Implementation notes: amd64, haswell, crypto_dh/surf127eps

Computer: haswell
Architecture: amd64
CPU ID: GenuineIntel-000306c3-bfebfbff
SUPERCOP version: 20130419
Operation: crypto_dh
Primitive: surf127eps

Time	Implementation	Compiler	Benchmark date	SUPERCOP version
889236	`mpfq`	`gcc -funroll-loops -fno-schedule-insns -O2 -fomit-frame-pointer`	20130609	20130419
891852	`mpfq`	`gcc -m64 -march=k8 -O3 -fomit-frame-pointer`	20130609	20130419
914708	`mpfq`	`gcc -m64 -march=corei7-avx -O -fomit-frame-pointer`	20130609	20130419
915282	`mpfq`	`gcc -O3 -fomit-frame-pointer`	20130609	20130419
916139	`mpfq`	`gcc -funroll-loops -march=k8 -O3 -fomit-frame-pointer`	20130609	20130419
917875	`mpfq`	`gcc -funroll-loops -march=nocona -O2 -fomit-frame-pointer`	20130609	20130419
925953	`mpfq`	`gcc -m64 -march=core2 -msse4.1 -O2 -fomit-frame-pointer`	20130609	20130419
927744	`mpfq`	`gcc -m64 -march=core2 -msse4 -O2 -fomit-frame-pointer`	20130609	20130419
930734	`mpfq`	`gcc -m64 -march=nocona -O2 -fomit-frame-pointer`	20130609	20130419
930830	`mpfq`	`gcc -m64 -march=barcelona -O3 -fomit-frame-pointer`	20130609	20130419
934140	`mpfq`	`gcc -funroll-loops -m64 -march=barcelona -O -fomit-frame-pointer`	20130609	20130419
934160	`mpfq`	`gcc -march=barcelona -O2 -fomit-frame-pointer`	20130609	20130419
935024	`mpfq`	`clang -O3 -mavx`	20130609	20130419
936116	`mpfq`	`gcc -O2 -fomit-frame-pointer`	20130609	20130419
939778	`mpfq`	`gcc -m64 -march=core2 -msse4.1 -O -fomit-frame-pointer`	20130609	20130419
940992	`mpfq`	`clang -O3`	20130609	20130419
941008	`mpfq`	`gcc -m64 -march=corei7 -O -fomit-frame-pointer`	20130609	20130419
943234	`mpfq`	`gcc -funroll-loops -O -fomit-frame-pointer`	20130609	20130419
943952	`mpfq`	`gcc -funroll-loops -m64 -march=k8 -O3 -fomit-frame-pointer`	20130609	20130419
944184	`mpfq`	`gcc -funroll-loops -m64 -march=barcelona -O3 -fomit-frame-pointer`	20130609	20130419
945056	`mpfq`	`gcc -m64 -march=corei7 -O3 -fomit-frame-pointer`	20130609	20130419
946866	`mpfq`	`gcc -fno-schedule-insns -O -fomit-frame-pointer`	20130609	20130419
947824	`mpfq`	`gcc -m64 -march=nocona -O -fomit-frame-pointer`	20130609	20130419
948300	`mpfq`	`gcc -funroll-loops -march=barcelona -O3 -fomit-frame-pointer`	20130609	20130419
951060	`mpfq`	`gcc -funroll-loops -m64 -march=k8 -O2 -fomit-frame-pointer`	20130609	20130419
952640	`mpfq`	`gcc -march=k8 -O2 -fomit-frame-pointer`	20130609	20130419
956628	`mpfq`	`gcc -funroll-loops -march=barcelona -O -fomit-frame-pointer`	20130609	20130419
956692	`mpfq`	`gcc -fno-schedule-insns -O3 -fomit-frame-pointer`	20130609	20130419
956716	`mpfq`	`gcc -funroll-loops -fno-schedule-insns -O3 -fomit-frame-pointer`	20130609	20130419
958124	`mpfq`	`gcc -funroll-loops -m64 -march=nocona -O3 -fomit-frame-pointer`	20130609	20130419
958400	`mpfq`	`gcc -funroll-loops -march=barcelona -O2 -fomit-frame-pointer`	20130609	20130419
958720	`mpfq`	`gcc -funroll-loops -O3 -fomit-frame-pointer`	20130609	20130419
959968	`mpfq`	`gcc -m64 -march=nocona -O3 -fomit-frame-pointer`	20130609	20130419
960680	`mpfq`	`gcc -m64 -march=core-avx-i -O2 -fomit-frame-pointer`	20130609	20130419
961332	`mpfq`	`gcc -m64 -march=native -mtune=native -O2 -fomit-frame-pointer`	20130609	20130419
961536	`mpfq`	`gcc -m64 -march=k8 -O2 -fomit-frame-pointer`	20130609	20130419
962748	`mpfq`	`gcc -m64 -march=corei7-avx -O2 -fomit-frame-pointer`	20130609	20130419
966144	`mpfq`	`gcc -funroll-loops -fno-schedule-insns -O -fomit-frame-pointer`	20130609	20130419
967616	`mpfq`	`gcc -m64 -march=core2 -O2 -fomit-frame-pointer`	20130609	20130419
970596	`mpfq`	`gcc -funroll-loops -march=nocona -O -fomit-frame-pointer`	20130609	20130419
975980	`mpfq`	`gcc -march=nocona -O3 -fomit-frame-pointer`	20130609	20130419
985288	`mpfq`	`gcc -march=nocona -O -fomit-frame-pointer`	20130609	20130419
985362	`mpfq`	`gcc -funroll-loops -m64 -march=barcelona -O2 -fomit-frame-pointer`	20130609	20130419
985800	`mpfq`	`gcc -m64 -march=k8 -O -fomit-frame-pointer`	20130609	20130419
987566	`mpfq`	`gcc -m64 -march=core2 -O3 -fomit-frame-pointer`	20130609	20130419
988296	`mpfq`	`gcc -m64 -march=core2 -msse4.1 -O3 -fomit-frame-pointer`	20130609	20130419
989058	`mpfq`	`gcc -m64 -march=barcelona -O2 -fomit-frame-pointer`	20130609	20130419
991580	`mpfq`	`gcc -march=k8 -O -fomit-frame-pointer`	20130609	20130419
