Implementation notes: amd64, h3neo, crypto_aead/romulusn

Computer: h3neo
Microarchitecture: amd64; K10 45nm (100f63)
Architecture: amd64
CPU ID: AuthenticAMD-00100f63-078bfbff
SUPERCOP version: 20240107
Operation: crypto_aead
Primitive: romulusn

Time	Object size	Test size	Implementation	Compiler	Benchmark date	SUPERCOP version
494784	11640 592 0	26579 1444 1024	`T:opt32t`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231220	20231212
494886	10814 592 0	26659 1444 1024	`T:opt32t`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231220	20231212
496068	11326 592 0	27819 1444 1024	`T:opt32t`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231220	20231212
500447	13770 608 0	30629 1420 1088	`T:opt32t`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231220	20231212
503069	9525 608 0	25125 1420 1088	`T:opt32t`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231220	20231212
505773	9413 592 0	23403 1444 1024	`T:opt32t`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231220	20231212
518070	8043 608 0	21512 1396 1056	`T:opt32t`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231220	20231212
547458	7930 592 0	21237 1436 1024	`T:opt32t`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231220	20231212
556401	8985 608 0	23660 1412 1088	`T:opt32t`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231220	20231212
1170833	30572 640 0	46411 1492 1024	`aadomn/opt32`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231220	20231212
1171441	30849 640 0	46587 1492 1024	`T:fixslice_opt32`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231220	20231212
1193250	30218 640 0	45443 1492 1024	`aadomn/opt32`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231220	20231212
1197403	29898 640 0	44203 1492 1024	`aadomn/opt32`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231220	20231212
1203863	30616 640 0	45747 1492 1024	`T:fixslice_opt32`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231220	20231212
1219264	30456 640 0	44667 1492 1024	`T:fixslice_opt32`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231220	20231212
1385901	21770 640 0	37524 1460 1088	`T:fixslice_opt32`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231220	20231212
1393934	20652 640 0	36381 1452 1088	`aadomn/opt32`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231220	20231212
1394643	23422 640 0	37395 1492 1024	`T:fixslice_opt32`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231220	20231212
1397347	22848 640 0	36891 1492 1024	`aadomn/opt32`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231220	20231212
1435058	20695 640 0	35668 1460 1088	`T:fixslice_opt32`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231220	20231212
1450601	19438 640 0	34276 1444 1088	`aadomn/opt32`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231220	20231212
1702284	20097 640 0	33317 1484 1024	`aadomn/opt32`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231220	20231212
1743328	20902 640 0	34069 1484 1024	`T:fixslice_opt32`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231220	20231212
1757676	21107 640 0	34687 1436 1056	`T:fixslice_opt32`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231220	20231212
1764292	20001 640 0	33560 1428 1056	`aadomn/opt32`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231220	20231212
1827536	37270 640 0	54348 1460 1088	`T:fixslice_opt32`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231220	20231212
1828004	35882 640 0	52957 1452 1088	`aadomn/opt32`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231220	20231212
2669107	13096 12 0	29411 856 1024	`T:ref`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231220	20231212
3292495	21522 12 0	38557 816 1088	`T:ref`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231220	20231212
3859413	10544 12 0	25339 856 1024	`T:ref`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231220	20231212
3865897	10504 12 0	26219 856 1024	`T:ref`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231220	20231212
9666242	5200 12 0	18533 848 1024	`T:ref`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231220	20231212
10566857	7683 12 0	21731 856 1024	`T:ref`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231220	20231212
11942041	6618 12 0	22333 816 1088	`T:ref`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231220	20231212
13140027	5432 12 0	20212 808 1088	`T:ref`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231220	20231212
13237407	4532 12 0	18064 792 1056	`T:ref`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231220	20231212

