linux-kernel - Re: linux-next: build failure after merge of the final tree (v4l-dvb tree related)

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] [day] [month] [year] [list]

Message-Id: <20130327221128.db6d370b4c06ceaf600b6ab5@canb.auug.org.au>
Date:	Wed, 27 Mar 2013 22:11:28 +1100
From:	Stephen Rothwell <sfr@...b.auug.org.au>
To:	Mauro Carvalho Chehab <mchehab@...hat.com>
Cc:	linux-next@...r.kernel.org, linux-kernel@...r.kernel.org,
	Hans Verkuil <hans.verkuil@...co.com>, Greg KH <greg@...ah.com>
Subject: Re: linux-next: build failure after merge of the final tree
 (v4l-dvb tree related)

Hi Mauro,

On Wed, 27 Mar 2013 10:52:34 +1100 Stephen Rothwell <sfr@...b.auug.org.au> wrote:
>
> On Tue, 26 Mar 2013 09:04:52 -0300 Mauro Carvalho Chehab <mchehab@...hat.com> wrote:
> >
> > Em Tue, 26 Mar 2013 17:18:47 +1100
> > Stephen Rothwell <sfr@...b.auug.org.au> escreveu:
> > 
> > > After merging the final tree, today's linux-next build (powerpc
> > > allyesconfig) failed like this:
> > > 
> > > drivers/staging/media/solo6x10/solo6x10-v4l2-enc.c: In function 'solo_enc_default':
> > > drivers/staging/media/solo6x10/solo6x10-v4l2-enc.c:1031:7: error: case label does not reduce to an integer constant
> > > drivers/staging/media/solo6x10/solo6x10-v4l2-enc.c:1035:7: error: case label does not reduce to an integer constant
> > > 
> > > I am not sure why this has suddenly appeared, but I have disabled the
> > > driver for now using this patch:
> > 
> > Well, solo6x10 didn't use to have its own private ioctl's until now.
> > Still, that's strange, as there are other drivers also using vidioc_default.
> > 
> > I suspect that those _IO* have its highest bit equal to 1 on powerpc
> > being too big for int. If so, the enclosed patch should fix. 
> > I'm still wandering why it didn't cause any compilation problems here.
> > 
> > Ok, I didn't try to compile it on powerpc, but still integers have 32
> > bits on ppc, right?
> 
> yes.
> 
> > Anyway, could you please try the enclosed patch?
> 
> Doesn't help.
> 
> on powerpc, SOLO_IOC_G_MOTION_THRESHOLDS expands to
> 
> (((2U) << (((0 +8)+8)+13)) | ((('V')) << (0 +8)) | (((192 +0)) << 0) | (((((sizeof(struct solo_motion_thresholds) == sizeof(struct solo_motion_thresholds[1]) && sizeof(struct solo_motion_thresholds) < (1 << 13)) ? sizeof(struct solo_motion_thresholds) : __invalid_size_argument_for_IOC))) << ((0 +8)+8)))
> 
> sizeof(struct solo_motion_thresholds) is 64*64*2 == 8192 which is not
> less than 1 << 13

Note that this will also break on alpha and mips.  Sparc seems to have a
tricky workaround (to allow 14 bits for size) - I wonder if that could be
used for these other arches as well?

-- 
Cheers,
Stephen Rothwell                    sfr@...b.auug.org.au

Content of type "application/pgp-signature" skipped