aboutsummaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
AgeCommit message (Collapse)AuthorFilesLines
2018-05-18drm/amd/amdgpu: Code comments for the amdgpu_ttm.c driver. (v2)Tom St Denis1-7/+341
NFC just comments. (v2): Updated based on feedback from Alex Deucher. Signed-off-by: Tom St Denis <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-05-16Merge branch 'drm-next-4.18' of git://people.freedesktop.org/~agd5f/linux ↵Dave Airlie1-39/+71
into drm-next Main changes for 4.18. I'd like to do a separate pull for vega20 later this week or next. Highlights: - Reserve pre-OS scanout buffer during init for seemless transition from console to driver - VEGAM support - Improved GPU scheduler documentation - Initial gfxoff support for raven - SR-IOV fixes - Default to non-AGP on PowerPC for radeon - Fine grained clock voltage control for vega10 - Power profiles for vega10 - Further clean up of powerplay/driver interface - Underlay fixes - Display link bw updates - Gamma fixes - Scatter/Gather display support on CZ/ST - Misc bug fixes and clean ups [airlied: fixup v3d vs scheduler API change] Link: https://patchwork.freedesktop.org/patch/msgid/[email protected] Signed-off-by: Dave Airlie <[email protected]>
2018-05-15drm/scheduler: remove unused parameterNayan Deshmukh1-1/+1
this patch also effect the amdgpu and etnaviv drivers which use the function drm_sched_entity_init Signed-off-by: Nayan Deshmukh <[email protected]> Suggested-by: Christian König <[email protected]> Acked-by: Lucas Stach <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-05-15drm/amdgpu: Add support to change mtype for 2nd part of gart BOs on GFX9Yong Zhao1-10/+44
This change prepares for a workaround in amdkfd for a GFX9 HW bug. It requires the control stack memory of compute queues, which is allocated from the second page of MQD gart BOs, to have mtype NC, rather than the default UC. Signed-off-by: Yong Zhao <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-05-15drm/amdgpu: use amdgpu_bo_param for amdgpu_bo_create v2Chunming Zhou1-5/+10
After that, we can easily add new parameter when need. v2: a) rebase. b) Initialize struct amdgpu_bo_param, future new member could only be used in some one case, but all member should have its own initial value. Signed-off-by: Chunming Zhou <[email protected]> Reviewed-by: Huang Rui <[email protected]> (v1) Reviewed-by: Christian König <[email protected]> Reviewed-by: Junwei Zhang <[email protected]> Cc: [email protected] Cc: [email protected] Signed-off-by: Alex Deucher <[email protected]>
2018-05-15drm/amdgpu: Free VGA stolen memory as soon as possible.Andrey Grodzovsky1-1/+5
Reserved VRAM is used to avoid overriding pre OS FB. Once our display stack takes over we don't need the reserved VRAM anymore. v2: Remove comment, we know actually why we need to reserve the stolen VRAM. Fix return type for amdgpu_ttm_late_init. v3: Return 0 in amdgpu_bo_late_init, rebase on changes to previous patch v4: rebase v5: For GMC9 reserve always just 9M and keep the stolem memory around until GART table curruption on S3 resume is resolved. Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Andrey Grodzovsky <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-05-15drm/amdgpu/gmc: steal the appropriate amount of vram for fw hand-over (v3)Alex Deucher1-6/+8
Steal 9 MB for vga emulation and fb if vga is enabled, otherwise, steal enough to cover the current display size as set by the vbios. If no memory is used (e.g., secondary or headless card), skip stolen memory reserve. v2: skip reservation if vram is limited, address Christian's comments v3: squash in fix from Harry Reviewed-and-Tested-by: Andrey Grodzovsky <[email protected]> (v2) Signed-off-by: Alex Deucher <[email protected]>
2018-05-15drm/amdgpu: fix and cleanup cpu visible VRAM handlingChristian König1-16/+3
The detection if a BO was placed in CPU visible VRAM was incorrect. Fix it and merge it with the correct detection in amdgpu_ttm.c Signed-off-by: Christian König <[email protected]> Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Chunming Zhou <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-03-23drm/amdgpu: Enable amdgpu_ttm_tt_get_user_pages in worker threadsFelix Kuehling1-9/+29
This commit allows amdgpu_ttm_tt_get_user_pages to work in a worker thread rather than regular process context. This will be used when KFD userptr BOs are restored after an MMU-notifier eviction. v2: Manage task reference with get_task_struct/put_task_struct Signed-off-by: Felix Kuehling <[email protected]> Acked-by: Oded Gabbay <[email protected]> Acked-by: Alex Deucher <[email protected]> Signed-off-by: Oded Gabbay <[email protected]>
2018-03-20drm/amd/amdgpu: fix offset into page with amdgpu_iomem debugfs fileTom St Denis1-2/+2
The offset inside the page wasn't included in the copy call meaning the start of the page was being read/written instead. Reported-by: Jay Cornwall <[email protected]> Signed-off-by: Tom St Denis <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-03-14drm/amdgpu: explicit give BO type to amdgpu_bo_createChristian König1-5/+6
