------- Additional Comments From uros at kss-loka dot si 2005-02-03 16:17
-------
Promised timings, all on P4-3.2, 800MHz FSB:
gcc296 -O3:
real 0m0.978s
user 0m0.979s
sys 0m0.000s
gcc296 -O3 -march=i686:
real 0m1.061s
user 0m1.029s
sys 0m0.016s
gcc344-pre -03
real 0m0.962s
user 0m0.961s
sys 0m0.002s
gcc344-pre -03 -march=pentium4:
real 0m0.962s
user 0m0.961s
sys 0m0.000s
gcc-40 -O3 -march=i686: <- scheduler is enabled here!
real 0m0.605s
user 0m0.605s
sys 0m0.000s
gcc-40 -O3 -march=pentium4: <- scheduler disabled
real 0m0.599s
user 0m0.600s
sys 0m0.000s
gcc-40 -O3 -march=pentium4 -mfpmath=sse:
real 0m0.726s
user 0m0.727s
sys 0m0.000s
I didn't test 3.3 but I wouldn't call the results of 3.4 and 4.0 _FAR_SLOWER_
:)
gcc -O3 -march=i686 -msse2 -mfpmath=sse
real 0m0.726s
user 0m0.725s
sys 0m0.002s
BTW: It looks there is a problem with -mfpmath=sse. I'll open another PR for
this problem: "Floating point computation far slower for -mfpmath=sse" :).
Steven, I guess this PR should be marked as fixed.
--
What |Removed |Added
----------------------------------------------------------------------------
Status|WAITING |NEW
Last reconfirmed|2004-08-01 00:24:18 |2005-02-03 16:17:27
date| |
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=8126