992264	`mpfq`	`gcc -m64 -O3 -fomit-frame-pointer`	20130609	20130419
994004	`mpfq`	`gcc -m64 -march=core2 -msse4 -O3 -fomit-frame-pointer`	20130609	20130419
997106	`mpfq`	`gcc -funroll-loops -O2 -fomit-frame-pointer`	20130609	20130419
997576	`mpfq`	`gcc -funroll-loops -march=k8 -O2 -fomit-frame-pointer`	20130609	20130419
998026	`mpfq`	`gcc -m64 -O2 -fomit-frame-pointer`	20130609	20130419
1001458	`mpfq`	`gcc -m64 -march=corei7 -O2 -fomit-frame-pointer`	20130609	20130419
1006726	`mpfq`	`gcc -funroll-loops -m64 -march=nocona -O -fomit-frame-pointer`	20130609	20130419
1007158	`mpfq`	`gcc -m64 -march=barcelona -O -fomit-frame-pointer`	20130609	20130419
1008528	`mpfq`	`gcc -march=nocona -O2 -fomit-frame-pointer`	20130609	20130419
1008890	`mpfq`	`gcc -funroll-loops -march=k8 -O -fomit-frame-pointer`	20130609	20130419
1010572	`mpfq`	`gcc -O -fomit-frame-pointer`	20130609	20130419
1010986	`mpfq`	`gcc -m64 -march=core2 -msse4 -O -fomit-frame-pointer`	20130609	20130419
1011072	`mpfq`	`gcc -funroll-loops -m64 -O -fomit-frame-pointer`	20130609	20130419
1015046	`mpfq`	`gcc -march=barcelona -O -fomit-frame-pointer`	20130609	20130419
1016818	`mpfq`	`gcc -m64 -O -fomit-frame-pointer`	20130609	20130419
1017092	`mpfq`	`gcc -march=barcelona -O3 -fomit-frame-pointer`	20130609	20130419
1018176	`mpfq`	`gcc -march=k8 -O3 -fomit-frame-pointer`	20130609	20130419
1022166	`mpfq`	`gcc -m64 -march=core-avx-i -O -fomit-frame-pointer`	20130609	20130419
1027988	`mpfq`	`gcc -m64 -march=native -mtune=native -O -fomit-frame-pointer`	20130609	20130419
1029672	`mpfq`	`gcc -fno-schedule-insns -O2 -fomit-frame-pointer`	20130609	20130419
1043804	`mpfq`	`gcc -funroll-loops -m64 -march=nocona -O2 -fomit-frame-pointer`	20130609	20130419
1046428	`mpfq`	`gcc -m64 -march=core2 -O -fomit-frame-pointer`	20130609	20130419
1058760	`mpfq`	`gcc -funroll-loops -m64 -march=k8 -O -fomit-frame-pointer`	20130609	20130419
1064748	`mpfq`	`gcc -funroll-loops -m64 -O3 -fomit-frame-pointer`	20130609	20130419
1069860	`mpfq`	`gcc -m64 -Os -fomit-frame-pointer`	20130609	20130419
1072925	`mpfq`	`gcc -funroll-loops -m64 -O2 -fomit-frame-pointer`	20130609	20130419
1083291	`mpfq`	`gcc -funroll-loops -fno-schedule-insns -Os -fomit-frame-pointer`	20130609	20130419
1095736	`mpfq`	`gcc -m64 -march=core2 -msse4 -Os -fomit-frame-pointer`	20130609	20130419
1096920	`mpfq`	`gcc -m64 -march=corei7 -Os -fomit-frame-pointer`	20130609	20130419
1103694	`mpfq`	`gcc -funroll-loops -m64 -Os -fomit-frame-pointer`	20130609	20130419
1116147	`mpfq`	`gcc -funroll-loops -march=nocona -O3 -fomit-frame-pointer`	20130609	20130419
1136152	`mpfq`	`gcc -m64 -march=k8 -Os -fomit-frame-pointer`	20130609	20130419
1137924	`mpfq`	`gcc -m64 -march=core2 -msse4.1 -Os -fomit-frame-pointer`	20130609	20130419
1138736	`mpfq`	`gcc -m64 -march=core-avx-i -Os -fomit-frame-pointer`	20130609	20130419
1140100	`mpfq`	`gcc -funroll-loops -m64 -march=barcelona -Os -fomit-frame-pointer`	20130609	20130419
1140200	`mpfq`	`gcc -m64 -march=nocona -Os -fomit-frame-pointer`	20130609	20130419
1142280	`mpfq`	`gcc -m64 -march=corei7-avx -Os -fomit-frame-pointer`	20130609	20130419
1142832	`mpfq`	`gcc -funroll-loops -march=k8 -Os -fomit-frame-pointer`	20130609	20130419
1147880	`mpfq`	`gcc -m64 -march=native -mtune=native -Os -fomit-frame-pointer`	20130609	20130419
1148528	`mpfq`	`gcc -funroll-loops -m64 -march=nocona -Os -fomit-frame-pointer`	20130609	20130419
1149952	`mpfq`	`gcc -fno-schedule-insns -Os -fomit-frame-pointer`	20130609	20130419
1152480	`mpfq`	`gcc -funroll-loops -m64 -march=k8 -Os -fomit-frame-pointer`	20130609	20130419
1172470	`mpfq`	`gcc -m64 -march=barcelona -Os -fomit-frame-pointer`	20130609	20130419
1172548	`mpfq`	`gcc -funroll-loops -march=nocona -Os -fomit-frame-pointer`	20130609	20130419
1175802	`mpfq`	`gcc -Os -fomit-frame-pointer`	20130609	20130419
1177088	`mpfq`	`gcc -march=barcelona -Os -fomit-frame-pointer`	20130609	20130419
1178220	`mpfq`	`gcc -march=nocona -Os -fomit-frame-pointer`	20130609	20130419
1184144	`mpfq`	`gcc -funroll-loops -Os -fomit-frame-pointer`	20130609	20130419
1202336	`mpfq`	`gcc -march=k8 -Os -fomit-frame-pointer`	20130609	20130419
1205925	`mpfq`	`gcc -m64 -march=core2 -Os -fomit-frame-pointer`	20130609	20130419
1237876	`mpfq`	`gcc -funroll-loops -march=barcelona -Os -fomit-frame-pointer`	20130609	20130419

