Implementation notes: amd64, slide, crypto_hash/keccakc1024

Computer: slide
Architecture: amd64
CPU ID: GenuineIntel-00040651-bfebfbff
SUPERCOP version: 20160806
Operation: crypto_hash
Primitive: keccakc1024

Time	Implementation	Compiler	Benchmark date	SUPERCOP version
26548	`inplace`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
26704	`inplace`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
26916	`inplace`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
27212	`simple`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
27368	`simple`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
27636	`simple`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
27668	`opt64u6`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
27688	`opt64u6`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
27704	`simple`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
27840	`opt64u6`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
28000	`opt64u6`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
28704	`inplace`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
29612	`opt64lcu24`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
29668	`opt64lcu24shld`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
30184	`opt64lcu6`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
30324	`x86_64_shld`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
30424	`x86_64_shld`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
30460	`x86_64_shld`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
30468	`opt64lcu24`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
30476	`x86_64_shld`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
30508	`opt64lcu24`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
30616	`opt64lcu24`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
31176	`opt64lcu6`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
31216	`opt64lcu6`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
31320	`opt64lcu6`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
31752	`opt64lcu24shld`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
32148	`opt64lcu24shld`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
32156	`opt64lcu24shld`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
32396	`x86_64_asm`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
32452	`x86_64_asm`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
32524	`x86_64_asm`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
32552	`x86_64_asm`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
35408	`sphlib`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
37332	`sphlib-small`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
37448	`sphlib-small`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
37504	`sphlib-small`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
38720	`sphlib-small`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
39032	`sphlib`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
39040	`sphlib`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
39296	`sseu2`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
39604	`sphlib`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
39912	`sseu2`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
40084	`sseu2`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
40500	`sseu2`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
58604	`mmxu1`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
61788	`opt32bi-rvku2`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
62228	`mmxu1`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
62244	`mmxu1`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
62676	`opt32bi-rvku2`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
63024	`simple32bi`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
63256	`simple32bi`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
63312	`simple32bi`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
63588	`opt32bi-rvku2`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
63876	`opt32bi-rvku2`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
63960	`simple32bi`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
64248	`mmxu1`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
70060	`inplace32bi`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
70996	`opt32bi-s2lcu4`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
71808	`opt32biT-s2lcu4`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
72276	`inplace32bi`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
73120	`opt32bi-s2lcu4`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
73320	`inplace32bi`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
73440	`opt32biT-s2lcu4`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
73832	`opt32bi-s2lcu4`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
74128	`opt32biT-s2lcu4`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
74920	`opt32biT-s2lcu4`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
75020	`opt32bi-s2lcu4`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
75496	`inplace32bi`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
81872	`compact`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
135356	`compact`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
147104	`compact`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
158036	`compact`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
294736	`compact8`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
298888	`compact8`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
313032	`compact8`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
429760	`compact8`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806

Compiler output

Implementation: crypto_hash/keccakc1024/xopu24
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv

KeccakF-1600-opt64.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/4.9/include/x86intrin.h:52:0,
KeccakF-1600-opt64.c: from KeccakF-1600-opt64.c:74:
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c: In function 'KeccakPermutationOnWords':
KeccakF-1600-opt64.c: /usr/lib/gcc/x86_64-linux-gnu/4.9/include/xopintrin.h:266:1: error: inlining failed in call to always_inline '_mm_roti_epi64': target specific option mismatch
KeccakF-1600-opt64.c: _mm_roti_epi64(__m128i __A, const int __B)
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c:90:33: error: called from here
KeccakF-1600-opt64.c: #define XOR128(a, b) _mm_xor_si128(a, b)
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-xop.macros:40:11: note: in expansion of macro 'XOR128'
KeccakF-1600-opt64.c: Dei = XOR128(Cae, ROL6464same(Cio, 1)); \
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-xop.macros:53:5: note: in expansion of macro 'computeD'
KeccakF-1600-opt64.c: computeD \
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-unrolling.macros:17:5: note: in expansion of macro 'thetaRhoPiChiIotaPrepareTheta'
KeccakF-1600-opt64.c: thetaRhoPiChiIotaPrepareTheta( 0, A, E) \
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c:185:5: note: in expansion of macro 'rounds'
KeccakF-1600-opt64.c: rounds
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/4.9/include/x86intrin.h:52:0,
KeccakF-1600-opt64.c: from KeccakF-1600-opt64.c:74:
KeccakF-1600-opt64.c: /usr/lib/gcc/x86_64-linux-gnu/4.9/include/xopintrin.h:266:1: error: inlining failed in call to always_inline '_mm_roti_epi64': target specific option mismatch
KeccakF-1600-opt64.c: _mm_roti_epi64(__m128i __A, const int __B)
KeccakF-1600-opt64.c: ...

Number of similar (compiler,implementation) pairs: 4, namely:

Compiler	Implementations
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv	xopu24
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv	xopu24
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv	xopu24
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv	xopu24