This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: haifa-sched.c: Internal compiler error
- To: schwab at issan dot informatik dot uni-dortmund dot de (Andreas Schwab)
- Subject: Re: haifa-sched.c: Internal compiler error
- From: hjl at lucon dot org (H.J. Lu)
- Date: Mon, 15 Jun 1998 07:50:56 -0700 (PDT)
- Cc: hjl at lucon dot org, sopwith at redhat dot com, egcs at cygnus dot com, linuxppc-dev at cs dot wisc dot edu
>
> hjl@lucon.org (H.J. Lu) writes:
>
> |> >
> |> > gcc -D__KERNEL__ -I/usr/src/linux/include -Wall -Wstrict-prototypes -O2
> |> > -fomit-frame-pointer -D__powerpc__ -fsigned-char -msoft-float -pipe
> |> > -fno-builtin -ffixed-r2 -Wno-uninitialized -mmultiple -mstring -c -o
> |> > super.o super.c
> |> > ../../gcc/haifa-sched.c:7593: Internal compiler error in function
> |> > schedule_region
> |> > cpp: output pipe has been closed
> |> > make[2]: *** [super.o] Error 1
> |> >
> |>
> |> FYI, all egcs since the last gcc2 merge may be miscompiled due to
> |> a combine bug from gcc2.
>
> It may be fixed by a the following change in gcc2, which is part of the
> latest snapshot:
>
> * combine.c (simplify_comparison, case AND): Don't commute AND
> with SUBREG if constant is whole mode and don't do if lowpart
> and not WORD_REGISTER_OPERATIONS.
>
This patch doesn't fix the x86 bug caused the bug introduced by the
previous change.
--
H.J. Lu (hjl@gnu.org)