Implementation notes: amd64, bulldozer, crypto_hash/sarmal512

Computer: bulldozer
Architecture: amd64
CPU ID: AuthenticAMD-00600f12-178bfbff
SUPERCOP version: 20120709
Operation: crypto_hash
Primitive: sarmal512

Time	Implementation	Compiler	Benchmark date	SUPERCOP version
23260	`asm64`	`gcc -m64 -Os -fomit-frame-pointer`	20120530	20120525
23349	`asm64`	`gcc -funroll-loops -march=nocona -O -fomit-frame-pointer`	20120530	20120525
23361	`asm64`	`gcc -funroll-loops -m64 -march=k8 -Os -fomit-frame-pointer`	20120530	20120525
23371	`asm64`	`gcc -m64 -march=nocona -O -fomit-frame-pointer`	20120530	20120525
23376	`asm64`	`gcc -m64 -march=core-avx-i -Os -fomit-frame-pointer`	20120530	20120525
23435	`asm64`	`gcc -m64 -march=core2 -msse4 -Os -fomit-frame-pointer`	20120530	20120525
23447	`asm64`	`gcc -funroll-loops -m64 -march=nocona -O -fomit-frame-pointer`	20120530	20120525
23507	`asm64`	`gcc -funroll-loops -march=nocona -O3 -fomit-frame-pointer`	20120530	20120525
23550	`asm64`	`gcc -m64 -march=core-avx-i -O -fomit-frame-pointer`	20120530	20120525
23571	`asm64`	`gcc -funroll-loops -m64 -Os -fomit-frame-pointer`	20120530	20120525
23594	`asm64`	`gcc -march=nocona -O3 -fomit-frame-pointer`	20120530	20120525
23609	`asm64`	`gcc -O -fomit-frame-pointer`	20120530	20120525
23633	`asm64`	`gcc -funroll-loops -march=nocona -O2 -fomit-frame-pointer`	20120530	20120525
23651	`asm64`	`gcc -m64 -march=core2 -msse4 -O -fomit-frame-pointer`	20120530	20120525
23782	`asm64`	`gcc -funroll-loops -m64 -march=nocona -O2 -fomit-frame-pointer`	20120530	20120525
23793	`opt64`	`gcc -funroll-loops -m64 -march=nocona -O2 -fomit-frame-pointer`	20120530	20120525
23805	`asm64`	`gcc -m64 -march=corei7 -O -fomit-frame-pointer`	20120530	20120525
23810	`asm64`	`gcc -funroll-loops -m64 -march=barcelona -O3 -fomit-frame-pointer`	20120530	20120525
23814	`asm64`	`gcc -march=nocona -O -fomit-frame-pointer`	20120530	20120525
23838	`asm64`	`gcc -fno-schedule-insns -O -fomit-frame-pointer`	20120530	20120525
23859	`asm64`	`gcc -m64 -march=nocona -O2 -fomit-frame-pointer`	20120530	20120525
23873	`asm64`	`gcc -funroll-loops -m64 -march=barcelona -O -fomit-frame-pointer`	20120530	20120525
23892	`asm64`	`gcc -funroll-loops -march=barcelona -O3 -fomit-frame-pointer`	20120530	20120525
23904	`asm64`	`gcc -m64 -O -fomit-frame-pointer`	20120530	20120525
23912	`asm64`	`gcc -m64 -march=native -mtune=native -O -fomit-frame-pointer`	20120530	20120525
24013	`asm64`	`gcc -m64 -march=core2 -msse4.1 -O -fomit-frame-pointer`	20120530	20120525
24025	`asm64`	`gcc -funroll-loops -march=barcelona -O -fomit-frame-pointer`	20120530	20120525
24055	`asm64`	`gcc -funroll-loops -m64 -march=nocona -O3 -fomit-frame-pointer`	20120530	20120525
24082	`opt64`	`gcc -funroll-loops -march=barcelona -O3 -fomit-frame-pointer`	20120530	20120525
24087	`asm64`	`gcc -m64 -march=native -mtune=native -O3 -fomit-frame-pointer`	20120530	20120525
24117	`asm64`	`gcc -funroll-loops -march=barcelona -O2 -fomit-frame-pointer`	20120530	20120525
24199	`opt64`	`gcc -funroll-loops -fno-schedule-insns -O3 -fomit-frame-pointer`	20120530	20120525
