aboutsummaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
AgeCommit message (Collapse)AuthorFilesLines
2018-03-14drm/amdgpu: fix prime teardown orderChristian König1-0/+2
We unmapped imported DMA-bufs when the GEM handle was dropped, not when the hardware was done with the buffere. Signed-off-by: Christian König <[email protected]> Reviewed-by: Michel Dänzer <[email protected]> CC: [email protected] Signed-off-by: Alex Deucher <[email protected]>
2017-12-20drm/amdgpu: fix test for shadow page tablesChristian König1-1/+5
They don't work 100% correctly at the moment. Reviewed-and-Tested-by: Michel Dänzer <[email protected]> Signed-off-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-18drm/amdgpu: move amdgpu_need_backup to amdgpu_object.cAlex Deucher1-0/+8
It's the only place it's used. Acked-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-15drm/amdgpu: set allow_reserved_eviction and resv when bo allocation and csRoger He1-1/+6
enable eviction of other per VM BOs during allocation and allows reaping of deleted BOs during CS. Reviewed-by: Chunming Zhou <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Roger He <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-06drm/amdgpu: use the new TTM bytes moved counter v2Christian König1-7/+3
Instead of the global statistics use the per context bytes moved counter. v2: rebased 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 ctx for ttm_bo_init_reservedChristian König1-1/+2
Instead of specifying if sleeping should be interruptible. 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 operation ctx to ttm_bo_validate v2Christian König1-4/+8
Give moving a BO into place an operation context to work with. v2: rebased 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]>
2017-12-06amdgpu: Don't use DRM_ERROR when failing to allocate a BOMichel Dänzer1-2/+2
This can be triggered by userspace, e.g. trying to allocate too large a BO, so it shouldn't log anything by default. Callers need to handle failure anyway. Reviewed-by: Alex Deucher <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Michel Dänzer <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-06drm/amdgpu: Implement BO size validation V2Andrey Grodzovsky1-0/+41
Validates BO size against each requested domain's total memory. v2: Make GTT size check a MUST to allow fall back to GTT. Rmove redundant NULL check. Signed-off-by: Andrey Grodzovsky <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-04drm/amdgpu: rename amdgpu_ttm_bind to amdgpu_ttm_alloc_gartChristian König1-1/+1
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: nuke amdgpu_ttm_is_bound() v2Christian König1-1/+1
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 König1-1/+1
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: fix pin domain compatibility checkChristian König1-1/+1
We need to test if any domain fits, not all of them. Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-04drm/amdgpu: always bind pinned BOsChristian König1-7/+7
We always need to bind pinned BOs, not just when the caller requested the address. Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-04drm/amdgpu: use the actual placement for pin accountingChristian König1-0/+2
This allows us to specify multiple possible placements again. Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-04drm/amdgpu: change redundant init logs to debug levelpding1-2/+2
When this VF stays in exclusive mode for long, other VFs will be impacted. The redundant messages causes exclusive mode timeout when they're redirected. That is a normal use case for cloud service to redirect guest log to virtual serial port. Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: pding <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-10-31drm/amdgpu: fix error handling in amdgpu_bo_do_createChristian König1-3/+3
The bo structure is freed up in case of an error, so we can't do any accounting if that happens. Signed-off-by: Christian König <[email protected]> Reviewed-by: Michel Dänzer <[email protected]> CC: [email protected] Signed-off-by: Alex Deucher <[email protected]>
2017-10-09drm/amdgpu: add helper to convert a ttm bo to amdgpu_boAndres Rodriguez1-5/+3
Reviewed-by: Christian König <[email protected]> Signed-off-by: Andres Rodriguez <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-09-26drm/amdgpu: simplify pinning into visible VRAMChristian König1-15/+8
