Test results for amd64, trident, crypto_sort/int64

[Page version: 20260830 23:14:38]

Measurements for amd64, trident, crypto_sort Test results for amd64, trident, crypto_sort Test results for crypto_sort/int64
Computer: trident
Microarchitecture: amd64; Core 2 65nm (6fb)
Architecture: amd64
CPU ID: GenuineIntel-000006fb-bfebfbff
SUPERCOP version: 20260627
Operation: crypto_sort
Primitive: int64
TimeObject sizeTest sizeImplementationCompilerBenchmark dateSUPERCOP version
8212811528 0 034216 1288 1488x863clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall2026063020260627
8212911689 0 034136 1288 1488x863clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall2026063020260627
821849992 0 031536 1288 1488x863clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall2026063020260627
8222810445 0 030246 1208 1552x863gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall2026063020260627
829218644 0 028006 1280 1488x863clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall2026063020260627
841721365 0 025661 1312 1584T:stdsortg++ -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall2026063020260627
848049498 0 029680 1288 1488x863clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall2026063020260627
849741297 0 024621 1312 1584T:stdsortg++ -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall2026063020260627
873971475 0 025460 1368 1488T:stdsortclang++ -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall2026063020260627
881631315 0 023281 1368 1488T:stdsortclang++ -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall2026063020260627
8822311364 0 032679 1216 1552x863gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall2026063020260627
883481558 0 024247 1376 1488T:stdsortclang++ -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall2026063020260627
895061712 0 026936 1368 1488T:stdsortclang++ -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall2026063020260627
90505925 0 021986 1288 1552T:stdsortg++ -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall2026063020260627
914841173 0 023642 1304 1584T:stdsortg++ -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall2026063020260627
9315310868 0 031351 1216 1552x863gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall2026063020260627
97716918 0 023368 1288 1488portable4clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall2026063020260627
97753917 0 023608 1288 1488portable4clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall2026063020260627
98110678 0 020046 1280 1488portable4clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall2026063020260627
99351837 0 022384 1288 1488portable4clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall2026063020260627
99490853 0 021056 1288 1488portable4clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall2026063020260627
100701865 0 021343 1216 1552portable4gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall2026063020260627
1036663909 0 022586 1192 1520x863gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall2026063020260627
103836865 0 022175 1216 1552portable4gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall2026063020260627
145141957 0 020750 1208 1552portable4gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall2026063020260627
174595664 0 019338 1192 1520portable4gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall2026063020260627

Compiler output

sort.c: sort.c:123:8: error: always_inline function '_mm256_loadu_si256' requires target feature 'avx', but would be inlined into function 'int64_sort_8through16' that is compiled without support for 'avx'
sort.c:   x0 = int64x4_load(x+0); // 0 4 8 12
sort.c:        ^
sort.c: sort.c:17:25: note: expanded from macro 'int64x4_load'
sort.c: #define int64x4_load(z) _mm256_loadu_si256((__m256i *) (z))
sort.c:                         ^
sort.c: sort.c:123:8: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
sort.c: sort.c:17:25: note: expanded from macro 'int64x4_load'
sort.c: #define int64x4_load(z) _mm256_loadu_si256((__m256i *) (z))
sort.c:                         ^
sort.c: sort.c:124:8: error: always_inline function '_mm256_loadu_si256' requires target feature 'avx', but would be inlined into function 'int64_sort_8through16' that is compiled without support for 'avx'
sort.c:   x1 = int64x4_load(x+4); // 1 5 9 13
sort.c:        ^
sort.c: sort.c:17:25: note: expanded from macro 'int64x4_load'
sort.c: #define int64x4_load(z) _mm256_loadu_si256((__m256i *) (z))
sort.c:                         ^
sort.c: sort.c:124:8: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
sort.c: sort.c:17:25: note: expanded from macro 'int64x4_load'
sort.c: #define int64x4_load(z) _mm256_loadu_si256((__m256i *) (z))
sort.c:                         ^
sort.c: sort.c:125:11: error: always_inline function '_mm256_set1_epi64x' requires target feature 'avx', but would be inlined into function 'int64_sort_8through16' that is compiled without support for 'avx'
sort.c:   infty = int64x4_broadcast(int64_largest);
sort.c:           ^
sort.c: sort.c:41:27: note: expanded from macro 'int64x4_broadcast'
sort.c: #define int64x4_broadcast _mm256_set1_epi64x
sort.c: ...
Number of similar (implementation,compiler) pairs: 5, namely:
ImplementationCompiler
2026avx2clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_14.0.6)
2026avx2clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_14.0.6)
2026avx2clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_14.0.6)
2026avx2clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_14.0.6)
2026avx2clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_14.0.6)

