This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: fix for 991118-1 failure on big-endian machines
- To: geoffk at ozemail dot com dot au (Geoff Keating)
- Subject: Re: fix for 991118-1 failure on big-endian machines
- From: Herman ten Brugge <Haj dot Ten dot Brugge at net dot HCC dot nl>
- Date: Fri, 19 Nov 1999 22:17:34 +0000 (WET)
- Cc: gcc-patches at gcc dot gnu dot org
Geoff Keating wrote :
>
>
> This fixes the 991118-1 failure on ppc, taking it back down to 0
> c-torture failures.
>
> OK to commit?
>
> --
> Geoffrey Keating <geoffk@cygnus.com>
>
> ===File ~/patches/cygnus/egcs-llbitfield.patch==============
> md5sum: 73ad925b50e68865 4bbde29a57ef9484 447230
> Index: egcs/gcc/ChangeLog
> 0a
> Fri Nov 19 12:57:10 1999 Geoffrey Keating <geoffk@cygnus.com>
>
> * varasm.c (output_constructor): Solve problem with long long
> bitfields, even on BYTES_BIG_ENDIAN machines (testcase 991118-1).
>
> .
> md5sum: 37422032ea13bf1e 619c410cae643677 124086
> Index: egcs/gcc/testsuite/ChangeLog
> 0a
> 1999-11-19 Geoffrey Keating <geoffk@cygnus.com>
>
> * gcc.c-torture/execute/991118-1.c: Also test case
> where the word boundary does not split a byte evenly.
>
> .
It seems I made an error. I really dont know why I did change the code
for big endian machines. In my original patch I mentioned that I did
check it on sparc/m68k. But the patch doesn't work there :-(
I must have done something really stupid when I tested the patch on
those targets.
The patch you suggest works for sparc/m68k.
Herman.