Mark Mitchell wrote on 01/01/07 14:46:
What a thread this has turned out to be.
Indeed.
In general, I'm not too thrilled with the idea of disabling
transformations for the sake of non-conforming code. However, I would
not mind a -fconforming flag similar to -fstrict-aliasing.
I haven't yet seen that anyone has actually tried the obvious: run SPEC
with and without -fwrapv. Would someone please do that? Or, pick your
favorite high-performance application and do the same. But, let's get
some concrete data as to how much this optimization helps.
On x86_64:
SPEC2000int is almost identical.
SPEC2000fp shows a -6% drop when using -fwrapv.
-----------------------------------------------------------------------------
HARDWARE
--------
CPU: Intel(R) Core(TM)2 CPU 6400 @ 2.13GHz
CPU MHz: 2128.001
FPU: Integrated
CPU(s) enabled: 2
Secondary Cache: 2048 KB
Memory: 2053792 kB
SOFTWARE
--------
Operating System: Linux 2.6.18-1.2868.fc6
Compiler: GNU C version 4.3.0 20070101 (experimental) (x86_64-unknow
n-linux-gnu)
NOTES
-----
Base is -O2 -march=nocona -mtune=generic -fwrapv
Peak is -O2 -march=nocona -mtune=generic
SPEC CINT2000 Summary
Estimated Estimated
Base Base Base Peak Peak Peak
Benchmarks Ref Time Run Time Ratio Ref Time Run Time Ratio
------------ -------- -------- -------- -------- -------- --------
164.gzip 1400 127 1099* 1400 127 1106*
175.vpr 1400 114 1227* 1400 116 1209*
176.gcc X X
181.mcf 1800 195 921* 1800 194 927*
186.crafty 1000 48.7 2054* 1000 48.2 2076*
197.parser 1800 197 915* 1800 195 923*
252.eon 1300 64.7 2011* 1300 64.8 2005*
253.perlbmk X X
254.gap 1100 64.9 1696* 1100 65.3 1685*
255.vortex X X
256.bzip2 1500 108 1384* 1500 108 1395*
300.twolf 3000 169 1771* 3000 169 1775*
Est. SPECint_base2000 1391
Est. SPECint2000 1394
SPEC CFP2000 Summary
Estimated Estimated
Base Base Base Peak Peak Peak
Benchmarks Ref Time Run Time Ratio Ref Time Run Time Ratio
------------ -------- -------- -------- -------- -------- --------
168.wupwise 1600 103 1547* 1600 88.3 1812*
171.swim 3100 152 2033* 3100 145 2131*
172.mgrid 1800 193 935* 1800 131 1376*
173.applu 2100 195 1076* 2100 188 1116*
177.mesa 1400 71.1 1968* 1400 71.3 1964*
178.galgel 2900 107 2699* 2900 99.1 2927*
179.art 2600 96.7 2689* 2600 94.6 2749*
183.equake 1300 69.0 1884* 1300 67.1 1939*
187.facerec 1900 149 1273* 1900 146 1302*
188.ammp 2200 170 1292* 2200 168 1312*
189.lucas 2000 102 1965* 2000 98.7 2025*
191.fma3d 2100 195 1079* 2100 192 1092*
200.sixtrack 1100 199 553* 1100 198 556*
301.apsi 2600 208 1248* 2600 196 1329*
Est. SPECfp_base2000 1463
Est. SPECfp2000 1560
-----------------------------------------------------------------------------
