[PATCH] drm/amdgpu: ignore duplicate BOs again

Alex Deucher alexdeucher at gmail.com
Tue Oct 17 13:03:22 UTC 2023


On Tue, Oct 17, 2023 at 8:43 AM Hamza Mahfooz <hamza.mahfooz at amd.com> wrote:
>
> On 10/17/23 08:10, Christian König wrote:
> > Looks like RADV is actually hitting this.
> >
> > Signed-off-by: Christian König <christian.koenig at amd.com>
> > Fixes: ca6c1e210aa7 ("drm/amdgpu: use the new drm_exec object for CS v3")
>
> Do you think this will fix the following issue?:
> https://gitlab.freedesktop.org/drm/amd/-/issues/2893

No.  That's a known issue with data migration back to VRAM after
suspend.  There are TTM changes required to fix that properly.

Alex


>
> > ---
> >   drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 3 ++-
> >   1 file changed, 2 insertions(+), 1 deletion(-)
> >
> > diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
> > index efdb1c48f431..d93a8961274c 100644
> > --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
> > +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
> > @@ -65,7 +65,8 @@ static int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p,
> >       }
> >
> >       amdgpu_sync_create(&p->sync);
> > -     drm_exec_init(&p->exec, DRM_EXEC_INTERRUPTIBLE_WAIT);
> > +     drm_exec_init(&p->exec, DRM_EXEC_INTERRUPTIBLE_WAIT |
> > +                   DRM_EXEC_IGNORE_DUPLICATES);
> >       return 0;
> >   }
> >
> --
> Hamza
>


More information about the amd-gfx mailing list