[PATCH] drm/virtio: populate .set_busid callback

2016-10-03 Thread Laszlo Ersek
On 10/03/16 19:28, Emil Velikov wrote: > On 3 October 2016 at 18:08, Laszlo Ersek wrote: >> On 10/03/16 18:43, Emil Velikov wrote: >>> Earlier commit was removing all the users of drm_platform_set_busid and >>> removed the virtio hunk (which uses the PCI version of the function) by >>> mistake. >>

[PATCH] drm/virtio: populate .set_busid callback

2016-10-03 Thread Laszlo Ersek
On 10/03/16 19:08, Laszlo Ersek wrote: > On 10/03/16 18:43, Emil Velikov wrote: >> Earlier commit was removing all the users of drm_platform_set_busid and >> removed the virtio hunk (which uses the PCI version of the function) by >> mistake. >> >> This in itself resulted in user space receiving an

[PATCH] drm/virtio: populate .set_busid callback

2016-10-03 Thread Laszlo Ersek
On 10/03/16 18:43, Emil Velikov wrote: > Earlier commit was removing all the users of drm_platform_set_busid and > removed the virtio hunk (which uses the PCI version of the function) by > mistake. > > This in itself resulted in user space receiving an incorrect value for > the busid, which in the

[PATCH] drm/virtio: populate .set_busid callback

2016-10-03 Thread Emil Velikov
On 3 October 2016 at 18:08, Laszlo Ersek wrote: > On 10/03/16 18:43, Emil Velikov wrote: >> Earlier commit was removing all the users of drm_platform_set_busid and >> removed the virtio hunk (which uses the PCI version of the function) by >> mistake. >> >> This in itself resulted in user space rec

[PATCH] drm/virtio: populate .set_busid callback

2016-10-03 Thread Emil Velikov
Earlier commit was removing all the users of drm_platform_set_busid and removed the virtio hunk (which uses the PCI version of the function) by mistake. This in itself resulted in user space receiving an incorrect value for the busid, which in the case below lead to the failure to detect the (corr