Test results for amd64, gemini, crypto_xof/bitwrite16

[Page version: 20241028 05:30:40]

Measurements for amd64, gemini, crypto_xof Test results for amd64, gemini, crypto_xof Test results for crypto_xof/bitwrite16

Computer: gemini
Microarchitecture: amd64; Goldmont Plus (706a8)
Architecture: amd64
CPU ID: GenuineIntel-000706a8-bfebfbff
SUPERCOP version: 20241022
Operation: crypto_xof
Primitive: bitwrite16

Time	Object size	Test size	Implementation	Compiler	Benchmark date	SUPERCOP version
388068	802 0 0	11352 780 952	`128`	`gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
396385	645 0 0	10783 772 952	`128`	`gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
419047	1168 0 0	12888 780 952	`128`	`gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
462182	481 0 0	9635 756 920	`128`	`gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
497321	1323 0 0	13712 812 920	`128`	`clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
537779	603 0 0	10262 804 920	`128`	`clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
538544	960 0 0	14720 812 920	`128`	`clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
582035	741 0 0	12208 812 920	`128`	`clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
934159	979 0 0	11182 804 920	`128`	`clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
1547881	1297 0 0	15080 812 920	`64x4`	`clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
1547912	1297 0 0	12776 812 920	`64x4`	`clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
1596859	1279 0 0	13664 812 920	`64x4`	`clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
1626901	2007 0 0	13720 780 952	`64x4`	`gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
1646938	1442 0 0	11984 780 952	`64x4`	`gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
1652402	1572 0 0	13048 812 920	`64x4x2`	`clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
1693202	1205 0 0	10862 804 920	`64x4`	`clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
1726222	1590 0 0	13976 812 920	`64x4x2`	`clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
1755817	1454 0 0	11118 804 920	`64x4x2`	`clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
1814021	1572 0 0	15352 812 920	`64x4x2`	`clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
1887598	1770 0 0	12312 780 952	`64x4x2`	`gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
1981403	2596 0 0	14312 780 952	`64x4x2`	`gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
2075135	1021 0 0	12744 780 952	`64`	`gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
2082388	484 0 0	9643 756 920	`64`	`gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
2086295	786 0 0	9955 756 920	`64x4`	`gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
2136598	802 0 0	11344 780 952	`64`	`gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
2379449	759 0 0	12224 812 920	`64`	`clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
2423046	1060 0 0	10227 756 920	`64x4x2`	`gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
2423892	1331 0 0	11542 804 920	`64x4`	`clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
2753147	1656 0 0	11799 772 952	`64x4`	`gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
2874238	1683 0 0	11894 804 920	`64x4x2`	`clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
2971241	748 0 0	10887 772 952	`64`	`gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
3062321	1984 0 0	12127 772 952	`64x4x2`	`gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
3116465	682 0 0	10342 804 920	`64`	`clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
3225503	833 0 0	14616 812 920	`64`	`clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
3448272	847 0 0	13232 812 920	`64`	`clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
3687907	812 0 0	11006 804 920	`64`	`clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
4736200	222 0 0	10760 780 952	`ref`	`gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
4737048	233 0 0	14008 812 920	`ref`	`clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
4737074	236 0 0	12624 812 920	`ref`	`clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
4743157	220 0 0	10359 772 952	`ref`	`gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
5109819	225 0 0	11944 780 952	`ref`	`gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
5114078	256 0 0	11720 812 920	`ref`	`clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
5325948	235 0 0	9886 804 920	`ref`	`clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
5332133	197 0 0	9339 756 920	`ref`	`gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011
11821660	383 0 0	10542 804 920	`ref`	`clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall`	20241016	20241011

