Implementation notes: amd64, gcc14, crypto_kem/firesaber2

Computer: gcc14
Architecture: amd64
CPU ID: GenuineIntel-00010676-bfebfbff
SUPERCOP version: 20220506
Operation: crypto_kem
Primitive: firesaber2
TimeObject sizeTest sizeImplementationCompilerBenchmark dateSUPERCOP version
162444029846 0 047558 836 1760T:refclang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE2021011820210114
167152129846 0 047558 836 1760T:refclang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE2021011820210114
184101529958 0 048590 836 1760T:refclang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE2021011820210114
190043831567 0 050222 836 1760T:refclang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE2021011820210114
255757564782 0 084597 860 1792T:refgcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE2021011820210114
276871312638 0 029264 828 1760T:refclang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE2021011820210114
281877312740 0 030437 860 1792T:refgcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE2021011820210114
336433413777 0 031901 860 1792T:refgcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE2021011820210114
460015412096 0 028741 852 1760T:refgcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE2021011820210114

Compiler output

Implementation: T:avx2
Security model: timingleaks
Compiler: clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE
SABER_indcpa.c: In file included from SABER_indcpa.c:9:
SABER_indcpa.c: In file included from ././polymul/toom-cook_4way.c:6:
SABER_indcpa.c: ././polymul/scm_avx.c:43:9: error: always_inline function '_mm256_mullo_epi16' requires target feature 'xsave', but would be inlined into function 'schoolbook_avx_new3_acc' that is compiled without support for 'xsave'
SABER_indcpa.c: temp = _mm256_mullo_epi16 (a0, b1);
SABER_indcpa.c: ^
SABER_indcpa.c: ././polymul/scm_avx.c:45:13: error: always_inline function '_mm256_add_epi16' requires target feature 'xsave', but would be inlined into function 'schoolbook_avx_new3_acc' that is compiled without support for 'xsave'
SABER_indcpa.c: c_avx[1] = _mm256_add_epi16(temp, c_avx[1]);
SABER_indcpa.c: ^
SABER_indcpa.c: ././polymul/scm_avx.c:48:9: error: always_inline function '_mm256_mullo_epi16' requires target feature 'xsave', but would be inlined into function 'schoolbook_avx_new3_acc' that is compiled without support for 'xsave'
SABER_indcpa.c: temp = _mm256_mullo_epi16 (a0, b2);
SABER_indcpa.c: ^
SABER_indcpa.c: ././polymul/scm_avx.c:51:13: error: always_inline function '_mm256_add_epi16' requires target feature 'xsave', but would be inlined into function 'schoolbook_avx_new3_acc' that is compiled without support for 'xsave'
SABER_indcpa.c: c_avx[2] = _mm256_add_epi16(temp, c_avx[2]);
SABER_indcpa.c: ^
SABER_indcpa.c: ././polymul/scm_avx.c:54:9: error: always_inline function '_mm256_mullo_epi16' requires target feature 'xsave', but would be inlined into function 'schoolbook_avx_new3_acc' that is compiled without support for 'xsave'
SABER_indcpa.c: temp = _mm256_mullo_epi16 (a0, b3);
SABER_indcpa.c: ^
SABER_indcpa.c: ././polymul/scm_avx.c:58:13: error: always_inline function '_mm256_add_epi16' requires target feature 'xsave', but would be inlined into function 'schoolbook_avx_new3_acc' that is compiled without support for 'xsave'
SABER_indcpa.c: c_avx[3] = _mm256_add_epi16(temp, c_avx[3]);
SABER_indcpa.c: ^
SABER_indcpa.c: ././polymul/scm_avx.c:60:9: error: always_inline function '_mm256_mullo_epi16' requires target feature 'xsave', but would be inlined into function 'schoolbook_avx_new3_acc' that is compiled without support for 'xsave'
SABER_indcpa.c: temp = _mm256_mullo_epi16 (a0, b4);
SABER_indcpa.c: ^
SABER_indcpa.c: ././polymul/scm_avx.c:65:13: error: always_inline function '_mm256_add_epi16' requires target feature 'xsave', but would be inlined into function 'schoolbook_avx_new3_acc' that is compiled without support for 'xsave'
SABER_indcpa.c: c_avx[4] = _mm256_add_epi16(temp, c_avx[4]);
SABER_indcpa.c: ...

