Implementation notes: amd64, dali, crypto_hash/keccakc256treed2

Computer: dali
Microarchitecture: amd64; Zen (820f01)
Architecture: amd64
CPU ID: AuthenticAMD-00820f01-178bfbff
SUPERCOP version: 20240625
Operation: crypto_hash
Primitive: keccakc256treed2
TimeObject sizeTest sizeImplementationCompilerBenchmark dateSUPERCOP version
11371136409 0 0147096 780 952T:sseu24gcc_-march=native_-mtune=native_-O_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
1143327376 0 038072 780 952T:sseu4gcc_-march=native_-mtune=native_-O_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
11466136779 0 0148816 780 952T:sseu24gcc_-march=native_-mtune=native_-O3_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
11485136437 0 0147456 780 952T:sseu24gcc_-march=native_-mtune=native_-O2_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
1149627851 0 039888 780 952T:sseu4gcc_-march=native_-mtune=native_-O3_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
1155727717 0 038736 780 952T:sseu4gcc_-march=native_-mtune=native_-O2_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
1188827938 0 038326 804 888T:sseu4clang_-march=native_-O_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
1192128240 0 040720 812 920T:sseu4clang_-march=native_-O2_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
1193128240 0 040832 812 920T:sseu4clang_-march=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
11942136460 0 0146115 756 920T:sseu24gcc_-march=native_-mtune=native_-Os_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
1195627529 0 037678 804 888T:sseu4clang_-march=native_-Os_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
1198327110 0 036755 756 920T:sseu4gcc_-march=native_-mtune=native_-Os_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625

Test failure

error 111
Number of similar (implementation,compiler) pairs: 4, namely:
ImplementationCompiler
T:sseu24clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)
T:sseu24clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)
T:sseu24clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)
T:sseu24clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)

Compiler output

DoublePermutation.c: DoublePermutation.c:507:5: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'KeccakDoublePermutationOnWords' that is compiled without support for 'ssse3'
DoublePermutation.c:     rounds
DoublePermutation.c:     ^
DoublePermutation.c: ./KeccakF-1600-unrolling.macros:17:5: note: expanded from macro 'rounds'
DoublePermutation.c:     thetaRhoPiChiIotaPrepareTheta( 0, A, E) \
DoublePermutation.c:     ^
DoublePermutation.c: DoublePermutation.c:136:11: note: expanded from macro 'thetaRhoPiChiIotaPrepareTheta'
DoublePermutation.c:     Bko = ROL64in128_8(A##mu); \
DoublePermutation.c:           ^
DoublePermutation.c: DoublePermutation.c:38:37: note: expanded from macro 'ROL64in128_8'
DoublePermutation.c:         #define ROL64in128_8(a)     _mm_shuffle_epi8(a, CONST128(rho8))
DoublePermutation.c:                                     ^
DoublePermutation.c: DoublePermutation.c:507:5: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'KeccakDoublePermutationOnWords' that is compiled without support for 'ssse3'
DoublePermutation.c: ./KeccakF-1600-unrolling.macros:17:5: note: expanded from macro 'rounds'
DoublePermutation.c:     thetaRhoPiChiIotaPrepareTheta( 0, A, E) \
DoublePermutation.c:     ^
DoublePermutation.c: DoublePermutation.c:161:11: note: expanded from macro 'thetaRhoPiChiIotaPrepareTheta'
DoublePermutation.c:     Bmu = ROL64in128_56(A##so); \
DoublePermutation.c:           ^
DoublePermutation.c: DoublePermutation.c:39:37: note: expanded from macro 'ROL64in128_56'
DoublePermutation.c:         #define ROL64in128_56(a)    _mm_shuffle_epi8(a, CONST128(rho56))
DoublePermutation.c:                                     ^
DoublePermutation.c: DoublePermutation.c:507:5: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'KeccakDoublePermutationOnWords' that is compiled without support for 'ssse3'
DoublePermutation.c: ./KeccakF-1600-unrolling.macros:18:5: note: expanded from macro 'rounds'
DoublePermutation.c:     thetaRhoPiChiIotaPrepareTheta( 1, E, A) \
DoublePermutation.c: ...
Number of similar (implementation,compiler) pairs: 1, namely:
ImplementationCompiler
T:sseu24clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)

