Implementation notes: amd64, speed2supercop, crypto_stream/speck128128ctr
Computer: speed2supercop
Microarchitecture: amd64; Haswell+AES (306c3)
Architecture: amd64
CPU ID: GenuineIntel-000306c3-1fc9cbf5
SUPERCOP version: 20240625
Operation: crypto_stream
Primitive: speck128128ctr
| Time | Object size | Test size | Implementation | Compiler | Benchmark date | SUPERCOP version |
|---|
| 2092 | 33840 0 0 | 44976 728 832 | T:avx2 | gcc_-march=native_-mtune=native_-Os_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall | 20240703 | 20240625 |
| 2128 | 33489 0 0 | 45446 792 792 | T:avx2 | clang_-march=native_-O_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall | 20240703 | 20240625 |
| 2136 | 33840 0 0 | 49630 792 808 | T:avx2 | clang_-march=native_-O2_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall | 20240703 | 20240625 |
| 2136 | 34032 0 0 | 49998 792 808 | T:avx2 | clang_-march=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall | 20240703 | 20240625 |
| 2136 | 33481 0 0 | 46263 784 856 | T:avx2 | clang_-march=native_-Os_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall | 20240703 | 20240625 |
| 2136 | 34700 0 0 | 46988 744 864 | T:avx2 | gcc_-march=native_-mtune=native_-O_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall | 20240703 | 20240625 |
| 2164 | 34864 0 0 | 47477 752 864 | T:avx2 | gcc_-march=native_-mtune=native_-O2_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall | 20240703 | 20240625 |
| 2168 | 36155 0 0 | 50813 752 864 | T:avx2 | gcc_-march=native_-mtune=native_-O3_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall | 20240703 | 20240625 |
| 4104 | 28069 0 0 | 40372 744 864 | T:sse4 | gcc_-march=native_-mtune=native_-O_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall | 20240703 | 20240625 |
| 4212 | 26887 0 0 | 42846 792 808 | T:sse4 | clang_-march=native_-O2_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall | 20240703 | 20240625 |
| 4212 | 26887 0 0 | 43038 792 808 | T:sse4 | clang_-march=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall | 20240703 | 20240625 |
| 4224 | 26492 0 0 | 39463 784 856 | T:sse4 | clang_-march=native_-Os_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall | 20240703 | 20240625 |
| 4228 | 26761 0 0 | 38830 792 792 | T:sse4 | clang_-march=native_-O_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall | 20240703 | 20240625 |
| 6420 | 16012 0 0 | 30693 752 864 | T:sse4 | gcc_-march=native_-mtune=native_-O3_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall | 20240703 | 20240625 |
| 6424 | 14864 0 0 | 27493 752 864 | T:sse4 | gcc_-march=native_-mtune=native_-O2_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall | 20240703 | 20240625 |
| 6728 | 13903 0 0 | 25040 728 832 | T:sse4 | gcc_-march=native_-mtune=native_-Os_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall | 20240703 | 20240625 |
Compiler output
stream.c: stream.c:288:3: error: always_inline function '_mm256_set_epi64x' requires target feature 'avx', but would be inlined into function 'ExpandKey' that is compiled without support for 'avx'
stream.c: EK(A,B,rk,key);
stream.c: ^
stream.c: ./Speck128128AVX2.h:54:24: note: expanded from macro 'EK'
stream.c: #define EK(A,B,k,key) (RK(B,A,k,key,0), RK(B,A,k,key,1), RK(B,A,k,key,2), RK(B,A,k,key,3), RK(B,A,k,key,4), RK(B,A,k,key,5), RK(B,A,k,key,6), \
stream.c: ^
stream.c: ./Speck128128AVX2.h:52:28: note: expanded from macro 'RK'
stream.c: #define RK(X,Y,k,key,i) (SET1(k[i],Y), key[i]=Y, X=RCS(X,8), X+=Y, X^=i, Y=LCS(Y,3), Y^=X)
stream.c: ^
stream.c: ./Intrinsics_AVX2_128block.h:25:22: note: expanded from macro 'SET1'
stream.c: #define SET1(X,c) (X=SET(c,c,c,c))
stream.c: ^
stream.c: ./Intrinsics_AVX2_128block.h:24:13: note: expanded from macro 'SET'
stream.c: #define SET _mm256_set_epi64x
stream.c: ^
stream.c: stream.c:288:3: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
stream.c: ./Speck128128AVX2.h:54:24: note: expanded from macro 'EK'
stream.c: #define EK(A,B,k,key) (RK(B,A,k,key,0), RK(B,A,k,key,1), RK(B,A,k,key,2), RK(B,A,k,key,3), RK(B,A,k,key,4), RK(B,A,k,key,5), RK(B,A,k,key,6), \
stream.c: ^
stream.c: ./Speck128128AVX2.h:52:28: note: expanded from macro 'RK'
stream.c: #define RK(X,Y,k,key,i) (SET1(k[i],Y), key[i]=Y, X=RCS(X,8), X+=Y, X^=i, Y=LCS(Y,3), Y^=X)
stream.c: ^
stream.c: ./Intrinsics_AVX2_128block.h:25:22: note: expanded from macro 'SET1'
stream.c: #define SET1(X,c) (X=SET(c,c,c,c))
stream.c: ^
stream.c: ...