Number of similar (compiler,implementation) pairs: 4, namely:
CompilerImplementations
clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE T:avx2
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE T:avx2
clang -march=native -O -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE T:avx2
clang -march=native -Os -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE T:avx2

Compiler output

Implementation: T:avx2
Security model: timingleaks
Compiler: clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE
SABER_indcpa.c: In file included from SABER_indcpa.c:9:
SABER_indcpa.c: In file included from ././polymul/toom-cook_4way.c:6:
SABER_indcpa.c: ././polymul/scm_avx.c:43:9: error: always_inline function '_mm256_mullo_epi16' requires target feature 'sse4.2', but would be inlined into function 'schoolbook_avx_new3_acc' that is compiled without support for 'sse4.2'
SABER_indcpa.c: temp = _mm256_mullo_epi16 (a0, b1);
SABER_indcpa.c: ^
SABER_indcpa.c: ././polymul/scm_avx.c:45:13: error: always_inline function '_mm256_add_epi16' requires target feature 'sse4.2', but would be inlined into function 'schoolbook_avx_new3_acc' that is compiled without support for 'sse4.2'
SABER_indcpa.c: c_avx[1] = _mm256_add_epi16(temp, c_avx[1]);
SABER_indcpa.c: ^
SABER_indcpa.c: ././polymul/scm_avx.c:48:9: error: always_inline function '_mm256_mullo_epi16' requires target feature 'sse4.2', but would be inlined into function 'schoolbook_avx_new3_acc' that is compiled without support for 'sse4.2'
SABER_indcpa.c: temp = _mm256_mullo_epi16 (a0, b2);
SABER_indcpa.c: ^
SABER_indcpa.c: ././polymul/scm_avx.c:51:13: error: always_inline function '_mm256_add_epi16' requires target feature 'sse4.2', but would be inlined into function 'schoolbook_avx_new3_acc' that is compiled without support for 'sse4.2'
SABER_indcpa.c: c_avx[2] = _mm256_add_epi16(temp, c_avx[2]);
SABER_indcpa.c: ^
SABER_indcpa.c: ././polymul/scm_avx.c:54:9: error: always_inline function '_mm256_mullo_epi16' requires target feature 'sse4.2', but would be inlined into function 'schoolbook_avx_new3_acc' that is compiled without support for 'sse4.2'
SABER_indcpa.c: temp = _mm256_mullo_epi16 (a0, b3);
SABER_indcpa.c: ^
SABER_indcpa.c: ././polymul/scm_avx.c:58:13: error: always_inline function '_mm256_add_epi16' requires target feature 'sse4.2', but would be inlined into function 'schoolbook_avx_new3_acc' that is compiled without support for 'sse4.2'
SABER_indcpa.c: c_avx[3] = _mm256_add_epi16(temp, c_avx[3]);
SABER_indcpa.c: ^
SABER_indcpa.c: ././polymul/scm_avx.c:60:9: error: always_inline function '_mm256_mullo_epi16' requires target feature 'sse4.2', but would be inlined into function 'schoolbook_avx_new3_acc' that is compiled without support for 'sse4.2'
SABER_indcpa.c: temp = _mm256_mullo_epi16 (a0, b4);
SABER_indcpa.c: ^
SABER_indcpa.c: ././polymul/scm_avx.c:65:13: error: always_inline function '_mm256_add_epi16' requires target feature 'sse4.2', but would be inlined into function 'schoolbook_avx_new3_acc' that is compiled without support for 'sse4.2'
SABER_indcpa.c: c_avx[4] = _mm256_add_epi16(temp, c_avx[4]);
SABER_indcpa.c: ...

Number of similar (compiler,implementation) pairs: 1, namely:
CompilerImplementations
clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE T:avx2

