Am 2021-04-22 um 8:35 a.m. schrieb Nirmoy Das:
> Fix vm->is_compute_context initialization in amdgpu_vm_init().
>
> Signed-off-by: Nirmoy Das <nirmoy.das@xxxxxxx>
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 6 +++---
> 1 file changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
> index f95bcda8463f..6f0a6011cb3d 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
> @@ -2815,9 +2815,9 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm,
> goto error_free_immediate;
>
> vm->pte_support_ats = false;
> - vm->is_compute_context = false;
> + vm->is_compute_context = vm_context == AMDGPU_VM_CONTEXT_COMPUTE;
>
> - if (vm_context == AMDGPU_VM_CONTEXT_COMPUTE) {
> + if (vm->is_compute_context) {
A weak ago or so, I submitted a patch that removed the last call to
amdgpu_vm_init with vm_context == AMDGPU_VM_CONTEXT_COMPUTE. We could
probably clean this up now and remove the vm_context parameter and the
AMDGPU_VM_CONTEXT* definitions. The only way to get a compute VM now is
through amdgpu_vm_make_compute.
Regards,
Felix
> vm->use_cpu_for_update = !!(adev->vm_manager.vm_update_mode &
> AMDGPU_VM_USE_CPU_FOR_COMPUTE);
>
> @@ -2844,7 +2844,7 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm,
> vm->evicting = false;
>
> amdgpu_vm_bo_param(adev, vm, adev->vm_manager.root_level, false, &bp);
> - if (vm_context == AMDGPU_VM_CONTEXT_COMPUTE)
> + if (vm->is_compute_context)
> bp.flags &= ~AMDGPU_GEM_CREATE_SHADOW;
> r = amdgpu_bo_create(adev, &bp, &root);
> if (r)
_______________________________________________
amd-gfx mailing list
amd-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/amd-gfx