Number of similar (implementation,compiler) pairs: 1, namely:
| Implementation | Compiler |
| T:avx2 | clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_16.0.6_(27+b1)) |
Compiler output
stream.c: stream.c:260:8: warning: unused variable 'i' [-Wunused-variable]
stream.c: u64 i,j,x[2],y[2];
stream.c: ^
stream.c: stream.c:260:10: warning: unused variable 'j' [-Wunused-variable]
stream.c: u64 i,j,x[2],y[2];
stream.c: ^
stream.c: stream.c:329:3: error: always_inline function '_mm512_set_epi64' requires target feature 'avx512f', but would be inlined into function 'ExpandKey' that is compiled without support for 'avx512f'
stream.c: EK(A,B,rk,key);
stream.c: ^
stream.c: ./Speck128128AVX512.h:59:24: note: expanded from macro 'EK'
stream.c: #define EK(A,B,k,key) (RK(B,A,k,key,0), RK(B,A,k,key,1), RK(B,A,k,key,2), RK(B,A,k,key,3), RK(B,A,k,key,4), RK(B,A,k,key,5), RK(B,A,k,key,6), \
stream.c: ^
stream.c: ./Speck128128AVX512.h:57:26: note: expanded from macro 'RK'
stream.c: #define RK(X,Y,k,key,i) (SET1(k[i],Y), key[i]=Y, X=RCS(X,8), X+=Y, X^=i, Y=LCS(Y,3), Y^=X)
stream.c: ^
stream.c: ./Intrinsics_AVX512_128block.h:25:22: note: expanded from macro 'SET1'
stream.c: #define SET1(X,c) (X=SET(c,c,c,c,c,c,c,c))
stream.c: ^
stream.c: ./Intrinsics_AVX512_128block.h:16:13: note: expanded from macro 'SET'
stream.c: #define SET _mm512_set_epi64
stream.c: ^
stream.c: stream.c:329:3: error: AVX vector return of type '__m512i' (vector of 8 'long long' values) without 'avx512f' enabled changes the ABI
stream.c: ./Speck128128AVX512.h:59:24: note: expanded from macro 'EK'
stream.c: #define EK(A,B,k,key) (RK(B,A,k,key,0), RK(B,A,k,key,1), RK(B,A,k,key,2), RK(B,A,k,key,3), RK(B,A,k,key,4), RK(B,A,k,key,5), RK(B,A,k,key,6), \
stream.c: ^
stream.c: ...