Compiler output

Implementation: T:avx2
Security model: timingleaks
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE
SABER_indcpa.c: In file included from ./polymul/toom-cook_4way.c:6:0,
SABER_indcpa.c: from SABER_indcpa.c:9:
SABER_indcpa.c: ./polymul/scm_avx.c: In function 'mul_add':
SABER_indcpa.c: ./polymul/scm_avx.c:6:1: warning: AVX vector return without AVX enabled changes the ABI [-Wpsabi]
SABER_indcpa.c: inline __m256i mul_add(__m256i a, __m256i b, __m256i c) {
SABER_indcpa.c: ^~~~~~
SABER_indcpa.c: ./polymul/scm_avx.c:6:16: note: The ABI for passing parameters with 32-byte alignment has changed in GCC 4.6
SABER_indcpa.c: inline __m256i mul_add(__m256i a, __m256i b, __m256i c) {
SABER_indcpa.c: ^~~~~~~
SABER_indcpa.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/6/include/immintrin.h:43:0,
SABER_indcpa.c: from SABER_indcpa.h:4,
SABER_indcpa.c: from SABER_indcpa.c:5:
SABER_indcpa.c: /usr/lib/gcc/x86_64-linux-gnu/6/include/avx2intrin.h:112:1: error: inlining failed in call to always_inline '_mm256_add_epi16': target specific option mismatch
SABER_indcpa.c: _mm256_add_epi16 (__m256i __A, __m256i __B)
SABER_indcpa.c: ^~~~~~~~~~~~~~~~
SABER_indcpa.c: In file included from ./polymul/toom-cook_4way.c:6:0,
SABER_indcpa.c: from SABER_indcpa.c:9:
SABER_indcpa.c: ./polymul/scm_avx.c:7:12: note: called from here
SABER_indcpa.c: return _mm256_add_epi16(_mm256_mullo_epi16(a, b), c);
SABER_indcpa.c: ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
SABER_indcpa.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/6/include/immintrin.h:43:0,
SABER_indcpa.c: from SABER_indcpa.h:4,
SABER_indcpa.c: from SABER_indcpa.c:5:
SABER_indcpa.c: /usr/lib/gcc/x86_64-linux-gnu/6/include/avx2intrin.h:553:1: error: inlining failed in call to always_inline '_mm256_mullo_epi16': target specific option mismatch
SABER_indcpa.c: _mm256_mullo_epi16 (__m256i __A, __m256i __B)
SABER_indcpa.c: ...

Number of similar (compiler,implementation) pairs: 4, namely:
CompilerImplementations
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE T:avx2
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv -fPIC -fPIE T:avx2
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv -fPIC -fPIE T:avx2
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv -fPIC -fPIE T:avx2

Compiler output

Implementation: T:avx2_nttmul
Security model: timingleaks
Compiler: clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE
poly.c: poly.c:31:26: error: always_inline function '_mm256_set1_epi16' requires target feature 'xsave', but would be inlined into function 'nttmul_poly_crt' that is compiled without support for 'xsave'
poly.c: const __m256i u_pinv = _mm256_set1_epi16(CRT_U_PINV);
poly.c: ^
poly.c: poly.c:32:21: error: always_inline function '_mm256_set1_epi16' requires target feature 'xsave', but would be inlined into function 'nttmul_poly_crt' that is compiled without support for 'xsave'
poly.c: const __m256i u = _mm256_set1_epi16(CRT_U);
poly.c: ^
poly.c: poly.c:33:22: error: always_inline function '_mm256_load_si256' requires target feature 'xsave', but would be inlined into function 'nttmul_poly_crt' that is compiled without support for 'xsave'
poly.c: const __m256i p0 = _mm256_load_si256((__m256i *)&PDATA0[_16XP]);
poly.c: ^
poly.c: poly.c:34:22: error: always_inline function '_mm256_load_si256' requires target feature 'xsave', but would be inlined into function 'nttmul_poly_crt' that is compiled without support for 'xsave'
poly.c: const __m256i p1 = _mm256_load_si256((__m256i *)&PDATA1[_16XP]);
poly.c: ^
poly.c: poly.c:35:23: error: always_inline function '_mm256_set1_epi16' requires target feature 'xsave', but would be inlined into function 'nttmul_poly_crt' that is compiled without support for 'xsave'
poly.c: const __m256i mod = _mm256_set1_epi16(KEM_Q-1);
poly.c: ^
poly.c: poly.c:36:30: error: always_inline function '_mm256_load_si256' requires target feature 'xsave', but would be inlined into function 'nttmul_poly_crt' that is compiled without support for 'xsave'
poly.c: const __m256i mont0_pinv = _mm256_load_si256((__m256i *)&PDATA0[_16XMONT_PINV]);
poly.c: ^
poly.c: poly.c:37:25: error: always_inline function '_mm256_load_si256' requires target feature 'xsave', but would be inlined into function 'nttmul_poly_crt' that is compiled without support for 'xsave'
poly.c: const __m256i mont0 = _mm256_load_si256((__m256i *)&PDATA0[_16XMONT]);
poly.c: ^
poly.c: poly.c:40:10: error: always_inline function '_mm256_load_si256' requires target feature 'xsave', but would be inlined into function 'nttmul_poly_crt' that is compiled without support for 'xsave'
poly.c: f0 = _mm256_load_si256((__m256i *)&a->coeffs[16*i]);
poly.c: ^
poly.c: poly.c:41:10: error: always_inline function '_mm256_load_si256' requires target feature 'xsave', but would be inlined into function 'nttmul_poly_crt' that is compiled without support for 'xsave'
poly.c: ...

