tu117/gr/sw_bundle_init.bin"); +MODULE_FIRMWARE("nvidia/tu117/gr/sw_method_init.bin"); + +MODULE_FIRMWARE("nvidia/tu116/gr/fecs_bl.bin"); +MODULE_FIRMWARE("nvidia/tu116/gr/fecs_inst.bin"); +MODULE_FIRMWARE("nvidia/tu116/gr/fecs_data.bin"); +MODULE_FIRMWARE("nvidia/tu116/gr/fecs_sig.bin"); +MODULE_FIRMWARE("nvidia/tu116/gr/gpccs_bl.bin"); +MODULE_FIRMWARE("nvidia/tu116/gr/gpccs_inst.bin"); +MODULE_FIRMWARE("nvidia/tu116/gr/gpccs_data.bin"); +MODULE_FIRMWARE("nvidia/tu116/gr/gpccs_sig.bin"); +MODULE_FIRMWARE("nvidia/tu116/gr/sw_ctx.bin"); +MODULE_FIRMWARE("nvidia/tu116/gr/sw_nonctx.bin"); +MODULE_FIRMWARE("nvidia/tu116/gr/sw_bundle_init.bin"); +MODULE_FIRMWARE("nvidia/tu116/gr/sw_method_init.bin"); + static const struct gf100_gr_fwif tu102_gr_fwif[] = { { 0, gm200_gr_load, &tu102_gr, &gp108_gr_fecs_acr, &gp108_gr_gpccs_acr }, diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/acr/tu102.c b/drivers/gpu/drm/nouveau/nvkm/subdev/acr/tu102.c index 7f4b89d82d32..d28d8f36ae24 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/acr/tu102.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/acr/tu102.c @@ -107,6 +107,12 @@ MODULE_FIRMWARE("nvidia/tu104/acr/ucode_unload.bin"); MODULE_FIRMWARE("nvidia/tu106/acr/unload_bl.bin"); MODULE_FIRMWARE("nvidia/tu106/acr/ucode_unload.bin"); +MODULE_FIRMWARE("nvidia/tu116/acr/unload_bl.bin"); +MODULE_FIRMWARE("nvidia/tu116/acr/ucode_unload.bin"); + +MODULE_FIRMWARE("nvidia/tu117/acr/unload_bl.bin"); +MODULE_FIRMWARE("nvidia/tu117/acr/ucode_unload.bin"); + static const struct nvkm_acr_hsf_fwif tu102_acr_unload_fwif[] = { { 0, nvkm_acr_hsfw_load, &gp108_acr_unload_0 }, @@ -130,6 +136,8 @@ tu102_acr_asb_0 = { MODULE_FIRMWARE("nvidia/tu102/acr/ucode_asb.bin"); MODULE_FIRMWARE("nvidia/tu104/acr/ucode_asb.bin"); MODULE_FIRMWARE("nvidia/tu106/acr/ucode_asb.bin"); +MODULE_FIRMWARE("nvidia/tu116/acr/ucode_asb.bin"); +MODULE_FIRMWARE("nvidia/tu117/acr/ucode_asb.bin"); static const struct nvkm_acr_hsf_fwif tu102_acr_asb_fwif[] = { @@ -154,6 +162,12 @@ MODULE_FIRMWARE("nvidia/tu104/acr/ucode_ahesasc.bin"); MODULE_FIRMWARE("nvidia/tu106/acr/bl.bin"); MODULE_FIRMWARE("nvidia/tu106/acr/ucode_ahesasc.bin"); +MODULE_FIRMWARE("nvidia/tu116/acr/bl.bin"); +MODULE_FIRMWARE("nvidia/tu116/acr/ucode_ahesasc.bin"); + +MODULE_FIRMWARE("nvidia/tu117/acr/bl.bin"); +MODULE_FIRMWARE("nvidia/tu117/acr/ucode_ahesasc.bin"); + static const struct nvkm_acr_hsf_fwif tu102_acr_ahesasc_fwif[] = { { 0, nvkm_acr_hsfw_load, &tu102_acr_ahesasc_0 }, diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/shadowpci.c b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/shadowpci.c index 9b91da09dc5f..8d9812a51ef6 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/shadowpci.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/shadowpci.c @@ -101,9 +101,13 @@ platform_init(struct nvkm_bios *bios, const char *name) else return ERR_PTR(-ENODEV); + if (!pdev->rom || pdev->romlen == 0) + return ERR_PTR(-ENODEV); + if ((priv = kmalloc(sizeof(*priv), GFP_KERNEL))) { + priv->size = pdev->romlen; if (ret = -ENODEV, - (priv->rom = pci_platform_rom(pdev, &priv->size))) + (priv->rom = ioremap(pdev->rom, pdev->romlen))) return priv; kfree(priv); } @@ -111,11 +115,20 @@ platform_init(struct nvkm_bios *bios, const char *name) return ERR_PTR(ret); } +static void +platform_fini(void *data) +{ + struct priv *priv = data; + + iounmap(priv->rom); + kfree(priv); +} + const struct nvbios_source nvbios_platform = { .name = "PLATFORM", .init = platform_init, - .fini = (void(*)(void *))kfree, + .fini = platform_fini, .read = pcirom_read, .rw = true, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gv100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gv100.c index 389bad312bf2..10ff5d053f7e 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gv100.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gv100.c @@ -51,3 +51,5 @@ MODULE_FIRMWARE("nvidia/gv100/nvdec/scrubber.bin"); MODULE_FIRMWARE("nvidia/tu102/nvdec/scrubber.bin"); MODULE_FIRMWARE("nvidia/tu104/nvdec/scrubber.bin"); MODULE_FIRMWARE("nvidia/tu106/nvdec/scrubber.bin"); +MODULE_FIRMWARE("nvidia/tu116/nvdec/scrubber.bin"); +MODULE_FIRMWARE("nvidia/tu117/nvdec/scrubber.bin"); diff --git a/drivers/gpu/drm/omapdrm/displays/Kconfig b/drivers/gpu/drm/omapdrm/displays/Kconfig index b562a8cd61bf..f2be594c7eff 100644 --- a/drivers/gpu/drm/omapdrm/displays/Kconfig +++ b/drivers/gpu/drm/omapdrm/displays/Kconfig @@ -1,28 +1,6 @@ # SPDX-License-Identifier: GPL-2.0-only menu "OMAPDRM External Display Device Drivers" -config DRM_OMAP_ENCODER_OPA362 - tristate "OPA362 external analog amplifier" - help - Driver for OPA362 external analog TV amplifier controlled - through a GPIO. - -config DRM_OMAP_ENCODER_TPD12S015 - tristate "TPD12S015 HDMI ESD protection and level shifter" - help - Driver for TPD12S015, which offers HDMI ESD protection and level - shifting. - -config DRM_OMAP_CONNECTOR_HDMI - tristate "HDMI Connector" - help - Driver for a generic HDMI connector. - -config DRM_OMAP_CONNECTOR_ANALOG_TV - tristate "Analog TV Connector" - help - Driver for a generic analog TV connector. - config DRM_OMAP_PANEL_DSI_CM tristate "Generic DSI Command Mode Panel" depends on BACKLIGHT_CLASS_DEVICE diff --git a/drivers/gpu/drm/omapdrm/displays/Makefile b/drivers/gpu/drm/omapdrm/displays/Makefile index cb76859dc574..488ddf153613 100644 --- a/drivers/gpu/drm/omapdrm/displays/Makefile +++ b/drivers/gpu/drm/omapdrm/displays/Makefile @@ -1,6 +1,2 @@ # SPDX-License-Identifier: GPL-2.0 -obj-$(CONFIG_DRM_OMAP_ENCODER_OPA362) += encoder-opa362.o -obj-$(CONFIG_DRM_OMAP_ENCODER_TPD12S015) += encoder-tpd12s015.o -obj-$(CONFIG_DRM_OMAP_CONNECTOR_HDMI) += connector-hdmi.o -obj-$(CONFIG_DRM_OMAP_CONNECTOR_ANALOG_TV) += connector-analog-tv.o obj-$(CONFIG_DRM_OMAP_PANEL_DSI_CM) += panel-dsi-cm.o diff --git a/drivers/gpu/drm/omapdrm/displays/connector-analog-tv.c b/drivers/gpu/drm/omapdrm/displays/connector-analog-tv.c deleted file mode 100644 index 0d20fab605d7..000000000000 --- a/drivers/gpu/drm/omapdrm/displays/connector-analog-tv.c +++ /dev/null @@ -1,97 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Analog TV Connector driver - * - * Copyright (C) 2013 Texas Instruments Incorporated - http://www.ti.com/ - * Author: Tomi Valkeinen <tomi.valkeinen@ti.com> - */ - -#include <linux/slab.h> -#include <linux/module.h> -#include <linux/platform_device.h> -#include <linux/of.h> - -#include "../dss/omapdss.h" - -struct panel_drv_data { - struct omap_dss_device dssdev; - - struct device *dev; -}; - -#define to_panel_data(x) container_of(x, struct panel_drv_data, dssdev) - -static int tvc_connect(struct omap_dss_device *src, - struct omap_dss_device *dst) -{ - return 0; -} - -static void tvc_disconnect(struct omap_dss_device *src, - struct omap_dss_device *dst) -{ -} - -static const struct omap_dss_device_ops tvc_ops = { - .connect = tvc_connect, - .disconnect = tvc_disconnect, -}; - -static int tvc_probe(struct platform_device *pdev) -{ - struct panel_drv_data *ddata; - struct omap_dss_device *dssdev; - - ddata = devm_kzalloc(&pdev->dev, sizeof(*ddata), GFP_KERNEL); - if (!ddata) - return -ENOMEM; - - platform_set_drvdata(pdev, ddata); - ddata->dev = &pdev->dev; - - dssdev = &ddata->dssdev; - dssdev->ops = &tvc_ops; - dssdev->dev = &pdev->dev; - dssdev->type = OMAP_DISPLAY_TYPE_VENC; - dssdev->display = true; - dssdev->owner = THIS_MODULE; - dssdev->of_ports = BIT(0); - - omapdss_display_init(dssdev); - omapdss_device_register(dssdev); - - return 0; -} - -static int __exit tvc_remove(struct platform_device *pdev) -{ - struct panel_drv_data *ddata = platform_get_drvdata(pdev); - - omapdss_device_unregister(&ddata->dssdev); - - return 0; -} - -static const struct of_device_id tvc_of_match[] = { - { .compatible = "omapdss,svideo-connector", }, - { .compatible = "omapdss,composite-video-connector", }, - {}, -}; - -MODULE_DEVICE_TABLE(of, tvc_of_match); - -static struct platform_driver tvc_connector_driver = { - .probe = tvc_probe, - .remove = __exit_p(tvc_remove), - .driver = { - .name = "connector-analog-tv", - .of_match_table = tvc_of_match, - .suppress_bind_attrs = true, - }, -}; - -module_platform_driver(tvc_connector_driver); - -MODULE_AUTHOR("Tomi Valkeinen <tomi.valkeinen@ti.com>"); -MODULE_DESCRIPTION("Analog TV Connector driver"); -MODULE_LICENSE("GPL"); diff --git a/drivers/gpu/drm/omapdrm/displays/connector-hdmi.c b/drivers/gpu/drm/omapdrm/displays/connector-hdmi.c deleted file mode 100644 index f5d69d810bb8..000000000000 --- a/drivers/gpu/drm/omapdrm/displays/connector-hdmi.c +++ /dev/null @@ -1,183 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * HDMI Connector driver - * - * Copyright (C) 2013 Texas Instruments Incorporated - http://www.ti.com/ - * Author: Tomi Valkeinen <tomi.valkeinen@ti.com> - */ - -#include <linux/gpio/consumer.h> -#include <linux/module.h> -#include <linux/mutex.h> -#include <linux/platform_device.h> -#include <linux/slab.h> - -#include "../dss/omapdss.h" - -struct panel_drv_data { - struct omap_dss_device dssdev; - void (*hpd_cb)(void *cb_data, enum drm_connector_status status); - void *hpd_cb_data; - struct mutex hpd_lock; - - struct device *dev; - - struct gpio_desc *hpd_gpio; -}; - -#define to_panel_data(x) container_of(x, struct panel_drv_data, dssdev) - -static int hdmic_connect(struct omap_dss_device *src, - struct omap_dss_device *dst) -{ - return 0; -} - -static void hdmic_disconnect(struct omap_dss_device *src, - struct omap_dss_device *dst) -{ -} - -static bool hdmic_detect(struct omap_dss_device *dssdev) -{ - struct panel_drv_data *ddata = to_panel_data(dssdev); - - return gpiod_get_value_cansleep(ddata->hpd_gpio); -} - -static void hdmic_register_hpd_cb(struct omap_dss_device *dssdev, - void (*cb)(void *cb_data, - enum drm_connector_status status), - void *cb_data) -{ - struct panel_drv_data *ddata = to_panel_data(dssdev); - - mutex_lock(&ddata->hpd_lock); - ddata->hpd_cb = cb; - ddata->hpd_cb_data = cb_data; - mutex_unlock(&ddata->hpd_lock); -} - -static void hdmic_unregister_hpd_cb(struct omap_dss_device *dssdev) -{ - struct panel_drv_data *ddata = to_panel_data(dssdev); - - mutex_lock(&ddata->hpd_lock); - ddata->hpd_cb = NULL; - ddata->hpd_cb_data = NULL; - mutex_unlock(&ddata->hpd_lock); -} - -static const struct omap_dss_device_ops hdmic_ops = { - .connect = hdmic_connect, - .disconnect = hdmic_disconnect, - - .detect = hdmic_detect, - .register_hpd_cb = hdmic_register_hpd_cb, - .unregister_hpd_cb = hdmic_unregister_hpd_cb, -}; - -static irqreturn_t hdmic_hpd_isr(int irq, void *data) -{ - struct panel_drv_data *ddata = data; - - mutex_lock(&ddata->hpd_lock); - if (ddata->hpd_cb) { - enum drm_connector_status status; - - if (hdmic_detect(&ddata->dssdev)) - status = connector_status_connected; - else - status = connector_status_disconnected; - - ddata->hpd_cb(ddata->hpd_cb_data, status); - } - mutex_unlock(&ddata->hpd_lock); - - return IRQ_HANDLED; -} - -static int hdmic_probe(struct platform_device *pdev) -{ - struct panel_drv_data *ddata; - struct omap_dss_device *dssdev; - struct gpio_desc *gpio; - int r; - - ddata = devm_kzalloc(&pdev->dev, sizeof(*ddata), GFP_KERNEL); - if (!ddata) - return -ENOMEM; - - platform_set_drvdata(pdev, ddata); - ddata->dev = &pdev->dev; - - mutex_init(&ddata->hpd_lock); - - /* HPD GPIO */ - gpio = devm_gpiod_get_optional(&pdev->dev, "hpd", GPIOD_IN); - if (IS_ERR(gpio)) { - dev_err(&pdev->dev, "failed to parse HPD gpio\n"); - return PTR_ERR(gpio); - } - - ddata->hpd_gpio = gpio; - - if (ddata->hpd_gpio) { - r = devm_request_threaded_irq(&pdev->dev, - gpiod_to_irq(ddata->hpd_gpio), - NULL, hdmic_hpd_isr, - IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING | - IRQF_ONESHOT, - "hdmic hpd", ddata); - if (r) - return r; - } - - dssdev = &ddata->dssdev; - dssdev->ops = &hdmic_ops; - dssdev->dev = &pdev->dev; - dssdev->type = OMAP_DISPLAY_TYPE_HDMI; - dssdev->display = true; - dssdev->owner = THIS_MODULE; - dssdev->of_ports = BIT(0); - dssdev->ops_flags = ddata->hpd_gpio - ? OMAP_DSS_DEVICE_OP_DETECT | OMAP_DSS_DEVICE_OP_HPD - : 0; - - omapdss_display_init(dssdev); - omapdss_device_register(dssdev); - - return 0; -} - -static int __exit hdmic_remove(struct platform_device *pdev) -{ - struct panel_drv_data *ddata = platform_get_drvdata(pdev); - - omapdss_device_unregister(&ddata->dssdev); - - return 0; -} - -static const struct of_device_id hdmic_of_match[] = { - { .compatible = "omapdss,hdmi-connector", }, - {}, -}; - -MODULE_DEVICE_TABLE(of, hdmic_of_match); - -static struct platform_driver hdmi_connector_driver = { - .probe = hdmic_probe, - .remove = __exit_p(hdmic_remove), - .driver = { - .name = "connector-hdmi", - .of_match_table = hdmic_of_match, - .suppress_bind_attrs = true, - }, -}; - -module_platform_driver(hdmi_connector_driver); - -MODULE_AUTHOR("Tomi Valkeinen <tomi.valkeinen@ti.com>"); -MODULE_DESCRIPTION("HDMI Connector driver"); -MODULE_LICENSE("GPL"); diff --git a/drivers/gpu/drm/omapdrm/displays/encoder-opa362.c b/drivers/gpu/drm/omapdrm/displays/encoder-opa362.c deleted file mode 100644 index b992387ed674..000000000000 --- a/drivers/gpu/drm/omapdrm/displays/encoder-opa362.c +++ /dev/null @@ -1,137 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * OPA362 analog video amplifier with output/power control - * - * Copyright (C) 2014 Golden Delicious Computers - * Author: H. Nikolaus Schaller <hns@goldelico.com> - * - * based on encoder-tfp410 - * - * Copyright (C) 2013 Texas Instruments Incorporated - http://www.ti.com/ - * Author: Tomi Valkeinen <tomi.valkeinen@ti.com> - */ - -#include <linux/gpio/consumer.h> -#include <linux/module.h> -#include <linux/platform_device.h> -#include <linux/slab.h> - -#include "../dss/omapdss.h" - -struct panel_drv_data { - struct omap_dss_device dssdev; - - struct gpio_desc *enable_gpio; -}; - -#define to_panel_data(x) container_of(x, struct panel_drv_data, dssdev) - -static int opa362_connect(struct omap_dss_device *src, - struct omap_dss_device *dst) -{ - return omapdss_device_connect(dst->dss, dst, dst->next); -} - -static void opa362_disconnect(struct omap_dss_device *src, - struct omap_dss_device *dst) -{ - omapdss_device_disconnect(dst, dst->next); -} - -static void opa362_enable(struct omap_dss_device *dssdev) -{ - struct panel_drv_data *ddata = to_panel_data(dssdev); - - if (ddata->enable_gpio) - gpiod_set_value_cansleep(ddata->enable_gpio, 1); -} - -static void opa362_disable(struct omap_dss_device *dssdev) -{ - struct panel_drv_data *ddata = to_panel_data(dssdev); - - if (ddata->enable_gpio) - gpiod_set_value_cansleep(ddata->enable_gpio, 0); -} - -static const struct omap_dss_device_ops opa362_ops = { - .connect = opa362_connect, - .disconnect = opa362_disconnect, - .enable = opa362_enable, - .disable = opa362_disable, -}; - -static int opa362_probe(struct platform_device *pdev) -{ - struct panel_drv_data *ddata; - struct omap_dss_device *dssdev; - struct gpio_desc *gpio; - - dev_dbg(&pdev->dev, "probe\n"); - - ddata = devm_kzalloc(&pdev->dev, sizeof(*ddata), GFP_KERNEL); - if (!ddata) - return -ENOMEM; - - platform_set_drvdata(pdev, ddata); - - gpio = devm_gpiod_get_optional(&pdev->dev, "enable", GPIOD_OUT_LOW); - if (IS_ERR(gpio)) - return PTR_ERR(gpio); - - ddata->enable_gpio = gpio; - - dssdev = &ddata->dssdev; - dssdev->ops = &opa362_ops; - dssdev->dev = &pdev->dev; - dssdev->type = OMAP_DISPLAY_TYPE_VENC; - dssdev->owner = THIS_MODULE; - dssdev->of_ports = BIT(1) | BIT(0); - - dssdev->next = omapdss_of_find_connected_device(pdev->dev.of_node, 1); - if (IS_ERR(dssdev->next)) { - if (PTR_ERR(dssdev->next) != -EPROBE_DEFER) - dev_err(&pdev->dev, "failed to find video sink\n"); - return PTR_ERR(dssdev->next); - } - - omapdss_device_register(dssdev); - - return 0; -} - -static int __exit opa362_remove(struct platform_device *pdev) -{ - struct panel_drv_data *ddata = platform_get_drvdata(pdev); - struct omap_dss_device *dssdev = &ddata->dssdev; - - if (dssdev->next) - omapdss_device_put(dssdev->next); - omapdss_device_unregister(&ddata->dssdev); - - opa362_disable(dssdev); - - return 0; -} - -static const struct of_device_id opa362_of_match[] = { - { .compatible = "omapdss,ti,opa362", }, - {}, -}; -MODULE_DEVICE_TABLE(of, opa362_of_match); - -static struct platform_driver opa362_driver = { - .probe = opa362_probe, - .remove = __exit_p(opa362_remove), - .driver = { - .name = "amplifier-opa362", - .of_match_table = opa362_of_match, - .suppress_bind_attrs = true, - }, -}; - -module_platform_driver(opa362_driver); - -MODULE_AUTHOR("H. Nikolaus Schaller <hns@goldelico.com>"); -MODULE_DESCRIPTION("OPA362 analog video amplifier with output/power control"); -MODULE_LICENSE("GPL v2"); diff --git a/drivers/gpu/drm/omapdrm/displays/encoder-tpd12s015.c b/drivers/gpu/drm/omapdrm/displays/encoder-tpd12s015.c deleted file mode 100644 index 089105c5aa0a..000000000000 --- a/drivers/gpu/drm/omapdrm/displays/encoder-tpd12s015.c +++ /dev/null @@ -1,217 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * TPD12S015 HDMI ESD protection & level shifter chip driver - * - * Copyright (C) 2013 Texas Instruments Incorporated - http://www.ti.com/ - * Author: Tomi Valkeinen <tomi.valkeinen@ti.com> - */ - -#include <linux/completion.h> -#include <linux/delay.h> -#include <linux/module.h> -#include <linux/slab.h> -#include <linux/platform_device.h> -#include <linux/gpio/consumer.h> -#include <linux/mutex.h> - -#include "../dss/omapdss.h" - -struct panel_drv_data { - struct omap_dss_device dssdev; - void (*hpd_cb)(void *cb_data, enum drm_connector_status status); - void *hpd_cb_data; - struct mutex hpd_lock; - - struct gpio_desc *ct_cp_hpd_gpio; - struct gpio_desc *ls_oe_gpio; - struct gpio_desc *hpd_gpio; -}; - -#define to_panel_data(x) container_of(x, struct panel_drv_data, dssdev) - -static int tpd_connect(struct omap_dss_device *src, - struct omap_dss_device *dst) -{ - struct panel_drv_data *ddata = to_panel_data(dst); - int r; - - r = omapdss_device_connect(dst->dss, dst, dst->next); - if (r) - return r; - - gpiod_set_value_cansleep(ddata->ct_cp_hpd_gpio, 1); - gpiod_set_value_cansleep(ddata->ls_oe_gpio, 1); - - /* DC-DC converter needs at max 300us to get to 90% of 5V */ - udelay(300); - - return 0; -} - -static void tpd_disconnect(struct omap_dss_device *src, - struct omap_dss_device *dst) -{ - struct panel_drv_data *ddata = to_panel_data(dst); - - gpiod_set_value_cansleep(ddata->ct_cp_hpd_gpio, 0); - gpiod_set_value_cansleep(ddata->ls_oe_gpio, 0); - - omapdss_device_disconnect(dst, dst->next); -} - -static bool tpd_detect(struct omap_dss_device *dssdev) -{ - struct panel_drv_data *ddata = to_panel_data(dssdev); - - return gpiod_get_value_cansleep(ddata->hpd_gpio); -} - -static void tpd_register_hpd_cb(struct omap_dss_device *dssdev, - void (*cb)(void *cb_data, - enum drm_connector_status status), - void *cb_data) -{ - struct panel_drv_data *ddata = to_panel_data(dssdev); - - mutex_lock(&ddata->hpd_lock); - ddata->hpd_cb = cb; - ddata->hpd_cb_data = cb_data; - mutex_unlock(&ddata->hpd_lock); -} - -static void tpd_unregister_hpd_cb(struct omap_dss_device *dssdev) -{ - struct panel_drv_data *ddata = to_panel_data(dssdev); - - mutex_lock(&ddata->hpd_lock); - ddata->hpd_cb = NULL; - ddata->hpd_cb_data = NULL; - mutex_unlock(&ddata->hpd_lock); -} - -static const struct omap_dss_device_ops tpd_ops = { - .connect = tpd_connect, - .disconnect = tpd_disconnect, - .detect = tpd_detect, - .register_hpd_cb = tpd_register_hpd_cb, - .unregister_hpd_cb = tpd_unregister_hpd_cb, -}; - -static irqreturn_t tpd_hpd_isr(int irq, void *data) -{ - struct panel_drv_data *ddata = data; - - mutex_lock(&ddata->hpd_lock); - if (ddata->hpd_cb) { - enum drm_connector_status status; - - if (tpd_detect(&ddata->dssdev)) - status = connector_status_connected; - else - status = connector_status_disconnected; - - ddata->hpd_cb(ddata->hpd_cb_data, status); - } - mutex_unlock(&ddata->hpd_lock); - - return IRQ_HANDLED; -} - -static int tpd_probe(struct platform_device *pdev) -{ - struct omap_dss_device *dssdev; - struct panel_drv_data *ddata; - int r; - struct gpio_desc *gpio; - - ddata = devm_kzalloc(&pdev->dev, sizeof(*ddata), GFP_KERNEL); - if (!ddata) - return -ENOMEM; - - platform_set_drvdata(pdev, ddata); - - gpio = devm_gpiod_get_index_optional(&pdev->dev, NULL, 0, - GPIOD_OUT_LOW); - if (IS_ERR(gpio)) - return PTR_ERR(gpio); - - ddata->ct_cp_hpd_gpio = gpio; - - gpio = devm_gpiod_get_index_optional(&pdev->dev, NULL, 1, - GPIOD_OUT_LOW); - if (IS_ERR(gpio)) - return PTR_ERR(gpio); - - ddata->ls_oe_gpio = gpio; - - gpio = devm_gpiod_get_index(&pdev->dev, NULL, 2, - GPIOD_IN); - if (IS_ERR(gpio)) - return PTR_ERR(gpio); - - ddata->hpd_gpio = gpio; - - mutex_init(&ddata->hpd_lock); - - r = devm_request_threaded_irq(&pdev->dev, gpiod_to_irq(ddata->hpd_gpio), - NULL, tpd_hpd_isr, - IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING | IRQF_ONESHOT, - "tpd12s015 hpd", ddata); - if (r) - return r; - - dssdev = &ddata->dssdev; - dssdev->ops = &tpd_ops; - dssdev->dev = &pdev->dev; - dssdev->type = OMAP_DISPLAY_TYPE_HDMI; - dssdev->owner = THIS_MODULE; - dssdev->of_ports = BIT(1) | BIT(0); - dssdev->ops_flags = OMAP_DSS_DEVICE_OP_DETECT - | OMAP_DSS_DEVICE_OP_HPD; - - dssdev->next = omapdss_of_find_connected_device(pdev->dev.of_node, 1); - if (IS_ERR(dssdev->next)) { - if (PTR_ERR(dssdev->next) != -EPROBE_DEFER) - dev_err(&pdev->dev, "failed to find video sink\n"); - return PTR_ERR(dssdev->next); - } - - omapdss_device_register(dssdev); - - return 0; -} - -static int __exit tpd_remove(struct platform_device *pdev) -{ - struct panel_drv_data *ddata = platform_get_drvdata(pdev); - struct omap_dss_device *dssdev = &ddata->dssdev; - - if (dssdev->next) - omapdss_device_put(dssdev->next); - omapdss_device_unregister(&ddata->dssdev); - - return 0; -} - -static const struct of_device_id tpd_of_match[] = { - { .compatible = "omapdss,ti,tpd12s015", }, - {}, -}; - -MODULE_DEVICE_TABLE(of, tpd_of_match); - -static struct platform_driver tpd_driver = { - .probe = tpd_probe, - .remove = __exit_p(tpd_remove), - .driver = { - .name = "tpd12s015", - .of_match_table = tpd_of_match, - .suppress_bind_attrs = true, - }, -}; - -module_platform_driver(tpd_driver); - -MODULE_AUTHOR("Tomi Valkeinen <tomi.valkeinen@ti.com>"); -MODULE_DESCRIPTION("TPD12S015 driver"); -MODULE_LICENSE("GPL"); diff --git a/drivers/gpu/drm/omapdrm/displays/panel-dsi-cm.c b/drivers/gpu/drm/omapdrm/displays/panel-dsi-cm.c index 564e3e1a1891..3484b5d4a91c 100644 --- a/drivers/gpu/drm/omapdrm/displays/panel-dsi-cm.c +++ b/drivers/gpu/drm/omapdrm/displays/panel-dsi-cm.c @@ -678,7 +678,7 @@ static int dsicm_power_on(struct panel_drv_data *ddata) if (r) goto err; - ddata->enabled = 1; + ddata->enabled = true; if (!ddata->intro_printed) { dev_info(&ddata->pdev->dev, "panel revision %02x.%02x.%02x\n", @@ -729,7 +729,7 @@ static void dsicm_power_off(struct panel_drv_data *ddata) if (ddata->vpnl) regulator_disable(ddata->vpnl); - ddata->enabled = 0; + ddata->enabled = false; } static int dsicm_panel_reset(struct panel_drv_data *ddata) @@ -1265,7 +1265,7 @@ static int dsicm_probe(struct platform_device *pdev) dssdev->type = OMAP_DISPLAY_TYPE_DSI; dssdev->display = true; dssdev->owner = THIS_MODULE; - dssdev->of_ports = BIT(0); + dssdev->of_port = 0; dssdev->ops_flags = OMAP_DSS_DEVICE_OP_MODES; dssdev->caps = OMAP_DSS_DISPLAY_CAP_MANUAL_UPDATE | diff --git a/drivers/gpu/drm/omapdrm/dss/Makefile b/drivers/gpu/drm/omapdrm/dss/Makefile index 5950c3f52c2e..f967e6948f2e 100644 --- a/drivers/gpu/drm/omapdrm/dss/Makefile +++ b/drivers/gpu/drm/omapdrm/dss/Makefile @@ -2,7 +2,7 @@ obj-$(CONFIG_OMAP2_DSS_INIT) += omapdss-boot-init.o obj-$(CONFIG_OMAP_DSS_BASE) += omapdss-base.o -omapdss-base-y := base.o display.o dss-of.o output.o +omapdss-base-y := base.o display.o output.o obj-$(CONFIG_OMAP2_DSS) += omapdss.o # Core DSS files diff --git a/drivers/gpu/drm/omapdrm/dss/base.c b/drivers/gpu/drm/omapdrm/dss/base.c index a1970b9db6ab..c7650a7c155d 100644 --- a/drivers/gpu/drm/omapdrm/dss/base.c +++ b/drivers/gpu/drm/omapdrm/dss/base.c @@ -149,8 +149,7 @@ struct omap_dss_device *omapdss_device_next_output(struct omap_dss_device *from) goto done; } - if (dssdev->id && - (dssdev->next || dssdev->bridge || dssdev->panel)) + if (dssdev->id && (dssdev->next || dssdev->bridge)) goto done; } @@ -185,11 +184,10 @@ int omapdss_device_connect(struct dss_device *dss, if (!dst) { /* * The destination is NULL when the source is connected to a - * bridge or panel instead of a DSS device. Stop here, we will - * attach the bridge or panel later when we will have a DRM - * encoder. + * bridge instead of a DSS device. Stop here, we will attach + * the bridge later when we will have a DRM encoder. */ - return src && (src->bridge || src->panel) ? 0 : -EINVAL; + return src && src->bridge ? 0 : -EINVAL; } if (omapdss_device_is_connected(dst)) @@ -197,10 +195,12 @@ int omapdss_device_connect(struct dss_device *dss, dst->dss = dss; - ret = dst->ops->connect(src, dst); - if (ret < 0) { - dst->dss = NULL; - return ret; + if (dst->ops && dst->ops->connect) { + ret = dst->ops->connect(src, dst); + if (ret < 0) { + dst->dss = NULL; + return ret; + } } return 0; @@ -217,7 +217,7 @@ void omapdss_device_disconnect(struct omap_dss_device *src, dst ? dev_name(dst->dev) : "NULL"); if (!dst) { - WARN_ON(!src->bridge && !src->panel); + WARN_ON(!src->bridge); return; } @@ -228,29 +228,18 @@ void omapdss_device_disconnect(struct omap_dss_device *src, WARN_ON(dst->state != OMAP_DSS_DISPLAY_DISABLED); - dst->ops->disconnect(src, dst); + if (dst->ops && dst->ops->disconnect) + dst->ops->disconnect(src, dst); dst->dss = NULL; } EXPORT_SYMBOL_GPL(omapdss_device_disconnect); -void omapdss_device_pre_enable(struct omap_dss_device *dssdev) -{ - if (!dssdev) - return; - - omapdss_device_pre_enable(dssdev->next); - - if (dssdev->ops->pre_enable) - dssdev->ops->pre_enable(dssdev); -} -EXPORT_SYMBOL_GPL(omapdss_device_pre_enable); - void omapdss_device_enable(struct omap_dss_device *dssdev) { if (!dssdev) return; - if (dssdev->ops->enable) + if (dssdev->ops && dssdev->ops->enable) dssdev->ops->enable(dssdev); omapdss_device_enable(dssdev->next); @@ -266,25 +255,11 @@ void omapdss_device_disable(struct omap_dss_device *dssdev) omapdss_device_disable(dssdev->next); - if (dssdev->ops->disable) + if (dssdev->ops && dssdev->ops->disable) dssdev->ops->disable(dssdev); } EXPORT_SYMBOL_GPL(omapdss_device_disable); -void omapdss_device_post_disable(struct omap_dss_device *dssdev) -{ - if (!dssdev) - return; - - if (dssdev->ops->post_disable) - dssdev->ops->post_disable(dssdev); - - omapdss_device_post_disable(dssdev->next); - - dssdev->state = OMAP_DSS_DISPLAY_DISABLED; -} -EXPORT_SYMBOL_GPL(omapdss_device_post_disable); - /* ----------------------------------------------------------------------------- * Components Handling */ diff --git a/drivers/gpu/drm/omapdrm/dss/display.c b/drivers/gpu/drm/omapdrm/dss/display.c index 8a3f61f5825f..3b82158b1bfd 100644 --- a/drivers/gpu/drm/omapdrm/dss/display.c +++ b/drivers/gpu/drm/omapdrm/dss/display.c @@ -40,15 +40,6 @@ void omapdss_display_init(struct omap_dss_device *dssdev) } EXPORT_SYMBOL_GPL(omapdss_display_init); -struct omap_dss_device *omapdss_display_get(struct omap_dss_device *output) -{ - while (output->next) - output = output->next; - - return omapdss_device_get(output); -} -EXPORT_SYMBOL_GPL(omapdss_display_get); - int omapdss_display_get_modes(struct drm_connector *connector, const struct videomode *vm) { diff --git a/drivers/gpu/drm/omapdrm/dss/dpi.c b/drivers/gpu/drm/omapdrm/dss/dpi.c index 95147437b990..5110acb0c6c1 100644 --- a/drivers/gpu/drm/omapdrm/dss/dpi.c +++ b/drivers/gpu/drm/omapdrm/dss/dpi.c @@ -9,20 +9,22 @@ #define DSS_SUBSYS_NAME "DPI" -#include <linux/kernel.h> +#include <linux/clk.h> #include <linux/delay.h> -#include <linux/export.h> #include <linux/err.h> #include <linux/errno.h> +#include <linux/export.h> +#include <linux/kernel.h> +#include <linux/of.h> #include <linux/platform_device.h> #include <linux/regulator/consumer.h> #include <linux/string.h> -#include <linux/of.h> -#include <linux/clk.h> #include <linux/sys_soc.h> -#include "omapdss.h" +#include <drm/drm_bridge.h> + #include "dss.h" +#include "omapdss.h" struct dpi_data { struct platform_device *pdev; @@ -34,19 +36,19 @@ struct dpi_data { enum dss_clk_source clk_src; struct dss_pll *pll; - struct mutex lock; - struct dss_lcd_mgr_config mgr_config; unsigned long pixelclock; int data_lines; struct omap_dss_device output; + struct drm_bridge bridge; }; -static struct dpi_data *dpi_get_data_from_dssdev(struct omap_dss_device *dssdev) -{ - return container_of(dssdev, struct dpi_data, output); -} +#define drm_bridge_to_dpi(bridge) container_of(bridge, struct dpi_data, bridge) + +/* ----------------------------------------------------------------------------- + * Clock Handling and PLL + */ static enum dss_clk_source dpi_get_clk_src_dra7xx(struct dpi_data *dpi, enum omap_channel channel) @@ -283,9 +285,7 @@ static bool dpi_dss_clk_calc(struct dpi_data *dpi, unsigned long pck, -static int dpi_set_pll_clk(struct dpi_data *dpi, enum omap_channel channel, - unsigned long pck_req, unsigned long *fck, int *lck_div, - int *pck_div) +static int dpi_set_pll_clk(struct dpi_data *dpi, unsigned long pck_req) { struct dpi_clk_calc_ctx ctx; int r; @@ -299,19 +299,15 @@ static int dpi_set_pll_clk(struct dpi_data *dpi, enum omap_channel channel, if (r) return r; - dss_select_lcd_clk_source(dpi->dss, channel, dpi->clk_src); + dss_select_lcd_clk_source(dpi->dss, dpi->output.dispc_channel, + dpi->clk_src); dpi->mgr_config.clock_info = ctx.dispc_cinfo; - *fck = ctx.pll_cinfo.clkout[ctx.clkout_idx]; - *lck_div = ctx.dispc_cinfo.lck_div; - *pck_div = ctx.dispc_cinfo.pck_div; - return 0; } -static int dpi_set_dispc_clk(struct dpi_data *dpi, unsigned long pck_req, - unsigned long *fck, int *lck_div, int *pck_div) +static int dpi_set_dispc_clk(struct dpi_data *dpi, unsigned long pck_req) { struct dpi_clk_calc_ctx ctx; int r; @@ -327,29 +323,19 @@ static int dpi_set_dispc_clk(struct dpi_data *dpi, unsigned long pck_req, dpi->mgr_config.clock_info = ctx.dispc_cinfo; - *fck = ctx.fck; - *lck_div = ctx.dispc_cinfo.lck_div; - *pck_div = ctx.dispc_cinfo.pck_div; - return 0; } static int dpi_set_mode(struct dpi_data *dpi) { - int lck_div = 0, pck_div = 0; - unsigned long fck = 0; - int r = 0; + int r; if (dpi->pll) - r = dpi_set_pll_clk(dpi, dpi->output.dispc_channel, - dpi->pixelclock, &fck, &lck_div, &pck_div); + r = dpi_set_pll_clk(dpi, dpi->pixelclock); else - r = dpi_set_dispc_clk(dpi, dpi->pixelclock, &fck, - &lck_div, &pck_div); - if (r) - return r; + r = dpi_set_dispc_clk(dpi, dpi->pixelclock); - return 0; + return r; } static void dpi_config_lcd_manager(struct dpi_data *dpi) @@ -366,25 +352,149 @@ static void dpi_config_lcd_manager(struct dpi_data *dpi) dss_mgr_set_lcd_config(&dpi->output, &dpi->mgr_config); } -static void dpi_display_enable(struct omap_dss_device *dssdev) +static int dpi_clock_update(struct dpi_data *dpi, unsigned long *clock) +{ + int lck_div, pck_div; + unsigned long fck; + struct dpi_clk_calc_ctx ctx; + + if (dpi->pll) { + if (!dpi_pll_clk_calc(dpi, *clock, &ctx)) + return -EINVAL; + + fck = ctx.pll_cinfo.clkout[ctx.clkout_idx]; + } else { + if (!dpi_dss_clk_calc(dpi, *clock, &ctx)) + return -EINVAL; + + fck = ctx.fck; + } + + lck_div = ctx.dispc_cinfo.lck_div; + pck_div = ctx.dispc_cinfo.pck_div; + + *clock = fck / lck_div / pck_div; + + return 0; +} + +static int dpi_verify_pll(struct dss_pll *pll) { - struct dpi_data *dpi = dpi_get_data_from_dssdev(dssdev); - struct omap_dss_device *out = &dpi->output; int r; - mutex_lock(&dpi->lock); + /* do initial setup with the PLL to see if it is operational */ + + r = dss_pll_enable(pll); + if (r) + return r; + + dss_pll_disable(pll); + + return 0; +} + +static void dpi_init_pll(struct dpi_data *dpi) +{ + struct dss_pll *pll; + + if (dpi->pll) + return; + + dpi->clk_src = dpi_get_clk_src(dpi); + + pll = dss_pll_find_by_src(dpi->dss, dpi->clk_src); + if (!pll) + return; + + if (dpi_verify_pll(pll)) { + DSSWARN("PLL not operational\n"); + return; + } + + dpi->pll = pll; +} + +/* ----------------------------------------------------------------------------- + * DRM Bridge Operations + */ + +static int dpi_bridge_attach(struct drm_bridge *bridge, + enum drm_bridge_attach_flags flags) +{ + struct dpi_data *dpi = drm_bridge_to_dpi(bridge); + + if (!(flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR)) + return -EINVAL; + + dpi_init_pll(dpi); + + return drm_bridge_attach(bridge->encoder, dpi->output.next_bridge, + bridge, flags); +} + +static enum drm_mode_status +dpi_bridge_mode_valid(struct drm_bridge *bridge, + const struct drm_display_mode *mode) +{ + struct dpi_data *dpi = drm_bridge_to_dpi(bridge); + unsigned long clock = mode->clock * 1000; + int ret; + + if (mode->hdisplay % 8 != 0) + return MODE_BAD_WIDTH; + + if (mode->clock == 0) + return MODE_NOCLOCK; + + ret = dpi_clock_update(dpi, &clock); + if (ret < 0) + return MODE_CLOCK_RANGE; + + return MODE_OK; +} + +static bool dpi_bridge_mode_fixup(struct drm_bridge *bridge, + const struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode) +{ + struct dpi_data *dpi = drm_bridge_to_dpi(bridge); + unsigned long clock = mode->clock * 1000; + int ret; + + ret = dpi_clock_update(dpi, &clock); + if (ret < 0) + return false; + + adjusted_mode->clock = clock / 1000; + + return true; +} + +static void dpi_bridge_mode_set(struct drm_bridge *bridge, + const struct drm_display_mode *mode, + const struct drm_display_mode *adjusted_mode) +{ + struct dpi_data *dpi = drm_bridge_to_dpi(bridge); + + dpi->pixelclock = adjusted_mode->clock * 1000; +} + +static void dpi_bridge_enable(struct drm_bridge *bridge) +{ + struct dpi_data *dpi = drm_bridge_to_dpi(bridge); + int r; if (dpi->vdds_dsi_reg) { r = regulator_enable(dpi->vdds_dsi_reg); if (r) - goto err_reg_enable; + return; } r = dispc_runtime_get(dpi->dss->dispc); if (r) goto err_get_dispc; - r = dss_dpi_select_source(dpi->dss, dpi->id, out->dispc_channel); + r = dss_dpi_select_source(dpi->dss, dpi->id, dpi->output.dispc_channel); if (r) goto err_src_sel; @@ -406,8 +516,6 @@ static void dpi_display_enable(struct omap_dss_device *dssdev) if (r) goto err_mgr_enable; - mutex_unlock(&dpi->lock); - return; err_mgr_enable: @@ -420,15 +528,11 @@ err_src_sel: err_get_dispc: if (dpi->vdds_dsi_reg) regulator_disable(dpi->vdds_dsi_reg); -err_reg_enable: - mutex_unlock(&dpi->lock); } -static void dpi_display_disable(struct omap_dss_device *dssdev) +static void dpi_bridge_disable(struct drm_bridge *bridge) { - struct dpi_data *dpi = dpi_get_data_from_dssdev(dssdev); - - mutex_lock(&dpi->lock); + struct dpi_data *dpi = drm_bridge_to_dpi(bridge); dss_mgr_disable(&dpi->output); @@ -442,99 +546,34 @@ static void dpi_display_disable(struct omap_dss_device *dssdev) if (dpi->vdds_dsi_reg) regulator_disable(dpi->vdds_dsi_reg); - - mutex_unlock(&dpi->lock); } -static void dpi_set_timings(struct omap_dss_device *dssdev, - const struct drm_display_mode *mode) -{ - struct dpi_data *dpi = dpi_get_data_from_dssdev(dssdev); - - DSSDBG("dpi_set_timings\n"); - - mutex_lock(&dpi->lock); - - dpi->pixelclock = mode->clock * 1000; - - mutex_unlock(&dpi->lock); -} +static const struct drm_bridge_funcs dpi_bridge_funcs = { + .attach = dpi_bridge_attach, + .mode_valid = dpi_bridge_mode_valid, + .mode_fixup = dpi_bridge_mode_fixup, + .mode_set = dpi_bridge_mode_set, + .enable = dpi_bridge_enable, + .disable = dpi_bridge_disable, +}; -static int dpi_check_timings(struct omap_dss_device *dssdev, - struct drm_display_mode *mode) +static void dpi_bridge_init(struct dpi_data *dpi) { - struct dpi_data *dpi = dpi_get_data_from_dssdev(dssdev); - int lck_div, pck_div; - unsigned long fck; - unsigned long pck; - struct dpi_clk_calc_ctx ctx; - bool ok; + dpi->bridge.funcs = &dpi_bridge_funcs; + dpi->bridge.of_node = dpi->pdev->dev.of_node; + dpi->bridge.type = DRM_MODE_CONNECTOR_DPI; - if (mode->hdisplay % 8 != 0) - return -EINVAL; - - if (mode->clock == 0) - return -EINVAL; - - if (dpi->pll) { - ok = dpi_pll_clk_calc(dpi, mode->clock * 1000, &ctx); - if (!ok) - return -EINVAL; - - fck = ctx.pll_cinfo.clkout[ctx.clkout_idx]; - } else { - ok = dpi_dss_clk_calc(dpi, mode->clock * 1000, &ctx); - if (!ok) - return -EINVAL; - - fck = ctx.fck; - } - - lck_div = ctx.dispc_cinfo.lck_div; - pck_div = ctx.dispc_cinfo.pck_div; - - pck = fck / lck_div / pck_div; - - mode->clock = pck / 1000; - - return 0; + drm_bridge_add(&dpi->bridge); } -static int dpi_verify_pll(struct dss_pll *pll) +static void dpi_bridge_cleanup(struct dpi_data *dpi) { - int r; - - /* do initial setup with the PLL to see if it is operational */ - - r = dss_pll_enable(pll); - if (r) - return r; - - dss_pll_disable(pll); - - return 0; + drm_bridge_remove(&dpi->bridge); } -static void dpi_init_pll(struct dpi_data *dpi) -{ - struct dss_pll *pll; - - if (dpi->pll) - return; - - dpi->clk_src = dpi_get_clk_src(dpi); - - pll = dss_pll_find_by_src(dpi->dss, dpi->clk_src); - if (!pll) - return; - - if (dpi_verify_pll(pll)) { - DSSWARN("PLL not operational\n"); - return; - } - - dpi->pll = pll; -} +/* ----------------------------------------------------------------------------- + * Initialisation and Cleanup + */ /* * Return a hardcoded channel for the DPI output. This should work for @@ -572,39 +611,14 @@ static enum omap_channel dpi_get_channel(struct dpi_data *dpi) } } -static int dpi_connect(struct omap_dss_device *src, - struct omap_dss_device *dst) -{ - struct dpi_data *dpi = dpi_get_data_from_dssdev(dst); - - dpi_init_pll(dpi); - - return omapdss_device_connect(dst->dss, dst, dst->next); -} - -static void dpi_disconnect(struct omap_dss_device *src, - struct omap_dss_device *dst) -{ - omapdss_device_disconnect(dst, dst->next); -} - -static const struct omap_dss_device_ops dpi_ops = { - .connect = dpi_connect, - .disconnect = dpi_disconnect, - - .enable = dpi_display_enable, - .disable = dpi_display_disable, - - .check_timings = dpi_check_timings, - .set_timings = dpi_set_timings, -}; - static int dpi_init_output_port(struct dpi_data *dpi, struct device_node *port) { struct omap_dss_device *out = &dpi->output; u32 port_num = 0; int r; + dpi_bridge_init(dpi); + of_property_read_u32(port, "reg", &port_num); dpi->id = port_num <= 2 ? port_num : 0; @@ -625,13 +639,14 @@ static int dpi_init_output_port(struct dpi_data *dpi, struct device_node *port) out->id = OMAP_DSS_OUTPUT_DPI; out->type = OMAP_DISPLAY_TYPE_DPI; out->dispc_channel = dpi_get_channel(dpi); - out->of_ports = BIT(port_num); - out->ops = &dpi_ops; + out->of_port = port_num; out->owner = THIS_MODULE; - r = omapdss_device_init_output(out); - if (r < 0) + r = omapdss_device_init_output(out, &dpi->bridge); + if (r < 0) { + dpi_bridge_cleanup(dpi); return r; + } omapdss_device_register(out); @@ -645,8 +660,14 @@ static void dpi_uninit_output_port(struct device_node *port) omapdss_device_unregister(out); omapdss_device_cleanup_output(out); + + dpi_bridge_cleanup(dpi); } +/* ----------------------------------------------------------------------------- + * Initialisation and Cleanup + */ + static const struct soc_device_attribute dpi_soc_devices[] = { { .machine = "OMAP3[456]*" }, { .machine = "[AD]M37*" }, @@ -706,8 +727,6 @@ int dpi_init_port(struct dss_device *dss, struct platform_device *pdev, dpi->dss = dss; port->data = dpi; - mutex_init(&dpi->lock); - r = dpi_init_regulator(dpi); if (r) return r; diff --git a/drivers/gpu/drm/omapdrm/dss/dsi.c b/drivers/gpu/drm/omapdrm/dss/dsi.c index da16ea095f13..79ddfbfd1b58 100644 --- a/drivers/gpu/drm/omapdrm/dss/dsi.c +++ b/drivers/gpu/drm/omapdrm/dss/dsi.c @@ -5116,12 +5116,12 @@ static int dsi_init_output(struct dsi_data *dsi) out->dispc_channel = dsi_get_channel(dsi); out->ops = &dsi_ops; out->owner = THIS_MODULE; - out->of_ports = BIT(0); + out->of_port = 0; out->bus_flags = DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE | DRM_BUS_FLAG_DE_HIGH | DRM_BUS_FLAG_SYNC_DRIVE_NEGEDGE; - r = omapdss_device_init_output(out); + r = omapdss_device_init_output(out, NULL); if (r < 0) return r; diff --git a/drivers/gpu/drm/omapdrm/dss/dss-of.c b/drivers/gpu/drm/omapdrm/dss/dss-of.c deleted file mode 100644 index b7981f3b80ad..000000000000 --- a/drivers/gpu/drm/omapdrm/dss/dss-of.c +++ /dev/null @@ -1,28 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Copyright (C) 2013 Texas Instruments Incorporated - http://www.ti.com/ - * Author: Tomi Valkeinen <tomi.valkeinen@ti.com> - */ - -#include <linux/err.h> -#include <linux/of.h> -#include <linux/of_graph.h> - -#include "omapdss.h" - -struct omap_dss_device * -omapdss_of_find_connected_device(struct device_node *node, unsigned int port) -{ - struct device_node *remote_node; - struct omap_dss_device *dssdev; - - remote_node = of_graph_get_remote_node(node, port, 0); - if (!remote_node) - return NULL; - - dssdev = omapdss_find_device_by_node(remote_node); - of_node_put(remote_node); - - return dssdev ? dssdev : ERR_PTR(-EPROBE_DEFER); -} -EXPORT_SYMBOL_GPL(omapdss_of_find_connected_device); diff --git a/drivers/gpu/drm/omapdrm/dss/dss.c b/drivers/gpu/drm/omapdrm/dss/dss.c index 225ec808b01a..4d5739fa4a5d 100644 --- a/drivers/gpu/drm/omapdrm/dss/dss.c +++ b/drivers/gpu/drm/omapdrm/dss/dss.c @@ -1151,46 +1151,38 @@ static const struct dss_features dra7xx_dss_feats = { .has_lcd_clk_src = true, }; -static int dss_init_ports(struct dss_device *dss) +static void __dss_uninit_ports(struct dss_device *dss, unsigned int num_ports) { struct platform_device *pdev = dss->pdev; struct device_node *parent = pdev->dev.of_node; struct device_node *port; unsigned int i; - int r; - for (i = 0; i < dss->feat->num_ports; i++) { + for (i = 0; i < num_ports; i++) { port = of_graph_get_port_by_id(parent, i); if (!port) continue; switch (dss->feat->ports[i]) { case OMAP_DISPLAY_TYPE_DPI: - r = dpi_init_port(dss, pdev, port, dss->feat->model); - if (r) - return r; + dpi_uninit_port(port); break; - case OMAP_DISPLAY_TYPE_SDI: - r = sdi_init_port(dss, pdev, port); - if (r) - return r; + sdi_uninit_port(port); break; - default: break; } } - - return 0; } -static void dss_uninit_ports(struct dss_device *dss) +static int dss_init_ports(struct dss_device *dss) { struct platform_device *pdev = dss->pdev; struct device_node *parent = pdev->dev.of_node; struct device_node *port; - int i; + unsigned int i; + int r; for (i = 0; i < dss->feat->num_ports; i++) { port = of_graph_get_port_by_id(parent, i); @@ -1199,15 +1191,32 @@ static void dss_uninit_ports(struct dss_device *dss) switch (dss->feat->ports[i]) { case OMAP_DISPLAY_TYPE_DPI: - dpi_uninit_port(port); + r = dpi_init_port(dss, pdev, port, dss->feat->model); + if (r) + goto error; break; + case OMAP_DISPLAY_TYPE_SDI: - sdi_uninit_port(port); + r = sdi_init_port(dss, pdev, port); + if (r) + goto error; break; + default: break; } } + + return 0; + +error: + __dss_uninit_ports(dss, i); + return r; +} + +static void dss_uninit_ports(struct dss_device *dss) +{ + __dss_uninit_ports(dss, dss->feat->num_ports); } static int dss_video_pll_probe(struct dss_device *dss) @@ -1339,9 +1348,15 @@ static int dss_component_compare(struct device *dev, void *data) return dev == child; } +struct dss_component_match_data { + struct device *dev; + struct component_match **match; +}; + static int dss_add_child_component(struct device *dev, void *data) { - struct component_match **match = data; + struct dss_component_match_data *cmatch = data; + struct component_match **match = cmatch->match; /* * HACK @@ -1352,7 +1367,17 @@ static int dss_add_child_component(struct device *dev, void *data) if (strstr(dev_name(dev), "rfbi")) return 0; - component_match_add(dev->parent, match, dss_component_compare, dev); + /* + * Handle possible interconnect target modules defined within the DSS. + * The DSS components can be children of an interconnect target module + * after the device tree has been updated for the module data. + * See also omapdss_boot_init() for compatible fixup. + */ + if (strstr(dev_name(dev), "target-module")) + return device_for_each_child(dev, cmatch, + dss_add_child_component); + + component_match_add(cmatch->dev, match, dss_component_compare, dev); return 0; } @@ -1395,6 +1420,7 @@ static int dss_probe_hardware(struct dss_device *dss) static int dss_probe(struct platform_device *pdev) { const struct soc_device_attribute *soc; + struct dss_component_match_data cmatch; struct component_match *match = NULL; struct resource *dss_mem; struct dss_device *dss; @@ -1472,7 +1498,9 @@ static int dss_probe(struct platform_device *pdev) omapdss_gather_components(&pdev->dev); - device_for_each_child(&pdev->dev, &match, dss_add_child_component); + cmatch.dev = &pdev->dev; + cmatch.match = &match; + device_for_each_child(&pdev->dev, &cmatch, dss_add_child_component); r = component_master_add_with_match(&pdev->dev, &dss_component_ops, match); if (r) @@ -1543,7 +1571,8 @@ static void dss_shutdown(struct platform_device *pdev) DSSDBG("shutdown\n"); for_each_dss_output(dssdev) { - if (dssdev->state == OMAP_DSS_DISPLAY_ACTIVE) + if (dssdev->state == OMAP_DSS_DISPLAY_ACTIVE && + dssdev->ops && dssdev->ops->disable) dssdev->ops->disable(dssdev); } } diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi.h b/drivers/gpu/drm/omapdrm/dss/hdmi.h index c867552c925c..3a40833d3368 100644 --- a/drivers/gpu/drm/omapdrm/dss/hdmi.h +++ b/drivers/gpu/drm/omapdrm/dss/hdmi.h @@ -14,6 +14,7 @@ #include <linux/hdmi.h> #include <sound/omap-hdmi-audio.h> #include <media/cec.h> +#include <drm/drm_bridge.h> #include "omapdss.h" #include "dss.h" @@ -364,6 +365,7 @@ struct omap_hdmi { bool core_enabled; struct omap_dss_device output; + struct drm_bridge bridge; struct platform_device *audio_pdev; void (*audio_abort_cb)(struct device *dev); @@ -378,6 +380,6 @@ struct omap_hdmi { bool display_enabled; }; -#define dssdev_to_hdmi(dssdev) container_of(dssdev, struct omap_hdmi, output) +#define drm_bridge_to_hdmi(b) container_of(b, struct omap_hdmi, bridge) #endif diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi4.c b/drivers/gpu/drm/omapdrm/dss/hdmi4.c index 0f557fad4513..2578c95570f6 100644 --- a/drivers/gpu/drm/omapdrm/dss/hdmi4.c +++ b/drivers/gpu/drm/omapdrm/dss/hdmi4.c @@ -28,6 +28,9 @@ #include <sound/omap-hdmi-audio.h> #include <media/cec.h> +#include <drm/drm_atomic.h> +#include <drm/drm_atomic_state_helper.h> + #include "omapdss.h" #include "hdmi4_core.h" #include "hdmi4_cec.h" @@ -237,20 +240,6 @@ static void hdmi_power_off_full(struct omap_hdmi *hdmi) hdmi_power_off_core(hdmi); } -static void hdmi_display_set_timings(struct omap_dss_device *dssdev, - const struct drm_display_mode *mode) -{ - struct omap_hdmi *hdmi = dssdev_to_hdmi(dssdev); - - mutex_lock(&hdmi->lock); - - drm_display_mode_to_videomode(mode, &hdmi->cfg.vm); - - dispc_set_tv_pclk(hdmi->dss->dispc, mode->clock * 1000); - - mutex_unlock(&hdmi->lock); -} - static int hdmi_dump_regs(struct seq_file *s, void *p) { struct omap_hdmi *hdmi = s->private; @@ -272,57 +261,139 @@ static int hdmi_dump_regs(struct seq_file *s, void *p) return 0; } -static int read_edid(struct omap_hdmi *hdmi, u8 *buf, int len) +static void hdmi_start_audio_stream(struct omap_hdmi *hd) { - int r; + hdmi_wp_audio_enable(&hd->wp, true); + hdmi4_audio_start(&hd->core, &hd->wp); +} - mutex_lock(&hdmi->lock); +static void hdmi_stop_audio_stream(struct omap_hdmi *hd) +{ + hdmi4_audio_stop(&hd->core, &hd->wp); + hdmi_wp_audio_enable(&hd->wp, false); +} - r = hdmi_runtime_get(hdmi); - BUG_ON(r); +int hdmi4_core_enable(struct hdmi_core_data *core) +{ + struct omap_hdmi *hdmi = container_of(core, struct omap_hdmi, core); + int r = 0; - r = hdmi4_read_edid(&hdmi->core, buf, len); + DSSDBG("ENTER omapdss_hdmi4_core_enable\n"); + + mutex_lock(&hdmi->lock); + + r = hdmi_power_on_core(hdmi); + if (r) { + DSSERR("failed to power on device\n"); + goto err0; + } - hdmi_runtime_put(hdmi); mutex_unlock(&hdmi->lock); + return 0; +err0: + mutex_unlock(&hdmi->lock); return r; } -static void hdmi_start_audio_stream(struct omap_hdmi *hd) +void hdmi4_core_disable(struct hdmi_core_data *core) { - hdmi_wp_audio_enable(&hd->wp, true); - hdmi4_audio_start(&hd->core, &hd->wp); + struct omap_hdmi *hdmi = container_of(core, struct omap_hdmi, core); + + DSSDBG("Enter omapdss_hdmi4_core_disable\n"); + + mutex_lock(&hdmi->lock); + + hdmi_power_off_core(hdmi); + + mutex_unlock(&hdmi->lock); } -static void hdmi_stop_audio_stream(struct omap_hdmi *hd) +/* ----------------------------------------------------------------------------- + * DRM Bridge Operations + */ + +static int hdmi4_bridge_attach(struct drm_bridge *bridge, + enum drm_bridge_attach_flags flags) { - hdmi4_audio_stop(&hd->core, &hd->wp); - hdmi_wp_audio_enable(&hd->wp, false); + struct omap_hdmi *hdmi = drm_bridge_to_hdmi(bridge); + + if (!(flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR)) + return -EINVAL; + + return drm_bridge_attach(bridge->encoder, hdmi->output.next_bridge, + bridge, flags); } -static void hdmi_display_enable(struct omap_dss_device *dssdev) +static void hdmi4_bridge_mode_set(struct drm_bridge *bridge, + const struct drm_display_mode *mode, + const struct drm_display_mode *adjusted_mode) { - struct omap_hdmi *hdmi = dssdev_to_hdmi(dssdev); + struct omap_hdmi *hdmi = drm_bridge_to_hdmi(bridge); + + mutex_lock(&hdmi->lock); + + drm_display_mode_to_videomode(adjusted_mode, &hdmi->cfg.vm); + + dispc_set_tv_pclk(hdmi->dss->dispc, adjusted_mode->clock * 1000); + + mutex_unlock(&hdmi->lock); +} + +static void hdmi4_bridge_enable(struct drm_bridge *bridge, + struct drm_bridge_state *bridge_state) +{ + struct omap_hdmi *hdmi = drm_bridge_to_hdmi(bridge); + struct drm_atomic_state *state = bridge_state->base.state; + struct drm_connector_state *conn_state; + struct drm_connector *connector; + struct drm_crtc_state *crtc_state; unsigned long flags; - int r; + int ret; + + /* + * None of these should fail, as the bridge can't be enabled without a + * valid CRTC to connector path with fully populated new states. + */ + connector = drm_atomic_get_new_connector_for_encoder(state, + bridge->encoder); + if (WARN_ON(!connector)) + return; + conn_state = drm_atomic_get_new_connector_state(state, connector); + if (WARN_ON(!conn_state)) + return; + crtc_state = drm_atomic_get_new_crtc_state(state, conn_state->crtc); + if (WARN_ON(!crtc_state)) + return; + + hdmi->cfg.hdmi_dvi_mode = connector->display_info.is_hdmi + ? HDMI_HDMI : HDMI_DVI; - DSSDBG("ENTER hdmi_display_enable\n"); + if (connector->display_info.is_hdmi) { + const struct drm_display_mode *mode; + struct hdmi_avi_infoframe avi; + + mode = &crtc_state->adjusted_mode; + ret = drm_hdmi_avi_infoframe_from_display_mode(&avi, connector, + mode); + if (ret == 0) + hdmi->cfg.infoframe = avi; + } mutex_lock(&hdmi->lock); - r = hdmi_power_on_full(hdmi); - if (r) { + ret = hdmi_power_on_full(hdmi); + if (ret) { DSSERR("failed to power on device\n"); goto done; } if (hdmi->audio_configured) { - r = hdmi4_audio_config(&hdmi->core, &hdmi->wp, - &hdmi->audio_config, - hdmi->cfg.vm.pixelclock); - if (r) { - DSSERR("Error restoring audio configuration: %d", r); + ret = hdmi4_audio_config(&hdmi->core, &hdmi->wp, + &hdmi->audio_config, + hdmi->cfg.vm.pixelclock); + if (ret) { + DSSERR("Error restoring audio configuration: %d", ret); hdmi->audio_abort_cb(&hdmi->pdev->dev); hdmi->audio_configured = false; } @@ -338,13 +409,12 @@ done: mutex_unlock(&hdmi->lock); } -static void hdmi_display_disable(struct omap_dss_device *dssdev) +static void hdmi4_bridge_disable(struct drm_bridge *bridge, + struct drm_bridge_state *bridge_state) { - struct omap_hdmi *hdmi = dssdev_to_hdmi(dssdev); + struct omap_hdmi *hdmi = drm_bridge_to_hdmi(bridge); unsigned long flags; - DSSDBG("Enter hdmi_display_disable\n"); - mutex_lock(&hdmi->lock); spin_lock_irqsave(&hdmi->audio_playing_lock, flags); @@ -357,58 +427,21 @@ static void hdmi_display_disable(struct omap_dss_device *dssdev) mutex_unlock(&hdmi->lock); } -int hdmi4_core_enable(struct hdmi_core_data *core) +static void hdmi4_bridge_hpd_notify(struct drm_bridge *bridge, + enum drm_connector_status status) { - struct omap_hdmi *hdmi = container_of(core, struct omap_hdmi, core); - int r = 0; - - DSSDBG("ENTER omapdss_hdmi4_core_enable\n"); - - mutex_lock(&hdmi->lock); - - r = hdmi_power_on_core(hdmi); - if (r) { - DSSERR("failed to power on device\n"); - goto err0; - } - - mutex_unlock(&hdmi->lock); - return 0; - -err0: - mutex_unlock(&hdmi->lock); - return r; -} - -void hdmi4_core_disable(struct hdmi_core_data *core) -{ - struct omap_hdmi *hdmi = container_of(core, struct omap_hdmi, core); - - DSSDBG("Enter omapdss_hdmi4_core_disable\n"); - - mutex_lock(&hdmi->lock); - - hdmi_power_off_core(hdmi); - - mutex_unlock(&hdmi->lock); -} - -static int hdmi_connect(struct omap_dss_device *src, - struct omap_dss_device *dst) -{ - return omapdss_device_connect(dst->dss, dst, dst->next); -} + struct omap_hdmi *hdmi = drm_bridge_to_hdmi(bridge); -static void hdmi_disconnect(struct omap_dss_device *src, - struct omap_dss_device *dst) -{ - omapdss_device_disconnect(dst, dst->next); + if (status == connector_status_disconnected) + hdmi4_cec_set_phys_addr(&hdmi->core, CEC_PHYS_ADDR_INVALID); } -static int hdmi_read_edid(struct omap_dss_device *dssdev, - u8 *edid, int len) +static struct edid *hdmi4_bridge_get_edid(struct drm_bridge *bridge, + struct drm_connector *connector) { - struct omap_hdmi *hdmi = dssdev_to_hdmi(dssdev); + struct omap_hdmi *hdmi = drm_bridge_to_hdmi(bridge); + struct edid *edid = NULL; + unsigned int cec_addr; bool need_enable; int r; @@ -417,63 +450,65 @@ static int hdmi_read_edid(struct omap_dss_device *dssdev, if (need_enable) { r = hdmi4_core_enable(&hdmi->core); if (r) - return r; + return NULL; } - r = read_edid(hdmi, edid, len); - if (r >= 256) - hdmi4_cec_set_phys_addr(&hdmi->core, - cec_get_edid_phys_addr(edid, r, NULL)); - else - hdmi4_cec_set_phys_addr(&hdmi->core, CEC_PHYS_ADDR_INVALID); - if (need_enable) - hdmi4_core_disable(&hdmi->core); + mutex_lock(&hdmi->lock); + r = hdmi_runtime_get(hdmi); + BUG_ON(r); - return r; -} + r = hdmi4_core_ddc_init(&hdmi->core); + if (r) + goto done; -static void hdmi_lost_hotplug(struct omap_dss_device *dssdev) -{ - struct omap_hdmi *hdmi = dssdev_to_hdmi(dssdev); + edid = drm_do_get_edid(connector, hdmi4_core_ddc_read, &hdmi->core); - hdmi4_cec_set_phys_addr(&hdmi->core, CEC_PHYS_ADDR_INVALID); -} +done: + hdmi_runtime_put(hdmi); + mutex_unlock(&hdmi->lock); -static int hdmi_set_infoframe(struct omap_dss_device *dssdev, - const struct hdmi_avi_infoframe *avi) -{ - struct omap_hdmi *hdmi = dssdev_to_hdmi(dssdev); + if (edid && edid->extensions) { + unsigned int len = (edid->extensions + 1) * EDID_LENGTH; - hdmi->cfg.infoframe = *avi; - return 0; -} + cec_addr = cec_get_edid_phys_addr((u8 *)edid, len, NULL); + } else { + cec_addr = CEC_PHYS_ADDR_INVALID; + } -static int hdmi_set_hdmi_mode(struct omap_dss_device *dssdev, - bool hdmi_mode) -{ - struct omap_hdmi *hdmi = dssdev_to_hdmi(dssdev); + hdmi4_cec_set_phys_addr(&hdmi->core, cec_addr); - hdmi->cfg.hdmi_dvi_mode = hdmi_mode ? HDMI_HDMI : HDMI_DVI; - return 0; -} + if (need_enable) + hdmi4_core_disable(&hdmi->core); -static const struct omap_dss_device_ops hdmi_ops = { - .connect = hdmi_connect, - .disconnect = hdmi_disconnect, + return edid; +} - .enable = hdmi_display_enable, - .disable = hdmi_display_disable, +static const struct drm_bridge_funcs hdmi4_bridge_funcs = { + .attach = hdmi4_bridge_attach, + .mode_set = hdmi4_bridge_mode_set, + .atomic_duplicate_state = drm_atomic_helper_bridge_duplicate_state, + .atomic_destroy_state = drm_atomic_helper_bridge_destroy_state, + .atomic_reset = drm_atomic_helper_bridge_reset, + .atomic_enable = hdmi4_bridge_enable, + .atomic_disable = hdmi4_bridge_disable, + .hpd_notify = hdmi4_bridge_hpd_notify, + .get_edid = hdmi4_bridge_get_edid, +}; - .set_timings = hdmi_display_set_timings, +static void hdmi4_bridge_init(struct omap_hdmi *hdmi) +{ + hdmi->bridge.funcs = &hdmi4_bridge_funcs; + hdmi->bridge.of_node = hdmi->pdev->dev.of_node; + hdmi->bridge.ops = DRM_BRIDGE_OP_EDID; + hdmi->bridge.type = DRM_MODE_CONNECTOR_HDMIA; - .read_edid = hdmi_read_edid, + drm_bridge_add(&hdmi->bridge); +} - .hdmi = { - .lost_hotplug = hdmi_lost_hotplug, - .set_infoframe = hdmi_set_infoframe, - .set_hdmi_mode = hdmi_set_hdmi_mode, - }, -}; +static void hdmi4_bridge_cleanup(struct omap_hdmi *hdmi) +{ + drm_bridge_remove(&hdmi->bridge); +} /* ----------------------------------------------------------------------------- * Audio Callbacks @@ -666,19 +701,21 @@ static int hdmi4_init_output(struct omap_hdmi *hdmi) struct omap_dss_device *out = &hdmi->output; int r; + hdmi4_bridge_init(hdmi); + out->dev = &hdmi->pdev->dev; out->id = OMAP_DSS_OUTPUT_HDMI; out->type = OMAP_DISPLAY_TYPE_HDMI; out->name = "hdmi.0"; out->dispc_channel = OMAP_DSS_CHANNEL_DIGIT; - out->ops = &hdmi_ops; out->owner = THIS_MODULE; - out->of_ports = BIT(0); - out->ops_flags = OMAP_DSS_DEVICE_OP_EDID; + out->of_port = 0; - r = omapdss_device_init_output(out); - if (r < 0) + r = omapdss_device_init_output(out, &hdmi->bridge); + if (r < 0) { + hdmi4_bridge_cleanup(hdmi); return r; + } omapdss_device_register(out); @@ -691,6 +728,8 @@ static void hdmi4_uninit_output(struct omap_hdmi *hdmi) omapdss_device_unregister(out); omapdss_device_cleanup_output(out); + + hdmi4_bridge_cleanup(hdmi); } static int hdmi4_probe_of(struct omap_hdmi *hdmi) diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi4_core.c b/drivers/gpu/drm/omapdrm/dss/hdmi4_core.c index ea5d5c228534..751985a2679a 100644 --- a/drivers/gpu/drm/omapdrm/dss/hdmi4_core.c +++ b/drivers/gpu/drm/omapdrm/dss/hdmi4_core.c @@ -32,7 +32,7 @@ static inline void __iomem *hdmi_av_base(struct hdmi_core_data *core) return core->base + HDMI_CORE_AV; } -static int hdmi_core_ddc_init(struct hdmi_core_data *core) +int hdmi4_core_ddc_init(struct hdmi_core_data *core) { void __iomem *base = core->base; @@ -74,13 +74,11 @@ static int hdmi_core_ddc_init(struct hdmi_core_data *core) return 0; } -static int hdmi_core_ddc_edid(struct hdmi_core_data *core, - u8 *pedid, int ext) +int hdmi4_core_ddc_read(void *data, u8 *buf, unsigned int block, size_t len) { + struct hdmi_core_data *core = data; void __iomem *base = core->base; u32 i; - char checksum; - u32 offset = 0; /* HDMI_CORE_DDC_STATUS_IN_PROG */ if (hdmi_wait_for_bit_change(base, HDMI_CORE_DDC_STATUS, @@ -89,24 +87,21 @@ static int hdmi_core_ddc_edid(struct hdmi_core_data *core, return -ETIMEDOUT; } - if (ext % 2 != 0) - offset = 0x80; - /* Load Segment Address Register */ - REG_FLD_MOD(base, HDMI_CORE_DDC_SEGM, ext / 2, 7, 0); + REG_FLD_MOD(base, HDMI_CORE_DDC_SEGM, block / 2, 7, 0); /* Load Slave Address Register */ REG_FLD_MOD(base, HDMI_CORE_DDC_ADDR, 0xA0 >> 1, 7, 1); /* Load Offset Address Register */ - REG_FLD_MOD(base, HDMI_CORE_DDC_OFFSET, offset, 7, 0); + REG_FLD_MOD(base, HDMI_CORE_DDC_OFFSET, block % 2 ? 0x80 : 0, 7, 0); /* Load Byte Count */ - REG_FLD_MOD(base, HDMI_CORE_DDC_COUNT1, 0x80, 7, 0); + REG_FLD_MOD(base, HDMI_CORE_DDC_COUNT1, len, 7, 0); REG_FLD_MOD(base, HDMI_CORE_DDC_COUNT2, 0x0, 1, 0); /* Set DDC_CMD */ - if (ext) + if (block) REG_FLD_MOD(base, HDMI_CORE_DDC_CMD, 0x4, 3, 0); else REG_FLD_MOD(base, HDMI_CORE_DDC_CMD, 0x2, 3, 0); @@ -122,7 +117,7 @@ static int hdmi_core_ddc_edid(struct hdmi_core_data *core, return -EIO; } - for (i = 0; i < 0x80; ++i) { + for (i = 0; i < len; ++i) { int t; /* IN_PROG */ @@ -141,48 +136,12 @@ static int hdmi_core_ddc_edid(struct hdmi_core_data *core, udelay(1); } - pedid[i] = REG_GET(base, HDMI_CORE_DDC_DATA, 7, 0); - } - - checksum = 0; - for (i = 0; i < 0x80; ++i) - checksum += pedid[i]; - - if (checksum != 0) { - DSSERR("E-EDID checksum failed!!\n"); - return -EIO; + buf[i] = REG_GET(base, HDMI_CORE_DDC_DATA, 7, 0); } return 0; } -int hdmi4_read_edid(struct hdmi_core_data *core, u8 *edid, int len) -{ - int r, l; - - if (len < 128) - return -EINVAL; - - r = hdmi_core_ddc_init(core); - if (r) - return r; - - r = hdmi_core_ddc_edid(core, edid, 0); - if (r) - return r; - - l = 128; - - if (len >= 128 * 2 && edid[0x7e] > 0) { - r = hdmi_core_ddc_edid(core, edid + 0x80, 1); - if (r) - return r; - l += 128; - } - - return l; -} - static void hdmi_core_init(struct hdmi_core_video_config *video_cfg) { DSSDBG("Enter hdmi_core_init\n"); diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi4_core.h b/drivers/gpu/drm/omapdrm/dss/hdmi4_core.h index 11c4b7ba1eee..dc64ae2aa300 100644 --- a/drivers/gpu/drm/omapdrm/dss/hdmi4_core.h +++ b/drivers/gpu/drm/omapdrm/dss/hdmi4_core.h @@ -249,7 +249,9 @@ struct hdmi_core_packet_enable_repeat { u32 generic_pkt_repeat; }; -int hdmi4_read_edid(struct hdmi_core_data *core, u8 *edid, int len); +int hdmi4_core_ddc_init(struct hdmi_core_data *core); +int hdmi4_core_ddc_read(void *data, u8 *buf, unsigned int block, size_t len); + void hdmi4_configure(struct hdmi_core_data *core, struct hdmi_wp_data *wp, struct hdmi_config *cfg); void hdmi4_core_dump(struct hdmi_core_data *core, struct seq_file *s); diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi5.c b/drivers/gpu/drm/omapdrm/dss/hdmi5.c index d9463b332554..4d4c1fabd0a1 100644 --- a/drivers/gpu/drm/omapdrm/dss/hdmi5.c +++ b/drivers/gpu/drm/omapdrm/dss/hdmi5.c @@ -31,6 +31,9 @@ #include <linux/of_graph.h> #include <sound/omap-hdmi-audio.h> +#include <drm/drm_atomic.h> +#include <drm/drm_atomic_state_helper.h> + #include "omapdss.h" #include "hdmi5_core.h" #include "dss.h" @@ -236,20 +239,6 @@ static void hdmi_power_off_full(struct omap_hdmi *hdmi) hdmi_power_off_core(hdmi); } -static void hdmi_display_set_timings(struct omap_dss_device *dssdev, - const struct drm_display_mode *mode) -{ - struct omap_hdmi *hdmi = dssdev_to_hdmi(dssdev); - - mutex_lock(&hdmi->lock); - - drm_display_mode_to_videomode(mode, &hdmi->cfg.vm); - - dispc_set_tv_pclk(hdmi->dss->dispc, mode->clock * 1000); - - mutex_unlock(&hdmi->lock); -} - static int hdmi_dump_regs(struct seq_file *s, void *p) { struct omap_hdmi *hdmi = s->private; @@ -271,66 +260,138 @@ static int hdmi_dump_regs(struct seq_file *s, void *p) return 0; } -static int read_edid(struct omap_hdmi *hdmi, u8 *buf, int len) +static void hdmi_start_audio_stream(struct omap_hdmi *hd) { - int r; - int idlemode; + REG_FLD_MOD(hd->wp.base, HDMI_WP_SYSCONFIG, 1, 3, 2); + hdmi_wp_audio_enable(&hd->wp, true); + hdmi_wp_audio_core_req_enable(&hd->wp, true); +} - mutex_lock(&hdmi->lock); +static void hdmi_stop_audio_stream(struct omap_hdmi *hd) +{ + hdmi_wp_audio_core_req_enable(&hd->wp, false); + hdmi_wp_audio_enable(&hd->wp, false); + REG_FLD_MOD(hd->wp.base, HDMI_WP_SYSCONFIG, hd->wp_idlemode, 3, 2); +} - r = hdmi_runtime_get(hdmi); - BUG_ON(r); +static int hdmi_core_enable(struct omap_hdmi *hdmi) +{ + int r = 0; - idlemode = REG_GET(hdmi->wp.base, HDMI_WP_SYSCONFIG, 3, 2); - /* No-idle mode */ - REG_FLD_MOD(hdmi->wp.base, HDMI_WP_SYSCONFIG, 1, 3, 2); + DSSDBG("ENTER omapdss_hdmi_core_enable\n"); - r = hdmi5_read_edid(&hdmi->core, buf, len); + mutex_lock(&hdmi->lock); - REG_FLD_MOD(hdmi->wp.base, HDMI_WP_SYSCONFIG, idlemode, 3, 2); + r = hdmi_power_on_core(hdmi); + if (r) { + DSSERR("failed to power on device\n"); + goto err0; + } - hdmi_runtime_put(hdmi); mutex_unlock(&hdmi->lock); + return 0; +err0: + mutex_unlock(&hdmi->lock); return r; } -static void hdmi_start_audio_stream(struct omap_hdmi *hd) +static void hdmi_core_disable(struct omap_hdmi *hdmi) { - REG_FLD_MOD(hd->wp.base, HDMI_WP_SYSCONFIG, 1, 3, 2); - hdmi_wp_audio_enable(&hd->wp, true); - hdmi_wp_audio_core_req_enable(&hd->wp, true); + DSSDBG("Enter omapdss_hdmi_core_disable\n"); + + mutex_lock(&hdmi->lock); + + hdmi_power_off_core(hdmi); + + mutex_unlock(&hdmi->lock); } -static void hdmi_stop_audio_stream(struct omap_hdmi *hd) +/* ----------------------------------------------------------------------------- + * DRM Bridge Operations + */ + +static int hdmi5_bridge_attach(struct drm_bridge *bridge, + enum drm_bridge_attach_flags flags) { - hdmi_wp_audio_core_req_enable(&hd->wp, false); - hdmi_wp_audio_enable(&hd->wp, false); - REG_FLD_MOD(hd->wp.base, HDMI_WP_SYSCONFIG, hd->wp_idlemode, 3, 2); + struct omap_hdmi *hdmi = drm_bridge_to_hdmi(bridge); + + if (!(flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR)) + return -EINVAL; + + return drm_bridge_attach(bridge->encoder, hdmi->output.next_bridge, + bridge, flags); } -static void hdmi_display_enable(struct omap_dss_device *dssdev) +static void hdmi5_bridge_mode_set(struct drm_bridge *bridge, + const struct drm_display_mode *mode, + const struct drm_display_mode *adjusted_mode) { - struct omap_hdmi *hdmi = dssdev_to_hdmi(dssdev); - unsigned long flags; - int r; + struct omap_hdmi *hdmi = drm_bridge_to_hdmi(bridge); + + mutex_lock(&hdmi->lock); - DSSDBG("ENTER hdmi_display_enable\n"); + drm_display_mode_to_videomode(adjusted_mode, &hdmi->cfg.vm); + + dispc_set_tv_pclk(hdmi->dss->dispc, adjusted_mode->clock * 1000); + + mutex_unlock(&hdmi->lock); +} + +static void hdmi5_bridge_enable(struct drm_bridge *bridge, + struct drm_bridge_state *bridge_state) +{ + struct omap_hdmi *hdmi = drm_bridge_to_hdmi(bridge); + struct drm_atomic_state *state = bridge_state->base.state; + struct drm_connector_state *conn_state; + struct drm_connector *connector; + struct drm_crtc_state *crtc_state; + unsigned long flags; + int ret; + + /* + * None of these should fail, as the bridge can't be enabled without a + * valid CRTC to connector path with fully populated new states. + */ + connector = drm_atomic_get_new_connector_for_encoder(state, + bridge->encoder); + if (WARN_ON(!connector)) + return; + conn_state = drm_atomic_get_new_connector_state(state, connector); + if (WARN_ON(!conn_state)) + return; + crtc_state = drm_atomic_get_new_crtc_state(state, conn_state->crtc); + if (WARN_ON(!crtc_state)) + return; + + hdmi->cfg.hdmi_dvi_mode = connector->display_info.is_hdmi + ? HDMI_HDMI : HDMI_DVI; + + if (connector->display_info.is_hdmi) { + const struct drm_display_mode *mode; + struct hdmi_avi_infoframe avi; + + mode = &crtc_state->adjusted_mode; + ret = drm_hdmi_avi_infoframe_from_display_mode(&avi, connector, + mode); + if (ret == 0) + hdmi->cfg.infoframe = avi; + } mutex_lock(&hdmi->lock); - r = hdmi_power_on_full(hdmi); - if (r) { + ret = hdmi_power_on_full(hdmi); + if (ret) { DSSERR("failed to power on device\n"); goto done; } if (hdmi->audio_configured) { - r = hdmi5_audio_config(&hdmi->core, &hdmi->wp, - &hdmi->audio_config, - hdmi->cfg.vm.pixelclock); - if (r) { - DSSERR("Error restoring audio configuration: %d", r); + ret = hdmi5_audio_config(&hdmi->core, &hdmi->wp, + &hdmi->audio_config, + hdmi->cfg.vm.pixelclock); + if (ret) { + DSSERR("Error restoring audio configuration: %d", ret); hdmi->audio_abort_cb(&hdmi->pdev->dev); hdmi->audio_configured = false; } @@ -346,13 +407,12 @@ done: mutex_unlock(&hdmi->lock); } -static void hdmi_display_disable(struct omap_dss_device *dssdev) +static void hdmi5_bridge_disable(struct drm_bridge *bridge, + struct drm_bridge_state *bridge_state) { - struct omap_hdmi *hdmi = dssdev_to_hdmi(dssdev); + struct omap_hdmi *hdmi = drm_bridge_to_hdmi(bridge); unsigned long flags; - DSSDBG("Enter hdmi_display_disable\n"); - mutex_lock(&hdmi->lock); spin_lock_irqsave(&hdmi->audio_playing_lock, flags); @@ -365,109 +425,74 @@ static void hdmi_display_disable(struct omap_dss_device *dssdev) mutex_unlock(&hdmi->lock); } -static int hdmi_core_enable(struct omap_hdmi *hdmi) +static struct edid *hdmi5_bridge_get_edid(struct drm_bridge *bridge, + struct drm_connector *connector) { - int r = 0; - - DSSDBG("ENTER omapdss_hdmi_core_enable\n"); + struct omap_hdmi *hdmi = drm_bridge_to_hdmi(bridge); + struct edid *edid; + bool need_enable; + int idlemode; + int r; - mutex_lock(&hdmi->lock); + need_enable = hdmi->core_enabled == false; - r = hdmi_power_on_core(hdmi); - if (r) { - DSSERR("failed to power on device\n"); - goto err0; + if (need_enable) { + r = hdmi_core_enable(hdmi); + if (r) + return NULL; } - mutex_unlock(&hdmi->lock); - return 0; - -err0: - mutex_unlock(&hdmi->lock); - return r; -} - -static void hdmi_core_disable(struct omap_hdmi *hdmi) -{ - DSSDBG("Enter omapdss_hdmi_core_disable\n"); - mutex_lock(&hdmi->lock); + r = hdmi_runtime_get(hdmi); + BUG_ON(r); - hdmi_power_off_core(hdmi); - - mutex_unlock(&hdmi->lock); -} - -static int hdmi_connect(struct omap_dss_device *src, - struct omap_dss_device *dst) -{ - return omapdss_device_connect(dst->dss, dst, dst->next); -} + idlemode = REG_GET(hdmi->wp.base, HDMI_WP_SYSCONFIG, 3, 2); + /* No-idle mode */ + REG_FLD_MOD(hdmi->wp.base, HDMI_WP_SYSCONFIG, 1, 3, 2); -static void hdmi_disconnect(struct omap_dss_device *src, - struct omap_dss_device *dst) -{ - omapdss_device_disconnect(dst, dst->next); -} + hdmi5_core_ddc_init(&hdmi->core); -static int hdmi_read_edid(struct omap_dss_device *dssdev, - u8 *edid, int len) -{ - struct omap_hdmi *hdmi = dssdev_to_hdmi(dssdev); - bool need_enable; - int r; + edid = drm_do_get_edid(connector, hdmi5_core_ddc_read, &hdmi->core); - need_enable = hdmi->core_enabled == false; + hdmi5_core_ddc_uninit(&hdmi->core); - if (need_enable) { - r = hdmi_core_enable(hdmi); - if (r) - return r; - } + REG_FLD_MOD(hdmi->wp.base, HDMI_WP_SYSCONFIG, idlemode, 3, 2); - r = read_edid(hdmi, edid, len); + hdmi_runtime_put(hdmi); + mutex_unlock(&hdmi->lock); if (need_enable) hdmi_core_disable(hdmi); - return r; + return (struct edid *)edid; } -static int hdmi_set_infoframe(struct omap_dss_device *dssdev, - const struct hdmi_avi_infoframe *avi) +static const struct drm_bridge_funcs hdmi5_bridge_funcs = { + .attach = hdmi5_bridge_attach, + .mode_set = hdmi5_bridge_mode_set, + .atomic_duplicate_state = drm_atomic_helper_bridge_duplicate_state, + .atomic_destroy_state = drm_atomic_helper_bridge_destroy_state, + .atomic_reset = drm_atomic_helper_bridge_reset, + .atomic_enable = hdmi5_bridge_enable, + .atomic_disable = hdmi5_bridge_disable, + .get_edid = hdmi5_bridge_get_edid, +}; + +static void hdmi5_bridge_init(struct omap_hdmi *hdmi) { - struct omap_hdmi *hdmi = dssdev_to_hdmi(dssdev); + hdmi->bridge.funcs = &hdmi5_bridge_funcs; + hdmi->bridge.of_node = hdmi->pdev->dev.of_node; + hdmi->bridge.ops = DRM_BRIDGE_OP_EDID; + hdmi->bridge.type = DRM_MODE_CONNECTOR_HDMIA; - hdmi->cfg.infoframe = *avi; - return 0; + drm_bridge_add(&hdmi->bridge); } -static int hdmi_set_hdmi_mode(struct omap_dss_device *dssdev, - bool hdmi_mode) +static void hdmi5_bridge_cleanup(struct omap_hdmi *hdmi) { - struct omap_hdmi *hdmi = dssdev_to_hdmi(dssdev); - - hdmi->cfg.hdmi_dvi_mode = hdmi_mode ? HDMI_HDMI : HDMI_DVI; - return 0; + drm_bridge_remove(&hdmi->bridge); } -static const struct omap_dss_device_ops hdmi_ops = { - .connect = hdmi_connect, - .disconnect = hdmi_disconnect, - - .enable = hdmi_display_enable, - .disable = hdmi_display_disable, - - .set_timings = hdmi_display_set_timings, - - .read_edid = hdmi_read_edid, - - .hdmi = { - .set_infoframe = hdmi_set_infoframe, - .set_hdmi_mode = hdmi_set_hdmi_mode, - }, -}; - /* ----------------------------------------------------------------------------- * Audio Callbacks */ @@ -650,19 +675,21 @@ static int hdmi5_init_output(struct omap_hdmi *hdmi) struct omap_dss_device *out = &hdmi->output; int r; + hdmi5_bridge_init(hdmi); + out->dev = &hdmi->pdev->dev; out->id = OMAP_DSS_OUTPUT_HDMI; out->type = OMAP_DISPLAY_TYPE_HDMI; out->name = "hdmi.0"; out->dispc_channel = OMAP_DSS_CHANNEL_DIGIT; - out->ops = &hdmi_ops; out->owner = THIS_MODULE; - out->of_ports = BIT(0); - out->ops_flags = OMAP_DSS_DEVICE_OP_EDID; + out->of_port = 0; - r = omapdss_device_init_output(out); - if (r < 0) + r = omapdss_device_init_output(out, &hdmi->bridge); + if (r < 0) { + hdmi5_bridge_cleanup(hdmi); return r; + } omapdss_device_register(out); @@ -675,6 +702,8 @@ static void hdmi5_uninit_output(struct omap_hdmi *hdmi) omapdss_device_unregister(out); omapdss_device_cleanup_output(out); + + hdmi5_bridge_cleanup(hdmi); } static int hdmi5_probe_of(struct omap_hdmi *hdmi) diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi5_core.c b/drivers/gpu/drm/omapdrm/dss/hdmi5_core.c index ff4d35c8771f..7dd587035160 100644 --- a/drivers/gpu/drm/omapdrm/dss/hdmi5_core.c +++ b/drivers/gpu/drm/omapdrm/dss/hdmi5_core.c @@ -23,7 +23,7 @@ #include "hdmi5_core.h" -static void hdmi_core_ddc_init(struct hdmi_core_data *core) +void hdmi5_core_ddc_init(struct hdmi_core_data *core) { void __iomem *base = core->base; const unsigned long long iclk = 266000000; /* DSS L3 ICLK */ @@ -102,7 +102,7 @@ static void hdmi_core_ddc_init(struct hdmi_core_data *core) REG_FLD_MOD(base, HDMI_CORE_I2CM_INT, 0x0, 2, 2); } -static void hdmi_core_ddc_uninit(struct hdmi_core_data *core) +void hdmi5_core_ddc_uninit(struct hdmi_core_data *core) { void __iomem *base = core->base; @@ -112,14 +112,14 @@ static void hdmi_core_ddc_uninit(struct hdmi_core_data *core) REG_FLD_MOD(base, HDMI_CORE_I2CM_INT, 0x1, 2, 2); } -static int hdmi_core_ddc_edid(struct hdmi_core_data *core, u8 *pedid, u8 ext) +int hdmi5_core_ddc_read(void *data, u8 *buf, unsigned int block, size_t len) { + struct hdmi_core_data *core = data; void __iomem *base = core->base; u8 cur_addr; - char checksum = 0; const int retries = 1000; - u8 seg_ptr = ext / 2; - u8 edidbase = ((ext % 2) * 0x80); + u8 seg_ptr = block / 2; + u8 edidbase = ((block % 2) * EDID_LENGTH); REG_FLD_MOD(base, HDMI_CORE_I2CM_SEGPTR, seg_ptr, 7, 0); @@ -127,7 +127,7 @@ static int hdmi_core_ddc_edid(struct hdmi_core_data *core, u8 *pedid, u8 ext) * TODO: We use polling here, although we probably should use proper * interrupts. */ - for (cur_addr = 0; cur_addr < 128; ++cur_addr) { + for (cur_addr = 0; cur_addr < len; ++cur_addr) { int i; /* clear ERROR and DONE */ @@ -164,45 +164,13 @@ static int hdmi_core_ddc_edid(struct hdmi_core_data *core, u8 *pedid, u8 ext) return -EIO; } - pedid[cur_addr] = REG_GET(base, HDMI_CORE_I2CM_DATAI, 7, 0); - checksum += pedid[cur_addr]; + buf[cur_addr] = REG_GET(base, HDMI_CORE_I2CM_DATAI, 7, 0); } return 0; } -int hdmi5_read_edid(struct hdmi_core_data *core, u8 *edid, int len) -{ - int r, n, i; - int max_ext_blocks = (len / 128) - 1; - - if (len < 128) - return -EINVAL; - - hdmi_core_ddc_init(core); - - r = hdmi_core_ddc_edid(core, edid, 0); - if (r) - goto out; - - n = edid[0x7e]; - - if (n > max_ext_blocks) - n = max_ext_blocks; - - for (i = 1; i <= n; i++) { - r = hdmi_core_ddc_edid(core, edid + i * EDID_LENGTH, i); - if (r) - goto out; - } - -out: - hdmi_core_ddc_uninit(core); - - return r ? r : len; -} - void hdmi5_core_dump(struct hdmi_core_data *core, struct seq_file *s) { diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi5_core.h b/drivers/gpu/drm/omapdrm/dss/hdmi5_core.h index f10b8a283011..65eadefdb3f9 100644 --- a/drivers/gpu/drm/omapdrm/dss/hdmi5_core.h +++ b/drivers/gpu/drm/omapdrm/dss/hdmi5_core.h @@ -281,7 +281,10 @@ struct csc_table { u16 c1, c2, c3, c4; }; -int hdmi5_read_edid(struct hdmi_core_data *core, u8 *edid, int len); +void hdmi5_core_ddc_init(struct hdmi_core_data *core); +int hdmi5_core_ddc_read(void *data, u8 *buf, unsigned int block, size_t len); +void hdmi5_core_ddc_uninit(struct hdmi_core_data *core); + void hdmi5_core_dump(struct hdmi_core_data *core, struct seq_file *s); int hdmi5_core_handle_irqs(struct hdmi_core_data *core); void hdmi5_configure(struct hdmi_core_data *core, struct hdmi_wp_data *wp, diff --git a/drivers/gpu/drm/omapdrm/dss/omapdss-boot-init.c b/drivers/gpu/drm/omapdrm/dss/omapdss-boot-init.c index 31502857f013..72a7da7bfff1 100644 --- a/drivers/gpu/drm/omapdrm/dss/omapdss-boot-init.c +++ b/drivers/gpu/drm/omapdrm/dss/omapdss-boot-init.c @@ -174,34 +174,38 @@ static const struct of_device_id omapdss_of_match[] __initconst = { }; static const struct of_device_id omapdss_of_fixups_whitelist[] __initconst = { - { .compatible = "composite-video-connector" }, - { .compatible = "hdmi-connector" }, { .compatible = "panel-dsi-cm" }, - { .compatible = "svideo-connector" }, - { .compatible = "ti,opa362" }, - { .compatible = "ti,tpd12s015" }, {}, }; +static void __init omapdss_find_children(struct device_node *np) +{ + struct device_node *child; + + for_each_available_child_of_node(np, child) { + if (!of_find_property(child, "compatible", NULL)) + continue; + + omapdss_walk_device(child, true); + + if (of_device_is_compatible(child, "ti,sysc")) + omapdss_find_children(child); + } +} + static int __init omapdss_boot_init(void) { - struct device_node *dss, *child; + struct device_node *dss; INIT_LIST_HEAD(&dss_conv_list); dss = of_find_matching_node(NULL, omapdss_of_match); if (dss == NULL || !of_device_is_available(dss)) - return 0; + goto put_node; omapdss_walk_device(dss, true); - - for_each_available_child_of_node(dss, child) { - if (!of_find_property(child, "compatible", NULL)) - continue; - - omapdss_walk_device(child, true); - } + omapdss_find_children(dss); while (!list_empty(&dss_conv_list)) { struct dss_conv_node *n; @@ -217,6 +221,8 @@ static int __init omapdss_boot_init(void) kfree(n); } +put_node: + of_node_put(dss); return 0; } diff --git a/drivers/gpu/drm/omapdrm/dss/omapdss.h b/drivers/gpu/drm/omapdrm/dss/omapdss.h index 79f6b195c7cf..ab19d4af8de7 100644 --- a/drivers/gpu/drm/omapdrm/dss/omapdss.h +++ b/drivers/gpu/drm/omapdrm/dss/omapdss.h @@ -285,13 +285,6 @@ struct omap_dss_writeback_info { u8 pre_mult_alpha; }; -struct omapdss_hdmi_ops { - void (*lost_hotplug)(struct omap_dss_device *dssdev); - int (*set_hdmi_mode)(struct omap_dss_device *dssdev, bool hdmi_mode); - int (*set_infoframe)(struct omap_dss_device *dssdev, - const struct hdmi_avi_infoframe *avi); -}; - struct omapdss_dsi_ops { void (*disable)(struct omap_dss_device *dssdev, bool disconnect_lanes, bool enter_ulps); @@ -349,46 +342,23 @@ struct omap_dss_device_ops { void (*disconnect)(struct omap_dss_device *dssdev, struct omap_dss_device *dst); - void (*pre_enable)(struct omap_dss_device *dssdev); void (*enable)(struct omap_dss_device *dssdev); void (*disable)(struct omap_dss_device *dssdev); - void (*post_disable)(struct omap_dss_device *dssdev); int (*check_timings)(struct omap_dss_device *dssdev, struct drm_display_mode *mode); - void (*set_timings)(struct omap_dss_device *dssdev, - const struct drm_display_mode *mode); - - bool (*detect)(struct omap_dss_device *dssdev); - - void (*register_hpd_cb)(struct omap_dss_device *dssdev, - void (*cb)(void *cb_data, - enum drm_connector_status status), - void *cb_data); - void (*unregister_hpd_cb)(struct omap_dss_device *dssdev); - - int (*read_edid)(struct omap_dss_device *dssdev, u8 *buf, int len); int (*get_modes)(struct omap_dss_device *dssdev, struct drm_connector *connector); - union { - const struct omapdss_hdmi_ops hdmi; - const struct omapdss_dsi_ops dsi; - }; + const struct omapdss_dsi_ops dsi; }; /** * enum omap_dss_device_ops_flag - Indicates which device ops are supported - * @OMAP_DSS_DEVICE_OP_DETECT: The device supports output connection detection - * @OMAP_DSS_DEVICE_OP_HPD: The device supports all hot-plug-related operations - * @OMAP_DSS_DEVICE_OP_EDID: The device supports reading EDID * @OMAP_DSS_DEVICE_OP_MODES: The device supports reading modes */ enum omap_dss_device_ops_flag { - OMAP_DSS_DEVICE_OP_DETECT = BIT(0), - OMAP_DSS_DEVICE_OP_HPD = BIT(1), - OMAP_DSS_DEVICE_OP_EDID = BIT(2), OMAP_DSS_DEVICE_OP_MODES = BIT(3), }; @@ -400,6 +370,7 @@ struct omap_dss_device { struct dss_device *dss; struct omap_dss_device *next; struct drm_bridge *bridge; + struct drm_bridge *next_bridge; struct drm_panel *panel; struct list_head list; @@ -436,8 +407,8 @@ struct omap_dss_device { /* output instance */ enum omap_dss_output_id id; - /* bitmask of port numbers in DT */ - unsigned int of_ports; + /* port number in DT */ + unsigned int of_port; }; struct omap_dss_driver { @@ -461,7 +432,6 @@ static inline bool omapdss_is_initialized(void) } void omapdss_display_init(struct omap_dss_device *dssdev); -struct omap_dss_device *omapdss_display_get(struct omap_dss_device *output); int omapdss_display_get_modes(struct drm_connector *connector, const struct videomode *vm); @@ -475,10 +445,8 @@ int omapdss_device_connect(struct dss_device *dss, struct omap_dss_device *dst); void omapdss_device_disconnect(struct omap_dss_device *src, struct omap_dss_device *dst); -void omapdss_device_pre_enable(struct omap_dss_device *dssdev); void omapdss_device_enable(struct omap_dss_device *dssdev); void omapdss_device_disable(struct omap_dss_device *dssdev); -void omapdss_device_post_disable(struct omap_dss_device *dssdev); int omap_dss_get_num_overlay_managers(void); @@ -487,7 +455,8 @@ int omap_dss_get_num_overlays(void); #define for_each_dss_output(d) \ while ((d = omapdss_device_next_output(d)) != NULL) struct omap_dss_device *omapdss_device_next_output(struct omap_dss_device *from); -int omapdss_device_init_output(struct omap_dss_device *out); +int omapdss_device_init_output(struct omap_dss_device *out, + struct drm_bridge *local_bridge); void omapdss_device_cleanup_output(struct omap_dss_device *out); typedef void (*omap_dispc_isr_t) (void *arg, u32 mask); @@ -502,9 +471,6 @@ static inline bool omapdss_device_is_enabled(struct omap_dss_device *dssdev) return dssdev->state == OMAP_DSS_DISPLAY_ACTIVE; } -struct omap_dss_device * -omapdss_of_find_connected_device(struct device_node *node, unsigned int port); - enum dss_writeback_channel { DSS_WB_LCD1_MGR = 0, DSS_WB_LCD2_MGR = 1, diff --git a/drivers/gpu/drm/omapdrm/dss/output.c b/drivers/gpu/drm/omapdrm/dss/output.c index 0693d34fca1b..ce21c798cca6 100644 --- a/drivers/gpu/drm/omapdrm/dss/output.c +++ b/drivers/gpu/drm/omapdrm/dss/output.c @@ -4,7 +4,6 @@ * Author: Archit Taneja <archit@ti.com> */ -#include <linux/bitops.h> #include <linux/kernel.h> #include <linux/module.h> #include <linux/platform_device.h> @@ -18,12 +17,14 @@ #include "dss.h" #include "omapdss.h" -int omapdss_device_init_output(struct omap_dss_device *out) +int omapdss_device_init_output(struct omap_dss_device *out, + struct drm_bridge *local_bridge) { struct device_node *remote_node; + int ret; remote_node = of_graph_get_remote_node(out->dev->of_node, - ffs(out->of_ports) - 1, 0); + out->of_port, 0); if (!remote_node) { dev_dbg(out->dev, "failed to find video sink\n"); return 0; @@ -39,17 +40,55 @@ int omapdss_device_init_output(struct omap_dss_device *out) if (out->next && out->type != out->next->type) { dev_err(out->dev, "output type and display type don't match\n"); - omapdss_device_put(out->next); - out->next = NULL; - return -EINVAL; + ret = -EINVAL; + goto error; } - return out->next || out->bridge || out->panel ? 0 : -EPROBE_DEFER; + if (out->panel) { + struct drm_bridge *bridge; + + bridge = drm_panel_bridge_add(out->panel); + if (IS_ERR(bridge)) { + dev_err(out->dev, + "unable to create panel bridge (%ld)\n", + PTR_ERR(bridge)); + ret = PTR_ERR(bridge); + goto error; + } + + out->bridge = bridge; + } + + if (local_bridge) { + if (!out->bridge) { + ret = -EPROBE_DEFER; + goto error; + } + + out->next_bridge = out->bridge; + out->bridge = local_bridge; + } + + if (!out->next && !out->bridge) { + ret = -EPROBE_DEFER; + goto error; + } + + return 0; + +error: + omapdss_device_cleanup_output(out); + out->next = NULL; + return ret; } EXPORT_SYMBOL(omapdss_device_init_output); void omapdss_device_cleanup_output(struct omap_dss_device *out) { + if (out->bridge && out->panel) + drm_panel_bridge_remove(out->next_bridge ? + out->next_bridge : out->bridge); + if (out->next) omapdss_device_put(out->next); } diff --git a/drivers/gpu/drm/omapdrm/dss/sdi.c b/drivers/gpu/drm/omapdrm/dss/sdi.c index 3b447c01fa2a..417a8740ad0a 100644 --- a/drivers/gpu/drm/omapdrm/dss/sdi.c +++ b/drivers/gpu/drm/omapdrm/dss/sdi.c @@ -6,17 +6,19 @@ #define DSS_SUBSYS_NAME "SDI" -#include <linux/kernel.h> #include <linux/delay.h> #include <linux/err.h> -#include <linux/regulator/consumer.h> #include <linux/export.h> +#include <linux/kernel.h> +#include <linux/of.h> #include <linux/platform_device.h> +#include <linux/regulator/consumer.h> #include <linux/string.h> -#include <linux/of.h> -#include "omapdss.h" +#include <drm/drm_bridge.h> + #include "dss.h" +#include "omapdss.h" struct sdi_device { struct platform_device *pdev; @@ -30,9 +32,11 @@ struct sdi_device { int datapairs; struct omap_dss_device output; + struct drm_bridge bridge; }; -#define dssdev_to_sdi(dssdev) container_of(dssdev, struct sdi_device, output) +#define drm_bridge_to_sdi(bridge) \ + container_of(bridge, struct sdi_device, bridge) struct sdi_clk_calc_ctx { struct sdi_device *sdi; @@ -118,9 +122,82 @@ static void sdi_config_lcd_manager(struct sdi_device *sdi) dss_mgr_set_lcd_config(&sdi->output, &sdi->mgr_config); } -static void sdi_display_enable(struct omap_dss_device *dssdev) +/* ----------------------------------------------------------------------------- + * DRM Bridge Operations + */ + +static int sdi_bridge_attach(struct drm_bridge *bridge, + enum drm_bridge_attach_flags flags) +{ + struct sdi_device *sdi = drm_bridge_to_sdi(bridge); + + if (!(flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR)) + return -EINVAL; + + return drm_bridge_attach(bridge->encoder, sdi->output.next_bridge, + bridge, flags); +} + +static enum drm_mode_status +sdi_bridge_mode_valid(struct drm_bridge *bridge, + const struct drm_display_mode *mode) +{ + struct sdi_device *sdi = drm_bridge_to_sdi(bridge); + unsigned long pixelclock = mode->clock * 1000; + struct dispc_clock_info dispc_cinfo; + unsigned long fck; + int ret; + + if (pixelclock == 0) + return MODE_NOCLOCK; + + ret = sdi_calc_clock_div(sdi, pixelclock, &fck, &dispc_cinfo); + if (ret < 0) + return MODE_CLOCK_RANGE; + + return MODE_OK; +} + +static bool sdi_bridge_mode_fixup(struct drm_bridge *bridge, + const struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode) +{ + struct sdi_device *sdi = drm_bridge_to_sdi(bridge); + unsigned long pixelclock = mode->clock * 1000; + struct dispc_clock_info dispc_cinfo; + unsigned long fck; + unsigned long pck; + int ret; + + ret = sdi_calc_clock_div(sdi, pixelclock, &fck, &dispc_cinfo); + if (ret < 0) + return false; + + pck = fck / dispc_cinfo.lck_div / dispc_cinfo.pck_div; + + if (pck != pixelclock) + dev_dbg(&sdi->pdev->dev, + "pixel clock adjusted from %lu Hz to %lu Hz\n", + pixelclock, pck); + + adjusted_mode->clock = pck / 1000; + + return true; +} + +static void sdi_bridge_mode_set(struct drm_bridge *bridge, + const struct drm_display_mode *mode, + const struct drm_display_mode *adjusted_mode) +{ + struct sdi_device *sdi = drm_bridge_to_sdi(bridge); + + sdi->pixelclock = adjusted_mode->clock * 1000; +} + +static void sdi_bridge_enable(struct drm_bridge *bridge, + struct drm_bridge_state *bridge_state) { - struct sdi_device *sdi = dssdev_to_sdi(dssdev); + struct sdi_device *sdi = drm_bridge_to_sdi(bridge); struct dispc_clock_info dispc_cinfo; unsigned long fck; int r; @@ -181,9 +258,10 @@ err_get_dispc: regulator_disable(sdi->vdds_sdi_reg); } -static void sdi_display_disable(struct omap_dss_device *dssdev) +static void sdi_bridge_disable(struct drm_bridge *bridge, + struct drm_bridge_state *bridge_state) { - struct sdi_device *sdi = dssdev_to_sdi(dssdev); + struct sdi_device *sdi = drm_bridge_to_sdi(bridge); dss_mgr_disable(&sdi->output); @@ -194,86 +272,56 @@ static void sdi_display_disable(struct omap_dss_device *dssdev) regulator_disable(sdi->vdds_sdi_reg); } -static void sdi_set_timings(struct omap_dss_device *dssdev, - const struct drm_display_mode *mode) -{ - struct sdi_device *sdi = dssdev_to_sdi(dssdev); - - sdi->pixelclock = mode->clock * 1000; -} +static const struct drm_bridge_funcs sdi_bridge_funcs = { + .attach = sdi_bridge_attach, + .mode_valid = sdi_bridge_mode_valid, + .mode_fixup = sdi_bridge_mode_fixup, + .mode_set = sdi_bridge_mode_set, + .atomic_enable = sdi_bridge_enable, + .atomic_disable = sdi_bridge_disable, +}; -static int sdi_check_timings(struct omap_dss_device *dssdev, - struct drm_display_mode *mode) +static void sdi_bridge_init(struct sdi_device *sdi) { - struct sdi_device *sdi = dssdev_to_sdi(dssdev); - struct dispc_clock_info dispc_cinfo; - unsigned long pixelclock = mode->clock * 1000; - unsigned long fck; - unsigned long pck; - int r; + sdi->bridge.funcs = &sdi_bridge_funcs; + sdi->bridge.of_node = sdi->pdev->dev.of_node; + sdi->bridge.type = DRM_MODE_CONNECTOR_LVDS; - if (pixelclock == 0) - return -EINVAL; - - r = sdi_calc_clock_div(sdi, pixelclock, &fck, &dispc_cinfo); - if (r) - return r; - - pck = fck / dispc_cinfo.lck_div / dispc_cinfo.pck_div; - - if (pck != pixelclock) { - DSSWARN("Pixel clock adjusted from %lu Hz to %lu Hz\n", - pixelclock, pck); - - mode->clock = pck / 1000; - } - - return 0; + drm_bridge_add(&sdi->bridge); } -static int sdi_connect(struct omap_dss_device *src, - struct omap_dss_device *dst) +static void sdi_bridge_cleanup(struct sdi_device *sdi) { - return omapdss_device_connect(dst->dss, dst, dst->next); + drm_bridge_remove(&sdi->bridge); } -static void sdi_disconnect(struct omap_dss_device *src, - struct omap_dss_device *dst) -{ - omapdss_device_disconnect(dst, dst->next); -} - -static const struct omap_dss_device_ops sdi_ops = { - .connect = sdi_connect, - .disconnect = sdi_disconnect, - - .enable = sdi_display_enable, - .disable = sdi_display_disable, - - .check_timings = sdi_check_timings, - .set_timings = sdi_set_timings, -}; +/* ----------------------------------------------------------------------------- + * Initialisation and Cleanup + */ static int sdi_init_output(struct sdi_device *sdi) { struct omap_dss_device *out = &sdi->output; int r; + sdi_bridge_init(sdi); + out->dev = &sdi->pdev->dev; out->id = OMAP_DSS_OUTPUT_SDI; out->type = OMAP_DISPLAY_TYPE_SDI; out->name = "sdi.0"; out->dispc_channel = OMAP_DSS_CHANNEL_LCD; /* We have SDI only on OMAP3, where it's on port 1 */ - out->of_ports = BIT(1); - out->ops = &sdi_ops; + out->of_port = 1; out->owner = THIS_MODULE; out->bus_flags = DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE /* 15.5.9.1.2 */ | DRM_BUS_FLAG_SYNC_DRIVE_POSEDGE; - r = omapdss_device_init_output(out); - if (r < 0) + r = omapdss_device_init_output(out, &sdi->bridge); + if (r < 0) { + sdi_bridge_cleanup(sdi); return r; + } omapdss_device_register(out); @@ -284,6 +332,8 @@ static void sdi_uninit_output(struct sdi_device *sdi) { omapdss_device_unregister(&sdi->output); omapdss_device_cleanup_output(&sdi->output); + + sdi_bridge_cleanup(sdi); } int sdi_init_port(struct dss_device *dss, struct platform_device *pdev, diff --git a/drivers/gpu/drm/omapdrm/dss/venc.c b/drivers/gpu/drm/omapdrm/dss/venc.c index 596a297d5813..766553bb2f87 100644 --- a/drivers/gpu/drm/omapdrm/dss/venc.c +++ b/drivers/gpu/drm/omapdrm/dss/venc.c @@ -13,7 +13,6 @@ #include <linux/clk.h> #include <linux/err.h> #include <linux/io.h> -#include <linux/mutex.h> #include <linux/completion.h> #include <linux/delay.h> #include <linux/string.h> @@ -26,6 +25,8 @@ #include <linux/component.h> #include <linux/sys_soc.h> +#include <drm/drm_bridge.h> + #include "omapdss.h" #include "dss.h" @@ -289,7 +290,6 @@ static const struct drm_display_mode omap_dss_ntsc_mode = { struct venc_device { struct platform_device *pdev; void __iomem *base; - struct mutex venc_lock; struct regulator *vdda_dac_reg; struct dss_device *dss; @@ -303,9 +303,10 @@ struct venc_device { bool requires_tv_dac_clk; struct omap_dss_device output; + struct drm_bridge bridge; }; -#define dssdev_to_venc(dssdev) container_of(dssdev, struct venc_device, output) +#define drm_bridge_to_venc(b) container_of(b, struct venc_device, bridge) static inline void venc_write_reg(struct venc_device *venc, int idx, u32 val) { @@ -477,56 +478,6 @@ static void venc_power_off(struct venc_device *venc) venc_runtime_put(venc); } -static void venc_display_enable(struct omap_dss_device *dssdev) -{ - struct venc_device *venc = dssdev_to_venc(dssdev); - - DSSDBG("venc_display_enable\n"); - - mutex_lock(&venc->venc_lock); - - venc_power_on(venc); - - mutex_unlock(&venc->venc_lock); -} - -static void venc_display_disable(struct omap_dss_device *dssdev) -{ - struct venc_device *venc = dssdev_to_venc(dssdev); - - DSSDBG("venc_display_disable\n"); - - mutex_lock(&venc->venc_lock); - - venc_power_off(venc); - - mutex_unlock(&venc->venc_lock); -} - -static int venc_get_modes(struct omap_dss_device *dssdev, - struct drm_connector *connector) -{ - static const struct drm_display_mode *modes[] = { - &omap_dss_pal_mode, - &omap_dss_ntsc_mode, - }; - unsigned int i; - - for (i = 0; i < ARRAY_SIZE(modes); ++i) { - struct drm_display_mode *mode; - - mode = drm_mode_duplicate(connector->dev, modes[i]); - if (!mode) - return i; - - mode->type = DRM_MODE_TYPE_DRIVER | DRM_MODE_TYPE_PREFERRED; - drm_mode_set_name(mode); - drm_mode_probed_add(connector, mode); - } - - return ARRAY_SIZE(modes); -} - static enum venc_videomode venc_get_videomode(const struct drm_display_mode *mode) { if (!(mode->flags & DRM_MODE_FLAG_INTERLACE)) @@ -545,57 +496,6 @@ static enum venc_videomode venc_get_videomode(const struct drm_display_mode *mod return VENC_MODE_UNKNOWN; } -static void venc_set_timings(struct omap_dss_device *dssdev, - const struct drm_display_mode *mode) -{ - struct venc_device *venc = dssdev_to_venc(dssdev); - enum venc_videomode venc_mode = venc_get_videomode(mode); - - DSSDBG("venc_set_timings\n"); - - mutex_lock(&venc->venc_lock); - - switch (venc_mode) { - default: - WARN_ON_ONCE(1); - /* Fall-through */ - case VENC_MODE_PAL: - venc->config = &venc_config_pal_trm; - break; - - case VENC_MODE_NTSC: - venc->config = &venc_config_ntsc_trm; - break; - } - - dispc_set_tv_pclk(venc->dss->dispc, 13500000); - - mutex_unlock(&venc->venc_lock); -} - -static int venc_check_timings(struct omap_dss_device *dssdev, - struct drm_display_mode *mode) -{ - DSSDBG("venc_check_timings\n"); - - switch (venc_get_videomode(mode)) { - case VENC_MODE_PAL: - drm_mode_copy(mode, &omap_dss_pal_mode); - break; - - case VENC_MODE_NTSC: - drm_mode_copy(mode, &omap_dss_ntsc_mode); - break; - - default: - return -EINVAL; - } - - drm_mode_set_crtcinfo(mode, CRTC_INTERLACE_HALVE_V); - drm_mode_set_name(mode); - return 0; -} - static int venc_dump_regs(struct seq_file *s, void *p) { struct venc_device *venc = s->private; @@ -673,31 +573,149 @@ static int venc_get_clocks(struct venc_device *venc) return 0; } -static int venc_connect(struct omap_dss_device *src, - struct omap_dss_device *dst) +/* ----------------------------------------------------------------------------- + * DRM Bridge Operations + */ + +static int venc_bridge_attach(struct drm_bridge *bridge, + enum drm_bridge_attach_flags flags) +{ + struct venc_device *venc = drm_bridge_to_venc(bridge); + + if (!(flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR)) + return -EINVAL; + + return drm_bridge_attach(bridge->encoder, venc->output.next_bridge, + bridge, flags); +} + +static enum drm_mode_status +venc_bridge_mode_valid(struct drm_bridge *bridge, + const struct drm_display_mode *mode) +{ + switch (venc_get_videomode(mode)) { + case VENC_MODE_PAL: + case VENC_MODE_NTSC: + return MODE_OK; + + default: + return MODE_BAD; + } +} + +static bool venc_bridge_mode_fixup(struct drm_bridge *bridge, + const struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode) +{ + const struct drm_display_mode *venc_mode; + + switch (venc_get_videomode(adjusted_mode)) { + case VENC_MODE_PAL: + venc_mode = &omap_dss_pal_mode; + break; + + case VENC_MODE_NTSC: + venc_mode = &omap_dss_ntsc_mode; + break; + + default: + return false; + } + + drm_mode_copy(adjusted_mode, venc_mode); + drm_mode_set_crtcinfo(adjusted_mode, CRTC_INTERLACE_HALVE_V); + drm_mode_set_name(adjusted_mode); + + return true; +} + +static void venc_bridge_mode_set(struct drm_bridge *bridge, + const struct drm_display_mode *mode, + const struct drm_display_mode *adjusted_mode) +{ + struct venc_device *venc = drm_bridge_to_venc(bridge); + enum venc_videomode venc_mode = venc_get_videomode(adjusted_mode); + + switch (venc_mode) { + default: + WARN_ON_ONCE(1); + /* Fall-through */ + case VENC_MODE_PAL: + venc->config = &venc_config_pal_trm; + break; + + case VENC_MODE_NTSC: + venc->config = &venc_config_ntsc_trm; + break; + } + + dispc_set_tv_pclk(venc->dss->dispc, 13500000); +} + +static void venc_bridge_enable(struct drm_bridge *bridge) { - return omapdss_device_connect(dst->dss, dst, dst->next); + struct venc_device *venc = drm_bridge_to_venc(bridge); + + venc_power_on(venc); } -static void venc_disconnect(struct omap_dss_device *src, - struct omap_dss_device *dst) +static void venc_bridge_disable(struct drm_bridge *bridge) { - omapdss_device_disconnect(dst, dst->next); + struct venc_device *venc = drm_bridge_to_venc(bridge); + + venc_power_off(venc); } -static const struct omap_dss_device_ops venc_ops = { - .connect = venc_connect, - .disconnect = venc_disconnect, +static int venc_bridge_get_modes(struct drm_bridge *bridge, + struct drm_connector *connector) +{ + static const struct drm_display_mode *modes[] = { + &omap_dss_pal_mode, + &omap_dss_ntsc_mode, + }; + unsigned int i; + + for (i = 0; i < ARRAY_SIZE(modes); ++i) { + struct drm_display_mode *mode; + + mode = drm_mode_duplicate(connector->dev, modes[i]); + if (!mode) + return i; - .enable = venc_display_enable, - .disable = venc_display_disable, + mode->type = DRM_MODE_TYPE_DRIVER | DRM_MODE_TYPE_PREFERRED; + drm_mode_set_name(mode); + drm_mode_probed_add(connector, mode); + } - .check_timings = venc_check_timings, - .set_timings = venc_set_timings, + return ARRAY_SIZE(modes); +} - .get_modes = venc_get_modes, +static const struct drm_bridge_funcs venc_bridge_funcs = { + .attach = venc_bridge_attach, + .mode_valid = venc_bridge_mode_valid, + .mode_fixup = venc_bridge_mode_fixup, + .mode_set = venc_bridge_mode_set, + .enable = venc_bridge_enable, + .disable = venc_bridge_disable, + .get_modes = venc_bridge_get_modes, }; +static void venc_bridge_init(struct venc_device *venc) +{ + venc->bridge.funcs = &venc_bridge_funcs; + venc->bridge.of_node = venc->pdev->dev.of_node; + venc->bridge.ops = DRM_BRIDGE_OP_MODES; + venc->bridge.type = DRM_MODE_CONNECTOR_SVIDEO; + venc->bridge.interlace_allowed = true; + + drm_bridge_add(&venc->bridge); +} + +static void venc_bridge_cleanup(struct venc_device *venc) +{ + drm_bridge_remove(&venc->bridge); +} + /* ----------------------------------------------------------------------------- * Component Bind & Unbind */ @@ -747,19 +765,22 @@ static int venc_init_output(struct venc_device *venc) struct omap_dss_device *out = &venc->output; int r; + venc_bridge_init(venc); + out->dev = &venc->pdev->dev; out->id = OMAP_DSS_OUTPUT_VENC; out->type = OMAP_DISPLAY_TYPE_VENC; out->name = "venc.0"; out->dispc_channel = OMAP_DSS_CHANNEL_DIGIT; - out->ops = &venc_ops; out->owner = THIS_MODULE; - out->of_ports = BIT(0); + out->of_port = 0; out->ops_flags = OMAP_DSS_DEVICE_OP_MODES; - r = omapdss_device_init_output(out); - if (r < 0) + r = omapdss_device_init_output(out, &venc->bridge); + if (r < 0) { + venc_bridge_cleanup(venc); return r; + } omapdss_device_register(out); @@ -770,6 +791,8 @@ static void venc_uninit_output(struct venc_device *venc) { omapdss_device_unregister(&venc->output); omapdss_device_cleanup_output(&venc->output); + + venc_bridge_cleanup(venc); } static int venc_probe_of(struct venc_device *venc) @@ -839,8 +862,6 @@ static int venc_probe(struct platform_device *pdev) if (soc_device_match(venc_soc_devices)) venc->requires_tv_dac_clk = true; - mutex_init(&venc->venc_lock); - venc->config = &venc_config_pal_trm; venc_mem = platform_get_resource(venc->pdev, IORESOURCE_MEM, 0); diff --git a/drivers/gpu/drm/omapdrm/omap_connector.c b/drivers/gpu/drm/omapdrm/omap_connector.c index 94cded387174..528764566b17 100644 --- a/drivers/gpu/drm/omapdrm/omap_connector.c +++ b/drivers/gpu/drm/omapdrm/omap_connector.c @@ -6,7 +6,6 @@ #include <drm/drm_atomic_helper.h> #include <drm/drm_crtc.h> -#include <drm/drm_panel.h> #include <drm/drm_probe_helper.h> #include "omap_drv.h" @@ -20,124 +19,12 @@ struct omap_connector { struct drm_connector base; struct omap_dss_device *output; - struct omap_dss_device *hpd; - bool hdmi_mode; }; -static void omap_connector_hpd_notify(struct drm_connector *connector, - enum drm_connector_status status) -{ - struct omap_connector *omap_connector = to_omap_connector(connector); - struct omap_dss_device *dssdev; - - if (status != connector_status_disconnected) - return; - - /* - * Notify all devics in the pipeline of disconnection. This is required - * to let the HDMI encoders reset their internal state related to - * connection status, such as the CEC address. - */ - for (dssdev = omap_connector->output; dssdev; dssdev = dssdev->next) { - if (dssdev->ops && dssdev->ops->hdmi.lost_hotplug) - dssdev->ops->hdmi.lost_hotplug(dssdev); - } -} - -static void omap_connector_hpd_cb(void *cb_data, - enum drm_connector_status status) -{ - struct omap_connector *omap_connector = cb_data; - struct drm_connector *connector = &omap_connector->base; - struct drm_device *dev = connector->dev; - enum drm_connector_status old_status; - - mutex_lock(&dev->mode_config.mutex); - old_status = connector->status; - connector->status = status; - mutex_unlock(&dev->mode_config.mutex); - - if (old_status == status) - return; - - omap_connector_hpd_notify(connector, status); - - drm_kms_helper_hotplug_event(dev); -} - -void omap_connector_enable_hpd(struct drm_connector *connector) -{ - struct omap_connector *omap_connector = to_omap_connector(connector); - struct omap_dss_device *hpd = omap_connector->hpd; - - if (hpd) - hpd->ops->register_hpd_cb(hpd, omap_connector_hpd_cb, - omap_connector); -} - -void omap_connector_disable_hpd(struct drm_connector *connector) -{ - struct omap_connector *omap_connector = to_omap_connector(connector); - struct omap_dss_device *hpd = omap_connector->hpd; - - if (hpd) - hpd->ops->unregister_hpd_cb(hpd); -} - -bool omap_connector_get_hdmi_mode(struct drm_connector *connector) -{ - struct omap_connector *omap_connector = to_omap_connector(connector); - - return omap_connector->hdmi_mode; -} - -static struct omap_dss_device * -omap_connector_find_device(struct drm_connector *connector, - enum omap_dss_device_ops_flag op) -{ - struct omap_connector *omap_connector = to_omap_connector(connector); - struct omap_dss_device *dssdev = NULL; - struct omap_dss_device *d; - - for (d = omap_connector->output; d; d = d->next) { - if (d->ops_flags & op) - dssdev = d; - } - - return dssdev; -} - static enum drm_connector_status omap_connector_detect( struct drm_connector *connector, bool force) { - struct omap_dss_device *dssdev; - enum drm_connector_status status; - - dssdev = omap_connector_find_device(connector, - OMAP_DSS_DEVICE_OP_DETECT); - - if (dssdev) { - status = dssdev->ops->detect(dssdev) - ? connector_status_connected - : connector_status_disconnected; - - omap_connector_hpd_notify(connector, status); - } else { - switch (connector->connector_type) { - case DRM_MODE_CONNECTOR_DPI: - case DRM_MODE_CONNECTOR_LVDS: - case DRM_MODE_CONNECTOR_DSI: - status = connector_status_connected; - break; - default: - status = connector_status_unknown; - break; - } - } - - VERB("%s: %d (force=%d)", connector->name, status, force); - - return status; + return connector_status_connected; } static void omap_connector_destroy(struct drm_connector *connector) @@ -146,14 +33,6 @@ static void omap_connector_destroy(struct drm_connector *connector) DBG("%s", connector->name); - if (omap_connector->hpd) { - struct omap_dss_device *hpd = omap_connector->hpd; - - hpd->ops->unregister_hpd_cb(hpd); - omapdss_device_put(hpd); - omap_connector->hpd = NULL; - } - drm_connector_unregister(connector); drm_connector_cleanup(connector); @@ -162,81 +41,27 @@ static void omap_connector_destroy(struct drm_connector *connector) kfree(omap_connector); } -#define MAX_EDID 512 - -static int omap_connector_get_modes_edid(struct drm_connector *connector, - struct omap_dss_device *dssdev) -{ - struct omap_connector *omap_connector = to_omap_connector(connector); - enum drm_connector_status status; - void *edid; - int n; - - status = omap_connector_detect(connector, false); - if (status != connector_status_connected) - goto no_edid; - - edid = kzalloc(MAX_EDID, GFP_KERNEL); - if (!edid) - goto no_edid; - - if (dssdev->ops->read_edid(dssdev, edid, MAX_EDID) <= 0 || - !drm_edid_is_valid(edid)) { - kfree(edid); - goto no_edid; - } - - drm_connector_update_edid_property(connector, edid); - n = drm_add_edid_modes(connector, edid); - - omap_connector->hdmi_mode = drm_detect_hdmi_monitor(edid); - - kfree(edid); - return n; - -no_edid: - drm_connector_update_edid_property(connector, NULL); - return 0; -} - static int omap_connector_get_modes(struct drm_connector *connector) { struct omap_connector *omap_connector = to_omap_connector(connector); - struct omap_dss_device *dssdev; + struct omap_dss_device *dssdev = NULL; + struct omap_dss_device *d; DBG("%s", connector->name); /* - * If display exposes EDID, then we parse that in the normal way to - * build table of supported modes. + * If the display pipeline reports modes (e.g. with a fixed resolution + * panel or an analog TV output), query it. */ - dssdev = omap_connector_find_device(connector, - OMAP_DSS_DEVICE_OP_EDID); - if (dssdev) - return omap_connector_get_modes_edid(connector, dssdev); + for (d = omap_connector->output; d; d = d->next) { + if (d->ops_flags & OMAP_DSS_DEVICE_OP_MODES) + dssdev = d; + } - /* - * Otherwise if the display pipeline reports modes (e.g. with a fixed - * resolution panel or an analog TV output), query it. - */ - dssdev = omap_connector_find_device(connector, - OMAP_DSS_DEVICE_OP_MODES); if (dssdev) return dssdev->ops->get_modes(dssdev, connector); - /* - * Otherwise if the display pipeline uses a drm_panel, we delegate the - * operation to the panel API. - */ - if (omap_connector->output->panel) - return drm_panel_get_modes(omap_connector->output->panel, - connector); - - /* - * We can't retrieve modes, which can happen for instance for a DVI or - * VGA output with the DDC bus unconnected. The KMS core will add the - * default modes. - */ + /* We can't retrieve modes. The KMS core will add the default modes. */ return 0; } @@ -249,7 +74,7 @@ enum drm_mode_status omap_connector_mode_fixup(struct omap_dss_device *dssdev, drm_mode_copy(adjusted_mode, mode); for (; dssdev; dssdev = dssdev->next) { - if (!dssdev->ops->check_timings) + if (!dssdev->ops || !dssdev->ops->check_timings) continue; ret = dssdev->ops->check_timings(dssdev, adjusted_mode); @@ -298,35 +123,6 @@ static const struct drm_connector_helper_funcs omap_connector_helper_funcs = { .mode_valid = omap_connector_mode_valid, }; -static int omap_connector_get_type(struct omap_dss_device *output) -{ - struct omap_dss_device *display; - enum omap_display_type type; - - display = omapdss_display_get(output); - type = display->type; - omapdss_device_put(display); - - switch (type) { - case OMAP_DISPLAY_TYPE_HDMI: - return DRM_MODE_CONNECTOR_HDMIA; - case OMAP_DISPLAY_TYPE_DVI: - return DRM_MODE_CONNECTOR_DVID; - case OMAP_DISPLAY_TYPE_DSI: - return DRM_MODE_CONNECTOR_DSI; - case OMAP_DISPLAY_TYPE_DPI: - case OMAP_DISPLAY_TYPE_DBI: - return DRM_MODE_CONNECTOR_DPI; - case OMAP_DISPLAY_TYPE_VENC: - /* TODO: This could also be composite */ - return DRM_MODE_CONNECTOR_SVIDEO; - case OMAP_DISPLAY_TYPE_SDI: - return DRM_MODE_CONNECTOR_LVDS; - default: - return DRM_MODE_CONNECTOR_Unknown; - } -} - /* initialize connector */ struct drm_connector *omap_connector_init(struct drm_device *dev, struct omap_dss_device *output, @@ -334,7 +130,6 @@ struct drm_connector *omap_connector_init(struct drm_device *dev, { struct drm_connector *connector = NULL; struct omap_connector *omap_connector; - struct omap_dss_device *dssdev; DBG("%s", output->name); @@ -349,27 +144,9 @@ struct drm_connector *omap_connector_init(struct drm_device *dev, connector->doublescan_allowed = 0; drm_connector_init(dev, connector, &omap_connector_funcs, - omap_connector_get_type(output)); + DRM_MODE_CONNECTOR_DSI); drm_connector_helper_add(connector, &omap_connector_helper_funcs); - /* - * Initialize connector status handling. First try to find a device that - * supports hot-plug reporting. If it fails, fall back to a device that - * support polling. If that fails too, we don't support hot-plug - * detection at all. - */ - dssdev = omap_connector_find_device(connector, OMAP_DSS_DEVICE_OP_HPD); - if (dssdev) { - omap_connector->hpd = omapdss_device_get(dssdev); - connector->polled = DRM_CONNECTOR_POLL_HPD; - } else { - dssdev = omap_connector_find_device(connector, - OMAP_DSS_DEVICE_OP_DETECT); - if (dssdev) - connector->polled = DRM_CONNECTOR_POLL_CONNECT | - DRM_CONNECTOR_POLL_DISCONNECT; - } - return connector; fail: diff --git a/drivers/gpu/drm/omapdrm/omap_connector.h b/drivers/gpu/drm/omapdrm/omap_connector.h index 13607bda33d8..0ecd4f1655b7 100644 --- a/drivers/gpu/drm/omapdrm/omap_connector.h +++ b/drivers/gpu/drm/omapdrm/omap_connector.h @@ -21,9 +21,6 @@ struct omap_dss_device; struct drm_connector *omap_connector_init(struct drm_device *dev, struct omap_dss_device *output, struct drm_encoder *encoder); -bool omap_connector_get_hdmi_mode(struct drm_connector *connector); -void omap_connector_enable_hpd(struct drm_connector *connector); -void omap_connector_disable_hpd(struct drm_connector *connector); enum drm_mode_status omap_connector_mode_fixup(struct omap_dss_device *dssdev, const struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode); diff --git a/drivers/gpu/drm/omapdrm/omap_crtc.c b/drivers/gpu/drm/omapdrm/omap_crtc.c index 3c5ddbf30e97..fce7e944a280 100644 --- a/drivers/gpu/drm/omapdrm/omap_crtc.c +++ b/drivers/gpu/drm/omapdrm/omap_crtc.c @@ -831,7 +831,7 @@ struct drm_crtc *omap_crtc_init(struct drm_device *dev, * OMAP_DSS_CHANNEL_DIGIT. X server assumes 256 element gamma * tables so lets use that. Size of HW gamma table can be * extracted with dispc_mgr_gamma_size(). If it returns 0 - * gamma table is not supprted. + * gamma table is not supported. */ if (priv->dispc_ops->mgr_gamma_size(priv->dispc, channel)) { unsigned int gamma_lut_size = 256; diff --git a/drivers/gpu/drm/omapdrm/omap_dmm_tiler.c b/drivers/gpu/drm/omapdrm/omap_dmm_tiler.c index 252f5ebb1acc..42ec51bb7b1b 100644 --- a/drivers/gpu/drm/omapdrm/omap_dmm_tiler.c +++ b/drivers/gpu/drm/omapdrm/omap_dmm_tiler.c @@ -82,12 +82,11 @@ static const u32 reg[][4] = { static int dmm_dma_copy(struct dmm *dmm, dma_addr_t src, dma_addr_t dst) { - struct dma_device *dma_dev = dmm->wa_dma_chan->device; struct dma_async_tx_descriptor *tx; enum dma_status status; dma_cookie_t cookie; - tx = dma_dev->device_prep_dma_memcpy(dmm->wa_dma_chan, dst, src, 4, 0); + tx = dmaengine_prep_dma_memcpy(dmm->wa_dma_chan, dst, src, 4, 0); if (!tx) { dev_err(dmm->dev, "Failed to prepare DMA memcpy\n"); return -EIO; @@ -99,7 +98,6 @@ static int dmm_dma_copy(struct dmm *dmm, dma_addr_t src, dma_addr_t dst) return -EIO; } - dma_async_issue_pending(dmm->wa_dma_chan); status = dma_sync_wait(dmm->wa_dma_chan, cookie); if (status != DMA_COMPLETE) dev_err(dmm->dev, "i878 wa DMA copy failure\n"); diff --git a/drivers/gpu/drm/omapdrm/omap_drv.c b/drivers/gpu/drm/omapdrm/omap_drv.c index d2750f60f519..cdafd7ef1c32 100644 --- a/drivers/gpu/drm/omapdrm/omap_drv.c +++ b/drivers/gpu/drm/omapdrm/omap_drv.c @@ -12,6 +12,7 @@ #include <drm/drm_atomic.h> #include <drm/drm_atomic_helper.h> #include <drm/drm_bridge.h> +#include <drm/drm_bridge_connector.h> #include <drm/drm_drv.h> #include <drm/drm_fb_helper.h> #include <drm/drm_file.h> @@ -134,9 +135,6 @@ static void omap_disconnect_pipelines(struct drm_device *ddev) for (i = 0; i < priv->num_pipes; i++) { struct omap_drm_pipeline *pipe = &priv->pipes[i]; - if (pipe->output->panel) - drm_panel_detach(pipe->output->panel); - omapdss_device_disconnect(NULL, pipe->output); omapdss_device_put(pipe->output); @@ -209,11 +207,12 @@ static int omap_display_id(struct omap_dss_device *output) struct device_node *node = NULL; if (output->next) { - struct omap_dss_device *display; + struct omap_dss_device *display = output; + + while (display->next) + display = display->next; - display = omapdss_display_get(output); node = display->dev->of_node; - omapdss_device_put(display); } else if (output->bridge) { struct drm_bridge *bridge = output->bridge; @@ -221,8 +220,6 @@ static int omap_display_id(struct omap_dss_device *output) bridge = drm_bridge_get_next_bridge(bridge); node = bridge->of_node; - } else if (output->panel) { - node = output->panel->dev->of_node; } return node ? of_alias_get_id(node, "display") : -ENODEV; @@ -297,9 +294,14 @@ static int omap_modeset_init(struct drm_device *dev) if (pipe->output->bridge) { ret = drm_bridge_attach(pipe->encoder, - pipe->output->bridge, NULL); - if (ret < 0) + pipe->output->bridge, NULL, + DRM_BRIDGE_ATTACH_NO_CONNECTOR); + if (ret < 0) { + dev_err(priv->dev, + "unable to attach bridge %pOF\n", + pipe->output->bridge->of_node); return ret; + } } id = omap_display_id(pipe->output); @@ -330,20 +332,28 @@ static int omap_modeset_init(struct drm_device *dev) struct drm_encoder *encoder = pipe->encoder; struct drm_crtc *crtc; - if (!pipe->output->bridge) { + if (pipe->output->next) { pipe->connector = omap_connector_init(dev, pipe->output, encoder); if (!pipe->connector) return -ENOMEM; + } else { + pipe->connector = drm_bridge_connector_init(dev, encoder); + if (IS_ERR(pipe->connector)) { + dev_err(priv->dev, + "unable to create bridge connector for %s\n", + pipe->output->name); + return PTR_ERR(pipe->connector); + } + } - drm_connector_attach_encoder(pipe->connector, encoder); + drm_connector_attach_encoder(pipe->connector, encoder); - if (pipe->output->panel) { - ret = drm_panel_attach(pipe->output->panel, - pipe->connector); - if (ret < 0) - return ret; - } + if (pipe->output->panel) { + ret = drm_panel_attach(pipe->output->panel, + pipe->connector); + if (ret < 0) + return ret; } crtc = omap_crtc_init(dev, pipe, priv->planes[i]); @@ -382,6 +392,23 @@ static int omap_modeset_init(struct drm_device *dev) return 0; } +static void omap_modeset_fini(struct drm_device *ddev) +{ + struct omap_drm_private *priv = ddev->dev_private; + unsigned int i; + + omap_drm_irq_uninstall(ddev); + + for (i = 0; i < priv->num_pipes; i++) { + struct omap_drm_pipeline *pipe = &priv->pipes[i]; + + if (pipe->output->panel) + drm_panel_detach(pipe->output->panel); + } + + drm_mode_config_cleanup(ddev); +} + /* * Enable the HPD in external components if supported */ @@ -391,8 +418,13 @@ static void omap_modeset_enable_external_hpd(struct drm_device *ddev) unsigned int i; for (i = 0; i < priv->num_pipes; i++) { - if (priv->pipes[i].connector) - omap_connector_enable_hpd(priv->pipes[i].connector); + struct drm_connector *connector = priv->pipes[i].connector; + + if (!connector) + continue; + + if (priv->pipes[i].output->bridge) + drm_bridge_connector_enable_hpd(connector); } } @@ -405,8 +437,13 @@ static void omap_modeset_disable_external_hpd(struct drm_device *ddev) unsigned int i; for (i = 0; i < priv->num_pipes; i++) { - if (priv->pipes[i].connector) - omap_connector_disable_hpd(priv->pipes[i].connector); + struct drm_connector *connector = priv->pipes[i].connector; + + if (!connector) + continue; + + if (priv->pipes[i].output->bridge) + drm_bridge_connector_disable_hpd(connector); } } @@ -629,8 +666,7 @@ err_cleanup_helpers: omap_fbdev_fini(ddev); err_cleanup_modeset: - drm_mode_config_cleanup(ddev); - omap_drm_irq_uninstall(ddev); + omap_modeset_fini(ddev); err_gem_deinit: omap_gem_deinit(ddev); destroy_workqueue(priv->wq); @@ -655,9 +691,7 @@ static void omapdrm_cleanup(struct omap_drm_private *priv) drm_atomic_helper_shutdown(ddev); - drm_mode_config_cleanup(ddev); - - omap_drm_irq_uninstall(ddev); + omap_modeset_fini(ddev); omap_gem_deinit(ddev); destroy_workqueue(priv->wq); diff --git a/drivers/gpu/drm/omapdrm/omap_encoder.c b/drivers/gpu/drm/omapdrm/omap_encoder.c index 4f2165a37795..ae4b867a67a3 100644 --- a/drivers/gpu/drm/omapdrm/omap_encoder.c +++ b/drivers/gpu/drm/omapdrm/omap_encoder.c @@ -10,7 +10,6 @@ #include <drm/drm_crtc.h> #include <drm/drm_modeset_helper_vtables.h> #include <drm/drm_edid.h> -#include <drm/drm_panel.h> #include "omap_drv.h" @@ -70,30 +69,6 @@ static void omap_encoder_update_videomode_flags(struct videomode *vm, } } -static void omap_encoder_hdmi_mode_set(struct drm_connector *connector, - struct drm_encoder *encoder, - struct drm_display_mode *adjusted_mode) -{ - struct omap_encoder *omap_encoder = to_omap_encoder(encoder); - struct omap_dss_device *dssdev = omap_encoder->output; - bool hdmi_mode; - - hdmi_mode = omap_connector_get_hdmi_mode(connector); - - if (dssdev->ops->hdmi.set_hdmi_mode) - dssdev->ops->hdmi.set_hdmi_mode(dssdev, hdmi_mode); - - if (hdmi_mode && dssdev->ops->hdmi.set_infoframe) { - struct hdmi_avi_infoframe avi; - int r; - - r = drm_hdmi_avi_infoframe_from_display_mode(&avi, connector, - adjusted_mode); - if (r == 0) - dssdev->ops->hdmi.set_infoframe(dssdev, &avi); - } -} - static void omap_encoder_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode) @@ -138,17 +113,8 @@ static void omap_encoder_mode_set(struct drm_encoder *encoder, bus_flags = connector->display_info.bus_flags; omap_encoder_update_videomode_flags(&vm, bus_flags); - /* Set timings for all devices in the display pipeline. */ + /* Set timings for the dss manager. */ dss_mgr_set_timings(output, &vm); - - for (dssdev = output; dssdev; dssdev = dssdev->next) { - if (dssdev->ops->set_timings) - dssdev->ops->set_timings(dssdev, adjusted_mode); - } - - /* Set the HDMI mode and HDMI infoframe if applicable. */ - if (output->type == OMAP_DISPLAY_TYPE_HDMI) - omap_encoder_hdmi_mode_set(connector, encoder, adjusted_mode); } static void omap_encoder_disable(struct drm_encoder *encoder) @@ -159,33 +125,12 @@ static void omap_encoder_disable(struct drm_encoder *encoder) dev_dbg(dev->dev, "disable(%s)\n", dssdev->name); - /* Disable the panel if present. */ - if (dssdev->panel) { - drm_panel_disable(dssdev->panel); - drm_panel_unprepare(dssdev->panel); - } - /* * Disable the chain of external devices, starting at the one at the - * internal encoder's output. + * internal encoder's output. This is used for DSI outputs only, as + * dssdev->next is NULL for all other outputs. */ omapdss_device_disable(dssdev->next); - - /* - * Disable the internal encoder. This will disable the DSS output. The - * DSI is treated as an exception as DSI pipelines still use the legacy - * flow where the pipeline output controls the encoder. - */ - if (dssdev->type != OMAP_DISPLAY_TYPE_DSI) { - dssdev->ops->disable(dssdev); - dssdev->state = OMAP_DSS_DISPLAY_DISABLED; - } - - /* - * Perform the post-disable operations on the chain of external devices - * to complete the display pipeline disable. - */ - omapdss_device_post_disable(dssdev->next); } static void omap_encoder_enable(struct drm_encoder *encoder) @@ -196,30 +141,12 @@ static void omap_encoder_enable(struct drm_encoder *encoder) dev_dbg(dev->dev, "enable(%s)\n", dssdev->name); - /* Prepare the chain of external devices for pipeline enable. */ - omapdss_device_pre_enable(dssdev->next); - - /* - * Enable the internal encoder. This will enable the DSS output. The - * DSI is treated as an exception as DSI pipelines still use the legacy - * flow where the pipeline output controls the encoder. - */ - if (dssdev->type != OMAP_DISPLAY_TYPE_DSI) { - dssdev->ops->enable(dssdev); - dssdev->state = OMAP_DSS_DISPLAY_ACTIVE; - } - /* * Enable the chain of external devices, starting at the one at the - * internal encoder's output. + * internal encoder's output. This is used for DSI outputs only, as + * dssdev->next is NULL for all other outputs. */ omapdss_device_enable(dssdev->next); - - /* Enable the panel if present. */ - if (dssdev->panel) { - drm_panel_prepare(dssdev->panel); - drm_panel_enable(dssdev->panel); - } } static int omap_encoder_atomic_check(struct drm_encoder *encoder, diff --git a/drivers/gpu/drm/omapdrm/omap_fbdev.c b/drivers/gpu/drm/omapdrm/omap_fbdev.c index b06e5cbfd03a..09a84919ef73 100644 --- a/drivers/gpu/drm/omapdrm/omap_fbdev.c +++ b/drivers/gpu/drm/omapdrm/omap_fbdev.c @@ -242,14 +242,10 @@ void omap_fbdev_init(struct drm_device *dev) drm_fb_helper_prepare(dev, helper, &omap_fb_helper_funcs); - ret = drm_fb_helper_init(dev, helper, priv->num_pipes); + ret = drm_fb_helper_init(dev, helper); if (ret) goto fail; - ret = drm_fb_helper_single_add_all_connectors(helper); - if (ret) - goto fini; - ret = drm_fb_helper_initial_config(helper, 32); if (ret) goto fini; diff --git a/drivers/gpu/drm/panel/Kconfig b/drivers/gpu/drm/panel/Kconfig index da3b84602cdd..a1723c1b5fbf 100644 --- a/drivers/gpu/drm/panel/Kconfig +++ b/drivers/gpu/drm/panel/Kconfig @@ -59,6 +59,16 @@ config DRM_PANEL_SIMPLE that it can be automatically turned off when the panel goes into a low power state. +config DRM_PANEL_ELIDA_KD35T133 + tristate "Elida KD35T133 panel driver" + depends on OF + depends on DRM_MIPI_DSI + depends on BACKLIGHT_CLASS_DEVICE + help + Say Y here if you want to enable support for the Elida + KD35T133 controller for 320x480 LCD panels with MIPI-DSI + system interfaces. + config DRM_PANEL_FEIXIN_K101_IM2BA02 tristate "Feixin K101 IM2BA02 panel" depends on OF @@ -167,6 +177,16 @@ config DRM_PANEL_NEC_NL8048HL11 panel (found on the Zoom2/3/3630 SDP boards). To compile this driver as a module, choose M here. +config DRM_PANEL_NOVATEK_NT35510 + tristate "Novatek NT35510 RGB panel driver" + depends on OF + depends on DRM_MIPI_DSI + depends on BACKLIGHT_CLASS_DEVICE + help + Say Y here if you want to enable support for the panels built + around the Novatek NT35510 display controller, such as some + Hydis panels. + config DRM_PANEL_NOVATEK_NT39016 tristate "Novatek NT39016 RGB/SPI panel" depends on OF && SPI diff --git a/drivers/gpu/drm/panel/Makefile b/drivers/gpu/drm/panel/Makefile index af1e2a3cc5fc..96a883cd6630 100644 --- a/drivers/gpu/drm/panel/Makefile +++ b/drivers/gpu/drm/panel/Makefile @@ -4,6 +4,7 @@ obj-$(CONFIG_DRM_PANEL_BOE_HIMAX8279D) += panel-boe-himax8279d.o obj-$(CONFIG_DRM_PANEL_BOE_TV101WUM_NL6) += panel-boe-tv101wum-nl6.o obj-$(CONFIG_DRM_PANEL_LVDS) += panel-lvds.o obj-$(CONFIG_DRM_PANEL_SIMPLE) += panel-simple.o +obj-$(CONFIG_DRM_PANEL_ELIDA_KD35T133) += panel-elida-kd35t133.o obj-$(CONFIG_DRM_PANEL_FEIXIN_K101_IM2BA02) += panel-feixin-k101-im2ba02.o obj-$(CONFIG_DRM_PANEL_FEIYANG_FY07024DI26A30D) += panel-feiyang-fy07024di26a30d.o obj-$(CONFIG_DRM_PANEL_ILITEK_IL9322) += panel-ilitek-ili9322.o @@ -15,6 +16,7 @@ obj-$(CONFIG_DRM_PANEL_LEADTEK_LTK500HD1829) += panel-leadtek-ltk500hd1829.o obj-$(CONFIG_DRM_PANEL_LG_LB035Q02) += panel-lg-lb035q02.o obj-$(CONFIG_DRM_PANEL_LG_LG4573) += panel-lg-lg4573.o obj-$(CONFIG_DRM_PANEL_NEC_NL8048HL11) += panel-nec-nl8048hl11.o +obj-$(CONFIG_DRM_PANEL_NOVATEK_NT35510) += panel-novatek-nt35510.o obj-$(CONFIG_DRM_PANEL_NOVATEK_NT39016) += panel-novatek-nt39016.o obj-$(CONFIG_DRM_PANEL_OLIMEX_LCD_OLINUXINO) += panel-olimex-lcd-olinuxino.o obj-$(CONFIG_DRM_PANEL_ORISETECH_OTM8009A) += panel-orisetech-otm8009a.o diff --git a/drivers/gpu/drm/panel/panel-elida-kd35t133.c b/drivers/gpu/drm/panel/panel-elida-kd35t133.c new file mode 100644 index 000000000000..711ded453c44 --- /dev/null +++ b/drivers/gpu/drm/panel/panel-elida-kd35t133.c @@ -0,0 +1,352 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Elida kd35t133 5.5" MIPI-DSI panel driver + * Copyright (C) 2020 Theobroma Systems Design und Consulting GmbH + * + * based on + * + * Rockteck jh057n00900 5.5" MIPI-DSI panel driver + * Copyright (C) Purism SPC 2019 + */ + +#include <linux/delay.h> +#include <linux/gpio/consumer.h> +#include <linux/media-bus-format.h> +#include <linux/module.h> +#include <linux/of.h> +#include <linux/regulator/consumer.h> + +#include <video/display_timing.h> +#include <video/mipi_display.h> + +#include <drm/drm_mipi_dsi.h> +#include <drm/drm_modes.h> +#include <drm/drm_panel.h> +#include <drm/drm_print.h> + +/* Manufacturer specific Commands send via DSI */ +#define KD35T133_CMD_INTERFACEMODECTRL 0xb0 +#define KD35T133_CMD_FRAMERATECTRL 0xb1 +#define KD35T133_CMD_DISPLAYINVERSIONCTRL 0xb4 +#define KD35T133_CMD_DISPLAYFUNCTIONCTRL 0xb6 +#define KD35T133_CMD_POWERCONTROL1 0xc0 +#define KD35T133_CMD_POWERCONTROL2 0xc1 +#define KD35T133_CMD_VCOMCONTROL 0xc5 +#define KD35T133_CMD_POSITIVEGAMMA 0xe0 +#define KD35T133_CMD_NEGATIVEGAMMA 0xe1 +#define KD35T133_CMD_SETIMAGEFUNCTION 0xe9 +#define KD35T133_CMD_ADJUSTCONTROL3 0xf7 + +struct kd35t133 { + struct device *dev; + struct drm_panel panel; + struct gpio_desc *reset_gpio; + struct regulator *vdd; + struct regulator *iovcc; + bool prepared; +}; + +static inline struct kd35t133 *panel_to_kd35t133(struct drm_panel *panel) +{ + return container_of(panel, struct kd35t133, panel); +} + +#define dsi_dcs_write_seq(dsi, cmd, seq...) do { \ + static const u8 d[] = { seq }; \ + int ret; \ + ret = mipi_dsi_dcs_write(dsi, cmd, d, ARRAY_SIZE(d)); \ + if (ret < 0) \ + return ret; \ + } while (0) + +static int kd35t133_init_sequence(struct kd35t133 *ctx) +{ + struct mipi_dsi_device *dsi = to_mipi_dsi_device(ctx->dev); + struct device *dev = ctx->dev; + + /* + * Init sequence was supplied by the panel vendor with minimal + * documentation. + */ + dsi_dcs_write_seq(dsi, KD35T133_CMD_POSITIVEGAMMA, + 0x00, 0x13, 0x18, 0x04, 0x0f, 0x06, 0x3a, 0x56, + 0x4d, 0x03, 0x0a, 0x06, 0x30, 0x3e, 0x0f); + dsi_dcs_write_seq(dsi, KD35T133_CMD_NEGATIVEGAMMA, + 0x00, 0x13, 0x18, 0x01, 0x11, 0x06, 0x38, 0x34, + 0x4d, 0x06, 0x0d, 0x0b, 0x31, 0x37, 0x0f); + dsi_dcs_write_seq(dsi, KD35T133_CMD_POWERCONTROL1, 0x18, 0x17); + dsi_dcs_write_seq(dsi, KD35T133_CMD_POWERCONTROL2, 0x41); + dsi_dcs_write_seq(dsi, KD35T133_CMD_VCOMCONTROL, 0x00, 0x1a, 0x80); + dsi_dcs_write_seq(dsi, MIPI_DCS_SET_ADDRESS_MODE, 0x48); + dsi_dcs_write_seq(dsi, MIPI_DCS_SET_PIXEL_FORMAT, 0x55); + dsi_dcs_write_seq(dsi, KD35T133_CMD_INTERFACEMODECTRL, 0x00); + dsi_dcs_write_seq(dsi, KD35T133_CMD_FRAMERATECTRL, 0xa0); + dsi_dcs_write_seq(dsi, KD35T133_CMD_DISPLAYINVERSIONCTRL, 0x02); + dsi_dcs_write_seq(dsi, KD35T133_CMD_DISPLAYFUNCTIONCTRL, + 0x20, 0x02); + dsi_dcs_write_seq(dsi, KD35T133_CMD_SETIMAGEFUNCTION, 0x00); + dsi_dcs_write_seq(dsi, KD35T133_CMD_ADJUSTCONTROL3, + 0xa9, 0x51, 0x2c, 0x82); + mipi_dsi_dcs_write(dsi, MIPI_DCS_ENTER_INVERT_MODE, NULL, 0); + + DRM_DEV_DEBUG_DRIVER(dev, "Panel init sequence done\n"); + return 0; +} + +static int kd35t133_unprepare(struct drm_panel *panel) +{ + struct kd35t133 *ctx = panel_to_kd35t133(panel); + struct mipi_dsi_device *dsi = to_mipi_dsi_device(ctx->dev); + int ret; + + if (!ctx->prepared) + return 0; + + ret = mipi_dsi_dcs_set_display_off(dsi); + if (ret < 0) + DRM_DEV_ERROR(ctx->dev, "failed to set display off: %d\n", + ret); + + ret = mipi_dsi_dcs_enter_sleep_mode(dsi); + if (ret < 0) { + DRM_DEV_ERROR(ctx->dev, "failed to enter sleep mode: %d\n", + ret); + return ret; + } + + regulator_disable(ctx->iovcc); + regulator_disable(ctx->vdd); + + ctx->prepared = false; + + return 0; +} + +static int kd35t133_prepare(struct drm_panel *panel) +{ + struct kd35t133 *ctx = panel_to_kd35t133(panel); + struct mipi_dsi_device *dsi = to_mipi_dsi_device(ctx->dev); + int ret; + + if (ctx->prepared) + return 0; + + DRM_DEV_DEBUG_DRIVER(ctx->dev, "Resetting the panel\n"); + ret = regulator_enable(ctx->vdd); + if (ret < 0) { + DRM_DEV_ERROR(ctx->dev, + "Failed to enable vdd supply: %d\n", ret); + return ret; + } + + ret = regulator_enable(ctx->iovcc); + if (ret < 0) { + DRM_DEV_ERROR(ctx->dev, + "Failed to enable iovcc supply: %d\n", ret); + goto disable_vdd; + } + + msleep(20); + + gpiod_set_value_cansleep(ctx->reset_gpio, 1); + usleep_range(10, 20); + gpiod_set_value_cansleep(ctx->reset_gpio, 0); + + msleep(20); + + ret = mipi_dsi_dcs_exit_sleep_mode(dsi); + if (ret < 0) { + DRM_DEV_ERROR(ctx->dev, "Failed to exit sleep mode: %d\n", ret); + goto disable_iovcc; + } + + msleep(250); + + ret = kd35t133_init_sequence(ctx); + if (ret < 0) { + DRM_DEV_ERROR(ctx->dev, "Panel init sequence failed: %d\n", + ret); + goto disable_iovcc; + } + + ret = mipi_dsi_dcs_set_display_on(dsi); + if (ret < 0) { + DRM_DEV_ERROR(ctx->dev, "Failed to set display on: %d\n", ret); + goto disable_iovcc; + } + + msleep(50); + + ctx->prepared = true; + + return 0; + +disable_iovcc: + regulator_disable(ctx->iovcc); +disable_vdd: + regulator_disable(ctx->vdd); + return ret; +} + +static const struct drm_display_mode default_mode = { + .hdisplay = 320, + .hsync_start = 320 + 130, + .hsync_end = 320 + 130 + 4, + .htotal = 320 + 130 + 4 + 130, + .vdisplay = 480, + .vsync_start = 480 + 2, + .vsync_end = 480 + 2 + 1, + .vtotal = 480 + 2 + 1 + 2, + .vrefresh = 60, + .clock = 17000, + .width_mm = 42, + .height_mm = 82, +}; + +static int kd35t133_get_modes(struct drm_panel *panel, + struct drm_connector *connector) +{ + struct kd35t133 *ctx = panel_to_kd35t133(panel); + struct drm_display_mode *mode; + + mode = drm_mode_duplicate(connector->dev, &default_mode); + if (!mode) { + DRM_DEV_ERROR(ctx->dev, "Failed to add mode %ux%u@%u\n", + default_mode.hdisplay, default_mode.vdisplay, + default_mode.vrefresh); + return -ENOMEM; + } + + drm_mode_set_name(mode); + + mode->type = DRM_MODE_TYPE_DRIVER | DRM_MODE_TYPE_PREFERRED; + connector->display_info.width_mm = mode->width_mm; + connector->display_info.height_mm = mode->height_mm; + drm_mode_probed_add(connector, mode); + + return 1; +} + +static const struct drm_panel_funcs kd35t133_funcs = { + .unprepare = kd35t133_unprepare, + .prepare = kd35t133_prepare, + .get_modes = kd35t133_get_modes, +}; + +static int kd35t133_probe(struct mipi_dsi_device *dsi) +{ + struct device *dev = &dsi->dev; + struct kd35t133 *ctx; + int ret; + + ctx = devm_kzalloc(dev, sizeof(*ctx), GFP_KERNEL); + if (!ctx) + return -ENOMEM; + + ctx->reset_gpio = devm_gpiod_get_optional(dev, "reset", GPIOD_OUT_LOW); + if (IS_ERR(ctx->reset_gpio)) { + DRM_DEV_ERROR(dev, "cannot get reset gpio\n"); + return PTR_ERR(ctx->reset_gpio); + } + + ctx->vdd = devm_regulator_get(dev, "vdd"); + if (IS_ERR(ctx->vdd)) { + ret = PTR_ERR(ctx->vdd); + if (ret != -EPROBE_DEFER) + DRM_DEV_ERROR(dev, + "Failed to request vdd regulator: %d\n", + ret); + return ret; + } + + ctx->iovcc = devm_regulator_get(dev, "iovcc"); + if (IS_ERR(ctx->iovcc)) { + ret = PTR_ERR(ctx->iovcc); + if (ret != -EPROBE_DEFER) + DRM_DEV_ERROR(dev, + "Failed to request iovcc regulator: %d\n", + ret); + return ret; + } + + mipi_dsi_set_drvdata(dsi, ctx); + + ctx->dev = dev; + + dsi->lanes = 1; + dsi->format = MIPI_DSI_FMT_RGB888; + dsi->mode_flags = MIPI_DSI_MODE_VIDEO | MIPI_DSI_MODE_VIDEO_BURST | + MIPI_DSI_MODE_LPM | MIPI_DSI_MODE_EOT_PACKET; + + drm_panel_init(&ctx->panel, &dsi->dev, &kd35t133_funcs, + DRM_MODE_CONNECTOR_DSI); + + ret = drm_panel_of_backlight(&ctx->panel); + if (ret) + return ret; + + drm_panel_add(&ctx->panel); + + ret = mipi_dsi_attach(dsi); + if (ret < 0) { + DRM_DEV_ERROR(dev, "mipi_dsi_attach failed: %d\n", ret); + drm_panel_remove(&ctx->panel); + return ret; + } + + return 0; +} + +static void kd35t133_shutdown(struct mipi_dsi_device *dsi) +{ + struct kd35t133 *ctx = mipi_dsi_get_drvdata(dsi); + int ret; + + ret = drm_panel_unprepare(&ctx->panel); + if (ret < 0) + DRM_DEV_ERROR(&dsi->dev, "Failed to unprepare panel: %d\n", + ret); + + ret = drm_panel_disable(&ctx->panel); + if (ret < 0) + DRM_DEV_ERROR(&dsi->dev, "Failed to disable panel: %d\n", + ret); +} + +static int kd35t133_remove(struct mipi_dsi_device *dsi) +{ + struct kd35t133 *ctx = mipi_dsi_get_drvdata(dsi); + int ret; + + kd35t133_shutdown(dsi); + + ret = mipi_dsi_detach(dsi); + if (ret < 0) + DRM_DEV_ERROR(&dsi->dev, "Failed to detach from DSI host: %d\n", + ret); + + drm_panel_remove(&ctx->panel); + + return 0; +} + +static const struct of_device_id kd35t133_of_match[] = { + { .compatible = "elida,kd35t133" }, + { /* sentinel */ } +}; +MODULE_DEVICE_TABLE(of, kd35t133_of_match); + +static struct mipi_dsi_driver kd35t133_driver = { + .driver = { + .name = "panel-elida-kd35t133", + .of_match_table = kd35t133_of_match, + }, + .probe = kd35t133_probe, + .remove = kd35t133_remove, + .shutdown = kd35t133_shutdown, +}; +module_mipi_dsi_driver(kd35t133_driver); + +MODULE_AUTHOR("Heiko Stuebner <heiko.stuebner@theobroma-systems.com>"); +MODULE_DESCRIPTION("DRM driver for Elida kd35t133 MIPI DSI panel"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/gpu/drm/panel/panel-ilitek-ili9322.c b/drivers/gpu/drm/panel/panel-ilitek-ili9322.c index f394d53a7da4..09935520e606 100644 --- a/drivers/gpu/drm/panel/panel-ilitek-ili9322.c +++ b/drivers/gpu/drm/panel/panel-ilitek-ili9322.c @@ -540,7 +540,7 @@ static int ili9322_enable(struct drm_panel *panel) /* Serial RGB modes */ static const struct drm_display_mode srgb_320x240_mode = { - .clock = 2453500, + .clock = 24535, .hdisplay = 320, .hsync_start = 320 + 359, .hsync_end = 320 + 359 + 1, @@ -554,7 +554,7 @@ static const struct drm_display_mode srgb_320x240_mode = { }; static const struct drm_display_mode srgb_360x240_mode = { - .clock = 2700000, + .clock = 27000, .hdisplay = 360, .hsync_start = 360 + 35, .hsync_end = 360 + 35 + 1, @@ -569,7 +569,7 @@ static const struct drm_display_mode srgb_360x240_mode = { /* This is the only mode listed for parallel RGB in the datasheet */ static const struct drm_display_mode prgb_320x240_mode = { - .clock = 6400000, + .clock = 64000, .hdisplay = 320, .hsync_start = 320 + 38, .hsync_end = 320 + 38 + 1, @@ -584,7 +584,7 @@ static const struct drm_display_mode prgb_320x240_mode = { /* YUV modes */ static const struct drm_display_mode yuv_640x320_mode = { - .clock = 2454000, + .clock = 24540, .hdisplay = 640, .hsync_start = 640 + 252, .hsync_end = 640 + 252 + 1, @@ -598,7 +598,7 @@ static const struct drm_display_mode yuv_640x320_mode = { }; static const struct drm_display_mode yuv_720x360_mode = { - .clock = 2700000, + .clock = 27000, .hdisplay = 720, .hsync_start = 720 + 252, .hsync_end = 720 + 252 + 1, @@ -613,7 +613,7 @@ static const struct drm_display_mode yuv_720x360_mode = { /* BT.656 VGA mode, 640x480 */ static const struct drm_display_mode itu_r_bt_656_640_mode = { - .clock = 2454000, + .clock = 24540, .hdisplay = 640, .hsync_start = 640 + 3, .hsync_end = 640 + 3 + 1, @@ -628,7 +628,7 @@ static const struct drm_display_mode itu_r_bt_656_640_mode = { /* BT.656 D1 mode 720x480 */ static const struct drm_display_mode itu_r_bt_656_720_mode = { - .clock = 2700000, + .clock = 27000, .hdisplay = 720, .hsync_start = 720 + 3, .hsync_end = 720 + 3 + 1, diff --git a/drivers/gpu/drm/panel/panel-lg-lg4573.c b/drivers/gpu/drm/panel/panel-lg-lg4573.c index b262b53dbd85..5907f2503755 100644 --- a/drivers/gpu/drm/panel/panel-lg-lg4573.c +++ b/drivers/gpu/drm/panel/panel-lg-lg4573.c @@ -197,7 +197,7 @@ static int lg4573_enable(struct drm_panel *panel) } static const struct drm_display_mode default_mode = { - .clock = 27000, + .clock = 28341, .hdisplay = 480, .hsync_start = 480 + 10, .hsync_end = 480 + 10 + 59, diff --git a/drivers/gpu/drm/panel/panel-novatek-nt35510.c b/drivers/gpu/drm/panel/panel-novatek-nt35510.c new file mode 100644 index 000000000000..4a8fa908a2cf --- /dev/null +++ b/drivers/gpu/drm/panel/panel-novatek-nt35510.c @@ -0,0 +1,1098 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Novatek NT35510 panel driver + * Copyright (C) 2020 Linus Walleij <linus.walleij@linaro.org> + * Based on code by Robert Teather (C) 2012 Samsung + * + * This display driver (and I refer to the physical component NT35510, + * not this Linux kernel software driver) can handle: + * 480x864, 480x854, 480x800, 480x720 and 480x640 pixel displays. + * It has 480x840x24bit SRAM embedded for storing a frame. + * When powered on the display is by default in 480x800 mode. + * + * The actual panels using this component have different names, but + * the code needed to set up and configure the panel will be similar, + * so they should all use the NT35510 driver with appropriate configuration + * per-panel, e.g. for physical size. + * + * This driver is for the DSI interface to panels using the NT35510. + * + * The NT35510 can also use an RGB (DPI) interface combined with an + * I2C or SPI interface for setting up the NT35510. If this is needed + * this panel driver should be refactored to also support that use + * case. + */ +#include <linux/backlight.h> +#include <linux/bitops.h> +#include <linux/gpio/consumer.h> +#include <linux/module.h> +#include <linux/of_device.h> +#include <linux/regmap.h> +#include <linux/regulator/consumer.h> + +#include <video/mipi_display.h> + +#include <drm/drm_mipi_dsi.h> +#include <drm/drm_modes.h> +#include <drm/drm_panel.h> +#include <drm/drm_print.h> + +#define MCS_CMD_MAUCCTR 0xF0 /* Manufacturer command enable */ +#define MCS_CMD_READ_ID1 0xDA +#define MCS_CMD_READ_ID2 0xDB +#define MCS_CMD_READ_ID3 0xDC +#define MCS_CMD_MTP_READ_SETTING 0xF8 /* Uncertain about name */ +#define MCS_CMD_MTP_READ_PARAM 0xFF /* Uncertain about name */ + +/* + * These manufacturer commands are available after we enable manufacturer + * command set (MCS) for page 0. + */ +#define NT35510_P0_DOPCTR 0xB1 +#define NT35510_P0_SDHDTCTR 0xB6 +#define NT35510_P0_GSEQCTR 0xB7 +#define NT35510_P0_SDEQCTR 0xB8 +#define NT35510_P0_SDVPCTR 0xBA +#define NT35510_P0_DPFRCTR1 0xBD +#define NT35510_P0_DPFRCTR2 0xBE +#define NT35510_P0_DPFRCTR3 0xBF +#define NT35510_P0_DPMCTR12 0xCC + +#define NT35510_P0_DOPCTR_LEN 2 +#define NT35510_P0_GSEQCTR_LEN 2 +#define NT35510_P0_SDEQCTR_LEN 4 +#define NT35510_P0_SDVPCTR_LEN 1 +#define NT35510_P0_DPFRCTR1_LEN 5 +#define NT35510_P0_DPFRCTR2_LEN 5 +#define NT35510_P0_DPFRCTR3_LEN 5 +#define NT35510_P0_DPMCTR12_LEN 3 + +#define NT35510_DOPCTR_0_RAMKP BIT(7) /* Contents kept in sleep */ +#define NT35510_DOPCTR_0_DSITE BIT(6) /* Enable TE signal */ +#define NT35510_DOPCTR_0_DSIG BIT(5) /* Enable generic read/write */ +#define NT35510_DOPCTR_0_DSIM BIT(4) /* Enable video mode on DSI */ +#define NT35510_DOPCTR_0_EOTP BIT(3) /* Support EoTP */ +#define NT35510_DOPCTR_0_N565 BIT(2) /* RGB or BGR pixel format */ +#define NT35510_DOPCTR_1_TW_PWR_SEL BIT(4) /* TE power selector */ +#define NT35510_DOPCTR_1_CRGB BIT(3) /* RGB or BGR byte order */ +#define NT35510_DOPCTR_1_CTB BIT(2) /* Vertical scanning direction */ +#define NT35510_DOPCTR_1_CRL BIT(1) /* Source driver data shift */ +#define NT35510_P0_SDVPCTR_PRG BIT(2) /* 0 = normal operation, 1 = VGLO */ +#define NT35510_P0_SDVPCTR_AVDD 0 /* source driver output = AVDD */ +#define NT35510_P0_SDVPCTR_OFFCOL 1 /* source driver output = off color */ +#define NT35510_P0_SDVPCTR_AVSS 2 /* source driver output = AVSS */ +#define NT35510_P0_SDVPCTR_HI_Z 3 /* source driver output = High impedance */ + +/* + * These manufacturer commands are available after we enable manufacturer + * command set (MCS) for page 1. + */ +#define NT35510_P1_SETAVDD 0xB0 +#define NT35510_P1_SETAVEE 0xB1 +#define NT35510_P1_SETVCL 0xB2 +#define NT35510_P1_SETVGH 0xB3 +#define NT35510_P1_SETVRGH 0xB4 +#define NT35510_P1_SETVGL 0xB5 +#define NT35510_P1_BT1CTR 0xB6 +#define NT35510_P1_BT2CTR 0xB7 +#define NT35510_P1_BT3CTR 0xB8 +#define NT35510_P1_BT4CTR 0xB9 /* VGH boosting times/freq */ +#define NT35510_P1_BT5CTR 0xBA +#define NT35510_P1_PFMCTR 0xBB +#define NT35510_P1_SETVGP 0xBC +#define NT35510_P1_SETVGN 0xBD +#define NT35510_P1_SETVCMOFF 0xBE +#define NT35510_P1_VGHCTR 0xBF /* VGH output ctrl */ +#define NT35510_P1_SET_GAMMA_RED_POS 0xD1 +#define NT35510_P1_SET_GAMMA_GREEN_POS 0xD2 +#define NT35510_P1_SET_GAMMA_BLUE_POS 0xD3 +#define NT35510_P1_SET_GAMMA_RED_NEG 0xD4 +#define NT35510_P1_SET_GAMMA_GREEN_NEG 0xD5 +#define NT35510_P1_SET_GAMMA_BLUE_NEG 0xD6 + +/* AVDD and AVEE setting 3 bytes */ +#define NT35510_P1_AVDD_LEN 3 +#define NT35510_P1_AVEE_LEN 3 +#define NT35510_P1_VGH_LEN 3 +#define NT35510_P1_VGL_LEN 3 +#define NT35510_P1_VGP_LEN 3 +#define NT35510_P1_VGN_LEN 3 +/* BT1CTR thru BT5CTR setting 3 bytes */ +#define NT35510_P1_BT1CTR_LEN 3 +#define NT35510_P1_BT2CTR_LEN 3 +#define NT35510_P1_BT4CTR_LEN 3 +#define NT35510_P1_BT5CTR_LEN 3 +/* 52 gamma parameters times two per color: positive and negative */ +#define NT35510_P1_GAMMA_LEN 52 + +/** + * struct nt35510_config - the display-specific NT35510 configuration + * + * Some of the settings provide an array of bytes, A, B C which mean: + * A = normal / idle off mode + * B = idle on mode + * C = partial / idle off mode + * + * Gamma correction arrays are 10bit numbers, two consecutive bytes + * makes out one point on the gamma correction curve. The points are + * not linearly placed along the X axis, we get points 0, 1, 3, 5 + * 7, 11, 15, 23, 31, 47, 63, 95, 127, 128, 160, 192, 208, 224, 232, + * 240, 244, 248, 250, 252, 254, 255. The voltages tuples form + * V0, V1, V3 ... V255, with 0x0000 being the lowest voltage and + * 0x03FF being the highest voltage. + * + * Each value must be strictly higher than the previous value forming + * a rising curve like this: + * + * ^ + * | V255 + * | V254 + * | .... + * | V5 + * | V3 + * | V1 + * | V0 + * +-------------------------------------------> + * + * The details about all settings can be found in the NT35510 Application + * Note. + */ +struct nt35510_config { + /** + * @width_mm: physical panel width [mm] + */ + u32 width_mm; + /** + * @height_mm: physical panel height [mm] + */ + u32 height_mm; + /** + * @mode: the display mode. This is only relevant outside the panel + * in video mode: in command mode this is configuring the internal + * timing in the display controller. + */ + const struct drm_display_mode mode; + /** + * @avdd: setting for AVDD ranging from 0x00 = 6.5V to 0x14 = 4.5V + * in 0.1V steps the default is 0x05 which means 6.0V + */ + u8 avdd[NT35510_P1_AVDD_LEN]; + /** + * @bt1ctr: setting for boost power control for the AVDD step-up + * circuit (1) + * bits 0..2 in the lower nibble controls PCK, the booster clock + * frequency for the step-up circuit: + * 0 = Hsync/32 + * 1 = Hsync/16 + * 2 = Hsync/8 + * 3 = Hsync/4 + * 4 = Hsync/2 + * 5 = Hsync + * 6 = Hsync x 2 + * 7 = Hsync x 4 + * bits 4..6 in the upper nibble controls BTP, the boosting + * amplification for the the step-up circuit: + * 0 = Disable + * 1 = 1.5 x VDDB + * 2 = 1.66 x VDDB + * 3 = 2 x VDDB + * 4 = 2.5 x VDDB + * 5 = 3 x VDDB + * The defaults are 4 and 4 yielding 0x44 + */ + u8 bt1ctr[NT35510_P1_BT1CTR_LEN]; + /** + * @avee: setting for AVEE ranging from 0x00 = -6.5V to 0x14 = -4.5V + * in 0.1V steps the default is 0x05 which means -6.0V + */ + u8 avee[NT35510_P1_AVEE_LEN]; + /** + * @bt2ctr: setting for boost power control for the AVEE step-up + * circuit (2) + * bits 0..2 in the lower nibble controls NCK, the booster clock + * frequency, the values are the same as for PCK in @bt1ctr. + * bits 4..5 in the upper nibble controls BTN, the boosting + * amplification for the the step-up circuit. + * 0 = Disable + * 1 = -1.5 x VDDB + * 2 = -2 x VDDB + * 3 = -2.5 x VDDB + * 4 = -3 x VDDB + * The defaults are 4 and 3 yielding 0x34 + */ + u8 bt2ctr[NT35510_P1_BT2CTR_LEN]; + /** + * @vgh: setting for VGH ranging from 0x00 = 7.0V to 0x0B = 18.0V + * in 1V steps, the default is 0x08 which means 15V + */ + u8 vgh[NT35510_P1_VGH_LEN]; + /** + * @bt4ctr: setting for boost power control for the VGH step-up + * circuit (4) + * bits 0..2 in the lower nibble controls HCK, the booster clock + * frequency, the values are the same as for PCK in @bt1ctr. + * bits 4..5 in the upper nibble controls BTH, the boosting + * amplification for the the step-up circuit. + * 0 = AVDD + VDDB + * 1 = AVDD - AVEE + * 2 = AVDD - AVEE + VDDB + * 3 = AVDD x 2 - AVEE + * The defaults are 4 and 3 yielding 0x34 + */ + u8 bt4ctr[NT35510_P1_BT4CTR_LEN]; + /** + * @vgl: setting for VGL ranging from 0x00 = -2V to 0x0f = -15V in + * 1V steps, the default is 0x08 which means -10V + */ + u8 vgl[NT35510_P1_VGL_LEN]; + /** + * @bt5ctr: setting for boost power control for the VGL step-up + * circuit (5) + * bits 0..2 in the lower nibble controls LCK, the booster clock + * frequency, the values are the same as for PCK in @bt1ctr. + * bits 4..5 in the upper nibble controls BTL, the boosting + * amplification for the the step-up circuit. + * 0 = AVEE + VCL + * 1 = AVEE - AVDD + * 2 = AVEE + VCL - AVDD + * 3 = AVEE x 2 - AVDD + * The defaults are 3 and 2 yielding 0x32 + */ + u8 bt5ctr[NT35510_P1_BT5CTR_LEN]; + /** + * @vgp: setting for VGP, the positive gamma divider voltages + * VGMP the high voltage and VGSP the low voltage. + * The first byte contains bit 8 of VGMP and VGSP in bits 4 and 0 + * The second byte contains bit 0..7 of VGMP + * The third byte contains bit 0..7 of VGSP + * VGMP 0x00 = 3.0V .. 