diff options
Diffstat (limited to 'drivers/gpu')
108 files changed, 2772 insertions, 671 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile index f8c58c425eb9..fdd0ca4b0f0b 100644 --- a/drivers/gpu/drm/amd/amdgpu/Makefile +++ b/drivers/gpu/drm/amd/amdgpu/Makefile @@ -23,7 +23,7 @@ # Makefile for the drm device driver. This driver provides support for the # Direct Rendering Infrastructure (DRI) in XFree86 4.1.0 and higher. -FULL_AMD_PATH=$(src)/.. +FULL_AMD_PATH=$(srctree)/$(src)/.. DISPLAY_FOLDER_NAME=display FULL_AMD_DISPLAY_PATH = $(FULL_AMD_PATH)/$(DISPLAY_FOLDER_NAME) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c index 4376b17ca594..56f8ca2a3bb4 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c @@ -464,8 +464,7 @@ static int amdgpu_atif_handler(struct amdgpu_device *adev, } } if (req.pending & ATIF_DGPU_DISPLAY_EVENT) { - if ((adev->flags & AMD_IS_PX) && - amdgpu_atpx_dgpu_req_power_for_displays()) { + if (adev->flags & AMD_IS_PX) { pm_runtime_get_sync(adev->ddev->dev); /* Just fire off a uevent and let userspace tell us what to do */ drm_helper_hpd_irq_event(adev->ddev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c index acf8ae0cee9a..aeead072fa79 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c @@ -335,6 +335,43 @@ void amdgpu_amdkfd_free_gtt_mem(struct kgd_dev *kgd, void *mem_obj) amdgpu_bo_unref(&(bo)); } +uint32_t amdgpu_amdkfd_get_fw_version(struct kgd_dev *kgd, + enum kgd_engine_type type) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)kgd; + + switch (type) { + case KGD_ENGINE_PFP: + return adev->gfx.pfp_fw_version; + + case KGD_ENGINE_ME: + return adev->gfx.me_fw_version; + + case KGD_ENGINE_CE: + return adev->gfx.ce_fw_version; + + case KGD_ENGINE_MEC1: + return adev->gfx.mec_fw_version; + + case KGD_ENGINE_MEC2: + return adev->gfx.mec2_fw_version; + + case KGD_ENGINE_RLC: + return adev->gfx.rlc_fw_version; + + case KGD_ENGINE_SDMA1: + return adev->sdma.instance[0].fw_version; + + case KGD_ENGINE_SDMA2: + return adev->sdma.instance[1].fw_version; + + default: + return 0; + } + + return 0; +} + void amdgpu_amdkfd_get_local_mem_info(struct kgd_dev *kgd, struct kfd_local_mem_info *mem_info) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h index 775f815f9521..4e37fa7e85b1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h @@ -81,6 +81,18 @@ struct amdgpu_kfd_dev { uint64_t vram_used; }; +enum kgd_engine_type { + KGD_ENGINE_PFP = 1, + KGD_ENGINE_ME, + KGD_ENGINE_CE, + KGD_ENGINE_MEC1, + KGD_ENGINE_MEC2, + KGD_ENGINE_RLC, + KGD_ENGINE_SDMA1, + KGD_ENGINE_SDMA2, + KGD_ENGINE_MAX +}; + 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); @@ -142,6 +154,8 @@ 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 amdgpu_amdkfd_free_gtt_mem(struct kgd_dev *kgd, void *mem_obj); +uint32_t amdgpu_amdkfd_get_fw_version(struct kgd_dev *kgd, + enum kgd_engine_type type); void amdgpu_amdkfd_get_local_mem_info(struct kgd_dev *kgd, struct kfd_local_mem_info *mem_info); uint64_t amdgpu_amdkfd_get_gpu_clock_counter(struct kgd_dev *kgd); 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 ff7fac7df34b..fa09e11a600c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c @@ -22,14 +22,12 @@ #include <linux/fdtable.h> #include <linux/uaccess.h> -#include <linux/firmware.h> #include <linux/mmu_context.h> #include <drm/drmP.h> #include "amdgpu.h" #include "amdgpu_amdkfd.h" #include "cikd.h" #include "cik_sdma.h" -#include "amdgpu_ucode.h" #include "gfx_v7_0.h" #include "gca/gfx_7_2_d.h" #include "gca/gfx_7_2_enum.h" @@ -139,7 +137,6 @@ static bool get_atc_vmid_pasid_mapping_valid(struct kgd_dev *kgd, uint8_t vmid); static uint16_t get_atc_vmid_pasid_mapping_pasid(struct kgd_dev *kgd, uint8_t vmid); -static uint16_t get_fw_version(struct kgd_dev *kgd, enum kgd_engine_type type); static void set_scratch_backing_va(struct kgd_dev *kgd, uint64_t va, uint32_t vmid); static void set_vm_context_page_table_base(struct kgd_dev *kgd, uint32_t vmid, @@ -191,7 +188,6 @@ static const struct kfd2kgd_calls kfd2kgd = { .address_watch_get_offset = kgd_address_watch_get_offset, .get_atc_vmid_pasid_mapping_pasid = get_atc_vmid_pasid_mapping_pasid, .get_atc_vmid_pasid_mapping_valid = get_atc_vmid_pasid_mapping_valid, - .get_fw_version = get_fw_version, .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, @@ -792,63 +788,6 @@ static void set_scratch_backing_va(struct kgd_dev *kgd, unlock_srbm(kgd); } -static uint16_t get_fw_version(struct kgd_dev *kgd, enum kgd_engine_type type) -{ - struct amdgpu_device *adev = (struct amdgpu_device *) kgd; - const union amdgpu_firmware_header *hdr; - - switch (type) { - case KGD_ENGINE_PFP: - hdr = (const union amdgpu_firmware_header *) - adev->gfx.pfp_fw->data; - break; - - case KGD_ENGINE_ME: - hdr = (const union amdgpu_firmware_header *) - adev->gfx.me_fw->data; - break; - - case KGD_ENGINE_CE: - hdr = (const union amdgpu_firmware_header *) - adev->gfx.ce_fw->data; - break; - - case KGD_ENGINE_MEC1: - hdr = (const union amdgpu_firmware_header *) - adev->gfx.mec_fw->data; - break; - - case KGD_ENGINE_MEC2: - hdr = (const union amdgpu_firmware_header *) - adev->gfx.mec2_fw->data; - break; - - case KGD_ENGINE_RLC: - hdr = (const union amdgpu_firmware_header *) - adev->gfx.rlc_fw->data; - break; - - case KGD_ENGINE_SDMA1: - hdr = (const union amdgpu_firmware_header *) - adev->sdma.instance[0].fw->data; - break; - - case KGD_ENGINE_SDMA2: - hdr = (const union amdgpu_firmware_header *) - adev->sdma.instance[1].fw->data; - break; - - default: - return 0; - } - - if (hdr == NULL) - return 0; - - /* Only 12 bit in use*/ - return hdr->common.ucode_version; -} - static void set_vm_context_page_table_base(struct kgd_dev *kgd, uint32_t vmid, uint64_t page_table_base) { 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 56ea929f524b..fec3a6aa1de6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c @@ -23,12 +23,10 @@ #include <linux/module.h> #include <linux/fdtable.h> #include <linux/uaccess.h> -#include <linux/firmware.h> #include <linux/mmu_context.h> #include <drm/drmP.h> #include "amdgpu.h" #include "amdgpu_amdkfd.h" -#include "amdgpu_ucode.h" #include "gfx_v8_0.h" #include "gca/gfx_8_0_sh_mask.h" #include "gca/gfx_8_0_d.h" @@ -95,7 +93,6 @@ static bool get_atc_vmid_pasid_mapping_valid(struct kgd_dev *kgd, uint8_t vmid); static uint16_t get_atc_vmid_pasid_mapping_pasid(struct kgd_dev *kgd, uint8_t vmid); -static uint16_t get_fw_version(struct kgd_dev *kgd, enum kgd_engine_type type); static void set_scratch_backing_va(struct kgd_dev *kgd, uint64_t va, uint32_t vmid); static void set_vm_context_page_table_base(struct kgd_dev *kgd, uint32_t vmid, @@ -148,7 +145,6 @@ static const struct kfd2kgd_calls kfd2kgd = { get_atc_vmid_pasid_mapping_pasid, .get_atc_vmid_pasid_mapping_valid = get_atc_vmid_pasid_mapping_valid, - .get_fw_version = get_fw_version, .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, @@ -751,63 +747,6 @@ static void set_scratch_backing_va(struct kgd_dev *kgd, unlock_srbm(kgd); } -static uint16_t get_fw_version(struct kgd_dev *kgd, enum kgd_engine_type type) -{ - struct amdgpu_device *adev = (struct amdgpu_device *) kgd; - const union amdgpu_firmware_header *hdr; - - switch (type) { - case KGD_ENGINE_PFP: - hdr = (const union amdgpu_firmware_header *) - adev->gfx.pfp_fw->data; - break; - - case KGD_ENGINE_ME: - hdr = (const union amdgpu_firmware_header *) - adev->gfx.me_fw->data; - break; - - case KGD_ENGINE_CE: - hdr = (const union amdgpu_firmware_header *) - adev->gfx.ce_fw->data; - break; - - case KGD_ENGINE_MEC1: - hdr = (const union amdgpu_firmware_header *) - adev->gfx.mec_fw->data; - break; - - case KGD_ENGINE_MEC2: - hdr = (const union amdgpu_firmware_header *) - adev->gfx.mec2_fw->data; - break; - - case KGD_ENGINE_RLC: - hdr = (const union amdgpu_firmware_header *) - adev->gfx.rlc_fw->data; - break; - - case KGD_ENGINE_SDMA1: - hdr = (const union amdgpu_firmware_header *) - adev->sdma.instance[0].fw->data; - break; - - case KGD_ENGINE_SDMA2: - hdr = (const union amdgpu_firmware_header *) - adev->sdma.instance[1].fw->data; - break; - - default: - return 0; - } - - if (hdr == NULL) - return 0; - - /* Only 12 bit in use*/ - return hdr->common.ucode_version; -} - static void set_vm_context_page_table_base(struct kgd_dev *kgd, uint32_t vmid, uint64_t 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 5c51d4910650..ef3d93b995b2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c @@ -25,12 +25,10 @@ #include <linux/module.h> #include <linux/fdtable.h> #include <linux/uaccess.h> -#include <linux/firmware.h> #include <linux/mmu_context.h> #include <drm/drmP.h> #include "amdgpu.h" #include "amdgpu_amdkfd.h" -#include "amdgpu_ucode.h" #include "soc15_hw_ip.h" #include "gc/gc_9_0_offset.h" #include "gc/gc_9_0_sh_mask.h" @@ -111,7 +109,6 @@ static uint16_t get_atc_vmid_pasid_mapping_pasid(struct kgd_dev *kgd, uint8_t vmid); static void set_vm_context_page_table_base(struct kgd_dev *kgd, uint32_t vmid, uint64_t page_table_base); -static uint16_t get_fw_version(struct kgd_dev *kgd, enum kgd_engine_type type); static void set_scratch_backing_va(struct kgd_dev *kgd, uint64_t va, uint32_t vmid); static int invalidate_tlbs(struct kgd_dev *kgd, uint16_t pasid); @@ -158,7 +155,6 @@ static const struct kfd2kgd_calls kfd2kgd = { get_atc_vmid_pasid_mapping_pasid, .get_atc_vmid_pasid_mapping_valid = get_atc_vmid_pasid_mapping_valid, - .get_fw_version = get_fw_version, .set_scratch_backing_va = set_scratch_backing_va, .get_tile_config = amdgpu_amdkfd_get_tile_config, .set_vm_context_page_table_base = set_vm_context_page_table_base, @@ -874,56 +870,6 @@ static void set_scratch_backing_va(struct kgd_dev *kgd, */ } -/* FIXME: Does this need to be ASIC-specific code? */ -static uint16_t get_fw_version(struct kgd_dev *kgd, enum kgd_engine_type type) -{ - struct amdgpu_device *adev = (struct amdgpu_device *) kgd; - const union amdgpu_firmware_header *hdr; - - switch (type) { - case KGD_ENGINE_PFP: - hdr = (const union amdgpu_firmware_header *)adev->gfx.pfp_fw->data; - break; - - case KGD_ENGINE_ME: - hdr = (const union amdgpu_firmware_header *)adev->gfx.me_fw->data; - break; - - case KGD_ENGINE_CE: - hdr = (const union amdgpu_firmware_header *)adev->gfx.ce_fw->data; - break; - - case KGD_ENGINE_MEC1: - hdr = (const union amdgpu_firmware_header *)adev->gfx.mec_fw->data; - break; - - case KGD_ENGINE_MEC2: - hdr = (const union amdgpu_firmware_header *)adev->gfx.mec2_fw->data; - break; - - case KGD_ENGINE_RLC: - hdr = (const union amdgpu_firmware_header *)adev->gfx.rlc_fw->data; - break; - - case KGD_ENGINE_SDMA1: - hdr = (const union amdgpu_firmware_header *)adev->sdma.instance[0].fw->data; - break; - - case KGD_ENGINE_SDMA2: - hdr = (const union amdgpu_firmware_header *)adev->sdma.instance[1].fw->data; - break; - - default: - return 0; - } - - if (hdr == NULL) - return 0; - - /* Only 12 bit in use*/ - return hdr->common.ucode_version; -} - static void set_vm_context_page_table_base(struct kgd_dev *kgd, uint32_t vmid, uint64_t page_table_base) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index d47354997e3c..9ec6356d3f0b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -3437,7 +3437,7 @@ static int amdgpu_do_asic_reset(struct amdgpu_hive_info *hive, vram_lost = amdgpu_device_check_vram_lost(tmp_adev); if (vram_lost) { - DRM_ERROR("VRAM is lost!\n"); + DRM_INFO("VRAM is lost due to GPU reset!\n"); atomic_inc(&tmp_adev->vram_lost_counter); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c index ec9e45004bff..93b2c5a48a71 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c @@ -88,12 +88,14 @@ static void amdgpu_bo_destroy(struct ttm_buffer_object *tbo) if (bo->gem_base.import_attach) drm_prime_gem_destroy(&bo->gem_base, bo->tbo.sg); drm_gem_object_release(&bo->gem_base); - amdgpu_bo_unref(&bo->parent); + /* in case amdgpu_device_recover_vram got NULL of bo->parent */ if (!list_empty(&bo->shadow_list)) { mutex_lock(&adev->shadow_list_lock); list_del_init(&bo->shadow_list); mutex_unlock(&adev->shadow_list_lock); } + amdgpu_bo_unref(&bo->parent); + kfree(bo->metadata); kfree(bo); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c index 4b7a076eea9c..34471dbaa872 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c @@ -144,7 +144,7 @@ static ssize_t amdgpu_get_dpm_state(struct device *dev, struct amdgpu_device *adev = ddev->dev_private; enum amd_pm_state_type pm; - if (adev->smu.ppt_funcs->get_current_power_state) + if (is_support_sw_smu(adev) && adev->smu.ppt_funcs->get_current_power_state) pm = amdgpu_smu_get_current_power_state(adev); else if (adev->powerplay.pp_funcs->get_current_power_state) pm = amdgpu_dpm_get_current_power_state(adev); @@ -342,6 +342,16 @@ static ssize_t amdgpu_set_dpm_forced_performance_level(struct device *dev, if (current_level == level) return count; + /* profile_exit setting is valid only when current mode is in profile mode */ + if (!(current_level & (AMD_DPM_FORCED_LEVEL_PROFILE_STANDARD | + AMD_DPM_FORCED_LEVEL_PROFILE_MIN_SCLK | + AMD_DPM_FORCED_LEVEL_PROFILE_MIN_MCLK | + AMD_DPM_FORCED_LEVEL_PROFILE_PEAK)) && + (level == AMD_DPM_FORCED_LEVEL_PROFILE_EXIT)) { + pr_err("Currently not in any profile mode!\n"); + return -EINVAL; + } + if (is_support_sw_smu(adev)) { mutex_lock(&adev->pm.mutex); if (adev->pm.dpm.thermal_active) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c index 905cce1814f3..05897b05766b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c @@ -38,18 +38,10 @@ static void psp_set_funcs(struct amdgpu_device *adev); static int psp_early_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct psp_context *psp = &adev->psp; psp_set_funcs(adev); - return 0; -} - -static int psp_sw_init(void *handle) -{ - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - struct psp_context *psp = &adev->psp; - int ret; - switch (adev->asic_type) { case CHIP_VEGA10: case CHIP_VEGA12: @@ -67,6 +59,15 @@ static int psp_sw_init(void *handle) psp->adev = adev; + return 0; +} + +static int psp_sw_init(void *handle) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct psp_context *psp = &adev->psp; + int ret; + ret = psp_init_microcode(psp); if (ret) { DRM_ERROR("Failed to load psp firmware!\n"); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c index 7e7f9ed89ee1..7d484fad3909 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c @@ -36,6 +36,7 @@ void amdgpu_virt_init_setting(struct amdgpu_device *adev) /* enable virtual display */ adev->mode_info.num_crtc = 1; adev->enable_virtual_display = true; + adev->ddev->driver->driver_features &= ~DRIVER_ATOMIC; adev->cg_flags = 0; adev->pg_flags = 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c index a07c85815b7a..4f10f5aba00b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c @@ -2757,6 +2757,37 @@ error_free_sched_entity: } /** + * amdgpu_vm_check_clean_reserved - check if a VM is clean + * + * @adev: amdgpu_device pointer + * @vm: the VM to check + * + * check all entries of the root PD, if any subsequent PDs are allocated, + * it means there are page table creating and filling, and is no a clean + * VM + * + * Returns: + * 0 if this VM is clean + */ +static int amdgpu_vm_check_clean_reserved(struct amdgpu_device *adev, + struct amdgpu_vm *vm) +{ + enum amdgpu_vm_level root = adev->vm_manager.root_level; + unsigned int entries = amdgpu_vm_num_entries(adev, root); + unsigned int i = 0; + + if (!(vm->root.entries)) + return 0; + + for (i = 0; i < entries; i++) { + if (vm->root.entries[i].base.bo) + return -EINVAL; + } + + return 0; +} + +/** * amdgpu_vm_make_compute - Turn a GFX VM into a compute VM * * @adev: amdgpu_device pointer @@ -2786,10 +2817,9 @@ int amdgpu_vm_make_compute(struct amdgpu_device *adev, struct amdgpu_vm *vm, uns return r; /* Sanity checks */ - if (!RB_EMPTY_ROOT(&vm->va.rb_root) || vm->root.entries) { - r = -EINVAL; + r = amdgpu_vm_check_clean_reserved(adev, vm); + if (r) goto unreserve_bo; - } if (pasid) { unsigned long flags; diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c index 8dbad496b29f..2471e7cf75ea 100644 --- a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c +++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c @@ -372,6 +372,9 @@ static int xgpu_ai_mailbox_rcv_irq(struct amdgpu_device *adev, if (amdgpu_sriov_runtime(adev)) schedule_work(&adev->virt.flr_work); break; + case IDH_QUERY_ALIVE: + xgpu_ai_mailbox_send_ack(adev); + break; /* READY_TO_ACCESS_GPU is fetched by kernel polling, IRQ can ignore * it byfar since that polling thread will handle it, * other msg like flr complete is not handled here. diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.h b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.h index 39d151b79153..077e91a33d62 100644 --- a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.h +++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.h @@ -49,6 +49,7 @@ enum idh_event { IDH_FLR_NOTIFICATION_CMPL, IDH_SUCCESS, IDH_FAIL, + IDH_QUERY_ALIVE, IDH_EVENT_MAX }; diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c b/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c index 6a0fcd67662a..aef9d059ae52 100644 --- a/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c +++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c @@ -515,7 +515,7 @@ static void xgpu_vi_mailbox_flr_work(struct work_struct *work) /* wait until RCV_MSG become 3 */ if (xgpu_vi_poll_msg(adev, IDH_FLR_NOTIFICATION_CMPL)) { - pr_err("failed to recieve FLR_CMPL\n"); + pr_err("failed to receive FLR_CMPL\n"); return; } diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c index 1ec60f54b992..9c88ce513d78 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c @@ -156,7 +156,6 @@ static const struct soc15_reg_golden golden_settings_sdma0_4_2[] = SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_RLC7_RB_RPTR_ADDR_LO, 0xfffffffd, 0x00000001), SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_RLC7_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_UTCL1_PAGE, 0x000003ff, 0x000003c0), - SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_UTCL1_WATERMK, 0xFE000000, 0x00000000), }; static const struct soc15_reg_golden golden_settings_sdma1_4_2[] = { @@ -186,7 +185,6 @@ static const struct soc15_reg_golden golden_settings_sdma1_4_2[] = { SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_RLC7_RB_RPTR_ADDR_LO, 0xfffffffd, 0x00000001), SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_RLC7_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_UTCL1_PAGE, 0x000003ff, 0x000003c0), - SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_UTCL1_WATERMK, 0xFE000000, 0x00000000), }; static const struct soc15_reg_golden golden_settings_sdma_rv1[] = @@ -851,7 +849,7 @@ static void sdma_v4_0_gfx_resume(struct amdgpu_device *adev, unsigned int i) wptr_poll_cntl = RREG32_SDMA(i, mmSDMA0_GFX_RB_WPTR_POLL_CNTL); wptr_poll_cntl = REG_SET_FIELD(wptr_poll_cntl, SDMA0_GFX_RB_WPTR_POLL_CNTL, - F32_POLL_ENABLE, amdgpu_sriov_vf(adev)); + F32_POLL_ENABLE, amdgpu_sriov_vf(adev)? 1 : 0); WREG32_SDMA(i, mmSDMA0_GFX_RB_WPTR_POLL_CNTL, wptr_poll_cntl); /* enable DMA RB */ @@ -942,7 +940,7 @@ static void sdma_v4_0_page_resume(struct amdgpu_device *adev, unsigned int i) wptr_poll_cntl = RREG32_SDMA(i, mmSDMA0_PAGE_RB_WPTR_POLL_CNTL); wptr_poll_cntl = REG_SET_FIELD(wptr_poll_cntl, SDMA0_PAGE_RB_WPTR_POLL_CNTL, - F32_POLL_ENABLE, amdgpu_sriov_vf(adev)); + F32_POLL_ENABLE, amdgpu_sriov_vf(adev)? 1 : 0); WREG32_SDMA(i, mmSDMA0_PAGE_RB_WPTR_POLL_CNTL, wptr_poll_cntl); /* enable DMA RB */ diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c index bdb5ad93990d..4900e4958dec 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15.c +++ b/drivers/gpu/drm/amd/amdgpu/soc15.c @@ -470,6 +470,12 @@ static int soc15_asic_reset(struct amdgpu_device *adev) case CHIP_VEGA12: soc15_asic_get_baco_capability(adev, &baco_reset); break; + case CHIP_VEGA20: + if (adev->psp.sos_fw_version >= 0x80067) + soc15_asic_get_baco_capability(adev, &baco_reset); + else + baco_reset = false; + break; default: baco_reset = false; break; @@ -895,7 +901,8 @@ static int soc15_common_early_init(void *handle) adev->pg_flags = AMD_PG_SUPPORT_SDMA | AMD_PG_SUPPORT_VCN; } else if (adev->pdev->device == 0x15d8) { - adev->cg_flags = AMD_CG_SUPPORT_GFX_MGLS | + adev->cg_flags = AMD_CG_SUPPORT_GFX_MGCG | + AMD_CG_SUPPORT_GFX_MGLS | AMD_CG_SUPPORT_GFX_CP_LS | AMD_CG_SUPPORT_GFX_3D_CGCG | AMD_CG_SUPPORT_GFX_3D_CGLS | diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c index dc461df48da0..2191d3d0a219 100644 --- a/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c @@ -787,10 +787,13 @@ static int uvd_v7_0_sriov_start(struct amdgpu_device *adev) 0xFFFFFFFF, 0x00000004); /* mc resume*/ if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) { - MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW), - lower_32_bits(adev->firmware.ucode[AMDGPU_UCODE_ID_UVD].mc_addr)); - MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH), - upper_32_bits(adev->firmware.ucode[AMDGPU_UCODE_ID_UVD].mc_addr)); + MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, i, + mmUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW), + adev->firmware.ucode[AMDGPU_UCODE_ID_UVD].tmr_mc_addr_lo); + MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, i, + mmUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH), + adev->firmware.ucode[AMDGPU_UCODE_ID_UVD].tmr_mc_addr_hi); + MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_VCPU_CACHE_OFFSET0), 0); offset = 0; } else { MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW), @@ -798,10 +801,11 @@ static int uvd_v7_0_sriov_start(struct amdgpu_device *adev) MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH), upper_32_bits(adev->uvd.inst[i].gpu_addr)); offset = size; + MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_VCPU_CACHE_OFFSET0), + AMDGPU_UVD_FIRMWARE_OFFSET >> 3); + } - MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_VCPU_CACHE_OFFSET0), - AMDGPU_UVD_FIRMWARE_OFFSET >> 3); MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_VCPU_CACHE_SIZE0), size); MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW), diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c index bed78a778e3f..40363ca6c5f1 100644 --- a/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c @@ -283,7 +283,7 @@ static int vce_v2_0_stop(struct amdgpu_device *adev) } if (vce_v2_0_wait_for_idle(adev)) { - DRM_INFO("VCE is busy, Can't set clock gateing"); + DRM_INFO("VCE is busy, Can't set clock gating"); return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c index aadc3e66ebd7..c0ec27991c22 100644 --- a/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c @@ -244,13 +244,18 @@ static int vce_v4_0_sriov_start(struct amdgpu_device *adev) MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCE, 0, mmVCE_LMI_SWAP_CNTL1), 0); MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCE, 0, mmVCE_LMI_VM_CTRL), 0); + offset = AMDGPU_VCE_FIRMWARE_OFFSET; if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) { + uint32_t low = adev->firmware.ucode[AMDGPU_UCODE_ID_VCE].tmr_mc_addr_lo; + uint32_t hi = adev->firmware.ucode[AMDGPU_UCODE_ID_VCE].tmr_mc_addr_hi; + uint64_t tmr_mc_addr = (uint64_t)(hi) << 32 | low; + MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCE, 0, - mmVCE_LMI_VCPU_CACHE_40BIT_BAR0), - adev->firmware.ucode[AMDGPU_UCODE_ID_VCE].mc_addr >> 8); + mmVCE_LMI_VCPU_CACHE_40BIT_BAR0), tmr_mc_addr >> 8); MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCE, 0, mmVCE_LMI_VCPU_CACHE_64BIT_BAR0), - (adev->firmware.ucode[AMDGPU_UCODE_ID_VCE].mc_addr >> 40) & 0xff); + (tmr_mc_addr >> 40) & 0xff); + MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCE, 0, mmVCE_VCPU_CACHE_OFFSET0), 0); } else { MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCE, 0, mmVCE_LMI_VCPU_CACHE_40BIT_BAR0), @@ -258,6 +263,9 @@ static int vce_v4_0_sriov_start(struct amdgpu_device *adev) MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCE, 0, mmVCE_LMI_VCPU_CACHE_64BIT_BAR0), (adev->vce.gpu_addr >> 40) & 0xff); + MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCE, 0, mmVCE_VCPU_CACHE_OFFSET0), + offset & ~0x0f000000); + } MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCE, 0, mmVCE_LMI_VCPU_CACHE_40BIT_BAR1), @@ -272,10 +280,7 @@ static int vce_v4_0_sriov_start(struct amdgpu_device *adev) mmVCE_LMI_VCPU_CACHE_64BIT_BAR2), (adev->vce.gpu_addr >> 40) & 0xff); - offset = AMDGPU_VCE_FIRMWARE_OFFSET; size = VCE_V4_0_FW_SIZE; - MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCE, 0, mmVCE_VCPU_CACHE_OFFSET0), - offset & ~0x0f000000); MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCE, 0, mmVCE_VCPU_CACHE_SIZE0), size); offset = (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) ? offset + size : 0; @@ -382,6 +387,7 @@ static int vce_v4_0_start(struct amdgpu_device *adev) static int vce_v4_0_stop(struct amdgpu_device *adev) { + /* Disable VCPU */ WREG32_P(SOC15_REG_OFFSET(VCE, 0, mmVCE_VCPU_CNTL), 0, ~0x200001); /* hold on ECPU */ @@ -389,8 +395,8 @@ static int vce_v4_0_stop(struct amdgpu_device *adev) VCE_SOFT_RESET__ECPU_SOFT_RESET_MASK, ~VCE_SOFT_RESET__ECPU_SOFT_RESET_MASK); - /* clear BUSY flag */ - WREG32_P(SOC15_REG_OFFSET(VCE, 0, mmVCE_STATUS), 0, ~VCE_STATUS__JOB_BUSY_MASK); + /* clear VCE_STATUS */ + WREG32(SOC15_REG_OFFSET(VCE, 0, mmVCE_STATUS), 0); /* Set Clock-Gating off */ /* if (adev->cg_flags & AMD_CG_SUPPORT_VCE_MGCG) @@ -922,6 +928,7 @@ static int vce_v4_0_set_clockgating_state(void *handle, return 0; } +#endif static int vce_v4_0_set_powergating_state(void *handle, enum amd_powergating_state state) @@ -935,16 +942,11 @@ static int vce_v4_0_set_powergating_state(void *handle, */ struct amdgpu_device *adev = (struct amdgpu_device *)handle; - if (!