Bug#234236: grave

[Thiemo Seufer <ica2_ts@csv.ica.uni-stuttgart.de>]

GOTO Masanori wrote:
> At Mon, 10 May 2004 07:21:18 +0200,
> Thiemo Seufer wrote:
> > GOTO Masanori wrote:
> > > At Mon, 3 May 2004 18:50:18 +0100,
> > > Martin Michlmayr wrote:
> > > > severity 234236 grave
> > > > thanks
> > > > 
> > > > This is a grave bug since it keeps about 10-15 packages from building
> > > > on mips.
> > > 
> > > Which packages did this bug affect?  I agreed with Thiemo's and
> > > Goswin's opinion:
> > >
> > > > I wonder if it would be time to convert fdisk to _FILE_OFFSET_BITS=64 for
> > > > linux on all archs instead of 32 bit + _llseek hacks.
> > > 
> > > > But then, util-linux should use more portable means than kernel headers,
> > > > like the syscall() interface.
> > > 
> > > So I don't think it's grave for linux-kernel-headers.
> > 
> > Glibc itself might break as well, probably in non-obvious ways, if built
> > against this version of l-k-h.

I remembered incorrectly. Glibc has its own macros, it doesn't use the
kernel ones any more.

> > The patch is obviously correct, and already in the upstream repository
> > (www.linux-mips.org).
> 
> I don't discuss the correctness of this patch; my question is: which
> packages are this patch needed.

util-linux and raidtools are listed as RC bugs. Martin can probably
provide a longer list of such packages. In the larger view it affects
every application which includes <linux/unistd.h> and tries to use a
syscall with more than 4 parameters.


Thiemo