Re: current CVS crashes in find_basic_blocks during bootstrap

[Zack Weinberg <zack at wolery dot cumb dot org>]

On Sat, May 06, 2000 at 10:09:35AM -0700, Zack Weinberg wrote:
>
> I see it only with --enable-checking, too, but I think the bug is
> unrelated; it's just that --enable-checking changes the code in
> function.c and makes it tickle the bug.  I'll see if I can come up
> with a smaller test case.

Sure enough, an unchecked stage1 crashes in the same place on an
--enable-checking function.c.

I stripped this down as much as I could but it's still huge.  It
stopped crashing when I did a more aggressive prune.  Anyway:

$ stage1/cc1 function.i -dumpbase function.c -g -O2 -W -Wall -Wtraditional -Wno-long-long -Wwrite-strings -W -Wall -Wtraditional -Wno-long-long -Wwrite-strings -pedantic -pedantic -version -o function.s
GNU C version 2.96 20000506 (experimental) (i686-pc-linux-gnu) compiled by GNU C version 2.95.2 20000313 (Debian GNU/Linux).
options passed:  -g -O2 -W -Wall -Wtraditional -Wno-long-long
 -Wwrite-strings -W -Wall -Wtraditional -Wno-long-long -Wwrite-strings
 -pedantic -pedantic
options enabled:  -fdefer-pop -foptimize-sibling-calls -fcse-follow-jumps
 -fcse-skip-blocks -fexpensive-optimizations -fthread-jumps
 -fstrength-reduce -fpeephole -fforce-mem -ffunction-cse -finline
 -fkeep-static-consts -fcaller-saves -fpcc-struct-return -fgcse
 -frerun-cse-after-loop -frerun-loop-opt -fdelete-null-pointer-checks
 -fschedule-insns2 -fsched-interblock -fsched-spec -fnew-exceptions
 -fcommon -fgnu-linker -fregmove -foptimize-register-move -fargument-alias
 -fstrict-aliasing -fident -fpeephole2 -fmath-errno -m80387 -mhard-float
 -mno-soft-float -mieee-fp -mfp-ret-in-387
 fixup_var_refs_1Segmentation fault

Test case is attached.

zw

function.i.bz2