Implementation notes: x86, gpu, crypto_dh

Computer: gpu
Architecture: x86
CPU ID: GenuineIntel-000206d7-bfebfbff
CPU cycles/second: 3800000000...3800000000 (x86cpuinfo)
SUPERCOP version: 20190816
Benchmark dates: 20190820...20191015

crypto_dh

Time	Relative time	Primitive	Implementation	Compiler
27662004	1.00	`claus`	`cryptopp`	`g++ -m32 -march=k6-3 -O3 -fomit-frame-pointer (7.4.0)`
44050356	1.59	`claus`	`ntl`	`g++ -m32 -march=k6 -O3 -fomit-frame-pointer`
46860248	1.69	`claus`	`gmp`	`g++ -m32 -march=native -mtune=native -O -fomit-frame-pointer`
		`claus`	`openssl`	`gcc -m32 -O3 -fomit-frame-pointer`
8392756	1.00	`curve2251`	`mpfq`	`gcc -funroll-loops -m32 -march=pentium-m -O -fomit-frame-pointer (7.4.0)`
4575420	1.00	`curve25519`	`ref`	`gcc -funroll-loops -m32 -march=i386 -O2 -fomit-frame-pointer (7.4.0)`
7700304	1.68	`curve25519`	`mpfq`	`gcc -m32 -march=native -mtune=native -O3 -fomit-frame-pointer`
17592296	1.00	`ed448goldilocks`	`32`	`gcc -m32 -march=core2 -msse4 -O2 -fomit-frame-pointer (7.4.0)`
		`ed448goldilocks`	`64`	`gcc -m32 -O3 -fomit-frame-pointer`
		`ed448goldilocks`	`amd64`	`gcc -m32 -O3 -fomit-frame-pointer`
		`ed521gs`	`ref`	`gcc -m32 -O3 -fomit-frame-pointer`
		`gls254`	`opt`	`gcc -m32 -O3 -fomit-frame-pointer`
		`gls254prot`	`opt`	`gcc -m32 -O3 -fomit-frame-pointer`
		`hecfp61e2bk`	`v01/varglv8`	`gcc -m32 -O3 -fomit-frame-pointer`
		`hecfp61e2bk`	`v01/w8s01glv8`	`gcc -m32 -O3 -fomit-frame-pointer`
		`hecfp61e2bk`	`v01/w8s02glv8`	`gcc -m32 -O3 -fomit-frame-pointer`
		`hecfp61e2bk`	`v01/w8s04glv8`	`gcc -m32 -O3 -fomit-frame-pointer`
		`hecfp61e2bk`	`v01/w8s08glv8`	`gcc -m32 -O3 -fomit-frame-pointer`
		`hecfp61e2bk`	`v01/w8s16glv8`	`gcc -m32 -O3 -fomit-frame-pointer`
		`hecfp61e2bk`	`v01/w8s32glv8`	`gcc -m32 -O3 -fomit-frame-pointer`
		`hecfp61e2i`	`v01/var`	`gcc -m32 -O3 -fomit-frame-pointer`
		`hecfp61e2i`	`v01/w8s01`	`gcc -m32 -O3 -fomit-frame-pointer`
		`hecfp61e2i`	`v01/w8s02`	`gcc -m32 -O3 -fomit-frame-pointer`
		`hecfp61e2i`	`v01/w8s04`	`gcc -m32 -O3 -fomit-frame-pointer`
		`hecfp61e2i`	`v01/w8s08`	`gcc -m32 -O3 -fomit-frame-pointer`
		`hecfp61e2i`	`v01/w8s16`	`gcc -m32 -O3 -fomit-frame-pointer`
		`hecfp61e2i`	`v01/w8s32`	`gcc -m32 -O3 -fomit-frame-pointer`
		`hecfp64e2bk`	`v01/varglv8`	`gcc -m32 -O3 -fomit-frame-pointer`
		`hecfp64e2bk`	`v01/w8s01glv8`	`gcc -m32 -O3 -fomit-frame-pointer`
		`hecfp64e2bk`	`v01/w8s02glv8`	`gcc -m32 -O3 -fomit-frame-pointer`
		`hecfp64e2bk`	`v01/w8s04glv8`	`gcc -m32 -O3 -fomit-frame-pointer`
		`hecfp64e2bk`	`v01/w8s08glv8`	`gcc -m32 -O3 -fomit-frame-pointer`
		`hecfp64e2bk`	`v01/w8s16glv8`	`gcc -m32 -O3 -fomit-frame-pointer`
		`hecfp64e2bk`	`v01/w8s32glv8`	`gcc -m32 -O3 -fomit-frame-pointer`
		`hecfp64e2i`	`v01/var`	`gcc -m32 -O3 -fomit-frame-pointer`
		`hecfp64e2i`	`v01/w8s01`	`gcc -m32 -O3 -fomit-frame-pointer`
		`hecfp64e2i`	`v01/w8s02`	`gcc -m32 -O3 -fomit-frame-pointer`
