Implementation notes: amd64, bigmem, crypto_dh/ecfp256i

Computer: bigmem
Architecture: amd64
CPU ID: GenuineIntel-000206e6-bfebfbff
SUPERCOP version: 20141124
Operation: crypto_dh
Primitive: ecfp256i

Time	Implementation	Compiler	Benchmark date	SUPERCOP version
1097052	`v01/var`	`gcc -funroll-loops -march=k8 -O3 -fomit-frame-pointer`	20150216	20141124
1097899	`v01/var`	`gcc -m64 -march=core2 -msse4.1 -O2 -fomit-frame-pointer`	20150216	20141124
1098526	`v01/var`	`gcc -m64 -march=k8 -O2 -fomit-frame-pointer`	20150216	20141124
1098611	`v01/var`	`gcc -m64 -O2 -fomit-frame-pointer`	20150216	20141124
1098696	`v01/var`	`gcc -m64 -march=core2 -O2 -fomit-frame-pointer`	20150216	20141124
1099650	`v01/var`	`gcc -m64 -O3 -fomit-frame-pointer`	20150216	20141124
1099707	`v01/var`	`gcc -m64 -march=core2 -msse4 -O2 -fomit-frame-pointer`	20150216	20141124
1099937	`v01/var`	`gcc -funroll-loops -march=barcelona -O3 -fomit-frame-pointer`	20150216	20141124
1100322	`v01/var`	`gcc -funroll-loops -fno-schedule-insns -O3 -fomit-frame-pointer`	20150216	20141124
1100756	`v01/var`	`gcc -m64 -march=k8 -O3 -fomit-frame-pointer`	20150216	20141124
1101130	`v01/var`	`gcc -funroll-loops -O2 -fomit-frame-pointer`	20150216	20141124
1101444	`v01/var`	`gcc -O3 -fomit-frame-pointer`	20150216	20141124
1101532	`v01/var`	`gcc -march=k8 -O3 -fomit-frame-pointer`	20150216	20141124
1101577	`v01/var`	`gcc -O2 -fomit-frame-pointer`	20150216	20141124
1101640	`v01/var`	`gcc -funroll-loops -fno-schedule-insns -O2 -fomit-frame-pointer`	20150216	20141124
1101691	`v01/var`	`gcc -funroll-loops -m64 -O3 -fomit-frame-pointer`	20150216	20141124
1101694	`v01/var`	`gcc -funroll-loops -march=k8 -O2 -fomit-frame-pointer`	20150216	20141124
1101719	`v01/var`	`gcc -march=k8 -O2 -fomit-frame-pointer`	20150216	20141124
1101897	`v01/var`	`gcc -funroll-loops -O3 -fomit-frame-pointer`	20150216	20141124
1102054	`v01/var`	`gcc -fno-schedule-insns -O2 -fomit-frame-pointer`	20150216	20141124
1102156	`v01/var`	`gcc -funroll-loops -m64 -march=k8 -O2 -fomit-frame-pointer`	20150216	20141124
1102325	`v01/var`	`gcc -m64 -march=native -mtune=native -O3 -fomit-frame-pointer`	20150216	20141124
1102353	`v01/var`	`gcc -funroll-loops -m64 -march=k8 -O3 -fomit-frame-pointer`	20150216	20141124
1102547	`v01/var`	`gcc -funroll-loops -m64 -O2 -fomit-frame-pointer`	20150216	20141124
1102799	`v01/var`	`gcc -m64 -march=native -mtune=native -O2 -fomit-frame-pointer`	20150216	20141124
1102946	`v01/var`	`gcc -m64 -march=core2 -msse4.1 -O3 -fomit-frame-pointer`	20150216	20141124
1102986	`v01/var`	`gcc -funroll-loops -m64 -march=barcelona -O3 -fomit-frame-pointer`	20150216	20141124
1103065	`v01/var`	`gcc -m64 -march=barcelona -O2 -fomit-frame-pointer`	20150216	20141124
1103416	`v01/var`	`gcc -funroll-loops -m64 -march=barcelona -O2 -fomit-frame-pointer`	20150216	20141124
1103790	`v01/var`	`gcc -march=nocona -O2 -fomit-frame-pointer`	20150216	20141124
1103887	`v01/var`	`gcc -m64 -march=core2 -O3 -fomit-frame-pointer`	20150216	20141124
1103989	`v01/var`	`gcc -funroll-loops -march=barcelona -O2 -fomit-frame-pointer`	20150216	20141124
1105357	`v01/var`	`gcc -m64 -march=barcelona -O3 -fomit-frame-pointer`	20150216	20141124
1105898	`v01/var`	`gcc -m64 -march=nocona -O2 -fomit-frame-pointer`	20150216	20141124
1105921	`v01/var`	`gcc -march=barcelona -O2 -fomit-frame-pointer`	20150216	20141124
1107340	`v01/var`	`gcc -march=barcelona -O3 -fomit-frame-pointer`	20150216	20141124
1108295	`v01/var`	`gcc -funroll-loops -m64 -march=nocona -O2 -fomit-frame-pointer`	20150216	20141124
1109284	`v01/var`	`gcc -m64 -march=nocona -O3 -fomit-frame-pointer`	20150216	20141124
1109655	`v01/var`	`gcc -fno-schedule-insns -O3 -fomit-frame-pointer`	20150216	20141124
1110293	`v01/var`	`gcc -m64 -march=core2 -msse4 -O3 -fomit-frame-pointer`	20150216	20141124
1110471	`v01/var`	`gcc -funroll-loops -march=nocona -O2 -fomit-frame-pointer`	20150216	20141124
1111828	`v01/var`	`gcc -funroll-loops -march=nocona -O3 -fomit-frame-pointer`	20150216	20141124
1113517	`v01/var`	`gcc -march=nocona -O3 -fomit-frame-pointer`	20150216	20141124
