Implementation notes: amd64, mangetsu, crypto_dh

Computer: mangetsu
Architecture: amd64
CPU ID: GenuineIntel-000206a7-bfebfbff
CPU cycles/second: 3400000000...3400000000 (amd64cpuinfo)
SUPERCOP version: 20141014
Benchmark dates: 20140911...20141018

crypto_dh

Time	Relative time	Primitive	Implementation	Compiler
8252628	1.00	`claus`	`crypto_dh/claus/cryptopp`	`g++ -m64 -march=k8 -O3 -fomit-frame-pointer (4.8.1 20130909 [gcc-4 8-branch revision 202388])`
		`claus`	`crypto_dh/claus/gmp`	`g++ -m64 -O3 -fomit-frame-pointer`
		`claus`	`crypto_dh/claus/ntl`	`g++ -m64 -O3 -fomit-frame-pointer`
		`claus`	`crypto_dh/claus/openssl`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
584868	1.00	`curve2251`	`crypto_dh/curve2251/relic/amd64-avx`	`gcc-4.7 -m64 -march=native -mtune=native -O3 -fomit-frame-pointer (4.7.4)`
656056	1.12	`curve2251`	`crypto_dh/curve2251/relic/amd64-clmul`	`gcc-4.7 -funroll-loops -march=k8 -O3 -fomit-frame-pointer`
1217056	2.08	`curve2251`	`crypto_dh/curve2251/relic/amd64-ssse3`	`gcc-4.7 -m64 -march=core2 -msse4 -O3 -fomit-frame-pointer`
		`curve2251`	`crypto_dh/curve2251/mpfq`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
777020	1.00	`curve25519`	`crypto_dh/curve25519/ref`	`gcc-4.7 -funroll-loops -fno-schedule-insns -O3 -fomit-frame-pointer (4.7.4)`
		`curve25519`	`crypto_dh/curve25519/mpfq`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
605380	1.00	`ecfp256e`	`crypto_dh/ecfp256e/v01/w8s8`	`gcc-4.7 -O3 -fomit-frame-pointer (4.7.4)`
611992	1.01	`ecfp256e`	`crypto_dh/ecfp256e/v01/w8s4`	`gcc-4.7 -m64 -march=core2 -O3 -fomit-frame-pointer`
623564	1.03	`ecfp256e`	`crypto_dh/ecfp256e/v01/w8s2`	`gcc-4.7 -O3 -fomit-frame-pointer (4.7.4)`
649560	1.07	`ecfp256e`	`crypto_dh/ecfp256e/v01/w8s1`	`gcc-4.7 -m64 -march=corei7 -O3 -fomit-frame-pointer`
997820	1.65	`ecfp256e`	`crypto_dh/ecfp256e/v01/var`	`gcc-4.7 -O3 -fomit-frame-pointer (4.7.4)`
716080	1.00	`ecfp256h`	`crypto_dh/ecfp256h/v01/w8s8`	`gcc-4.7 -fno-schedule-insns -O3 -fomit-frame-pointer (4.7.4)`
723676	1.01	`ecfp256h`	`crypto_dh/ecfp256h/v01/w8s4`	`gcc-4.7 -O3 -fomit-frame-pointer`
735656	1.03	`ecfp256h`	`crypto_dh/ecfp256h/v01/w8s2`	`gcc-4.7 -fno-schedule-insns -O3 -fomit-frame-pointer (4.7.4)`
764392	1.07	`ecfp256h`	`crypto_dh/ecfp256h/v01/w8s1`	`gcc-4.7 -m64 -march=core2 -msse4 -O3 -fomit-frame-pointer`
1210704	1.69	`ecfp256h`	`crypto_dh/ecfp256h/v01/var`	`gcc-4.7 -m64 -march=k8 -O3 -fomit-frame-pointer`
1045940	1.00	`ecfp256i`	`crypto_dh/ecfp256i/v01/var`	`gcc-4.7 -m64 -march=core2 -O3 -fomit-frame-pointer (4.7.4)`
		`ecfp256i`	`crypto_dh/ecfp256i/v01/w8s1`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
		`ecfp256i`	`crypto_dh/ecfp256i/v01/w8s2`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
		`ecfp256i`	`crypto_dh/ecfp256i/v01/w8s4`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
		`ecfp256i`	`crypto_dh/ecfp256i/v01/w8s8`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