(adev->pg_flags & AMD_PG_SUPPORT_VCE)) - return 0; - if (state == AMD_PG_STATE_GATE) - /* XXX do we need a vce_v4_0_stop()? */ - return 0; + return vce_v4_0_stop(adev); else return vce_v4_0_start(adev); } -#endif static void vce_v4_0_ring_emit_ib(struct amdgpu_ring *ring, struct amdgpu_job *job, struct amdgpu_ib *ib, uint32_t flags) @@ -1059,7 +1061,7 @@ const struct amd_ip_funcs vce_v4_0_ip_funcs = { .soft_reset = NULL /* vce_v4_0_soft_reset */, .post_soft_reset = NULL /* vce_v4_0_post_soft_reset */, .set_clockgating_state = vce_v4_0_set_clockgating_state, - .set_powergating_state = NULL /* vce_v4_0_set_powergating_state */, + .set_powergating_state = vce_v4_0_set_powergating_state, }; static const struct amdgpu_ring_funcs vce_v4_0_ring_vm_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/vega10_ih.c b/drivers/gpu/drm/amd/amdgpu/vega10_ih.c index 1b2f69a9a24e..8d89ab7f0ae8 100644 --- a/drivers/gpu/drm/amd/amdgpu/vega10_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/vega10_ih.c @@ -31,7 +31,7 @@ #include "soc15_common.h" #include "vega10_ih.h" - +#define MAX_REARM_RETRY 10 static void vega10_ih_set_interrupt_funcs(struct amdgpu_device *adev); @@ -382,6 +382,38 @@ static void vega10_ih_decode_iv(struct amdgpu_device *adev, } /** + * vega10_ih_irq_rearm - rearm IRQ if lost + * + * @adev: amdgpu_device pointer + * + */ +static void vega10_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-wwrite 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; + } +} + +/** * vega10_ih_set_rptr - set the IH ring buffer rptr * * @adev: amdgpu_device pointer @@ -395,6 +427,9 @@ static void vega10_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)) + vega10_ih_irq_rearm(adev, ih); } else if (ih == &adev->irq.ih) { WREG32_SOC15(OSSSYS, 0, mmIH_RB_RPTR, ih->rptr); } else if (ih == &adev->irq.ih1) { diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c index 2fee3063a0d6..c1e4d44d6137 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c @@ -494,9 +494,9 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd, { unsigned int size; - kfd->mec_fw_version = kfd->kfd2kgd->get_fw_version(kfd->kgd, + kfd->mec_fw_version = amdgpu_amdkfd_get_fw_version(kfd->kgd, KGD_ENGINE_MEC1); - kfd->sdma_fw_version = kfd->kfd2kgd->get_fw_version(kfd->kgd, + kfd->sdma_fw_version = amdgpu_amdkfd_get_fw_version(kfd->kgd, KGD_ENGINE_SDMA1); kfd->shared_resources = *gpu_resources; 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 923cb5ca5a57..995f9df66142 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -364,6 +364,7 @@ static void dm_vupdate_high_irq(void *interrupt_params) struct amdgpu_device *adev = irq_params->adev; struct amdgpu_crtc *acrtc; struct dm_crtc_state *acrtc_state; + unsigned long flags; acrtc = get_crtc_by_otg_inst(adev, irq_params->irq_src - IRQ_TYPE_VUPDATE); @@ -379,8 +380,25 @@ static void dm_vupdate_high_irq(void *interrupt_params) * page-flip completion events that have been queued to us * if a pageflip happened inside front-porch. */ - if (amdgpu_dm_vrr_active(acrtc_state)) + if (amdgpu_dm_vrr_active(acrtc_state)) { drm_crtc_handle_vblank(&acrtc->base); + + /* BTR processing for pre-DCE12 ASICs */ + if (acrtc_state->stream && + adev->family < AMDGPU_FAMILY_AI) { + spin_lock_irqsave(&adev->ddev->event_lock, flags); + mod_freesync_handle_v_update( + adev->dm.freesync_module, + acrtc_state->stream, + &acrtc_state->vrr_params); + + dc_stream_adjust_vmin_vmax( + adev->dm.dc, + acrtc_state->stream, + &acrtc_state->vrr_params.adjust); + spin_unlock_irqrestore(&adev->ddev->event_lock, flags); + } + } } } @@ -390,6 +408,7 @@ static void dm_crtc_high_irq(void *interrupt_params) struct amdgpu_device *adev = irq_params->adev; struct amdgpu_crtc *acrtc; struct dm_crtc_state *acrtc_state; + unsigned long flags; acrtc = get_crtc_by_otg_inst(adev, irq_params->irq_src - IRQ_TYPE_VBLANK); @@ -412,9 +431,10 @@ static void dm_crtc_high_irq(void *interrupt_params) */ amdgpu_dm_crtc_handle_crc_irq(&acrtc->base); - if (acrtc_state->stream && + if (acrtc_state->stream && adev->family >= AMDGPU_FAMILY_AI && acrtc_state->vrr_params.supported && acrtc_state->freesync_config.state == VRR_STATE_ACTIVE_VARIABLE) { + spin_lock_irqsave(&adev->ddev->event_lock, flags); mod_freesync_handle_v_update( adev->dm.freesync_module, acrtc_state->stream, @@ -424,6 +444,7 @@ static void dm_crtc_high_irq(void *interrupt_params) adev->dm.dc, acrtc_state->stream, &acrtc_state->vrr_params.adjust); + spin_unlock_irqrestore(&adev->ddev->event_lock, flags); } } } @@ -534,6 +555,8 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) if (amdgpu_dc_feature_mask & DC_FBC_MASK) init_data.flags.fbc_support = true; + init_data.flags.power_down_display_on_boot = true; + /* Display Core create. */ adev->dm.dc = dc_create(&init_data); @@ -3438,6 +3461,8 @@ dm_crtc_duplicate_state(struct drm_crtc *crtc) dc_stream_retain(state->stream); } + state->active_planes = cur->active_planes; + state->interrupts_enabled = cur->interrupts_enabled; state->vrr_params = cur->vrr_params; state->vrr_infopacket = cur->vrr_infopacket; state->abm_level = cur->abm_level; @@ -3862,7 +3887,20 @@ static void dm_crtc_helper_disable(struct drm_crtc *crtc) { } -static bool does_crtc_have_active_plane(struct drm_crtc_state *new_crtc_state) +static bool does_crtc_have_active_cursor(struct drm_crtc_state *new_crtc_state) +{ + struct drm_device *dev = new_crtc_state->crtc->dev; + struct drm_plane *plane; + + drm_for_each_plane_mask(plane, dev, new_crtc_state->plane_mask) { + if (plane->type == DRM_PLANE_TYPE_CURSOR) + return true; + } + + return false; +} + +static int count_crtc_active_planes(struct drm_crtc_state *new_crtc_state) { struct drm_atomic_state *state = new_crtc_state->state; struct drm_plane *plane; @@ -3891,7 +3929,32 @@ static bool does_crtc_have_active_plane(struct drm_crtc_state *new_crtc_state) num_active += (new_plane_state->fb != NULL); } - return num_active > 0; + return num_active; +} + +/* + * Sets whether interrupts should be enabled on a specific CRTC. + * We require that the stream be enabled and that there exist active + * DC planes on the stream. + */ +static void +dm_update_crtc_interrupt_state(struct drm_crtc *crtc, + struct drm_crtc_state *new_crtc_state) +{ + struct dm_crtc_state *dm_new_crtc_state = + to_dm_crtc_state(new_crtc_state); + + dm_new_crtc_state->active_planes = 0; + dm_new_crtc_state->interrupts_enabled = false; + + if (!dm_new_crtc_state->stream) + return; + + dm_new_crtc_state->active_planes = + count_crtc_active_planes(new_crtc_state); + + dm_new_crtc_state->interrupts_enabled = + dm_new_crtc_state->active_planes > 0; } static int dm_crtc_helper_atomic_check(struct drm_crtc *crtc, @@ -3902,6 +3965,14 @@ static int dm_crtc_helper_atomic_check(struct drm_crtc *crtc, struct dm_crtc_state *dm_crtc_state = to_dm_crtc_state(state); int ret = -EINVAL; + /* + * Update interrupt state for the CRTC. This needs to happen whenever + * the CRTC has changed or whenever any of its planes have changed. + * Atomic check satisfies both of these requirements since the CRTC + * is added to the state by DRM during drm_atomic_helper_check_planes. + */ + dm_update_crtc_interrupt_state(crtc, state); + if (unlikely(!dm_crtc_state->stream && modeset_required(state, NULL, dm_crtc_state->stream))) { WARN_ON(1); @@ -3912,9 +3983,13 @@ static int dm_crtc_helper_atomic_check(struct drm_crtc *crtc, if (!dm_crtc_state->stream) return 0; - /* We want at least one hardware plane enabled to use the stream. */ + /* + * We want at least one hardware plane enabled to use + * the stream with a cursor enabled. + */ if (state->enable && state->active && - !does_crtc_have_active_plane(state)) + does_crtc_have_active_cursor(state) && + dm_crtc_state->active_planes == 0) return -EINVAL; if (dc_validate_stream(dc, dm_crtc_state->stream) == DC_OK) @@ -4188,6 +4263,7 @@ static const uint32_t rgb_formats[] = { DRM_FORMAT_ABGR2101010, DRM_FORMAT_XBGR8888, DRM_FORMAT_ABGR8888, + DRM_FORMAT_RGB565, }; static const uint32_t overlay_formats[] = { @@ -4196,6 +4272,7 @@ static const uint32_t overlay_formats[] = { DRM_FORMAT_RGBA8888, DRM_FORMAT_XBGR8888, DRM_FORMAT_ABGR8888, + DRM_FORMAT_RGB565 }; static const u32 cursor_formats[] = { @@ -4999,8 +5076,10 @@ static void update_freesync_state_on_stream( struct dc_plane_state *surface, u32 flip_timestamp_in_us) { - struct mod_vrr_params vrr_params = new_crtc_state->vrr_params; + struct mod_vrr_params vrr_params; struct dc_info_packet vrr_infopacket = {0}; + struct amdgpu_device *adev = dm->adev; + unsigned long flags; if (!new_stream) return; @@ -5013,6 +5092,9 @@ static void update_freesync_state_on_stream( if (!new_stream->timing.h_total || !new_stream->timing.v_total) return; + spin_lock_irqsave(&adev->ddev->event_lock, flags); + vrr_params = new_crtc_state->vrr_params; + if (surface) { mod_freesync_handle_preflip( dm->freesync_module, @@ -5020,6 +5102,12 @@ static void update_freesync_state_on_stream( new_stream, flip_timestamp_in_us, &vrr_params); + + if (adev->family < AMDGPU_FAMILY_AI && + amdgpu_dm_vrr_active(new_crtc_state)) { + mod_freesync_handle_v_update(dm->freesync_module, + new_stream, &vrr_params); + } } mod_freesync_build_vrr_infopacket( @@ -5051,6 +5139,8 @@ static void update_freesync_state_on_stream( new_crtc_state->base.crtc->base.id, (int)new_crtc_state->base.vrr_enabled, (int)vrr_params.state); + + spin_unlock_irqrestore(&adev->ddev->event_lock, flags); } static void pre_update_freesync_state_on_stream( @@ -5058,8 +5148,10 @@ static void pre_update_freesync_state_on_stream( struct dm_crtc_state *new_crtc_state) { struct dc_stream_state *new_stream = new_crtc_state->stream; - struct mod_vrr_params vrr_params = new_crtc_state->vrr_params; + struct mod_vrr_params vrr_params; struct mod_freesync_config config = new_crtc_state->freesync_config; + struct amdgpu_device *adev = dm->adev; + unsigned long flags; if (!new_stream) return; @@ -5071,6 +5163,9 @@ static void pre_update_freesync_state_on_stream( if (!new_stream->timing.h_total || !new_stream->timing.v_total) return; + spin_lock_irqsave(&adev->ddev->event_lock, flags); + vrr_params = new_crtc_state->vrr_params; + if (new_crtc_state->vrr_supported && config.min_refresh_in_uhz && config.max_refresh_in_uhz) { @@ -5091,6 +5186,7 @@ static void pre_update_freesync_state_on_stream( sizeof(vrr_params.adjust)) != 0); new_crtc_state->vrr_params = vrr_params; + spin_unlock_irqrestore(&adev->ddev->event_lock, flags); } static void amdgpu_dm_handle_vrr_transition(struct dm_crtc_state *old_state, @@ -5123,6 +5219,22 @@ static void amdgpu_dm_handle_vrr_transition(struct dm_crtc_state *old_state, } } +static void amdgpu_dm_commit_cursors(struct drm_atomic_state *state) +{ + struct drm_plane *plane; + struct drm_plane_state *old_plane_state, *new_plane_state; + int i; + + /* + * TODO: Make this per-stream so we don't issue redundant updates for + * commits with multiple streams. + */ + for_each_oldnew_plane_in_state(state, plane, old_plane_state, + new_plane_state, i) + if (plane->type == DRM_PLANE_TYPE_CURSOR) + handle_cursor_update(plane, old_plane_state); +} + static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, struct dc_state *dc_state, struct drm_device *dev, @@ -5130,7 +5242,7 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, struct drm_crtc *pcrtc, bool wait_for_vblank) { - uint32_t i, r; + uint32_t i; uint64_t timestamp_ns; struct drm_plane *plane; struct drm_plane_state *old_plane_state, *new_plane_state; @@ -5141,6 +5253,7 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, struct dm_crtc_state *dm_old_crtc_state = to_dm_crtc_state(drm_atomic_get_old_crtc_state(state, pcrtc)); int planes_count = 0, vpos, hpos; + long r; unsigned long flags; struct amdgpu_bo *abo; uint64_t tiling_flags; @@ -5162,6 +5275,14 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, goto cleanup; } + /* + * Disable the cursor first if we're disabling all the planes. + * It'll remain on the screen after the planes are re-enabled + * if we don't. + */ + if (acrtc_state->active_planes == 0) + amdgpu_dm_commit_cursors(state); + /* update planes when needed */ for_each_oldnew_plane_in_state(state, plane, old_plane_state, new_plane_state, i) { struct drm_crtc *crtc = new_plane_state->crtc; @@ -5205,22 +5326,28 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, continue; } + abo = gem_to_amdgpu_bo(fb->obj[0]); + + /* + * Wait for all fences on this FB. Do limited wait to avoid + * deadlock during GPU reset when this fence will not signal + * but we hold reservation lock for the BO. + */ + r = reservation_object_wait_timeout_rcu(abo->tbo.resv, true, + false, + msecs_to_jiffies(5000)); + if (unlikely(r <= 0)) + DRM_ERROR("Waiting for fences timed out or interrupted!"); + /* * TODO This might fail and hence better not used, wait * explicitly on fences instead * and in general should be called for * blocking commit to as per framework helpers */ - abo = gem_to_amdgpu_bo(fb->obj[0]); r = amdgpu_bo_reserve(abo, true); - if (unlikely(r != 0)) { + if (unlikely(r != 0)) DRM_ERROR("failed to reserve buffer before flip\n"); - WARN_ON(1); - } - - /* Wait for all fences on this FB */ - WARN_ON(reservation_object_wait_timeout_rcu(abo->tbo.resv, true, false, - MAX_SCHEDULE_TIMEOUT) < 0); amdgpu_bo_get_tiling_flags(abo, &tiling_flags); @@ -5329,7 +5456,8 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, } } - if (planes_count) { + /* Update the planes if changed or disable if we don't have any. */ + if (planes_count || acrtc_state->active_planes == 0) { if (new_pcrtc_state->mode_changed) { bundle->stream_update.src = acrtc_state->stream->src; bundle->stream_update.dst = acrtc_state->stream->dst; @@ -5352,15 +5480,72 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, mutex_unlock(&dm->dc_lock); } - for_each_oldnew_plane_in_state(state, plane, old_plane_state, new_plane_state, i) - if (plane->type == DRM_PLANE_TYPE_CURSOR) - handle_cursor_update(plane, old_plane_state); + /* + * Update cursor state *after* programming all the planes. + * This avoids redundant programming in the case where we're going + * to be disabling a single plane - those pipes are being disabled. + */ + if (acrtc_state->active_planes) + amdgpu_dm_commit_cursors(state); cleanup: kfree(bundle); } /* + * Enable interrupts on CRTCs that are newly active, undergone + * a modeset, or have active planes again. + * + * Done in two passes, based on the for_modeset flag: + * Pass 1: For CRTCs going through modeset + * Pass 2: For CRTCs going from 0 to n active planes + * + * Interrupts can only be enabled after the planes are programmed, + * so this requires a two-pass approach since we don't want to + * just defer the interrupts until after commit planes every time. + */ +static void amdgpu_dm_enable_crtc_interrupts(struct drm_device *dev, + struct drm_atomic_state *state, + bool for_modeset) +{ + struct amdgpu_device *adev = dev->dev_private; + struct drm_crtc *crtc; + struct drm_crtc_state *old_crtc_state, *new_crtc_state; + int i; + + for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, + new_crtc_state, i) { + struct amdgpu_crtc *acrtc = to_amdgpu_crtc(crtc); + struct dm_crtc_state *dm_new_crtc_state = + to_dm_crtc_state(new_crtc_state); + struct dm_crtc_state *dm_old_crtc_state = + to_dm_crtc_state(old_crtc_state); + bool modeset = drm_atomic_crtc_needs_modeset(new_crtc_state); + bool run_pass; + + run_pass = (for_modeset && modeset) || + (!for_modeset && !modeset && + !dm_old_crtc_state->interrupts_enabled); + + if (!run_pass) + continue; + + if (!dm_new_crtc_state->interrupts_enabled) + continue; + + manage_dm_interrupts(adev, acrtc, true); + +#ifdef CONFIG_DEBUG_FS + /* The stream has changed so CRC capture needs to re-enabled. */ + if (dm_new_crtc_state->crc_enabled) { + dm_new_crtc_state->crc_enabled = false; + amdgpu_dm_crtc_set_crc_source(crtc, "auto"); + } +#endif + } +} + +/* * amdgpu_dm_crtc_copy_transient_flags - copy mirrored flags from DRM to DC * @crtc_state: the DRM CRTC state * @stream_state: the DC stream state. @@ -5384,30 +5569,41 @@ static int amdgpu_dm_atomic_commit(struct drm_device *dev, int i; /* - * We evade vblanks and pflips on crtc that - * should be changed. We do it here to flush & disable - * interrupts before drm_swap_state is called in drm_atomic_helper_commit - * it will update crtc->dm_crtc_state->stream pointer which is used in - * the ISRs. + * We evade vblank and pflip interrupts on CRTCs that are undergoing + * a modeset, being disabled, or have no active planes. + * + * It's done in atomic commit rather than commit tail for now since + * some of these interrupt handlers access the current CRTC state and + * potentially the stream pointer itself. + * + * Since the atomic state is swapped within atomic commit and not within + * commit tail this would leave to new state (that hasn't been committed yet) + * being accesssed from within the handlers. + * + * TODO: Fix this so we can do this in commit tail and not have to block + * in atomic check. */ for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) { struct dm_crtc_state *dm_old_crtc_state = to_dm_crtc_state(old_crtc_state); struct dm_crtc_state *dm_new_crtc_state = to_dm_crtc_state(new_crtc_state); struct amdgpu_crtc *acrtc = to_amdgpu_crtc(crtc); - if (drm_atomic_crtc_needs_modeset(new_crtc_state) - && dm_old_crtc_state->stream) { + if (dm_old_crtc_state->interrupts_enabled && + (!dm_new_crtc_state->interrupts_enabled || + drm_atomic_crtc_needs_modeset(new_crtc_state))) { /* - * If the stream is removed and CRC capture was - * enabled on the CRTC the extra vblank reference - * needs to be dropped since CRC capture will be - * disabled. + * Drop the extra vblank reference added by CRC + * capture if applicable. */ - if (!dm_new_crtc_state->stream - && dm_new_crtc_state->crc_enabled) { + if (dm_new_crtc_state->crc_enabled) drm_crtc_vblank_put(crtc); + + /* + * Only keep CRC capture enabled if there's + * still a stream for the CRTC. + */ + if (!dm_new_crtc_state->stream) dm_new_crtc_state->crc_enabled = false; - } manage_dm_interrupts(adev, acrtc, false); } @@ -5623,47 +5819,26 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) mutex_unlock(&dm->dc_lock); } - /* Update freesync state before amdgpu_dm_handle_vrr_transition(). */ - for_each_new_crtc_in_state(state, crtc, new_crtc_state, i) { - dm_new_crtc_state = to_dm_crtc_state(new_crtc_state); - pre_update_freesync_state_on_stream(dm, dm_new_crtc_state); - } - + /* Count number of newly disabled CRTCs for dropping PM refs later. */ for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, - new_crtc_state, i) { - /* - * loop to enable interrupts on newly arrived crtc - */ - struct amdgpu_crtc *acrtc = to_amdgpu_crtc(crtc); - bool modeset_needed; - + new_crtc_state, i) { if (old_crtc_state->active && !new_crtc_state->active) crtc_disable_count++; dm_new_crtc_state = to_dm_crtc_state(new_crtc_state); dm_old_crtc_state = to_dm_crtc_state(old_crtc_state); + /* Update freesync active state. */ + pre_update_freesync_state_on_stream(dm, dm_new_crtc_state); + /* Handle vrr on->off / off->on transitions */ amdgpu_dm_handle_vrr_transition(dm_old_crtc_state, dm_new_crtc_state); - - modeset_needed = modeset_required( - new_crtc_state, - dm_new_crtc_state->stream, - dm_old_crtc_state->stream); - - if (dm_new_crtc_state->stream == NULL || !modeset_needed) - continue; - - manage_dm_interrupts(adev, acrtc, true); - -#ifdef CONFIG_DEBUG_FS - /* The stream has changed so CRC capture needs to re-enabled. */ - if (dm_new_crtc_state->crc_enabled) - amdgpu_dm_crtc_set_crc_source(crtc, "auto"); -#endif } + /* Enable interrupts for CRTCs going through a modeset. */ + amdgpu_dm_enable_crtc_interrupts(dev, state, true); + for_each_new_crtc_in_state(state, crtc, new_crtc_state, j) if (new_crtc_state->pageflip_flags & DRM_MODE_PAGE_FLIP_ASYNC) wait_for_vblank = false; @@ -5677,6 +5852,8 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) dm, crtc, wait_for_vblank); } + /* Enable interrupts for CRTCs going from 0 to n active planes. */ + amdgpu_dm_enable_crtc_interrupts(dev, state, false); /* * send vblank event on all events not handled in flip and 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 3a0b6164c755..978ff14a7d45 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h @@ -271,6 +271,9 @@ struct dm_crtc_state { struct drm_crtc_state base; struct dc_stream_state *stream; + int active_planes; + bool interrupts_enabled; + int crc_skip_count; bool crc_enabled; 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 3ef68a249f4d..b37ecc3ede61 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link.c @@ -514,6 +514,40 @@ static void link_disconnect_remap(struct dc_sink *prev_sink, struct dc_link *lin } +static void read_edp_current_link_settings_on_detect(struct dc_link *link) +{ + union lane_count_set lane_count_set = { {0} }; + uint8_t link_bw_set; + uint8_t link_rate_set; + + // Read DPCD 00101h to find out the number of lanes currently set + core_link_read_dpcd(link, DP_LANE_COUNT_SET, + &lane_count_set.raw, sizeof(lane_count_set)); + link->cur_link_settings.lane_count = lane_count_set.bits.LANE_COUNT_SET; + + // Read DPCD 00100h to find if standard link rates are set + core_link_read_dpcd(link, DP_LINK_BW_SET, + &link_bw_set, sizeof(link_bw_set)); + + if (link_bw_set == 0) { + /* If standard link rates are not being used, + * Read DPCD 00115h to find the link rate set used + */ + core_link_read_dpcd(link, DP_LINK_RATE_SET, + &link_rate_set, sizeof(link_rate_set)); + + if (link_rate_set < link->dpcd_caps.edp_supported_link_rates_count) { + link->cur_link_settings.link_rate = + link->dpcd_caps.edp_supported_link_rates[link_rate_set]; + link->cur_link_settings.link_rate_set = link_rate_set; + link->cur_link_settings.use_link_rate_set = true; + } + } else { + link->cur_link_settings.link_rate = link_bw_set; + link->cur_link_settings.use_link_rate_set = false; + } +} + static bool detect_dp( struct dc_link *link, struct display_sink_capability *sink_caps, @@ -648,9 +682,14 @@ bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason) return false; } - if (link->connector_signal == SIGNAL_TYPE_EDP && - link->local_sink) - return true; + if (link->connector_signal == SIGNAL_TYPE_EDP) { + /* On detect, we want to make sure current link settings are + * up to date, especially if link was powered on by GOP. + */ + read_edp_current_link_settings_on_detect(link); + if (link->local_sink) + return true; + } if (link->connector_signal == SIGNAL_TYPE_LVDS && link->local_sink) @@ -1396,13 +1435,19 @@ static enum dc_status enable_link_dp( /* get link settings for video mode timing */ decide_link_settings(stream, &link_settings); - /* If link settings are different than current and link already enabled - * then need to disable before programming to new rate. - */ - if (link->link_status.link_active && - (link->cur_link_settings.lane_count != link_settings.lane_count || - link->cur_link_settings.link_rate != link_settings.link_rate)) { - dp_disable_link_phy(link, pipe_ctx->stream->signal); + if (pipe_ctx->stream->signal == SIGNAL_TYPE_EDP) { + /* If link settings are different than current and link already enabled + * then need to disable before programming to new rate. + */ + if (link->link_status.link_active && + (link->cur_link_settings.lane_count != link_settings.lane_count || + link->cur_link_settings.link_rate != link_settings.link_rate)) { + dp_disable_link_phy(link, pipe_ctx->stream->signal); + } + + /*in case it is not on*/ + link->dc->hwss.edp_power_control(link, true); + link->dc->hwss.edp_wait_for_hpd_ready(link, true); } pipe_ctx->stream_res.pix_clk_params.requested_sym_clk = @@ -1448,15 +1493,9 @@ static enum dc_status enable_link_edp( struct pipe_ctx *pipe_ctx) { enum dc_status status; - struct dc_stream_state *stream = pipe_ctx->stream; - struct dc_link *link = stream->link; - /*in case it is not on*/ - link->dc->hwss.edp_power_control(link, true); - link->dc->hwss.edp_wait_for_hpd_ready(link, true); status = enable_link_dp(state, pipe_ctx); - return status; } 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 a6424c70f4c5..1ee544a32ebb 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 @@ -2185,6 +2185,30 @@ static int translate_dpcd_max_bpc(enum dpcd_downstream_port_max_bpc bpc) return -1; } +static void read_dp_device_vendor_id(struct dc_link *link) +{ + struct dp_device_vendor_id dp_id; + + /* read IEEE branch device id */ + core_link_read_dpcd( + link, + DP_BRANCH_OUI, + (uint8_t *)&dp_id, + sizeof(dp_id)); + + link->dpcd_caps.branch_dev_id = + (dp_id.ieee_oui[0] << 16) + + (dp_id.ieee_oui[1] << 8) + + dp_id.ieee_oui[2]; + + memmove( + link->dpcd_caps.branch_dev_name, + dp_id.ieee_device_id, + sizeof(dp_id.ieee_device_id)); +} + + + static