1117251	`v01/var`	`gcc -m64 -march=core2 -msse4 -Os -fomit-frame-pointer`	20150216	20141124
1117277	`v01/var`	`gcc -funroll-loops -m64 -march=k8 -Os -fomit-frame-pointer`	20150216	20141124
1117736	`v01/var`	`gcc -m64 -march=native -mtune=native -Os -fomit-frame-pointer`	20150216	20141124
1118348	`v01/var`	`gcc -m64 -march=core2 -msse4.1 -Os -fomit-frame-pointer`	20150216	20141124
1118643	`v01/var`	`gcc -march=k8 -Os -fomit-frame-pointer`	20150216	20141124
1118821	`v01/var`	`gcc -Os -fomit-frame-pointer`	20150216	20141124
1119192	`v01/var`	`gcc -m64 -march=barcelona -Os -fomit-frame-pointer`	20150216	20141124
1119569	`v01/var`	`gcc -m64 -march=core2 -Os -fomit-frame-pointer`	20150216	20141124
1119589	`v01/var`	`gcc -march=barcelona -Os -fomit-frame-pointer`	20150216	20141124
1119886	`v01/var`	`gcc -funroll-loops -m64 -march=barcelona -Os -fomit-frame-pointer`	20150216	20141124
1120314	`v01/var`	`gcc -funroll-loops -march=k8 -Os -fomit-frame-pointer`	20150216	20141124
1120575	`v01/var`	`gcc -funroll-loops -march=nocona -Os -fomit-frame-pointer`	20150216	20141124
1120722	`v01/var`	`gcc -m64 -march=k8 -Os -fomit-frame-pointer`	20150216	20141124
1121019	`v01/var`	`gcc -fno-schedule-insns -Os -fomit-frame-pointer`	20150216	20141124
1121076	`v01/var`	`gcc -funroll-loops -m64 -march=nocona -O3 -fomit-frame-pointer`	20150216	20141124
1121167	`v01/var`	`gcc -funroll-loops -m64 -Os -fomit-frame-pointer`	20150216	20141124
1121354	`v01/var`	`gcc -m64 -Os -fomit-frame-pointer`	20150216	20141124
1121533	`v01/var`	`gcc -funroll-loops -march=barcelona -Os -fomit-frame-pointer`	20150216	20141124
1122643	`v01/var`	`gcc -funroll-loops -fno-schedule-insns -Os -fomit-frame-pointer`	20150216	20141124
1123825	`v01/var`	`gcc -march=nocona -Os -fomit-frame-pointer`	20150216	20141124
1124026	`v01/var`	`gcc -m64 -march=nocona -Os -fomit-frame-pointer`	20150216	20141124
1124572	`v01/var`	`gcc -funroll-loops -Os -fomit-frame-pointer`	20150216	20141124
1125065	`v01/var`	`gcc -funroll-loops -m64 -march=nocona -Os -fomit-frame-pointer`	20150216	20141124
1193420	`v01/var`	`gcc -m64 -march=nocona -O -fomit-frame-pointer`	20150216	20141124
1194652	`v01/var`	`gcc -m64 -march=core2 -msse4.1 -O -fomit-frame-pointer`	20150216	20141124
1194947	`v01/var`	`gcc -march=nocona -O -fomit-frame-pointer`	20150216	20141124
1195168	`v01/var`	`gcc -m64 -march=core2 -O -fomit-frame-pointer`	20150216	20141124
1195400	`v01/var`	`gcc -funroll-loops -m64 -march=barcelona -O -fomit-frame-pointer`	20150216	20141124
1196199	`v01/var`	`gcc -m64 -march=native -mtune=native -O -fomit-frame-pointer`	20150216	20141124
1197265	`v01/var`	`gcc -m64 -march=k8 -O -fomit-frame-pointer`	20150216	20141124
1197432	`v01/var`	`gcc -m64 -march=core2 -msse4 -O -fomit-frame-pointer`	20150216	20141124
1197505	`v01/var`	`gcc -march=k8 -O -fomit-frame-pointer`	20150216	20141124
1197939	`v01/var`	`gcc -fno-schedule-insns -O -fomit-frame-pointer`	20150216	20141124
1198347	`v01/var`	`gcc -m64 -march=barcelona -O -fomit-frame-pointer`	20150216	20141124
1199330	`v01/var`	`gcc -m64 -O -fomit-frame-pointer`	20150216	20141124
1199529	`v01/var`	`gcc -funroll-loops -m64 -march=nocona -O -fomit-frame-pointer`	20150216	20141124
1199656	`v01/var`	`gcc -O -fomit-frame-pointer`	20150216	20141124
1201121	`v01/var`	`gcc -funroll-loops -O -fomit-frame-pointer`	20150216	20141124
1201823	`v01/var`	`gcc -funroll-loops -m64 -march=k8 -O -fomit-frame-pointer`	20150216	20141124
1202928	`v01/var`	`gcc -funroll-loops -fno-schedule-insns -O -fomit-frame-pointer`	20150216	20141124
1203229	`v01/var`	`gcc -funroll-loops -m64 -O -fomit-frame-pointer`	20150216	20141124
1203691	`v01/var`	`gcc -funroll-loops -march=barcelona -O -fomit-frame-pointer`	20150216	20141124
1204377	`v01/var`	`gcc -funroll-loops -march=k8 -O -fomit-frame-pointer`	20150216	20141124
1205527	`v01/var`	`gcc -funroll-loops -march=nocona -O -fomit-frame-pointer`	20150216	20141124
1206762	`v01/var`	`gcc -march=barcelona -O -fomit-frame-pointer`	20150216	20141124
1392484	`v01/var`	`cc`	20150216	20141124
1395034	`v01/var`	`gcc`	20150216	20141124
1396037	`v01/var`	`gcc -funroll-loops`	20150216	20141124

