Am 06.09.23 um 21:55 schrieb Alex Deucher:
Commit 254986e324ad ("drm/radeon: Use the drm suballocation manager implementation.")
made the fence wait in amdgpu_sa_bo_new() interruptible but there is no
code to handle an interrupt. This caused the kernel to randomly explode
in high-VRAM-pressure situations so make it uninterruptible again.
Fixes: 254986e324ad ("drm/radeon: Use the drm suballocation manager implementation.")
Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2769
Signed-off-by: Alex Deucher <alexander.deucher@xxxxxxx>
CC: stable@xxxxxxxxxxxxxxx # 6.4+
CC: Simon Pilkington <simonp.git@xxxxxxxxx>
Reviewed-by: Christian König <christian.koenig@xxxxxxx>
Going to push this to drm-misc-fixes in a minute.
Regards,
Christian.
---
drivers/gpu/drm/radeon/radeon_sa.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/radeon/radeon_sa.c b/drivers/gpu/drm/radeon/radeon_sa.c
index c87a57c9c592..22dd8b445685 100644
--- a/drivers/gpu/drm/radeon/radeon_sa.c
+++ b/drivers/gpu/drm/radeon/radeon_sa.c
@@ -123,7 +123,7 @@ int radeon_sa_bo_new(struct radeon_sa_manager *sa_manager,
unsigned int size, unsigned int align)
{
struct drm_suballoc *sa = drm_suballoc_new(&sa_manager->base, size,
- GFP_KERNEL, true, align);
+ GFP_KERNEL, false, align);
if (IS_ERR(sa)) {
*sa_bo = NULL;