Number of similar (implementation,compiler) pairs: 5, namely:
| Implementation | Compiler |
| T:avx512 | clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_16.0.6_(27+b1)) |
| T:avx512 | clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_16.0.6_(27+b1)) |
| T:avx512 | clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_16.0.6_(27+b1)) |
| T:avx512 | clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_16.0.6_(27+b1)) |
| T:avx512 | clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_16.0.6_(27+b1)) |
Compiler output
stream.c: stream.c: In function 'Encrypt_Xor':
stream.c: stream.c:260:10: warning: unused variable 'j' [-Wunused-variable]
stream.c: 260 | u64 i,j,x[2],y[2];
stream.c: | ^
stream.c: stream.c:260:8: warning: unused variable 'i' [-Wunused-variable]
stream.c: 260 | u64 i,j,x[2],y[2];
stream.c: | ^
stream.c: In file included from Speck128128AVX512.h:17,
stream.c: from stream.c:21:
stream.c: stream.c: In function 'Encrypt':
stream.c: Intrinsics_AVX512_128block.h:25:21: warning: AVX512F vector return without AVX512F enabled changes the ABI [-Wpsabi]
stream.c: 25 | #define SET1(X,c) (X=SET(c,c,c,c,c,c,c,c))
stream.c: | ~~^~~~~~~~~~~~~~~~~~~~~~
stream.c: stream.c:135:5: note: in expansion of macro 'SET1'
stream.c: 135 | SET1(X[0],nonce[1]); SET8(Y[0],nonce[0]);
stream.c: | ^~~~
stream.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/13/include/immintrin.h:53,
stream.c: from Intrinsics_AVX512_128block.h:5:
stream.c: /usr/lib/gcc/x86_64-linux-gnu/13/include/avx512fintrin.h:6532:1: error: inlining failed in call to 'always_inline' '_mm512_storeu_si512': target specific option mismatch
stream.c: 6532 | _mm512_storeu_si512 (void *__P, __m512i __A)
stream.c: | ^~~~~~~~~~~~~~~~~~~
stream.c: Intrinsics_AVX512_128block.h:32:18: note: called from here
stream.c: 32 | #define ST(ip,X) _mm512_storeu_si512((void *)(ip),X)
stream.c: | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
stream.c: Intrinsics_AVX512_128block.h:33:43: note: in expansion of macro 'ST'
stream.c: ...
Number of similar (implementation,compiler) pairs: 4, namely:
| Implementation | Compiler |
| T:avx512 | gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (13.3.0) |
| T:avx512 | gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (13.3.0) |
| T:avx512 | gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (13.3.0) |
| T:avx512 | gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (13.3.0) |
Compiler output
stream.c: stream.c:114:21: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'Encrypt' that is compiled without support for 'ssse3'
stream.c: if (numbytes==32) Enc(X,Y,rk,2);
stream.c: ^
stream.c: ./Speck128128SSE4.h:42:23: note: expanded from macro 'Enc'
stream.c: #define Enc(X,Y,k,n) (Rx##n(X,Y,k[0]), Rx##n(X,Y,k[1]), Rx##n(X,Y,k[2]), Rx##n(X,Y,k[3]), Rx##n(X,Y,k[4]), Rx##n(X,Y,k[5]), Rx##n(X,Y,k[6]), Rx##n(X,Y,k[7]), \
stream.c: ^
stream.c: <scratch space>:110:1: note: expanded from here
stream.c: Rx2
stream.c: ^
stream.c: ./Speck128128SSE4.h:25:21: note: expanded from macro 'Rx2'
stream.c: #define Rx2(X,Y,k) (R(X[0],Y[0],k))
stream.c: ^
stream.c: ./Speck128128SSE4.h:23:29: note: expanded from macro 'R'
stream.c: #define R(X,Y,k) (X=XOR(ADD(ROR8(X),Y),k), Y=XOR(ROL(Y,3),X))
stream.c: ^
stream.c: ./Intrinsics_SSE4_128block.h:41:19: note: expanded from macro 'ROR8'
stream.c: #define ROR8(X) (SHFL(X,R8))
stream.c: ^
stream.c: ./Intrinsics_SSE4_128block.h:37:14: note: expanded from macro 'SHFL'
stream.c: #define SHFL _mm_shuffle_epi8
stream.c: ^
stream.c: stream.c:114:21: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'Encrypt' that is compiled without support for 'ssse3'
stream.c: ./Speck128128SSE4.h:42:41: note: expanded from macro 'Enc'
stream.c: #define Enc(X,Y,k,n) (Rx##n(X,Y,k[0]), Rx##n(X,Y,k[1]), Rx##n(X,Y,k[2]), Rx##n(X,Y,k[3]), Rx##n(X,Y,k[4]), Rx##n(X,Y,k[5]), Rx##n(X,Y,k[6]), Rx##n(X,Y,k[7]), \
stream.c: ^
stream.c: ...