24234	`asm64`	`gcc -march=nocona -O2 -fomit-frame-pointer`	20120530	20120525
24239	`opt64`	`gcc -m64 -march=core2 -msse4 -O2 -fomit-frame-pointer`	20120530	20120525
24258	`asm64`	`gcc -funroll-loops -m64 -march=barcelona -O2 -fomit-frame-pointer`	20120530	20120525
24262	`asm64`	`gcc -m64 -march=core2 -O -fomit-frame-pointer`	20120530	20120525
24278	`asm64`	`gcc -m64 -march=corei7-avx -O -fomit-frame-pointer`	20120530	20120525
24329	`asm64`	`gcc -m64 -march=nocona -O3 -fomit-frame-pointer`	20120530	20120525
24437	`asm64`	`gcc -m64 -march=barcelona -O3 -fomit-frame-pointer`	20120530	20120525
24468	`asm64`	`gcc -funroll-loops -march=nocona -Os -fomit-frame-pointer`	20120530	20120525
24494	`asm64`	`gcc -m64 -march=native -mtune=native -O2 -fomit-frame-pointer`	20120530	20120525
24544	`asm64`	`gcc -march=barcelona -O3 -fomit-frame-pointer`	20120530	20120525
24561	`asm64`	`gcc -m64 -march=barcelona -O2 -fomit-frame-pointer`	20120530	20120525
24562	`asm64`	`gcc -funroll-loops -fno-schedule-insns -Os -fomit-frame-pointer`	20120530	20120525
24612	`asm64`	`gcc -march=barcelona -O2 -fomit-frame-pointer`	20120530	20120525
24630	`opt64`	`gcc -march=nocona -O -fomit-frame-pointer`	20120530	20120525
24680	`opt64`	`gcc -m64 -march=barcelona -O2 -fomit-frame-pointer`	20120530	20120525
24683	`opt64`	`gcc -m64 -march=k8 -O2 -fomit-frame-pointer`	20120530	20120525
24693	`asm64`	`gcc -m64 -march=corei7 -Os -fomit-frame-pointer`	20120530	20120525
24705	`asm64`	`gcc -march=barcelona -O -fomit-frame-pointer`	20120530	20120525
24709	`asm64`	`gcc -funroll-loops -m64 -march=barcelona -Os -fomit-frame-pointer`	20120530	20120525
24713	`asm64`	`gcc -march=nocona -Os -fomit-frame-pointer`	20120530	20120525
24727	`opt64`	`gcc -funroll-loops -m64 -march=nocona -O3 -fomit-frame-pointer`	20120530	20120525
24738	`opt64`	`gcc -m64 -march=core-avx-i -O2 -fomit-frame-pointer`	20120530	20120525
24759	`opt64`	`gcc -funroll-loops -march=nocona -O -fomit-frame-pointer`	20120530	20120525
24780	`opt64`	`gcc -m64 -march=core-avx-i -O -fomit-frame-pointer`	20120530	20120525
24781	`asm64`	`gcc -m64 -march=barcelona -O -fomit-frame-pointer`	20120530	20120525
24787	`asm64`	`gcc -m64 -march=barcelona -Os -fomit-frame-pointer`	20120530	20120525
24809	`asm64`	`gcc -m64 -march=k8 -Os -fomit-frame-pointer`	20120530	20120525
24864	`asm64`	`gcc -funroll-loops -Os -fomit-frame-pointer`	20120530	20120525
24871	`asm64`	`gcc -funroll-loops -march=k8 -Os -fomit-frame-pointer`	20120530	20120525
24981	`asm64`	`gcc -march=k8 -Os -fomit-frame-pointer`	20120530	20120525
25003	`opt64`	`gcc -m64 -march=core-avx-i -O3 -fomit-frame-pointer`	20120530	20120525
25013	`opt64`	`gcc -m64 -march=nocona -O2 -fomit-frame-pointer`	20120530	20120525
25016	`opt64`	`gcc -funroll-loops -fno-schedule-insns -O -fomit-frame-pointer`	20120530	20120525
25027	`asm64`	`gcc -fno-schedule-insns -Os -fomit-frame-pointer`	20120530	20120525
25030	`asm64`	`gcc -funroll-loops -m64 -march=nocona -Os -fomit-frame-pointer`	20120530	20120525
25048	`asm64`	`gcc -m64 -march=native -mtune=native -Os -fomit-frame-pointer`	20120530	20120525
