drm/amdgpu: don't reemit ring contents more than once
authorAlex Deucher <alexander.deucher@amd.com>
Thu, 13 Nov 2025 18:24:10 +0000 (13:24 -0500)
committerAlex Deucher <alexander.deucher@amd.com>
Mon, 5 Jan 2026 22:28:45 +0000 (17:28 -0500)
commit9fc27cbabee6d2e63b0268ca709ad3129b3ac50d
treecb9af6f4265bb547c068b522a25c667b8060fe21
parentdc8a887de1a7d397ab4131f45676e89565417aa8
drm/amdgpu: don't reemit ring contents more than once

If we cancel a bad job and reemit the ring contents, and
we get another timeout, cancel everything rather than reemitting.
The wptr markers are only relevant for the original emit.  If
we reemit, the wptr markers are no longer correct.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit fb62a2067ca4555a6572d911e05919a311c010aa)
drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h