Just set the CPU access required flag when we pin it. Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-09-13drm/amdgpu: fix and cleanup amdgpu_bo_create v2Christian König1-60/+23
We adjusted the BO flags for USWC handling, but those never took effect because the placement was passed in instead of generated inside this function. v2: better commit message Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-08-29drm/amdgpu: track evicted page tables v2Christian König1-1/+1
Instead of validating all page tables when one was evicted, track which one needs a validation. v2: simplify amdgpu_vm_ready as well Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]> (v1) Reviewed-by: Chunming Zhou <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-08-29drm/amdgpu: fix and cleanup shadow handlingChristian König1-23/+23
Set the shadow flag on the shadow and not the parent, always bind shadow BOs during allocation instead of manually, use the reservation_object wrappers to grab the lock. This fixes a couple of issues with binding the shadow BOs as well as correctly evicting them when memory becomes tight. Signed-off-by: Christian König <[email protected]> Reviewed-by: Chunming Zhou <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-08-17drm/amdgpu: move vram usage tracking into the vram manager v2Christian König1-50/+0
Looks like a better place for this. v2: use atomic64_t members instead Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-08-17drm/amdgpu: move gtt usage tracking into the gtt manager v2Christian König1-2/+0
It doesn't make much sense to count those numbers twice. v2: use and atomic64_t instead Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-08-15drm/amdgpu: Fix preferred typoKent Russell1-3/+3
Change "prefered" to "preferred" Signed-off-by: Kent Russell <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-08-15drm/amdgpu: Add a parameter to amdgpu_bo_create()Yong Zhao1-3/+9
The parameter init_value contains the value to which we initialized VRAM bo when AMDGPU_GEM_CREATE_VRAM_CLEARED flag is set. Signed-off-by: Yong Zhao <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-08-15drm/amdgpu: add amdgpu_bo_create_reservedChristian König1-8/+40
Same as amdgpu_bo_create_kernel, but keeps the BO reserved. Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-08-15drm/amdgpu: improve amdgpu_bo_create_kernelChristian König1-8/+14
Make allocating the new BO optional. Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-08-15drm/amdgpu: cleanup kptr handlingChristian König1-11/+15
Don't keep around the same pointer twice. Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-07-25drm/amdgpu: ttm_bind only when user needs gpu_addr in bo pinChunming Zhou1-6/+7
Signed-off-by: Chunming Zhou <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-07-14drm/amdgpu: make sure BOs are always kunmappedChristian König1-0/+3
When a BO is moved or destroyed it shouldn't be kmapped any more. Signed-off-by: Christian König <[email protected]> Reviewed-by: Felix Kuehling <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-07-14drm/amdgpu: Don't force BOs into visible VRAM for page faultsJohn Brooks1-7/+10
There is no need for page faults to force BOs into visible VRAM if it's full, and the time it takes to do so is great enough to cause noticeable stuttering. Add GTT as a possible placement so that if visible VRAM is full, page faults move BOs to GTT instead of evicting other BOs from VRAM. Suggested-by: Michel Dänzer <[email protected]> Signed-off-by: John Brooks <[email protected]> Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Michel Dänzer <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-07-14drm/amdgpu: Set/clear CPU_ACCESS flag on page fault and move to VRAMJohn Brooks1-10/+10
When a BO is moved to VRAM, clear AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED. This allows it to potentially later move to invisible VRAM if the CPU does not access it again. Setting the CPU_ACCESS flag in amdgpu_bo_fault_reserve_notify() also means that we can remove the loop to restrict lpfn to the end of visible VRAM, because amdgpu_ttm_placement_init() will do it for us. v3 [Michel Dänzer] * Use AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED instead of a new flag (Christian König) * Clear flag in amdgpu_bo_move instead of amdgpu_move_ram_vram (Christian) * Explicitly mention amdgpu_bo_fault_reserve_notify in amdgpu_bo_move * Also clear flag in amdgpu_bo_create_restricted Suggested-by: Michel Dänzer <[email protected]> Signed-off-by: John Brooks <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Michel Dänzer <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-07-14drm/amdgpu: Throttle visible VRAM moves separatelyJohn Brooks1-3/+9