void get_active_converter_info( uint8_t data, struct dc_link *link) { @@ -2271,27 +2295,6 @@ static void get_active_converter_info( ddc_service_set_dongle_type(link->ddc, link->dpcd_caps.dongle_type); { - struct dp_device_vendor_id dp_id; - - /* read IEEE branch device id */ - core_link_read_dpcd( - link, - DP_BRANCH_OUI, - (uint8_t *)&dp_id, - sizeof(dp_id)); - - link->dpcd_caps.branch_dev_id = - (dp_id.ieee_oui[0] << 16) + - (dp_id.ieee_oui[1] << 8) + - dp_id.ieee_oui[2]; - - memmove( - link->dpcd_caps.branch_dev_name, - dp_id.ieee_device_id, - sizeof(dp_id.ieee_device_id)); - } - - { struct dp_sink_hw_fw_revision dp_hw_fw_revision; core_link_read_dpcd( @@ -2455,6 +2458,8 @@ static bool retrieve_link_cap(struct dc_link *link) ds_port.byte = dpcd_data[DP_DOWNSTREAMPORT_PRESENT - DP_DPCD_REV]; + read_dp_device_vendor_id(link); + get_active_converter_info(ds_port.byte, link); dp_wa_power_up_0010FA(link, dpcd_data, sizeof(dpcd_data)); @@ -2586,9 +2591,6 @@ void detect_edp_sink_caps(struct dc_link *link) uint32_t entry; uint32_t link_rate_in_khz; enum dc_link_rate link_rate = LINK_RATE_UNKNOWN; - union lane_count_set lane_count_set = { {0} }; - uint8_t link_bw_set; - uint8_t link_rate_set; retrieve_link_cap(link); link->dpcd_caps.edp_supported_link_rates_count = 0; @@ -2614,33 +2616,6 @@ void detect_edp_sink_caps(struct dc_link *link) } } link->verified_link_cap = link->reported_link_cap; - - // Read DPCD 00101h to find out the number of lanes currently set - core_link_read_dpcd(link, DP_LANE_COUNT_SET, - &lane_count_set.raw, sizeof(lane_count_set)); - link->cur_link_settings.lane_count = lane_count_set.bits.LANE_COUNT_SET; - - // Read DPCD 00100h to find if standard link rates are set - core_link_read_dpcd(link, DP_LINK_BW_SET, - &link_bw_set, sizeof(link_bw_set)); - - if (link_bw_set == 0) { - /* If standard link rates are not being used, - * Read DPCD 00115h to find the link rate set used - */ - core_link_read_dpcd(link, DP_LINK_RATE_SET, - &link_rate_set, sizeof(link_rate_set)); - - if (link_rate_set < link->dpcd_caps.edp_supported_link_rates_count) { - link->cur_link_settings.link_rate = - link->dpcd_caps.edp_supported_link_rates[link_rate_set]; - link->cur_link_settings.link_rate_set = link_rate_set; - link->cur_link_settings.use_link_rate_set = true; - } - } else { - link->cur_link_settings.link_rate = link_bw_set; - link->cur_link_settings.use_link_rate_set = false; - } } void dc_link_dp_enable_hpd(const struct dc_link *link) 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 f7f7515f65f4..b0dea759cd86 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 @@ -58,6 +58,8 @@ void dp_enable_link_phy( const struct dc_link_settings *link_settings) { struct link_encoder *link_enc = link->link_enc; + struct dc *core_dc = link->ctx->dc; + struct dmcu *dmcu = core_dc->res_pool->dmcu; struct pipe_ctx *pipes = link->dc->current_state->res_ctx.pipe_ctx; @@ -84,6 +86,9 @@ void dp_enable_link_phy( } } + if (dmcu != NULL && dmcu->funcs->lock_phy) + dmcu->funcs->lock_phy(dmcu); + if (dc_is_dp_sst_signal(signal)) { link_enc->funcs->enable_dp_output( link_enc, @@ -95,6 +100,10 @@ void dp_enable_link_phy( link_settings, clock_source); } + + if (dmcu != NULL && dmcu->funcs->unlock_phy) + dmcu->funcs->unlock_phy(dmcu); + link->cur_link_settings = *link_settings; dp_receiver_power_ctrl(link, true); @@ -150,15 +159,25 @@ bool edp_receiver_ready_T7(struct dc_link *link) void dp_disable_link_phy(struct dc_link *link, enum signal_type signal) { + struct dc *core_dc = link->ctx->dc; + struct dmcu *dmcu = core_dc->res_pool->dmcu; + if (!link->wa_flags.dp_keep_receiver_powered) dp_receiver_power_ctrl(link, false); if (signal == SIGNAL_TYPE_EDP) { link->link_enc->funcs->disable_output(link->link_enc, signal); link->dc->hwss.edp_power_control(link, false); - } else + } else { + if (dmcu != NULL && dmcu->funcs->lock_phy) + dmcu->funcs->lock_phy(dmcu); + link->link_enc->funcs->disable_output(link->link_enc, signal); + if (dmcu != NULL && dmcu->funcs->unlock_phy) + dmcu->funcs->unlock_phy(dmcu); + } + /* Clear current link setting.*/ memset(&link->cur_link_settings, 0, sizeof(link->cur_link_settings)); diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c index e10479d58c11..96e97d25d639 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c @@ -163,6 +163,27 @@ struct dc_stream_state *dc_create_stream_for_sink( return stream; } +struct dc_stream_state *dc_copy_stream(const struct dc_stream_state *stream) +{ + struct dc_stream_state *new_stream; + + new_stream = kzalloc(sizeof(struct dc_stream_state), GFP_KERNEL); + if (!new_stream) + return NULL; + + memcpy(new_stream, stream, sizeof(struct dc_stream_state)); + + if (new_stream->sink) + dc_sink_retain(new_stream->sink); + + if (new_stream->out_transfer_func) + dc_transfer_func_retain(new_stream->out_transfer_func); + + kref_init(&new_stream->refcount); + + return new_stream; +} + /** * dc_stream_get_status_from_state - Get stream status from given dc state * @state: DC state to find the stream status in @@ -312,7 +333,7 @@ bool dc_stream_set_cursor_position( (!pipe_ctx->plane_res.mi && !pipe_ctx->plane_res.hubp) || !pipe_ctx->plane_state || (!pipe_ctx->plane_res.xfm && !pipe_ctx->plane_res.dpp) || - !pipe_ctx->plane_res.ipp) + (!pipe_ctx->plane_res.ipp && !pipe_ctx->plane_res.dpp)) continue; if (!pipe_to_program) { diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h index 3459e39714bc..70edd9ea5afe 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 "3.2.26" +#define DC_VER "3.2.27" #define MAX_SURFACES 3 #define MAX_PLANES 6 @@ -204,6 +204,7 @@ struct dc_config { bool optimize_edp_link_rate; bool disable_fractional_pwm; bool allow_seamless_boot_optimization; + bool power_down_display_on_boot; }; enum visual_confirm { diff --git a/drivers/gpu/drm/amd/display/dc/dc_link.h b/drivers/gpu/drm/amd/display/dc/dc_link.h index cc7ffac64c96..7b9429e30d82 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_link.h +++ b/drivers/gpu/drm/amd/display/dc/dc_link.h @@ -120,6 +120,7 @@ struct dc_link { /* MST record stream using this link */ struct link_flags { bool dp_keep_receiver_powered; + bool dp_skip_DID2; } wa_flags; struct link_mst_stream_allocation_table mst_stream_alloc_table; diff --git a/drivers/gpu/drm/amd/display/dc/dc_stream.h b/drivers/gpu/drm/amd/display/dc/dc_stream.h index 17fa3bf6cf7b..189bdab929a5 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_stream.h +++ b/drivers/gpu/drm/amd/display/dc/dc_stream.h @@ -307,6 +307,8 @@ enum surface_update_type dc_check_update_surfaces_for_stream( */ struct dc_stream_state *dc_create_stream_for_sink(struct dc_sink *dc_sink); +struct dc_stream_state *dc_copy_stream(const struct dc_stream_state *stream); + void update_stream_signal(struct dc_stream_state *stream, struct dc_sink *sink); void dc_stream_retain(struct dc_stream_state *dc_stream); diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_abm.c b/drivers/gpu/drm/amd/display/dc/dce/dce_abm.c index 855360b1414f..da96229db53a 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_abm.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_abm.c @@ -50,7 +50,6 @@ #define MCP_ABM_LEVEL_SET 0x65 #define MCP_ABM_PIPE_SET 0x66 #define MCP_BL_SET 0x67 -#define MCP_BL_SET_PWM_FRAC 0x6A /* Enable or disable Fractional PWM */ #define MCP_DISABLE_ABM_IMMEDIATELY 255 @@ -391,23 +390,6 @@ static bool dce_abm_init_backlight(struct abm *abm) REG_UPDATE(BL_PWM_GRP1_REG_LOCK, BL_PWM_GRP1_REG_LOCK, 0); - /* Wait until microcontroller is ready to process interrupt */ - REG_WAIT(MASTER_COMM_CNTL_REG, MASTER_COMM_INTERRUPT, 0, 100, 800); - - /* Set PWM fractional enable/disable */ - value = (abm->ctx->dc->config.disable_fractional_pwm == false) ? 1 : 0; - REG_WRITE(MASTER_COMM_DATA_REG1, value); - - /* Set command to enable or disable fractional PWM microcontroller */ - REG_UPDATE(MASTER_COMM_CMD_REG, MASTER_COMM_CMD_REG_BYTE0, - MCP_BL_SET_PWM_FRAC); - - /* Notify microcontroller of new command */ - REG_UPDATE(MASTER_COMM_CNTL_REG, MASTER_COMM_INTERRUPT, 1); - - /* Ensure command has been executed before continuing */ - REG_WAIT(MASTER_COMM_CNTL_REG, MASTER_COMM_INTERRUPT, 0, 100, 800); - return true; } 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 aa586672e8cd..818536eea00a 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c @@ -51,6 +51,9 @@ #define PSR_SET_WAITLOOP 0x31 #define MCP_INIT_DMCU 0x88 #define MCP_INIT_IRAM 0x89 +#define MCP_SYNC_PHY_LOCK 0x90 +#define MCP_SYNC_PHY_UNLOCK 0x91 +#define MCP_BL_SET_PWM_FRAC 0x6A /* Enable or disable Fractional PWM */ #define MASTER_COMM_CNTL_REG__MASTER_COMM_INTERRUPT_MASK 0x00000001L static bool dce_dmcu_init(struct dmcu *dmcu) @@ -339,9 +342,32 @@ static void dcn10_get_dmcu_version(struct dmcu *dmcu) IRAM_RD_ADDR_AUTO_INC, 0); } +static void dcn10_dmcu_enable_fractional_pwm(struct dmcu *dmcu, + uint32_t fractional_pwm) +{ + struct dce_dmcu *dmcu_dce = TO_DCE_DMCU(dmcu); + + /* Wait until microcontroller is ready to process interrupt */ + REG_WAIT(MASTER_COMM_CNTL_REG, MASTER_COMM_INTERRUPT, 0, 100, 800); + + /* Set PWM fractional enable/disable */ + REG_WRITE(MASTER_COMM_DATA_REG1, fractional_pwm); + + /* Set command to enable or disable fractional PWM microcontroller */ + REG_UPDATE(MASTER_COMM_CMD_REG, MASTER_COMM_CMD_REG_BYTE0, + MCP_BL_SET_PWM_FRAC); + + /* Notify microcontroller of new command */ + REG_UPDATE(MASTER_COMM_CNTL_REG, MASTER_COMM_INTERRUPT, 1); + + /* Ensure command has been executed before continuing */ + REG_WAIT(MASTER_COMM_CNTL_REG, MASTER_COMM_INTERRUPT, 0, 100, 800); +} + 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; /* Definition of DC_DMCU_SCRATCH @@ -379,9 +405,14 @@ static bool dcn10_dmcu_init(struct dmcu *dmcu) if (dmcu->dmcu_state == DMCU_RUNNING) { /* Retrieve and cache the DMCU firmware version. */ dcn10_get_dmcu_version(dmcu); + + /* Initialize DMCU to use fractional PWM or not */ + dcn10_dmcu_enable_fractional_pwm(dmcu, + (config->disable_fractional_pwm == false) ? 1 : 0); status = true; - } else + } else { status = false; + } break; case DMCU_RUNNING: @@ -690,7 +721,7 @@ static bool dcn10_is_dmcu_initialized(struct dmcu *dmcu) return true; } -#endif +#endif //(CONFIG_DRM_AMD_DC_DCN1_0) static const struct dmcu_funcs dce_funcs = { .dmcu_init = dce_dmcu_init, diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_clk_mgr.c index 0d9bee8d5ab9..2b2de1d913c9 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_clk_mgr.c @@ -151,9 +151,6 @@ static void dcn1_update_clocks(struct clk_mgr *clk_mgr, struct dc *dc = clk_mgr->ctx->dc; struct dc_debug_options *debug = &dc->debug; struct dc_clocks *new_clocks = &context->bw_ctx.bw.dcn.clk; - struct pp_smu_display_requirement_rv *smu_req_cur = - &dc->res_pool->pp_smu_req; - struct pp_smu_display_requirement_rv smu_req = *smu_req_cur; struct pp_smu_funcs_rv *pp_smu = NULL; bool send_request_to_increase = false; bool send_request_to_lower = false; @@ -175,8 +172,6 @@ static void dcn1_update_clocks(struct clk_mgr *clk_mgr, */ if (pp_smu && pp_smu->set_display_count) pp_smu->set_display_count(&pp_smu->pp_smu, display_count); - - smu_req.display_count = display_count; } if (new_clocks->dispclk_khz > clk_mgr->clks.dispclk_khz @@ -187,7 +182,6 @@ static void dcn1_update_clocks(struct clk_mgr *clk_mgr, if (should_set_clock(safe_to_lower, new_clocks->phyclk_khz, clk_mgr->clks.phyclk_khz)) { clk_mgr->clks.phyclk_khz = new_clocks->phyclk_khz; - send_request_to_lower = true; } @@ -197,24 +191,18 @@ static void dcn1_update_clocks(struct clk_mgr *clk_mgr, if (should_set_clock(safe_to_lower, new_clocks->fclk_khz, clk_mgr->clks.fclk_khz)) { clk_mgr->clks.fclk_khz = new_clocks->fclk_khz; - smu_req.hard_min_fclk_mhz = new_clocks->fclk_khz / 1000; - send_request_to_lower = true; } //DCF Clock if (should_set_clock(safe_to_lower, new_clocks->dcfclk_khz, clk_mgr->clks.dcfclk_khz)) { clk_mgr->clks.dcfclk_khz = new_clocks->dcfclk_khz; - smu_req.hard_min_dcefclk_mhz = new_clocks->dcfclk_khz / 1000; - send_request_to_lower = true; } if (should_set_clock(safe_to_lower, new_clocks->dcfclk_deep_sleep_khz, clk_mgr->clks.dcfclk_deep_sleep_khz)) { clk_mgr->clks.dcfclk_deep_sleep_khz = new_clocks->dcfclk_deep_sleep_khz; - smu_req.min_deep_sleep_dcefclk_mhz = (new_clocks->dcfclk_deep_sleep_khz + 999) / 1000; - send_request_to_lower = true; } @@ -227,9 +215,9 @@ static void dcn1_update_clocks(struct clk_mgr *clk_mgr, pp_smu->set_hard_min_dcfclk_by_freq && pp_smu->set_min_deep_sleep_dcfclk) { - pp_smu->set_hard_min_fclk_by_freq(&pp_smu->pp_smu, smu_req.hard_min_fclk_mhz); - pp_smu->set_hard_min_dcfclk_by_freq(&pp_smu->pp_smu, smu_req.hard_min_dcefclk_mhz); - pp_smu->set_min_deep_sleep_dcfclk(&pp_smu->pp_smu, smu_req.min_deep_sleep_dcefclk_mhz); + pp_smu->set_hard_min_fclk_by_freq(&pp_smu->pp_smu, new_clocks->fclk_khz / 1000); + pp_smu->set_hard_min_dcfclk_by_freq(&pp_smu->pp_smu, new_clocks->dcfclk_khz / 1000); + pp_smu->set_min_deep_sleep_dcfclk(&pp_smu->pp_smu, (new_clocks->dcfclk_deep_sleep_khz + 999) / 1000); } } @@ -239,7 +227,6 @@ static void dcn1_update_clocks(struct clk_mgr *clk_mgr, || new_clocks->dispclk_khz == clk_mgr->clks.dispclk_khz) { dcn1_ramp_up_dispclk_with_dpp(clk_mgr, new_clocks); clk_mgr->clks.dispclk_khz = new_clocks->dispclk_khz; - send_request_to_lower = true; } @@ -249,13 +236,11 @@ static void dcn1_update_clocks(struct clk_mgr *clk_mgr, pp_smu->set_hard_min_dcfclk_by_freq && pp_smu->set_min_deep_sleep_dcfclk) { - pp_smu->set_hard_min_fclk_by_freq(&pp_smu->pp_smu, smu_req.hard_min_fclk_mhz); - pp_smu->set_hard_min_dcfclk_by_freq(&pp_smu->pp_smu, smu_req.hard_min_dcefclk_mhz); - pp_smu->set_min_deep_sleep_dcfclk(&pp_smu->pp_smu, smu_req.min_deep_sleep_dcefclk_mhz); + pp_smu->set_hard_min_fclk_by_freq(&pp_smu->pp_smu, new_clocks->fclk_khz / 1000); + pp_smu->set_hard_min_dcfclk_by_freq(&pp_smu->pp_smu, new_clocks->dcfclk_khz / 1000); + pp_smu->set_min_deep_sleep_dcfclk(&pp_smu->pp_smu, (new_clocks->dcfclk_deep_sleep_khz + 999) / 1000); } } - - *smu_req_cur = smu_req; } static const struct clk_mgr_funcs dcn1_funcs = { .get_dp_ref_clk_frequency = dce12_get_dp_ref_freq_khz, 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 295cbd5b843f..0db2a6e96fc0 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c @@ -283,7 +283,8 @@ void hubbub1_program_watermarks( hubbub1->watermarks.a.urgent_ns = watermarks->a.urgent_ns; prog_wm_value = convert_and_clamp(watermarks->a.urgent_ns, refclk_mhz, 0x1fffff); - REG_WRITE(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_A, prog_wm_value); + REG_SET(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_A, 0, + DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_A, prog_wm_value); DC_LOG_BANDWIDTH_CALCS("URGENCY_WATERMARK_A calculated =%d\n" "HW register value = 0x%x\n", @@ -310,7 +311,8 @@ void hubbub1_program_watermarks( prog_wm_value = convert_and_clamp( watermarks->a.cstate_pstate.cstate_enter_plus_exit_ns, refclk_mhz, 0x1fffff); - REG_WRITE(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_A, prog_wm_value); + REG_SET(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_A, 0, + DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_A, prog_wm_value); 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); @@ -323,7 +325,8 @@ void hubbub1_program_watermarks( prog_wm_value = convert_and_clamp( watermarks->a.cstate_pstate.cstate_exit_ns, refclk_mhz, 0x1fffff); - REG_WRITE(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_A, prog_wm_value); + REG_SET(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_A, 0, + DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_A, prog_wm_value); 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); @@ -337,7 +340,8 @@ void hubbub1_program_watermarks( prog_wm_value = convert_and_clamp( watermarks->a.cstate_pstate.pstate_change_ns, refclk_mhz, 0x1fffff); - REG_WRITE(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_A, prog_wm_value); + REG_SET(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_A, 0, + DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_A, prog_wm_value); 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); @@ -348,7 +352,8 @@ void hubbub1_program_watermarks( hubbub1->watermarks.b.urgent_ns = watermarks->b.urgent_ns; prog_wm_value = convert_and_clamp(watermarks->b.urgent_ns, refclk_mhz, 0x1fffff); - REG_WRITE(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_B, prog_wm_value); + REG_SET(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_B, 0, + DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_B, prog_wm_value); DC_LOG_BANDWIDTH_CALCS("URGENCY_WATERMARK_B calculated =%d\n" "HW register value = 0x%x\n", @@ -375,7 +380,8 @@ void hubbub1_program_watermarks( prog_wm_value = convert_and_clamp( watermarks->b.cstate_pstate.cstate_enter_plus_exit_ns, refclk_mhz, 0x1fffff); - REG_WRITE(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_B, prog_wm_value); + REG_SET(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_B, 0, + DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_B, prog_wm_value); 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); @@ -388,7 +394,8 @@ void hubbub1_program_watermarks( prog_wm_value = convert_and_clamp( watermarks->b.cstate_pstate.cstate_exit_ns, refclk_mhz, 0x1fffff); - REG_WRITE(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_B, prog_wm_value); + REG_SET(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_B, 0, + DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_B, prog_wm_value); 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); @@ -402,7 +409,8 @@ void hubbub1_program_watermarks( prog_wm_value = convert_and_clamp( watermarks->b.cstate_pstate.pstate_change_ns, refclk_mhz, 0x1fffff); - REG_WRITE(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_B, prog_wm_value); + REG_SET(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_B, 0, + DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_B, prog_wm_value); 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); @@ -413,7 +421,8 @@ void hubbub1_program_watermarks( hubbub1->watermarks.c.urgent_ns = watermarks->c.urgent_ns; prog_wm_value = convert_and_clamp(watermarks->c.urgent_ns, refclk_mhz, 0x1fffff); - REG_WRITE(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_C, prog_wm_value); + REG_SET(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_C, 0, + DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_C, prog_wm_value); DC_LOG_BANDWIDTH_CALCS("URGENCY_WATERMARK_C calculated =%d\n" "HW register value = 0x%x\n", @@ -440,7 +449,8 @@ void hubbub1_program_watermarks( prog_wm_value = convert_and_clamp( watermarks->c.cstate_pstate.cstate_enter_plus_exit_ns, refclk_mhz, 0x1fffff); - REG_WRITE(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_C, prog_wm_value); + REG_SET(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_C, 0, + DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_C, prog_wm_value); 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); @@ -453,7 +463,8 @@ void hubbub1_program_watermarks( prog_wm_value = convert_and_clamp( watermarks->c.cstate_pstate.cstate_exit_ns, refclk_mhz, 0x1fffff); - REG_WRITE(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_C, prog_wm_value); + REG_SET(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_C, 0, + DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_C, prog_wm_value); 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); @@ -467,7 +478,8 @@ void hubbub1_program_watermarks( prog_wm_value = convert_and_clamp( watermarks->c.cstate_pstate.pstate_change_ns, refclk_mhz, 0x1fffff); - REG_WRITE(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_C, prog_wm_value); + REG_SET(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_C, 0, + DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_C, prog_wm_value); 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); @@ -478,7 +490,8 @@ void hubbub1_program_watermarks( hubbub1->watermarks.d.urgent_ns = watermarks->d.urgent_ns; prog_wm_value = convert_and_clamp(watermarks->d.urgent_ns, refclk_mhz, 0x1fffff); - REG_WRITE(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_D, prog_wm_value); + REG_SET(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_D, 0, + DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_D, prog_wm_value); DC_LOG_BANDWIDTH_CALCS("URGENCY_WATERMARK_D calculated =%d\n" "HW register value = 0x%x\n", @@ -505,7 +518,8 @@ void hubbub1_program_watermarks( prog_wm_value = convert_and_clamp( watermarks->d.cstate_pstate.cstate_enter_plus_exit_ns, refclk_mhz, 0x1fffff); - REG_WRITE(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_D, prog_wm_value); + REG_SET(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_D, 0, + DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_D, prog_wm_value); 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); @@ -518,7 +532,8 @@ void hubbub1_program_watermarks( prog_wm_value = convert_and_clamp( watermarks->d.cstate_pstate.cstate_exit_ns, refclk_mhz, 0x1fffff); - REG_WRITE(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_D, prog_wm_value); + REG_SET(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_D, 0, + DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_D, prog_wm_value); 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); @@ -532,7 +547,8 @@ void hubbub1_program_watermarks( prog_wm_value = convert_and_clamp( watermarks->d.cstate_pstate.pstate_change_ns, refclk_mhz, 0x1fffff); - REG_WRITE(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D, prog_wm_value); + REG_SET(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D, 0, + DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D, prog_wm_value); 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); @@ -867,6 +883,7 @@ static const struct hubbub_funcs hubbub1_funcs = { .dcc_support_pixel_format = hubbub1_dcc_support_pixel_format, .get_dcc_compression_cap = hubbub1_get_dcc_compression_cap, .wm_read_state = hubbub1_wm_read_state, + .program_watermarks = hubbub1_program_watermarks, }; void hubbub1_construct(struct hubbub *hubbub, 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 9cd4a5194154..85811b24a497 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.h @@ -32,18 +32,14 @@ #define TO_DCN10_HUBBUB(hubbub)\ container_of(hubbub, struct dcn10_hubbub, base) -#define HUBHUB_REG_LIST_DCN()\ +#define HUBBUB_REG_LIST_DCN_COMMON()\ SR(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_A),\ - SR(DCHUBBUB_ARB_PTE_META_URGENCY_WATERMARK_A),\ SR(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_A),\ SR(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_B),\ - SR(DCHUBBUB_ARB_PTE_META_URGENCY_WATERMARK_B),\ SR(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_B),\ SR(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_C),\ - SR(DCHUBBUB_ARB_PTE_META_URGENCY_WATERMARK_C),\ SR(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_C),\ SR(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_D),\ - SR(DCHUBBUB_ARB_PTE_META_URGENCY_WATERMARK_D),\ SR(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D),\ SR(DCHUBBUB_ARB_WATERMARK_CHANGE_CNTL),\ SR(DCHUBBUB_ARB_DRAM_STATE_CNTL),\ @@ -54,6 +50,12 @@ SR(DCHUBBUB_TEST_DEBUG_DATA),\ SR(DCHUBBUB_SOFT_RESET) +#define HUBBUB_VM_REG_LIST() \ + SR(DCHUBBUB_ARB_PTE_META_URGENCY_WATERMARK_A),\ + SR(DCHUBBUB_ARB_PTE_META_URGENCY_WATERMARK_B),\ + SR(DCHUBBUB_ARB_PTE_META_URGENCY_WATERMARK_C),\ + SR(DCHUBBUB_ARB_PTE_META_URGENCY_WATERMARK_D) + #define HUBBUB_SR_WATERMARK_REG_LIST()\ SR(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_A),\ SR(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_A),\ @@ -65,7 +67,8 @@ SR(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_D) #define HUBBUB_REG_LIST_DCN10(id)\ - HUBHUB_REG_LIST_DCN(), \ + HUBBUB_REG_LIST_DCN_COMMON(), \ + HUBBUB_VM_REG_LIST(), \ HUBBUB_SR_WATERMARK_REG_LIST(), \ SR(DCHUBBUB_SDPIF_FB_TOP),\ SR(DCHUBBUB_SDPIF_FB_BASE),\ @@ -122,8 +125,7 @@ struct dcn_hubbub_registers { #define HUBBUB_SF(reg_name, field_name, post_fix)\ .field_name = reg_name ## __ ## field_name ## post_fix - -#define HUBBUB_MASK_SH_LIST_DCN(mask_sh)\ +#define HUBBUB_MASK_SH_LIST_DCN_COMMON(mask_sh)\ HUBBUB_SF(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_ENABLE, mask_sh), \ HUBBUB_SF(DCHUBBUB_SOFT_RESET, DCHUBBUB_GLOBAL_SOFT_RESET, mask_sh), \ HUBBUB_SF(DCHUBBUB_ARB_WATERMARK_CHANGE_CNTL, DCHUBBUB_ARB_WATERMARK_CHANGE_REQUEST, mask_sh), \ @@ -133,10 +135,29 @@ struct dcn_hubbub_registers { HUBBUB_SF(DCHUBBUB_ARB_DRAM_STATE_CNTL, DCHUBBUB_ARB_ALLOW_PSTATE_CHANGE_FORCE_VALUE, mask_sh), \ HUBBUB_SF(DCHUBBUB_ARB_DRAM_STATE_CNTL, DCHUBBUB_ARB_ALLOW_PSTATE_CHANGE_FORCE_ENABLE, mask_sh), \ HUBBUB_SF(DCHUBBUB_ARB_SAT_LEVEL, DCHUBBUB_ARB_SAT_LEVEL, mask_sh), \ - HUBBUB_SF(DCHUBBUB_ARB_DF_REQ_OUTSTAND, DCHUBBUB_ARB_MIN_REQ_OUTSTAND, mask_sh) + HUBBUB_SF(DCHUBBUB_ARB_DF_REQ_OUTSTAND, DCHUBBUB_ARB_MIN_REQ_OUTSTAND, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_A, DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_A, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_B, DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_B, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_C, DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_C, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_D, DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_D, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_A, DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_A, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_B, DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_B, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_C, DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_C, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D, DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D, mask_sh) + +#define HUBBUB_MASK_SH_LIST_STUTTER(mask_sh) \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_A, DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_A, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_B, DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_B, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_C, DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_C, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_D, DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_D, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_A, DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_A, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_B, DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_B, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_C, DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_C, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_D, DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_D, mask_sh) #define HUBBUB_MASK_SH_LIST_DCN10(mask_sh)\ - HUBBUB_MASK_SH_LIST_DCN(mask_sh), \ + HUBBUB_MASK_SH_LIST_DCN_COMMON(mask_sh), \ + HUBBUB_MASK_SH_LIST_STUTTER(mask_sh), \ HUBBUB_SF(DCHUBBUB_SDPIF_FB_TOP, SDPIF_FB_TOP, mask_sh), \ HUBBUB_SF(DCHUBBUB_SDPIF_FB_BASE, SDPIF_FB_BASE, mask_sh), \ HUBBUB_SF(DCHUBBUB_SDPIF_FB_OFFSET, SDPIF_FB_OFFSET, mask_sh), \ @@ -167,15 +188,35 @@ struct dcn_hubbub_registers { type FB_OFFSET;\ type AGP_BOT;\ type AGP_TOP;\ - type AGP_BASE + type AGP_BASE;\ + type DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_A;\ + type DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_B;\ + type DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_C;\ + type DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_D;\ + type DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_A;\ + type DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_B;\ + type DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_C;\ + type DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D + +#define HUBBUB_STUTTER_REG_FIELD_LIST(type) \ + type DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_A;\ + type DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_B;\ + type DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_C;\ + type DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_D;\ + type DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_A;\ + type DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_B;\ + type DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_C;\ + type DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_D struct dcn_hubbub_shift { DCN_HUBBUB_REG_FIELD_LIST(uint8_t); + HUBBUB_STUTTER_REG_FIELD_LIST(uint8_t); }; struct dcn_hubbub_mask { DCN_HUBBUB_REG_FIELD_LIST(uint32_t); + HUBBUB_STUTTER_REG_FIELD_LIST(uint32_t); }; struct dc; diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c index 0ba68d41b9c3..54b219a710d8 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c @@ -1178,6 +1178,10 @@ void hubp1_vtg_sel(struct hubp *hubp, uint32_t otg_inst) REG_UPDATE(DCHUBP_CNTL, HUBP_VTG_SEL, otg_inst); } +void hubp1_init(struct hubp *hubp) +{ + //do nothing +} static const struct hubp_funcs dcn10_hubp_funcs = { .hubp_program_surface_flip_and_addr = hubp1_program_surface_flip_and_addr, @@ -1201,7 +1205,7 @@ static const struct hubp_funcs dcn10_hubp_funcs = { .hubp_clear_underflow = hubp1_clear_underflow, .hubp_disable_control = hubp1_disable_control, .hubp_get_underflow_status = hubp1_get_underflow_status, - + .hubp_init = hubp1_init, }; /*****************************************/ diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h index db98ba361686..99d2b7e2a578 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h @@ -34,6 +34,7 @@ #define HUBP_REG_LIST_DCN(id)\ SRI(DCHUBP_CNTL, HUBP, id),\ SRI(HUBPREQ_DEBUG_DB, HUBP, id),\ + SRI(HUBPREQ_DEBUG, HUBP, id),\ SRI(DCSURF_ADDR_CONFIG, HUBP, id),\ SRI(DCSURF_TILING_CONFIG, HUBP, id),\ SRI(DCSURF_SURFACE_PITCH, HUBPREQ, id),\ @@ -138,6 +139,7 @@ #define HUBP_COMMON_REG_VARIABLE_LIST \ uint32_t DCHUBP_CNTL; \ uint32_t HUBPREQ_DEBUG_DB; \ + uint32_t HUBPREQ_DEBUG; \ uint32_t DCSURF_ADDR_CONFIG; \ uint32_t DCSURF_TILING_CONFIG; \ uint32_t DCSURF_SURFACE_PITCH; \ @@ -749,4 +751,6 @@ enum cursor_pitch hubp1_get_cursor_pitch(unsigned int pitch); void hubp1_vready_workaround(struct hubp *hubp, struct _vcs_dpi_display_pipe_dest_params_st *pipe_dest); +void hubp1_init(struct hubp *hubp); + #endif 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 dab370676bfa..33d311cea28c 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 @@ -1118,14 +1118,17 @@ static void dcn10_init_hw(struct dc *dc) * Otherwise, if taking control is not possible, we need to power * everything down. */ - if (dcb->funcs->is_accelerated_mode(dcb)) { + if (dcb->funcs->is_accelerated_mode(dcb) || dc->config.power_down_display_on_boot) { for (i = 0; i < dc->res_pool->pipe_count; i++) { struct hubp *hubp = dc->res_pool->hubps[i]; struct dpp *dpp = dc->res_pool->dpps[i]; + hubp->funcs->hubp_init(hubp); dc->res_pool->opps[i]->mpc_tree_params.opp_id = dc->res_pool->opps[i]->inst; plane_atomic_power_down(dc, dpp, hubp); } + + apply_DEGVIDCN10_253_wa(dc); } for (i = 0; i < dc->res_pool->audio_count; i++) { @@ -2436,6 +2439,8 @@ static void dcn10_prepare_bandwidth( struct dc *dc, struct dc_state *context) { + struct hubbub *hubbub = dc->res_pool->hubbub; + if (dc->debug.sanity_checks) dcn10_verify_allow_pstate_change_high(dc); @@ -2449,7 +2454,7 @@ static void dcn10_prepare_bandwidth( false); } - hubbub1_program_watermarks(dc->res_pool->hubbub, + hubbub->funcs->program_watermarks(hubbub, &context->bw_ctx.bw.dcn.watermarks, dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000, true); @@ -2466,6 +2471,8 @@ static void dcn10_optimize_bandwidth( struct dc *dc, struct dc_state *context) { + struct hubbub *hubbub = dc->res_pool->hubbub; + if (dc->debug.sanity_checks) dcn10_verify_allow_pstate_change_high(dc); @@ -2479,7 +2486,7 @@ static void dcn10_optimize_bandwidth( true); } - hubbub1_program_watermarks(dc->res_pool->hubbub, + hubbub->funcs->program_watermarks(hubbub, &context->bw_ctx.bw.dcn.watermarks, dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000, true); diff --git a/drivers/gpu/drm/amd/display/dc/dm_pp_smu.h b/drivers/gpu/drm/amd/display/dc/dm_pp_smu.h index cc6891b8ea69..4fc4208d1472 100644 --- a/drivers/gpu/drm/amd/display/dc/dm_pp_smu.h +++ b/drivers/gpu/drm/amd/display/dc/dm_pp_smu.h @@ -74,29 +74,6 @@ struct pp_smu_wm_range_sets { struct pp_smu_wm_set_range writer_wm_sets[MAX_WATERMARK_SETS]; }; -struct pp_smu_display_requirement_rv { - /* PPSMC_MSG_SetDisplayCount: count - * 0 triggers S0i2 optimization - */ - unsigned int display_count; - - /* PPSMC_MSG_SetHardMinFclkByFreq: mhz - * FCLK will vary with DPM, but never below requested hard min - */ - unsigned int hard_min_fclk_mhz; - - /* PPSMC_MSG_SetHardMinDcefclkByFreq: mhz - * fixed clock at requested freq, either from FCH bypass or DFS - */ - unsigned int hard_min_dcefclk_mhz; - - /* PPSMC_MSG_SetMinDeepSleepDcefclk: mhz - * when DF is in cstate, dcf clock is further divided down - * to just above given frequency - */ - unsigned int min_deep_sleep_dcefclk_mhz; -}; - struct pp_smu_funcs_rv { struct pp_smu pp_smu; 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 88a82a23d259..6f5ab05d6467 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/core_types.h +++ b/drivers/gpu/drm/amd/display/dc/inc/core_types.h @@ -145,7 +145,6 @@ struct resource_pool { struct hubbub *hubbub; struct mpc *mpc; struct pp_smu_funcs *pp_smu; - struct pp_smu_display_requirement_rv pp_smu_req; struct dce_aux *engines[MAX_PIPES]; struct dce_i2c_hw *hw_i2cs[MAX_PIPES]; struct dce_i2c_sw *sw_i2cs[MAX_PIPES]; 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 5e8fead3c09a..93667e8b23b3 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h @@ -77,6 +77,12 @@ struct hubbub_funcs { void (*get_dchub_ref_freq)(struct hubbub *hubbub, unsigned int dccg_ref_freq_inKhz, unsigned int *dchub_ref_freq_inKhz); + + void (*program_watermarks)( + struct hubbub *hubbub, + struct dcn_watermark_set *watermarks, + unsigned int refclk_mhz, + bool safe_to_lower); }; struct hubbub { diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dmcu.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dmcu.h index cbaa43853611..c68f0ce346c7 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/dmcu.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dmcu.h @@ -70,6 +70,8 @@ struct dmcu_funcs { void (*get_psr_wait_loop)(struct dmcu *dmcu, unsigned int *psr_wait_loop_number); bool (*is_dmcu_initialized)(struct dmcu *dmcu); + bool (*lock_phy)(struct dmcu *dmcu); + bool (*unlock_phy)(struct dmcu *dmcu); }; #endif diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h b/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h index 1cd07e94ee63..455df4999797 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h @@ -130,6 +130,7 @@ struct hubp_funcs { void (*hubp_clear_underflow)(struct hubp *hubp); void (*hubp_disable_control)(struct hubp *hubp, bool disable_hubp); unsigned int (*hubp_get_underflow_status)(struct hubp *hubp); + void (*hubp_init)(struct hubp *hubp); }; diff --git a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c index 3d867e34f8b3..19b1eaebe484 100644 --- a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c +++ b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c @@ -437,10 +437,8 @@ static void apply_below_the_range(struct core_freesync *core_freesync, inserted_frame_duration_in_us = last_render_time_in_us / frames_to_insert; - if (inserted_frame_duration_in_us < - (1000000 / in_out_vrr->max_refresh_in_uhz)) - inserted_frame_duration_in_us = - (1000000 / in_out_vrr->max_refresh_in_uhz); + if (inserted_frame_duration_in_us < in_out_vrr->min_duration_in_us) + inserted_frame_duration_in_us = in_out_vrr->min_duration_in_us; /* Cache the calculated variables */ in_out_vrr->btr.inserted_duration_in_us = diff --git a/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_1_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_1_0_offset.h index 721c61171045..5a44e614ab7e 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_1_0_offset.h +++ b/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_1_0_offset.h @@ -2347,6 +2347,8 @@ #define mmHUBP0_DCHUBP_VMPG_CONFIG_BASE_IDX 2 #define mmHUBP0_HUBPREQ_DEBUG_DB 0x0569 #define mmHUBP0_HUBPREQ_DEBUG_DB_BASE_IDX 2 +#define mmHUBP0_HUBPREQ_DEBUG 0x056a +#define mmHUBP0_HUBPREQ_DEBUG_BASE_IDX 2 #define mmHUBP0_HUBP_MEASURE_WIN_CTRL_DCFCLK 0x056e #define mmHUBP0_HUBP_MEASURE_WIN_CTRL_DCFCLK_BASE_IDX 2 #define mmHUBP0_HUBP_MEASURE_WIN_CTRL_DPPCLK 0x056f @@ -2631,6 +2633,8 @@ #define mmHUBP1_DCHUBP_VMPG_CONFIG_BASE_IDX 2 #define mmHUBP1_HUBPREQ_DEBUG_DB 0x062d #define mmHUBP1_HUBPREQ_DEBUG_DB_BASE_IDX 2 +#define mmHUBP1_HUBPREQ_DEBUG 0x062e +#define mmHUBP1_HUBPREQ_DEBUG_BASE_IDX 2 #define mmHUBP1_HUBP_MEASURE_WIN_CTRL_DCFCLK 0x0632 #define mmHUBP1_HUBP_MEASURE_WIN_CTRL_DCFCLK_BASE_IDX 2 #define mmHUBP1_HUBP_MEASURE_WIN_CTRL_DPPCLK 0x0633 @@ -2915,6 +2919,8 @@ #define mmHUBP2_DCHUBP_VMPG_CONFIG_BASE_IDX 2 #define mmHUBP2_HUBPREQ_DEBUG_DB 0x06f1 #define mmHUBP2_HUBPREQ_DEBUG_DB_BASE_IDX 2 +#define mmHUBP2_HUBPREQ_DEBUG 0x06f2 +#define mmHUBP2_HUBPREQ_DEBUG_BASE_IDX 2 #define mmHUBP2_HUBP_MEASURE_WIN_CTRL_DCFCLK 0x06f6 #define mmHUBP2_HUBP_MEASURE_WIN_CTRL_DCFCLK_BASE_IDX 2 #define mmHUBP2_HUBP_MEASURE_WIN_CTRL_DPPCLK 0x06f7 @@ -3199,6 +3205,8 @@ #define mmHUBP3_DCHUBP_VMPG_CONFIG_BASE_IDX 2 #define mmHUBP3_HUBPREQ_DEBUG_DB 0x07b5 #define mmHUBP3_HUBPREQ_DEBUG_DB_BASE_IDX 2 +#define mmHUBP3_HUBPREQ_DEBUG 0x07b6 +#define mmHUBP3_HUBPREQ_DEBUG_BASE_IDX 2 #define mmHUBP3_HUBP_MEASURE_WIN_CTRL_DCFCLK 0x07ba #define mmHUBP3_HUBP_MEASURE_WIN_CTRL_DCFCLK_BASE_IDX 2 #define mmHUBP3_HUBP_MEASURE_WIN_CTRL_DPPCLK 0x07bb diff --git a/drivers/gpu/drm/amd/include/atomfirmware.h b/drivers/gpu/drm/amd/include/atomfirmware.h index 08769b4b7a74..d3075adb3297 100644 --- a/drivers/gpu/drm/amd/include/atomfirmware.h +++ b/drivers/gpu/drm/amd/include/atomfirmware.h @@ -718,6 +718,7 @@ enum atom_encoder_caps_def ATOM_ENCODER_CAP_RECORD_HBR2_EN =0x02, // DP1.2 HBR2 setting is qualified and HBR2 can be enabled ATOM_ENCODER_CAP_RECORD_HDMI6Gbps_EN =0x04, // HDMI2.0 6Gbps enable or not. ATOM_ENCODER_CAP_RECORD_HBR3_EN =0x08, // DP1.3 HBR3 is supported by board. + ATOM_ENCODER_CAP_RECORD_USB_C_TYPE =0x100, // the DP connector is a USB-C type. }; struct atom_encoder_caps_record diff --git a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h index 5f3c10ebff08..b897aca9b4c9 100644 --- a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h +++ b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h @@ -85,18 +85,6 @@ enum kgd_memory_pool { KGD_POOL_FRAMEBUFFER = 3, }; -enum kgd_engine_type { - KGD_ENGINE_PFP = 1, - KGD_ENGINE_ME, - KGD_ENGINE_CE, - KGD_ENGINE_MEC1, - KGD_ENGINE_MEC2, - KGD_ENGINE_RLC, - KGD_ENGINE_SDMA1, - KGD_ENGINE_SDMA2, - KGD_ENGINE_MAX -}; - /** * enum kfd_sched_policy * @@ -230,8 +218,6 @@ struct tile_config { * @hqd_sdma_destroy: Destructs and preempts the SDMA queue assigned to that * SDMA hqd slot. * - * @get_fw_version: Returns FW versions from the header - * * @set_scratch_backing_va: Sets VA for scratch backing memory of a VMID. * Only used for no cp scheduling mode * @@ -311,8 +297,6 @@ struct kfd2kgd_calls { struct kgd_dev *kgd, uint8_t vmid); - uint16_t (*get_fw_version)(struct kgd_dev *kgd, - enum kgd_engine_type type); 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); diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.c index f32e3d0aaea6..9a595f7525e6 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.c @@ -35,6 +35,7 @@ #include "smu10_hwmgr.h" #include "power_state.h" #include "soc15_common.h" +#include "smu10.h" #define SMU10_MAX_DEEPSLEEP_DIVIDER_ID 5 #define SMU10_MINIMUM_ENGINE_CLOCK 800 /* 8Mhz, the low boundary of engine clock allowed on this chip */ @@ -204,18 +205,13 @@ static int smu10_set_clock_limit(struct pp_hwmgr *hwmgr, const void *input) return 0; } -static inline uint32_t convert_10k_to_mhz(uint32_t clock) -{ - return (clock + 99) / 100; -} - static int smu10_set_min_deep_sleep_dcefclk(struct pp_hwmgr *hwmgr, uint32_t clock) { struct smu10_hwmgr *smu10_data = (struct smu10_hwmgr *)(hwmgr->backend); if (smu10_data->need_min_deep_sleep_dcefclk && - smu10_data->deep_sleep_dcefclk != convert_10k_to_mhz(clock)) { - smu10_data->deep_sleep_dcefclk = convert_10k_to_mhz(clock); + smu10_data->deep_sleep_dcefclk != clock) { + smu10_data->deep_sleep_dcefclk = clock; smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetMinDeepSleepDcefclk, smu10_data->deep_sleep_dcefclk); @@ -228,8 +224,8 @@ static int smu10_set_hard_min_dcefclk_by_freq(struct pp_hwmgr *hwmgr, uint32_t c struct smu10_hwmgr *smu10_data = (struct smu10_hwmgr *)(hwmgr->backend); if (smu10_data->dcf_actual_hard_min_freq && - smu10_data->dcf_actual_hard_min_freq != convert_10k_to_mhz(clock)) { - smu10_data->dcf_actual_hard_min_freq = convert_10k_to_mhz(clock); + smu10_data->dcf_actual_hard_min_freq != clock) { + smu10_data->dcf_actual_hard_min_freq = clock; smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetHardMinDcefclkByFreq, smu10_data->dcf_actual_hard_min_freq); @@ -242,8 +238,8 @@ static int smu10_set_hard_min_fclk_by_freq(struct pp_hwmgr *hwmgr, uint32_t cloc struct smu10_hwmgr *smu10_data = (struct smu10_hwmgr *)(hwmgr->backend); if (smu10_data->f_actual_hard_min_freq && - smu10_data->f_actual_hard_min_freq != convert_10k_to_mhz(clock)) { - smu10_data->f_actual_hard_min_freq = convert_10k_to_mhz(clock); + smu10_data->f_actual_hard_min_freq != clock) { + smu10_data->f_actual_hard_min_freq = clock; smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetHardMinFclkByFreq, smu10_data->f_actual_hard_min_freq); @@ -572,7 +568,6 @@ static int smu10_dpm_force_dpm_level(struct pp_hwmgr *hwmgr, enum amd_dpm_forced_level level) { struct smu10_hwmgr *data = hwmgr->backend; - struct amdgpu_device *adev = hwmgr->adev; uint32_t min_sclk = hwmgr->display_config->min_core_set_clock; uint32_t min_mclk = hwmgr->display_config->min_mem_set_clock/100; @@ -581,11 +576,6 @@ static int smu10_dpm_force_dpm_level(struct pp_hwmgr *hwmgr, return 0; } - /* Disable UMDPSTATE support on rv2 temporarily */ - if ((adev->asic_type == CHIP_RAVEN) && - (adev->rev_id >= 8)) - return 0; - if (min_sclk < data->gfx_min_freq_limit) min_sclk = data->gfx_min_freq_limit; @@ -1200,6 +1190,94 @@ static void smu10_powergate_vcn(struct pp_hwmgr *hwmgr, bool bgate) } } +static int conv_power_profile_to_pplib_workload(int power_profile) +{ + int pplib_workload = 0; + + switch (power_profile) { + case PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT: + pplib_workload = WORKLOAD_DEFAULT_BIT; + break; + case PP_SMC_POWER_PROFILE_FULLSCREEN3D: + pplib_workload = WORKLOAD_PPLIB_FULL_SCREEN_3D_BIT; + break; + case PP_SMC_POWER_PROFILE_POWERSAVING: + pplib_workload = WORKLOAD_PPLIB_POWER_SAVING_BIT; + break; + case PP_SMC_POWER_PROFILE_VIDEO: + pplib_workload = WORKLOAD_PPLIB_VIDEO_BIT; + break; + case PP_SMC_POWER_PROFILE_VR: + pplib_workload = WORKLOAD_PPLIB_VR_BIT; + break; + case PP_SMC_POWER_PROFILE_COMPUTE: + pplib_workload = WORKLOAD_PPLIB_COMPUTE_BIT; + break; + } + + return pplib_workload; +} + +static int smu10_get_power_profile_mode(struct pp_hwmgr *hwmgr, char *buf) +{ + uint32_t i, size = 0; + static const uint8_t + profile_mode_setting[6][4] = {{70, 60, 0, 0,}, + {70, 60, 1, 3,}, + {90, 60, 0, 0,}, + {70, 60, 0, 0,}, + {70, 90, 0, 0,}, + {30, 60, 0, 6,}, + }; + static const char *profile_name[6] = { + "BOOTUP_DEFAULT", + "3D_FULL_SCREEN", + "POWER_SAVING", + "VIDEO", + "VR", + "COMPUTE"}; + static const char *title[6] = {"NUM", + "MODE_NAME", + "BUSY_SET_POINT", + "FPS", + "USE_RLC_BUSY", + "MIN_ACTIVE_LEVEL"}; + + if (!buf) + return -EINVAL; + + size += sprintf(buf + size, "%s %16s %s %s %s %s\n",title[0], + title[1], title[2], title[3], title[4], title[5]); + + for (i = 0; i <= PP_SMC_POWER_PROFILE_COMPUTE; i++) + size += sprintf(buf + size, "%3d %14s%s: %14d %3d %10d %14d\n", + i, profile_name[i], (i == hwmgr->power_profile_mode) ? "*" : " ", + profile_mode_setting[i][0], profile_mode_setting[i][1], + profile_mode_setting[i][2], profile_mode_setting[i][3]); + + return size; +} + +static int smu10_set_power_profile_mode(struct pp_hwmgr *hwmgr, long *input, uint32_t size) +{ + int workload_type = 0; + + if (input[size] > PP_SMC_POWER_PROFILE_COMPUTE) { + pr_err("Invalid power profile mode %ld\n", input[size]); + return -EINVAL; + } + hwmgr->power_profile_mode = input[size]; + + /* conv PP_SMC_POWER_PROFILE* to WORKLOAD_PPLIB_*_BIT */ + workload_type = + conv_power_profile_to_pplib_workload(hwmgr->power_profile_mode); + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_ActiveProcessNotify, + 1 << workload_type); + + return 0; +} + + static const struct pp_hwmgr_func smu10_hwmgr_funcs = { .backend_init = smu10_hwmgr_backend_init, .backend_fini = smu10_hwmgr_backend_fini, @@ -1241,6 +1319,8 @@ static const struct pp_hwmgr_func smu10_hwmgr_funcs = { .powergate_sdma = smu10_powergate_sdma, .set_hard_min_dcefclk_by_freq = smu10_set_hard_min_dcefclk_by_freq, .set_hard_min_fclk_by_freq = smu10_set_hard_min_fclk_by_freq, + .get_power_profile_mode = smu10_get_power_profile_mode, + .set_power_profile_mode = smu10_set_power_profile_mode, }; int smu10_init_function_pointers(struct pp_hwmgr *hwmgr) diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c index 39a547084e90..9b9f87b84910 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c @@ -3460,7 +3460,18 @@ static void vega20_power_gate_vce(struct pp_hwmgr *hwmgr, bool bgate) return ; data->vce_power_gated = bgate; - vega20_enable_disable_vce_dpm(hwmgr, !bgate); + if (bgate) { + vega20_enable_disable_vce_dpm(hwmgr, !bgate); + amdgpu_device_ip_set_powergating_state(hwmgr->adev, + AMD_IP_BLOCK_TYPE_VCE, + AMD_PG_STATE_GATE); + } else { + amdgpu_device_ip_set_powergating_state(hwmgr->adev, + AMD_IP_BLOCK_TYPE_VCE, + AMD_PG_STATE_UNGATE); + vega20_enable_disable_vce_dpm(hwmgr, !bgate); + } + } static void vega20_power_gate_uvd(struct pp_hwmgr *hwmgr, bool bgate) diff --git a/drivers/gpu/drm/amd/powerplay/inc/rv_ppsmc.h b/drivers/gpu/drm/amd/powerplay/inc/rv_ppsmc.h index a2991fa2e6f8..90879e4092a3 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/rv_ppsmc.h +++ b/drivers/gpu/drm/amd/powerplay/inc/rv_ppsmc.h @@ -85,7 +85,6 @@ #define PPSMC_MSG_SetRccPfcPmeRestoreRegister 0x36 #define PPSMC_Message_Count 0x37 - typedef uint16_t PPSMC_Result; typedef int PPSMC_Msg; diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu10.h b/drivers/gpu/drm/amd/powerplay/inc/smu10.h index 9e837a5014c5..b96520528240 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/smu10.h +++ b/drivers/gpu/drm/amd/powerplay/inc/smu10.h @@ -136,12 +136,14 @@ #define FEATURE_CORE_CSTATES_MASK (1 << FEATURE_CORE_CSTATES_BIT) /* Workload bits */ -#define WORKLOAD_PPLIB_FULL_SCREEN_3D_BIT 0 -#define WORKLOAD_PPLIB_VIDEO_BIT 2 -#define WORKLOAD_PPLIB_VR_BIT 3 -#define WORKLOAD_PPLIB_COMPUTE_BIT 4 -#define WORKLOAD_PPLIB_CUSTOM_BIT 5 -#define WORKLOAD_PPLIB_COUNT 6 +#define WORKLOAD_DEFAULT_BIT 0 +#define WORKLOAD_PPLIB_FULL_SCREEN_3D_BIT 1 +#define WORKLOAD_PPLIB_POWER_SAVING_BIT 2 +#define WORKLOAD_PPLIB_VIDEO_BIT 3 +#define WORKLOAD_PPLIB_VR_BIT 4 +#define WORKLOAD_PPLIB_COMPUTE_BIT 5 +#define WORKLOAD_PPLIB_CUSTOM_BIT 6 +#define WORKLOAD_PPLIB_COUNT 7 typedef struct { /* MP1_EXT_SCRATCH0 */ diff --git a/drivers/gpu/drm/amd/powerplay/smu_v11_0.c b/drivers/gpu/drm/amd/powerplay/smu_v11_0.c index c478b38662d0..92903a4cc4d8 100644 --- a/drivers/gpu/drm/amd/powerplay/smu_v11_0.c +++ b/drivers/gpu/drm/amd/powerplay/smu_v11_0.c @@ -1896,8 +1896,13 @@ set_fan_speed_rpm_failed: static int smu_v11_0_set_xgmi_pstate(struct smu_context *smu, uint32_t pstate) { - /* send msg to SMU to set pstate */ - return 0; + int ret = 0; + mutex_lock(&(smu->mutex)); + ret = smu_send_smc_msg_with_param(smu, + SMU_MSG_SetXgmiMode, + pstate ? XGMI_STATE_D0 : XGMI_STATE_D3); + mutex_unlock(&(smu->mutex)); + return ret; } static const struct smu_funcs smu_v11_0_funcs = { diff --git a/drivers/gpu/drm/arm/display/include/malidp_product.h b/drivers/gpu/drm/arm/display/include/malidp_product.h index b35fc5db866b..1053b11352eb 100644 --- a/drivers/gpu/drm/arm/display/include/malidp_product.h +++ b/drivers/gpu/drm/arm/display/include/malidp_product.h @@ -20,4 +20,16 @@ /* Mali-display product IDs */ #define MALIDP_D71_PRODUCT_ID 0x0071 +union komeda_config_id { + struct { + __u32 max_line_sz:16, + n_pipelines:2, + n_scalers:2, /* number of scalers per pipeline */ + n_layers:3, /* number of layers per pipeline */ + n_richs:3, /* number of rich layers per pipeline */ + reserved_bits:6; + }; + __u32 value; +}; + #endif /* _MALIDP_PRODUCT_H_ */ diff --git a/drivers/gpu/drm/arm/display/komeda/Makefile b/drivers/gpu/drm/arm/display/komeda/Makefile index d593125236ae..412eeba8c39f 100644 --- a/drivers/gpu/drm/arm/display/komeda/Makefile +++ b/drivers/gpu/drm/arm/display/komeda/Makefile @@ -1,14 +1,15 @@ # SPDX-License-Identifier: GPL-2.0 ccflags-y := \ - -I$(src)/../include \ - -I$(src) + -I $(srctree)/$(src)/../include \ + -I $(srctree)/$(src) komeda-y := \ komeda_drv.o \ komeda_dev.o \ komeda_format_caps.o \ komeda_pipeline.o \ + komeda_pipeline_state.o \ komeda_framebuffer.o \ komeda_kms.o \ komeda_crtc.o \ diff --git a/drivers/gpu/drm/arm/display/komeda/d71/d71_component.c b/drivers/gpu/drm/arm/display/komeda/d71/d71_component.c index c56cfc2de147..031e5f305a3c 100644 --- a/drivers/gpu/drm/arm/display/komeda/d71/d71_component.c +++ b/drivers/gpu/drm/arm/display/komeda/d71/d71_component.c @@ -391,7 +391,7 @@ static void d71_compiz_dump(struct komeda_component *c, struct seq_file *sf) seq_printf(sf, "CU_USER_HIGH:\t\t0x%X\n", v[1]); } -struct komeda_component_funcs d71_compiz_funcs = { +static struct komeda_component_funcs d71_compiz_funcs = { .update = d71_compiz_update, .disable = d71_component_disable, .dump_register = d71_compiz_dump, @@ -467,7 +467,7 @@ static void d71_improc_dump(struct komeda_component *c, struct seq_file *sf) seq_printf(sf, "IPS_RGB_YUV_COEFF%u:\t0x%X\n", i, v[i]); } -struct komeda_component_funcs d71_improc_funcs = { +static struct komeda_component_funcs d71_improc_funcs = { .update = d71_improc_update, .disable = d71_component_disable, .dump_register = d71_improc_dump, @@ -543,7 +543,8 @@ static void d71_timing_ctrlr_update(struct komeda_component *c, malidp_write32(reg, BLK_CONTROL, value); } -void d71_timing_ctrlr_dump(struct komeda_component *c, struct seq_file *sf) +static void d71_timing_ctrlr_dump(struct komeda_component *c, + struct seq_file *sf) { u32 v[8], i; @@ -579,7 +580,7 @@ void d71_timing_ctrlr_dump(struct komeda_component *c, struct seq_file *sf) seq_printf(sf, "BS_USER:\t\t0x%X\n", v[4]); } -struct komeda_component_funcs d71_timing_ctrlr_funcs = { +static struct komeda_component_funcs d71_timing_ctrlr_funcs = { .update = d71_timing_ctrlr_update, .disable = d71_timing_ctrlr_disable, .dump_register = d71_timing_ctrlr_dump, diff --git a/drivers/gpu/drm/arm/display/komeda/d71/d71_dev.c b/drivers/gpu/drm/arm/display/komeda/d71/d71_dev.c index 72631d673f85..34506ef7ad40 100644 --- a/drivers/gpu/drm/arm/display/komeda/d71/d71_dev.c +++ b/drivers/gpu/drm/arm/display/komeda/d71/d71_dev.c @@ -243,6 +243,56 @@ static int d71_disable_irq(struct komeda_dev *mdev) return 0; } +static void d71_on_off_vblank(struct komeda_dev *mdev, int master_pipe, bool on) +{ + struct d71_dev *d71 = mdev->chip_data; + struct d71_pipeline *pipe = d71->pipes[master_pipe]; + + malidp_write32_mask(pipe->dou_addr, BLK_IRQ_MASK, + DOU_IRQ_PL0, on ? DOU_IRQ_PL0 : 0); +} + +static int to_d71_opmode(int core_mode) +{ + switch (core_mode) { + case KOMEDA_MODE_DISP0: + return DO0_ACTIVE_MODE; + case KOMEDA_MODE_DISP1: + return DO1_ACTIVE_MODE; + case KOMEDA_MODE_DUAL_DISP: + return DO01_ACTIVE_MODE; + case KOMEDA_MODE_INACTIVE: + return INACTIVE_MODE; + default: + WARN(1, "Unknown operation mode"); + return INACTIVE_MODE; + } +} + +static int d71_change_opmode(struct komeda_dev *mdev, int new_mode) +{ + struct d71_dev *d71 = mdev->chip_data; + u32 opmode = to_d71_opmode(new_mode); + int ret; + + malidp_write32_mask(d71->gcu_addr, BLK_CONTROL, 0x7, opmode); + + ret = dp_wait_cond(((malidp_read32(d71->gcu_addr, BLK_CONTROL) & 0x7) == opmode), + 100, 1000, 10000); + + return ret > 0 ? 