635704	1.00	`ecfp256q`	`crypto_dh/ecfp256q/v01/w8s8`	`gcc-4.7 -m64 -march=core2 -msse4 -O3 -fomit-frame-pointer (4.7.4)`
640816	1.01	`ecfp256q`	`crypto_dh/ecfp256q/v01/w8s4`	`gcc-4.7 -m64 -march=core2 -O3 -fomit-frame-pointer`
652648	1.03	`ecfp256q`	`crypto_dh/ecfp256q/v01/w8s2`	`gcc-4.7 -m64 -march=core2 -msse4 -O3 -fomit-frame-pointer (4.7.4)`
678000	1.07	`ecfp256q`	`crypto_dh/ecfp256q/v01/w8s1`	`gcc-4.7 -m64 -march=core2 -msse4 -O3 -fomit-frame-pointer (4.7.4)`
1024592	1.61	`ecfp256q`	`crypto_dh/ecfp256q/v01/var`	`gcc-4.7 -O3 -fomit-frame-pointer`
761856	1.00	`ecfp256s`	`crypto_dh/ecfp256s/v01/w8s8`	`gcc-4.7 -m64 -march=core2 -msse4 -O3 -fomit-frame-pointer (4.7.4)`
770636	1.01	`ecfp256s`	`crypto_dh/ecfp256s/v01/w8s4`	`gcc-4.7 -O3 -fomit-frame-pointer`
784112	1.03	`ecfp256s`	`crypto_dh/ecfp256s/v01/w8s2`	`gcc-4.7 -m64 -march=core2 -O3 -fomit-frame-pointer`
813712	1.07	`ecfp256s`	`crypto_dh/ecfp256s/v01/w8s1`	`gcc-4.7 -O3 -fomit-frame-pointer`
1292220	1.70	`ecfp256s`	`crypto_dh/ecfp256s/v01/var`	`gcc-4.8 -fno-schedule-insns -O3 -fomit-frame-pointer`
1754608	1.00	`ed448goldilocks`	`crypto_dh/ed448goldilocks/amd64 (2014-10-12)`	`clang -O3 -fomit-frame-pointer (4.2.1 Compatible Clang 3.3 (branches/release 33 183898))`
2276080	1.30	`ed448goldilocks`	`crypto_dh/ed448goldilocks/64`	`gcc-4.7 -funroll-loops -m64 -march=k8 -O2 -fomit-frame-pointer`
3842920	2.19	`ed448goldilocks`	`crypto_dh/ed448goldilocks/arm32`	`clang -O3 -fomit-frame-pointer (4.2.1 Compatible Clang 3.3 (branches/release 33 183898))`
5822496	3.32	`ed448goldilocks`	`crypto_dh/ed448goldilocks/32`	`gcc-4.8 -m64 -march=barcelona -O3 -fomit-frame-pointer`
		`ed448goldilocks`	`crypto_dh/ed448goldilocks/neon`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
708256	1.00	`gls1271`	`crypto_dh/gls1271/ref4`	`gcc-4.7 -fno-schedule-insns -O3 -fomit-frame-pointer (4.7.4)`
311328	1.00	`gls254`	`crypto_dh/gls254/ref`	`clang -O3 -fwrapv -march=native -fomit-frame-pointer (4.2.1 Compatible Clang 3.3 (branches/release 33 183898))`
463172	1.49	`gls254`	`crypto_dh/gls254/prot`	`gcc-4.8 -m64 -march=corei7-avx -O3 -fomit-frame-pointer`
463256	1.00	`gls254prot`	`crypto_dh/gls254prot/prot`	`gcc-4.8 -m64 -march=native -mtune=native -O3 -fomit-frame-pointer (4.8.1 20130909 [gcc-4 8-branch revision 202388])`
604348	1.00	`hecfp127i`	`crypto_dh/hecfp127i/v02/w8s32`	`gcc-4.8 -m64 -march=core2 -msse4.1 -O3 -fomit-frame-pointer (4.8.1 20130909 [gcc-4 8-branch revision 202388])`
605172	1.00	`hecfp127i`	`crypto_dh/hecfp127i/v02/w8s16`	`gcc-4.8 -m64 -march=k8 -O3 -fomit-frame-pointer`
608228	1.01	`hecfp127i`	`crypto_dh/hecfp127i/v02/w8s08`	`gcc-4.8 -march=nocona -O2 -fomit-frame-pointer`
614692	1.02	`hecfp127i`	`crypto_dh/hecfp127i/v02/w8s04`	`gcc-4.8 -m64 -march=barcelona -O2 -fomit-frame-pointer`
627256	1.04	`hecfp127i`	`crypto_dh/hecfp127i/v02/w8s02`	`gcc-4.8 -m64 -march=k8 -O3 -fomit-frame-pointer`
654140	1.08	`hecfp127i`	`crypto_dh/hecfp127i/v02/w8s01`	`gcc-4.8 -m64 -march=k8 -O3 -fomit-frame-pointer`
