[FFmpeg-devel] [PATCH] avutil/hwcontext_drm: Check ioctl in drm_map_frame() for failure

[Michael Niedermayer]

On Mon, May 20, 2024 at 03:36:55PM +0200, Michael Niedermayer wrote:
> On Mon, May 20, 2024 at 11:33:41AM +0200, Andreas Rheinhardt wrote:
> > Michael Niedermayer:
> > > Fixes: CID1583742 Unchecked return value
> > > 
> > > Sponsored-by: Sovereign Tech Fund
> > > Signed-off-by: Michael Niedermayer <michael at niedermayer.cc>
> > > ---
> > >  libavutil/hwcontext_drm.c | 5 ++++-
> > >  1 file changed, 4 insertions(+), 1 deletion(-)
> > > 
> > > diff --git a/libavutil/hwcontext_drm.c b/libavutil/hwcontext_drm.c
> > > index 0847db09a08..e080c0597b8 100644
> > > --- a/libavutil/hwcontext_drm.c
> > > +++ b/libavutil/hwcontext_drm.c
> > > @@ -166,7 +166,10 @@ static int drm_map_frame(AVHWFramesContext *hwfc,
> > >  #if HAVE_LINUX_DMA_BUF_H
> > >          /* We're not checking for errors here because the kernel may not
> > >           * support the ioctl, in which case its okay to carry on */
> > > -        ioctl(desc->objects[i].fd, DMA_BUF_IOCTL_SYNC, &sync_start);
> > > +        if (ioctl(desc->objects[i].fd, DMA_BUF_IOCTL_SYNC, &sync_start) == -1) {
> > > +            err = AVERROR(errno);
> > > +            goto fail;
> > > +        }
> > >  #endif
> > >      }
> > >      map->nb_regions = i;
> > 
> > Did you read the comment above the code?
> 
> Apparently not
> 
> patch droped, will mark this as intentional

theres a 2nd ioctl() call in the file (drm_unmap_frame()), that has no comment
and no check
Is that as intended or does it need some change ?
(its CID1583743 Unchecked return value)

thx

[...]

-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

The educated differ from the uneducated as much as the living from the
dead. -- Aristotle 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 195 bytes
Desc: not available
URL: <https://ffmpeg.org/pipermail/ffmpeg-devel/attachments/20240713/c83b82ad/attachment.sig>