Implementation notes: amd64, hydra5, crypto_hash/fugue512

Computer: hydra5
Microarchitecture: amd64; K10 32nm (300f10)
Architecture: amd64
CPU ID: AuthenticAMD-00300f10-178bfbff
SUPERCOP version: 20231107
Operation: crypto_hash
Primitive: fugue512

Time	Object size	Test size	Implementation	Compiler	Benchmark date	SUPERCOP version
99396	37741 0 0	48313 796 928	`T:sphlib`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231109	20231107
99828	40081 0 0	51858 868 896	`T:sphlib`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20230612	20230530
99875	39844 0 0	50706 868 896	`T:sphlib`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20230612	20230530
99880	40081 0 0	52746 868 896	`T:sphlib`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20230612	20230530
99936	40081 0 0	52746 868 896	`T:sphlib`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20230612	20230530
100183	42460 0 0	56230 820 960	`T:sphlib`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231109	20231107
100496	40491 0 0	52101 812 960	`T:sphlib`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231109	20231107
100525	42108 0 0	54542 820 960	`T:sphlib`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231109	20231107
112330	41465 0 0	51564 860 896	`T:sphlib`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20230612	20230530

Compiler output

Implementation: T:ccalik/aesni
Security model: timingleaks
Compiler: clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE

hash.c: hash.c:449:4: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'Compress256' that is compiled without support for 'ssse3'
hash.c: SUBROUND256_2(ctx->state[8], ctx->state[9], ctx->state[7], ctx->state[2], ctx->state[7], ctx->state[8], ctx->state[6], ctx->state[1]);
hash.c: ^
hash.c: hash.c:340:2: note: expanded from macro 'SUBROUND256_2'
hash.c: SUBSTITUTE(r0, _t1, _t2, _t3, _t0);\
hash.c: ^
hash.c: hash.c:249:10: note: expanded from macro 'SUBSTITUTE'
hash.c: _t2 = _mm_shuffle_epi8(r0, M128(_inv_shift_rows));\
hash.c: ^
hash.c: hash.c:449:4: error: always_inline function '_mm_aesenclast_si128' requires target feature 'aes', but would be inlined into function 'Compress256' that is compiled without support for 'aes'
hash.c: hash.c:340:2: note: expanded from macro 'SUBROUND256_2'
hash.c: SUBSTITUTE(r0, _t1, _t2, _t3, _t0);\
hash.c: ^
hash.c: hash.c:250:10: note: expanded from macro 'SUBSTITUTE'
hash.c: _t2 = _mm_aesenclast_si128(_t2, M128(_zero))
hash.c: ^
hash.c: hash.c:449:4: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'Compress256' that is compiled without support for 'ssse3'
hash.c: hash.c:341:2: note: expanded from macro 'SUBROUND256_2'
hash.c: SUPERMIX(_t2, _t3, _t0, _t1, r0);\
hash.c: ^
hash.c: hash.c:253:2: note: expanded from macro 'SUPERMIX'
hash.c: PRESUPERMIX(t0, t1, t2, t3, t4);\
hash.c: ^
hash.c: hash.c:245:25: note: expanded from macro 'PRESUPERMIX'
hash.c: s2 = _mm_xor_si128(s2, _mm_shuffle_epi8(M128(_mul2mask), t1));\
hash.c: ...

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

Compiler	Implementations
clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:ccalik/aesni
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:ccalik/aesni
clang -march=native -O -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:ccalik/aesni
clang -march=native -Os -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:ccalik/aesni
clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:ccalik/aesni

Compiler output

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

hash.c: In file included from hash_api.h:29,
hash.c: from hash.c:20:
hash.c: hash.c: In function 'Compress256':
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:301:22: note: called from here
hash.c: 301 | t0 = _mm_shuffle_epi8(t0, M128(_supermix4c));\
hash.c: | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
hash.c: hash.c:254:9: note: in expansion of macro 'POSTSUPERMIX'
hash.c: 254 | POSTSUPERMIX(t0, t1, t2, t3, t4)
hash.c: | ^~~~~~~~~~~~
hash.c: hash.c:348:9: note: in expansion of macro 'SUPERMIX'
hash.c: 348 | SUPERMIX(_t2, _t3, _t0, _t1, s0);\
hash.c: | ^~~~~~~~
hash.c: hash.c:449:25: note: in expansion of macro 'SUBROUND256_2'
hash.c: 449 | SUBROUND256_2(ctx->state[8], ctx->state[9], ctx->state[7], ctx->state[2], ctx->state[7], ctx->state[8], ctx->state[6], ctx->state[1]);
hash.c: | ^~~~~~~~~~~~~
hash.c: In file included from hash_api.h:29,
hash.c: from hash.c:20:
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:298:22: note: called from here
hash.c: 298 | t0 = _mm_shuffle_epi8(t0, M128(_supermix4b));\
hash.c: ...

