Implementation notes: amd64, bolero, crypto_kem/ntruhps2048509

Computer: bolero
Architecture: amd64
CPU ID: GenuineIntel-000406f1-bfebfbff
SUPERCOP version: 20210114
Operation: crypto_kem
Primitive: ntruhps2048509

Time	Object size	Test size	Implementation	Compiler	Benchmark date	SUPERCOP version
65712	83784 25984 0	175917 26784 1608	`avx2`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20201018	20201018
66180	91294 25984 0	182748 26784 1576	`avx2`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20201018	20201018
66660	85401 25984 0	176780 26784 1576	`avx2`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20201018	20201018
67412	85401 25984 0	176780 26784 1576	`avx2`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20201018	20201018
78612	74223 25984 0	163429 26784 1608	`avx2`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20201018	20201018
82132	73288 25984 0	160834 26776 1576	`avx2`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20201018	20201018
85108	74419 25984 0	163341 26784 1608	`avx2`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20201018	20201018
90168	73025 25984 0	161045 26776 1576	`avx2`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20201018	20201018
1802696	29274 0 0	120436 792 1576	`ref`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20201018	20201018
1834340	23099 0 0	114228 792 1576	`ref`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20201018	20201018
1841808	23099 0 0	114228 792 1576	`ref`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20201018	20201018
1923376	26855 0 0	117556 792 1576	`ref`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20201018	20201018
1961836	6086 0 0	93562 784 1576	`ref`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20201018	20201018
2353912	21599 0 0	113735 800 1608	`ref`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20201018	20201018
2410456	7240 0 0	96437 792 1608	`ref`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20201018	20201018
6540120	6468 0 0	94381 784 1576	`ref`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20201018	20201018
6707732	7171 0 0	96029 792 1608	`ref`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20201018	20201018

Compiler output

Implementation: avx2
Security model: constbranchindex
Compiler: clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE

poly_r2_inv.c: poly_r2_inv.c:19:14: error: always_inline function '_pext_u64' requires target feature 'bmi2', but would be inlined into function 'crypto_kem_ntruhps2048509_avx2_constbranchindex_poly_R2_tobytes' that is compiled without support for 'bmi2'
poly_r2_inv.c: out[i] = _pext_u64(coeff_pointer[2*i], 0x1000100010001);
poly_r2_inv.c: ^
poly_r2_inv.c: poly_r2_inv.c:20:15: error: always_inline function '_pext_u64' requires target feature 'bmi2', but would be inlined into function 'crypto_kem_ntruhps2048509_avx2_constbranchindex_poly_R2_tobytes' that is compiled without support for 'bmi2'
poly_r2_inv.c: out[i] |= _pext_u64(coeff_pointer[2*i+1], 0x1000100010001) << 4;
poly_r2_inv.c: ^
poly_r2_inv.c: poly_r2_inv.c:22:12: error: always_inline function '_pext_u64' requires target feature 'bmi2', but would be inlined into function 'crypto_kem_ntruhps2048509_avx2_constbranchindex_poly_R2_tobytes' that is compiled without support for 'bmi2'
poly_r2_inv.c: out[i] = _pext_u64(coeff_pointer[2*63], 0x1000100010001);
poly_r2_inv.c: ^
poly_r2_inv.c: poly_r2_inv.c:23:13: error: always_inline function '_pext_u64' requires target feature 'bmi2', but would be inlined into function 'crypto_kem_ntruhps2048509_avx2_constbranchindex_poly_R2_tobytes' that is compiled without support for 'bmi2'
poly_r2_inv.c: out[i] |= _pext_u64(coeff_pointer[2*63+1], 0x1) << 4;
poly_r2_inv.c: ^
poly_r2_inv.c: 4 errors generated.

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

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