Compiler output

write.c: write.c:36:19: error: always_inline function '_mm256_set1_epi64x' requires target feature 'avx', but would be inlined into function 'atmost8192' that is compiled without support for 'avx'
write.c:   __m256i hlast = _mm256_set1_epi64x(0);
write.c:                   ^
write.c: write.c:36:19: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
write.c: write.c:42:23: warning: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI [-Wpsabi]
write.c:     __m256i word256 = vec256_1shlmod(pos);
write.c:                       ^
write.c: write.c:44:20: error: always_inline function '_mm256_loadu_si256' requires target feature 'avx', but would be inlined into function 'atmost8192' that is compiled without support for 'avx'
write.c:       __m256i hi = _mm256_loadu_si256((__m256i *) (h+32*i));
write.c:                    ^
write.c: write.c:44:20: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
write.c: write.c:45:23: warning: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI [-Wpsabi]
write.c:       hi |= word256 & vec256_equal_mask(i,pos>>8);
write.c:                       ^
write.c: write.c:46:7: error: always_inline function '_mm256_storeu_si256' requires target feature 'avx', but would be inlined into function 'atmost8192' that is compiled without support for 'avx'
write.c:       _mm256_storeu_si256((__m256i *) (h+32*i),hi);
write.c:       ^
write.c: write.c:46:7: error: AVX vector argument of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
write.c: write.c:48:24: warning: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI [-Wpsabi]
write.c:     hlast |= word256 & vec256_equal_mask(words,pos>>8);
write.c:                        ^
write.c: write.c:25:21: error: always_inline function '_mm256_set_m128i' requires target feature 'avx', but would be inlined into function 'vec256_1shlmod' that is compiled without support for 'avx'
write.c:   __m256i word256 = _mm256_set_m128i(word128&mask7,word128&~mask7);
write.c:                     ^
write.c: write.c:25:21: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
write.c: ...

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

Implementation	Compiler
`256`	`clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`256`	`clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`256`	`clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`256`	`clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`256`	`clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`

Compiler output

write.c: write.c: In function 'vec256_equal_mask':
write.c: write.c:12:1: warning: AVX vector return without AVX enabled changes the ABI [-Wpsabi]
write.c:    12 | {
write.c:       | ^
write.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/10/include/immintrin.h:51,
write.c:                  from write.c:5:
write.c: /usr/lib/gcc/x86_64-linux-gnu/10/include/avxintrin.h:1342:1: error: inlining failed in call to 'always_inline' '_mm256_set1_epi64x': target specific option mismatch
write.c:  1342 | _mm256_set1_epi64x (long long __A)
write.c:       | ^~~~~~~~~~~~~~~~~~
write.c: write.c:16:10: note: called from here
write.c:    16 |   return _mm256_set1_epi64x(mask);
write.c:       |          ^~~~~~~~~~~~~~~~~~~~~~~~

Number of similar (implementation,compiler) pairs: 12, namely:

Implementation	Compiler
`256`	`gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`256`	`gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`256`	`gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`256`	`gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`256x4`	`gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`256x4`	`gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`256x4`	`gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`256x4`	`gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`256x4x2`	`gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`256x4x2`	`gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`256x4x2`	`gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`256x4x2`	`gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`

Compiler output

write.c: write.c:36:19: error: always_inline function '_mm256_set1_epi64x' requires target feature 'avx', but would be inlined into function 'atmost8192' that is compiled without support for 'avx'
write.c:   __m256i hlast = _mm256_set1_epi64x(0);
write.c:                   ^
write.c: write.c:36:19: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
write.c: write.c:42:23: warning: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI [-Wpsabi]
write.c:     __m256i word256 = vec256_1shlmod(pos);
write.c:                       ^
write.c: write.c:44:21: error: always_inline function '_mm256_loadu_si256' requires target feature 'avx', but would be inlined into function 'atmost8192' that is compiled without support for 'avx'
write.c:       __m256i hi0 = _mm256_loadu_si256((__m256i *) (h+32*i));
write.c:                     ^
write.c: write.c:44:21: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
write.c: write.c:45:21: error: always_inline function '_mm256_loadu_si256' requires target feature 'avx', but would be inlined into function 'atmost8192' that is compiled without support for 'avx'
write.c:       __m256i hi1 = _mm256_loadu_si256((__m256i *) (h+32*i+32));
write.c:                     ^
write.c: write.c:45:21: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
write.c: write.c:46:21: error: always_inline function '_mm256_loadu_si256' requires target feature 'avx', but would be inlined into function 'atmost8192' that is compiled without support for 'avx'
write.c:       __m256i hi2 = _mm256_loadu_si256((__m256i *) (h+32*i+64));
write.c:                     ^
write.c: write.c:46:21: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
write.c: write.c:47:21: error: always_inline function '_mm256_loadu_si256' requires target feature 'avx', but would be inlined into function 'atmost8192' that is compiled without support for 'avx'
write.c:       __m256i hi3 = _mm256_loadu_si256((__m256i *) (h+32*i+96));
write.c:                     ^
write.c: write.c:47:21: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
write.c: write.c:48:32: warning: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI [-Wpsabi]
write.c:       __m256i base = word256 & vec256_equal_mask(i>>2,pos>>10);
write.c: ...

