Libav 0.7 FTBFS on mipsel: Error: opcode not supported on this processor: mips2 (mips2) `ldl $2, 7($13)'

[Aurelien Jarno]

On Sat, Jul 23, 2011 at 11:37:08PM +0200, Kurt Roeckx wrote:
> On Sat, Jul 23, 2011 at 10:44:22PM +0200, Aurelien Jarno wrote:
> > On Wed, Jul 20, 2011 at 08:52:45PM +0200, Philipp Kern wrote:
> > > [ mips* buildds Cc'ed ]
> > > 
> > > On Tue, Jul 19, 2011 at 09:58:34PM +0200, David Kuehling wrote:
> > > > The complete userspace *is* o32, just the kernel is not.  I think that
> > > > is a pretty valid way to run a system, compiling the kernel for mips64
> > > > gives better performance on those machines that can run mips64 code.
> > > > 
> > > > A mips64 kernel can run 32 and 64-bit architecture binaries, and has to
> > > > pick one description when asked via 'uname -m'.  The 'setarch' tool can
> > > > be used to configure which architecture that is.
> > > > 
> > > > I.e. athough my machine usually returns 'mips64' on 'uname -m', after
> > > > running, 'setarch mips' it returns just 'mips'.  Maybe that'd be a a
> > > > cleaner way to fix the problem for all package builds?
> > > 
> > > I think that's a misconfiguration of the buildds.  They're supposed to
> > > put linux32 into the schroot configuration if the kernel arch differs
> > > from the userspace arch, to get the right entry into uname.
> > 
> > It should be fixed right now, but it would be nice to have that in
> > create-chroot.sh as otherwise we might forget next time we create a
> > chroot or setup a machine.
> 
> It at least does the right thing on the i386 buildds.  Maybe the
> chroot was older?
> 

It's what I thought first, but the script [1] doesn't seems to do that,
or at least I fail to see where.

[1] https://buildd.debian.org/docs/create-chroot.sh

-- 
Aurelien Jarno	                        GPG: 1024D/F1BCDB73
aurelien at aurel32.net                 http://www.aurel32.net