Drop the "kernel" and sg parameter and give the BO type to create explicit to amdgpu_bo_create instead of figuring it out from the parameters. Signed-off-by: Christian König <[email protected]> Reviewed-by: Roger He <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-03-14drm/amdgpu: drop the backing store when DMA-buf imports are evictedChristian König1-0/+6
Instead of moving this to the SYSTEM domain just drop the backing store and let the resulting allocation be freed. Signed-off-by: Christian König <[email protected]> Acked-by: Roger He <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-03-14drm/ttm: add bo as parameter to the ttm_tt_create callbackChristian König1-4/+4
Instead of calculating the size in bytes just to recalculate the number of pages from it pass the BO directly to the function. Signed-off-by: Christian König <[email protected]> Reviewed-by: Roger He <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-03-14drm/amdgpu: stop allocating a page array for prime shared BOsChristian König1-2/+3
We don't need the page array for prime shared BOs, stop allocating it. Signed-off-by: Christian König <[email protected]> Reviewed-by: Roger He <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-02-06drm/amdgpu: Add GPUVM memory management functions for KFDFelix Kuehling1-0/+7
v2: * Removed unused flags from struct kgd_mem * Updated some comments * Added a check to unmap_memory_from_gpu whether BO was mapped v3: add mutex_destroy in relevant places Signed-off-by: Felix Kuehling <[email protected]> Acked-by: Oded Gabbay <[email protected]> Signed-off-by: Oded Gabbay <[email protected]>
2018-02-06drm/amdgpu: Add KFD eviction fenceFelix Kuehling1-0/+18
This fence is used by KFD to keep memory resident while user mode queues are enabled. Trying to evict memory will trigger the enable_signaling callback, which starts a KFD eviction, which involves preempting user mode queues before signaling the fence. There is one such fence per process. v2: * Grab a reference to mm_struct * Dereference fence after NULL check * Simplify fence release, no need to signal without anyone waiting * Added signed-off-by Harish, who is the original author of this code v3: * update MAINTAINERS file * change amd_kfd_ prefix to amdkfd_ * remove useless initialization of variable to NULL v4: * set amdkfd_fence_ops to be static * Suggested by: Fengguang Wu <[email protected]> Signed-off-by: Harish Kasiviswanathan <[email protected]> Signed-off-by: Felix Kuehling <[email protected]> Reviewed-by: Oded Gabbay <[email protected]> Signed-off-by: Oded Gabbay <[email protected]>
2018-03-07drm/amdgpu: replace iova debugfs file with iomem (v3)Tom St Denis1-21/+81
This allows access to pages allocated through the driver with optional IOMMU mapping. v2: Fix number of bytes copied and add write method v3: drop check for kmap return Original-by: Christian König <[email protected]> Signed-off-by: Tom St Denis <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-03-07drm/amdgpu: drop gtt->adevChristian König1-5/+5
We can use ttm->bdev instead. Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-03-05drm/amdgpu: use separate status for buffer funcs availability v2Christian König1-10/+11
The ring status can change during GPU reset, but we still need to be able to schedule TTM buffer moves in the meantime. Otherwise we can ran into problems because of aborted move/fill operations during GPU resets. v2: still check if ring is available during direct submit. Signed-off-by: Christian König <[email protected]> Acked-by: Chunming zhou <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-03-05drm/amdgpu: ignore changes of buffer function status because of GPU resetsChristian König1-1/+1
When we reset the GPU we also disable/enable the SDMA, but we don't want to change TTM idea of the VRAM size in the middle of that. Signed-off-by: Christian König <[email protected]> Acked-by: Chunming zhou <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-03-05drm/amdgpu: change amdgpu_ttm_set_active_vram_sizeChristian König1-6/+17
Instead of setting the active VRAM size directly provide a the info if we can use the buffer functions or not. Signed-off-by: Christian König <[email protected]> Acked-by: Chunming zhou <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-02-28drm/amdgpu: Map all visible VRAM at startupAmber Lin1-0/+17
When using CPU to update page table, we need to kmap all the PDs/PTs after they are allocated and that requires a TLB shot down on each CPU, which is quite heavy. Instead, we map the whole visible VRAM to a kernel address at once. Pages can be obtained from the offset. v2: move the mapping base from gmc to amdgpu_mman structure, and the implementation in amdgpu_ttm_* functions Signed-off-by: Amber Lin <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-02-26drm/ttm: drop ttm->dummy_read_pageChristian König1-3/+2