Compiler output

sort.c: sort.c: In function 'int64_sort_8through16':
sort.c: sort.c:123:6: warning: AVX vector return without AVX enabled changes the ABI [-Wpsabi]
sort.c:   123 |   x0 = int64x4_load(x+0); // 0 4 8 12
sort.c:       |      ^
sort.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/12/include/immintrin.h:43,
sort.c:                  from sort.c:4:
sort.c: /usr/lib/gcc/x86_64-linux-gnu/12/include/avxintrin.h:933:1: error: inlining failed in call to 'always_inline' '_mm256_storeu_si256': target specific option mismatch
sort.c:   933 | _mm256_storeu_si256 (__m256i_u *__P, __m256i __A)
sort.c:       | ^~~~~~~~~~~~~~~~~~~
sort.c: sort.c:18:28: note: called from here
sort.c:    18 | #define int64x4_store(z,i) _mm256_storeu_si256((__m256i *) (z),(i))
sort.c:       |                            ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
sort.c: sort.c:224:3: note: in expansion of macro 'int64x4_store'
sort.c:   224 |   int64x4_store(x+0,y2);
sort.c:       |   ^~~~~~~~~~~~~
sort.c: /usr/lib/gcc/x86_64-linux-gnu/12/include/avxintrin.h:933:1: error: inlining failed in call to 'always_inline' '_mm256_storeu_si256': target specific option mismatch
sort.c:   933 | _mm256_storeu_si256 (__m256i_u *__P, __m256i __A)
sort.c:       | ^~~~~~~~~~~~~~~~~~~
sort.c: sort.c:18:28: note: called from here
sort.c:    18 | #define int64x4_store(z,i) _mm256_storeu_si256((__m256i *) (z),(i))
sort.c:       |                            ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
sort.c: sort.c:223:3: note: in expansion of macro 'int64x4_store'
sort.c:   223 |   int64x4_store(x+4,y1);
sort.c:       |   ^~~~~~~~~~~~~
sort.c: /usr/lib/gcc/x86_64-linux-gnu/12/include/avxintrin.h:933:1: error: inlining failed in call to 'always_inline' '_mm256_storeu_si256': target specific option mismatch
sort.c: ...
Number of similar (implementation,compiler) pairs: 4, namely:
ImplementationCompiler
2026avx2gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (12.2.0)
2026avx2gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (12.2.0)
2026avx2gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (12.2.0)
2026avx2gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (12.2.0)

