Re: [PATCH] drm/virtio: Add pixel blend mode property to cursor plane

[Dmitry Osipenko <dmitry.osipenko@collabora.com>]

On 10/1/26 21:00, Marek Szyprowski wrote:
> On 20.09.2026 20:21, Dmitry Osipenko wrote:
>> On 8/28/26 12:01, oushixiong1025@163.com wrote:
>>> From: Shixiong Ou <oushixiong@kylinos.cn>
>>>
>>> The cursor plane exposes a format with an alpha channel
>>> (DRM_FORMAT_ARGB8888) without a pixel blend mode property. Since
>>> commit 860e748bddcc ("drm: ensure blend mode supported if pixel
>>> format with alpha exposed") this triggers a warning during
>>> drm_mode_config_validate():
>>>
>>> [    0.649020] ------------[ cut here ]------------
>>> [    0.649040] [PLANE:36:plane-1] pixel format with alpha exposed but blend mode not setup
>>> [    0.649081] WARNING: drivers/gpu/drm/drm_mode_config.c:872 at drm_mode_config_validate
>>> ......
>>> [    0.649761] Call trace:
>>> [    0.649764]  drm_mode_config_validate+0x398/0x558 [drm] (P)
>>> [    0.649912]  drm_dev_register+0x1cc/0x2a0 [drm]
>>> [    0.650058]  virtio_gpu_probe+0xd4/0x1c0 [virtio_gpu]
>>> [    0.650088]  virtio_dev_probe+0x1c8/0x310
>>> ......
>>> [    0.650261] ---[ end trace 0000000000000000 ]---
>>>
>>>
>>> Create the property with the only supported blend mode,
>>> DRM_MODE_BLEND_PREMULTI, which is also the property's default and
>>> matches what userspace had to assume before the property existed.
>>>
>>> Reported-by: Ye Liu <liuye@kylinos.cn>
>>> Signed-off-by: Shixiong Ou <oushixiong@kylinos.cn>
>>> ---
>>>  drivers/gpu/drm/virtio/virtgpu_plane.c | 14 ++++++++++++++
>>>  1 file changed, 14 insertions(+)
>>>
>>> diff --git a/drivers/gpu/drm/virtio/virtgpu_plane.c b/drivers/gpu/drm/virtio/virtgpu_plane.c
>>> index 1d1b27ece62a..11771c943ea6 100644
>>> --- a/drivers/gpu/drm/virtio/virtgpu_plane.c
>>> +++ b/drivers/gpu/drm/virtio/virtgpu_plane.c
>>> @@ -24,6 +24,7 @@
>>>   */
>>>  
>>>  #include <drm/drm_atomic_helper.h>
>>> +#include <drm/drm_blend.h>
>>>  #include <drm/drm_damage_helper.h>
>>>  #include <drm/drm_fourcc.h>
>>>  #include <drm/drm_gem_atomic_helper.h>
>>> @@ -588,6 +589,7 @@ struct drm_plane *virtio_gpu_plane_init(struct virtio_gpu_device *vgdev,
>>>  	struct drm_plane *plane;
>>>  	const uint32_t *formats;
>>>  	int nformats;
>>> +	int ret;
>>>  
>>>  	if (type == DRM_PLANE_TYPE_CURSOR) {
>>>  		formats = virtio_gpu_cursor_formats;
>>> @@ -610,5 +612,17 @@ struct drm_plane *virtio_gpu_plane_init(struct virtio_gpu_device *vgdev,
>>>  	if (type == DRM_PLANE_TYPE_PRIMARY)
>>>  		drm_plane_enable_fb_damage_clips(plane);
>>>  
>>> +	if (type == DRM_PLANE_TYPE_CURSOR) {
>>> +		/*
>>> +		 * The cursor plane exposes a format with an alpha channel,
>>> +		 * which requires a blend mode property. The host blends
>>> +		 * premultiplied alpha, matching the property's default.
>>> +		 */
>>> +		ret = drm_plane_create_blend_mode_property(plane,
>>> +							   BIT(DRM_MODE_BLEND_PREMULTI));
>>> +		if (ret)
>>> +			return ERR_PTR(ret);
>>> +	}
>>> +
>>>  	return plane;
>>>  }
>> Applied to misc-fixes, thanks!
> 
> 
> This duplicates commit f2e64f450c16 ("drm/virtio: create blend mode property on cursor
> plane") merged a bit earlier, what in turn results in double call to
> drm_plane_create_blend_mode_property() what confuses userspace (modetest tool crashes).
Good catch, will revert this patch. Thanks!

-- 
Best regards,
Dmitry