25055	`opt64`	`gcc -m64 -O3 -fomit-frame-pointer`	20120530	20120525
25078	`opt64`	`gcc -funroll-loops -O -fomit-frame-pointer`	20120530	20120525
25090	`opt64`	`gcc -m64 -march=core2 -msse4 -O3 -fomit-frame-pointer`	20120530	20120525
25099	`asm64`	`cc`	20120530	20120525
25110	`opt64`	`gcc -m64 -march=core2 -O -fomit-frame-pointer`	20120530	20120525
25120	`opt64`	`gcc -funroll-loops -m64 -O2 -fomit-frame-pointer`	20120530	20120525
25135	`asm64`	`gcc -funroll-loops -march=barcelona -Os -fomit-frame-pointer`	20120530	20120525
25148	`asm64`	`gcc -Os -fomit-frame-pointer`	20120530	20120525
25181	`opt64`	`gcc -m64 -O2 -fomit-frame-pointer`	20120530	20120525
25183	`opt64`	`gcc -march=nocona -O3 -fomit-frame-pointer`	20120530	20120525
25210	`asm64`	`gcc -m64 -march=corei7-avx -Os -fomit-frame-pointer`	20120530	20120525
25214	`opt64`	`gcc -m64 -march=nocona -O3 -fomit-frame-pointer`	20120530	20120525
25222	`opt64`	`gcc -m64 -O -fomit-frame-pointer`	20120530	20120525
25309	`opt64`	`gcc -funroll-loops -m64 -O3 -fomit-frame-pointer`	20120530	20120525
25319	`opt64`	`gcc -m64 -march=corei7 -O3 -fomit-frame-pointer`	20120530	20120525
25322	`opt64`	`gcc -m64 -march=nocona -O -fomit-frame-pointer`	20120530	20120525
25356	`opt64`	`gcc -fno-schedule-insns -O -fomit-frame-pointer`	20120530	20120525
25357	`asm64`	`gcc -march=barcelona -Os -fomit-frame-pointer`	20120530	20120525
25379	`asm64`	`gcc -m64 -march=core2 -msse4.1 -Os -fomit-frame-pointer`	20120530	20120525
25418	`opt64`	`gcc -funroll-loops -O2 -fomit-frame-pointer`	20120530	20120525
25455	`opt64`	`gcc -m64 -march=core2 -O3 -fomit-frame-pointer`	20120530	20120525
25478	`opt64`	`gcc -m64 -march=core2 -msse4 -O -fomit-frame-pointer`	20120530	20120525
25483	`opt64`	`gcc -funroll-loops -m64 -march=nocona -O -fomit-frame-pointer`	20120530	20120525
25575	`opt64`	`gcc -funroll-loops -O3 -fomit-frame-pointer`	20120530	20120525
25585	`opt64`	`gcc -m64 -march=core2 -msse4.1 -O2 -fomit-frame-pointer`	20120530	20120525
25618	`asm64`	`gcc -m64 -march=core2 -Os -fomit-frame-pointer`	20120530	20120525
25638	`opt64`	`gcc -O3 -fomit-frame-pointer`	20120530	20120525
25643	`opt64`	`gcc -fno-schedule-insns -O3 -fomit-frame-pointer`	20120530	20120525
25652	`asm64`	`gcc -funroll-loops`	20120530	20120525
25680	`opt64`	`gcc -m64 -march=core2 -O2 -fomit-frame-pointer`	20120530	20120525
25692	`asm64`	`gcc`	20120530	20120525
25692	`opt64`	`gcc -funroll-loops -march=nocona -O2 -fomit-frame-pointer`	20120530	20120525
25714	`opt64`	`gcc -funroll-loops -m64 -march=k8 -O -fomit-frame-pointer`	20120530	20120525
25823	`opt64`	`gcc -fno-schedule-insns -O2 -fomit-frame-pointer`	20120530	20120525
25824	`opt64`	`gcc -O2 -fomit-frame-pointer`	20120530	20120525
25831	`opt64`	`gcc -funroll-loops -march=barcelona -O -fomit-frame-pointer`	20120530	20120525
25863	`opt64`	`gcc -funroll-loops -m64 -O -fomit-frame-pointer`	20120530	20120525
25867	`opt64`	`gcc -funroll-loops -march=k8 -O -fomit-frame-pointer`	20120530	20120525
25867	`opt64`	`gcc -m64 -march=native -mtune=native -O3 -fomit-frame-pointer`	20120530	20120525
