It was changed to GFP_ATOMIC in commit ec2f0577c (add & use virtio_gpu_queue_fenced_ctrl_buffer) because the allocation happened with a spinlock held. That was no longer true after commit 9fdd90c0f (add virtio_gpu_alloc_fence()).
Signed-off-by: Chia-I Wu olvaffe@gmail.com Cc: Gerd Hoffmann kraxel@redhat.com Cc: Gustavo Padovan gustavo.padovan@collabora.com Cc: Robert Foss robert.foss@collabora.com --- drivers/gpu/drm/virtio/virtgpu_fence.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/virtio/virtgpu_fence.c b/drivers/gpu/drm/virtio/virtgpu_fence.c index 87d1966192f4..b63ea4797afc 100644 --- a/drivers/gpu/drm/virtio/virtgpu_fence.c +++ b/drivers/gpu/drm/virtio/virtgpu_fence.c @@ -71,7 +71,7 @@ struct virtio_gpu_fence *virtio_gpu_fence_alloc(struct virtio_gpu_device *vgdev) { struct virtio_gpu_fence_driver *drv = &vgdev->fence_drv; struct virtio_gpu_fence *fence = kzalloc(sizeof(struct virtio_gpu_fence), - GFP_ATOMIC); + GFP_KERNEL); if (!fence) return fence;
Hey Chia-I,
This looks good to me, I can't find any spinlocks being held during that allocation.
Reviewed-by: Robert Foss robert.foss@collabora.com
On 30.04.19 00:10, Chia-I Wu wrote:
It was changed to GFP_ATOMIC in commit ec2f0577c (add & use virtio_gpu_queue_fenced_ctrl_buffer) because the allocation happened with a spinlock held. That was no longer true after commit 9fdd90c0f (add virtio_gpu_alloc_fence()).
Signed-off-by: Chia-I Wu olvaffe@gmail.com Cc: Gerd Hoffmann kraxel@redhat.com Cc: Gustavo Padovan gustavo.padovan@collabora.com Cc: Robert Foss robert.foss@collabora.com
drivers/gpu/drm/virtio/virtgpu_fence.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/virtio/virtgpu_fence.c b/drivers/gpu/drm/virtio/virtgpu_fence.c index 87d1966192f4..b63ea4797afc 100644 --- a/drivers/gpu/drm/virtio/virtgpu_fence.c +++ b/drivers/gpu/drm/virtio/virtgpu_fence.c @@ -71,7 +71,7 @@ struct virtio_gpu_fence *virtio_gpu_fence_alloc(struct virtio_gpu_device *vgdev) { struct virtio_gpu_fence_driver *drv = &vgdev->fence_drv; struct virtio_gpu_fence *fence = kzalloc(sizeof(struct virtio_gpu_fence),
GFP_ATOMIC);
GFP_KERNEL);
On Mon, 29 Apr 2019 at 23:10, Chia-I Wu olvaffe@gmail.com wrote:
It was changed to GFP_ATOMIC in commit ec2f0577c (add & use virtio_gpu_queue_fenced_ctrl_buffer) because the allocation happened with a spinlock held. That was no longer true after commit 9fdd90c0f (add virtio_gpu_alloc_fence()).
Signed-off-by: Chia-I Wu olvaffe@gmail.com Cc: Gerd Hoffmann kraxel@redhat.com Cc: Gustavo Padovan gustavo.padovan@collabora.com Cc: Robert Foss robert.foss@collabora.com
Reviewed-by: Emil Velikov emil.velikov@collabora.com
Side: Rob, should we follow-up on 9fdd90c0f and drop the virtio_gpu_fence_emit() return type?
HTH -Emil
On 03.05.19 16:31, Emil Velikov wrote:
On Mon, 29 Apr 2019 at 23:10, Chia-I Wu olvaffe@gmail.com wrote:
It was changed to GFP_ATOMIC in commit ec2f0577c (add & use virtio_gpu_queue_fenced_ctrl_buffer) because the allocation happened with a spinlock held. That was no longer true after commit 9fdd90c0f (add virtio_gpu_alloc_fence()).
Signed-off-by: Chia-I Wu olvaffe@gmail.com Cc: Gerd Hoffmann kraxel@redhat.com Cc: Gustavo Padovan gustavo.padovan@collabora.com Cc: Robert Foss robert.foss@collabora.com
Reviewed-by: Emil Velikov emil.velikov@collabora.com
Side: Rob, should we follow-up on 9fdd90c0f and drop the virtio_gpu_fence_emit() return type?
Good idea. None of the called functions return anything. I'll get right on it.
Rob.
On Mon, Apr 29, 2019 at 03:10:21PM -0700, Chia-I Wu wrote:
It was changed to GFP_ATOMIC in commit ec2f0577c (add & use virtio_gpu_queue_fenced_ctrl_buffer) because the allocation happened with a spinlock held. That was no longer true after commit 9fdd90c0f (add virtio_gpu_alloc_fence()).
Added to drm-misc-next.
thanks, Gerd
dri-devel@lists.freedesktop.org
-
Chia-I Wu -
Emil Velikov -
Gerd Hoffmann -
Robert Foss