0 : -ETIMEDOUT; +} + +static void d71_flush(struct komeda_dev *mdev, + int master_pipe, u32 active_pipes) +{ + struct d71_dev *d71 = mdev->chip_data; + u32 reg_offset = (master_pipe == 0) ? + GCU_CONFIG_VALID0 : GCU_CONFIG_VALID1; + + malidp_write32(d71->gcu_addr, reg_offset, GCU_CONFIG_CVAL); +} + static int d71_reset(struct d71_dev *d71) { u32 __iomem *gcu = d71->gcu_addr; @@ -459,6 +509,9 @@ static struct komeda_dev_funcs d71_chip_funcs = { .irq_handler = d71_irq_handler, .enable_irq = d71_enable_irq, .disable_irq = d71_disable_irq, + .on_off_vblank = d71_on_off_vblank, + .change_opmode = d71_change_opmode, + .flush = d71_flush, }; struct komeda_dev_funcs * @@ -467,6 +520,7 @@ d71_identify(u32 __iomem *reg_base, struct komeda_chip_info *chip) chip->arch_id = malidp_read32(reg_base, GLB_ARCH_ID); chip->core_id = malidp_read32(reg_base, GLB_CORE_ID); chip->core_info = malidp_read32(reg_base, GLB_CORE_INFO); + chip->bus_width = D71_BUS_WIDTH_16_BYTES; return &d71_chip_funcs; } diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_crtc.c b/drivers/gpu/drm/arm/display/komeda/komeda_crtc.c index f88a14927be9..62fad59f5a6a 100644 --- a/drivers/gpu/drm/arm/display/komeda/komeda_crtc.c +++ b/drivers/gpu/drm/arm/display/komeda/komeda_crtc.c @@ -18,6 +18,144 @@ #include "komeda_dev.h" #include "komeda_kms.h" +/** + * komeda_crtc_atomic_check - build display output data flow + * @crtc: DRM crtc + * @state: the crtc state object + * + * crtc_atomic_check is the final check stage, so beside build a display data + * pipeline according to the crtc_state, but still needs to release or disable + * the unclaimed pipeline resources. + * + * RETURNS: + * Zero for success or -errno + */ +static int +komeda_crtc_atomic_check(struct drm_crtc *crtc, + struct drm_crtc_state *state) +{ + struct komeda_crtc *kcrtc = to_kcrtc(crtc); + struct komeda_crtc_state *kcrtc_st = to_kcrtc_st(state); + int err; + + if (state->active) { + err = komeda_build_display_data_flow(kcrtc, kcrtc_st); + if (err) + return err; + } + + /* release unclaimed pipeline resources */ + err = komeda_release_unclaimed_resources(kcrtc->master, kcrtc_st); + if (err) + return err; + + return 0; +} + +static u32 komeda_calc_mclk(struct komeda_crtc_state *kcrtc_st) +{ + unsigned long mclk = kcrtc_st->base.adjusted_mode.clock * 1000; + + return mclk; +} + +/* For active a crtc, mainly need two parts of preparation + * 1. adjust display operation mode. + * 2. enable needed clk + */ +static int +komeda_crtc_prepare(struct komeda_crtc *kcrtc) +{ + struct komeda_dev *mdev = kcrtc->base.dev->dev_private; + struct komeda_pipeline *master = kcrtc->master; + struct komeda_crtc_state *kcrtc_st = to_kcrtc_st(kcrtc->base.state); + unsigned long pxlclk_rate = kcrtc_st->base.adjusted_mode.clock * 1000; + u32 new_mode; + int err; + + mutex_lock(&mdev->lock); + + new_mode = mdev->dpmode | BIT(master->id); + if (WARN_ON(new_mode == mdev->dpmode)) { + err = 0; + goto unlock; + } + + err = mdev->funcs->change_opmode(mdev, new_mode); + if (err) { + DRM_ERROR("failed to change opmode: 0x%x -> 0x%x.\n,", + mdev->dpmode, new_mode); + goto unlock; + } + + mdev->dpmode = new_mode; + /* Only need to enable mclk on single display mode, but no need to + * enable mclk it on dual display mode, since the dual mode always + * switch from single display mode, the mclk already enabled, no need + * to enable it again. + */ + if (new_mode != KOMEDA_MODE_DUAL_DISP) { + err = clk_set_rate(mdev->mclk, komeda_calc_mclk(kcrtc_st)); + if (err) + DRM_ERROR("failed to set mclk.\n"); + err = clk_prepare_enable(mdev->mclk); + if (err) + DRM_ERROR("failed to enable mclk.\n"); + } + + err = clk_prepare_enable(master->aclk); + if (err) + DRM_ERROR("failed to enable axi clk for pipe%d.\n", master->id); + err = clk_set_rate(master->pxlclk, pxlclk_rate); + if (err) + DRM_ERROR("failed to set pxlclk for pipe%d\n", master->id); + err = clk_prepare_enable(master->pxlclk); + if (err) + DRM_ERROR("failed to enable pxl clk for pipe%d.\n", master->id); + +unlock: + mutex_unlock(&mdev->lock); + + return err; +} + +static int +komeda_crtc_unprepare(struct komeda_crtc *kcrtc) +{ + struct komeda_dev *mdev = kcrtc->base.dev->dev_private; + struct komeda_pipeline *master = kcrtc->master; + u32 new_mode; + int err; + + mutex_lock(&mdev->lock); + + new_mode = mdev->dpmode & (~BIT(master->id)); + + if (WARN_ON(new_mode == mdev->dpmode)) { + err = 0; + goto unlock; + } + + err = mdev->funcs->change_opmode(mdev, new_mode); + if (err) { + DRM_ERROR("failed to change opmode: 0x%x -> 0x%x.\n,", + mdev->dpmode, new_mode); + goto unlock; + } + + mdev->dpmode = new_mode; + + clk_disable_unprepare(master->pxlclk); + clk_disable_unprepare(master->aclk); + if (new_mode == KOMEDA_MODE_INACTIVE) + clk_disable_unprepare(mdev->mclk); + +unlock: + mutex_unlock(&mdev->lock); + + return err; +} + void komeda_crtc_handle_event(struct komeda_crtc *kcrtc, struct komeda_events *evts) { @@ -31,15 +169,264 @@ void komeda_crtc_handle_event(struct komeda_crtc *kcrtc, if (events & KOMEDA_EVENT_EOW) DRM_DEBUG("EOW.\n"); - /* will handle it with crtc->flush */ - if (events & KOMEDA_EVENT_FLIP) - DRM_DEBUG("FLIP Done.\n"); + if (events & KOMEDA_EVENT_FLIP) { + unsigned long flags; + struct drm_pending_vblank_event *event; + + spin_lock_irqsave(&crtc->dev->event_lock, flags); + if (kcrtc->disable_done) { + complete_all(kcrtc->disable_done); + kcrtc->disable_done = NULL; + } else if (crtc->state->event) { + event = crtc->state->event; + /* + * Consume event before notifying drm core that flip + * happened. + */ + crtc->state->event = NULL; + drm_crtc_send_vblank_event(crtc, event); + } else { + DRM_WARN("CRTC[%d]: FLIP happen but no pending commit.\n", + drm_crtc_index(&kcrtc->base)); + } + spin_unlock_irqrestore(&crtc->dev->event_lock, flags); + } +} + +static void +komeda_crtc_do_flush(struct drm_crtc *crtc, + struct drm_crtc_state *old) +{ + struct komeda_crtc *kcrtc = to_kcrtc(crtc); + struct komeda_crtc_state *kcrtc_st = to_kcrtc_st(crtc->state); + struct komeda_dev *mdev = kcrtc->base.dev->dev_private; + struct komeda_pipeline *master = kcrtc->master; + + DRM_DEBUG_ATOMIC("CRTC%d_FLUSH: active_pipes: 0x%x, affected: 0x%x.\n", + drm_crtc_index(crtc), + kcrtc_st->active_pipes, kcrtc_st->affected_pipes); + + /* step 1: update the pipeline/component state to HW */ + if (has_bit(master->id, kcrtc_st->affected_pipes)) + komeda_pipeline_update(master, old->state); + + /* step 2: notify the HW to kickoff the update */ + mdev->funcs->flush(mdev, master->id, kcrtc_st->active_pipes); } -struct drm_crtc_helper_funcs komeda_crtc_helper_funcs = { +static void +komeda_crtc_atomic_enable(struct drm_crtc *crtc, + struct drm_crtc_state *old) +{ + komeda_crtc_prepare(to_kcrtc(crtc)); + drm_crtc_vblank_on(crtc); + komeda_crtc_do_flush(crtc, old); +} + +static void +komeda_crtc_atomic_disable(struct drm_crtc *crtc, + struct drm_crtc_state *old) +{ + struct komeda_crtc *kcrtc = to_kcrtc(crtc); + struct komeda_crtc_state *old_st = to_kcrtc_st(old); + struct komeda_dev *mdev = crtc->dev->dev_private; + struct komeda_pipeline *master = kcrtc->master; + struct completion *disable_done = &crtc->state->commit->flip_done; + struct completion temp; + int timeout; + + DRM_DEBUG_ATOMIC("CRTC%d_DISABLE: active_pipes: 0x%x, affected: 0x%x.\n", + drm_crtc_index(crtc), + old_st->active_pipes, old_st->affected_pipes); + + if (has_bit(master->id, old_st->active_pipes)) + komeda_pipeline_disable(master, old->state); + + /* crtc_disable has two scenarios according to the state->active switch. + * 1. active -> inactive + * this commit is a disable commit. and the commit will be finished + * or done after the disable operation. on this case we can directly + * use the crtc->state->event to tracking the HW disable operation. + * 2. active -> active + * the crtc->commit is not for disable, but a modeset operation when + * crtc is active, such commit actually has been completed by 3 + * DRM operations: + * crtc_disable, update_planes(crtc_flush), crtc_enable + * so on this case the crtc->commit is for the whole process. + * we can not use it for tracing the disable, we need a temporary + * flip_done for tracing the disable. and crtc->state->event for + * the crtc_enable operation. + * That's also the reason why skip modeset commit in + * komeda_crtc_atomic_flush() + */ + if (crtc->state->active) { + struct komeda_pipeline_state *pipe_st; + /* clear the old active_comps to zero */ + pipe_st = komeda_pipeline_get_old_state(master, old->state); + pipe_st->active_comps = 0; + + init_completion(&temp); + kcrtc->disable_done = &temp; + disable_done = &temp; + } + + mdev->funcs->flush(mdev, master->id, 0); + + /* wait the disable take affect.*/ + timeout = wait_for_completion_timeout(disable_done, HZ); + if (timeout == 0) { + DRM_ERROR("disable pipeline%d timeout.\n", kcrtc->master->id); + if (crtc->state->active) { + unsigned long flags; + + spin_lock_irqsave(&crtc->dev->event_lock, flags); + kcrtc->disable_done = NULL; + spin_unlock_irqrestore(&crtc->dev->event_lock, flags); + } + } + + drm_crtc_vblank_off(crtc); + komeda_crtc_unprepare(kcrtc); +} + +static void +komeda_crtc_atomic_flush(struct drm_crtc *crtc, + struct drm_crtc_state *old) +{ + /* commit with modeset will be handled in enable/disable */ + if (drm_atomic_crtc_needs_modeset(crtc->state)) + return; + + komeda_crtc_do_flush(crtc, old); +} + +static enum drm_mode_status +komeda_crtc_mode_valid(struct drm_crtc *crtc, const struct drm_display_mode *m) +{ + struct komeda_dev *mdev = crtc->dev->dev_private; + struct komeda_crtc *kcrtc = to_kcrtc(crtc); + struct komeda_pipeline *master = kcrtc->master; + long mode_clk, pxlclk; + + if (m->flags & DRM_MODE_FLAG_INTERLACE) + return MODE_NO_INTERLACE; + + /* main clock/AXI clk must be faster than pxlclk*/ + mode_clk = m->clock * 1000; + pxlclk = clk_round_rate(master->pxlclk, mode_clk); + if (pxlclk != mode_clk) { + DRM_DEBUG_ATOMIC("pxlclk doesn't support %ld Hz\n", mode_clk); + + return MODE_NOCLOCK; + } + + if (clk_round_rate(mdev->mclk, mode_clk) < pxlclk) { + DRM_DEBUG_ATOMIC("mclk can't satisfy the requirement of %s-clk: %ld.\n", + m->name, pxlclk); + + return MODE_CLOCK_HIGH; + } + + if (clk_round_rate(master->aclk, mode_clk) < pxlclk) { + DRM_DEBUG_ATOMIC("aclk can't satisfy the requirement of %s-clk: %ld.\n", + m->name, pxlclk); + + return MODE_CLOCK_HIGH; + } + + return MODE_OK; +} + +static bool komeda_crtc_mode_fixup(struct drm_crtc *crtc, + const struct drm_display_mode *m, + struct drm_display_mode *adjusted_mode) +{ + struct komeda_crtc *kcrtc = to_kcrtc(crtc); + struct komeda_pipeline *master = kcrtc->master; + long mode_clk = m->clock * 1000; + + adjusted_mode->clock = clk_round_rate(master->pxlclk, mode_clk) / 1000; + + return true; +} + +static struct drm_crtc_helper_funcs komeda_crtc_helper_funcs = { + .atomic_check = komeda_crtc_atomic_check, + .atomic_flush = komeda_crtc_atomic_flush, + .atomic_enable = komeda_crtc_atomic_enable, + .atomic_disable = komeda_crtc_atomic_disable, + .mode_valid = komeda_crtc_mode_valid, + .mode_fixup = komeda_crtc_mode_fixup, }; +static void komeda_crtc_reset(struct drm_crtc *crtc) +{ + struct komeda_crtc_state *state; + + if (crtc->state) + __drm_atomic_helper_crtc_destroy_state(crtc->state); + + kfree(to_kcrtc_st(crtc->state)); + crtc->state = NULL; + + state = kzalloc(sizeof(*state), GFP_KERNEL); + if (state) { + crtc->state = &state->base; + crtc->state->crtc = crtc; + } +} + +static struct drm_crtc_state * +komeda_crtc_atomic_duplicate_state(struct drm_crtc *crtc) +{ + struct komeda_crtc_state *old = to_kcrtc_st(crtc->state); + struct komeda_crtc_state *new; + + new = kzalloc(sizeof(*new), GFP_KERNEL); + if (!new) + return NULL; + + __drm_atomic_helper_crtc_duplicate_state(crtc, &new->base); + + new->affected_pipes = old->active_pipes; + + return &new->base; +} + +static void komeda_crtc_atomic_destroy_state(struct drm_crtc *crtc, + struct drm_crtc_state *state) +{ + __drm_atomic_helper_crtc_destroy_state(state); + kfree(to_kcrtc_st(state)); +} + +static int komeda_crtc_vblank_enable(struct drm_crtc *crtc) +{ + struct komeda_dev *mdev = crtc->dev->dev_private; + struct komeda_crtc *kcrtc = to_kcrtc(crtc); + + mdev->funcs->on_off_vblank(mdev, kcrtc->master->id, true); + return 0; +} + +static void komeda_crtc_vblank_disable(struct drm_crtc *crtc) +{ + struct komeda_dev *mdev = crtc->dev->dev_private; + struct komeda_crtc *kcrtc = to_kcrtc(crtc); + + mdev->funcs->on_off_vblank(mdev, kcrtc->master->id, false); +} + static const struct drm_crtc_funcs komeda_crtc_funcs = { + .gamma_set = drm_atomic_helper_legacy_gamma_set, + .destroy = drm_crtc_cleanup, + .set_config = drm_atomic_helper_set_config, + .page_flip = drm_atomic_helper_page_flip, + .reset = komeda_crtc_reset, + .atomic_duplicate_state = komeda_crtc_atomic_duplicate_state, + .atomic_destroy_state = komeda_crtc_atomic_destroy_state, + .enable_vblank = komeda_crtc_vblank_enable, + .disable_vblank = komeda_crtc_vblank_disable, }; int komeda_kms_setup_crtcs(struct komeda_kms_dev *kms, diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_dev.c b/drivers/gpu/drm/arm/display/komeda/komeda_dev.c index 24548b87e182..ca3599e4a4d3 100644 --- a/drivers/gpu/drm/arm/display/komeda/komeda_dev.c +++ b/drivers/gpu/drm/arm/display/komeda/komeda_dev.c @@ -59,6 +59,48 @@ static void komeda_debugfs_init(struct komeda_dev *mdev) } #endif +static ssize_t +core_id_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct komeda_dev *mdev = dev_to_mdev(dev); + + return snprintf(buf, PAGE_SIZE, "0x%08x\n", mdev->chip.core_id); +} +static DEVICE_ATTR_RO(core_id); + +static ssize_t +config_id_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct komeda_dev *mdev = dev_to_mdev(dev); + struct komeda_pipeline *pipe = mdev->pipelines[0]; + union komeda_config_id config_id; + int i; + + memset(&config_id, 0, sizeof(config_id)); + + config_id.max_line_sz = pipe->layers[0]->hsize_in.end; + config_id.n_pipelines = mdev->n_pipelines; + config_id.n_scalers = pipe->n_scalers; + config_id.n_layers = pipe->n_layers; + config_id.n_richs = 0; + for (i = 0; i < pipe->n_layers; i++) { + if (pipe->layers[i]->layer_type == KOMEDA_FMT_RICH_LAYER) + config_id.n_richs++; + } + return snprintf(buf, PAGE_SIZE, "0x%08x\n", config_id.value); +} +static DEVICE_ATTR_RO(config_id); + +static struct attribute *komeda_sysfs_entries[] = { + &dev_attr_core_id.attr, + &dev_attr_config_id.attr, + NULL, +}; + +static struct attribute_group komeda_sysfs_attr_group = { + .attrs = komeda_sysfs_entries, +}; + static int komeda_parse_pipe_dt(struct komeda_dev *mdev, struct device_node *np) { struct komeda_pipeline *pipe; @@ -151,6 +193,8 @@ struct komeda_dev *komeda_dev_create(struct device *dev) if (!mdev) return ERR_PTR(-ENOMEM); + mutex_init(&mdev->lock); + mdev->dev = dev; mdev->reg_base = devm_ioremap_resource(dev, io_res); if (IS_ERR(mdev->reg_base)) { @@ -205,6 +249,12 @@ struct komeda_dev *komeda_dev_create(struct device *dev) goto err_cleanup; } + err = sysfs_create_group(&dev->kobj, &komeda_sysfs_attr_group); + if (err) { + DRM_ERROR("create sysfs group failed.\n"); + goto err_cleanup; + } + #ifdef CONFIG_DEBUG_FS komeda_debugfs_init(mdev); #endif @@ -222,6 +272,8 @@ void komeda_dev_destroy(struct komeda_dev *mdev) struct komeda_dev_funcs *funcs = mdev->funcs; int i; + sysfs_remove_group(&dev->kobj, &komeda_sysfs_attr_group); + #ifdef CONFIG_DEBUG_FS debugfs_remove_recursive(mdev->debugfs_root); #endif diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_dev.h b/drivers/gpu/drm/arm/display/komeda/komeda_dev.h index 8eae2620ce77..29e03c4e1ffc 100644 --- a/drivers/gpu/drm/arm/display/komeda/komeda_dev.h +++ b/drivers/gpu/drm/arm/display/komeda/komeda_dev.h @@ -103,9 +103,38 @@ struct komeda_dev_funcs { int (*enable_irq)(struct komeda_dev *mdev); /** @disable_irq: disable irq */ int (*disable_irq)(struct komeda_dev *mdev); + /** @on_off_vblank: notify HW to on/off vblank */ + void (*on_off_vblank)(struct komeda_dev *mdev, + int master_pipe, bool on); /** @dump_register: Optional, dump registers to seq_file */ void (*dump_register)(struct komeda_dev *mdev, struct seq_file *seq); + /** + * @change_opmode: + * + * Notify HW to switch to a new display operation mode. + */ + int (*change_opmode)(struct komeda_dev *mdev, int new_mode); + /** @flush: Notify the HW to flush or kickoff the update */ + void (*flush)(struct komeda_dev *mdev, + int master_pipe, u32 active_pipes); +}; + +/* + * DISPLAY_MODE describes how many display been enabled, and which will be + * passed to CHIP by &komeda_dev_funcs->change_opmode(), then CHIP can do the + * pipeline resources assignment according to this usage hint. + * - KOMEDA_MODE_DISP0: Only one display enabled, pipeline-0 work as master. + * - KOMEDA_MODE_DISP1: Only one display enabled, pipeline-0 work as master. + * - KOMEDA_MODE_DUAL_DISP: Dual display mode, both display has been enabled. + * And D71 supports assign two pipelines to one single display on mode + * KOMEDA_MODE_DISP0/DISP1 + */ +enum { + KOMEDA_MODE_INACTIVE = 0, + KOMEDA_MODE_DISP0 = BIT(0), + KOMEDA_MODE_DISP1 = BIT(1), + KOMEDA_MODE_DUAL_DISP = KOMEDA_MODE_DISP0 | KOMEDA_MODE_DISP1, }; /** @@ -116,21 +145,31 @@ struct komeda_dev_funcs { * control-abilites of device. */ struct komeda_dev { + /** @dev: the base device structure */ struct device *dev; + /** @reg_base: the base address of komeda io space */ u32 __iomem *reg_base; + /** @chip: the basic chip information */ struct komeda_chip_info chip; /** @fmt_tbl: initialized by &komeda_dev_funcs->init_format_table */ struct komeda_format_caps_table fmt_tbl; /** @pclk: APB clock for register access */ struct clk *pclk; - /** @mck: HW main engine clk */ + /** @mclk: HW main engine clk */ struct clk *mclk; /** @irq: irq number */ int irq; + /** @lock: used to protect dpmode */ + struct mutex lock; + /** @dpmode: current display mode */ + u32 dpmode; + + /** @n_pipelines: the number of pipe in @pipelines */ int n_pipelines; + /** @pipelines: the komeda pipelines */ struct komeda_pipeline *pipelines[KOMEDA_MAX_PIPELINES]; /** @funcs: chip funcs to access to HW */ @@ -143,6 +182,7 @@ struct komeda_dev { */ void *chip_data; + /** @debugfs_root: root directory of komeda debugfs */ struct dentry *debugfs_root; }; @@ -158,4 +198,6 @@ d71_identify(u32 __iomem *reg, struct komeda_chip_info *chip); struct komeda_dev *komeda_dev_create(struct device *dev); void komeda_dev_destroy(struct komeda_dev *mdev); +struct komeda_dev *dev_to_mdev(struct device *dev); + #endif /*_KOMEDA_DEV_H_*/ diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_drv.c b/drivers/gpu/drm/arm/display/komeda/komeda_drv.c index 2bdd189b041d..cfa5068d9d1e 100644 --- a/drivers/gpu/drm/arm/display/komeda/komeda_drv.c +++ b/drivers/gpu/drm/arm/display/komeda/komeda_drv.c @@ -17,6 +17,13 @@ struct komeda_drv { struct komeda_kms_dev *kms; }; +struct komeda_dev *dev_to_mdev(struct device *dev) +{ + struct komeda_drv *mdrv = dev_get_drvdata(dev); + + return mdrv ? mdrv->mdev : NULL; +} + static void komeda_unbind(struct device *dev) { struct komeda_drv *mdrv = dev_get_drvdata(dev); @@ -120,7 +127,7 @@ static const struct komeda_product_data komeda_products[] = { }, }; -const struct of_device_id komeda_of_match[] = { +static const struct of_device_id komeda_of_match[] = { { .compatible = "arm,mali-d71", .data = &komeda_products[MALI_D71], }, {}, }; diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_framebuffer.h b/drivers/gpu/drm/arm/display/komeda/komeda_framebuffer.h index 0de2e4a2afd2..ea2fe190c1e3 100644 --- a/drivers/gpu/drm/arm/display/komeda/komeda_framebuffer.h +++ b/drivers/gpu/drm/arm/display/komeda/komeda_framebuffer.h @@ -10,11 +10,16 @@ #include <drm/drm_framebuffer.h> #include "komeda_format_caps.h" -/** struct komeda_fb - entend drm_framebuffer with komeda attribute */ +/** + * struct komeda_fb - Entending drm_framebuffer with komeda attribute + */ struct komeda_fb { /** @base: &drm_framebuffer */ struct drm_framebuffer base; - /* @format_caps: &komeda_format_caps */ + /** + * @format_caps: + * extends drm_format_info for komeda specific information + */ const struct komeda_format_caps *format_caps; /** @aligned_w: aligned frame buffer width */ u32 aligned_w; diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_kms.c b/drivers/gpu/drm/arm/display/komeda/komeda_kms.c index b214edbfbbc6..86f6542afb40 100644 --- a/drivers/gpu/drm/arm/display/komeda/komeda_kms.c +++ b/drivers/gpu/drm/arm/display/komeda/komeda_kms.c @@ -26,10 +26,10 @@ static int komeda_gem_cma_dumb_create(struct drm_file *file, struct drm_device *dev, struct drm_mode_create_dumb *args) { - u32 alignment = 16; /* TODO get alignment from dev */ + struct komeda_dev *mdev = dev->dev_private; + u32 pitch = DIV_ROUND_UP(args->width * args->bpp, 8); - args->pitch = ALIGN(DIV_ROUND_UP(args->width * args->bpp, 8), - alignment); + args->pitch = ALIGN(pitch, mdev->chip.bus_width); return drm_gem_cma_dumb_create_internal(file, dev, args); } @@ -100,9 +100,37 @@ static const struct drm_mode_config_helper_funcs komeda_mode_config_helpers = { .atomic_commit_tail = komeda_kms_commit_tail, }; +static int komeda_kms_check(struct drm_device *dev, + struct drm_atomic_state *state) +{ + struct drm_crtc *crtc; + struct drm_crtc_state *old_crtc_st, *new_crtc_st; + int i, err; + + err = drm_atomic_helper_check_modeset(dev, state); + if (err) + return err; + + /* komeda need to re-calculate resource assumption in every commit + * so need to add all affected_planes (even unchanged) to + * drm_atomic_state. + */ + for_each_oldnew_crtc_in_state(state, crtc, old_crtc_st, new_crtc_st, i) { + err = drm_atomic_add_affected_planes(state, crtc); + if (err) + return err; + } + + err = drm_atomic_helper_check_planes(dev, state); + if (err) + return err; + + return 0; +} + static const struct drm_mode_config_funcs komeda_mode_config_funcs = { .fb_create = komeda_fb_create, - .atomic_check = drm_atomic_helper_check, + .atomic_check = komeda_kms_check, .atomic_commit = drm_atomic_helper_commit, }; @@ -184,6 +212,7 @@ uninstall_irq: drm_irq_uninstall(drm); cleanup_mode_config: drm_mode_config_cleanup(drm); + komeda_kms_cleanup_private_objs(kms); free_kms: kfree(kms); return ERR_PTR(err); @@ -198,7 +227,7 @@ void komeda_kms_detach(struct komeda_kms_dev *kms) drm_dev_unregister(drm); drm_irq_uninstall(drm); component_unbind_all(mdev->dev, drm); - komeda_kms_cleanup_private_objs(mdev); + komeda_kms_cleanup_private_objs(kms); drm_mode_config_cleanup(drm); drm->dev_private = NULL; drm_dev_put(drm); diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_kms.h b/drivers/gpu/drm/arm/display/komeda/komeda_kms.h index 15ac8b85506c..ac3d9209b4d9 100644 --- a/drivers/gpu/drm/arm/display/komeda/komeda_kms.h +++ b/drivers/gpu/drm/arm/display/komeda/komeda_kms.h @@ -15,7 +15,9 @@ #include <video/videomode.h> #include <video/display_timing.h> -/** struct komeda_plane - komeda instance of drm_plane */ +/** + * struct komeda_plane - komeda instance of drm_plane + */ struct komeda_plane { /** @base: &drm_plane */ struct drm_plane base; @@ -70,9 +72,14 @@ struct komeda_crtc { * merge into the master. */ struct komeda_pipeline *slave; + + /** @disable_done: this flip_done is for tracing the disable */ + struct completion *disable_done; }; -/** struct komeda_crtc_state */ +/** + * struct komeda_crtc_state + */ struct