Compiler output

DoublePermutation.c: DoublePermutation.c:507:5: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'KeccakDoublePermutationOnWords' that is compiled without support for 'ssse3'
DoublePermutation.c:     rounds
DoublePermutation.c:     ^
DoublePermutation.c: ./KeccakF-1600-unrolling.macros:90:9: note: expanded from macro 'rounds'
DoublePermutation.c:         thetaRhoPiChiIotaPrepareTheta(i  , A, E) \
DoublePermutation.c:         ^
DoublePermutation.c: DoublePermutation.c:136:11: note: expanded from macro 'thetaRhoPiChiIotaPrepareTheta'
DoublePermutation.c:     Bko = ROL64in128_8(A##mu); \
DoublePermutation.c:           ^
DoublePermutation.c: DoublePermutation.c:38:37: note: expanded from macro 'ROL64in128_8'
DoublePermutation.c:         #define ROL64in128_8(a)     _mm_shuffle_epi8(a, CONST128(rho8))
DoublePermutation.c:                                     ^
DoublePermutation.c: DoublePermutation.c:507:5: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'KeccakDoublePermutationOnWords' that is compiled without support for 'ssse3'
DoublePermutation.c: ./KeccakF-1600-unrolling.macros:90:9: note: expanded from macro 'rounds'
DoublePermutation.c:         thetaRhoPiChiIotaPrepareTheta(i  , A, E) \
DoublePermutation.c:         ^
DoublePermutation.c: DoublePermutation.c:161:11: note: expanded from macro 'thetaRhoPiChiIotaPrepareTheta'
DoublePermutation.c:     Bmu = ROL64in128_56(A##so); \
DoublePermutation.c:           ^
DoublePermutation.c: DoublePermutation.c:39:37: note: expanded from macro 'ROL64in128_56'
DoublePermutation.c:         #define ROL64in128_56(a)    _mm_shuffle_epi8(a, CONST128(rho56))
DoublePermutation.c:                                     ^
DoublePermutation.c: DoublePermutation.c:507:5: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'KeccakDoublePermutationOnWords' that is compiled without support for 'ssse3'
DoublePermutation.c: ./KeccakF-1600-unrolling.macros:91:9: note: expanded from macro 'rounds'
DoublePermutation.c:         thetaRhoPiChiIotaPrepareTheta(i+1, E, A) \
DoublePermutation.c: ...
Number of similar (implementation,compiler) pairs: 1, namely:
ImplementationCompiler
T:sseu4clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)