1069380	1.77	`hecfp127i`	`crypto_dh/hecfp127i/v02/var`	`gcc-4.8 -march=k8 -O2 -fomit-frame-pointer`
483980	1.00	`hecfp128bk`	`crypto_dh/hecfp128bk/v02/w8s32glv4`	`gcc-4.7 -m64 -march=k8 -O2 -fomit-frame-pointer (4.7.4)`
486204	1.00	`hecfp128bk`	`crypto_dh/hecfp128bk/v02/w8s16glv4`	`gcc-4.8 -m64 -march=nocona -O2 -fomit-frame-pointer`
488848	1.01	`hecfp128bk`	`crypto_dh/hecfp128bk/v02/w8s08glv4`	`gcc-4.7 -march=barcelona -O2 -fomit-frame-pointer`
497224	1.03	`hecfp128bk`	`crypto_dh/hecfp128bk/v02/w8s04glv4`	`gcc-4.7 -O3 -fomit-frame-pointer`
517180	1.07	`hecfp128bk`	`crypto_dh/hecfp128bk/v02/w8s02glv4`	`gcc-4.8 -march=barcelona -O2 -fomit-frame-pointer`
556560	1.15	`hecfp128bk`	`crypto_dh/hecfp128bk/v02/w8s01glv4`	`gcc-4.7 -march=nocona -O3 -fomit-frame-pointer`
755872	1.56	`hecfp128bk`	`crypto_dh/hecfp128bk/v02/varglv4`	`gcc-4.8 -m64 -march=barcelona -O2 -fomit-frame-pointer`
483164	1.00	`hecfp128fkt`	`crypto_dh/hecfp128fkt/v02/w8s32glv4`	`gcc-4.8 -m64 -march=core2 -O3 -fomit-frame-pointer (4.8.1 20130909 [gcc-4 8-branch revision 202388])`
483424	1.00	`hecfp128fkt`	`crypto_dh/hecfp128fkt/v02/w8s16glv4`	`gcc-4.7 -m64 -march=core2 -msse4 -O3 -fomit-frame-pointer`
487572	1.01	`hecfp128fkt`	`crypto_dh/hecfp128fkt/v02/w8s08glv4`	`gcc-4.7 -fno-schedule-insns -O3 -fomit-frame-pointer`
496004	1.03	`hecfp128fkt`	`crypto_dh/hecfp128fkt/v02/w8s04glv4`	`gcc-4.7 -march=barcelona -O2 -fomit-frame-pointer`
515208	1.07	`hecfp128fkt`	`crypto_dh/hecfp128fkt/v02/w8s02glv4`	`gcc-4.8 -march=k8 -O3 -fomit-frame-pointer`
555596	1.15	`hecfp128fkt`	`crypto_dh/hecfp128fkt/v02/w8s01glv4`	`gcc-4.8 -m64 -march=barcelona -O2 -fomit-frame-pointer`
752584	1.56	`hecfp128fkt`	`crypto_dh/hecfp128fkt/v02/varglv4`	`gcc-4.7 -march=barcelona -O2 -fomit-frame-pointer`
936440	1.00	`hecfp128i`	`crypto_dh/hecfp128i/v02/w8s32`	`gcc-4.7 -O3 -fomit-frame-pointer (4.7.4)`
936492	1.00	`hecfp128i`	`crypto_dh/hecfp128i/v02/w8s16`	`gcc-4.7 -m64 -march=nocona -O3 -fomit-frame-pointer`
940772	1.00	`hecfp128i`	`crypto_dh/hecfp128i/v02/w8s08`	`gcc-4.7 -m64 -march=k8 -O3 -fomit-frame-pointer`
951408	1.02	`hecfp128i`	`crypto_dh/hecfp128i/v02/w8s04`	`gcc-4.7 -fno-schedule-insns -O3 -fomit-frame-pointer`
972016	1.04	`hecfp128i`	`crypto_dh/hecfp128i/v02/w8s02`	`gcc-4.7 -m64 -march=core2 -msse4 -O3 -fomit-frame-pointer`
1012252	1.08	`hecfp128i`	`crypto_dh/hecfp128i/v02/w8s01`	`gcc-4.7 -march=k8 -O3 -fomit-frame-pointer`
1660332	1.77	`hecfp128i`	`crypto_dh/hecfp128i/v02/var`	`gcc-4.8 -O2 -fomit-frame-pointer`
		`hecfp61e2bk`	`crypto_dh/hecfp61e2bk/v01/varglv8`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
		`hecfp61e2bk`	`crypto_dh/hecfp61e2bk/v01/w8s01glv8`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
		`hecfp61e2bk`	`crypto_dh/hecfp61e2bk/v01/w8s02glv8`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
		`hecfp61e2bk`	`crypto_dh/hecfp61e2bk/v01/w8s04glv8`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