25926	`opt64`	`gcc -march=k8 -O3 -fomit-frame-pointer`	20120530	20120525
25952	`opt64`	`gcc -O -fomit-frame-pointer`	20120530	20120525
25959	`opt64`	`gcc -m64 -march=native -mtune=native -O2 -fomit-frame-pointer`	20120530	20120525
25982	`asm64`	`gcc -m64 -march=nocona -Os -fomit-frame-pointer`	20120530	20120525
25984	`opt64`	`gcc -m64 -march=core2 -msse4.1 -O3 -fomit-frame-pointer`	20120530	20120525
26020	`opt64`	`gcc -m64 -march=corei7-avx -O2 -fomit-frame-pointer`	20120530	20120525
26034	`opt64`	`gcc -m64 -march=k8 -O3 -fomit-frame-pointer`	20120530	20120525
26083	`opt64`	`gcc -funroll-loops -fno-schedule-insns -O2 -fomit-frame-pointer`	20120530	20120525
26086	`opt64`	`gcc -funroll-loops -m64 -march=barcelona -O3 -fomit-frame-pointer`	20120530	20120525
26110	`opt64`	`gcc -march=nocona -O2 -fomit-frame-pointer`	20120530	20120525
26184	`opt64`	`gcc -m64 -march=native -mtune=native -O -fomit-frame-pointer`	20120530	20120525
26320	`opt64`	`gcc -m64 -march=corei7-avx -O3 -fomit-frame-pointer`	20120530	20120525
26362	`opt64`	`gcc -funroll-loops -march=k8 -O2 -fomit-frame-pointer`	20120530	20120525
26428	`opt64`	`gcc -funroll-loops -m64 -march=k8 -O2 -fomit-frame-pointer`	20120530	20120525
26487	`opt64`	`gcc -march=k8 -O -fomit-frame-pointer`	20120530	20120525
26515	`opt64`	`gcc -m64 -march=corei7 -O -fomit-frame-pointer`	20120530	20120525
26553	`opt64`	`gcc -march=barcelona -O -fomit-frame-pointer`	20120530	20120525
26603	`opt64`	`gcc -funroll-loops -march=k8 -O3 -fomit-frame-pointer`	20120530	20120525
26749	`opt64`	`gcc -m64 -march=k8 -O -fomit-frame-pointer`	20120530	20120525
26820	`opt64`	`gcc -m64 -march=corei7-avx -O -fomit-frame-pointer`	20120530	20120525
26822	`opt64`	`gcc -m64 -march=barcelona -O -fomit-frame-pointer`	20120530	20120525
26959	`opt64`	`gcc -funroll-loops -m64 -march=barcelona -O2 -fomit-frame-pointer`	20120530	20120525
26972	`opt64`	`gcc -funroll-loops -m64 -march=barcelona -O -fomit-frame-pointer`	20120530	20120525
26979	`opt64`	`gcc -m64 -march=corei7 -O2 -fomit-frame-pointer`	20120530	20120525
27103	`opt64`	`gcc -funroll-loops -march=barcelona -O2 -fomit-frame-pointer`	20120530	20120525
27103	`opt64`	`gcc -march=k8 -O2 -fomit-frame-pointer`	20120530	20120525
27179	`opt64`	`gcc -funroll-loops -m64 -march=k8 -O3 -fomit-frame-pointer`	20120530	20120525
27276	`opt64`	`gcc -march=barcelona -O3 -fomit-frame-pointer`	20120530	20120525
27309	`opt64`	`gcc -funroll-loops -march=nocona -O3 -fomit-frame-pointer`	20120530	20120525
27312	`opt64`	`gcc -m64 -march=corei7-avx -Os -fomit-frame-pointer`	20120530	20120525
27318	`opt64`	`gcc -funroll-loops -m64 -march=barcelona -Os -fomit-frame-pointer`	20120530	20120525
27352	`opt64`	`gcc -m64 -march=core2 -msse4.1 -O -fomit-frame-pointer`	20120530	20120525
27422	`opt64`	`gcc -funroll-loops -m64 -Os -fomit-frame-pointer`	20120530	20120525
27466	`opt64`	`gcc -funroll-loops -Os -fomit-frame-pointer`	20120530	20120525
27467	`opt64`	`gcc -m64 -Os -fomit-frame-pointer`	20120530	20120525
27524	`opt64`	`gcc -funroll-loops -fno-schedule-insns -Os -fomit-frame-pointer`	20120530	20120525
