------- Comment #24 from tkho at ucla dot edu 2005-11-11 01:26 -------
For comparison, here's the code from gcc 2.95.3. It generates the same 18
instructions for both march=i386 and march=pentiumpro.
`gcc -c test3.c -save-temps -O2 -momit-leaf-frame-pointer -march=pentiumpro`:
pushl %ebx
movl 8(%esp),%ecx
movl 12(%esp),%ebx
movl 16(%esp),%eax
movl 20(%esp),%edx
shldl $8,%ecx,%ebx
sall $8,%ecx
movl %edx,%eax
xorl %edx,%edx
shrl $16,%eax
andl $255,%eax
andl $0,%edx
orl %eax,%ecx
orl %edx,%ebx
movl %ecx,%eax
movl %ebx,%edx
popl %ebx
ret
Also, in comment #23, I erronously used g++. Luckily, the same code was
generated with gcc.
On another note, Mark, I tried your patch in comment #10. I grabbed gcc-head
from 2005-09-28 and compared a clean build with a build that had your patches
applied. There was no difference in the assembly for the test case in comment
#23, and there was no performance gain in our benchmark application.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17886