Compiler output

Implementation: crypto_dh/surf127eps/mpfq
Compiler: cc

surf127eps.c: In file included from mpfq_p_127_735.h:5:0,
surf127eps.c: from surf127eps.c:18:
surf127eps.c: x86_64/mpfq_p_127_735.h: In function 'mpfq_p_127_735_hadamard':
surf127eps.c: x86_64/mpfq_p_127_735.h:360:3: error: 'asm' operand has impossible constraints
surf127eps.c: __asm__ volatile (
surf127eps.c: ^

Number of similar (compiler,implementation) pairs: 3, namely:

Compiler	Implementations
cc	mpfq
gcc	mpfq
gcc -funroll-loops	mpfq

Compiler output

Implementation: crypto_dh/surf127eps/mpfq
Compiler: clang -O3

mpfq_p_127_735.c: In file included from mpfq_p_127_735.c:5:
mpfq_p_127_735.c: ./x86_64/mpfq_p_127_735.c:111:1: warning: control reaches end of non-void function [-Wreturn-type]
mpfq_p_127_735.c: }
mpfq_p_127_735.c: ^
mpfq_p_127_735.c: 1 warning generated.

Number of similar (compiler,implementation) pairs: 2, namely:

Compiler	Implementations
clang -O3	mpfq
clang -O3 -mavx	mpfq

Compiler output

Implementation: crypto_dh/surf127eps/mpfq
Compiler: gcc -m64 -march=core-avx-i -O3 -fomit-frame-pointer

surf127eps.c: In file included from mpfq_p_127_735.h:5:0,
surf127eps.c: from surf127eps.c:18:
surf127eps.c: surf127eps.c: In function 'KShadamard':
surf127eps.c: x86_64/mpfq_p_127_735.h:360:3: error: 'asm' operand has impossible constraints
surf127eps.c: __asm__ volatile (
surf127eps.c: ^

Number of similar (compiler,implementation) pairs: 3, namely:

Compiler	Implementations
gcc -m64 -march=core-avx-i -O3 -fomit-frame-pointer	mpfq
gcc -m64 -march=corei7-avx -O3 -fomit-frame-pointer	mpfq
gcc -m64 -march=native -mtune=native -O3 -fomit-frame-pointer	mpfq