Implementation notes: amd64, hydra8, crypto_sign/rainbow5cclassic963664

Computer: hydra8
Architecture: amd64
CPU ID: GenuineIntel-000306a9-bfebfbff
SUPERCOP version: 20200618
Operation: crypto_sign
Primitive: rainbow5cclassic963664

Time	Object size	Test size	Implementation	Compiler	Benchmark date	SUPERCOP version
1805736	142529 8 0	162275 920 1688	`ssse3`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200618	20200618
1818412	116623 8 0	137725 912 1656	`ssse3`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200618	20200618
1840548	100045 8 0	119621 912 1656	`ssse3`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200618	20200618
1841300	100045 8 0	119621 912 1656	`ssse3`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200618	20200618
1933400	90592 8 0	108459 920 1688	`ssse3`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200618	20200618
2017168	71055 8 0	87003 904 1656	`ssse3`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200618	20200618
2042620	89110 8 0	106899 920 1688	`ssse3`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200618	20200618
2364424	50891 8 0	68059 912 1656	`ssse3`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200618	20200618
5123268	147601 0 0	167112 912 1688	`ref`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200618	20200618
5329540	225431 0 0	244696 904 1688	`amd64`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200618	20200618
5646532	121624 0 0	139730 896 1656	`amd64`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200618	20200618
5763824	109528 0 0	126954 904 1656	`ref`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200618	20200618
6335416	126370 0 0	143530 896 1656	`amd64`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200618	20200618
6420148	126370 0 0	143530 896 1656	`amd64`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200618	20200618
6562588	113619 0 0	129186 904 1656	`ref`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200618	20200618
6566432	113619 0 0	129186 904 1656	`ref`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200618	20200618
6645416	133766 0 0	152474 896 1656	`amd64`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200618	20200618
6991916	120532 0 0	137618 904 1656	`ref`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200618	20200618
9383268	106539 0 0	124248 904 1688	`amd64`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200618	20200618
9462980	71179 0 0	87624 888 1656	`amd64`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200618	20200618
9724976	101040 0 0	118744 904 1688	`amd64`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200618	20200618
15768296	60757 0 0	78480 912 1688	`ref`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200618	20200618
16015440	61693 0 0	79464 912 1688	`ref`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200618	20200618
16675564	54229 0 0	70720 896 1656	`ref`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200618	20200618
18437036	32332 0 0	49544 896 1656	`amd64`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200618	20200618
28311416	28957 0 0	46064 904 1656	`ref`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200618	20200618

Compiler output

Implementation: crypto_sign/rainbow5cclassic963664/avx2
Compiler: clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE

blas_comm.c: In file included from blas_comm.c:6:
blas_comm.c: In file included from ./blas.h:23:
blas_comm.c: ./blas_avx2.h:171:15: error: '__builtin_ia32_permti256' needs target feature avx2
blas_comm.c: __m256i ml = _mm256_permute2x128_si256( m_tab , m_tab , 0 );
blas_comm.c: ^
blas_comm.c: /usr/lib/llvm-6.0/lib/clang/6.0.0/include/avx2intrin.h:975:12: note: expanded from macro '_mm256_permute2x128_si256'
blas_comm.c: (__m256i)__builtin_ia32_permti256((__m256i)(V1), (__m256i)(V2), (M)); })
blas_comm.c: ^
blas_comm.c: In file included from blas_comm.c:6:
blas_comm.c: In file included from ./blas.h:23:
blas_comm.c: ./blas_avx2.h:172:15: error: '__builtin_ia32_permti256' needs target feature avx2
blas_comm.c: __m256i mh = _mm256_permute2x128_si256( m_tab , m_tab , 0x11 );
blas_comm.c: ^
blas_comm.c: /usr/lib/llvm-6.0/lib/clang/6.0.0/include/avx2intrin.h:975:12: note: expanded from macro '_mm256_permute2x128_si256'
blas_comm.c: (__m256i)__builtin_ia32_permti256((__m256i)(V1), (__m256i)(V2), (M)); })
blas_comm.c: ^
blas_comm.c: In file included from blas_comm.c:6:
blas_comm.c: In file included from ./blas.h:23:
blas_comm.c: In file included from ./blas_avx2.h:15:
blas_comm.c: ./gf16_avx2.h:29:9: error: always_inline function '_mm256_shuffle_epi8' requires target feature 'avx2', but would be inlined into function 'linear_transform_8x8_256b' that is compiled without support for 'avx2'
blas_comm.c: return _mm256_shuffle_epi8(tab_l,v&mask_f)^_mm256_shuffle_epi8(tab_h,_mm256_srli_epi16(v,4)&mask_f);
blas_comm.c: ^
blas_comm.c: ./gf16_avx2.h:29:45: error: always_inline function '_mm256_shuffle_epi8' requires target feature 'avx2', but would be inlined into function 'linear_transform_8x8_256b' that is compiled without support for 'avx2'
blas_comm.c: return _mm256_shuffle_epi8(tab_l,v&mask_f)^_mm256_shuffle_epi8(tab_h,_mm256_srli_epi16(v,4)&mask_f);
blas_comm.c: ^
blas_comm.c: ...

