[PATCH v2 0/2] drm/vboxvideo: Use struct drm_shadow_plane_state for cursor plane

Hans de Goede hdegoede at redhat.com
Tue Feb 9 11:04:51 UTC 2021


Hi,

On 2/9/21 9:13 AM, Thomas Zimmermann wrote:
> Hi
> 
> Am 08.02.21 um 18:43 schrieb Hans de Goede:
>> Hi,
>>
>> On 2/8/21 2:50 PM, Thomas Zimmermann wrote:
>>> (was: drm/vboxvideo: Vmap/vunmap cursor BO in prepare_fb and cleanup_fb)
>>>
>>> Functions in the atomic commit tail are not allowed to acquire the
>>> dmabuf's reservation lock. So we cannot legally call the GEM object's
>>> vmap functionality in atomic_update.
>>>
>>> But, much better, we can use drm_shadow_plane_state to do all the mapping
>>> for us. Patch 1 exports the helpers for shadow-buffered planes from the
>>> DRM KMS helper library and adds documentation on how to use them. Patch 2
>>> replaces the vmap code in vbox' cursor update code with a the helpers for
>>> shadow-buffered planes.
>>>
>>> Thomas Zimmermann (2):
>>>    drm/gem: Export helpers for shadow-buffered planes
>>>    drm/vboxvideo: Implement cursor plane with struct
>>>      drm_shadow_plane_state
>>
>> I've given this a test spin in a virtualbox vm using VboxSVGA graphics
>> and I've not found any problems:
>>
>> Tested-by: Hans de Goede <hdegoede at redhat.com>
> 
> Thanks a lot. Can I add your Acked-by as well?

Patch 1/2 is a bit outside of my area of expertise, but I see you already have
a Reviewed-by from Daniel there, so that one should be good to go.

I've just reviewed 2/2 and it looks good to me, so for 2/2 you may add my:

Reviewed-by: Hans de Goede <hdegoede at redhat.com>

Regards,

Hans



>>>
>>>   drivers/gpu/drm/drm_gem_atomic_helper.c | 148 +++++++++++++++++++++++-
>>>   drivers/gpu/drm/vboxvideo/vbox_mode.c   |  28 ++---
>>>   include/drm/drm_gem_atomic_helper.h     |  32 +++++
>>>   3 files changed, 181 insertions(+), 27 deletions(-)
>>>
>>> -- 
>>> 2.30.0
>>>
>>
>> _______________________________________________
>> dri-devel mailing list
>> dri-devel at lists.freedesktop.org
>> https://lists.freedesktop.org/mailman/listinfo/dri-devel
>>
> 



More information about the dri-devel mailing list