[PATCH v2] libcamera: udev: Catch udev notification errors

[Kieran Bingham]

Quoting Laurent Pinchart (2024-08-07 16:01:59)
> On Wed, Aug 07, 2024 at 03:33:20PM +0100, Kieran Bingham wrote:
> > Quoting Laurent Pinchart (2024-08-07 15:24:55)
> > > On Wed, Aug 07, 2024 at 02:49:50PM +0100, Kieran Bingham wrote:
> > > > The udev_monitor_receive_device() can return NULL on an error as
> > > > detailed in the man pages for the function.
> > > > 
> > > > The udevNotify() handler in the DeviceEnumeratorUdev directly uses the
> > > > return value of udev_monitor_receive_device() in successive calls to
> > > > process the event without having first checked the udev_device.
> > > > 
> > > > Ensure we identify, and handle events where the udev_device can not be
> > > > returned successfully.
> > > > 
> > > > Bug: https://bugs.libcamera.org/show_bug.cgi?id=230
> > > > Reviewed-by: Stefan Klug <stefan.klug at ideasonboard.com>
> > > > Signed-off-by: Kieran Bingham <kieran.bingham at ideasonboard.com>
> > > > 
> > > > ---
> > > > v2:
> > > >  - Report strerror based on errno from the udev_monitor_receive_device()
> > > >    call.
> > > > 
> > > >  src/libcamera/device_enumerator_udev.cpp | 8 ++++++++
> > > >  1 file changed, 8 insertions(+)
> > > > 
> > > > diff --git a/src/libcamera/device_enumerator_udev.cpp b/src/libcamera/device_enumerator_udev.cpp
> > > > index 01c70b6daa82..53eeb772d900 100644
> > > > --- a/src/libcamera/device_enumerator_udev.cpp
> > > > +++ b/src/libcamera/device_enumerator_udev.cpp
> > > > @@ -332,6 +332,14 @@ int DeviceEnumeratorUdev::addV4L2Device(dev_t devnum)
> > > >  void DeviceEnumeratorUdev::udevNotify()
> > > >  {
> > > >       struct udev_device *dev = udev_monitor_receive_device(monitor_);
> > > > +     if (!dev) {
> > > > +             int error = errno;
> > > 
> > > grepping for ' = errno' gives me 12 variables named ret, and one named
> > > err. I'd go for 'ret' here too for consitency.
> > 
> > I explicitly didn't use 'ret' because it's not a ret value...
> > 
> > And 'err' was just shorthand. I'll rename if you wish, but please
> > confirm.
> 
> I'd prefer 'ret' but I'm also fine with 'err'.
> 
> > > Reviewed-by: Laurent Pinchart <laurent.pinchart at ideasonboard.com>
> > > 
> > > Could you ask the bug reporter to test this patch and tell what error
> > > number they get ?
> > 
> > Yes, but I don't know how they'll test it, as it's part of a separate CI
> > system, so I don't think we'll get a response (until it filters into a
> > release, that then gets utilised in that CI system).
> 
> Can't they push a test branch to that CI system, or run the failing
> tests locally ?

You tell me ;-) But I would suspect that as they were testing "something
else entirely" [0] it would require updating various CI docker images to
make a different version of libcamera be built inside that test
container (which I would anticipate would have been a binary package
installation) - so that they could run their secondary test which was
probably a race condition as 7 out of 8 other runs [1][2] on different
runners worked successfully.

I've asked [3] - I'm just saying that I'm not expecting any 'quick'
answer. I think this is like asking them to test a systemd patch in our
libcamera CI...

[0] Update syscall tables and linux headers #33930 (systemd)
    https://github.com/systemd/systemd/pull/33930

[1] https://github.com/systemd/systemd/actions/runs/10233929013
[2] https://github.com/systemd/systemd/actions/runs/10233929013/job/28312766461

[3] https://github.com/systemd/systemd/pull/33930#issuecomment-2273739561

--
Kieran


> 
> > > > +             LOG(DeviceEnumerator, Warning)
> > > > +                     << "Ignoring notfication received without a device: "
> > > > +                     << strerror(error);
> > > > +             return;
> > > > +     }
> > > > +
> > > >       std::string_view action(udev_device_get_action(dev));
> > > >       std::string_view deviceNode(udev_device_get_devnode(dev));
> > > >  
> 
> -- 
> Regards,
> 
> Laurent Pinchart