27590	`opt64`	`gcc -m64 -march=core-avx-i -Os -fomit-frame-pointer`	20120530	20120525
27602	`opt64`	`gcc -march=barcelona -O2 -fomit-frame-pointer`	20120530	20120525
27658	`opt64`	`gcc -funroll-loops -march=nocona -Os -fomit-frame-pointer`	20120530	20120525
27729	`opt64`	`gcc -march=barcelona -Os -fomit-frame-pointer`	20120530	20120525
27741	`opt64`	`gcc -Os -fomit-frame-pointer`	20120530	20120525
27765	`opt64`	`gcc -m64 -march=barcelona -Os -fomit-frame-pointer`	20120530	20120525
27775	`opt64`	`gcc -fno-schedule-insns -Os -fomit-frame-pointer`	20120530	20120525
27797	`opt64`	`gcc -funroll-loops -m64 -march=k8 -Os -fomit-frame-pointer`	20120530	20120525
27826	`opt64`	`gcc -m64 -march=core2 -Os -fomit-frame-pointer`	20120530	20120525
27861	`opt64`	`gcc -m64 -march=nocona -Os -fomit-frame-pointer`	20120530	20120525
27991	`opt64`	`gcc -m64 -march=core2 -msse4.1 -Os -fomit-frame-pointer`	20120530	20120525
28180	`opt64`	`gcc -m64 -march=native -mtune=native -Os -fomit-frame-pointer`	20120530	20120525
28256	`opt64`	`gcc -m64 -march=k8 -Os -fomit-frame-pointer`	20120530	20120525
28283	`opt64`	`gcc -funroll-loops -march=barcelona -Os -fomit-frame-pointer`	20120530	20120525
28489	`opt64`	`gcc -funroll-loops -m64 -march=nocona -Os -fomit-frame-pointer`	20120530	20120525
28494	`opt64`	`gcc -march=k8 -Os -fomit-frame-pointer`	20120530	20120525
28639	`opt64`	`gcc -m64 -march=core2 -msse4 -Os -fomit-frame-pointer`	20120530	20120525
28851	`opt64`	`gcc -march=nocona -Os -fomit-frame-pointer`	20120530	20120525
28925	`opt64`	`gcc -m64 -march=barcelona -O3 -fomit-frame-pointer`	20120530	20120525
29604	`opt64`	`gcc -m64 -march=corei7 -Os -fomit-frame-pointer`	20120530	20120525
30817	`opt64`	`gcc -funroll-loops -march=k8 -Os -fomit-frame-pointer`	20120530	20120525
48437	`opt64`	`cc`	20120530	20120525
49470	`opt64`	`gcc -funroll-loops`	20120530	20120525
49832	`opt64`	`gcc`	20120530	20120525

Checksum failure

Implementation: crypto_hash/sarmal512/asm64
Compiler: gcc -m64 -march=k8 -O2 -fomit-frame-pointer

e2cf2e8e623fb36d83bfe3608cfeb6450ad15b9d9c6c1e4bea2695ad7f5282a009fb4aeac03cf5ad582722ed2d653432de9380c3156eab6bbcfbc3787ceebd17

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/sarmal512/asm64
Compiler: gcc -O2 -fomit-frame-pointer

error 111

Number of similar (compiler,implementation) pairs: 32, 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=core-avx-i -O2 -fomit-frame-pointer	asm64
gcc -m64 -march=core-avx-i -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 -msse4.1 -O3 -fomit-frame-pointer	asm64
gcc -m64 -march=core2 -msse4 -O2 -fomit-frame-pointer	asm64
gcc -m64 -march=core2 -msse4 -O3 -fomit-frame-pointer	asm64
gcc -m64 -march=corei7-avx -O2 -fomit-frame-pointer	asm64
gcc -m64 -march=corei7-avx -O3 -fomit-frame-pointer	asm64
gcc -m64 -march=corei7 -O2 -fomit-frame-pointer	asm64
gcc -m64 -march=corei7 -O3 -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/sarmal512/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/sarmal512/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