Number of similar (implementation,compiler) pairs: 1, namely:
| Implementation | Compiler |
| T:sse4 | clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_16.0.6_(27+b1)) |
Compiler output
stream.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/13/include/pmmintrin.h:31,
stream.c: from /usr/lib/gcc/x86_64-linux-gnu/13/include/tmmintrin.h:31,
stream.c: from /usr/lib/gcc/x86_64-linux-gnu/13/include/smmintrin.h:32,
stream.c: from Intrinsics_SSE4_128block.h:5,
stream.c: from Speck128128SSE4.h:17,
stream.c: from stream.c:20:
stream.c: In function '_mm_unpackhi_epi64',
stream.c: inlined from 'Encrypt_Xor' at stream.c:245:22:
stream.c: /usr/lib/gcc/x86_64-linux-gnu/13/include/emmintrin.h:1030:19: warning: 'X[2]' may be used uninitialized [-Wmaybe-uninitialized]
stream.c: 1030 | return (__m128i)__builtin_ia32_punpckhqdq128 ((__v2di)__A, (__v2di)__B);
stream.c: | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
stream.c: stream.c: In function 'Encrypt_Xor':
stream.c: stream.c:215:8: note: 'X[2]' was declared here
stream.c: 215 | u128 X[4],Y[4],Z[4];
stream.c: | ^
stream.c: In function '_mm_unpackhi_epi64',
stream.c: inlined from 'Encrypt_Xor' at stream.c:245:22:
stream.c: /usr/lib/gcc/x86_64-linux-gnu/13/include/emmintrin.h:1030:19: warning: 'Y[2]' may be used uninitialized [-Wmaybe-uninitialized]
stream.c: 1030 | return (__m128i)__builtin_ia32_punpckhqdq128 ((__v2di)__A, (__v2di)__B);
stream.c: | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
stream.c: stream.c: In function 'Encrypt_Xor':
stream.c: stream.c:215:13: note: 'Y[2]' was declared here
stream.c: 215 | u128 X[4],Y[4],Z[4];
stream.c: | ^
Number of similar (implementation,compiler) pairs: 2, namely:
| Implementation | Compiler |
| T:sse4 | gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (13.3.0) |
| T:sse4 | gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (13.3.0) |
Compiler output
stream.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/13/include/pmmintrin.h:31,
stream.c: from /usr/lib/gcc/x86_64-linux-gnu/13/include/tmmintrin.h:31,
stream.c: from /usr/lib/gcc/x86_64-linux-gnu/13/include/smmintrin.h:32,
stream.c: from Intrinsics_SSE4_128block.h:5,
stream.c: from Speck128128SSE4.h:17,
stream.c: from stream.c:20:
stream.c: In function '_mm_unpackhi_epi64',
stream.c: inlined from 'Encrypt' at stream.c:132:22:
stream.c: /usr/lib/gcc/x86_64-linux-gnu/13/include/emmintrin.h:1030:19: warning: 'X[2]' may be used uninitialized [-Wmaybe-uninitialized]
stream.c: 1030 | return (__m128i)__builtin_ia32_punpckhqdq128 ((__v2di)__A, (__v2di)__B);
stream.c: | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
stream.c: stream.c: In function 'Encrypt':
stream.c: stream.c:102:8: note: 'X[2]' was declared here
stream.c: 102 | u128 X[4],Y[4],Z[4];
stream.c: | ^
stream.c: In function '_mm_unpackhi_epi64',
stream.c: inlined from 'Encrypt' at stream.c:132:22:
stream.c: /usr/lib/gcc/x86_64-linux-gnu/13/include/emmintrin.h:1030:19: warning: 'Y[2]' may be used uninitialized [-Wmaybe-uninitialized]
stream.c: 1030 | return (__m128i)__builtin_ia32_punpckhqdq128 ((__v2di)__A, (__v2di)__B);
stream.c: | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
stream.c: stream.c: In function 'Encrypt':
stream.c: stream.c:102:13: note: 'Y[2]' was declared here
stream.c: 102 | u128 X[4],Y[4],Z[4];
stream.c: | ^
stream.c: In function '_mm_unpackhi_epi64',
stream.c: ...
Number of similar (implementation,compiler) pairs: 2, namely:
| Implementation | Compiler |
| T:sse4 | gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (13.3.0) |
| T:sse4 | gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (13.3.0) |