		`hecfp64e2i`	`v01/w8s04`	`gcc -m32 -O3 -fomit-frame-pointer`
		`hecfp64e2i`	`v01/w8s08`	`gcc -m32 -O3 -fomit-frame-pointer`
		`hecfp64e2i`	`v01/w8s16`	`gcc -m32 -O3 -fomit-frame-pointer`
		`hecfp64e2i`	`v01/w8s32`	`gcc -m32 -O3 -fomit-frame-pointer`
		`hector`	`ref`	`gcc -m32 -O3 -fomit-frame-pointer`
		`jacfp127i`	`v01/var`	`gcc -m32 -O3 -fomit-frame-pointer`
		`jacfp127i`	`v01/w4s01`	`gcc -m32 -O3 -fomit-frame-pointer`
		`jacfp127i`	`v01/w4s02`	`gcc -m32 -O3 -fomit-frame-pointer`
		`jacfp127i`	`v01/w4s04`	`gcc -m32 -O3 -fomit-frame-pointer`
		`jacfp127i`	`v01/w4s08`	`gcc -m32 -O3 -fomit-frame-pointer`
		`jacfp127i`	`v01/w4s16`	`gcc -m32 -O3 -fomit-frame-pointer`
		`jacfp127i`	`v01/w4s32`	`gcc -m32 -O3 -fomit-frame-pointer`
		`jacfp127i`	`v01/w8s01`	`gcc -m32 -O3 -fomit-frame-pointer`
		`jacfp127i`	`v01/w8s02`	`gcc -m32 -O3 -fomit-frame-pointer`
		`jacfp127i`	`v01/w8s04`	`gcc -m32 -O3 -fomit-frame-pointer`
		`jacfp127i`	`v01/w8s08`	`gcc -m32 -O3 -fomit-frame-pointer`
		`jacfp127i`	`v01/w8s16`	`gcc -m32 -O3 -fomit-frame-pointer`
		`jacfp127i`	`v01/w8s32`	`gcc -m32 -O3 -fomit-frame-pointer`
		`jacfp128bk`	`v01/varglv4`	`gcc -m32 -O3 -fomit-frame-pointer`
		`jacfp128bk`	`v01/w4s01glv4`	`gcc -m32 -O3 -fomit-frame-pointer`
		`jacfp128bk`	`v01/w4s02glv4`	`gcc -m32 -O3 -fomit-frame-pointer`
		`jacfp128bk`	`v01/w4s04glv4`	`gcc -m32 -O3 -fomit-frame-pointer`
		`jacfp128bk`	`v01/w4s08glv4`	`gcc -m32 -O3 -fomit-frame-pointer`
		`jacfp128bk`	`v01/w4s16glv4`	`gcc -m32 -O3 -fomit-frame-pointer`
		`jacfp128bk`	`v01/w4s32glv4`	`gcc -m32 -O3 -fomit-frame-pointer`
		`jacfp128bk`	`v01/w8s01glv4`	`gcc -m32 -O3 -fomit-frame-pointer`
		`jacfp128bk`	`v01/w8s02glv4`	`gcc -m32 -O3 -fomit-frame-pointer`
		`jacfp128bk`	`v01/w8s04glv4`	`gcc -m32 -O3 -fomit-frame-pointer`
		`jacfp128bk`	`v01/w8s08glv4`	`gcc -m32 -O3 -fomit-frame-pointer`
		`jacfp128bk`	`v01/w8s16glv4`	`gcc -m32 -O3 -fomit-frame-pointer`
		`jacfp128bk`	`v01/w8s32glv4`	`gcc -m32 -O3 -fomit-frame-pointer`
1844252	1.00	`k277mon`	`ref`	`gcc -m32 -march=core-avx-i -O3 -fomit-frame-pointer (7.4.0)`
		`k277taa`	`ref`	`gcc -m32 -O3 -fomit-frame-pointer`
		`k298`	`ref`	`gcc -m32 -O3 -fomit-frame-pointer`
		`kumfp61e2g`	`v01/var`	`gcc -m32 -O3 -fomit-frame-pointer`
		`kumfp64e2g`	`v01/var`	`gcc -m32 -O3 -fomit-frame-pointer`
		`kumjacfp127g`	`v01/var`	`gcc -m32 -O3 -fomit-frame-pointer`
		`kumjacfp127g`	`v01/w4s01`	`gcc -m32 -O3 -fomit-frame-pointer`
		`kumjacfp127g`	`v01/w4s02`	`gcc -m32 -O3 -fomit-frame-pointer`
		`kumjacfp127g`	`v01/w4s04`	`gcc -m32 -O3 -fomit-frame-pointer`
		`kumjacfp127g`	`v01/w4s08`	`gcc -m32 -O3 -fomit-frame-pointer`
		`kumjacfp127g`	`v01/w4s16`	`gcc -m32 -O3 -fomit-frame-pointer`
		`kumjacfp127g`	`v01/w4s32`	`gcc -m32 -O3 -fomit-frame-pointer`
		`kumjacfp127g`	`v01/w8s01`	`gcc -m32 -O3 -fomit-frame-pointer`
