Implementation notes: amd64, phenom, crypto_hash/sarmal256

Computer: phenom
Architecture: amd64
CPU ID: AuthenticAMD-00100fa0-178bfbff
SUPERCOP version: 20120812
Operation: crypto_hash
Primitive: sarmal256

Time	Implementation	Compiler	Benchmark date	SUPERCOP version
11504	`opt64`	`gcc -funroll-loops -march=k8 -O -fomit-frame-pointer`	20120523	20120521
11561	`opt64`	`gcc -m64 -march=corei7 -O -fomit-frame-pointer`	20120523	20120521
12846	`opt64`	`gcc -fno-schedule-insns -O2 -fomit-frame-pointer`	20120523	20120521
13106	`opt64`	`gcc -funroll-loops -fno-schedule-insns -O -fomit-frame-pointer`	20120523	20120521
13111	`opt64`	`gcc -funroll-loops -march=nocona -O -fomit-frame-pointer`	20120523	20120521
13118	`opt64`	`gcc -funroll-loops -m64 -O -fomit-frame-pointer`	20120523	20120521
13127	`opt64`	`gcc -funroll-loops -m64 -march=nocona -O -fomit-frame-pointer`	20120523	20120521
13433	`opt64`	`gcc -funroll-loops -march=barcelona -O -fomit-frame-pointer`	20120523	20120521
13462	`opt64`	`gcc -funroll-loops -O -fomit-frame-pointer`	20120523	20120521
13464	`opt64`	`gcc -funroll-loops -m64 -march=k8 -O -fomit-frame-pointer`	20120523	20120521
13472	`opt64`	`gcc -m64 -march=nocona -O -fomit-frame-pointer`	20120523	20120521
13477	`opt64`	`gcc -march=nocona -O -fomit-frame-pointer`	20120523	20120521
13480	`opt64`	`gcc -funroll-loops -m64 -march=barcelona -O -fomit-frame-pointer`	20120523	20120521
13576	`opt64`	`gcc -m64 -O -fomit-frame-pointer`	20120523	20120521
13608	`opt64`	`gcc -m64 -march=core2 -msse4.1 -O -fomit-frame-pointer`	20120523	20120521
13615	`opt64`	`gcc -fno-schedule-insns -O -fomit-frame-pointer`	20120523	20120521
13615	`opt64`	`gcc -m64 -march=core2 -msse4 -O -fomit-frame-pointer`	20120523	20120521
13618	`opt64`	`gcc -m64 -march=core2 -O -fomit-frame-pointer`	20120523	20120521
13620	`opt64`	`gcc -O -fomit-frame-pointer`	20120523	20120521
13871	`asm64`	`gcc -funroll-loops -m64 -march=nocona -O -fomit-frame-pointer`	20120523	20120521
13875	`asm64`	`gcc -funroll-loops -march=nocona -O2 -fomit-frame-pointer`	20120523	20120521
13923	`asm64`	`gcc -march=nocona -O -fomit-frame-pointer`	20120523	20120521
13933	`opt64`	`gcc -funroll-loops -march=nocona -O3 -fomit-frame-pointer`	20120523	20120521
13972	`asm64`	`gcc -m64 -march=nocona -O -fomit-frame-pointer`	20120523	20120521
13977	`opt64`	`gcc -funroll-loops -m64 -march=nocona -O3 -fomit-frame-pointer`	20120523	20120521
13987	`asm64`	`gcc -march=barcelona -O3 -fomit-frame-pointer`	20120523	20120521
14015	`asm64`	`gcc -O -fomit-frame-pointer`	20120523	20120521
14016	`asm64`	`gcc -m64 -march=core2 -msse4 -O -fomit-frame-pointer`	20120523	20120521
14036	`asm64`	`gcc -m64 -O -fomit-frame-pointer`	20120523	20120521
14055	`opt64`	`gcc -funroll-loops -march=nocona -O2 -fomit-frame-pointer`	20120523	20120521
14061	`opt64`	`gcc -funroll-loops -m64 -O3 -fomit-frame-pointer`	20120523	20120521
14063	`opt64`	`gcc -funroll-loops -fno-schedule-insns -O3 -fomit-frame-pointer`	20120523	20120521
14070	`opt64`	`gcc -funroll-loops -O3 -fomit-frame-pointer`	20120523	20120521
14098	`asm64`	`gcc -m64 -march=corei7 -O -fomit-frame-pointer`	20120523	20120521
14105	`opt64`	`gcc -funroll-loops -O2 -fomit-frame-pointer`	20120523	20120521