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

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

Compiler output

Implementation: T:ccalik/vperm
Security model: timingleaks
Compiler: clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE

hash.c: hash.c:447:4: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'Compress256' that is compiled without support for 'ssse3'
hash.c: TIX256(pmsg, ctx->state[1], ctx->state[0], ctx->state[6], ctx->state[8], _t0, _t1, _t2);
hash.c: ^
hash.c: hash.c:168:2: note: expanded from macro 'TIX256'
hash.c: TRANSFORM(t1, _k_ipt, t2, t3);\
hash.c: ^
hash.c: ./vperm.h:52:7: note: expanded from macro 'TRANSFORM'
hash.c: t1 = _mm_shuffle_epi8(*((__m128i*)table + 1), t1);\
hash.c: ^
hash.c: hash.c:447:4: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'Compress256' that is compiled without support for 'ssse3'
hash.c: hash.c:168:2: note: expanded from macro 'TIX256'
hash.c: TRANSFORM(t1, _k_ipt, t2, t3);\
hash.c: ^
hash.c: ./vperm.h:53:7: note: expanded from macro 'TRANSFORM'
hash.c: x = _mm_shuffle_epi8(*((__m128i*)table + 0), x);\
hash.c: ^
hash.c: hash.c:448:4: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'Compress256' that is compiled without support for 'ssse3'
hash.c: SUBROUND256_2(ctx->state[8], ctx->state[9], ctx->state[7], ctx->state[2], ctx->state[7], ctx->state[8], ctx->state[6], ctx->state[1]);
hash.c: ^
hash.c: hash.c:339:2: note: expanded from macro 'SUBROUND256_2'
hash.c: SUBSTITUTE(r0, _t1, _t2, _t3, _t0);\
hash.c: ^
hash.c: hash.c:264:20: note: expanded from macro 'SUBSTITUTE'
hash.c: #define SUBSTITUTE SUBSTITUTE_VPERM_CORE
hash.c: ^
hash.c: ...

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

Compiler	Implementations
clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:ccalik/vperm
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:ccalik/vperm
clang -march=native -O -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:ccalik/vperm
clang -march=native -Os -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:ccalik/vperm
clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:ccalik/vperm

Compiler output

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

hash.c: In file included from hash_api.h:29,
hash.c: from hash.c:20:
hash.c: hash.c: In function 'Compress256':
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:300:22: note: called from here
hash.c: 300 | t0 = _mm_shuffle_epi8(t0, M128(_supermix4c));\
hash.c: | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
hash.c: hash.c:268:9: note: in expansion of macro 'POSTSUPERMIX'
hash.c: 268 | POSTSUPERMIX(t0, t1, t2, t3, t4);\
hash.c: | ^~~~~~~~~~~~
hash.c: hash.c:347:9: note: in expansion of macro 'SUPERMIX'
hash.c: 347 | SUPERMIX(_t2, _t3, _t0, _t1, s0);\
hash.c: | ^~~~~~~~
hash.c: hash.c:448:25: note: in expansion of macro 'SUBROUND256_2'
hash.c: 448 | SUBROUND256_2(ctx->state[8], ctx->state[9], ctx->state[7], ctx->state[2], ctx->state[7], ctx->state[8], ctx->state[6], ctx->state[1]);
hash.c: | ^~~~~~~~~~~~~
hash.c: In file included from hash_api.h:29,
hash.c: from hash.c:20:
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:297:22: note: called from here
hash.c: 297 | t0 = _mm_shuffle_epi8(t0, M128(_supermix4b));\
hash.c: ...

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

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