https://bugs.kde.org/show_bug.cgi?id=498492
--- Comment #2 from Paul Floyd <pjfl...@wanadoo.fr> ---
GCC/Linux
0000000000001139 <do_lzcnt64>:
1139: 55 push %rbp
113a: 48 89 e5 mov %rsp,%rbp
113d: 48 89 7d d8 mov %rdi,-0x28(%rbp)
1141: 48 89 75 d0 mov %rsi,-0x30(%rbp)
1145: 48 89 55 c8 mov %rdx,-0x38(%rbp)
1149: 48 8b 45 c8 mov -0x38(%rbp),%rax
114d: 48 89 45 e0 mov %rax,-0x20(%rbp)
1151: 48 c7 45 e8 00 00 00 movq $0x0,-0x18(%rbp)
1158: 00
1159: 48 c7 45 f0 00 00 00 movq $0x0,-0x10(%rbp)
1160: 00
1161: 48 8d 45 e0 lea -0x20(%rbp),%rax
1165: 49 bb 55 55 55 55 55 movabs $0x5555555555555555,%r11
116c: 55 55 55
116f: f3 4c 0f bd 18 lzcnt (%rax),%r11
1174: 4c 89 58 08 mov %r11,0x8(%rax)
1178: 9c pushf
1179: 41 5b pop %r11
117b: 4c 89 58 10 mov %r11,0x10(%rax)
117f: 48 8b 55 e8 mov -0x18(%rbp),%rdx
1183: 48 8b 45 d0 mov -0x30(%rbp),%rax
1187: 48 89 10 mov %rdx,(%rax)
118a: 48 8b 45 f0 mov -0x10(%rbp),%rax
118e: 25 d5 08 00 00 and $0x8d5,%eax
1193: 89 c2 mov %eax,%edx
1195: 48 8b 45 d8 mov -0x28(%rbp),%rax
1199: 89 10 mov %edx,(%rax)
119b: 90 nop
119c: 5d pop %rbp
119d: c3 ret
I think that it was working by chance with GCC, which happens to leave 8 bytes
of space on the stack that doesn't overwrite any outputs.
--
You are receiving this mail because:
You are watching all bug changes.
