[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20080226160439.5941cc31.akpm@linux-foundation.org>
Date: Tue, 26 Feb 2008 16:04:39 -0800
From: Andrew Morton <akpm@...ux-foundation.org>
To: Evgeniy Polyakov <johnpol@....mipt.ru>
Cc: Adrian Bunk <bunk@...nel.org>, Martin Michlmayr <tbm@...ius.com>,
linux-kernel@...r.kernel.org,
Herbert Xu <herbert@...dor.apana.org.au>,
Ralf Baechle <ralf@...ux-mips.org>,
linux-crypto@...r.kernel.org
Subject: Re: [2.6.25 patch] drivers/crypto/hifn_795x.c: fix 64bit division
On Tue, 26 Feb 2008 21:52:40 +0300 Evgeniy Polyakov <johnpol@....mipt.ru> wrote:
> Hi Adrian.
>
> On Tue, Feb 26, 2008 at 05:34:21PM +0200, Adrian Bunk (bunk@...nel.org) wrote:
> > Using ndelay() with a 64bit variable as parameter can result in build
> > errors like the following on some 32bit systems when it results in a
> > 64bit division:
> >
> > <-- snip -->
> >
> > ...
> > MODPOST 759 modules
> > ERROR: "__divdi3" [drivers/crypto/hifn_795x.ko] undefined!
> >
> > <-- snip -->
> >
> > Reported by Martin Michlmayr.
> >
> > Signed-off-by: Adrian Bunk <bunk@...nel.org>
>
> Yep, ndelay() uses division, thanks a lot Adrian for spotting this.
hm. Where?
> Herbert, please apply.
>
> ACK.
udelay() might be exposed to the same problem. It would be better to fix
ndelay() and udelay() rather than callers. It is reasonable to pass a u64
into ndelay() and to expect the build to not explode.
(Geeze macros suck)
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists