Re: crosstool 0.28rc36 'glibc-vfp.patch'

[Lennert Buytenhek <buytenh at wantstofly dot org>]

On Sat, Sep 25, 2004 at 08:28:08PM -0700, Dan Kegel wrote:

> >It's definitely needed for hardfloat.  So I'd think it's needed for
> >big-endian systems in any case, and for VFP on little-endian systems
> >too.  Someone would have to verify that though.
> 
> Can you folks have a look at glibc-20040827 and verify
> whether this patch is still needed and useful with that version?
> Let's get ready to get this into glibc CVS if in fact it's
> really a good patch for everyone.

I just had a look at glibc-20040830, and it's still needed and useful
for this version.  glibc-20040830 out-of-the-box still contains the
following wrong assumptions:
- sysdeps/arm/bits/endian.h: float word order is big endian (which it is
  not on vfp systems)
- sysdeps/arm/gmp-mparam.h: IEEE doubles are mixed endian (which they
  are not on big endian systems, neither on vfp systems)
- sysdeps/arm/ieee754.h: IEEE doubles are in little endian byte order
  (which they are not on big endian systems)

glibc-vfp.patch from crosstool 0.28rc36 seems the right solution for
all of these issues.


--L

------
Want more information?  See the CrossGCC FAQ, http://www.objsw.com/CrossGCC/
Want to unsubscribe? Send a note to crossgcc-unsubscribe@sources.redhat.com