aboutsummaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
AgeCommit message (Collapse)AuthorFilesLines
2018-06-13drm/amdgpu: fix clear_all and replace handling in the VM (v2)Junwei Zhang1-2/+4
v2: assign bo_va as well We need to put the lose ends on the invalid list because it is possible that we need to split up huge pages for them. Cc: stable@vger.kernel.org Signed-off-by: Christian König <christian.koenig@amd.com> Signed-off-by: Junwei Zhang <Jerry.Zhang@amd.com> (v2) Reviewed-by: David Zhou <david1.zhou@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-05-24drm/amdgpu: move VM BOs on LRU againChristian König1-5/+23
Move all BOs belonging to a VM on the LRU with every submission. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-05-24drm/amdgpu: consistenly use VM moved flagChristian König1-13/+15
Instead of sometimes checking if the vm_status is empty use the moved flag and also reset it when the BO leaves the state machine. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com> Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-05-24drm/amdgpu: kmap PDs/PTs in amdgpu_vm_update_directoriesChristian König1-7/+8
In theory it is possible that PDs/PTs can move without eviction. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com> Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-05-24drm/amdgpu: further optimize amdgpu_vm_handle_movedChristian König1-13/+14
Splice the moved list to a local one to avoid taking the lock over and over again. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com> Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-05-24drm/amdgpu: cleanup amdgpu_vm_validate_pt_bos v2Christian König1-12/+7
Use list_for_each_entry_safe here. v2: Drop the optimization, it doesn't work as expected. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com> Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-05-24drm/amdgpu: rework VM state machine lock handling v2Christian König1-46/+20
Only the moved state needs a separate spin lock protection. All other states are protected by reserving the VM anyway. v2: fix some more incorrect cases Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
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 <nayan26deshmukh@gmail.com> Suggested-by: Christian König <christian.koenig@amd.com> Acked-by: Lucas Stach <l.stach@pengutronix.de> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-05-15drm/amdgpu: invalidate parent bo when shadow bo was invalidatedChunming Zhou1-0/+4
Shadow BO is located on GTT and its parent (PT and PD) BO could located on VRAM. In some case, the BO on GTT could be evicted but the parent did not. This may cause the shadow BO not be put in the evict list and could not be invalidate correctly. v2: suggested by Christian Signed-off-by: Chunming Zhou <david1.zhou@amd.com> Reported-by: Shaoyun Liu <Shaoyun.Liu@amd.com> Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-05-15drm/amdgpu: abstract bo_base init functionChunming Zhou1-35/+38
Signed-off-by: Chunming Zhou <david1.zhou@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-05-15drm/amdgpu: bo could be null when access in vm bo updateJunwei Zhang1-3/+2
Signed-off-by: Junwei Zhang <Jerry.Zhang@amd.com> Reviewed-by: David Zhou <david1.zhou@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-05-15drm/amdgpu: simplify bo_va list when vm bo update (v2)Junwei Zhang1-12/+10
v2: fix compiling warning 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>
2018-05-15drm/amdgpu: fix list not initializedChunming Zhou1-2/+1
Otherwise, cpu stuck for 22s with kernel panic. Signed-off-by: Chunming Zhou <david1.zhou@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-05-15drm/amdgpu: optionally do a writeback but don't invalidate TC for IB fencesMarek Olšák1-1/+1
There is a new IB flag that enables this new behavior. Full invalidation is unnecessary for RELEASE_MEM and doesn't make sense when draw calls from two adjacent gfx IBs run in parallel. This will be the new default for Mesa. v2: bump the version Signed-off-by: Marek Olšák <marek.olsak@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-05-15drm/amdgpu: use amdgpu_bo_param for amdgpu_bo_create v2Chunming Zhou1-7/+19
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 <david1.zhou@amd.com> Reviewed-by: Huang Rui <ray.huang@amd.com> (v1) Reviewed-by: Christian König <christian.koenig@amd.com> Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com> Cc: christian.koenig@amd.com Cc: Felix.Kuehling@amd.com Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-05-15drm/amdgpu: re-validate per VM BOs if required v2Christian König1-1/+14