		`kumjacfp127g`	`v01/w8s02`	`gcc -m32 -O3 -fomit-frame-pointer`
		`kumjacfp127g`	`v01/w8s04`	`gcc -m32 -O3 -fomit-frame-pointer`
		`kumjacfp127g`	`v01/w8s08`	`gcc -m32 -O3 -fomit-frame-pointer`
		`kumjacfp127g`	`v01/w8s16`	`gcc -m32 -O3 -fomit-frame-pointer`
		`kumjacfp127g`	`v01/w8s32`	`gcc -m32 -O3 -fomit-frame-pointer`
3688840	1.00	`kummer`	`ref`	`gcc -funroll-loops -m32 -march=i486 -O2 -fomit-frame-pointer (7.4.0)`
		`nist521gs`	`ref`	`gcc -m32 -O3 -fomit-frame-pointer`
12868872	1.00	`nistp256`	`ref`	`gcc -funroll-loops -m32 -march=k8 -O2 -fomit-frame-pointer (7.4.0)`
		`nistp256`	`wbl`	`gcc -m32 -O3 -fomit-frame-pointer`
		`prjfp127i`	`v01/var`	`gcc -m32 -O3 -fomit-frame-pointer`
		`prjfp127i`	`v01/w4s01`	`gcc -m32 -O3 -fomit-frame-pointer`
		`prjfp127i`	`v01/w4s02`	`gcc -m32 -O3 -fomit-frame-pointer`
		`prjfp127i`	`v01/w4s04`	`gcc -m32 -O3 -fomit-frame-pointer`
		`prjfp127i`	`v01/w4s08`	`gcc -m32 -O3 -fomit-frame-pointer`
		`prjfp127i`	`v01/w4s16`	`gcc -m32 -O3 -fomit-frame-pointer`
		`prjfp127i`	`v01/w4s32`	`gcc -m32 -O3 -fomit-frame-pointer`
		`prjfp127i`	`v01/w8s01`	`gcc -m32 -O3 -fomit-frame-pointer`
		`prjfp127i`	`v01/w8s02`	`gcc -m32 -O3 -fomit-frame-pointer`
		`prjfp127i`	`v01/w8s04`	`gcc -m32 -O3 -fomit-frame-pointer`
		`prjfp127i`	`v01/w8s08`	`gcc -m32 -O3 -fomit-frame-pointer`
		`prjfp127i`	`v01/w8s16`	`gcc -m32 -O3 -fomit-frame-pointer`
		`prjfp127i`	`v01/w8s32`	`gcc -m32 -O3 -fomit-frame-pointer`
		`prjfp128bk`	`v01/varglv4`	`gcc -m32 -O3 -fomit-frame-pointer`
		`prjfp128bk`	`v01/w4s01glv4`	`gcc -m32 -O3 -fomit-frame-pointer`
		`prjfp128bk`	`v01/w4s02glv4`	`gcc -m32 -O3 -fomit-frame-pointer`
		`prjfp128bk`	`v01/w4s04glv4`	`gcc -m32 -O3 -fomit-frame-pointer`
		`prjfp128bk`	`v01/w4s08glv4`	`gcc -m32 -O3 -fomit-frame-pointer`
		`prjfp128bk`	`v01/w4s16glv4`	`gcc -m32 -O3 -fomit-frame-pointer`
		`prjfp128bk`	`v01/w4s32glv4`	`gcc -m32 -O3 -fomit-frame-pointer`
		`prjfp128bk`	`v01/w8s01glv4`	`gcc -m32 -O3 -fomit-frame-pointer`
		`prjfp128bk`	`v01/w8s02glv4`	`gcc -m32 -O3 -fomit-frame-pointer`
		`prjfp128bk`	`v01/w8s04glv4`	`gcc -m32 -O3 -fomit-frame-pointer`
		`prjfp128bk`	`v01/w8s08glv4`	`gcc -m32 -O3 -fomit-frame-pointer`
		`prjfp128bk`	`v01/w8s16glv4`	`gcc -m32 -O3 -fomit-frame-pointer`
		`prjfp128bk`	`v01/w8s32glv4`	`gcc -m32 -O3 -fomit-frame-pointer`
4843275	1.00	`sclaus1024`	`cryptopp`	`g++ -m32 -march=pentium-m -O -fomit-frame-pointer (7.4.0)`
6713288	1.39	`sclaus1024`	`gmp`	`g++ -m32 -march=pentiumpro -O2 -fomit-frame-pointer`
25840204	1.00	`sclaus2048`	`cryptopp`	`g++ -m32 -march=core2 -msse4 -Os -fomit-frame-pointer (7.4.0)`
32474452	1.26	`sclaus2048`	`gmp`	`g++ -m32 -march=nocona -O -fomit-frame-pointer`
8854888	1.00	`surf127eps`	`mpfq`	`gcc -funroll-loops -m32 -march=pentium3 -O3 -fomit-frame-pointer (7.4.0)`
3611079	1.00	`surf2113`	`mpfq`	`gcc -funroll-loops -m32 -march=pentium4 -O3 -fomit-frame-pointer (7.4.0)`