Implementation notes: amd64, renoir, crypto_kem/hqc128
Computer: renoir
Microarchitecture: amd64; Zen 2 (860f01)
Architecture: amd64
CPU ID: AuthenticAMD-00860f01-178bfbff
SUPERCOP version: 20240625
Operation: crypto_kem
Primitive: hqc128
| Time | Object size | Test size | Implementation | Compiler | Benchmark date | SUPERCOP version |
|---|
| 887854 | 39290 212424 36052 | 93035 213316 37816 | T:avx | gcc_-march=native_-mtune=native_-O3_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall | 20240629 | 20240625 |
| 948333 | 24735 212424 36052 | 76835 213316 37816 | T:avx | gcc_-march=native_-mtune=native_-O2_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall | 20240629 | 20240625 |
| 1098428 | 23788 212424 36052 | 75195 213316 37816 | T:avx | gcc_-march=native_-mtune=native_-O_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall | 20240629 | 20240625 |
| 1177393 | 18940 212424 36052 | 69227 213308 37784 | T:avx | gcc_-march=native_-mtune=native_-Os_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall | 20240629 | 20240625 |
Test failure
error 111
Number of similar (implementation,compiler) pairs: 4, namely:
| Implementation | Compiler |
| T:avx | clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1) |
| T:avx | clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1) |
| T:avx | clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1) |
| T:avx | clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1) |
Compiler output
bch.c: bch.c:291:7: error: always_inline function '_mm256_set1_epi32' requires target feature 'avx', but would be inlined into function 'compute_syndromes' that is compiled without support for 'avx'
bch.c: y = _mm256_set1_epi32(*aux);
bch.c: ^
bch.c: bch.c:291:7: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
bch.c: bch.c:294:7: error: always_inline function '_mm256_shuffle_epi8' requires target feature 'avx2', but would be inlined into function 'compute_syndromes' that is compiled without support for 'avx2'
bch.c: y = _mm256_shuffle_epi8(y, mask_one);
bch.c: ^
bch.c: bch.c:294:7: error: AVX vector argument of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
bch.c: bch.c:296:15: error: always_inline function '_mm256_and_si256' requires target feature 'avx2', but would be inlined into function 'compute_syndromes' that is compiled without support for 'avx2'
bch.c: z[i >> 2] = _mm256_and_si256(y, mask_two);
bch.c: ^
bch.c: bch.c:296:15: error: AVX vector argument of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
bch.c: bch.c:306:17: error: always_inline function '_mm256_set1_epi64x' requires target feature 'avx', but would be inlined into function 'compute_syndromes' that is compiled without support for 'avx'
bch.c: tmp_repeat = _mm256_set1_epi64x((long long)(tmp_array[i]!=0));
bch.c: ^
bch.c: bch.c:306:17: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
bch.c: bch.c:307:8: error: always_inline function '_mm256_cmpeq_epi64' requires target feature 'avx2', but would be inlined into function 'compute_syndromes' that is compiled without support for 'avx2'
bch.c: L = _mm256_cmpeq_epi64(tmp_repeat,un_256);
bch.c: ^
bch.c: bch.c:307:8: error: AVX vector argument of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
bch.c: bch.c:308:17: error: always_inline function '_mm256_lddqu_si256' requires target feature 'avx', but would be inlined into function 'compute_syndromes' that is compiled without support for 'avx'
bch.c: tmp_repeat = _mm256_lddqu_si256((__m256i *)(alpha_tmp + i * (PARAM_DELTA << 1)));
bch.c: ^
bch.c: bch.c:308:17: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
bch.c: bch.c:309:8: error: always_inline function '_mm256_and_si256' requires target feature 'avx2', but would be inlined into function 'compute_syndromes' that is compiled without support for 'avx2'
bch.c: ...
Number of similar (implementation,compiler) pairs: 1, namely:
| Implementation | Compiler |
| T:avx | clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1) |
Compiler output
gf2x.c: gf2x.c: In function 'TOOM3Mult':
gf2x.c: gf2x.c:397:8: warning: array subscript 0 is outside array bounds of '__m256i[32]' [-Warray-bounds]
gf2x.c: 397 | U1_64 = ((uint64_t *) U1) - 1;
gf2x.c: | ~~~~~~^~~~~~~~~~~~~~~~~~~~~~~
gf2x.c: gf2x.c:346:53: note: while referencing 'U1'
gf2x.c: 346 | static __m256i U0[T_TM3_3W_256], V0[T_TM3_3W_256], U1[T_TM3_3W_256], V1[T_TM3_3W_256], U2[T_TM3_3W_256], V2[T_TM3_3W_256];
gf2x.c: | ^~
gf2x.c: gf2x.c:398:8: warning: array subscript 0 is outside array bounds of '__m256i[32]' [-Warray-bounds]
gf2x.c: 398 | U2_64 = ((uint64_t *) U2) - 2;
gf2x.c: | ~~~~~~^~~~~~~~~~~~~~~~~~~~~~~
gf2x.c: gf2x.c:346:89: note: while referencing 'U2'
gf2x.c: 346 | static __m256i U0[T_TM3_3W_256], V0[T_TM3_3W_256], U1[T_TM3_3W_256], V1[T_TM3_3W_256], U2[T_TM3_3W_256], V2[T_TM3_3W_256];
gf2x.c: | ^~
gf2x.c: gf2x.c:400:8: warning: array subscript 0 is outside array bounds of '__m256i[32]' [-Warray-bounds]
gf2x.c: 400 | V1_64 = ((uint64_t *) V1) - 1;
gf2x.c: | ~~~~~~^~~~~~~~~~~~~~~~~~~~~~~
gf2x.c: gf2x.c:346:71: note: while referencing 'V1'
gf2x.c: 346 | static __m256i U0[T_TM3_3W_256], V0[T_TM3_3W_256], U1[T_TM3_3W_256], V1[T_TM3_3W_256], U2[T_TM3_3W_256], V2[T_TM3_3W_256];
gf2x.c: | ^~
gf2x.c: gf2x.c:401:8: warning: array subscript 0 is outside array bounds of '__m256i[32]' [-Warray-bounds]
gf2x.c: 401 | V2_64 = ((uint64_t *) V2) - 2;
gf2x.c: | ~~~~~~^~~~~~~~~~~~~~~~~~~~~~~
gf2x.c: gf2x.c:346:107: note: while referencing 'V2'
gf2x.c: 346 | static __m256i U0[T_TM3_3W_256], V0[T_TM3_3W_256], U1[T_TM3_3W_256], V1[T_TM3_3W_256], U2[T_TM3_3W_256], V2[T_TM3_3W_256];
gf2x.c: | ^~
rng.c: rng.c: In function 'AES256_ECB':
rng.c: rng.c:125:9: warning: variable 'ciphertext_len' set but not used [-Wunused-but-set-variable]
rng.c: 125 | int ciphertext_len;
rng.c: | ^~~~~~~~~~~~~~
Number of similar (implementation,compiler) pairs: 3, namely:
| Implementation | Compiler |
| T:avx | gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110) |
| T:avx | gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110) |
| T:avx | gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110) |
Compiler output
rng.c: rng.c: In function 'AES256_ECB':
rng.c: rng.c:125:9: warning: variable 'ciphertext_len' set but not used [-Wunused-but-set-variable]
rng.c: 125 | int ciphertext_len;
rng.c: | ^~~~~~~~~~~~~~
Number of similar (implementation,compiler) pairs: 1, namely:
| Implementation | Compiler |
| T:avx | gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110) |