Compiler output

sort.c: sort.c:130:30: error: always_inline function '_mm_cmpgt_epi64' requires target feature 'sse4.2', but would be inlined into function 'int64_sort_8through16' that is compiled without support for 'sse4.2'
sort.c:   x4 = int8x16_iftopthenelse(int64x2_smaller_mask(int64x2_set(8,9),int64x2_broadcast(pos4)),int8x16_varextract(int64x2_load(x+pos4-2),int8x16_sub(int8x16_set(16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31),int8x16_broadcast(8*(pos4&1)))),infty); // 4 12
sort.c:                              ^
sort.c: sort.c:20:35: note: expanded from macro 'int64x2_smaller_mask'
sort.c: #define int64x2_smaller_mask(a,b) _mm_cmpgt_epi64(b,a)
sort.c:                                   ^
sort.c: sort.c:130:8: error: always_inline function '_mm_blendv_epi8' requires target feature 'sse4.1', but would be inlined into function 'int64_sort_8through16' that is compiled without support for 'sse4.1'
sort.c:   x4 = int8x16_iftopthenelse(int64x2_smaller_mask(int64x2_set(8,9),int64x2_broadcast(pos4)),int8x16_varextract(int64x2_load(x+pos4-2),int8x16_sub(int8x16_set(16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31),int8x16_broadcast(8*(pos4&1)))),infty); // 4 12
sort.c:        ^
sort.c: sort.c:23:38: note: expanded from macro 'int8x16_iftopthenelse'
sort.c: #define int8x16_iftopthenelse(c,t,e) _mm_blendv_epi8(e,t,c)
sort.c:                                      ^
sort.c: sort.c:132:30: error: always_inline function '_mm_cmpgt_epi64' requires target feature 'sse4.2', but would be inlined into function 'int64_sort_8through16' that is compiled without support for 'sse4.2'
sort.c:   x5 = int8x16_iftopthenelse(int64x2_smaller_mask(int64x2_set(10,11),int64x2_broadcast(pos5)),int8x16_varextract(int64x2_load(x+pos5-2),int8x16_sub(int8x16_set(16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31),int8x16_broadcast(8*(pos5&1)))),infty); // 5 13
sort.c:                              ^
sort.c: sort.c:20:35: note: expanded from macro 'int64x2_smaller_mask'
sort.c: #define int64x2_smaller_mask(a,b) _mm_cmpgt_epi64(b,a)
sort.c:                                   ^
sort.c: sort.c:132:8: error: always_inline function '_mm_blendv_epi8' requires target feature 'sse4.1', but would be inlined into function 'int64_sort_8through16' that is compiled without support for 'sse4.1'
sort.c:   x5 = int8x16_iftopthenelse(int64x2_smaller_mask(int64x2_set(10,11),int64x2_broadcast(pos5)),int8x16_varextract(int64x2_load(x+pos5-2),int8x16_sub(int8x16_set(16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31),int8x16_broadcast(8*(pos5&1)))),infty); // 5 13
sort.c:        ^
sort.c: sort.c:23:38: note: expanded from macro 'int8x16_iftopthenelse'
sort.c: #define int8x16_iftopthenelse(c,t,e) _mm_blendv_epi8(e,t,c)
sort.c:                                      ^
sort.c: sort.c:134:30: error: always_inline function '_mm_cmpgt_epi64' requires target feature 'sse4.2', but would be inlined into function 'int64_sort_8through16' that is compiled without support for 'sse4.2'
sort.c: ...
Number of similar (implementation,compiler) pairs: 4, namely:
ImplementationCompiler
sse42clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_14.0.6)
sse42clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_14.0.6)
sse42clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_14.0.6)
sse42clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_14.0.6)

Compiler output

sort.c: sort.c:130:93: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'int64_sort_8through16' that is compiled without support for 'ssse3'
sort.c:   x4 = int8x16_iftopthenelse(int64x2_smaller_mask(int64x2_set(8,9),int64x2_broadcast(pos4)),int8x16_varextract(int64x2_load(x+pos4-2),int8x16_sub(int8x16_set(16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31),int8x16_broadcast(8*(pos4&1)))),infty); // 4 12
sort.c:                                                                                             ^
sort.c: sort.c:39:28: note: expanded from macro 'int8x16_varextract'
sort.c: #define int8x16_varextract _mm_shuffle_epi8
sort.c:                            ^
sort.c: sort.c:130:30: error: always_inline function '_mm_cmpgt_epi64' requires target feature 'sse4.2', but would be inlined into function 'int64_sort_8through16' that is compiled without support for 'sse4.2'
sort.c:   x4 = int8x16_iftopthenelse(int64x2_smaller_mask(int64x2_set(8,9),int64x2_broadcast(pos4)),int8x16_varextract(int64x2_load(x+pos4-2),int8x16_sub(int8x16_set(16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31),int8x16_broadcast(8*(pos4&1)))),infty); // 4 12
sort.c:                              ^
sort.c: sort.c:20:35: note: expanded from macro 'int64x2_smaller_mask'
sort.c: #define int64x2_smaller_mask(a,b) _mm_cmpgt_epi64(b,a)
sort.c:                                   ^
sort.c: sort.c:130:8: error: always_inline function '_mm_blendv_epi8' requires target feature 'sse4.1', but would be inlined into function 'int64_sort_8through16' that is compiled without support for 'sse4.1'
sort.c:   x4 = int8x16_iftopthenelse(int64x2_smaller_mask(int64x2_set(8,9),int64x2_broadcast(pos4)),int8x16_varextract(int64x2_load(x+pos4-2),int8x16_sub(int8x16_set(16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31),int8x16_broadcast(8*(pos4&1)))),infty); // 4 12
sort.c:        ^
sort.c: sort.c:23:38: note: expanded from macro 'int8x16_iftopthenelse'
sort.c: #define int8x16_iftopthenelse(c,t,e) _mm_blendv_epi8(e,t,c)
sort.c:                                      ^
sort.c: sort.c:132:95: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'int64_sort_8through16' that is compiled without support for 'ssse3'
sort.c:   x5 = int8x16_iftopthenelse(int64x2_smaller_mask(int64x2_set(10,11),int64x2_broadcast(pos5)),int8x16_varextract(int64x2_load(x+pos5-2),int8x16_sub(int8x16_set(16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31),int8x16_broadcast(8*(pos5&1)))),infty); // 5 13
sort.c:                                                                                               ^
sort.c: sort.c:39:28: note: expanded from macro 'int8x16_varextract'
sort.c: #define int8x16_varextract _mm_shuffle_epi8
sort.c:                            ^
sort.c: sort.c:132:30: error: always_inline function '_mm_cmpgt_epi64' requires target feature 'sse4.2', but would be inlined into function 'int64_sort_8through16' that is compiled without support for 'sse4.2'
sort.c: ...
Number of similar (implementation,compiler) pairs: 1, namely:
ImplementationCompiler
sse42clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_14.0.6)

