GCC build failed for powerpc-eabisim with your patch on 2001-06-04T15:31:45Z.

[Geoff Keating]

> Mailing-List: contact gcc-regression-help@gcc.gnu.org; run by ezmlm
> list-help: < mailto:gcc-regression-help@gcc.gnu.org >
> list-unsubscribe: < mailto:gcc-regression-unsubscribe@gcc.gnu.org >
> list-post: < mailto:gcc-regression@gcc.gnu.org >
> Date: Mon, 4 Jun 2001 19:46:40 +0200
> From: Jan Hubicka <jh@suse.cz>
> Content-Disposition: inline
> User-Agent: Mutt/1.3.15i
> 
> > With your recent patch, GCC does not compile on:
> >  powerpc-eabisim
> > /maat/heart/tbox/cvs-gcc/gcc/gcc/libgcc2.c: In function `__gcc_bcmp':
> > /maat/heart/tbox/cvs-gcc/gcc/gcc/libgcc2.c:1233: Internal compiler error in gen_lowpart, at emit-rtl.c:1110
> > Please submit a full bug report,
> > with preprocessed source if appropriate.
> > See <URL: http://www.gnu.org/software/gcc/bugs.html > for instructions.
> 
> This is latent bug caught by my checking code.
> Basically we do SImode lowpart of (subreg:QI (reg:SI) 3).
> The second is already lowpart of SImode value, so the result should be (reg:SI),
> but most of the code combines offsets by adding the offsets.  This is not
> true for paradoxical subregs on big endian machines in our current representation.
> 
> To simplify_subreg I've copied code that seemed to handle, but it is not correct,
> as it checks only old subreg to be paradoxical, so I've rewrote the code to cleaner
> form and I've checked that it fixes the problem.
> 
> I am now running i386 checking, but it is mostly useless, as the code in
> question triggers only for big endian.
> 
> OK assuming it passes?
> (it makes gcc to compile few randomly choosen PPC files, that seems to be
> good test given the amount of sanity checking I do have)

You can always build a cross-compiler to powerpc-eabisim.  That is
what the automated tester does (it's running on an x86 linux box).

-- 
- Geoffrey Keating <geoffk@geoffk.org>