Re: Bug#269989: ghc6: FTBFS: Passes -input to ar

[Ian Lynagh <igloo@earth.li>]

On Sat, Sep 04, 2004 at 12:40:50PM -0700, Daniel Schepler wrote:
> Package: ghc6
> Severity: serious
> Version: 6.2.1-5
> 
> >From my build log:
> 
> ...
> /usr/bin/ar q    libHSbase.a -input libHSbase.a.list
> /usr/bin/ar: creating libHSbase.a
> /usr/bin/ar: '-input': No such file
> rm -f libHSbase.a.list
> : libHSbase.a
> ...
> ../../ghc/compiler/stage1/ghc-inplace [...]
> /tmp/buildd/ghc6-6.2.1/libraries/readline/libHSreadline.a: could not read symbols: Archive has no index; run ranlib to add one
> collect2: ld returned 1 exit status
> make[2]: *** [stage2/ghc-6.2.1] Error 1
> make[2]: Leaving directory `/tmp/buildd/ghc6-6.2.1/ghc/compiler'
> make[1]: *** [stage2] Error 2
> make[1]: Leaving directory `/tmp/buildd/ghc6-6.2.1'
> make: *** [build-stamp] Error 2
> 
> The same patch as given in bug #266446 for ghc5 fixes the bug for
> ghc6.

Thanks for the bug report.

This was previously reported as #267139, but was since reassigned to
binutils. I'll leave this one here should anyone else come across the
problem.

I am told the new ar is destined for sarge, so according to this plan
ghc6 would be unbuildable in sarge.

A sourceful upload with the above patch would mean about 6 days of
compilation on mips and m68k, so I'd rather not do that if a fixed ar
will be appearing.

I will thus do nothing unless told otherwise by a RM or James.


Thanks
Ian