Implementation notes: armeabi, cubox, crypto_hash/keccak

Computer: cubox
Architecture: armeabi
CPU ID: unknown CPU ID
SUPERCOP version: 20161026
Operation: crypto_hash
Primitive: keccak

Time	Implementation	Compiler	Benchmark date	SUPERCOP version
100411	`simple32bi`	`gcc -funroll-loops -mcpu=marvell-pj4 -Os`	20161209	20161026
100734	`inplace32bi-armv7a`	`gcc -funroll-loops -mcpu=marvell-pj4 -O3`	20161209	20161026
100804	`inplace32bi-armv7a`	`gcc -mcpu=marvell-pj4 -O2`	20161209	20161026
100952	`inplace32bi-armv7a`	`gcc -funroll-loops -mcpu=marvell-pj4 -Os`	20161209	20161026
100952	`inplace32bi-armv7a`	`gcc -mcpu=marvell-pj4 -Os`	20161209	20161026
100960	`inplace32bi-armv7a`	`gcc -mcpu=marvell-pj4 -O3`	20161209	20161026
100962	`inplace32bi-armv7a`	`gcc -funroll-loops -mcpu=marvell-pj4 -O2`	20161209	20161026
106512	`opt32bi-rvku2`	`gcc -funroll-loops -mcpu=marvell-pj4 -Os`	20161209	20161026
111446	`simple32bi`	`gcc -mcpu=marvell-pj4 -Os`	20161209	20161026
111484	`inplace32bi`	`gcc -funroll-loops -mcpu=marvell-pj4 -Os`	20161209	20161026
119556	`opt32bi-rvku2`	`gcc -mcpu=marvell-pj4 -Os`	20161209	20161026
123400	`inplace32bi`	`gcc -mcpu=marvell-pj4 -Os`	20161209	20161026
126208	`opt32bi-s2lcu4`	`gcc -funroll-loops -mcpu=marvell-pj4 -Os`	20161209	20161026
131554	`arm11`	`gcc -funroll-loops -mcpu=marvell-pj4 -O2`	20161209	20161026
131633	`arm11`	`gcc -funroll-loops -mcpu=marvell-pj4 -O3`	20161209	20161026
131919	`arm11`	`gcc -mcpu=marvell-pj4 -O3`	20161209	20161026
132088	`arm11`	`gcc -mcpu=marvell-pj4 -O2`	20161209	20161026
132716	`arm11`	`gcc -mcpu=marvell-pj4 -Os`	20161209	20161026
132768	`arm11`	`gcc -funroll-loops -mcpu=marvell-pj4 -Os`	20161209	20161026
136966	`opt32biT-s2lcu4`	`gcc -funroll-loops -mcpu=marvell-pj4 -Os`	20161209	20161026
149468	`opt32bi-s2lcu4`	`gcc -mcpu=marvell-pj4 -Os`	20161209	20161026
159690	`opt32biT-s2lcu4`	`gcc -mcpu=marvell-pj4 -Os`	20161209	20161026
166829	`opt32bi-rvku2`	`gcc -funroll-loops -mcpu=marvell-pj4 -O2`	20161209	20161026
168977	`opt32bi-s2lcu4`	`gcc -funroll-loops -mcpu=marvell-pj4 -O3`	20161209	20161026
172470	`opt32bi-s2lcu4`	`gcc -funroll-loops -mcpu=marvell-pj4 -O2`	20161209	20161026
175925	`simple32bi`	`gcc -funroll-loops -mcpu=marvell-pj4 -O3`	20161209	20161026
177568	`simple32bi`	`gcc -funroll-loops -mcpu=marvell-pj4 -O2`	20161209	20161026
177795	`opt32biT-s2lcu4`	`gcc -funroll-loops -mcpu=marvell-pj4 -O3`	20161209	20161026
181984	`opt32bi-s2lcu4`	`gcc -mcpu=marvell-pj4 -O3`	20161209	20161026
182932	`opt32bi-rvku2`	`gcc -funroll-loops -mcpu=marvell-pj4 -O3`	20161209	20161026
184404	`opt32bi-rvku2`	`gcc -mcpu=marvell-pj4 -O2`	20161209	20161026
186237	`inplace32bi`	`gcc -funroll-loops -mcpu=marvell-pj4 -O2`	20161209	20161026
187807	`inplace32bi`	`gcc -funroll-loops -mcpu=marvell-pj4 -O3`	20161209	20161026
188651	`opt32bi-s2lcu4`	`gcc -mcpu=marvell-pj4 -O2`	20161209	20161026
188821	`opt32biT-s2lcu4`	`gcc -funroll-loops -mcpu=marvell-pj4 -O2`	20161209	20161026
190158	`simple32bi`	`gcc -mcpu=marvell-pj4 -O2`	20161209	20161026
190284	`simple32bi`	`gcc -mcpu=marvell-pj4 -O3`	20161209	20161026
191222	`opt32biT-s2lcu4`	`gcc -mcpu=marvell-pj4 -O3`	20161209	20161026
194418	`inplace32bi`	`gcc -mcpu=marvell-pj4 -O2`	20161209	20161026
