[Public]
-----Original Message----- From: Greg Kroah-Hartman gregkh@linuxfoundation.org Sent: Wednesday, September 17, 2025 8:35 AM To: stable@vger.kernel.org Cc: Greg Kroah-Hartman gregkh@linuxfoundation.org; patches@lists.linux.dev; cao, lin lin.cao@amd.com; Prosyak, Vitaly Vitaly.Prosyak@amd.com; Koenig, Christian Christian.Koenig@amd.com; Deucher, Alexander Alexander.Deucher@amd.com; Sasha Levin sashal@kernel.org Subject: [PATCH 6.12 140/140] drm/amdgpu: fix a memory leak in fence cleanup when unloading
6.12-stable review patch. If anyone has any objections, please let me know.
From: Alex Deucher alexander.deucher@amd.com
commit 7838fb5f119191403560eca2e23613380c0e425e upstream.
Commit b61badd20b44 ("drm/amdgpu: fix usage slab after free") reordered when amdgpu_fence_driver_sw_fini() was called after that patch, amdgpu_fence_driver_sw_fini() effectively became a no-op as the sched entities we never freed because the ring pointers were already set to NULL. Remove the NULL setting.
Reported-by: Lin.Cao lincao12@amd.com Cc: Vitaly Prosyak vitaly.prosyak@amd.com Cc: Christian König christian.koenig@amd.com Fixes: b61badd20b44 ("drm/amdgpu: fix usage slab after free")
Does 6.12 contain b61badd20b44 or a backport of it? If not, then this patch is not applicable.
Alex
Reviewed-by: Christian König christian.koenig@amd.com Signed-off-by: Alex Deucher alexander.deucher@amd.com (cherry picked from commit a525fa37aac36c4591cc8b07ae8957862415fbd5) Cc: stable@vger.kernel.org [ Adapt to conditional check ] Signed-off-by: Sasha Levin sashal@kernel.org Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org
drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c | 3 --- 1 file changed, 3 deletions(-)
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c @@ -400,9 +400,6 @@ void amdgpu_ring_fini(struct amdgpu_ring dma_fence_put(ring->vmid_wait); ring->vmid_wait = NULL; ring->me = 0;
if (!ring->is_mes_queue)ring->adev->rings[ring->idx] = NULL;}
/**