Number of similar (compiler,implementation) pairs: 4, namely:

Compiler	Implementations
clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	avx2
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	avx2
clang -march=native -O -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	avx2
clang -march=native -Os -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	avx2

Compiler output

Implementation: crypto_sign/rainbow5cclassic963664/avx2
Compiler: clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE

blas_comm.c: In file included from blas_comm.c:6:
blas_comm.c: In file included from ./blas.h:23:
blas_comm.c: ./blas_avx2.h:88:17: error: always_inline function '_mm256_loadu_si256' requires target feature 'sse4.2', but would be inlined into function 'gf256v_add_avx2' that is compiled without support for 'sse4.2'
blas_comm.c: __m256i inp = _mm256_loadu_si256( (__m256i*) (a+i*32) );
blas_comm.c: ^
blas_comm.c: ./blas_avx2.h:89:17: error: always_inline function '_mm256_loadu_si256' requires target feature 'sse4.2', but would be inlined into function 'gf256v_add_avx2' that is compiled without support for 'sse4.2'
blas_comm.c: __m256i out = _mm256_loadu_si256( (__m256i*) (accu_b+i*32) );
blas_comm.c: ^
blas_comm.c: ./blas_avx2.h:91:3: error: always_inline function '_mm256_storeu_si256' requires target feature 'sse4.2', but would be inlined into function 'gf256v_add_avx2' that is compiled without support for 'sse4.2'
blas_comm.c: _mm256_storeu_si256( (__m256i*) (accu_b+i*32) , out );
blas_comm.c: ^
blas_comm.c: ./blas_avx2.h:170:18: error: always_inline function '_mm256_load_si256' requires target feature 'sse4.2', but would be inlined into function 'gf256v_madd_avx2' that is compiled without support for 'sse4.2'
blas_comm.c: __m256i m_tab = _mm256_load_si256( (__m256i*) (__gf256_mul + 32*b) );
blas_comm.c: ^
blas_comm.c: ./blas_avx2.h:171:15: error: '__builtin_ia32_permti256' needs target feature avx2
blas_comm.c: __m256i ml = _mm256_permute2x128_si256( m_tab , m_tab , 0 );
blas_comm.c: ^
blas_comm.c: /usr/lib/llvm-6.0/lib/clang/6.0.0/include/avx2intrin.h:975:12: note: expanded from macro '_mm256_permute2x128_si256'
blas_comm.c: (__m256i)__builtin_ia32_permti256((__m256i)(V1), (__m256i)(V2), (M)); })
blas_comm.c: ^
blas_comm.c: In file included from blas_comm.c:6:
blas_comm.c: In file included from ./blas.h:23:
blas_comm.c: ./blas_avx2.h:172:15: error: '__builtin_ia32_permti256' needs target feature avx2
blas_comm.c: __m256i mh = _mm256_permute2x128_si256( m_tab , m_tab , 0x11 );
blas_comm.c: ^
blas_comm.c: ...

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

Compiler	Implementations
clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	avx2

Compiler output

Implementation: crypto_sign/rainbow5cclassic963664/ssse3
Compiler: clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE

blas_comm.c: In file included from blas_comm.c:6:
blas_comm.c: In file included from ./blas.h:23:
blas_comm.c: In file included from ./blas_sse.h:16:
blas_comm.c: ./gf16_sse.h:34:9: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'linear_transform_8x8_128b' that is compiled without support for 'ssse3'
blas_comm.c: return _mm_shuffle_epi8(tab_l,v&mask_f)^_mm_shuffle_epi8(tab_h,_mm_srli_epi16(v,4)&mask_f);
blas_comm.c: ^
blas_comm.c: ./gf16_sse.h:34:42: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'linear_transform_8x8_128b' that is compiled without support for 'ssse3'
blas_comm.c: return _mm_shuffle_epi8(tab_l,v&mask_f)^_mm_shuffle_epi8(tab_h,_mm_srli_epi16(v,4)&mask_f);
blas_comm.c: ^
blas_comm.c: 2 errors generated.

