Implementation notes: x86, thoth, crypto_hash/keccakc512

Computer: thoth
Architecture: x86
CPU ID: AuthenticAMD-00000622-0183f9ff
SUPERCOP version: 20160806
Operation: crypto_hash
Primitive: keccakc512
TimeImplementationCompilerBenchmark dateSUPERCOP version
56397mmxu1gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv2016072620160724
59130mmxu1gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv2016072620160724
60119mmxu1gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv2016072620160724
62135mmxu1gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv2016072620160724
73289opt32bi-s2lcu4gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv2016072620160724
74472simple32bigcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv2016072620160724
75151opt32bi-s2lcu4clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments2016072620160724
75510opt32bi-rvku2gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv2016072620160724
75738inplace32bigcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv2016072620160724
76555opt32biT-s2lcu4clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments2016072620160724
77031opt32biT-s2lcu4gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv2016072620160724
77222simple32biclang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments2016072620160724
78198opt32biT-s2lcu4gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv2016072620160724
79785opt32bi-rvku2clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments2016072620160724
80327opt32bi-s2lcu4gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv2016072620160724
80468opt32bi-rvku2gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv2016072620160724
81031opt32bi-s2lcu4gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv2016072620160724
81083simple32bigcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv2016072620160724
81810opt32biT-s2lcu4gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv2016072620160724
82861simple32bigcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv2016072620160724
84155inplace32bigcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv2016072620160724
84298inplace32bigcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv2016072620160724
84577sphlibclang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments2016072620160724
84622opt32bi-s2lcu4gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv2016072620160724
84636inplace32bigcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv2016072620160724
85185opt32biT-s2lcu4gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv2016072620160724
85572opt32bi-rvku2gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv2016072620160724
87836sphlib-smallclang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments2016072620160724
89608opt32bi-rvku2gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv2016072620160724
91129simple32bigcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv2016072620160724
92459sphlibgcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv2016072620160724
93038sphlib-smallgcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv2016072620160724
93204sphlibgcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv2016072620160724
93547sphlib-smallgcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv2016072620160724
94077sphlibgcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv2016072620160724
94865opt64lcu6clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments2016072620160724
96588sphlib-smallgcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv2016072620160724
101160opt64u6clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments2016072620160724
101428simpleclang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments2016072620160724
106257sphlibgcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv2016072620160724
106402sphlib-smallgcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv2016072620160724
106404inplaceclang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments2016072620160724
121689opt64lcu24gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv2016072620160724
129810opt64lcu24gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv2016072620160724
130083opt64lcu24gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv2016072620160724
135297opt64lcu24gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv2016072620160724
135839opt64lcu6gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv2016072620160724
136425simplegcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv2016072620160724
137310inplacegcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv2016072620160724
138787opt64u6gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv2016072620160724
141978inplacegcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv2016072620160724
142047inplacegcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv2016072620160724
142917inplace32biclang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments2016072620160724
145352inplacegcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv2016072620160724
145534opt64lcu6gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv2016072620160724
146321simplegcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv2016072620160724
149019opt64lcu6gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv2016072620160724
150536simplegcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv2016072620160724
150555opt64lcu6gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv2016072620160724
150641opt64u6gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv2016072620160724
152241opt64u6gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv2016072620160724
153291simplegcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv2016072620160724
154426opt64u6gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv2016072620160724
154617opt64lcu24clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments2016072620160724
248400compactgcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv2016072620160724
263696compactclang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments2016072620160724
342513mmxu1clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments2016072620160724
348930compactgcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv2016072620160724
358721compactgcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv2016072620160724
369264compactgcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv2016072620160724
502176compact8clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments2016072620160724
503341compact8gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv2016072620160724
595654compact8gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv2016072620160724
660319compact8gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv2016072620160724
672664compact8gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv2016072620160724

Checksum failure

Implementation: crypto_hash/keccakc512/opt64lcu24shld
Compiler: clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments
025c47113c7aef56cd4b370d3cc96a8324fa4dedfe7fe5c89296200e47ff97cb
Number of similar (compiler,implementation) pairs: 5, namely:
CompilerImplementations
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments opt64lcu24shld
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv opt64lcu24shld
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv opt64lcu24shld
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv opt64lcu24shld
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv opt64lcu24shld

