summaryrefslogtreecommitdiffstats
path: root/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
AgeCommit message (Collapse)AuthorFilesLines
2017-03-29drm/amdgpu:changes in gfx DMAframe scheme (v2)Monk Liu1-11/+18
1) Adapt to vulkan: Now use double SWITCH BUFFER to replace the 128 nops w/a, because when vulkan introduced, umd can insert 7 ~ 16 IBs per submit which makes 256 DW size cannot hold the whole DMAframe (if we still insert those 128 nops), CP team suggests use double SWITCH_BUFFERs, instead of tricky 128 NOPs w/a. 2) To fix the CE VM fault issue when MCBP introduced: Need one more COND_EXEC wrapping IB part (original one us for VM switch part). this change can fix vm fault issue caused by below scenario without this change: >CE passed original COND_EXEC (no MCBP issued this moment), proceed as normal. >DE catch up to this COND_EXEC, but this time MCBP issued, thus DE treats all following packages as NOP. The following VM switch packages now looks just as NOP to DE, so DE dosen't do VM flush at all. >Now CE proceeds to the first IBc, and triggers VM fault, because DE didn't do VM flush for this DMAframe. 3) change estimated alloc size for gfx9. with new DMAframe scheme, we need modify emit_frame_size for gfx9 4) No need to insert 128 nops after gfx8 vm flush anymore because there was double SWITCH_BUFFER append to vm flush, and for gfx7 we already use double SWITCH_BUFFER following after vm_flush so no change needed for it. 5) Change emit_frame_size for gfx8 v2: squash in BUG removal from Monk Signed-off-by: Monk Liu <Monk.Liu@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-03-29drm/amdgpu:enable MCBP for SR-IOV (v2)Monk Liu1-0/+3
Apply the new IB during IB emit for SRIOV with MCBP v2: agd: use define instead of magic number Signed-off-by: Monk Liu <Monk.Liu@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-03-29drm/amdgpu:implement cond_exec for gfx8Monk Liu1-0/+30
when MCBP enabled for gfx8, the cond_exec must also be implemented, otherwise there will be odds to meet cross engine (ce and me) deadlock when world switch happens. Signed-off-by: Monk Liu <Monk.Liu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-03-29drm/amdgpu: use new flag to handle different firmware loading methodHuang Rui1-3/+3
This patch introduces a new flag named "amdgpu_firmware_load_type" to handle different firmware loading method. Since Vega10, there are three ways to load firmware. It would be better to use a flag and a fw_load_type kernel parameter to configure it. Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Huang Rui <ray.huang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-03-29drm/amdgpu/gfx8: further KIQ parameter cleanupAlex Deucher1-13/+11
The ring structure already has what we need. Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-03-29drm/amdgpu/gfx8: store the eop gpu addr in the ring structureAlex Deucher1-8/+6
Avoids passing around additional parameters during setup. Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-03-29drm/amdgpu/gfx8: reduce the functon params for mpq setupAlex Deucher1-18/+14
Everything we need is in the ring structure. No need to pass all the bits explicitly. Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-03-29drm/amdgpu/gfx8: reserve kiq eop object before unmapping itAlex Deucher1-0/+4
It's required. Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-03-29drm/amdgpu/gfx8: fold loops in kiq_resume()Alex Deucher1-4/+0
No need to loop through the compute queues twice. Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-03-29drm/amdgpu/gfx8: test KIQ before compute ringsAlex Deucher1-6/+7
If KIQ isn't working, the compute rings won't work either. Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-03-29drm/amdgpu/gfx8: reserve mqd objects before mapping themAlex Deucher1-11/+21
It's required. Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-03-29drm/amdgpu/gfx8: rename some functionsAlex Deucher1-6/+6
To better match where they are used. Called from sw_init and sw_fini. Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-03-29drm/amdgpu/gfx8: whitespace cleanupAlex Deucher1-6/+5
Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-03-29drm/amdgpu: enable gfx/system/vce clockgating on Polars12.Rex Zhu1-0/+1
Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-03-29drm/amdgpu: remove unneeded conversions to boolAndrew F. Davis1-2/+2
Found with scripts/coccinelle/misc/boolconv.cocci. Signed-off-by: Andrew F. Davis <afd@ti.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-03-29drm/amdgpu: switch ih handling to two levels (v3)Alex Deucher1-4/+6
Newer asics have a two levels of irq ids now: client id - the IP src id - the interrupt src within the IP v2: integrated Christian's comments. v3: fix rebase fail in SI and CIK Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Ken Wang <Qingqing.Wang@amd.com> Reviewed-by: Ken Wang <Qingqing.Wang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-03-29drm/amdgpu: change wptr to 64 bits (v2)Ken Wang1-10/+13
Newer asics need 64 bit wptrs. If the wptr is now smaller than the rptr that doesn't indicate a wrap-around anymore. v2: integrate Christian's comments. Signed-off-by: Ken Wang <Qingqing.Wang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-03-29drm/amdgpu: init aperture definitions (v2)Junwei Zhang1-4/+13
