drm/amdkfd: Remove skiping userptr buffer mapping when mmu notifier marks it as invalid
Subsystems (3)
PATCH 1/1 drm/amdkfd: Remove skiping userptr buffer mapping when mmu notifier marks it as invalid
From: Xiaogang Chen <[email protected]>
mmu notifier does not always hold mm->sem during call back. That causes a race condition between kfd userprt buffer mapping and mmu notifier which leds to gpu shadder or SDMA access userptr buffer before it has been mapped to gpu VM. Always map userptr buffer to avoid that though it may make some userprt buffers mapped two times.
--- drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c | 10 ---------- 1 file changed, 10 deletions(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c +0 -10
@@ -1907,16 +1907,6 @@ int amdgpu_amdkfd_gpuvm_map_memory_to_gpu(*/ mutex_lock (&mem->process_info->lock); -/* Lock mmap-sem. If we find an invalid userptr BO, we can be -* sure that the MMU notifier is no longer running -* concurrently and the queues are actually stopped -*/ -if (amdgpu_ttm_tt_get_usermm (bo->tbo.ttm)) { -mmap_write_lock (current->mm); - is_invalid_userptr =atomic_read (&mem->invalid); -mmap_write_unlock (current->mm); - } -mutex_lock (&mem->lock); domain = mem->domain;
Signature · 1 line
2.25.1
Am 2022-10-28 um 20:26 schrieb Xiaogang.Chen:
Quoted 9 lines
From: Xiaogang Chen <[email protected]>
mmu notifier does not always hold
mm->semduring call back. That causes a race condition between kfd userprt buffer mapping and mmu notifier which leds to gpu shadder or SDMA access userptr buffer before it has been mapped to gpu VM. Always map userptr buffer to avoid that though it may make some userprt buffers mapped two times.Suggested-by Felix Kuehling <[email protected]>Signed-off-by Xiaogang Chen <[email protected]>
Quoted 25 lines
--- drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c | 10 ---------- 1 file changed, 10 deletions(-)drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c +0
-10@@ -1907,16 +1907,6 @@ int amdgpu_amdkfd_gpuvm_map_memory_to_gpu(*/ mutex_lock (&mem->process_info->lock); -/* Lock mmap-sem. If we find an invalid userptr BO, we can be -* sure that the MMU notifier is no longer running -* concurrently and the queues are actually stopped -*/ -if (amdgpu_ttm_tt_get_usermm (bo->tbo.ttm)) { -mmap_write_lock (current->mm); - is_invalid_userptr =atomic_read (&mem->invalid); -mmap_write_unlock (current->mm); - } -mutex_lock (&mem->lock); domain = mem->domain;