| Message ID | 20230302202219.846011-1-sakari.ailus@linux.intel.com |
|---|---|
| State | Superseded |
| Headers | show |
| Series | [1/1] media: v4l: subdev: Make link validation safer | expand |
Hi, On Fri, Mar 03, 2023 at 04:06:11PM +0100, Hans de Goede wrote: > Hi, > > On 3/3/23 12:36, Tomi Valkeinen wrote: > > On 03/03/2023 13:30, Sakari Ailus wrote: > >> Hi Tomi, > >> > >> On Fri, Mar 03, 2023 at 10:41:27AM +0200, Tomi Valkeinen wrote: > >>> On 02/03/2023 22:22, Sakari Ailus wrote: > >>>> Link validation currently accesses invalid pointers if the link passed to it > >>>> is not between two sub-devices. This is of course a driver bug. > >>>> > >>>> Ignore the error but print a debug message, as this is how it used to work > >>>> previously. > >>>> > >>>> Fixes: a6b995ed03ff ("media: subdev: use streams in v4l2_subdev_link_validate()") > >>>> Reported-by: Hans de Goede <hdegoede@redhat.com> > >>>> Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com> > >>>> --- > >>>> Hi Hans, > >>>> > >>>> Could you test this? > >>>> > >>>> The bug is of course in the ImgU driver and this reverts to the old > >>>> pre-streams behaviour. It silently fails instead of oopsing. The ImgU driver > >>>> needs to be fixed and I think we could make this return an error at the same > >>>> time. Right now I can't be sure the ImgU driver is the only one suffering > >>>> from this, but if so, it's likely to be broken anyway. > >>> > >>> Maybe it should be at least a warn? How do we catch other broken drivers > >>> otherwise? > >> > >> The purpose of this patch is just to restore the old behaviour, and merge > >> it as a fix to v6.3 (via Cc'ing stable). I agree this should be made an > >> error but I'd like that change to be present in the media tree for some > >> time first. > > > > I meant that keep it returning 0 (no error), but instead of a debug print, use pr_warn. Or maybe pr_warn_once for now. > > Switching to pr_warn_once() sounds reasonable to me. I'll send v2 with that.
diff --git a/drivers/media/v4l2-core/v4l2-subdev.c b/drivers/media/v4l2-core/v4l2-subdev.c index dff1d9be7841..a6c80096586e 100644 --- a/drivers/media/v4l2-core/v4l2-subdev.c +++ b/drivers/media/v4l2-core/v4l2-subdev.c @@ -1224,6 +1224,17 @@ int v4l2_subdev_link_validate(struct media_link *link) struct v4l2_subdev_state *source_state, *sink_state; int ret; + if (!is_media_entity_v4l2_subdev(link->sink->entity)) { + pr_debug("entity \"%s\" not a V4L2 sub-device, driver bug!\n", + link->sink->entity->name); + return 0; + } + if (!is_media_entity_v4l2_subdev(link->source->entity)) { + pr_debug("entity \"%s\" not a V4L2 sub-device, driver bug!\n", + link->source->entity->name); + return 0; + } + sink_sd = media_entity_to_v4l2_subdev(link->sink->entity); source_sd = media_entity_to_v4l2_subdev(link->source->entity);
Link validation currently accesses invalid pointers if the link passed to it is not between two sub-devices. This is of course a driver bug. Ignore the error but print a debug message, as this is how it used to work previously. Fixes: a6b995ed03ff ("media: subdev: use streams in v4l2_subdev_link_validate()") Reported-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com> --- Hi Hans, Could you test this? The bug is of course in the ImgU driver and this reverts to the old pre-streams behaviour. It silently fails instead of oopsing. The ImgU driver needs to be fixed and I think we could make this return an error at the same time. Right now I can't be sure the ImgU driver is the only one suffering from this, but if so, it's likely to be broken anyway. - Sakari drivers/media/v4l2-core/v4l2-subdev.c | 11 +++++++++++ 1 file changed, 11 insertions(+)