		`hecfp61e2bk`	`crypto_dh/hecfp61e2bk/v01/w8s08glv8`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
		`hecfp61e2bk`	`crypto_dh/hecfp61e2bk/v01/w8s16glv8`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
		`hecfp61e2bk`	`crypto_dh/hecfp61e2bk/v01/w8s32glv8`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
		`hecfp61e2i`	`crypto_dh/hecfp61e2i/v01/var`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
		`hecfp61e2i`	`crypto_dh/hecfp61e2i/v01/w8s01`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
		`hecfp61e2i`	`crypto_dh/hecfp61e2i/v01/w8s02`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
		`hecfp61e2i`	`crypto_dh/hecfp61e2i/v01/w8s04`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
		`hecfp61e2i`	`crypto_dh/hecfp61e2i/v01/w8s08`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
		`hecfp61e2i`	`crypto_dh/hecfp61e2i/v01/w8s16`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
		`hecfp61e2i`	`crypto_dh/hecfp61e2i/v01/w8s32`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
		`hecfp64e2bk`	`crypto_dh/hecfp64e2bk/v01/varglv8`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
		`hecfp64e2bk`	`crypto_dh/hecfp64e2bk/v01/w8s01glv8`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
		`hecfp64e2bk`	`crypto_dh/hecfp64e2bk/v01/w8s02glv8`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
		`hecfp64e2bk`	`crypto_dh/hecfp64e2bk/v01/w8s04glv8`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
		`hecfp64e2bk`	`crypto_dh/hecfp64e2bk/v01/w8s08glv8`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
		`hecfp64e2bk`	`crypto_dh/hecfp64e2bk/v01/w8s16glv8`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
		`hecfp64e2bk`	`crypto_dh/hecfp64e2bk/v01/w8s32glv8`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
		`hecfp64e2i`	`crypto_dh/hecfp64e2i/v01/var`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
		`hecfp64e2i`	`crypto_dh/hecfp64e2i/v01/w8s01`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
		`hecfp64e2i`	`crypto_dh/hecfp64e2i/v01/w8s02`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
		`hecfp64e2i`	`crypto_dh/hecfp64e2i/v01/w8s04`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
		`hecfp64e2i`	`crypto_dh/hecfp64e2i/v01/w8s08`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
		`hecfp64e2i`	`crypto_dh/hecfp64e2i/v01/w8s16`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
		`hecfp64e2i`	`crypto_dh/hecfp64e2i/v01/w8s32`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
		`hector`	`crypto_dh/hector/ref`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
484468	1.00	`kumfp127g`	`crypto_dh/kumfp127g/v02/var`	`gcc-4.8 -m64 -march=nocona -O3 -fomit-frame-pointer (4.8.1 20130909 [gcc-4 8-branch revision 202388])`
752652	1.00	`kumfp128g`	`crypto_dh/kumfp128g/v02/var`	`gcc-4.7 -march=nocona -Os -fomit-frame-pointer (4.7.4)`
		`kumfp61e2g`	`crypto_dh/kumfp61e2g/v01/var`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
