This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Bootstrap failure on powerpc64 (mainline)
- From: Hans-Peter Nilsson <hp at bitrange dot com>
- To: Josef Zlomek <zlomj9am at artax dot karlin dot mff dot cuni dot cz>
- Cc: gcc at gcc dot gnu dot org
- Date: Thu, 6 Mar 2003 09:24:40 -0500 (EST)
- Subject: Re: Bootstrap failure on powerpc64 (mainline)
On Thu, 6 Mar 2003, Josef Zlomek wrote:
> Backtrace:
>
> #0 fancy_abort (file=0x105ee8c8 "../../gcc/expr.c", line=3495, function=0x105eeb70 "emit_move_insn_1") at ../../gcc/diagnostic.c:1367
> #1 0x0000000010190af4 in emit_move_insn_1 (x=0x7fe3874420, y=0x7fe3946f80) at ../../gcc/expr.c:3495
> #2 0x000000001018fe28 in emit_move_insn (x=0x7fe3874420, y=0x7fe3946f80) at ../../gcc/expr.c:3196
> #3 0x000000001034e5e8 in move_movables (loop=0x109fce70, movables=0x109faab0, threshold=71, insn_count=163) at ../../gcc/loop.c:2226
> When I try command "frame 1" and "p debug_rtx (x)" I get following error,
> so it's hard for me to debug it.
Might be an unrelated bug in debugging info or gdb. Work around
it by using "up" until you get to the emit_move_insn call in
move_movables, then write out the rtx for the src and dest for
the move, something like:
(gdb) up
(gdb) up
(gdb) call debug_rtx (newreg)
(gdb) call debug_rtx (m->set_dest)
If any of these are MODE_CC (for example CCmode), I guess the
rs6000 port better add some movecc patterns. :-) See recent
mmix.md patch and optabs.c/expr.c patch.
brgds, H-P