The BO move throttling code is designed to allow VRAM to fill quickly if it is relatively empty. However, this does not take into account situations where the visible VRAM is smaller than total VRAM, and total VRAM may not be close to full but the visible VRAM segment is under pressure. In such situations, visible VRAM would experience unrestricted swapping and performance would drop. Add a separate counter specifically for moves involving visible VRAM, and check it before moving BOs there. v2: Only perform calculations for separate counter if visible VRAM is smaller than total VRAM. (Michel Dänzer) v3: [Michel Dänzer] * Use BO's location rather than the AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED flag to determine whether to account a move for visible VRAM in most cases. * Use a single if (adev->mc.visible_vram_size < adev->mc.real_vram_size) { block in amdgpu_cs_get_threshold_for_moves. Fixes: 95844d20ae02 (drm/amdgpu: throttle buffer migrations at CS using a fixed MBps limit (v2)) Signed-off-by: John Brooks <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Michel Dänzer <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-07-14drm/amdgpu: remove stale TODO commentChristian König1-1/+0
That is already fixed. Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]>
2017-07-14drm/amdgpu: add vm_needs_flush parameter to amdgpu_copy_bufferChristian König1-2/+2
This allows us to flush the system VM here. Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Acked-by: Felix Kuehling <[email protected]>
2017-05-24drm/amdgpu: add an INFO query for monitoring VRAM CPU page faultsMarek Olšák1-0/+1
v2: bump the DRM version Signed-off-by: Marek Olšák <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-04-28drm/amdgpu: validate shadow before restoring from itRoger.He1-0/+21
Reviewed-by: Chunming Zhou <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Roger.He <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-04-28drm/amdgpu: Fix use of interruptible waitingAlex Xie1-1/+1
1. The signal interrupt can affect the expected behaviour. 2. There is no good mechanism to handle the corresponding error. Signed-off-by: Alex Xie <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-04-07drm/amdgpu: Refuse to pin or change acceptable domains of prime BOs to VRAM. ↵Christopher James Halse Rogers1-0/+4
(v2) Migration to VRAM will break the sharing, resulting in rendering on the exporting GPU never becoming visible on the importing GPU. v2: Don't pin BOs to GTT. Instead, refuse to migrate them out of GTT. Reviewed-by: Christian König <[email protected]> Signed-off-by: Christopher James Halse Rogers <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-04-06drm/amdgpu: handle CPU access for split VRAM buffers (v2)Christian König1-3/+1
This avoids merging them together on page fault. v2: squash in 64-bit division fix Signed-off-by: Christian König <[email protected]> Acked-by: Michel Dänzer <[email protected]> Reviewed-by: Nicolai Hähnle <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-04-04drm/amdgpu: use TTM_PL_FLAG_CONTIGUOUS v2Christian König1-6/+5
Implement AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS using TTM_PL_FLAG_CONTIGUOUS instead of a placement limit. That allows us to better handle CPU accessible placements. v2: prevent virtual BO start address from overflowing Signed-off-by: Christian König <[email protected]> Acked-by: Michel Dänzer <[email protected]> Reviewed-by: Nicolai Hähnle <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-03-29drm/amd/amdgpu: fix performance drop when VRAM pressureRoger.He1-3/+1
When VRAM pressue and trigger huge evictions there is performance drop, this patch fix it. Signed-off-by: Roger.He <[email protected]> Reviewed-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-03-29drm/amdgpu: don't validate TILE_SPLIT on GFX9Marek Olšák1-1/+4
Signed-off-by: Marek Olšák <[email protected]> Acked-by: Alex Deucher <[email protected]> Acked-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-03-29drm/amdgpu: simplify reservation handling during buffer creationNicolai Hähnle1-13/+4