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

Implementation	Compiler
`256x4`	`clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`256x4`	`clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`256x4`	`clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`256x4`	`clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`256x4`	`clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`

Compiler output

write.c: write.c:36:19: error: always_inline function '_mm256_set1_epi64x' requires target feature 'avx', but would be inlined into function 'atmost8192' that is compiled without support for 'avx'
write.c:   __m256i hlast = _mm256_set1_epi64x(0);
write.c:                   ^
write.c: write.c:36:19: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
write.c: write.c:44:17: warning: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI [-Wpsabi]
write.c:     word256_0 = vec256_1shlmod(pos_0);
write.c:                 ^
write.c: write.c:47:19: warning: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI [-Wpsabi]
write.c:       word256_1 = vec256_1shlmod(pos_1);
write.c:                   ^
write.c: write.c:50:19: error: always_inline function '_mm256_set1_epi64x' requires target feature 'avx', but would be inlined into function 'atmost8192' that is compiled without support for 'avx'
write.c:       word256_1 = _mm256_set1_epi64x(0);
write.c:                   ^
write.c: write.c:50:19: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
write.c: write.c:53:21: error: always_inline function '_mm256_loadu_si256' requires target feature 'avx', but would be inlined into function 'atmost8192' that is compiled without support for 'avx'
write.c:       __m256i hi0 = _mm256_loadu_si256((__m256i *) (h+32*i));
write.c:                     ^
write.c: write.c:53:21: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
write.c: write.c:54:21: error: always_inline function '_mm256_loadu_si256' requires target feature 'avx', but would be inlined into function 'atmost8192' that is compiled without support for 'avx'
write.c:       __m256i hi1 = _mm256_loadu_si256((__m256i *) (h+32*i+32));
write.c:                     ^
write.c: write.c:54:21: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
write.c: write.c:55:21: error: always_inline function '_mm256_loadu_si256' requires target feature 'avx', but would be inlined into function 'atmost8192' that is compiled without support for 'avx'
write.c:       __m256i hi2 = _mm256_loadu_si256((__m256i *) (h+32*i+64));
write.c:                     ^
write.c: ...

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

Implementation	Compiler
`256x4x2`	`clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`256x4x2`	`clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`256x4x2`	`clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`256x4x2`	`clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`256x4x2`	`clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`

Compiler output

write.c: write.c: In function 'crypto_xof_bitwrite16_64x4_constbranchindex':
write.c: write.c:51:46: warning: iteration 2305843009213693952 invokes undefined behavior [-Waggressive-loop-optimizations]
write.c:    51 |       crypto_uint64 hi = crypto_uint64_load(h+8*i);
write.c:       |                                             ~^~~~
write.c: write.c:50:5: note: within this loop
write.c:    50 |     for (;i < words;++i) {
write.c:       |     ^~~

Number of similar (implementation,compiler) pairs: 1, namely:

Implementation	Compiler
`64x4`	`gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`

Compiler output

write.c: write.c: In function 'crypto_xof_bitwrite16_64x4x2_constbranchindex':
write.c: write.c:69:46: warning: iteration 2305843009213693952 invokes undefined behavior [-Waggressive-loop-optimizations]
write.c:    69 |       crypto_uint64 hi = crypto_uint64_load(h+8*i);
write.c:       |                                             ~^~~~
write.c: write.c:68:5: note: within this loop
write.c:    68 |     for (;i < words;++i) {
write.c:       |     ^~~

Number of similar (implementation,compiler) pairs: 1, namely:

Implementation	Compiler
`64x4x2`	`gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`

Passed TIMECOP

TIMECOP iterations: 10

Number of similar (implementation,compiler) pairs: 45, namely:

Implementation	Compiler
`128`	`clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`128`	`clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`128`	`clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`128`	`clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`128`	`clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`128`	`gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`128`	`gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`128`	`gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`128`	`gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`64`	`clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`64`	`clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`64`	`clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`64`	`clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`64`	`clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`64`	`gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`64`	`gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`64`	`gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`64`	`gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`64x4`	`clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`64x4`	`clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`64x4`	`clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`64x4`	`clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`64x4`	`clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`64x4`	`gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`64x4`	`gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`64x4`	`gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`64x4`	`gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`64x4x2`	`clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`64x4x2`	`clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`64x4x2`	`clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`64x4x2`	`clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`64x4x2`	`clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`64x4x2`	`gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`64x4x2`	`gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`64x4x2`	`gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`64x4x2`	`gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`ref`	`clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`ref`	`clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`ref`	`clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`ref`	`clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`ref`	`clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`ref`	`gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`ref`	`gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`ref`	`gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`ref`	`gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`