komeda_crtc_state { /** @base: &drm_crtc_state */ struct drm_crtc_state base; @@ -80,7 +87,15 @@ struct komeda_crtc_state { /* private properties */ /* computed state which are used by validate/check */ + /** + * @affected_pipes: + * the affected pipelines in once display instance + */ u32 affected_pipes; + /** + * @active_pipes: + * the active pipelines in once display instance + */ u32 active_pipes; }; @@ -108,7 +123,7 @@ int komeda_kms_add_crtcs(struct komeda_kms_dev *kms, struct komeda_dev *mdev); int komeda_kms_add_planes(struct komeda_kms_dev *kms, struct komeda_dev *mdev); int komeda_kms_add_private_objs(struct komeda_kms_dev *kms, struct komeda_dev *mdev); -void komeda_kms_cleanup_private_objs(struct komeda_dev *mdev); +void komeda_kms_cleanup_private_objs(struct komeda_kms_dev *kms); void komeda_crtc_handle_event(struct komeda_crtc *kcrtc, struct komeda_events *evts); diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_pipeline.c b/drivers/gpu/drm/arm/display/komeda/komeda_pipeline.c index 07398efc40f5..c379439c6194 100644 --- a/drivers/gpu/drm/arm/display/komeda/komeda_pipeline.c +++ b/drivers/gpu/drm/arm/display/komeda/komeda_pipeline.c @@ -62,7 +62,7 @@ void komeda_pipeline_destroy(struct komeda_dev *mdev, devm_kfree(mdev->dev, pipe); } -struct komeda_component ** +static struct komeda_component ** komeda_pipeline_get_component_pos(struct komeda_pipeline *pipe, int id) { struct komeda_dev *mdev = pipe->mdev; diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_pipeline.h b/drivers/gpu/drm/arm/display/komeda/komeda_pipeline.h index c30a790d0712..b1f813a349a4 100644 --- a/drivers/gpu/drm/arm/display/komeda/komeda_pipeline.h +++ b/drivers/gpu/drm/arm/display/komeda/komeda_pipeline.h @@ -90,32 +90,35 @@ struct komeda_component { u32 __iomem *reg; /** @id: component id */ u32 id; - /** @hw_ic: component hw id, - * which is initialized by chip and used by chip only + /** + * @hw_id: component hw id, + * which is initialized by chip and used by chip only */ u32 hw_id; /** * @max_active_inputs: - * @max_active_outpus: + * @max_active_outputs: * - * maximum number of inputs/outputs that can be active in the same time + * maximum number of inputs/outputs that can be active at the same time * Note: * the number isn't the bit number of @supported_inputs or * @supported_outputs, but may be less than it, since component may not * support enabling all @supported_inputs/outputs at the same time. */ u8 max_active_inputs; + /** @max_active_outputs: maximum number of outputs */ u8 max_active_outputs; /** * @supported_inputs: * @supported_outputs: * - * bitmask of BIT(component->id) for the supported inputs/outputs + * bitmask of BIT(component->id) for the supported inputs/outputs, * describes the possibilities of how a component is linked into a * pipeline. */ u32 supported_inputs; + /** @supported_outputs: bitmask of supported output componenet ids */ u32 supported_outputs; /** @@ -134,7 +137,8 @@ struct komeda_component { struct komeda_component_output { /** @component: indicate which component the data comes from */ struct komeda_component *component; - /** @output_port: + /** + * @output_port: * the output port of the &komeda_component_output.component */ u8 output_port; @@ -150,11 +154,12 @@ struct komeda_component_output { struct komeda_component_state { /** @obj: tracking component_state by drm_atomic_state */ struct drm_private_state obj; + /** @component: backpointer to the component */ struct komeda_component *component; /** * @binding_user: - * currently bound user, the user can be crtc/plane/wb_conn, which is - * valid decided by @component and @inputs + * currently bound user, the user can be @crtc, @plane or @wb_conn, + * which is valid decided by @component and @inputs * * - Layer: its user always is plane. * - compiz/improc/timing_ctrlr: the user is crtc. @@ -162,20 +167,24 @@ struct komeda_component_state { * - scaler: plane when input is layer, wb_conn if input is compiz. */ union { + /** @crtc: backpointer for user crtc */ struct drm_crtc *crtc; + /** @plane: backpointer for user plane */ struct drm_plane *plane; + /** @wb_conn: backpointer for user wb_connector */ struct drm_connector *wb_conn; void *binding_user; }; + /** * @active_inputs: * * active_inputs is bitmask of @inputs index * - * - active_inputs = changed_active_inputs + unchanged_active_inputs - * - affected_inputs = old->active_inputs + new->active_inputs; + * - active_inputs = changed_active_inputs | unchanged_active_inputs + * - affected_inputs = old->active_inputs | new->active_inputs; * - disabling_inputs = affected_inputs ^ active_inputs; - * - changed_inputs = disabling_inputs + changed_active_inputs; + * - changed_inputs = disabling_inputs | changed_active_inputs; * * NOTE: * changed_inputs doesn't include all active_input but only @@ -183,7 +192,9 @@ struct komeda_component_state { * level for dirty update. */ u16 active_inputs; + /** @changed_active_inputs: bitmask of the changed @active_inputs */ u16 changed_active_inputs; + /** @affected_inputs: bitmask for affected @inputs */ u16 affected_inputs; /** * @inputs: @@ -278,6 +289,22 @@ struct komeda_timing_ctrlr_state { struct komeda_component_state base; }; +/* Why define A separated structure but not use plane_state directly ? + * 1. Komeda supports layer_split which means a plane_state can be split and + * handled by two layers, one layer only handle half of plane image. + * 2. Fix up the user properties according to HW's capabilities, like user + * set rotation to R180, but HW only supports REFLECT_X+Y. the rot here is + * after drm_rotation_simplify() + */ +struct komeda_data_flow_cfg { + struct komeda_component_output input; + u16 in_x, in_y, in_w, in_h; + u32 out_x, out_y, out_w, out_h; + u32 rot; + int blending_zorder; + u8 pixel_blend_mode, layer_alpha; +}; + /** struct komeda_pipeline_funcs */ struct komeda_pipeline_funcs { /* dump_register: Optional, dump registers to seq_file */ @@ -303,14 +330,23 @@ struct komeda_pipeline { int id; /** @avail_comps: available components mask of pipeline */ u32 avail_comps; + /** @n_layers: the number of layer on @layers */ int n_layers; + /** @layers: the pipeline layers */ struct komeda_layer *layers[KOMEDA_PIPELINE_MAX_LAYERS]; + /** @n_scalers: the number of scaler on @scalers */ int n_scalers; + /** @scalers: the pipeline scalers */ struct komeda_scaler *scalers[KOMEDA_PIPELINE_MAX_SCALERS]; + /** @compiz: compositor */ struct komeda_compiz *compiz; + /** @wb_layer: writeback layer */ struct komeda_layer *wb_layer; + /** @improc: post image processor */ struct komeda_improc *improc; + /** @ctrlr: timing controller */ struct komeda_timing_ctrlr *ctrlr; + /** @funcs: chip pipeline functions */ struct komeda_pipeline_funcs *funcs; /* private pipeline functions */ /** @of_node: pipeline dt node */ @@ -331,6 +367,7 @@ struct komeda_pipeline { struct komeda_pipeline_state { /** @obj: tracking pipeline_state by drm_atomic_state */ struct drm_private_state obj; + /** @pipe: backpointer to the pipeline */ struct komeda_pipeline *pipe; /** @crtc: currently bound crtc */ struct drm_crtc *crtc; @@ -382,4 +419,26 @@ komeda_component_add(struct komeda_pipeline *pipe, void komeda_component_destroy(struct komeda_dev *mdev, struct komeda_component *c); +struct komeda_plane_state; +struct komeda_crtc_state; +struct komeda_crtc; + +int komeda_build_layer_data_flow(struct komeda_layer *layer, + struct komeda_plane_state *kplane_st, + struct komeda_crtc_state *kcrtc_st, + struct komeda_data_flow_cfg *dflow); +int komeda_build_display_data_flow(struct komeda_crtc *kcrtc, + struct komeda_crtc_state *kcrtc_st); + +int komeda_release_unclaimed_resources(struct komeda_pipeline *pipe, + struct komeda_crtc_state *kcrtc_st); + +struct komeda_pipeline_state * +komeda_pipeline_get_old_state(struct komeda_pipeline *pipe, + struct drm_atomic_state *state); +void komeda_pipeline_disable(struct komeda_pipeline *pipe, + struct drm_atomic_state *old_state); +void komeda_pipeline_update(struct komeda_pipeline *pipe, + struct drm_atomic_state *old_state); + #endif /* _KOMEDA_PIPELINE_H_*/ diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_pipeline_state.c b/drivers/gpu/drm/arm/display/komeda/komeda_pipeline_state.c new file mode 100644 index 000000000000..36570d7dad61 --- /dev/null +++ b/drivers/gpu/drm/arm/display/komeda/komeda_pipeline_state.c @@ -0,0 +1,610 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * (C) COPYRIGHT 2018 ARM Limited. All rights reserved. + * Author: James.Qian.Wang <james.qian.wang@arm.com> + * + */ + +#include <drm/drm_print.h> +#include <linux/clk.h> +#include "komeda_dev.h" +#include "komeda_kms.h" +#include "komeda_pipeline.h" +#include "komeda_framebuffer.h" + +static inline bool is_switching_user(void *old, void *new) +{ + if (!old || !new) + return false; + + return old != new; +} + +static struct komeda_pipeline_state * +komeda_pipeline_get_state(struct komeda_pipeline *pipe, + struct drm_atomic_state *state) +{ + struct drm_private_state *priv_st; + + priv_st = drm_atomic_get_private_obj_state(state, &pipe->obj); + if (IS_ERR(priv_st)) + return ERR_CAST(priv_st); + + return priv_to_pipe_st(priv_st); +} + +struct komeda_pipeline_state * +komeda_pipeline_get_old_state(struct komeda_pipeline *pipe, + struct drm_atomic_state *state) +{ + struct drm_private_state *priv_st; + + priv_st = drm_atomic_get_old_private_obj_state(state, &pipe->obj); + if (priv_st) + return priv_to_pipe_st(priv_st); + return NULL; +} + +static struct komeda_pipeline_state * +komeda_pipeline_get_new_state(struct komeda_pipeline *pipe, + struct drm_atomic_state *state) +{ + struct drm_private_state *priv_st; + + priv_st = drm_atomic_get_new_private_obj_state(state, &pipe->obj); + if (priv_st) + return priv_to_pipe_st(priv_st); + return NULL; +} + +/* Assign pipeline for crtc */ +static struct komeda_pipeline_state * +komeda_pipeline_get_state_and_set_crtc(struct komeda_pipeline *pipe, + struct drm_atomic_state *state, + struct drm_crtc *crtc) +{ + struct komeda_pipeline_state *st; + + st = komeda_pipeline_get_state(pipe, state); + if (IS_ERR(st)) + return st; + + if (is_switching_user(crtc, st->crtc)) { + DRM_DEBUG_ATOMIC("CRTC%d required pipeline%d is busy.\n", + drm_crtc_index(crtc), pipe->id); + return ERR_PTR(-EBUSY); + } + + /* pipeline only can be disabled when the it is free or unused */ + if (!crtc && st->active_comps) { + DRM_DEBUG_ATOMIC("Disabling a busy pipeline:%d.\n", pipe->id); + return ERR_PTR(-EBUSY); + } + + st->crtc = crtc; + + if (crtc) { + struct komeda_crtc_state *kcrtc_st; + + kcrtc_st = to_kcrtc_st(drm_atomic_get_new_crtc_state(state, + crtc)); + + kcrtc_st->active_pipes |= BIT(pipe->id); + kcrtc_st->affected_pipes |= BIT(pipe->id); + } + return st; +} + +static struct komeda_component_state * +komeda_component_get_state(struct komeda_component *c, + struct drm_atomic_state *state) +{ + struct drm_private_state *priv_st; + + WARN_ON(!drm_modeset_is_locked(&c->pipeline->obj.lock)); + + priv_st = drm_atomic_get_private_obj_state(state, &c->obj); + if (IS_ERR(priv_st)) + return ERR_CAST(priv_st); + + return priv_to_comp_st(priv_st); +} + +static struct komeda_component_state * +komeda_component_get_old_state(struct komeda_component *c, + struct drm_atomic_state *state) +{ + struct drm_private_state *priv_st; + + priv_st = drm_atomic_get_old_private_obj_state(state, &c->obj); + if (priv_st) + return priv_to_comp_st(priv_st); + return NULL; +} + +/** + * komeda_component_get_state_and_set_user() + * + * @c: component to get state and set user + * @state: global atomic state + * @user: direct user, the binding user + * @crtc: the CRTC user, the big boss :) + * + * This function accepts two users: + * - The direct user: can be plane/crtc/wb_connector depends on component + * - The big boss (CRTC) + * CRTC is the big boss (the final user), because all component resources + * eventually will be assigned to CRTC, like the layer will be binding to + * kms_plane, but kms plane will be binding to a CRTC eventually. + * + * The big boss (CRTC) is for pipeline assignment, since &komeda_component isn't + * independent and can be assigned to CRTC freely, but belongs to a specific + * pipeline, only pipeline can be shared between crtc, and pipeline as a whole + * (include all the internal components) assigned to a specific CRTC. + * + * So when set a user to komeda_component, need first to check the status of + * component->pipeline to see if the pipeline is available on this specific + * CRTC. if the pipeline is busy (assigned to another CRTC), even the required + * component is free, the component still cannot be assigned to the direct user. + */ +static struct komeda_component_state * +komeda_component_get_state_and_set_user(struct komeda_component *c, + struct drm_atomic_state *state, + void *user, + struct drm_crtc *crtc) +{ + struct komeda_pipeline_state *pipe_st; + struct komeda_component_state *st; + + /* First check if the pipeline is available */ + pipe_st = komeda_pipeline_get_state_and_set_crtc(c->pipeline, + state, crtc); + if (IS_ERR(pipe_st)) + return ERR_CAST(pipe_st); + + st = komeda_component_get_state(c, state); + if (IS_ERR(st)) + return st; + + /* check if the component has been occupied */ + if (is_switching_user(user, st->binding_user)) { + DRM_DEBUG_ATOMIC("required %s is busy.\n", c->name); + return ERR_PTR(-EBUSY); + } + + st->binding_user = user; + /* mark the component as active if user is valid */ + if (st->binding_user) + pipe_st->active_comps |= BIT(c->id); + + return st; +} + +static void +komeda_component_add_input(struct komeda_component_state *state, + struct komeda_component_output *input, + int idx) +{ + struct komeda_component *c = state->component; + + WARN_ON((idx < 0 || idx >= c->max_active_inputs)); + + /* since the inputs[i] is only valid when it is active. So if a input[i] + * is a newly enabled input which switches from disable to enable, then + * the old inputs[i] is undefined (NOT zeroed), we can not rely on + * memcmp, but directly mark it changed + */ + if (!has_bit(idx, state->affected_inputs) || + memcmp(&state->inputs[idx], input, sizeof(*input))) { + memcpy(&state->inputs[idx], input, sizeof(*input)); + state->changed_active_inputs |= BIT(idx); + } + state->active_inputs |= BIT(idx); + state->affected_inputs |= BIT(idx); +} + +static int +komeda_component_check_input(struct komeda_component_state *state, + struct komeda_component_output *input, + int idx) +{ + struct komeda_component *c = state->component; + + if ((idx < 0) || (idx >= c->max_active_inputs)) { + DRM_DEBUG_ATOMIC("%s invalid input id: %d.\n", c->name, idx); + return -EINVAL; + } + + if (has_bit(idx, state->active_inputs)) { + DRM_DEBUG_ATOMIC("%s required input_id: %d has been occupied already.\n", + c->name, idx); + return -EINVAL; + } + + return 0; +} + +static void +komeda_component_set_output(struct komeda_component_output *output, + struct komeda_component *comp, + u8 output_port) +{ + output->component = comp; + output->output_port = output_port; +} + +static int +komeda_component_validate_private(struct komeda_component *c, + struct komeda_component_state *st) +{ + int err; + + if (!c->funcs->validate) + return 0; + + err = c->funcs->validate(c, st); + if (err) + DRM_DEBUG_ATOMIC("%s validate private failed.\n", c->name); + + return err; +} + +static int +komeda_layer_check_cfg(struct komeda_layer *layer, + struct komeda_plane_state *kplane_st, + struct komeda_data_flow_cfg *dflow) +{ + if (!in_range(&layer->hsize_in, dflow->in_w)) { + DRM_DEBUG_ATOMIC("src_w: %d is out of range.\n", dflow->in_w); + return -EINVAL; + } + + if (!in_range(&layer->vsize_in, dflow->in_h)) { + DRM_DEBUG_ATOMIC("src_h: %d is out of range.\n", dflow->in_h); + return -EINVAL; + } + + return 0; +} + +static int +komeda_layer_validate(struct komeda_layer *layer, + struct komeda_plane_state *kplane_st, + struct komeda_data_flow_cfg *dflow) +{ + struct drm_plane_state *plane_st = &kplane_st->base; + struct drm_framebuffer *fb = plane_st->fb; + struct komeda_fb *kfb = to_kfb(fb); + struct komeda_component_state *c_st; + struct komeda_layer_state *st; + int i, err; + + err = komeda_layer_check_cfg(layer, kplane_st, dflow); + if (err) + return err; + + c_st = komeda_component_get_state_and_set_user(&layer->base, + plane_st->state, plane_st->plane, plane_st->crtc); + if (IS_ERR(c_st)) + return PTR_ERR(c_st); + + st = to_layer_st(c_st); + + st->rot = dflow->rot; + st->hsize = kfb->aligned_w; + st->vsize = kfb->aligned_h; + + for (i = 0; i < fb->format->num_planes; i++) + st->addr[i] = komeda_fb_get_pixel_addr(kfb, dflow->in_x, + dflow->in_y, i); + + err = komeda_component_validate_private(&layer->base, c_st); + if (err) + return err; + + /* update the data flow for the next stage */ + komeda_component_set_output(&dflow->input, &layer->base, 0); + + return 0; +} + +static void pipeline_composition_size(struct komeda_crtc_state *kcrtc_st, + u16 *hsize, u16 *vsize) +{ + struct drm_display_mode *m = &kcrtc_st->base.adjusted_mode; + + if (hsize) + *hsize = m->hdisplay; + if (vsize) + *vsize = m->vdisplay; +} + +static int +komeda_compiz_set_input(struct komeda_compiz *compiz, + struct komeda_crtc_state *kcrtc_st, + struct komeda_data_flow_cfg *dflow) +{ + struct drm_atomic_state *drm_st = kcrtc_st->base.state; + struct komeda_component_state *c_st, *old_st; + struct komeda_compiz_input_cfg *cin; + u16 compiz_w, compiz_h; + int idx = dflow->blending_zorder; + + pipeline_composition_size(kcrtc_st, &compiz_w, &compiz_h); + /* check display rect */ + if ((dflow->out_x + dflow->out_w > compiz_w) || + (dflow->out_y + dflow->out_h > compiz_h) || + dflow->out_w == 0 || dflow->out_h == 0) { + DRM_DEBUG_ATOMIC("invalid disp rect [x=%d, y=%d, w=%d, h=%d]\n", + dflow->out_x, dflow->out_y, + dflow->out_w, dflow->out_h); + return -EINVAL; + } + + c_st = komeda_component_get_state_and_set_user(&compiz->base, drm_st, + kcrtc_st->base.crtc, kcrtc_st->base.crtc); + if (IS_ERR(c_st)) + return PTR_ERR(c_st); + + if (komeda_component_check_input(c_st, &dflow->input, idx)) + return -EINVAL; + + cin = &(to_compiz_st(c_st)->cins[idx]); + + cin->hsize = dflow->out_w; + cin->vsize = dflow->out_h; + cin->hoffset = dflow->out_x; + cin->voffset = dflow->out_y; + cin->pixel_blend_mode = dflow->pixel_blend_mode; + cin->layer_alpha = dflow->layer_alpha; + + old_st = komeda_component_get_old_state(&compiz->base, drm_st); + WARN_ON(!old_st); + + /* compare with old to check if this input has been changed */ + if (memcmp(&(to_compiz_st(old_st)->cins[idx]), cin, sizeof(*cin))) + c_st->changed_active_inputs |= BIT(idx); + + komeda_component_add_input(c_st, &dflow->input, idx); + + return 0; +} + +static int +komeda_compiz_validate(struct komeda_compiz *compiz, + struct komeda_crtc_state *state, + struct komeda_data_flow_cfg *dflow) +{ + struct komeda_component_state *c_st; + struct komeda_compiz_state *st; + + c_st = komeda_component_get_state_and_set_user(&compiz->base, + state->base.state, state->base.crtc, state->base.crtc); + if (IS_ERR(c_st)) + return PTR_ERR(c_st); + + st = to_compiz_st(c_st); + + pipeline_composition_size(state, &st->hsize, &st->vsize); + + komeda_component_set_output(&dflow->input, &compiz->base, 0); + + /* compiz output dflow will be fed to the next pipeline stage, prepare + * the data flow configuration for the next stage + */ + if (dflow) { + dflow->in_w = st->hsize; + dflow->in_h = st->vsize; + dflow->out_w = dflow->in_w; + dflow->out_h = dflow->in_h; + /* the output data of compiz doesn't have alpha, it only can be + * used as bottom layer when blend it with master layers + */ + dflow->pixel_blend_mode = DRM_MODE_BLEND_PIXEL_NONE; + dflow->layer_alpha = 0xFF; + dflow->blending_zorder = 0; + } + + return 0; +} + +static int +komeda_improc_validate(struct komeda_improc *improc, + struct komeda_crtc_state *kcrtc_st, + struct komeda_data_flow_cfg *dflow) +{ + struct drm_crtc *crtc = kcrtc_st->base.crtc; + struct komeda_component_state *c_st; + struct komeda_improc_state *st; + + c_st = komeda_component_get_state_and_set_user(&improc->base, + kcrtc_st->base.state, crtc, crtc); + if (IS_ERR(c_st)) + return PTR_ERR(c_st); + + st = to_improc_st(c_st); + + st->hsize = dflow->in_w; + st->vsize = dflow->in_h; + + komeda_component_add_input(&st->base, &dflow->input, 0); + komeda_component_set_output(&dflow->input, &improc->base, 0); + + return 0; +} + +static int +komeda_timing_ctrlr_validate(struct komeda_timing_ctrlr *ctrlr, + struct komeda_crtc_state *kcrtc_st, + struct komeda_data_flow_cfg *dflow) +{ + struct drm_crtc *crtc = kcrtc_st->base.crtc; + struct komeda_timing_ctrlr_state *st; + struct komeda_component_state *c_st; + + c_st = komeda_component_get_state_and_set_user(&ctrlr->base, + kcrtc_st->base.state, crtc, crtc); + if (IS_ERR(c_st)) + return PTR_ERR(c_st); + + st = to_ctrlr_st(c_st); + + komeda_component_add_input(&st->base, &dflow->input, 0); + komeda_component_set_output(&dflow->input, &ctrlr->base, 0); + + return 0; +} + +int komeda_build_layer_data_flow(struct komeda_layer *layer, + struct komeda_plane_state *kplane_st, + struct komeda_crtc_state *kcrtc_st, + struct komeda_data_flow_cfg *dflow) +{ + struct drm_plane *plane = kplane_st->base.plane; + struct komeda_pipeline *pipe = layer->base.pipeline; + int err; + + DRM_DEBUG_ATOMIC("%s handling [PLANE:%d:%s]: src[x/y:%d/%d, w/h:%d/%d] disp[x/y:%d/%d, w/h:%d/%d]", + layer->base.name, plane->base.id, plane->name, + dflow->in_x, dflow->in_y, dflow->in_w, dflow->in_h, + dflow->out_x, dflow->out_y, dflow->out_w, dflow->out_h); + + err = komeda_layer_validate(layer, kplane_st, dflow); + if (err) + return err; + + err = komeda_compiz_set_input(pipe->compiz, kcrtc_st, dflow); + + return err; +} + +/* build display output data flow, the data path is: + * compiz -> improc -> timing_ctrlr + */ +int komeda_build_display_data_flow(struct komeda_crtc *kcrtc, + struct komeda_crtc_state *kcrtc_st) +{ + struct komeda_pipeline *master = kcrtc->master; + struct komeda_data_flow_cfg m_dflow; /* master data flow */ + int err; + + memset(&m_dflow, 0, sizeof(m_dflow)); + + err = komeda_compiz_validate(master->compiz, kcrtc_st, &m_dflow); + if (err) + return err; + + err = komeda_improc_validate(master->improc, kcrtc_st, &m_dflow); + if (err) + return err; + + err = komeda_timing_ctrlr_validate(master->ctrlr, kcrtc_st, &m_dflow); + if (err) + return err; + + return 0; +} + +static void +komeda_pipeline_unbound_components(struct komeda_pipeline *pipe, + struct komeda_pipeline_state *new) +{ + struct drm_atomic_state *drm_st = new->obj.state; + struct komeda_pipeline_state *old = priv_to_pipe_st(pipe->obj.state); + struct komeda_component_state *c_st; + struct komeda_component *c; + u32 disabling_comps, id; + + WARN_ON(!old); + + disabling_comps = (~new->active_comps) & old->active_comps; + + /* unbound all disabling component */ + dp_for_each_set_bit(id, disabling_comps) { + c = komeda_pipeline_get_component(pipe, id); + c_st = komeda_component_get_state_and_set_user(c, + drm_st, NULL, new->crtc); + WARN_ON(IS_ERR(c_st)); + } +} + +/* release unclaimed pipeline resource */ +int komeda_release_unclaimed_resources(struct komeda_pipeline *pipe, + struct komeda_crtc_state *kcrtc_st) +{ + struct drm_atomic_state *drm_st = kcrtc_st->base.state; + struct komeda_pipeline_state *st; + + /* ignore the pipeline which is not affected */ + if (!pipe || !has_bit(pipe->id, kcrtc_st->affected_pipes)) + return 0; + + if (has_bit(pipe->id, kcrtc_st->active_pipes)) + st = komeda_pipeline_get_new_state(pipe, drm_st); + else + st = komeda_pipeline_get_state_and_set_crtc(pipe, drm_st, NULL); + + if (WARN_ON(IS_ERR_OR_NULL(st))) + return -EINVAL; + + komeda_pipeline_unbound_components(pipe, st); + + return 0; +} + +void komeda_pipeline_disable(struct komeda_pipeline *pipe, + struct drm_atomic_state *old_state) +{ + struct komeda_pipeline_state *old; + struct komeda_component *c; + struct komeda_component_state *c_st; + u32 id, disabling_comps = 0; + + old = komeda_pipeline_get_old_state(pipe, old_state); + + disabling_comps = old->active_comps; + DRM_DEBUG_ATOMIC("PIPE%d: disabling_comps: 0x%x.\n", + pipe->id, disabling_comps); + + dp_for_each_set_bit(id, disabling_comps) { + c = komeda_pipeline_get_component(pipe, id); + c_st = priv_to_comp_st(c->obj.state); + + /* + * If we disabled a component then all active_inputs should be + * put in the list of changed_active_inputs, so they get + * re-enabled. + * This usually happens during a modeset when the pipeline is + * first disabled and then the actual state gets committed + * again. + */ + c_st->changed_active_inputs |= c_st->active_inputs; + + c->funcs->disable(c); + } +} + +void komeda_pipeline_update(struct komeda_pipeline *pipe, + struct drm_atomic_state *old_state) +{ + struct komeda_pipeline_state *new = priv_to_pipe_st(pipe->obj.state); + struct komeda_pipeline_state *old; + struct komeda_component *c; + u32 id, changed_comps = 0; + + old = komeda_pipeline_get_old_state(pipe, old_state); + + changed_comps = new->active_comps | old->active_comps; + + DRM_DEBUG_ATOMIC("PIPE%d: active_comps: 0x%x, changed: 0x%x.