------- Comment #5 from pinskia at gcc dot gnu dot org 2006-01-15 20:26 -------
Hmm, we get:
foo:
movl 4(%esp), %edx
xorl %ecx, %ecx
movl $1, %eax
cmpl g+4, %ecx
jb .L3
cmpl g, %edx
jb .L3
ret
.p2align 2,,3
.L3:
xorl %eax, %eax
ret
-----
The only thing which is semi wrong now is:
xorl %ecx, %ecx
cmpl g+4, %ecx
Everything else looks good (well except on ppc where we still get the different
compares).
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=14237