198638	`inplace32bi`	`gcc -mcpu=marvell-pj4 -O3`	20161209	20161026
198788	`opt32bi-rvku2`	`gcc -mcpu=marvell-pj4 -O3`	20161209	20161026
206834	`opt32biT-s2lcu4`	`gcc -mcpu=marvell-pj4 -O2`	20161209	20161026
230282	`opt64lcu6`	`gcc -funroll-loops -mcpu=marvell-pj4 -Os`	20161209	20161026
232276	`simple`	`gcc -funroll-loops -mcpu=marvell-pj4 -O3`	20161209	20161026
232707	`inplace`	`gcc -funroll-loops -mcpu=marvell-pj4 -O3`	20161209	20161026
237228	`opt64u6`	`gcc -funroll-loops -mcpu=marvell-pj4 -Os`	20161209	20161026
238818	`simple`	`gcc -funroll-loops -mcpu=marvell-pj4 -O2`	20161209	20161026
243097	`inplace`	`gcc -funroll-loops -mcpu=marvell-pj4 -O2`	20161209	20161026
252108	`simple`	`gcc -funroll-loops -mcpu=marvell-pj4 -Os`	20161209	20161026
252328	`simple`	`gcc -mcpu=marvell-pj4 -Os`	20161209	20161026
255845	`opt64lcu6`	`gcc -funroll-loops -mcpu=marvell-pj4 -O3`	20161209	20161026
257691	`opt64u6`	`gcc -funroll-loops -mcpu=marvell-pj4 -O3`	20161209	20161026
258117	`simple`	`gcc -mcpu=marvell-pj4 -O3`	20161209	20161026
259628	`opt64lcu6`	`gcc -funroll-loops -mcpu=marvell-pj4 -O2`	20161209	20161026
260287	`opt64lcu6`	`gcc -mcpu=marvell-pj4 -Os`	20161209	20161026
260752	`simple`	`gcc -mcpu=marvell-pj4 -O2`	20161209	20161026
261331	`inplace`	`gcc -mcpu=marvell-pj4 -O3`	20161209	20161026
262944	`opt64u6`	`gcc -mcpu=marvell-pj4 -Os`	20161209	20161026
266208	`inplace`	`gcc -funroll-loops -mcpu=marvell-pj4 -Os`	20161209	20161026
269344	`opt64u6`	`gcc -funroll-loops -mcpu=marvell-pj4 -O2`	20161209	20161026
273278	`inplace`	`gcc -mcpu=marvell-pj4 -Os`	20161209	20161026
279418	`inplace`	`gcc -mcpu=marvell-pj4 -O2`	20161209	20161026
280912	`opt64lcu24`	`gcc -funroll-loops -mcpu=marvell-pj4 -Os`	20161209	20161026
287013	`opt64lcu6`	`gcc -mcpu=marvell-pj4 -O3`	20161209	20161026
287878	`opt64lcu24`	`gcc -mcpu=marvell-pj4 -Os`	20161209	20161026
299151	`opt64u6`	`gcc -mcpu=marvell-pj4 -O3`	20161209	20161026
302295	`compact`	`gcc -funroll-loops -mcpu=marvell-pj4 -O2`	20161209	20161026
303650	`opt64lcu6`	`gcc -mcpu=marvell-pj4 -O2`	20161209	20161026
308852	`compact`	`gcc -mcpu=marvell-pj4 -O3`	20161209	20161026
309126	`opt64u6`	`gcc -mcpu=marvell-pj4 -O2`	20161209	20161026
328642	`compact8`	`gcc -funroll-loops -mcpu=marvell-pj4 -O3`	20161209	20161026
331656	`compact8`	`gcc -mcpu=marvell-pj4 -O3`	20161209	20161026
379440	`compact`	`gcc -funroll-loops -mcpu=marvell-pj4 -O3`	20161209	20161026
478000	`compact`	`gcc -mcpu=marvell-pj4 -O2`	20161209	20161026
480504	`opt64lcu24`	`gcc -funroll-loops -mcpu=marvell-pj4 -O2`	20161209	20161026
481534	`opt64lcu24`	`gcc -funroll-loops -mcpu=marvell-pj4 -O3`	20161209	20161026
499997	`opt64lcu24`	`gcc -mcpu=marvell-pj4 -O2`	20161209	20161026
500952	`opt64lcu24`	`gcc -mcpu=marvell-pj4 -O3`	20161209	20161026
584118	`compact8`	`gcc -funroll-loops -mcpu=marvell-pj4 -O2`	20161209	20161026
614977	`compact8`	`gcc -mcpu=marvell-pj4 -O2`	20161209	20161026
643645	`compact`	`gcc -funroll-loops -mcpu=marvell-pj4 -Os`	20161209	20161026
692056	`compact`	`gcc -mcpu=marvell-pj4 -Os`	20161209	20161026
916619	`compact8`	`gcc -mcpu=marvell-pj4 -Os`	20161209	20161026
931155	`compact8`	`gcc -funroll-loops -mcpu=marvell-pj4 -Os`	20161209	20161026