Number of similar (compiler,implementation) pairs: 4, namely:
CompilerImplementations
clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE T:avx2_nttmul
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE T:avx2_nttmul
clang -march=native -O -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE T:avx2_nttmul
clang -march=native -Os -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE T:avx2_nttmul

Compiler output

Implementation: T:avx2_nttmul
Security model: timingleaks
Compiler: clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE
poly.c: poly.c:31:26: error: always_inline function '_mm256_set1_epi16' requires target feature 'sse4.2', but would be inlined into function 'nttmul_poly_crt' that is compiled without support for 'sse4.2'
poly.c: const __m256i u_pinv = _mm256_set1_epi16(CRT_U_PINV);
poly.c: ^
poly.c: poly.c:32:21: error: always_inline function '_mm256_set1_epi16' requires target feature 'sse4.2', but would be inlined into function 'nttmul_poly_crt' that is compiled without support for 'sse4.2'
poly.c: const __m256i u = _mm256_set1_epi16(CRT_U);
poly.c: ^
poly.c: poly.c:33:22: error: always_inline function '_mm256_load_si256' requires target feature 'sse4.2', but would be inlined into function 'nttmul_poly_crt' that is compiled without support for 'sse4.2'
poly.c: const __m256i p0 = _mm256_load_si256((__m256i *)&PDATA0[_16XP]);
poly.c: ^
poly.c: poly.c:34:22: error: always_inline function '_mm256_load_si256' requires target feature 'sse4.2', but would be inlined into function 'nttmul_poly_crt' that is compiled without support for 'sse4.2'
poly.c: const __m256i p1 = _mm256_load_si256((__m256i *)&PDATA1[_16XP]);
poly.c: ^
poly.c: poly.c:35:23: error: always_inline function '_mm256_set1_epi16' requires target feature 'sse4.2', but would be inlined into function 'nttmul_poly_crt' that is compiled without support for 'sse4.2'
poly.c: const __m256i mod = _mm256_set1_epi16(KEM_Q-1);
poly.c: ^
poly.c: poly.c:36:30: error: always_inline function '_mm256_load_si256' requires target feature 'sse4.2', but would be inlined into function 'nttmul_poly_crt' that is compiled without support for 'sse4.2'
poly.c: const __m256i mont0_pinv = _mm256_load_si256((__m256i *)&PDATA0[_16XMONT_PINV]);
poly.c: ^
poly.c: poly.c:37:25: error: always_inline function '_mm256_load_si256' requires target feature 'sse4.2', but would be inlined into function 'nttmul_poly_crt' that is compiled without support for 'sse4.2'
poly.c: const __m256i mont0 = _mm256_load_si256((__m256i *)&PDATA0[_16XMONT]);
poly.c: ^
poly.c: poly.c:40:10: error: always_inline function '_mm256_load_si256' requires target feature 'sse4.2', but would be inlined into function 'nttmul_poly_crt' that is compiled without support for 'sse4.2'
poly.c: f0 = _mm256_load_si256((__m256i *)&a->coeffs[16*i]);
poly.c: ^
poly.c: poly.c:41:10: error: always_inline function '_mm256_load_si256' requires target feature 'sse4.2', but would be inlined into function 'nttmul_poly_crt' that is compiled without support for 'sse4.2'
poly.c: ...

