Implementation notes: amd64, hydra7, crypto_hash/keccakc448

Computer: hydra7
Microarchitecture: amd64; Sandy Bridge+AES (206a7)
Architecture: amd64
CPU ID: GenuineIntel-000206a7-bfebfbff
SUPERCOP version: 20240107
Operation: crypto_hash
Primitive: keccakc448

Time	Object size	Test size	Implementation	Compiler	Benchmark date	SUPERCOP version
16502	166588 0 0	179037 804 960	`T:x86_64_shld`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
16515	166580 0 0	177781 804 960	`T:x86_64_shld`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
16535	166322 0 0	177197 804 960	`T:x86_64_shld`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
17049	166114 0 0	175896 780 928	`T:x86_64_shld`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
17686	60743 0 0	71032 780 928	`T:opt64lcu24shld`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
17849	63341 0 0	75101 804 960	`T:opt64lcu24shld`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
18135	63272 0 0	74709 804 960	`T:opt64lcu24shld`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
18484	65661 0 0	78661 804 960	`T:opt64lcu24shld`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
19419	159868 0 0	172317 804 960	`T:x86_64_asm`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
19514	159602 0 0	170477 804 960	`T:x86_64_asm`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
19926	159394 0 0	169176 780 928	`T:x86_64_asm`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
20016	159860 0 0	171061 804 960	`T:x86_64_asm`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
20072	11372 0 0	24373 804 960	`T:sseu2`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
20285	11308 0 0	23077 804 960	`T:sseu2`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
20386	58347 0 0	68648 780 928	`T:opt64lcu24`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
20681	19197 0 0	30957 804 960	`T:opt64lcu6`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
20749	18285 0 0	28584 780 928	`T:opt64lcu6`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
20838	61565 0 0	73325 804 960	`T:opt64lcu24`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
21014	10964 0 0	22405 804 960	`T:sseu2`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
21147	61101 0 0	72533 804 960	`T:opt64lcu24`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
21241	10490 0 0	20792 780 928	`T:sseu2`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
21316	18948 0 0	30389 804 960	`T:opt64lcu6`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
21435	63485 0 0	76485 804 960	`T:opt64lcu24`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
21721	2763 0 0	13143 788 928	`T:simple`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
21770	4358 0 0	14743 788 928	`T:inplace`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
22426	19401 0 0	29688 780 928	`T:opt64u6`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
22484	20148 0 0	31909 804 960	`T:opt64u6`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
22528	20557 0 0	33557 804 960	`T:opt64lcu6`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
22733	4675 0 0	16492 812 960	`T:inplace`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
22775	9332 0 0	22404 812 960	`T:inplace`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
22822	5924 0 0	18996 812 960	`T:simple`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
23170	3043 0 0	14860 812 960	`T:simple`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
23469	4607 0 0	16140 812 960	`T:inplace`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
23612	20303 0 0	31749 804 960	`T:opt64u6`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
23959	8562 0 0	19997 804 960	`T:mmxu1`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
24017	21268 0 0	34269 804 960	`T:opt64u6`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
24049	9140 0 0	20901 804 960	`T:mmxu1`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
24444	9412 0 0	22413 804 960	`T:mmxu1`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
24510	2963 0 0	14492 812 960	`T:simple`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
26444	8629 0 0	18928 780 928	`T:mmxu1`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
40782	6694 192 0	19717 1004 960	`T:compact8`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
47567	32264 0 0	45245 804 960	`T:opt32bi-s2lcu4`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
48557	26541 0 262148	36848 780 263104	`T:opt32biT-s2lcu4`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
49669	29824 0 262148	42837 804 263136	`T:opt32biT-s2lcu4`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
49761	28111 0 262148	39557 804 263136	`T:opt32biT-s2lcu4`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
50372	25356 0 0	35648 780 928	`T:opt32bi-s2lcu4`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
50952	29208 0 262148	40981 804 263136	`T:opt32biT-s2lcu4`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
51014	19912 0 0	32901 804 960	`T:opt32bi-rvku2`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
51994	7329 0 0	20436 812 960	`T:compact`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
52579	15830 0 0	26128 780 928	`T:opt32bi-rvku2`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
52778	26259 0 0	37693 804 960	`T:opt32bi-s2lcu4`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
52965	27587 0 0	39349 804 960	`T:opt32bi-s2lcu4`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
53177	16017 0 0	27781 804 960	`T:opt32bi-rvku2`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
56298	16180 0 0	27613 804 960	`T:opt32bi-rvku2`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
73131	2001 0 0	12399 788 928	`T:compact`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
79625	2541 0 0	14388 812 960	`T:compact`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
112232	2250 0 0	13804 812 960	`T:compact`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
174070	1792 192 0	13180 996 960	`T:compact8`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
176022	1974 192 0	13741 1004 960	`T:compact8`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
207433	1579 192 0	11912 980 928	`T:compact8`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212