0x108 = 6.3V in steps of 12.5mV + * VGSP 0x00 = 0V .. 0x111 = 3.7V in steps of 12.5mV + */ + u8 vgp[NT35510_P1_VGP_LEN]; + /** + * @vgn: setting for VGN, the negative gamma divider voltages, + * same layout of bytes as @vgp. + */ + u8 vgn[NT35510_P1_VGN_LEN]; + /** + * @sdeqctr: Source driver control settings, first byte is + * 0 for mode 1 and 1 for mode 2. Mode 1 uses two steps and + * mode 2 uses three steps meaning EQS3 is not used in mode + * 1. Mode 2 is default. The last three parameters are EQS1, EQS2 + * and EQS3, setting the rise time for each equalizer step: + * 0x00 = 0.0 us to 0x0f = 7.5 us in steps of 0.5us. The default + * is 0x07 = 3.5 us. + */ + u8 sdeqctr[NT35510_P0_SDEQCTR_LEN]; + /** + * @sdvpctr: power/voltage behaviour during vertical porch time + */ + u8 sdvpctr; + /** + * @t1: the number of pixel clocks on one scanline, range + * 0x100 (258 ticks) .. 0x3FF (1024 ticks) so the value + 1 + * clock ticks. + */ + u16 t1; + /** + * @vbp: vertical back porch toward the PANEL note: not toward + * the DSI host; these are separate interfaces, in from DSI host + * and out to the panel. + */ + u8 vbp; + /** + * @vfp: vertical front porch toward the PANEL. + */ + u8 vfp; + /** + * @psel: pixel clock divisor: 0 = 1, 1 = 2, 2 = 4, 3 = 8. + */ + u8 psel; + /** + * @dpmctr12: Display timing control 12 + * Byte 1 bit 4 selects LVGL voltage level: 0 = VGLX, 1 = VGL_REG + * Byte 1 bit 1 selects gate signal mode: 0 = non-overlap, 1 = overlap + * Byte 1 bit 0 selects output signal control R/L swap, 0 = normal + * 1 = swap all O->E, L->R + * Byte 2 is CLW delay clock for CK O/E and CKB O/E signals: + * 0x00 = 0us .. 0xFF = 12.75us in 0.05us steps + * Byte 3 is FTI_H0 delay time for STP O/E signals: + * 0x00 = 0us .. 0xFF = 12.75us in 0.05us steps + */ + u8 dpmctr12[NT35510_P0_DPMCTR12_LEN]; + /** + * @gamma_corr_pos_r: Red gamma correction parameters, positive + */ + u8 gamma_corr_pos_r[NT35510_P1_GAMMA_LEN]; + /** + * @gamma_corr_pos_g: Green gamma correction parameters, positive + */ + u8 gamma_corr_pos_g[NT35510_P1_GAMMA_LEN]; + /** + * @gamma_corr_pos_b: Blue gamma correction parameters, positive + */ + u8 gamma_corr_pos_b[NT35510_P1_GAMMA_LEN]; + /** + * @gamma_corr_neg_r: Red gamma correction parameters, negative + */ + u8 gamma_corr_neg_r[NT35510_P1_GAMMA_LEN]; + /** + * @gamma_corr_neg_g: Green gamma correction parameters, negative + */ + u8 gamma_corr_neg_g[NT35510_P1_GAMMA_LEN]; + /** + * @gamma_corr_neg_b: Blue gamma correction parameters, negative + */ + u8 gamma_corr_neg_b[NT35510_P1_GAMMA_LEN]; +}; + +/** + * struct nt35510 - state container for the NT35510 panel + */ +struct nt35510 { + /** + * @dev: the container device + */ + struct device *dev; + /** + * @conf: the specific panel configuration, as the NT35510 + * can be combined with many physical panels, they can have + * different physical dimensions and gamma correction etc, + * so this is stored in the config. + */ + const struct nt35510_config *conf; + /** + * @panel: the DRM panel object for the instance + */ + struct drm_panel panel; + /** + * @supplies: regulators supplying the panel + */ + struct regulator_bulk_data supplies[2]; + /** + * @reset_gpio: the reset line + */ + struct gpio_desc *reset_gpio; +}; + +/* Manufacturer command has strictly this byte sequence */ +static const u8 nt35510_mauc_select_page_0[] = { 0x55, 0xAA, 0x52, 0x08, 0x00 }; +static const u8 nt35510_mauc_select_page_1[] = { 0x55, 0xAA, 0x52, 0x08, 0x01 }; +static const u8 nt35510_vgh_on[] = { 0x01 }; + +static inline struct nt35510 *panel_to_nt35510(struct drm_panel *panel) +{ + return container_of(panel, struct nt35510, panel); +} + +#define NT35510_ROTATE_0_SETTING 0x02 +#define NT35510_ROTATE_180_SETTING 0x00 + +static int nt35510_send_long(struct nt35510 *nt, struct mipi_dsi_device *dsi, + u8 cmd, u8 cmdlen, const u8 *seq) +{ + const u8 *seqp = seq; + int cmdwritten = 0; + int chunk = cmdlen; + int ret; + + if (chunk > 15) + chunk = 15; + ret = mipi_dsi_dcs_write(dsi, cmd, seqp, chunk); + if (ret < 0) { + DRM_DEV_ERROR(nt->dev, + "error sending DCS command seq cmd %02x\n", + cmd); + return ret; + } + cmdwritten += chunk; + seqp += chunk; + + while (cmdwritten < cmdlen) { + chunk = cmdlen - cmdwritten; + if (chunk > 15) + chunk = 15; + ret = mipi_dsi_generic_write(dsi, seqp, chunk); + if (ret < 0) { + DRM_DEV_ERROR(nt->dev, + "error sending generic write seq %02x\n", + cmd); + return ret; + } + cmdwritten += chunk; + seqp += chunk; + } + DRM_DEV_DEBUG(nt->dev, "sent command %02x %02x bytes\n", + cmd, cmdlen); + return 0; +} + +static int nt35510_read_id(struct nt35510 *nt) +{ + struct mipi_dsi_device *dsi = to_mipi_dsi_device(nt->dev); + u8 id1, id2, id3; + int ret; + + ret = mipi_dsi_dcs_read(dsi, MCS_CMD_READ_ID1, &id1, 1); + if (ret < 0) { + DRM_DEV_ERROR(nt->dev, "could not read MTP ID1\n"); + return ret; + } + ret = mipi_dsi_dcs_read(dsi, MCS_CMD_READ_ID2, &id2, 1); + if (ret < 0) { + DRM_DEV_ERROR(nt->dev, "could not read MTP ID2\n"); + return ret; + } + ret = mipi_dsi_dcs_read(dsi, MCS_CMD_READ_ID3, &id3, 1); + if (ret < 0) { + DRM_DEV_ERROR(nt->dev, "could not read MTP ID3\n"); + return ret; + } + + /* + * Multi-Time Programmable (?) memory contains manufacturer + * ID (e.g. Hydis 0x55), driver ID (e.g. NT35510 0xc0) and + * version. + */ + DRM_DEV_INFO(nt->dev, + "MTP ID manufacturer: %02x version: %02x driver: %02x\n", + id1, id2, id3); + + return 0; +} + +/** + * nt35510_setup_power() - set up power config in page 1 + * @nt: the display instance to set up + */ +static int nt35510_setup_power(struct nt35510 *nt) +{ + struct mipi_dsi_device *dsi = to_mipi_dsi_device(nt->dev); + int ret; + + ret = nt35510_send_long(nt, dsi, NT35510_P1_SETAVDD, + NT35510_P1_AVDD_LEN, + nt->conf->avdd); + if (ret) + return ret; + ret = nt35510_send_long(nt, dsi, NT35510_P1_BT1CTR, + NT35510_P1_BT1CTR_LEN, + nt->conf->bt1ctr); + if (ret) + return ret; + ret = nt35510_send_long(nt, dsi, NT35510_P1_SETAVEE, + NT35510_P1_AVEE_LEN, + nt->conf->avee); + if (ret) + return ret; + ret = nt35510_send_long(nt, dsi, NT35510_P1_BT2CTR, + NT35510_P1_BT2CTR_LEN, + nt->conf->bt2ctr); + if (ret) + return ret; + ret = nt35510_send_long(nt, dsi, NT35510_P1_SETVGH, + NT35510_P1_VGH_LEN, + nt->conf->vgh); + if (ret) + return ret; + ret = nt35510_send_long(nt, dsi, NT35510_P1_BT4CTR, + NT35510_P1_BT4CTR_LEN, + nt->conf->bt4ctr); + if (ret) + return ret; + ret = nt35510_send_long(nt, dsi, NT35510_P1_VGHCTR, + ARRAY_SIZE(nt35510_vgh_on), + nt35510_vgh_on); + if (ret) + return ret; + ret = nt35510_send_long(nt, dsi, NT35510_P1_SETVGL, + NT35510_P1_VGL_LEN, + nt->conf->vgl); + if (ret) + return ret; + ret = nt35510_send_long(nt, dsi, NT35510_P1_BT5CTR, + NT35510_P1_BT5CTR_LEN, + nt->conf->bt5ctr); + if (ret) + return ret; + ret = nt35510_send_long(nt, dsi, NT35510_P1_SETVGP, + NT35510_P1_VGP_LEN, + nt->conf->vgp); + if (ret) + return ret; + ret = nt35510_send_long(nt, dsi, NT35510_P1_SETVGN, + NT35510_P1_VGN_LEN, + nt->conf->vgn); + if (ret) + return ret; + + /* Typically 10 ms */ + usleep_range(10000, 20000); + + return 0; +} + +/** + * nt35510_setup_display() - set up display config in page 0 + * @nt: the display instance to set up + */ +static int nt35510_setup_display(struct nt35510 *nt) +{ + struct mipi_dsi_device *dsi = to_mipi_dsi_device(nt->dev); + const struct nt35510_config *conf = nt->conf; + u8 dopctr[NT35510_P0_DOPCTR_LEN]; + u8 gseqctr[NT35510_P0_GSEQCTR_LEN]; + u8 dpfrctr[NT35510_P0_DPFRCTR1_LEN]; + /* FIXME: set up any rotation (assume none for now) */ + u8 addr_mode = NT35510_ROTATE_0_SETTING; + u8 val; + int ret; + + /* Enable TE, EoTP and RGB pixel format */ + dopctr[0] = NT35510_DOPCTR_0_DSITE | NT35510_DOPCTR_0_EOTP | + NT35510_DOPCTR_0_N565; + dopctr[1] = NT35510_DOPCTR_1_CTB; + ret = nt35510_send_long(nt, dsi, NT35510_P0_DOPCTR, + NT35510_P0_DOPCTR_LEN, + dopctr); + if (ret) + return ret; + + ret = mipi_dsi_dcs_write(dsi, MIPI_DCS_SET_ADDRESS_MODE, &addr_mode, + sizeof(addr_mode)); + if (ret < 0) + return ret; + + /* + * Source data hold time, default 0x05 = 2.5us + * 0x00..0x3F = 0 .. 31.5us in steps of 0.5us + * 0x0A = 5us + */ + val = 0x0A; + ret = mipi_dsi_dcs_write(dsi, NT35510_P0_SDHDTCTR, &val, + sizeof(val)); + if (ret < 0) + return ret; + + /* EQ control for gate signals, 0x00 = 0 us */ + gseqctr[0] = 0x00; + gseqctr[1] = 0x00; + ret = nt35510_send_long(nt, dsi, NT35510_P0_GSEQCTR, + NT35510_P0_GSEQCTR_LEN, + gseqctr); + if (ret) + return ret; + + ret = nt35510_send_long(nt, dsi, NT35510_P0_SDEQCTR, + NT35510_P0_SDEQCTR_LEN, + conf->sdeqctr); + if (ret) + return ret; + + ret = mipi_dsi_dcs_write(dsi, NT35510_P0_SDVPCTR, + &conf->sdvpctr, 1); + if (ret < 0) + return ret; + + /* + * Display timing control for active and idle off mode: + * the first byte contains + * the two high bits of T1A and second byte the low 8 bits, and + * the valid range is 0x100 (257) to 0x3ff (1023) representing + * 258..1024 (+1) pixel clock ticks for one scanline. At 20MHz pixel + * clock this covers the range of 12.90us .. 51.20us in steps of + * 0.05us, the default is 0x184 (388) representing 389 ticks. + * The third byte is VBPDA, vertical back porch display active + * and the fourth VFPDA, vertical front porch display active, + * both given in number of scanlines in the range 0x02..0xff + * for 2..255 scanlines. The fifth byte is 2 bits selecting + * PSEL for active and idle off mode, how much the 20MHz clock + * is divided by 0..3. This needs to be adjusted to get the right + * frame rate. + */ + dpfrctr[0] = (conf->t1 >> 8) & 0xFF; + dpfrctr[1] = conf->t1 & 0xFF; + /* Vertical back porch */ + dpfrctr[2] = conf->vbp; + /* Vertical front porch */ + dpfrctr[3] = conf->vfp; + dpfrctr[4] = conf->psel; + ret = nt35510_send_long(nt, dsi, NT35510_P0_DPFRCTR1, + NT35510_P0_DPFRCTR1_LEN, + dpfrctr); + if (ret) + return ret; + /* For idle and partial idle off mode we decrease front porch by one */ + dpfrctr[3]--; + ret = nt35510_send_long(nt, dsi, NT35510_P0_DPFRCTR2, + NT35510_P0_DPFRCTR2_LEN, + dpfrctr); + if (ret) + return ret; + ret = nt35510_send_long(nt, dsi, NT35510_P0_DPFRCTR3, + NT35510_P0_DPFRCTR3_LEN, + dpfrctr); + if (ret) + return ret; + + /* Enable TE on vblank */ + ret = mipi_dsi_dcs_set_tear_on(dsi, MIPI_DSI_DCS_TEAR_MODE_VBLANK); + if (ret) + return ret; + + /* Turn on the pads? */ + ret = nt35510_send_long(nt, dsi, NT35510_P0_DPMCTR12, + NT35510_P0_DPMCTR12_LEN, + conf->dpmctr12); + if (ret) + return ret; + + return 0; +} + +static int nt35510_set_brightness(struct backlight_device *bl) +{ + struct nt35510 *nt = bl_get_data(bl); + struct mipi_dsi_device *dsi = to_mipi_dsi_device(nt->dev); + u8 brightness = bl->props.brightness; + int ret; + + DRM_DEV_DEBUG(nt->dev, "set brightness %d\n", brightness); + ret = mipi_dsi_dcs_write(dsi, MIPI_DCS_SET_DISPLAY_BRIGHTNESS, + &brightness, + sizeof(brightness)); + if (ret < 0) + return ret; + + return 0; +} + +static const struct backlight_ops nt35510_bl_ops = { + .update_status = nt35510_set_brightness, +}; + +/* + * This power-on sequence + */ +static int nt35510_power_on(struct nt35510 *nt) +{ + struct mipi_dsi_device *dsi = to_mipi_dsi_device(nt->dev); + int ret; + + ret = regulator_bulk_enable(ARRAY_SIZE(nt->supplies), nt->supplies); + if (ret < 0) { + dev_err(nt->dev, "unable to enable regulators\n"); + return ret; + } + + /* Toggle RESET in accordance with datasheet page 370 */ + if (nt->reset_gpio) { + gpiod_set_value(nt->reset_gpio, 1); + /* Active min 10 us according to datasheet, let's say 20 */ + usleep_range(20, 1000); + gpiod_set_value(nt->reset_gpio, 0); + /* + * 5 ms during sleep mode, 120 ms during sleep out mode + * according to datasheet, let's use 120-140 ms. + */ + usleep_range(120000, 140000); + } + + ret = nt35510_read_id(nt); + if (ret) + return ret; + + /* Set up stuff in manufacturer control, page 1 */ + ret = nt35510_send_long(nt, dsi, MCS_CMD_MAUCCTR, + ARRAY_SIZE(nt35510_mauc_select_page_1), + nt35510_mauc_select_page_1); + if (ret) + return ret; + + ret = nt35510_setup_power(nt); + if (ret) + return ret; + + ret = nt35510_send_long(nt, dsi, NT35510_P1_SET_GAMMA_RED_POS, + NT35510_P1_GAMMA_LEN, + nt->conf->gamma_corr_pos_r); + if (ret) + return ret; + ret = nt35510_send_long(nt, dsi, NT35510_P1_SET_GAMMA_GREEN_POS, + NT35510_P1_GAMMA_LEN, + nt->conf->gamma_corr_pos_g); + if (ret) + return ret; + ret = nt35510_send_long(nt, dsi, NT35510_P1_SET_GAMMA_BLUE_POS, + NT35510_P1_GAMMA_LEN, + nt->conf->gamma_corr_pos_b); + if (ret) + return ret; + ret = nt35510_send_long(nt, dsi, NT35510_P1_SET_GAMMA_RED_NEG, + NT35510_P1_GAMMA_LEN, + nt->conf->gamma_corr_neg_r); + if (ret) + return ret; + ret = nt35510_send_long(nt, dsi, NT35510_P1_SET_GAMMA_GREEN_NEG, + NT35510_P1_GAMMA_LEN, + nt->conf->gamma_corr_neg_g); + if (ret) + return ret; + ret = nt35510_send_long(nt, dsi, NT35510_P1_SET_GAMMA_BLUE_NEG, + NT35510_P1_GAMMA_LEN, + nt->conf->gamma_corr_neg_b); + if (ret) + return ret; + + /* Set up stuff in manufacturer control, page 0 */ + ret = nt35510_send_long(nt, dsi, MCS_CMD_MAUCCTR, + ARRAY_SIZE(nt35510_mauc_select_page_0), + nt35510_mauc_select_page_0); + if (ret) + return ret; + + ret = nt35510_setup_display(nt); + if (ret) + return ret; + + return 0; +} + +static int nt35510_power_off(struct nt35510 *nt) +{ + int ret; + + ret = regulator_bulk_disable(ARRAY_SIZE(nt->supplies), nt->supplies); + if (ret) + return ret; + + if (nt->reset_gpio) + gpiod_set_value(nt->reset_gpio, 1); + + return 0; +} + +static int nt35510_unprepare(struct drm_panel *panel) +{ + struct nt35510 *nt = panel_to_nt35510(panel); + struct mipi_dsi_device *dsi = to_mipi_dsi_device(nt->dev); + int ret; + + ret = mipi_dsi_dcs_set_display_off(dsi); + if (ret) { + DRM_DEV_ERROR(nt->dev, "failed to turn display off (%d)\n", + ret); + return ret; + } + usleep_range(10000, 20000); + + /* Enter sleep mode */ + ret = mipi_dsi_dcs_enter_sleep_mode(dsi); + if (ret) { + DRM_DEV_ERROR(nt->dev, "failed to enter sleep mode (%d)\n", + ret); + return ret; + } + + /* Wait 4 frames, how much is that 5ms in the vendor driver */ + usleep_range(5000, 10000); + + ret = nt35510_power_off(nt); + if (ret) + return ret; + + return 0; +} + +static int nt35510_prepare(struct drm_panel *panel) +{ + struct nt35510 *nt = panel_to_nt35510(panel); + struct mipi_dsi_device *dsi = to_mipi_dsi_device(nt->dev); + int ret; + + ret = nt35510_power_on(nt); + if (ret) + return ret; + + /* Exit sleep mode */ + ret = mipi_dsi_dcs_exit_sleep_mode(dsi); + if (ret) { + DRM_DEV_ERROR(nt->dev, "failed to exit sleep mode (%d)\n", + ret); + return ret; + } + /* Up to 120 ms */ + usleep_range(120000, 150000); + + ret = mipi_dsi_dcs_set_display_on(dsi); + if (ret) { + DRM_DEV_ERROR(nt->dev, "failed to turn display on (%d)\n", + ret); + return ret; + } + /* Some 10 ms */ + usleep_range(10000, 20000); + + return 0; +} + +static int nt35510_get_modes(struct drm_panel *panel, + struct drm_connector *connector) +{ + struct nt35510 *nt = panel_to_nt35510(panel); + struct drm_display_mode *mode; + struct drm_display_info *info; + + info = &connector->display_info; + info->width_mm = nt->conf->width_mm; + info->height_mm = nt->conf->height_mm; + mode = drm_mode_duplicate(connector->dev, &nt->conf->mode); + if (!mode) { + DRM_ERROR("bad mode or failed to add mode\n"); + return -EINVAL; + } + drm_mode_set_name(mode); + mode->type = DRM_MODE_TYPE_DRIVER | DRM_MODE_TYPE_PREFERRED; + + mode->width_mm = nt->conf->width_mm; + mode->height_mm = nt->conf->height_mm; + drm_mode_probed_add(connector, mode); + + return 1; /* Number of modes */ +} + +static const struct drm_panel_funcs nt35510_drm_funcs = { + .unprepare = nt35510_unprepare, + .prepare = nt35510_prepare, + .get_modes = nt35510_get_modes, +}; + +static int nt35510_probe(struct mipi_dsi_device *dsi) +{ + struct device *dev = &dsi->dev; + struct nt35510 *nt; + int ret; + + nt = devm_kzalloc(dev, sizeof(struct nt35510), GFP_KERNEL); + if (!nt) + return -ENOMEM; + mipi_dsi_set_drvdata(dsi, nt); + nt->dev = dev; + + dsi->lanes = 2; + dsi->format = MIPI_DSI_FMT_RGB888; + /* + * Datasheet suggests max HS rate for NT35510 is 250 MHz + * (period time 4ns, see figure 7.6.4 page 365) and max LP rate is + * 20 MHz (period time 50ns, see figure 7.6.6. page 366). + * However these frequencies appear in source code for the Hydis + * HVA40WV1 panel and setting up the LP frequency makes the panel + * not work. + * + * TODO: if other panels prove to be closer to the datasheet, + * maybe make this a per-panel config in struct nt35510_config? + */ + dsi->hs_rate = 349440000; + dsi->lp_rate = 9600000; + dsi->mode_flags = MIPI_DSI_CLOCK_NON_CONTINUOUS | + MIPI_DSI_MODE_EOT_PACKET; + + /* + * Every new incarnation of this display must have a unique + * data entry for the system in this driver. + */ + nt->conf = of_device_get_match_data(dev); + if (!nt->conf) { + dev_err(dev, "missing device configuration\n"); + return -ENODEV; + } + + nt->supplies[0].supply = "vdd"; /* 2.3-4.8 V */ + nt->supplies[1].supply = "vddi"; /* 1.65-3.3V */ + ret = devm_regulator_bulk_get(dev, ARRAY_SIZE(nt->supplies), + nt->supplies); + if (ret < 0) + return ret; + ret = regulator_set_voltage(nt->supplies[0].consumer, + 2300000, 4800000); + if (ret) + return ret; + ret = regulator_set_voltage(nt->supplies[1].consumer, + 1650000, 3300000); + if (ret) + return ret; + + nt->reset_gpio = devm_gpiod_get_optional(dev, "reset", GPIOD_ASIS); + if (IS_ERR(nt->reset_gpio)) { + dev_err(dev, "error getting RESET GPIO\n"); + return PTR_ERR(nt->reset_gpio); + } + + drm_panel_init(&nt->panel, dev, &nt35510_drm_funcs, + DRM_MODE_CONNECTOR_DSI); + + /* + * First, try to locate an external backlight (such as on GPIO) + * if this fails, assume we will want to use the internal backlight + * control. + */ + ret = drm_panel_of_backlight(&nt->panel); + if (ret) { + dev_err(dev, "error getting external backlight %d\n", ret); + return ret; + } + if (!nt->panel.backlight) { + struct backlight_device *bl; + + bl = devm_backlight_device_register(dev, "nt35510", dev, nt, + &nt35510_bl_ops, NULL); + if (IS_ERR(bl)) { + DRM_DEV_ERROR(dev, "failed to register backlight device\n"); + return PTR_ERR(bl); + } + bl->props.max_brightness = 255; + bl->props.brightness = 255; + bl->props.power = FB_BLANK_POWERDOWN; + nt->panel.backlight = bl; + } + + ret = drm_panel_add(&nt->panel); + if (ret < 0) + return ret; + + ret = mipi_dsi_attach(dsi); + if (ret < 0) + drm_panel_remove(&nt->panel); + + return 0; +} + +static int nt35510_remove(struct mipi_dsi_device *dsi) +{ + struct nt35510 *nt = mipi_dsi_get_drvdata(dsi); + int ret; + + mipi_dsi_detach(dsi); + /* Power off */ + ret = nt35510_power_off(nt); + drm_panel_remove(&nt->panel); + + return ret; +} + +/* + * These gamma correction values are 10bit tuples, so only bits 0 and 1 is + * ever used in the first byte. They form a positive and negative gamma + * correction curve for each color, values must be strictly higher for each + * step on the curve. As can be seen these default curves goes from 0x0001 + * to 0x03FE. + */ +#define NT35510_GAMMA_POS_DEFAULT 0x00, 0x01, 0x00, 0x43, 0x00, \ + 0x6B, 0x00, 0x87, 0x00, 0xA3, 0x00, 0xCE, 0x00, 0xF1, 0x01, \ + 0x27, 0x01, 0x53, 0x01, 0x98, 0x01, 0xCE, 0x02, 0x22, 0x02, \ + 0x83, 0x02, 0x78, 0x02, 0x9E, 0x02, 0xDD, 0x03, 0x00, 0x03, \ + 0x2E, 0x03, 0x54, 0x03, 0x7F, 0x03, 0x95, 0x03, 0xB3, 0x03, \ + 0xC2, 0x03, 0xE1, 0x03, 0xF1, 0x03, 0xFE + +#define NT35510_GAMMA_NEG_DEFAULT 0x00, 0x01, 0x00, 0x43, 0x00, \ + 0x6B, 0x00, 0x87, 0x00, 0xA3, 0x00, 0xCE, 0x00, 0xF1, 0x01, \ + 0x27, 0x01, 0x53, 0x01, 0x98, 0x01, 0xCE, 0x02, 0x22, 0x02, \ + 0x43, 0x02, 0x50, 0x02, 0x9E, 0x02, 0xDD, 0x03, 0x00, 0x03, \ + 0x2E, 0x03, 0x54, 0x03, 0x7F, 0x03, 0x95, 0x03, 0xB3, 0x03, \ + 0xC2, 0x03, 0xE1, 0x03, 0xF1, 0x03, 0xFE + +/* + * The Hydis HVA40WV1 panel + */ +static const struct nt35510_config nt35510_hydis_hva40wv1 = { + .width_mm = 52, + .height_mm = 86, + /** + * As the Hydis panel is used in command mode, the porches etc + * are settings programmed internally into the NT35510 controller + * and generated toward the physical display. As the panel is not + * used in video mode, these are not really exposed to the DSI + * host. + * + * Display frame rate control: + * Frame rate = (20 MHz / 1) / (389 * (7 + 50 + 800)) ~= 60 Hz + */ + .mode = { + /* The internal pixel clock of the NT35510 is 20 MHz */ + .clock = 20000, + .hdisplay = 480, + .hsync_start = 480 + 2, /* HFP = 2 */ + .hsync_end = 480 + 2 + 0, /* HSync = 0 */ + .htotal = 480 + 2 + 0 + 5, /* HFP = 5 */ + .vdisplay = 800, + .vsync_start = 800 + 2, /* VFP = 2 */ + .vsync_end = 800 + 2 + 0, /* VSync = 0 */ + .vtotal = 800 + 2 + 0 + 5, /* VBP = 5 */ + .vrefresh = 60, /* Calculated */ + .flags = 0, + }, + /* 0x09: AVDD = 5.6V */ + .avdd = { 0x09, 0x09, 0x09 }, + /* 0x34: PCK = Hsync/2, BTP = 2 x VDDB */ + .bt1ctr = { 0x34, 0x34, 0x34 }, + /* 0x09: AVEE = -5.6V */ + .avee = { 0x09, 0x09, 0x09 }, + /* 0x24: NCK = Hsync/2, BTN = -2 x VDDB */ + .bt2ctr = { 0x24, 0x24, 0x24 }, + /* 0x05 = 12V */ + .vgh = { 0x05, 0x05, 0x05 }, + /* 0x24: NCKA = Hsync/2, VGH = 2 x AVDD - AVEE */ + .bt4ctr = { 0x24, 0x24, 0x24 }, + /* 0x0B = -13V */ + .vgl = { 0x0B, 0x0B, 0x0B }, + /* 0x24: LCKA = Hsync, VGL = AVDD + VCL - AVDD */ + .bt5ctr = { 0x24, 0x24, 0x24 }, + /* VGMP: 0x0A3 = 5.0375V, VGSP = 0V */ + .vgp = { 0x00, 0xA3, 0x00 }, + /* VGMP: 0x0A3 = 5.0375V, VGSP = 0V */ + .vgn = { 0x00, 0xA3, 0x00 }, + /* SDEQCTR: source driver EQ mode 2, 2.5 us rise time on each step */ + .sdeqctr = { 0x01, 0x05, 0x05, 0x05 }, + /* SDVPCTR: Normal operation off color during v porch */ + .sdvpctr = 0x01, + /* T1: number of pixel clocks on one scanline: 0x184 = 389 clocks */ + .t1 = 0x0184, + /* VBP: vertical back porch toward the panel */ + .vbp = 7, + /* VFP: vertical front porch toward the panel */ + .vfp = 50, + /* PSEL: divide pixel clock 20MHz with 1 (no clock downscaling) */ + .psel = 0, + /* DPTMCTR12: 0x03: LVGL = VGLX, overlap mode, swap R->L O->E */ + .dpmctr12 = { 0x03, 0x00, 0x00, }, + /* Default gamma correction values */ + .gamma_corr_pos_r = { NT35510_GAMMA_POS_DEFAULT }, + .gamma_corr_pos_g = { NT35510_GAMMA_POS_DEFAULT }, + .gamma_corr_pos_b = { NT35510_GAMMA_POS_DEFAULT }, + .gamma_corr_neg_r = { NT35510_GAMMA_NEG_DEFAULT }, + .gamma_corr_neg_g = { NT35510_GAMMA_NEG_DEFAULT }, + .gamma_corr_neg_b = { NT35510_GAMMA_NEG_DEFAULT }, +}; + +static const struct of_device_id nt35510_of_match[] = { + { + .compatible = "hydis,hva40wv1", + .data = &nt35510_hydis_hva40wv1, + }, + { } +}; +MODULE_DEVICE_TABLE(of, nt35510_of_match); + +static struct mipi_dsi_driver nt35510_driver = { + .probe = nt35510_probe, + .remove = nt35510_remove, + .driver = { + .name = "panel-novatek-nt35510", + .of_match_table = nt35510_of_match, + }, +}; +module_mipi_dsi_driver(nt35510_driver); + +MODULE_AUTHOR("Linus Walleij <linus.walleij@linaro.org>"); +MODULE_DESCRIPTION("NT35510-based panel driver"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/gpu/drm/panel/panel-samsung-ld9040.c b/drivers/gpu/drm/panel/panel-samsung-ld9040.c index 3c52f15f7a1c..9bb2e8c7934a 100644 --- a/drivers/gpu/drm/panel/panel-samsung-ld9040.c +++ b/drivers/gpu/drm/panel/panel-samsung-ld9040.c @@ -373,6 +373,12 @@ static const struct of_device_id ld9040_of_match[] = { }; MODULE_DEVICE_TABLE(of, ld9040_of_match); +static const struct spi_device_id ld9040_ids[] = { + { "ld9040", }, + { /* sentinel */ } +}; +MODULE_DEVICE_TABLE(spi, ld9040_ids); + static struct spi_driver ld9040_driver = { .probe = ld9040_probe, .remove = ld9040_remove, diff --git a/drivers/gpu/drm/panel/panel-simple.c b/drivers/gpu/drm/panel/panel-simple.c index d901ccb5cf9b..3ad828eaefe1 100644 --- a/drivers/gpu/drm/panel/panel-simple.c +++ b/drivers/gpu/drm/panel/panel-simple.c @@ -351,6 +351,54 @@ static const struct drm_panel_funcs panel_simple_funcs = { .get_timings = panel_simple_get_timings, }; +static struct panel_desc panel_dpi; + +static int panel_dpi_probe(struct device *dev, + struct panel_simple *panel) +{ + struct display_timing *timing; + const struct device_node *np; + struct panel_desc *desc; + unsigned int bus_flags; + struct videomode vm; + int ret; + + np = dev->of_node; + desc = devm_kzalloc(dev, sizeof(*desc), GFP_KERNEL); + if (!desc) + return -ENOMEM; + + timing = devm_kzalloc(dev, sizeof(*timing), GFP_KERNEL); + if (!timing) + return -ENOMEM; + + ret = of_get_display_timing(np, "panel-timing", timing); + if (ret < 0) { + dev_err(dev, "%pOF: no panel-timing node found for \"panel-dpi\" binding\n", + np); + return ret; + } + + desc->timings = timing; + desc->num_timings = 1; + + of_property_read_u32(np, "width-mm", &desc->size.width); + of_property_read_u32(np, "height-mm", &desc->size.height); + + /* Extract bus_flags from display_timing */ + bus_flags = 0; + vm.flags = timing->flags; + drm_bus_flags_from_videomode(&vm, &bus_flags); + desc->bus_flags = bus_flags; + + /* We do not know the connector for the DT node, so guess it */ + desc->connector_type = DRM_MODE_CONNECTOR_DPI; + + panel->desc = desc; + + return 0; +} + #define PANEL_SIMPLE_BOUNDS_CHECK(to_check, bounds, field) \ (to_check->field.typ >= bounds->field.min && \ to_check->field.typ <= bounds->field.max) @@ -437,8 +485,15 @@ static int panel_simple_probe(struct device *dev, const struct panel_desc *desc) return -EPROBE_DEFER; } - if (!of_get_display_timing(dev->of_node, "panel-timing", &dt)) - panel_simple_parse_panel_timing_node(dev, panel, &dt); + if (desc == &panel_dpi) { + /* Handle the generic panel-dpi binding */ + err = panel_dpi_probe(dev, panel); + if (err) + goto free_ddc; + } else { + if (!of_get_display_timing(dev->of_node, "panel-timing", &dt)) + panel_simple_parse_panel_timing_node(dev, panel, &dt); + } drm_panel_init(&panel->base, dev, &panel_simple_funcs, desc->connector_type); @@ -2211,7 +2266,7 @@ static const struct drm_display_mode mitsubishi_aa070mc01_mode = { }; static const struct drm_display_mode logicpd_type_28_mode = { - .clock = 9000, + .clock = 9107, .hdisplay = 480, .hsync_start = 480 + 3, .hsync_end = 480 + 3 + 42, @@ -2340,6 +2395,51 @@ static const struct panel_desc netron_dy_e231732 = { .bus_format = MEDIA_BUS_FMT_RGB666_1X18, }; +static const struct drm_display_mode neweast_wjfh116008a_modes[] = { + { + .clock = 138500, + .hdisplay = 1920, + .hsync_start = 1920 + 48, + .hsync_end = 1920 + 48 + 32, + .htotal = 1920 + 48 + 32 + 80, + .vdisplay = 1080, + .vsync_start = 1080 + 3, + .vsync_end = 1080 + 3 + 5, + .vtotal = 1080 + 3 + 5 + 23, + .vrefresh = 60, + .flags = DRM_MODE_FLAG_NVSYNC | DRM_MODE_FLAG_NHSYNC, + }, { + .clock = 110920, + .hdisplay = 1920, + .hsync_start = 1920 + 48, + .hsync_end = 1920 + 48 + 32, + .htotal = 1920 + 48 + 32 + 80, + .vdisplay = 1080, + .vsync_start = 1080 + 3, + .vsync_end = 1080 + 3 + 5, + .vtotal = 1080 + 3 + 5 + 23, + .vrefresh = 48, + .flags = DRM_MODE_FLAG_NVSYNC | DRM_MODE_FLAG_NHSYNC, + } +}; + +static const struct panel_desc neweast_wjfh116008a = { + .modes = neweast_wjfh116008a_modes, + .num_modes = 2, + .bpc = 6, + .size = { + .width = 260, + .height = 150, + }, + .delay = { + .prepare = 110, + .enable = 20, + .unprepare = 500, + }, + .bus_format = MEDIA_BUS_FMT_RGB666_1X18, + .connector_type = DRM_MODE_CONNECTOR_eDP, +}; + static const struct drm_display_mode newhaven_nhd_43_480272ef_atxl_mode = { .clock = 9000, .hdisplay = 480, @@ -2506,15 +2606,15 @@ static const struct panel_desc ontat_yx700wv03 = { }; static const struct drm_display_mode ortustech_com37h3m_mode = { - .clock = 22153, + .clock = 22230, .hdisplay = 480, - .hsync_start = 480 + 8, - .hsync_end = 480 + 8 + 10, - .htotal = 480 + 8 + 10 + 10, + .hsync_start = 480 + 40, + .hsync_end = 480 + 40 + 10, + .htotal = 480 + 40 + 10 + 40, .vdisplay = 640, .vsync_start = 640 + 4, - .vsync_end = 640 + 4 + 3, - .vtotal = 640 + 4 + 3 + 4, + .vsync_end = 640 + 4 + 2, + .vtotal = 640 + 4 + 2 + 4, .vrefresh = 60, .flags = DRM_MODE_FLAG_NVSYNC | DRM_MODE_FLAG_NHSYNC, }; @@ -2555,6 +2655,7 @@ static const struct panel_desc ortustech_com43h4m85ulc = { }, .bus_format = MEDIA_BUS_FMT_RGB888_1X24, .bus_flags = DRM_BUS_FLAG_DE_HIGH | DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE, + .connector_type = DRM_MODE_CONNECTOR_DPI, }; static const struct drm_display_mode osddisplays_osd070t1718_19ts_mode = { @@ -2580,7 +2681,8 @@ static const struct panel_desc osddisplays_osd070t1718_19ts = { .height = 91, }, .bus_format = MEDIA_BUS_FMT_RGB888_1X24, - .bus_flags = DRM_BUS_FLAG_DE_HIGH | DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE, + .bus_flags = DRM_BUS_FLAG_DE_HIGH | DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE | + DRM_BUS_FLAG_SYNC_DRIVE_POSEDGE, .connector_type = DRM_MODE_CONNECTOR_DPI, }; @@ -2662,6 +2764,35 @@ static const struct panel_desc rocktech_rk070er9427 = { .bus_format = MEDIA_BUS_FMT_RGB666_1X18, }; +static const struct drm_display_mode rocktech_rk101ii01d_ct_mode = { + .clock = 71100, + .hdisplay = 1280, + .hsync_start = 1280 + 48, + .hsync_end = 1280 + 48 + 32, + .htotal = 1280 + 48 + 32 + 80, + .vdisplay = 800, + .vsync_start = 800 + 2, + .vsync_end = 800 + 2 + 5, + .vtotal = 800 + 2 + 5 + 16, + .vrefresh = 60, +}; + +static const struct panel_desc rocktech_rk101ii01d_ct = { + .modes = &rocktech_rk101ii01d_ct_mode, + .num_modes = 1, + .size = { + .width = 217, + .height = 136, + }, + .delay = { + .prepare = 50, + .disable = 50, + }, + .bus_flags = DRM_BUS_FLAG_DE_HIGH, + .bus_format = MEDIA_BUS_FMT_RGB888_1X7X4_SPWG, + .connector_type = DRM_MODE_CONNECTOR_LVDS, +}; + static const struct drm_display_mode samsung_lsn122dl01_c01_mode = { .clock = 271560, .hdisplay = 2560, @@ -2884,30 +3015,6 @@ static const struct panel_desc sharp_lq123p1jx31 = { }, }; -static const struct drm_display_mode sharp_lq150x1lg11_mode = { - .clock = 71100, - .hdisplay = 1024, - .hsync_start = 1024 + 168, - .hsync_end = 1024 + 168 + 64, - .htotal = 1024 + 168 + 64 + 88, - .vdisplay = 768, - .vsync_start = 768 + 37, - .vsync_end = 768 + 37 + 2, - .vtotal = 768 + 37 + 2 + 8, - .vrefresh = 60, -}; - -static const struct panel_desc sharp_lq150x1lg11 = { - .modes = &sharp_lq150x1lg11_mode, - .num_modes = 1, - .bpc = 6, - .size = { - .width = 304, - .height = 228, - }, - .bus_format = MEDIA_BUS_FMT_RGB565_1X16, -}; - static const struct display_timing sharp_ls020b1dd01d_timing = { .pixelclock = { 2000000, 4200000, 5000000 }, .hactive = { 240, 240, 240 }, @@ -3531,6 +3638,9 @@ static const struct of_device_id platform_of_match[] = { .compatible = "netron-dy,e231732", .data = &netron_dy_e231732, }, { + .compatible = "neweast,wjfh116008a", + .data = &neweast_wjfh116008a, + }, { .compatible = "newhaven,nhd-4.3-480272ef-atxl", .data = &newhaven_nhd_43_480272ef_atxl, }, { @@ -3570,6 +3680,9 @@ static const struct of_device_id platform_of_match[] = { .compatible = "rocktech,rk070er9427", .data = &rocktech_rk070er9427, }, { + .compatible = "rocktech,rk101ii01d-ct", + .data = &rocktech_rk101ii01d_ct, + }, { .compatible = "samsung,lsn122dl01-c01", .data = &samsung_lsn122dl01_c01, }, { @@ -3597,9 +3710,6 @@ static const struct of_device_id platform_of_match[] = { .compatible = "sharp,lq123p1jx31", .data = &sharp_lq123p1jx31, }, { - .compatible = "sharp,lq150x1lg11", - .data = &sharp_lq150x1lg11, - }, { .compatible = "sharp,ls020b1dd01d", .data = &sharp_ls020b1dd01d, }, { @@ -3657,6 +3767,10 @@ static const struct of_device_id platform_of_match[] = { .compatible = "winstar,wf35ltiacd", .data = &winstar_wf35ltiacd, }, { + /* Must be the last entry */ + .compatible = "panel-dpi", + .data = &panel_dpi, + }, { /* sentinel */ } }; diff --git a/drivers/gpu/drm/panel/panel-sony-acx424akp.c b/drivers/gpu/drm/panel/panel-sony-acx424akp.c index de0abf76ae6f..c91e55b2d7a3 100644 --- a/drivers/gpu/drm/panel/panel-sony-acx424akp.c +++ b/drivers/gpu/drm/panel/panel-sony-acx424akp.c @@ -48,7 +48,7 @@ struct acx424akp { }; static const struct drm_display_mode sony_acx424akp_vid_mode = { - .clock = 330000, + .clock = 27234, .hdisplay = 480, .hsync_start = 480 + 15, .hsync_end = 480 + 15 + 0, @@ -68,7 +68,7 @@ static const struct drm_display_mode sony_acx424akp_vid_mode = { * command mode using the maximum HS frequency. */ static const struct drm_display_mode sony_acx424akp_cmd_mode = { - .clock = 420160, + .clock = 35478, .hdisplay = 480, .hsync_start = 480 + 154, .hsync_end = 480 + 154 + 16, diff --git a/drivers/gpu/drm/panfrost/panfrost_device.c b/drivers/gpu/drm/panfrost/panfrost_device.c index 238fb6d54df4..8136babd3ba9 100644 --- a/drivers/gpu/drm/panfrost/panfrost_device.c +++ b/drivers/gpu/drm/panfrost/panfrost_device.c @@ -5,6 +5,7 @@ #include <linux/clk.h> #include <linux/reset.h> #include <linux/platform_device.h> +#include <linux/pm_domain.h> #include <linux/regulator/consumer.h> #include "panfrost_device.h" @@ -87,18 +88,27 @@ static void panfrost_clk_fini(struct panfrost_device *pfdev) static int panfrost_regulator_init(struct panfrost_device *pfdev) { - int ret; + int ret, i; - pfdev->regulator = devm_regulator_get(pfdev->dev, "mali"); - if (IS_ERR(pfdev->regulator)) { - ret = PTR_ERR(pfdev->regulator); - dev_err(pfdev->dev, "failed to get regulator: %d\n", ret); + if (WARN(pfdev->comp->num_supplies > ARRAY_SIZE(pfdev->regulators), + "Too many supplies in compatible structure.\n")) + return -EINVAL; + + for (i = 0; i < pfdev->comp->num_supplies; i++) + pfdev->regulators[i].supply = pfdev->comp->supply_names[i]; + + ret = devm_regulator_bulk_get(pfdev->dev, + pfdev->comp->num_supplies, + pfdev->regulators); + if (ret < 0) { + dev_err(pfdev->dev, "failed to get regulators: %d\n", ret); return ret; } - ret = regulator_enable(pfdev->regulator); + ret = regulator_bulk_enable(pfdev->comp->num_supplies, + pfdev->regulators); if (ret < 0) { - dev_err(pfdev->dev, "failed to enable regulator: %d\n", ret); + dev_err(pfdev->dev, "failed to enable regulators: %d\n", ret); return ret; } @@ -107,7 +117,81 @@ static int panfrost_regulator_init(struct panfrost_device *pfdev) static void panfrost_regulator_fini(struct panfrost_device *pfdev) { - regulator_disable(pfdev->regulator); + regulator_bulk_disable(pfdev->comp->num_supplies, + pfdev->regulators); +} + +static void panfrost_pm_domain_fini(struct panfrost_device *pfdev) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(pfdev->pm_domain_devs); i++) { + if (!pfdev->pm_domain_devs[i]) + break; + + if (pfdev->pm_domain_links[i]) + device_link_del(pfdev->pm_domain_links[i]); + + dev_pm_domain_detach(pfdev->pm_domain_devs[i], true); + } +} + +static int panfrost_pm_domain_init(struct panfrost_device *pfdev) +{ + int err; + int i, num_domains; + + num_domains = of_count_phandle_with_args(pfdev->dev->of_node, + "power-domains", + "#power-domain-cells"); + + /* + * Single domain is handled by the core, and, if only a single power + * the power domain is requested, the property is optional. + */ + if (num_domains < 2 && pfdev->comp->num_pm_domains < 2) + return 0; + + if (num_domains != pfdev->comp->num_pm_domains) { + dev_err(pfdev->dev, + "Incorrect number of power domains: %d provided, %d needed\n", + num_domains, pfdev->comp->num_pm_domains); + return -EINVAL; + } + + if (WARN(num_domains > ARRAY_SIZE(pfdev->pm_domain_devs), + "Too many supplies in compatible structure.\n")) + return -EINVAL; + + for (i = 0; i < num_domains; i++) { + pfdev->pm_domain_devs[i] = + dev_pm_domain_attach_by_name(pfdev->dev, + pfdev->comp->pm_domain_names[i]); + if (IS_ERR_OR_NULL(pfdev->pm_domain_devs[i])) { + err = PTR_ERR(pfdev->pm_domain_devs[i]) ? : -ENODATA; + pfdev->pm_domain_devs[i] = NULL; + dev_err(pfdev->dev, + "failed to get pm-domain %s(%d): %d\n", + pfdev->comp->pm_domain_names[i], i, err); + goto err; + } + + pfdev->pm_domain_links[i] = device_link_add(pfdev->dev, + pfdev->pm_domain_devs[i], DL_FLAG_PM_RUNTIME | + DL_FLAG_STATELESS | DL_FLAG_RPM_ACTIVE); + if (!pfdev->pm_domain_links[i]) { + dev_err(pfdev->pm_domain_devs[i], + "adding device link failed!\n"); + err = -ENODEV; + goto err; + } + } + + return 0; + +err: + panfrost_pm_domain_fini(pfdev); + return err; } int panfrost_device_init(struct panfrost_device *pfdev) @@ -140,37 +224,43 @@ int panfrost_device_init(struct panfrost_device *pfdev) goto err_out1; } + err = panfrost_pm_domain_init(pfdev); + if (err) + goto err_out2; + res = platform_get_resource(pfdev->pdev, IORESOURCE_MEM, 0); pfdev->iomem = devm_ioremap_resource(pfdev->dev, res); if (IS_ERR(pfdev->iomem)) { dev_err(pfdev->dev, "failed to ioremap iomem\n"); err = PTR_ERR(pfdev->iomem); - goto err_out2; + goto err_out3; } err = panfrost_gpu_init(pfdev); if (err) - goto err_out2; + goto err_out3; err = panfrost_mmu_init(pfdev); if (err) - goto err_out3; + goto err_out4; err = panfrost_job_init(pfdev); if (err) - goto err_out4; + goto err_out5; err = panfrost_perfcnt_init(pfdev); if (err) - goto err_out5; + goto err_out6; return 0; -err_out5: +err_out6: panfrost_job_fini(pfdev); -err_out4: +err_out5: panfrost_mmu_fini(pfdev); -err_out3: +err_out4: panfrost_gpu_fini(pfdev); +err_out3: + panfrost_pm_domain_fini(pfdev); err_out2: panfrost_reset_fini(pfdev); err_out1: @@ -186,6 +276,7 @@ void panfrost_device_fini(struct panfrost_device *pfdev) panfrost_job_fini(pfdev); panfrost_mmu_fini(pfdev); panfrost_gpu_fini(pfdev); + panfrost_pm_domain_fini(pfdev); panfrost_reset_fini(pfdev); panfrost_regulator_fini(pfdev); panfrost_clk_fini(pfdev); diff --git a/drivers/gpu/drm/panfrost/panfrost_device.h b/drivers/gpu/drm/panfrost/panfrost_device.h index 06713811b92c..c30c719a8059 100644 --- a/drivers/gpu/drm/panfrost/panfrost_device.h +++ b/drivers/gpu/drm/panfrost/panfrost_device.h @@ -7,6 +7,7 @@ #include <linux/atomic.h> #include <linux/io-pgtable.h> +#include <linux/regulator/consumer.h> #include <linux/spinlock.h> #include <drm/drm_device.h> #include <drm/drm_mm.h> @@ -19,6 +20,8 @@ struct panfrost_job; struct panfrost_perfcnt; #define NUM_JOB_SLOTS 3 +#define MAX_REGULATORS 2 +#define MAX_PM_DOMAINS 3 struct panfrost_features { u16 id; @@ -51,6 +54,23 @@ struct panfrost_features { unsigned long hw_issues[64 / BITS_PER_LONG]; }; +/* + * Features that cannot be automatically detected and need matching using the + * compatible string, typically SoC-specific. + */ +struct panfrost_compatible { + /* Supplies count and names. */ + int num_supplies; + const char * const *supply_names; + /* + * Number of power domains required, note that values 0 and 1 are + * handled identically, as only values > 1 need special handling. + */ + int num_pm_domains; + /* Only required if num_pm_domains > 1. */ + const char * const *pm_domain_names; +}; + struct panfrost_device { struct device *dev; struct drm_device *ddev; @@ -59,10 +79,14 @@ struct panfrost_device { void __iomem *iomem; struct clk *clock; struct clk *bus_clock; - struct regulator *regulator; + struct regulator_bulk_data regulators[MAX_REGULATORS]; struct reset_control *rstc; + /* pm_domains for devices with more than one. */ + struct device *pm_domain_devs[MAX_PM_DOMAINS]; + struct device_link *pm_domain_links[MAX_PM_DOMAINS]; struct panfrost_features features; + const struct panfrost_compatible *comp; spinlock_t as_lock; unsigned long as_in_use_mask; diff --git a/drivers/gpu/drm/panfrost/panfrost_drv.c b/drivers/gpu/drm/panfrost/panfrost_drv.c index b7a618db3ee2..882fecc33fdb 100644 --- a/drivers/gpu/drm/panfrost/panfrost_drv.c +++ b/drivers/gpu/drm/panfrost/panfrost_drv.c @@ -584,6 +584,10 @@ static int panfrost_probe(struct platform_device *pdev) platform_set_drvdata(pdev, pfdev); + pfdev->comp = of_device_get_match_data(&pdev->dev); + if (!pfdev->comp) + return -ENODEV; + /* Allocate and initialze the DRM device. */ ddev = drm_dev_alloc(&panfrost_drm_driver, &pdev->dev); if (IS_ERR(ddev)) @@ -655,16 +659,24 @@ static int panfrost_remove(struct platform_device *pdev) return 0; } +static const char * const default_supplies[] = { "mali" }; +static const struct panfrost_compatible default_data = { + .num_supplies = ARRAY_SIZE(default_supplies), + .supply_names = default_supplies, + .num_pm_domains = 1, /* optional */ + .pm_domain_names = NULL, +}; + static const struct of_device_id dt_match[] = { - { .compatible = "arm,mali-t604" }, - { .compatible = "arm,mali-t624" }, - { .compatible = "arm,mali-t628" }, - { .compatible = "arm,mali-t720" }, - { .compatible = "arm,mali-t760" }, - { .compatible = "arm,mali-t820" }, - { .compatible = "arm,mali-t830" }, - { .compatible = "arm,mali-t860" }, - { .compatible = "arm,mali-t880" }, + { .compatible = "arm,mali-t604", .data = &default_data, }, + { .compatible = "arm,mali-t624", .data = &default_data, }, + { .compatible = "arm,mali-t628", .data = &default_data, }, + { .compatible = "arm,mali-t720", .data = &default_data, }, + { .compatible = "arm,mali-t760", .data = &default_data, }, + { .compatible = "arm,mali-t820", .data = &default_data, }, + { .compatible = "arm,mali-t830", .data = &default_data, }, + { .compatible = "arm,mali-t860", .data = &default_data, }, + { .compatible = "arm,mali-t880", .data = &default_data, }, {} }; MODULE_DEVICE_TABLE(of, dt_match); diff --git a/drivers/gpu/drm/panfrost/panfrost_gpu.c b/drivers/gpu/drm/panfrost/panfrost_gpu.c index 1b9b79cd5804..f2c1ddc41a9b 100644 --- a/drivers/gpu/drm/panfrost/panfrost_gpu.c +++ b/drivers/gpu/drm/panfrost/panfrost_gpu.c @@ -308,17 +308,20 @@ void panfrost_gpu_power_on(struct panfrost_device *pfdev) gpu_write(pfdev, L2_PWRON_LO, pfdev->features.l2_present); ret = readl_relaxed_poll_timeout(pfdev->iomem + L2_READY_LO, val, val == pfdev->features.l2_present, 100, 1000); + if (ret) + dev_err(pfdev->dev, "error powering up gpu L2"); gpu_write(pfdev, SHADER_PWRON_LO, pfdev->features.shader_present); - ret |= readl_relaxed_poll_timeout(pfdev->iomem + SHADER_READY_LO, + ret = readl_relaxed_poll_timeout(pfdev->iomem + SHADER_READY_LO, val, val == pfdev->features.shader_present, 100, 1000); + if (ret) + dev_err(pfdev->dev, "error powering up gpu shader"); gpu_write(pfdev, TILER_PWRON_LO, pfdev->features.tiler_present); - ret |= readl_relaxed_poll_timeout(pfdev->iomem + TILER_READY_LO, + ret = readl_relaxed_poll_timeout(pfdev->iomem + TILER_READY_LO, val, val == pfdev->features.tiler_present, 100, 1000); - if (ret) - dev_err(pfdev->dev, "error powering up gpu"); + dev_err(pfdev->dev, "error powering up gpu tiler"); } void panfrost_gpu_power_off(struct panfrost_device *pfdev) diff --git a/drivers/gpu/drm/panfrost/panfrost_job.c b/drivers/gpu/drm/panfrost/panfrost_job.c index 8fd8726313ae..7914b1570841 100644 --- a/drivers/gpu/drm/panfrost/panfrost_job.c +++ b/drivers/gpu/drm/panfrost/panfrost_job.c @@ -280,12 +280,8 @@ static void panfrost_job_cleanup(struct kref *ref) } if (job->bos) { - struct panfrost_gem_object *bo; - - for (i = 0; i < job->bo_count; i++) { - bo = to_panfrost_bo(job->bos[i]); + for (i = 0; i < job->bo_count; i++) drm_gem_object_put_unlocked(job->bos[i]); - } kvfree(job->bos); } diff --git a/drivers/gpu/drm/panfrost/panfrost_mmu.c b/drivers/gpu/drm/panfrost/panfrost_mmu.c index 23314f41717b..ed28aeba6d59 100644 --- a/drivers/gpu/drm/panfrost/panfrost_mmu.c +++ b/drivers/gpu/drm/panfrost/panfrost_mmu.c @@ -151,7 +151,12 @@ u32 panfrost_mmu_as_get(struct panfrost_device *pfdev, struct panfrost_mmu *mmu) as = mmu->as; if (as >= 0) { int en = atomic_inc_return(&mmu->as_count); - WARN_ON(en >= NUM_JOB_SLOTS); + + /* + * AS can be retained by active jobs or a perfcnt context, + * hence the '+ 1' here. + */ + WARN_ON(en >= (NUM_JOB_SLOTS + 1)); list_move(&mmu->list, &pfdev->as_lru_list); goto out; @@ -596,33 +601,27 @@ static irqreturn_t panfrost_mmu_irq_handler_thread(int irq, void *data) source_id = (fault_status >> 16); /* Page fault only */ - if ((status & mask) == BIT(i)) { - WARN_ON(exception_type < 0xC1 || exception_type > 0xC4); - + ret = -1; + if ((status & mask) == BIT(i) && (exception_type & 0xF8) == 0xC0) ret = panfrost_mmu_map_fault_addr(pfdev, i, addr); - if (!ret) { - mmu_write(pfdev, MMU_INT_CLEAR, BIT(i)); - status &= ~mask; - continue; - } - } - /* terminal fault, print info about the fault */ - dev_err(pfdev->dev, - "Unhandled Page fault in AS%d at VA 0x%016llX\n" - "Reason: %s\n" - "raw fault status: 0x%X\n" - "decoded fault status: %s\n" - "exception type 0x%X: %s\n" - "access type 0x%X: %s\n" - "source id 0x%X\n", - i, addr, - "TODO", - fault_status, - (fault_status & (1 << 10) ? "DECODER FAULT" : "SLAVE FAULT"), - exception_type, panfrost_exception_name(pfdev, exception_type), - access_type, access_type_name(pfdev, fault_status), - source_id); + if (ret) + /* terminal fault, print info about the fault */ + dev_err(pfdev->dev, + "Unhandled Page fault in AS%d at VA 0x%016llX\n" + "Reason: %s\n" + "raw fault status: 0x%X\n" + "decoded fault status: %s\n" + "exception type 0x%X: %s\n" + "access type 0x%X: %s\n" + "source id 0x%X\n", + i, addr, + "TODO", + fault_status, + (fault_status & (1 << 10) ? "DECODER FAULT" : "SLAVE FAULT"), + exception_type, panfrost_exception_name(pfdev, exception_type), + access_type, access_type_name(pfdev, fault_status), + source_id); mmu_write(pfdev, MMU_INT_CLEAR, mask); diff --git a/drivers/gpu/drm/panfrost/panfrost_perfcnt.c b/drivers/gpu/drm/panfrost/panfrost_perfcnt.c index 684820448be3..6913578d5aa7 100644 --- a/drivers/gpu/drm/panfrost/panfrost_perfcnt.c +++ b/drivers/gpu/drm/panfrost/panfrost_perfcnt.c @@ -73,7 +73,7 @@ static int panfrost_perfcnt_enable_locked(struct panfrost_device *pfdev, struct panfrost_file_priv *user = file_priv->driver_priv; struct panfrost_perfcnt *perfcnt = pfdev->perfcnt; struct drm_gem_shmem_object *bo; - u32 cfg; + u32 cfg, as; int ret; if (user == perfcnt->user) @@ -126,12 +126,8 @@ static int panfrost_perfcnt_enable_locked(struct panfrost_device *pfdev, perfcnt->user = user; - /* - * Always use address space 0 for now. - * FIXME: this needs to be updated when we start using different - * address space. - */ - cfg = GPU_PERFCNT_CFG_AS(0) | + as = panfrost_mmu_as_get(pfdev, perfcnt->mapping->mmu); + cfg = GPU_PERFCNT_CFG_AS(as) | GPU_PERFCNT_CFG_MODE(GPU_PERFCNT_CFG_MODE_MANUAL); /* @@ -195,6 +191,7 @@ static int panfrost_perfcnt_disable_locked(struct panfrost_device *pfdev, drm_gem_shmem_vunmap(&perfcnt->mapping->obj->base.base, perfcnt->buf); perfcnt->buf = NULL; panfrost_gem_close(&perfcnt->mapping->obj->base.base, file_priv); + panfrost_mmu_as_put(pfdev, perfcnt->mapping->mmu); panfrost_gem_mapping_put(perfcnt->mapping); perfcnt->mapping = NULL; pm_runtime_mark_last_busy(pfdev->dev); diff --git a/drivers/gpu/drm/pl111/pl111_versatile.c b/drivers/gpu/drm/pl111/pl111_versatile.c index 09aeaffb7660..4f325c410b5d 100644 --- a/drivers/gpu/drm/pl111/pl111_versatile.c +++ b/drivers/gpu/drm/pl111/pl111_versatile.c @@ -19,6 +19,7 @@ static struct regmap *versatile_syscon_map; * We detect the different syscon types from the compatible strings. */ enum versatile_clcd { + INTEGRATOR_IMPD1, INTEGRATOR_CLCD_CM, VERSATILE_CLCD, REALVIEW_CLCD_EB, @@ -65,6 +66,14 @@ static const struct of_device_id versatile_clcd_of_match[] = { {}, }; +static const struct of_device_id impd1_clcd_of_match[] = { + { + .compatible = "arm,im-pd1-syscon", + .data = (void *)INTEGRATOR_IMPD1, + }, + {}, +}; + /* * Core module CLCD control on the Integrator/CP, bits * 8 thru 19 of the CM_CONTROL register controls a bunch @@ -125,6 +134,36 @@ static void pl111_integrator_enable(struct drm_device *drm, u32 format) val); } +#define IMPD1_CTRL_OFFSET 0x18 +#define IMPD1_CTRL_DISP_LCD (0 << 0) +#define IMPD1_CTRL_DISP_VGA (1 << 0) +#define IMPD1_CTRL_DISP_LCD1 (2 << 0) +#define IMPD1_CTRL_DISP_ENABLE (1 << 2) +#define IMPD1_CTRL_DISP_MASK (7 << 0) + +static void pl111_impd1_enable(struct drm_device *drm, u32 format) +{ + u32 val; + + dev_info(drm->dev, "enable IM-PD1 CLCD connectors\n"); + val = IMPD1_CTRL_DISP_VGA | IMPD1_CTRL_DISP_ENABLE; + + regmap_update_bits(versatile_syscon_map, + IMPD1_CTRL_OFFSET, + IMPD1_CTRL_DISP_MASK, + val); +} + +static void pl111_impd1_disable(struct drm_device *drm) +{ + dev_info(drm->dev, "disable IM-PD1 CLCD connectors\n"); + + regmap_update_bits(versatile_syscon_map, + IMPD1_CTRL_OFFSET, + IMPD1_CTRL_DISP_MASK, + 0); +} + /* * This configuration register in the Versatile and RealView * family is uniformly present but appears more and more @@ -271,6 +310,20 @@ static const struct pl111_variant_data pl110_integrator = { }; /* + * The IM-PD1 variant is a PL110 with a bunch of broken, or not + * yet implemented features + */ +static const struct pl111_variant_data pl110_impd1 = { + .name = "PL110 IM-PD1", + .is_pl110 = true, + .broken_clockdivider = true, + .broken_vblank = true, + .formats = pl110_integrator_pixel_formats, + .nformats = ARRAY_SIZE(pl110_integrator_pixel_formats), + .fb_bpp = 16, +}; + +/* * This is the in-between PL110 variant found in the ARM Versatile, * supporting RGB565/BGR565 */ @@ -322,8 +375,21 @@ int pl111_versatile_init(struct device *dev, struct pl111_drm_dev_private *priv) /* Non-ARM reference designs, just bail out */ return 0; } + versatile_clcd_type = (enum versatile_clcd)clcd_id->data; + /* + * On the Integrator, check if we should use the IM-PD1 instead, + * if we find it, it will take precedence. This is on the Integrator/AP + * which only has this option for PL110 graphics. + */ + if (versatile_clcd_type == INTEGRATOR_CLCD_CM) { + np = of_find_matching_node_and_match(NULL, impd1_clcd_of_match, + &clcd_id); + if (np) + versatile_clcd_type = (enum versatile_clcd)clcd_id->data; + } + /* Versatile Express special handling */ if (versatile_clcd_type == VEXPRESS_CLCD_V2M) { struct platform_device *pdev; @@ -367,6 +433,13 @@ int pl111_versatile_init(struct device *dev, struct pl111_drm_dev_private *priv) priv->variant_display_enable = pl111_integrator_enable; dev_info(dev, "set up callbacks for Integrator PL110\n"); break; + case INTEGRATOR_IMPD1: + versatile_syscon_map = map; + priv->variant = &pl110_impd1; + priv->variant_display_enable = pl111_impd1_enable; + priv->variant_display_disable = pl111_impd1_disable; + dev_info(dev, "set up callbacks for IM-PD1 PL110\n"); + break; case VERSATILE_CLCD: versatile_syscon_map = map; /* This can do RGB565 with external PLD */ diff --git a/drivers/gpu/drm/qxl/qxl_cmd.c b/drivers/gpu/drm/qxl/qxl_cmd.c index ef09dc6bc635..d1086b2a6892 100644 --- a/drivers/gpu/drm/qxl/qxl_cmd.c +++ b/drivers/gpu/drm/qxl/qxl_cmd.c @@ -36,7 +36,7 @@ static int qxl_reap_surface_id(struct qxl_device *qdev, int max_to_reap); struct ring { struct qxl_ring_header header; - uint8_t elements[0]; + uint8_t elements[]; }; struct qxl_ring { diff --git a/drivers/gpu/drm/qxl/qxl_display.c b/drivers/gpu/drm/qxl/qxl_display.c index ab4f8dd00400..09583a08e141 100644 --- a/drivers/gpu/drm/qxl/qxl_display.c +++ b/drivers/gpu/drm/qxl/qxl_display.c @@ -31,6 +31,7 @@ #include <drm/drm_gem_framebuffer_helper.h> #include <drm/drm_plane_helper.h> #include <drm/drm_probe_helper.h> +#include <drm/drm_simple_kms_helper.h> #include "qxl_drv.h" #include "qxl_object.h" @@ -1007,9 +1008,6 @@ static struct drm_encoder *qxl_best_encoder(struct drm_connector *connector) return &qxl_output->enc; } -static const struct drm_encoder_helper_funcs qxl_enc_helper_funcs = { -}; - static const struct drm_connector_helper_funcs qxl_connector_helper_funcs = { .get_modes = qxl_conn_get_modes, .mode_valid = qxl_conn_mode_valid, @@ -1059,15 +1057,6 @@ static const struct drm_connector_funcs qxl_connector_funcs = { .atomic_destroy_state = drm_atomic_helper_connector_destroy_state, }; -static void qxl_enc_destroy(struct drm_encoder *encoder) -{ - drm_encoder_cleanup(encoder); -} - -static const struct drm_encoder_funcs qxl_enc_funcs = { - .destroy = qxl_enc_destroy, -}; - static int qxl_mode_create_hotplug_mode_update_property(struct qxl_device *qdev) { if (qdev->hotplug_mode_update_property) @@ -1086,6 +1075,7 @@ static int qdev_output_init(struct drm_device *dev, int num_output) struct qxl_output *qxl_output; struct drm_connector *connector; struct drm_encoder *encoder; + int ret; qxl_output = kzalloc(sizeof(struct qxl_output), GFP_KERNEL); if (!qxl_output) @@ -1098,15 +1088,19 @@ static int qdev_output_init(struct drm_device *dev, int num_output) drm_connector_init(dev, &qxl_output->base, &qxl_connector_funcs, DRM_MODE_CONNECTOR_VIRTUAL); - drm_encoder_init(dev, &qxl_output->enc, &qxl_enc_funcs, - DRM_MODE_ENCODER_VIRTUAL, NULL); + ret = drm_simple_encoder_init(dev, &qxl_output->enc, + DRM_MODE_ENCODER_VIRTUAL); + if (ret) { + drm_err(dev, "drm_simple_encoder_init() failed, error %d\n", + ret); + goto err_drm_connector_cleanup; + } /* we get HPD via client monitors config */ connector->polled = DRM_CONNECTOR_POLL_HPD; encoder->possible_crtcs = 1 << num_output; drm_connector_attach_encoder(&qxl_output->base, &qxl_output->enc); - drm_encoder_helper_add(encoder, &qxl_enc_helper_funcs); drm_connector_helper_add(connector, &qxl_connector_helper_funcs); drm_object_attach_property(&connector->base, @@ -1116,6 +1110,11 @@ static int qdev_output_init(struct drm_device *dev, int num_output) drm_object_attach_property(&connector->base, dev->mode_config.suggested_y_property, 0); return 0; + +err_drm_connector_cleanup: + drm_connector_cleanup(&qxl_output->base); + kfree(qxl_output); + return ret; } static struct drm_framebuffer * diff --git a/drivers/gpu/drm/qxl/qxl_drv.c b/drivers/gpu/drm/qxl/qxl_drv.c index 1d601f57a6ba..4fda3f9b29f4 100644 --- a/drivers/gpu/drm/qxl/qxl_drv.c +++ b/drivers/gpu/drm/qxl/qxl_drv.c @@ -34,6 +34,7 @@ #include <linux/pci.h> #include <drm/drm.h> +#include <drm/drm_atomic_helper.h> #include <drm/drm_drv.h> #include <drm/drm_file.h> #include <drm/drm_modeset_helper.h> @@ -132,21 +133,30 @@ free_dev: return ret; } +static void qxl_drm_release(struct drm_device *dev) +{ + struct qxl_device *qdev = dev->dev_private; + + /* + * TODO: qxl_device_fini() call should be in qxl_pci_remove(), + * reodering qxl_modeset_fini() + qxl_device_fini() calls is + * non-trivial though. + */ + qxl_modeset_fini(qdev); + qxl_device_fini(qdev); + dev->dev_private = NULL; + kfree(qdev); +} + static void qxl_pci_remove(struct pci_dev *pdev) { struct drm_device *dev = pci_get_drvdata(pdev); - struct qxl_device *qdev = dev->dev_private; drm_dev_unregister(dev); - - qxl_modeset_fini(qdev); - qxl_device_fini(qdev); + drm_atomic_helper_shutdown(dev); if (is_vga(pdev)) vga_put(pdev, VGA_RSRC_LEGACY_IO); - - dev->dev_private = NULL; - kfree(qdev); drm_dev_put(dev); } @@ -279,6 +289,8 @@ static struct drm_driver qxl_driver = { .major = 0, .minor = 1, .patchlevel = 0, + + .release = qxl_drm_release, }; static int __init qxl_init(void) diff --git a/drivers/gpu/drm/qxl/qxl_kms.c b/drivers/gpu/drm/qxl/qxl_kms.c index bfc1631093e9..70b20ee4741a 100644 --- a/drivers/gpu/drm/qxl/qxl_kms.c +++ b/drivers/gpu/drm/qxl/qxl_kms.c @@ -299,12 +299,12 @@ void qxl_device_fini(struct qxl_device *qdev) { qxl_bo_unref(&qdev->current_release_bo[0]); qxl_bo_unref(&qdev->current_release_bo[1]); + qxl_gem_fini(qdev); + qxl_bo_fini(qdev); flush_work(&qdev->gc_work); qxl_ring_free(qdev->command_ring); qxl_ring_free(qdev->cursor_ring); qxl_ring_free(qdev->release_ring); - qxl_gem_fini(qdev); - qxl_bo_fini(qdev); io_mapping_free(qdev->surface_mapping); io_mapping_free(qdev->vram_mapping); iounmap(qdev->ram_header); diff --git a/drivers/gpu/drm/radeon/.gitignore b/drivers/gpu/drm/radeon/.gitignore index 403eb3a5891f..9c1a94153983 100644 --- a/drivers/gpu/drm/radeon/.gitignore +++ b/drivers/gpu/drm/radeon/.gitignore @@ -1,3 +1,4 @@ +# SPDX-License-Identifier: GPL-2.0-only mkregtable *_reg_safe.h diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c index be583695427a..91811757104c 100644 --- a/drivers/gpu/drm/radeon/atombios_crtc.c +++ b/drivers/gpu/drm/radeon/atombios_crtc.c @@ -2231,6 +2231,7 @@ static const struct drm_crtc_helper_funcs atombios_helper_funcs = { .prepare = atombios_crtc_prepare, .commit = atombios_crtc_commit, .disable = atombios_crtc_disable, + .get_scanout_position = radeon_get_crtc_scanout_position, }; void radeon_atombios_init_crtc(struct drm_device *dev, diff --git a/drivers/gpu/drm/radeon/radeon_bios.c b/drivers/gpu/drm/radeon/radeon_bios.c index c42f73fad3e3..bb29cf02974d 100644 --- a/drivers/gpu/drm/radeon/radeon_bios.c +++ b/drivers/gpu/drm/radeon/radeon_bios.c @@ -108,25 +108,33 @@ static bool radeon_read_bios(struct radeon_device *rdev) static bool radeon_read_platform_bios(struct radeon_device *rdev) { - uint8_t __iomem *bios; - size_t size; + phys_addr_t rom = rdev->pdev->rom; + size_t romlen = rdev->pdev->romlen; + void __iomem *bios; rdev->bios = NULL; - bios = pci_platform_rom(rdev->pdev, &size); - if (!bios) { + if (!rom || romlen == 0) return false; - } - if (size == 0 || bios[0] != 0x55 || bios[1] != 0xaa) { + rdev->bios = kzalloc(romlen, GFP_KERNEL); + if (!rdev->bios) return false; - } - rdev->bios = kmemdup(bios, size, GFP_KERNEL); - if (rdev->bios == NULL) { - return false; - } + + bios = ioremap(rom, romlen); + if (!bios) + goto free_bios; + + memcpy_fromio(rdev->bios, bios, romlen); + iounmap(bios); + + if (rdev->bios[0] != 0x55 || rdev->bios[1] != 0xaa) + goto free_bios; return true; +free_bios: + kfree(rdev->bios); + return false; } #ifdef CONFIG_ACPI diff --git a/drivers/gpu/drm/radeon/radeon_display.c b/drivers/gpu/drm/radeon/radeon_display.c index d07c7db0c815..35db79a168bf 100644 --- a/drivers/gpu/drm/radeon/radeon_display.c +++ b/drivers/gpu/drm/radeon/radeon_display.c @@ -45,6 +45,10 @@ #include "atom.h" #include "radeon.h" +u32 radeon_get_vblank_counter_kms(struct drm_crtc *crtc); +int radeon_enable_vblank_kms(struct drm_crtc *crtc); +void radeon_disable_vblank_kms(struct drm_crtc *crtc); + static void avivo_crtc_load_lut(struct drm_crtc *crtc) { struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc); @@ -460,7 +464,7 @@ static void radeon_flip_work_func(struct work_struct *__work) (DRM_SCANOUTPOS_VALID | DRM_SCANOUTPOS_IN_VBLANK) && (!ASIC_IS_AVIVO(rdev) || ((int) (work->target_vblank - - dev->driver->get_vblank_counter(dev, work->crtc_id)) > 0))) + crtc->funcs->get_vblank_counter(crtc)) > 0))) usleep_range(1000, 2000); /* We borrow the event spin lock for protecting flip_status */ @@ -576,7 +580,7 @@ static int radeon_crtc_page_flip_target(struct drm_crtc *crtc, } work->base = base; work->target_vblank = target - (uint32_t)drm_crtc_vblank_count(crtc) + - dev->driver->get_vblank_counter(dev, work->crtc_id); + crtc->funcs->get_vblank_counter(crtc); /* We borrow the event spin lock for protecting flip_work */ spin_lock_irqsave(&crtc->dev->event_lock, flags); @@ -668,6 +672,10 @@ static const struct drm_crtc_funcs radeon_crtc_funcs = { .set_config = radeon_crtc_set_config, .destroy = radeon_crtc_destroy, .page_flip_target = radeon_crtc_page_flip_target, + .get_vblank_counter = radeon_get_vblank_counter_kms, + .enable_vblank = radeon_enable_vblank_kms, + .disable_vblank = radeon_disable_vblank_kms, + .get_vblank_timestamp = drm_crtc_vblank_helper_get_vblank_timestamp, }; static void radeon_crtc_init(struct drm_device *dev, int index) @@ -1973,3 +1981,16 @@ int radeon_get_crtc_scanoutpos(struct drm_device *dev, unsigned int pipe, return ret; } + +bool +radeon_get_crtc_scanout_position(struct drm_crtc *crtc, + bool in_vblank_irq, int *vpos, int *hpos, + ktime_t *stime, ktime_t *etime, + const struct drm_display_mode *mode) +{ + struct drm_device *dev = crtc->dev; + unsigned int pipe = crtc->index; + + return radeon_get_crtc_scanoutpos(dev, pipe, 0, vpos, hpos, + stime, etime, mode); +} diff --git a/drivers/gpu/drm/radeon/radeon_dp_mst.c b/drivers/gpu/drm/radeon/radeon_dp_mst.c index 28eef9282874..008308780443 100644 --- a/drivers/gpu/drm/radeon/radeon_dp_mst.c +++ b/drivers/gpu/drm/radeon/radeon_dp_mst.c @@ -301,35 +301,8 @@ static struct drm_connector *radeon_dp_add_mst_connector(struct drm_dp_mst_topol return connector; } -static void radeon_dp_register_mst_connector(struct drm_connector *connector) -{ - struct drm_device *dev = connector->dev; - struct radeon_device *rdev = dev->dev_private; - - radeon_fb_add_connector(rdev, connector); - - drm_connector_register(connector); -} - -static void radeon_dp_destroy_mst_connector(struct drm_dp_mst_topology_mgr *mgr, - struct drm_connector *connector) -{ - struct radeon_connector *master = container_of(mgr, struct radeon_connector, mst_mgr); - struct drm_device *dev = master->base.dev; - struct radeon_device *rdev = dev->dev_private; - - drm_connector_unregister(connector); - radeon_fb_remove_connector(rdev, connector); - drm_connector_cleanup(connector); - - kfree(connector); - DRM_DEBUG_KMS("\n"); -} - static const struct drm_dp_mst_topology_cbs mst_cbs = { .add_connector = radeon_dp_add_mst_connector, - .register_connector = radeon_dp_register_mst_connector, - .destroy_connector = radeon_dp_destroy_mst_connector, }; static struct diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c index fd74e2611185..59f8186a2415 100644 --- a/drivers/gpu/drm/radeon/radeon_drv.c +++ b/drivers/gpu/drm/radeon/radeon_drv.c @@ -37,6 +37,7 @@ #include <linux/vga_switcheroo.h> #include <linux/mmu_notifier.h> +#include <drm/drm_agpsupport.h> #include <drm/drm_crtc_helper.h> #include <drm/drm_drv.h> #include <drm/drm_fb_helper.h> @@ -119,9 +120,6 @@ void radeon_driver_postclose_kms(struct drm_device *dev, int radeon_suspend_kms(struct drm_device *dev, bool suspend, bool fbcon, bool freeze); int radeon_resume_kms(struct drm_device *dev, bool resume, bool fbcon); -u32 radeon_get_vblank_counter_kms(struct drm_device *dev, unsigned int pipe); -int radeon_enable_vblank_kms(struct drm_device *dev, unsigned int pipe); -void radeon_disable_vblank_kms(struct drm_device *dev, unsigned int pipe); void radeon_driver_irq_preinstall_kms(struct drm_device *dev); int radeon_driver_irq_postinstall_kms(struct drm_device *dev); void radeon_driver_irq_uninstall_kms(struct drm_device *dev); @@ -325,6 +323,7 @@ static int radeon_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent) { unsigned long flags = 0; + struct drm_device *dev; int ret; if (!ent) @@ -365,7 +364,44 @@ static int radeon_pci_probe(struct pci_dev *pdev, if (ret) return ret; - return drm_get_pci_dev(pdev, ent, &kms_driver); + dev = drm_dev_alloc(&kms_driver, &pdev->dev); + if (IS_ERR(dev)) + return PTR_ERR(dev); + + ret = pci_enable_device(pdev); + if (ret) + goto err_free; + + dev->pdev = pdev; +#ifdef __alpha__ + dev->hose = pdev->sysdata; +#endif + + pci_set_drvdata(pdev, dev); + + if (pci_find_capability(dev->pdev, PCI_CAP_ID_AGP)) + dev->agp = drm_agp_init(dev); + if (dev->agp) { + dev->agp->agp_mtrr = arch_phys_wc_add( + dev->agp->agp_info.aper_base, + dev->agp->agp_info.aper_size * + 1024 * 1024); + } + + ret = drm_dev_register(dev, ent->driver_data); + if (ret) + goto err_agp; + + return 0; + +err_agp: + if (dev->agp) + arch_phys_wc_del(dev->agp->agp_mtrr); + kfree(dev->agp); + pci_disable_device(pdev); +err_free: + drm_dev_put(dev); + return ret; } static void @@ -563,29 +599,14 @@ static const struct file_operations radeon_driver_kms_fops = { #endif }; -static bool -radeon_get_crtc_scanout_position(struct drm_device *dev, unsigned int pipe, - bool in_vblank_irq, int *vpos, int *hpos, - ktime_t *stime, ktime_t *etime, - const struct drm_display_mode *mode) -{ - return radeon_get_crtc_scanoutpos(dev, pipe, 0, vpos, hpos, - stime, etime, mode); -} - static struct drm_driver kms_driver = { .driver_features = - DRIVER_USE_AGP | DRIVER_GEM | DRIVER_RENDER, + DRIVER_GEM | DRIVER_RENDER, .load = radeon_driver_load_kms, .open = radeon_driver_open_kms, .postclose = radeon_driver_postclose_kms, .lastclose = radeon_driver_lastclose_kms, .unload = radeon_driver_unload_kms, - .get_vblank_counter = radeon_get_vblank_counter_kms, - .enable_vblank = radeon_enable_vblank_kms, - .disable_vblank = radeon_disable_vblank_kms, - .get_vblank_timestamp = drm_calc_vbltimestamp_from_scanoutpos, - .get_scanout_position = radeon_get_crtc_scanout_position, .irq_preinstall = radeon_driver_irq_preinstall_kms, .irq_postinstall = radeon_driver_irq_postinstall_kms, .irq_uninstall = radeon_driver_irq_uninstall_kms, diff --git a/drivers/gpu/drm/radeon/radeon_fb.c b/drivers/gpu/drm/radeon/radeon_fb.c index ec0b7d6c994d..cf3156a65fc1 100644 --- a/drivers/gpu/drm/radeon/radeon_fb.c +++ b/drivers/gpu/drm/radeon/radeon_fb.c @@ -354,15 +354,10 @@ int radeon_fbdev_init(struct radeon_device *rdev) drm_fb_helper_prepare(rdev->ddev, &rfbdev->helper, &radeon_fb_helper_funcs); - ret = drm_fb_helper_init(rdev->ddev, &rfbdev->helper, - RADEONFB_CONN_LIMIT); + ret = drm_fb_helper_init(rdev->ddev, &rfbdev->helper); if (ret) goto free; - ret = drm_fb_helper_single_add_all_connectors(&rfbdev->helper); - if (ret) - goto fini; - /* disable all the possible outputs/crtcs before entering KMS mode */ drm_helper_disable_unused_functions(rdev->ddev); @@ -404,15 +399,3 @@ bool radeon_fbdev_robj_is_fb(struct radeon_device *rdev, struct radeon_bo *robj) return true; return false; } - -void radeon_fb_add_connector(struct radeon_device *rdev, struct drm_connector *connector) -{ - if (rdev->mode_info.rfbdev) - drm_fb_helper_add_one_connector(&rdev->mode_info.rfbdev->helper, connector); -} - -void radeon_fb_remove_connector(struct radeon_device *rdev, struct drm_connector *connector) -{ - if (rdev->mode_info.rfbdev) - drm_fb_helper_remove_one_connector(&rdev->mode_info.rfbdev->helper, connector); -} diff --git a/drivers/gpu/drm/radeon/radeon_kms.c b/drivers/gpu/drm/radeon/radeon_kms.c index d24f23a81656..58176db85952 100644 --- a/drivers/gpu/drm/radeon/radeon_kms.c +++ b/drivers/gpu/drm/radeon/radeon_kms.c @@ -32,6 +32,7 @@ #include <linux/uaccess.h> #include <linux/vga_switcheroo.h> +#include <drm/drm_agpsupport.h> #include <drm/drm_fb_helper.h> #include <drm/drm_file.h> #include <drm/drm_ioctl.h> @@ -77,6 +78,11 @@ void radeon_driver_unload_kms(struct drm_device *dev) radeon_modeset_fini(rdev); radeon_device_fini(rdev); + if (dev->agp) + arch_phys_wc_del(dev->agp->agp_mtrr); + kfree(dev->agp); + dev->agp = NULL; + done_free: kfree(rdev); dev->dev_private = NULL; @@ -739,14 +745,15 @@ void radeon_driver_postclose_kms(struct drm_device *dev, /** * radeon_get_vblank_counter_kms - get frame count * - * @dev: drm dev pointer - * @pipe: crtc to get the frame count from + * @crtc: crtc to get the frame count from * * Gets the frame count on the requested crtc (all asics). * Returns frame count on success, -EINVAL on failure. */ -u32 radeon_get_vblank_counter_kms(struct drm_device *dev, unsigned int pipe) +u32 radeon_get_vblank_counter_kms(struct drm_crtc *crtc) { + struct drm_device *dev = crtc->dev; + unsigned int pipe = crtc->index; int vpos, hpos, stat; u32 count; struct radeon_device *rdev = dev->dev_private; @@ -808,25 +815,26 @@ u32 radeon_get_vblank_counter_kms(struct drm_device *dev, unsigned int pipe) /** * radeon_enable_vblank_kms - enable vblank interrupt * - * @dev: drm dev pointer * @crtc: crtc to enable vblank interrupt for * * Enable the interrupt on the requested crtc (all asics). * Returns 0 on success, -EINVAL on failure. */ -int radeon_enable_vblank_kms(struct drm_device *dev, int crtc) +int radeon_enable_vblank_kms(struct drm_crtc *crtc) { + struct drm_device *dev = crtc->dev; + unsigned int pipe = crtc->index; struct radeon_device *rdev = dev->dev_private; unsigned long irqflags; int r; - if (crtc < 0 || crtc >= rdev->num_crtc) { - DRM_ERROR("Invalid crtc %d\n", crtc); + if (pipe < 0 || pipe >= rdev->num_crtc) { + DRM_ERROR("Invalid crtc %d\n", pipe); return -EINVAL; } spin_lock_irqsave(&rdev->irq.lock, irqflags); - rdev->irq.crtc_vblank_int[crtc] = true; + rdev->irq.crtc_vblank_int[pipe] = true; r = radeon_irq_set(rdev); spin_unlock_irqrestore(&rdev->irq.lock, irqflags); return r; @@ -835,23 +843,24 @@ int radeon_enable_vblank_kms(struct drm_device *dev, int crtc) /** * radeon_disable_vblank_kms - disable vblank interrupt * - * @dev: drm dev pointer * @crtc: crtc to disable vblank interrupt for * * Disable the interrupt on the requested crtc (all asics). */ -void radeon_disable_vblank_kms(struct drm_device *dev, int crtc) +void radeon_disable_vblank_kms(struct drm_crtc *crtc) { + struct drm_device *dev = crtc->dev; + unsigned int pipe = crtc->index; struct radeon_device *rdev = dev->dev_private; unsigned long irqflags; - if (crtc < 0 || crtc >= rdev->num_crtc) { - DRM_ERROR("Invalid crtc %d\n", crtc); + if (pipe < 0 || pipe >= rdev->num_crtc) { + DRM_ERROR("Invalid crtc %d\n", pipe); return; } spin_lock_irqsave(&rdev->irq.lock, irqflags); - rdev->irq.crtc_vblank_int[crtc] = false; + rdev->irq.crtc_vblank_int[pipe] = false; radeon_irq_set(rdev); spin_unlock_irqrestore(&rdev->irq.lock, irqflags); } diff --git a/drivers/gpu/drm/radeon/radeon_legacy_crtc.c b/drivers/gpu/drm/radeon/radeon_legacy_crtc.c index a1985a552794..8817fd033cd0 100644 --- a/drivers/gpu/drm/radeon/radeon_legacy_crtc.c +++ b/drivers/gpu/drm/radeon/radeon_legacy_crtc.c @@ -1111,7 +1111,8 @@ static const struct drm_crtc_helper_funcs legacy_helper_funcs = { .mode_set_base_atomic = radeon_crtc_set_base_atomic, .prepare = radeon_crtc_prepare, .commit = radeon_crtc_commit, - .disable = radeon_crtc_disable + .disable = radeon_crtc_disable, + .get_scanout_position = radeon_get_crtc_scanout_position, }; diff --git a/drivers/gpu/drm/radeon/radeon_mode.h b/drivers/gpu/drm/radeon/radeon_mode.h index 96565171d13e..c7f223743d46 100644 --- a/drivers/gpu/drm/radeon/radeon_mode.h +++ b/drivers/gpu/drm/radeon/radeon_mode.h @@ -880,6 +880,12 @@ extern int radeon_get_crtc_scanoutpos(struct drm_device *dev, unsigned int pipe, ktime_t *stime, ktime_t *etime, const struct drm_display_mode *mode); +extern bool +radeon_get_crtc_scanout_position(struct drm_crtc *crtc, bool in_vblank_irq, + int *vpos, int *hpos, + ktime_t *stime, ktime_t *etime, + const struct drm_display_mode *mode); + extern bool radeon_combios_check_hardcoded_edid(struct radeon_device *rdev); extern struct edid * radeon_bios_get_hardcoded_edid(struct radeon_device *rdev); @@ -980,9 +986,6 @@ bool radeon_fbdev_robj_is_fb(struct radeon_device *rdev, struct radeon_bo *robj) void radeon_crtc_handle_vblank(struct radeon_device *rdev, int crtc_id); -void radeon_fb_add_connector(struct radeon_device *rdev, struct drm_connector *connector); -void radeon_fb_remove_connector(struct radeon_device *rdev, struct drm_connector *connector); - void radeon_crtc_handle_flip(struct radeon_device *rdev, int crtc_id); int radeon_align_pitch(struct radeon_device *rdev, int width, int bpp, bool tiled); diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c index badf1b6d1549..5d50c9edbe80 100644 --- a/drivers/gpu/drm/radeon/radeon_ttm.c +++ b/drivers/gpu/drm/radeon/radeon_ttm.c @@ -523,7 +523,7 @@ static int radeon_ttm_tt_pin_userptr(struct ttm_tt *ttm) r = -ENOMEM; nents = dma_map_sg(rdev->dev, ttm->sg->sgl, ttm->sg->nents, direction); - if (nents != ttm->sg->nents) + if (nents == 0) goto release_sg; drm_prime_sg_to_page_addr_arrays(ttm->sg, ttm->pages, diff --git a/drivers/gpu/drm/radeon/si_dpm.c b/drivers/gpu/drm/radeon/si_dpm.c index 05e8b4d0af3f..2cb85dbe728f 100644 --- a/drivers/gpu/drm/radeon/si_dpm.c +++ b/drivers/gpu/drm/radeon/si_dpm.c @@ -2979,7 +2979,6 @@ static void si_apply_state_adjust_rules(struct radeon_device *rdev, if (rdev->family == CHIP_HAINAN) { if ((rdev->pdev->revision == 0x81) || - (rdev->pdev->revision == 0x83) || (rdev->pdev->revision == 0xC3) || (rdev->pdev->device == 0x6664) || (rdev->pdev->device == 0x6665) || diff --git a/drivers/gpu/drm/rcar-du/rcar_du_encoder.c b/drivers/gpu/drm/rcar-du/rcar_du_encoder.c index 3cd83a030a04..c07c6a88aff0 100644 --- a/drivers/gpu/drm/rcar-du/rcar_du_encoder.c +++ b/drivers/gpu/drm/rcar-du/rcar_du_encoder.c @@ -121,7 +121,7 @@ int rcar_du_encoder_init(struct rcar_du_device *rcdu, * Attach the bridge to the encoder. The bridge will create the * connector. */ - ret = drm_bridge_attach(encoder, bridge, NULL); + ret = drm_bridge_attach(encoder, bridge, NULL, 0); if (ret) { drm_encoder_cleanup(encoder); return ret; diff --git a/drivers/gpu/drm/rcar-du/rcar_lvds.c b/drivers/gpu/drm/rcar-du/rcar_lvds.c index 06432c881e07..ab0d49618cf9 100644 --- a/drivers/gpu/drm/rcar-du/rcar_lvds.c +++ b/drivers/gpu/drm/rcar-du/rcar_lvds.c @@ -23,6 +23,7 @@ #include <drm/drm_bridge.h> #include <drm/drm_of.h> #include <drm/drm_panel.h> +#include <drm/drm_print.h> #include <drm/drm_probe_helper.h> #include "rcar_lvds.h" @@ -643,7 +644,8 @@ static bool rcar_lvds_mode_fixup(struct drm_bridge *bridge, return true; } -static int rcar_lvds_attach(struct drm_bridge *bridge) +static int rcar_lvds_attach(struct drm_bridge *bridge, + enum drm_bridge_attach_flags flags) { struct rcar_lvds *lvds = bridge_to_rcar_lvds(bridge); struct drm_connector *connector = &lvds->connector; @@ -653,7 +655,12 @@ static int rcar_lvds_attach(struct drm_bridge *bridge) /* If we have a next bridge just attach it. */ if (lvds->next_bridge) return drm_bridge_attach(bridge->encoder, lvds->next_bridge, - bridge); + bridge, flags); + + if (flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR) { + DRM_ERROR("Fix bridge driver to make connector optional!"); + return -EINVAL; + } /* Otherwise if we have a panel, create a connector. */ if (!lvds->panel) diff --git a/drivers/gpu/drm/rockchip/analogix_dp-rockchip.c b/drivers/gpu/drm/rockchip/analogix_dp-rockchip.c index f38f5e113c6b..ce98c08aa8b4 100644 --- a/drivers/gpu/drm/rockchip/analogix_dp-rockchip.c +++ b/drivers/gpu/drm/rockchip/analogix_dp-rockchip.c @@ -325,15 +325,9 @@ static int rockchip_dp_bind(struct device *dev, struct device *master, void *data) { struct rockchip_dp_device *dp = dev_get_drvdata(dev); - const struct rockchip_dp_chip_data *dp_data; struct drm_device *drm_dev = data; int ret; - dp_data = of_device_get_match_data(dev); - if (!dp_data) - return -ENODEV; - - dp->data = dp_data; dp->drm_dev = drm_dev; ret = rockchip_dp_drm_create_encoder(dp); @@ -344,16 +338,9 @@ static int rockchip_dp_bind(struct device *dev, struct device *master, dp->plat_data.encoder = &dp->encoder; - dp->plat_data.dev_type = dp->data->chip_type; - dp->plat_data.power_on_start = rockchip_dp_poweron_start; - dp->plat_data.power_off = rockchip_dp_powerdown; - dp->plat_data.get_modes = rockchip_dp_get_modes; - - dp->adp = analogix_dp_bind(dev, dp->drm_dev, &dp->plat_data); - if (IS_ERR(dp->adp)) { - ret = PTR_ERR(dp->adp); + ret = analogix_dp_bind(dp->adp, drm_dev); + if (ret) goto err_cleanup_encoder; - } return 0; err_cleanup_encoder: @@ -368,8 +355,6 @@ static void rockchip_dp_unbind(struct device *dev, struct device *master, analogix_dp_unbind(dp->adp); dp->encoder.funcs->destroy(&dp->encoder); - - dp->adp = ERR_PTR(-ENODEV); } static const struct component_ops rockchip_dp_component_ops = { @@ -380,10 +365,15 @@ static const struct component_ops rockchip_dp_component_ops = { static int rockchip_dp_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; + const struct rockchip_dp_chip_data *dp_data; struct drm_panel *panel = NULL; struct rockchip_dp_device *dp; int ret; + dp_data = of_device_get_match_data(dev); + if (!dp_data) + return -ENODEV; + ret = drm_of_find_panel_or_bridge(dev->of_node, 1, 0, &panel, NULL); if (ret < 0) return ret; @@ -394,7 +384,12 @@ static int rockchip_dp_probe(struct platform_device *pdev) dp->dev = dev; dp->adp = ERR_PTR(-ENODEV); + dp->data = dp_data; dp->plat_data.panel = panel; + dp->plat_data.dev_type = dp->data->chip_type; + dp->plat_data.power_on_start = rockchip_dp_poweron_start; + dp->plat_data.power_off = rockchip_dp_powerdown; + dp->plat_data.get_modes = rockchip_dp_get_modes; ret = rockchip_dp_of_probe(dp); if (ret < 0) @@ -402,12 +397,19 @@ static int rockchip_dp_probe(struct platform_device *pdev) platform_set_drvdata(pdev, dp); + dp->adp = analogix_dp_probe(dev, &dp->plat_data); + if (IS_ERR(dp->adp)) + return PTR_ERR(dp->adp); + return component_add(dev, &rockchip_dp_component_ops); } static int rockchip_dp_remove(struct platform_device *pdev) { + struct rockchip_dp_device *dp = platform_get_drvdata(pdev); + component_del(&pdev->dev, &rockchip_dp_component_ops); + analogix_dp_remove(dp->adp); return 0; } diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_fbdev.c b/drivers/gpu/drm/rockchip/rockchip_drm_fbdev.c index 521fe42ac5e2..2fdc455c4ad7 100644 --- a/drivers/gpu/drm/rockchip/rockchip_drm_fbdev.c +++ b/drivers/gpu/drm/rockchip/rockchip_drm_fbdev.c @@ -124,7 +124,7 @@ int rockchip_drm_fbdev_init(struct drm_device *dev) drm_fb_helper_prepare(dev, helper, &rockchip_drm_fb_helper_funcs); - ret = drm_fb_helper_init(dev, helper, ROCKCHIP_MAX_CONNECTOR); + ret = drm_fb_helper_init(dev, helper); if (ret < 0) { DRM_DEV_ERROR(dev->dev, "Failed to initialize drm fb helper - %d.