\n", + pipe->id, new->active_comps, changed_comps); + + dp_for_each_set_bit(id, changed_comps) { + c = komeda_pipeline_get_component(pipe, id); + + if (new->active_comps & BIT(c->id)) + c->funcs->update(c, priv_to_comp_st(c->obj.state)); + else + c->funcs->disable(c); + } +} diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_plane.c b/drivers/gpu/drm/arm/display/komeda/komeda_plane.c index 0a4953a9a909..07ed0cc1bc44 100644 --- a/drivers/gpu/drm/arm/display/komeda/komeda_plane.c +++ b/drivers/gpu/drm/arm/display/komeda/komeda_plane.c @@ -7,10 +7,96 @@ #include <drm/drm_atomic.h> #include <drm/drm_atomic_helper.h> #include <drm/drm_plane_helper.h> +#include <drm/drm_print.h> #include "komeda_dev.h" #include "komeda_kms.h" +static int +komeda_plane_init_data_flow(struct drm_plane_state *st, + struct komeda_data_flow_cfg *dflow) +{ + struct drm_framebuffer *fb = st->fb; + + memset(dflow, 0, sizeof(*dflow)); + + dflow->blending_zorder = st->zpos; + + /* if format doesn't have alpha, fix blend mode to PIXEL_NONE */ + dflow->pixel_blend_mode = fb->format->has_alpha ? + st->pixel_blend_mode : DRM_MODE_BLEND_PIXEL_NONE; + dflow->layer_alpha = st->alpha >> 8; + + dflow->out_x = st->crtc_x; + dflow->out_y = st->crtc_y; + dflow->out_w = st->crtc_w; + dflow->out_h = st->crtc_h; + + dflow->in_x = st->src_x >> 16; + dflow->in_y = st->src_y >> 16; + dflow->in_w = st->src_w >> 16; + dflow->in_h = st->src_h >> 16; + + return 0; +} + +/** + * komeda_plane_atomic_check - build input data flow + * @plane: DRM plane + * @state: the plane state object + * + * RETURNS: + * Zero for success or -errno + */ +static int +komeda_plane_atomic_check(struct drm_plane *plane, + struct drm_plane_state *state) +{ + struct komeda_plane *kplane = to_kplane(plane); + struct komeda_plane_state *kplane_st = to_kplane_st(state); + struct komeda_layer *layer = kplane->layer; + struct drm_crtc_state *crtc_st; + struct komeda_crtc *kcrtc; + struct komeda_crtc_state *kcrtc_st; + struct komeda_data_flow_cfg dflow; + int err; + + if (!state->crtc || !state->fb) + return 0; + + crtc_st = drm_atomic_get_crtc_state(state->state, state->crtc); + if (!crtc_st->enable) { + DRM_DEBUG_ATOMIC("Cannot update plane on a disabled CRTC.\n"); + return -EINVAL; + } + + /* crtc is inactive, skip the resource assignment */ + if (!crtc_st->active) + return 0; + + kcrtc = to_kcrtc(state->crtc); + kcrtc_st = to_kcrtc_st(crtc_st); + + err = komeda_plane_init_data_flow(state, &dflow); + if (err) + return err; + + err = komeda_build_layer_data_flow(layer, kplane_st, kcrtc_st, &dflow); + + return err; +} + +/* plane doesn't represent a real HW, so there is no HW update for plane. + * komeda handles all the HW update in crtc->atomic_flush + */ +static void +komeda_plane_atomic_update(struct drm_plane *plane, + struct drm_plane_state *old_state) +{ +} + static const struct drm_plane_helper_funcs komeda_plane_helper_funcs = { + .atomic_check = komeda_plane_atomic_check, + .atomic_update = komeda_plane_atomic_update, }; static void komeda_plane_destroy(struct drm_plane *plane) @@ -20,7 +106,60 @@ static void komeda_plane_destroy(struct drm_plane *plane) kfree(to_kplane(plane)); } +static void komeda_plane_reset(struct drm_plane *plane) +{ + struct komeda_plane_state *state; + struct komeda_plane *kplane = to_kplane(plane); + + if (plane->state) + __drm_atomic_helper_plane_destroy_state(plane->state); + + kfree(plane->state); + plane->state = NULL; + + state = kzalloc(sizeof(*state), GFP_KERNEL); + if (state) { + state->base.rotation = DRM_MODE_ROTATE_0; + state->base.pixel_blend_mode = DRM_MODE_BLEND_PREMULTI; + state->base.alpha = DRM_BLEND_ALPHA_OPAQUE; + state->base.zpos = kplane->layer->base.id; + plane->state = &state->base; + plane->state->plane = plane; + } +} + +static struct drm_plane_state * +komeda_plane_atomic_duplicate_state(struct drm_plane *plane) +{ + struct komeda_plane_state *new; + + if (WARN_ON(!plane->state)) + return NULL; + + new = kzalloc(sizeof(*new), GFP_KERNEL); + if (!new) + return NULL; + + __drm_atomic_helper_plane_duplicate_state(plane, &new->base); + + return &new->base; +} + +static void +komeda_plane_atomic_destroy_state(struct drm_plane *plane, + struct drm_plane_state *state) +{ + __drm_atomic_helper_plane_destroy_state(state); + kfree(to_kplane_st(state)); +} + static const struct drm_plane_funcs komeda_plane_funcs = { + .update_plane = drm_atomic_helper_update_plane, + .disable_plane = drm_atomic_helper_disable_plane, + .destroy = komeda_plane_destroy, + .reset = komeda_plane_reset, + .atomic_duplicate_state = komeda_plane_atomic_duplicate_state, + .atomic_destroy_state = komeda_plane_atomic_destroy_state, }; /* for komeda, which is pipeline can be share between crtcs */ diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_private_obj.c b/drivers/gpu/drm/arm/display/komeda/komeda_private_obj.c index f1c9e3fefa86..a54878cbd6e4 100644 --- a/drivers/gpu/drm/arm/display/komeda/komeda_private_obj.c +++ b/drivers/gpu/drm/arm/display/komeda/komeda_private_obj.c @@ -7,6 +7,188 @@ #include "komeda_dev.h" #include "komeda_kms.h" +static void +komeda_component_state_reset(struct komeda_component_state *st) +{ + st->binding_user = NULL; + st->affected_inputs = st->active_inputs; + st->active_inputs = 0; + st->changed_active_inputs = 0; +} + +static struct drm_private_state * +komeda_layer_atomic_duplicate_state(struct drm_private_obj *obj) +{ + struct komeda_layer_state *st; + + st = kmemdup(obj->state, sizeof(*st), GFP_KERNEL); + if (!st) + return NULL; + + komeda_component_state_reset(&st->base); + __drm_atomic_helper_private_obj_duplicate_state(obj, &st->base.obj); + + return &st->base.obj; +} + +static void +komeda_layer_atomic_destroy_state(struct drm_private_obj *obj, + struct drm_private_state *state) +{ + struct komeda_layer_state *st = to_layer_st(priv_to_comp_st(state)); + + kfree(st); +} + +static const struct drm_private_state_funcs komeda_layer_obj_funcs = { + .atomic_duplicate_state = komeda_layer_atomic_duplicate_state, + .atomic_destroy_state = komeda_layer_atomic_destroy_state, +}; + +static int komeda_layer_obj_add(struct komeda_kms_dev *kms, + struct komeda_layer *layer) +{ + struct komeda_layer_state *st; + + st = kzalloc(sizeof(*st), GFP_KERNEL); + if (!st) + return -ENOMEM; + + st->base.component = &layer->base; + drm_atomic_private_obj_init(&kms->base, &layer->base.obj, &st->base.obj, + &komeda_layer_obj_funcs); + return 0; +} + +static struct drm_private_state * +komeda_compiz_atomic_duplicate_state(struct drm_private_obj *obj) +{ + struct komeda_compiz_state *st; + + st = kmemdup(obj->state, sizeof(*st), GFP_KERNEL); + if (!st) + return NULL; + + komeda_component_state_reset(&st->base); + __drm_atomic_helper_private_obj_duplicate_state(obj, &st->base.obj); + + return &st->base.obj; +} + +static void +komeda_compiz_atomic_destroy_state(struct drm_private_obj *obj, + struct drm_private_state *state) +{ + kfree(to_compiz_st(priv_to_comp_st(state))); +} + +static const struct drm_private_state_funcs komeda_compiz_obj_funcs = { + .atomic_duplicate_state = komeda_compiz_atomic_duplicate_state, + .atomic_destroy_state = komeda_compiz_atomic_destroy_state, +}; + +static int komeda_compiz_obj_add(struct komeda_kms_dev *kms, + struct komeda_compiz *compiz) +{ + struct komeda_compiz_state *st; + + st = kzalloc(sizeof(*st), GFP_KERNEL); + if (!st) + return -ENOMEM; + + st->base.component = &compiz->base; + drm_atomic_private_obj_init(&kms->base, &compiz->base.obj, &st->base.obj, + &komeda_compiz_obj_funcs); + + return 0; +} + +static struct drm_private_state * +komeda_improc_atomic_duplicate_state(struct drm_private_obj *obj) +{ + struct komeda_improc_state *st; + + st = kmemdup(obj->state, sizeof(*st), GFP_KERNEL); + if (!st) + return NULL; + + komeda_component_state_reset(&st->base); + __drm_atomic_helper_private_obj_duplicate_state(obj, &st->base.obj); + + return &st->base.obj; +} + +static void +komeda_improc_atomic_destroy_state(struct drm_private_obj *obj, + struct drm_private_state *state) +{ + kfree(to_improc_st(priv_to_comp_st(state))); +} + +static const struct drm_private_state_funcs komeda_improc_obj_funcs = { + .atomic_duplicate_state = komeda_improc_atomic_duplicate_state, + .atomic_destroy_state = komeda_improc_atomic_destroy_state, +}; + +static int komeda_improc_obj_add(struct komeda_kms_dev *kms, + struct komeda_improc *improc) +{ + struct komeda_improc_state *st; + + st = kzalloc(sizeof(*st), GFP_KERNEL); + if (!st) + return -ENOMEM; + + st->base.component = &improc->base; + drm_atomic_private_obj_init(&kms->base, &improc->base.obj, &st->base.obj, + &komeda_improc_obj_funcs); + + return 0; +} + +static struct drm_private_state * +komeda_timing_ctrlr_atomic_duplicate_state(struct drm_private_obj *obj) +{ + struct komeda_timing_ctrlr_state *st; + + st = kmemdup(obj->state, sizeof(*st), GFP_KERNEL); + if (!st) + return NULL; + + komeda_component_state_reset(&st->base); + __drm_atomic_helper_private_obj_duplicate_state(obj, &st->base.obj); + + return &st->base.obj; +} + +static void +komeda_timing_ctrlr_atomic_destroy_state(struct drm_private_obj *obj, + struct drm_private_state *state) +{ + kfree(to_ctrlr_st(priv_to_comp_st(state))); +} + +static const struct drm_private_state_funcs komeda_timing_ctrlr_obj_funcs = { + .atomic_duplicate_state = komeda_timing_ctrlr_atomic_duplicate_state, + .atomic_destroy_state = komeda_timing_ctrlr_atomic_destroy_state, +}; + +static int komeda_timing_ctrlr_obj_add(struct komeda_kms_dev *kms, + struct komeda_timing_ctrlr *ctrlr) +{ + struct komeda_compiz_state *st; + + st = kzalloc(sizeof(*st), GFP_KERNEL); + if (!st) + return -ENOMEM; + + st->base.component = &ctrlr->base; + drm_atomic_private_obj_init(&kms->base, &ctrlr->base.obj, &st->base.obj, + &komeda_timing_ctrlr_obj_funcs); + + return 0; +} + static struct drm_private_state * komeda_pipeline_atomic_duplicate_state(struct drm_private_obj *obj) { @@ -55,7 +237,7 @@ int komeda_kms_add_private_objs(struct komeda_kms_dev *kms, struct komeda_dev *mdev) { struct komeda_pipeline *pipe; - int i, err; + int i, j, err; for (i = 0; i < mdev->n_pipelines; i++) { pipe = mdev->pipelines[i]; @@ -64,25 +246,33 @@ int komeda_kms_add_private_objs(struct komeda_kms_dev *kms, if (err) return err; - /* Add component */ + for (j = 0; j < pipe->n_layers; j++) { + err = komeda_layer_obj_add(kms, pipe->layers[j]); + if (err) + return err; + } + + err = komeda_compiz_obj_add(kms, pipe->compiz); + if (err) + return err; + + err = komeda_improc_obj_add(kms, pipe->improc); + if (err) + return err; + + err = komeda_timing_ctrlr_obj_add(kms, pipe->ctrlr); + if (err) + return err; } return 0; } -void komeda_kms_cleanup_private_objs(struct komeda_dev *mdev) +void komeda_kms_cleanup_private_objs(struct komeda_kms_dev *kms) { - struct komeda_pipeline *pipe; - struct komeda_component *c; - int i, id; + struct drm_mode_config *config = &kms->base.mode_config; + struct drm_private_obj *obj, *next; - for (i = 0; i < mdev->n_pipelines; i++) { - pipe = mdev->pipelines[i]; - dp_for_each_set_bit(id, pipe->avail_comps) { - c = komeda_pipeline_get_component(pipe, id); - - drm_atomic_private_obj_fini(&c->obj); - } - drm_atomic_private_obj_fini(&pipe->obj); - } + list_for_each_entry_safe(obj, next, &config->privobj_list, head) + drm_atomic_private_obj_fini(obj); } diff --git a/drivers/gpu/drm/drm_atomic.c b/drivers/gpu/drm/drm_atomic.c index 5eb40130fafb..f4924cb7f495 100644 --- a/drivers/gpu/drm/drm_atomic.c +++ b/drivers/gpu/drm/drm_atomic.c @@ -798,6 +798,50 @@ drm_atomic_get_private_obj_state(struct drm_atomic_state *state, EXPORT_SYMBOL(drm_atomic_get_private_obj_state); /** + * drm_atomic_get_old_private_obj_state + * @state: global atomic state object + * @obj: private_obj to grab + * + * This function returns the old private object state for the given private_obj, + * or NULL if the private_obj is not part of the global atomic state. + */ +struct drm_private_state * +drm_atomic_get_old_private_obj_state(struct drm_atomic_state *state, + struct drm_private_obj *obj) +{ + int i; + + for (i = 0; i < state->num_private_objs; i++) + if (obj == state->private_objs[i].ptr) + return state->private_objs[i].old_state; + + return NULL; +} +EXPORT_SYMBOL(drm_atomic_get_old_private_obj_state); + +/** + * drm_atomic_get_new_private_obj_state + * @state: global atomic state object + * @obj: private_obj to grab + * + * This function returns the new private object state for the given private_obj, + * or NULL if the private_obj is not part of the global atomic state. + */ +struct drm_private_state * +drm_atomic_get_new_private_obj_state(struct drm_atomic_state *state, + struct drm_private_obj *obj) +{ + int i; + + for (i = 0; i < state->num_private_objs; i++) + if (obj == state->private_objs[i].ptr) + return state->private_objs[i].new_state; + + return NULL; +} +EXPORT_SYMBOL(drm_atomic_get_new_private_obj_state); + +/** * drm_atomic_get_connector_state - get connector state * @state: global atomic state object * @connector: connector to get state object for @@ -1236,4 +1280,3 @@ int drm_atomic_debugfs_init(struct drm_minor *minor) minor->debugfs_root, minor); } #endif - diff --git a/drivers/gpu/drm/drm_connector.c b/drivers/gpu/drm/drm_connector.c index 2355124849db..b34c3d38bf15 100644 --- a/drivers/gpu/drm/drm_connector.c +++ b/drivers/gpu/drm/drm_connector.c @@ -1416,12 +1416,6 @@ EXPORT_SYMBOL(drm_mode_create_scaling_mode_property); * * The driver may place further restrictions within these minimum * and maximum bounds. - * - * The semantics for the vertical blank timestamp differ when - * variable refresh rate is active. The vertical blank timestamp - * is defined to be an estimate using the current mode's fixed - * refresh rate timings. The semantics for the page-flip event - * timestamp remain the same. */ /** diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c index 2c22ea446075..649cfd8b4200 100644 --- a/drivers/gpu/drm/drm_edid.c +++ b/drivers/gpu/drm/drm_edid.c @@ -162,6 +162,25 @@ static const struct edid_quirk { /* Rotel RSX-1058 forwards sink's EDID but only does HDMI 1.1*/ { "ETR", 13896, EDID_QUIRK_FORCE_8BPC }, + /* Valve Index Headset */ + { "VLV", 0x91a8, EDID_QUIRK_NON_DESKTOP }, + { "VLV", 0x91b0, EDID_QUIRK_NON_DESKTOP }, + { "VLV", 0x91b1, EDID_QUIRK_NON_DESKTOP }, + { "VLV", 0x91b2, EDID_QUIRK_NON_DESKTOP }, + { "VLV", 0x91b3, EDID_QUIRK_NON_DESKTOP }, + { "VLV", 0x91b4, EDID_QUIRK_NON_DESKTOP }, + { "VLV", 0x91b5, EDID_QUIRK_NON_DESKTOP }, + { "VLV", 0x91b6, EDID_QUIRK_NON_DESKTOP }, + { "VLV", 0x91b7, EDID_QUIRK_NON_DESKTOP }, + { "VLV", 0x91b8, EDID_QUIRK_NON_DESKTOP }, + { "VLV", 0x91b9, EDID_QUIRK_NON_DESKTOP }, + { "VLV", 0x91ba, EDID_QUIRK_NON_DESKTOP }, + { "VLV", 0x91bb, EDID_QUIRK_NON_DESKTOP }, + { "VLV", 0x91bc, EDID_QUIRK_NON_DESKTOP }, + { "VLV", 0x91bd, EDID_QUIRK_NON_DESKTOP }, + { "VLV", 0x91be, EDID_QUIRK_NON_DESKTOP }, + { "VLV", 0x91bf, EDID_QUIRK_NON_DESKTOP }, + /* HTC Vive and Vive Pro VR Headsets */ { "HVR", 0xaa01, EDID_QUIRK_NON_DESKTOP }, { "HVR", 0xaa02, EDID_QUIRK_NON_DESKTOP }, diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c index fae4676707b6..50de138c89e0 100644 --- a/drivers/gpu/drm/drm_gem.c +++ b/drivers/gpu/drm/drm_gem.c @@ -1372,8 +1372,8 @@ EXPORT_SYMBOL(drm_gem_unlock_reservations); * drm_gem_fence_array_add - Adds the fence to an array of fences to be * waited on, deduplicating fences from the same context. * - * @fence_array array of dma_fence * for the job to block on. - * @fence the dma_fence to add to the list of dependencies. + * @fence_array: array of dma_fence * for the job to block on. + * @fence: the dma_fence to add to the list of dependencies. * * Returns: * 0 on success, or an error on failing to expand the array. @@ -1423,9 +1423,9 @@ EXPORT_SYMBOL(drm_gem_fence_array_add); * GEM objects used in the job but before updating the reservations with your * own fences. * - * @fence_array array of dma_fence * for the job to block on. - * @obj the gem object to add new dependencies from. - * @write whether the job might write the object (so we need to depend on + * @fence_array: array of dma_fence * for the job to block on. + * @obj: the gem object to add new dependencies from. + * @write: whether the job might write the object (so we need to depend on * shared fences in the reservation object). */ int drm_gem_fence_array_add_implicit(struct xarray *fence_array, diff --git a/drivers/gpu/drm/drm_gem_cma_helper.c b/drivers/gpu/drm/drm_gem_cma_helper.c index cc26625b4b33..e01ceed09e67 100644 --- a/drivers/gpu/drm/drm_gem_cma_helper.c +++ b/drivers/gpu/drm/drm_gem_cma_helper.c @@ -186,13 +186,13 @@ void drm_gem_cma_free_object(struct drm_gem_object *gem_obj) cma_obj = to_drm_gem_cma_obj(gem_obj); - if (cma_obj->vaddr) { - dma_free_wc(gem_obj->dev->dev, cma_obj->base.size, - cma_obj->vaddr, cma_obj->paddr); - } else if (gem_obj->import_attach) { + if (gem_obj->import_attach) { if (cma_obj->vaddr) dma_buf_vunmap(gem_obj->import_attach->dmabuf, cma_obj->vaddr); drm_prime_gem_destroy(gem_obj, cma_obj->sgt); + } else if (cma_obj->vaddr) { + dma_free_wc(gem_obj->dev->dev, cma_obj->base.size, + cma_obj->vaddr, cma_obj->paddr); } drm_gem_object_release(gem_obj); diff --git a/drivers/gpu/drm/drm_ioctl.c b/drivers/gpu/drm/drm_ioctl.c index 5878145077d0..2263e3ddd822 100644 --- a/drivers/gpu/drm/drm_ioctl.c +++ b/drivers/gpu/drm/drm_ioctl.c @@ -245,6 +245,9 @@ static int drm_getcap(struct drm_device *dev, void *data, struct drm_file *file_ case DRM_CAP_SYNCOBJ: req->value = drm_core_check_feature(dev, DRIVER_SYNCOBJ); return 0; + case DRM_CAP_SYNCOBJ_TIMELINE: + req->value = drm_core_check_feature(dev, DRIVER_SYNCOBJ_TIMELINE); + return 0; } /* Other caps only work with KMS drivers */ diff --git a/drivers/gpu/drm/drm_syncobj.c b/drivers/gpu/drm/drm_syncobj.c index f3ceeb504e6c..361a01a08c18 100644 --- a/drivers/gpu/drm/drm_syncobj.c +++ b/drivers/gpu/drm/drm_syncobj.c @@ -740,8 +740,8 @@ drm_syncobj_transfer_ioctl(struct drm_device *dev, void *data, struct drm_syncobj_transfer *args = data; int ret; - if (!drm_core_check_feature(dev, DRIVER_SYNCOBJ)) - return -ENODEV; + if (!drm_core_check_feature(dev, DRIVER_SYNCOBJ_TIMELINE)) + return -EOPNOTSUPP; if (args->pad) return -EINVAL; @@ -1091,8 +1091,8 @@ drm_syncobj_timeline_wait_ioctl(struct drm_device *dev, void *data, struct drm_syncobj **syncobjs; int ret = 0; - if (!drm_core_check_feature(dev, DRIVER_SYNCOBJ)) - return -ENODEV; + if (!drm_core_check_feature(dev, DRIVER_SYNCOBJ_TIMELINE)) + return -EOPNOTSUPP; if (args->flags & ~(DRM_SYNCOBJ_WAIT_FLAGS_WAIT_ALL | DRM_SYNCOBJ_WAIT_FLAGS_WAIT_FOR_SUBMIT | @@ -1195,7 +1195,7 @@ drm_syncobj_timeline_signal_ioctl(struct drm_device *dev, void *data, uint32_t i, j; int ret; - if (!drm_core_check_feature(dev, DRIVER_SYNCOBJ)) + if (!drm_core_check_feature(dev, DRIVER_SYNCOBJ_TIMELINE)) return -EOPNOTSUPP; if (args->pad != 0) @@ -1266,8 +1266,8 @@ int drm_syncobj_query_ioctl(struct drm_device *dev, void *data, uint32_t i; int ret; - if (!drm_core_check_feature(dev, DRIVER_SYNCOBJ)) - return -ENODEV; + if (!drm_core_check_feature(dev, DRIVER_SYNCOBJ_TIMELINE)) + return -EOPNOTSUPP; if (args->pad != 0) return -EINVAL; diff --git a/drivers/gpu/drm/etnaviv/etnaviv_drv.c b/drivers/gpu/drm/etnaviv/etnaviv_drv.c index 9f42f7538236..7eb7cf9c3fa8 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_drv.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_drv.c @@ -514,6 +514,9 @@ static int etnaviv_bind(struct device *dev) } drm->dev_private = priv; + dev->dma_parms = &priv->dma_parms; + dma_set_max_seg_size(dev, SZ_2G); + mutex_init(&priv->gem_lock); INIT_LIST_HEAD(&priv->gem_list); priv->num_gpus = 0; @@ -551,6 +554,8 @@ static void etnaviv_unbind(struct device *dev) component_unbind_all(dev, drm); + dev->dma_parms = NULL; + drm->dev_private = NULL; kfree(priv); diff --git a/drivers/gpu/drm/etnaviv/etnaviv_drv.h b/drivers/gpu/drm/etnaviv/etnaviv_drv.h index 6044ace6bb3e..8798423705e1 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_drv.h +++ b/drivers/gpu/drm/etnaviv/etnaviv_drv.h @@ -42,6 +42,7 @@ struct etnaviv_file_private { struct etnaviv_drm_private { int num_gpus; + struct device_dma_parameters dma_parms; struct etnaviv_gpu *gpu[ETNA_MAX_PIPES]; /* list of GEM objects: */ diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem.c b/drivers/gpu/drm/etnaviv/etnaviv_gem.c index c60752ef7324..e8778ebb72e6 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_gem.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_gem.c @@ -622,24 +622,18 @@ int etnaviv_gem_new_handle(struct drm_device *dev, struct drm_file *file, lockdep_set_class(&to_etnaviv_bo(obj)->lock, &etnaviv_shm_lock_class); ret = drm_gem_object_init(dev, obj, size); - if (ret == 0) { - struct address_space *mapping; - - /* - * Our buffers are kept pinned, so allocating them - * from the MOVABLE zone is a really bad idea, and - * conflicts with CMA. See comments above new_inode() - * why this is required _and_ expected if you're - * going to pin these pages. - */ - mapping = obj->filp->f_mapping; - mapping_set_gfp_mask(mapping, GFP_HIGHUSER | - __GFP_RETRY_MAYFAIL | __GFP_NOWARN); - } - if (ret) goto fail; + /* + * Our buffers are kept pinned, so allocating them from the MOVABLE + * zone is a really bad idea, and conflicts with CMA. See comments + * 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); + etnaviv_gem_obj_add(dev, obj); ret = drm_gem_handle_create(file, obj, handle); diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gpu.c b/drivers/gpu/drm/etnaviv/etnaviv_gpu.c index 6904535475de..72d01e873160 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_gpu.