Implementation notes: amd64, speed2supercop, crypto_sign/sphincss256harakasimple

Computer: speed2supercop
Microarchitecture: amd64; Haswell+AES (306c3)
Architecture: amd64
CPU ID: GenuineIntel-000306c3-1fc9cbf5
SUPERCOP version: 20240625
Operation: crypto_sign
Primitive: sphincss256harakasimple

Time	Object size	Test size	Implementation	Compiler	Benchmark date	SUPERCOP version
787727020	36847 0 1280	58568 808 2880	`T:aesni`	`clang_-march=native_-O2_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240710	20240625
795234584	31649 0 1280	54128 808 2880	`T:aesni`	`clang_-march=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240710	20240625
828266544	29556 0 1280	50943 768 2912	`T:aesni`	`gcc_-march=native_-mtune=native_-O3_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240710	20240625
863160980	25365 0 1280	44791 768 2912	`T:aesni`	`gcc_-march=native_-mtune=native_-O2_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240710	20240625
913420264	23255 0 1280	42367 768 2912	`T:aesni`	`gcc_-march=native_-mtune=native_-O_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240710	20240625
7765585132	16324 0 1280	35905 800 2912	`T:ref`	`clang_-march=native_-Os_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240710	20240625
7971264604	27561 0 1280	49663 808 2880	`T:ref`	`clang_-march=native_-O2_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240710	20240625
8089557776	17779 0 1280	37207 768 2912	`T:ref`	`gcc_-march=native_-mtune=native_-O2_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240710	20240625
8204060012	24469 0 1280	47223 808 2880	`T:ref`	`clang_-march=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240710	20240625
8386428416	27845 0 1280	49336 808 2848	`T:ref`	`clang_-mcpu=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240710	20240625
8819777220	23302 0 1280	44727 768 2912	`T:ref`	`gcc_-march=native_-mtune=native_-O3_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240710	20240625
11500296884	15942 0 1280	34888 808 2848	`T:ref`	`clang_-march=native_-O_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240710	20240625
13254846296	15391 0 1280	33479 760 2880	`T:ref`	`gcc_-march=native_-mtune=native_-Os_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240710	20240625
14101215716	15822 0 1280	34975 768 2912	`T:ref`	`gcc_-march=native_-mtune=native_-O_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240710	20240625

Test failure

error 111

Number of similar (implementation,compiler) pairs: 3, namely:

Implementation	Compiler
`T:aesni`	`clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_16.0.6_(27+b1))`
`T:aesni`	`clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_16.0.6_(27+b1))`
`T:aesni`	`gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (13.3.0)`

Compiler output

haraka.c: haraka.c:421:5: error: always_inline function '_mm_aesenc_si128' requires target feature 'aes', but would be inlined into function 'haraka512_perm' that is compiled without support for 'aes'
haraka.c:     AES4(s[0], s[1], s[2], s[3], rc);
haraka.c:     ^
haraka.c: haraka.c:36:8: note: expanded from macro 'AES4'
haraka.c:   s0 = _mm_aesenc_si128(s0, *(rci)); \
haraka.c:        ^
haraka.c: haraka.c:421:5: error: always_inline function '_mm_aesenc_si128' requires target feature 'aes', but would be inlined into function 'haraka512_perm' that is compiled without support for 'aes'
haraka.c: haraka.c:37:8: note: expanded from macro 'AES4'
haraka.c:   s1 = _mm_aesenc_si128(s1, *(rci + 1)); \
haraka.c:        ^
haraka.c: haraka.c:421:5: error: always_inline function '_mm_aesenc_si128' requires target feature 'aes', but would be inlined into function 'haraka512_perm' that is compiled without support for 'aes'
haraka.c: haraka.c:38:8: note: expanded from macro 'AES4'
haraka.c:   s2 = _mm_aesenc_si128(s2, *(rci + 2)); \
haraka.c:        ^
haraka.c: haraka.c:421:5: error: always_inline function '_mm_aesenc_si128' requires target feature 'aes', but would be inlined into function 'haraka512_perm' that is compiled without support for 'aes'
haraka.c: haraka.c:39:8: note: expanded from macro 'AES4'
haraka.c:   s3 = _mm_aesenc_si128(s3, *(rci + 3)); \
haraka.c:        ^
haraka.c: haraka.c:421:5: error: always_inline function '_mm_aesenc_si128' requires target feature 'aes', but would be inlined into function 'haraka512_perm' that is compiled without support for 'aes'
haraka.c: haraka.c:40:8: note: expanded from macro 'AES4'
haraka.c:   s0 = _mm_aesenc_si128(s0, *(rci + 4)); \
haraka.c:        ^
haraka.c: haraka.c:421:5: error: always_inline function '_mm_aesenc_si128' requires target feature 'aes', but would be inlined into function 'haraka512_perm' that is compiled without support for 'aes'
haraka.c: haraka.c:41:8: note: expanded from macro 'AES4'
haraka.c:   s1 = _mm_aesenc_si128(s1, *(rci + 5)); \
haraka.c: ...

Number of similar (implementation,compiler) pairs: 1, namely:

Implementation	Compiler
`T:aesni`	`clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_16.0.6_(27+b1))`