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

Compiler	Implementations
clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	ssse3

Compiler output

Implementation: crypto_sign/rainbow5cclassic963664/avx2
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE

blas_comm.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/7/include/immintrin.h:43:0,
blas_comm.c: from blas_avx2.h:10,
blas_comm.c: from blas.h:23,
blas_comm.c: from blas_comm.c:6:
blas_comm.c: gf16_avx2.h: In function 'linear_transform_8x8_256b':
blas_comm.c: /usr/lib/gcc/x86_64-linux-gnu/7/include/avx2intrin.h:588:1: error: inlining failed in call to always_inline '_mm256_shuffle_epi8': target specific option mismatch
blas_comm.c: _mm256_shuffle_epi8 (__m256i __X, __m256i __Y)
blas_comm.c: ^~~~~~~~~~~~~~~~~~~
blas_comm.c: In file included from blas_avx2.h:15:0,
blas_comm.c: from blas.h:23,
blas_comm.c: from blas_comm.c:6:
blas_comm.c: gf16_avx2.h:29:45: note: called from here
blas_comm.c: return _mm256_shuffle_epi8(tab_l,v&mask_f)^_mm256_shuffle_epi8(tab_h,_mm256_srli_epi16(v,4)&mask_f);
blas_comm.c: ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
blas_comm.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/7/include/immintrin.h:43:0,
blas_comm.c: from blas_avx2.h:10,
blas_comm.c: from blas.h:23,
blas_comm.c: from blas_comm.c:6:
blas_comm.c: /usr/lib/gcc/x86_64-linux-gnu/7/include/avx2intrin.h:759:1: error: inlining failed in call to always_inline '_mm256_srli_epi16': target specific option mismatch
blas_comm.c: _mm256_srli_epi16 (__m256i __A, int __B)
blas_comm.c: ^~~~~~~~~~~~~~~~~
blas_comm.c: In file included from blas_avx2.h:15:0,
blas_comm.c: from blas.h:23,
blas_comm.c: from blas_comm.c:6:
blas_comm.c: gf16_avx2.h:29:71: note: called from here
blas_comm.c: ...

Number of similar (compiler,implementation) pairs: 4, namely:

Compiler	Implementations
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE	avx2
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv -fPIC -fPIE	avx2
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv -fPIC -fPIE	avx2
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv -fPIC -fPIE	avx2

Namespace violations

Implementation: crypto_sign/rainbow5cclassic963664/ssse3
Compiler: clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE

blas_comm.o gf16mat_gauss_elim T
blas_comm.o gf16mat_inv T
blas_comm.o gf16mat_mul T
blas_comm.o gf16mat_prod T
blas_comm.o gf16mat_solve_linear_eq T
blas_comm.o gf256mat_gauss_elim T
blas_comm.o gf256mat_inv T
blas_comm.o gf256mat_mul T
blas_comm.o gf256mat_prod T
blas_comm.o gf256mat_solve_linear_eq T
blas_comm.o gf256v_is_zero T
blas_comm.o gf256v_polymul T
blas_comm.o gf256v_set_zero T
blas_comm_sse.o gf16mat_gauss_elim_sse T
blas_comm_sse.o gf16mat_prod_16_sse T
blas_comm_sse.o gf16mat_prod_add_multab_sse T
blas_comm_sse.o gf16mat_prod_multab_sse T
blas_comm_sse.o gf16mat_prod_sse T
blas_comm_sse.o gf16mat_solve_linear_eq_sse T
blas_comm_sse.o gf256mat_gauss_elim_sse T
blas_comm_sse.o gf256mat_prod_add_multab_sse T
blas_comm_sse.o gf256mat_prod_add_sse T
blas_comm_sse.o gf256mat_prod_multab_sse T
blas_comm_sse.o gf256mat_prod_sse T
blas_comm_sse.o gf256mat_solve_linear_eq_sse T
gf16.o __gf16_exp R
gf16.o __gf16_inv R
gf16.o __gf16_log R
gf16.o __gf16_mul D
gf16.o __gf16_mulx2 R
gf16.o __gf16_squ R
gf16.o __gf16_squ_sl4 R
gf16.o __gf16_squ_x8 R
gf16.o __gf256_mul R
gf16.o __mask_0x55 R
gf16.o __mask_16 R
gf16.o __mask_low R
parallel_matrix_op.o UpperTrianglize T
parallel_matrix_op.o batch_2trimat_madd_gf16 T
parallel_matrix_op.o batch_2trimat_madd_gf256 T
parallel_matrix_op.o batch_bmatTr_madd_gf16 T
parallel_matrix_op.o batch_bmatTr_madd_gf256 T
parallel_matrix_op.o batch_matTr_madd_gf16 T
parallel_matrix_op.o batch_matTr_madd_gf256 T
parallel_matrix_op.o batch_mat_madd_gf16 T
parallel_matrix_op.o batch_mat_madd_gf256 T
parallel_matrix_op.o batch_quad_recmat_eval_gf16 T
parallel_matrix_op.o batch_quad_recmat_eval_gf256 T
parallel_matrix_op.o batch_quad_trimat_eval_gf16 T
parallel_matrix_op.o batch_quad_trimat_eval_gf256 T
parallel_matrix_op.o batch_trimatTr_madd_gf16 T
parallel_matrix_op.o batch_trimatTr_madd_gf256 T
parallel_matrix_op.o batch_trimat_madd_gf16 T
parallel_matrix_op.o batch_trimat_madd_gf256 T
parallel_matrix_op_sse.o batch_2trimat_madd_multab_gf16_sse T
parallel_matrix_op_sse.o batch_2trimat_madd_multab_gf256_sse T
parallel_matrix_op_sse.o batch_bmatTr_madd_multab_gf16_sse T
parallel_matrix_op_sse.o batch_bmatTr_madd_multab_gf256_sse T
parallel_matrix_op_sse.o batch_matTr_madd_multab_gf16_sse T
parallel_matrix_op_sse.o batch_matTr_madd_multab_gf256_sse T
parallel_matrix_op_sse.o batch_mat_madd_multab_gf16_sse T
parallel_matrix_op_sse.o batch_mat_madd_multab_gf256_sse T
parallel_matrix_op_sse.o batch_quad_trimat_eval_multab_gf16_sse T
parallel_matrix_op_sse.o batch_quad_trimat_eval_multab_gf256_sse T
parallel_matrix_op_sse.o batch_trimatTr_madd_multab_gf16_sse T
parallel_matrix_op_sse.o batch_trimatTr_madd_multab_gf256_sse T
parallel_matrix_op_sse.o batch_trimat_madd_multab_gf16_sse T
parallel_matrix_op_sse.o batch_trimat_madd_multab_gf256_sse T
rainbow_blas.o public_map T
rainbow_keypair.o cpk_to_pk T
rainbow_keypair.o generate_compact_keypair_cyclic T
rainbow_keypair.o generate_keypair T
rainbow_keypair.o generate_keypair_cyclic T
rainbow_keypair.o generate_secretkey T
rainbow_keypair.o generate_secretkey_cyclic T
rainbow_keypair.o rainbow_evaluate_cpk T
rainbow_keypair.o sk_to_pk T
rainbow_keypair_computation.o calculate_F_from_Q T
rainbow_keypair_computation.o calculate_Q_from_F T
rainbow_keypair_computation.o calculate_Q_from_F_cyclic T
rainbow_keypair_computation.o extcpk_to_pk T
rainbow_keypair_computation_simd.o calculate_F_from_Q_simd T
rainbow_keypair_computation_simd.o calculate_Q_from_F_cyclic_simd T
rainbow_keypair_computation_simd.o calculate_Q_from_F_simd T
rainbow_simd.o rainbow_sign T
rainbow_simd.o rainbow_sign_cyclic T
rainbow_simd.o rainbow_verify T
rainbow_simd.o rainbow_verify_cyclic T
rng.o AES256_CTR_DRBG_Update T
rng.o AES256_ECB T
rng.o DRBG_ctx C
rng.o handleErrors T
rng.o randombytes_init T
rng.o randombytes_init_with_state T
rng.o randombytes_with_state T
rng.o seedexpander T
rng.o seedexpander_init T
utils.o byte_fdump T
utils.o byte_fget T
utils.o byte_from_binfile T
utils.o byte_from_file T
utils.o byte_read_file T
utils_hash.o hash_msg T
utils_prng.o prng_gen T
utils_prng.o prng_set T

Number of similar (compiler,implementation) pairs: 8, namely:

Compiler	Implementations
clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	ssse3
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	ssse3
clang -march=native -O -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	ssse3
clang -march=native -Os -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	ssse3
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE	ssse3
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv -fPIC -fPIE	ssse3
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv -fPIC -fPIE	ssse3
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv -fPIC -fPIE	ssse3

Namespace violations

Implementation: crypto_sign/rainbow5cclassic963664/amd64
Compiler: clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE

Number of similar (compiler,implementation) pairs: 18, namely:

Compiler	Implementations
clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	amd64 ref
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	amd64 ref
clang -march=native -O -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	amd64 ref
clang -march=native -Os -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	amd64 ref
clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	amd64 ref
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE	amd64 ref
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv -fPIC -fPIE	amd64 ref
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv -fPIC -fPIE	amd64 ref
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv -fPIC -fPIE	amd64 ref