\n", @@ -132,13 +132,6 @@ int rockchip_drm_fbdev_init(struct drm_device *dev) return ret; } - ret = drm_fb_helper_single_add_all_connectors(helper); - if (ret < 0) { - DRM_DEV_ERROR(dev->dev, - "Failed to add connectors - %d.\n", ret); - goto err_drm_fb_helper_fini; - } - ret = drm_fb_helper_initial_config(helper, PREFERRED_BPP); if (ret < 0) { DRM_DEV_ERROR(dev->dev, diff --git a/drivers/gpu/drm/rockchip/rockchip_lvds.c b/drivers/gpu/drm/rockchip/rockchip_lvds.c index f25a36743cbd..449a62908d21 100644 --- a/drivers/gpu/drm/rockchip/rockchip_lvds.c +++ b/drivers/gpu/drm/rockchip/rockchip_lvds.c @@ -646,7 +646,7 @@ static int rockchip_lvds_bind(struct device *dev, struct device *master, goto err_free_connector; } } else { - ret = drm_bridge_attach(encoder, lvds->bridge, NULL); + ret = drm_bridge_attach(encoder, lvds->bridge, NULL, 0); if (ret) { DRM_DEV_ERROR(drm_dev->dev, "failed to attach bridge: %d\n", ret); diff --git a/drivers/gpu/drm/rockchip/rockchip_rgb.c b/drivers/gpu/drm/rockchip/rockchip_rgb.c index ae730275a34f..90784781e515 100644 --- a/drivers/gpu/drm/rockchip/rockchip_rgb.c +++ b/drivers/gpu/drm/rockchip/rockchip_rgb.c @@ -98,7 +98,8 @@ struct rockchip_rgb *rockchip_rgb_init(struct device *dev, if (of_property_read_u32(endpoint, "reg", &endpoint_id)) endpoint_id = 0; - if (rockchip_drm_endpoint_is_subdriver(endpoint) > 0) + /* if subdriver (> 0) or error case (< 0), ignore entry */ + if (rockchip_drm_endpoint_is_subdriver(endpoint) != 0) continue; child_count++; @@ -144,7 +145,7 @@ struct rockchip_rgb *rockchip_rgb_init(struct device *dev, rgb->bridge = bridge; - ret = drm_bridge_attach(encoder, rgb->bridge, NULL); + ret = drm_bridge_attach(encoder, rgb->bridge, NULL, 0); if (ret) { DRM_DEV_ERROR(drm_dev->dev, "failed to attach bridge: %d\n", ret); diff --git a/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h b/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h index d79086498aff..877ce9b127f1 100644 --- a/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h +++ b/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h @@ -59,6 +59,33 @@ TRACE_EVENT(drm_sched_job, __entry->job_count, __entry->hw_job_count) ); +TRACE_EVENT(drm_run_job, + TP_PROTO(struct drm_sched_job *sched_job, struct drm_sched_entity *entity), + TP_ARGS(sched_job, entity), + TP_STRUCT__entry( + __field(struct drm_sched_entity *, entity) + __field(struct dma_fence *, fence) + __field(const char *, name) + __field(uint64_t, id) + __field(u32, job_count) + __field(int, hw_job_count) + ), + + TP_fast_assign( + __entry->entity = entity; + __entry->id = sched_job->id; + __entry->fence = &sched_job->s_fence->finished; + __entry->name = sched_job->sched->name; + __entry->job_count = spsc_queue_count(&entity->job_queue); + __entry->hw_job_count = atomic_read( + &sched_job->sched->hw_rq_count); + ), + TP_printk("entity=%p, id=%llu, fence=%p, ring=%s, job count:%u, hw job count:%d", + __entry->entity, __entry->id, + __entry->fence, __entry->name, + __entry->job_count, __entry->hw_job_count) +); + TRACE_EVENT(drm_sched_process_job, TP_PROTO(struct drm_sched_fence *fence), TP_ARGS(fence), diff --git a/drivers/gpu/drm/scheduler/sched_entity.c b/drivers/gpu/drm/scheduler/sched_entity.c index 63bccd201b97..c803e14eed91 100644 --- a/drivers/gpu/drm/scheduler/sched_entity.c +++ b/drivers/gpu/drm/scheduler/sched_entity.c @@ -84,6 +84,24 @@ int drm_sched_entity_init(struct drm_sched_entity *entity, EXPORT_SYMBOL(drm_sched_entity_init); /** + * drm_sched_entity_modify_sched - Modify sched of an entity + * @entity: scheduler entity to init + * @sched_list: the list of new drm scheds which will replace + * existing entity->sched_list + * @num_sched_list: number of drm sched in sched_list + */ +void drm_sched_entity_modify_sched(struct drm_sched_entity *entity, + struct drm_gpu_scheduler **sched_list, + unsigned int num_sched_list) +{ + WARN_ON(!num_sched_list || !sched_list); + + entity->sched_list = sched_list; + entity->num_sched_list = num_sched_list; +} +EXPORT_SYMBOL(drm_sched_entity_modify_sched); + +/** * drm_sched_entity_is_idle - Check if entity is idle * * @entity: scheduler entity @@ -120,38 +138,6 @@ bool drm_sched_entity_is_ready(struct drm_sched_entity *entity) } /** - * drm_sched_entity_get_free_sched - Get the rq from rq_list with least load - * - * @entity: scheduler entity - * - * Return the pointer to the rq with least load. - */ -static struct drm_sched_rq * -drm_sched_entity_get_free_sched(struct drm_sched_entity *entity) -{ - struct drm_sched_rq *rq = NULL; - unsigned int min_score = UINT_MAX, num_score; - int i; - - for (i = 0; i < entity->num_sched_list; ++i) { - struct drm_gpu_scheduler *sched = entity->sched_list[i]; - - if (!entity->sched_list[i]->ready) { - DRM_WARN("sched%s is not ready, skipping", sched->name); - continue; - } - - num_score = atomic_read(&sched->score); - if (num_score < min_score) { - min_score = num_score; - rq = &entity->sched_list[i]->sched_rq[entity->priority]; - } - } - - return rq; -} - -/** * drm_sched_entity_flush - Flush a context entity * * @entity: scheduler entity @@ -461,6 +447,7 @@ struct drm_sched_job *drm_sched_entity_pop_job(struct drm_sched_entity *entity) void drm_sched_entity_select_rq(struct drm_sched_entity *entity) { struct dma_fence *fence; + struct drm_gpu_scheduler *sched; struct drm_sched_rq *rq; if (spsc_queue_count(&entity->job_queue) || entity->num_sched_list <= 1) @@ -471,7 +458,8 @@ void drm_sched_entity_select_rq(struct drm_sched_entity *entity) return; spin_lock(&entity->rq_lock); - rq = drm_sched_entity_get_free_sched(entity); + sched = drm_sched_pick_best(entity->sched_list, entity->num_sched_list); + rq = sched ? &sched->sched_rq[entity->priority] : NULL; if (rq != entity->rq) { drm_sched_rq_remove_entity(entity->rq, entity); entity->rq = rq; @@ -498,7 +486,7 @@ void drm_sched_entity_push_job(struct drm_sched_job *sched_job, bool first; trace_drm_sched_job(sched_job, entity); - atomic_inc(&entity->rq->sched->score); + atomic_inc(&entity->rq->sched->num_jobs); WRITE_ONCE(entity->last_user, current->group_leader); first = spsc_queue_push(&entity->job_queue, &sched_job->queue_node); diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/scheduler/sched_main.c index 71ce6215956f..8e731ed0d9d9 100644 --- a/drivers/gpu/drm/scheduler/sched_main.c +++ b/drivers/gpu/drm/scheduler/sched_main.c @@ -92,7 +92,6 @@ void drm_sched_rq_add_entity(struct drm_sched_rq *rq, if (!list_empty(&entity->list)) return; spin_lock(&rq->lock); - atomic_inc(&rq->sched->score); list_add_tail(&entity->list, &rq->entities); spin_unlock(&rq->lock); } @@ -111,7 +110,6 @@ void drm_sched_rq_remove_entity(struct drm_sched_rq *rq, if (list_empty(&entity->list)) return; spin_lock(&rq->lock); - atomic_dec(&rq->sched->score); list_del_init(&entity->list); if (rq->current_entity == entity) rq->current_entity = NULL; @@ -222,8 +220,7 @@ EXPORT_SYMBOL(drm_sched_fault); * * Suspend the delayed work timeout for the scheduler. This is done by * modifying the delayed work timeout to an arbitrary large value, - * MAX_SCHEDULE_TIMEOUT in this case. Note that this function can be - * called from an IRQ context. + * MAX_SCHEDULE_TIMEOUT in this case. * * Returns the timeout remaining * @@ -252,46 +249,41 @@ EXPORT_SYMBOL(drm_sched_suspend_timeout); * @sched: scheduler instance for which to resume the timeout * @remaining: remaining timeout * - * Resume the delayed work timeout for the scheduler. Note that - * this function can be called from an IRQ context. + * Resume the delayed work timeout for the scheduler. */ void drm_sched_resume_timeout(struct drm_gpu_scheduler *sched, unsigned long remaining) { - unsigned long flags; - - spin_lock_irqsave(&sched->job_list_lock, flags); + spin_lock(&sched->job_list_lock); if (list_empty(&sched->ring_mirror_list)) cancel_delayed_work(&sched->work_tdr); else mod_delayed_work(system_wq, &sched->work_tdr, remaining); - spin_unlock_irqrestore(&sched->job_list_lock, flags); + spin_unlock(&sched->job_list_lock); } EXPORT_SYMBOL(drm_sched_resume_timeout); static void drm_sched_job_begin(struct drm_sched_job *s_job) { struct drm_gpu_scheduler *sched = s_job->sched; - unsigned long flags; - spin_lock_irqsave(&sched->job_list_lock, flags); + spin_lock(&sched->job_list_lock); list_add_tail(&s_job->node, &sched->ring_mirror_list); drm_sched_start_timeout(sched); - spin_unlock_irqrestore(&sched->job_list_lock, flags); + spin_unlock(&sched->job_list_lock); } static void drm_sched_job_timedout(struct work_struct *work) { struct drm_gpu_scheduler *sched; struct drm_sched_job *job; - unsigned long flags; sched = container_of(work, struct drm_gpu_scheduler, work_tdr.work); /* Protects against concurrent deletion in drm_sched_get_cleanup_job */ - spin_lock_irqsave(&sched->job_list_lock, flags); + spin_lock(&sched->job_list_lock); job = list_first_entry_or_null(&sched->ring_mirror_list, struct drm_sched_job, node); @@ -302,7 +294,7 @@ static void drm_sched_job_timedout(struct work_struct *work) * is parked at which point it's safe. */ list_del_init(&job->node); - spin_unlock_irqrestore(&sched->job_list_lock, flags); + spin_unlock(&sched->job_list_lock); job->sched->ops->timedout_job(job); @@ -315,12 +307,12 @@ static void drm_sched_job_timedout(struct work_struct *work) sched->free_guilty = false; } } else { - spin_unlock_irqrestore(&sched->job_list_lock, flags); + spin_unlock(&sched->job_list_lock); } - spin_lock_irqsave(&sched->job_list_lock, flags); + spin_lock(&sched->job_list_lock); drm_sched_start_timeout(sched); - spin_unlock_irqrestore(&sched->job_list_lock, flags); + spin_unlock(&sched->job_list_lock); } /** @@ -383,7 +375,6 @@ EXPORT_SYMBOL(drm_sched_increase_karma); void drm_sched_stop(struct drm_gpu_scheduler *sched, struct drm_sched_job *bad) { struct drm_sched_job *s_job, *tmp; - unsigned long flags; kthread_park(sched->thread); @@ -417,9 +408,9 @@ void drm_sched_stop(struct drm_gpu_scheduler *sched, struct drm_sched_job *bad) * remove job from ring_mirror_list. * Locking here is for concurrent resume timeout */ - spin_lock_irqsave(&sched->job_list_lock, flags); + spin_lock(&sched->job_list_lock); list_del_init(&s_job->node); - spin_unlock_irqrestore(&sched->job_list_lock, flags); + spin_unlock(&sched->job_list_lock); /* * Wait for job's HW fence callback to finish using s_job @@ -462,7 +453,6 @@ EXPORT_SYMBOL(drm_sched_stop); void drm_sched_start(struct drm_gpu_scheduler *sched, bool full_recovery) { struct drm_sched_job *s_job, *tmp; - unsigned long flags; int r; /* @@ -491,9 +481,9 @@ void drm_sched_start(struct drm_gpu_scheduler *sched, bool full_recovery) } if (full_recovery) { - spin_lock_irqsave(&sched->job_list_lock, flags); + spin_lock(&sched->job_list_lock); drm_sched_start_timeout(sched); - spin_unlock_irqrestore(&sched->job_list_lock, flags); + spin_unlock(&sched->job_list_lock); } kthread_unpark(sched->thread); @@ -657,11 +647,13 @@ static void drm_sched_process_job(struct dma_fence *f, struct dma_fence_cb *cb) struct drm_gpu_scheduler *sched = s_fence->sched; atomic_dec(&sched->hw_rq_count); - atomic_dec(&sched->score); + atomic_dec(&sched->num_jobs); trace_drm_sched_process_job(s_fence); + dma_fence_get(&s_fence->finished); drm_sched_fence_finished(s_fence); + dma_fence_put(&s_fence->finished); wake_up_interruptible(&sched->wake_up_worker); } @@ -677,7 +669,6 @@ static struct drm_sched_job * drm_sched_get_cleanup_job(struct drm_gpu_scheduler *sched) { struct drm_sched_job *job; - unsigned long flags; /* * Don't destroy jobs while the timeout worker is running OR thread @@ -688,7 +679,7 @@ drm_sched_get_cleanup_job(struct drm_gpu_scheduler *sched) __kthread_should_park(sched->thread)) return NULL; - spin_lock_irqsave(&sched->job_list_lock, flags); + spin_lock(&sched->job_list_lock); job = list_first_entry_or_null(&sched->ring_mirror_list, struct drm_sched_job, node); @@ -702,12 +693,48 @@ drm_sched_get_cleanup_job(struct drm_gpu_scheduler *sched) drm_sched_start_timeout(sched); } - spin_unlock_irqrestore(&sched->job_list_lock, flags); + spin_unlock(&sched->job_list_lock); return job; } /** + * drm_sched_pick_best - Get a drm sched from a sched_list with the least load + * @sched_list: list of drm_gpu_schedulers + * @num_sched_list: number of drm_gpu_schedulers in the sched_list + * + * Returns pointer of the sched with the least load or NULL if none of the + * drm_gpu_schedulers are ready + */ +struct drm_gpu_scheduler * +drm_sched_pick_best(struct drm_gpu_scheduler **sched_list, + unsigned int num_sched_list) +{ + struct drm_gpu_scheduler *sched, *picked_sched = NULL; + int i; + unsigned int min_jobs = UINT_MAX, num_jobs; + + for (i = 0; i < num_sched_list; ++i) { + sched = sched_list[i]; + + if (!sched->ready) { + DRM_WARN("scheduler %s is not ready, skipping", + sched->name); + continue; + } + + num_jobs = atomic_read(&sched->num_jobs); + if (num_jobs < min_jobs) { + min_jobs = num_jobs; + picked_sched = sched; + } + } + + return picked_sched; +} +EXPORT_SYMBOL(drm_sched_pick_best); + +/** * drm_sched_blocked - check if the scheduler is blocked * * @sched: scheduler instance @@ -773,6 +800,7 @@ static int drm_sched_main(void *param) atomic_inc(&sched->hw_rq_count); drm_sched_job_begin(sched_job); + trace_drm_run_job(sched_job, entity); fence = sched->ops->run_job(sched_job); drm_sched_fence_scheduled(s_fence); @@ -832,7 +860,7 @@ int drm_sched_init(struct drm_gpu_scheduler *sched, spin_lock_init(&sched->job_list_lock); atomic_set(&sched->hw_rq_count, 0); INIT_DELAYED_WORK(&sched->work_tdr, drm_sched_job_timedout); - atomic_set(&sched->score, 0); + atomic_set(&sched->num_jobs, 0); atomic64_set(&sched->job_id_count, 0); /* Each scheduler will run on a seperate kernel thread */ diff --git a/drivers/gpu/drm/selftests/drm_cmdline_selftests.h b/drivers/gpu/drm/selftests/drm_cmdline_selftests.h index ceac7af9a172..29e367db6118 100644 --- a/drivers/gpu/drm/selftests/drm_cmdline_selftests.h +++ b/drivers/gpu/drm/selftests/drm_cmdline_selftests.h @@ -53,6 +53,7 @@ cmdline_test(drm_cmdline_test_rotate_0) cmdline_test(drm_cmdline_test_rotate_90) cmdline_test(drm_cmdline_test_rotate_180) cmdline_test(drm_cmdline_test_rotate_270) +cmdline_test(drm_cmdline_test_rotate_multiple) cmdline_test(drm_cmdline_test_rotate_invalid_val) cmdline_test(drm_cmdline_test_rotate_truncated) cmdline_test(drm_cmdline_test_hmirror) diff --git a/drivers/gpu/drm/selftests/test-drm_cmdline_parser.c b/drivers/gpu/drm/selftests/test-drm_cmdline_parser.c index 520f3e66a384..d96cd890def6 100644 --- a/drivers/gpu/drm/selftests/test-drm_cmdline_parser.c +++ b/drivers/gpu/drm/selftests/test-drm_cmdline_parser.c @@ -856,6 +856,17 @@ static int drm_cmdline_test_rotate_270(void *ignored) return 0; } +static int drm_cmdline_test_rotate_multiple(void *ignored) +{ + struct drm_cmdline_mode mode = { }; + + FAIL_ON(drm_mode_parse_command_line_for_connector("720x480,rotate=0,rotate=90", + &no_connector, + &mode)); + + return 0; +} + static int drm_cmdline_test_rotate_invalid_val(void *ignored) { struct drm_cmdline_mode mode = { }; @@ -888,7 +899,7 @@ static int drm_cmdline_test_hmirror(void *ignored) FAIL_ON(!mode.specified); FAIL_ON(mode.xres != 720); FAIL_ON(mode.yres != 480); - FAIL_ON(mode.rotation_reflection != DRM_MODE_REFLECT_X); + FAIL_ON(mode.rotation_reflection != (DRM_MODE_ROTATE_0 | DRM_MODE_REFLECT_X)); FAIL_ON(mode.refresh_specified); @@ -913,7 +924,7 @@ static int drm_cmdline_test_vmirror(void *ignored) FAIL_ON(!mode.specified); FAIL_ON(mode.xres != 720); FAIL_ON(mode.yres != 480); - FAIL_ON(mode.rotation_reflection != DRM_MODE_REFLECT_Y); + FAIL_ON(mode.rotation_reflection != (DRM_MODE_ROTATE_0 | DRM_MODE_REFLECT_Y)); FAIL_ON(mode.refresh_specified); diff --git a/drivers/gpu/drm/sti/sti_crtc.c b/drivers/gpu/drm/sti/sti_crtc.c index dc64fbfc4e61..49e6cb8f5836 100644 --- a/drivers/gpu/drm/sti/sti_crtc.c +++ b/drivers/gpu/drm/sti/sti_crtc.c @@ -279,12 +279,13 @@ int sti_crtc_vblank_cb(struct notifier_block *nb, return 0; } -int sti_crtc_enable_vblank(struct drm_device *dev, unsigned int pipe) +static int sti_crtc_enable_vblank(struct drm_crtc *crtc) { + struct drm_device *dev = crtc->dev; + unsigned int pipe = crtc->index; struct sti_private *dev_priv = dev->dev_private; struct sti_compositor *compo = dev_priv->compo; struct notifier_block *vtg_vblank_nb = &compo->vtg_vblank_nb[pipe]; - struct drm_crtc *crtc = &compo->mixer[pipe]->drm_crtc; struct sti_vtg *vtg = compo->vtg[pipe]; DRM_DEBUG_DRIVER("\n"); @@ -297,8 +298,10 @@ int sti_crtc_enable_vblank(struct drm_device *dev, unsigned int pipe) return 0; } -void sti_crtc_disable_vblank(struct drm_device *drm_dev, unsigned int pipe) +static void sti_crtc_disable_vblank(struct drm_crtc *crtc) { + struct drm_device *drm_dev = crtc->dev; + unsigned int pipe = crtc->index; struct sti_private *priv = drm_dev->dev_private; struct sti_compositor *compo = priv->compo; struct notifier_block *vtg_vblank_nb = &compo->vtg_vblank_nb[pipe]; @@ -330,6 +333,8 @@ static const struct drm_crtc_funcs sti_crtc_funcs = { .atomic_duplicate_state = drm_atomic_helper_crtc_duplicate_state, .atomic_destroy_state = drm_atomic_helper_crtc_destroy_state, .late_register = sti_crtc_late_register, + .enable_vblank = sti_crtc_enable_vblank, + .disable_vblank = sti_crtc_disable_vblank, }; bool sti_crtc_is_main(struct drm_crtc *crtc) diff --git a/drivers/gpu/drm/sti/sti_crtc.h b/drivers/gpu/drm/sti/sti_crtc.h index df489ab14e2b..1132b4586712 100644 --- a/drivers/gpu/drm/sti/sti_crtc.h +++ b/drivers/gpu/drm/sti/sti_crtc.h @@ -15,8 +15,6 @@ struct sti_mixer; int sti_crtc_init(struct drm_device *drm_dev, struct sti_mixer *mixer, struct drm_plane *primary, struct drm_plane *cursor); -int sti_crtc_enable_vblank(struct drm_device *dev, unsigned int pipe); -void sti_crtc_disable_vblank(struct drm_device *dev, unsigned int pipe); int sti_crtc_vblank_cb(struct notifier_block *nb, unsigned long event, void *data); bool sti_crtc_is_main(struct drm_crtc *drm_crtc); diff --git a/drivers/gpu/drm/sti/sti_drv.c b/drivers/gpu/drm/sti/sti_drv.c index a39fc36f815b..50870d8cbb76 100644 --- a/drivers/gpu/drm/sti/sti_drv.c +++ b/drivers/gpu/drm/sti/sti_drv.c @@ -21,7 +21,6 @@ #include <drm/drm_of.h> #include <drm/drm_probe_helper.h> -#include "sti_crtc.h" #include "sti_drv.h" #include "sti_plane.h" @@ -146,9 +145,6 @@ static struct drm_driver sti_driver = { .dumb_create = drm_gem_cma_dumb_create, .fops = &sti_driver_fops, - .enable_vblank = sti_crtc_enable_vblank, - .disable_vblank = sti_crtc_disable_vblank, - .prime_handle_to_fd = drm_gem_prime_handle_to_fd, .prime_fd_to_handle = drm_gem_prime_fd_to_handle, .gem_prime_get_sg_table = drm_gem_cma_prime_get_sg_table, diff --git a/drivers/gpu/drm/sti/sti_dvo.c b/drivers/gpu/drm/sti/sti_dvo.c index b2778ec1cdd7..3d04bfca21a0 100644 --- a/drivers/gpu/drm/sti/sti_dvo.c +++ b/drivers/gpu/drm/sti/sti_dvo.c @@ -467,7 +467,7 @@ static int sti_dvo_bind(struct device *dev, struct device *master, void *data) bridge->of_node = dvo->dev.of_node; drm_bridge_add(bridge); - err = drm_bridge_attach(encoder, bridge, NULL); + err = drm_bridge_attach(encoder, bridge, NULL, 0); if (err) { DRM_ERROR("Failed to attach bridge\n"); return err; diff --git a/drivers/gpu/drm/sti/sti_hda.c b/drivers/gpu/drm/sti/sti_hda.c index 2bb32009d117..f3f28d79b0e4 100644 --- a/drivers/gpu/drm/sti/sti_hda.c +++ b/drivers/gpu/drm/sti/sti_hda.c @@ -701,7 +701,7 @@ static int sti_hda_bind(struct device *dev, struct device *master, void *data) bridge->driver_private = hda; bridge->funcs = &sti_hda_bridge_funcs; - drm_bridge_attach(encoder, bridge, NULL); + drm_bridge_attach(encoder, bridge, NULL, 0); connector->encoder = encoder; diff --git a/drivers/gpu/drm/sti/sti_hdmi.c b/drivers/gpu/drm/sti/sti_hdmi.c index 64ed102033c8..18eaf786ffa4 100644 --- a/drivers/gpu/drm/sti/sti_hdmi.c +++ b/drivers/gpu/drm/sti/sti_hdmi.c @@ -1281,7 +1281,7 @@ static int sti_hdmi_bind(struct device *dev, struct device *master, void *data) bridge->driver_private = hdmi; bridge->funcs = &sti_hdmi_bridge_funcs; - drm_bridge_attach(encoder, bridge, NULL); + drm_bridge_attach(encoder, bridge, NULL, 0); connector->encoder = encoder; diff --git a/drivers/gpu/drm/stm/drv.c b/drivers/gpu/drm/stm/drv.c index 5a9f9aca8bc2..ea9fcbdc68b3 100644 --- a/drivers/gpu/drm/stm/drv.c +++ b/drivers/gpu/drm/stm/drv.c @@ -72,8 +72,6 @@ static struct drm_driver drv_driver = { .gem_prime_vmap = drm_gem_cma_prime_vmap, .gem_prime_vunmap = drm_gem_cma_prime_vunmap, .gem_prime_mmap = drm_gem_cma_prime_mmap, - .get_scanout_position = ltdc_crtc_scanoutpos, - .get_vblank_timestamp = drm_calc_vbltimestamp_from_scanoutpos, }; static int drv_load(struct drm_device *ddev) diff --git a/drivers/gpu/drm/stm/ltdc.c b/drivers/gpu/drm/stm/ltdc.c index 8750ee831bea..df585fe64f61 100644 --- a/drivers/gpu/drm/stm/ltdc.c +++ b/drivers/gpu/drm/stm/ltdc.c @@ -636,43 +636,13 @@ static void ltdc_crtc_atomic_flush(struct drm_crtc *crtc, } } -static const struct drm_crtc_helper_funcs ltdc_crtc_helper_funcs = { - .mode_valid = ltdc_crtc_mode_valid, - .mode_fixup = ltdc_crtc_mode_fixup, - .mode_set_nofb = ltdc_crtc_mode_set_nofb, - .atomic_flush = ltdc_crtc_atomic_flush, - .atomic_enable = ltdc_crtc_atomic_enable, - .atomic_disable = ltdc_crtc_atomic_disable, -}; - -static int ltdc_crtc_enable_vblank(struct drm_crtc *crtc) -{ - struct ltdc_device *ldev = crtc_to_ltdc(crtc); - struct drm_crtc_state *state = crtc->state; - - DRM_DEBUG_DRIVER("\n"); - - if (state->enable) - reg_set(ldev->regs, LTDC_IER, IER_LIE); - else - return -EPERM; - - return 0; -} - -static void ltdc_crtc_disable_vblank(struct drm_crtc *crtc) -{ - struct ltdc_device *ldev = crtc_to_ltdc(crtc); - - DRM_DEBUG_DRIVER("\n"); - reg_clear(ldev->regs, LTDC_IER, IER_LIE); -} - -bool ltdc_crtc_scanoutpos(struct drm_device *ddev, unsigned int pipe, - bool in_vblank_irq, int *vpos, int *hpos, - ktime_t *stime, ktime_t *etime, - const struct drm_display_mode *mode) +static bool ltdc_crtc_get_scanout_position(struct drm_crtc *crtc, + bool in_vblank_irq, + int *vpos, int *hpos, + ktime_t *stime, ktime_t *etime, + const struct drm_display_mode *mode) { + struct drm_device *ddev = crtc->dev; struct ltdc_device *ldev = ddev->dev_private; int line, vactive_start, vactive_end, vtotal; @@ -715,6 +685,39 @@ bool ltdc_crtc_scanoutpos(struct drm_device *ddev, unsigned int pipe, return true; } +static const struct drm_crtc_helper_funcs ltdc_crtc_helper_funcs = { + .mode_valid = ltdc_crtc_mode_valid, + .mode_fixup = ltdc_crtc_mode_fixup, + .mode_set_nofb = ltdc_crtc_mode_set_nofb, + .atomic_flush = ltdc_crtc_atomic_flush, + .atomic_enable = ltdc_crtc_atomic_enable, + .atomic_disable = ltdc_crtc_atomic_disable, + .get_scanout_position = ltdc_crtc_get_scanout_position, +}; + +static int ltdc_crtc_enable_vblank(struct drm_crtc *crtc) +{ + struct ltdc_device *ldev = crtc_to_ltdc(crtc); + struct drm_crtc_state *state = crtc->state; + + DRM_DEBUG_DRIVER("\n"); + + if (state->enable) + reg_set(ldev->regs, LTDC_IER, IER_LIE); + else + return -EPERM; + + return 0; +} + +static void ltdc_crtc_disable_vblank(struct drm_crtc *crtc) +{ + struct ltdc_device *ldev = crtc_to_ltdc(crtc); + + DRM_DEBUG_DRIVER("\n"); + reg_clear(ldev->regs, LTDC_IER, IER_LIE); +} + static const struct drm_crtc_funcs ltdc_crtc_funcs = { .destroy = drm_crtc_cleanup, .set_config = drm_atomic_helper_set_config, @@ -724,6 +727,7 @@ static const struct drm_crtc_funcs ltdc_crtc_funcs = { .atomic_destroy_state = drm_atomic_helper_crtc_destroy_state, .enable_vblank = ltdc_crtc_enable_vblank, .disable_vblank = ltdc_crtc_disable_vblank, + .get_vblank_timestamp = drm_crtc_vblank_helper_get_vblank_timestamp, .gamma_set = drm_atomic_helper_legacy_gamma_set, }; @@ -1105,7 +1109,7 @@ static int ltdc_encoder_init(struct drm_device *ddev, struct drm_bridge *bridge) drm_encoder_helper_add(encoder, <dc_encoder_helper_funcs); - ret = drm_bridge_attach(encoder, bridge, NULL); + ret = drm_bridge_attach(encoder, bridge, NULL, 0); if (ret) { drm_encoder_cleanup(encoder); return -EINVAL; diff --git a/drivers/gpu/drm/stm/ltdc.h b/drivers/gpu/drm/stm/ltdc.h index 310e87f0667c..f153b908c70e 100644 --- a/drivers/gpu/drm/stm/ltdc.h +++ b/drivers/gpu/drm/stm/ltdc.h @@ -40,11 +40,6 @@ struct ltdc_device { struct drm_atomic_state *suspend_state; }; -bool ltdc_crtc_scanoutpos(struct drm_device *dev, unsigned int pipe, - bool in_vblank_irq, int *vpos, int *hpos, - ktime_t *stime, ktime_t *etime, - const struct drm_display_mode *mode); - int ltdc_load(struct drm_device *ddev); void ltdc_unload(struct drm_device *ddev); void ltdc_suspend(struct drm_device *ddev); diff --git a/drivers/gpu/drm/sun4i/sun4i_lvds.c b/drivers/gpu/drm/sun4i/sun4i_lvds.c index 65b7a8739666..26e5c7ceb8ff 100644 --- a/drivers/gpu/drm/sun4i/sun4i_lvds.c +++ b/drivers/gpu/drm/sun4i/sun4i_lvds.c @@ -156,7 +156,7 @@ int sun4i_lvds_init(struct drm_device *drm, struct sun4i_tcon *tcon) } if (bridge) { - ret = drm_bridge_attach(encoder, bridge, NULL); + ret = drm_bridge_attach(encoder, bridge, NULL, 0); if (ret) { dev_err(drm->dev, "Couldn't attach our bridge\n"); goto err_cleanup_connector; diff --git a/drivers/gpu/drm/sun4i/sun4i_rgb.c b/drivers/gpu/drm/sun4i/sun4i_rgb.c index b27f16af50f5..3b23d5be3cf3 100644 --- a/drivers/gpu/drm/sun4i/sun4i_rgb.c +++ b/drivers/gpu/drm/sun4i/sun4i_rgb.c @@ -253,7 +253,7 @@ int sun4i_rgb_init(struct drm_device *drm, struct sun4i_tcon *tcon) } if (rgb->bridge) { - ret = drm_bridge_attach(encoder, rgb->bridge, NULL); + ret = drm_bridge_attach(encoder, rgb->bridge, NULL, 0); if (ret) { dev_err(drm->dev, "Couldn't attach our bridge\n"); goto err_cleanup_connector; diff --git a/drivers/gpu/drm/sun4i/sun4i_tcon.c b/drivers/gpu/drm/sun4i/sun4i_tcon.c index c81cdce6ed55..624437b27cdc 100644 --- a/drivers/gpu/drm/sun4i/sun4i_tcon.c +++ b/drivers/gpu/drm/sun4i/sun4i_tcon.c @@ -114,46 +114,71 @@ static void sun4i_tcon_channel_set_status(struct sun4i_tcon *tcon, int channel, } } +static void sun4i_tcon_setup_lvds_phy(struct sun4i_tcon *tcon, + const struct drm_encoder *encoder) +{ + regmap_write(tcon->regs, SUN4I_TCON0_LVDS_ANA0_REG, + SUN4I_TCON0_LVDS_ANA0_CK_EN | + SUN4I_TCON0_LVDS_ANA0_REG_V | + SUN4I_TCON0_LVDS_ANA0_REG_C | + SUN4I_TCON0_LVDS_ANA0_EN_MB | + SUN4I_TCON0_LVDS_ANA0_PD | + SUN4I_TCON0_LVDS_ANA0_DCHS); + + udelay(2); /* delay at least 1200 ns */ + regmap_update_bits(tcon->regs, SUN4I_TCON0_LVDS_ANA1_REG, + SUN4I_TCON0_LVDS_ANA1_INIT, + SUN4I_TCON0_LVDS_ANA1_INIT); + udelay(1); /* delay at least 120 ns */ + regmap_update_bits(tcon->regs, SUN4I_TCON0_LVDS_ANA1_REG, + SUN4I_TCON0_LVDS_ANA1_UPDATE, + SUN4I_TCON0_LVDS_ANA1_UPDATE); + regmap_update_bits(tcon->regs, SUN4I_TCON0_LVDS_ANA0_REG, + SUN4I_TCON0_LVDS_ANA0_EN_MB, + SUN4I_TCON0_LVDS_ANA0_EN_MB); +} + +static void sun6i_tcon_setup_lvds_phy(struct sun4i_tcon *tcon, + const struct drm_encoder *encoder) +{ + u8 val; + + regmap_write(tcon->regs, SUN4I_TCON0_LVDS_ANA0_REG, + SUN6I_TCON0_LVDS_ANA0_C(2) | + SUN6I_TCON0_LVDS_ANA0_V(3) | + SUN6I_TCON0_LVDS_ANA0_PD(2) | + SUN6I_TCON0_LVDS_ANA0_EN_LDO); + udelay(2); + + regmap_update_bits(tcon->regs, SUN4I_TCON0_LVDS_ANA0_REG, + SUN6I_TCON0_LVDS_ANA0_EN_MB, + SUN6I_TCON0_LVDS_ANA0_EN_MB); + udelay(2); + + regmap_update_bits(tcon->regs, SUN4I_TCON0_LVDS_ANA0_REG, + SUN6I_TCON0_LVDS_ANA0_EN_DRVC, + SUN6I_TCON0_LVDS_ANA0_EN_DRVC); + + if (sun4i_tcon_get_pixel_depth(encoder) == 18) + val = 7; + else + val = 0xf; + + regmap_write_bits(tcon->regs, SUN4I_TCON0_LVDS_ANA0_REG, + SUN6I_TCON0_LVDS_ANA0_EN_DRVD(0xf), + SUN6I_TCON0_LVDS_ANA0_EN_DRVD(val)); +} + static void sun4i_tcon_lvds_set_status(struct sun4i_tcon *tcon, const struct drm_encoder *encoder, bool enabled) { if (enabled) { - u8 val; - regmap_update_bits(tcon->regs, SUN4I_TCON0_LVDS_IF_REG, SUN4I_TCON0_LVDS_IF_EN, SUN4I_TCON0_LVDS_IF_EN); - - /* - * As their name suggest, these values only apply to the A31 - * and later SoCs. We'll have to rework this when merging - * support for the older SoCs. - */ - regmap_write(tcon->regs, SUN4I_TCON0_LVDS_ANA0_REG, - SUN6I_TCON0_LVDS_ANA0_C(2) | - SUN6I_TCON0_LVDS_ANA0_V(3) | - SUN6I_TCON0_LVDS_ANA0_PD(2) | - SUN6I_TCON0_LVDS_ANA0_EN_LDO); - udelay(2); - - regmap_update_bits(tcon->regs, SUN4I_TCON0_LVDS_ANA0_REG, - SUN6I_TCON0_LVDS_ANA0_EN_MB, - SUN6I_TCON0_LVDS_ANA0_EN_MB); - udelay(2); - - regmap_update_bits(tcon->regs, SUN4I_TCON0_LVDS_ANA0_REG, - SUN6I_TCON0_LVDS_ANA0_EN_DRVC, - SUN6I_TCON0_LVDS_ANA0_EN_DRVC); - - if (sun4i_tcon_get_pixel_depth(encoder) == 18) - val = 7; - else - val = 0xf; - - regmap_write_bits(tcon->regs, SUN4I_TCON0_LVDS_ANA0_REG, - SUN6I_TCON0_LVDS_ANA0_EN_DRVD(0xf), - SUN6I_TCON0_LVDS_ANA0_EN_DRVD(val)); + if (tcon->quirks->setup_lvds_phy) + tcon->quirks->setup_lvds_phy(tcon, encoder); } else { regmap_update_bits(tcon->regs, SUN4I_TCON0_LVDS_IF_REG, SUN4I_TCON0_LVDS_IF_EN, 0); @@ -1453,6 +1478,16 @@ static const struct sun4i_tcon_quirks sun6i_a31s_quirks = { .dclk_min_div = 1, }; +static const struct sun4i_tcon_quirks sun7i_a20_tcon0_quirks = { + .supports_lvds = true, + .has_channel_0 = true, + .has_channel_1 = true, + .dclk_min_div = 4, + /* Same display pipeline structure as A10 */ + .set_mux = sun4i_a10_tcon_set_mux, + .setup_lvds_phy = sun4i_tcon_setup_lvds_phy, +}; + static const struct sun4i_tcon_quirks sun7i_a20_quirks = { .has_channel_0 = true, .has_channel_1 = true, @@ -1465,12 +1500,15 @@ static const struct sun4i_tcon_quirks sun8i_a33_quirks = { .has_channel_0 = true, .has_lvds_alt = true, .dclk_min_div = 1, + .setup_lvds_phy = sun6i_tcon_setup_lvds_phy, + .supports_lvds = true, }; static const struct sun4i_tcon_quirks sun8i_a83t_lcd_quirks = { .supports_lvds = true, .has_channel_0 = true, .dclk_min_div = 1, + .setup_lvds_phy = sun6i_tcon_setup_lvds_phy, }; static const struct sun4i_tcon_quirks sun8i_a83t_tv_quirks = { @@ -1505,6 +1543,8 @@ const struct of_device_id sun4i_tcon_of_table[] = { { .compatible = "allwinner,sun6i-a31-tcon", .data = &sun6i_a31_quirks }, { .compatible = "allwinner,sun6i-a31s-tcon", .data = &sun6i_a31s_quirks }, { .compatible = "allwinner,sun7i-a20-tcon", .data = &sun7i_a20_quirks }, + { .compatible = "allwinner,sun7i-a20-tcon0", .data = &sun7i_a20_tcon0_quirks }, + { .compatible = "allwinner,sun7i-a20-tcon1", .data = &sun7i_a20_quirks }, { .compatible = "allwinner,sun8i-a23-tcon", .data = &sun8i_a33_quirks }, { .compatible = "allwinner,sun8i-a33-tcon", .data = &sun8i_a33_quirks }, { .compatible = "allwinner,sun8i-a83t-tcon-lcd", .data = &sun8i_a83t_lcd_quirks }, diff --git a/drivers/gpu/drm/sun4i/sun4i_tcon.h b/drivers/gpu/drm/sun4i/sun4i_tcon.h index a62ec826ae71..cfbf4e6c1679 100644 --- a/drivers/gpu/drm/sun4i/sun4i_tcon.h +++ b/drivers/gpu/drm/sun4i/sun4i_tcon.h @@ -193,6 +193,13 @@ #define SUN4I_TCON_MUX_CTRL_REG 0x200 #define SUN4I_TCON0_LVDS_ANA0_REG 0x220 +#define SUN4I_TCON0_LVDS_ANA0_DCHS BIT(16) +#define SUN4I_TCON0_LVDS_ANA0_PD (BIT(20) | BIT(21)) +#define SUN4I_TCON0_LVDS_ANA0_EN_MB BIT(22) +#define SUN4I_TCON0_LVDS_ANA0_REG_C (BIT(24) | BIT(25)) +#define SUN4I_TCON0_LVDS_ANA0_REG_V (BIT(26) | BIT(27)) +#define SUN4I_TCON0_LVDS_ANA0_CK_EN (BIT(29) | BIT(28)) + #define SUN6I_TCON0_LVDS_ANA0_EN_MB BIT(31) #define SUN6I_TCON0_LVDS_ANA0_EN_LDO BIT(30) #define SUN6I_TCON0_LVDS_ANA0_EN_DRVC BIT(24) @@ -201,6 +208,10 @@ #define SUN6I_TCON0_LVDS_ANA0_V(x) (((x) & 3) << 8) #define SUN6I_TCON0_LVDS_ANA0_PD(x) (((x) & 3) << 4) +#define SUN4I_TCON0_LVDS_ANA1_REG 0x224 +#define SUN4I_TCON0_LVDS_ANA1_INIT (0x1f << 26 | 0x1f << 10) +#define SUN4I_TCON0_LVDS_ANA1_UPDATE (0x1f << 16 | 0x1f << 00) + #define SUN4I_TCON1_FILL_CTL_REG 0x300 #define SUN4I_TCON1_FILL_BEG0_REG 0x304 #define SUN4I_TCON1_FILL_END0_REG 0x308 @@ -228,6 +239,9 @@ struct sun4i_tcon_quirks { /* callback to handle tcon muxing options */ int (*set_mux)(struct sun4i_tcon *, const struct drm_encoder *); + /* handler for LVDS setup routine */ + void (*setup_lvds_phy)(struct sun4i_tcon *tcon, + const struct drm_encoder *encoder); }; struct sun4i_tcon { diff --git a/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c b/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c index a75fcb113172..059939789730 100644 --- a/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c +++ b/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c @@ -14,7 +14,6 @@ #include <linux/phy/phy-mipi-dphy.h> #include <linux/phy/phy.h> #include <linux/platform_device.h> -#include <linux/pm_runtime.h> #include <linux/regmap.h> #include <linux/regulator/consumer.h> #include <linux/reset.h> @@ -27,7 +26,6 @@ #include <drm/drm_probe_helper.h> #include "sun4i_crtc.h" -#include "sun4i_drv.h" #include "sun4i_tcon.h" #include "sun6i_mipi_dsi.h" @@ -722,10 +720,31 @@ static void sun6i_dsi_encoder_enable(struct drm_encoder *encoder) union phy_configure_opts opts = { 0 }; struct phy_configure_opts_mipi_dphy *cfg = &opts.mipi_dphy; u16 delay; + int err; DRM_DEBUG_DRIVER("Enabling DSI output\n"); - pm_runtime_get_sync(dsi->dev); + err = regulator_enable(dsi->regulator); + if (err) + dev_warn(dsi->dev, "failed to enable VCC-DSI supply: %d\n", err); + + reset_control_deassert(dsi->reset); + clk_prepare_enable(dsi->mod_clk); + + /* + * Enable the DSI block. + */ + regmap_write(dsi->regs, SUN6I_DSI_CTL_REG, SUN6I_DSI_CTL_EN); + + regmap_write(dsi->regs, SUN6I_DSI_BASIC_CTL0_REG, + SUN6I_DSI_BASIC_CTL0_ECC_EN | SUN6I_DSI_BASIC_CTL0_CRC_EN); + + regmap_write(dsi->regs, SUN6I_DSI_TRANS_START_REG, 10); + regmap_write(dsi->regs, SUN6I_DSI_TRANS_ZERO_REG, 0); + + sun6i_dsi_inst_init(dsi, dsi->device); + + regmap_write(dsi->regs, SUN6I_DSI_DEBUG_DATA_REG, 0xff); delay = sun6i_dsi_get_video_start_delay(dsi, mode); regmap_write(dsi->regs, SUN6I_DSI_BASIC_CTL1_REG, @@ -749,7 +768,7 @@ static void sun6i_dsi_encoder_enable(struct drm_encoder *encoder) phy_configure(dsi->dphy, &opts); phy_power_on(dsi->dphy); - if (!IS_ERR(dsi->panel)) + if (dsi->panel) drm_panel_prepare(dsi->panel); /* @@ -764,7 +783,7 @@ static void sun6i_dsi_encoder_enable(struct drm_encoder *encoder) * ordering on the panels I've tested it with, so I guess this * will do for now, until that IP is better understood. */ - if (!IS_ERR(dsi->panel)) + if (dsi->panel) drm_panel_enable(dsi->panel); sun6i_dsi_start(dsi, DSI_START_HSC); @@ -780,7 +799,7 @@ static void sun6i_dsi_encoder_disable(struct drm_encoder *encoder) DRM_DEBUG_DRIVER("Disabling DSI output\n"); - if (!IS_ERR(dsi->panel)) { + if (dsi->panel) { drm_panel_disable(dsi->panel); drm_panel_unprepare(dsi->panel); } @@ -788,7 +807,9 @@ static void sun6i_dsi_encoder_disable(struct drm_encoder *encoder) phy_power_off(dsi->dphy); phy_exit(dsi->dphy); - pm_runtime_put(dsi->dev); + clk_disable_unprepare(dsi->mod_clk); + reset_control_assert(dsi->reset); + regulator_disable(dsi->regulator); } static int sun6i_dsi_get_modes(struct drm_connector *connector) @@ -805,7 +826,10 @@ static struct drm_connector_helper_funcs sun6i_dsi_connector_helper_funcs = { static enum drm_connector_status sun6i_dsi_connector_detect(struct drm_connector *connector, bool force) { - return connector_status_connected; + struct sun6i_dsi *dsi = connector_to_sun6i_dsi(connector); + + return dsi->panel ? connector_status_connected : + connector_status_disconnected; } static const struct drm_connector_funcs sun6i_dsi_connector_funcs = { @@ -942,11 +966,18 @@ static int sun6i_dsi_attach(struct mipi_dsi_host *host, struct mipi_dsi_device *device) { struct sun6i_dsi *dsi = host_to_sun6i_dsi(host); + struct drm_panel *panel = of_drm_find_panel(device->dev.of_node); + if (IS_ERR(panel)) + return PTR_ERR(panel); + if (!dsi->drm || !dsi->drm->registered) + return -EPROBE_DEFER; + + dsi->panel = panel; dsi->device = device; - dsi->panel = of_drm_find_panel(device->dev.of_node); - if (IS_ERR(dsi->panel)) - return PTR_ERR(dsi->panel); + + drm_panel_attach(dsi->panel, &dsi->connector); + drm_kms_helper_hotplug_event(dsi->drm); dev_info(host->dev, "Attached device %s\n", device->name); @@ -957,10 +988,14 @@ static int sun6i_dsi_detach(struct mipi_dsi_host *host, struct mipi_dsi_device *device) { struct sun6i_dsi *dsi = host_to_sun6i_dsi(host); + struct drm_panel *panel = dsi->panel; dsi->panel = NULL; dsi->device = NULL; + drm_panel_detach(panel); + drm_kms_helper_hotplug_event(dsi->drm); + return 0; } @@ -1022,15 +1057,9 @@ static int sun6i_dsi_bind(struct device *dev, struct device *master, void *data) { struct drm_device *drm = data; - struct sun4i_drv *drv = drm->dev_private; struct sun6i_dsi *dsi = dev_get_drvdata(dev); int ret; - if (!dsi->panel) - return -EPROBE_DEFER; - - dsi->drv = drv; - drm_encoder_helper_add(&dsi->encoder, &sun6i_dsi_enc_helper_funcs); ret = drm_encoder_init(drm, @@ -1056,7 +1085,8 @@ static int sun6i_dsi_bind(struct device *dev, struct device *master, } drm_connector_attach_encoder(&dsi->connector, &dsi->encoder); - drm_panel_attach(dsi->panel, &dsi->connector); + + dsi->drm = drm; return 0; @@ -1070,7 +1100,7 @@ static void sun6i_dsi_unbind(struct device *dev, struct device *master, { struct sun6i_dsi *dsi = dev_get_drvdata(dev); - drm_panel_detach(dsi->panel); + dsi->drm = NULL; } static const struct component_ops sun6i_dsi_ops = { @@ -1157,12 +1187,10 @@ static int sun6i_dsi_probe(struct platform_device *pdev) goto err_unprotect_clk; } - pm_runtime_enable(dev); - ret = mipi_dsi_host_register(&dsi->host); if (ret) { dev_err(dev, "Couldn't register MIPI-DSI host\n"); - goto err_pm_disable; + goto err_unprotect_clk; } ret = component_add(&pdev->dev, &sun6i_dsi_ops); @@ -1175,8 +1203,6 @@ static int sun6i_dsi_probe(struct platform_device *pdev) err_remove_dsi_host: mipi_dsi_host_unregister(&dsi->host); -err_pm_disable: - pm_runtime_disable(dev); err_unprotect_clk: clk_rate_exclusive_put(dsi->mod_clk); err_attach_clk: @@ -1192,7 +1218,6 @@ static int sun6i_dsi_remove(struct platform_device *pdev) component_del(&pdev->dev, &sun6i_dsi_ops); mipi_dsi_host_unregister(&dsi->host); - pm_runtime_disable(dev); clk_rate_exclusive_put(dsi->mod_clk); if (!IS_ERR(dsi->bus_clk)) @@ -1201,59 +1226,6 @@ static int sun6i_dsi_remove(struct platform_device *pdev) return 0; } -static int __maybe_unused sun6i_dsi_runtime_resume(struct device *dev) -{ - struct sun6i_dsi *dsi = dev_get_drvdata(dev); - int err; - - err = regulator_enable(dsi->regulator); - if (err) { - dev_err(dsi->dev, "failed to enable VCC-DSI supply: %d\n", err); - return err; - } - - reset_control_deassert(dsi->reset); - clk_prepare_enable(dsi->mod_clk); - - /* - * Enable the DSI block. - * - * Some part of it can only be done once we get a number of - * lanes, see sun6i_dsi_inst_init - */ - regmap_write(dsi->regs, SUN6I_DSI_CTL_REG, SUN6I_DSI_CTL_EN); - - regmap_write(dsi->regs, SUN6I_DSI_BASIC_CTL0_REG, - SUN6I_DSI_BASIC_CTL0_ECC_EN | SUN6I_DSI_BASIC_CTL0_CRC_EN); - - regmap_write(dsi->regs, SUN6I_DSI_TRANS_START_REG, 10); - regmap_write(dsi->regs, SUN6I_DSI_TRANS_ZERO_REG, 0); - - if (dsi->device) - sun6i_dsi_inst_init(dsi, dsi->device); - - regmap_write(dsi->regs, SUN6I_DSI_DEBUG_DATA_REG, 0xff); - - return 0; -} - -static int __maybe_unused sun6i_dsi_runtime_suspend(struct device *dev) -{ - struct sun6i_dsi *dsi = dev_get_drvdata(dev); - - clk_disable_unprepare(dsi->mod_clk); - reset_control_assert(dsi->reset); - regulator_disable(dsi->regulator); - - return 0; -} - -static const struct dev_pm_ops sun6i_dsi_pm_ops = { - SET_RUNTIME_PM_OPS(sun6i_dsi_runtime_suspend, - sun6i_dsi_runtime_resume, - NULL) -}; - static const struct of_device_id sun6i_dsi_of_table[] = { { .compatible = "allwinner,sun6i-a31-mipi-dsi" }, { .compatible = "allwinner,sun50i-a64-mipi-dsi" }, @@ -1267,7 +1239,6 @@ static struct platform_driver sun6i_dsi_platform_driver = { .driver = { .name = "sun6i-mipi-dsi", .of_match_table = sun6i_dsi_of_table, - .pm = &sun6i_dsi_pm_ops, }, }; module_platform_driver(sun6i_dsi_platform_driver); diff --git a/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.h b/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.h index 3f4846f581ef..c863900ae3b4 100644 --- a/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.h +++ b/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.h @@ -28,8 +28,8 @@ struct sun6i_dsi { struct phy *dphy; struct device *dev; - struct sun4i_drv *drv; struct mipi_dsi_device *device; + struct drm_device *drm; struct drm_panel *panel; }; diff --git a/drivers/gpu/drm/sun4i/sun8i_mixer.c b/drivers/gpu/drm/sun4i/sun8i_mixer.c index 7c24f8f832a5..4a64f7ae437a 100644 --- a/drivers/gpu/drm/sun4i/sun8i_mixer.c +++ b/drivers/gpu/drm/sun4i/sun8i_mixer.c @@ -107,48 +107,128 @@ static const struct de2_fmt_info de2_formats[] = { .csc = SUN8I_CSC_MODE_OFF, }, { + /* for DE2 VI layer which ignores alpha */ + .drm_fmt = DRM_FORMAT_XRGB4444, + .de2_fmt = SUN8I_MIXER_FBFMT_ARGB4444, + .rgb = true, + .csc = SUN8I_CSC_MODE_OFF, + }, + { .drm_fmt = DRM_FORMAT_ABGR4444, .de2_fmt = SUN8I_MIXER_FBFMT_ABGR4444, .rgb = true, .csc = SUN8I_CSC_MODE_OFF, }, { + /* for DE2 VI layer which ignores alpha */ + .drm_fmt = DRM_FORMAT_XBGR4444, + .de2_fmt = SUN8I_MIXER_FBFMT_ABGR4444, + .rgb = true, + .csc = SUN8I_CSC_MODE_OFF, + }, + { .drm_fmt = DRM_FORMAT_RGBA4444, .de2_fmt = SUN8I_MIXER_FBFMT_RGBA4444, .rgb = true, .csc = SUN8I_CSC_MODE_OFF, }, { + /* for DE2 VI layer which ignores alpha */ + .drm_fmt = DRM_FORMAT_RGBX4444, + .de2_fmt = SUN8I_MIXER_FBFMT_RGBA4444, + .rgb = true, + .csc = SUN8I_CSC_MODE_OFF, + }, + { .drm_fmt = DRM_FORMAT_BGRA4444, .de2_fmt = SUN8I_MIXER_FBFMT_BGRA4444, .rgb = true, .csc = SUN8I_CSC_MODE_OFF, }, { + /* for DE2 VI layer which ignores alpha */ + .drm_fmt = DRM_FORMAT_BGRX4444, + .de2_fmt = SUN8I_MIXER_FBFMT_BGRA4444, + .rgb = true, + .csc = SUN8I_CSC_MODE_OFF, + }, + { .drm_fmt = DRM_FORMAT_ARGB1555, .de2_fmt = SUN8I_MIXER_FBFMT_ARGB1555, .rgb = true, .csc = SUN8I_CSC_MODE_OFF, }, { + /* for DE2 VI layer which ignores alpha */ + .drm_fmt = DRM_FORMAT_XRGB1555, + .de2_fmt = SUN8I_MIXER_FBFMT_ARGB1555, + .rgb = true, + .csc = SUN8I_CSC_MODE_OFF, + }, + { .drm_fmt = DRM_FORMAT_ABGR1555, .de2_fmt = SUN8I_MIXER_FBFMT_ABGR1555, .rgb = true, .csc = SUN8I_CSC_MODE_OFF, }, { + /* for DE2 VI layer which ignores alpha */ + .drm_fmt = DRM_FORMAT_XBGR1555, + .de2_fmt = SUN8I_MIXER_FBFMT_ABGR1555, + .rgb = true, + .csc = SUN8I_CSC_MODE_OFF, + }, + { .drm_fmt = DRM_FORMAT_RGBA5551, .de2_fmt = SUN8I_MIXER_FBFMT_RGBA5551, .rgb = true, .csc = SUN8I_CSC_MODE_OFF, }, { + /* for DE2 VI layer which ignores alpha */ + .drm_fmt = DRM_FORMAT_RGBX5551, + .de2_fmt = SUN8I_MIXER_FBFMT_RGBA5551, + .rgb = true, + .csc = SUN8I_CSC_MODE_OFF, + }, + { .drm_fmt = DRM_FORMAT_BGRA5551, .de2_fmt = SUN8I_MIXER_FBFMT_BGRA5551, .rgb = true, .csc = SUN8I_CSC_MODE_OFF, }, { + /* for DE2 VI layer which ignores alpha */ + .drm_fmt = DRM_FORMAT_BGRX5551, + .de2_fmt = SUN8I_MIXER_FBFMT_BGRA5551, + .rgb = true, + .csc = SUN8I_CSC_MODE_OFF, + }, + { + .drm_fmt = DRM_FORMAT_ARGB2101010, + .de2_fmt = SUN8I_MIXER_FBFMT_ARGB2101010, + .rgb = true, + .csc = SUN8I_CSC_MODE_OFF, + }, + { + .drm_fmt = DRM_FORMAT_ABGR2101010, + .de2_fmt = SUN8I_MIXER_FBFMT_ABGR2101010, + .rgb = true, + .csc = SUN8I_CSC_MODE_OFF, + }, + { + .drm_fmt = DRM_FORMAT_RGBA1010102, + .de2_fmt = SUN8I_MIXER_FBFMT_RGBA1010102, + .rgb = true, + .csc = SUN8I_CSC_MODE_OFF, + }, + { + .drm_fmt = DRM_FORMAT_BGRA1010102, + .de2_fmt = SUN8I_MIXER_FBFMT_BGRA1010102, + .rgb = true, + .csc = SUN8I_CSC_MODE_OFF, + }, + { .drm_fmt = DRM_FORMAT_UYVY, .de2_fmt = SUN8I_MIXER_FBFMT_UYVY, .rgb = false, @@ -197,12 +277,6 @@ static const struct de2_fmt_info de2_formats[] = { .csc = SUN8I_CSC_MODE_YUV2RGB, }, { - .drm_fmt = DRM_FORMAT_YUV444, - .de2_fmt = SUN8I_MIXER_FBFMT_RGB888, - .rgb = true, - .csc = SUN8I_CSC_MODE_YUV2RGB, - }, - { .drm_fmt = DRM_FORMAT_YUV422, .de2_fmt = SUN8I_MIXER_FBFMT_YUV422, .rgb = false, @@ -221,12 +295,6 @@ static const struct de2_fmt_info de2_formats[] = { .csc = SUN8I_CSC_MODE_YUV2RGB, }, { - .drm_fmt = DRM_FORMAT_YVU444, - .de2_fmt = SUN8I_MIXER_FBFMT_RGB888, - .rgb = true, - .csc = SUN8I_CSC_MODE_YVU2RGB, - }, - { .drm_fmt = DRM_FORMAT_YVU422, .de2_fmt = SUN8I_MIXER_FBFMT_YUV422, .rgb = false, @@ -244,6 +312,18 @@ static const struct de2_fmt_info de2_formats[] = { .rgb = false, .csc = SUN8I_CSC_MODE_YVU2RGB, }, + { + .drm_fmt = DRM_FORMAT_P010, + .de2_fmt = SUN8I_MIXER_FBFMT_P010_YUV, + .rgb = false, + .csc = SUN8I_CSC_MODE_YUV2RGB, + }, + { + .drm_fmt = DRM_FORMAT_P210, + .de2_fmt = SUN8I_MIXER_FBFMT_P210_YUV, + .rgb = false, + .csc = SUN8I_CSC_MODE_YUV2RGB, + }, }; const struct de2_fmt_info *sun8i_mixer_format_info(u32 format) diff --git a/drivers/gpu/drm/sun4i/sun8i_mixer.h b/drivers/gpu/drm/sun4i/sun8i_mixer.h index c6cc94057faf..345b28b0a80a 100644 --- a/drivers/gpu/drm/sun4i/sun8i_mixer.h +++ b/drivers/gpu/drm/sun4i/sun8i_mixer.h @@ -93,6 +93,10 @@ #define SUN8I_MIXER_FBFMT_ABGR1555 17 #define SUN8I_MIXER_FBFMT_RGBA5551 18 #define SUN8I_MIXER_FBFMT_BGRA5551 19 +#define SUN8I_MIXER_FBFMT_ARGB2101010 20 +#define SUN8I_MIXER_FBFMT_ABGR2101010 21 +#define SUN8I_MIXER_FBFMT_RGBA1010102 22 +#define SUN8I_MIXER_FBFMT_BGRA1010102 23 #define SUN8I_MIXER_FBFMT_YUYV 0 #define SUN8I_MIXER_FBFMT_UYVY 1 @@ -109,6 +113,13 @@ /* format 12 is semi-planar YUV411 UVUV */ /* format 13 is semi-planar YUV411 VUVU */ #define SUN8I_MIXER_FBFMT_YUV411 14 +/* format 15 doesn't exist */ +/* format 16 is P010 YVU */ +#define SUN8I_MIXER_FBFMT_P010_YUV 17 +/* format 18 is P210 YVU */ +#define SUN8I_MIXER_FBFMT_P210_YUV 19 +/* format 20 is packed YVU444 10-bit */ +/* format 21 is packed YUV444 10-bit */ /* * Sub-engines listed bellow are unused for now. The EN registers are here only diff --git a/drivers/gpu/drm/sun4i/sun8i_vi_layer.c b/drivers/gpu/drm/sun4i/sun8i_vi_layer.c index 42d445d23773..b8398ca18b0f 100644 --- a/drivers/gpu/drm/sun4i/sun8i_vi_layer.c +++ b/drivers/gpu/drm/sun4i/sun8i_vi_layer.c @@ -398,24 +398,66 @@ static const struct drm_plane_funcs sun8i_vi_layer_funcs = { }; /* - * While all RGB formats are supported, VI planes don't support - * alpha blending, so there is no point having formats with alpha - * channel if their opaque analog exist. + * While DE2 VI layer supports same RGB formats as UI layer, alpha + * channel is ignored. This structure lists all unique variants + * where alpha channel is replaced with "don't care" (X) channel. */ static const u32 sun8i_vi_layer_formats[] = { + DRM_FORMAT_BGR565, + DRM_FORMAT_BGR888, + DRM_FORMAT_BGRX4444, + DRM_FORMAT_BGRX5551, + DRM_FORMAT_BGRX8888, + DRM_FORMAT_RGB565, + DRM_FORMAT_RGB888, + DRM_FORMAT_RGBX4444, + DRM_FORMAT_RGBX5551, + DRM_FORMAT_RGBX8888, + DRM_FORMAT_XBGR1555, + DRM_FORMAT_XBGR4444, + DRM_FORMAT_XBGR8888, + DRM_FORMAT_XRGB1555, + DRM_FORMAT_XRGB4444, + DRM_FORMAT_XRGB8888, + + DRM_FORMAT_NV16, + DRM_FORMAT_NV12, + DRM_FORMAT_NV21, + DRM_FORMAT_NV61, + DRM_FORMAT_UYVY, + DRM_FORMAT_VYUY, + DRM_FORMAT_YUYV, + DRM_FORMAT_YVYU, + DRM_FORMAT_YUV411, + DRM_FORMAT_YUV420, + DRM_FORMAT_YUV422, + DRM_FORMAT_YVU411, + DRM_FORMAT_YVU420, + DRM_FORMAT_YVU422, +}; + +static const u32 sun8i_vi_layer_de3_formats[] = { DRM_FORMAT_ABGR1555, + DRM_FORMAT_ABGR2101010, DRM_FORMAT_ABGR4444, + DRM_FORMAT_ABGR8888, DRM_FORMAT_ARGB1555, + DRM_FORMAT_ARGB2101010, DRM_FORMAT_ARGB4444, + DRM_FORMAT_ARGB8888, DRM_FORMAT_BGR565, DRM_FORMAT_BGR888, + DRM_FORMAT_BGRA1010102, DRM_FORMAT_BGRA5551, DRM_FORMAT_BGRA4444, + DRM_FORMAT_BGRA8888, DRM_FORMAT_BGRX8888, DRM_FORMAT_RGB565, DRM_FORMAT_RGB888, + DRM_FORMAT_RGBA1010102, DRM_FORMAT_RGBA4444, DRM_FORMAT_RGBA5551, + DRM_FORMAT_RGBA8888, DRM_FORMAT_RGBX8888, DRM_FORMAT_XBGR8888, DRM_FORMAT_XRGB8888, @@ -424,6 +466,8 @@ static const u32 sun8i_vi_layer_formats[] = { DRM_FORMAT_NV12, DRM_FORMAT_NV21, DRM_FORMAT_NV61, + DRM_FORMAT_P010, + DRM_FORMAT_P210, DRM_FORMAT_UYVY, DRM_FORMAT_VYUY, DRM_FORMAT_YUYV, @@ -431,11 +475,9 @@ static const u32 sun8i_vi_layer_formats[] = { DRM_FORMAT_YUV411, DRM_FORMAT_YUV420, DRM_FORMAT_YUV422, - DRM_FORMAT_YUV444, DRM_FORMAT_YVU411, DRM_FORMAT_YVU420, DRM_FORMAT_YVU422, - DRM_FORMAT_YVU444, }; struct sun8i_vi_layer *sun8i_vi_layer_init_one(struct drm_device *drm, @@ -443,19 +485,27 @@ struct sun8i_vi_layer *sun8i_vi_layer_init_one(struct drm_device *drm, int index) { u32 supported_encodings, supported_ranges; + unsigned int plane_cnt, format_count; struct sun8i_vi_layer *layer; - unsigned int plane_cnt; + const u32 *formats; int ret; layer = devm_kzalloc(drm->dev, sizeof(*layer), GFP_KERNEL); if (!layer) return ERR_PTR(-ENOMEM); + if (mixer->cfg->is_de3) { + formats = sun8i_vi_layer_de3_formats; + format_count = ARRAY_SIZE(sun8i_vi_layer_de3_formats); + } else { + formats = sun8i_vi_layer_formats; + format_count = ARRAY_SIZE(sun8i_vi_layer_formats); + } + /* possible crtcs are set later */ ret = drm_universal_plane_init(drm, &layer->plane, 0, &sun8i_vi_layer_funcs, - sun8i_vi_layer_formats, - ARRAY_SIZE(sun8i_vi_layer_formats), + formats, format_count, NULL, DRM_PLANE_TYPE_OVERLAY, NULL); if (ret) { dev_err(drm->dev, "Couldn't initialize layer\n"); diff --git a/drivers/gpu/drm/tegra/dc.c b/drivers/gpu/drm/tegra/dc.c index 7c70fd31a4c2..1a7b08f35776 100644 --- a/drivers/gpu/drm/tegra/dc.c +++ b/drivers/gpu/drm/tegra/dc.c @@ -2503,7 +2503,6 @@ static int tegra_dc_couple(struct tegra_dc *dc) static int tegra_dc_probe(struct platform_device *pdev) { - struct resource *regs; struct tegra_dc *dc; int err; @@ -2560,8 +2559,7 @@ static int tegra_dc_probe(struct platform_device *pdev) tegra_powergate_power_off(dc->powergate); } - regs = platform_get_resource(pdev, IORESOURCE_MEM, 0); - dc->regs = devm_ioremap_resource(&pdev->dev, regs); + dc->regs = devm_platform_ioremap_resource(pdev, 0); if (IS_ERR(dc->regs)) return PTR_ERR(dc->regs); @@ -2573,7 +2571,13 @@ static int tegra_dc_probe(struct platform_device *pdev) err = tegra_dc_rgb_probe(dc); if (err < 0 && err != -ENODEV) { - dev_err(&pdev->dev, "failed to probe RGB output: %d\n", err); + const char *level = KERN_ERR; + + if (err == -EPROBE_DEFER) + level = KERN_DEBUG; + + dev_printk(level, dc->dev, "failed to probe RGB output: %d\n", + err); return err; } @@ -2588,10 +2592,16 @@ static int tegra_dc_probe(struct platform_device *pdev) if (err < 0) { dev_err(&pdev->dev, "failed to register host1x client: %d\n", err); - return err; + goto disable_pm; } return 0; + +disable_pm: + pm_runtime_disable(&pdev->dev); + tegra_dc_rgb_remove(dc); + + return err; } static int tegra_dc_remove(struct platform_device *pdev) diff --git a/drivers/gpu/drm/tegra/fb.c b/drivers/gpu/drm/tegra/fb.c index 84f0e01e3428..b8a328f53862 100644 --- a/drivers/gpu/drm/tegra/fb.c +++ b/drivers/gpu/drm/tegra/fb.c @@ -314,19 +314,13 @@ static int tegra_fbdev_init(struct tegra_fbdev *fbdev, struct drm_device *drm = fbdev->base.dev; int err; - err = drm_fb_helper_init(drm, &fbdev->base, max_connectors); + err = drm_fb_helper_init(drm, &fbdev->base); if (err < 0) { dev_err(drm->dev, "failed to initialize DRM FB helper: %d\n", err); return err; } - err = drm_fb_helper_single_add_all_connectors(&fbdev->base); - if (err < 0) { - dev_err(drm->dev, "failed to add connectors: %d\n", err); - goto fini; - } - err = drm_fb_helper_initial_config(&fbdev->base, preferred_bpp); if (err < 0) { dev_err(drm->dev, "failed to set initial configuration: %d\n", diff --git a/drivers/gpu/drm/tegra/hdmi.c b/drivers/gpu/drm/tegra/hdmi.c index 6f117628f257..38252c0f068d 100644 --- a/drivers/gpu/drm/tegra/hdmi.c +++ b/drivers/gpu/drm/tegra/hdmi.c @@ -1648,6 +1648,7 @@ static irqreturn_t tegra_hdmi_irq(int irq, void *data) static int tegra_hdmi_probe(struct platform_device *pdev) { + const char *level = KERN_ERR; struct tegra_hdmi *hdmi; struct resource *regs; int err; @@ -1686,21 +1687,36 @@ static int tegra_hdmi_probe(struct platform_device *pdev) } hdmi->hdmi = devm_regulator_get(&pdev->dev, "hdmi"); - if (IS_ERR(hdmi->hdmi)) { - dev_err(&pdev->dev, "failed to get HDMI regulator\n"); - return PTR_ERR(hdmi->hdmi); + err = PTR_ERR_OR_ZERO(hdmi->hdmi); + if (err) { + if (err == -EPROBE_DEFER) + level = KERN_DEBUG; + + dev_printk(level, &pdev->dev, + "failed to get HDMI regulator: %d\n", err); + return err; } hdmi->pll = devm_regulator_get(&pdev->dev, "pll"); - if (IS_ERR(hdmi->pll)) { - dev_err(&pdev->dev, "failed to get PLL regulator\n"); - return PTR_ERR(hdmi->pll); + err = PTR_ERR_OR_ZERO(hdmi->pll); + if (err) { + if (err == -EPROBE_DEFER) + level = KERN_DEBUG; + + dev_printk(level, &pdev->dev, + "failed to get PLL regulator: %d\n", err); + return err; } hdmi->vdd = devm_regulator_get(&pdev->dev, "vdd"); - if (IS_ERR(hdmi->vdd)) { - dev_err(&pdev->dev, "failed to get VDD regulator\n"); - return PTR_ERR(hdmi->vdd); + err = PTR_ERR_OR_ZERO(hdmi->vdd); + if (err) { + if (err == -EPROBE_DEFER) + level = KERN_DEBUG; + + dev_printk(level, &pdev->dev, + "failed to get VDD regulator: %d\n", err); + return err; } hdmi->output.dev = &pdev->dev; diff --git a/drivers/gpu/drm/tidss/tidss_crtc.c b/drivers/gpu/drm/tidss/tidss_crtc.c index 032c31ee2820..d4ce9bab8c7e 100644 --- a/drivers/gpu/drm/tidss/tidss_crtc.c +++ b/drivers/gpu/drm/tidss/tidss_crtc.c @@ -17,6 +17,7 @@ #include "tidss_dispc.h" #include "tidss_drv.h" #include "tidss_irq.h" +#include "tidss_plane.h" /* Page flip and frame done IRQs */ @@ -111,6 +112,54 @@ static int tidss_crtc_atomic_check(struct drm_crtc *crtc, return dispc_vp_bus_check(dispc, hw_videoport, state); } +/* + * This needs all affected planes to be present in the atomic + * state. The untouched planes are added to the state in + * tidss_atomic_check(). + */ +static void tidss_crtc_position_planes(struct tidss_device *tidss, + struct drm_crtc *crtc, + struct drm_crtc_state *old_state, + bool newmodeset) +{ + struct drm_atomic_state *ostate = old_state->state; + struct tidss_crtc *tcrtc = to_tidss_crtc(crtc); + struct drm_crtc_state *cstate = crtc->state; + int layer; + + if (!newmodeset && !cstate->zpos_changed && + !to_tidss_crtc_state(cstate)->plane_pos_changed) + return; + + for (layer = 0; layer < tidss->feat->num_planes; layer++) { + struct drm_plane_state *pstate; + struct drm_plane *plane; + bool layer_active = false; + int i; + + for_each_new_plane_in_state(ostate, plane, pstate, i) { + if (pstate->crtc != crtc || !pstate->visible) + continue; + + if (pstate->normalized_zpos == layer) { + layer_active = true; + break; + } + } + + if (layer_active) { + struct tidss_plane *tplane = to_tidss_plane(plane); + + dispc_ovr_set_plane(tidss->dispc, tplane->hw_plane_id, + tcrtc->hw_videoport, + pstate->crtc_x, pstate->crtc_y, + layer); + } + dispc_ovr_enable_layer(tidss->dispc, tcrtc->hw_videoport, layer, + layer_active); + } +} + static void tidss_crtc_atomic_flush(struct drm_crtc *crtc, struct drm_crtc_state *old_crtc_state) { @@ -146,6 +195,9 @@ static void tidss_crtc_atomic_flush(struct drm_crtc *crtc, /* Write vp properties to HW if needed. */ dispc_vp_setup(tidss->dispc, tcrtc->hw_videoport, crtc->state, false); + /* Update plane positions if needed. */ + tidss_crtc_position_planes(tidss, crtc, old_crtc_state, false); + WARN_ON(drm_crtc_vblank_get(crtc) != 0); spin_lock_irqsave(&ddev->event_lock, flags); @@ -183,6 +235,7 @@ static void tidss_crtc_atomic_enable(struct drm_crtc *crtc, return; dispc_vp_setup(tidss->dispc, tcrtc->hw_videoport, crtc->state, true); + tidss_crtc_position_planes(tidss, crtc, old_state, true); /* Turn vertical blanking interrupt reporting on. */ drm_crtc_vblank_on(crtc); @@ -318,6 +371,8 @@ static struct drm_crtc_state *tidss_crtc_duplicate_state(struct drm_crtc *crtc) __drm_atomic_helper_crtc_duplicate_state(crtc, &state->base); + state->plane_pos_changed = false; + state->bus_format = current_state->bus_format; state->bus_flags = current_state->bus_flags; diff --git a/drivers/gpu/drm/tidss/tidss_crtc.h b/drivers/gpu/drm/tidss/tidss_crtc.h index df9d90b1ad2d..09e773666228 100644 --- a/drivers/gpu/drm/tidss/tidss_crtc.h +++ b/drivers/gpu/drm/tidss/tidss_crtc.h @@ -32,6 +32,8 @@ struct tidss_crtc_state { /* Must be first. */ struct drm_crtc_state base; + bool plane_pos_changed; + u32 bus_format; u32 bus_flags; }; diff --git a/drivers/gpu/drm/tidss/tidss_dispc.c b/drivers/gpu/drm/tidss/tidss_dispc.c index eeb160dc047b..29f42768e294 100644 --- a/drivers/gpu/drm/tidss/tidss_dispc.c +++ b/drivers/gpu/drm/tidss/tidss_dispc.c @@ -281,11 +281,6 @@ struct dss_vp_data { u32 *gamma_table; }; -struct dss_plane_data { - u32 zorder; - u32 hw_videoport; -}; - struct dispc_device { struct tidss_device *tidss; struct device *dev; @@ -307,8 +302,6 @@ struct dispc_device { struct dss_vp_data vp_data[TIDSS_MAX_PORTS]; - struct dss_plane_data plane_data[TIDSS_MAX_PLANES]; - u32 *fourccs; u32 num_fourccs; @@ -1235,7 +1228,7 @@ int dispc_vp_set_clk_rate(struct dispc_device *dispc, u32 hw_videoport, if (dispc_pclk_diff(rate, new_rate) > 5) dev_warn(dispc->dev, - "vp%d: Clock rate %lu differs over 5%% from requsted %lu\n", + "vp%d: Clock rate %lu differs over 5%% from requested %lu\n", hw_videoport, new_rate, rate); dev_dbg(dispc->dev, "vp%d: new rate %lu Hz (requested %lu Hz)\n", @@ -1247,7 +1240,7 @@ int dispc_vp_set_clk_rate(struct dispc_device *dispc, u32 hw_videoport, /* OVR */ static void dispc_k2g_ovr_set_plane(struct dispc_device *dispc, u32 hw_plane, u32 hw_videoport, - u32 x, u32 y, u32 zpos) + u32 x, u32 y, u32 layer) { /* On k2g there is only one plane and no need for ovr */ dispc_vid_write(dispc, hw_plane, DISPC_VID_K2G_POSITION, @@ -1256,44 +1249,43 @@ static void dispc_k2g_ovr_set_plane(struct dispc_device *dispc, static void dispc_am65x_ovr_set_plane(struct dispc_device *dispc, u32 hw_plane, u32 hw_videoport, - u32 x, u32 y, u32 zpos) + u32 x, u32 y, u32 layer) { - OVR_REG_FLD_MOD(dispc, hw_videoport, DISPC_OVR_ATTRIBUTES(zpos), + OVR_REG_FLD_MOD(dispc, hw_videoport, DISPC_OVR_ATTRIBUTES(layer), hw_plane, 4, 1); - OVR_REG_FLD_MOD(dispc, hw_videoport, DISPC_OVR_ATTRIBUTES(zpos), + OVR_REG_FLD_MOD(dispc, hw_videoport, DISPC_OVR_ATTRIBUTES(layer), x, 17, 6); - OVR_REG_FLD_MOD(dispc, hw_videoport, DISPC_OVR_ATTRIBUTES(zpos), + OVR_REG_FLD_MOD(dispc, hw_videoport, DISPC_OVR_ATTRIBUTES(layer), y, 30, 19); } static void dispc_j721e_ovr_set_plane(struct dispc_device *dispc, u32 hw_plane, u32 hw_videoport, - u32 x, u32 y, u32 zpos) + u32 x, u32 y, u32 layer) { - OVR_REG_FLD_MOD(dispc, hw_videoport, DISPC_OVR_ATTRIBUTES(zpos), + OVR_REG_FLD_MOD(dispc, hw_videoport, DISPC_OVR_ATTRIBUTES(layer), hw_plane, 4, 1); - OVR_REG_FLD_MOD(dispc, hw_videoport, DISPC_OVR_ATTRIBUTES2(zpos), + OVR_REG_FLD_MOD(dispc, hw_videoport, DISPC_OVR_ATTRIBUTES2(layer), x, 13, 0); - OVR_REG_FLD_MOD(dispc, hw_videoport, DISPC_OVR_ATTRIBUTES2(zpos), + OVR_REG_FLD_MOD(dispc, hw_videoport, DISPC_OVR_ATTRIBUTES2(layer), y, 29, 16); } -static void dispc_ovr_set_plane(struct dispc_device *dispc, - u32 hw_plane, u32 hw_videoport, - u32 x, u32 y, u32 zpos) +void dispc_ovr_set_plane(struct dispc_device *dispc, u32 hw_plane, + u32 hw_videoport, u32 x, u32 y, u32 layer) { switch (dispc->feat->subrev) { case DISPC_K2G: dispc_k2g_ovr_set_plane(dispc, hw_plane, hw_videoport, - x, y, zpos); + x, y, layer); break; case DISPC_AM65X: dispc_am65x_ovr_set_plane(dispc, hw_plane, hw_videoport, - x, y, zpos); + x, y, layer); break; case DISPC_J721E: dispc_j721e_ovr_set_plane(dispc, hw_plane, hw_videoport, - x, y, zpos); + x, y, layer); break; default: WARN_ON(1); @@ -1301,10 +1293,13 @@ static void dispc_ovr_set_plane(struct dispc_device *dispc, } } -static void dispc_ovr_enable_plane(struct dispc_device *dispc, - u32 hw_videoport, u32 zpos, bool enable) +void dispc_ovr_enable_layer(struct dispc_device *dispc, + u32 hw_videoport, u32 layer, bool enable) { - OVR_REG_FLD_MOD(dispc, hw_videoport, DISPC_OVR_ATTRIBUTES(zpos), + if (dispc->feat->subrev == DISPC_K2G) + return; + + OVR_REG_FLD_MOD(dispc, hw_videoport, DISPC_OVR_ATTRIBUTES(layer), !!enable, 0, 0); } @@ -1510,7 +1505,7 @@ struct dispc_csc_coef *dispc_find_csc(enum drm_color_encoding encoding, static void dispc_vid_csc_setup(struct dispc_device *dispc, u32 hw_plane, const struct drm_plane_state *state) { - static const struct dispc_csc_coef *coef; + const struct dispc_csc_coef *coef; coef = dispc_find_csc(state->color_encoding, state->color_range); if (!coef) { @@ -1699,7 +1694,7 @@ static int dispc_vid_calc_scaling(struct dispc_device *dispc, if (sp->xinc > f->xinc_max) { dev_dbg(dispc->dev, - "%s: Too wide input bufer %u > %u\n", __func__, + "%s: Too wide input buffer %u > %u\n", __func__, state->src_w >> 16, in_width_max * f->xinc_max); return -EINVAL; } @@ -2070,21 +2065,11 @@ int dispc_plane_setup(struct dispc_device *dispc, u32 hw_plane, VID_REG_FLD_MOD(dispc, hw_plane, DISPC_VID_ATTRIBUTES, 0, 28, 28); - dispc_ovr_set_plane(dispc, hw_plane, hw_videoport, - state->crtc_x, state->crtc_y, - state->normalized_zpos); - - dispc->plane_data[hw_plane].zorder = state->normalized_zpos; - dispc->plane_data[hw_plane].hw_videoport = hw_videoport; - return 0; } int dispc_plane_enable(struct dispc_device *dispc, u32 hw_plane, bool enable) { - dispc_ovr_enable_plane(dispc, dispc->plane_data[hw_plane].hw_videoport, - dispc->plane_data[hw_plane].zorder, enable); - VID_REG_FLD_MOD(dispc, hw_plane, DISPC_VID_ATTRIBUTES, !!enable, 0, 0); return 0; diff --git a/drivers/gpu/drm/tidss/tidss_dispc.h b/drivers/gpu/drm/tidss/tidss_dispc.h index e65e6a2bb821..a4a68249e44b 100644 --- a/drivers/gpu/drm/tidss/tidss_dispc.h +++ b/drivers/gpu/drm/tidss/tidss_dispc.h @@ -94,6 +94,11 @@ extern const struct dispc_features dispc_j721e_feats; void dispc_set_irqenable(struct dispc_device *dispc, dispc_irq_t mask); dispc_irq_t dispc_read_and_clear_irqstatus(struct dispc_device *dispc); +void dispc_ovr_set_plane(struct dispc_device *dispc, u32 hw_plane, + u32 hw_videoport, u32 x, u32 y, u32 layer); +void dispc_ovr_enable_layer(struct dispc_device *dispc, + u32 hw_videoport, u32 layer, bool enable); + void dispc_vp_prepare(struct dispc_device *dispc, u32 hw_videoport, const struct drm_crtc_state *state); void dispc_vp_enable(struct dispc_device *dispc, u32 hw_videoport, diff --git a/drivers/gpu/drm/tidss/tidss_encoder.c b/drivers/gpu/drm/tidss/tidss_encoder.c index f7fe3a43ead0..83785b0a66a9 100644 --- a/drivers/gpu/drm/tidss/tidss_encoder.c +++ b/drivers/gpu/drm/tidss/tidss_encoder.c @@ -32,7 +32,7 @@ static int tidss_encoder_atomic_check(struct drm_encoder *encoder, * bridge timings, or from the connector's display_info if no * bridge defines the timings. */ - list_for_each_entry(bridge, &encoder->bridge_chain, chain_node) { + drm_for_each_bridge_in_chain(encoder, bridge) { if (!bridge->timings) continue; diff --git a/drivers/gpu/drm/tidss/tidss_kms.c b/drivers/gpu/drm/tidss/tidss_kms.c index 5311e0f1c551..7d419960b030 100644 --- a/drivers/gpu/drm/tidss/tidss_kms.c +++ b/drivers/gpu/drm/tidss/tidss_kms.c @@ -47,9 +47,59 @@ static const struct drm_mode_config_helper_funcs mode_config_helper_funcs = { .atomic_commit_tail = tidss_atomic_commit_tail, }; +static int tidss_atomic_check(struct drm_device *ddev, + struct drm_atomic_state *state) +{ + struct drm_plane_state *opstate; + struct drm_plane_state *npstate; + struct drm_plane *plane; + struct drm_crtc_state *cstate; + struct drm_crtc *crtc; + int ret, i; + + ret = drm_atomic_helper_check(ddev, state); + if (ret) + return ret; + + /* + * Add all active planes on a CRTC to the atomic state, if + * x/y/z position or activity of any plane on that CRTC + * changes. This is needed for updating the plane positions in + * tidss_crtc_position_planes() which is called from + * crtc_atomic_enable() and crtc_atomic_flush(). We have an + * extra flag to to mark x,y-position changes and together + * with zpos_changed the condition recognizes all the above + * cases. + */ + for_each_oldnew_plane_in_state(state, plane, opstate, npstate, i) { + if (!npstate->crtc || !npstate->visible) + continue; + + if (!opstate->crtc || opstate->crtc_x != npstate->crtc_x || + opstate->crtc_y != npstate->crtc_y) { + cstate = drm_atomic_get_crtc_state(state, + npstate->crtc); + if (IS_ERR(cstate)) + return PTR_ERR(cstate); + to_tidss_crtc_state(cstate)->plane_pos_changed = true; + } + } + + for_each_new_crtc_in_state(state, crtc, cstate, i) { + if (to_tidss_crtc_state(cstate)->plane_pos_changed || + cstate->zpos_changed) { + ret = drm_atomic_add_affected_planes(state, crtc); + if (ret) + return ret; + } + } + + return 0; +} + static const struct drm_mode_config_funcs mode_config_funcs = { .fb_create = drm_gem_fb_create, - .atomic_check = drm_atomic_helper_check, + .atomic_check = tidss_atomic_check, .atomic_commit = drm_atomic_helper_commit, }; @@ -172,7 +222,7 @@ static int tidss_dispc_modeset_init(struct tidss_device *tidss) return PTR_ERR(enc); } - ret = drm_bridge_attach(enc, pipes[i].bridge, NULL); + ret = drm_bridge_attach(enc, pipes[i].bridge, NULL, 0); if (ret) { dev_err(tidss->dev, "bridge attach failed: %d\n", ret); return ret; diff --git a/drivers/gpu/drm/tilcdc/tilcdc_external.c b/drivers/gpu/drm/tilcdc/tilcdc_external.c index 51d034e095f4..28b7f703236e 100644 --- a/drivers/gpu/drm/tilcdc/tilcdc_external.c +++ b/drivers/gpu/drm/tilcdc/tilcdc_external.c @@ -95,7 +95,7 @@ int tilcdc_attach_bridge(struct drm_device *ddev, struct drm_bridge *bridge) priv->external_encoder->possible_crtcs = BIT(0); - ret = drm_bridge_attach(priv->external_encoder, bridge, NULL); + ret = drm_bridge_attach(priv->external_encoder, bridge, NULL, 0); if (ret) { dev_err(ddev->dev, "drm_bridge_attach() failed %d\n", ret); return ret; diff --git a/drivers/gpu/drm/tiny/ili9486.c b/drivers/gpu/drm/tiny/ili9486.c index 5084b38c1a71..532560aebb1e 100644 --- a/drivers/gpu/drm/tiny/ili9486.c +++ b/drivers/gpu/drm/tiny/ili9486.c @@ -45,7 +45,7 @@ static int waveshare_command(struct mipi_dbi *mipi, u8 *cmd, u8 *par, void *data = par; u32 speed_hz; int i, ret; - u16 *buf; + __be16 *buf; buf = kmalloc(32 * sizeof(u16), GFP_KERNEL); if (!buf) diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c index 1fbc36f05d89..9e07c3f75156 100644 --- a/drivers/gpu/drm/ttm/ttm_bo.c +++ b/drivers/gpu/drm/ttm/ttm_bo.c @@ -145,34 +145,12 @@ static inline uint32_t ttm_bo_type_flags(unsigned type) return 1 << (type); } -static void ttm_bo_release_list(struct kref *list_kref) -{ - struct ttm_buffer_object *bo = - container_of(list_kref, struct ttm_buffer_object, list_kref); - size_t acc_size = bo->acc_size; - - BUG_ON(kref_read(&bo->list_kref)); - BUG_ON(kref_read(&bo->kref)); - BUG_ON(bo->mem.mm_node != NULL); - BUG_ON(!list_empty(&bo->lru)); - BUG_ON(!list_empty(&bo->ddestroy)); - ttm_tt_destroy(bo->ttm); - atomic_dec(&ttm_bo_glob.bo_count); - dma_fence_put(bo->moving); - if (!ttm_bo_uses_embedded_gem_object(bo)) - dma_resv_fini(&bo->base._resv); - bo->destroy(bo); - ttm_mem_global_free(&ttm_mem_glob, acc_size); -} - static void ttm_bo_add_mem_to_lru(struct ttm_buffer_object *bo, struct ttm_mem_reg *mem) { struct ttm_bo_device *bdev = bo->bdev; struct ttm_mem_type_manager *man; - dma_resv_assert_held(bo->base.resv); - if (!list_empty(&bo->lru)) return; @@ -181,21 +159,14 @@ static void ttm_bo_add_mem_to_lru(struct ttm_buffer_object *bo, man = &bdev->man[mem->mem_type]; list_add_tail(&bo->lru, &man->lru[bo->priority]); - kref_get(&bo->list_kref); if (!(man->flags & TTM_MEMTYPE_FLAG_FIXED) && bo->ttm && !