Implementation notes: amd64, h3neo, crypto_hash/blake64

Computer: h3neo
Microarchitecture: amd64; K10 45nm (100f63)
Architecture: amd64
CPU ID: AuthenticAMD-00100f63-078bfbff
SUPERCOP version: 20240625
Operation: crypto_hash
Primitive: blake64

Time	Object size	Test size	Implementation	Compiler	Benchmark date	SUPERCOP version
12463	21995 0 0	35541 804 960	`T:sphlib`	`gcc_-march=native_-mtune=native_-O3_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240702	20240625
12476	21685 0 0	33885 804 960	`T:sphlib`	`gcc_-march=native_-mtune=native_-O2_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240702	20240625
13427	21923 0 0	31965 836 896	`T:sphlib`	`clang_-march=native_-Os_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240702	20240625
13490	21712 0 0	33315 844 896	`T:sphlib`	`clang_-march=native_-O2_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240702	20240625
13510	21744 0 0	32395 844 896	`T:sphlib`	`clang_-march=native_-O_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240702	20240625
13517	21744 0 0	34267 844 896	`T:sphlib`	`clang_-march=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240702	20240625
13520	21568 0 0	34531 844 896	`T:sphlib`	`clang_-mcpu=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240702	20240625
13793	20651 0 0	32125 804 960	`T:sphlib`	`gcc_-march=native_-mtune=native_-O_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240702	20240625
15405	19599 0 0	29968 780 928	`T:sphlib`	`gcc_-march=native_-mtune=native_-Os_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240702	20240625
18609	8035 0 0	19525 804 960	`T:sphlib-small`	`gcc_-march=native_-mtune=native_-O_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240702	20240625
19238	7449 0 0	17808 780 928	`T:sphlib-small`	`gcc_-march=native_-mtune=native_-Os_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240702	20240625
21559	3766 0 0	14048 780 928	`T:ref`	`gcc_-march=native_-mtune=native_-Os_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240701	20240625
22840	4792 0 0	16227 844 896	`T:ref`	`clang_-march=native_-O2_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240701	20240625
22926	4792 0 0	17147 844 896	`T:ref`	`clang_-march=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240701	20240625
22953	8576 0 0	21099 844 896	`T:sphlib-small`	`clang_-march=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240702	20240625
23021	8496 0 0	20115 844 896	`T:sphlib-small`	`clang_-march=native_-O2_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240702	20240625
23159	4936 0 0	17731 844 896	`T:ref`	`clang_-mcpu=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240701	20240625
23215	4267 0 0	15588 796 960	`T:ref`	`gcc_-march=native_-mtune=native_-O_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240701	20240625
23258	8303 0 0	18349 836 896	`T:sphlib-small`	`clang_-march=native_-Os_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240702	20240625
23278	8528 0 0	19195 844 896	`T:sphlib-small`	`clang_-march=native_-O_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240702	20240625
23662	11093 0 0	23325 804 960	`T:sphlib-small`	`gcc_-march=native_-mtune=native_-O2_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240702	20240625
23960	5562 0 0	16091 844 896	`T:ref`	`clang_-march=native_-O_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240701	20240625
24251	11531 0 0	25077 804 960	`T:sphlib-small`	`gcc_-march=native_-mtune=native_-O3_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240701	20240625
24980	3674 0 0	13549 836 896	`T:ref`	`clang_-march=native_-Os_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240701	20240625
27202	8592 0 0	21587 844 896	`T:sphlib-small`	`clang_-mcpu=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240702	20240625
27846	15139 0 0	28637 804 960	`T:sse2`	`gcc_-march=native_-mtune=native_-O3_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240702	20240625
28339	15039 0 0	27181 804 960	`T:sse2`	`gcc_-march=native_-mtune=native_-O2_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240702	20240625
29505	6764 0 0	20229 804 960	`T:ref`	`gcc_-march=native_-mtune=native_-O3_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240701	20240625
29666	5303 0 0	17445 804 960	`T:ref`	`gcc_-march=native_-mtune=native_-O2_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240701	20240625
33041	13243 0 0	23536 780 928	`T:sse2`	`gcc_-march=native_-mtune=native_-Os_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240702	20240625
33571	15844 0 0	27180 796 960	`T:sse2`	`gcc_-march=native_-mtune=native_-O_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240702	20240625
33643	17352 0 0	30163 844 896	`T:sse2`	`clang_-mcpu=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240702	20240625
35683	18122 0 0	28651 844 896	`T:sse2`	`clang_-march=native_-O_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240702	20240625
35694	17592 0 0	29931 844 896	`T:sse2`	`clang_-march=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240702	20240625
35718	17592 0 0	29011 844 896	`T:sse2`	`clang_-march=native_-O2_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240702	20240625
35802	16424 0 0	26269 836 896	`T:sse2`	`clang_-march=native_-Os_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240702	20240625

