mips-sgi-irix6.5SE doesn't bootstrap

Jeffrey A Law law@cygnus.com
Tue May 19 23:38:00 GMT 1998


  In message < 199805200606.XAA00719@mail.earthlink.net >you write:
  > 
  > I'm now getting:
  > 
  >   ./xgcc -B./  -DIN_GCC    -W -Wall -O2 -g -O2 -I./include  -I. -I../../gcc
  >  -I../\
  >   ../gcc/config \
  >   -c ../../gcc/objc/hash.c -o objc/hash.o    
  >     xgcc: Internal compiler error: program cc1 got fatal signal 11 
  > 
  > That's the stage2 xgcc.  It's dying in recog_5 in the generated
  > insn-recog.c at:
  > 
  >   >*4006    if (GET_MODE (x1) == SImode && GET_CODE (x1) == PLUS && 1)
I believe there's a bug in Joern's last reload1.c change -- I can't
get an x86 to bootstrap with that patch installed.

You might try backing that patch out and bootstrapping again.

Joern -- there's definitely a bug of some sort, because I do get
a clean 3-stage on the x86 when I take it out:

Tue May 19 17:19:16 1998  J"orn Rennecke <amylaar@cygnus.co.uk>

        * reload1.c (reload_reg_free_for_value_p): New function.
        (allocate_reload_reg, choose_reload_regs): Use it.


If you've got an x86 in the UK, you should probably try a 3-stage,
expect cccp to be mis-compiled by the stage1 compiler -- the stage2
cpp will core dump when it's used late in stage2.


jeff



More information about the Gcc-bugs mailing list