		`kumfp64e2g`	`crypto_dh/kumfp64e2g/v01/var`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
357264	1.00	`kummer`	`crypto_dh/kummer/ref`	`gcc-4.7 -funroll-loops -m64 -O2 -fomit-frame-pointer (4.7.4)`
3107908	1.00	`nistp256`	`crypto_dh/nistp256/ref`	`gcc-4.7 -m64 -march=core2 -msse4.1 -O3 -fomit-frame-pointer (4.7.4)`
3390928	1.09	`nistp256`	`crypto_dh/nistp256/wbl`	`clang -O3 -fwrapv -mavx -fomit-frame-pointer`
1471600	1.00	`sclaus1024`	`crypto_dh/sclaus1024/cryptopp`	`g++ -march=barcelona -O3 -fomit-frame-pointer (4.8.1 20130909 [gcc-4 8-branch revision 202388])`
		`sclaus1024`	`crypto_dh/sclaus1024/gmp`	`g++ -m64 -O3 -fomit-frame-pointer`
7272728	1.00	`sclaus2048`	`crypto_dh/sclaus2048/cryptopp`	`g++ -m64 -march=native -mtune=native -O3 -fomit-frame-pointer (4.8.1 20130909 [gcc-4 8-branch revision 202388])`
		`sclaus2048`	`crypto_dh/sclaus2048/gmp`	`g++ -m64 -O3 -fomit-frame-pointer`
		`surf127eps`	`crypto_dh/surf127eps/mpfq`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
		`surf2113`	`crypto_dh/surf2113/mpfq`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`

crypto_scalarmult

Time	Relative time	Primitive	Implementation	Compiler
775304	1.00	`curve25519`	`crypto_scalarmult/curve25519/amd64-51`	`gcc-4.8 -funroll-loops -m64 -O -fomit-frame-pointer (4.8.1 20130909 [gcc-4 8-branch revision 202388])`
805640	1.04	`curve25519`	`crypto_scalarmult/curve25519/donna_c64`	`gcc-4.7 -m64 -march=corei7 -O3 -fomit-frame-pointer`
812012	1.05	`curve25519`	`crypto_scalarmult/curve25519/amd64-64`	`clang -O3 -fwrapv -mavx -fpolly -funroll-loops -fvectorize -fslp-vectorize -fslp-vectorize-aggressive -fomit-frame-pointer`
1435636	1.85	`curve25519`	`crypto_scalarmult/curve25519/donna`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
1705156	2.20	`curve25519`	`crypto_scalarmult/curve25519/ref10`	`gcc-4.7 -funroll-loops -march=barcelona -Os -fomit-frame-pointer`
26729196	34.48	`curve25519`	`crypto_scalarmult/curve25519/ref`	`clang -O3 -fwrapv -mavx -maes -mpclmul -fpolly -funroll-loops -fvectorize -fslp-vectorize -fslp-vectorize-aggressive -fomit-frame-pointer`
355492	1.00	`kummer`	`crypto_scalarmult/kummer/avx`	`clang -O3 -fwrapv -march=x86-64 -mcpu=core-avx2 -mavx2 -maes -mpclmul -fomit-frame-pointer (4.2.1 Compatible Clang 3.3 (branches/release 33 183898))`
1188628	3.34	`kummer`	`crypto_scalarmult/kummer/ref5`	`clang -O3 -fwrapv -march=native -fpolly -funroll-loops -fvectorize -fslp-vectorize -fslp-vectorize-aggressive -fomit-frame-pointer`
1274736	3.59	`kummer`	`crypto_scalarmult/kummer/ref5u`	`clang -O3 -fwrapv -mavx -fpolly -funroll-loops -fvectorize -fslp-vectorize -fslp-vectorize-aggressive -fomit-frame-pointer`
		`kummer`	`crypto_scalarmult/kummer/avx2`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
		`kummer`	`crypto_scalarmult/kummer/avx2int`	`gcc-4.7 -m64 -O3 -fomit-frame-pointer`
3110184	1.00	`nistp256`	`crypto_scalarmult/nistp256/mj32`	`gcc-4.7 -funroll-loops -march=nocona -O3 -fomit-frame-pointer (4.7.4)`