v2: agd: move apertures to mc structure Signed-off-by: Flora Cui <Flora.Cui@amd.com> Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-03-29drm/amdgpu/gfx8: move CE&DE meta data structure to vi_structs.hXiangliang Yu1-8/+8
Because different HWs have different definition for CE & DE meta data, follow mqd design to move the structures to vi_structs.h. And change the prefix from amdgpu to vi as the structures is only for VI family. Signed-off-by: Xiangliang Yu <Xiangliang.Yu@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-03-29drm/amdgpu: Fix module unload hang by KIQ IRQ setTrigger Huang1-6/+4
In some cases, manually insmod/rmmod amdgpu is necessary. When unloading amdgpu, the KIQ IRQ enable/disable function will case system hang. The root cause is, in the sequence of function amdgpu_fini, the sw_fini of IP block AMD_IP_BLOCK_TYPE_GFX will be invoked earlier than that of AMD_IP_BLOCK_TYPE_IH. So continue to use the variable freed by AMD_IP_BLOCK_TYPE_GFX will cause system hang. Signed-off-by: Trigger Huang <trigger.huang@amd.com> Reviewed-by: Xiangliang Yu < Xiangliang.Yu@amd.com> Reviewed-by: Monk Liu <monk.liu@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-03-29drm/amdgpu: export gfx config double offchip LDS buffers (v3)Junwei Zhang1-0/+14
v2: move the config struct to drm_amdgpu_info_device v3: move the config feature to amdgpu_gca_config Signed-off-by: Junwei Zhang <Jerry.Zhang@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-03-29drm/amdgpu/gfx: free memory of mqd backupXiangliang Yu1-0/+2
Need to free mqd backup when destroying ring. Signed-off-by: Xiangliang Yu <Xiangliang.Yu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Monk Liu <Monk.Liu@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-03-29drm/amdgpu: change pointer of mqd_ptr & mqd_backup to voidXiangliang Yu1-9/+17
vi_mqd is only used by VI family but mqd_ptr and mqd_backup is common for all ASIC, so change the pointer to void. Signed-off-by: Xiangliang Yu <Xiangliang.Yu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Reviewed-by: Monk Liu <Monk.Liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-03-29drm/amdgpu:fix kiq_resume routine (V2)Monk Liu1-11/+33
v2: use in_rest to fix compute ring test failure issue which occured after FLR/gpu_reset. we need backup a clean status of MQD which was created in drv load stage, and use it in resume stage, otherwise KCQ and KIQ all may faild in ring/ib test. Signed-off-by: Monk Liu <Monk.Liu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Xiangliang Yu <Xiangliang.Yu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-03-29drm/amdgpu:use clear_ring to clr RBMonk Liu1-0/+1
In resume routine, we need clr RB prior to the ring test of engine, otherwise some engine hang duplicated during GPU reset. Signed-off-by: Monk Liu <Monk.Liu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-03-29drm/amdgpu:alloc mqd backupMonk Liu1-0/+10
this is required for restoring the mqds after GPU reset. Signed-off-by: Monk Liu <Monk.Liu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Xiangliang Yu <Xiangliang.Yu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-03-29drm/amdgpu:use MACRO like other placesMonk Liu1-4/+2
Change-Id: Ica8f86577a50d817119de4b4fb95068dc72652a9 Signed-off-by: Monk Liu <Monk.Liu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-03-29drm/amdgpu:divide KCQ mqd init to sw and hwMonk Liu1-65/+42
sw part only invoked once during sw_init. hw part invoked during first drv load and resume later. that way we cannot alloc mqd in hw/resume, we only keep mqd allocted in sw_init routine. and hw_init routine only kmap and set it. Signed-off-by: Monk Liu <Monk.Liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-03-29drm/amdgpu:minor cleanupMonk Liu1-4/+1
Signed-off-by: Monk Liu <Monk.Liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-03-29drm/amdgpu:no need use sriov vf checksMonk Liu1-9/+4
We ultimately want to re-use this for bare metal, so no need to have vf checks in the KIQ code itself since kiq itself is currently only used in VF cases. Signed-off-by: Monk Liu <Monk.Liu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-03-29drm/amdgpu:bo_free_kernel will set ptr to NULL if freedMonk Liu1-1/+0
Signed-off-by: Monk Liu <Monk.Liu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-03-29drm/amdgpu:imple mqd soft ini/finiMonk Liu1-0/+52
this is for SRIOV fix: mqd soft init/fini will be invoked by sw_init to allocate BO for compute MQD resource, instead of original scheme that hw_init allocates MQD. because if hw_init allocates MQD, then resume will allocate MQD, and that lead to memory leak after driver recovered from hang. Signed-off-by: Monk Liu <Monk.Liu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-03-29drm/damdgpu:add new mqd member in ringMonk Liu1-0/+2
introduce a new mqd member in ring is for later usage. we need keep a clean version of MQD for the purpose of recovering compute rings from hang. Signed-off-by: Monk Liu <Monk.Liu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-03-29drm/amdgpu:cg & pg shouldn't active on VF deviceMonk Liu1-0/+9