14130	`opt64`	`gcc -funroll-loops -m64 -O2 -fomit-frame-pointer`	20120523	20120521
14139	`opt64`	`gcc -funroll-loops -fno-schedule-insns -O2 -fomit-frame-pointer`	20120523	20120521
14157	`asm64`	`gcc -march=barcelona -O -fomit-frame-pointer`	20120523	20120521
14160	`asm64`	`gcc -m64 -march=native -mtune=native -O2 -fomit-frame-pointer`	20120523	20120521
14166	`opt64`	`gcc -march=nocona -O3 -fomit-frame-pointer`	20120523	20120521
14181	`opt64`	`gcc -m64 -march=nocona -O3 -fomit-frame-pointer`	20120523	20120521
14182	`opt64`	`gcc -funroll-loops -m64 -march=nocona -O2 -fomit-frame-pointer`	20120523	20120521
14196	`opt64`	`gcc -m64 -march=barcelona -O -fomit-frame-pointer`	20120523	20120521
14205	`opt64`	`gcc -m64 -march=k8 -O -fomit-frame-pointer`	20120523	20120521
14226	`opt64`	`gcc -march=barcelona -O -fomit-frame-pointer`	20120523	20120521
14242	`opt64`	`gcc -march=k8 -O -fomit-frame-pointer`	20120523	20120521
14265	`opt64`	`gcc -funroll-loops -march=barcelona -O3 -fomit-frame-pointer`	20120523	20120521
14275	`opt64`	`gcc -m64 -march=core2 -O3 -fomit-frame-pointer`	20120523	20120521
14296	`opt64`	`gcc -funroll-loops -m64 -march=barcelona -O3 -fomit-frame-pointer`	20120523	20120521
14314	`opt64`	`gcc -funroll-loops -m64 -march=k8 -O3 -fomit-frame-pointer`	20120523	20120521
14319	`opt64`	`gcc -funroll-loops -m64 -march=k8 -O2 -fomit-frame-pointer`	20120523	20120521
14338	`opt64`	`gcc -funroll-loops -march=k8 -O3 -fomit-frame-pointer`	20120523	20120521
14342	`opt64`	`gcc -funroll-loops -m64 -march=barcelona -O2 -fomit-frame-pointer`	20120523	20120521
14344	`asm64`	`gcc -m64 -Os -fomit-frame-pointer`	20120523	20120521
14363	`opt64`	`gcc -m64 -march=native -mtune=native -O -fomit-frame-pointer`	20120523	20120521
14366	`opt64`	`gcc -funroll-loops -march=barcelona -O2 -fomit-frame-pointer`	20120523	20120521
14371	`opt64`	`gcc -funroll-loops -march=k8 -O2 -fomit-frame-pointer`	20120523	20120521
14382	`asm64`	`gcc -fno-schedule-insns -Os -fomit-frame-pointer`	20120523	20120521
14386	`asm64`	`gcc -Os -fomit-frame-pointer`	20120523	20120521
14423	`opt64`	`gcc -m64 -O3 -fomit-frame-pointer`	20120523	20120521
14426	`opt64`	`gcc -fno-schedule-insns -O3 -fomit-frame-pointer`	20120523	20120521
14476	`opt64`	`gcc -O3 -fomit-frame-pointer`	20120523	20120521
14491	`asm64`	`gcc -march=k8 -Os -fomit-frame-pointer`	20120523	20120521
14500	`asm64`	`gcc -funroll-loops -m64 -march=barcelona -Os -fomit-frame-pointer`	20120523	20120521
14562	`opt64`	`gcc -m64 -march=native -mtune=native -O3 -fomit-frame-pointer`	20120523	20120521
14606	`opt64`	`gcc -m64 -march=barcelona -O3 -fomit-frame-pointer`	20120523	20120521
14624	`opt64`	`gcc -m64 -march=k8 -O3 -fomit-frame-pointer`	20120523	20120521
14633	`opt64`	`gcc -march=k8 -O3 -fomit-frame-pointer`	20120523	20120521
14819	`opt64`	`gcc -m64 -march=nocona -O2 -fomit-frame-pointer`	20120523	20120521
14850	`opt64`	`gcc -m64 -march=core2 -msse4.1 -O2 -fomit-frame-pointer`	20120523	20120521
14852	`opt64`	`gcc -march=barcelona -O3 -fomit-frame-pointer`	20120523	20120521
14882	`opt64`	`gcc -m64 -march=corei7 -O2 -fomit-frame-pointer`	20120523	20120521
14886	`opt64`	`gcc -m64 -march=core2 -O2 -fomit-frame-pointer`	20120523	20120521
14920	`opt64`	`gcc -march=nocona -O2 -fomit-frame-pointer`	20120523	20120521