Compiler output

hash.c: hash.c:268:3: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'blake64_compress' that is compiled without support for 'ssse3'
hash.c:   round(0);
hash.c:   ^
hash.c: hash.c:172:13: note: expanded from macro 'round'
hash.c:     row4a = _mm_shuffle_epi8(row4a, r16); \
hash.c:             ^
hash.c: hash.c:268:3: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'blake64_compress' that is compiled without support for 'ssse3'
hash.c: hash.c:193:13: note: expanded from macro 'round'
hash.c:     row4b = _mm_shuffle_epi8(row4b, r16); \
hash.c:             ^
hash.c: hash.c:268:3: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'blake64_compress' that is compiled without support for 'ssse3'
hash.c: hash.c:228:13: note: expanded from macro 'round'
hash.c:     row4a = _mm_shuffle_epi8(row4a, r16); \
hash.c:             ^
hash.c: hash.c:268:3: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'blake64_compress' that is compiled without support for 'ssse3'
hash.c: hash.c:249:13: note: expanded from macro 'round'
hash.c:     row4b = _mm_shuffle_epi8(row4b, r16); \
hash.c:             ^
hash.c: hash.c:269:3: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'blake64_compress' that is compiled without support for 'ssse3'
hash.c:   round(1);
hash.c:   ^
hash.c: hash.c:172:13: note: expanded from macro 'round'
hash.c:     row4a = _mm_shuffle_epi8(row4a, r16); \
hash.c:             ^
hash.c: hash.c:269:3: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'blake64_compress' that is compiled without support for 'ssse3'
hash.c: ...

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

Implementation	Compiler
`T:ssse3`	`clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)`
`T:ssse3`	`clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)`
`T:ssse3`	`clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)`
`T:ssse3`	`clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)`
`T:ssse3`	`clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)`

Compiler output

hash.c: In file included from hash.c:5:
hash.c: hash.c: In function 'blake64_compress':
hash.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
hash.c:   136 | _mm_shuffle_epi8 (__m128i __X, __m128i __Y)
hash.c:       | ^~~~~~~~~~~~~~~~
hash.c: hash.c:249:13: note: called from here
hash.c:   249 |     row4b = _mm_shuffle_epi8(row4b, r16); \
hash.c:       |             ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
hash.c: hash.c:281:3: note: in expansion of macro 'round'
hash.c:   281 |   round(13);
hash.c:       |   ^~~~~
hash.c: In file included from hash.c:5:
hash.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
hash.c:   136 | _mm_shuffle_epi8 (__m128i __X, __m128i __Y)
hash.c:       | ^~~~~~~~~~~~~~~~
hash.c: hash.c:228:13: note: called from here
hash.c:   228 |     row4a = _mm_shuffle_epi8(row4a, r16); \
hash.c:       |             ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
hash.c: hash.c:281:3: note: in expansion of macro 'round'
hash.c:   281 |   round(13);
hash.c:       |   ^~~~~
hash.c: In file included from hash.c:5:
hash.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
hash.c:   136 | _mm_shuffle_epi8 (__m128i __X, __m128i __Y)
hash.c:       | ^~~~~~~~~~~~~~~~
hash.c: ...

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

Implementation	Compiler
`T:ssse3`	`gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)`
`T:ssse3`	`gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)`
`T:ssse3`	`gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)`
`T:ssse3`	`gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)`