Number of similar (compiler,implementation) pairs: 1, namely:
CompilerImplementations
clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE T:avx2_nttmul

Compiler output

Implementation: T:avx2_nttmul
Security model: timingleaks
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE
poly.c: poly.c: In function 'mulmod':
poly.c: poly.c:7:1: warning: AVX vector return without AVX enabled changes the ABI [-Wpsabi]
poly.c: static inline __m256i mulmod(const __m256i a, const __m256i b_pinv, const __m256i b, const __m256i p) {
poly.c: ^~~~~~
poly.c: poly.c:7:23: note: The ABI for passing parameters with 32-byte alignment has changed in GCC 4.6
poly.c: static inline __m256i mulmod(const __m256i a, const __m256i b_pinv, const __m256i b, const __m256i p) {
poly.c: ^~~~~~
poly.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/6/include/immintrin.h:43:0,
poly.c: from poly.c:3:
poly.c: /usr/lib/gcc/x86_64-linux-gnu/6/include/avx2intrin.h:808:1: error: inlining failed in call to always_inline '_mm256_sub_epi16': target specific option mismatch
poly.c: _mm256_sub_epi16 (__m256i __A, __m256i __B)
poly.c: ^~~~~~~~~~~~~~~~
poly.c: poly.c:12:5: note: called from here
poly.c: t = _mm256_sub_epi16(u,t);
poly.c: ~~^~~~~~~~~~~~~~~~~~~~~~~
poly.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/6/include/immintrin.h:43:0,
poly.c: from poly.c:3:
poly.c: /usr/lib/gcc/x86_64-linux-gnu/6/include/avx2intrin.h:546:1: error: inlining failed in call to always_inline '_mm256_mulhi_epi16': target specific option mismatch
poly.c: _mm256_mulhi_epi16 (__m256i __A, __m256i __B)
poly.c: ^~~~~~~~~~~~~~~~~~
poly.c: poly.c:11:5: note: called from here
poly.c: t = _mm256_mulhi_epi16(t,p);
poly.c: ~~^~~~~~~~~~~~~~~~~~~~~~~~~
poly.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/6/include/immintrin.h:43:0,
poly.c: from poly.c:3:
poly.c: ...

Number of similar (compiler,implementation) pairs: 4, namely:
CompilerImplementations
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE T:avx2_nttmul
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv -fPIC -fPIE T:avx2_nttmul
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv -fPIC -fPIE T:avx2_nttmul
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv -fPIC -fPIE T:avx2_nttmul

Namespace violations

Implementation: T:ref
Security model: timingleaks
Compiler: clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE
SABER_indcpa.o GenMatrix T
SABER_indcpa.o InnerProd T
SABER_indcpa.o MatrixVectorMul T
SABER_indcpa.o POL2MSG T
SABER_indcpa.o clock1 C
SABER_indcpa.o clock2 C
SABER_indcpa.o clock_cl_mv C
SABER_indcpa.o clock_cl_sm C
SABER_indcpa.o clock_kp_mv C
SABER_indcpa.o clock_kp_sm C
SABER_indcpa.o indcpa_kem_dec T
SABER_indcpa.o indcpa_kem_enc T
SABER_indcpa.o indcpa_kem_keypair T
SABER_indcpa.o karatsuba_simple T
SABER_indcpa.o pol_mul T
SABER_indcpa.o print_poly2 T
SABER_indcpa.o reduce T
SABER_indcpa.o toom_cook_4way T
cbd.o cbd T
fips202.o KeccakF1600_StatePermute T
fips202.o cshake128_simple T
fips202.o cshake128_simple_absorb T
fips202.o cshake128_simple_squeezeblocks T
fips202.o sha3_256 T
fips202.o sha3_512 T
fips202.o shake128 T
kem.o clock1 C
kem.o clock2 C
kem.o clock_cl_mv C
kem.o clock_cl_sm C
kem.o clock_kp_mv C
kem.o clock_kp_sm C
pack_unpack.o BS2POL T
pack_unpack.o BS2POLVEC T
pack_unpack.o BS2POLVECp T
pack_unpack.o BS2POLVECq T
pack_unpack.o POLVEC2BS T
pack_unpack.o POLVECp2BS T
pack_unpack.o POLVECq2BS T
pack_unpack.o SABER_pack_3bit T
pack_unpack.o SABER_pack_4bit T
pack_unpack.o SABER_pack_6bit T
pack_unpack.o SABER_un_pack3bit T
pack_unpack.o SABER_un_pack4bit T
pack_unpack.o SABER_un_pack6bit T
poly.o GenSecret T
verify.o cmov T
verify.o verify T

Number of similar (compiler,implementation) pairs: 9, namely:
CompilerImplementations
clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE T:ref
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE T:ref
clang -march=native -O -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE T:ref
clang -march=native -Os -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE T:ref
clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE T:ref
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE T:ref
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv -fPIC -fPIE T:ref
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv -fPIC -fPIE T:ref
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv -fPIC -fPIE T:ref