If a per VM BO ends up in a allowed domain it never moves back into the prefered domain. v2: move the extra handling into amdgpu_vm_bo_update when we exit the state machine. Make memory type handling generic. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-03-23drm/amdgpu: Fix acquiring VM on large-BAR systemsFelix Kuehling1-2/+1
On large-BAR systems the VM page tables for compute are accessed by the CPU. Always allow CPU access to the page directory so that it can be used later by the CPU when a VM is converted to a compute VM. Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
2018-03-15drm/amdgpu: Add kfd2kgd interface to acquire an existing VMFelix Kuehling1-0/+3
This allows acquiring an existing VM from a render node FD to use it for a compute process. Such VMs get destroyed when the original file descriptor is released. Added a callback from amdgpu_vm_fini to handle KFD VM destruction correctly in this case. v2: * Removed vm->vm_context check in amdgpu_amdkfd_gpuvm_destroy_cb, check vm->process_info earlier instead Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
2018-03-15drm/amdgpu: Add helper to turn an existing VM into a compute VMFelix Kuehling1-0/+67
v2: Removed updating and checking of vm->vm_context v3: Enable amdgpu_vm_clear_bo in amdgpu_vm_make_compute Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
2018-03-14drm/amdgpu: explicit give BO type to amdgpu_bo_createChristian König1-4/+4
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 <christian.koenig@amd.com> Reviewed-by: Roger He <Hongbo.He@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-02-19drm/amdgpu: update the PASID mapping only on demandChristian König1-9/+28
Updating the PASID is rather heavyweight and shouldn't be done all the time. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-02-19drm/amdgpu: separate PASID mapping from VM flush v2Christian König1-2/+5
Stuffing the PASID mapping into the VM flush isn't flexible enough since the PASID mapping changes not as often as we need a VM flush. v2: add missing use of gmc_v7_0_emit_pasid_mapping Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-02-19drm/amdgpu: sync the VM PD/PT before clearing itChristian König1-0/+5
Otherwise we might overwrite stuff which is still in use. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-02-19drm/amdgpu: clear the shadow fence as wellChristian König1-0/+5
It also needs to be initialized. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-02-19drm/amdgpu: release the VM shadow in the error path as wellChristian König1-0/+2
Without it we run into a memory leak. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-02-19drm/amdgpu: fill only the lower range with ATS entries v2Christian König1-29/+53
At least on x86-64 the upper range is purely used by the kernel, avoid creating any ATS mappings there as security precaution and to allow proper page fault reporting in the upper range. v2: remove unused variable Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-02-19drm/amdgpu: revert "Add support for filling a buffer with 64 bit value"Christian König1-3/+2
This reverts commit 7bdc53f925af085ffa0580f10489f82b36cc2f1c and commit 330df03b3abf944f8f5180f2abc61367749984c0. Neither are needed any more. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-02-19drm/amdgpu: revert "Add a parameter to amdgpu_bo_create()"Christian König1-3/+2
This reverts commit 2046d46db9166bddc84778f0b3477f6d1e9068ea. Not needed any more. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-02-19drm/amdgpu: drop root shadow syncChristian König1-5/+0
Completely pointless, it is the same reservation object as the root PD anyway. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Roger He <Hongbo.He@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-02-19drm/amdgpu: revert "drm/amdgpu: use AMDGPU_GEM_CREATE_VRAM_CLEARED for VM ↵Christian König1-24/+92
PD/PTs" v2 Using the standard clear turned out to be to inflexible. First of all it is executed on the system queue, together with buffer moves instead on the per VM queue. And second we need to fill in the page tables with more than just zero. We keep the new functionality of initializing the PDEs/PTEs with ATC routing entries intact. v2: update commit message. Signed-off-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-02-19drm/amdgpu: add optional ring to *_hdp callbacksChristian König1-2/+2