Compiler output

Implementation: aadomn/x86
Security model: constbranchindex
Compiler: clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE

skinny128.c: skinny128.c:115:5: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'skinny128_384_plus' that is compiled without support for 'ssse3'
skinny128.c: DOUBLE_ROUND(rtk_23);
skinny128.c: ^
skinny128.c: skinny128.c:78:5: note: expanded from macro 'DOUBLE_ROUND'
skinny128.c: SBOX_ARK_EVEN(rtk_23); \
skinny128.c: ^
skinny128.c: skinny128.c:23:13: note: expanded from macro 'SBOX_ARK_EVEN'
skinny128.c: state = _mm_shuffle_epi8(s1, state); /* apply inner S-box S1 */ \
skinny128.c: ^
skinny128.c: skinny128.c:115:5: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'skinny128_384_plus' that is compiled without support for 'ssse3'
skinny128.c: skinny128.c:78:5: note: expanded from macro 'DOUBLE_ROUND'
skinny128.c: SBOX_ARK_EVEN(rtk_23); \
skinny128.c: ^
skinny128.c: skinny128.c:24:13: note: expanded from macro 'SBOX_ARK_EVEN'
skinny128.c: tmp0 = _mm_shuffle_epi8(s0, tmp0); /* apply inner S-box S0 */ \
skinny128.c: ^
skinny128.c: skinny128.c:115:5: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'skinny128_384_plus' that is compiled without support for 'ssse3'
skinny128.c: skinny128.c:78:5: note: expanded from macro 'DOUBLE_ROUND'
skinny128.c: SBOX_ARK_EVEN(rtk_23); \
skinny128.c: ^
skinny128.c: skinny128.c:32:13: note: expanded from macro 'SBOX_ARK_EVEN'
skinny128.c: tmp0 = _mm_shuffle_epi8(s3, tmp0); /* apply inner S-box S3 */ \
skinny128.c: ^
skinny128.c: skinny128.c:115:5: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'skinny128_384_plus' that is compiled without support for 'ssse3'
skinny128.c: skinny128.c:78:5: note: expanded from macro 'DOUBLE_ROUND'
skinny128.c: ...

Number of similar (compiler,implementation) pairs: 5, namely:

Compiler	Implementations
clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	aadomn/x86
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	aadomn/x86
clang -march=native -O -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	aadomn/x86
clang -march=native -Os -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	aadomn/x86
clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	aadomn/x86

Compiler output

Implementation: aadomn/x86
Security model: constbranchindex
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE

skinny128.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/11/include/immintrin.h:37,
skinny128.c: from skinny128.c:10:
skinny128.c: skinny128.c: In function 'skinny128_384_plus':
skinny128.c: /usr/lib/gcc/x86_64-linux-gnu/11/include/tmmintrin.h:136:1: error: inlining failed in call to 'always_inline' '_mm_shuffle_epi8': target specific option mismatch
skinny128.c: 136 | _mm_shuffle_epi8 (__m128i __X, __m128i __Y)
skinny128.c: | ^~~~~~~~~~~~~~~~
skinny128.c: skinny128.c:70:13: note: called from here
skinny128.c: 70 | state = _mm_shuffle_epi8(state, m1); /* state <- (r2, - , r2, r0) */ \
skinny128.c: | ^~~~~~~~~~~~~~~~~~~~~~~~~~~
skinny128.c: skinny128.c:81:5: note: in expansion of macro 'SR_MC'
skinny128.c: 81 | SR_MC(); \
skinny128.c: | ^~~~~
skinny128.c: skinny128.c:134:5: note: in expansion of macro 'DOUBLE_ROUND'
skinny128.c: 134 | DOUBLE_ROUND(rtk_23+304);
skinny128.c: | ^~~~~~~~~~~~
skinny128.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/11/include/immintrin.h:37,
skinny128.c: from skinny128.c:10:
skinny128.c: /usr/lib/gcc/x86_64-linux-gnu/11/include/tmmintrin.h:136:1: error: inlining failed in call to 'always_inline' '_mm_shuffle_epi8': target specific option mismatch
skinny128.c: 136 | _mm_shuffle_epi8 (__m128i __X, __m128i __Y)
skinny128.c: | ^~~~~~~~~~~~~~~~
skinny128.c: skinny128.c:68:14: note: called from here
skinny128.c: 68 | tmp0 = _mm_shuffle_epi8(state, m0); /* tmp0 <- (r3, r0, r1, r2) */ \
skinny128.c: | ^~~~~~~~~~~~~~~~~~~~~~~~~~~
skinny128.c: skinny128.c:81:5: note: in expansion of macro 'SR_MC'
skinny128.c: 81 | SR_MC(); \
skinny128.c: ...

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

Compiler	Implementations
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE	aadomn/x86
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv -fPIC -fPIE	aadomn/x86
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv -fPIC -fPIE	aadomn/x86
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv -fPIC -fPIE	aadomn/x86