diff options
| author |   Christian König <christian.koenig@amd.com> | 2025-09-19 09:27:03 +0200 |
|---|---|---|
| committer |   Alex Deucher <alexander.deucher@amd.com> | 2025-09-25 15:39:00 -0400 |
| commit | 90e09ea4cfd4aaaf07ababa6d8c880035587e7e9 (patch) | |
| tree | eaa937d2799aaaa8a1898bfeda63e4a1488cd44f /drivers/gpu/drm/amd/amdgpu/amdgpu_ids.h | |
| parent | drm/amdgpu: remove leftover from enforcing isolation by VMID (diff) | |
| download | wireguard-linux-90e09ea4cfd4aaaf07ababa6d8c880035587e7e9.tar.xz wireguard-linux-90e09ea4cfd4aaaf07ababa6d8c880035587e7e9.zip | |
drm/amdgpu: revert "rework reserved VMID handling" v2
This reverts commit e44a0fe630c58b0a87d8281f5c1077a3479e5fce.
Initially we used VMID reservation to enforce isolation between
processes. That has now been replaced by proper fence handling.
Both OpenGL, RADV and ROCm developers requested a way to reserve a VMID
for SPM, so restore that approach by reverting back to only allowing a
single process to use the reserved VMID.
Only compile tested for now.
v2: use -ENOENT instead of -EINVAL if VMID is not available
Signed-off-by: Christian König <christian.koenig@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Diffstat (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_ids.h')
| -rw-r--r-- | drivers/gpu/drm/amd/amdgpu/amdgpu_ids.h | 11 |
1 files changed, 5 insertions, 6 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.h index 240fa6751260..b3649cd3af56 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.h @@ -67,8 +67,7 @@ struct amdgpu_vmid_mgr { unsigned num_ids; struct list_head ids_lru; struct amdgpu_vmid ids[AMDGPU_NUM_VMID]; - struct amdgpu_vmid *reserved; - unsigned int reserved_use_count; + bool reserved_vmid; }; int amdgpu_pasid_alloc(unsigned int bits); @@ -79,10 +78,10 @@ void amdgpu_pasid_free_delayed(struct dma_resv *resv, bool amdgpu_vmid_had_gpu_reset(struct amdgpu_device *adev, struct amdgpu_vmid *id); bool amdgpu_vmid_uses_reserved(struct amdgpu_vm *vm, unsigned int vmhub); -int amdgpu_vmid_alloc_reserved(struct amdgpu_device *adev, - unsigned vmhub); -void amdgpu_vmid_free_reserved(struct amdgpu_device *adev, - unsigned vmhub); +int amdgpu_vmid_alloc_reserved(struct amdgpu_device *adev, struct amdgpu_vm *vm, + unsigned vmhub); +void amdgpu_vmid_free_reserved(struct amdgpu_device *adev, struct amdgpu_vm *vm, + unsigned vmhub); int amdgpu_vmid_grab(struct amdgpu_vm *vm, struct amdgpu_ring *ring, struct amdgpu_job *job, struct dma_fence **fence); void amdgpu_vmid_reset(struct amdgpu_device *adev, unsigned vmhub, |