By using ttm_bo_init_reserved instead of the manual initialization of the reservation object, the reservation lock will be properly unlocked and destroyed when the TTM BO initialization fails. Actual deadlocks caused by the missing unlock should have been fixed by "drm/ttm: never add BO that failed to validate to the LRU list", superseding the flawed fix in commit 38fc4856ad98 ("drm/amdgpu: fix a potential deadlock in amdgpu_bo_create_restricted()"). This change fixes remaining recursive locking errors that can be seen with lock debugging enabled, and avoids the error of freeing a locked mutex. As an additional minor bonus, buffers created with resv == NULL and the AMDGPU_GEM_CREATE_VRAM_CLEARED flag are now only added to the global LRU list after the fill commands have been issued. v2: use amdgpu_bo_unreserve instead of ttm_bo_unreserve Fixes: 12a852219583 ("drm/amdgpu: improve AMDGPU_GEM_CREATE_VRAM_CLEARED handling (v2)") Signed-off-by: Nicolai Hähnle <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-03-29Revert "drm/amdgpu: fix a potential deadlock in amdgpu_bo_create_restricted()"Nicolai Hähnle1-4/+1
This reverts commit 38fc4856ad98f230bc91da0421dec69e4aee40f8, which introduces a use-after-free. The underlying bug should be properly fixed with "drm/ttm: never add BO that failed to validate to the LRU list". Cc: zhoucm1 <[email protected]> Signed-off-by: Nicolai Hähnle <[email protected]> Tested-by: Samuel Pitoiset <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-02-09drm/amdgpu: report the number of bytes moved at buffer creationSamuel Pitoiset1-0/+6
Like ttm_bo_validate(), ttm_bo_init() might need to move BO and the number of bytes moved by TTM should be reported. This can help the throttle buffer migration mechanism to make a better decision. v2: fix computation Reviewed-by: Christian König <[email protected]> Signed-off-by: Samuel Pitoiset <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-02-09drm/amdgpu: fix a potential deadlock in amdgpu_bo_create_restricted()Samuel Pitoiset1-1/+4
When ttm_bo_init() fails, the reservation mutex should be unlocked. In debug build, the kernel reported "possible recursive locking detected" in this codepath. For debugging purposes, I also added a "WARN_ON(ww_mutex_is_locked())" when ttm_bo_init() fails and the mutex was locked as expected. This should fix (random) GPU hangs. The easy way to reproduce the issue is to change the "Super Sampling" option from 1.0 to 2.0 in Hitman. It will create a huge buffer, evict a bunch of buffers (around ~5k) and deadlock. This regression has been introduced pretty recently. v2: only release the mutex if resv is NULL Fixes: 12a852219583 ("drm/amdgpu: improve AMDGPU_GEM_CREATE_VRAM_CLEARED handling (v2)") Reviewed-by: Christian König <[email protected]> Signed-off-by: Samuel Pitoiset <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-02-09drm/amdgpu: shut up #warning for compile testingArnd Bergmann1-0/+2
My randconfig tests on linux-next showed a newly introduced warning: drivers/gpu/drm/amd/amdgpu/amdgpu_object.c: In function 'amdgpu_bo_create_restricted': drivers/gpu/drm/amd/amdgpu/amdgpu_object.c:377:2: error: #warning Please enable CONFIG_MTRR and CONFIG_X86_PAT for better performance thanks to write-combining [-Werror=cpp] Generally speaking, warnings about bad kernel configuration are not particularly helpful. We could enforce the selection of X86_PAT through Kconfig, so the driver cannot even be used unless it is enabled, or we could just rely on the runtime warning that is also there. In this version, I'm making the warning conditional on CONFIG_COMPILE_TEST, which shuts it up for me, but not people that may actually want to run the kernel as a compromize. Fixes: a2e2f29970aa ("drm/amdgpu: Bring bo creation in line with radeon driver (v2)") Reviewed-by: Michel Dänzer <[email protected]> Signed-off-by: Arnd Bergmann <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-01-27drm/amdgpu: Bring bo creation in line with radeon driver (v2)Nils Holland1-0/+20
Add the bo creation changes that have been done to the radeon driver in recent times, e.g. disable GTT WC on 32 bit because it is broken there, and also disable it generally (and print a warning message) when CONFIG_X86_PAT is not set. v2: agd: fix warning in defined(CONFIG_X86) && !defined(CONFIG_X86_PAT) case Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Nils Holland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
Powered by cgit, Gruvboxed by Blaster4385.