[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20181030135242.doccralyqaqzshyx@sirius.home.kraxel.org>
Date: Tue, 30 Oct 2018 14:52:42 +0100
From: Gerd Hoffmann <kraxel@...hat.com>
To: Emil Velikov <emil.l.velikov@...il.com>
Cc: Robert Foss <robert.foss@...labora.com>,
David Airlie <airlied@...ux.ie>,
"Linux-Kernel@...r. Kernel. Org" <linux-kernel@...r.kernel.org>,
ML dri-devel <dri-devel@...ts.freedesktop.org>,
"open list:VIRTIO GPU DRIVER"
<virtualization@...ts.linux-foundation.org>,
Gustavo Padovan <gustavo.padovan@...labora.com>,
Emil Velikov <emil.velikov@...labora.com>
Subject: Re: [PATCH 2/5] drm/virtio: add uapi for in and out explicit fences
On Tue, Oct 30, 2018 at 11:31:04AM +0000, Emil Velikov wrote:
> HI Gerd,
>
> On Tue, 30 Oct 2018 at 06:11, Gerd Hoffmann <kraxel@...hat.com> wrote:
> >
> > Hi,
> >
> > > The execbuffer IOCTL is now read-write to allow the userspace to read the
> > > out-fence.
> >
> > > #define DRM_IOCTL_VIRTGPU_EXECBUFFER \
> > > - DRM_IOW(DRM_COMMAND_BASE + DRM_VIRTGPU_EXECBUFFER,\
> > > + DRM_IOWR(DRM_COMMAND_BASE + DRM_VIRTGPU_EXECBUFFER,\
> > > struct drm_virtgpu_execbuffer)
> >
> > That changes the ioctl number and breaks the userspace api.
> >
> Have you looked at the drm_ioctl() implementation? AFAICT it
> explicitly caters for this kind of changes.
Looking ...
The direction bits are not used to lookup the ioctl functions,
so it should work indeed.
Series doesn't apply to drm-misc-next and needs a rebase.
cheers,
Gerd
Powered by blists - more mailing lists