Checksum failure

Implementation: crypto_hash/keccakc512/gil/singlefile
Compiler: clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments
ebbed3a47232047ad399cb77b6324d6c902f4715de871750545b1a3b12046ebe
Number of similar (compiler,implementation) pairs: 10, namely:
CompilerImplementations
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments gil/singlefile gil/singlefile_unrolled
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv gil/singlefile gil/singlefile_unrolled
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv gil/singlefile gil/singlefile_unrolled
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv gil/singlefile gil/singlefile_unrolled
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv gil/singlefile gil/singlefile_unrolled

Compiler output

Implementation: crypto_hash/keccakc512/sseu2
Compiler: clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c:184:5: error: always_inline function '_mm_load_si128' requires target feature 'sse2', but would be inlined into function 'KeccakPermutationOnWords' that is compiled without support for 'sse2'
KeccakF-1600-opt64.c: copyFromState(A, state)
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: ./KeccakF-1600-simd128.macros:547:19: note: expanded from macro 'copyFromState'
KeccakF-1600-opt64.c: X##bae.v128 = LOAD128(state[ 0]); \
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c:51:33: note: expanded from macro 'LOAD128'
KeccakF-1600-opt64.c: #define LOAD128(a) _mm_load_si128((const V128 *)&(a))
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c:184:5: error: always_inline function '_mm_unpackhi_epi64' requires target feature 'sse2', but would be inlined into function 'KeccakPermutationOnWords' that is compiled without support for 'sse2'
KeccakF-1600-opt64.c: ./KeccakF-1600-simd128.macros:549:13: note: expanded from macro 'copyFromState'
KeccakF-1600-opt64.c: X##be = GET64HIHI(X##bae.v128, X##bae.v128); \
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c:58:33: note: expanded from macro 'GET64HIHI'
KeccakF-1600-opt64.c: #define GET64HIHI(a, b) _mm_unpackhi_epi64(a, b)
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c:184:5: error: always_inline function '_mm_load_si128' requires target feature 'sse2', but would be inlined into function 'KeccakPermutationOnWords' that is compiled without support for 'sse2'
KeccakF-1600-opt64.c: ./KeccakF-1600-simd128.macros:551:19: note: expanded from macro 'copyFromState'
KeccakF-1600-opt64.c: X##bio.v128 = LOAD128(state[ 2]); \
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c:51:33: note: expanded from macro 'LOAD128'
KeccakF-1600-opt64.c: #define LOAD128(a) _mm_load_si128((const V128 *)&(a))
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c:184:5: error: always_inline function '_mm_unpackhi_epi64' requires target feature 'sse2', but would be inlined into function 'KeccakPermutationOnWords' that is compiled without support for 'sse2'
KeccakF-1600-opt64.c: ./KeccakF-1600-simd128.macros:553:13: note: expanded from macro 'copyFromState'
KeccakF-1600-opt64.c: ...

Number of similar (compiler,implementation) pairs: 1, namely:
CompilerImplementations
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments sseu2

Compiler output

Implementation: crypto_hash/keccakc512/xopu24
Compiler: clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c:184:5: error: always_inline function '_mm_load_si128' requires target feature 'sse2', but would be inlined into function 'KeccakPermutationOnWords' that is compiled without support for 'sse2'
KeccakF-1600-opt64.c: copyFromState(A, state)
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: ./KeccakF-1600-xop.macros:482:14: note: expanded from macro 'copyFromState'
KeccakF-1600-opt64.c: X##bae = LOAD128(state[ 0]); \
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c:87:33: note: expanded from macro 'LOAD128'
KeccakF-1600-opt64.c: #define LOAD128(a) _mm_load_si128((const V128 *)&(a))
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c:184:5: error: always_inline function '_mm_unpackhi_epi64' requires target feature 'sse2', but would be inlined into function 'KeccakPermutationOnWords' that is compiled without support for 'sse2'
KeccakF-1600-opt64.c: ./KeccakF-1600-xop.macros:484:13: note: expanded from macro 'copyFromState'
KeccakF-1600-opt64.c: X##be = GET64HIHI(X##bae, X##bae); \
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c:96:33: note: expanded from macro 'GET64HIHI'
KeccakF-1600-opt64.c: #define GET64HIHI(a, b) _mm_unpackhi_epi64(a, b)
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c:184:5: error: always_inline function '_mm_load_si128' requires target feature 'sse2', but would be inlined into function 'KeccakPermutationOnWords' that is compiled without support for 'sse2'
KeccakF-1600-opt64.c: ./KeccakF-1600-xop.macros:486:14: note: expanded from macro 'copyFromState'
KeccakF-1600-opt64.c: X##bio = LOAD128(state[ 2]); \
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c:87:33: note: expanded from macro 'LOAD128'
KeccakF-1600-opt64.c: #define LOAD128(a) _mm_load_si128((const V128 *)&(a))
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c:184:5: error: always_inline function '_mm_unpackhi_epi64' requires target feature 'sse2', but would be inlined into function 'KeccakPermutationOnWords' that is compiled without support for 'sse2'
KeccakF-1600-opt64.c: ./KeccakF-1600-xop.macros:488:13: note: expanded from macro 'copyFromState'
KeccakF-1600-opt64.c: ...