(bo->ttm->page_flags & (TTM_PAGE_FLAG_SG | TTM_PAGE_FLAG_SWAPPED))) { list_add_tail(&bo->swap, &ttm_bo_glob.swap_lru[bo->priority]); - kref_get(&bo->list_kref); } } -static void ttm_bo_ref_bug(struct kref *list_kref) -{ - BUG(); -} - static void ttm_bo_del_from_lru(struct ttm_buffer_object *bo) { struct ttm_bo_device *bdev = bo->bdev; @@ -203,12 +174,10 @@ static void ttm_bo_del_from_lru(struct ttm_buffer_object *bo) if (!list_empty(&bo->swap)) { list_del_init(&bo->swap); - kref_put(&bo->list_kref, ttm_bo_ref_bug); notify = true; } if (!list_empty(&bo->lru)) { list_del_init(&bo->lru); - kref_put(&bo->list_kref, ttm_bo_ref_bug); notify = true; } @@ -421,94 +390,49 @@ static int ttm_bo_individualize_resv(struct ttm_buffer_object *bo) BUG_ON(!dma_resv_trylock(&bo->base._resv)); r = dma_resv_copy_fences(&bo->base._resv, bo->base.resv); + dma_resv_unlock(&bo->base._resv); if (r) - dma_resv_unlock(&bo->base._resv); + return r; + + if (bo->type != ttm_bo_type_sg) { + /* This works because the BO is about to be destroyed and nobody + * reference it any more. The only tricky case is the trylock on + * the resv object while holding the lru_lock. + */ + spin_lock(&ttm_bo_glob.lru_lock); + bo->base.resv = &bo->base._resv; + spin_unlock(&ttm_bo_glob.lru_lock); + } return r; } static void ttm_bo_flush_all_fences(struct ttm_buffer_object *bo) { + struct dma_resv *resv = &bo->base._resv; struct dma_resv_list *fobj; struct dma_fence *fence; int i; - fobj = dma_resv_get_list(&bo->base._resv); - fence = dma_resv_get_excl(&bo->base._resv); + rcu_read_lock(); + fobj = rcu_dereference(resv->fence); + fence = rcu_dereference(resv->fence_excl); if (fence && !fence->ops->signaled) dma_fence_enable_sw_signaling(fence); for (i = 0; fobj && i < fobj->shared_count; ++i) { - fence = rcu_dereference_protected(fobj->shared[i], - dma_resv_held(bo->base.resv)); + fence = rcu_dereference(fobj->shared[i]); if (!fence->ops->signaled) dma_fence_enable_sw_signaling(fence); } -} - -static void ttm_bo_cleanup_refs_or_queue(struct ttm_buffer_object *bo) -{ - struct ttm_bo_device *bdev = bo->bdev; - int ret; - - ret = ttm_bo_individualize_resv(bo); - if (ret) { - /* Last resort, if we fail to allocate memory for the - * fences block for the BO to become idle - */ - dma_resv_wait_timeout_rcu(bo->base.resv, true, false, - 30 * HZ); - spin_lock(&ttm_bo_glob.lru_lock); - goto error; - } - - spin_lock(&ttm_bo_glob.lru_lock); - ret = dma_resv_trylock(bo->base.resv) ? 0 : -EBUSY; - if (!ret) { - if (dma_resv_test_signaled_rcu(&bo->base._resv, true)) { - ttm_bo_del_from_lru(bo); - spin_unlock(&ttm_bo_glob.lru_lock); - if (bo->base.resv != &bo->base._resv) - dma_resv_unlock(&bo->base._resv); - - ttm_bo_cleanup_memtype_use(bo); - dma_resv_unlock(bo->base.resv); - return; - } - - ttm_bo_flush_all_fences(bo); - - /* - * Make NO_EVICT bos immediately available to - * shrinkers, now that they are queued for - * destruction. - */ - if (bo->mem.placement & TTM_PL_FLAG_NO_EVICT) { - bo->mem.placement &= ~TTM_PL_FLAG_NO_EVICT; - ttm_bo_move_to_lru_tail(bo, NULL); - } - - dma_resv_unlock(bo->base.resv); - } - if (bo->base.resv != &bo->base._resv) { - ttm_bo_flush_all_fences(bo); - dma_resv_unlock(&bo->base._resv); - } - -error: - kref_get(&bo->list_kref); - list_add_tail(&bo->ddestroy, &bdev->ddestroy); - spin_unlock(&ttm_bo_glob.lru_lock); - - schedule_delayed_work(&bdev->wq, - ((HZ / 100) < 1) ? 1 : HZ / 100); + rcu_read_unlock(); } /** * function ttm_bo_cleanup_refs - * If bo idle, remove from delayed- and lru lists, and unref. - * If not idle, do nothing. + * If bo idle, remove from lru lists, and unref. + * If not idle, block if possible. * * Must be called with lru_lock and reservation held, this function * will drop the lru lock and optionally the reservation lock before returning. @@ -522,14 +446,9 @@ static int ttm_bo_cleanup_refs(struct ttm_buffer_object *bo, bool interruptible, bool no_wait_gpu, bool unlock_resv) { - struct dma_resv *resv; + struct dma_resv *resv = &bo->base._resv; int ret; - if (unlikely(list_empty(&bo->ddestroy))) - resv = bo->base.resv; - else - resv = &bo->base._resv; - if (dma_resv_test_signaled_rcu(resv, true)) ret = 0; else @@ -542,9 +461,8 @@ static int ttm_bo_cleanup_refs(struct ttm_buffer_object *bo, dma_resv_unlock(bo->base.resv); spin_unlock(&ttm_bo_glob.lru_lock); - lret = dma_resv_wait_timeout_rcu(resv, true, - interruptible, - 30 * HZ); + lret = dma_resv_wait_timeout_rcu(resv, true, interruptible, + 30 * HZ); if (lret < 0) return lret; @@ -576,14 +494,14 @@ static int ttm_bo_cleanup_refs(struct ttm_buffer_object *bo, ttm_bo_del_from_lru(bo); list_del_init(&bo->ddestroy); - kref_put(&bo->list_kref, ttm_bo_ref_bug); - spin_unlock(&ttm_bo_glob.lru_lock); ttm_bo_cleanup_memtype_use(bo); if (unlock_resv) dma_resv_unlock(bo->base.resv); + ttm_bo_put(bo); + return 0; } @@ -605,8 +523,9 @@ static bool ttm_bo_delayed_delete(struct ttm_bo_device *bdev, bool remove_all) bo = list_first_entry(&bdev->ddestroy, struct ttm_buffer_object, ddestroy); - kref_get(&bo->list_kref); list_move_tail(&bo->ddestroy, &removed); + if (!ttm_bo_get_unless_zero(bo)) + continue; if (remove_all || bo->base.resv != &bo->base._resv) { spin_unlock(&glob->lru_lock); @@ -621,7 +540,7 @@ static bool ttm_bo_delayed_delete(struct ttm_bo_device *bdev, bool remove_all) spin_unlock(&glob->lru_lock); } - kref_put(&bo->list_kref, ttm_bo_release_list); + ttm_bo_put(bo); spin_lock(&glob->lru_lock); } list_splice_tail(&removed, &bdev->ddestroy); @@ -647,16 +566,69 @@ static void ttm_bo_release(struct kref *kref) container_of(kref, struct ttm_buffer_object, kref); struct ttm_bo_device *bdev = bo->bdev; struct ttm_mem_type_manager *man = &bdev->man[bo->mem.mem_type]; + size_t acc_size = bo->acc_size; + int ret; - if (bo->bdev->driver->release_notify) - bo->bdev->driver->release_notify(bo); + if (!bo->deleted) { + ret = ttm_bo_individualize_resv(bo); + if (ret) { + /* Last resort, if we fail to allocate memory for the + * fences block for the BO to become idle + */ + dma_resv_wait_timeout_rcu(bo->base.resv, true, false, + 30 * HZ); + } - drm_vma_offset_remove(bdev->vma_manager, &bo->base.vma_node); - ttm_mem_io_lock(man, false); - ttm_mem_io_free_vm(bo); - ttm_mem_io_unlock(man); - ttm_bo_cleanup_refs_or_queue(bo); - kref_put(&bo->list_kref, ttm_bo_release_list); + if (bo->bdev->driver->release_notify) + bo->bdev->driver->release_notify(bo); + + drm_vma_offset_remove(bdev->vma_manager, &bo->base.vma_node); + ttm_mem_io_lock(man, false); + ttm_mem_io_free_vm(bo); + ttm_mem_io_unlock(man); + } + + if (!dma_resv_test_signaled_rcu(bo->base.resv, true)) { + /* The BO is not idle, resurrect it for delayed destroy */ + ttm_bo_flush_all_fences(bo); + bo->deleted = true; + + spin_lock(&ttm_bo_glob.lru_lock); + + /* + * Make NO_EVICT bos immediately available to + * shrinkers, now that they are queued for + * destruction. + */ + if (bo->mem.placement & TTM_PL_FLAG_NO_EVICT) { + bo->mem.placement &= ~TTM_PL_FLAG_NO_EVICT; + ttm_bo_del_from_lru(bo); + ttm_bo_add_mem_to_lru(bo, &bo->mem); + } + + kref_init(&bo->kref); + list_add_tail(&bo->ddestroy, &bdev->ddestroy); + spin_unlock(&ttm_bo_glob.lru_lock); + + schedule_delayed_work(&bdev->wq, + ((HZ / 100) < 1) ? 1 : HZ / 100); + return; + } + + spin_lock(&ttm_bo_glob.lru_lock); + ttm_bo_del_from_lru(bo); + list_del(&bo->ddestroy); + spin_unlock(&ttm_bo_glob.lru_lock); + + ttm_bo_cleanup_memtype_use(bo); + + BUG_ON(bo->mem.mm_node != NULL); + atomic_dec(&ttm_bo_glob.bo_count); + dma_fence_put(bo->moving); + if (!ttm_bo_uses_embedded_gem_object(bo)) + dma_resv_fini(&bo->base._resv); + bo->destroy(bo); + ttm_mem_global_free(&ttm_mem_glob, acc_size); } void ttm_bo_put(struct ttm_buffer_object *bo) @@ -759,8 +731,7 @@ static bool ttm_bo_evict_swapout_allowable(struct ttm_buffer_object *bo, if (bo->base.resv == ctx->resv) { dma_resv_assert_held(bo->base.resv); - if (ctx->flags & TTM_OPT_FLAG_ALLOW_RES_EVICT - || !list_empty(&bo->ddestroy)) + if (ctx->flags & TTM_OPT_FLAG_ALLOW_RES_EVICT) ret = true; *locked = false; if (busy) @@ -841,6 +812,11 @@ static int ttm_mem_evict_first(struct ttm_bo_device *bdev, dma_resv_unlock(bo->base.resv); continue; } + if (!ttm_bo_get_unless_zero(bo)) { + if (locked) + dma_resv_unlock(bo->base.resv); + continue; + } break; } @@ -852,21 +828,19 @@ static int ttm_mem_evict_first(struct ttm_bo_device *bdev, } if (!bo) { - if (busy_bo) - kref_get(&busy_bo->list_kref); + if (busy_bo && !ttm_bo_get_unless_zero(busy_bo)) + busy_bo = NULL; spin_unlock(&ttm_bo_glob.lru_lock); ret = ttm_mem_evict_wait_busy(busy_bo, ctx, ticket); if (busy_bo) - kref_put(&busy_bo->list_kref, ttm_bo_release_list); + ttm_bo_put(busy_bo); return ret; } - kref_get(&bo->list_kref); - - if (!list_empty(&bo->ddestroy)) { + if (bo->deleted) { ret = ttm_bo_cleanup_refs(bo, ctx->interruptible, ctx->no_wait_gpu, locked); - kref_put(&bo->list_kref, ttm_bo_release_list); + ttm_bo_put(bo); return ret; } @@ -876,7 +850,7 @@ static int ttm_mem_evict_first(struct ttm_bo_device *bdev, if (locked) ttm_bo_unreserve(bo); - kref_put(&bo->list_kref, ttm_bo_release_list); + ttm_bo_put(bo); return ret; } @@ -1221,6 +1195,18 @@ int ttm_bo_validate(struct ttm_buffer_object *bo, uint32_t new_flags; dma_resv_assert_held(bo->base.resv); + + /* + * Remove the backing store if no placement is given. + */ + if (!placement->num_placement && !placement->num_busy_placement) { + ret = ttm_bo_pipeline_gutting(bo); + if (ret) + return ret; + + return ttm_tt_create(bo, false); + } + /* * Check whether we need to move buffer. */ @@ -1288,7 +1274,6 @@ int ttm_bo_init_reserved(struct ttm_bo_device *bdev, bo->destroy = destroy ? destroy : ttm_bo_default_destroy; kref_init(&bo->kref); - kref_init(&bo->list_kref); INIT_LIST_HEAD(&bo->lru); INIT_LIST_HEAD(&bo->ddestroy); INIT_LIST_HEAD(&bo->swap); @@ -1808,11 +1793,18 @@ int ttm_bo_swapout(struct ttm_bo_global *glob, struct ttm_operation_ctx *ctx) spin_lock(&glob->lru_lock); for (i = 0; i < TTM_MAX_BO_PRIORITY; ++i) { list_for_each_entry(bo, &glob->swap_lru[i], swap) { - if (ttm_bo_evict_swapout_allowable(bo, ctx, &locked, - NULL)) { - ret = 0; - break; + if (!ttm_bo_evict_swapout_allowable(bo, ctx, &locked, + NULL)) + continue; + + if (!ttm_bo_get_unless_zero(bo)) { + if (locked) + dma_resv_unlock(bo->base.resv); + continue; } + + ret = 0; + break; } if (!ret) break; @@ -1823,11 +1815,9 @@ int ttm_bo_swapout(struct ttm_bo_global *glob, struct ttm_operation_ctx *ctx) return ret; } - kref_get(&bo->list_kref); - - if (!list_empty(&bo->ddestroy)) { + if (bo->deleted) { ret = ttm_bo_cleanup_refs(bo, false, false, locked); - kref_put(&bo->list_kref, ttm_bo_release_list); + ttm_bo_put(bo); return ret; } @@ -1881,7 +1871,7 @@ out: */ if (locked) dma_resv_unlock(bo->base.resv); - kref_put(&bo->list_kref, ttm_bo_release_list); + ttm_bo_put(bo); return ret; } EXPORT_SYMBOL(ttm_bo_swapout); diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c index 49ed55779128..52d2b71f1588 100644 --- a/drivers/gpu/drm/ttm/ttm_bo_util.c +++ b/drivers/gpu/drm/ttm/ttm_bo_util.c @@ -507,14 +507,14 @@ static int ttm_buffer_object_transfer(struct ttm_buffer_object *bo, fbo->base.moving = NULL; drm_vma_node_reset(&fbo->base.base.vma_node); - kref_init(&fbo->base.list_kref); kref_init(&fbo->base.kref); fbo->base.destroy = &ttm_transfered_destroy; fbo->base.acc_size = 0; - if (bo->base.resv == &bo->base._resv) + if (bo->type != ttm_bo_type_sg) fbo->base.base.resv = &fbo->base.base._resv; dma_resv_init(&fbo->base.base._resv); + fbo->base.base.dev = NULL; ret = dma_resv_trylock(&fbo->base.base._resv); WARN_ON(!ret); diff --git a/drivers/gpu/drm/ttm/ttm_bo_vm.c b/drivers/gpu/drm/ttm/ttm_bo_vm.c index 389128b8c4dd..0ad30b112982 100644 --- a/drivers/gpu/drm/ttm/ttm_bo_vm.c +++ b/drivers/gpu/drm/ttm/ttm_bo_vm.c @@ -59,9 +59,10 @@ static vm_fault_t ttm_bo_vm_fault_idle(struct ttm_buffer_object *bo, /* * If possible, avoid waiting for GPU with mmap_sem - * held. + * held. We only do this if the fault allows retry and this + * is the first attempt. */ - if (vmf->flags & FAULT_FLAG_ALLOW_RETRY) { + if (fault_flag_allow_retry_first(vmf->flags)) { ret = VM_FAULT_RETRY; if (vmf->flags & FAULT_FLAG_RETRY_NOWAIT) goto out_unlock; @@ -135,7 +136,12 @@ vm_fault_t ttm_bo_vm_reserve(struct ttm_buffer_object *bo, * for the buffer to become unreserved. */ if (unlikely(!dma_resv_trylock(bo->base.resv))) { - if (vmf->flags & FAULT_FLAG_ALLOW_RETRY) { + /* + * If the fault allows retry and this is the first + * fault attempt, we try to release the mmap_sem + * before waiting + */ + if (fault_flag_allow_retry_first(vmf->flags)) { if (!(vmf->flags & FAULT_FLAG_RETRY_NOWAIT)) { ttm_bo_get(bo); up_read(&vmf->vma->vm_mm->mmap_sem); @@ -156,6 +162,89 @@ vm_fault_t ttm_bo_vm_reserve(struct ttm_buffer_object *bo, } EXPORT_SYMBOL(ttm_bo_vm_reserve); +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +/** + * ttm_bo_vm_insert_huge - Insert a pfn for PUD or PMD faults + * @vmf: Fault data + * @bo: The buffer object + * @page_offset: Page offset from bo start + * @fault_page_size: The size of the fault in pages. + * @pgprot: The page protections. + * Does additional checking whether it's possible to insert a PUD or PMD + * pfn and performs the insertion. + * + * Return: VM_FAULT_NOPAGE on successful insertion, VM_FAULT_FALLBACK if + * a huge fault was not possible, or on insertion error. + */ +static vm_fault_t ttm_bo_vm_insert_huge(struct vm_fault *vmf, + struct ttm_buffer_object *bo, + pgoff_t page_offset, + pgoff_t fault_page_size, + pgprot_t pgprot) +{ + pgoff_t i; + vm_fault_t ret; + unsigned long pfn; + pfn_t pfnt; + struct ttm_tt *ttm = bo->ttm; + bool write = vmf->flags & FAULT_FLAG_WRITE; + + /* Fault should not cross bo boundary. */ + page_offset &= ~(fault_page_size - 1); + if (page_offset + fault_page_size > bo->num_pages) + goto out_fallback; + + if (bo->mem.bus.is_iomem) + pfn = ttm_bo_io_mem_pfn(bo, page_offset); + else + pfn = page_to_pfn(ttm->pages[page_offset]); + + /* pfn must be fault_page_size aligned. */ + if ((pfn & (fault_page_size - 1)) != 0) + goto out_fallback; + + /* Check that memory is contiguous. */ + if (!bo->mem.bus.is_iomem) { + for (i = 1; i < fault_page_size; ++i) { + if (page_to_pfn(ttm->pages[page_offset + i]) != pfn + i) + goto out_fallback; + } + } else if (bo->bdev->driver->io_mem_pfn) { + for (i = 1; i < fault_page_size; ++i) { + if (ttm_bo_io_mem_pfn(bo, page_offset + i) != pfn + i) + goto out_fallback; + } + } + + pfnt = __pfn_to_pfn_t(pfn, PFN_DEV); + if (fault_page_size == (HPAGE_PMD_SIZE >> PAGE_SHIFT)) + ret = vmf_insert_pfn_pmd_prot(vmf, pfnt, pgprot, write); +#ifdef CONFIG_HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD + else if (fault_page_size == (HPAGE_PUD_SIZE >> PAGE_SHIFT)) + ret = vmf_insert_pfn_pud_prot(vmf, pfnt, pgprot, write); +#endif + else + WARN_ON_ONCE(ret = VM_FAULT_FALLBACK); + + if (ret != VM_FAULT_NOPAGE) + goto out_fallback; + + return VM_FAULT_NOPAGE; +out_fallback: + count_vm_event(THP_FAULT_FALLBACK); + return VM_FAULT_FALLBACK; +} +#else +static vm_fault_t ttm_bo_vm_insert_huge(struct vm_fault *vmf, + struct ttm_buffer_object *bo, + pgoff_t page_offset, + pgoff_t fault_page_size, + pgprot_t pgprot) +{ + return VM_FAULT_FALLBACK; +} +#endif + /** * ttm_bo_vm_fault_reserved - TTM fault helper * @vmf: The struct vm_fault given as argument to the fault callback @@ -163,6 +252,7 @@ EXPORT_SYMBOL(ttm_bo_vm_reserve); * @num_prefault: Maximum number of prefault pages. The caller may want to * specify this based on madvice settings and the size of the GPU object * backed by the memory. + * @fault_page_size: The size of the fault in pages. * * This function inserts one or more page table entries pointing to the * memory backing the buffer object, and then returns a return code @@ -176,7 +266,8 @@ EXPORT_SYMBOL(ttm_bo_vm_reserve); */ vm_fault_t ttm_bo_vm_fault_reserved(struct vm_fault *vmf, pgprot_t prot, - pgoff_t num_prefault) + pgoff_t num_prefault, + pgoff_t fault_page_size) { struct vm_area_struct *vma = vmf->vma; struct ttm_buffer_object *bo = vma->vm_private_data; @@ -268,6 +359,13 @@ vm_fault_t ttm_bo_vm_fault_reserved(struct vm_fault *vmf, prot = pgprot_decrypted(prot); } + /* We don't prefault on huge faults. Yet. */ + if (IS_ENABLED(CONFIG_TRANSPARENT_HUGEPAGE) && fault_page_size != 1) { + ret = ttm_bo_vm_insert_huge(vmf, bo, page_offset, + fault_page_size, prot); + goto out_io_unlock; + } + /* * Speculatively prefault a number of pages. Only error on * first page. @@ -334,7 +432,7 @@ vm_fault_t ttm_bo_vm_fault(struct vm_fault *vmf) return ret; prot = vma->vm_page_prot; - ret = ttm_bo_vm_fault_reserved(vmf, prot, TTM_BO_VM_NUM_PREFAULT); + ret = ttm_bo_vm_fault_reserved(vmf, prot, TTM_BO_VM_NUM_PREFAULT, 1); if (ret == VM_FAULT_RETRY && !(vmf->flags & FAULT_FLAG_RETRY_NOWAIT)) return ret; @@ -445,7 +543,7 @@ static const struct vm_operations_struct ttm_bo_vm_ops = { .fault = ttm_bo_vm_fault, .open = ttm_bo_vm_open, .close = ttm_bo_vm_close, - .access = ttm_bo_vm_access + .access = ttm_bo_vm_access, }; static struct ttm_buffer_object *ttm_bo_vm_lookup(struct ttm_bo_device *bdev, diff --git a/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c b/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c index bf876faea592..faefaaef7909 100644 --- a/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c +++ b/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c @@ -604,7 +604,7 @@ static struct dma_pool *ttm_dma_pool_init(struct device *dev, gfp_t flags, p = pool->name; for (i = 0; i < ARRAY_SIZE(t); i++) { if (type & t[i]) { - p += snprintf(p, sizeof(pool->name) - (p - pool->name), + p += scnprintf(p, sizeof(pool->name) - (p - pool->name), "%s", n[i]); } } diff --git a/drivers/gpu/drm/v3d/v3d_drv.h b/drivers/gpu/drm/v3d/v3d_drv.h index 9a35c555ec52..ac2603334587 100644 --- a/drivers/gpu/drm/v3d/v3d_drv.h +++ b/drivers/gpu/drm/v3d/v3d_drv.h @@ -254,27 +254,42 @@ struct v3d_csd_job { }; /** - * _wait_for - magic (register) wait macro + * __wait_for - magic wait macro * - * Does the right thing for modeset paths when run under kdgb or similar atomic - * contexts. Note that it's important that we check the condition again after - * having timed out, since the timeout could be due to preemption or similar and - * we've never had a chance to check the condition before the timeout. + * Macro to help avoid open coding check/wait/timeout patterns. Note that it's + * important that we check the condition again after having timed out, since the + * timeout could be due to preemption or similar and we've never had a chance to + * check the condition before the timeout. */ -#define wait_for(COND, MS) ({ \ - unsigned long timeout__ = jiffies + msecs_to_jiffies(MS) + 1; \ - int ret__ = 0; \ - while (!(COND)) { \ - if (time_after(jiffies, timeout__)) { \ - if (!(COND)) \ - ret__ = -ETIMEDOUT; \ +#define __wait_for(OP, COND, US, Wmin, Wmax) ({ \ + const ktime_t end__ = ktime_add_ns(ktime_get_raw(), 1000ll * (US)); \ + long wait__ = (Wmin); /* recommended min for usleep is 10 us */ \ + int ret__; \ + might_sleep(); \ + for (;;) { \ + const bool expired__ = ktime_after(ktime_get_raw(), end__); \ + OP; \ + /* Guarantee COND check prior to timeout */ \ + barrier(); \ + if (COND) { \ + ret__ = 0; \ break; \ } \ - msleep(1); \ + if (expired__) { \ + ret__ = -ETIMEDOUT; \ + break; \ + } \ + usleep_range(wait__, wait__ * 2); \ + if (wait__ < (Wmax)) \ + wait__ <<= 1; \ } \ ret__; \ }) +#define _wait_for(COND, US, Wmin, Wmax) __wait_for(, (COND), (US), (Wmin), \ + (Wmax)) +#define wait_for(COND, MS) _wait_for((COND), (MS) * 1000, 10, 1000) + static inline unsigned long nsecs_to_jiffies_timeout(const u64 n) { /* nsecs_to_jiffies64() does not guard against overflow */ diff --git a/drivers/gpu/drm/vboxvideo/vbox_drv.c b/drivers/gpu/drm/vboxvideo/vbox_drv.c index 8512d970a09f..ac8f75db2ecd 100644 --- a/drivers/gpu/drm/vboxvideo/vbox_drv.c +++ b/drivers/gpu/drm/vboxvideo/vbox_drv.c @@ -41,6 +41,10 @@ static int vbox_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent) if (!vbox_check_supported(VBE_DISPI_ID_HGSMI)) return -ENODEV; + ret = drm_fb_helper_remove_conflicting_pci_framebuffers(pdev, "vboxvideodrmfb"); + if (ret) + return ret; + vbox = kzalloc(sizeof(*vbox), GFP_KERNEL); if (!vbox) return -ENOMEM; diff --git a/drivers/gpu/drm/vboxvideo/vboxvideo.h b/drivers/gpu/drm/vboxvideo/vboxvideo.h index 0592004f71aa..a5de40fe1a76 100644 --- a/drivers/gpu/drm/vboxvideo/vboxvideo.h +++ b/drivers/gpu/drm/vboxvideo/vboxvideo.h @@ -138,7 +138,7 @@ struct vbva_buffer { u32 data_len; /* variable size for the rest of the vbva_buffer area in VRAM. */ - u8 data[0]; + u8 data[]; } __packed; #define VBVA_MAX_RECORD_SIZE (128 * 1024 * 1024) diff --git a/drivers/gpu/drm/vc4/vc4_crtc.c b/drivers/gpu/drm/vc4/vc4_crtc.c index b00e20f5ce05..1208258ad3b2 100644 --- a/drivers/gpu/drm/vc4/vc4_crtc.c +++ b/drivers/gpu/drm/vc4/vc4_crtc.c @@ -84,13 +84,14 @@ static const struct debugfs_reg32 crtc_regs[] = { VC4_REG32(PV_HACT_ACT), }; -bool vc4_crtc_get_scanoutpos(struct drm_device *dev, unsigned int crtc_id, - bool in_vblank_irq, int *vpos, int *hpos, - ktime_t *stime, ktime_t *etime, - const struct drm_display_mode *mode) +static bool vc4_crtc_get_scanout_position(struct drm_crtc *crtc, + bool in_vblank_irq, + int *vpos, int *hpos, + ktime_t *stime, ktime_t *etime, + const struct drm_display_mode *mode) { + struct drm_device *dev = crtc->dev; struct vc4_dev *vc4 = to_vc4_dev(dev); - struct drm_crtc *crtc = drm_crtc_from_index(dev, crtc_id); struct vc4_crtc *vc4_crtc = to_vc4_crtc(crtc); u32 val; int fifo_lines; @@ -1030,6 +1031,7 @@ static const struct drm_crtc_funcs vc4_crtc_funcs = { .gamma_set = drm_atomic_helper_legacy_gamma_set, .enable_vblank = vc4_enable_vblank, .disable_vblank = vc4_disable_vblank, + .get_vblank_timestamp = drm_crtc_vblank_helper_get_vblank_timestamp, }; static const struct drm_crtc_helper_funcs vc4_crtc_helper_funcs = { @@ -1039,6 +1041,7 @@ static const struct drm_crtc_helper_funcs vc4_crtc_helper_funcs = { .atomic_flush = vc4_crtc_atomic_flush, .atomic_enable = vc4_crtc_atomic_enable, .atomic_disable = vc4_crtc_atomic_disable, + .get_scanout_position = vc4_crtc_get_scanout_position, }; static const struct vc4_crtc_data pv0_data = { diff --git a/drivers/gpu/drm/vc4/vc4_dpi.c b/drivers/gpu/drm/vc4/vc4_dpi.c index c586325de2a5..6dfede03396e 100644 --- a/drivers/gpu/drm/vc4/vc4_dpi.c +++ b/drivers/gpu/drm/vc4/vc4_dpi.c @@ -252,7 +252,7 @@ static int vc4_dpi_init_bridge(struct vc4_dpi *dpi) bridge = drm_panel_bridge_add_typed(panel, DRM_MODE_CONNECTOR_DPI); - return drm_bridge_attach(dpi->encoder, bridge, NULL); + return drm_bridge_attach(dpi->encoder, bridge, NULL, 0); } static int vc4_dpi_bind(struct device *dev, struct device *master, void *data) diff --git a/drivers/gpu/drm/vc4/vc4_drv.c b/drivers/gpu/drm/vc4/vc4_drv.c index 5e6fb6c2307f..76f93b662766 100644 --- a/drivers/gpu/drm/vc4/vc4_drv.c +++ b/drivers/gpu/drm/vc4/vc4_drv.c @@ -190,9 +190,6 @@ static struct drm_driver vc4_drm_driver = { .irq_postinstall = vc4_irq_postinstall, .irq_uninstall = vc4_irq_uninstall, - .get_scanout_position = vc4_crtc_get_scanoutpos, - .get_vblank_timestamp = drm_calc_vbltimestamp_from_scanoutpos, - #if defined(CONFIG_DEBUG_FS) .debugfs_init = vc4_debugfs_init, #endif diff --git a/drivers/gpu/drm/vc4/vc4_drv.h b/drivers/gpu/drm/vc4/vc4_drv.h index 6627b20c99e9..139d25a8328e 100644 --- a/drivers/gpu/drm/vc4/vc4_drv.h +++ b/drivers/gpu/drm/vc4/vc4_drv.h @@ -65,7 +65,7 @@ struct vc4_perfmon { * Note that counter values can't be reset, but you can fake a reset by * destroying the perfmon and creating a new one. */ - u64 counters[0]; + u64 counters[]; }; struct vc4_dev { @@ -677,32 +677,41 @@ struct vc4_validated_shader_info { }; /** - * _wait_for - magic (register) wait macro + * __wait_for - magic wait macro * - * Does the right thing for modeset paths when run under kdgb or similar atomic - * contexts. Note that it's important that we check the condition again after - * having timed out, since the timeout could be due to preemption or similar and - * we've never had a chance to check the condition before the timeout. + * Macro to help avoid open coding check/wait/timeout patterns. Note that it's + * important that we check the condition again after having timed out, since the + * timeout could be due to preemption or similar and we've never had a chance to + * check the condition before the timeout. */ -#define _wait_for(COND, MS, W) ({ \ - unsigned long timeout__ = jiffies + msecs_to_jiffies(MS) + 1; \ - int ret__ = 0; \ - while (!(COND)) { \ - if (time_after(jiffies, timeout__)) { \ - if (!(COND)) \ - ret__ = -ETIMEDOUT; \ +#define __wait_for(OP, COND, US, Wmin, Wmax) ({ \ + const ktime_t end__ = ktime_add_ns(ktime_get_raw(), 1000ll * (US)); \ + long wait__ = (Wmin); /* recommended min for usleep is 10 us */ \ + int ret__; \ + might_sleep(); \ + for (;;) { \ + const bool expired__ = ktime_after(ktime_get_raw(), end__); \ + OP; \ + /* Guarantee COND check prior to timeout */ \ + barrier(); \ + if (COND) { \ + ret__ = 0; \ break; \ } \ - if (W && drm_can_sleep()) { \ - msleep(W); \ - } else { \ - cpu_relax(); \ + if (expired__) { \ + ret__ = -ETIMEDOUT; \ + break; \ } \ + usleep_range(wait__, wait__ * 2); \ + if (wait__ < (Wmax)) \ + wait__ <<= 1; \ } \ ret__; \ }) -#define wait_for(COND, MS) _wait_for(COND, MS, 1) +#define _wait_for(COND, US, Wmin, Wmax) __wait_for(, (COND), (US), (Wmin), \ + (Wmax)) +#define wait_for(COND, MS) _wait_for((COND), (MS) * 1000, 10, 1000) /* vc4_bo.c */ struct drm_gem_object *vc4_create_object(struct drm_device *dev, size_t size); @@ -743,10 +752,6 @@ void vc4_bo_remove_from_purgeable_pool(struct vc4_bo *bo); /* vc4_crtc.c */ extern struct platform_driver vc4_crtc_driver; -bool vc4_crtc_get_scanoutpos(struct drm_device *dev, unsigned int crtc_id, - bool in_vblank_irq, int *vpos, int *hpos, - ktime_t *stime, ktime_t *etime, - const struct drm_display_mode *mode); void vc4_crtc_handle_vblank(struct vc4_crtc *crtc); void vc4_crtc_txp_armed(struct drm_crtc_state *state); void vc4_crtc_get_margins(struct drm_crtc_state *state, diff --git a/drivers/gpu/drm/vc4/vc4_dsi.c b/drivers/gpu/drm/vc4/vc4_dsi.c index fd8a2eb60505..d99b1d526651 100644 --- a/drivers/gpu/drm/vc4/vc4_dsi.c +++ b/drivers/gpu/drm/vc4/vc4_dsi.c @@ -1619,7 +1619,7 @@ static int vc4_dsi_bind(struct device *dev, struct device *master, void *data) DRM_MODE_ENCODER_DSI, NULL); drm_encoder_helper_add(dsi->encoder, &vc4_dsi_encoder_helper_funcs); - ret = drm_bridge_attach(dsi->encoder, dsi->bridge, NULL); + ret = drm_bridge_attach(dsi->encoder, dsi->bridge, NULL, 0); if (ret) { dev_err(dev, "bridge attach failed: %d\n", ret); return ret; diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.c b/drivers/gpu/drm/vc4/vc4_hdmi.c index cea18dc15f77..340719238753 100644 --- a/drivers/gpu/drm/vc4/vc4_hdmi.c +++ b/drivers/gpu/drm/vc4/vc4_hdmi.c @@ -681,11 +681,23 @@ static enum drm_mode_status vc4_hdmi_encoder_mode_valid(struct drm_encoder *crtc, const struct drm_display_mode *mode) { - /* HSM clock must be 108% of the pixel clock. Additionally, - * the AXI clock needs to be at least 25% of pixel clock, but - * HSM ends up being the limiting factor. + /* + * As stated in RPi's vc4 firmware "HDMI state machine (HSM) clock must + * be faster than pixel clock, infinitesimally faster, tested in + * simulation. Otherwise, exact value is unimportant for HDMI + * operation." This conflicts with bcm2835's vc4 documentation, which + * states HSM's clock has to be at least 108% of the pixel clock. + * + * Real life tests reveal that vc4's firmware statement holds up, and + * users are able to use pixel clocks closer to HSM's, namely for + * 1920x1200@60Hz. So it was decided to have leave a 1% margin between + * both clocks. Which, for RPi0-3 implies a maximum pixel clock of + * 162MHz. + * + * Additionally, the AXI clock needs to be at least 25% of + * pixel clock, but HSM ends up being the limiting factor. */ - if (mode->clock > HSM_CLOCK_FREQ / (1000 * 108 / 100)) + if (mode->clock > HSM_CLOCK_FREQ / (1000 * 101 / 100)) return MODE_CLOCK_HIGH; return MODE_OK; diff --git a/drivers/gpu/drm/virtio/virtgpu_display.c b/drivers/gpu/drm/virtio/virtgpu_display.c index 7b0f0643bb2d..2b7e6ae65546 100644 --- a/drivers/gpu/drm/virtio/virtgpu_display.c +++ b/drivers/gpu/drm/virtio/virtgpu_display.c @@ -90,6 +90,7 @@ static void virtio_gpu_crtc_mode_set_nofb(struct drm_crtc *crtc) virtio_gpu_cmd_set_scanout(vgdev, output->index, 0, crtc->mode.hdisplay, crtc->mode.vdisplay, 0, 0); + virtio_gpu_notify(vgdev); } static void virtio_gpu_crtc_atomic_enable(struct drm_crtc *crtc, @@ -108,6 +109,7 @@ static void virtio_gpu_crtc_atomic_disable(struct drm_crtc *crtc, struct virtio_gpu_output *output = drm_crtc_to_virtio_gpu_output(crtc); virtio_gpu_cmd_set_scanout(vgdev, output->index, 0, 0, 0, 0, 0); + virtio_gpu_notify(vgdev); output->enabled = false; } @@ -368,6 +370,5 @@ void virtio_gpu_modeset_fini(struct virtio_gpu_device *vgdev) for (i = 0 ; i < vgdev->num_scanouts; ++i) kfree(vgdev->outputs[i].edid); - drm_atomic_helper_shutdown(vgdev->ddev); drm_mode_config_cleanup(vgdev->ddev); } diff --git a/drivers/gpu/drm/virtio/virtgpu_drv.c b/drivers/gpu/drm/virtio/virtgpu_drv.c index 8cf27af3ad53..ab4bed78e656 100644 --- a/drivers/gpu/drm/virtio/virtgpu_drv.c +++ b/drivers/gpu/drm/virtio/virtgpu_drv.c @@ -31,6 +31,7 @@ #include <linux/pci.h> #include <drm/drm.h> +#include <drm/drm_atomic_helper.h> #include <drm/drm_drv.h> #include <drm/drm_file.h> @@ -135,7 +136,8 @@ static void virtio_gpu_remove(struct virtio_device *vdev) { struct drm_device *dev = vdev->priv; - drm_dev_unregister(dev); + drm_dev_unplug(dev); + drm_atomic_helper_shutdown(dev); virtio_gpu_deinit(dev); drm_dev_put(dev); } @@ -214,4 +216,6 @@ static struct drm_driver driver = { .major = DRIVER_MAJOR, .minor = DRIVER_MINOR, .patchlevel = DRIVER_PATCHLEVEL, + + .release = virtio_gpu_release, }; diff --git a/drivers/gpu/drm/virtio/virtgpu_drv.h b/drivers/gpu/drm/virtio/virtgpu_drv.h index d278c8c50f39..c1824bdf2418 100644 --- a/drivers/gpu/drm/virtio/virtgpu_drv.h +++ b/drivers/gpu/drm/virtio/virtgpu_drv.h @@ -32,6 +32,7 @@ #include <linux/virtio_gpu.h> #include <drm/drm_atomic.h> +#include <drm/drm_drv.h> #include <drm/drm_encoder.h> #include <drm/drm_fb_helper.h> #include <drm/drm_gem.h> @@ -68,15 +69,21 @@ struct virtio_gpu_object_params { struct virtio_gpu_object { struct drm_gem_shmem_object base; uint32_t hw_res_handle; - - struct sg_table *pages; - uint32_t mapped; bool dumb; bool created; }; #define gem_to_virtio_gpu_obj(gobj) \ container_of((gobj), struct virtio_gpu_object, base.base) +struct virtio_gpu_object_shmem { + struct virtio_gpu_object base; + struct sg_table *pages; + uint32_t mapped; +}; + +#define to_virtio_gpu_shmem(virtio_gpu_object) \ + container_of((virtio_gpu_object), struct virtio_gpu_object_shmem, base) + struct virtio_gpu_object_array { struct ww_acquire_ctx ticket; struct list_head next; @@ -114,6 +121,7 @@ struct virtio_gpu_vbuffer { char *resp_buf; int resp_size; virtio_gpu_resp_cb resp_cb; + void *resp_cb_data; struct virtio_gpu_object_array *objs; struct list_head list; @@ -175,10 +183,8 @@ struct virtio_gpu_device { struct virtio_gpu_queue ctrlq; struct virtio_gpu_queue cursorq; struct kmem_cache *vbufs; - bool vqs_ready; - bool disable_notify; - bool pending_notify; + atomic_t pending_commands; struct ida resource_ida; @@ -208,6 +214,8 @@ struct virtio_gpu_device { struct virtio_gpu_fpriv { uint32_t ctx_id; + bool context_created; + struct mutex context_lock; }; /* virtio_ioctl.c */ @@ -217,6 +225,7 @@ extern struct drm_ioctl_desc virtio_gpu_ioctls[DRM_VIRTIO_NUM_IOCTLS]; /* virtio_kms.c */ int virtio_gpu_init(struct drm_device *dev); void virtio_gpu_deinit(struct drm_device *dev); +void virtio_gpu_release(struct drm_device *dev); int virtio_gpu_driver_open(struct drm_device *dev, struct drm_file *file); void virtio_gpu_driver_postclose(struct drm_device *dev, struct drm_file *file); @@ -263,7 +272,7 @@ void virtio_gpu_cmd_create_resource(struct virtio_gpu_device *vgdev, struct virtio_gpu_object_array *objs, struct virtio_gpu_fence *fence); void virtio_gpu_cmd_unref_resource(struct virtio_gpu_device *vgdev, - uint32_t resource_id); + struct virtio_gpu_object *bo); void virtio_gpu_cmd_transfer_to_host_2d(struct virtio_gpu_device *vgdev, uint64_t offset, uint32_t width, uint32_t height, @@ -280,9 +289,8 @@ void virtio_gpu_cmd_set_scanout(struct virtio_gpu_device *vgdev, uint32_t x, uint32_t y); int virtio_gpu_object_attach(struct virtio_gpu_device *vgdev, struct virtio_gpu_object *obj, - struct virtio_gpu_fence *fence); -void virtio_gpu_object_detach(struct virtio_gpu_device *vgdev, - struct virtio_gpu_object *obj); + struct virtio_gpu_mem_entry *ents, + unsigned int nents); int virtio_gpu_attach_status_page(struct virtio_gpu_device *vgdev); int virtio_gpu_detach_status_page(struct virtio_gpu_device *vgdev); void virtio_gpu_cursor_ping(struct virtio_gpu_device *vgdev, @@ -333,8 +341,7 @@ void virtio_gpu_dequeue_ctrl_func(struct work_struct *work); void virtio_gpu_dequeue_cursor_func(struct work_struct *work); void virtio_gpu_dequeue_fence_func(struct work_struct *work); -void virtio_gpu_disable_notify(struct virtio_gpu_device *vgdev); -void virtio_gpu_enable_notify(struct virtio_gpu_device *vgdev); +void virtio_gpu_notify(struct virtio_gpu_device *vgdev); /* virtio_gpu_display.c */ void virtio_gpu_modeset_init(struct virtio_gpu_device *vgdev); @@ -356,12 +363,16 @@ void virtio_gpu_fence_event_process(struct virtio_gpu_device *vdev, u64 last_seq); /* virtio_gpu_object */ +void virtio_gpu_cleanup_object(struct virtio_gpu_object *bo); struct drm_gem_object *virtio_gpu_create_object(struct drm_device *dev, size_t size); int virtio_gpu_object_create(struct virtio_gpu_device *vgdev, struct virtio_gpu_object_params *params, struct virtio_gpu_object **bo_ptr, struct virtio_gpu_fence *fence); + +bool virtio_gpu_is_shmem(struct virtio_gpu_object *bo); + /* virtgpu_prime.c */ struct drm_gem_object *virtgpu_gem_prime_import_sg_table( struct drm_device *dev, struct dma_buf_attachment *attach, diff --git a/drivers/gpu/drm/virtio/virtgpu_gem.c b/drivers/gpu/drm/virtio/virtgpu_gem.c index 0a2b62279647..0d6152c99a27 100644 --- a/drivers/gpu/drm/virtio/virtgpu_gem.c +++ b/drivers/gpu/drm/virtio/virtgpu_gem.c @@ -123,6 +123,7 @@ int virtio_gpu_gem_object_open(struct drm_gem_object *obj, virtio_gpu_cmd_context_attach_resource(vgdev, vfpriv->ctx_id, objs); + virtio_gpu_notify(vgdev); return 0; } @@ -143,6 +144,7 @@ void virtio_gpu_gem_object_close(struct drm_gem_object *obj, virtio_gpu_cmd_context_detach_resource(vgdev, vfpriv->ctx_id, objs); + virtio_gpu_notify(vgdev); } struct virtio_gpu_object_array *virtio_gpu_array_alloc(u32 nents) diff --git a/drivers/gpu/drm/virtio/virtgpu_ioctl.c b/drivers/gpu/drm/virtio/virtgpu_ioctl.c index 205ec4abae2b..336cc9143205 100644 --- a/drivers/gpu/drm/virtio/virtgpu_ioctl.c +++ b/drivers/gpu/drm/virtio/virtgpu_ioctl.c @@ -33,13 +33,34 @@ #include "virtgpu_drv.h" +static void virtio_gpu_create_context(struct drm_device *dev, + struct drm_file *file) +{ + struct virtio_gpu_device *vgdev = dev->dev_private; + struct virtio_gpu_fpriv *vfpriv = file->driver_priv; + char dbgname[TASK_COMM_LEN]; + + mutex_lock(&vfpriv->context_lock); + if (vfpriv->context_created) + goto out_unlock; + + get_task_comm(dbgname, current); + virtio_gpu_cmd_context_create(vgdev, vfpriv->ctx_id, + strlen(dbgname), dbgname); + virtio_gpu_notify(vgdev); + vfpriv->context_created = true; + +out_unlock: + mutex_unlock(&vfpriv->context_lock); +} + static int virtio_gpu_map_ioctl(struct drm_device *dev, void *data, - struct drm_file *file_priv) + struct drm_file *file) { struct virtio_gpu_device *vgdev = dev->dev_private; struct drm_virtgpu_map *virtio_gpu_map = data; - return virtio_gpu_mode_dumb_mmap(file_priv, vgdev->ddev, + return virtio_gpu_mode_dumb_mmap(file, vgdev->ddev, virtio_gpu_map->handle, &virtio_gpu_map->offset); } @@ -51,11 +72,11 @@ static int virtio_gpu_map_ioctl(struct drm_device *dev, void *data, * VIRTIO_GPUReleaseInfo struct (first XXX bytes) */ static int virtio_gpu_execbuffer_ioctl(struct drm_device *dev, void *data, - struct drm_file *drm_file) + struct drm_file *file) { struct drm_virtgpu_execbuffer *exbuf = data; struct virtio_gpu_device *vgdev = dev->dev_private; - struct virtio_gpu_fpriv *vfpriv = drm_file->driver_priv; + struct virtio_gpu_fpriv *vfpriv = file->driver_priv; struct virtio_gpu_fence *out_fence; int ret; uint32_t *bo_handles = NULL; @@ -74,6 +95,7 @@ static int virtio_gpu_execbuffer_ioctl(struct drm_device *dev, void *data, exbuf->fence_fd = -1; + virtio_gpu_create_context(dev, file); if (exbuf->flags & VIRTGPU_EXECBUF_FENCE_FD_IN) { struct dma_fence *in_fence; @@ -116,7 +138,7 @@ static int virtio_gpu_execbuffer_ioctl(struct drm_device *dev, void *data, goto out_unused_fd; } - buflist = virtio_gpu_array_from_handles(drm_file, bo_handles, + buflist = virtio_gpu_array_from_handles(file, bo_handles, exbuf->num_bo_handles); if (!buflist) { ret = -ENOENT; @@ -126,22 +148,22 @@ static int virtio_gpu_execbuffer_ioctl(struct drm_device *dev, void *data, bo_handles = NULL; } - if (buflist) { - ret = virtio_gpu_array_lock_resv(buflist); - if (ret) - goto out_unused_fd; - } - buf = vmemdup_user(u64_to_user_ptr(exbuf->command), exbuf->size); if (IS_ERR(buf)) { ret = PTR_ERR(buf); - goto out_unresv; + goto out_unused_fd; + } + + if (buflist) { + ret = virtio_gpu_array_lock_resv(buflist); + if (ret) + goto out_memdup; } out_fence = virtio_gpu_fence_alloc(vgdev); if(!out_fence) { ret = -ENOMEM; - goto out_memdup; + goto out_unresv; } if (out_fence_fd >= 0) { @@ -158,13 +180,14 @@ static int virtio_gpu_execbuffer_ioctl(struct drm_device *dev, void *data, virtio_gpu_cmd_submit(vgdev, buf, exbuf->size, vfpriv->ctx_id, buflist, out_fence); + virtio_gpu_notify(vgdev); return 0; -out_memdup: - kvfree(buf); out_unresv: if (buflist) virtio_gpu_array_unlock_resv(buflist); +out_memdup: + kvfree(buf); out_unused_fd: kvfree(bo_handles); if (buflist) @@ -177,7 +200,7 @@ out_unused_fd: } static int virtio_gpu_getparam_ioctl(struct drm_device *dev, void *data, - struct drm_file *file_priv) + struct drm_file *file) { struct virtio_gpu_device *vgdev = dev->dev_private; struct drm_virtgpu_getparam *param = data; @@ -200,7 +223,7 @@ static int virtio_gpu_getparam_ioctl(struct drm_device *dev, void *data, } static int virtio_gpu_resource_create_ioctl(struct drm_device *dev, void *data, - struct drm_file *file_priv) + struct drm_file *file) { struct virtio_gpu_device *vgdev = dev->dev_private; struct drm_virtgpu_resource_create *rc = data; @@ -211,7 +234,17 @@ static int virtio_gpu_resource_create_ioctl(struct drm_device *dev, void *data, uint32_t handle = 0; struct virtio_gpu_object_params params = { 0 }; - if (vgdev->has_virgl_3d == false) { + if (vgdev->has_virgl_3d) { + virtio_gpu_create_context(dev, file); + params.virgl = true; + params.target = rc->target; + params.bind = rc->bind; + params.depth = rc->depth; + params.array_size = rc->array_size; + params.last_level = rc->last_level; + params.nr_samples = rc->nr_samples; + params.flags = rc->flags; + } else { if (rc->depth > 1) return -EINVAL; if (rc->nr_samples > 1) @@ -228,16 +261,6 @@ static int virtio_gpu_resource_create_ioctl(struct drm_device *dev, void *data, params.width = rc->width; params.height = rc->height; params.size = rc->size; - if (vgdev->has_virgl_3d) { - params.virgl = true; - params.target = rc->target; - params.bind = rc->bind; - params.depth = rc->depth; - params.array_size = rc->array_size; - params.last_level = rc->last_level; - params.nr_samples = rc->nr_samples; - params.flags = rc->flags; - } /* allocate a single page size object */ if (params.size == 0) params.size = PAGE_SIZE; @@ -251,7 +274,7 @@ static int virtio_gpu_resource_create_ioctl(struct drm_device *dev, void *data, return ret; obj = &qobj->base.base; - ret = drm_gem_handle_create(file_priv, obj, &handle); + ret = drm_gem_handle_create(file, obj, &handle); if (ret) { drm_gem_object_release(obj); return ret; @@ -264,13 +287,13 @@ static int virtio_gpu_resource_create_ioctl(struct drm_device *dev, void *data, } static int virtio_gpu_resource_info_ioctl(struct drm_device *dev, void *data, - struct drm_file *file_priv) + struct drm_file *file) { struct drm_virtgpu_resource_info *ri = data; struct drm_gem_object *gobj = NULL; struct virtio_gpu_object *qobj = NULL; - gobj = drm_gem_object_lookup(file_priv, ri->bo_handle); + gobj = drm_gem_object_lookup(file, ri->bo_handle); if (gobj == NULL) return -ENOENT; @@ -297,6 +320,7 @@ static int virtio_gpu_transfer_from_host_ioctl(struct drm_device *dev, if (vgdev->has_virgl_3d == false) return -ENOSYS; + virtio_gpu_create_context(dev, file); objs = virtio_gpu_array_from_handles(file, &args->bo_handle, 1); if (objs == NULL) return -ENOENT; @@ -314,6 +338,7 @@ static int virtio_gpu_transfer_from_host_ioctl(struct drm_device *dev, (vgdev, vfpriv->ctx_id, offset, args->level, &args->box, objs, fence); dma_fence_put(&fence->f); + virtio_gpu_notify(vgdev); return 0; err_unlock: @@ -344,6 +369,7 @@ static int virtio_gpu_transfer_to_host_ioctl(struct drm_device *dev, void *data, args->box.w, args->box.h, args->box.x, args->box.y, objs, NULL); } else { + virtio_gpu_create_context(dev, file); ret = virtio_gpu_array_lock_resv(objs); if (ret != 0) goto err_put_free; @@ -359,6 +385,7 @@ static int virtio_gpu_transfer_to_host_ioctl(struct drm_device *dev, void *data, args->level, &args->box, objs, fence); dma_fence_put(&fence->f); } + virtio_gpu_notify(vgdev); return 0; err_unlock: @@ -445,6 +472,7 @@ static int virtio_gpu_get_caps_ioctl(struct drm_device *dev, /* not in cache - need to talk to hw */ virtio_gpu_cmd_get_capset(vgdev, found_valid, args->cap_set_ver, &cache_ent); + virtio_gpu_notify(vgdev); copy_exit: ret = wait_event_timeout(vgdev->resp_wq, diff --git a/drivers/gpu/drm/virtio/virtgpu_kms.c b/drivers/gpu/drm/virtio/virtgpu_kms.c index c1086df49816..023a030ca7b9 100644 --- a/drivers/gpu/drm/virtio/virtgpu_kms.c +++ b/drivers/gpu/drm/virtio/virtgpu_kms.c @@ -44,6 +44,7 @@ static void virtio_gpu_config_changed_work_func(struct work_struct *work) if (vgdev->has_edid) virtio_gpu_cmd_get_edids(vgdev); virtio_gpu_cmd_get_display_info(vgdev); + virtio_gpu_notify(vgdev); drm_helper_hpd_irq_event(vgdev->ddev); events_clear |= VIRTIO_GPU_EVENT_DISPLAY; } @@ -51,22 +52,11 @@ static void virtio_gpu_config_changed_work_func(struct work_struct *work) events_clear, &events_clear); } -static int virtio_gpu_context_create(struct virtio_gpu_device *vgdev, - uint32_t nlen, const char *name) -{ - int handle = ida_alloc(&vgdev->ctx_id_ida, GFP_KERNEL); - - if (handle < 0) - return handle; - handle += 1; - virtio_gpu_cmd_context_create(vgdev, handle, nlen, name); - return handle; -} - static void virtio_gpu_context_destroy(struct virtio_gpu_device *vgdev, uint32_t ctx_id) { virtio_gpu_cmd_context_destroy(vgdev, ctx_id); + virtio_gpu_notify(vgdev); ida_free(&vgdev->ctx_id_ida, ctx_id - 1); } @@ -92,6 +82,7 @@ static void virtio_gpu_get_capsets(struct virtio_gpu_device *vgdev, } for (i = 0; i < num_capsets; i++) { virtio_gpu_cmd_get_capset_info(vgdev, i); + virtio_gpu_notify(vgdev); ret = wait_event_timeout(vgdev->resp_wq, vgdev->capsets[i].id > 0, 5 * HZ); if (ret == 0) { @@ -199,13 +190,13 @@ int virtio_gpu_init(struct drm_device *dev) virtio_gpu_modeset_init(vgdev); virtio_device_ready(vgdev->vdev); - vgdev->vqs_ready = true; if (num_capsets) virtio_gpu_get_capsets(vgdev, num_capsets); if (vgdev->has_edid) virtio_gpu_cmd_get_edids(vgdev); virtio_gpu_cmd_get_display_info(vgdev); + virtio_gpu_notify(vgdev); wait_event_timeout(vgdev->resp_wq, !vgdev->display_info_pending, 5 * HZ); return 0; @@ -234,12 +225,16 @@ void virtio_gpu_deinit(struct drm_device *dev) struct virtio_gpu_device *vgdev = dev->dev_private; flush_work(&vgdev->obj_free_work); - vgdev->vqs_ready = false; flush_work(&vgdev->ctrlq.dequeue_work); flush_work(&vgdev->cursorq.dequeue_work); flush_work(&vgdev->config_changed_work); vgdev->vdev->config->reset(vgdev->vdev); vgdev->vdev->config->del_vqs(vgdev->vdev); +} + +void virtio_gpu_release(struct drm_device *dev) +{ + struct virtio_gpu_device *vgdev = dev->dev_private; virtio_gpu_modeset_fini(vgdev); virtio_gpu_free_vbufs(vgdev); @@ -252,8 +247,7 @@ int virtio_gpu_driver_open(struct drm_device *dev, struct drm_file *file) { struct virtio_gpu_device *vgdev = dev->dev_private; struct virtio_gpu_fpriv *vfpriv; - int id; - char dbgname[TASK_COMM_LEN]; + int handle; /* can't create contexts without 3d renderer */ if (!vgdev->has_virgl_3d) @@ -264,14 +258,15 @@ int virtio_gpu_driver_open(struct drm_device *dev, struct drm_file *file) if (!vfpriv) return -ENOMEM; - get_task_comm(dbgname, current); - id = virtio_gpu_context_create(vgdev, strlen(dbgname), dbgname); - if (id < 0) { + mutex_init(&vfpriv->context_lock); + + handle = ida_alloc(&vgdev->ctx_id_ida, GFP_KERNEL); + if (handle < 0) { kfree(vfpriv); - return id; + return handle; } - vfpriv->ctx_id = id; + vfpriv->ctx_id = handle + 1; file->driver_priv = vfpriv; return 0; } @@ -287,6 +282,7 @@ void virtio_gpu_driver_postclose(struct drm_device *dev, struct drm_file *file) vfpriv = file->driver_priv; virtio_gpu_context_destroy(vgdev, vfpriv->ctx_id); + mutex_destroy(&vfpriv->context_lock); kfree(vfpriv); file->driver_priv = NULL; } diff --git a/drivers/gpu/drm/virtio/virtgpu_object.c b/drivers/gpu/drm/virtio/virtgpu_object.c index 017a9e0fc3bb..d9039bb7c5e3 100644 --- a/drivers/gpu/drm/virtio/virtgpu_object.c +++ b/drivers/gpu/drm/virtio/virtgpu_object.c @@ -23,6 +23,7 @@ * WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ +#include <linux/dma-mapping.h> #include <linux/moduleparam.h> #include "virtgpu_drv.h" @@ -42,8 +43,8 @@ static int virtio_gpu_resource_id_get(struct virtio_gpu_device *vgdev, * "f91a9dd35715 Fix unlinking resources from hash * table." (Feb 2019) fixes the bug. */ - static int handle; - handle++; + static atomic_t seqno = ATOMIC_INIT(0); + int handle = atomic_inc_return(&seqno); *resid = handle + 1; } else { int handle = ida_alloc(&vgdev->resource_ida, GFP_KERNEL); @@ -61,21 +62,46 @@ static void virtio_gpu_resource_id_put(struct virtio_gpu_device *vgdev, uint32_t } } -static void virtio_gpu_free_object(struct drm_gem_object *obj) +void virtio_gpu_cleanup_object(struct virtio_gpu_object *bo) { - struct virtio_gpu_object *bo = gem_to_virtio_gpu_obj(obj); struct virtio_gpu_device *vgdev = bo->base.base.dev->dev_private; - if (bo->pages) - virtio_gpu_object_detach(vgdev, bo); - if (bo->created) - virtio_gpu_cmd_unref_resource(vgdev, bo->hw_res_handle); virtio_gpu_resource_id_put(vgdev, bo->hw_res_handle); + if (virtio_gpu_is_shmem(bo)) { + struct virtio_gpu_object_shmem *shmem = to_virtio_gpu_shmem(bo); + + if (shmem->pages) { + if (shmem->mapped) { + dma_unmap_sg(vgdev->vdev->dev.parent, + shmem->pages->sgl, shmem->mapped, + DMA_TO_DEVICE); + shmem->mapped = 0; + } + + sg_free_table(shmem->pages); + shmem->pages = NULL; + drm_gem_shmem_unpin(&bo->base.base); + } + + drm_gem_shmem_free_object(&bo->base.base); + } +} + +static void virtio_gpu_free_object(struct drm_gem_object *obj) +{ + struct virtio_gpu_object *bo = gem_to_virtio_gpu_obj(obj); + struct virtio_gpu_device *vgdev = bo->base.base.dev->dev_private; - drm_gem_shmem_free_object(obj); + if (bo->created) { + virtio_gpu_cmd_unref_resource(vgdev, bo); + virtio_gpu_notify(vgdev); + /* completion handler calls virtio_gpu_cleanup_object() */ + return; + } + virtio_gpu_cleanup_object(bo); } -static const struct drm_gem_object_funcs virtio_gpu_gem_funcs = { +static const struct drm_gem_object_funcs virtio_gpu_shmem_funcs = { .free = virtio_gpu_free_object, .open = virtio_gpu_gem_object_open, .close = virtio_gpu_gem_object_close, @@ -86,20 +112,75 @@ static const struct drm_gem_object_funcs virtio_gpu_gem_funcs = { .get_sg_table = drm_gem_shmem_get_sg_table, .vmap = drm_gem_shmem_vmap, .vunmap = drm_gem_shmem_vunmap, - .mmap = &drm_gem_shmem_mmap, + .mmap = drm_gem_shmem_mmap, }; +bool virtio_gpu_is_shmem(struct virtio_gpu_object *bo) +{ + return bo->base.base.funcs == &virtio_gpu_shmem_funcs; +} + struct drm_gem_object *virtio_gpu_create_object(struct drm_device *dev, size_t size) { - struct virtio_gpu_object *bo; + struct virtio_gpu_object_shmem *shmem; + struct drm_gem_shmem_object *dshmem; - bo = kzalloc(sizeof(*bo), GFP_KERNEL); - if (!bo) + shmem = kzalloc(sizeof(*shmem), GFP_KERNEL); + if (!shmem) return NULL; - bo->base.base.funcs = &virtio_gpu_gem_funcs; - return &bo->base.base; + dshmem = &shmem->base.base; + dshmem->base.funcs = &virtio_gpu_shmem_funcs; + dshmem->map_cached = true; + return &dshmem->base; +} + +static int virtio_gpu_object_shmem_init(struct virtio_gpu_device *vgdev, + struct virtio_gpu_object *bo, + struct virtio_gpu_mem_entry **ents, + unsigned int *nents) +{ + bool use_dma_api = !virtio_has_iommu_quirk(vgdev->vdev); + struct virtio_gpu_object_shmem *shmem = to_virtio_gpu_shmem(bo); + struct scatterlist *sg; + int si, ret; + + ret = drm_gem_shmem_pin(&bo->base.base); + if (ret < 0) + return -EINVAL; + + shmem->pages = drm_gem_shmem_get_sg_table(&bo->base.base); + if (!shmem->pages) { + drm_gem_shmem_unpin(&bo->base.base); + return -EINVAL; + } + + if (use_dma_api) { + shmem->mapped = dma_map_sg(vgdev->vdev->dev.parent, + shmem->pages->sgl, + shmem->pages->nents, + DMA_TO_DEVICE); + *nents = shmem->mapped; + } else { + *nents = shmem->pages->nents; + } + + *ents = kmalloc_array(*nents, sizeof(struct virtio_gpu_mem_entry), + GFP_KERNEL); + if (!(*ents)) { + DRM_ERROR("failed to allocate ent list\n"); + return -ENOMEM; + } + + for_each_sg(shmem->pages->sgl, sg, *nents, si) { + (*ents)[si].addr = cpu_to_le64(use_dma_api + ? sg_dma_address(sg) + : sg_phys(sg)); + (*ents)[si].length = cpu_to_le32(sg->length); + (*ents)[si].padding = 0; + } + return 0; } int virtio_gpu_object_create(struct virtio_gpu_device *vgdev, @@ -110,6 +191,8 @@ int virtio_gpu_object_create(struct virtio_gpu_device *vgdev, struct virtio_gpu_object_array *objs = NULL; struct drm_gem_shmem_object *shmem_obj; struct virtio_gpu_object *bo; + struct virtio_gpu_mem_entry *ents; + unsigned int nents; int ret; *bo_ptr = NULL; @@ -146,12 +229,19 @@ int virtio_gpu_object_create(struct virtio_gpu_device *vgdev, objs, fence); } - ret = virtio_gpu_object_attach(vgdev, bo, NULL); + ret = virtio_gpu_object_shmem_init(vgdev, bo, &ents, &nents); + if (ret != 0) { + virtio_gpu_free_object(&shmem_obj->base); + return ret; + } + + ret = virtio_gpu_object_attach(vgdev, bo, ents, nents); if (ret != 0) { virtio_gpu_free_object(&shmem_obj->base); return ret; } + virtio_gpu_notify(vgdev); *bo_ptr = bo; return 0; diff --git a/drivers/gpu/drm/virtio/virtgpu_plane.c b/drivers/gpu/drm/virtio/virtgpu_plane.c index d1c3f5fbfee4..52d24179bcec 100644 --- a/drivers/gpu/drm/virtio/virtgpu_plane.c +++ b/drivers/gpu/drm/virtio/virtgpu_plane.c @@ -148,14 +148,13 @@ static void virtio_gpu_primary_plane_update(struct drm_plane *plane, plane->state->src_w >> 16, plane->state->src_h >> 16, 0, 0); + virtio_gpu_notify(vgdev); return; } if (!drm_atomic_helper_damage_merged(old_state, plane->state, &rect)) return; - virtio_gpu_disable_notify(vgdev); - bo = gem_to_virtio_gpu_obj(plane->state->fb->obj[0]); if (bo->dumb) virtio_gpu_update_dumb_bo(vgdev, plane->state, &rect); @@ -186,8 +185,7 @@ static void virtio_gpu_primary_plane_update(struct drm_plane *plane, rect.y1, rect.x2 - rect.x1, rect.y2 - rect.y1); - - virtio_gpu_enable_notify(vgdev); + virtio_gpu_notify(vgdev); } static int virtio_gpu_cursor_prepare_fb(struct drm_plane *plane, @@ -266,6 +264,7 @@ static void virtio_gpu_cursor_plane_update(struct drm_plane *plane, plane->state->crtc_w, plane->state->crtc_h, 0, 0, objs, vgfb->fence); + virtio_gpu_notify(vgdev); dma_fence_wait(&vgfb->fence->f, true); dma_fence_put(&vgfb->fence->f); vgfb->fence = NULL; diff --git a/drivers/gpu/drm/virtio/virtgpu_vq.c b/drivers/gpu/drm/virtio/virtgpu_vq.c index cc02fc4bab2a..73854915ec34 100644 --- a/drivers/gpu/drm/virtio/virtgpu_vq.c +++ b/drivers/gpu/drm/virtio/virtgpu_vq.c @@ -120,23 +120,6 @@ virtio_gpu_vbuf_ctrl_hdr(struct virtio_gpu_vbuffer *vbuf) return (struct virtio_gpu_ctrl_hdr *)vbuf->buf; } -static void *virtio_gpu_alloc_cmd(struct virtio_gpu_device *vgdev, - struct virtio_gpu_vbuffer **vbuffer_p, - int size) -{ - struct virtio_gpu_vbuffer *vbuf; - - vbuf = virtio_gpu_get_vbuf(vgdev, size, - sizeof(struct virtio_gpu_ctrl_hdr), - NULL, NULL); - if (IS_ERR(vbuf)) { - *vbuffer_p = NULL; - return ERR_CAST(vbuf); - } - *vbuffer_p = vbuf; - return vbuf->buf; -} - static struct virtio_gpu_update_cursor* virtio_gpu_alloc_cursor(struct virtio_gpu_device *vgdev, struct virtio_gpu_vbuffer **vbuffer_p) @@ -172,6 +155,25 @@ static void *virtio_gpu_alloc_cmd_resp(struct virtio_gpu_device *vgdev, return (struct virtio_gpu_command *)vbuf->buf; } +static void *virtio_gpu_alloc_cmd(struct virtio_gpu_device *vgdev, + struct virtio_gpu_vbuffer **vbuffer_p, + int size) +{ + return virtio_gpu_alloc_cmd_resp(vgdev, NULL, vbuffer_p, size, + sizeof(struct virtio_gpu_ctrl_hdr), + NULL); +} + +static void *virtio_gpu_alloc_cmd_cb(struct virtio_gpu_device *vgdev, + struct virtio_gpu_vbuffer **vbuffer_p, + int size, + virtio_gpu_resp_cb cb) +{ + return virtio_gpu_alloc_cmd_resp(vgdev, cb, vbuffer_p, size, + sizeof(struct virtio_gpu_ctrl_hdr), + NULL); +} + static void free_vbuf(struct virtio_gpu_device *vgdev, struct virtio_gpu_vbuffer *vbuf) { @@ -220,7 +222,7 @@ void virtio_gpu_dequeue_ctrl_func(struct work_struct *work) trace_virtio_gpu_cmd_response(vgdev->ctrlq.vq, resp); if (resp->type != cpu_to_le32(VIRTIO_GPU_RESP_OK_NODATA)) { - if (resp->type >= cpu_to_le32(VIRTIO_GPU_RESP_ERR_UNSPEC)) { + if (le32_to_cpu(resp->type) >= VIRTIO_GPU_RESP_ERR_UNSPEC) { struct virtio_gpu_ctrl_hdr *cmd; cmd = virtio_gpu_vbuf_ctrl_hdr(entry); DRM_ERROR_RATELIMITED("response 0x%x (command 0x%x)\n", @@ -327,8 +329,14 @@ static void virtio_gpu_queue_ctrl_sgs(struct virtio_gpu_device *vgdev, int incnt) { struct virtqueue *vq = vgdev->ctrlq.vq; - bool notify = false; - int ret; + int ret, idx; + + if (!drm_dev_enter(vgdev->ddev, &idx)) { + if (fence && vbuf->objs) + virtio_gpu_array_unlock_resv(vbuf->objs); + free_vbuf(vgdev, vbuf); + return; + } if (vgdev->has_indirect) elemcnt = 1; @@ -336,16 +344,9 @@ static void virtio_gpu_queue_ctrl_sgs(struct virtio_gpu_device *vgdev, again: spin_lock(&vgdev->ctrlq.qlock); - if (!vgdev->vqs_ready) { - spin_unlock(&vgdev->ctrlq.qlock); - - if (fence && vbuf->objs) - virtio_gpu_array_unlock_resv(vbuf->objs); - return; - } - if (vq->num_free < elemcnt) { spin_unlock(&vgdev->ctrlq.qlock); + virtio_gpu_notify(vgdev); wait_event(vgdev->ctrlq.ack_queue, vq->num_free >= elemcnt); goto again; } @@ -367,16 +368,11 @@ again: trace_virtio_gpu_cmd_queue(vq, virtio_gpu_vbuf_ctrl_hdr(vbuf)); - notify = virtqueue_kick_prepare(vq); + atomic_inc(&vgdev->pending_commands); spin_unlock(&vgdev->ctrlq.qlock); - if (notify) { - if (vgdev->disable_notify) - vgdev->pending_notify = true; - else - virtqueue_notify(vq); - } + drm_dev_exit(idx); } static void virtio_gpu_queue_fenced_ctrl_buffer(struct virtio_gpu_device *vgdev, @@ -432,19 +428,20 @@ static void virtio_gpu_queue_fenced_ctrl_buffer(struct virtio_gpu_device *vgdev, } } -void virtio_gpu_disable_notify(struct virtio_gpu_device *vgdev) -{ - vgdev->disable_notify = true; -} - -void virtio_gpu_enable_notify(struct virtio_gpu_device *vgdev) +void virtio_gpu_notify(struct virtio_gpu_device *vgdev) { - vgdev->disable_notify = false; + bool notify; - if (!vgdev->pending_notify) + if (!atomic_read(&vgdev->pending_commands)) return; - vgdev->pending_notify = false; - virtqueue_notify(vgdev->ctrlq.vq); + + spin_lock(&vgdev->ctrlq.qlock); + atomic_set(&vgdev->pending_commands, 0); + notify = virtqueue_kick_prepare(vgdev->ctrlq.vq); + spin_unlock(&vgdev->ctrlq.qlock); + + if (notify) + virtqueue_notify(vgdev->ctrlq.vq); } static void virtio_gpu_queue_ctrl_buffer(struct virtio_gpu_device *vgdev, @@ -458,12 +455,13 @@ static void virtio_gpu_queue_cursor(struct virtio_gpu_device *vgdev, { struct virtqueue *vq = vgdev->cursorq.vq; struct scatterlist *sgs[1], ccmd; + int idx, ret, outcnt; bool notify; - int ret; - int outcnt; - if (!vgdev->vqs_ready) + if (!drm_dev_enter(vgdev->ddev, &idx)) { + free_vbuf(vgdev, vbuf); return; + } sg_init_one(&ccmd, vbuf->buf, vbuf->size); sgs[0] = &ccmd; @@ -488,6 +486,8 @@ retry: if (notify) virtqueue_notify(vq); + + drm_dev_exit(idx); } /* just create gem objects for userspace and long lived objects, @@ -518,35 +518,32 @@ void virtio_gpu_cmd_create_resource(struct virtio_gpu_device *vgdev, bo->created = true; } -void virtio_gpu_cmd_unref_resource(struct virtio_gpu_device *vgdev, - uint32_t resource_id) +static void virtio_gpu_cmd_unref_cb(struct virtio_gpu_device *vgdev, + struct virtio_gpu_vbuffer *vbuf) { - struct virtio_gpu_resource_unref *cmd_p; - struct virtio_gpu_vbuffer *vbuf; - - cmd_p = virtio_gpu_alloc_cmd(vgdev, &vbuf, sizeof(*cmd_p)); - memset(cmd_p, 0, sizeof(*cmd_p)); + struct virtio_gpu_object *bo; - cmd_p->hdr.type = cpu_to_le32(VIRTIO_GPU_CMD_RESOURCE_UNREF); - cmd_p->resource_id = cpu_to_le32(resource_id); + bo = vbuf->resp_cb_data; + vbuf->resp_cb_data = NULL; - virtio_gpu_queue_ctrl_buffer(vgdev, vbuf); + virtio_gpu_cleanup_object(bo); } -static void virtio_gpu_cmd_resource_inval_backing(struct virtio_gpu_device *vgdev, - uint32_t resource_id, - struct virtio_gpu_fence *fence) +void virtio_gpu_cmd_unref_resource(struct virtio_gpu_device *vgdev, + struct virtio_gpu_object *bo) { - struct virtio_gpu_resource_detach_backing *cmd_p; + struct virtio_gpu_resource_unref *cmd_p; struct virtio_gpu_vbuffer *vbuf; - cmd_p = virtio_gpu_alloc_cmd(vgdev, &vbuf, sizeof(*cmd_p)); + cmd_p = virtio_gpu_alloc_cmd_cb(vgdev, &vbuf, sizeof(*cmd_p), + virtio_gpu_cmd_unref_cb); memset(cmd_p, 0, sizeof(*cmd_p)); - cmd_p->hdr.type = cpu_to_le32(VIRTIO_GPU_CMD_RESOURCE_DETACH_BACKING); - cmd_p->resource_id = cpu_to_le32(resource_id); + cmd_p->hdr.type = cpu_to_le32(VIRTIO_GPU_CMD_RESOURCE_UNREF); + cmd_p->resource_id = cpu_to_le32(bo->hw_res_handle); - virtio_gpu_queue_fenced_ctrl_buffer(vgdev, vbuf, fence); + vbuf->resp_cb_data = bo; + virtio_gpu_queue_ctrl_buffer(vgdev, vbuf); } void virtio_gpu_cmd_set_scanout(struct virtio_gpu_device *vgdev, @@ -603,10 +600,11 @@ void virtio_gpu_cmd_transfer_to_host_2d(struct virtio_gpu_device *vgdev, struct virtio_gpu_transfer_to_host_2d *cmd_p; struct virtio_gpu_vbuffer *vbuf; bool use_dma_api = !virtio_has_iommu_quirk(vgdev->vdev); + struct virtio_gpu_object_shmem *shmem = to_virtio_gpu_shmem(bo); if (use_dma_api) dma_sync_sg_for_device(vgdev->vdev->dev.parent, - bo->pages->sgl, bo->pages->nents, + shmem->pages->sgl, shmem->pages->nents, DMA_TO_DEVICE); cmd_p = virtio_gpu_alloc_cmd(vgdev, &vbuf, sizeof(*cmd_p)); @@ -954,7 +952,6 @@ void virtio_gpu_cmd_context_attach_resource(struct virtio_gpu_device *vgdev, cmd_p->hdr.ctx_id = cpu_to_le32(ctx_id); cmd_p->resource_id = cpu_to_le32(bo->hw_res_handle); virtio_gpu_queue_ctrl_buffer(vgdev, vbuf); - } void virtio_gpu_cmd_context_detach_resource(struct virtio_gpu_device *vgdev, @@ -1004,6 +1001,7 @@ virtio_gpu_cmd_resource_create_3d(struct virtio_gpu_device *vgdev, cmd_p->flags = cpu_to_le32(params->flags); virtio_gpu_queue_fenced_ctrl_buffer(vgdev, vbuf, fence); + bo->created = true; } @@ -1018,10 +1016,11 @@ void virtio_gpu_cmd_transfer_to_host_3d(struct virtio_gpu_device *vgdev, struct virtio_gpu_transfer_host_3d *cmd_p; struct virtio_gpu_vbuffer *vbuf; bool use_dma_api = !virtio_has_iommu_quirk(vgdev->vdev); + struct virtio_gpu_object_shmem *shmem = to_virtio_gpu_shmem(bo); if (use_dma_api) dma_sync_sg_for_device(vgdev->vdev->dev.parent, - bo->pages->sgl, bo->pages->nents, + shmem->pages->sgl, shmem->pages->nents, DMA_TO_DEVICE); cmd_p = virtio_gpu_alloc_cmd(vgdev, &vbuf, sizeof(*cmd_p)); @@ -1090,89 +1089,14 @@ void virtio_gpu_cmd_submit(struct virtio_gpu_device *vgdev, int virtio_gpu_object_attach(struct virtio_gpu_device *vgdev, struct virtio_gpu_object *obj, - struct virtio_gpu_fence *fence) + struct virtio_gpu_mem_entry *ents, + unsigned int nents) { - bool use_dma_api = !virtio_has_iommu_quirk(vgdev->vdev); - struct virtio_gpu_mem_entry *ents; - struct scatterlist *sg; - int si, nents, ret; - - if (WARN_ON_ONCE(!obj->created)) - return -EINVAL; - if (WARN_ON_ONCE(obj->pages)) - return -EINVAL; - - ret = drm_gem_shmem_pin(&obj->base.base); - if (ret < 0) - return -EINVAL; - - obj->pages = drm_gem_shmem_get_sg_table(&obj->base.base); - if (obj->pages == NULL) { - drm_gem_shmem_unpin(&obj->base.base); - return -EINVAL; - } - - if (use_dma_api) { - obj->mapped = dma_map_sg(vgdev->vdev->dev.parent, - obj->pages->sgl, obj->pages->nents, - DMA_TO_DEVICE); - nents = obj->mapped; - } else { - nents = obj->pages->nents; - } - - /* gets freed when the ring has consumed it */ - ents = kmalloc_array(nents, sizeof(struct virtio_gpu_mem_entry), - GFP_KERNEL); - if (!ents) { - DRM_ERROR("failed to allocate ent list\n"); - return -ENOMEM; - } - - for_each_sg(obj->pages->sgl, sg, nents, si) { - ents[si].addr = cpu_to_le64(use_dma_api - ? sg_dma_address(sg) - : sg_phys(sg)); - ents[si].length = cpu_to_le32(sg->length); - ents[si].padding = 0; - } - virtio_gpu_cmd_resource_attach_backing(vgdev, obj->hw_res_handle, - ents, nents, - fence); + ents, nents, NULL); return 0; } -void virtio_gpu_object_detach(struct virtio_gpu_device *vgdev, - struct virtio_gpu_object *obj) -{ - bool use_dma_api = !virtio_has_iommu_quirk(vgdev->vdev); - - if (WARN_ON_ONCE(!obj->pages)) - return; - - if (use_dma_api && obj->mapped) { - struct virtio_gpu_fence *fence = virtio_gpu_fence_alloc(vgdev); - /* detach backing and wait for the host process it ... */ - virtio_gpu_cmd_resource_inval_backing(vgdev, obj->hw_res_handle, fence); - dma_fence_wait(&fence->f, true); - dma_fence_put(&fence->f); - - /* ... then tear down iommu mappings */ - dma_unmap_sg(vgdev->vdev->dev.parent, - obj->pages->sgl, obj->mapped, - DMA_TO_DEVICE); - obj->mapped = 0; - } else { - virtio_gpu_cmd_resource_inval_backing(vgdev, obj->hw_res_handle, NULL); - } - - sg_free_table(obj->pages); - obj->pages = NULL; - - drm_gem_shmem_unpin(&obj->base.base); -} - void virtio_gpu_cursor_ping(struct virtio_gpu_device *vgdev, struct virtio_gpu_output *output) { diff --git a/drivers/gpu/drm/vkms/vkms_crtc.c b/drivers/gpu/drm/vkms/vkms_crtc.c index 74f703b8d22a..ac85e17428f8 100644 --- a/drivers/gpu/drm/vkms/vkms_crtc.c +++ b/drivers/gpu/drm/vkms/vkms_crtc.c @@ -76,10 +76,12 @@ static void vkms_disable_vblank(struct drm_crtc *crtc) hrtimer_cancel(&out->vblank_hrtimer); } -bool vkms_get_vblank_timestamp(struct drm_device *dev, unsigned int pipe, - int *max_error, ktime_t *vblank_time, - bool in_vblank_irq) +static bool vkms_get_vblank_timestamp(struct drm_crtc *crtc, + int *max_error, ktime_t *vblank_time, + bool in_vblank_irq) { + struct drm_device *dev = crtc->dev; + unsigned int pipe = crtc->index; struct vkms_device *vkmsdev = drm_device_to_vkms_device(dev); struct vkms_output *output = &vkmsdev->output; struct drm_vblank_crtc *vblank = &dev->vblank[pipe]; @@ -154,6 +156,7 @@ static const struct drm_crtc_funcs vkms_crtc_funcs = { .atomic_destroy_state = vkms_atomic_crtc_destroy_state, .enable_vblank = vkms_enable_vblank, .disable_vblank = vkms_disable_vblank, + .get_vblank_timestamp = vkms_get_vblank_timestamp, .get_crc_sources = vkms_get_crc_sources, .set_crc_source = vkms_set_crc_source, .verify_crc_source = vkms_verify_crc_source, diff --git a/drivers/gpu/drm/vkms/vkms_drv.c b/drivers/gpu/drm/vkms/vkms_drv.c index 25bd7519295f..860de052e820 100644 --- a/drivers/gpu/drm/vkms/vkms_drv.c +++ b/drivers/gpu/drm/vkms/vkms_drv.c @@ -103,7 +103,6 @@ static struct drm_driver vkms_driver = { .dumb_create = vkms_dumb_create, .gem_vm_ops = &vkms_gem_vm_ops, .gem_free_object_unlocked = vkms_gem_free_object, - .get_vblank_timestamp = vkms_get_vblank_timestamp, .prime_fd_to_handle = drm_gem_prime_fd_to_handle, .gem_prime_import_sg_table = vkms_prime_import_sg_table, diff --git a/drivers/gpu/drm/vkms/vkms_drv.h b/drivers/gpu/drm/vkms/vkms_drv.h index 7d52e24564db..eda04ffba7b1 100644 --- a/drivers/gpu/drm/vkms/vkms_drv.h +++ b/drivers/gpu/drm/vkms/vkms_drv.h @@ -111,10 +111,6 @@ struct vkms_gem_object { int vkms_crtc_init(struct drm_device *dev, struct drm_crtc *crtc, struct drm_plane *primary, struct drm_plane *cursor); -bool vkms_get_vblank_timestamp(struct drm_device *dev, unsigned int pipe, - int *max_error, ktime_t *vblank_time, - bool in_vblank_irq); - int vkms_output_init(struct vkms_device *vkmsdev, int index); struct drm_plane *vkms_plane_init(struct vkms_device *vkmsdev, diff --git a/drivers/gpu/drm/vmwgfx/Makefile b/drivers/gpu/drm/vmwgfx/Makefile index c877a21a0739..31f85f09f1fc 100644 --- a/drivers/gpu/drm/vmwgfx/Makefile +++ b/drivers/gpu/drm/vmwgfx/Makefile @@ -8,7 +8,8 @@ vmwgfx-y := vmwgfx_execbuf.o vmwgfx_gmr.o vmwgfx_kms.o vmwgfx_drv.o \ vmwgfx_cmdbuf_res.o vmwgfx_cmdbuf.o vmwgfx_stdu.o \ vmwgfx_cotable.o vmwgfx_so.o vmwgfx_binding.o vmwgfx_msg.o \ vmwgfx_simple_resource.o vmwgfx_va.o vmwgfx_blit.o \ - vmwgfx_validation.o vmwgfx_page_dirty.o \ + vmwgfx_validation.o vmwgfx_page_dirty.o vmwgfx_streamoutput.o \ ttm_object.o ttm_lock.o +vmwgfx-$(CONFIG_TRANSPARENT_HUGEPAGE) += vmwgfx_thp.o obj-$(CONFIG_DRM_VMWGFX) := vmwgfx.o diff --git a/drivers/gpu/drm/vmwgfx/device_include/svga3d_cmd.h b/drivers/gpu/drm/vmwgfx/device_include/svga3d_cmd.h index 9cbba0e8ce6a..799bc0963f7a 100644 --- a/drivers/gpu/drm/vmwgfx/device_include/svga3d_cmd.h +++ b/drivers/gpu/drm/vmwgfx/device_include/svga3d_cmd.h @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: GPL-2.0 OR MIT */ /********************************************************** - * Copyright 1998-2015 VMware, Inc. + * Copyright 1998-2020 VMware, Inc. * * Permission is hereby granted, free of charge, to any person * obtaining a copy of this software and associated documentation @@ -104,12 +104,12 @@ typedef enum { SVGA_3D_CMD_DEAD1 = 1083, SVGA_3D_CMD_DEAD2 = 1084, - SVGA_3D_CMD_LOGICOPS_BITBLT = 1085, - SVGA_3D_CMD_LOGICOPS_TRANSBLT = 1086, - SVGA_3D_CMD_LOGICOPS_STRETCHBLT = 1087, - SVGA_3D_CMD_LOGICOPS_COLORFILL = 1088, - SVGA_3D_CMD_LOGICOPS_ALPHABLEND = 1089, - SVGA_3D_CMD_LOGICOPS_CLEARTYPEBLEND = 1090, + SVGA_3D_CMD_DEAD12 = 1085, + SVGA_3D_CMD_DEAD13 = 1086, + SVGA_3D_CMD_DEAD14 = 1087, + SVGA_3D_CMD_DEAD15 = 1088, + SVGA_3D_CMD_DEAD16 = 1089, + SVGA_3D_CMD_DEAD17 = 1090, SVGA_3D_CMD_SET_OTABLE_BASE = 1091, SVGA_3D_CMD_READBACK_OTABLE = 1092, @@ -261,30 +261,23 @@ typedef enum { SVGA_3D_CMD_DX_SET_VS_CONSTANT_BUFFER_OFFSET = 1220, SVGA_3D_CMD_DX_SET_PS_CONSTANT_BUFFER_OFFSET = 1221, SVGA_3D_CMD_DX_SET_GS_CONSTANT_BUFFER_OFFSET = 1222, - - /* - * Reserve some IDs to be used for the SM5 shader types. - */ - SVGA_3D_CMD_DX_RESERVED1 = 1223, - SVGA_3D_CMD_DX_RESERVED2 = 1224, - SVGA_3D_CMD_DX_RESERVED3 = 1225, + SVGA_3D_CMD_DX_SET_HS_CONSTANT_BUFFER_OFFSET = 1223, + SVGA_3D_CMD_DX_SET_DS_CONSTANT_BUFFER_OFFSET = 1224, + SVGA_3D_CMD_DX_SET_CS_CONSTANT_BUFFER_OFFSET = 1225, SVGA_3D_CMD_DX_COND_BIND_ALL_SHADER = 1226, SVGA_3D_CMD_DX_MAX = 1227, SVGA_3D_CMD_SCREEN_COPY = 1227, - /* - * Reserve some IDs to be used for video. - */ - SVGA_3D_CMD_VIDEO_RESERVED1 = 1228, - SVGA_3D_CMD_VIDEO_RESERVED2 = 1229, - SVGA_3D_CMD_VIDEO_RESERVED3 = 1230, - SVGA_3D_CMD_VIDEO_RESERVED4 = 1231, - SVGA_3D_CMD_VIDEO_RESERVED5 = 1232, - SVGA_3D_CMD_VIDEO_RESERVED6 = 1233, - SVGA_3D_CMD_VIDEO_RESERVED7 = 1234, - SVGA_3D_CMD_VIDEO_RESERVED8 = 1235, + SVGA_3D_CMD_RESERVED1 = 1228, + SVGA_3D_CMD_RESERVED2 = 1229, + SVGA_3D_CMD_RESERVED3 = 1230, + SVGA_3D_CMD_RESERVED4 = 1231, + SVGA_3D_CMD_RESERVED5 = 1232, + SVGA_3D_CMD_RESERVED6 = 1233, + SVGA_3D_CMD_RESERVED7 = 1234, + SVGA_3D_CMD_RESERVED8 = 1235, SVGA_3D_CMD_GROW_OTABLE = 1236, SVGA_3D_CMD_DX_GROW_COTABLE = 1237, @@ -298,7 +291,46 @@ typedef enum { SVGA_3D_CMD_DX_PRED_CONVERT = 1243, SVGA_3D_CMD_WHOLE_SURFACE_COPY = 1244, - SVGA_3D_CMD_MAX = 1245, + SVGA_3D_CMD_DX_DEFINE_UA_VIEW = 1245, + SVGA_3D_CMD_DX_DESTROY_UA_VIEW = 1246, + SVGA_3D_CMD_DX_CLEAR_UA_VIEW_UINT = 1247, + SVGA_3D_CMD_DX_CLEAR_UA_VIEW_FLOAT = 1248, + SVGA_3D_CMD_DX_COPY_STRUCTURE_COUNT = 1249, + SVGA_3D_CMD_DX_SET_UA_VIEWS = 1250, + + SVGA_3D_CMD_DX_DRAW_INDEXED_INSTANCED_INDIRECT = 1251, + SVGA_3D_CMD_DX_DRAW_INSTANCED_INDIRECT = 1252, + SVGA_3D_CMD_DX_DISPATCH = 1253, + SVGA_3D_CMD_DX_DISPATCH_INDIRECT = 1254, + + SVGA_3D_CMD_WRITE_ZERO_SURFACE = 1255, + SVGA_3D_CMD_HINT_ZERO_SURFACE = 1256, + SVGA_3D_CMD_DX_TRANSFER_TO_BUFFER = 1257, + SVGA_3D_CMD_DX_SET_STRUCTURE_COUNT = 1258, + + SVGA_3D_CMD_LOGICOPS_BITBLT = 1259, + SVGA_3D_CMD_LOGICOPS_TRANSBLT = 1260, + SVGA_3D_CMD_LOGICOPS_STRETCHBLT = 1261, + SVGA_3D_CMD_LOGICOPS_COLORFILL = 1262, + SVGA_3D_CMD_LOGICOPS_ALPHABLEND = 1263, + SVGA_3D_CMD_LOGICOPS_CLEARTYPEBLEND = 1264, + + SVGA_3D_CMD_RESERVED2_1 = 1265, + + SVGA_3D_CMD_RESERVED2_2 = 1266, + SVGA_3D_CMD_DEFINE_GB_SURFACE_V4 = 1267, + SVGA_3D_CMD_DX_SET_CS_UA_VIEWS = 1268, + SVGA_3D_CMD_DX_SET_MIN_LOD = 1269, + SVGA_3D_CMD_RESERVED2_3 = 1270, + SVGA_3D_CMD_RESERVED2_4 = 1271, + SVGA_3D_CMD_DX_DEFINE_DEPTHSTENCIL_VIEW_V2 = 1272, + SVGA_3D_CMD_DX_DEFINE_STREAMOUTPUT_WITH_MOB = 1273, + SVGA_3D_CMD_DX_SET_SHADER_IFACE = 1274, + SVGA_3D_CMD_DX_BIND_STREAMOUTPUT = 1275, + SVGA_3D_CMD_SURFACE_STRETCHBLT_NON_MS_TO_MS = 1276, + SVGA_3D_CMD_DX_BIND_SHADER_IFACE = 1277, + + SVGA_3D_CMD_MAX = 1278, SVGA_3D_CMD_FUTURE_MAX = 3000 } SVGAFifo3dCmdId; @@ -334,6 +366,7 @@ struct { uint32 sid; SVGA3dSurface1Flags surfaceFlags; SVGA3dSurfaceFormat format; + /* * If surfaceFlags has SVGA3D_SURFACE_CUBEMAP bit set, all SVGA3dSurfaceFace * structures must have the same value of numMipLevels field. @@ -341,6 +374,7 @@ struct { * numMipLevels set to 0. */ SVGA3dSurfaceFace face[SVGA3D_MAX_SURFACE_FACES]; + /* * Followed by an SVGA3dSize structure for each mip level in each face. * @@ -360,6 +394,7 @@ struct { uint32 sid; SVGA3dSurface1Flags surfaceFlags; SVGA3dSurfaceFormat format; + /* * If surfaceFlags has SVGA3D_SURFACE_CUBEMAP bit set, all SVGA3dSurfaceFace * structures must have the same value of numMipLevels field. @@ -369,6 +404,7 @@ struct { SVGA3dSurfaceFace face[SVGA3D_MAX_SURFACE_FACES]; uint32 multisampleCount; SVGA3dTextureFilter autogenFilter; + /* * Followed by an SVGA3dSize structure for each mip level in each face. * @@ -517,6 +553,18 @@ typedef struct { SVGA3dSurfaceImageId src; SVGA3dSurfaceImageId dest; + SVGA3dBox boxSrc; + SVGA3dBox boxDest; +} +#include "vmware_pack_end.h" +SVGA3dCmdSurfaceStretchBltNonMSToMS; +/* SVGA_3D_CMD_SURFACE_STRETCHBLT_NON_MS_TO_MS */ + +typedef +#include "vmware_pack_begin.h" +struct { + SVGA3dSurfaceImageId src; + SVGA3dSurfaceImageId dest; SVGA3dBox boxSrc; SVGA3dBox boxDest; SVGA3dStretchBltMode mode; @@ -555,6 +603,7 @@ struct { SVGAGuestImage guest; SVGA3dSurfaceImageId host; SVGA3dTransferType transfer; + /* * Followed by variable number of SVGA3dCopyBox structures. For consistency * in all clipping logic and coordinate translation, we define the @@ -789,7 +838,7 @@ struct { uint32 indexBufferSid; /* Valid index buffer sid. */ uint32 indexBufferOffset; /* Byte offset into the vertex buffer, almost */ - /* always 0 for DX9 guests, non-zero for OpenGL */ + /* always 0 for pre SM guests, non-zero for OpenGL */ /* guests. We can't represent non-multiple of */ /* stride offsets in D3D9Renderer... */ uint8 indexBufferStride; /* Allowable values = 1, 2, or 4 */ @@ -1228,6 +1277,7 @@ struct SVGA3dCmdLogicOpsBitBlt { SVGA3dSurfaceImageId src; SVGA3dSurfaceImageId dst; SVGA3dLogicOp logicOp; + SVGA3dLogicOpRop3 logicOpRop3; /* Followed by variable number of SVGA3dCopyBox structures */ } #include "vmware_pack_end.h" @@ -1247,7 +1297,8 @@ struct SVGA3dCmdLogicOpsTransBlt { uint32 color; uint32 flags; SVGA3dBox srcBox; - SVGA3dBox dstBox; + SVGA3dSignedBox dstBox; + SVGA3dBox clipBox; } #include "vmware_pack_end.h" SVGA3dCmdLogicOpsTransBlt; /* SVGA_3D_CMD_LOGICOPS_TRANSBLT */ @@ -1266,7 +1317,8 @@ struct SVGA3dCmdLogicOpsStretchBlt { uint16 mode; uint16 flags; SVGA3dBox srcBox; - SVGA3dBox dstBox; + SVGA3dSignedBox dstBox; + SVGA3dBox clipBox; } #include "vmware_pack_end.h" SVGA3dCmdLogicOpsStretchBlt; /* SVGA_3D_CMD_LOGICOPS_STRETCHBLT */ @@ -1283,6 +1335,7 @@ struct SVGA3dCmdLogicOpsColorFill { SVGA3dSurfaceImageId dst; uint32 color; SVGA3dLogicOp logicOp; + SVGA3dLogicOpRop3 logicOpRop3; /* Followed by variable number of SVGA3dRect structures. */ } #include "vmware_pack_end.h" @@ -1302,7 +1355,8 @@ struct SVGA3dCmdLogicOpsAlphaBlend { uint32 alphaVal; uint32 flags; SVGA3dBox srcBox; - SVGA3dBox dstBox; + SVGA3dSignedBox dstBox; + SVGA3dBox clipBox; } #include "vmware_pack_end.h" SVGA3dCmdLogicOpsAlphaBlend; /* SVGA_3D_CMD_LOGICOPS_ALPHABLEND */ @@ -1365,8 +1419,9 @@ struct { SVGA3dSurface2Flags surface2Flags; uint8 multisamplePattern; uint8 qualityLevel; - uint8 pad0[2]; - uint32 pad1[3]; + uint16 bufferByteStride; + float minLOD; + uint32 pad0[2]; } #include "vmware_pack_end.h" SVGAOTableSurfaceEntry; @@ -1543,7 +1598,7 @@ typedef #include "vmware_pack_begin.h" struct { SVGAOTableType type; - PPN baseAddress; + PPN32 baseAddress; uint32 sizeInBytes; uint32 validSizeInBytes; SVGAMobFormat ptDepth; @@ -1599,7 +1654,7 @@ typedef struct SVGA3dCmdDefineGBMob { SVGAMobId mobid; SVGAMobFormat ptDepth; - PPN base; + PPN32 base; uint32 sizeInBytes; } #include "vmware_pack_end.h" @@ -1618,7 +1673,6 @@ struct SVGA3dCmdDestroyGBMob { #include "vmware_pack_end.h" SVGA3dCmdDestroyGBMob; /* SVGA_3D_CMD_DESTROY_GB_MOB */ - /* * Define a memory object (Mob) in the OTable with a PPN64 base. */ @@ -1719,6 +1773,27 @@ struct SVGA3dCmdDefineGBSurface_v3 { SVGA3dCmdDefineGBSurface_v3; /* SVGA_3D_CMD_DEFINE_GB_SURFACE_V3 */ /* + * Defines a guest-backed surface, adding buffer byte stride. + */ +typedef +#include "vmware_pack_begin.h" +struct SVGA3dCmdDefineGBSurface_v4 { + uint32 sid; + SVGA3dSurfaceAllFlags surfaceFlags; + SVGA3dSurfaceFormat format; + uint32 numMipLevels; + uint32 multisampleCount; + SVGA3dMSPattern multisamplePattern; + SVGA3dMSQualityLevel qualityLevel; + SVGA3dTextureFilter autogenFilter; + SVGA3dSize size; + uint32 arraySize; + uint32 bufferByteStride; +} +#include "vmware_pack_end.h" +SVGA3dCmdDefineGBSurface_v4; /* SVGA_3D_CMD_DEFINE_GB_SURFACE_V4 */ + +/* * Destroy a guest-backed surface. */ @@ -2181,4 +2256,20 @@ SVGA3dCmdScreenCopy; /* SVGA_3D_CMD_SCREEN_COPY */ #define SVGA_SCREEN_COPY_STATUS_SUCCESS 0x01 #define SVGA_SCREEN_COPY_STATUS_INVALID 0xFFFFFFFF +typedef +#include "vmware_pack_begin.h" +struct { + uint32 sid; +} +#include "vmware_pack_end.h" +SVGA3dCmdWriteZeroSurface; /* SVGA_3D_CMD_WRITE_ZERO_SURFACE */ + +typedef +#include "vmware_pack_begin.h" +struct { + uint32 sid; +} +#include "vmware_pack_end.h" +SVGA3dCmdHintZeroSurface; /* SVGA_3D_CMD_HINT_ZERO_SURFACE */ + #endif /* _SVGA3D_CMD_H_ */ diff --git a/drivers/gpu/drm/vmwgfx/device_include/svga3d_devcaps.h b/drivers/gpu/drm/vmwgfx/device_include/svga3d_devcaps.h index f256560049bf..617b468c626c 100644 --- a/drivers/gpu/drm/vmwgfx/device_include/svga3d_devcaps.h +++ b/drivers/gpu/drm/vmwgfx/device_include/svga3d_devcaps.h @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: GPL-2.0 OR MIT */ /********************************************************** - * Copyright 1998-2015 VMware, Inc. + * Copyright 1998-2019 VMware, Inc. * * Permission is hereby granted, free of charge, to any person * obtaining a copy of this software and associated documentation @@ -39,6 +39,8 @@ #include "includeCheck.h" +#include "svga3d_types.h" + /* * 3D Hardware Version * @@ -69,381 +71,408 @@ typedef enum { * DevCap indexes. */ -typedef enum { - SVGA3D_DEVCAP_INVALID = ((uint32)-1), - SVGA3D_DEVCAP_3D = 0, - SVGA3D_DEVCAP_MAX_LIGHTS = 1, - - /* - * SVGA3D_DEVCAP_MAX_TEXTURES reflects the maximum number of - * fixed-function texture units available. Each of these units - * work in both FFP and Shader modes, and they support texture - * transforms and texture coordinates. The host may have additional - * texture image units that are only usable with shaders. - */ - SVGA3D_DEVCAP_MAX_TEXTURES = 2, - SVGA3D_DEVCAP_MAX_CLIP_PLANES = 3, - SVGA3D_DEVCAP_VERTEX_SHADER_VERSION = 4, - SVGA3D_DEVCAP_VERTEX_SHADER = 5, - SVGA3D_DEVCAP_FRAGMENT_SHADER_VERSION = 6, - SVGA3D_DEVCAP_FRAGMENT_SHADER = 7, - SVGA3D_DEVCAP_MAX_RENDER_TARGETS = 8, - SVGA3D_DEVCAP_S23E8_TEXTURES = 9, - SVGA3D_DEVCAP_S10E5_TEXTURES = 10, - SVGA3D_DEVCAP_MAX_FIXED_VERTEXBLEND = 11, - SVGA3D_DEVCAP_D16_BUFFER_FORMAT = 12, - SVGA3D_DEVCAP_D24S8_BUFFER_FORMAT = 13, - SVGA3D_DEVCAP_D24X8_BUFFER_FORMAT = 14, - SVGA3D_DEVCAP_QUERY_TYPES = 15, - SVGA3D_DEVCAP_TEXTURE_GRADIENT_SAMPLING = 16, - SVGA3D_DEVCAP_MAX_POINT_SIZE = 17, - SVGA3D_DEVCAP_MAX_SHADER_TEXTURES = 18, - SVGA3D_DEVCAP_MAX_TEXTURE_WIDTH = 19, - SVGA3D_DEVCAP_MAX_TEXTURE_HEIGHT = 20, - SVGA3D_DEVCAP_MAX_VOLUME_EXTENT = 21, - SVGA3D_DEVCAP_MAX_TEXTURE_REPEAT = 22, - SVGA3D_DEVCAP_MAX_TEXTURE_ASPECT_RATIO = 23, - SVGA3D_DEVCAP_MAX_TEXTURE_ANISOTROPY = 24, - SVGA3D_DEVCAP_MAX_PRIMITIVE_COUNT = 25, - SVGA3D_DEVCAP_MAX_VERTEX_INDEX = 26, - SVGA3D_DEVCAP_MAX_VERTEX_SHADER_INSTRUCTIONS = 27, - SVGA3D_DEVCAP_MAX_FRAGMENT_SHADER_INSTRUCTIONS = 28, - SVGA3D_DEVCAP_MAX_VERTEX_SHADER_TEMPS = 29, - SVGA3D_DEVCAP_MAX_FRAGMENT_SHADER_TEMPS = 30, - SVGA3D_DEVCAP_TEXTURE_OPS = 31, - SVGA3D_DEVCAP_SURFACEFMT_X8R8G8B8 = 32, - SVGA3D_DEVCAP_SURFACEFMT_A8R8G8B8 = 33, - SVGA3D_DEVCAP_SURFACEFMT_A2R10G10B10 = 34, - SVGA3D_DEVCAP_SURFACEFMT_X1R5G5B5 = 35, - SVGA3D_DEVCAP_SURFACEFMT_A1R5G5B5 = 36, - SVGA3D_DEVCAP_SURFACEFMT_A4R4G4B4 = 37, - SVGA3D_DEVCAP_SURFACEFMT_R5G6B5 = 38, - SVGA3D_DEVCAP_SURFACEFMT_LUMINANCE16 = 39, - SVGA3D_DEVCAP_SURFACEFMT_LUMINANCE8_ALPHA8 = 40, - SVGA3D_DEVCAP_SURFACEFMT_ALPHA8 = 41, - SVGA3D_DEVCAP_SURFACEFMT_LUMINANCE8 = 42, - SVGA3D_DEVCAP_SURFACEFMT_Z_D16 = 43, - SVGA3D_DEVCAP_SURFACEFMT_Z_D24S8 = 44, - SVGA3D_DEVCAP_SURFACEFMT_Z_D24X8 = 45, - SVGA3D_DEVCAP_SURFACEFMT_DXT1 = 46, - SVGA3D_DEVCAP_SURFACEFMT_DXT2 = 47, - SVGA3D_DEVCAP_SURFACEFMT_DXT3 = 48, - SVGA3D_DEVCAP_SURFACEFMT_DXT4 = 49, - SVGA3D_DEVCAP_SURFACEFMT_DXT5 = 50, - SVGA3D_DEVCAP_SURFACEFMT_BUMPX8L8V8U8 = 51, - SVGA3D_DEVCAP_SURFACEFMT_A2W10V10U10 = 52, - SVGA3D_DEVCAP_SURFACEFMT_BUMPU8V8 = 53, - SVGA3D_DEVCAP_SURFACEFMT_Q8W8V8U8 = 54, - SVGA3D_DEVCAP_SURFACEFMT_CxV8U8 = 55, - SVGA3D_DEVCAP_SURFACEFMT_R_S10E5 = 56, - SVGA3D_DEVCAP_SURFACEFMT_R_S23E8 = 57, - SVGA3D_DEVCAP_SURFACEFMT_RG_S10E5 = 58, - SVGA3D_DEVCAP_SURFACEFMT_RG_S23E8 = 59, - SVGA3D_DEVCAP_SURFACEFMT_ARGB_S10E5 = 60, - SVGA3D_DEVCAP_SURFACEFMT_ARGB_S23E8 = 61, - - /* - * There is a hole in our devcap definitions for - * historical reasons. - * - * Define a constant just for completeness. - */ - SVGA3D_DEVCAP_MISSING62 = 62, - - SVGA3D_DEVCAP_MAX_VERTEX_SHADER_TEXTURES = 63, - - /* - * Note that MAX_SIMULTANEOUS_RENDER_TARGETS is a maximum count of color - * render targets. This does not include the depth or stencil targets. - */ - SVGA3D_DEVCAP_MAX_SIMULTANEOUS_RENDER_TARGETS = 64, - - SVGA3D_DEVCAP_SURFACEFMT_V16U16 = 65, - SVGA3D_DEVCAP_SURFACEFMT_G16R16 = 66, - SVGA3D_DEVCAP_SURFACEFMT_A16B16G16R16 = 67, - SVGA3D_DEVCAP_SURFACEFMT_UYVY = 68, - SVGA3D_DEVCAP_SURFACEFMT_YUY2 = 69, - SVGA3D_DEVCAP_MULTISAMPLE_NONMASKABLESAMPLES = 70, - SVGA3D_DEVCAP_MULTISAMPLE_MASKABLESAMPLES = 71, - SVGA3D_DEVCAP_ALPHATOCOVERAGE = 72, - SVGA3D_DEVCAP_SUPERSAMPLE = 73, - SVGA3D_DEVCAP_AUTOGENMIPMAPS = 74, - SVGA3D_DEVCAP_SURFACEFMT_NV12 = 75, - SVGA3D_DEVCAP_SURFACEFMT_AYUV = 76, - - /* - * This is the maximum number of SVGA context IDs that the guest - * can define using SVGA_3D_CMD_CONTEXT_DEFINE. - */ - SVGA3D_DEVCAP_MAX_CONTEXT_IDS = 77, - - /* - * This is the maximum number of SVGA surface IDs that the guest - * can define using SVGA_3D_CMD_SURFACE_DEFINE*. - */ - SVGA3D_DEVCAP_MAX_SURFACE_IDS = 78, - - SVGA3D_DEVCAP_SURFACEFMT_Z_DF16 = 79, - SVGA3D_DEVCAP_SURFACEFMT_Z_DF24 = 80, - SVGA3D_DEVCAP_SURFACEFMT_Z_D24S8_INT = 81, - - SVGA3D_DEVCAP_SURFACEFMT_ATI1 = 82, - SVGA3D_DEVCAP_SURFACEFMT_ATI2 = 83, - - /* - * Deprecated. - */ - SVGA3D_DEVCAP_DEAD1 = 84, - - /* - * This contains several SVGA_3D_CAPS_VIDEO_DECODE elements - * ored together, one for every type of video decoding supported. - */ - SVGA3D_DEVCAP_VIDEO_DECODE = 85, - - /* - * This contains several SVGA_3D_CAPS_VIDEO_PROCESS elements - * ored together, one for every type of video processing supported. - */ - SVGA3D_DEVCAP_VIDEO_PROCESS = 86, - - SVGA3D_DEVCAP_LINE_AA = 87, /* boolean */ - SVGA3D_DEVCAP_LINE_STIPPLE = 88, /* boolean */ - SVGA3D_DEVCAP_MAX_LINE_WIDTH = 89, /* float */ - SVGA3D_DEVCAP_MAX_AA_LINE_WIDTH = 90, /* float */ - - SVGA3D_DEVCAP_SURFACEFMT_YV12 = 91, - - /* - * Does the host support the SVGA logic ops commands? - */ - SVGA3D_DEVCAP_LOGICOPS = 92, - - /* - * Are TS_CONSTANT, TS_COLOR_KEY, and TS_COLOR_KEY_ENABLE supported? - */ - SVGA3D_DEVCAP_TS_COLOR_KEY = 93, /* boolean */ - - /* - * Deprecated. - */ - SVGA3D_DEVCAP_DEAD2 = 94, - - /* - * Does the device support DXContexts? - */ - SVGA3D_DEVCAP_DXCONTEXT = 95, - - /* - * What is the maximum size of a texture array? - * - * (Even if this cap is zero, cubemaps are still allowed.) - */ - SVGA3D_DEVCAP_MAX_TEXTURE_ARRAY_SIZE = 96, - - /* - * What is the maximum number of vertex buffers or vertex input registers - * that can be expected to work correctly with a DXContext? - * - * The guest is allowed to set up to SVGA3D_DX_MAX_VERTEXBUFFERS, but - * anything in excess of this cap is not guaranteed to render correctly. - * - * Similarly, the guest can set up to SVGA3D_DX_MAX_VERTEXINPUTREGISTERS - * input registers without the SVGA3D_DEVCAP_SM4_1 cap, or - * SVGA3D_DX_SM41_MAX_VERTEXINPUTREGISTERS with the SVGA3D_DEVCAP_SM4_1, - * but only the registers up to this cap value are guaranteed to render - * correctly. - * - * If guest-drivers are able to expose a lower-limit, it's recommended - * that they clamp to this value. Otherwise, the host will make a - * best-effort on case-by-case basis if guests exceed this. - */ - SVGA3D_DEVCAP_DX_MAX_VERTEXBUFFERS = 97, - - /* - * What is the maximum number of constant buffers that can be expected to - * work correctly with a DX context? - * - * The guest is allowed to set up to SVGA3D_DX_MAX_CONSTBUFFERS, but - * anything in excess of this cap is not guaranteed to render correctly. - * - * If guest-drivers are able to expose a lower-limit, it's recommended - * that they clamp to this value. Otherwise, the host will make a - * best-effort on case-by-case basis if guests exceed this. - */ - SVGA3D_DEVCAP_DX_MAX_CONSTANT_BUFFERS = 98, - - /* - * Does the device support provoking vertex control? - * - * If this cap is present, the provokingVertexLast field in the - * rasterizer state is enabled. (Guests can then set it to FALSE, - * meaning that the first vertex is the provoking vertex, or TRUE, - * meaning that the last verteix is the provoking vertex.) - * - * If this cap is FALSE, then guests should set the provokingVertexLast - * to FALSE, otherwise rendering behavior is undefined. - */ - SVGA3D_DEVCAP_DX_PROVOKING_VERTEX = 99, - - SVGA3D_DEVCAP_DXFMT_X8R8G8B8 = 100, - SVGA3D_DEVCAP_DXFMT_A8R8G8B8 = 101, - SVGA3D_DEVCAP_DXFMT_R5G6B5 = 102, - SVGA3D_DEVCAP_DXFMT_X1R5G5B5 = 103, - SVGA3D_DEVCAP_DXFMT_A1R5G5B5 = 104, - SVGA3D_DEVCAP_DXFMT_A4R4G4B4 = 105, - SVGA3D_DEVCAP_DXFMT_Z_D32 = 106, - SVGA3D_DEVCAP_DXFMT_Z_D16 = 107, - SVGA3D_DEVCAP_DXFMT_Z_D24S8 = 108, - SVGA3D_DEVCAP_DXFMT_Z_D15S1 = 109, - SVGA3D_DEVCAP_DXFMT_LUMINANCE8 = 110, - SVGA3D_DEVCAP_DXFMT_LUMINANCE4_ALPHA4 = 111, - SVGA3D_DEVCAP_DXFMT_LUMINANCE16 = 112, - SVGA3D_DEVCAP_DXFMT_LUMINANCE8_ALPHA8 = 113, - SVGA3D_DEVCAP_DXFMT_DXT1 = 114, - SVGA3D_DEVCAP_DXFMT_DXT2 = 115, - SVGA3D_DEVCAP_DXFMT_DXT3 = 116, - SVGA3D_DEVCAP_DXFMT_DXT4 = 117, - SVGA3D_DEVCAP_DXFMT_DXT5 = 118, - SVGA3D_DEVCAP_DXFMT_BUMPU8V8 = 119, - SVGA3D_DEVCAP_DXFMT_BUMPL6V5U5 = 120, - SVGA3D_DEVCAP_DXFMT_BUMPX8L8V8U8 = 121, - SVGA3D_DEVCAP_DXFMT_FORMAT_DEAD1 = 122, - SVGA3D_DEVCAP_DXFMT_ARGB_S10E5 = 123, - SVGA3D_DEVCAP_DXFMT_ARGB_S23E8 = 124, - SVGA3D_DEVCAP_DXFMT_A2R10G10B10 = 125, - SVGA3D_DEVCAP_DXFMT_V8U8 = 126, - SVGA3D_DEVCAP_DXFMT_Q8W8V8U8 = 127, - SVGA3D_DEVCAP_DXFMT_CxV8U8 = 128, - SVGA3D_DEVCAP_DXFMT_X8L8V8U8 = 129, - SVGA3D_DEVCAP_DXFMT_A2W10V10U10 = 130, - SVGA3D_DEVCAP_DXFMT_ALPHA8 = 131, - SVGA3D_DEVCAP_DXFMT_R_S10E5 = 132, - SVGA3D_DEVCAP_DXFMT_R_S23E8 = 133, - SVGA3D_DEVCAP_DXFMT_RG_S10E5 = 134, - SVGA3D_DEVCAP_DXFMT_RG_S23E8 = 135, - SVGA3D_DEVCAP_DXFMT_BUFFER = 136, - SVGA3D_DEVCAP_DXFMT_Z_D24X8 = 137, - SVGA3D_DEVCAP_DXFMT_V16U16 = 138, - SVGA3D_DEVCAP_DXFMT_G16R16 = 139, - SVGA3D_DEVCAP_DXFMT_A16B16G16R16 = 140, - SVGA3D_DEVCAP_DXFMT_UYVY = 141, - SVGA3D_DEVCAP_DXFMT_YUY2 = 142, - SVGA3D_DEVCAP_DXFMT_NV12 = 143, - SVGA3D_DEVCAP_DXFMT_AYUV = 144, - SVGA3D_DEVCAP_DXFMT_R32G32B32A32_TYPELESS = 145, - SVGA3D_DEVCAP_DXFMT_R32G32B32A32_UINT = 146, - SVGA3D_DEVCAP_DXFMT_R32G32B32A32_SINT = 147, - SVGA3D_DEVCAP_DXFMT_R32G32B32_TYPELESS = 148, - SVGA3D_DEVCAP_DXFMT_R32G32B32_FLOAT = 149, - SVGA3D_DEVCAP_DXFMT_R32G32B32_UINT = 150, - SVGA3D_DEVCAP_DXFMT_R32G32B32_SINT = 151, - SVGA3D_DEVCAP_DXFMT_R16G16B16A16_TYPELESS = 152, - SVGA3D_DEVCAP_DXFMT_R16G16B16A16_UINT = 153, - SVGA3D_DEVCAP_DXFMT_R16G16B16A16_SNORM = 154, - SVGA3D_DEVCAP_DXFMT_R16G16B16A16_SINT = 155, - SVGA3D_DEVCAP_DXFMT_R32G32_TYPELESS = 156, - SVGA3D_DEVCAP_DXFMT_R32G32_UINT = 157, - SVGA3D_DEVCAP_DXFMT_R32G32_SINT = 158, - SVGA3D_DEVCAP_DXFMT_R32G8X24_TYPELESS = 159, - SVGA3D_DEVCAP_DXFMT_D32_FLOAT_S8X24_UINT = 160, - SVGA3D_DEVCAP_DXFMT_R32_FLOAT_X8X24 = 161, - SVGA3D_DEVCAP_DXFMT_X32_G8X24_UINT = 162, - SVGA3D_DEVCAP_DXFMT_R10G10B10A2_TYPELESS = 163, - SVGA3D_DEVCAP_DXFMT_R10G10B10A2_UINT = 164, - SVGA3D_DEVCAP_DXFMT_R11G11B10_FLOAT = 165, - SVGA3D_DEVCAP_DXFMT_R8G8B8A8_TYPELESS = 166, - SVGA3D_DEVCAP_DXFMT_R8G8B8A8_UNORM = 167, - SVGA3D_DEVCAP_DXFMT_R8G8B8A8_UNORM_SRGB = 168, - SVGA3D_DEVCAP_DXFMT_R8G8B8A8_UINT = 169, - SVGA3D_DEVCAP_DXFMT_R8G8B8A8_SINT = 170, - SVGA3D_DEVCAP_DXFMT_R16G16_TYPELESS = 171, - SVGA3D_DEVCAP_DXFMT_R16G16_UINT = 172, - SVGA3D_DEVCAP_DXFMT_R16G16_SINT = 173, - SVGA3D_DEVCAP_DXFMT_R32_TYPELESS = 174, - SVGA3D_DEVCAP_DXFMT_D32_FLOAT = 175, - SVGA3D_DEVCAP_DXFMT_R32_UINT = 176, - SVGA3D_DEVCAP_DXFMT_R32_SINT = 177, - SVGA3D_DEVCAP_DXFMT_R24G8_TYPELESS = 178, - SVGA3D_DEVCAP_DXFMT_D24_UNORM_S8_UINT = 179, - SVGA3D_DEVCAP_DXFMT_R24_UNORM_X8 = 180, - SVGA3D_DEVCAP_DXFMT_X24_G8_UINT = 181, - SVGA3D_DEVCAP_DXFMT_R8G8_TYPELESS = 182, - SVGA3D_DEVCAP_DXFMT_R8G8_UNORM = 183, - SVGA3D_DEVCAP_DXFMT_R8G8_UINT = 184, - SVGA3D_DEVCAP_DXFMT_R8G8_SINT = 185, - SVGA3D_DEVCAP_DXFMT_R16_TYPELESS = 186, - SVGA3D_DEVCAP_DXFMT_R16_UNORM = 187, - SVGA3D_DEVCAP_DXFMT_R16_UINT = 188, - SVGA3D_DEVCAP_DXFMT_R16_SNORM = 189, - SVGA3D_DEVCAP_DXFMT_R16_SINT = 190, - SVGA3D_DEVCAP_DXFMT_R8_TYPELESS = 191, - SVGA3D_DEVCAP_DXFMT_R8_UNORM = 192, - SVGA3D_DEVCAP_DXFMT_R8_UINT = 193, - SVGA3D_DEVCAP_DXFMT_R8_SNORM = 194, - SVGA3D_DEVCAP_DXFMT_R8_SINT = 195, - SVGA3D_DEVCAP_DXFMT_P8 = 196, - SVGA3D_DEVCAP_DXFMT_R9G9B9E5_SHAREDEXP = 197, - SVGA3D_DEVCAP_DXFMT_R8G8_B8G8_UNORM = 198, - SVGA3D_DEVCAP_DXFMT_G8R8_G8B8_UNORM = 199, - SVGA3D_DEVCAP_DXFMT_BC1_TYPELESS = 200, - SVGA3D_DEVCAP_DXFMT_BC1_UNORM_SRGB = 201, - SVGA3D_DEVCAP_DXFMT_BC2_TYPELESS = 202, - SVGA3D_DEVCAP_DXFMT_BC2_UNORM_SRGB = 203, - SVGA3D_DEVCAP_DXFMT_BC3_TYPELESS = 204, - SVGA3D_DEVCAP_DXFMT_BC3_UNORM_SRGB = 205, - SVGA3D_DEVCAP_DXFMT_BC4_TYPELESS = 206, - SVGA3D_DEVCAP_DXFMT_ATI1 = 207, - SVGA3D_DEVCAP_DXFMT_BC4_SNORM = 208, - SVGA3D_DEVCAP_DXFMT_BC5_TYPELESS = 209, - SVGA3D_DEVCAP_DXFMT_ATI2 = 210, - SVGA3D_DEVCAP_DXFMT_BC5_SNORM = 211, - SVGA3D_DEVCAP_DXFMT_R10G10B10_XR_BIAS_A2_UNORM = 212, - SVGA3D_DEVCAP_DXFMT_B8G8R8A8_TYPELESS = 213, - SVGA3D_DEVCAP_DXFMT_B8G8R8A8_UNORM_SRGB = 214, - SVGA3D_DEVCAP_DXFMT_B8G8R8X8_TYPELESS = 215, - SVGA3D_DEVCAP_DXFMT_B8G8R8X8_UNORM_SRGB = 216, - SVGA3D_DEVCAP_DXFMT_Z_DF16 = 217, - SVGA3D_DEVCAP_DXFMT_Z_DF24 = 218, - SVGA3D_DEVCAP_DXFMT_Z_D24S8_INT = 219, - SVGA3D_DEVCAP_DXFMT_YV12 = 220, - SVGA3D_DEVCAP_DXFMT_R32G32B32A32_FLOAT = 221, - SVGA3D_DEVCAP_DXFMT_R16G16B16A16_FLOAT = 222, - SVGA3D_DEVCAP_DXFMT_R16G16B16A16_UNORM = 223, - SVGA3D_DEVCAP_DXFMT_R32G32_FLOAT = 224, - SVGA3D_DEVCAP_DXFMT_R10G10B10A2_UNORM = 225, - SVGA3D_DEVCAP_DXFMT_R8G8B8A8_SNORM = 226, - SVGA3D_DEVCAP_DXFMT_R16G16_FLOAT = 227, - SVGA3D_DEVCAP_DXFMT_R16G16_UNORM = 228, - SVGA3D_DEVCAP_DXFMT_R16G16_SNORM = 229, - SVGA3D_DEVCAP_DXFMT_R32_FLOAT = 230, - SVGA3D_DEVCAP_DXFMT_R8G8_SNORM = 231, - SVGA3D_DEVCAP_DXFMT_R16_FLOAT = 232, - SVGA3D_DEVCAP_DXFMT_D16_UNORM = 233, - SVGA3D_DEVCAP_DXFMT_A8_UNORM = 234, - SVGA3D_DEVCAP_DXFMT_BC1_UNORM = 235, - SVGA3D_DEVCAP_DXFMT_BC2_UNORM = 236, - SVGA3D_DEVCAP_DXFMT_BC3_UNORM = 237, - SVGA3D_DEVCAP_DXFMT_B5G6R5_UNORM = 238, - SVGA3D_DEVCAP_DXFMT_B5G5R5A1_UNORM = 239, - SVGA3D_DEVCAP_DXFMT_B8G8R8A8_UNORM = 240, - SVGA3D_DEVCAP_DXFMT_B8G8R8X8_UNORM = 241, - SVGA3D_DEVCAP_DXFMT_BC4_UNORM = 242, - SVGA3D_DEVCAP_DXFMT_BC5_UNORM = 243, - - /* - * Advertises shaderModel 4.1 support, independent blend-states, - * cube-map arrays, and a higher vertex input registers limit. - * - * (See documentation on SVGA3D_DEVCAP_DX_MAX_VERTEXBUFFERS.) - */ - SVGA3D_DEVCAP_SM41 = 244, - - SVGA3D_DEVCAP_MULTISAMPLE_2X = 245, - SVGA3D_DEVCAP_MULTISAMPLE_4X = 246, - - SVGA3D_DEVCAP_MAX /* This must be the last index. */ -} SVGA3dDevCapIndex; +typedef uint32 SVGA3dDevCapIndex; + +#define SVGA3D_DEVCAP_INVALID ((uint32)-1) +#define SVGA3D_DEVCAP_3D 0 +#define SVGA3D_DEVCAP_MAX_LIGHTS 1 + +/* + * SVGA3D_DEVCAP_MAX_TEXTURES reflects the maximum number of + * fixed-function texture units available. Each of these units + * work in both FFP and Shader modes, and they support texture + * transforms and texture coordinates. The host may have additional + * texture image units that are only usable with shaders. + */ +#define SVGA3D_DEVCAP_MAX_TEXTURES 2 +#define SVGA3D_DEVCAP_MAX_CLIP_PLANES 3 +#define SVGA3D_DEVCAP_VERTEX_SHADER_VERSION 4 +#define SVGA3D_DEVCAP_VERTEX_SHADER 5 +#define SVGA3D_DEVCAP_FRAGMENT_SHADER_VERSION 6 +#define SVGA3D_DEVCAP_FRAGMENT_SHADER 7 +#define SVGA3D_DEVCAP_MAX_RENDER_TARGETS 8 +#define SVGA3D_DEVCAP_S23E8_TEXTURES 9 +#define SVGA3D_DEVCAP_S10E5_TEXTURES 10 +#define SVGA3D_DEVCAP_MAX_FIXED_VERTEXBLEND 11 +#define SVGA3D_DEVCAP_D16_BUFFER_FORMAT 12 +#define SVGA3D_DEVCAP_D24S8_BUFFER_FORMAT 13 +#define SVGA3D_DEVCAP_D24X8_BUFFER_FORMAT 14 +#define SVGA3D_DEVCAP_QUERY_TYPES 15 +#define SVGA3D_DEVCAP_TEXTURE_GRADIENT_SAMPLING 16 +#define SVGA3D_DEVCAP_MAX_POINT_SIZE 17 +#define SVGA3D_DEVCAP_MAX_SHADER_TEXTURES 18 +#define SVGA3D_DEVCAP_MAX_TEXTURE_WIDTH 19 +#define SVGA3D_DEVCAP_MAX_TEXTURE_HEIGHT 20 +#define SVGA3D_DEVCAP_MAX_VOLUME_EXTENT 21 +#define SVGA3D_DEVCAP_MAX_TEXTURE_REPEAT 22 +#define SVGA3D_DEVCAP_MAX_TEXTURE_ASPECT_RATIO 23 +#define SVGA3D_DEVCAP_MAX_TEXTURE_ANISOTROPY 24 +#define SVGA3D_DEVCAP_MAX_PRIMITIVE_COUNT 25 +#define SVGA3D_DEVCAP_MAX_VERTEX_INDEX 26 +#define SVGA3D_DEVCAP_MAX_VERTEX_SHADER_INSTRUCTIONS 27 +#define SVGA3D_DEVCAP_MAX_FRAGMENT_SHADER_INSTRUCTIONS 28 +#define SVGA3D_DEVCAP_MAX_VERTEX_SHADER_TEMPS 29 +#define SVGA3D_DEVCAP_MAX_FRAGMENT_SHADER_TEMPS 30 +#define SVGA3D_DEVCAP_TEXTURE_OPS 31 +#define SVGA3D_DEVCAP_SURFACEFMT_X8R8G8B8 32 +#define SVGA3D_DEVCAP_SURFACEFMT_A8R8G8B8 33 +#define SVGA3D_DEVCAP_SURFACEFMT_A2R10G10B10 34 +#define SVGA3D_DEVCAP_SURFACEFMT_X1R5G5B5 35 +#define SVGA3D_DEVCAP_SURFACEFMT_A1R5G5B5 36 +#define SVGA3D_DEVCAP_SURFACEFMT_A4R4G4B4 37 +#define SVGA3D_DEVCAP_SURFACEFMT_R5G6B5 38 +#define SVGA3D_DEVCAP_SURFACEFMT_LUMINANCE16 39 +#define SVGA3D_DEVCAP_SURFACEFMT_LUMINANCE8_ALPHA8 40 +#define SVGA3D_DEVCAP_SURFACEFMT_ALPHA8 41 +#define SVGA3D_DEVCAP_SURFACEFMT_LUMINANCE8 42 +#define SVGA3D_DEVCAP_SURFACEFMT_Z_D16 43 +#define SVGA3D_DEVCAP_SURFACEFMT_Z_D24S8 44 +#define SVGA3D_DEVCAP_SURFACEFMT_Z_D24X8 45 +#define SVGA3D_DEVCAP_SURFACEFMT_DXT1 46 +#define SVGA3D_DEVCAP_SURFACEFMT_DXT2 47 +#define SVGA3D_DEVCAP_SURFACEFMT_DXT3 48 +#define SVGA3D_DEVCAP_SURFACEFMT_DXT4 49 +#define SVGA3D_DEVCAP_SURFACEFMT_DXT5 50 +#define SVGA3D_DEVCAP_SURFACEFMT_BUMPX8L8V8U8 51 +#define SVGA3D_DEVCAP_SURFACEFMT_A2W10V10U10 52 +#define SVGA3D_DEVCAP_SURFACEFMT_BUMPU8V8 53 +#define SVGA3D_DEVCAP_SURFACEFMT_Q8W8V8U8 54 +#define SVGA3D_DEVCAP_SURFACEFMT_CxV8U8 55 +#define SVGA3D_DEVCAP_SURFACEFMT_R_S10E5 56 +#define SVGA3D_DEVCAP_SURFACEFMT_R_S23E8 57 +#define SVGA3D_DEVCAP_SURFACEFMT_RG_S10E5 58 +#define SVGA3D_DEVCAP_SURFACEFMT_RG_S23E8 59 +#define SVGA3D_DEVCAP_SURFACEFMT_ARGB_S10E5 60 +#define SVGA3D_DEVCAP_SURFACEFMT_ARGB_S23E8 61 + +/* + * There is a hole in our devcap definitions for + * historical reasons. + * + * Define a constant just for completeness. + */ +#define SVGA3D_DEVCAP_MISSING62 62 + +#define SVGA3D_DEVCAP_MAX_VERTEX_SHADER_TEXTURES 63 + +/* + * Note that MAX_SIMULTANEOUS_RENDER_TARGETS is a maximum count of color + * render targets. This does not include the depth or stencil targets. + */ +#define SVGA3D_DEVCAP_MAX_SIMULTANEOUS_RENDER_TARGETS 64 + +#define SVGA3D_DEVCAP_SURFACEFMT_V16U16 65 +#define SVGA3D_DEVCAP_SURFACEFMT_G16R16 66 +#define SVGA3D_DEVCAP_SURFACEFMT_A16B16G16R16 67 +#define SVGA3D_DEVCAP_SURFACEFMT_UYVY 68 +#define SVGA3D_DEVCAP_SURFACEFMT_YUY2 69 + +/* + * Deprecated. + */ +#define SVGA3D_DEVCAP_DEAD4 70 +#define SVGA3D_DEVCAP_DEAD5 71 +#define SVGA3D_DEVCAP_DEAD7 72 +#define SVGA3D_DEVCAP_DEAD6 73 + +#define SVGA3D_DEVCAP_AUTOGENMIPMAPS 74 +#define SVGA3D_DEVCAP_SURFACEFMT_NV12 75 +#define SVGA3D_DEVCAP_DEAD10 76 + +/* + * This is the maximum number of SVGA context IDs that the guest + * can define using SVGA_3D_CMD_CONTEXT_DEFINE. + */ +#define SVGA3D_DEVCAP_MAX_CONTEXT_IDS 77 + +/* + * This is the maximum number of SVGA surface IDs that the guest + * can define using SVGA_3D_CMD_SURFACE_DEFINE*. + */ +#define SVGA3D_DEVCAP_MAX_SURFACE_IDS 78 + +#define SVGA3D_DEVCAP_SURFACEFMT_Z_DF16 79 +#define SVGA3D_DEVCAP_SURFACEFMT_Z_DF24 80 +#define SVGA3D_DEVCAP_SURFACEFMT_Z_D24S8_INT 81 + +#define SVGA3D_DEVCAP_SURFACEFMT_ATI1 82 +#define SVGA3D_DEVCAP_SURFACEFMT_ATI2 83 + +/* + * Deprecated. + */ +#define SVGA3D_DEVCAP_DEAD1 84 +#define SVGA3D_DEVCAP_DEAD8 85 +#define SVGA3D_DEVCAP_DEAD9 86 + +#define SVGA3D_DEVCAP_LINE_AA 87 /* boolean */ +#define SVGA3D_DEVCAP_LINE_STIPPLE 88 /* boolean */ +#define SVGA3D_DEVCAP_MAX_LINE_WIDTH 89 /* float */ +#define SVGA3D_DEVCAP_MAX_AA_LINE_WIDTH 90 /* float */ + +#define SVGA3D_DEVCAP_SURFACEFMT_YV12 91 + +/* + * Deprecated. + */ +#define SVGA3D_DEVCAP_DEAD3 92 + +/* + * Are TS_CONSTANT, TS_COLOR_KEY, and TS_COLOR_KEY_ENABLE supported? + */ +#define SVGA3D_DEVCAP_TS_COLOR_KEY 93 /* boolean */ + +/* + * Deprecated. + */ +#define SVGA3D_DEVCAP_DEAD2 94 + +/* + * Does the device support DXContexts? + */ +#define SVGA3D_DEVCAP_DXCONTEXT 95 + +/* + * Deprecated. + */ +#define SVGA3D_DEVCAP_DEAD11 96 + +/* + * What is the maximum number of vertex buffers or vertex input registers + * that can be expected to work correctly with a DXContext? + * + * The guest is allowed to set up to SVGA3D_DX_MAX_VERTEXBUFFERS, but + * anything in excess of this cap is not guaranteed to render correctly. + * + * Similarly, the guest can set up to SVGA3D_DX_MAX_VERTEXINPUTREGISTERS + * input registers without the SVGA3D_DEVCAP_SM4_1 cap, or + * SVGA3D_DX_SM41_MAX_VERTEXINPUTREGISTERS with the SVGA3D_DEVCAP_SM4_1, + * but only the registers up to this cap value are guaranteed to render + * correctly. + * + * If guest-drivers are able to expose a lower-limit, it's recommended + * that they clamp to this value. Otherwise, the host will make a + * best-effort on case-by-case basis if guests exceed this. + */ +#define SVGA3D_DEVCAP_DX_MAX_VERTEXBUFFERS 97 + +/* + * What is the maximum number of constant buffers that can be expected to + * work correctly with a DX context? + * + * The guest is allowed to set up to SVGA3D_DX_MAX_CONSTBUFFERS, but + * anything in excess of this cap is not guaranteed to render correctly. + * + * If guest-drivers are able to expose a lower-limit, it's recommended + * that they clamp to this value. Otherwise, the host will make a + * best-effort on case-by-case basis if guests exceed this. + */ +#define SVGA3D_DEVCAP_DX_MAX_CONSTANT_BUFFERS 98 + +/* + * Does the device support provoking vertex control? + * + * If this cap is present, the provokingVertexLast field in the + * rasterizer state is enabled. (Guests can then set it to FALSE, + * meaning that the first vertex is the provoking vertex, or TRUE, + * meaning that the last verteix is the provoking vertex.) + * + * If this cap is FALSE, then guests should set the provokingVertexLast + * to FALSE, otherwise rendering behavior is undefined. + */ +#define SVGA3D_DEVCAP_DX_PROVOKING_VERTEX 99 + +#define SVGA3D_DEVCAP_DXFMT_X8R8G8B8 100 +#define SVGA3D_DEVCAP_DXFMT_A8R8G8B8 101 +#define SVGA3D_DEVCAP_DXFMT_R5G6B5 102 +#define SVGA3D_DEVCAP_DXFMT_X1R5G5B5 103 +#define SVGA3D_DEVCAP_DXFMT_A1R5G5B5 104 +#define SVGA3D_DEVCAP_DXFMT_A4R4G4B4 105 +#define SVGA3D_DEVCAP_DXFMT_Z_D32 106 +#define SVGA3D_DEVCAP_DXFMT_Z_D16 107 +#define SVGA3D_DEVCAP_DXFMT_Z_D24S8 108 +#define SVGA3D_DEVCAP_DXFMT_Z_D15S1 109 +#define SVGA3D_DEVCAP_DXFMT_LUMINANCE8 110 +#define SVGA3D_DEVCAP_DXFMT_LUMINANCE4_ALPHA4 111 +#define SVGA3D_DEVCAP_DXFMT_LUMINANCE16 112 +#define SVGA3D_DEVCAP_DXFMT_LUMINANCE8_ALPHA8 113 +#define SVGA3D_DEVCAP_DXFMT_DXT1 114 +#define SVGA3D_DEVCAP_DXFMT_DXT2 115 +#define SVGA3D_DEVCAP_DXFMT_DXT3 116 +#define SVGA3D_DEVCAP_DXFMT_DXT4 117 +#define SVGA3D_DEVCAP_DXFMT_DXT5 118 +#define SVGA3D_DEVCAP_DXFMT_BUMPU8V8 119 +#define SVGA3D_DEVCAP_DXFMT_BUMPL6V5U5 120 +#define SVGA3D_DEVCAP_DXFMT_BUMPX8L8V8U8 121 +#define SVGA3D_DEVCAP_DXFMT_FORMAT_DEAD1 122 +#define SVGA3D_DEVCAP_DXFMT_ARGB_S10E5 123 +#define SVGA3D_DEVCAP_DXFMT_ARGB_S23E8 124 +#define SVGA3D_DEVCAP_DXFMT_A2R10G10B10 125 +#define SVGA3D_DEVCAP_DXFMT_V8U8 126 +#define SVGA3D_DEVCAP_DXFMT_Q8W8V8U8 127 +#define SVGA3D_DEVCAP_DXFMT_CxV8U8 128 +#define SVGA3D_DEVCAP_DXFMT_X8L8V8U8 129 +#define SVGA3D_DEVCAP_DXFMT_A2W10V10U10 130 +#define SVGA3D_DEVCAP_DXFMT_ALPHA8 131 +#define SVGA3D_DEVCAP_DXFMT_R_S10E5 132 +#define SVGA3D_DEVCAP_DXFMT_R_S23E8 133 +#define SVGA3D_DEVCAP_DXFMT_RG_S10E5 134 +#define SVGA3D_DEVCAP_DXFMT_RG_S23E8 135 +#define SVGA3D_DEVCAP_DXFMT_BUFFER 136 +#define SVGA3D_DEVCAP_DXFMT_Z_D24X8 137 +#define SVGA3D_DEVCAP_DXFMT_V16U16 138 +#define SVGA3D_DEVCAP_DXFMT_G16R16 139 +#define SVGA3D_DEVCAP_DXFMT_A16B16G16R16 140 +#define SVGA3D_DEVCAP_DXFMT_UYVY 141 +#define SVGA3D_DEVCAP_DXFMT_YUY2 142 +#define SVGA3D_DEVCAP_DXFMT_NV12 143 +#define SVGA3D_DEVCAP_DXFMT_FORMAT_DEAD2 144 +#define SVGA3D_DEVCAP_DXFMT_R32G32B32A32_TYPELESS 145 +#define SVGA3D_DEVCAP_DXFMT_R32G32B32A32_UINT 146 +#define SVGA3D_DEVCAP_DXFMT_R32G32B32A32_SINT 147 +#define SVGA3D_DEVCAP_DXFMT_R32G32B32_TYPELESS 148 +#define SVGA3D_DEVCAP_DXFMT_R32G32B32_FLOAT 149 +#define SVGA3D_DEVCAP_DXFMT_R32G32B32_UINT 150 +#define SVGA3D_DEVCAP_DXFMT_R32G32B32_SINT 151 +#define SVGA3D_DEVCAP_DXFMT_R16G16B16A16_TYPELESS 152 +#define SVGA3D_DEVCAP_DXFMT_R16G16B16A16_UINT 153 +#define SVGA3D_DEVCAP_DXFMT_R16G16B16A16_SNORM 154 +#define SVGA3D_DEVCAP_DXFMT_R16G16B16A16_SINT 155 +#define SVGA3D_DEVCAP_DXFMT_R32G32_TYPELESS 156 +#define SVGA3D_DEVCAP_DXFMT_R32G32_UINT 157 +#define SVGA3D_DEVCAP_DXFMT_R32G32_SINT 158 +#define SVGA3D_DEVCAP_DXFMT_R32G8X24_TYPELESS 159 +#define SVGA3D_DEVCAP_DXFMT_D32_FLOAT_S8X24_UINT 160 +#define SVGA3D_DEVCAP_DXFMT_R32_FLOAT_X8X24 161 +#define SVGA3D_DEVCAP_DXFMT_X32_G8X24_UINT 162 +#define SVGA3D_DEVCAP_DXFMT_R10G10B10A2_TYPELESS 163 +#define SVGA3D_DEVCAP_DXFMT_R10G10B10A2_UINT 164 +#define SVGA3D_DEVCAP_DXFMT_R11G11B10_FLOAT 165 +#define SVGA3D_DEVCAP_DXFMT_R8G8B8A8_TYPELESS 166 +#define SVGA3D_DEVCAP_DXFMT_R8G8B8A8_UNORM 167 +#define SVGA3D_DEVCAP_DXFMT_R8G8B8A8_UNORM_SRGB 168 +#define SVGA3D_DEVCAP_DXFMT_R8G8B8A8_UINT 169 +#define SVGA3D_DEVCAP_DXFMT_R8G8B8A8_SINT 170 +#define SVGA3D_DEVCAP_DXFMT_R16G16_TYPELESS 171 +#define SVGA3D_DEVCAP_DXFMT_R16G16_UINT 172 +#define SVGA3D_DEVCAP_DXFMT_R16G16_SINT 173 +#define SVGA3D_DEVCAP_DXFMT_R32_TYPELESS 174 +#define SVGA3D_DEVCAP_DXFMT_D32_FLOAT 175 +#define SVGA3D_DEVCAP_DXFMT_R32_UINT 176 +#define SVGA3D_DEVCAP_DXFMT_R32_SINT 177 +#define SVGA3D_DEVCAP_DXFMT_R24G8_TYPELESS 178 +#define SVGA3D_DEVCAP_DXFMT_D24_UNORM_S8_UINT 179 +#define SVGA3D_DEVCAP_DXFMT_R24_UNORM_X8 180 +#define SVGA3D_DEVCAP_DXFMT_X24_G8_UINT 181 +#define SVGA3D_DEVCAP_DXFMT_R8G8_TYPELESS 182 +#define SVGA3D_DEVCAP_DXFMT_R8G8_UNORM 183 +#define SVGA3D_DEVCAP_DXFMT_R8G8_UINT 184 +#define SVGA3D_DEVCAP_DXFMT_R8G8_SINT 185 +#define SVGA3D_DEVCAP_DXFMT_R16_TYPELESS 186 +#define SVGA3D_DEVCAP_DXFMT_R16_UNORM 187 +#define SVGA3D_DEVCAP_DXFMT_R16_UINT 188 +#define SVGA3D_DEVCAP_DXFMT_R16_SNORM 189 +#define SVGA3D_DEVCAP_DXFMT_R16_SINT 190 +#define SVGA3D_DEVCAP_DXFMT_R8_TYPELESS 191 +#define SVGA3D_DEVCAP_DXFMT_R8_UNORM 192 +#define SVGA3D_DEVCAP_DXFMT_R8_UINT 193 +#define SVGA3D_DEVCAP_DXFMT_R8_SNORM 194 +#define SVGA3D_DEVCAP_DXFMT_R8_SINT 195 +#define SVGA3D_DEVCAP_DXFMT_P8 196 +#define SVGA3D_DEVCAP_DXFMT_R9G9B9E5_SHAREDEXP 197 +#define SVGA3D_DEVCAP_DXFMT_R8G8_B8G8_UNORM 198 +#define SVGA3D_DEVCAP_DXFMT_G8R8_G8B8_UNORM 199 +#define SVGA3D_DEVCAP_DXFMT_BC1_TYPELESS 200 +#define SVGA3D_DEVCAP_DXFMT_BC1_UNORM_SRGB 201 +#define SVGA3D_DEVCAP_DXFMT_BC2_TYPELESS 202 +#define SVGA3D_DEVCAP_DXFMT_BC2_UNORM_SRGB 203 +#define SVGA3D_DEVCAP_DXFMT_BC3_TYPELESS 204 +#define SVGA3D_DEVCAP_DXFMT_BC3_UNORM_SRGB 205 +#define SVGA3D_DEVCAP_DXFMT_BC4_TYPELESS 206 +#define SVGA3D_DEVCAP_DXFMT_ATI1 207 +#define SVGA3D_DEVCAP_DXFMT_BC4_SNORM 208 +#define SVGA3D_DEVCAP_DXFMT_BC5_TYPELESS 209 +#define SVGA3D_DEVCAP_DXFMT_ATI2 210 +#define SVGA3D_DEVCAP_DXFMT_BC5_SNORM 211 +#define SVGA3D_DEVCAP_DXFMT_R10G10B10_XR_BIAS_A2_UNORM 212 +#define SVGA3D_DEVCAP_DXFMT_B8G8R8A8_TYPELESS 213 +#define SVGA3D_DEVCAP_DXFMT_B8G8R8A8_UNORM_SRGB 214 +#define SVGA3D_DEVCAP_DXFMT_B8G8R8X8_TYPELESS 215 +#define SVGA3D_DEVCAP_DXFMT_B8G8R8X8_UNORM_SRGB 216 +#define SVGA3D_DEVCAP_DXFMT_Z_DF16 217 +#define SVGA3D_DEVCAP_DXFMT_Z_DF24 218 +#define SVGA3D_DEVCAP_DXFMT_Z_D24S8_INT 219 +#define SVGA3D_DEVCAP_DXFMT_YV12 220 +#define SVGA3D_DEVCAP_DXFMT_R32G32B32A32_FLOAT 221 +#define SVGA3D_DEVCAP_DXFMT_R16G16B16A16_FLOAT 222 +#define SVGA3D_DEVCAP_DXFMT_R16G16B16A16_UNORM 223 +#define SVGA3D_DEVCAP_DXFMT_R32G32_FLOAT 224 +#define SVGA3D_DEVCAP_DXFMT_R10G10B10A2_UNORM 225 +#define SVGA3D_DEVCAP_DXFMT_R8G8B8A8_SNORM 226 +#define SVGA3D_DEVCAP_DXFMT_R16G16_FLOAT 227 +#define SVGA3D_DEVCAP_DXFMT_R16G16_UNORM 228 +#define SVGA3D_DEVCAP_DXFMT_R16G16_SNORM 229 +#define SVGA3D_DEVCAP_DXFMT_R32_FLOAT 230 +#define SVGA3D_DEVCAP_DXFMT_R8G8_SNORM 231 +#define SVGA3D_DEVCAP_DXFMT_R16_FLOAT 232 +#define SVGA3D_DEVCAP_DXFMT_D16_UNORM 233 +#define SVGA3D_DEVCAP_DXFMT_A8_UNORM 234 +#define SVGA3D_DEVCAP_DXFMT_BC1_UNORM 235 +#define SVGA3D_DEVCAP_DXFMT_BC2_UNORM 236 +#define SVGA3D_DEVCAP_DXFMT_BC3_UNORM 237 +#define SVGA3D_DEVCAP_DXFMT_B5G6R5_UNORM 238 +#define SVGA3D_DEVCAP_DXFMT_B5G5R5A1_UNORM 239 +#define SVGA3D_DEVCAP_DXFMT_B8G8R8A8_UNORM 240 +#define SVGA3D_DEVCAP_DXFMT_B8G8R8X8_UNORM 241 +#define SVGA3D_DEVCAP_DXFMT_BC4_UNORM 242 +#define SVGA3D_DEVCAP_DXFMT_BC5_UNORM 243 + +/* + * Advertises shaderModel 4.1 support, independent blend-states, + * cube-map arrays, and a higher vertex input registers limit. + * + * (See documentation on SVGA3D_DEVCAP_DX_MAX_VERTEXBUFFERS.) + */ +#define SVGA3D_DEVCAP_SM41 244 +#define SVGA3D_DEVCAP_MULTISAMPLE_2X 245 +#define SVGA3D_DEVCAP_MULTISAMPLE_4X 246 + +/* + * Indicates that the device has rendering support for + * the full multisample quality. If this cap is not present, + * the host may or may not support full quality rendering. + * + * See also SVGA_REG_MS_HINT_RESOLVED. + */ +#define SVGA3D_DEVCAP_MS_FULL_QUALITY 247 + +/* + * Advertises support for the SVGA3D LogicOps commands. + */ +#define SVGA3D_DEVCAP_LOGICOPS 248 + +/* + * Advertises support for using logicOps in the DXBlendStates. + */ +#define SVGA3D_DEVCAP_LOGIC_BLENDOPS 249 + +/* +* Note DXFMT range is now non-contiguous. +*/ +#define SVGA3D_DEVCAP_RESERVED_1 250 +#define SVGA3D_DEVCAP_DXFMT_BC6H_TYPELESS 251 +#define SVGA3D_DEVCAP_DXFMT_BC6H_UF16 252 +#define SVGA3D_DEVCAP_DXFMT_BC6H_SF16 253 +#define SVGA3D_DEVCAP_DXFMT_BC7_TYPELESS 254 +#define SVGA3D_DEVCAP_DXFMT_BC7_UNORM 255 +#define SVGA3D_DEVCAP_DXFMT_BC7_UNORM_SRGB 256 +#define SVGA3D_DEVCAP_RESERVED_2 257 + +#define SVGA3D_DEVCAP_SM5 258 +#define SVGA3D_DEVCAP_MULTISAMPLE_8X 259 + +/* This must be the last index. */ +#define SVGA3D_DEVCAP_MAX 260 /* * Bit definitions for DXFMT devcaps @@ -472,10 +501,10 @@ typedef enum { #define SVGA3D_DXFMT_MAX (1 << 10) typedef union { - Bool b; + SVGA3dBool b; uint32 u; - int32 i; - float f; + int32 i; + float f; } SVGA3dDevCapResult; #endif /* _SVGA3D_DEVCAPS_H_ */ diff --git a/drivers/gpu/drm/vmwgfx/device_include/svga3d_dx.h b/drivers/gpu/drm/vmwgfx/device_include/svga3d_dx.h index 7a49c94df221..f703ac2b1768 100644 --- a/drivers/gpu/drm/vmwgfx/device_include/svga3d_dx.h +++ b/drivers/gpu/drm/vmwgfx/device_include/svga3d_dx.h @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: GPL-2.0 OR MIT */ /********************************************************** - * Copyright 2012-2015 VMware, Inc. + * Copyright 2012-2019 VMware, Inc. * * Permission is hereby granted, free of charge, to any person * obtaining a copy of this software and associated documentation @@ -118,12 +118,14 @@ typedef uint8 SVGA3dMultisampleRastEnable; #define SVGA3D_DX_MAX_SRVIEWS 128 #define SVGA3D_DX_MAX_CONSTBUFFERS 16 #define SVGA3D_DX_MAX_SAMPLERS 16 +#define SVGA3D_DX_MAX_CLASS_INSTANCES 253 #define SVGA3D_DX_MAX_CONSTBUF_BINDING_SIZE (4096 * 4 * (uint32)sizeof(uint32)) typedef uint32 SVGA3dShaderResourceViewId; typedef uint32 SVGA3dRenderTargetViewId; typedef uint32 SVGA3dDepthStencilViewId; +typedef uint32 SVGA3dUAViewId; typedef uint32 SVGA3dShaderId; typedef uint32 SVGA3dElementLayoutId; @@ -145,6 +147,17 @@ typedef union { float value[4]; } SVGA3dRGBAFloat; +typedef union { + struct { + uint32 r; + uint32 g; + uint32 b; + uint32 a; + }; + + uint32 value[4]; +} SVGA3dRGBAUint32; + typedef #include "vmware_pack_begin.h" struct { @@ -249,6 +262,39 @@ struct SVGA3dCmdDXSetShader { #include "vmware_pack_end.h" SVGA3dCmdDXSetShader; /* SVGA_3D_CMD_DX_SET_SHADER */ +typedef union { + struct { + uint32 cbOffset : 12; + uint32 cbId : 4; + uint32 baseSamp : 4; + uint32 baseTex : 7; + uint32 reserved : 5; + }; + uint32 value; +} SVGA3dIfaceData; + +typedef +#include "vmware_pack_begin.h" +struct SVGA3dCmdDXSetShaderIface { + SVGA3dShaderType type; + uint32 numClassInstances; + uint32 index; + uint32 iface; + SVGA3dIfaceData data; +} +#include "vmware_pack_end.h" +SVGA3dCmdDXSetShaderIface; /* SVGA_3D_CMD_DX_SET_SHADER_IFACE */ + +typedef +#include "vmware_pack_begin.h" +struct SVGA3dCmdDXBindShaderIface { + uint32 cid; + SVGAMobId mobid; + uint32 offsetInBytes; +} +#include "vmware_pack_end.h" +SVGA3dCmdDXBindShaderIface; /* SVGA_3D_CMD_DX_BIND_SHADER_IFACE */ + typedef #include "vmware_pack_begin.h" struct SVGA3dCmdDXSetSamplers { @@ -306,6 +352,26 @@ SVGA3dCmdDXDrawIndexedInstanced; /* SVGA_3D_CMD_DX_DRAW_INDEXED_INSTANCED */ typedef #include "vmware_pack_begin.h" +struct SVGA3dCmdDXDrawIndexedInstancedIndirect { + SVGA3dSurfaceId argsBufferSid; + uint32 byteOffsetForArgs; +} +#include "vmware_pack_end.h" +SVGA3dCmdDXDrawIndexedInstancedIndirect; +/* SVGA_3D_CMD_DX_DRAW_INDEXED_INSTANCED_INDIRECT */ + +typedef +#include "vmware_pack_begin.h" +struct SVGA3dCmdDXDrawInstancedIndirect { + SVGA3dSurfaceId argsBufferSid; + uint32 byteOffsetForArgs; +} +#include "vmware_pack_end.h" +SVGA3dCmdDXDrawInstancedIndirect; +/* SVGA_3D_CMD_DX_DRAW_INSTANCED_INDIRECT */ + +typedef +#include "vmware_pack_begin.h" struct SVGA3dCmdDXDrawAuto { uint32 pad0; } @@ -314,6 +380,27 @@ SVGA3dCmdDXDrawAuto; /* SVGA_3D_CMD_DX_DRAW_AUTO */ typedef #include "vmware_pack_begin.h" +struct SVGA3dCmdDXDispatch { + uint32 threadGroupCountX; + uint32 threadGroupCountY; + uint32 threadGroupCountZ; +} +#include "vmware_pack_end.h" +SVGA3dCmdDXDispatch; +/* SVGA_3D_CMD_DX_DISPATCH */ + +typedef +#include "vmware_pack_begin.h" +struct SVGA3dCmdDXDispatchIndirect { + SVGA3dSurfaceId argsBufferSid; + uint32 byteOffsetForArgs; +} +#include "vmware_pack_end.h" +SVGA3dCmdDXDispatchIndirect; +/* SVGA_3D_CMD_DX_DISPATCH_INDIRECT */ + +typedef +#include "vmware_pack_begin.h" struct SVGA3dCmdDXSetInputLayout { SVGA3dElementLayoutId elementLayoutId; } @@ -525,7 +612,7 @@ struct MKS3dDXSOState { uint32 offset; /* Starting offset */ uint32 intOffset; /* Internal offset */ uint32 vertexCount; /* vertices written */ - uint32 sizeInBytes; /* max bytes to write */ + uint32 dead; } #include "vmware_pack_end.h" SVGA3dDXSOState; @@ -786,6 +873,31 @@ struct SVGA3dCmdDXTransferFromBuffer { SVGA3dCmdDXTransferFromBuffer; /* SVGA_3D_CMD_DX_TRANSFER_FROM_BUFFER */ +#define SVGA3D_TRANSFER_TO_BUFFER_READBACK (1 << 0) +#define SVGA3D_TRANSFER_TO_BUFFER_FLAGS_MASK (1 << 0) +typedef uint32 SVGA3dTransferToBufferFlags; + +/* + * Raw byte wise transfer to a buffer surface from another surface + * of the requested box. Supported if SVGA_CAP_DX2 is set. This + * command does not take a context. + */ +typedef +#include "vmware_pack_begin.h" +struct SVGA3dCmdDXTransferToBuffer { + SVGA3dSurfaceId srcSid; + uint32 srcSubResource; + SVGA3dBox srcBox; + SVGA3dSurfaceId destSid; + uint32 destOffset; + uint32 destPitch; + uint32 destSlicePitch; + SVGA3dTransferToBufferFlags flags; +} +#include "vmware_pack_end.h" +SVGA3dCmdDXTransferToBuffer; /* SVGA_3D_CMD_DX_TRANSFER_TO_BUFFER */ + + /* * Raw byte wise transfer from a buffer surface into another surface * of the requested box. Supported if SVGA3D_DEVCAP_DXCONTEXT is set. @@ -905,6 +1017,20 @@ typedef SVGA3dCmdDXSetConstantBufferOffset SVGA3dCmdDXSetPSConstantBufferOffset; typedef SVGA3dCmdDXSetConstantBufferOffset SVGA3dCmdDXSetGSConstantBufferOffset; /* SVGA_3D_CMD_DX_SET_GS_CONSTANT_BUFFER_OFFSET */ +typedef SVGA3dCmdDXSetConstantBufferOffset SVGA3dCmdDXSetHSConstantBufferOffset; +/* SVGA_3D_CMD_DX_SET_HS_CONSTANT_BUFFER_OFFSET */ + +typedef SVGA3dCmdDXSetConstantBufferOffset SVGA3dCmdDXSetDSConstantBufferOffset; +/* SVGA_3D_CMD_DX_SET_DS_CONSTANT_BUFFER_OFFSET */ + +typedef SVGA3dCmdDXSetConstantBufferOffset SVGA3dCmdDXSetCSConstantBufferOffset; +/* SVGA_3D_CMD_DX_SET_CS_CONSTANT_BUFFER_OFFSET */ + + +#define SVGA3D_BUFFEREX_SRV_RAW (1 << 0) +#define SVGA3D_BUFFEREX_SRV_FLAGS_MAX (1 << 1) +#define SVGA3D_BUFFEREX_SRV_FLAGS_MASK (SVGA3D_BUFFEREX_SRV_FLAGS_MAX - 1) +typedef uint32 SVGA3dBufferExFlags; typedef #include "vmware_pack_begin.h" @@ -925,7 +1051,7 @@ struct { struct { uint32 firstElement; uint32 numElements; - uint32 flags; + SVGA3dBufferExFlags flags; uint32 pad0; } bufferex; }; @@ -1072,6 +1198,32 @@ struct SVGA3dCmdDXDefineDepthStencilView { SVGA3dCmdDXDefineDepthStencilView; /* SVGA_3D_CMD_DX_DEFINE_DEPTHSTENCIL_VIEW */ +/* + * Version 2 needed in order to start validating and using the flags + * field. Unfortunately the device wasn't validating or using the + * flags field and the driver wasn't initializing it in shipped code, + * so a new version of the command is needed to allow that code to + * continue to work. + */ +typedef +#include "vmware_pack_begin.h" +struct SVGA3dCmdDXDefineDepthStencilView_v2 { + SVGA3dDepthStencilViewId depthStencilViewId; + + SVGA3dSurfaceId sid; + SVGA3dSurfaceFormat format; + SVGA3dResourceType resourceDimension; + uint32 mipSlice; + uint32 firstArraySlice; + uint32 arraySize; + SVGA3DCreateDSViewFlags flags; + uint8 pad0; + uint16 pad1; +} +#include "vmware_pack_end.h" +SVGA3dCmdDXDefineDepthStencilView_v2; +/* SVGA_3D_CMD_DX_DEFINE_DEPTHSTENCIL_VIEW_V2 */ + typedef #include "vmware_pack_begin.h" struct SVGA3dCmdDXDestroyDepthStencilView { @@ -1081,6 +1233,138 @@ struct SVGA3dCmdDXDestroyDepthStencilView { SVGA3dCmdDXDestroyDepthStencilView; /* SVGA_3D_CMD_DX_DESTROY_DEPTHSTENCIL_VIEW */ + +#define SVGA3D_UABUFFER_RAW (1 << 0) +#define SVGA3D_UABUFFER_APPEND (1 << 1) +#define SVGA3D_UABUFFER_COUNTER (1 << 2) +typedef uint32 SVGA3dUABufferFlags; + +typedef +#include "vmware_pack_begin.h" +struct { + union { + struct { + uint32 firstElement; + uint32 numElements; + SVGA3dUABufferFlags flags; + uint32 padding0; + uint32 padding1; + } buffer; + struct { + uint32 mipSlice; + uint32 firstArraySlice; + uint32 arraySize; + uint32 padding0; + uint32 padding1; + } tex; /* 1d, 2d */ + struct { + uint32 mipSlice; + uint32 firstW; + uint32 wSize; + uint32 padding0; + uint32 padding1; + } tex3D; + }; +} +#include "vmware_pack_end.h" +SVGA3dUAViewDesc; + +typedef +#include "vmware_pack_begin.h" +struct { + SVGA3dSurfaceId sid; + SVGA3dSurfaceFormat format; + SVGA3dResourceType resourceDimension; + SVGA3dUAViewDesc desc; + uint32 structureCount; + uint32 pad[7]; +} +#include "vmware_pack_end.h" +SVGACOTableDXUAViewEntry; + +typedef +#include "vmware_pack_begin.h" +struct SVGA3dCmdDXDefineUAView { + SVGA3dUAViewId uaViewId; + + SVGA3dSurfaceId sid; + SVGA3dSurfaceFormat format; + SVGA3dResourceType resourceDimension; + + SVGA3dUAViewDesc desc; +} +#include "vmware_pack_end.h" +SVGA3dCmdDXDefineUAView; +/* SVGA_3D_CMD_DX_DEFINE_UA_VIEW */ + +typedef +#include "vmware_pack_begin.h" +struct SVGA3dCmdDXDestroyUAView { + SVGA3dUAViewId uaViewId; +} +#include "vmware_pack_end.h" +SVGA3dCmdDXDestroyUAView; +/* SVGA_3D_CMD_DX_DESTROY_UA_VIEW */ + +typedef +#include "vmware_pack_begin.h" +struct SVGA3dCmdDXClearUAViewUint { + SVGA3dUAViewId uaViewId; + SVGA3dRGBAUint32 value; +} +#include "vmware_pack_end.h" +SVGA3dCmdDXClearUAViewUint; +/* SVGA_3D_CMD_DX_CLEAR_UA_VIEW_UINT */ + +typedef +#include "vmware_pack_begin.h" +struct SVGA3dCmdDXClearUAViewFloat { + SVGA3dUAViewId uaViewId; + SVGA3dRGBAFloat value; +} +#include "vmware_pack_end.h" +SVGA3dCmdDXClearUAViewFloat; +/* SVGA_3D_CMD_DX_CLEAR_UA_VIEW_FLOAT */ + +typedef +#include "vmware_pack_begin.h" +struct SVGA3dCmdDXCopyStructureCount { + SVGA3dUAViewId srcUAViewId; + SVGA3dSurfaceId destSid; + uint32 destByteOffset; +} +#include "vmware_pack_end.h" +SVGA3dCmdDXCopyStructureCount; +/* SVGA_3D_CMD_DX_COPY_STRUCTURE_COUNT */ + +typedef +#include "vmware_pack_begin.h" +struct SVGA3dCmdDXSetStructureCount { + SVGA3dUAViewId uaViewId; + uint32 structureCount; +} +#include "vmware_pack_end.h" +SVGA3dCmdDXSetStructureCount; +/* SVGA_3D_CMD_DX_SET_STRUCTURE_COUNT */ + +typedef +#include "vmware_pack_begin.h" +struct SVGA3dCmdDXSetUAViews { + uint32 uavSpliceIndex; + /* Followed by a variable number of SVGA3dUAViewId's. */ +} +#include "vmware_pack_end.h" +SVGA3dCmdDXSetUAViews; /* SVGA_3D_CMD_DX_SET_UA_VIEWS */ + +typedef +#include "vmware_pack_begin.h" +struct SVGA3dCmdDXSetCSUAViews { + uint32 startIndex; + /* Followed by a variable number of SVGA3dUAViewId's. */ +} +#include "vmware_pack_end.h" +SVGA3dCmdDXSetCSUAViews; /* SVGA_3D_CMD_DX_SET_CS_UA_VIEWS */ + typedef #include "vmware_pack_begin.h" struct SVGA3dInputElementDesc { @@ -1099,7 +1383,7 @@ typedef struct { uint32 elid; uint32 numDescs; - SVGA3dInputElementDesc desc[32]; + SVGA3dInputElementDesc descs[32]; uint32 pad[62]; } #include "vmware_pack_end.h" @@ -1261,7 +1545,8 @@ struct { uint8 lineStippleEnable; uint8 lineStippleFactor; uint16 lineStipplePattern; - uint32 forcedSampleCount; + uint8 forcedSampleCount; + uint8 mustBeZero[3]; } #include "vmware_pack_end.h" SVGACOTableDXRasterizerStateEntry; @@ -1352,6 +1637,71 @@ struct SVGA3dCmdDXDestroySamplerState { #include "vmware_pack_end.h" SVGA3dCmdDXDestroySamplerState; /* SVGA_3D_CMD_DX_DESTROY_SAMPLER_STATE */ + +#define SVGADX_SIGNATURE_SEMANTIC_NAME_UNDEFINED 0 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_POSITION 1 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_CLIP_DISTANCE 2 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_CULL_DISTANCE 3 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_RENDER_TARGET_ARRAY_INDEX 4 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_VIEWPORT_ARRAY_INDEX 5 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_VERTEX_ID 6 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_PRIMITIVE_ID 7 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_INSTANCE_ID 8 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_IS_FRONT_FACE 9 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_SAMPLE_INDEX 10 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_FINAL_QUAD_U_EQ_0_EDGE_TESSFACTOR 11 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_FINAL_QUAD_V_EQ_0_EDGE_TESSFACTOR 12 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_FINAL_QUAD_U_EQ_1_EDGE_TESSFACTOR 13 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_FINAL_QUAD_V_EQ_1_EDGE_TESSFACTOR 14 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_FINAL_QUAD_U_INSIDE_TESSFACTOR 15 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_FINAL_QUAD_V_INSIDE_TESSFACTOR 16 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_FINAL_TRI_U_EQ_0_EDGE_TESSFACTOR 17 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_FINAL_TRI_V_EQ_0_EDGE_TESSFACTOR 18 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_FINAL_TRI_W_EQ_0_EDGE_TESSFACTOR 19 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_FINAL_TRI_INSIDE_TESSFACTOR 20 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_FINAL_LINE_DETAIL_TESSFACTOR 21 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_FINAL_LINE_DENSITY_TESSFACTOR 22 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_MAX 23 +typedef uint32 SVGA3dDXSignatureSemanticName; + +#define SVGADX_SIGNATURE_REGISTER_COMPONENT_UNKNOWN 0 +typedef uint32 SVGA3dDXSignatureRegisterComponentType; + +#define SVGADX_SIGNATURE_MIN_PRECISION_DEFAULT 0 +typedef uint32 SVGA3dDXSignatureMinPrecision; + +typedef +#include "vmware_pack_begin.h" +struct SVGA3dDXSignatureEntry { + uint32 registerIndex; + SVGA3dDXSignatureSemanticName semanticName; + uint32 mask; /* Lower 4 bits represent X, Y, Z, W channels */ + SVGA3dDXSignatureRegisterComponentType componentType; + SVGA3dDXSignatureMinPrecision minPrecision; +} +#include "vmware_pack_end.h" +SVGA3dDXShaderSignatureEntry; + +#define SVGADX_SIGNATURE_HEADER_VERSION_0 0x08a92d12 + +/* + * The SVGA3dDXSignatureHeader structure is added after the shader + * body in the mob that is bound to the shader. It is followed by the + * specified number of SVGA3dDXSignatureEntry structures for each of + * the three types of signatures in the order (input, output, patch + * constants). + */ +typedef +#include "vmware_pack_begin.h" +struct SVGA3dDXSignatureHeader { + uint32 headerVersion; + uint32 numInputSignatures; + uint32 numOutputSignatures; + uint32 numPatchConstantSignatures; +} +#include "vmware_pack_end.h" +SVGA3dDXShaderSignatureHeader; + typedef #include "vmware_pack_begin.h" struct SVGA3dCmdDXDefineShader { @@ -1415,7 +1765,8 @@ SVGA3dCmdDXCondBindAllShader; /* SVGA_3D_CMD_DX_COND_BIND_ALL_SHADER */ /* * The maximum number of streamout decl's in each streamout entry. */ -#define SVGA3D_MAX_STREAMOUT_DECLS 64 +#define SVGA3D_MAX_DX10_STREAMOUT_DECLS 64 +#define SVGA3D_MAX_STREAMOUT_DECLS 512 typedef #include "vmware_pack_begin.h" @@ -1434,10 +1785,16 @@ typedef #include "vmware_pack_begin.h" struct SVGAOTableStreamOutputEntry { uint32 numOutputStreamEntries; - SVGA3dStreamOutputDeclarationEntry decl[SVGA3D_MAX_STREAMOUT_DECLS]; + SVGA3dStreamOutputDeclarationEntry decl[SVGA3D_MAX_DX10_STREAMOUT_DECLS]; uint32 streamOutputStrideInBytes[SVGA3D_DX_MAX_SOTARGETS]; uint32 rasterizedStream; - uint32 pad[250]; + uint32 numOutputStreamStrides; + uint32 mobid; + uint32 offsetInBytes; + uint8 usesMob; + uint8 pad0; + uint16 pad1; + uint32 pad2[246]; } #include "vmware_pack_end.h" SVGACOTableDXStreamOutputEntry; @@ -1447,13 +1804,47 @@ typedef struct SVGA3dCmdDXDefineStreamOutput { SVGA3dStreamOutputId soid; uint32 numOutputStreamEntries; - SVGA3dStreamOutputDeclarationEntry decl[SVGA3D_MAX_STREAMOUT_DECLS]; + SVGA3dStreamOutputDeclarationEntry decl[SVGA3D_MAX_DX10_STREAMOUT_DECLS]; uint32 streamOutputStrideInBytes[SVGA3D_DX_MAX_SOTARGETS]; uint32 rasterizedStream; } #include "vmware_pack_end.h" SVGA3dCmdDXDefineStreamOutput; /* SVGA_3D_CMD_DX_DEFINE_STREAMOUTPUT */ +/* + * Version 2 needed in order to start validating and using the + * rasterizedStream field. Unfortunately the device wasn't validating + * or using this field and the driver wasn't initializing it in shipped + * code, so a new version of the command is needed to allow that code + * to continue to work. Also added new numOutputStreamStrides field. + */ + +#define SVGA3D_DX_SO_NO_RASTERIZED_STREAM 0xFFFFFFFF + +typedef +#include "vmware_pack_begin.h" +struct SVGA3dCmdDXDefineStreamOutputWithMob { + SVGA3dStreamOutputId soid; + uint32 numOutputStreamEntries; + uint32 numOutputStreamStrides; + uint32 streamOutputStrideInBytes[SVGA3D_DX_MAX_SOTARGETS]; + uint32 rasterizedStream; +} +#include "vmware_pack_end.h" +SVGA3dCmdDXDefineStreamOutputWithMob; +/* SVGA_3D_CMD_DX_DEFINE_STREAMOUTPUT_WITH_MOB */ + +typedef +#include "vmware_pack_begin.h" +struct SVGA3dCmdDXBindStreamOutput { + SVGA3dStreamOutputId soid; + uint32 mobid; + uint32 offsetInBytes; + uint32 sizeInBytes; +} +#include "vmware_pack_end.h" +SVGA3dCmdDXBindStreamOutput; /* SVGA_3D_CMD_DX_BIND_STREAMOUTPUT */ + typedef #include "vmware_pack_begin.h" struct SVGA3dCmdDXDestroyStreamOutput { @@ -1472,6 +1863,15 @@ SVGA3dCmdDXSetStreamOutput; /* SVGA_3D_CMD_DX_SET_STREAMOUTPUT */ typedef #include "vmware_pack_begin.h" +struct SVGA3dCmdDXSetMinLOD { + SVGA3dSurfaceId sid; + float minLOD; +} +#include "vmware_pack_end.h" +SVGA3dCmdDXSetMinLOD; /* SVGA_3D_CMD_DX_SET_MIN_LOD */ + +typedef +#include "vmware_pack_begin.h" struct { uint64 value; uint32 mobId; @@ -1581,33 +1981,38 @@ struct SVGADXContextMobFormat { uint32 rasterizerStateId; uint32 depthStencilViewId; uint32 renderTargetViewIds[SVGA3D_MAX_SIMULTANEOUS_RENDER_TARGETS]; - uint32 unorderedAccessViewIds[SVGA3D_MAX_UAVIEWS]; } renderState; + uint32 pad0[8]; + struct { uint32 targets[SVGA3D_DX_MAX_SOTARGETS]; uint32 soid; } streamOut; - uint32 pad0[11]; + + uint32 pad1[10]; + + uint32 uavSpliceIndex; uint8 numViewports; uint8 numScissorRects; - uint16 pad1[1]; + uint16 pad2[1]; - uint32 pad2[3]; + uint32 pad3[3]; SVGA3dViewport viewports[SVGA3D_DX_MAX_VIEWPORTS]; - uint32 pad3[32]; + uint32 pad4[32]; SVGASignedRect scissorRects[SVGA3D_DX_MAX_SCISSORRECTS]; - uint32 pad4[64]; + uint32 pad5[64]; struct { uint32 queryID; uint32 value; } predication; - uint32 pad5[2]; + SVGAMobId shaderIfaceMobid; + uint32 shaderIfaceOffset; struct { uint32 shaderId; SVGA3dConstantBufferBinding constantBuffers[SVGA3D_DX_MAX_CONSTBUFFERS]; @@ -1619,11 +2024,38 @@ struct SVGADXContextMobFormat { SVGA3dQueryId queryID[SVGA3D_MAX_QUERY]; SVGA3dCOTableData cotables[SVGA_COTABLE_MAX]; - uint32 pad7[380]; + + uint32 pad7[64]; + + uint32 uaViewIds[SVGA3D_DX11_1_MAX_UAVIEWS]; + uint32 csuaViewIds[SVGA3D_DX11_1_MAX_UAVIEWS]; + + uint32 pad8[188]; } #include "vmware_pack_end.h" SVGADXContextMobFormat; +/* + * There is conflicting documentation on max class instances (253 vs 256). The + * lower value is the one used throughout the device, but since mob format is + * more involved to increase if needed, conservatively use the higher one here. + */ +#define SVGA3D_DX_MAX_CLASS_INSTANCES_PADDED 256 + +typedef +#include "vmware_pack_begin.h" +struct SVGADXShaderIfaceMobFormat { + struct { + uint32 numClassInstances; + uint32 iface[SVGA3D_DX_MAX_CLASS_INSTANCES_PADDED]; + SVGA3dIfaceData data[SVGA3D_DX_MAX_CLASS_INSTANCES_PADDED]; + } shaderIfaceState[SVGA3D_NUM_SHADERTYPE]; + + uint32 pad0[1018]; +} +#include "vmware_pack_end.h" +SVGADXShaderIfaceMobFormat; + typedef #include "vmware_pack_begin.h" struct SVGA3dCmdDXTempSetContext { diff --git a/drivers/gpu/drm/vmwgfx/device_include/svga3d_limits.h b/drivers/gpu/drm/vmwgfx/device_include/svga3d_limits.h index b22a67f15660..f4375a41b3aa 100644 --- a/drivers/gpu/drm/vmwgfx/device_include/svga3d_limits.h +++ b/drivers/gpu/drm/vmwgfx/device_include/svga3d_limits.h @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: GPL-2.0 OR MIT */ /********************************************************** - * Copyright 2007-2015 VMware, Inc. + * Copyright 2007-2019 VMware, Inc. * * Permission is hereby granted, free of charge, to any person * obtaining a copy of this software and associated documentation @@ -40,11 +40,25 @@ #include "includeCheck.h" #define SVGA3D_NUM_CLIPPLANES 6 +#define SVGA3D_MAX_CONTEXT_IDS 256 +#define SVGA3D_MAX_SURFACE_IDS (32 * 1024) + +/* + * While there are separate bind-points for RenderTargetViews and + * UnorderedAccessViews in a DXContext, there is in fact one shared + * semantic space that the guest-driver can use on any given draw call. + * So there are really only 8 slots that can be spilt up between them, with the + * spliceIndex controlling where the UAV's sit in the collapsed array. + */ #define SVGA3D_MAX_RENDER_TARGETS 8 #define SVGA3D_MAX_SIMULTANEOUS_RENDER_TARGETS (SVGA3D_MAX_RENDER_TARGETS) #define SVGA3D_MAX_UAVIEWS 8 -#define SVGA3D_MAX_CONTEXT_IDS 256 -#define SVGA3D_MAX_SURFACE_IDS (32 * 1024) +#define SVGA3D_DX11_1_MAX_UAVIEWS 64 + +/* + * Maximum canonical size of a surface in host-backed mode (pre-GBObjects). + */ +#define SVGA3D_HB_MAX_SURFACE_SIZE MBYTES_2_BYTES(128) /* * Maximum ID a shader can be assigned on a given context. @@ -59,6 +73,8 @@ #define SVGA3D_NUM_TEXTURE_UNITS 32 #define SVGA3D_NUM_LIGHTS 8 +#define SVGA3D_MAX_VIDEOPROCESSOR_SAMPLERS 32 + /* * Maximum size in dwords of shader text the SVGA device will allow. * Currently 8 MB. @@ -67,6 +83,11 @@ #define SVGA3D_MAX_SHADER_MEMORY (SVGA3D_MAX_SHADER_MEMORY_BYTES / \ sizeof(uint32)) +/* + * The maximum value of threadGroupCount in each dimension + */ +#define SVGA3D_MAX_SHADER_THREAD_GROUPS 65535 + #define SVGA3D_MAX_CLIP_PLANES 6 /* @@ -85,7 +106,9 @@ /* * Maximum number of array indexes in a GB surface (with DX enabled). */ -#define SVGA3D_MAX_SURFACE_ARRAYSIZE 512 +#define SVGA3D_SM4_MAX_SURFACE_ARRAYSIZE 512 +#define SVGA3D_SM5_MAX_SURFACE_ARRAYSIZE 2048 +#define SVGA3D_MAX_SURFACE_ARRAYSIZE SVGA3D_SM5_MAX_SURFACE_ARRAYSIZE /* * The maximum number of vertex arrays we're guaranteed to support in @@ -99,4 +122,9 @@ */ #define SVGA3D_MAX_DRAW_PRIMITIVE_RANGES 32 +/* + * The maximum number of samples that can be contained in a surface. + */ +#define SVGA3D_MAX_SAMPLES 8 + #endif /* _SVGA3D_LIMITS_H_ */ diff --git a/drivers/gpu/drm/vmwgfx/device_include/svga3d_surfacedefs.h b/drivers/gpu/drm/vmwgfx/device_include/svga3d_surfacedefs.h index 61414f105c67..4db25bd9fa22 100644 --- a/drivers/gpu/drm/vmwgfx/device_include/svga3d_surfacedefs.h +++ b/drivers/gpu/drm/vmwgfx/device_include/svga3d_surfacedefs.h @@ -131,6 +131,8 @@ enum svga3d_block_desc { SVGA3DBLOCKDESC_BC3 = 1 << 26, SVGA3DBLOCKDESC_BC4 = 1 << 27, SVGA3DBLOCKDESC_BC5 = 1 << 28, + SVGA3DBLOCKDESC_BC6H = 1 << 29, + SVGA3DBLOCKDESC_BC7 = 1 << 30, SVGA3DBLOCKDESC_A_UINT = SVGA3DBLOCKDESC_ALPHA | SVGA3DBLOCKDESC_UINT | @@ -290,6 +292,18 @@ enum svga3d_block_desc { SVGA3DBLOCKDESC_COMP_UNORM, SVGA3DBLOCKDESC_BC5_COMP_SNORM = SVGA3DBLOCKDESC_BC5 | SVGA3DBLOCKDESC_COMP_SNORM, + SVGA3DBLOCKDESC_BC6H_COMP_TYPELESS = SVGA3DBLOCKDESC_BC6H | + SVGA3DBLOCKDESC_COMP_TYPELESS, + SVGA3DBLOCKDESC_BC6H_COMP_UF16 = SVGA3DBLOCKDESC_BC6H | + SVGA3DBLOCKDESC_COMPRESSED, + SVGA3DBLOCKDESC_BC6H_COMP_SF16 = SVGA3DBLOCKDESC_BC6H | + SVGA3DBLOCKDESC_COMPRESSED, + SVGA3DBLOCKDESC_BC7_COMP_TYPELESS = SVGA3DBLOCKDESC_BC7 | + SVGA3DBLOCKDESC_COMP_TYPELESS, + SVGA3DBLOCKDESC_BC7_COMP_UNORM = SVGA3DBLOCKDESC_BC7 | + SVGA3DBLOCKDESC_COMP_UNORM, + SVGA3DBLOCKDESC_BC7_COMP_UNORM_SRGB = SVGA3DBLOCKDESC_BC7_COMP_UNORM | + SVGA3DBLOCKDESC_SRGB, SVGA3DBLOCKDESC_NV12 = SVGA3DBLOCKDESC_YUV_VIDEO | SVGA3DBLOCKDESC_PLANAR_YUV | @@ -494,7 +508,7 @@ static const struct svga3d_surface_desc svga3d_surface_descs[] = { {{8}, {8}, {8}, {0}}, {{16}, {8}, {0}, {0}}}, - {SVGA3D_FORMAT_DEAD1, SVGA3DBLOCKDESC_UVL, + {SVGA3D_FORMAT_DEAD1, SVGA3DBLOCKDESC_NONE, {1, 1, 1}, 3, 3, {{8}, {8}, {8}, {0}}, {{16}, {8}, {0}, {0}}}, @@ -604,7 +618,7 @@ static const struct svga3d_surface_desc svga3d_surface_descs[] = { {{0}, {0}, {48}, {0}}, {{0}, {0}, {0}, {0}}}, - {SVGA3D_AYUV, SVGA3DBLOCKDESC_AYUV, + {SVGA3D_FORMAT_DEAD2, SVGA3DBLOCKDESC_NONE, {1, 1, 1}, 4, 4, {{8}, {8}, {8}, {8}}, {{0}, {8}, {16}, {24}}}, @@ -1103,6 +1117,46 @@ static const struct svga3d_surface_desc svga3d_surface_descs[] = { {4, 4, 1}, 16, 16, {{0}, {0}, {128}, {0}}, {{0}, {0}, {0}, {0}}}, + + {SVGA3D_B4G4R4A4_UNORM, SVGA3DBLOCKDESC_RGBA_UNORM, + {1, 1, 1}, 2, 2, + {{4}, {4}, {4}, {4}}, + {{0}, {4}, {8}, {12}}}, + + {SVGA3D_BC6H_TYPELESS, SVGA3DBLOCKDESC_BC6H_COMP_TYPELESS, + {4, 4, 1}, 16, 16, + {{0}, {0}, {128}, {0}}, + {{0}, {0}, {0}, {0}}}, + + {SVGA3D_BC6H_UF16, SVGA3DBLOCKDESC_BC6H_COMP_UF16, + {4, 4, 1}, 16, 16, + {{0}, {0}, {128}, {0}}, + {{0}, {0}, {0}, {0}}}, + + {SVGA3D_BC6H_SF16, SVGA3DBLOCKDESC_BC6H_COMP_SF16, + {4, 4, 1}, 16, 16, + {{0}, {0}, {128}, {0}}, + {{0}, {0}, {0}, {0}}}, + + {SVGA3D_BC7_TYPELESS, SVGA3DBLOCKDESC_BC7_COMP_TYPELESS, + {4, 4, 1}, 16, 16, + {{0}, {0}, {128}, {0}}, + {{0}, {0}, {0}, {0}}}, + + {SVGA3D_BC7_UNORM, SVGA3DBLOCKDESC_BC7_COMP_UNORM, + {4, 4, 1}, 16, 16, + {{0}, {0}, {128}, {0}}, + {{0}, {0}, {0}, {0}}}, + + {SVGA3D_BC7_UNORM_SRGB, SVGA3DBLOCKDESC_BC7_COMP_UNORM_SRGB, + {4, 4, 1}, 16, 16, + {{0}, {0}, {128}, {0}}, + {{0}, {0}, {0}, {0}}}, + + {SVGA3D_AYUV, SVGA3DBLOCKDESC_AYUV, + {1, 1, 1}, 4, 4, + {{8}, {8}, {8}, {8}}, + {{0}, {8}, {16}, {24}}}, }; static inline u32 clamped_umul32(u32 a, u32 b) diff --git a/drivers/gpu/drm/vmwgfx/device_include/svga3d_types.h b/drivers/gpu/drm/vmwgfx/device_include/svga3d_types.h index 308370665a8e..77e338a65791 100644 --- a/drivers/gpu/drm/vmwgfx/device_include/svga3d_types.h +++ b/drivers/gpu/drm/vmwgfx/device_include/svga3d_types.h @@ -116,6 +116,19 @@ SVGA3dBox; typedef #include "vmware_pack_begin.h" struct { + int32 x; + int32 y; + int32 z; + int32 w; + int32 h; + int32 d; +} +#include "vmware_pack_end.h" +SVGA3dSignedBox; + +typedef +#include "vmware_pack_begin.h" +struct { uint32 x; uint32 y; uint32 z; @@ -198,8 +211,7 @@ typedef enum SVGA3dSurfaceFormat { /* Planar video formats */ SVGA3D_NV12 = 44, - /* Video format with alpha */ - SVGA3D_AYUV = 45, + SVGA3D_FORMAT_DEAD2 = 45, SVGA3D_R32G32B32A32_TYPELESS = 46, SVGA3D_R32G32B32A32_UINT = 47, @@ -305,6 +317,18 @@ typedef enum SVGA3dSurfaceFormat { SVGA3D_B8G8R8X8_UNORM = 142, SVGA3D_BC4_UNORM = 143, SVGA3D_BC5_UNORM = 144, + SVGA3D_B4G4R4A4_UNORM = 145, + + /* DX11 compressed formats */ + SVGA3D_BC6H_TYPELESS = 146, + SVGA3D_BC6H_UF16 = 147, + SVGA3D_BC6H_SF16 = 148, + SVGA3D_BC7_TYPELESS = 149, + SVGA3D_BC7_UNORM = 150, + SVGA3D_BC7_UNORM_SRGB = 151, + + /* Video format with alpha */ + SVGA3D_AYUV = 152, SVGA3D_FORMAT_MAX } SVGA3dSurfaceFormat; @@ -326,10 +350,10 @@ typedef enum SVGA3dSurfaceFormat { #define SVGA3D_SURFACE_HINT_RENDERTARGET (CONST64U(1) << 6) #define SVGA3D_SURFACE_HINT_DEPTHSTENCIL (CONST64U(1) << 7) #define SVGA3D_SURFACE_HINT_WRITEONLY (CONST64U(1) << 8) -#define SVGA3D_SURFACE_MASKABLE_ANTIALIAS (CONST64U(1) << 9) +#define SVGA3D_SURFACE_DEAD2 (CONST64U(1) << 9) #define SVGA3D_SURFACE_AUTOGENMIPMAPS (CONST64U(1) << 10) -#define SVGA3D_SURFACE_DECODE_RENDERTARGET (CONST64U(1) << 11) +#define SVGA3D_SURFACE_DEAD1 (CONST64U(1) << 11) /* * Is this surface using a base-level pitch for it's mob backing? @@ -387,7 +411,7 @@ typedef enum SVGA3dSurfaceFormat { * Setting this flag allow this surface to be used with the * SVGA_3D_CMD_DX_TRANSFER_FROM_BUFFER command. It is only valid for * buffer surfaces, and no bind flags are allowed to be set on surfaces - * with this flag. + * with this flag except SVGA3D_SURFACE_TRANSFER_TO_BUFFER. */ #define SVGA3D_SURFACE_TRANSFER_FROM_BUFFER (CONST64U(1) << 30) @@ -402,7 +426,31 @@ typedef enum SVGA3dSurfaceFormat { */ #define SVGA3D_SURFACE_MULTISAMPLE (CONST64U(1) << 32) -#define SVGA3D_SURFACE_FLAG_MAX (CONST64U(1) << 33) +/* + * Specified that the surface is allowed to be bound to a UAView. + */ +#define SVGA3D_SURFACE_BIND_UAVIEW (CONST64U(1) << 33) + +/* + * Setting this flag allow this surface to be used with the + * SVGA_3D_CMD_DX_TRANSFER_TO_BUFFER command. It is only valid for + * buffer surfaces, and no bind flags are allowed to be set on surfaces + * with this flag except SVGA3D_SURFACE_TRANSFER_FROM_BUFFER. + */ +#define SVGA3D_SURFACE_TRANSFER_TO_BUFFER (CONST64U(1) << 34) + +#define SVGA3D_SURFACE_BIND_LOGICOPS (CONST64U(1) << 35) + +/* + * Optional flags for use with SVGA3D_SURFACE_BIND_UAVIEW + */ +#define SVGA3D_SURFACE_BIND_RAW_VIEWS (CONST64U(1) << 36) +#define SVGA3D_SURFACE_BUFFER_STRUCTURED (CONST64U(1) << 37) + +#define SVGA3D_SURFACE_DRAWINDIRECT_ARGS (CONST64U(1) << 38) +#define SVGA3D_SURFACE_RESOURCE_CLAMP (CONST64U(1) << 39) + +#define SVGA3D_SURFACE_FLAG_MAX (CONST64U(1) << 40) /* * Surface flags types: @@ -428,17 +476,25 @@ typedef uint64 SVGA3dSurfaceAllFlags; SVGA3D_SURFACE_STAGING_DOWNLOAD | \ SVGA3D_SURFACE_HINT_INDIRECT_UPDATE | \ SVGA3D_SURFACE_TRANSFER_FROM_BUFFER | \ - SVGA3D_SURFACE_MULTISAMPLE \ + SVGA3D_SURFACE_RESERVED1 | \ + SVGA3D_SURFACE_MULTISAMPLE | \ + SVGA3D_SURFACE_BIND_UAVIEW | \ + SVGA3D_SURFACE_TRANSFER_TO_BUFFER | \ + SVGA3D_SURFACE_BIND_LOGICOPS | \ + SVGA3D_SURFACE_BIND_RAW_VIEWS | \ + SVGA3D_SURFACE_BUFFER_STRUCTURED | \ + SVGA3D_SURFACE_DRAWINDIRECT_ARGS | \ + SVGA3D_SURFACE_RESOURCE_CLAMP \ ) #define SVGA3D_SURFACE_HB_PRESENT_DISALLOWED_MASK \ ( SVGA3D_SURFACE_1D | \ + SVGA3D_SURFACE_RESERVED1 | \ SVGA3D_SURFACE_MULTISAMPLE \ ) #define SVGA3D_SURFACE_2D_DISALLOWED_MASK \ ( SVGA3D_SURFACE_CUBEMAP | \ - SVGA3D_SURFACE_MASKABLE_ANTIALIAS | \ SVGA3D_SURFACE_AUTOGENMIPMAPS | \ SVGA3D_SURFACE_VOLUME | \ SVGA3D_SURFACE_1D | \ @@ -448,7 +504,14 @@ typedef uint64 SVGA3dSurfaceAllFlags; SVGA3D_SURFACE_BIND_DEPTH_STENCIL | \ SVGA3D_SURFACE_BIND_STREAM_OUTPUT | \ SVGA3D_SURFACE_TRANSFER_FROM_BUFFER | \ - SVGA3D_SURFACE_MULTISAMPLE \ + SVGA3D_SURFACE_RESERVED1 | \ + SVGA3D_SURFACE_MULTISAMPLE | \ + SVGA3D_SURFACE_BIND_UAVIEW | \ + SVGA3D_SURFACE_TRANSFER_TO_BUFFER | \ + SVGA3D_SURFACE_BIND_RAW_VIEWS | \ + SVGA3D_SURFACE_BUFFER_STRUCTURED | \ + SVGA3D_SURFACE_DRAWINDIRECT_ARGS | \ + SVGA3D_SURFACE_RESOURCE_CLAMP \ ) #define SVGA3D_SURFACE_BASICOPS_DISALLOWED_MASK \ @@ -456,6 +519,7 @@ typedef uint64 SVGA3dSurfaceAllFlags; SVGA3D_SURFACE_AUTOGENMIPMAPS | \ SVGA3D_SURFACE_VOLUME | \ SVGA3D_SURFACE_1D | \ + SVGA3D_SURFACE_RESERVED1 | \ SVGA3D_SURFACE_MULTISAMPLE \ ) @@ -474,7 +538,14 @@ typedef uint64 SVGA3dSurfaceAllFlags; SVGA3D_SURFACE_STAGING_DOWNLOAD | \ SVGA3D_SURFACE_HINT_INDIRECT_UPDATE | \ SVGA3D_SURFACE_TRANSFER_FROM_BUFFER | \ - SVGA3D_SURFACE_MULTISAMPLE \ + SVGA3D_SURFACE_RESERVED1 | \ + SVGA3D_SURFACE_MULTISAMPLE | \ + SVGA3D_SURFACE_BIND_UAVIEW | \ + SVGA3D_SURFACE_TRANSFER_TO_BUFFER | \ + SVGA3D_SURFACE_BIND_RAW_VIEWS | \ + SVGA3D_SURFACE_BUFFER_STRUCTURED | \ + SVGA3D_SURFACE_DRAWINDIRECT_ARGS | \ + SVGA3D_SURFACE_RESOURCE_CLAMP \ ) #define SVGA3D_SURFACE_BUFFER_DISALLOWED_MASK \ @@ -482,10 +553,11 @@ typedef uint64 SVGA3dSurfaceAllFlags; SVGA3D_SURFACE_AUTOGENMIPMAPS | \ SVGA3D_SURFACE_VOLUME | \ SVGA3D_SURFACE_1D | \ - SVGA3D_SURFACE_MASKABLE_ANTIALIAS | \ + SVGA3D_SURFACE_DEAD2 | \ SVGA3D_SURFACE_ARRAY | \ SVGA3D_SURFACE_MULTISAMPLE | \ - SVGA3D_SURFACE_MOB_PITCH \ + SVGA3D_SURFACE_MOB_PITCH | \ + SVGA3D_SURFACE_RESOURCE_CLAMP \ ) #define SVGA3D_SURFACE_MULTISAMPLE_DISALLOWED_MASK \ @@ -494,14 +566,23 @@ typedef uint64 SVGA3dSurfaceAllFlags; SVGA3D_SURFACE_VOLUME | \ SVGA3D_SURFACE_1D | \ SVGA3D_SURFACE_SCREENTARGET | \ - SVGA3D_SURFACE_MOB_PITCH \ + SVGA3D_SURFACE_MOB_PITCH | \ + SVGA3D_SURFACE_TRANSFER_FROM_BUFFER | \ + SVGA3D_SURFACE_RESERVED1 | \ + SVGA3D_SURFACE_BIND_UAVIEW | \ + SVGA3D_SURFACE_TRANSFER_TO_BUFFER | \ + SVGA3D_SURFACE_BIND_LOGICOPS | \ + SVGA3D_SURFACE_BIND_RAW_VIEWS | \ + SVGA3D_SURFACE_BUFFER_STRUCTURED | \ + SVGA3D_SURFACE_DRAWINDIRECT_ARGS \ ) -#define SVGA3D_SURFACE_DX_ONLY_MASK \ - ( SVGA3D_SURFACE_BIND_STREAM_OUTPUT | \ - SVGA3D_SURFACE_STAGING_UPLOAD | \ - SVGA3D_SURFACE_STAGING_DOWNLOAD | \ - SVGA3D_SURFACE_TRANSFER_FROM_BUFFER \ +#define SVGA3D_SURFACE_DX_ONLY_MASK \ + ( SVGA3D_SURFACE_BIND_STREAM_OUTPUT | \ + SVGA3D_SURFACE_STAGING_UPLOAD | \ + SVGA3D_SURFACE_STAGING_DOWNLOAD | \ + SVGA3D_SURFACE_TRANSFER_FROM_BUFFER | \ + SVGA3D_SURFACE_TRANSFER_TO_BUFFER \ ) #define SVGA3D_SURFACE_STAGING_MASK \ @@ -516,9 +597,135 @@ typedef uint64 SVGA3dSurfaceAllFlags; SVGA3D_SURFACE_BIND_SHADER_RESOURCE | \ SVGA3D_SURFACE_BIND_RENDER_TARGET | \ SVGA3D_SURFACE_BIND_DEPTH_STENCIL | \ - SVGA3D_SURFACE_BIND_STREAM_OUTPUT \ + SVGA3D_SURFACE_BIND_STREAM_OUTPUT | \ + SVGA3D_SURFACE_BIND_UAVIEW | \ + SVGA3D_SURFACE_BIND_LOGICOPS | \ + SVGA3D_SURFACE_BIND_RAW_VIEWS \ + ) + +#define SVGA3D_SURFACE_VADECODE_DISALLOWED_MASK \ + ( SVGA3D_SURFACE_CUBEMAP | \ + SVGA3D_SURFACE_HINT_STATIC | \ + SVGA3D_SURFACE_HINT_DYNAMIC | \ + SVGA3D_SURFACE_HINT_INDEXBUFFER | \ + SVGA3D_SURFACE_HINT_VERTEXBUFFER | \ + SVGA3D_SURFACE_HINT_TEXTURE | \ + SVGA3D_SURFACE_HINT_RENDERTARGET | \ + SVGA3D_SURFACE_HINT_DEPTHSTENCIL | \ + SVGA3D_SURFACE_HINT_WRITEONLY | \ + SVGA3D_SURFACE_DEAD2 | \ + SVGA3D_SURFACE_AUTOGENMIPMAPS | \ + SVGA3D_SURFACE_HINT_RT_LOCKABLE | \ + SVGA3D_SURFACE_VOLUME | \ + SVGA3D_SURFACE_SCREENTARGET | \ + SVGA3D_SURFACE_1D | \ + SVGA3D_SURFACE_BIND_VERTEX_BUFFER | \ + SVGA3D_SURFACE_BIND_INDEX_BUFFER | \ + SVGA3D_SURFACE_BIND_CONSTANT_BUFFER | \ + SVGA3D_SURFACE_BIND_RENDER_TARGET | \ + SVGA3D_SURFACE_BIND_SHADER_RESOURCE | \ + SVGA3D_SURFACE_BIND_DEPTH_STENCIL | \ + SVGA3D_SURFACE_BIND_STREAM_OUTPUT | \ + SVGA3D_SURFACE_INACTIVE | \ + SVGA3D_SURFACE_STAGING_UPLOAD | \ + SVGA3D_SURFACE_STAGING_DOWNLOAD | \ + SVGA3D_SURFACE_HINT_INDIRECT_UPDATE | \ + SVGA3D_SURFACE_TRANSFER_FROM_BUFFER | \ + SVGA3D_SURFACE_MULTISAMPLE | \ + SVGA3D_SURFACE_BIND_UAVIEW | \ + SVGA3D_SURFACE_TRANSFER_TO_BUFFER | \ + SVGA3D_SURFACE_BIND_LOGICOPS | \ + SVGA3D_SURFACE_BIND_RAW_VIEWS | \ + SVGA3D_SURFACE_BUFFER_STRUCTURED | \ + SVGA3D_SURFACE_DRAWINDIRECT_ARGS | \ + SVGA3D_SURFACE_RESOURCE_CLAMP \ + ) + +#define SVGA3D_SURFACE_VAPROCESSFRAME_OUTPUT_DISALLOWED_MASK \ + ( SVGA3D_SURFACE_HINT_INDEXBUFFER | \ + SVGA3D_SURFACE_HINT_VERTEXBUFFER | \ + SVGA3D_SURFACE_HINT_DEPTHSTENCIL | \ + SVGA3D_SURFACE_DEAD2 | \ + SVGA3D_SURFACE_VOLUME | \ + SVGA3D_SURFACE_1D | \ + SVGA3D_SURFACE_BIND_VERTEX_BUFFER | \ + SVGA3D_SURFACE_BIND_INDEX_BUFFER | \ + SVGA3D_SURFACE_BIND_CONSTANT_BUFFER | \ + SVGA3D_SURFACE_BIND_DEPTH_STENCIL | \ + SVGA3D_SURFACE_BIND_STREAM_OUTPUT | \ + SVGA3D_SURFACE_INACTIVE | \ + SVGA3D_SURFACE_STAGING_UPLOAD | \ + SVGA3D_SURFACE_STAGING_DOWNLOAD | \ + SVGA3D_SURFACE_TRANSFER_FROM_BUFFER | \ + SVGA3D_SURFACE_VADECODE | \ + SVGA3D_SURFACE_MULTISAMPLE | \ + SVGA3D_SURFACE_BIND_UAVIEW | \ + SVGA3D_SURFACE_TRANSFER_TO_BUFFER | \ + SVGA3D_SURFACE_BIND_LOGICOPS | \ + SVGA3D_SURFACE_BIND_RAW_VIEWS | \ + SVGA3D_SURFACE_BUFFER_STRUCTURED | \ + SVGA3D_SURFACE_DRAWINDIRECT_ARGS | \ + SVGA3D_SURFACE_RESOURCE_CLAMP \ + ) + +#define SVGA3D_SURFACE_VAPROCESSFRAME_INPUT_DISALLOWED_MASK \ + ( SVGA3D_SURFACE_CUBEMAP | \ + SVGA3D_SURFACE_HINT_INDEXBUFFER | \ + SVGA3D_SURFACE_HINT_VERTEXBUFFER | \ + SVGA3D_SURFACE_HINT_DEPTHSTENCIL | \ + SVGA3D_SURFACE_DEAD2 | \ + SVGA3D_SURFACE_VOLUME | \ + SVGA3D_SURFACE_SCREENTARGET | \ + SVGA3D_SURFACE_1D | \ + SVGA3D_SURFACE_BIND_VERTEX_BUFFER | \ + SVGA3D_SURFACE_BIND_INDEX_BUFFER | \ + SVGA3D_SURFACE_BIND_CONSTANT_BUFFER | \ + SVGA3D_SURFACE_BIND_DEPTH_STENCIL | \ + SVGA3D_SURFACE_BIND_STREAM_OUTPUT | \ + SVGA3D_SURFACE_STAGING_UPLOAD | \ + SVGA3D_SURFACE_STAGING_DOWNLOAD | \ + SVGA3D_SURFACE_TRANSFER_FROM_BUFFER | \ + SVGA3D_SURFACE_MULTISAMPLE | \ + SVGA3D_SURFACE_BIND_UAVIEW | \ + SVGA3D_SURFACE_TRANSFER_TO_BUFFER | \ + SVGA3D_SURFACE_BIND_LOGICOPS | \ + SVGA3D_SURFACE_BIND_RAW_VIEWS | \ + SVGA3D_SURFACE_BUFFER_STRUCTURED | \ + SVGA3D_SURFACE_DRAWINDIRECT_ARGS | \ + SVGA3D_SURFACE_RESOURCE_CLAMP \ + ) + +#define SVGA3D_SURFACE_LOGICOPS_DISALLOWED_MASK \ + ( SVGA3D_SURFACE_CUBEMAP | \ + SVGA3D_SURFACE_DEAD2 | \ + SVGA3D_SURFACE_AUTOGENMIPMAPS | \ + SVGA3D_SURFACE_VOLUME | \ + SVGA3D_SURFACE_1D | \ + SVGA3D_SURFACE_BIND_VERTEX_BUFFER | \ + SVGA3D_SURFACE_BIND_INDEX_BUFFER | \ + SVGA3D_SURFACE_BIND_CONSTANT_BUFFER | \ + SVGA3D_SURFACE_BIND_DEPTH_STENCIL | \ + SVGA3D_SURFACE_BIND_STREAM_OUTPUT | \ + SVGA3D_SURFACE_TRANSFER_FROM_BUFFER | \ + SVGA3D_SURFACE_VADECODE | \ + SVGA3D_SURFACE_MULTISAMPLE | \ + SVGA3D_SURFACE_BIND_UAVIEW | \ + SVGA3D_SURFACE_TRANSFER_TO_BUFFER | \ + SVGA3D_SURFACE_BIND_RAW_VIEWS | \ + SVGA3D_SURFACE_BUFFER_STRUCTURED | \ + SVGA3D_SURFACE_DRAWINDIRECT_ARGS | \ + SVGA3D_SURFACE_RESOURCE_CLAMP \ ) +#define SVGA3D_BUFFER_STRUCTURED_STRIDE_MAX 2048 + + +/* + * These are really the D3DFORMAT_OP defines from the wdk. We need + * them so that we can query the host for what the supported surface + * operations are (when we're using the D3D backend, in particular), + * and so we can send those operations to the guest. + */ typedef enum { SVGA3DFORMAT_OP_TEXTURE = 0x00000001, SVGA3DFORMAT_OP_VOLUMETEXTURE = 0x00000002, @@ -1338,7 +1545,40 @@ typedef enum { SVGA3D_PRIMITIVE_LINESTRIP_ADJ = 8, SVGA3D_PRIMITIVE_TRIANGLELIST_ADJ = 9, SVGA3D_PRIMITIVE_TRIANGLESTRIP_ADJ = 10, - SVGA3D_PRIMITIVE_MAX + SVGA3D_PRIMITIVE_DX10_MAX = 11, + SVGA3D_PRIMITIVE_1_CONTROL_POINT_PATCH = 11, + SVGA3D_PRIMITIVE_2_CONTROL_POINT_PATCH = 12, + SVGA3D_PRIMITIVE_3_CONTROL_POINT_PATCH = 13, + SVGA3D_PRIMITIVE_4_CONTROL_POINT_PATCH = 14, + SVGA3D_PRIMITIVE_5_CONTROL_POINT_PATCH = 15, + SVGA3D_PRIMITIVE_6_CONTROL_POINT_PATCH = 16, + SVGA3D_PRIMITIVE_7_CONTROL_POINT_PATCH = 17, + SVGA3D_PRIMITIVE_8_CONTROL_POINT_PATCH = 18, + SVGA3D_PRIMITIVE_9_CONTROL_POINT_PATCH = 19, + SVGA3D_PRIMITIVE_10_CONTROL_POINT_PATCH = 20, + SVGA3D_PRIMITIVE_11_CONTROL_POINT_PATCH = 21, + SVGA3D_PRIMITIVE_12_CONTROL_POINT_PATCH = 22, + SVGA3D_PRIMITIVE_13_CONTROL_POINT_PATCH = 23, + SVGA3D_PRIMITIVE_14_CONTROL_POINT_PATCH = 24, + SVGA3D_PRIMITIVE_15_CONTROL_POINT_PATCH = 25, + SVGA3D_PRIMITIVE_16_CONTROL_POINT_PATCH = 26, + SVGA3D_PRIMITIVE_17_CONTROL_POINT_PATCH = 27, + SVGA3D_PRIMITIVE_18_CONTROL_POINT_PATCH = 28, + SVGA3D_PRIMITIVE_19_CONTROL_POINT_PATCH = 29, + SVGA3D_PRIMITIVE_20_CONTROL_POINT_PATCH = 30, + SVGA3D_PRIMITIVE_21_CONTROL_POINT_PATCH = 31, + SVGA3D_PRIMITIVE_22_CONTROL_POINT_PATCH = 32, + SVGA3D_PRIMITIVE_23_CONTROL_POINT_PATCH = 33, + SVGA3D_PRIMITIVE_24_CONTROL_POINT_PATCH = 34, + SVGA3D_PRIMITIVE_25_CONTROL_POINT_PATCH = 35, + SVGA3D_PRIMITIVE_26_CONTROL_POINT_PATCH = 36, + SVGA3D_PRIMITIVE_27_CONTROL_POINT_PATCH = 37, + SVGA3D_PRIMITIVE_28_CONTROL_POINT_PATCH = 38, + SVGA3D_PRIMITIVE_29_CONTROL_POINT_PATCH = 39, + SVGA3D_PRIMITIVE_30_CONTROL_POINT_PATCH = 40, + SVGA3D_PRIMITIVE_31_CONTROL_POINT_PATCH = 41, + SVGA3D_PRIMITIVE_32_CONTROL_POINT_PATCH = 42, + SVGA3D_PRIMITIVE_MAX = 43 } SVGA3dPrimitiveType; typedef enum { @@ -1442,16 +1682,15 @@ typedef enum { SVGA3D_QUERYTYPE_STREAMOUTPUTSTATS = 5, SVGA3D_QUERYTYPE_STREAMOVERFLOWPREDICATE = 6, SVGA3D_QUERYTYPE_OCCLUSION64 = 7, - SVGA3D_QUERYTYPE_EVENT = 8, - SVGA3D_QUERYTYPE_DX10_MAX = 9, - SVGA3D_QUERYTYPE_SOSTATS_STREAM0 = 9, - SVGA3D_QUERYTYPE_SOSTATS_STREAM1 = 10, - SVGA3D_QUERYTYPE_SOSTATS_STREAM2 = 11, - SVGA3D_QUERYTYPE_SOSTATS_STREAM3 = 12, - SVGA3D_QUERYTYPE_SOP_STREAM0 = 13, - SVGA3D_QUERYTYPE_SOP_STREAM1 = 14, - SVGA3D_QUERYTYPE_SOP_STREAM2 = 15, - SVGA3D_QUERYTYPE_SOP_STREAM3 = 16, + SVGA3D_QUERYTYPE_DX10_MAX = 8, + SVGA3D_QUERYTYPE_SOSTATS_STREAM0 = 8, + SVGA3D_QUERYTYPE_SOSTATS_STREAM1 = 9, + SVGA3D_QUERYTYPE_SOSTATS_STREAM2 = 10, + SVGA3D_QUERYTYPE_SOSTATS_STREAM3 = 11, + SVGA3D_QUERYTYPE_SOP_STREAM0 = 12, + SVGA3D_QUERYTYPE_SOP_STREAM1 = 13, + SVGA3D_QUERYTYPE_SOP_STREAM2 = 14, + SVGA3D_QUERYTYPE_SOP_STREAM3 = 15, SVGA3D_QUERYTYPE_MAX } SVGA3dQueryType; @@ -1584,28 +1823,33 @@ typedef enum { SVGA3D_READ_HOST_VRAM = 2, } SVGA3dTransferType; -typedef enum { - SVGA3D_LOGICOP_INVALID = 0, - SVGA3D_LOGICOP_MIN = 1, - SVGA3D_LOGICOP_COPY = 1, - SVGA3D_LOGICOP_NOT = 2, - SVGA3D_LOGICOP_AND = 3, - SVGA3D_LOGICOP_OR = 4, - SVGA3D_LOGICOP_XOR = 5, - SVGA3D_LOGICOP_NXOR = 6, - SVGA3D_LOGICOP_ROP3MIN = 30, /* 7-29 are reserved for future logic ops. */ - SVGA3D_LOGICOP_ROP3MAX = (SVGA3D_LOGICOP_ROP3MIN + 255), - SVGA3D_LOGICOP_MAX = (SVGA3D_LOGICOP_ROP3MAX + 1), -} SVGA3dLogicOp; +#define SVGA3D_LOGICOP_INVALID 0 +#define SVGA3D_LOGICOP_MIN 1 +#define SVGA3D_LOGICOP_COPY 1 +#define SVGA3D_LOGICOP_NOT 2 +#define SVGA3D_LOGICOP_AND 3 +#define SVGA3D_LOGICOP_OR 4 +#define SVGA3D_LOGICOP_XOR 5 +#define SVGA3D_LOGICOP_NXOR 6 +#define SVGA3D_LOGICOP_ROP3 7 +#define SVGA3D_LOGICOP_MAX 8 + +typedef uint16 SVGA3dLogicOp; + +#define SVGA3D_LOGICOP_ROP3_INVALID ((uint16) -1) +#define SVGA3D_LOGICOP_ROP3_MIN 0 +#define SVGA3D_LOGICOP_ROP3_MAX 256 + +typedef uint16 SVGA3dLogicOpRop3; typedef #include "vmware_pack_begin.h" struct { union { struct { - uint16 function; /* SVGA3dFogFunction */ - uint8 type; /* SVGA3dFogType */ - uint8 base; /* SVGA3dFogBase */ + uint16 function; // SVGA3dFogFunction + uint8 type; // SVGA3dFogType + uint8 base; // SVGA3dFogBase }; uint32 uintValue; }; @@ -1742,4 +1986,15 @@ typedef enum SVGA3dMSQualityLevel { SVGA3D_MS_QUALITY_MAX = 2, } SVGA3dMSQualityLevel; +/* + * Screen Target Update Flags + */ + +typedef enum SVGA3dFrameUpdateType { + SVGA3D_FRAME_END = 0, + SVGA3D_FRAME_PARTIAL = 1, + SVGA3D_FRAME_UNKNOWN = 2, + SVGA3D_FRAME_MAX = 3, +} SVGA3dFrameUpdateType; + #endif /* _SVGA3D_TYPES_H_ */ diff --git a/drivers/gpu/drm/vmwgfx/device_include/svga_reg.h b/drivers/gpu/drm/vmwgfx/device_include/svga_reg.h index 056f54b35d73..19fb9e3299e7 100644 --- a/drivers/gpu/drm/vmwgfx/device_include/svga_reg.h +++ b/drivers/gpu/drm/vmwgfx/device_include/svga_reg.h @@ -70,8 +70,7 @@ typedef uint32 SVGAMobId; /* * Legal values for the SVGA_REG_CURSOR_ON register in old-fashioned - * cursor bypass mode. This is still supported, but no new guest - * drivers should use it. + * cursor bypass mode. */ #define SVGA_CURSOR_ON_HIDE 0x0 #define SVGA_CURSOR_ON_SHOW 0x1 @@ -137,6 +136,17 @@ typedef uint32 SVGAMobId; #define SVGA_IRQFLAG_ERROR 0x10 /* Error while processing commands */ /* + * The byte-size is the size of the actual cursor data, + * possibly after expanding it to the current bit depth. + * + * 40K is sufficient memory for two 32-bit planes for a 64 x 64 cursor. + * + * The dimension limit is a bound on the maximum width or height. + */ +#define SVGA_MAX_CURSOR_CMD_BYTES (40 * 1024) +#define SVGA_MAX_CURSOR_CMD_DIMENSION 1024 + +/* * Registers */ @@ -169,7 +179,7 @@ enum { SVGA_REG_SYNC = 21, /* See "FIFO Synchronization Registers" */ SVGA_REG_BUSY = 22, /* See "FIFO Synchronization Registers" */ SVGA_REG_GUEST_ID = 23, /* (Deprecated) */ - SVGA_REG_CURSOR_ID = 24, /* (Deprecated) */ + SVGA_REG_DEAD = 24, /* Drivers should never write this. */ SVGA_REG_CURSOR_X = 25, /* (Deprecated) */ SVGA_REG_CURSOR_Y = 26, /* (Deprecated) */ SVGA_REG_CURSOR_ON = 27, /* (Deprecated) */ @@ -208,7 +218,13 @@ enum { SVGA_REG_MAX_PRIMARY_MEM = 50, SVGA_REG_MAX_PRIMARY_BOUNDING_BOX_MEM = 50, - SVGA_REG_SUGGESTED_GBOBJECT_MEM_SIZE_KB = 51, /* Sugested limit on mob mem */ + /* + * Legacy version of SVGA_REG_GBOBJECT_MEM_SIZE_KB for drivers that + * don't know how to convert to a 64-bit byte value without overflowing. + * (See SVGA_REG_GBOBJECT_MEM_SIZE_KB). + */ + SVGA_REG_SUGGESTED_GBOBJECT_MEM_SIZE_KB = 51, + SVGA_REG_DEV_CAP = 52, /* Write dev cap index, read value */ SVGA_REG_CMD_PREPEND_LOW = 53, SVGA_REG_CMD_PREPEND_HIGH = 54, @@ -218,7 +234,59 @@ enum { SVGA_REG_BLANK_SCREEN_TARGETS = 58, SVGA_REG_CAP2 = 59, SVGA_REG_DEVEL_CAP = 60, - SVGA_REG_TOP = 61, /* Must be 1 more than the last register */ + + /* + * Allow the guest to hint to the device which driver is running. + * + * This should not generally change device behavior, but might be + * convenient to work-around specific bugs in guest drivers. + * + * Drivers should first write their id value into SVGA_REG_GUEST_DRIVER_ID, + * and then fill out all of the version registers that they have defined. + * + * After the driver has written all of the registers, they should + * then write the value SVGA_REG_GUEST_DRIVER_ID_SUBMIT to the + * SVGA_REG_GUEST_DRIVER_ID register, to signal that they have finished. + * + * The SVGA_REG_GUEST_DRIVER_ID values are defined below by the + * SVGARegGuestDriverId enum. + * + * The SVGA_REG_GUEST_DRIVER_VERSION fields are driver-specific, + * but ideally should encode a monotonically increasing number that allows + * the device to perform inequality checks against ranges of driver versions. + */ + SVGA_REG_GUEST_DRIVER_ID = 61, + SVGA_REG_GUEST_DRIVER_VERSION1 = 62, + SVGA_REG_GUEST_DRIVER_VERSION2 = 63, + SVGA_REG_GUEST_DRIVER_VERSION3 = 64, + SVGA_REG_CURSOR_MOBID = 65, + SVGA_REG_CURSOR_MAX_BYTE_SIZE = 66, + SVGA_REG_CURSOR_MAX_DIMENSION = 67, + + SVGA_REG_FIFO_CAPS = 68, + SVGA_REG_FENCE = 69, + + SVGA_REG_RESERVED1 = 70, + SVGA_REG_RESERVED2 = 71, + SVGA_REG_RESERVED3 = 72, + SVGA_REG_RESERVED4 = 73, + SVGA_REG_RESERVED5 = 74, + SVGA_REG_SCREENDMA = 75, + + /* + * The maximum amount of guest-backed objects that the device can have + * resident at a time. Guest-drivers should keep their working set size + * below this limit for best performance. + * + * Note that this value is in kilobytes, and not bytes, because the actual + * number of bytes might be larger than can fit in a 32-bit register. + * + * PLEASE USE A 64-BIT VALUE WHEN CONVERTING THIS INTO BYTES. + * (See SVGA_REG_SUGGESTED_GBOBJECT_MEM_SIZE_KB). + */ + SVGA_REG_GBOBJECT_MEM_SIZE_KB = 76, + + SVGA_REG_TOP = 77, /* Must be 1 more than the last register */ SVGA_PALETTE_BASE = 1024, /* Base of SVGA color map */ /* Next 768 (== 256*3) registers exist for colormap */ @@ -229,6 +297,20 @@ enum { the use of the current SVGA driver. */ }; + +/* + * Values for SVGA_REG_GUEST_DRIVER_ID. + */ +typedef enum SVGARegGuestDriverId { + SVGA_REG_GUEST_DRIVER_ID_UNKNOWN = 0, + SVGA_REG_GUEST_DRIVER_ID_WDDM = 1, + SVGA_REG_GUEST_DRIVER_ID_LINUX = 2, + SVGA_REG_GUEST_DRIVER_ID_MAX, + + SVGA_REG_GUEST_DRIVER_ID_SUBMIT = MAX_UINT32, +} SVGARegGuestDriverId; + + /* * Guest memory regions (GMRs): * @@ -416,7 +498,6 @@ typedef enum { SVGA_CB_CONTEXT_0 = 0x0, SVGA_CB_CONTEXT_1 = 0x1, /* Supported with SVGA_CAP_HP_CMD_QUEUE */ SVGA_CB_CONTEXT_MAX = 0x2, - SVGA_CB_CONTEXT_HP_MAX = 0x2, } SVGACBContext; @@ -733,9 +814,6 @@ SVGASignedPoint; * and must not be reused. Those capabilities will never be reported * by new versions of the SVGA device. * - * XXX: Add longer descriptions for each capability, including a list - * of the new features that each capability provides. - * * SVGA_CAP_IRQMASK -- * Provides device interrupts. Adds device register SVGA_REG_IRQMASK * to set interrupt mask and direct I/O port SVGA_IRQSTATUS_PORT to @@ -842,17 +920,51 @@ SVGASignedPoint; * Allow the IntraSurfaceCopy command. * * SVGA_CAP2_DX2 -- - * Allow the DefineGBSurface_v3, WholeSurfaceCopy. + * Allow the DefineGBSurface_v3, WholeSurfaceCopy, WriteZeroSurface, and + * HintZeroSurface commands, and the SVGA_REG_GUEST_DRIVER_ID register. + * + * SVGA_CAP2_GB_MEMSIZE_2 -- + * Allow the SVGA_REG_GBOBJECT_MEM_SIZE_KB register. + * + * SVGA_CAP2_SCREENDMA_REG -- + * Allow the SVGA_REG_SCREENDMA register. + * + * SVGA_CAP2_OTABLE_PTDEPTH_2 -- + * Allow 2 level page tables for OTable commands. + * + * SVGA_CAP2_NON_MS_TO_MS_STRETCHBLT -- + * Allow a stretch blt from a non-multisampled surface to a multisampled + * surface. + * + * SVGA_CAP2_CURSOR_MOB -- + * Allow the SVGA_REG_CURSOR_MOBID register. + * + * SVGA_CAP2_MSHINT -- + * Allow the SVGA_REG_MSHINT register. + * + * SVGA_CAP2_DX3 -- + * Allows the DefineGBSurface_v4 command. + * Allows the DXDefineDepthStencilView_v2, DXDefineStreamOutputWithMob, + * and DXBindStreamOutput commands if 3D is also available. + * Allows the DXPredStagingCopy and DXStagingCopy commands if SM41 + * is also available. * * SVGA_CAP2_RESERVED -- * Reserve the last bit for extending the SVGA capabilities to some * future mechanisms. */ -#define SVGA_CAP2_NONE 0x00000000 -#define SVGA_CAP2_GROW_OTABLE 0x00000001 -#define SVGA_CAP2_INTRA_SURFACE_COPY 0x00000002 -#define SVGA_CAP2_DX2 0x00000004 -#define SVGA_CAP2_RESERVED 0x80000000 +#define SVGA_CAP2_NONE 0x00000000 +#define SVGA_CAP2_GROW_OTABLE 0x00000001 +#define SVGA_CAP2_INTRA_SURFACE_COPY 0x00000002 +#define SVGA_CAP2_DX2 0x00000004 +#define SVGA_CAP2_GB_MEMSIZE_2 0x00000008 +#define SVGA_CAP2_SCREENDMA_REG 0x00000010 +#define SVGA_CAP2_OTABLE_PTDEPTH_2 0x00000020 +#define SVGA_CAP2_NON_MS_TO_MS_STRETCHBLT 0x00000040 +#define SVGA_CAP2_CURSOR_MOB 0x00000080 +#define SVGA_CAP2_MSHINT 0x00000100 +#define SVGA_CAP2_DX3 0x00000400 +#define SVGA_CAP2_RESERVED 0x80000000 /* @@ -875,7 +987,9 @@ typedef enum { SVGABackdoorCapFifoCaps = 1, SVGABackdoorCap3dHWVersion = 2, SVGABackdoorCapDeviceCaps2 = 3, - SVGABackdoorCapMax = 4, + SVGABackdoorCapDevelCaps = 4, + SVGABackdoorDevelRenderer = 5, + SVGABackdoorCapMax = 6, } SVGABackdoorCapType; @@ -1055,103 +1169,80 @@ enum { /* * FIFO Synchronization Registers * - * This explains the relationship between the various FIFO - * sync-related registers in IOSpace and in FIFO space. - * * SVGA_REG_SYNC -- * - * The SYNC register can be used in two different ways by the guest: - * - * 1. If the guest wishes to fully sync (drain) the FIFO, - * it will write once to SYNC then poll on the BUSY - * register. The FIFO is sync'ed once BUSY is zero. - * - * 2. If the guest wants to asynchronously wake up the host, - * it will write once to SYNC without polling on BUSY. - * Ideally it will do this after some new commands have - * been placed in the FIFO, and after reading a zero - * from SVGA_FIFO_BUSY. - * - * (1) is the original behaviour that SYNC was designed to - * support. Originally, a write to SYNC would implicitly - * trigger a read from BUSY. This causes us to synchronously - * process the FIFO. - * - * This behaviour has since been changed so that writing SYNC - * will *not* implicitly cause a read from BUSY. Instead, it - * makes a channel call which asynchronously wakes up the MKS - * thread. - * - * New guests can use this new behaviour to implement (2) - * efficiently. This lets guests get the host's attention - * without waiting for the MKS to poll, which gives us much - * better CPU utilization on SMP hosts and on UP hosts while - * we're blocked on the host GPU. - * - * Old guests shouldn't notice the behaviour change. SYNC was - * never guaranteed to process the entire FIFO, since it was - * bounded to a particular number of CPU cycles. Old guests will - * still loop on the BUSY register until the FIFO is empty. - * - * Writing to SYNC currently has the following side-effects: - * - * - Sets SVGA_REG_BUSY to TRUE (in the monitor) - * - Asynchronously wakes up the MKS thread for FIFO processing - * - The value written to SYNC is recorded as a "reason", for - * stats purposes. - * - * If SVGA_FIFO_BUSY is available, drivers are advised to only - * write to SYNC if SVGA_FIFO_BUSY is FALSE. Drivers should set - * SVGA_FIFO_BUSY to TRUE after writing to SYNC. The MKS will - * eventually set SVGA_FIFO_BUSY on its own, but this approach - * lets the driver avoid sending multiple asynchronous wakeup - * messages to the MKS thread. + * The SYNC register can be used by the guest driver to signal to the + * device that the guest driver is waiting for previously submitted + * commands to complete. + * + * When the guest driver writes to the SYNC register, the device sets + * the BUSY register to TRUE, and starts processing the submitted commands + * (if it was not already doing so). When all previously submitted + * commands are finished and the device is idle again, it sets the BUSY + * register back to FALSE. (If the guest driver submits new commands + * after writing the SYNC register, the new commands are not guaranteed + * to have been procesesd.) + * + * When guest drivers are submitting commands using the FIFO, the device + * periodically polls to check for new FIFO commands when idle, which may + * introduce a delay in command processing. If the guest-driver wants + * the commands to be processed quickly (which it typically does), it + * should write SYNC after each batch of commands is committed to the + * FIFO to immediately wake up the device. For even better performance, + * the guest can use the SVGA_FIFO_BUSY register to avoid these extra + * SYNC writes if the device is already active, using the technique known + * as "Ringing the Doorbell" (described below). (Note that command + * buffer submission implicitly wakes up the device, and so doesn't + * suffer from this problem.) + * + * The SYNC register can also be used in combination with BUSY to + * synchronously ensure that all SVGA commands are processed (with both + * the FIFO and command-buffers). To do this, the guest driver should + * write to SYNC, and then loop reading BUSY until BUSY returns FALSE. + * This technique is known as a "Legacy Sync". * * SVGA_REG_BUSY -- * * This register is set to TRUE when SVGA_REG_SYNC is written, - * and it reads as FALSE when the FIFO has been completely - * drained. - * - * Every read from this register causes us to synchronously - * process FIFO commands. There is no guarantee as to how many - * commands each read will process. + * and is set back to FALSE when the device has finished processing + * all commands and is idle again. * - * CPU time spent processing FIFO commands will be billed to - * the guest. + * Every read from the BUSY reigster will block for an undefined + * amount of time (normally until the device finishes some interesting + * work unit), or the device is idle. * - * New drivers should avoid using this register unless they - * need to guarantee that the FIFO is completely drained. It - * is overkill for performing a sync-to-fence. Older drivers - * will use this register for any type of synchronization. + * Guest drivers can also do a partial Legacy Sync to check for some + * particular condition, for instance by stopping early when a fence + * passes before BUSY has been set back to FALSE. This is particularly + * useful if the guest-driver knows that it is blocked waiting on the + * device, because it will yield CPU time back to the host. * * SVGA_FIFO_BUSY -- * - * This register is a fast way for the guest driver to check - * whether the FIFO is already being processed. It reads and - * writes at normal RAM speeds, with no monitor intervention. - * - * If this register reads as TRUE, the host is guaranteeing that - * any new commands written into the FIFO will be noticed before - * the MKS goes back to sleep. + * The SVGA_FIFO_BUSY register is a fast way for the guest driver to check + * whether the device is actively processing FIFO commands before writing + * the more expensive SYNC register. * - * If this register reads as FALSE, no such guarantee can be - * made. + * If this register reads as TRUE, the device is actively processing + * FIFO commands. * - * The guest should use this register to quickly determine - * whether or not it needs to wake up the host. If the guest - * just wrote a command or group of commands that it would like - * the host to begin processing, it should: + * If this register reads as FALSE, the device may not be actively + * processing commands, and the guest driver should try + * "Ringing the Doorbell". * - * 1. Read SVGA_FIFO_BUSY. If it reads as TRUE, no further - * action is necessary. + * To Ring the Doorbell, the guest should: * - * 2. Write TRUE to SVGA_FIFO_BUSY. This informs future guest - * code that we've already sent a SYNC to the host and we - * don't need to send a duplicate. + * 1. Have already written their batch of commands into the FIFO. + * 2. Check if the SVGA_FIFO_BUSY register is available by reading + * SVGA_FIFO_MIN. + * 3. Read SVGA_FIFO_BUSY. If it reads as TRUE, the device is actively + * processing FIFO commands, and no further action is necessary. + * 4. If SVGA_FIFO_BUSY was FALSE, write TRUE to SVGA_REG_SYNC. * - * 3. Write a reason to SVGA_REG_SYNC. This will send an - * asynchronous wakeup to the MKS thread. + * For maximum performance, this procedure should be followed after + * every meaningful batch of commands has been written into the FIFO. + * (Normally when the underlying application signals it's finished a + * meaningful work unit by calling Flush.) */ @@ -1164,9 +1255,6 @@ enum { * Video -- SVGA Video overlay units are supported * Escape -- Escape command is supported * - * XXX: Add longer descriptions for each capability, including a list - * of the new features that each capability provides. - * * SVGA_FIFO_CAP_SCREEN_OBJECT -- * * Provides dynamic multi-screen rendering, for improved Unity and @@ -1279,6 +1367,15 @@ enum { /* + * ScreenDMA Register Values + */ + +#define SVGA_SCREENDMA_REG_UNDEFINED 0 +#define SVGA_SCREENDMA_REG_NOT_PRESENT 1 +#define SVGA_SCREENDMA_REG_PRESENT 2 +#define SVGA_SCREENDMA_REG_MAX 3 + +/* * Video overlay support */ @@ -1665,6 +1762,80 @@ SVGAFifoCmdDefineAlphaCursor; /* + * Provide a new large cursor image, as an AND/XOR mask. + * + * Should only be used for CursorMob functionality + */ + +typedef +#include "vmware_pack_begin.h" +struct { + uint32 hotspotX; + uint32 hotspotY; + uint32 width; + uint32 height; + uint32 andMaskDepth; + uint32 xorMaskDepth; + /* + * Followed by scanline data for AND mask, then XOR mask. + * Each scanline is padded to a 32-bit boundary. + */ +} +#include "vmware_pack_end.h" +SVGAGBColorCursorHeader; + + +/* + * Provide a new large cursor image, in 32-bit BGRA format. + * + * Should only be used for CursorMob functionality + */ + +typedef +#include "vmware_pack_begin.h" +struct { + uint32 hotspotX; + uint32 hotspotY; + uint32 width; + uint32 height; + /* Followed by scanline data */ +} +#include "vmware_pack_end.h" +SVGAGBAlphaCursorHeader; + + /* + * Define the SVGA guest backed cursor types + */ + +typedef enum { + SVGA_COLOR_CURSOR = 0, + SVGA_ALPHA_CURSOR = 1, +} SVGAGBCursorType; + +/* + * Provide a new large cursor image. + * + * Should only be used for CursorMob functionality + */ + +typedef +#include "vmware_pack_begin.h" +struct { + SVGAGBCursorType type; + union { + SVGAGBColorCursorHeader colorHeader; + SVGAGBAlphaCursorHeader alphaHeader; + } header; + uint32 sizeInBytes; + /* + * Followed by the cursor data + */ +} +#include "vmware_pack_end.h" +SVGAGBCursorHeader; + + +/* * SVGA_CMD_UPDATE_VERBOSE -- * * Just like SVGA_CMD_UPDATE, but also provide a per-rectangle @@ -2061,9 +2232,12 @@ SVGAFifoCmdRemapGMR2; #define SVGA_VRAM_MAX_SIZE (128 * 1024 * 1024) #define SVGA_MEMORY_SIZE_MAX (1024 * 1024 * 1024) #define SVGA_FIFO_SIZE_MAX (2 * 1024 * 1024) -#define SVGA_GRAPHICS_MEMORY_KB_MIN (32 * 1024) -#define SVGA_GRAPHICS_MEMORY_KB_MAX (2 * 1024 * 1024) -#define SVGA_GRAPHICS_MEMORY_KB_DEFAULT (256 * 1024) +#define SVGA_GRAPHICS_MEMORY_KB_MIN (32 * 1024) +#define SVGA_GRAPHICS_MEMORY_KB_MAX_2GB (2 * 1024 * 1024) +#define SVGA_GRAPHICS_MEMORY_KB_MAX_3GB (3 * 1024 * 1024) +#define SVGA_GRAPHICS_MEMORY_KB_MAX_4GB (4 * 1024 * 1024) +#define SVGA_GRAPHICS_MEMORY_KB_MAX_8GB (8 * 1024 * 1024) +#define SVGA_GRAPHICS_MEMORY_KB_DEFAULT (256 * 1024) #define SVGA_VRAM_SIZE_W2K (64 * 1024 * 1024) /* 64 MB */ @@ -2086,4 +2260,6 @@ SVGAFifoCmdRemapGMR2; #define SVGA_FIFO_SIZE_GBOBJECTS (256 * 1024) #define SVGA_VRAM_SIZE_GBOBJECTS (4 * 1024 * 1024) +#define SVGA_PCI_REGS_PAGES (1) + #endif diff --git a/drivers/gpu/drm/vmwgfx/device_include/svga_types.h b/drivers/gpu/drm/vmwgfx/device_include/svga_types.h index 350bbc6fab02..beddccee40f6 100644 --- a/drivers/gpu/drm/vmwgfx/device_include/svga_types.h +++ b/drivers/gpu/drm/vmwgfx/device_include/svga_types.h @@ -37,6 +37,7 @@ typedef s8 int8; typedef uint64 PA; typedef uint32 PPN; +typedef uint32 PPN32; typedef uint64 PPN64; typedef bool Bool; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_binding.c b/drivers/gpu/drm/vmwgfx/vmwgfx_binding.c index 66e14e38d5e8..f41550797970 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_binding.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_binding.c @@ -57,9 +57,11 @@ #define VMW_BINDING_RT_BIT 0 #define VMW_BINDING_PS_BIT 1 -#define VMW_BINDING_SO_BIT 2 +#define VMW_BINDING_SO_T_BIT 2 #define VMW_BINDING_VB_BIT 3 -#define VMW_BINDING_NUM_BITS 4 +#define VMW_BINDING_UAV_BIT 4 +#define VMW_BINDING_CS_UAV_BIT 5 +#define VMW_BINDING_NUM_BITS 6 #define VMW_BINDING_PS_SR_BIT 0 @@ -75,6 +77,8 @@ * @vertex_buffers: Vertex buffer bindings. * @index_buffer: Index buffer binding. * @per_shader: Per shader-type bindings. + * @ua_views: UAV bindings. + * @so_state: StreamOutput bindings. * @dirty: Bitmap tracking per binding-type changes that have not yet * been emitted to the device. * @dirty_vb: Bitmap tracking individual vertex buffer binding changes that @@ -95,10 +99,12 @@ struct vmw_ctx_binding_state { struct vmw_ctx_bindinfo_view render_targets[SVGA3D_RT_MAX]; struct vmw_ctx_bindinfo_tex texture_units[SVGA3D_NUM_TEXTURE_UNITS]; struct vmw_ctx_bindinfo_view ds_view; - struct vmw_ctx_bindinfo_so so_targets[SVGA3D_DX_MAX_SOTARGETS]; + struct vmw_ctx_bindinfo_so_target so_targets[SVGA3D_DX_MAX_SOTARGETS]; struct vmw_ctx_bindinfo_vb vertex_buffers[SVGA3D_DX_MAX_VERTEXBUFFERS]; struct vmw_ctx_bindinfo_ib index_buffer; - struct vmw_dx_shader_bindings per_shader[SVGA3D_NUM_SHADERTYPE_DX10]; + struct vmw_dx_shader_bindings per_shader[SVGA3D_NUM_SHADERTYPE]; + struct vmw_ctx_bindinfo_uav ua_views[VMW_MAX_UAV_BIND_TYPE]; + struct vmw_ctx_bindinfo_so so_state; unsigned long dirty; DECLARE_BITMAP(dirty_vb, SVGA3D_DX_MAX_VERTEXBUFFERS); @@ -115,12 +121,16 @@ static int vmw_binding_scrub_texture(struct vmw_ctx_bindinfo *bi, bool rebind); static int vmw_binding_scrub_cb(struct vmw_ctx_bindinfo *bi, bool rebind); static int vmw_binding_scrub_dx_rt(struct vmw_ctx_bindinfo *bi, bool rebind); static int vmw_binding_scrub_sr(struct vmw_ctx_bindinfo *bi, bool rebind); -static int vmw_binding_scrub_so(struct vmw_ctx_bindinfo *bi, bool rebind); +static int vmw_binding_scrub_so_target(struct vmw_ctx_bindinfo *bi, bool rebind); static int vmw_binding_emit_dirty(struct vmw_ctx_binding_state *cbs); static int vmw_binding_scrub_dx_shader(struct vmw_ctx_bindinfo *bi, bool rebind); static int vmw_binding_scrub_ib(struct vmw_ctx_bindinfo *bi, bool rebind); static int vmw_binding_scrub_vb(struct vmw_ctx_bindinfo *bi, bool rebind); +static int vmw_binding_scrub_uav(struct vmw_ctx_bindinfo *bi, bool rebind); +static int vmw_binding_scrub_cs_uav(struct vmw_ctx_bindinfo *bi, bool rebind); +static int vmw_binding_scrub_so(struct vmw_ctx_bindinfo *bi, bool rebind); + static void vmw_binding_build_asserts(void) __attribute__ ((unused)); typedef int (*vmw_scrub_func)(struct vmw_ctx_bindinfo *, bool); @@ -151,6 +161,9 @@ static const size_t vmw_binding_shader_offsets[] = { offsetof(struct vmw_ctx_binding_state, per_shader[0].shader), offsetof(struct vmw_ctx_binding_state, per_shader[1].shader), offsetof(struct vmw_ctx_binding_state, per_shader[2].shader), + offsetof(struct vmw_ctx_binding_state, per_shader[3].shader), + offsetof(struct vmw_ctx_binding_state, per_shader[4].shader), + offsetof(struct vmw_ctx_binding_state, per_shader[5].shader), }; static const size_t vmw_binding_rt_offsets[] = { offsetof(struct vmw_ctx_binding_state, render_targets), @@ -162,6 +175,9 @@ static const size_t vmw_binding_cb_offsets[] = { offsetof(struct vmw_ctx_binding_state, per_shader[0].const_buffers), offsetof(struct vmw_ctx_binding_state, per_shader[1].const_buffers), offsetof(struct vmw_ctx_binding_state, per_shader[2].const_buffers), + offsetof(struct vmw_ctx_binding_state, per_shader[3].const_buffers), + offsetof(struct vmw_ctx_binding_state, per_shader[4].const_buffers), + offsetof(struct vmw_ctx_binding_state, per_shader[5].const_buffers), }; static const size_t vmw_binding_dx_ds_offsets[] = { offsetof(struct vmw_ctx_binding_state, ds_view), @@ -170,8 +186,11 @@ static const size_t vmw_binding_sr_offsets[] = { offsetof(struct vmw_ctx_binding_state, per_shader[0].shader_res), offsetof(struct vmw_ctx_binding_state, per_shader[1].shader_res), offsetof(struct vmw_ctx_binding_state, per_shader[2].shader_res), + offsetof(struct vmw_ctx_binding_state, per_shader[3].shader_res), + offsetof(struct vmw_ctx_binding_state, per_shader[4].shader_res), + offsetof(struct vmw_ctx_binding_state, per_shader[5].shader_res), }; -static const size_t vmw_binding_so_offsets[] = { +static const size_t vmw_binding_so_target_offsets[] = { offsetof(struct vmw_ctx_binding_state, so_targets), }; static const size_t vmw_binding_vb_offsets[] = { @@ -180,6 +199,15 @@ static const size_t vmw_binding_vb_offsets[] = { static const size_t vmw_binding_ib_offsets[] = { offsetof(struct vmw_ctx_binding_state, index_buffer), }; +static const size_t vmw_binding_uav_offsets[] = { + offsetof(struct vmw_ctx_binding_state, ua_views[0].views), +}; +static const size_t vmw_binding_cs_uav_offsets[] = { + offsetof(struct vmw_ctx_binding_state, ua_views[1].views), +}; +static const size_t vmw_binding_so_offsets[] = { + offsetof(struct vmw_ctx_binding_state, so_state), +}; static const struct vmw_binding_info vmw_binding_infos[] = { [vmw_ctx_binding_shader] = { @@ -214,10 +242,10 @@ static const struct vmw_binding_info vmw_binding_infos[] = { .size = sizeof(struct vmw_ctx_bindinfo_view), .offsets = vmw_binding_dx_ds_offsets, .scrub_func = vmw_binding_scrub_dx_rt}, - [vmw_ctx_binding_so] = { - .size = sizeof(struct vmw_ctx_bindinfo_so), - .offsets = vmw_binding_so_offsets, - .scrub_func = vmw_binding_scrub_so}, + [vmw_ctx_binding_so_target] = { + .size = sizeof(struct vmw_ctx_bindinfo_so_target), + .offsets = vmw_binding_so_target_offsets, + .scrub_func = vmw_binding_scrub_so_target}, [vmw_ctx_binding_vb] = { .size = sizeof(struct vmw_ctx_bindinfo_vb), .offsets = vmw_binding_vb_offsets, @@ -226,6 +254,18 @@ static const struct vmw_binding_info vmw_binding_infos[] = { .size = sizeof(struct vmw_ctx_bindinfo_ib), .offsets = vmw_binding_ib_offsets, .scrub_func = vmw_binding_scrub_ib}, + [vmw_ctx_binding_uav] = { + .size = sizeof(struct vmw_ctx_bindinfo_view), + .offsets = vmw_binding_uav_offsets, + .scrub_func = vmw_binding_scrub_uav}, + [vmw_ctx_binding_cs_uav] = { + .size = sizeof(struct vmw_ctx_bindinfo_view), + .offsets = vmw_binding_cs_uav_offsets, + .scrub_func = vmw_binding_scrub_cs_uav}, + [vmw_ctx_binding_so] = { + .size = sizeof(struct vmw_ctx_bindinfo_so), + .offsets = vmw_binding_so_offsets, + .scrub_func = vmw_binding_scrub_so}, }; /** @@ -312,6 +352,18 @@ void vmw_binding_add(struct vmw_ctx_binding_state *cbs, } /** + * vmw_binding_add_uav_index - Add UAV index for tracking. + * @cbs: Pointer to the context binding state tracker. + * @slot: UAV type to which bind this index. + * @index: The splice index to track. + */ +void vmw_binding_add_uav_index(struct vmw_ctx_binding_state *cbs, uint32 slot, + uint32 index) +{ + cbs->ua_views[slot].index = index; +} + +/** * vmw_binding_transfer: Transfer a context binding tracking entry. * * @cbs: Pointer to the persistent context binding state tracker. @@ -450,6 +502,10 @@ void vmw_binding_state_commit(struct vmw_ctx_binding_state *to, vmw_binding_transfer(to, from, entry); vmw_binding_drop(entry); } + + /* Also transfer uav splice indices */ + to->ua_views[0].index = from->ua_views[0].index; + to->ua_views[1].index = from->ua_views[1].index; } /** @@ -828,8 +884,8 @@ static void vmw_collect_so_targets(struct vmw_ctx_binding_state *cbs, const struct vmw_ctx_bindinfo *bi, u32 max_num) { - const struct vmw_ctx_bindinfo_so *biso = - container_of(bi, struct vmw_ctx_bindinfo_so, bi); + const struct vmw_ctx_bindinfo_so_target *biso = + container_of(bi, struct vmw_ctx_bindinfo_so_target, bi); unsigned long i; SVGA3dSoTarget *so_buffer = (SVGA3dSoTarget *) cbs->bind_cmd_buffer; @@ -854,11 +910,11 @@ static void vmw_collect_so_targets(struct vmw_ctx_binding_state *cbs, } /** - * vmw_binding_emit_set_so - Issue delayed streamout binding commands + * vmw_emit_set_so_target - Issue delayed streamout binding commands * * @cbs: Pointer to the context's struct vmw_ctx_binding_state */ -static int vmw_emit_set_so(struct vmw_ctx_binding_state *cbs) +static int vmw_emit_set_so_target(struct vmw_ctx_binding_state *cbs) { const struct vmw_ctx_bindinfo *loc = &cbs->so_targets[0].bi; struct { @@ -1005,6 +1061,66 @@ static int vmw_emit_set_vb(struct vmw_ctx_binding_state *cbs) return 0; } +static int vmw_emit_set_uav(struct vmw_ctx_binding_state *cbs) +{ + const struct vmw_ctx_bindinfo *loc = &cbs->ua_views[0].views[0].bi; + struct { + SVGA3dCmdHeader header; + SVGA3dCmdDXSetUAViews body; + } *cmd; + size_t cmd_size, view_id_size; + const struct vmw_resource *ctx = vmw_cbs_context(cbs); + + vmw_collect_view_ids(cbs, loc, SVGA3D_MAX_UAVIEWS); + view_id_size = cbs->bind_cmd_count*sizeof(uint32); + cmd_size = sizeof(*cmd) + view_id_size; + cmd = VMW_FIFO_RESERVE_DX(ctx->dev_priv, cmd_size, ctx->id); + if (!cmd) + return -ENOMEM; + + cmd->header.id = SVGA_3D_CMD_DX_SET_UA_VIEWS; + cmd->header.size = sizeof(cmd->body) + view_id_size; + + /* Splice index is specified user-space */ + cmd->body.uavSpliceIndex = cbs->ua_views[0].index; + + memcpy(&cmd[1], cbs->bind_cmd_buffer, view_id_size); + + vmw_fifo_commit(ctx->dev_priv, cmd_size); + + return 0; +} + +static int vmw_emit_set_cs_uav(struct vmw_ctx_binding_state *cbs) +{ + const struct vmw_ctx_bindinfo *loc = &cbs->ua_views[1].views[0].bi; + struct { + SVGA3dCmdHeader header; + SVGA3dCmdDXSetCSUAViews body; + } *cmd; + size_t cmd_size, view_id_size; + const struct vmw_resource *ctx = vmw_cbs_context(cbs); + + vmw_collect_view_ids(cbs, loc, SVGA3D_MAX_UAVIEWS); + view_id_size = cbs->bind_cmd_count*sizeof(uint32); + cmd_size = sizeof(*cmd) + view_id_size; + cmd = VMW_FIFO_RESERVE_DX(ctx->dev_priv, cmd_size, ctx->id); + if (!cmd) + return -ENOMEM; + + cmd->header.id = SVGA_3D_CMD_DX_SET_CS_UA_VIEWS; + cmd->header.size = sizeof(cmd->body) + view_id_size; + + /* Start index is specified user-space */ + cmd->body.startIndex = cbs->ua_views[1].index; + + memcpy(&cmd[1], cbs->bind_cmd_buffer, view_id_size); + + vmw_fifo_commit(ctx->dev_priv, cmd_size); + + return 0; +} + /** * vmw_binding_emit_dirty - Issue delayed binding commands * @@ -1030,12 +1146,18 @@ static int vmw_binding_emit_dirty(struct vmw_ctx_binding_state *cbs) case VMW_BINDING_PS_BIT: ret = vmw_binding_emit_dirty_ps(cbs); break; - case VMW_BINDING_SO_BIT: - ret = vmw_emit_set_so(cbs); + case VMW_BINDING_SO_T_BIT: + ret = vmw_emit_set_so_target(cbs); break; case VMW_BINDING_VB_BIT: ret = vmw_emit_set_vb(cbs); break; + case VMW_BINDING_UAV_BIT: + ret = vmw_emit_set_uav(cbs); + break; + case VMW_BINDING_CS_UAV_BIT: + ret = vmw_emit_set_cs_uav(cbs); + break; default: BUG(); } @@ -1089,18 +1211,18 @@ static int vmw_binding_scrub_dx_rt(struct vmw_ctx_bindinfo *bi, bool rebind) } /** - * vmw_binding_scrub_so - Schedule a dx streamoutput buffer binding + * vmw_binding_scrub_so_target - Schedule a dx streamoutput buffer binding * scrub from a context * * @bi: single binding information. * @rebind: Whether to issue a bind instead of scrub command. */ -static int vmw_binding_scrub_so(struct vmw_ctx_bindinfo *bi, bool rebind) +static int vmw_binding_scrub_so_target(struct vmw_ctx_bindinfo *bi, bool rebind) { struct vmw_ctx_binding_state *cbs = vmw_context_binding_state(bi->ctx); - __set_bit(VMW_BINDING_SO_BIT, &cbs->dirty); + __set_bit(VMW_BINDING_SO_T_BIT, &cbs->dirty); return 0; } @@ -1162,6 +1284,49 @@ static int vmw_binding_scrub_ib(struct vmw_ctx_bindinfo *bi, bool rebind) return 0; } +static int vmw_binding_scrub_uav(struct vmw_ctx_bindinfo *bi, bool rebind) +{ + struct vmw_ctx_binding_state *cbs = vmw_context_binding_state(bi->ctx); + + __set_bit(VMW_BINDING_UAV_BIT, &cbs->dirty); + return 0; +} + +static int vmw_binding_scrub_cs_uav(struct vmw_ctx_bindinfo *bi, bool rebind) +{ + struct vmw_ctx_binding_state *cbs = vmw_context_binding_state(bi->ctx); + + __set_bit(VMW_BINDING_CS_UAV_BIT, &cbs->dirty); + return 0; +} + +/** + * vmw_binding_scrub_so - Scrub a streamoutput binding from context. + * @bi: Single binding information. + * @rebind: Whether to issue a bind instead of scrub command. + */ +static int vmw_binding_scrub_so(struct vmw_ctx_bindinfo *bi, bool rebind) +{ + struct vmw_ctx_bindinfo_so *binding = + container_of(bi, typeof(*binding), bi); + struct vmw_private *dev_priv = bi->ctx->dev_priv; + struct { + SVGA3dCmdHeader header; + SVGA3dCmdDXSetStreamOutput body; + } *cmd; + + cmd = VMW_FIFO_RESERVE_DX(dev_priv, sizeof(*cmd), bi->ctx->id); + if (!cmd) + return -ENOMEM; + + cmd->header.id = SVGA_3D_CMD_DX_SET_STREAMOUTPUT; + cmd->header.size = sizeof(cmd->body); + cmd->body.soid = rebind ? bi->res->id : SVGA3D_INVALID_ID; + vmw_fifo_commit(dev_priv, sizeof(*cmd)); + + return 0; +} + /** * vmw_binding_state_alloc - Allocate a struct vmw_ctx_binding_state with * memory accounting. @@ -1248,8 +1413,8 @@ void vmw_binding_state_reset(struct vmw_ctx_binding_state *cbs) * Each time a resource is put on the validation list as the result of a * context binding referencing it, we need to determine whether that resource * will be dirtied (written to by the GPU) as a result of the corresponding - * GPU operation. Currently rendertarget-, depth-stencil-, and - * stream-output-target bindings are capable of dirtying its resource. + * GPU operation. Currently rendertarget-, depth-stencil-, stream-output-target + * and unordered access view bindings are capable of dirtying its resource. * * Return: Whether the binding type dirties the resource its binding points to. */ @@ -1259,11 +1424,13 @@ u32 vmw_binding_dirtying(enum vmw_ctx_binding_type binding_type) [vmw_ctx_binding_rt] = VMW_RES_DIRTY_SET, [vmw_ctx_binding_dx_rt] = VMW_RES_DIRTY_SET, [vmw_ctx_binding_ds] = VMW_RES_DIRTY_SET, - [vmw_ctx_binding_so] = VMW_RES_DIRTY_SET, + [vmw_ctx_binding_so_target] = VMW_RES_DIRTY_SET, + [vmw_ctx_binding_uav] = VMW_RES_DIRTY_SET, + [vmw_ctx_binding_cs_uav] = VMW_RES_DIRTY_SET, }; /* Review this function as new bindings are added. */ - BUILD_BUG_ON(vmw_ctx_binding_max != 11); + BUILD_BUG_ON(vmw_ctx_binding_max != 14); return is_binding_dirtying[binding_type]; } diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_binding.h b/drivers/gpu/drm/vmwgfx/vmwgfx_binding.h index cd9805c045cb..dcb71fd0bb3b 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_binding.h +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_binding.h @@ -33,6 +33,8 @@ #define VMW_MAX_VIEW_BINDINGS 128 +#define VMW_MAX_UAV_BIND_TYPE 2 + struct vmw_private; struct vmw_ctx_binding_state; @@ -48,9 +50,12 @@ enum vmw_ctx_binding_type { vmw_ctx_binding_dx_rt, vmw_ctx_binding_sr, vmw_ctx_binding_ds, - vmw_ctx_binding_so, + vmw_ctx_binding_so_target, vmw_ctx_binding_vb, vmw_ctx_binding_ib, + vmw_ctx_binding_uav, + vmw_ctx_binding_cs_uav, + vmw_ctx_binding_so, vmw_ctx_binding_max }; @@ -128,14 +133,14 @@ struct vmw_ctx_bindinfo_view { }; /** - * struct vmw_ctx_bindinfo_so - StreamOutput binding metadata + * struct vmw_ctx_bindinfo_so_target - StreamOutput binding metadata * * @bi: struct vmw_ctx_bindinfo we derive from. * @offset: Device data used to reconstruct binding command. * @size: Device data used to reconstruct binding command. * @slot: Device data used to reconstruct binding command. */ -struct vmw_ctx_bindinfo_so { +struct vmw_ctx_bindinfo_so_target { struct vmw_ctx_bindinfo bi; uint32 offset; uint32 size; @@ -189,9 +194,31 @@ struct vmw_dx_shader_bindings { unsigned long dirty; }; +/** + * struct vmw_ctx_bindinfo_uav - UAV context binding state. + * @views: UAV view bindings. + * @splice_index: The device splice index set by user-space. + */ +struct vmw_ctx_bindinfo_uav { + struct vmw_ctx_bindinfo_view views[SVGA3D_MAX_UAVIEWS]; + uint32 index; +}; + +/** + * struct vmw_ctx_bindinfo_so - Stream output binding metadata. + * @bi: struct vmw_ctx_bindinfo we derive from. + * @slot: Device data used to reconstruct binding command. + */ +struct vmw_ctx_bindinfo_so { + struct vmw_ctx_bindinfo bi; + uint32 slot; +}; + extern void vmw_binding_add(struct vmw_ctx_binding_state *cbs, const struct vmw_ctx_bindinfo *ci, u32 shader_slot, u32 slot); +extern void vmw_binding_add_uav_index(struct vmw_ctx_binding_state *cbs, + uint32 slot, uint32 splice_index); extern void vmw_binding_state_commit(struct vmw_ctx_binding_state *to, struct vmw_ctx_binding_state *from); diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c index 065015d2a8f6..3b41cf63110a 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c @@ -1241,7 +1241,8 @@ int vmw_cmdbuf_set_pool_size(struct vmw_cmdbuf_man *man, * actually call into the already enabled manager, when * binding the MOB. */ - if (!(dev_priv->capabilities & SVGA_CAP_DX)) + if (!(dev_priv->capabilities & SVGA_CAP_DX) || + !dev_priv->has_mob) return -ENOMEM; ret = ttm_bo_create(&dev_priv->bdev, size, ttm_bo_type_device, diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_context.c b/drivers/gpu/drm/vmwgfx/vmwgfx_context.c index a56c9d802382..61c246335e66 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_context.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_context.c @@ -36,7 +36,7 @@ struct vmw_user_context { struct vmw_resource res; struct vmw_ctx_binding_state *cbs; struct vmw_cmdbuf_res_manager *man; - struct vmw_resource *cotables[SVGA_COTABLE_DX10_MAX]; + struct vmw_resource *cotables[SVGA_COTABLE_MAX]; spinlock_t cotable_lock; struct vmw_buffer_object *dx_query_mob; }; @@ -116,12 +116,15 @@ static const struct vmw_res_func vmw_dx_context_func = { * Context management: */ -static void vmw_context_cotables_unref(struct vmw_user_context *uctx) +static void vmw_context_cotables_unref(struct vmw_private *dev_priv, + struct vmw_user_context *uctx) { struct vmw_resource *res; int i; + u32 cotable_max = has_sm5_context(dev_priv) ? + SVGA_COTABLE_MAX : SVGA_COTABLE_DX10_MAX; - for (i = 0; i < SVGA_COTABLE_DX10_MAX; ++i) { + for (i = 0; i < cotable_max; ++i) { spin_lock(&uctx->cotable_lock); res = uctx->cotables[i]; uctx->cotables[i] = NULL; @@ -155,7 +158,7 @@ static void vmw_hw_context_destroy(struct vmw_resource *res) !dev_priv->query_cid_valid) __vmw_execbuf_release_pinned_bo(dev_priv, NULL); mutex_unlock(&dev_priv->cmdbuf_mutex); - vmw_context_cotables_unref(uctx); + vmw_context_cotables_unref(dev_priv, uctx); return; } @@ -208,7 +211,9 @@ static int vmw_gb_context_init(struct vmw_private *dev_priv, spin_lock_init(&uctx->cotable_lock); if (dx) { - for (i = 0; i < SVGA_COTABLE_DX10_MAX; ++i) { + u32 cotable_max = has_sm5_context(dev_priv) ? + SVGA_COTABLE_MAX : SVGA_COTABLE_DX10_MAX; + for (i = 0; i < cotable_max; ++i) { uctx->cotables[i] = vmw_cotable_alloc(dev_priv, &uctx->res, i); if (IS_ERR(uctx->cotables[i])) { @@ -222,7 +227,7 @@ static int vmw_gb_context_init(struct vmw_private *dev_priv, return 0; out_cotables: - vmw_context_cotables_unref(uctx); + vmw_context_cotables_unref(dev_priv, uctx); out_err: if (res_free) res_free(res); @@ -545,10 +550,12 @@ void vmw_dx_context_scrub_cotables(struct vmw_resource *ctx, { struct vmw_user_context *uctx = container_of(ctx, struct vmw_user_context, res); + u32 cotable_max = has_sm5_context(ctx->dev_priv) ? + SVGA_COTABLE_MAX : SVGA_COTABLE_DX10_MAX; int i; vmw_binding_state_scrub(uctx->cbs); - for (i = 0; i < SVGA_COTABLE_DX10_MAX; ++i) { + for (i = 0; i < cotable_max; ++i) { struct vmw_resource *res; /* Avoid racing with ongoing cotable destruction. */ @@ -731,7 +738,7 @@ static int vmw_context_define(struct drm_device *dev, void *data, }; int ret; - if (!dev_priv->has_dx && dx) { + if (!has_sm4_context(dev_priv) && dx) { VMW_DEBUG_USER("DX contexts not supported by device.\n"); return -EINVAL; } @@ -839,7 +846,10 @@ struct vmw_cmdbuf_res_manager *vmw_context_res_man(struct vmw_resource *ctx) struct vmw_resource *vmw_context_cotable(struct vmw_resource *ctx, SVGACOTableType cotable_type) { - if (cotable_type >= SVGA_COTABLE_DX10_MAX) + u32 cotable_max = has_sm5_context(ctx->dev_priv) ? + SVGA_COTABLE_MAX : SVGA_COTABLE_DX10_MAX; + + if (cotable_type >= cotable_max) return ERR_PTR(-EINVAL); return container_of(ctx, struct vmw_user_context, res)-> diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c b/drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c index 3ca5cf375b01..65e8e7a97724 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c @@ -80,9 +80,10 @@ static const struct vmw_cotable_info co_info[] = { {1, sizeof(SVGACOTableDXDepthStencilEntry), NULL}, {1, sizeof(SVGACOTableDXRasterizerStateEntry), NULL}, {1, sizeof(SVGACOTableDXSamplerEntry), NULL}, - {1, sizeof(SVGACOTableDXStreamOutputEntry), NULL}, + {1, sizeof(SVGACOTableDXStreamOutputEntry), &vmw_dx_streamoutput_cotable_list_scrub}, {1, sizeof(SVGACOTableDXQueryEntry), NULL}, - {1, sizeof(SVGACOTableDXShaderEntry), &vmw_dx_shader_cotable_list_scrub} + {1, sizeof(SVGACOTableDXShaderEntry), &vmw_dx_shader_cotable_list_scrub}, + {1, sizeof(SVGACOTableDXUAViewEntry), &vmw_view_cotable_list_destroy} }; /* @@ -102,6 +103,7 @@ const SVGACOTableType vmw_cotable_scrub_order[] = { SVGA_COTABLE_SAMPLER, SVGA_COTABLE_STREAMOUTPUT, SVGA_COTABLE_DXQUERY, + SVGA_COTABLE_UAVIEW, }; static int vmw_cotable_bind(struct vmw_resource *res, diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c index 827458f49112..c2247a893ed4 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c @@ -29,6 +29,7 @@ #include <linux/dma-mapping.h> #include <linux/module.h> #include <linux/pci.h> +#include <linux/mem_encrypt.h> #include <drm/drm_drv.h> #include <drm/drm_ioctl.h> @@ -289,6 +290,8 @@ static void vmw_print_capabilities2(uint32_t capabilities2) DRM_INFO(" Grow oTable.\n"); if (capabilities2 & SVGA_CAP2_INTRA_SURFACE_COPY) DRM_INFO(" IntraSurface copy.\n"); + if (capabilities2 & SVGA_CAP2_DX3) + DRM_INFO(" DX3.\n"); } static void vmw_print_capabilities(uint32_t capabilities) @@ -448,7 +451,7 @@ static int vmw_request_device(struct vmw_private *dev_priv) dev_priv->cman = vmw_cmdbuf_man_create(dev_priv); if (IS_ERR(dev_priv->cman)) { dev_priv->cman = NULL; - dev_priv->has_dx = false; + dev_priv->sm_type = VMW_SM_LEGACY; } ret = vmw_request_device_late(dev_priv); @@ -575,6 +578,10 @@ static int vmw_dma_select_mode(struct vmw_private *dev_priv) [vmw_dma_map_populate] = "Caching DMA mappings.", [vmw_dma_map_bind] = "Giving up DMA mappings early."}; + /* TTM currently doesn't fully support SEV encryption. */ + if (mem_encrypt_active()) + return -EINVAL; + if (vmw_force_coherent) dev_priv->map_mode = vmw_dma_alloc_coherent; else if (vmw_restrict_iommu) @@ -682,8 +689,10 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset) ret = vmw_dma_select_mode(dev_priv); if (unlikely(ret != 0)) { - DRM_INFO("Restricting capabilities due to IOMMU setup.\n"); + DRM_INFO("Restricting capabilities since DMA not available.\n"); refuse_dma = true; + if (dev_priv->capabilities & SVGA_CAP_GBOBJECTS) + DRM_INFO("Disabling 3D acceleration.\n"); } dev_priv->vram_size = vmw_read(dev_priv, SVGA_REG_VRAM_SIZE); @@ -711,9 +720,15 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset) dev_priv->max_mob_pages = 0; dev_priv->max_mob_size = 0; if (dev_priv->capabilities & SVGA_CAP_GBOBJECTS) { - uint64_t mem_size = - vmw_read(dev_priv, - SVGA_REG_SUGGESTED_GBOBJECT_MEM_SIZE_KB); + uint64_t mem_size; + + if (dev_priv->capabilities2 & SVGA_CAP2_GB_MEMSIZE_2) + mem_size = vmw_read(dev_priv, + SVGA_REG_GBOBJECT_MEM_SIZE_KB); + else + mem_size = + vmw_read(dev_priv, + SVGA_REG_SUGGESTED_GBOBJECT_MEM_SIZE_KB); /* * Workaround for low memory 2D VMs to compensate for the @@ -866,7 +881,7 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset) dev_priv->has_gmr = false; } - if (dev_priv->capabilities & SVGA_CAP_GBOBJECTS) { + if (dev_priv->capabilities & SVGA_CAP_GBOBJECTS && !refuse_dma) { dev_priv->has_mob = true; if (ttm_bo_init_mm(&dev_priv->bdev, VMW_PL_MOB, VMW_PL_MOB) != 0) { @@ -876,14 +891,32 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset) } } - if (dev_priv->has_mob) { + if (dev_priv->has_mob && (dev_priv->capabilities & SVGA_CAP_DX)) { spin_lock(&dev_priv->cap_lock); vmw_write(dev_priv, SVGA_REG_DEV_CAP, SVGA3D_DEVCAP_DXCONTEXT); - dev_priv->has_dx = !!vmw_read(dev_priv, SVGA_REG_DEV_CAP); + if (vmw_read(dev_priv, SVGA_REG_DEV_CAP)) + dev_priv->sm_type = VMW_SM_4; spin_unlock(&dev_priv->cap_lock); } vmw_validation_mem_init_ttm(dev_priv, VMWGFX_VALIDATION_MEM_GRAN); + + /* SVGA_CAP2_DX2 (DefineGBSurface_v3) is needed for SM4_1 support */ + if (has_sm4_context(dev_priv) && + (dev_priv->capabilities2 & SVGA_CAP2_DX2)) { + vmw_write(dev_priv, SVGA_REG_DEV_CAP, SVGA3D_DEVCAP_SM41); + + if (vmw_read(dev_priv, SVGA_REG_DEV_CAP)) + dev_priv->sm_type = VMW_SM_4_1; + + if (has_sm4_1_context(dev_priv) && + (dev_priv->capabilities2 & SVGA_CAP2_DX3)) { + vmw_write(dev_priv, SVGA_REG_DEV_CAP, SVGA3D_DEVCAP_SM5); + if (vmw_read(dev_priv, SVGA_REG_DEV_CAP)) + dev_priv->sm_type = VMW_SM_5; + } + } + ret = vmw_kms_init(dev_priv); if (unlikely(ret != 0)) goto out_no_kms; @@ -893,23 +926,14 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset) if (ret) goto out_no_fifo; - if (dev_priv->has_dx) { - /* - * SVGA_CAP2_DX2 (DefineGBSurface_v3) is needed for SM4_1 - * support - */ - if ((dev_priv->capabilities2 & SVGA_CAP2_DX2) != 0) { - vmw_write(dev_priv, SVGA_REG_DEV_CAP, - SVGA3D_DEVCAP_SM41); - dev_priv->has_sm4_1 = vmw_read(dev_priv, - SVGA_REG_DEV_CAP); - } - } - - DRM_INFO("DX: %s\n", dev_priv->has_dx ? "yes." : "no."); DRM_INFO("Atomic: %s\n", (dev->driver->driver_features & DRIVER_ATOMIC) ? "yes." : "no."); - DRM_INFO("SM4_1: %s\n", dev_priv->has_sm4_1 ? "yes." : "no."); + if (dev_priv->sm_type == VMW_SM_5) + DRM_INFO("SM5 support available.\n"); + if (dev_priv->sm_type == VMW_SM_4_1) + DRM_INFO("SM4_1 support available.\n"); + if (dev_priv->sm_type == VMW_SM_4) + DRM_INFO("SM4 support available.\n"); snprintf(host_log, sizeof(host_log), "vmwgfx: %s-%s", VMWGFX_REPO, VMWGFX_GIT_VERSION); @@ -1223,6 +1247,18 @@ static void vmw_remove(struct pci_dev *pdev) pci_disable_device(pdev); } +static unsigned long +vmw_get_unmapped_area(struct file *file, unsigned long uaddr, + unsigned long len, unsigned long pgoff, + unsigned long flags) +{ + struct drm_file *file_priv = file->private_data; + struct vmw_private *dev_priv = vmw_priv(file_priv->minor->dev); + + return drm_get_unmapped_area(file, uaddr, len, pgoff, flags, + &dev_priv->vma_manager); +} + static int vmwgfx_pm_notifier(struct notifier_block *nb, unsigned long val, void *ptr) { @@ -1394,14 +1430,12 @@ static const struct file_operations vmwgfx_driver_fops = { .compat_ioctl = vmw_compat_ioctl, #endif .llseek = noop_llseek, + .get_unmapped_area = vmw_get_unmapped_area, }; static struct drm_driver driver = { .driver_features = DRIVER_MODESET | DRIVER_RENDER | DRIVER_ATOMIC, - .get_vblank_counter = vmw_get_vblank_counter, - .enable_vblank = vmw_enable_vblank, - .disable_vblank = vmw_disable_vblank, .ioctls = vmw_ioctls, .num_ioctls = ARRAY_SIZE(vmw_ioctls), .master_set = vmw_master_set, diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h index 86b69397d166..8cdcd6e5f9e1 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h @@ -58,7 +58,7 @@ #define VMWGFX_DRIVER_NAME "vmwgfx" #define VMWGFX_DRIVER_DATE "20200114" #define VMWGFX_DRIVER_MAJOR 2 -#define VMWGFX_DRIVER_MINOR 17 +#define VMWGFX_DRIVER_MINOR 18 #define VMWGFX_DRIVER_PATCHLEVEL 0 #define VMWGFX_FIFO_STATIC_SIZE (1024*1024) #define VMWGFX_MAX_RELOCATIONS 2048 @@ -202,6 +202,7 @@ enum vmw_res_type { vmw_res_dx_context, vmw_res_cotable, vmw_res_view, + vmw_res_streamoutput, vmw_res_max }; @@ -210,7 +211,8 @@ enum vmw_res_type { */ enum vmw_cmdbuf_res_type { vmw_cmdbuf_res_shader, - vmw_cmdbuf_res_view + vmw_cmdbuf_res_view, + vmw_cmdbuf_res_streamoutput }; struct vmw_cmdbuf_res_manager; @@ -223,24 +225,58 @@ struct vmw_cursor_snooper { struct vmw_framebuffer; struct vmw_surface_offset; -struct vmw_surface { - struct vmw_resource res; - SVGA3dSurfaceAllFlags flags; - uint32_t format; - uint32_t mip_levels[DRM_VMW_MAX_SURFACE_FACES]; +/** + * struct vmw_surface_metadata - Metadata describing a surface. + * + * @flags: Device flags. + * @format: Surface SVGA3D_x format. + * @mip_levels: Mip level for each face. For GB first index is used only. + * @multisample_count: Sample count. + * @multisample_pattern: Sample patterns. + * @quality_level: Quality level. + * @autogen_filter: Filter for automatically generated mipmaps. + * @array_size: Number of array elements for a 1D/2D texture. For cubemap + texture number of faces * array_size. This should be 0 for pre + SM4 device. + * @buffer_byte_stride: Buffer byte stride. + * @num_sizes: Size of @sizes. For GB surface this should always be 1. + * @base_size: Surface dimension. + * @sizes: Array representing mip sizes. Legacy only. + * @scanout: Whether this surface will be used for scanout. + * + * This tracks metadata for both legacy and guest backed surface. + */ +struct vmw_surface_metadata { + u64 flags; + u32 format; + u32 mip_levels[DRM_VMW_MAX_SURFACE_FACES]; + u32 multisample_count; + u32 multisample_pattern; + u32 quality_level; + u32 autogen_filter; + u32 array_size; + u32 num_sizes; + u32 buffer_byte_stride; struct drm_vmw_size base_size; struct drm_vmw_size *sizes; - uint32_t num_sizes; bool scanout; - uint32_t array_size; - /* TODO so far just a extra pointer */ +}; + +/** + * struct vmw_surface: Resource structure for a surface. + * + * @res: The base resource for this surface. + * @metadata: Metadata for this surface resource. + * @snooper: Cursor data. Legacy surface only. + * @offsets: Legacy surface only. + * @view_list: List of views bound to this surface. + */ +struct vmw_surface { + struct vmw_resource res; + struct vmw_surface_metadata metadata; struct vmw_cursor_snooper snooper; struct vmw_surface_offset *offsets; - SVGA3dTextureFilter autogen_filter; - uint32_t multisample_count; struct list_head view_list; - SVGA3dMSPattern multisample_pattern; - SVGA3dMSQualityLevel quality_level; }; struct vmw_marker_queue { @@ -441,6 +477,22 @@ enum { VMW_IRQTHREAD_MAX }; +/** + * enum vmw_sm_type - Graphics context capability supported by device. + * @VMW_SM_LEGACY: Pre DX context. + * @VMW_SM_4: Context support upto SM4. + * @VMW_SM_4_1: Context support upto SM4_1. + * @VMW_SM_5: Context support up to SM5. + * @VMW_SM_MAX: Should be the last. + */ +enum vmw_sm_type { + VMW_SM_LEGACY = 0, + VMW_SM_4, + VMW_SM_4_1, + VMW_SM_5, + VMW_SM_MAX +}; + struct vmw_private { struct ttm_bo_device bdev; @@ -475,22 +527,9 @@ struct vmw_private { bool has_mob; spinlock_t hw_lock; spinlock_t cap_lock; - bool has_dx; bool assume_16bpp; - bool has_sm4_1; - - /* - * VGA registers. - */ - - struct vmw_vga_topology_state vga_save[VMWGFX_MAX_DISPLAYS]; - uint32_t vga_width; - uint32_t vga_height; - uint32_t vga_bpp; - uint32_t vga_bpl; - uint32_t vga_pitchlock; - uint32_t num_displays; + enum vmw_sm_type sm_type; /* * Framebuffer info. @@ -661,6 +700,39 @@ static inline uint32_t vmw_read(struct vmw_private *dev_priv, return val; } +/** + * has_sm4_context - Does the device support SM4 context. + * @dev_priv: Device private. + * + * Return: Bool value if device support SM4 context or not. + */ +static inline bool has_sm4_context(const struct vmw_private *dev_priv) +{ + return (dev_priv->sm_type >= VMW_SM_4); +} + +/** + * has_sm4_1_context - Does the device support SM4_1 context. + * @dev_priv: Device private. + * + * Return: Bool value if device support SM4_1 context or not. + */ +static inline bool has_sm4_1_context(const struct vmw_private *dev_priv) +{ + return (dev_priv->sm_type >= VMW_SM_4_1); +} + +/** + * has_sm5_context - Does the device support SM5 context. + * @dev_priv: Device private. + * + * Return: Bool value if device support SM5 context or not. + */ +static inline bool has_sm5_context(const struct vmw_private *dev_priv) +{ + return (dev_priv->sm_type >= VMW_SM_5); +} + extern void vmw_svga_enable(struct vmw_private *dev_priv); extern void vmw_svga_disable(struct vmw_private *dev_priv); @@ -900,7 +972,6 @@ extern void vmw_fifo_commit(struct vmw_private *dev_priv, uint32_t bytes); extern void vmw_fifo_commit_flush(struct vmw_private *dev_priv, uint32_t bytes); extern int vmw_fifo_send_fence(struct vmw_private *dev_priv, uint32_t *seqno); -extern void vmw_fifo_ping_host_locked(struct vmw_private *, uint32_t reason); extern void vmw_fifo_ping_host(struct vmw_private *dev_priv, uint32_t reason); extern bool vmw_fifo_have_3d(struct vmw_private *dev_priv); extern bool vmw_fifo_have_pitchlock(struct vmw_private *dev_priv); @@ -929,6 +1000,7 @@ extern int vmw_mmap(struct file *filp, struct vm_area_struct *vma); extern void vmw_validation_mem_init_ttm(struct vmw_private *dev_priv, size_t gran); + /** * TTM buffer object driver - vmwgfx_ttm_buffer.c */ @@ -947,7 +1019,6 @@ extern struct ttm_placement vmw_mob_placement; extern struct ttm_placement vmw_mob_ne_placement; extern struct ttm_placement vmw_nonfixed_placement; extern struct ttm_bo_driver vmw_bo_driver; -extern int vmw_dma_quiescent(struct drm_device *dev); extern int vmw_bo_map_dma(struct ttm_buffer_object *bo); extern void vmw_bo_unmap_dma(struct ttm_buffer_object *bo); extern const struct vmw_sg_table * @@ -1085,8 +1156,6 @@ int vmw_fb_on(struct vmw_private *vmw_priv); int vmw_kms_init(struct vmw_private *dev_priv); int vmw_kms_close(struct vmw_private *dev_priv); -int vmw_kms_save_vga(struct vmw_private *vmw_priv); -int vmw_kms_restore_vga(struct vmw_private *vmw_priv); int vmw_kms_cursor_bypass_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv); void vmw_kms_cursor_post_execbuf(struct vmw_private *dev_priv); @@ -1100,9 +1169,9 @@ int vmw_kms_write_svga(struct vmw_private *vmw_priv, bool vmw_kms_validate_mode_vram(struct vmw_private *dev_priv, uint32_t pitch, uint32_t height); -u32 vmw_get_vblank_counter(struct drm_device *dev, unsigned int pipe); -int vmw_enable_vblank(struct drm_device *dev, unsigned int pipe); -void vmw_disable_vblank(struct drm_device *dev, unsigned int pipe); +u32 vmw_get_vblank_counter(struct drm_crtc *crtc); +int vmw_enable_vblank(struct drm_crtc *crtc); +void vmw_disable_vblank(struct drm_crtc *crtc); int vmw_kms_present(struct vmw_private *dev_priv, struct drm_file *file_priv, struct vmw_framebuffer *vfb, @@ -1139,7 +1208,6 @@ int vmw_overlay_init(struct vmw_private *dev_priv); int vmw_overlay_close(struct vmw_private *dev_priv); int vmw_overlay_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv); -int vmw_overlay_stop_all(struct vmw_private *dev_priv); int vmw_overlay_resume_all(struct vmw_private *dev_priv); int vmw_overlay_pause_all(struct vmw_private *dev_priv); int vmw_overlay_claim(struct vmw_private *dev_priv, uint32_t *out); @@ -1186,10 +1254,6 @@ extern void vmw_otables_takedown(struct vmw_private *dev_priv); extern const struct vmw_user_resource_conv *user_context_converter; -extern int vmw_context_check(struct vmw_private *dev_priv, - struct ttm_object_file *tfile, - int id, - struct vmw_resource **p_res); extern int vmw_context_define_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv); extern int vmw_extended_context_define_ioctl(struct drm_device *dev, void *data, @@ -1219,7 +1283,6 @@ vmw_context_get_dx_query_mob(struct vmw_resource *ctx_res); extern const struct vmw_user_resource_conv *user_surface_converter; -extern void vmw_surface_res_free(struct vmw_resource *res); extern int vmw_surface_destroy_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv); extern int vmw_surface_define_ioctl(struct drm_device *dev, void *data, @@ -1230,11 +1293,6 @@ extern int vmw_gb_surface_define_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv); extern int vmw_gb_surface_reference_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv); -extern int vmw_surface_check(struct vmw_private *dev_priv, - struct ttm_object_file *tfile, - uint32_t handle, int *id); -extern int vmw_surface_validate(struct vmw_private *dev_priv, - struct vmw_surface *srf); int vmw_surface_gb_priv_define(struct drm_device *dev, uint32_t user_accounting_size, SVGA3dSurfaceAllFlags svga3d_flags, @@ -1254,6 +1312,11 @@ extern int vmw_gb_surface_reference_ext_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv); +int vmw_gb_surface_define(struct vmw_private *dev_priv, + uint32_t user_accounting_size, + const struct vmw_surface_metadata *req, + struct vmw_surface **srf_out); + /* * Shader management - vmwgfx_shader.c */ @@ -1287,6 +1350,24 @@ vmw_shader_lookup(struct vmw_cmdbuf_res_manager *man, u32 user_key, SVGA3dShaderType shader_type); /* + * Streamoutput management + */ +struct vmw_resource * +vmw_dx_streamoutput_lookup(struct vmw_cmdbuf_res_manager *man, + u32 user_key); +int vmw_dx_streamoutput_add(struct vmw_cmdbuf_res_manager *man, + struct vmw_resource *ctx, + SVGA3dStreamOutputId user_key, + struct list_head *list); +void vmw_dx_streamoutput_set_size(struct vmw_resource *res, u32 size); +int vmw_dx_streamoutput_remove(struct vmw_cmdbuf_res_manager *man, + SVGA3dStreamOutputId user_key, + struct list_head *list); +void vmw_dx_streamoutput_cotable_list_scrub(struct vmw_private *dev_priv, + struct list_head *list, + bool readback); + +/* * Command buffer managed resources - vmwgfx_cmdbuf_res.c */ @@ -1430,6 +1511,17 @@ void vmw_bo_dirty_unmap(struct vmw_buffer_object *vbo, pgoff_t start, pgoff_t end); vm_fault_t vmw_bo_vm_fault(struct vm_fault *vmf); vm_fault_t vmw_bo_vm_mkwrite(struct vm_fault *vmf); +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +vm_fault_t vmw_bo_vm_huge_fault(struct vm_fault *vmf, + enum page_entry_size pe_size); +#endif + +/* Transparent hugepage support - vmwgfx_thp.c */ +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +extern const struct ttm_mem_type_manager_func vmw_thp_func; +#else +#define vmw_thp_func ttm_bo_manager_func +#endif /** * VMW_DEBUG_KMS - Debug output for kernel mode-setting diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c index 73489a45decb..367d5b87ee6a 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c @@ -459,10 +459,13 @@ static int vmw_resource_context_res_add(struct vmw_private *dev_priv, int ret = 0; struct vmw_resource *res; u32 i; + u32 cotable_max = has_sm5_context(ctx->dev_priv) ? + SVGA_COTABLE_MAX : SVGA_COTABLE_DX10_MAX; /* Add all cotables to the validation list. */ - if (dev_priv->has_dx && vmw_res_type(ctx) == vmw_res_dx_context) { - for (i = 0; i < SVGA_COTABLE_DX10_MAX; ++i) { + if (has_sm4_context(dev_priv) && + vmw_res_type(ctx) == vmw_res_dx_context) { + for (i = 0; i < cotable_max; ++i) { res = vmw_context_cotable(ctx, i); if (IS_ERR(res)) continue; @@ -489,7 +492,8 @@ static int vmw_resource_context_res_add(struct vmw_private *dev_priv, break; } - if (dev_priv->has_dx && vmw_res_type(ctx) == vmw_res_dx_context) { + if (has_sm4_context(dev_priv) && + vmw_res_type(ctx) == vmw_res_dx_context) { struct vmw_buffer_object *dx_query_mob; dx_query_mob = vmw_context_get_dx_query_mob(ctx); @@ -2116,6 +2120,9 @@ vmw_cmd_dx_set_single_constant_buffer(struct vmw_private *dev_priv, SVGA3dCmdHeader *header) { VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXSetSingleConstantBuffer); + SVGA3dShaderType max_shader_num = has_sm5_context(dev_priv) ? + SVGA3D_NUM_SHADERTYPE : SVGA3D_NUM_SHADERTYPE_DX10; + struct vmw_resource *res = NULL; struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context); struct vmw_ctx_bindinfo_cb binding; @@ -2139,7 +2146,7 @@ vmw_cmd_dx_set_single_constant_buffer(struct vmw_private *dev_priv, binding.size = cmd->body.sizeInBytes; binding.slot = cmd->body.slot; - if (binding.shader_slot >= SVGA3D_NUM_SHADERTYPE_DX10 || + if (binding.shader_slot >= max_shader_num || binding.slot >= SVGA3D_DX_MAX_CONSTBUFFERS) { VMW_DEBUG_USER("Illegal const buffer shader %u slot %u.\n", (unsigned int) cmd->body.type, @@ -2167,12 +2174,15 @@ static int vmw_cmd_dx_set_shader_res(struct vmw_private *dev_priv, { VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXSetShaderResources) = container_of(header, typeof(*cmd), header); + SVGA3dShaderType max_allowed = has_sm5_context(dev_priv) ? + SVGA3D_SHADERTYPE_MAX : SVGA3D_SHADERTYPE_DX10_MAX; + u32 num_sr_view = (cmd->header.size - sizeof(cmd->body)) / sizeof(SVGA3dShaderResourceViewId); if ((u64) cmd->body.startView + (u64) num_sr_view > (u64) SVGA3D_DX_MAX_SRVIEWS || - cmd->body.type >= SVGA3D_SHADERTYPE_DX10_MAX) { + cmd->body.type >= max_allowed) { VMW_DEBUG_USER("Invalid shader binding.\n"); return -EINVAL; } @@ -2196,6 +2206,8 @@ static int vmw_cmd_dx_set_shader(struct vmw_private *dev_priv, SVGA3dCmdHeader *header) { VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXSetShader); + SVGA3dShaderType max_allowed = has_sm5_context(dev_priv) ? + SVGA3D_SHADERTYPE_MAX : SVGA3D_SHADERTYPE_DX10_MAX; struct vmw_resource *res = NULL; struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context); struct vmw_ctx_bindinfo_shader binding; @@ -2206,7 +2218,7 @@ static int vmw_cmd_dx_set_shader(struct vmw_private *dev_priv, cmd = container_of(header, typeof(*cmd), header); - if (cmd->body.type >= SVGA3D_SHADERTYPE_DX10_MAX || + if (cmd->body.type >= max_allowed || cmd->body.type < SVGA3D_SHADERTYPE_MIN) { VMW_DEBUG_USER("Illegal shader type %u.\n", (unsigned int) cmd->body.type); @@ -2467,7 +2479,7 @@ static int vmw_cmd_dx_set_so_targets(struct vmw_private *dev_priv, SVGA3dCmdHeader *header) { struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context); - struct vmw_ctx_bindinfo_so binding; + struct vmw_ctx_bindinfo_so_target binding; struct vmw_resource *res; struct { SVGA3dCmdHeader header; @@ -2497,7 +2509,7 @@ static int vmw_cmd_dx_set_so_targets(struct vmw_private *dev_priv, binding.bi.ctx = ctx_node->ctx; binding.bi.res = res; - binding.bi.bt = vmw_ctx_binding_so, + binding.bi.bt = vmw_ctx_binding_so_target, binding.offset = cmd->targets[i].offset; binding.size = cmd->targets[i].sizeInBytes; binding.slot = i; @@ -2804,6 +2816,352 @@ static int vmw_cmd_intra_surface_copy(struct vmw_private *dev_priv, &cmd->body.surface.sid, NULL); } +static int vmw_cmd_sm5(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + SVGA3dCmdHeader *header) +{ + if (!has_sm5_context(dev_priv)) + return -EINVAL; + + return 0; +} + +static int vmw_cmd_sm5_view_define(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + SVGA3dCmdHeader *header) +{ + if (!has_sm5_context(dev_priv)) + return -EINVAL; + + return vmw_cmd_dx_view_define(dev_priv, sw_context, header); +} + +static int vmw_cmd_sm5_view_remove(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + SVGA3dCmdHeader *header) +{ + if (!has_sm5_context(dev_priv)) + return -EINVAL; + + return vmw_cmd_dx_view_remove(dev_priv, sw_context, header); +} + +static int vmw_cmd_clear_uav_uint(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + SVGA3dCmdHeader *header) +{ + struct { + SVGA3dCmdHeader header; + SVGA3dCmdDXClearUAViewUint body; + } *cmd = container_of(header, typeof(*cmd), header); + struct vmw_resource *ret; + + if (!has_sm5_context(dev_priv)) + return -EINVAL; + + ret = vmw_view_id_val_add(sw_context, vmw_view_ua, + cmd->body.uaViewId); + + return PTR_ERR_OR_ZERO(ret); +} + +static int vmw_cmd_clear_uav_float(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + SVGA3dCmdHeader *header) +{ + struct { + SVGA3dCmdHeader header; + SVGA3dCmdDXClearUAViewFloat body; + } *cmd = container_of(header, typeof(*cmd), header); + struct vmw_resource *ret; + + if (!has_sm5_context(dev_priv)) + return -EINVAL; + + ret = vmw_view_id_val_add(sw_context, vmw_view_ua, + cmd->body.uaViewId); + + return PTR_ERR_OR_ZERO(ret); +} + +static int vmw_cmd_set_uav(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + SVGA3dCmdHeader *header) +{ + struct { + SVGA3dCmdHeader header; + SVGA3dCmdDXSetUAViews body; + } *cmd = container_of(header, typeof(*cmd), header); + u32 num_uav = (cmd->header.size - sizeof(cmd->body)) / + sizeof(SVGA3dUAViewId); + int ret; + + if (!has_sm5_context(dev_priv)) + return -EINVAL; + + if (num_uav > SVGA3D_MAX_UAVIEWS) { + VMW_DEBUG_USER("Invalid UAV binding.\n"); + return -EINVAL; + } + + ret = vmw_view_bindings_add(sw_context, vmw_view_ua, + vmw_ctx_binding_uav, 0, (void *)&cmd[1], + num_uav, 0); + if (ret) + return ret; + + vmw_binding_add_uav_index(sw_context->dx_ctx_node->staged, 0, + cmd->body.uavSpliceIndex); + + return ret; +} + +static int vmw_cmd_set_cs_uav(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + SVGA3dCmdHeader *header) +{ + struct { + SVGA3dCmdHeader header; + SVGA3dCmdDXSetCSUAViews body; + } *cmd = container_of(header, typeof(*cmd), header); + u32 num_uav = (cmd->header.size - sizeof(cmd->body)) / + sizeof(SVGA3dUAViewId); + int ret; + + if (!has_sm5_context(dev_priv)) + return -EINVAL; + + if (num_uav > SVGA3D_MAX_UAVIEWS) { + VMW_DEBUG_USER("Invalid UAV binding.\n"); + return -EINVAL; + } + + ret = vmw_view_bindings_add(sw_context, vmw_view_ua, + vmw_ctx_binding_cs_uav, 0, (void *)&cmd[1], + num_uav, 0); + if (ret) + return ret; + + vmw_binding_add_uav_index(sw_context->dx_ctx_node->staged, 1, + cmd->body.startIndex); + + return ret; +} + +static int vmw_cmd_dx_define_streamoutput(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + SVGA3dCmdHeader *header) +{ + struct vmw_ctx_validation_info *ctx_node = sw_context->dx_ctx_node; + struct vmw_resource *res; + struct { + SVGA3dCmdHeader header; + SVGA3dCmdDXDefineStreamOutputWithMob body; + } *cmd = container_of(header, typeof(*cmd), header); + int ret; + + if (!has_sm5_context(dev_priv)) + return -EINVAL; + + if (!ctx_node) { + DRM_ERROR("DX Context not set.\n"); + return -EINVAL; + } + + res = vmw_context_cotable(ctx_node->ctx, SVGA_COTABLE_STREAMOUTPUT); + ret = vmw_cotable_notify(res, cmd->body.soid); + if (ret) + return ret; + + return vmw_dx_streamoutput_add(sw_context->man, ctx_node->ctx, + cmd->body.soid, + &sw_context->staged_cmd_res); +} + +static int vmw_cmd_dx_destroy_streamoutput(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + SVGA3dCmdHeader *header) +{ + struct vmw_ctx_validation_info *ctx_node = sw_context->dx_ctx_node; + struct vmw_resource *res; + struct { + SVGA3dCmdHeader header; + SVGA3dCmdDXDestroyStreamOutput body; + } *cmd = container_of(header, typeof(*cmd), header); + + if (!ctx_node) { + DRM_ERROR("DX Context not set.\n"); + return -EINVAL; + } + + /* + * When device does not support SM5 then streamoutput with mob command is + * not available to user-space. Simply return in this case. + */ + if (!has_sm5_context(dev_priv)) + return 0; + + /* + * With SM5 capable device if lookup fails then user-space probably used + * old streamoutput define command. Return without an error. + */ + res = vmw_dx_streamoutput_lookup(vmw_context_res_man(ctx_node->ctx), + cmd->body.soid); + if (IS_ERR(res)) + return 0; + + return vmw_dx_streamoutput_remove(sw_context->man, cmd->body.soid, + &sw_context->staged_cmd_res); +} + +static int vmw_cmd_dx_bind_streamoutput(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + SVGA3dCmdHeader *header) +{ + struct vmw_ctx_validation_info *ctx_node = sw_context->dx_ctx_node; + struct vmw_resource *res; + struct { + SVGA3dCmdHeader header; + SVGA3dCmdDXBindStreamOutput body; + } *cmd = container_of(header, typeof(*cmd), header); + int ret; + + if (!has_sm5_context(dev_priv)) + return -EINVAL; + + if (!ctx_node) { + DRM_ERROR("DX Context not set.\n"); + return -EINVAL; + } + + res = vmw_dx_streamoutput_lookup(vmw_context_res_man(ctx_node->ctx), + cmd->body.soid); + if (IS_ERR(res)) { + DRM_ERROR("Cound not find streamoutput to bind.\n"); + return PTR_ERR(res); + } + + vmw_dx_streamoutput_set_size(res, cmd->body.sizeInBytes); + + ret = vmw_execbuf_res_noctx_val_add(sw_context, res, + VMW_RES_DIRTY_NONE); + if (ret) { + DRM_ERROR("Error creating resource validation node.\n"); + return ret; + } + + return vmw_cmd_res_switch_backup(dev_priv, sw_context, res, + &cmd->body.mobid, + cmd->body.offsetInBytes); +} + +static int vmw_cmd_dx_set_streamoutput(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + SVGA3dCmdHeader *header) +{ + struct vmw_ctx_validation_info *ctx_node = sw_context->dx_ctx_node; + struct vmw_resource *res; + struct vmw_ctx_bindinfo_so binding; + struct { + SVGA3dCmdHeader header; + SVGA3dCmdDXSetStreamOutput body; + } *cmd = container_of(header, typeof(*cmd), header); + int ret; + + if (!ctx_node) { + DRM_ERROR("DX Context not set.\n"); + return -EINVAL; + } + + if (cmd->body.soid == SVGA3D_INVALID_ID) + return 0; + + /* + * When device does not support SM5 then streamoutput with mob command is + * not available to user-space. Simply return in this case. + */ + if (!has_sm5_context(dev_priv)) + return 0; + + /* + * With SM5 capable device if lookup fails then user-space probably used + * old streamoutput define command. Return without an error. + */ + res = vmw_dx_streamoutput_lookup(vmw_context_res_man(ctx_node->ctx), + cmd->body.soid); + if (IS_ERR(res)) { + return 0; + } + + ret = vmw_execbuf_res_noctx_val_add(sw_context, res, + VMW_RES_DIRTY_NONE); + if (ret) { + DRM_ERROR("Error creating resource validation node.\n"); + return ret; + } + + binding.bi.ctx = ctx_node->ctx; + binding.bi.res = res; + binding.bi.bt = vmw_ctx_binding_so; + binding.slot = 0; /* Only one SO set to context at a time. */ + + vmw_binding_add(sw_context->dx_ctx_node->staged, &binding.bi, 0, + binding.slot); + + return ret; +} + +static int vmw_cmd_indexed_instanced_indirect(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + SVGA3dCmdHeader *header) +{ + struct vmw_draw_indexed_instanced_indirect_cmd { + SVGA3dCmdHeader header; + SVGA3dCmdDXDrawIndexedInstancedIndirect body; + } *cmd = container_of(header, typeof(*cmd), header); + + if (!has_sm5_context(dev_priv)) + return -EINVAL; + + return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, + VMW_RES_DIRTY_NONE, user_surface_converter, + &cmd->body.argsBufferSid, NULL); +} + +static int vmw_cmd_instanced_indirect(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + SVGA3dCmdHeader *header) +{ + struct vmw_draw_instanced_indirect_cmd { + SVGA3dCmdHeader header; + SVGA3dCmdDXDrawInstancedIndirect body; + } *cmd = container_of(header, typeof(*cmd), header); + + if (!has_sm5_context(dev_priv)) + return -EINVAL; + + return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, + VMW_RES_DIRTY_NONE, user_surface_converter, + &cmd->body.argsBufferSid, NULL); +} + +static int vmw_cmd_dispatch_indirect(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + SVGA3dCmdHeader *header) +{ + struct vmw_dispatch_indirect_cmd { + SVGA3dCmdHeader header; + SVGA3dCmdDXDispatchIndirect body; + } *cmd = container_of(header, typeof(*cmd), header); + + if (!has_sm5_context(dev_priv)) + return -EINVAL; + + return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, + VMW_RES_DIRTY_NONE, user_surface_converter, + &cmd->body.argsBufferSid, NULL); +} + static int vmw_cmd_check_not_3d(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, void *buf, uint32_t *size) @@ -2922,18 +3280,12 @@ static const struct vmw_cmd_entry vmw_cmd_entries[SVGA_3D_CMD_MAX] = { false, false, false), VMW_CMD_DEF(SVGA_3D_CMD_DEAD2, &vmw_cmd_invalid, false, false, false), - VMW_CMD_DEF(SVGA_3D_CMD_LOGICOPS_BITBLT, &vmw_cmd_invalid, - false, false, false), - VMW_CMD_DEF(SVGA_3D_CMD_LOGICOPS_TRANSBLT, &vmw_cmd_invalid, - false, false, false), - VMW_CMD_DEF(SVGA_3D_CMD_LOGICOPS_STRETCHBLT, &vmw_cmd_invalid, - false, false, false), - VMW_CMD_DEF(SVGA_3D_CMD_LOGICOPS_COLORFILL, &vmw_cmd_invalid, - false, false, false), - VMW_CMD_DEF(SVGA_3D_CMD_LOGICOPS_ALPHABLEND, &vmw_cmd_invalid, - false, false, false), - VMW_CMD_DEF(SVGA_3D_CMD_LOGICOPS_CLEARTYPEBLEND, &vmw_cmd_invalid, - false, false, false), + VMW_CMD_DEF(SVGA_3D_CMD_DEAD12, &vmw_cmd_invalid, false, false, false), + VMW_CMD_DEF(SVGA_3D_CMD_DEAD13, &vmw_cmd_invalid, false, false, false), + VMW_CMD_DEF(SVGA_3D_CMD_DEAD14, &vmw_cmd_invalid, false, false, false), + VMW_CMD_DEF(SVGA_3D_CMD_DEAD15, &vmw_cmd_invalid, false, false, false), + VMW_CMD_DEF(SVGA_3D_CMD_DEAD16, &vmw_cmd_invalid, false, false, false), + VMW_CMD_DEF(SVGA_3D_CMD_DEAD17, &vmw_cmd_invalid, false, false, false), VMW_CMD_DEF(SVGA_3D_CMD_SET_OTABLE_BASE, &vmw_cmd_invalid, false, false, true), VMW_CMD_DEF(SVGA_3D_CMD_READBACK_OTABLE, &vmw_cmd_invalid, @@ -3141,9 +3493,9 @@ static const struct vmw_cmd_entry vmw_cmd_entries[SVGA_3D_CMD_MAX] = { VMW_CMD_DEF(SVGA_3D_CMD_DX_DEFINE_STREAMOUTPUT, &vmw_cmd_dx_so_define, true, false, true), VMW_CMD_DEF(SVGA_3D_CMD_DX_DESTROY_STREAMOUTPUT, - &vmw_cmd_dx_cid_check, true, false, true), - VMW_CMD_DEF(SVGA_3D_CMD_DX_SET_STREAMOUTPUT, &vmw_cmd_dx_cid_check, - true, false, true), + &vmw_cmd_dx_destroy_streamoutput, true, false, true), + VMW_CMD_DEF(SVGA_3D_CMD_DX_SET_STREAMOUTPUT, + &vmw_cmd_dx_set_streamoutput, true, false, true), VMW_CMD_DEF(SVGA_3D_CMD_DX_SET_SOTARGETS, &vmw_cmd_dx_set_so_targets, true, false, true), VMW_CMD_DEF(SVGA_3D_CMD_DX_SET_INPUT_LAYOUT, @@ -3159,6 +3511,37 @@ static const struct vmw_cmd_entry vmw_cmd_entries[SVGA_3D_CMD_MAX] = { true, false, true), VMW_CMD_DEF(SVGA_3D_CMD_INTRA_SURFACE_COPY, &vmw_cmd_intra_surface_copy, true, false, true), + + /* + * SM5 commands + */ + VMW_CMD_DEF(SVGA_3D_CMD_DX_DEFINE_UA_VIEW, &vmw_cmd_sm5_view_define, + true, false, true), + VMW_CMD_DEF(SVGA_3D_CMD_DX_DESTROY_UA_VIEW, &vmw_cmd_sm5_view_remove, + true, false, true), + VMW_CMD_DEF(SVGA_3D_CMD_DX_CLEAR_UA_VIEW_UINT, &vmw_cmd_clear_uav_uint, + true, false, true), + VMW_CMD_DEF(SVGA_3D_CMD_DX_CLEAR_UA_VIEW_FLOAT, + &vmw_cmd_clear_uav_float, true, false, true), + VMW_CMD_DEF(SVGA_3D_CMD_DX_COPY_STRUCTURE_COUNT, &vmw_cmd_invalid, true, + false, true), + VMW_CMD_DEF(SVGA_3D_CMD_DX_SET_UA_VIEWS, &vmw_cmd_set_uav, true, false, + true), + VMW_CMD_DEF(SVGA_3D_CMD_DX_DRAW_INDEXED_INSTANCED_INDIRECT, + &vmw_cmd_indexed_instanced_indirect, true, false, true), + VMW_CMD_DEF(SVGA_3D_CMD_DX_DRAW_INSTANCED_INDIRECT, + &vmw_cmd_instanced_indirect, true, false, true), + VMW_CMD_DEF(SVGA_3D_CMD_DX_DISPATCH, &vmw_cmd_sm5, true, false, true), + VMW_CMD_DEF(SVGA_3D_CMD_DX_DISPATCH_INDIRECT, + &vmw_cmd_dispatch_indirect, true, false, true), + VMW_CMD_DEF(SVGA_3D_CMD_DX_SET_CS_UA_VIEWS, &vmw_cmd_set_cs_uav, true, + false, true), + VMW_CMD_DEF(SVGA_3D_CMD_DX_DEFINE_DEPTHSTENCIL_VIEW_V2, + &vmw_cmd_sm5_view_define, true, false, true), + VMW_CMD_DEF(SVGA_3D_CMD_DX_DEFINE_STREAMOUTPUT_WITH_MOB, + &vmw_cmd_dx_define_streamoutput, true, false, true), + VMW_CMD_DEF(SVGA_3D_CMD_DX_BIND_STREAMOUTPUT, + &vmw_cmd_dx_bind_streamoutput, true, false, true), }; bool vmw_cmd_describe(const void *buf, u32 *size, char const **cmd) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c b/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c index e5252ef3812f..6941689085ed 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c @@ -169,10 +169,8 @@ void vmw_fifo_ping_host(struct vmw_private *dev_priv, uint32_t reason) { u32 *fifo_mem = dev_priv->mmio_virt; - preempt_disable(); if (cmpxchg(fifo_mem + SVGA_FIFO_BUSY, 0, 1) == 0) vmw_write(dev_priv, SVGA_REG_SYNC, reason); - preempt_enable(); } void vmw_fifo_release(struct vmw_private *dev_priv, struct vmw_fifo_state *fifo) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c index a15375eb476e..f681b7b4df1b 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c @@ -114,10 +114,13 @@ int vmw_getparam_ioctl(struct drm_device *dev, void *data, (dev_priv->active_display_unit == vmw_du_screen_target); break; case DRM_VMW_PARAM_DX: - param->value = dev_priv->has_dx; + param->value = has_sm4_context(dev_priv); break; case DRM_VMW_PARAM_SM4_1: - param->value = dev_priv->has_sm4_1; + param->value = has_sm4_1_context(dev_priv); + break; + case DRM_VMW_PARAM_SM5: + param->value = has_sm5_context(dev_priv); break; default: return -EINVAL; @@ -126,14 +129,17 @@ int vmw_getparam_ioctl(struct drm_device *dev, void *data, return 0; } -static u32 vmw_mask_multisample(unsigned int cap, u32 fmt_value) +static u32 vmw_mask_legacy_multisample(unsigned int cap, u32 fmt_value) { /* * A version of user-space exists which use MULTISAMPLE_MASKABLESAMPLES * to check the sample count supported by virtual device. Since there * never was support for multisample count for backing MOB return 0. + * + * MULTISAMPLE_MASKABLESAMPLES devcap is marked as deprecated by virtual + * device. */ - if (cap == SVGA3D_DEVCAP_MULTISAMPLE_MASKABLESAMPLES) + if (cap == SVGA3D_DEVCAP_DEAD5) return 0; return fmt_value; @@ -164,7 +170,7 @@ static int vmw_fill_compat_cap(struct vmw_private *dev_priv, void *bounce, for (i = 0; i < max_size; ++i) { vmw_write(dev_priv, SVGA_REG_DEV_CAP, i); compat_cap->pairs[i][0] = i; - compat_cap->pairs[i][1] = vmw_mask_multisample + compat_cap->pairs[i][1] = vmw_mask_legacy_multisample (i, vmw_read(dev_priv, SVGA_REG_DEV_CAP)); } spin_unlock(&dev_priv->cap_lock); @@ -220,7 +226,7 @@ int vmw_get_cap_3d_ioctl(struct drm_device *dev, void *data, spin_lock(&dev_priv->cap_lock); for (i = 0; i < num; ++i) { vmw_write(dev_priv, SVGA_REG_DEV_CAP, i); - *bounce32++ = vmw_mask_multisample + *bounce32++ = vmw_mask_legacy_multisample (i, vmw_read(dev_priv, SVGA_REG_DEV_CAP)); } spin_unlock(&dev_priv->cap_lock); diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c index f47d5710cc95..04d66592f605 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c @@ -905,14 +905,14 @@ static int vmw_kms_new_framebuffer_surface(struct vmw_private *dev_priv, */ /* Surface must be marked as a scanout. */ - if (unlikely(!surface->scanout)) + if (unlikely(!surface->metadata.scanout)) return -EINVAL; - if (unlikely(surface->mip_levels[0] != 1 || - surface->num_sizes != 1 || - surface->base_size.width < mode_cmd->width || - surface->base_size.height < mode_cmd->height || - surface->base_size.depth != 1)) { + if (unlikely(surface->metadata.mip_levels[0] != 1 || + surface->metadata.num_sizes != 1 || + surface->metadata.base_size.width < mode_cmd->width || + surface->metadata.base_size.height < mode_cmd->height || + surface->metadata.base_size.depth != 1)) { DRM_ERROR("Incompatible surface dimensions " "for requested mode.