Compiler output

DoublePermutation.c: DoublePermutation.c:507:5: error: '__builtin_ia32_vprotqi' needs target feature xop
DoublePermutation.c:     rounds
DoublePermutation.c:     ^
DoublePermutation.c: ./KeccakF-1600-unrolling.macros:17:5: note: expanded from macro 'rounds'
DoublePermutation.c:     thetaRhoPiChiIotaPrepareTheta( 0, A, E) \
DoublePermutation.c:     ^
DoublePermutation.c: DoublePermutation.c:78:21: note: expanded from macro 'thetaRhoPiChiIotaPrepareTheta'
DoublePermutation.c:     Da = XOR128(Cu, ROL64in128(Ce, 1)); \
DoublePermutation.c:                     ^
DoublePermutation.c: DoublePermutation.c:33:37: note: expanded from macro 'ROL64in128'
DoublePermutation.c:         #define ROL64in128(a, o)    _mm_roti_epi64(a, o)
DoublePermutation.c:                                     ^
DoublePermutation.c: /usr/lib/llvm-11/lib/clang/11.0.1/include/xopintrin.h:237:12: note: expanded from macro '_mm_roti_epi64'
DoublePermutation.c:   (__m128i)__builtin_ia32_vprotqi((__v2di)(__m128i)(A), (N))
DoublePermutation.c:            ^
DoublePermutation.c: DoublePermutation.c:507:5: error: '__builtin_ia32_vprotqi' needs target feature xop
DoublePermutation.c: ./KeccakF-1600-unrolling.macros:17:5: note: expanded from macro 'rounds'
DoublePermutation.c:     thetaRhoPiChiIotaPrepareTheta( 0, A, E) \
DoublePermutation.c:     ^
DoublePermutation.c: DoublePermutation.c:79:21: note: expanded from macro 'thetaRhoPiChiIotaPrepareTheta'
DoublePermutation.c:     De = XOR128(Ca, ROL64in128(Ci, 1)); \
DoublePermutation.c:                     ^
DoublePermutation.c: DoublePermutation.c:33:37: note: expanded from macro 'ROL64in128'
DoublePermutation.c:         #define ROL64in128(a, o)    _mm_roti_epi64(a, o)
DoublePermutation.c:                                     ^
DoublePermutation.c: ...
Number of similar (implementation,compiler) pairs: 5, namely:
ImplementationCompiler
T:xopu24clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)
T:xopu24clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)
T:xopu24clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)
T:xopu24clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)
T:xopu24clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)

Compiler output

DoublePermutation.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/10/include/x86intrin.h:38,
DoublePermutation.c:                  from DoublePermutation.c:17:
DoublePermutation.c: DoublePermutation.c: In function 'KeccakDoublePermutationOnWords':
DoublePermutation.c: /usr/lib/gcc/x86_64-linux-gnu/10/include/xopintrin.h:272:1: error: inlining failed in call to 'always_inline' '_mm_roti_epi64': target specific option mismatch
DoublePermutation.c:   272 | _mm_roti_epi64(__m128i __A, const int __B)
DoublePermutation.c:       | ^~~~~~~~~~~~~~
DoublePermutation.c: DoublePermutation.c:33:37: note: called from here
DoublePermutation.c:    33 |         #define ROL64in128(a, o)    _mm_roti_epi64(a, o)
DoublePermutation.c:       |                                     ^~~~~~~~~~~~~~~~~~~~
DoublePermutation.c: DoublePermutation.c:276:11: note: in expansion of macro 'ROL64in128'
DoublePermutation.c:   276 |     Bsu = ROL64in128(A##se, 2); \
DoublePermutation.c:       |           ^~~~~~~~~~
DoublePermutation.c: KeccakF-1600-unrolling.macros:40:5: note: in expansion of macro 'thetaRhoPiChiIota'
DoublePermutation.c:    40 |     thetaRhoPiChiIota(23, E, A) \
DoublePermutation.c:       |     ^~~~~~~~~~~~~~~~~
DoublePermutation.c: DoublePermutation.c:507:5: note: in expansion of macro 'rounds'
DoublePermutation.c:   507 |     rounds
DoublePermutation.c:       |     ^~~~~~
DoublePermutation.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/10/include/x86intrin.h:38,
DoublePermutation.c:                  from DoublePermutation.c:17:
DoublePermutation.c: /usr/lib/gcc/x86_64-linux-gnu/10/include/xopintrin.h:272:1: error: inlining failed in call to 'always_inline' '_mm_roti_epi64': target specific option mismatch
DoublePermutation.c:   272 | _mm_roti_epi64(__m128i __A, const int __B)
DoublePermutation.c:       | ^~~~~~~~~~~~~~
DoublePermutation.c: DoublePermutation.c:33:37: note: called from here
DoublePermutation.c:    33 |         #define ROL64in128(a, o)    _mm_roti_epi64(a, o)
DoublePermutation.c: ...
Number of similar (implementation,compiler) pairs: 4, namely:
ImplementationCompiler
T:xopu24gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)
T:xopu24gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)
T:xopu24gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)
T:xopu24gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)