CG & PG function changes engine clock/gating, which is not appropriate for VF device, because one vf doesn't know the whole picture of engine's overall workload. Signed-off-by: Monk Liu <Monk.Liu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-02-16drm/amdgpu: fix warning on older gcc releasesArnd Bergmann1-2/+2
gcc-4.8 warns about '{0}' being used an an initializer for nested structures: drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c: In function ‘gfx_v8_0_ring_emit_ce_meta_init’: drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c:7263:2: warning: missing braces around initializer [-Wmissing-braces] } ce_payload = {0}; drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c: In function ‘gfx_v8_0_ring_emit_de_meta_init’: drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c:7290:2: warning: missing braces around initializer [-Wmissing-braces] } de_payload = {0}; Using an empty {} initializer however has the same effect and works on all versions. Fixes: acad2b2a7b70 ("drm/amdgpu:implement CE/DE meta-init routines") Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-02-13drm/amdgpu: Initialize pipe priority order on graphic initializationozeng1-0/+8
Initialized PIPE_ORDER_TS0/1/2/3 field of SPI_ARB_PRIORITY register to 2. This set the pipe priority order to: 02 - HP3D, CS_H, GFX, CS_M, CS_L Signed-off-by: Oak Zeng <Oak.Zeng@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-01-27drm/amd/amdgpu: remove the uncessary parameter for ib schedulerJunwei Zhang1-2/+2
Signed-off-by: Junwei Zhang <Jerry.Zhang@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-01-27drm/amdgpu:in cntx_ctrl we need insert meta-init for CE/DE(V2)Monk Liu1-1/+9
to support SRIOV preemption. v2: fix emit_frame_size Signed-off-by: Monk Liu <Monk.Liu@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-01-27drm/amdgpu:implement CE/DE meta-init routinesMonk Liu1-0/+61
those package need to insert into ring buffer for SRIOV case. they are used to let CP do preemption. Signed-off-by: Monk Liu <Monk.Liu@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-01-27drm/amdgpu:add META_DATA struct for CSA/SRIOV v2Monk Liu1-0/+7
META-DATA is used in GFX cmd submit, we have two format suit for META-DATA-init, one is legacy and another is for chained-ib preempt, which is used in vulkan UMD. v2: drop use CP version number to judge if chain-ib supports or not, we wait for it mature Signed-off-by: Monk Liu <Monk.Liu@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-01-27drm/amdgpu: simplify allocation of scratch regsNils Wallménius1-6/+1
The scratch regs are sequential so there's no need to keep them in an array, we can just return the index of the first free register + the base register. Also change the array of bools for keeping track of the free regs to a bitfield. Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Nils Wallménius <nils.wallmenius@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-01-27drm/amdgpu/gfx8: fix bugon.cocci warningsJulia Lawall1-2/+1
Use BUG_ON instead of a if condition followed by BUG. Generated by: scripts/coccinelle/misc/bugon.cocci Reviewed-by: Christian König <christian.koenig@amd.com> Reviewd-by: Xiangliang.Yu <Xiangliang.Yu@amd.com> CC: Xiangliang Yu <Xiangliang.Yu@amd.com> Signed-off-by: Julia Lawall <julia.lawall@lip6.fr> Signed-off-by: Fengguang Wu <fengguang.wu@intel.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-01-27drm/amdgpu/gfx8: implement emit_rreg/wreg functionXiangliang Yu1-0/+37
Implement emit_rreg/wreg function for kiq ring. Signed-off-by: Xiangliang Yu <Xiangliang.Yu@amd.com> Signed-off-by: Monk Liu <Monk.Liu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-01-27drm/amdgpu: add get clockgating_state method for gfx v8Huang Rui1-0/+40
Signed-off-by: Huang Rui <ray.huang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-01-27drm/amdgpu/gfx8: add support kernel interface queue(KIQ)Xiangliang Yu1-2/+605
KIQ is queue-memory based initialization method: setup KIQ queue firstly, then send command to KIQ to setup other queues, without accessing registers. For virtualization, need KIQ to access virtual function registers when running on guest mode. V2: use amdgpu_bo_create/free_kernel to allocate BO. Signed-off-by: Monk Liu <Monk.Liu@amd.com> Signed-off-by: Xiangliang Yu <Xiangliang.Yu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-01-27drm/amdgpu: use same enter/exit safe mode for gfx_8.Rex Zhu1-94/+1
Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-01-27drm/amdgpu: refine gfx_v8 pg code.Rex Zhu1-11/+12
move en/disable GFX CP/SMU_HS PG to function gfx_v8_0_set_powergating_state Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-01-06drm/amd/amdgpu: add Polaris12 support (v3)Junwei Zhang1-5/+26
v2: agd: squash in various fixes v3: agd: squash in: drm/amdgpu: remove unnecessary smc sk firmware for polaris12 Signed-off-by: Junwei Zhang <Jerry.Zhang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Reviewed-by: Ken Wang <Qingqing.Wang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-12-15drm/amdgpu: refine set clock gating for tonga/polarisRex Zhu1-43/+122
Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-12-15drm/amdgpu: fix pg can't be disabled by PG mask.Rex Zhu1-5/+2
Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
generated by cgit v1.2.3 (git 2.39.1) at 2025-04-13 10:38:26 +0000