15053	`opt64`	`gcc -m64 -march=core2 -msse4 -O2 -fomit-frame-pointer`	20120523	20120521
15161	`opt64`	`gcc -m64 -O2 -fomit-frame-pointer`	20120523	20120521
15218	`opt64`	`gcc -O2 -fomit-frame-pointer`	20120523	20120521
15396	`opt64`	`gcc -m64 -march=native -mtune=native -O2 -fomit-frame-pointer`	20120523	20120521
15399	`opt64`	`gcc -march=barcelona -O2 -fomit-frame-pointer`	20120523	20120521
15407	`opt64`	`gcc -m64 -march=barcelona -O2 -fomit-frame-pointer`	20120523	20120521
15424	`opt64`	`gcc -m64 -march=k8 -O2 -fomit-frame-pointer`	20120523	20120521
15434	`opt64`	`gcc -march=k8 -O2 -fomit-frame-pointer`	20120523	20120521
15613	`opt64`	`gcc -funroll-loops -march=k8 -Os -fomit-frame-pointer`	20120523	20120521
15620	`opt64`	`gcc -funroll-loops -m64 -march=k8 -Os -fomit-frame-pointer`	20120523	20120521
15620	`opt64`	`gcc -funroll-loops -march=barcelona -Os -fomit-frame-pointer`	20120523	20120521
15680	`opt64`	`gcc -funroll-loops -Os -fomit-frame-pointer`	20120523	20120521
15696	`opt64`	`gcc -funroll-loops -fno-schedule-insns -Os -fomit-frame-pointer`	20120523	20120521
15735	`opt64`	`gcc -funroll-loops -m64 -march=barcelona -Os -fomit-frame-pointer`	20120523	20120521
15823	`opt64`	`gcc -funroll-loops -m64 -Os -fomit-frame-pointer`	20120523	20120521
15833	`opt64`	`gcc -funroll-loops -m64 -march=nocona -Os -fomit-frame-pointer`	20120523	20120521
15872	`opt64`	`gcc -funroll-loops -march=nocona -Os -fomit-frame-pointer`	20120523	20120521
15950	`opt64`	`gcc -m64 -march=nocona -Os -fomit-frame-pointer`	20120523	20120521
16021	`opt64`	`gcc -Os -fomit-frame-pointer`	20120523	20120521
16044	`opt64`	`gcc -fno-schedule-insns -Os -fomit-frame-pointer`	20120523	20120521
16049	`opt64`	`gcc -m64 -march=k8 -Os -fomit-frame-pointer`	20120523	20120521
16051	`opt64`	`gcc -m64 -Os -fomit-frame-pointer`	20120523	20120521
16055	`opt64`	`gcc -march=barcelona -Os -fomit-frame-pointer`	20120523	20120521
16121	`opt64`	`gcc -m64 -march=core2 -msse4 -Os -fomit-frame-pointer`	20120523	20120521
16155	`opt64`	`gcc -m64 -march=corei7 -Os -fomit-frame-pointer`	20120523	20120521
16156	`opt64`	`gcc -m64 -march=core2 -Os -fomit-frame-pointer`	20120523	20120521
16156	`opt64`	`gcc -march=nocona -Os -fomit-frame-pointer`	20120523	20120521
16166	`opt64`	`gcc -m64 -march=core2 -msse4.1 -Os -fomit-frame-pointer`	20120523	20120521
16196	`opt64`	`gcc -m64 -march=barcelona -Os -fomit-frame-pointer`	20120523	20120521
16223	`opt64`	`gcc -m64 -march=native -mtune=native -Os -fomit-frame-pointer`	20120523	20120521
16245	`asm64`	`gcc -m64 -march=nocona -O3 -fomit-frame-pointer`	20120523	20120521
16274	`asm64`	`gcc -march=nocona -O3 -fomit-frame-pointer`	20120523	20120521
16308	`asm64`	`gcc -funroll-loops -m64 -march=nocona -O3 -fomit-frame-pointer`	20120523	20120521
16313	`asm64`	`gcc -funroll-loops -march=nocona -O3 -fomit-frame-pointer`	20120523	20120521
16318	`asm64`	`gcc -funroll-loops -m64 -march=nocona -O2 -fomit-frame-pointer`	20120523	20120521
16388	`asm64`	`gcc -funroll-loops -march=nocona -O -fomit-frame-pointer`	20120523	20120521
16409	`asm64`	`gcc -m64 -march=native -mtune=native -O3 -fomit-frame-pointer`	20120523	20120521
16426	`asm64`	`gcc -funroll-loops -m64 -march=barcelona -O3 -fomit-frame-pointer`	20120523	20120521