This adds an optional ring to the invalidate_hdp and flush_hdp callbacks. If the ring isn't specified or the emit_wreg function not available the HDP operation will be done with the CPU otherwise by writing on the ring. Signed-off-by: Christian König <christian.koenig@amd.com> Acked-by: Chunming Zhou <david1.zhou@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-02-19drm/amdgpu: forward pasid to backend flush implementationsChristian König1-1/+2
rd the pasid from the VM code to the emit_vm_flush function and update all implementations with the new parameter. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-02-19drm/amdgpu: move PD/PT address calculation into backend functionChristian König1-50/+29
This way we can better handle the differences for CPU based updates. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-02-19drm/amdgpu: move struct gart_funcs into amdgpu_gmc.hChristian König1-9/+8
And rename it to struct gmc_funcs. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Samuel Li <Samuel.Li@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-02-19drm/amdgpu: move struct amdgpu_mc into amdgpu_gmc.hChristian König1-1/+1
And rename it to amdgpu_gmc as well. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Samuel Li <Samuel.Li@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-02-19drm/amdgpu: drop extra tlb invalidation in gpuvmAlex Deucher1-2/+0
We only need to flush the HDP here, not invalidate the TLB. Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-02-19drm/amdgpu: adjust HDP write queue flushing for tlb invalidationAlex Deucher1-0/+2
Separate tlb invalidation and hdp flushing and move the HDP flush to the caller. Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-01-29drm/amdgpu: fix another potential cause of VM faultsChristian König1-20/+12
The root PD can be evicted directly after allocating it, just validate it on first use. Signed-off-by: Christian König <christian.koenig@amd.com> Acked-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-01-16drm/amdgpu: fix CPU based VM updatesChristian König1-5/+9
That got accidentially removed. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-01-16drm/amdgpu: fix amdgpu_vm_pasid_fault_creditChristian König1-3/+7
As soon as the lock is dropped the VM pointer can be invalid. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-01-10drm/amdgpu: optimize moved handling only when vm_debug is inactiveChristian König1-1/+1
Otherwise we would completely circumvent that debugging feature. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Roger He <Hongbo.He@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-01-10drm/amdgpu: simplify huge page handlingChristian König1-42/+18
Update the PDEs after resetting the huge flag. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com> Acked-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-01-10drm/amdgpu: minor optimize VM moved handling v2Christian König1-1/+14
Try to lock moved BOs if it's successful we can update the PTEs directly to the new location. v2: rebase Signed-off-by: Christian König <christian.koenig@amd.com> Acked-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-01-10drm/amdgpu: loosen the criteria for huge pages a bitChristian König1-5/+2
We can actually handle invalid huge pages perfectly fine now. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com> Acked-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-30drm/amdgpu: Correct the IB size of bo update mapping.Emily Deng1-2/+8
The amdgpu_vm_frag_ptes will call amdgpu_vm_update_ptes, and for buffer object that has shadow buffer, need twice commands. Signed-off-by: Emily Deng <Emily.Deng@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-27drm/amdgpu: fix VM faults with per VM BOsChristian König1-2/+20
There was a small window between unreserve and second reserve where the freshly allocated BO could have been evicted without the VM noticing it. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-27drm/amdgpu: drop client_id from VMChristian König1-2/+0
Use the fence context from the scheduler entity. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-27drm/amdgpu: rename vm_id to vmidChristian König1-7/+7
sed -i "s/vm_id/vmid/g" drivers/gpu/drm/amd/amdgpu/*.c sed -i "s/vm_id/vmid/g" drivers/gpu/drm/amd/amdgpu/*.h Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-27drm/amdgpu: separate VMID and PASID handlingChristian König1-410/+12
Move both into the new files amdgpu_ids.[ch]. No functional change. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-18drm/amdgpu: rename ip block helper functionsAlex Deucher1-1/+1
add device to the name for consistency. Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Powered by cgit, Gruvboxed by Blaster4385.