Only used by the AGP backend and there it can be easily accessed using ttm->bdev->glob. Signed-off-by: Christian König <[email protected]> Reviewed-by: Michel Dänzer <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-02-19drm/amdgpu: remove extra TT unpopulated checkChristian König1-3/+0
The subsystem should check that, not the driver. Signed-off-by: Christian König <[email protected]> Reviewed-by: Roger He <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-02-19drm/amdgpu: revert "Add support for filling a buffer with 64 bit value"Christian König1-12/+5
This reverts commit 7bdc53f925af085ffa0580f10489f82b36cc2f1c and commit 330df03b3abf944f8f5180f2abc61367749984c0. Neither are needed any more. Signed-off-by: Christian König <[email protected]> Reviewed-by: Felix Kuehling <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-02-19drm/amdgpu: revert "Add a parameter to amdgpu_bo_create()"Christian König1-1/+1
This reverts commit 2046d46db9166bddc84778f0b3477f6d1e9068ea. Not needed any more. Signed-off-by: Christian König <[email protected]> Reviewed-by: Felix Kuehling <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-02-19drm/amdgpu: Use new TTM flag to avoid OOM triggering.Andrey Grodzovsky1-0/+4
Avoid OOM on syatem pages allocations. v2: Remove modeprobe parameter, make this behaviour the only option. v3: Move setting no_retry flag into amdgpu_ttm_init. Signed-off-by: Andrey Grodzovsky <[email protected]> Reviewed-by: Christian König <[email protected]> Reviewed-by: Felix Kuehling <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-02-19drm/amdgpu: move struct amdgpu_mc into amdgpu_gmc.hChristian König1-23/+23
And rename it to amdgpu_gmc as well. Signed-off-by: Christian König <[email protected]> Reviewed-by: Samuel Li <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-02-13drm/amdgpu: only enable swiotlb alloc when need v2Chunming Zhou1-3/+3
get the max io mapping address of system memory to see if it is over our card accessing range. v2: move checking later Signed-off-by: Chunming Zhou <[email protected]> Reviewed-by: Monk Liu <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2017-12-28drm/ttm: use an operation ctx for ttm_tt_bindRoger He1-1/+1
forward the operation context to ttm_tt_bind as well, and the ultimate goal is swapout enablement for reserved BOs. v2: use common term rather than amd specific Reviewed-by: Thomas Hellström <[email protected]> Reviewed-by: Christian König <[email protected]> Reviewed-by: Chuming Zhou <[email protected]> Signed-off-by: Roger He <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-28drm/ttm: use an operation ctx for ttm_tt_populate in ttm_bo_driver (v2)Roger He1-3/+4
forward the operation context to ttm_tt_populate as well, and the ultimate goal is swapout enablement for reserved BOs. v2: squash in fix for vboxvideo Reviewed-by: Christian König <[email protected]> Signed-off-by: Roger He <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-18Revert "drm/amd/amdgpu: set gtt size according to system memory size only"Andrey Grodzovsky1-3/+5
This reverts commit ba851eed895c76be0eb4260bdbeb7e26f9ccfaa2. With that change piglit max size tests (running with -t max.*size) are causing OOM and hard hang on my CZ with 1GB RAM. Signed-off-by: Andrey Grodzovsky <[email protected]> Acked-by: Alex Deucher <[email protected]> Reviewed-by: Christian König <[email protected]> Reviewed-by: Roger He <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-18drm/amdgpu: move fw_reserve functions to amdgpu_ttm.cAlex Deucher1-2/+97
It's the only place they are used. Acked-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-15drm/ttm: use an ttm operation ctx for ttm_bo_move_xxxRoger He1-4/+3
include ttm_bo_move_memcpy and ttm_bo_move_ttm Signed-off-by: Roger He <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-07drm: move amd_gpu_scheduler into common locationLucas Stach1-4/+4
This moves and renames the AMDGPU scheduler to a common location in DRM in order to facilitate re-use by other drivers. This is mostly a straight forward rename with no code changes. One notable exception is the function to_drm_sched_fence(), which is no longer a inline header function to avoid the need to export the drm_sched_fence_ops_scheduled and drm_sched_fence_ops_finished structures. Reviewed-by: Chunming Zhou <[email protected]> Tested-by: Dieter Nützel <[email protected]> Acked-by: Alex Deucher <[email protected]> Signed-off-by: Lucas Stach <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-06drm/amd/amdgpu: set gtt size according to system memory size onlyRoger He1-5/+3
Reviewed-by: Christian König <[email protected]> Signed-off-by: Roger He <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-06drm/amdgpu: forward operation context to ttm_bo_mem_spaceChristian König1-18/+12
