>>>>> Tim Wilkinson writes:
> Okay this makes sense. This then means there's a problem with GLIBC since it
> excludes certain files when building a static library but unfortuantely
> needs them
> because it uses #ifdef PIC to include bits of code related to shared library
> workings (essentially it considers PIC to mean SHARED when this may not
> infact be
> the case - not for MIPS anyway).
> So what you're saying is pretty much how I have my gcc, binutils and glibc
> set up
> - I introduced a -DSHARED to glibc (which I suspect isn't really okay) when
> building shared libraries so static libraries didn't include the bad code.
> I may
> look for a better solution however.
I'm currently working and glibc - and the biggest problem is to get
binutils that work without any failures :-(. The SHARED patch is IMO
the right solution - but I need further to convince the glibc
maintainer. Please note that glibc 2.1 and 2.0 do need patches for
Mips/Linux. I've put most of my work (the clean patches;-) under CVS
for glibc 2.2 already - but some patches are still missing. If you
have any clean patches for the development version of glibc 2.2 ,
send them directly to me so that I can integrate them.
> I've also made a bunch of elf related changes to mips/linux so that the
> output from gcc is more like the elf output for the x86/linux targets. It's
> interesting to note that mips/linux doesn't even defined linux for CPP (and
> doesn't do a bunch of other stuff either come to that) which confused things
> end when compiling the linux kernel.
If you could get the gcc patches that are flying around integrated
into the official version of gcc (CVS version of gcc 2.96) that would
be fine. Please note that I've send already some patches to the GCC
folks and the MIPS/Linux support should be better in the development
version of gcc 2.96 than in 2.95.x. If time permits, I'll try to work
also on this.
 See http://sourceware.cygnus.com/glibc how to access our CVS archive.
SuSE Labs firstname.lastname@example.org