Compiler output

Implementation: crypto_hash/keccak/inplace-armv7a-neon
Compiler: gcc -funroll-loops -mcpu=marvell-pj4 -O2

keccak.s: keccak.s: Assembler messages:
keccak.s: keccak.s:48: Error: selected FPU does not support instruction -- `vmov.i64 q6,#0'
keccak.s: keccak.s:50: Error: selected FPU does not support instruction -- `vmov.i64 q7,#0'
keccak.s: keccak.s:51: Error: selected FPU does not support instruction -- `vmov.i64 q8,#0'
keccak.s: keccak.s:53: Error: selected FPU does not support instruction -- `vmov.i64 q9,#0'
keccak.s: keccak.s:54: Error: selected FPU does not support instruction -- `vmov.i64 q10,#0'
keccak.s: keccak.s:56: Error: selected FPU does not support instruction -- `vmov.i64 q11,#0'
keccak.s: keccak.s:57: Error: selected FPU does not support instruction -- `vmov.i64 q12,#0'
keccak.s: keccak.s:59: Error: selected FPU does not support instruction -- `vmov.i64 q13,#0'
keccak.s: keccak.s:60: Error: selected FPU does not support instruction -- `vmov.i64 q14,#0'
keccak.s: ...
keccak.s: keccak.s:347: Error: selected processor does not support `veor.64 d19,d0' in ARM mode
keccak.s: keccak.s:351: Error: selected processor does not support `veor.64 d24,d0' in ARM mode
keccak.s: keccak.s:355: Error: selected processor does not support `veor.64 d29,d0' in ARM mode
keccak.s: keccak.s:359: Error: selected processor does not support `veor.64 d10,d0' in ARM mode
keccak.s: keccak.s:363: Error: selected FPU does not support instruction -- `vmov.i8 d3,#0xFF'
keccak.s: keccak.s:364: Error: selected processor does not support `vshl.u64 d3,d3,#63' in ARM mode
keccak.s: keccak.s:365: Error: selected processor does not support `veor.64 d10,d10,d3' in ARM mode
keccak.s: keccak.s:369: Error: selected processor does not support `veor.64 q4,q5' in ARM mode
keccak.s: keccak.s:370: Error: selected processor does not support `veor.64 d5,d8,d9' in ARM mode
keccak.s: keccak.s:371: Error: selected processor does not support `veor.64 d5,d5,d7' in ARM mode

Number of similar (compiler,implementation) pairs: 6, namely:

Compiler	Implementations
gcc -funroll-loops -mcpu=marvell-pj4 -O2	inplace-armv7a-neon
gcc -funroll-loops -mcpu=marvell-pj4 -O3	inplace-armv7a-neon
gcc -funroll-loops -mcpu=marvell-pj4 -Os	inplace-armv7a-neon
gcc -mcpu=marvell-pj4 -O2	inplace-armv7a-neon
gcc -mcpu=marvell-pj4 -O3	inplace-armv7a-neon
gcc -mcpu=marvell-pj4 -Os	inplace-armv7a-neon

Compiler output

Implementation: crypto_hash/keccak/inplace32bi-armv6m
Compiler: gcc -funroll-loops -mcpu=marvell-pj4 -O2

try.c: /usr/bin/ld: error: crypto_hash_keccak.a(Keccak-inplace32BI-armgcc-ARMv6M.o): Conflicting architecture profiles M/A
try.c: /usr/bin/ld: failed to merge target specific data of file crypto_hash_keccak.a(Keccak-inplace32BI-armgcc-ARMv6M.o)
try.c: collect2: error: ld returned 1 exit status

Number of similar (compiler,implementation) pairs: 6, namely:

Compiler	Implementations
gcc -funroll-loops -mcpu=marvell-pj4 -O2	inplace32bi-armv6m
gcc -funroll-loops -mcpu=marvell-pj4 -O3	inplace32bi-armv6m
gcc -funroll-loops -mcpu=marvell-pj4 -Os	inplace32bi-armv6m
gcc -mcpu=marvell-pj4 -O2	inplace32bi-armv6m
gcc -mcpu=marvell-pj4 -O3	inplace32bi-armv6m
gcc -mcpu=marvell-pj4 -Os	inplace32bi-armv6m