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_gpu.c @@ -365,6 +365,7 @@ static void etnaviv_hw_identify(struct etnaviv_gpu *gpu) dev_info(gpu->dev, "model: GC%x, revision: %x\n", gpu->identity.model, gpu->identity.revision); + gpu->idle_mask = ~VIVS_HI_IDLE_STATE_AXI_LP; /* * If there is a match in the HWDB, we aren't interested in the * remaining register values, as they might be wrong. @@ -412,7 +413,7 @@ static void etnaviv_hw_identify(struct etnaviv_gpu *gpu) } /* GC600 idle register reports zero bits where modules aren't present */ - if (gpu->identity.model == chipModel_GC600) { + if (gpu->identity.model == chipModel_GC600) gpu->idle_mask = VIVS_HI_IDLE_STATE_TX | VIVS_HI_IDLE_STATE_RA | VIVS_HI_IDLE_STATE_SE | @@ -421,9 +422,6 @@ static void etnaviv_hw_identify(struct etnaviv_gpu *gpu) VIVS_HI_IDLE_STATE_PE | VIVS_HI_IDLE_STATE_DE | VIVS_HI_IDLE_STATE_FE; - } else { - gpu->idle_mask = ~VIVS_HI_IDLE_STATE_AXI_LP; - } etnaviv_hw_specs(gpu); } diff --git a/drivers/gpu/drm/i915/gvt/Makefile b/drivers/gpu/drm/i915/gvt/Makefile index 271fb46d4dd0..ea8324abc784 100644 --- a/drivers/gpu/drm/i915/gvt/Makefile +++ b/drivers/gpu/drm/i915/gvt/Makefile @@ -5,5 +5,5 @@ GVT_SOURCE := gvt.o aperture_gm.o handlers.o vgpu.o trace_points.o firmware.o \ execlist.o scheduler.o sched_policy.o mmio_context.o cmd_parser.o debugfs.o \ fb_decoder.o dmabuf.o page_track.o -ccflags-y += -I$(src) -I$(src)/$(GVT_DIR) +ccflags-y += -I $(srctree)/$(src) -I $(srctree)/$(src)/$(GVT_DIR)/ i915-y += $(addprefix $(GVT_DIR)/, $(GVT_SOURCE)) diff --git a/drivers/gpu/drm/i915/gvt/debugfs.c b/drivers/gpu/drm/i915/gvt/debugfs.c index 2ec89bcb59f1..8a9606f91e68 100644 --- a/drivers/gpu/drm/i915/gvt/debugfs.c +++ b/drivers/gpu/drm/i915/gvt/debugfs.c @@ -196,9 +196,9 @@ DEFINE_SIMPLE_ATTRIBUTE(vgpu_scan_nonprivbb_fops, int intel_gvt_debugfs_add_vgpu(struct intel_vgpu *vgpu) { struct dentry *ent; - char name[10] = ""; + char name[16] = ""; - sprintf(name, "vgpu%d", vgpu->id); + snprintf(name, 16, "vgpu%d", vgpu->id); vgpu->debugfs = debugfs_create_dir(name, vgpu->gvt->debugfs_root); if (!vgpu->debugfs) return -ENOMEM; diff --git a/drivers/gpu/drm/i915/gvt/dmabuf.c b/drivers/gpu/drm/i915/gvt/dmabuf.c index 4e1e425189ba..41c8ebc60c63 100644 --- a/drivers/gpu/drm/i915/gvt/dmabuf.c +++ b/drivers/gpu/drm/i915/gvt/dmabuf.c @@ -45,6 +45,7 @@ static int vgpu_gem_get_pages( int i, ret; gen8_pte_t __iomem *gtt_entries; struct intel_vgpu_fb_info *fb_info; + u32 page_num; fb_info = (struct intel_vgpu_fb_info *)obj->gvt_info; if (WARN_ON(!fb_info)) @@ -54,14 +55,15 @@ static int vgpu_gem_get_pages( if (unlikely(!st)) return -ENOMEM; - ret = sg_alloc_table(st, fb_info->size, GFP_KERNEL); + page_num = obj->base.size >> PAGE_SHIFT; + ret = sg_alloc_table(st, page_num, GFP_KERNEL); if (ret) { kfree(st); return ret; } gtt_entries = (gen8_pte_t __iomem *)dev_priv->ggtt.gsm + (fb_info->start >> PAGE_SHIFT); - for_each_sg(st->sgl, sg, fb_info->size, i) { + for_each_sg(st->sgl, sg, page_num, i) { sg->offset = 0; sg->length = PAGE_SIZE; sg_dma_address(sg) = @@ -158,7 +160,7 @@ static struct drm_i915_gem_object *vgpu_create_gem(struct drm_device *dev, return NULL; drm_gem_private_object_init(dev, &obj->base, - info->size << PAGE_SHIFT); + roundup(info->size, PAGE_SIZE)); i915_gem_object_init(obj, &intel_vgpu_gem_ops); obj->read_domains = I915_GEM_DOMAIN_GTT; @@ -206,11 +208,12 @@ static int vgpu_get_plane_info(struct drm_device *dev, struct intel_vgpu_fb_info *info, int plane_id) { - struct drm_i915_private *dev_priv = to_i915(dev); struct intel_vgpu_primary_plane_format p; struct intel_vgpu_cursor_plane_format c; int ret, tile_height = 1; + memset(info, 0, sizeof(*info)); + if (plane_id == DRM_PLANE_TYPE_PRIMARY) { ret = intel_vgpu_decode_primary_plane(vgpu, &p); if (ret) @@ -267,8 +270,7 @@ static int vgpu_get_plane_info(struct drm_device *dev, return -EINVAL; } - info->size = (info->stride * roundup(info->height, tile_height) - + PAGE_SIZE - 1) >> PAGE_SHIFT; + info->size = info->stride * roundup(info->height, tile_height); if (info->size == 0) { gvt_vgpu_err("fb size is zero\n"); return -EINVAL; @@ -278,11 +280,6 @@ static int vgpu_get_plane_info(struct drm_device *dev, gvt_vgpu_err("Not aligned fb address:0x%llx\n", info->start); return -EFAULT; } - if (((info->start >> PAGE_SHIFT) + info->size) > - ggtt_total_entries(&dev_priv->ggtt)) { - gvt_vgpu_err("Invalid GTT offset or size\n"); - return -EFAULT; - } if (!intel_gvt_ggtt_validate_range(vgpu, info->start, info->size)) { gvt_vgpu_err("invalid gma addr\n"); diff --git a/drivers/gpu/drm/i915/gvt/gtt.c b/drivers/gpu/drm/i915/gvt/gtt.c index c2f7d20f6346..08c74e65836b 100644 --- a/drivers/gpu/drm/i915/gvt/gtt.c +++ b/drivers/gpu/drm/i915/gvt/gtt.c @@ -811,7 +811,7 @@ static int reclaim_one_ppgtt_mm(struct intel_gvt *gvt); /* Allocate shadow page table without guest page. */ static struct intel_vgpu_ppgtt_spt *ppgtt_alloc_spt( - struct intel_vgpu *vgpu, intel_gvt_gtt_type_t type) + struct intel_vgpu *vgpu, enum intel_gvt_gtt_type type) { struct device *kdev = &vgpu->gvt->dev_priv->drm.pdev->dev; struct intel_vgpu_ppgtt_spt *spt = NULL; @@ -861,7 +861,7 @@ err_free_spt: /* Allocate shadow page table associated with specific gfn. */ static struct intel_vgpu_ppgtt_spt *ppgtt_alloc_spt_gfn( - struct intel_vgpu *vgpu, intel_gvt_gtt_type_t type, + struct intel_vgpu *vgpu, enum intel_gvt_gtt_type type, unsigned long gfn, bool guest_pde_ips) { struct intel_vgpu_ppgtt_spt *spt; @@ -936,7 +936,7 @@ static int ppgtt_invalidate_spt_by_shadow_entry(struct intel_vgpu *vgpu, { struct intel_gvt_gtt_pte_ops *ops = vgpu->gvt->gtt.pte_ops; struct intel_vgpu_ppgtt_spt *s; - intel_gvt_gtt_type_t cur_pt_type; + enum intel_gvt_gtt_type cur_pt_type; GEM_BUG_ON(!gtt_type_is_pt(get_next_pt_type(e->type))); @@ -1076,6 +1076,9 @@ static struct intel_vgpu_ppgtt_spt *ppgtt_populate_spt_by_guest_entry( } else { int type = get_next_pt_type(we->type); + if (!gtt_type_is_pt(type)) + goto err; + spt = ppgtt_alloc_spt_gfn(vgpu, type, ops->get_pfn(we), ips); if (IS_ERR(spt)) { ret = PTR_ERR(spt); @@ -1855,7 +1858,7 @@ static void vgpu_free_mm(struct intel_vgpu_mm *mm) * Zero on success, negative error code in pointer if failed. */ struct intel_vgpu_mm *intel_vgpu_create_ppgtt_mm(struct intel_vgpu *vgpu, - intel_gvt_gtt_type_t root_entry_type, u64 pdps[]) + enum intel_gvt_gtt_type root_entry_type, u64 pdps[]) { struct intel_gvt *gvt = vgpu->gvt; struct intel_vgpu_mm *mm; @@ -2309,7 +2312,7 @@ int intel_vgpu_emulate_ggtt_mmio_write(struct intel_vgpu *vgpu, } static int alloc_scratch_pages(struct intel_vgpu *vgpu, - intel_gvt_gtt_type_t type) + enum intel_gvt_gtt_type type) { struct intel_vgpu_gtt *gtt = &vgpu->gtt; struct intel_gvt_gtt_pte_ops *ops = vgpu->gvt->gtt.pte_ops; @@ -2594,7 +2597,7 @@ struct intel_vgpu_mm *intel_vgpu_find_ppgtt_mm(struct intel_vgpu *vgpu, * Zero on success, negative error code if failed. */ struct intel_vgpu_mm *intel_vgpu_get_ppgtt_mm(struct intel_vgpu *vgpu, - intel_gvt_gtt_type_t root_entry_type, u64 pdps[]) + enum intel_gvt_gtt_type root_entry_type, u64 pdps[]) { struct intel_vgpu_mm *mm; diff --git a/drivers/gpu/drm/i915/gvt/gtt.h b/drivers/gpu/drm/i915/gvt/gtt.h index 32c573aea494..42d0394f0de2 100644 --- a/drivers/gpu/drm/i915/gvt/gtt.h +++ b/drivers/gpu/drm/i915/gvt/gtt.h @@ -95,8 +95,8 @@ struct intel_gvt_gtt { unsigned long scratch_mfn; }; -typedef enum { - GTT_TYPE_INVALID = -1, +enum intel_gvt_gtt_type { + GTT_TYPE_INVALID = 0, GTT_TYPE_GGTT_PTE, @@ -124,7 +124,7 @@ typedef enum { GTT_TYPE_PPGTT_PML4_PT, GTT_TYPE_MAX, -} intel_gvt_gtt_type_t; +}; enum intel_gvt_mm_type { INTEL_GVT_MM_GGTT, @@ -148,7 +148,7 @@ struct intel_vgpu_mm { union { struct { - intel_gvt_gtt_type_t root_entry_type; + enum intel_gvt_gtt_type root_entry_type; /* * The 4 PDPs in ring context. For 48bit addressing, * only PDP0 is valid and point to PML4. For 32it @@ -169,7 +169,7 @@ struct intel_vgpu_mm { }; struct intel_vgpu_mm *intel_vgpu_create_ppgtt_mm(struct intel_vgpu *vgpu, - intel_gvt_gtt_type_t root_entry_type, u64 pdps[]); + enum intel_gvt_gtt_type root_entry_type, u64 pdps[]); static inline void intel_vgpu_mm_get(struct intel_vgpu_mm *mm) { @@ -233,7 +233,7 @@ struct intel_vgpu_ppgtt_spt { struct intel_vgpu *vgpu; struct { - intel_gvt_gtt_type_t type; + enum intel_gvt_gtt_type type; bool pde_ips; /* for 64KB PTEs */ void *vaddr; struct page *page; @@ -241,7 +241,7 @@ struct intel_vgpu_ppgtt_spt { } shadow_page; struct { - intel_gvt_gtt_type_t type; + enum intel_gvt_gtt_type type; bool pde_ips; /* for 64KB PTEs */ unsigned long gfn; unsigned long write_cnt; @@ -267,7 +267,7 @@ struct intel_vgpu_mm *intel_vgpu_find_ppgtt_mm(struct intel_vgpu *vgpu, u64 pdps[]); struct intel_vgpu_mm *intel_vgpu_get_ppgtt_mm(struct intel_vgpu *vgpu, - intel_gvt_gtt_type_t root_entry_type, u64 pdps[]); + enum intel_gvt_gtt_type root_entry_type, u64 pdps[]); int intel_vgpu_put_ppgtt_mm(struct intel_vgpu *vgpu, u64 pdps[]); diff --git a/drivers/gpu/drm/i915/gvt/handlers.c b/drivers/gpu/drm/i915/gvt/handlers.c index 18f01eeb2510..90673fca792f 100644 --- a/drivers/gpu/drm/i915/gvt/handlers.c +++ b/drivers/gpu/drm/i915/gvt/handlers.c @@ -1206,7 +1206,7 @@ static int pvinfo_mmio_read(struct intel_vgpu *vgpu, unsigned int offset, static int handle_g2v_notification(struct intel_vgpu *vgpu, int notification) { - intel_gvt_gtt_type_t root_entry_type = GTT_TYPE_PPGTT_ROOT_L4_ENTRY; + enum intel_gvt_gtt_type root_entry_type = GTT_TYPE_PPGTT_ROOT_L4_ENTRY; struct intel_vgpu_mm *mm; u64 *pdps; @@ -3303,7 +3303,7 @@ void intel_gvt_clean_mmio_info(struct intel_gvt *gvt) /* Special MMIO blocks. */ static struct gvt_mmio_block mmio_blocks[] = { {D_SKL_PLUS, _MMIO(CSR_MMIO_START_RANGE), 0x3000, NULL, NULL}, - {D_ALL, MCHBAR_MIRROR_REG_BASE, 0x4000, NULL, NULL}, + {D_ALL, _MMIO(MCHBAR_MIRROR_BASE_SNB), 0x40000, NULL, NULL}, {D_ALL, _MMIO(VGT_PVINFO_PAGE), VGT_PVINFO_SIZE, pvinfo_mmio_read, pvinfo_mmio_write}, {D_ALL, LGC_PALETTE(PIPE_A, 0), 1024, NULL, NULL}, diff --git a/drivers/gpu/drm/i915/gvt/mmio_context.c b/drivers/gpu/drm/i915/gvt/mmio_context.c index e7e14c842be4..edf6d646eb25 100644 --- a/drivers/gpu/drm/i915/gvt/mmio_context.c +++ b/drivers/gpu/drm/i915/gvt/mmio_context.c @@ -132,6 +132,7 @@ static struct engine_mmio gen9_engine_mmio_list[] __cacheline_aligned = { {RCS0, GEN9_GAMT_ECO_REG_RW_IA, 0x0, false}, /* 0x4ab0 */ {RCS0, GEN9_CSFE_CHICKEN1_RCS, 0xffff, false}, /* 0x20d4 */ + {RCS0, _MMIO(0x20D8), 0xffff, true}, /* 0x20d8 */ {RCS0, GEN8_GARBCNTL, 0x0, false}, /* 0xb004 */ {RCS0, GEN7_FF_THREAD_MODE, 0x0, false}, /* 0x20a0 */ diff --git a/drivers/gpu/drm/i915/gvt/reg.h b/drivers/gpu/drm/i915/gvt/reg.h index 3de5b643b266..33aaa14bfdde 100644 --- a/drivers/gpu/drm/i915/gvt/reg.h +++ b/drivers/gpu/drm/i915/gvt/reg.h @@ -126,7 +126,4 @@ #define RING_GFX_MODE(base) _MMIO((base) + 0x29c) #define VF_GUARDBAND _MMIO(0x83a4) -/* define the effective range of MCHBAR register on Sandybridge+ */ -#define MCHBAR_MIRROR_REG_BASE _MMIO(MCHBAR_MIRROR_BASE_SNB + 0x4000) - #endif diff --git a/drivers/gpu/drm/i915/gvt/scheduler.c b/drivers/gpu/drm/i915/gvt/scheduler.c index 8998fa5ab198..7c99bbc3e2b8 100644 --- a/drivers/gpu/drm/i915/gvt/scheduler.c +++ b/drivers/gpu/drm/i915/gvt/scheduler.c @@ -1343,7 +1343,7 @@ static int prepare_mm(struct intel_vgpu_workload *workload) struct execlist_ctx_descriptor_format *desc = &workload->ctx_desc; struct intel_vgpu_mm *mm; struct intel_vgpu *vgpu = workload->vgpu; - intel_gvt_gtt_type_t root_entry_type; + enum intel_gvt_gtt_type root_entry_type; u64 pdps[GVT_RING_CTX_NR_PDPS]; switch (desc->addressing_mode) { diff --git a/drivers/gpu/drm/i915/i915_request.c b/drivers/gpu/drm/i915/i915_request.c index b836721d3b13..ce342f7f7ddb 100644 --- a/drivers/gpu/drm/i915/i915_request.c +++ b/drivers/gpu/drm/i915/i915_request.c @@ -425,6 +425,26 @@ void __i915_request_submit(struct i915_request *request) if (i915_gem_context_is_banned(request->gem_context)) i915_request_skip(request, -EIO); + /* + * Are we using semaphores when the gpu is already saturated? + * + * Using semaphores incurs a cost in having the GPU poll a + * memory location, busywaiting for it to change. The continual + * memory reads can have a noticeable impact on the rest of the + * system with the extra bus traffic, stalling the cpu as it too + * tries to access memory across the bus (perf stat -e bus-cycles). + * + * If we installed a semaphore on this request and we only submit + * the request after the signaler completed, that indicates the + * system is overloaded and using semaphores at this time only + * increases the amount of work we are doing. If so, we disable + * further use of semaphores until we are idle again, whence we + * optimistically try again. + */ + if (request->sched.semaphores && + i915_sw_fence_signaled(&request->semaphore)) + request->hw_context->saturated |= request->sched.semaphores; + /* We may be recursing from the signal callback of another i915 fence */ spin_lock_nested(&request->lock, SINGLE_DEPTH_NESTING); @@ -799,6 +819,39 @@ err_unreserve: } static int +i915_request_await_start(struct i915_request *rq, struct i915_request *signal) +{ + if (list_is_first(&signal->ring_link, &signal->ring->request_list)) + return 0; + + signal = list_prev_entry(signal, ring_link); + if (i915_timeline_sync_is_later(rq->timeline, &signal->fence)) + return 0; + + return i915_sw_fence_await_dma_fence(&rq->submit, + &signal->fence, 0, + I915_FENCE_GFP); +} + +static intel_engine_mask_t +already_busywaiting(struct i915_request *rq) +{ + /* + * Polling a semaphore causes bus traffic, delaying other users of + * both the GPU and CPU. We want to limit the impact on others, + * while taking advantage of early submission to reduce GPU + * latency. Therefore we restrict ourselves to not using more + * than one semaphore from each source, and not using a semaphore + * if we have detected the engine is saturated (i.e. would not be + * submitted early and cause bus traffic reading an already passed + * semaphore). + * + * See the are-we-too-late? check in __i915_request_submit(). + */ + return rq->sched.semaphores | rq->hw_context->saturated; +} + +static int emit_semaphore_wait(struct i915_request *to, struct i915_request *from, gfp_t gfp) @@ -811,11 +864,15 @@ emit_semaphore_wait(struct i915_request *to, GEM_BUG_ON(INTEL_GEN(to->i915) < 8); /* Just emit the first semaphore we see as request space is limited. */ - if (to->sched.semaphores & from->engine->mask) + if (already_busywaiting(to) & from->engine->mask) return i915_sw_fence_await_dma_fence(&to->submit, &from->fence, 0, I915_FENCE_GFP); + err = i915_request_await_start(to, from); + if (err < 0) + return err; + err = i915_sw_fence_await_dma_fence(&to->semaphore, &from->fence, 0, I915_FENCE_GFP); diff --git a/drivers/gpu/drm/i915/intel_color.c b/drivers/gpu/drm/i915/intel_color.c index ca341a9e47e6..9093daabc290 100644 --- a/drivers/gpu/drm/i915/intel_color.c +++ b/drivers/gpu/drm/i915/intel_color.c @@ -173,13 +173,13 @@ static void icl_update_output_csc(struct intel_crtc *crtc, I915_WRITE(PIPE_CSC_OUTPUT_PREOFF_LO(pipe), preoff[2]); I915_WRITE(PIPE_CSC_OUTPUT_COEFF_RY_GY(pipe), coeff[0] << 16 | coeff[1]); - I915_WRITE(PIPE_CSC_OUTPUT_COEFF_BY(pipe), coeff[2]); + I915_WRITE(PIPE_CSC_OUTPUT_COEFF_BY(pipe), coeff[2] << 16); I915_WRITE(PIPE_CSC_OUTPUT_COEFF_RU_GU(pipe), coeff[3] << 16 | coeff[4]); - I915_WRITE(PIPE_CSC_OUTPUT_COEFF_BU(pipe), coeff[5]); + I915_WRITE(PIPE_CSC_OUTPUT_COEFF_BU(pipe), coeff[5] << 16); I915_WRITE(PIPE_CSC_OUTPUT_COEFF_RV_GV(pipe), coeff[6] << 16 | coeff[7]); - I915_WRITE(PIPE_CSC_OUTPUT_COEFF_BV(pipe), coeff[8]); + I915_WRITE(PIPE_CSC_OUTPUT_COEFF_BV(pipe), coeff[8] << 16); I915_WRITE(PIPE_CSC_OUTPUT_POSTOFF_HI(pipe), postoff[0]); I915_WRITE(PIPE_CSC_OUTPUT_POSTOFF_ME(pipe), postoff[1]); diff --git a/drivers/gpu/drm/i915/intel_context.c b/drivers/gpu/drm/i915/intel_context.c index 8931e0fee873..924cc556223a 100644 --- a/drivers/gpu/drm/i915/intel_context.c +++ b/drivers/gpu/drm/i915/intel_context.c @@ -230,6 +230,7 @@ intel_context_init(struct intel_context *ce, ce->gem_context = ctx; ce->engine = engine; ce->ops = engine->cops; + ce->saturated = 0; INIT_LIST_HEAD(&ce->signal_link); INIT_LIST_HEAD(&ce->signals); diff --git a/drivers/gpu/drm/i915/intel_context_types.h b/drivers/gpu/drm/i915/intel_context_types.h index 68b4ca1611e0..339c7437fe82 100644 --- a/drivers/gpu/drm/i915/intel_context_types.h +++ b/drivers/gpu/drm/i915/intel_context_types.h @@ -14,6 +14,7 @@ #include <linux/types.h> #include "i915_active_types.h" +#include "intel_engine_types.h" struct i915_gem_context; struct i915_vma; @@ -58,6 +59,8 @@ struct intel_context { atomic_t pin_count; struct mutex pin_mutex; /* guards pinning and associated on-gpuing */ + intel_engine_mask_t saturated; /* submitting semaphores too late? */ + /** * active_tracker: Active tracker for the external rq activity * on this intel_context object. diff --git a/drivers/gpu/drm/i915/intel_workarounds.c b/drivers/gpu/drm/i915/intel_workarounds.c index ccaf63679435..9682dd575152 100644 --- a/drivers/gpu/drm/i915/intel_workarounds.c +++ b/drivers/gpu/drm/i915/intel_workarounds.c @@ -541,10 +541,6 @@ static void icl_ctx_workarounds_init(struct intel_engine_cs *engine) WA_SET_BIT_MASKED(GEN7_ROW_CHICKEN2, GEN11_TDL_CLOCK_GATING_FIX_DISABLE); - /* WaEnableStateCacheRedirectToCS:icl */ - WA_SET_BIT_MASKED(GEN9_SLICE_COMMON_ECO_CHICKEN1, - GEN11_STATE_CACHE_REDIRECT_TO_CS); - /* Wa_2006665173:icl (pre-prod) */ if (IS_ICL_REVID(i915, ICL_REVID_A0, ICL_REVID_A0)) WA_SET_BIT_MASKED(GEN11_COMMON_SLICE_CHICKEN3, @@ -1050,6 +1046,9 @@ static void icl_whitelist_build(struct i915_wa_list *w) /* WaAllowUMDToModifySamplerMode:icl */ whitelist_reg(w, GEN10_SAMPLER_MODE); + + /* WaEnableStateCacheRedirectToCS:icl */ + whitelist_reg(w, GEN9_SLICE_COMMON_ECO_CHICKEN1); } void intel_engine_init_whitelist(struct intel_engine_cs *engine) diff --git a/drivers/gpu/drm/msm/Makefile b/drivers/gpu/drm/msm/Makefile index 72d1bfcaab7a..7a05cbf2f820 100644 --- a/drivers/gpu/drm/msm/Makefile +++ b/drivers/gpu/drm/msm/Makefile @@ -1,7 +1,7 @@ # SPDX-License-Identifier: GPL-2.0 -ccflags-y := -Idrivers/gpu/drm/msm -ccflags-y += -Idrivers/gpu/drm/msm/disp/dpu1 -ccflags-$(CONFIG_DRM_MSM_DSI) += -Idrivers/gpu/drm/msm/dsi +ccflags-y := -I $(srctree)/$(src) +ccflags-y += -I $(srctree)/$(src)/disp/dpu1 +ccflags-$(CONFIG_DRM_MSM_DSI) += -I $(srctree)/$(src)/dsi msm-y := \ adreno/adreno_device.o \ diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c index ec24508b9d68..e74dce474250 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c @@ -527,6 +527,7 @@ static int a6xx_hw_init(struct msm_gpu *gpu) dev_warn_once(gpu->dev->dev, "Zap shader not enabled - using SECVID_TRUST_CNTL instead\n"); gpu_write(gpu, REG_A6XX_RBBM_SECVID_TRUST_CNTL, 0x0); + ret = 0; } out: diff --git a/drivers/gpu/drm/nouveau/Kbuild b/drivers/gpu/drm/nouveau/Kbuild index 581404e6544d..378c5dd692b0 100644 --- a/drivers/gpu/drm/nouveau/Kbuild +++ b/drivers/gpu/drm/nouveau/Kbuild @@ -1,7 +1,7 @@ -ccflags-y += -I$(src)/include -ccflags-y += -I$(src)/include/nvkm -ccflags-y += -I$(src)/nvkm -ccflags-y += -I$(src) +ccflags-y += -I $(srctree)/$(src)/include +ccflags-y += -I $(srctree)/$(src)/include/nvkm +ccflags-y += -I $(srctree)/$(src)/nvkm +ccflags-y += -I $(srctree)/$(src) # NVKM - HW resource manager #- code also used by various userspace tools/tests diff --git a/drivers/gpu/drm/panfrost/Kconfig b/drivers/gpu/drm/panfrost/Kconfig index 7f5e572daa2d..591611dc4e34 100644 --- a/drivers/gpu/drm/panfrost/Kconfig +++ b/drivers/gpu/drm/panfrost/Kconfig @@ -3,7 +3,7 @@ config DRM_PANFROST tristate "Panfrost (DRM support for ARM Mali Midgard/Bifrost GPUs)" depends on DRM - depends on ARM || ARM64 || COMPILE_TEST + depends on ARM || ARM64 || (COMPILE_TEST && !GENERIC_ATOMIC64) depends on MMU select DRM_SCHED select IOMMU_SUPPORT diff --git a/drivers/gpu/drm/panfrost/panfrost_devfreq.c b/drivers/gpu/drm/panfrost/panfrost_devfreq.c index a8121ae67ee3..238bd1d89d43 100644 --- a/drivers/gpu/drm/panfrost/panfrost_devfreq.c +++ b/drivers/gpu/drm/panfrost/panfrost_devfreq.c @@ -7,6 +7,7 @@ #include <linux/regulator/consumer.h> #include "panfrost_device.h" +#include "panfrost_devfreq.h" #include "panfrost_features.h" #include "panfrost_issues.h" #include "panfrost_gpu.h" diff --git a/drivers/gpu/drm/panfrost/panfrost_device.c b/drivers/gpu/drm/panfrost/panfrost_device.c index 91e8fb0f2b25..970f669c6d29 100644 --- a/drivers/gpu/drm/panfrost/panfrost_device.c +++ b/drivers/gpu/drm/panfrost/panfrost_device.c @@ -98,6 +98,7 @@ int panfrost_device_init(struct panfrost_device *pfdev) struct resource *res; mutex_init(&pfdev->sched_lock); + mutex_init(&pfdev->reset_lock); INIT_LIST_HEAD(&pfdev->scheduled_jobs); spin_lock_init(&pfdev->hwaccess_lock); diff --git a/drivers/gpu/drm/panfrost/panfrost_device.h b/drivers/gpu/drm/panfrost/panfrost_device.h index 1ba48d105763..56f452dfb490 100644 --- a/drivers/gpu/drm/panfrost/panfrost_device.h +++ b/drivers/gpu/drm/panfrost/panfrost_device.h @@ -78,6 +78,7 @@ struct panfrost_device { struct list_head scheduled_jobs; struct mutex sched_lock; + struct mutex reset_lock; struct { struct devfreq *devfreq; diff --git a/drivers/gpu/drm/panfrost/panfrost_drv.c b/drivers/gpu/drm/panfrost/panfrost_drv.c index c06af78ab833..94b0819ad50b 100644 --- a/drivers/gpu/drm/panfrost/panfrost_drv.c +++ b/drivers/gpu/drm/panfrost/panfrost_drv.c @@ -3,8 +3,6 @@ /* Copyright 2019 Linaro, Ltd., Rob Herring <robh@kernel.org> */ /* Copyright 2019 Collabora ltd. */ -#include <linux/bitfield.h> -#include <linux/dma-mapping.h> #include <linux/module.h> #include <linux/of_platform.h> #include <linux/pagemap.h> @@ -172,13 +170,27 @@ static int panfrost_ioctl_submit(struct drm_device *dev, void *data, { struct panfrost_device *pfdev = dev->dev_private; struct drm_panfrost_submit *args = data; - struct drm_syncobj *sync_out; + struct drm_syncobj *sync_out = NULL; struct panfrost_job *job; int ret = 0; + if (!args->jc) + return -EINVAL; + + if (args->requirements && args->requirements != PANFROST_JD_REQ_FS) + return -EINVAL; + + if (args->out_sync > 0) { + sync_out = drm_syncobj_find(file, args->out_sync); + if (!sync_out) + return -ENODEV; + } + job = kzalloc(sizeof(*job), GFP_KERNEL); - if (!job) - return -ENOMEM; + if (!job) { + ret = -ENOMEM; + goto fail_out_sync; + } kref_init(&job->refcount); @@ -190,25 +202,24 @@ static int panfrost_ioctl_submit(struct drm_device *dev, void *data, ret = panfrost_copy_in_sync(dev, file, args, job); if (ret) - goto fail; + goto fail_job; ret = panfrost_lookup_bos(dev, file, args, job); if (ret) - goto fail; + goto fail_job; ret = panfrost_job_push(job); if (ret) - goto fail; + goto fail_job; /* Update the return sync object for the job */ - sync_out = drm_syncobj_find(file, args->out_sync); - if (sync_out) { + if (sync_out) drm_syncobj_replace_fence(sync_out, job->render_done_fence); - drm_syncobj_put(sync_out); - } -fail: +fail_job: panfrost_job_put(job); +fail_out_sync: + drm_syncobj_put(sync_out); return ret; } @@ -384,16 +395,15 @@ static int panfrost_probe(struct platform_device *pdev) err = panfrost_device_init(pfdev); if (err) { - dev_err(&pdev->dev, "Fatal error during GPU init\n"); + if (err != -EPROBE_DEFER) + dev_err(&pdev->dev, "Fatal error during GPU init\n"); goto err_out0; } - dma_set_mask_and_coherent(pfdev->dev, - DMA_BIT_MASK(FIELD_GET(0xff00, pfdev->features.mmu_features))); - err = panfrost_devfreq_init(pfdev); if (err) { - dev_err(&pdev->dev, "Fatal error during devfreq init\n"); + if (err != -EPROBE_DEFER) + dev_err(&pdev->dev, "Fatal error during devfreq init\n"); goto err_out1; } @@ -410,6 +420,7 @@ static int panfrost_probe(struct platform_device *pdev) err_out1: panfrost_device_fini(pfdev); err_out0: + pm_runtime_disable(pfdev->dev); drm_dev_put(ddev); return err; } diff --git a/drivers/gpu/drm/panfrost/panfrost_gem.c b/drivers/gpu/drm/panfrost/panfrost_gem.c index 8a0376283a21..a5528a360ef4 100644 --- a/drivers/gpu/drm/panfrost/panfrost_gem.c +++ b/drivers/gpu/drm/panfrost/panfrost_gem.c @@ -14,7 +14,7 @@ /* Called DRM core on the last userspace/kernel unreference of the * BO. */ -void panfrost_gem_free_object(struct drm_gem_object *obj) +static void panfrost_gem_free_object(struct drm_gem_object *obj) { struct panfrost_gem_object *bo = to_panfrost_bo(obj); struct panfrost_device *pfdev = obj->dev->dev_private; diff --git a/drivers/gpu/drm/panfrost/panfrost_gpu.c b/drivers/gpu/drm/panfrost/panfrost_gpu.c index aceaf6e44a09..58ef25573cda 100644 --- a/drivers/gpu/drm/panfrost/panfrost_gpu.c +++ b/drivers/gpu/drm/panfrost/panfrost_gpu.c @@ -2,8 +2,10 @@ /* Copyright 2018 Marty E. Plummer <hanetzer@startmail.com> */ /* Copyright 2019 Linaro, Ltd., Rob Herring <robh@kernel.org> */ /* Copyright 2019 Collabora ltd. */ +#include <linux/bitfield.h> #include <linux/bitmap.h> #include <linux/delay.h> +#include <linux/dma-mapping.h> #include <linux/interrupt.h> #include <linux/io.h> #include <linux/iopoll.h> @@ -276,13 +278,13 @@ static void panfrost_gpu_init_features(struct panfrost_device *pfdev) pfdev->features.hw_issues); dev_info(pfdev->dev, "Features: L2:0x%08x Shader:0x%08x Tiler:0x%08x Mem:0x%0x MMU:0x%08x AS:0x%x JS:0x%x", - gpu_read(pfdev, GPU_L2_FEATURES), - gpu_read(pfdev, GPU_CORE_FEATURES), - gpu_read(pfdev, GPU_TILER_FEATURES), - gpu_read(pfdev, GPU_MEM_FEATURES), - gpu_read(pfdev, GPU_MMU_FEATURES), - gpu_read(pfdev, GPU_AS_PRESENT), - gpu_read(pfdev, GPU_JS_PRESENT)); + pfdev->features.l2_features, + pfdev->features.core_features, + pfdev->features.tiler_features, + pfdev->features.mem_features, + pfdev->features.mmu_features, + pfdev->features.as_present, + pfdev->features.js_present); dev_info(pfdev->dev, "shader_present=0x%0llx l2_present=0x%0llx", pfdev->features.shader_present, pfdev->features.l2_present); @@ -332,6 +334,9 @@ int panfrost_gpu_init(struct panfrost_device *pfdev) panfrost_gpu_init_features(pfdev); + dma_set_mask_and_coherent(pfdev->dev, + DMA_BIT_MASK(FIELD_GET(0xff00, pfdev->features.mmu_features))); + irq = platform_get_irq_byname(to_platform_device(pfdev->dev), "gpu"); if (irq <= 0) return -ENODEV; diff --git a/drivers/gpu/drm/panfrost/panfrost_job.c b/drivers/gpu/drm/panfrost/panfrost_job.c index 0a7ed04f7d52..a5716c8fe8b3 100644 --- a/drivers/gpu/drm/panfrost/panfrost_job.c +++ b/drivers/gpu/drm/panfrost/panfrost_job.c @@ -384,6 +384,8 @@ static void panfrost_job_timedout(struct drm_sched_job *sched_job) job_read(pfdev, JS_TAIL_LO(js)), sched_job); + mutex_lock(&pfdev->reset_lock); + for (i = 0; i < NUM_JOB_SLOTS; i++) drm_sched_stop(&pfdev->js->queue[i].sched); @@ -406,6 +408,8 @@ static void panfrost_job_timedout(struct drm_sched_job *sched_job) /* restart scheduler after GPU is usable again */ for (i = 0; i < NUM_JOB_SLOTS; i++) drm_sched_start(&pfdev->js->queue[i].sched, true); + + mutex_unlock(&pfdev->reset_lock); } static const struct drm_sched_backend_ops panfrost_sched_ops = { diff --git a/drivers/gpu/drm/radeon/radeon_display.c b/drivers/gpu/drm/radeon/radeon_display.c index aa898c699101..433df7036f96 100644 --- a/drivers/gpu/drm/radeon/radeon_display.c +++ b/drivers/gpu/drm/radeon/radeon_display.c @@ -922,12 +922,12 @@ static void avivo_get_fb_ref_div(unsigned nom, unsigned den, unsigned post_div, ref_div_max = max(min(100 / post_div, ref_div_max), 1u); /* get matching reference and feedback divider */ - *ref_div = min(max(DIV_ROUND_CLOSEST(den, post_div), 1u), ref_div_max); + *ref_div = min(max(den/post_div, 1u), ref_div_max); *fb_div = DIV_ROUND_CLOSEST(nom * *ref_div * post_div, den); /* limit fb divider to its maximum */ if (*fb_div > fb_div_max) { - *ref_div = DIV_ROUND_CLOSEST(*ref_div * fb_div_max, *fb_div); + *ref_div = (*ref_div * fb_div_max)/(*fb_div); *fb_div = fb_div_max; } } diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_drv.c b/drivers/gpu/drm/rockchip/rockchip_drm_drv.c index 8d7a634c12c2..cb938d3cd3c2 100644 --- a/drivers/gpu/drm/rockchip/rockchip_drm_drv.c +++ b/drivers/gpu/drm/rockchip/rockchip_drm_drv.c @@ -448,6 +448,14 @@ static int rockchip_drm_platform_remove(struct platform_device *pdev) return 0; } +static void rockchip_drm_platform_shutdown(struct platform_device *pdev) +{ + struct drm_device *drm = platform_get_drvdata(pdev); + + if (drm) + drm_atomic_helper_shutdown(drm); +} + static const struct of_device_id rockchip_drm_dt_ids[] = { { .compatible = "rockchip,display-subsystem", }, { /* sentinel */ }, @@ -457,6 +465,7 @@ MODULE_DEVICE_TABLE(of, rockchip_drm_dt_ids); static struct platform_driver rockchip_drm_platform_driver = { .probe = rockchip_drm_platform_probe, .remove = rockchip_drm_platform_remove, + .shutdown = rockchip_drm_platform_shutdown, .driver = { .name = "rockchip-drm", .of_match_table = rockchip_drm_dt_ids, |