Test failure

Implementation: crypto_dh/ecfp256i/v01/w8s1
Compiler: cc

error 111
crypto_dh_keypair writes after output

Number of similar (compiler,implementation) pairs: 348, namely:

Compiler	Implementations
cc	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -O2 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -O3 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -O -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -Os -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -fno-schedule-insns -O2 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -fno-schedule-insns -O3 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -fno-schedule-insns -O -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -fno-schedule-insns -Os -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -O2 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -O3 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -O -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -Os -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -fno-schedule-insns -O2 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -fno-schedule-insns -O3 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -fno-schedule-insns -O -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -fno-schedule-insns -Os -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -m64 -O2 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -m64 -O3 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -m64 -O -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -m64 -Os -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -m64 -march=barcelona -O2 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -m64 -march=barcelona -O3 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -m64 -march=barcelona -O -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -m64 -march=barcelona -Os -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -m64 -march=k8 -O2 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -m64 -march=k8 -O3 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -m64 -march=k8 -O -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -m64 -march=k8 -Os -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -m64 -march=nocona -O2 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -m64 -march=nocona -O3 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -m64 -march=nocona -O -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -m64 -march=nocona -Os -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -march=barcelona -O2 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -march=barcelona -O3 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -march=barcelona -O -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -march=barcelona -Os -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -march=k8 -O2 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -march=k8 -O3 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -march=k8 -O -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -march=k8 -Os -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -march=nocona -O2 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -march=nocona -O3 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -march=nocona -O -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -funroll-loops -march=nocona -Os -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -m64 -O2 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -m64 -O3 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -m64 -O -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -m64 -Os -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -m64 -march=core2 -O2 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -m64 -march=core2 -O3 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -m64 -march=core2 -O -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -m64 -march=core2 -Os -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -m64 -march=core2 -msse4.1 -O2 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -m64 -march=core2 -msse4.1 -O3 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -m64 -march=core2 -msse4.1 -O -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -m64 -march=core2 -msse4.1 -Os -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -m64 -march=core2 -msse4 -O2 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -m64 -march=core2 -msse4 -O3 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -m64 -march=core2 -msse4 -O -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -m64 -march=core2 -msse4 -Os -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -m64 -march=k8 -O2 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -m64 -march=k8 -O3 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -m64 -march=k8 -O -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -m64 -march=k8 -Os -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -m64 -march=native -mtune=native -O2 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -m64 -march=native -mtune=native -O3 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -m64 -march=native -mtune=native -O -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -m64 -march=native -mtune=native -Os -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -m64 -march=nocona -O2 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -m64 -march=nocona -O3 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -m64 -march=nocona -O -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -m64 -march=nocona -Os -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -march=barcelona -O2 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -march=barcelona -O3 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -march=barcelona -O -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -march=barcelona -Os -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -march=k8 -O2 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -march=k8 -O3 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -march=k8 -O -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -march=k8 -Os -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -march=nocona -O2 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -march=nocona -O3 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -march=nocona -O -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -march=nocona -Os -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8

Test failure

Implementation: crypto_dh/ecfp256i/v01/w8s1
Compiler: gcc -m64 -march=barcelona -O2 -fomit-frame-pointer

error 111
crypto_dh_keypair writes after output
error 111
crypto_dh_keypair writes after output

Number of similar (compiler,implementation) pairs: 16, namely:

Compiler	Implementations
gcc -m64 -march=barcelona -O2 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -m64 -march=barcelona -O3 -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -m64 -march=barcelona -O -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8
gcc -m64 -march=barcelona -Os -fomit-frame-pointer	v01/w8s1 v01/w8s2 v01/w8s4 v01/w8s8