Compiler output

Implementation: T:inplace32bi
Security model: timingleaks
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE

Keccak-inplace32BI.c: Keccak-inplace32BI.c: In function 'crypto_hash_keccakc448_inplace32bi_timingleaks':
Keccak-inplace32BI.c: Keccak-inplace32BI.c:73:6: error: #error "The output size must be a multiple of the lane size in this simple implementation."
Keccak-inplace32BI.c: 73 | #error "The output size must be a multiple of the lane size in this simple implementation."
Keccak-inplace32BI.c: | ^~~~~

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

Compiler	Implementations
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:inplace32bi
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:inplace32bi
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:inplace32bi
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:inplace32bi

Compiler output

Implementation: T:simple32bi
Security model: timingleaks
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE

Keccak-simple32BI.c: Keccak-simple32BI.c: In function 'crypto_hash_keccakc448_simple32bi_timingleaks':
Keccak-simple32BI.c: Keccak-simple32BI.c:73:6: error: #error "The output size must be a multiple of the lane size in this simple implementation."
Keccak-simple32BI.c: 73 | #error "The output size must be a multiple of the lane size in this simple implementation."
Keccak-simple32BI.c: | ^~~~~

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

Compiler	Implementations
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:simple32bi
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:simple32bi
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:simple32bi
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:simple32bi

Compiler output

Implementation: T:xopu24
Security model: timingleaks
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE

KeccakF-1600-opt64.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/11/include/x86intrin.h:38,
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/11/include/xopintrin.h:272:1: error: inlining failed in call to 'always_inline' '_mm_roti_epi64': target specific option mismatch
KeccakF-1600-opt64.c: 272 | _mm_roti_epi64(__m128i __A, const int __B)
KeccakF-1600-opt64.c: | ^~~~~~~~~~~~~~
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c:102:33: note: called from here
KeccakF-1600-opt64.c: 102 | #define ROL6464same(a, o) _mm_roti_epi64(a, o)
KeccakF-1600-opt64.c: | ^~~~~~~~~~~~~~~~~~~~
KeccakF-1600-opt64.c: KeccakF-1600-xop.macros:103:13: note: in expansion of macro 'ROL6464same'
KeccakF-1600-opt64.c: 103 | Bsusa = ROL6464same(Bsusa, 2); \
KeccakF-1600-opt64.c: | ^~~~~~~~~~~
KeccakF-1600-opt64.c: KeccakF-1600-xop.macros:123:36: note: in expansion of macro 'thetaRhoPiChiIotaPrepareTheta'
KeccakF-1600-opt64.c: 123 | #define thetaRhoPiChiIota(i, A, E) thetaRhoPiChiIotaPrepareTheta(i, A, E)
KeccakF-1600-opt64.c: | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
KeccakF-1600-opt64.c: KeccakF-1600-unrolling.macros:40:5: note: in expansion of macro 'thetaRhoPiChiIota'
KeccakF-1600-opt64.c: 40 | thetaRhoPiChiIota(23, E, A) \
KeccakF-1600-opt64.c: | ^~~~~~~~~~~~~~~~~
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c:185:5: note: in expansion of macro 'rounds'
KeccakF-1600-opt64.c: 185 | rounds
KeccakF-1600-opt64.c: | ^~~~~~
KeccakF-1600-opt64.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/11/include/x86intrin.h:38,
KeccakF-1600-opt64.c: from KeccakF-1600-opt64.c:74:
KeccakF-1600-opt64.c: /usr/lib/gcc/x86_64-linux-gnu/11/include/xopintrin.h:245:1: error: inlining failed in call to 'always_inline' '_mm_rot_epi64': target specific option mismatch
KeccakF-1600-opt64.c: 245 | _mm_rot_epi64(__m128i __A, __m128i __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 -fPIC -fPIE	T:xopu24
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:xopu24
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:xopu24
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:xopu24