[PATCH] drm/amdgpu: Fix vram recover doesn't work after whole GPU reset
Hamza Mahfooz
hamza.mahfooz at amd.com
Fri May 5 13:19:05 UTC 2023
On 5/4/23 23:46, Lin.Cao wrote:
> v1: Vmbo->shadow is used to back vram bo up when vram lost. So that we
> should set shadow as vmbo->shadow to recover vmbo->bo
> v2: Modify if(vmbo->shadow) shadow = vmbo->shadow as if(!vmbo->shadow)
> continue;
>
> Fix: 'commit e18aaea733da ("drm/amdgpu: move shadow_list to amdgpu_bo_vm")'
Please change the previous line to:
Fixes: e18aaea733da ("drm/amdgpu: move shadow_list to amdgpu_bo_vm")
> Signed-off-by: Lin.Cao <lincao12 at amd.com>
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 7 ++++++-
> 1 file changed, 6 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> index 750eaffa81ba..0581b4fec001 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> @@ -4514,7 +4514,12 @@ static int amdgpu_device_recover_vram(struct amdgpu_device *adev)
> dev_info(adev->dev, "recover vram bo from shadow start\n");
> mutex_lock(&adev->shadow_list_lock);
> list_for_each_entry(vmbo, &adev->shadow_list, shadow_list) {
> - shadow = &vmbo->bo;
> +
> + /* If vm is compute context or adev is APU, shadow will be NULL */
> + if (!vmbo->shadow)
> + continue;
> + shadow = vmbo->shadow;
> +
> /* No need to recover an evicted BO */
> if (shadow->tbo.resource->mem_type != TTM_PL_TT ||
> shadow->tbo.resource->start == AMDGPU_BO_INVALID_OFFSET ||
--
Hamza
More information about the amd-gfx
mailing list