Number of similar (compiler,implementation) pairs: 1, namely:
CompilerImplementations
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments xopu24

Compiler output

Implementation: crypto_hash/keccakc512/xopu24
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv
KeccakF-1600-opt64.c: In file included from KeccakF-1600-opt64.c:130:0:
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c: In function 'KeccakPermutationOnWords':
KeccakF-1600-opt64.c: KeccakF-1600-xop.macros:482:12: warning: SSE vector return without SSE enabled changes the ABI [-Wpsabi]
KeccakF-1600-opt64.c: X##bae = LOAD128(state[ 0]); \
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c:184:5: note: in expansion of macro 'copyFromState'
KeccakF-1600-opt64.c: copyFromState(A, state)
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: In file included from /usr/lib/gcc/i686-linux-gnu/5/include/xmmintrin.h:1249:0,
KeccakF-1600-opt64.c: from /usr/lib/gcc/i686-linux-gnu/5/include/x86intrin.h:31,
KeccakF-1600-opt64.c: from KeccakF-1600-opt64.c:74:
KeccakF-1600-opt64.c: /usr/lib/gcc/i686-linux-gnu/5/include/emmintrin.h:720:1: error: inlining failed in call to always_inline '_mm_storel_epi64': target specific option mismatch
KeccakF-1600-opt64.c: _mm_storel_epi64 (__m128i *__P, __m128i __B)
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c:80:33: error: called from here
KeccakF-1600-opt64.c: #define STORE64(a, b) _mm_storel_epi64((V64 *)&(a), b)
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-xop.macros:557:5: note: in expansion of macro 'STORE64'
KeccakF-1600-opt64.c: STORE64(state[24], X##su); \
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-unrolling.macros:41:5: note: in expansion of macro 'copyToState'
KeccakF-1600-opt64.c: copyToState(state, 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: rounds
KeccakF-1600-opt64.c: ...

Number of similar (compiler,implementation) pairs: 4, namely:
CompilerImplementations
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv xopu24
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv xopu24
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv xopu24
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv xopu24

Compiler output

Implementation: crypto_hash/keccakc512/sseu2
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv
KeccakF-1600-opt64.c: In file included from KeccakF-1600-opt64.c:67:0:
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c: In function 'KeccakPermutationOnWords':
KeccakF-1600-opt64.c: KeccakF-1600-simd128.macros:547:17: warning: SSE vector return without SSE enabled changes the ABI [-Wpsabi]
KeccakF-1600-opt64.c: X##bae.v128 = LOAD128(state[ 0]); \
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c:184:5: note: in expansion of macro 'copyFromState'
KeccakF-1600-opt64.c: copyFromState(A, state)
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: In file included from /usr/lib/gcc/i686-linux-gnu/5/include/xmmintrin.h:1249:0,
KeccakF-1600-opt64.c: from /usr/lib/gcc/i686-linux-gnu/5/include/x86intrin.h:31,
KeccakF-1600-opt64.c: from KeccakF-1600-opt64.c:31:
KeccakF-1600-opt64.c: /usr/lib/gcc/i686-linux-gnu/5/include/emmintrin.h:1286:1: error: inlining failed in call to always_inline '_mm_xor_si128': target specific option mismatch
KeccakF-1600-opt64.c: _mm_xor_si128 (__m128i __A, __m128i __B)
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c:45:35: error: called from here
KeccakF-1600-opt64.c: #define XOReq64(a, b) a = _mm_xor_si128(a, b)
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-simd128.macros:600:5: note: in expansion of macro 'XOReq64'
KeccakF-1600-opt64.c: XOReq64(Cu, X##su); \
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c:184:5: note: in expansion of macro 'copyFromState'
KeccakF-1600-opt64.c: copyFromState(A, state)
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: In file included from /usr/lib/gcc/i686-linux-gnu/5/include/xmmintrin.h:1249:0,
KeccakF-1600-opt64.c: from /usr/lib/gcc/i686-linux-gnu/5/include/x86intrin.h:31,
KeccakF-1600-opt64.c: ...

Number of similar (compiler,implementation) pairs: 4, namely:
CompilerImplementations
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv sseu2
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv sseu2
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv sseu2
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv sseu2