This way we can finally use some more stats. Signed-off-by: Christian König <[email protected]> Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Chunming Zhou <[email protected]> Tested-by: Dieter Nützel <[email protected]> Tested-by: Michel Dänzer <[email protected]> Acked-by: Felix Kuehling <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-06drm/ttm: add context to driver move callback as wellChristian König1-10/+11
Instead of passing the parameters manually. Signed-off-by: Christian König <[email protected]> Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Chunming Zhou <[email protected]> Tested-by: Dieter Nützel <[email protected]> Tested-by: Michel Dänzer <[email protected]> Acked-by: Felix Kuehling <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-06drm/ttm: use an operation context for ttm_bo_mem_space v2Christian König1-5/+6
Instead of specifying interruptible and no_wait_gpu manually. v2: rebase Signed-off-by: Christian König <[email protected]> Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Chunming Zhou <[email protected]> Tested-by: Dieter Nützel <[email protected]> Tested-by: Michel Dänzer <[email protected]> Acked-by: Felix Kuehling <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-06drm/amdgpu:fix NULL pointer access during drv removeMonk Liu1-1/+1
NULL pointer is because original logic will step into set_pde_pte() even after the gart.ptr is freed due to there are twice gart_unbind() on all gart area. also, there are other minor fixes: 1,since gart_init only create dummy page, the corresponding gart_fini shouldn't do more like unbinding all GART, this is unnecessary because in driver fini stage all GART unbinding had already been done during each IP's SW_FINI (GMC's SW_FINI is the last one called), so remove the step for the GART unbinding in gart_fini(). 2,gart_fini() is already invoked during each GMC IP's gart_fini routine,e.g. gmc_vx_0_gart_fini(), so no need to manually call it during ttm_fini(). 3,amdgpu_gem_force_release() should be put ahead of amdgpu_vm_manager_fini() Signed-off-by: Monk Liu <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-06drm/amdgpu:cleanup GMC & gart garbage functionMonk Liu1-2/+0
for gart_ram_alloc/free, they are never used in driver thus ripe them out totally. for gart_vram_pin/unpin, they are not needed becuase we can use bo_creat_kernel/free to replace the original manual way in the gart_vram_alloc/free, thus gart_vram_pin/unpin can also be riped out. Signed-off-by: Monk Liu <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-06drm/amdgpu:cleanup stolen vga memory finishMonk Liu1-8/+3
Signed-off-by: Monk Liu <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-06drm/amd/amdgpu: not allow gtt size exceed 75%*system memory sizeRoger He1-3/+8
keep consistency with threshold of swapout Reviewed-by: Christian König <[email protected]> Signed-off-by: Roger He <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-04drm/amdgpu: don't use ttm_bo_move_ttm in amdgpu_ttm_bind v2Christian König1-6/+15
Just allocate the GART space and fill it. This prevents forcing the BO to be idle. v2: don't unbind/bind at all, just fill the allocated GART space Signed-off-by: Christian König <[email protected]> Acked-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-04drm/amdgpu: rename amdgpu_ttm_bind to amdgpu_ttm_alloc_gartChristian König1-2/+2
We actually don't bind here, but rather allocate GART space if necessary. Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-04drm/amdgpu: move GART recovery into GTT manager v2Christian König1-36/+15
The GTT manager handles the GART address space anyway, so it is completely pointless to keep the same information around twice. v2: rebased Signed-off-by: Christian König <[email protected]> Reviewed-by: Chunming Zhou <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-04drm/amdgpu: nuke amdgpu_ttm_is_bound() v2Christian König1-15/+9
Rename amdgpu_gtt_mgr_is_allocated() to amdgpu_gtt_mgr_has_gart_addr() and use that instead. v2: rename the function as well. Signed-off-by: Christian König <[email protected]> Reviewed-by: Chunming Zhou <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-04drm/amdgpu: remove extra parameter from amdgpu_ttm_bind() v2Christian König1-2/+2
We always use the BO mem now. v2: minor rebase Signed-off-by: Christian König <[email protected]> Reviewed-by: Michel Dänzer <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-04drm/amdgpu: don't wait interruptible while binding GART spaceChristian König1-1/+1
Display can't seem to handle this correctly. Signed-off-by: Christian König <[email protected]> Acked-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-04drm/amd/scheduler:introduce guilty pointer memberMonk Liu1-1/+1
this member will be used later, it will points to the real var inside of context and CS_SUBMIT & gpu schdduler can decide if skip a job depends on context->guilty or *entity->guilty Signed-off-by: Monk Liu <[email protected]> Reviewed-by: Chunming Zhou <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
Powered by cgit, Gruvboxed by Blaster4385.