16434	`asm64`	`gcc -funroll-loops -march=barcelona -O3 -fomit-frame-pointer`	20120523	20120521
16450	`opt64`	`gcc -march=k8 -Os -fomit-frame-pointer`	20120523	20120521
16512	`asm64`	`gcc -fno-schedule-insns -O -fomit-frame-pointer`	20120523	20120521
16516	`asm64`	`gcc -m64 -march=core2 -msse4.1 -O -fomit-frame-pointer`	20120523	20120521
16519	`asm64`	`gcc -m64 -march=barcelona -O3 -fomit-frame-pointer`	20120523	20120521
16520	`asm64`	`gcc -m64 -march=core2 -O -fomit-frame-pointer`	20120523	20120521
16538	`asm64`	`gcc -funroll-loops -march=barcelona -O2 -fomit-frame-pointer`	20120523	20120521
16565	`asm64`	`gcc -march=nocona -O2 -fomit-frame-pointer`	20120523	20120521
16574	`asm64`	`gcc -funroll-loops -march=barcelona -O -fomit-frame-pointer`	20120523	20120521
16577	`asm64`	`gcc -funroll-loops -m64 -march=barcelona -O -fomit-frame-pointer`	20120523	20120521
16580	`asm64`	`gcc -m64 -march=barcelona -O2 -fomit-frame-pointer`	20120523	20120521
16580	`asm64`	`gcc -m64 -march=nocona -O2 -fomit-frame-pointer`	20120523	20120521
16611	`asm64`	`gcc -march=barcelona -O2 -fomit-frame-pointer`	20120523	20120521
16633	`asm64`	`gcc -m64 -march=barcelona -O -fomit-frame-pointer`	20120523	20120521
16732	`asm64`	`gcc -m64 -march=native -mtune=native -O -fomit-frame-pointer`	20120523	20120521
16744	`asm64`	`gcc -funroll-loops -m64 -march=barcelona -O2 -fomit-frame-pointer`	20120523	20120521
16898	`asm64`	`gcc -m64 -march=corei7 -Os -fomit-frame-pointer`	20120523	20120521
16912	`asm64`	`gcc -funroll-loops -m64 -march=k8 -Os -fomit-frame-pointer`	20120523	20120521
16919	`asm64`	`gcc -funroll-loops -march=barcelona -Os -fomit-frame-pointer`	20120523	20120521
16924	`asm64`	`gcc -funroll-loops -fno-schedule-insns -Os -fomit-frame-pointer`	20120523	20120521
16924	`asm64`	`gcc -m64 -march=core2 -Os -fomit-frame-pointer`	20120523	20120521
16926	`asm64`	`gcc -funroll-loops -m64 -Os -fomit-frame-pointer`	20120523	20120521
16929	`asm64`	`gcc -funroll-loops -Os -fomit-frame-pointer`	20120523	20120521
16954	`asm64`	`gcc -funroll-loops -march=nocona -Os -fomit-frame-pointer`	20120523	20120521
16954	`asm64`	`gcc -m64 -march=barcelona -Os -fomit-frame-pointer`	20120523	20120521
16963	`asm64`	`gcc -funroll-loops -m64 -march=nocona -Os -fomit-frame-pointer`	20120523	20120521
16992	`asm64`	`gcc -march=nocona -Os -fomit-frame-pointer`	20120523	20120521
16996	`asm64`	`gcc -m64 -march=nocona -Os -fomit-frame-pointer`	20120523	20120521
17015	`asm64`	`gcc -march=barcelona -Os -fomit-frame-pointer`	20120523	20120521
17016	`asm64`	`gcc -m64 -march=core2 -msse4.1 -Os -fomit-frame-pointer`	20120523	20120521
17077	`asm64`	`gcc -m64 -march=native -mtune=native -Os -fomit-frame-pointer`	20120523	20120521
17092	`asm64`	`gcc -funroll-loops -march=k8 -Os -fomit-frame-pointer`	20120523	20120521
17099	`asm64`	`gcc -m64 -march=core2 -msse4 -Os -fomit-frame-pointer`	20120523	20120521
17472	`asm64`	`gcc -m64 -march=k8 -Os -fomit-frame-pointer`	20120523	20120521
17934	`asm64`	`cc`	20120523	20120521
17976	`asm64`	`gcc -funroll-loops`	20120523	20120521
17981	`asm64`	`gcc`	20120523	20120521
35689	`opt64`	`gcc -funroll-loops`	20120523	20120521
35692	`opt64`	`gcc`	20120523	20120521
35698	`opt64`	`cc`	20120523	20120521

