Implementation notes: amd64, slide, crypto_hash/keccakc768

Computer: slide
Architecture: amd64
CPU ID: GenuineIntel-00040651-bfebfbff
SUPERCOP version: 20160806
Operation: crypto_hash
Primitive: keccakc768

Time	Implementation	Compiler	Benchmark date	SUPERCOP version
18308	`inplace`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
18328	`inplace`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
18548	`inplace`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
18692	`simple`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
18740	`opt64u6`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
18800	`simple`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
19012	`simple`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
19048	`opt64u6`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
19076	`simple`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
19308	`opt64u6`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
19324	`opt64u6`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
19540	`inplace`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
20404	`opt64lcu24shld`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
20464	`opt64lcu24`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
20844	`x86_64_shld`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
20900	`x86_64_shld`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
20988	`opt64lcu6`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
21052	`opt64lcu24`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
21064	`opt64lcu24`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
21112	`x86_64_shld`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
21272	`opt64lcu24`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
21300	`opt64lcu6`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
21332	`opt64lcu6`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
21368	`opt64lcu6`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
21628	`opt64lcu24shld`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
21632	`opt64lcu24shld`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
21884	`x86_64_shld`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
22296	`x86_64_asm`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
22300	`x86_64_asm`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
22336	`x86_64_asm`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
22508	`x86_64_asm`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
22804	`opt64lcu24shld`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
26820	`sseu2`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
26896	`sseu2`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
26992	`sseu2`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
27220	`sseu2`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
39596	`mmxu1`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
42612	`mmxu1`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
42648	`mmxu1`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
43868	`mmxu1`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
44396	`opt32bi-rvku2`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
44432	`opt32bi-rvku2`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
44516	`simple32bi`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
44668	`simple32bi`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
44812	`simple32bi`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
44976	`opt32bi-rvku2`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
45340	`simple32bi`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
45416	`opt32bi-rvku2`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
49288	`opt32biT-s2lcu4`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
49488	`inplace32bi`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
49800	`compact`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
50156	`opt32bi-s2lcu4`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
50796	`inplace32bi`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
51516	`inplace32bi`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
51544	`opt32biT-s2lcu4`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
51728	`opt32bi-s2lcu4`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
52300	`opt32biT-s2lcu4`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
52936	`opt32bi-s2lcu4`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
52980	`inplace32bi`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
53004	`opt32bi-s2lcu4`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
53060	`opt32biT-s2lcu4`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
95636	`compact`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
100784	`compact`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
107344	`compact`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
198536	`compact8`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
204276	`compact8`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
213944	`compact8`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
292732	`compact8`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806

Compiler output

Implementation: crypto_hash/keccakc768/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