\n"); return -EINVAL; @@ -941,7 +941,7 @@ static int vmw_kms_new_framebuffer_surface(struct vmw_private *dev_priv, * For DX, surface format validation is done when surface->scanout * is set. */ - if (!dev_priv->has_dx && format != surface->format) { + if (!has_sm4_context(dev_priv) && format != surface->metadata.format) { DRM_ERROR("Invalid surface format for requested mode.\n"); return -EINVAL; } @@ -1144,8 +1144,8 @@ static int vmw_create_bo_proxy(struct drm_device *dev, struct vmw_buffer_object *bo_mob, struct vmw_surface **srf_out) { + struct vmw_surface_metadata metadata = {0}; uint32_t format; - struct drm_vmw_size content_base_size = {0}; struct vmw_resource *res; unsigned int bytes_pp; struct drm_format_name_buf format_name; @@ -1175,22 +1175,15 @@ static int vmw_create_bo_proxy(struct drm_device *dev, return -EINVAL; } - content_base_size.width = mode_cmd->pitches[0] / bytes_pp; - content_base_size.height = mode_cmd->height; - content_base_size.depth = 1; - - ret = vmw_surface_gb_priv_define(dev, - 0, /* kernel visible only */ - 0, /* flags */ - format, - true, /* can be a scanout buffer */ - 1, /* num of mip levels */ - 0, - 0, - content_base_size, - SVGA3D_MS_PATTERN_NONE, - SVGA3D_MS_QUALITY_NONE, - srf_out); + metadata.format = format; + metadata.mip_levels[0] = 1; + metadata.num_sizes = 1; + metadata.base_size.width = mode_cmd->pitches[0] / bytes_pp; + metadata.base_size.height = mode_cmd->height; + metadata.base_size.depth = 1; + metadata.scanout = true; + + ret = vmw_gb_surface_define(vmw_priv(dev), 0, &metadata, srf_out); if (ret) { DRM_ERROR("Failed to allocate proxy content buffer\n"); return ret; @@ -1897,87 +1890,6 @@ int vmw_kms_write_svga(struct vmw_private *vmw_priv, return 0; } -int vmw_kms_save_vga(struct vmw_private *vmw_priv) -{ - struct vmw_vga_topology_state *save; - uint32_t i; - - vmw_priv->vga_width = vmw_read(vmw_priv, SVGA_REG_WIDTH); - vmw_priv->vga_height = vmw_read(vmw_priv, SVGA_REG_HEIGHT); - vmw_priv->vga_bpp = vmw_read(vmw_priv, SVGA_REG_BITS_PER_PIXEL); - if (vmw_priv->capabilities & SVGA_CAP_PITCHLOCK) - vmw_priv->vga_pitchlock = - vmw_read(vmw_priv, SVGA_REG_PITCHLOCK); - else if (vmw_fifo_have_pitchlock(vmw_priv)) - vmw_priv->vga_pitchlock = vmw_mmio_read(vmw_priv->mmio_virt + - SVGA_FIFO_PITCHLOCK); - - if (!(vmw_priv->capabilities & SVGA_CAP_DISPLAY_TOPOLOGY)) - return 0; - - vmw_priv->num_displays = vmw_read(vmw_priv, - SVGA_REG_NUM_GUEST_DISPLAYS); - - if (vmw_priv->num_displays == 0) - vmw_priv->num_displays = 1; - - for (i = 0; i < vmw_priv->num_displays; ++i) { - save = &vmw_priv->vga_save[i]; - vmw_write(vmw_priv, SVGA_REG_DISPLAY_ID, i); - save->primary = vmw_read(vmw_priv, SVGA_REG_DISPLAY_IS_PRIMARY); - save->pos_x = vmw_read(vmw_priv, SVGA_REG_DISPLAY_POSITION_X); - save->pos_y = vmw_read(vmw_priv, SVGA_REG_DISPLAY_POSITION_Y); - save->width = vmw_read(vmw_priv, SVGA_REG_DISPLAY_WIDTH); - save->height = vmw_read(vmw_priv, SVGA_REG_DISPLAY_HEIGHT); - vmw_write(vmw_priv, SVGA_REG_DISPLAY_ID, SVGA_ID_INVALID); - if (i == 0 && vmw_priv->num_displays == 1 && - save->width == 0 && save->height == 0) { - - /* - * It should be fairly safe to assume that these - * values are uninitialized. - */ - - save->width = vmw_priv->vga_width - save->pos_x; - save->height = vmw_priv->vga_height - save->pos_y; - } - } - - return 0; -} - -int vmw_kms_restore_vga(struct vmw_private *vmw_priv) -{ - struct vmw_vga_topology_state *save; - uint32_t i; - - vmw_write(vmw_priv, SVGA_REG_WIDTH, vmw_priv->vga_width); - vmw_write(vmw_priv, SVGA_REG_HEIGHT, vmw_priv->vga_height); - vmw_write(vmw_priv, SVGA_REG_BITS_PER_PIXEL, vmw_priv->vga_bpp); - if (vmw_priv->capabilities & SVGA_CAP_PITCHLOCK) - vmw_write(vmw_priv, SVGA_REG_PITCHLOCK, - vmw_priv->vga_pitchlock); - else if (vmw_fifo_have_pitchlock(vmw_priv)) - vmw_mmio_write(vmw_priv->vga_pitchlock, - vmw_priv->mmio_virt + SVGA_FIFO_PITCHLOCK); - - if (!(vmw_priv->capabilities & SVGA_CAP_DISPLAY_TOPOLOGY)) - return 0; - - for (i = 0; i < vmw_priv->num_displays; ++i) { - save = &vmw_priv->vga_save[i]; - vmw_write(vmw_priv, SVGA_REG_DISPLAY_ID, i); - vmw_write(vmw_priv, SVGA_REG_DISPLAY_IS_PRIMARY, save->primary); - vmw_write(vmw_priv, SVGA_REG_DISPLAY_POSITION_X, save->pos_x); - vmw_write(vmw_priv, SVGA_REG_DISPLAY_POSITION_Y, save->pos_y); - vmw_write(vmw_priv, SVGA_REG_DISPLAY_WIDTH, save->width); - vmw_write(vmw_priv, SVGA_REG_DISPLAY_HEIGHT, save->height); - vmw_write(vmw_priv, SVGA_REG_DISPLAY_ID, SVGA_ID_INVALID); - } - - return 0; -} - bool vmw_kms_validate_mode_vram(struct vmw_private *dev_priv, uint32_t pitch, uint32_t height) @@ -1991,7 +1903,7 @@ bool vmw_kms_validate_mode_vram(struct vmw_private *dev_priv, /** * Function called by DRM code called with vbl_lock held. */ -u32 vmw_get_vblank_counter(struct drm_device *dev, unsigned int pipe) +u32 vmw_get_vblank_counter(struct drm_crtc *crtc) { return 0; } @@ -1999,7 +1911,7 @@ u32 vmw_get_vblank_counter(struct drm_device *dev, unsigned int pipe) /** * Function called by DRM code called with vbl_lock held. */ -int vmw_enable_vblank(struct drm_device *dev, unsigned int pipe) +int vmw_enable_vblank(struct drm_crtc *crtc) { return -EINVAL; } @@ -2007,7 +1919,7 @@ int vmw_enable_vblank(struct drm_device *dev, unsigned int pipe) /** * Function called by DRM code called with vbl_lock held. */ -void vmw_disable_vblank(struct drm_device *dev, unsigned int pipe) +void vmw_disable_vblank(struct drm_crtc *crtc) { } @@ -2597,7 +2509,7 @@ int vmw_kms_update_proxy(struct vmw_resource *res, int increment) { struct vmw_private *dev_priv = res->dev_priv; - struct drm_vmw_size *size = &vmw_res_to_srf(res)->base_size; + struct drm_vmw_size *size = &vmw_res_to_srf(res)->metadata.base_size; struct { SVGA3dCmdHeader header; SVGA3dCmdUpdateGBImage body; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c index 5702219ec38f..16dafff5cab1 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c @@ -236,6 +236,9 @@ static const struct drm_crtc_funcs vmw_legacy_crtc_funcs = { .atomic_duplicate_state = vmw_du_crtc_duplicate_state, .atomic_destroy_state = vmw_du_crtc_destroy_state, .set_config = drm_atomic_helper_set_config, + .get_vblank_counter = vmw_get_vblank_counter, + .enable_vblank = vmw_enable_vblank, + .disable_vblank = vmw_disable_vblank, }; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c b/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c index 0a6bbac00896..e8eb42933ca2 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c @@ -320,7 +320,7 @@ int vmw_otables_setup(struct vmw_private *dev_priv) struct vmw_otable **otables = &dev_priv->otable_batch.otables; int ret; - if (dev_priv->has_dx) { + if (has_sm4_context(dev_priv)) { *otables = kmemdup(dx_tables, sizeof(dx_tables), GFP_KERNEL); if (!(*otables)) return -ENOMEM; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_overlay.c b/drivers/gpu/drm/vmwgfx/vmwgfx_overlay.c index fdb52f6d29fb..cd7ed1650d60 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_overlay.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_overlay.c @@ -354,37 +354,6 @@ static int vmw_overlay_update_stream(struct vmw_private *dev_priv, } /** - * Stop all streams. - * - * Used by the fb code when starting. - * - * Takes the overlay lock. - */ -int vmw_overlay_stop_all(struct vmw_private *dev_priv) -{ - struct vmw_overlay *overlay = dev_priv->overlay_priv; - int i, ret; - - if (!overlay) - return 0; - - mutex_lock(&overlay->mutex); - - for (i = 0; i < VMW_MAX_NUM_STREAMS; i++) { - struct vmw_stream *stream = &overlay->stream[i]; - if (!stream->buf) - continue; - - ret = vmw_overlay_stop(dev_priv, i, false, false); - WARN_ON(ret != 0); - } - - mutex_unlock(&overlay->mutex); - - return 0; -} - -/** * Try to resume all paused streams. * * Used by the kms code after moving a new scanout buffer to vram. diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_page_dirty.c b/drivers/gpu/drm/vmwgfx/vmwgfx_page_dirty.c index f07aa857587c..d4d66532f9c9 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_page_dirty.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_page_dirty.c @@ -69,7 +69,7 @@ struct vmw_bo_dirty { unsigned int ref_count; unsigned long bitmap_size; size_t size; - unsigned long bitmap[0]; + unsigned long bitmap[]; }; /** @@ -473,11 +473,11 @@ vm_fault_t vmw_bo_vm_fault(struct vm_fault *vmf) * a lot of unnecessary write faults. */ if (vbo->dirty && vbo->dirty->method == VMW_BO_DIRTY_MKWRITE) - prot = vma->vm_page_prot; + prot = vm_get_page_prot(vma->vm_flags & ~VM_SHARED); else prot = vm_get_page_prot(vma->vm_flags); - ret = ttm_bo_vm_fault_reserved(vmf, prot, num_prefault); + ret = ttm_bo_vm_fault_reserved(vmf, prot, num_prefault, 1); if (ret == VM_FAULT_RETRY && !(vmf->flags & FAULT_FLAG_RETRY_NOWAIT)) return ret; @@ -486,3 +486,75 @@ out_unlock: return ret; } + +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +vm_fault_t vmw_bo_vm_huge_fault(struct vm_fault *vmf, + enum page_entry_size pe_size) +{ + struct vm_area_struct *vma = vmf->vma; + struct ttm_buffer_object *bo = (struct ttm_buffer_object *) + vma->vm_private_data; + struct vmw_buffer_object *vbo = + container_of(bo, struct vmw_buffer_object, base); + pgprot_t prot; + vm_fault_t ret; + pgoff_t fault_page_size; + bool write = vmf->flags & FAULT_FLAG_WRITE; + bool is_cow_mapping = + (vma->vm_flags & (VM_SHARED | VM_MAYWRITE)) == VM_MAYWRITE; + + switch (pe_size) { + case PE_SIZE_PMD: + fault_page_size = HPAGE_PMD_SIZE >> PAGE_SHIFT; + break; +#ifdef CONFIG_HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD + case PE_SIZE_PUD: + fault_page_size = HPAGE_PUD_SIZE >> PAGE_SHIFT; + break; +#endif + default: + WARN_ON_ONCE(1); + return VM_FAULT_FALLBACK; + } + + /* Always do write dirty-tracking and COW on PTE level. */ + if (write && (READ_ONCE(vbo->dirty) || is_cow_mapping)) + return VM_FAULT_FALLBACK; + + ret = ttm_bo_vm_reserve(bo, vmf); + if (ret) + return ret; + + if (vbo->dirty) { + pgoff_t allowed_prefault; + unsigned long page_offset; + + page_offset = vmf->pgoff - + drm_vma_node_start(&bo->base.vma_node); + if (page_offset >= bo->num_pages || + vmw_resources_clean(vbo, page_offset, + page_offset + PAGE_SIZE, + &allowed_prefault)) { + ret = VM_FAULT_SIGBUS; + goto out_unlock; + } + + /* + * Write protect, so we get a new fault on write, and can + * split. + */ + prot = vm_get_page_prot(vma->vm_flags & ~VM_SHARED); + } else { + prot = vm_get_page_prot(vma->vm_flags); + } + + ret = ttm_bo_vm_fault_reserved(vmf, prot, 1, fault_page_size); + if (ret == VM_FAULT_RETRY && !(vmf->flags & FAULT_FLAG_RETRY_NOWAIT)) + return ret; + +out_unlock: + dma_resv_unlock(bo->base.resv); + + return ret; +} +#endif diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c b/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c index e5a283263211..32a22e4eddb1 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c @@ -319,6 +319,9 @@ static const struct drm_crtc_funcs vmw_screen_object_crtc_funcs = { .atomic_destroy_state = vmw_du_crtc_destroy_state, .set_config = drm_atomic_helper_set_config, .page_flip = drm_atomic_helper_page_flip, + .get_vblank_counter = vmw_get_vblank_counter, + .enable_vblank = vmw_enable_vblank, + .disable_vblank = vmw_disable_vblank, }; /* diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_so.c b/drivers/gpu/drm/vmwgfx/vmwgfx_so.c index 63807361e16f..3f97b61dd5d8 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_so.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_so.c @@ -319,7 +319,8 @@ int vmw_view_add(struct vmw_cmdbuf_res_manager *man, static const size_t vmw_view_define_sizes[] = { [vmw_view_sr] = sizeof(SVGA3dCmdDXDefineShaderResourceView), [vmw_view_rt] = sizeof(SVGA3dCmdDXDefineRenderTargetView), - [vmw_view_ds] = sizeof(SVGA3dCmdDXDefineDepthStencilView) + [vmw_view_ds] = sizeof(SVGA3dCmdDXDefineDepthStencilView), + [vmw_view_ua] = sizeof(SVGA3dCmdDXDefineUAView) }; struct vmw_private *dev_priv = ctx->dev_priv; @@ -499,8 +500,8 @@ struct vmw_resource *vmw_view_lookup(struct vmw_cmdbuf_res_manager *man, * Each time a resource is put on the validation list as the result of a * view pointing to it, we need to determine whether that resource will * be dirtied (written to by the GPU) as a result of the corresponding - * GPU operation. Currently only rendertarget- and depth-stencil views are - * capable of dirtying its resource. + * GPU operation. Currently only rendertarget-, depth-stencil and unordered + * access views are capable of dirtying its resource. * * Return: Whether the view type of @res dirties the resource it points to. */ @@ -509,10 +510,11 @@ u32 vmw_view_dirtying(struct vmw_resource *res) static u32 view_is_dirtying[vmw_view_max] = { [vmw_view_rt] = VMW_RES_DIRTY_SET, [vmw_view_ds] = VMW_RES_DIRTY_SET, + [vmw_view_ua] = VMW_RES_DIRTY_SET, }; /* Update this function as we add more view types */ - BUILD_BUG_ON(vmw_view_max != 3); + BUILD_BUG_ON(vmw_view_max != 4); return view_is_dirtying[vmw_view(res)->view_type]; } @@ -520,12 +522,14 @@ const u32 vmw_view_destroy_cmds[] = { [vmw_view_sr] = SVGA_3D_CMD_DX_DESTROY_SHADERRESOURCE_VIEW, [vmw_view_rt] = SVGA_3D_CMD_DX_DESTROY_RENDERTARGET_VIEW, [vmw_view_ds] = SVGA_3D_CMD_DX_DESTROY_DEPTHSTENCIL_VIEW, + [vmw_view_ua] = SVGA_3D_CMD_DX_DESTROY_UA_VIEW, }; const SVGACOTableType vmw_view_cotables[] = { [vmw_view_sr] = SVGA_COTABLE_SRVIEW, [vmw_view_rt] = SVGA_COTABLE_RTVIEW, [vmw_view_ds] = SVGA_COTABLE_DSVIEW, + [vmw_view_ua] = SVGA_COTABLE_UAVIEW, }; const SVGACOTableType vmw_so_cotables[] = { diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_so.h b/drivers/gpu/drm/vmwgfx/vmwgfx_so.h index 12565047bc55..f48b84bfeeac 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_so.h +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_so.h @@ -30,6 +30,7 @@ enum vmw_view_type { vmw_view_sr, vmw_view_rt, vmw_view_ds, + vmw_view_ua, vmw_view_max, }; @@ -61,6 +62,7 @@ union vmw_view_destroy { struct SVGA3dCmdDXDestroyRenderTargetView rtv; struct SVGA3dCmdDXDestroyShaderResourceView srv; struct SVGA3dCmdDXDestroyDepthStencilView dsv; + struct SVGA3dCmdDXDestroyUAView uav; u32 view_id; }; @@ -87,6 +89,10 @@ static inline enum vmw_view_type vmw_view_cmd_to_type(u32 id) { u32 tmp = (id - SVGA_3D_CMD_DX_DEFINE_SHADERRESOURCE_VIEW) / 2; + if (id == SVGA_3D_CMD_DX_DEFINE_UA_VIEW || + id == SVGA_3D_CMD_DX_DESTROY_UA_VIEW) + return vmw_view_ua; + if (tmp > (u32)vmw_view_max) return vmw_view_max; @@ -123,6 +129,7 @@ static inline enum vmw_so_type vmw_so_cmd_to_type(u32 id) case SVGA_3D_CMD_DX_DESTROY_SAMPLER_STATE: return vmw_so_ss; case SVGA_3D_CMD_DX_DEFINE_STREAMOUTPUT: + case SVGA_3D_CMD_DX_DEFINE_STREAMOUTPUT_WITH_MOB: case SVGA_3D_CMD_DX_DESTROY_STREAMOUTPUT: return vmw_so_so; default: diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c b/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c index 41a96fb49835..9ffa9c75a5da 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c @@ -590,7 +590,7 @@ static void vmw_stdu_bo_cpu_commit(struct vmw_kms_dirty *dirty) return; /* Assume we are blitting from Guest (bo) to Host (display_srf) */ - dst_pitch = stdu->display_srf->base_size.width * stdu->cpp; + dst_pitch = stdu->display_srf->metadata.base_size.width * stdu->cpp; dst_bo = &stdu->display_srf->res.backup->base; dst_offset = ddirty->top * dst_pitch + ddirty->left * stdu->cpp; @@ -916,6 +916,9 @@ static const struct drm_crtc_funcs vmw_stdu_crtc_funcs = { .atomic_destroy_state = vmw_du_crtc_destroy_state, .set_config = drm_atomic_helper_set_config, .page_flip = drm_atomic_helper_page_flip, + .get_vblank_counter = vmw_get_vblank_counter, + .enable_vblank = vmw_enable_vblank, + .disable_vblank = vmw_disable_vblank, }; @@ -1038,7 +1041,6 @@ vmw_stdu_primary_plane_prepare_fb(struct drm_plane *plane, struct vmw_plane_state *vps = vmw_plane_state_to_vps(new_state); enum stdu_content_type new_content_type; struct vmw_framebuffer_surface *new_vfbs; - struct drm_crtc *crtc = new_state->crtc; uint32_t hdisplay = new_state->crtc_w, vdisplay = new_state->crtc_h; int ret; @@ -1055,8 +1057,9 @@ vmw_stdu_primary_plane_prepare_fb(struct drm_plane *plane, vfb = vmw_framebuffer_to_vfb(new_fb); new_vfbs = (vfb->bo) ? NULL : vmw_framebuffer_to_vfbs(new_fb); - if (new_vfbs && new_vfbs->surface->base_size.width == hdisplay && - new_vfbs->surface->base_size.height == vdisplay) + if (new_vfbs && + new_vfbs->surface->metadata.base_size.width == hdisplay && + new_vfbs->surface->metadata.base_size.height == vdisplay) new_content_type = SAME_AS_DISPLAY; else if (vfb->bo) new_content_type = SEPARATE_BO; @@ -1064,12 +1067,11 @@ vmw_stdu_primary_plane_prepare_fb(struct drm_plane *plane, new_content_type = SEPARATE_SURFACE; if (new_content_type != SAME_AS_DISPLAY) { - struct vmw_surface content_srf; - struct drm_vmw_size display_base_size = {0}; + struct vmw_surface_metadata metadata = {0}; - display_base_size.width = hdisplay; - display_base_size.height = vdisplay; - display_base_size.depth = 1; + metadata.base_size.width = hdisplay; + metadata.base_size.height = vdisplay; + metadata.base_size.depth = 1; /* * If content buffer is a buffer object, then we have to @@ -1079,15 +1081,15 @@ vmw_stdu_primary_plane_prepare_fb(struct drm_plane *plane, switch (new_fb->format->cpp[0]*8) { case 32: - content_srf.format = SVGA3D_X8R8G8B8; + metadata.format = SVGA3D_X8R8G8B8; break; case 16: - content_srf.format = SVGA3D_R5G6B5; + metadata.format = SVGA3D_R5G6B5; break; case 8: - content_srf.format = SVGA3D_P8; + metadata.format = SVGA3D_P8; break; default: @@ -1095,22 +1097,20 @@ vmw_stdu_primary_plane_prepare_fb(struct drm_plane *plane, return -EINVAL; } - content_srf.flags = 0; - content_srf.mip_levels[0] = 1; - content_srf.multisample_count = 0; - content_srf.multisample_pattern = - SVGA3D_MS_PATTERN_NONE; - content_srf.quality_level = SVGA3D_MS_QUALITY_NONE; + metadata.mip_levels[0] = 1; + metadata.num_sizes = 1; + metadata.scanout = true; } else { - content_srf = *new_vfbs->surface; + metadata = new_vfbs->surface->metadata; } if (vps->surf) { - struct drm_vmw_size cur_base_size = vps->surf->base_size; + struct drm_vmw_size cur_base_size = + vps->surf->metadata.base_size; - if (cur_base_size.width != display_base_size.width || - cur_base_size.height != display_base_size.height || - vps->surf->format != content_srf.format) { + if (cur_base_size.width != metadata.base_size.width || + cur_base_size.height != metadata.base_size.height || + vps->surf->metadata.format != metadata.format) { WARN_ON(vps->pinned != 0); vmw_surface_unreference(&vps->surf); } @@ -1118,20 +1118,8 @@ vmw_stdu_primary_plane_prepare_fb(struct drm_plane *plane, } if (!vps->surf) { - ret = vmw_surface_gb_priv_define - (crtc->dev, - /* Kernel visible only */ - 0, - content_srf.flags, - content_srf.format, - true, /* a scanout buffer */ - content_srf.mip_levels[0], - content_srf.multisample_count, - 0, - display_base_size, - content_srf.multisample_pattern, - content_srf.quality_level, - &vps->surf); + ret = vmw_gb_surface_define(dev_priv, 0, &metadata, + &vps->surf); if (ret != 0) { DRM_ERROR("Couldn't allocate STDU surface.\n"); return ret; @@ -1308,7 +1296,7 @@ vmw_stdu_bo_populate_update_cpu(struct vmw_du_update_plane *update, void *cmd, diff.cpp = stdu->cpp; dst_bo = &stdu->display_srf->res.backup->base; - dst_pitch = stdu->display_srf->base_size.width * stdu->cpp; + dst_pitch = stdu->display_srf->metadata.base_size.width * stdu->cpp; dst_offset = bb->y1 * dst_pitch + bb->x1 * stdu->cpp; src_bo = &vfbbo->buffer->base; @@ -1885,7 +1873,7 @@ int vmw_kms_stdu_init_display(struct vmw_private *dev_priv) /* Do nothing if Screen Target support is turned off */ - if (!VMWGFX_ENABLE_SCREEN_TARGET_OTABLE) + if (!VMWGFX_ENABLE_SCREEN_TARGET_OTABLE || !dev_priv->has_mob) return -ENOSYS; if (!(dev_priv->capabilities & SVGA_CAP_GBOBJECTS)) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_streamoutput.c b/drivers/gpu/drm/vmwgfx/vmwgfx_streamoutput.c new file mode 100644 index 000000000000..193192456663 --- /dev/null +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_streamoutput.c @@ -0,0 +1,387 @@ +// SPDX-License-Identifier: GPL-2.0 OR MIT +/************************************************************************** + * + * Copyright © 2018-2019 VMware, Inc., Palo Alto, CA., USA + * All Rights Reserved. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the + * "Software"), to deal in the Software without restriction, including + * without limitation the rights to use, copy, modify, merge, publish, + * distribute, sub license, and/or sell copies of the Software, and to + * permit persons to whom the Software is furnished to do so, subject to + * the following conditions: + * + * The above copyright notice and this permission notice (including the + * next paragraph) shall be included in all copies or substantial portions + * of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDERS, AUTHORS AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM, + * DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR + * OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE + * USE OR OTHER DEALINGS IN THE SOFTWARE. + * + **************************************************************************/ + +#include <drm/ttm/ttm_placement.h> + +#include "vmwgfx_drv.h" +#include "vmwgfx_resource_priv.h" +#include "vmwgfx_binding.h" + +/** + * struct vmw_dx_streamoutput - Streamoutput resource metadata. + * @res: Base resource struct. + * @ctx: Non-refcounted context to which @res belong. + * @cotable: Refcounted cotable holding this Streamoutput. + * @cotable_head: List head for cotable-so_res list. + * @id: User-space provided identifier. + * @size: User-space provided mob size. + * @committed: Whether streamoutput is actually created or pending creation. + */ +struct vmw_dx_streamoutput { + struct vmw_resource res; + struct vmw_resource *ctx; + struct vmw_resource *cotable; + struct list_head cotable_head; + u32 id; + u32 size; + bool committed; +}; + +static int vmw_dx_streamoutput_create(struct vmw_resource *res); +static int vmw_dx_streamoutput_bind(struct vmw_resource *res, + struct ttm_validate_buffer *val_buf); +static int vmw_dx_streamoutput_unbind(struct vmw_resource *res, bool readback, + struct ttm_validate_buffer *val_buf); +static void vmw_dx_streamoutput_commit_notify(struct vmw_resource *res, + enum vmw_cmdbuf_res_state state); + +static size_t vmw_streamoutput_size; + +static const struct vmw_res_func vmw_dx_streamoutput_func = { + .res_type = vmw_res_streamoutput, + .needs_backup = true, + .may_evict = false, + .type_name = "DX streamoutput", + .backup_placement = &vmw_mob_placement, + .create = vmw_dx_streamoutput_create, + .destroy = NULL, /* Command buffer managed resource. */ + .bind = vmw_dx_streamoutput_bind, + .unbind = vmw_dx_streamoutput_unbind, + .commit_notify = vmw_dx_streamoutput_commit_notify, +}; + +static inline struct vmw_dx_streamoutput * +vmw_res_to_dx_streamoutput(struct vmw_resource *res) +{ + return container_of(res, struct vmw_dx_streamoutput, res); +} + +/** + * vmw_dx_streamoutput_unscrub - Reattach the MOB to streamoutput. + * @res: The streamoutput resource. + * + * Return: 0 on success, negative error code on failure. + */ +static int vmw_dx_streamoutput_unscrub(struct vmw_resource *res) +{ + struct vmw_dx_streamoutput *so = vmw_res_to_dx_streamoutput(res); + struct vmw_private *dev_priv = res->dev_priv; + struct { + SVGA3dCmdHeader header; + SVGA3dCmdDXBindStreamOutput body; + } *cmd; + + if (!list_empty(&so->cotable_head) || !so->committed ) + return 0; + + cmd = VMW_FIFO_RESERVE_DX(dev_priv, sizeof(*cmd), so->ctx->id); + if (!cmd) + return -ENOMEM; + + cmd->header.id = SVGA_3D_CMD_DX_BIND_STREAMOUTPUT; + cmd->header.size = sizeof(cmd->body); + cmd->body.soid = so->id; + cmd->body.mobid = res->backup->base.mem.start; + cmd->body.offsetInBytes = res->backup_offset; + cmd->body.sizeInBytes = so->size; + vmw_fifo_commit(dev_priv, sizeof(*cmd)); + + vmw_cotable_add_resource(so->cotable, &so->cotable_head); + + return 0; +} + +static int vmw_dx_streamoutput_create(struct vmw_resource *res) +{ + struct vmw_private *dev_priv = res->dev_priv; + struct vmw_dx_streamoutput *so = vmw_res_to_dx_streamoutput(res); + int ret = 0; + + WARN_ON_ONCE(!so->committed); + + if (vmw_resource_mob_attached(res)) { + mutex_lock(&dev_priv->binding_mutex); + ret = vmw_dx_streamoutput_unscrub(res); + mutex_unlock(&dev_priv->binding_mutex); + } + + res->id = so->id; + + return ret; +} + +static int vmw_dx_streamoutput_bind(struct vmw_resource *res, + struct ttm_validate_buffer *val_buf) +{ + struct vmw_private *dev_priv = res->dev_priv; + struct ttm_buffer_object *bo = val_buf->bo; + int ret; + + if (WARN_ON(bo->mem.mem_type != VMW_PL_MOB)) + return -EINVAL; + + mutex_lock(&dev_priv->binding_mutex); + ret = vmw_dx_streamoutput_unscrub(res); + mutex_unlock(&dev_priv->binding_mutex); + + return ret; +} + +/** + * vmw_dx_streamoutput_scrub - Unbind the MOB from streamoutput. + * @res: The streamoutput resource. + * + * Return: 0 on success, negative error code on failure. + */ +static int vmw_dx_streamoutput_scrub(struct vmw_resource *res) +{ + struct vmw_private *dev_priv = res->dev_priv; + struct vmw_dx_streamoutput *so = vmw_res_to_dx_streamoutput(res); + struct { + SVGA3dCmdHeader header; + SVGA3dCmdDXBindStreamOutput body; + } *cmd; + + if (list_empty(&so->cotable_head)) + return 0; + + WARN_ON_ONCE(!so->committed); + + cmd = VMW_FIFO_RESERVE_DX(dev_priv, sizeof(*cmd), so->ctx->id); + if (!cmd) + return -ENOMEM; + + cmd->header.id = SVGA_3D_CMD_DX_BIND_STREAMOUTPUT; + cmd->header.size = sizeof(cmd->body); + cmd->body.soid = res->id; + cmd->body.mobid = SVGA3D_INVALID_ID; + cmd->body.offsetInBytes = 0; + cmd->body.sizeInBytes = so->size; + vmw_fifo_commit(dev_priv, sizeof(*cmd)); + + res->id = -1; + list_del_init(&so->cotable_head); + + return 0; +} + +static int vmw_dx_streamoutput_unbind(struct vmw_resource *res, bool readback, + struct ttm_validate_buffer *val_buf) +{ + struct vmw_private *dev_priv = res->dev_priv; + struct vmw_fence_obj *fence; + int ret; + + if (WARN_ON(res->backup->base.mem.mem_type != VMW_PL_MOB)) + return -EINVAL; + + mutex_lock(&dev_priv->binding_mutex); + ret = vmw_dx_streamoutput_scrub(res); + mutex_unlock(&dev_priv->binding_mutex); + + if (ret) + return ret; + + (void) vmw_execbuf_fence_commands(NULL, dev_priv, &fence, NULL); + vmw_bo_fence_single(val_buf->bo, fence); + + if (fence != NULL) + vmw_fence_obj_unreference(&fence); + + return 0; +} + +static void vmw_dx_streamoutput_commit_notify(struct vmw_resource *res, + enum vmw_cmdbuf_res_state state) +{ + struct vmw_private *dev_priv = res->dev_priv; + struct vmw_dx_streamoutput *so = vmw_res_to_dx_streamoutput(res); + + if (state == VMW_CMDBUF_RES_ADD) { + mutex_lock(&dev_priv->binding_mutex); + vmw_cotable_add_resource(so->cotable, &so->cotable_head); + so->committed = true; + res->id = so->id; + mutex_unlock(&dev_priv->binding_mutex); + } else { + mutex_lock(&dev_priv->binding_mutex); + list_del_init(&so->cotable_head); + so->committed = false; + res->id = -1; + mutex_unlock(&dev_priv->binding_mutex); + } +} + +/** + * vmw_dx_streamoutput_lookup - Do a streamoutput resource lookup by user key. + * @man: Command buffer managed resource manager for current context. + * @user_key: User-space identifier for lookup. + * + * Return: Valid refcounted vmw_resource on success, error pointer on failure. + */ +struct vmw_resource * +vmw_dx_streamoutput_lookup(struct vmw_cmdbuf_res_manager *man, + u32 user_key) +{ + return vmw_cmdbuf_res_lookup(man, vmw_cmdbuf_res_streamoutput, + user_key); +} + +static void vmw_dx_streamoutput_res_free(struct vmw_resource *res) +{ + struct vmw_private *dev_priv = res->dev_priv; + struct vmw_dx_streamoutput *so = vmw_res_to_dx_streamoutput(res); + + vmw_resource_unreference(&so->cotable); + kfree(so); + ttm_mem_global_free(vmw_mem_glob(dev_priv), vmw_streamoutput_size); +} + +static void vmw_dx_streamoutput_hw_destroy(struct vmw_resource *res) +{ + /* Destroyed by user-space cmd buf or as part of context takedown. */ + res->id = -1; +} + +/** + * vmw_dx_streamoutput_add - Add a streamoutput as a cmd buf managed resource. + * @man: Command buffer managed resource manager for current context. + * @ctx: Pointer to context resource. + * @user_key: The identifier for this streamoutput. + * @list: The list of staged command buffer managed resources. + * + * Return: 0 on success, negative error code on failure. + */ +int vmw_dx_streamoutput_add(struct vmw_cmdbuf_res_manager *man, + struct vmw_resource *ctx, u32 user_key, + struct list_head *list) +{ + struct vmw_dx_streamoutput *so; + struct vmw_resource *res; + struct vmw_private *dev_priv = ctx->dev_priv; + struct ttm_operation_ctx ttm_opt_ctx = { + .interruptible = true, + .no_wait_gpu = false + }; + int ret; + + if (!vmw_streamoutput_size) + vmw_streamoutput_size = ttm_round_pot(sizeof(*so)); + + ret = ttm_mem_global_alloc(vmw_mem_glob(dev_priv), + vmw_streamoutput_size, &ttm_opt_ctx); + if (ret) { + if (ret != -ERESTARTSYS) + DRM_ERROR("Out of graphics memory for streamout.\n"); + return ret; + } + + so = kmalloc(sizeof(*so), GFP_KERNEL); + if (!so) { + ttm_mem_global_free(vmw_mem_glob(dev_priv), + vmw_streamoutput_size); + return -ENOMEM; + } + + res = &so->res; + so->ctx = ctx; + so->cotable = vmw_resource_reference + (vmw_context_cotable(ctx, SVGA_COTABLE_STREAMOUTPUT)); + so->id = user_key; + so->committed = false; + INIT_LIST_HEAD(&so->cotable_head); + ret = vmw_resource_init(dev_priv, res, true, + vmw_dx_streamoutput_res_free, + &vmw_dx_streamoutput_func); + if (ret) + goto out_resource_init; + + ret = vmw_cmdbuf_res_add(man, vmw_cmdbuf_res_streamoutput, user_key, + res, list); + if (ret) + goto out_resource_init; + + res->id = so->id; + res->hw_destroy = vmw_dx_streamoutput_hw_destroy; + +out_resource_init: + vmw_resource_unreference(&res); + + return ret; +} + +/** + * vmw_dx_streamoutput_set_size - Sets streamoutput mob size in res struct. + * @res: The streamoutput res for which need to set size. + * @size: The size provided by user-space to set. + */ +void vmw_dx_streamoutput_set_size(struct vmw_resource *res, u32 size) +{ + struct vmw_dx_streamoutput *so = vmw_res_to_dx_streamoutput(res); + + so->size = size; +} + +/** + * vmw_dx_streamoutput_remove - Stage streamoutput for removal. + * @man: Command buffer managed resource manager for current context. + * @user_key: The identifier for this streamoutput. + * @list: The list of staged command buffer managed resources. + * + * Return: 0 on success, negative error code on failure. + */ +int vmw_dx_streamoutput_remove(struct vmw_cmdbuf_res_manager *man, + u32 user_key, + struct list_head *list) +{ + struct vmw_resource *r; + + return vmw_cmdbuf_res_remove(man, vmw_cmdbuf_res_streamoutput, + (u32)user_key, list, &r); +} + +/** + * vmw_dx_streamoutput_cotable_list_scrub - cotable unbind_func callback. + * @dev_priv: Device private. + * @list: The list of cotable resources. + * @readback: Whether the call was part of a readback unbind. + */ +void vmw_dx_streamoutput_cotable_list_scrub(struct vmw_private *dev_priv, + struct list_head *list, + bool readback) +{ + struct vmw_dx_streamoutput *entry, *next; + + lockdep_assert_held_once(&dev_priv->binding_mutex); + + list_for_each_entry_safe(entry, next, list, cotable_head) { + WARN_ON(vmw_dx_streamoutput_scrub(&entry->res)); + if (!readback) + entry->committed =false; + } +} diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c b/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c index 3ce630aa4fde..7ef51fa84b01 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c @@ -79,7 +79,7 @@ struct vmw_surface_dirty { struct svga3dsurface_cache cache; size_t size; u32 num_subres; - SVGA3dBox boxes[0]; + SVGA3dBox boxes[]; }; static void vmw_user_surface_free(struct vmw_resource *res); @@ -199,7 +199,7 @@ struct vmw_surface_destroy { */ static inline uint32_t vmw_surface_dma_size(const struct vmw_surface *srf) { - return srf->num_sizes * sizeof(struct vmw_surface_dma); + return srf->metadata.num_sizes * sizeof(struct vmw_surface_dma); } @@ -213,7 +213,7 @@ static inline uint32_t vmw_surface_dma_size(const struct vmw_surface *srf) */ static inline uint32_t vmw_surface_define_size(const struct vmw_surface *srf) { - return sizeof(struct vmw_surface_define) + srf->num_sizes * + return sizeof(struct vmw_surface_define) + srf->metadata.num_sizes * sizeof(SVGA3dSize); } @@ -262,7 +262,8 @@ static void vmw_surface_define_encode(const struct vmw_surface *srf, uint32_t cmd_len; int i; - cmd_len = sizeof(cmd->body) + srf->num_sizes * sizeof(SVGA3dSize); + cmd_len = sizeof(cmd->body) + srf->metadata.num_sizes * + sizeof(SVGA3dSize); cmd->header.id = SVGA_3D_CMD_SURFACE_DEFINE; cmd->header.size = cmd_len; @@ -272,16 +273,16 @@ static void vmw_surface_define_encode(const struct vmw_surface *srf, * since driver internally stores as 64 bit. * For legacy surface define only 32 bit flag is supported. */ - cmd->body.surfaceFlags = (SVGA3dSurface1Flags)srf->flags; - cmd->body.format = srf->format; + cmd->body.surfaceFlags = (SVGA3dSurface1Flags)srf->metadata.flags; + cmd->body.format = srf->metadata.format; for (i = 0; i < DRM_VMW_MAX_SURFACE_FACES; ++i) - cmd->body.face[i].numMipLevels = srf->mip_levels[i]; + cmd->body.face[i].numMipLevels = srf->metadata.mip_levels[i]; cmd += 1; cmd_size = (SVGA3dSize *) cmd; - src_size = srf->sizes; + src_size = srf->metadata.sizes; - for (i = 0; i < srf->num_sizes; ++i, cmd_size++, src_size++) { + for (i = 0; i < srf->metadata.num_sizes; ++i, cmd_size++, src_size++) { cmd_size->width = src_size->width; cmd_size->height = src_size->height; cmd_size->depth = src_size->depth; @@ -305,15 +306,15 @@ static void vmw_surface_dma_encode(struct vmw_surface *srf, uint32_t i; struct vmw_surface_dma *cmd = (struct vmw_surface_dma *)cmd_space; const struct svga3d_surface_desc *desc = - svga3dsurface_get_desc(srf->format); + svga3dsurface_get_desc(srf->metadata.format); - for (i = 0; i < srf->num_sizes; ++i) { + for (i = 0; i < srf->metadata.num_sizes; ++i) { SVGA3dCmdHeader *header = &cmd->header; SVGA3dCmdSurfaceDMA *body = &cmd->body; SVGA3dCopyBox *cb = &cmd->cb; SVGA3dCmdSurfaceDMASuffix *suffix = &cmd->suffix; const struct vmw_surface_offset *cur_offset = &srf->offsets[i]; - const struct drm_vmw_size *cur_size = &srf->sizes[i]; + const struct drm_vmw_size *cur_size = &srf->metadata.sizes[i]; header->id = SVGA_3D_CMD_SURFACE_DMA; header->size = sizeof(*body) + sizeof(*cb) + sizeof(*suffix); @@ -669,7 +670,7 @@ static void vmw_user_surface_free(struct vmw_resource *res) if (user_srf->master) drm_master_put(&user_srf->master); kfree(srf->offsets); - kfree(srf->sizes); + kfree(srf->metadata.sizes); kfree(srf->snooper.image); ttm_prime_object_kfree(user_srf, prime); ttm_mem_global_free(vmw_mem_glob(dev_priv), size); @@ -728,6 +729,7 @@ int vmw_surface_define_ioctl(struct drm_device *dev, void *data, struct vmw_private *dev_priv = vmw_priv(dev); struct vmw_user_surface *user_srf; struct vmw_surface *srf; + struct vmw_surface_metadata *metadata; struct vmw_resource *res; struct vmw_resource *tmp; union drm_vmw_surface_create_arg *arg = @@ -793,43 +795,45 @@ int vmw_surface_define_ioctl(struct drm_device *dev, void *data, } srf = &user_srf->srf; + metadata = &srf->metadata; res = &srf->res; /* Driver internally stores as 64-bit flags */ - srf->flags = (SVGA3dSurfaceAllFlags)req->flags; - srf->format = req->format; - srf->scanout = req->scanout; + metadata->flags = (SVGA3dSurfaceAllFlags)req->flags; + metadata->format = req->format; + metadata->scanout = req->scanout; - memcpy(srf->mip_levels, req->mip_levels, sizeof(srf->mip_levels)); - srf->num_sizes = num_sizes; + memcpy(metadata->mip_levels, req->mip_levels, + sizeof(metadata->mip_levels)); + metadata->num_sizes = num_sizes; user_srf->size = size; - srf->sizes = memdup_user((struct drm_vmw_size __user *)(unsigned long) - req->size_addr, - sizeof(*srf->sizes) * srf->num_sizes); - if (IS_ERR(srf->sizes)) { - ret = PTR_ERR(srf->sizes); + metadata->sizes = + memdup_user((struct drm_vmw_size __user *)(unsigned long) + req->size_addr, + sizeof(*metadata->sizes) * metadata->num_sizes); + if (IS_ERR(metadata->sizes)) { + ret = PTR_ERR(metadata->sizes); goto out_no_sizes; } - srf->offsets = kmalloc_array(srf->num_sizes, - sizeof(*srf->offsets), + srf->offsets = kmalloc_array(metadata->num_sizes, sizeof(*srf->offsets), GFP_KERNEL); if (unlikely(!srf->offsets)) { ret = -ENOMEM; goto out_no_offsets; } - srf->base_size = *srf->sizes; - srf->autogen_filter = SVGA3D_TEX_FILTER_NONE; - srf->multisample_count = 0; - srf->multisample_pattern = SVGA3D_MS_PATTERN_NONE; - srf->quality_level = SVGA3D_MS_QUALITY_NONE; + metadata->base_size = *srf->metadata.sizes; + metadata->autogen_filter = SVGA3D_TEX_FILTER_NONE; + metadata->multisample_count = 0; + metadata->multisample_pattern = SVGA3D_MS_PATTERN_NONE; + metadata->quality_level = SVGA3D_MS_QUALITY_NONE; cur_bo_offset = 0; cur_offset = srf->offsets; - cur_size = srf->sizes; + cur_size = metadata->sizes; for (i = 0; i < DRM_VMW_MAX_SURFACE_FACES; ++i) { - for (j = 0; j < srf->mip_levels[i]; ++j) { + for (j = 0; j < metadata->mip_levels[i]; ++j) { uint32_t stride = svga3dsurface_calculate_pitch (desc, cur_size); @@ -843,11 +847,11 @@ int vmw_surface_define_ioctl(struct drm_device *dev, void *data, } } res->backup_size = cur_bo_offset; - if (srf->scanout && - srf->num_sizes == 1 && - srf->sizes[0].width == 64 && - srf->sizes[0].height == 64 && - srf->format == SVGA3D_A8R8G8B8) { + if (metadata->scanout && + metadata->num_sizes == 1 && + metadata->sizes[0].width == 64 && + metadata->sizes[0].height == 64 && + metadata->format == SVGA3D_A8R8G8B8) { srf->snooper.image = kzalloc(64 * 64 * 4, GFP_KERNEL); if (!srf->snooper.image) { @@ -911,7 +915,7 @@ int vmw_surface_define_ioctl(struct drm_device *dev, void *data, out_no_copy: kfree(srf->offsets); out_no_offsets: - kfree(srf->sizes); + kfree(metadata->sizes); out_no_sizes: ttm_prime_object_kfree(user_srf, prime); out_no_user_srf: @@ -1031,18 +1035,19 @@ int vmw_surface_reference_ioctl(struct drm_device *dev, void *data, srf = &user_srf->srf; /* Downcast of flags when sending back to user space */ - rep->flags = (uint32_t)srf->flags; - rep->format = srf->format; - memcpy(rep->mip_levels, srf->mip_levels, sizeof(srf->mip_levels)); + rep->flags = (uint32_t)srf->metadata.flags; + rep->format = srf->metadata.format; + memcpy(rep->mip_levels, srf->metadata.mip_levels, + sizeof(srf->metadata.mip_levels)); user_sizes = (struct drm_vmw_size __user *)(unsigned long) rep->size_addr; if (user_sizes) - ret = copy_to_user(user_sizes, &srf->base_size, - sizeof(srf->base_size)); + ret = copy_to_user(user_sizes, &srf->metadata.base_size, + sizeof(srf->metadata.base_size)); if (unlikely(ret != 0)) { VMW_DEBUG_USER("copy_to_user failed %p %u\n", user_sizes, - srf->num_sizes); + srf->metadata.num_sizes); ttm_ref_object_base_unref(tfile, base->handle, TTM_REF_USAGE); ret = -EFAULT; } @@ -1062,6 +1067,7 @@ static int vmw_gb_surface_create(struct vmw_resource *res) { struct vmw_private *dev_priv = res->dev_priv; struct vmw_surface *srf = vmw_res_to_srf(res); + struct vmw_surface_metadata *metadata = &srf->metadata; uint32_t cmd_len, cmd_id, submit_len; int ret; struct { @@ -1076,6 +1082,10 @@ static int vmw_gb_surface_create(struct vmw_resource *res) SVGA3dCmdHeader header; SVGA3dCmdDefineGBSurface_v3 body; } *cmd3; + struct { + SVGA3dCmdHeader header; + SVGA3dCmdDefineGBSurface_v4 body; + } *cmd4; if (likely(res->id != -1)) return 0; @@ -1092,12 +1102,16 @@ static int vmw_gb_surface_create(struct vmw_resource *res) goto out_no_fifo; } - if (dev_priv->has_sm4_1 && srf->array_size > 0) { + if (has_sm5_context(dev_priv) && metadata->array_size > 0) { + cmd_id = SVGA_3D_CMD_DEFINE_GB_SURFACE_V4; + cmd_len = sizeof(cmd4->body); + submit_len = sizeof(*cmd4); + } else if (has_sm4_1_context(dev_priv) && metadata->array_size > 0) { cmd_id = SVGA_3D_CMD_DEFINE_GB_SURFACE_V3; cmd_len = sizeof(cmd3->body); submit_len = sizeof(*cmd3); - } else if (srf->array_size > 0) { - /* has_dx checked on creation time. */ + } else if (metadata->array_size > 0) { + /* VMW_SM_4 support verified at creation time. */ cmd_id = SVGA_3D_CMD_DEFINE_GB_SURFACE_V2; cmd_len = sizeof(cmd2->body); submit_len = sizeof(*cmd2); @@ -1110,51 +1124,68 @@ static int vmw_gb_surface_create(struct vmw_resource *res) cmd = VMW_FIFO_RESERVE(dev_priv, submit_len); cmd2 = (typeof(cmd2))cmd; cmd3 = (typeof(cmd3))cmd; + cmd4 = (typeof(cmd4))cmd; if (unlikely(!cmd)) { ret = -ENOMEM; goto out_no_fifo; } - if (dev_priv->has_sm4_1 && srf->array_size > 0) { + if (has_sm5_context(dev_priv) && metadata->array_size > 0) { + cmd4->header.id = cmd_id; + cmd4->header.size = cmd_len; + cmd4->body.sid = srf->res.id; + cmd4->body.surfaceFlags = metadata->flags; + cmd4->body.format = metadata->format; + cmd4->body.numMipLevels = metadata->mip_levels[0]; + cmd4->body.multisampleCount = metadata->multisample_count; + cmd4->body.multisamplePattern = metadata->multisample_pattern; + cmd4->body.qualityLevel = metadata->quality_level; + cmd4->body.autogenFilter = metadata->autogen_filter; + cmd4->body.size.width = metadata->base_size.width; + cmd4->body.size.height = metadata->base_size.height; + cmd4->body.size.depth = metadata->base_size.depth; + cmd4->body.arraySize = metadata->array_size; + cmd4->body.bufferByteStride = metadata->buffer_byte_stride; + } else if (has_sm4_1_context(dev_priv) && metadata->array_size > 0) { cmd3->header.id = cmd_id; cmd3->header.size = cmd_len; cmd3->body.sid = srf->res.id; - cmd3->body.surfaceFlags = srf->flags; - cmd3->body.format = srf->format; - cmd3->body.numMipLevels = srf->mip_levels[0]; - cmd3->body.multisampleCount = srf->multisample_count; - cmd3->body.multisamplePattern = srf->multisample_pattern; - cmd3->body.qualityLevel = srf->quality_level; - cmd3->body.autogenFilter = srf->autogen_filter; - cmd3->body.size.width = srf->base_size.width; - cmd3->body.size.height = srf->base_size.height; - cmd3->body.size.depth = srf->base_size.depth; - cmd3->body.arraySize = srf->array_size; - } else if (srf->array_size > 0) { + cmd3->body.surfaceFlags = metadata->flags; + cmd3->body.format = metadata->format; + cmd3->body.numMipLevels = metadata->mip_levels[0]; + cmd3->body.multisampleCount = metadata->multisample_count; + cmd3->body.multisamplePattern = metadata->multisample_pattern; + cmd3->body.qualityLevel = metadata->quality_level; + cmd3->body.autogenFilter = metadata->autogen_filter; + cmd3->body.size.width = metadata->base_size.width; + cmd3->body.size.height = metadata->base_size.height; + cmd3->body.size.depth = metadata->base_size.depth; + cmd3->body.arraySize = metadata->array_size; + } else if (metadata->array_size > 0) { cmd2->header.id = cmd_id; cmd2->header.size = cmd_len; cmd2->body.sid = srf->res.id; - cmd2->body.surfaceFlags = srf->flags; - cmd2->body.format = srf->format; - cmd2->body.numMipLevels = srf->mip_levels[0]; - cmd2->body.multisampleCount = srf->multisample_count; - cmd2->body.autogenFilter = srf->autogen_filter; - cmd2->body.size.width = srf->base_size.width; - cmd2->body.size.height = srf->base_size.height; - cmd2->body.size.depth = srf->base_size.depth; - cmd2->body.arraySize = srf->array_size; + cmd2->body.surfaceFlags = metadata->flags; + cmd2->body.format = metadata->format; + cmd2->body.numMipLevels = metadata->mip_levels[0]; + cmd2->body.multisampleCount = metadata->multisample_count; + cmd2->body.autogenFilter = metadata->autogen_filter; + cmd2->body.size.width = metadata->base_size.width; + cmd2->body.size.height = metadata->base_size.height; + cmd2->body.size.depth = metadata->base_size.depth; + cmd2->body.arraySize = metadata->array_size; } else { cmd->header.id = cmd_id; cmd->header.size = cmd_len; cmd->body.sid = srf->res.id; - cmd->body.surfaceFlags = srf->flags; - cmd->body.format = srf->format; - cmd->body.numMipLevels = srf->mip_levels[0]; - cmd->body.multisampleCount = srf->multisample_count; - cmd->body.autogenFilter = srf->autogen_filter; - cmd->body.size.width = srf->base_size.width; - cmd->body.size.height = srf->base_size.height; - cmd->body.size.depth = srf->base_size.depth; + cmd->body.surfaceFlags = metadata->flags; + cmd->body.format = metadata->format; + cmd->body.numMipLevels = metadata->mip_levels[0]; + cmd->body.multisampleCount = metadata->multisample_count; + cmd->body.autogenFilter = metadata->autogen_filter; + cmd->body.size.width = metadata->base_size.width; + cmd->body.size.height = metadata->base_size.height; + cmd->body.size.depth = metadata->base_size.depth; } vmw_fifo_commit(dev_priv, submit_len); @@ -1314,7 +1345,6 @@ static int vmw_gb_surface_destroy(struct vmw_resource *res) return 0; } - /** * vmw_gb_surface_define_ioctl - Ioctl function implementing * the user surface define functionality. @@ -1336,6 +1366,7 @@ int vmw_gb_surface_define_ioctl(struct drm_device *dev, void *data, req_ext.svga3d_flags_upper_32_bits = 0; req_ext.multisample_pattern = SVGA3D_MS_PATTERN_NONE; req_ext.quality_level = SVGA3D_MS_QUALITY_NONE; + req_ext.buffer_byte_stride = 0; req_ext.must_be_zero = 0; return vmw_gb_surface_define_internal(dev, &req_ext, rep, file_priv); @@ -1371,171 +1402,6 @@ int vmw_gb_surface_reference_ioctl(struct drm_device *dev, void *data, } /** - * vmw_surface_gb_priv_define - Define a private GB surface - * - * @dev: Pointer to a struct drm_device - * @user_accounting_size: Used to track user-space memory usage, set - * to 0 for kernel mode only memory - * @svga3d_flags: SVGA3d surface flags for the device - * @format: requested surface format - * @for_scanout: true if inteded to be used for scanout buffer - * @num_mip_levels: number of MIP levels - * @multisample_count: - * @array_size: Surface array size. - * @size: width, heigh, depth of the surface requested - * @multisample_pattern: Multisampling pattern when msaa is supported - * @quality_level: Precision settings - * @user_srf_out: allocated user_srf. Set to NULL on failure. - * - * GB surfaces allocated by this function will not have a user mode handle, and - * thus will only be visible to vmwgfx. For optimization reasons the - * surface may later be given a user mode handle by another function to make - * it available to user mode drivers. - */ -int vmw_surface_gb_priv_define(struct drm_device *dev, - uint32_t user_accounting_size, - SVGA3dSurfaceAllFlags svga3d_flags, - SVGA3dSurfaceFormat format, - bool for_scanout, - uint32_t num_mip_levels, - uint32_t multisample_count, - uint32_t array_size, - struct drm_vmw_size size, - SVGA3dMSPattern multisample_pattern, - SVGA3dMSQualityLevel quality_level, - struct vmw_surface **srf_out) -{ - struct vmw_private *dev_priv = vmw_priv(dev); - struct vmw_user_surface *user_srf; - struct ttm_operation_ctx ctx = { - .interruptible = true, - .no_wait_gpu = false - }; - struct vmw_surface *srf; - int ret; - u32 num_layers = 1; - u32 sample_count = 1; - - *srf_out = NULL; - - if (for_scanout) { - if (!svga3dsurface_is_screen_target_format(format)) { - VMW_DEBUG_USER("Invalid Screen Target surface format."); - return -EINVAL; - } - - if (size.width > dev_priv->texture_max_width || - size.height > dev_priv->texture_max_height) { - VMW_DEBUG_USER("%ux%u\n, exceeds max surface size %ux%u", - size.width, size.height, - dev_priv->texture_max_width, - dev_priv->texture_max_height); - return -EINVAL; - } - } else { - const struct svga3d_surface_desc *desc; - - desc = svga3dsurface_get_desc(format); - if (unlikely(desc->block_desc == SVGA3DBLOCKDESC_NONE)) { - VMW_DEBUG_USER("Invalid surface format.\n"); - return -EINVAL; - } - } - - /* array_size must be null for non-GL3 host. */ - if (array_size > 0 && !dev_priv->has_dx) { - VMW_DEBUG_USER("Tried to create DX surface on non-DX host.\n"); - return -EINVAL; - } - - ret = ttm_read_lock(&dev_priv->reservation_sem, true); - if (unlikely(ret != 0)) - return ret; - - ret = ttm_mem_global_alloc(vmw_mem_glob(dev_priv), - user_accounting_size, &ctx); - if (unlikely(ret != 0)) { - if (ret != -ERESTARTSYS) - DRM_ERROR("Out of graphics memory for surface" - " creation.\n"); - goto out_unlock; - } - - user_srf = kzalloc(sizeof(*user_srf), GFP_KERNEL); - if (unlikely(!user_srf)) { - ret = -ENOMEM; - goto out_no_user_srf; - } - - *srf_out = &user_srf->srf; - user_srf->size = user_accounting_size; - user_srf->prime.base.shareable = false; - user_srf->prime.base.tfile = NULL; - - srf = &user_srf->srf; - srf->flags = svga3d_flags; - srf->format = format; - srf->scanout = for_scanout; - srf->mip_levels[0] = num_mip_levels; - srf->num_sizes = 1; - srf->sizes = NULL; - srf->offsets = NULL; - srf->base_size = size; - srf->autogen_filter = SVGA3D_TEX_FILTER_NONE; - srf->array_size = array_size; - srf->multisample_count = multisample_count; - srf->multisample_pattern = multisample_pattern; - srf->quality_level = quality_level; - - if (array_size) - num_layers = array_size; - else if (svga3d_flags & SVGA3D_SURFACE_CUBEMAP) - num_layers = SVGA3D_MAX_SURFACE_FACES; - - if (srf->flags & SVGA3D_SURFACE_MULTISAMPLE) - sample_count = srf->multisample_count; - - srf->res.backup_size = - svga3dsurface_get_serialized_size_extended(srf->format, - srf->base_size, - srf->mip_levels[0], - num_layers, - sample_count); - - if (srf->flags & SVGA3D_SURFACE_BIND_STREAM_OUTPUT) - srf->res.backup_size += sizeof(SVGA3dDXSOState); - - /* - * Don't set SVGA3D_SURFACE_SCREENTARGET flag for a scanout surface with - * size greater than STDU max width/height. This is really a workaround - * to support creation of big framebuffer requested by some user-space - * for whole topology. That big framebuffer won't really be used for - * binding with screen target as during prepare_fb a separate surface is - * created so it's safe to ignore SVGA3D_SURFACE_SCREENTARGET flag. - */ - if (dev_priv->active_display_unit == vmw_du_screen_target && - for_scanout && size.width <= dev_priv->stdu_max_width && - size.height <= dev_priv->stdu_max_height) - srf->flags |= SVGA3D_SURFACE_SCREENTARGET; - - /* - * From this point, the generic resource management functions - * destroy the object on failure. - */ - ret = vmw_surface_init(dev_priv, srf, vmw_user_surface_free); - - ttm_read_unlock(&dev_priv->reservation_sem); - return ret; - -out_no_user_srf: - ttm_mem_global_free(vmw_mem_glob(dev_priv), user_accounting_size); - -out_unlock: - ttm_read_unlock(&dev_priv->reservation_sem); - return ret; -} - -/** * vmw_gb_surface_define_ext_ioctl - Ioctl function implementing * the user surface define functionality. * @@ -1588,43 +1454,60 @@ vmw_gb_surface_define_internal(struct drm_device *dev, struct drm_vmw_gb_surface_create_rep *rep, struct drm_file *file_priv) { + struct ttm_object_file *tfile = vmw_fpriv(file_priv)->tfile; struct vmw_private *dev_priv = vmw_priv(dev); struct vmw_user_surface *user_srf; + struct vmw_surface_metadata metadata = {0}; struct vmw_surface *srf; struct vmw_resource *res; struct vmw_resource *tmp; - struct ttm_object_file *tfile = vmw_fpriv(file_priv)->tfile; - int ret; + int ret = 0; uint32_t size; uint32_t backup_handle = 0; SVGA3dSurfaceAllFlags svga3d_flags_64 = SVGA3D_FLAGS_64(req->svga3d_flags_upper_32_bits, req->base.svga3d_flags); - if (!dev_priv->has_sm4_1) { - /* - * If SM4_1 is not support then cannot send 64-bit flag to - * device. - */ + /* array_size must be null for non-GL3 host. */ + if (req->base.array_size > 0 && !has_sm4_context(dev_priv)) { + VMW_DEBUG_USER("SM4 surface not supported.\n"); + return -EINVAL; + } + + if (!has_sm4_1_context(dev_priv)) { if (req->svga3d_flags_upper_32_bits != 0) - return -EINVAL; + ret = -EINVAL; if (req->base.multisample_count != 0) - return -EINVAL; + ret = -EINVAL; if (req->multisample_pattern != SVGA3D_MS_PATTERN_NONE) - return -EINVAL; + ret = -EINVAL; if (req->quality_level != SVGA3D_MS_QUALITY_NONE) - return -EINVAL; + ret = -EINVAL; + + if (ret) { + VMW_DEBUG_USER("SM4.1 surface not supported.\n"); + return ret; + } + } + + if (req->buffer_byte_stride > 0 && !has_sm5_context(dev_priv)) { + VMW_DEBUG_USER("SM5 surface not supported.\n"); + return -EINVAL; } if ((svga3d_flags_64 & SVGA3D_SURFACE_MULTISAMPLE) && - req->base.multisample_count == 0) + req->base.multisample_count == 0) { + VMW_DEBUG_USER("Invalid sample count.\n"); return -EINVAL; + } - if (req->base.mip_levels > DRM_VMW_MAX_MIP_LEVELS) + if (req->base.mip_levels > DRM_VMW_MAX_MIP_LEVELS) { + VMW_DEBUG_USER("Invalid mip level.\n"); return -EINVAL; + } if (unlikely(vmw_user_surface_size == 0)) vmw_user_surface_size = ttm_round_pot(sizeof(*user_srf)) + @@ -1632,22 +1515,25 @@ vmw_gb_surface_define_internal(struct drm_device *dev, size = vmw_user_surface_size; + metadata.flags = svga3d_flags_64; + metadata.format = req->base.format; + metadata.mip_levels[0] = req->base.mip_levels; + metadata.multisample_count = req->base.multisample_count; + metadata.multisample_pattern = req->multisample_pattern; + metadata.quality_level = req->quality_level; + metadata.array_size = req->base.array_size; + metadata.buffer_byte_stride = req->buffer_byte_stride; + metadata.num_sizes = 1; + metadata.base_size = req->base.base_size; + metadata.scanout = req->base.drm_surface_flags & + drm_vmw_surface_flag_scanout; + /* Define a surface based on the parameters. */ - ret = vmw_surface_gb_priv_define(dev, - size, - svga3d_flags_64, - req->base.format, - req->base.drm_surface_flags & - drm_vmw_surface_flag_scanout, - req->base.mip_levels, - req->base.multisample_count, - req->base.array_size, - req->base.base_size, - req->multisample_pattern, - req->quality_level, - &srf); - if (unlikely(ret != 0)) + ret = vmw_gb_surface_define(dev_priv, size, &metadata, &srf); + if (ret != 0) { + VMW_DEBUG_USER("Failed to define surface.\n"); return ret; + } user_srf = container_of(srf, struct vmw_user_surface, srf); if (drm_is_primary_client(file_priv)) @@ -1762,6 +1648,7 @@ vmw_gb_surface_reference_internal(struct drm_device *dev, struct ttm_object_file *tfile = vmw_fpriv(file_priv)->tfile; struct vmw_surface *srf; struct vmw_user_surface *user_srf; + struct vmw_surface_metadata *metadata; struct ttm_base_object *base; uint32_t backup_handle; int ret = -EINVAL; @@ -1777,6 +1664,7 @@ vmw_gb_surface_reference_internal(struct drm_device *dev, DRM_ERROR("Shared GB surface is missing a backup buffer.\n"); goto out_bad_resource; } + metadata = &srf->metadata; mutex_lock(&dev_priv->cmdbuf_mutex); /* Protect res->backup */ ret = vmw_user_bo_reference(tfile, srf->res.backup, &backup_handle); @@ -1790,15 +1678,15 @@ vmw_gb_surface_reference_internal(struct drm_device *dev, goto out_bad_resource; } - rep->creq.base.svga3d_flags = SVGA3D_FLAGS_LOWER_32(srf->flags); - rep->creq.base.format = srf->format; - rep->creq.base.mip_levels = srf->mip_levels[0]; + rep->creq.base.svga3d_flags = SVGA3D_FLAGS_LOWER_32(metadata->flags); + rep->creq.base.format = metadata->format; + rep->creq.base.mip_levels = metadata->mip_levels[0]; rep->creq.base.drm_surface_flags = 0; - rep->creq.base.multisample_count = srf->multisample_count; - rep->creq.base.autogen_filter = srf->autogen_filter; - rep->creq.base.array_size = srf->array_size; + rep->creq.base.multisample_count = metadata->multisample_count; + rep->creq.base.autogen_filter = metadata->autogen_filter; + rep->creq.base.array_size = metadata->array_size; rep->creq.base.buffer_handle = backup_handle; - rep->creq.base.base_size = srf->base_size; + rep->creq.base.base_size = metadata->base_size; rep->crep.handle = user_srf->prime.base.handle; rep->crep.backup_size = srf->res.backup_size; rep->crep.buffer_handle = backup_handle; @@ -1808,9 +1696,9 @@ vmw_gb_surface_reference_internal(struct drm_device *dev, rep->creq.version = drm_vmw_gb_surface_v1; rep->creq.svga3d_flags_upper_32_bits = - SVGA3D_FLAGS_UPPER_32(srf->flags); - rep->creq.multisample_pattern = srf->multisample_pattern; - rep->creq.quality_level = srf->quality_level; + SVGA3D_FLAGS_UPPER_32(metadata->flags); + rep->creq.multisample_pattern = metadata->multisample_pattern; + rep->creq.quality_level = metadata->quality_level; rep->creq.must_be_zero = 0; out_bad_resource: @@ -1968,7 +1856,7 @@ static void vmw_surface_dirty_range_add(struct vmw_resource *res, size_t start, start >= res->backup_offset + res->backup_size)) return; - if (srf->format == SVGA3D_BUFFER) + if (srf->metadata.format == SVGA3D_BUFFER) vmw_surface_buf_dirty_range_add(res, start, end); else vmw_surface_tex_dirty_range_add(res, start, end); @@ -2058,6 +1946,7 @@ static int vmw_surface_dirty_sync(struct vmw_resource *res) static int vmw_surface_dirty_alloc(struct vmw_resource *res) { struct vmw_surface *srf = vmw_res_to_srf(res); + const struct vmw_surface_metadata *metadata = &srf->metadata; struct vmw_surface_dirty *dirty; u32 num_layers = 1; u32 num_mip; @@ -2070,12 +1959,12 @@ static int vmw_surface_dirty_alloc(struct vmw_resource *res) }; int ret; - if (srf->array_size) - num_layers = srf->array_size; - else if (srf->flags & SVGA3D_SURFACE_CUBEMAP) + if (metadata->array_size) + num_layers = metadata->array_size; + else if (metadata->flags & SVGA3D_SURFACE_CUBEMAP) num_layers *= SVGA3D_MAX_SURFACE_FACES; - num_mip = srf->mip_levels[0]; + num_mip = metadata->mip_levels[0]; if (!num_mip) num_mip = 1; @@ -2096,9 +1985,10 @@ static int vmw_surface_dirty_alloc(struct vmw_resource *res) goto out_no_dirty; } - num_samples = max_t(u32, 1, srf->multisample_count); - ret = svga3dsurface_setup_cache(&srf->base_size, srf->format, num_mip, - num_layers, num_samples, &dirty->cache); + num_samples = max_t(u32, 1, metadata->multisample_count); + ret = svga3dsurface_setup_cache(&metadata->base_size, metadata->format, + num_mip, num_layers, num_samples, + &dirty->cache); if (ret) goto out_no_cache; @@ -2153,3 +2043,147 @@ static int vmw_surface_clean(struct vmw_resource *res) return 0; } + +/* + * vmw_gb_surface_define - Define a private GB surface + * + * @dev_priv: Pointer to a device private. + * @user_accounting_size: Used to track user-space memory usage, set + * to 0 for kernel mode only memory + * @metadata: Metadata representing the surface to create. + * @user_srf_out: allocated user_srf. Set to NULL on failure. + * + * GB surfaces allocated by this function will not have a user mode handle, and + * thus will only be visible to vmwgfx. For optimization reasons the + * surface may later be given a user mode handle by another function to make + * it available to user mode drivers. + */ +int vmw_gb_surface_define(struct vmw_private *dev_priv, + uint32_t user_accounting_size, + const struct vmw_surface_metadata *req, + struct vmw_surface **srf_out) +{ + struct vmw_surface_metadata *metadata; + struct vmw_user_surface *user_srf; + struct vmw_surface *srf; + struct ttm_operation_ctx ctx = { + .interruptible = true, + .no_wait_gpu = false + }; + u32 sample_count = 1; + u32 num_layers = 1; + int ret; + + *srf_out = NULL; + + if (req->scanout) { + if (!svga3dsurface_is_screen_target_format(req->format)) { + VMW_DEBUG_USER("Invalid Screen Target surface format."); + return -EINVAL; + } + + if (req->base_size.width > dev_priv->texture_max_width || + req->base_size.height > dev_priv->texture_max_height) { + VMW_DEBUG_USER("%ux%u\n, exceed max surface size %ux%u", + req->base_size.width, + req->base_size.height, + dev_priv->texture_max_width, + dev_priv->texture_max_height); + return -EINVAL; + } + } else { + const struct svga3d_surface_desc *desc = + svga3dsurface_get_desc(req->format); + + if (desc->block_desc == SVGA3DBLOCKDESC_NONE) { + VMW_DEBUG_USER("Invalid surface format.\n"); + return -EINVAL; + } + } + + if (req->autogen_filter != SVGA3D_TEX_FILTER_NONE) + return -EINVAL; + + if (req->num_sizes != 1) + return -EINVAL; + + if (req->sizes != NULL) + return -EINVAL; + + ret = ttm_read_lock(&dev_priv->reservation_sem, true); + if (unlikely(ret != 0)) + return ret; + + ret = ttm_mem_global_alloc(vmw_mem_glob(dev_priv), + user_accounting_size, &ctx); + if (ret != 0) { + if (ret != -ERESTARTSYS) + DRM_ERROR("Out of graphics memory for surface.\n"); + goto out_unlock; + } + + user_srf = kzalloc(sizeof(*user_srf), GFP_KERNEL); + if (unlikely(!user_srf)) { + ret = -ENOMEM; + goto out_no_user_srf; + } + + *srf_out = &user_srf->srf; + user_srf->size = user_accounting_size; + user_srf->prime.base.shareable = false; + user_srf->prime.base.tfile = NULL; + + srf = &user_srf->srf; + srf->metadata = *req; + srf->offsets = NULL; + + metadata = &srf->metadata; + + if (metadata->array_size) + num_layers = req->array_size; + else if (metadata->flags & SVGA3D_SURFACE_CUBEMAP) + num_layers = SVGA3D_MAX_SURFACE_FACES; + + if (metadata->flags & SVGA3D_SURFACE_MULTISAMPLE) + sample_count = metadata->multisample_count; + + srf->res.backup_size = + svga3dsurface_get_serialized_size_extended(metadata->format, + metadata->base_size, + metadata->mip_levels[0], + num_layers, + sample_count); + + if (metadata->flags & SVGA3D_SURFACE_BIND_STREAM_OUTPUT) + srf->res.backup_size += sizeof(SVGA3dDXSOState); + + /* + * Don't set SVGA3D_SURFACE_SCREENTARGET flag for a scanout surface with + * size greater than STDU max width/height. This is really a workaround + * to support creation of big framebuffer requested by some user-space + * for whole topology. That big framebuffer won't really be used for + * binding with screen target as during prepare_fb a separate surface is + * created so it's safe to ignore SVGA3D_SURFACE_SCREENTARGET flag. + */ + if (dev_priv->active_display_unit == vmw_du_screen_target && + metadata->scanout && + metadata->base_size.width <= dev_priv->stdu_max_width && + metadata->base_size.height <= dev_priv->stdu_max_height) + metadata->flags |= SVGA3D_SURFACE_SCREENTARGET; + + /* + * From this point, the generic resource management functions + * destroy the object on failure. + */ + ret = vmw_surface_init(dev_priv, srf, vmw_user_surface_free); + + ttm_read_unlock(&dev_priv->reservation_sem); + return ret; + +out_no_user_srf: + ttm_mem_global_free(vmw_mem_glob(dev_priv), user_accounting_size); + +out_unlock: + ttm_read_unlock(&dev_priv->reservation_sem); + return ret; +} diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_thp.c b/drivers/gpu/drm/vmwgfx/vmwgfx_thp.c new file mode 100644 index 000000000000..b7c816ba7166 --- /dev/null +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_thp.c @@ -0,0 +1,166 @@ +// SPDX-License-Identifier: GPL-2.0 OR MIT +/* + * Huge page-table-entry support for IO memory. + * + * Copyright (C) 2007-2019 Vmware, Inc. All rights reservedd. + */ +#include "vmwgfx_drv.h" +#include <drm/ttm/ttm_module.h> +#include <drm/ttm/ttm_bo_driver.h> +#include <drm/ttm/ttm_placement.h> + +/** + * struct vmw_thp_manager - Range manager implementing huge page alignment + * + * @mm: The underlying range manager. Protected by @lock. + * @lock: Manager lock. + */ +struct vmw_thp_manager { + struct drm_mm mm; + spinlock_t lock; +}; + +static int vmw_thp_insert_aligned(struct drm_mm *mm, struct drm_mm_node *node, + unsigned long align_pages, + const struct ttm_place *place, + struct ttm_mem_reg *mem, + unsigned long lpfn, + enum drm_mm_insert_mode mode) +{ + if (align_pages >= mem->page_alignment && + (!mem->page_alignment || align_pages % mem->page_alignment == 0)) { + return drm_mm_insert_node_in_range(mm, node, + mem->num_pages, + align_pages, 0, + place->fpfn, lpfn, mode); + } + + return -ENOSPC; +} + +static int vmw_thp_get_node(struct ttm_mem_type_manager *man, + struct ttm_buffer_object *bo, + const struct ttm_place *place, + struct ttm_mem_reg *mem) +{ + struct vmw_thp_manager *rman = (struct vmw_thp_manager *) man->priv; + struct drm_mm *mm = &rman->mm; + struct drm_mm_node *node; + unsigned long align_pages; + unsigned long lpfn; + enum drm_mm_insert_mode mode = DRM_MM_INSERT_BEST; + int ret; + + node = kzalloc(sizeof(*node), GFP_KERNEL); + if (!node) + return -ENOMEM; + + lpfn = place->lpfn; + if (!lpfn) + lpfn = man->size; + + mode = DRM_MM_INSERT_BEST; + if (place->flags & TTM_PL_FLAG_TOPDOWN) + mode = DRM_MM_INSERT_HIGH; + + spin_lock(&rman->lock); + if (IS_ENABLED(CONFIG_HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD)) { + align_pages = (HPAGE_PUD_SIZE >> PAGE_SHIFT); + if (mem->num_pages >= align_pages) { + ret = vmw_thp_insert_aligned(mm, node, align_pages, + place, mem, lpfn, mode); + if (!ret) + goto found_unlock; + } + } + + align_pages = (HPAGE_PMD_SIZE >> PAGE_SHIFT); + if (mem->num_pages >= align_pages) { + ret = vmw_thp_insert_aligned(mm, node, align_pages, place, mem, + lpfn, mode); + if (!ret) + goto found_unlock; + } + + ret = drm_mm_insert_node_in_range(mm, node, mem->num_pages, + mem->page_alignment, 0, + place->fpfn, lpfn, mode); +found_unlock: + spin_unlock(&rman->lock); + + if (unlikely(ret)) { + kfree(node); + } else { + mem->mm_node = node; + mem->start = node->start; + } + + return 0; +} + + + +static void vmw_thp_put_node(struct ttm_mem_type_manager *man, + struct ttm_mem_reg *mem) +{ + struct vmw_thp_manager *rman = (struct vmw_thp_manager *) man->priv; + + if (mem->mm_node) { + spin_lock(&rman->lock); + drm_mm_remove_node(mem->mm_node); + spin_unlock(&rman->lock); + + kfree(mem->mm_node); + mem->mm_node = NULL; + } +} + +static int vmw_thp_init(struct ttm_mem_type_manager *man, + unsigned long p_size) +{ + struct vmw_thp_manager *rman; + + rman = kzalloc(sizeof(*rman), GFP_KERNEL); + if (!rman) + return -ENOMEM; + + drm_mm_init(&rman->mm, 0, p_size); + spin_lock_init(&rman->lock); + man->priv = rman; + return 0; +} + +static int vmw_thp_takedown(struct ttm_mem_type_manager *man) +{ + struct vmw_thp_manager *rman = (struct vmw_thp_manager *) man->priv; + struct drm_mm *mm = &rman->mm; + + spin_lock(&rman->lock); + if (drm_mm_clean(mm)) { + drm_mm_takedown(mm); + spin_unlock(&rman->lock); + kfree(rman); + man->priv = NULL; + return 0; + } + spin_unlock(&rman->lock); + return -EBUSY; +} + +static void vmw_thp_debug(struct ttm_mem_type_manager *man, + struct drm_printer *printer) +{ + struct vmw_thp_manager *rman = (struct vmw_thp_manager *) man->priv; + + spin_lock(&rman->lock); + drm_mm_print(&rman->mm, printer); + spin_unlock(&rman->lock); +} + +const struct ttm_mem_type_manager_func vmw_thp_func = { + .init = vmw_thp_init, + .takedown = vmw_thp_takedown, + .get_node = vmw_thp_get_node, + .put_node = vmw_thp_put_node, + .debug = vmw_thp_debug +}; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c index 3f3b2c7a208a..bf0bc4697959 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c @@ -749,7 +749,7 @@ static int vmw_init_mem_type(struct ttm_bo_device *bdev, uint32_t type, break; case TTM_PL_VRAM: /* "On-card" video ram */ - man->func = &ttm_bo_manager_func; + man->func = &vmw_thp_func; man->gpu_offset = 0; man->flags = TTM_MEMTYPE_FLAG_FIXED | TTM_MEMTYPE_FLAG_MAPPABLE; man->available_caching = TTM_PL_FLAG_CACHED; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_glue.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_glue.c index aa7e50f63b94..3c03b1746661 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_glue.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_glue.c @@ -34,7 +34,10 @@ int vmw_mmap(struct file *filp, struct vm_area_struct *vma) .page_mkwrite = vmw_bo_vm_mkwrite, .fault = vmw_bo_vm_fault, .open = ttm_bo_vm_open, - .close = ttm_bo_vm_close + .close = ttm_bo_vm_close, +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + .huge_fault = vmw_bo_vm_huge_fault, +#endif }; struct drm_file *file_priv = filp->private_data; struct vmw_private *dev_priv = vmw_priv(file_priv->minor->dev); diff --git a/drivers/gpu/drm/xen/xen_drm_front.c b/drivers/gpu/drm/xen/xen_drm_front.c index 4be49c1aef51..374142018171 100644 --- a/drivers/gpu/drm/xen/xen_drm_front.c +++ b/drivers/gpu/drm/xen/xen_drm_front.c @@ -401,7 +401,7 @@ static int xen_drm_drv_dumb_create(struct drm_file *filp, obj = xen_drm_front_gem_create(dev, args->size); if (IS_ERR_OR_NULL(obj)) { - ret = PTR_ERR(obj); + ret = PTR_ERR_OR_ZERO(obj); goto fail; } |