Checksum failure

Implementation: crypto_hash/sarmal256/asm64
Compiler: gcc -m64 -march=k8 -O2 -fomit-frame-pointer

3eb9be88f1271c693d908177f324500a7befe6930fdff907df8c1fffbd8fe0ba

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

Compiler	Implementations
gcc -m64 -march=k8 -O2 -fomit-frame-pointer	asm64
gcc -march=k8 -O2 -fomit-frame-pointer	asm64

Test failure

Implementation: crypto_hash/sarmal256/asm64
Compiler: gcc -O2 -fomit-frame-pointer

error 111

Number of similar (compiler,implementation) pairs: 25, namely:

Compiler	Implementations
gcc -O2 -fomit-frame-pointer	asm64
gcc -O3 -fomit-frame-pointer	asm64
gcc -fno-schedule-insns -O2 -fomit-frame-pointer	asm64
gcc -fno-schedule-insns -O3 -fomit-frame-pointer	asm64
gcc -funroll-loops -O2 -fomit-frame-pointer	asm64
gcc -funroll-loops -O3 -fomit-frame-pointer	asm64
gcc -funroll-loops -fno-schedule-insns -O2 -fomit-frame-pointer	asm64
gcc -funroll-loops -fno-schedule-insns -O3 -fomit-frame-pointer	asm64
gcc -funroll-loops -m64 -O2 -fomit-frame-pointer	asm64
gcc -funroll-loops -m64 -O3 -fomit-frame-pointer	asm64
gcc -funroll-loops -m64 -march=k8 -O2 -fomit-frame-pointer	asm64
gcc -funroll-loops -m64 -march=k8 -O3 -fomit-frame-pointer	asm64
gcc -funroll-loops -march=k8 -O2 -fomit-frame-pointer	asm64
gcc -funroll-loops -march=k8 -O3 -fomit-frame-pointer	asm64
gcc -m64 -O2 -fomit-frame-pointer	asm64
gcc -m64 -O3 -fomit-frame-pointer	asm64
gcc -m64 -march=core2 -O2 -fomit-frame-pointer	asm64
gcc -m64 -march=core2 -O3 -fomit-frame-pointer	asm64
gcc -m64 -march=core2 -msse4.1 -O2 -fomit-frame-pointer	asm64
gcc -m64 -march=core2 -msse4 -O2 -fomit-frame-pointer	asm64
gcc -m64 -march=corei7 -O2 -fomit-frame-pointer	asm64
gcc -m64 -march=k8 -O3 -fomit-frame-pointer	asm64
gcc -m64 -march=k8 -O -fomit-frame-pointer	asm64
gcc -march=k8 -O3 -fomit-frame-pointer	asm64
gcc -march=k8 -O -fomit-frame-pointer	asm64

Test failure

Implementation: crypto_hash/sarmal256/asm64
Compiler: gcc -funroll-loops -O -fomit-frame-pointer

error 111
crypto_hash does not handle overlap

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

Compiler	Implementations
gcc -funroll-loops -O -fomit-frame-pointer	asm64
gcc -funroll-loops -fno-schedule-insns -O -fomit-frame-pointer	asm64
gcc -funroll-loops -m64 -O -fomit-frame-pointer	asm64

Test failure

Implementation: crypto_hash/sarmal256/asm64
Compiler: gcc -funroll-loops -m64 -march=k8 -O -fomit-frame-pointer

error 142
Alarm clock

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

Compiler	Implementations
gcc -funroll-loops -m64 -march=k8 -O -fomit-frame-pointer	asm64
gcc -funroll-loops -march=k8 -O -fomit-frame-pointer	asm64