linux-kernel - Re: [PATCH 10/10] bkl: Fix-up compile problems as a result of the bkl-pushdown.

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <alpine.LFD.2.00.1004271352510.3318@localhost>
Date:	Tue, 27 Apr 2010 13:54:36 +0200 (CEST)
From:	John Kacur <jkacur@...hat.com>
To:	Arnd Bergmann <arnd@...db.de>
cc:	John Kacur <jkacur@...hat.com>,
	lkml <linux-kernel@...r.kernel.org>,
	Linus Torvalds <torvalds@...ux-foundation.org>,
	Frederic Weisbecker <fweisbec@...il.com>,
	Jan Blunck <jblunck@...il.com>,
	Thomas Gleixner <tglx@...utronix.de>
Subject: Re: [PATCH 10/10] bkl: Fix-up compile problems as a result of the
 bkl-pushdown.



On Tue, 27 Apr 2010, Arnd Bergmann wrote:

> On Tuesday 27 April 2010, John Kacur wrote:
> > Fix-up compile problems as a result of the bkl-pushdown.
> > In particular,  the v4l2_ioctl should call an unlocked_ioctl
> > 
> > Signed-off-by: John Kacur <jkacur@...hat.com>
> > ---
> >  drivers/media/video/v4l2-dev.c |    4 ++--
> >  1 files changed, 2 insertions(+), 2 deletions(-)
> > 
> > diff --git a/drivers/media/video/v4l2-dev.c b/drivers/media/video/v4l2-dev.c
> > index 3606694..8fbfa61 100644
> > --- a/drivers/media/video/v4l2-dev.c
> > +++ b/drivers/media/video/v4l2-dev.c
> > @@ -222,12 +222,12 @@ static long v4l2_ioctl(struct file *filp,
> >         struct video_device *vdev = video_devdata(filp);
> >         int ret;
> >  
> > -       if (!vdev->fops->ioctl)
> > +       if (!vdev->fops->unlocked_ioctl)
> >                 return -ENOTTY;
> >         /* Allow ioctl to continue even if the device was unregistered.
> >            Things like dequeueing buffers might still be useful. */
> >         lock_kernel();
> > -       ret = vdev->fops->ioctl(filp->f_path.dentry->d_inode, filp, cmd, arg);
> > +       ret = vdev->fops->unlocked_ioctl(filp, cmd, arg);
> >         unlock_kernel();
> >  
> >         return ret;
> 
> No, this is what I thought as well, at first, but the version I posted is
> actually correct.
> 
> vdev->fops is not struct file_operations but struct v4l2_file_operations,
> and the v4l2_ioctl function is used only when fops->ioctl is set, see
> __video_register_device.
> 
> The v4l2 ioctl stuff probably can use a lot of cleanup itself, but so
> far I think we're not making it worse with my patch.
> 

Hi Arnd

Well it is certainly possible that my fixup is not correct too - your 
patch cannot be correct, because it doesn't compile! Here is what I get 
when I apply your patch to a recent linus/master

make O=/bld/arnd/ drivers/media/video/v4l2-dev.o
-----CUT A BUNCH OF STUFF OUT ---
  CC      drivers/media/video/v4l2-dev.o
/home/jkacur/jk-2.6/drivers/media/video/v4l2-dev.c: In function 
‘v4l2_ioctl’:
/home/jkacur/jk-2.6/drivers/media/video/v4l2-dev.c:230: warning: passing 
argument 1 of ‘vdev->fops->ioctl’ from incompatible pointer type
/home/jkacur/jk-2.6/drivers/media/video/v4l2-dev.c:230: note: expected 
‘struct file *’ but argument is of type ‘struct inode *’
/home/jkacur/jk-2.6/drivers/media/video/v4l2-dev.c:230: warning: passing 
argument 2 of ‘vdev->fops->ioctl’ makes integer from pointer without a 
cast
/home/jkacur/jk-2.6/drivers/media/video/v4l2-dev.c:230: note: expected 
‘unsigned int’ but argument is of type ‘struct file *’
/home/jkacur/jk-2.6/drivers/media/video/v4l2-dev.c:230: error: too many 
arguments to function ‘vdev->fops->ioctl’
make[2]: *** [drivers/media/video/v4l2-dev.o] Error 1
make[1]: *** [drivers/media/video/v4l2-dev.o] Error 2
make: *** [sub-make] Error 2


Thanks