Compiler output

sort.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/12/include/immintrin.h:39,
sort.c:                  from sort.c:4:
sort.c: /usr/lib/gcc/x86_64-linux-gnu/12/include/smmintrin.h: In function 'int64_sort_8through16':
sort.c: /usr/lib/gcc/x86_64-linux-gnu/12/include/smmintrin.h:179:1: error: inlining failed in call to 'always_inline' '_mm_blendv_epi8': target specific option mismatch
sort.c:   179 | _mm_blendv_epi8 (__m128i __X, __m128i __Y, __m128i __M)
sort.c:       | ^~~~~~~~~~~~~~~
sort.c: sort.c:23:38: note: called from here
sort.c:    23 | #define int8x16_iftopthenelse(c,t,e) _mm_blendv_epi8(e,t,c)
sort.c:       |                                      ^~~~~~~~~~~~~~~~~~~~~~
sort.c: sort.c:265:8: note: in expansion of macro 'int8x16_iftopthenelse'
sort.c:   265 |   x7 = int8x16_iftopthenelse(t,x7,y2); // 13 15
sort.c:       |        ^~~~~~~~~~~~~~~~~~~~~
sort.c: /usr/lib/gcc/x86_64-linux-gnu/12/include/smmintrin.h:179:1: error: inlining failed in call to 'always_inline' '_mm_blendv_epi8': target specific option mismatch
sort.c:   179 | _mm_blendv_epi8 (__m128i __X, __m128i __Y, __m128i __M)
sort.c:       | ^~~~~~~~~~~~~~~
sort.c: sort.c:23:38: note: called from here
sort.c:    23 | #define int8x16_iftopthenelse(c,t,e) _mm_blendv_epi8(e,t,c)
sort.c:       |                                      ^~~~~~~~~~~~~~~~~~~~~~
sort.c: sort.c:264:8: note: in expansion of macro 'int8x16_iftopthenelse'
sort.c:   264 |   x2 = int8x16_iftopthenelse(t,y2,x7); // 12 14
sort.c:       |        ^~~~~~~~~~~~~~~~~~~~~
sort.c: /usr/lib/gcc/x86_64-linux-gnu/12/include/smmintrin.h:796:1: error: inlining failed in call to 'always_inline' '_mm_cmpgt_epi64': target specific option mismatch
sort.c:   796 | _mm_cmpgt_epi64 (__m128i __X, __m128i __Y)
sort.c:       | ^~~~~~~~~~~~~~~
sort.c: sort.c:20:35: note: called from here
sort.c: ...
Number of similar (implementation,compiler) pairs: 4, namely:
ImplementationCompiler
sse42gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (12.2.0)
sse42gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (12.2.0)
sse42gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (12.2.0)
sse42gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (12.2.0)

Passed TIMECOP

TIMECOP iterations: 10
Number of similar (implementation,compiler) pairs: 18, namely:
ImplementationCompiler
portable4clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_14.0.6)
portable4clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_14.0.6)
portable4clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_14.0.6)
portable4clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_14.0.6)
portable4clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_14.0.6)
portable4gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (12.2.0)
portable4gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (12.2.0)
portable4gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (12.2.0)
portable4gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (12.2.0)
x863clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_14.0.6)
x863clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_14.0.6)
x863clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_14.0.6)
x863clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_14.0.6)
x863clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_14.0.6)
x863gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (12.2.0)
x863gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (12.2.0)
x863gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (12.2.0)
x863gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (12.2.0)