Implementation notes: amd64, ghf51, crypto_sign/dilithium3

Computer: ghf51
Architecture: amd64
CPU ID: AuthenticAMD-00810f81-178bfbff
SUPERCOP version: 20200702
Operation: crypto_sign
Primitive: dilithium3

Time	Object size	Test size	Implementation	Compiler	Benchmark date	SUPERCOP version
836290	148429 0 0	169170 784 1600	`avx2`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200713	20200702
837148	142805 0 0	163658 784 1600	`avx2`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200713	20200702
852384	169993 0 0	191700 792 1632	`avx2`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200713	20200702
852748	142805 0 0	163658 784 1600	`avx2`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200713	20200702
876668	132225 0 0	150228 776 1600	`avx2`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200713	20200702
911690	133368 0 0	153548 792 1632	`avx2`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200713	20200702
916786	131930 0 0	151932 792 1632	`avx2`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200713	20200702
924040	128907 0 0	147844 784 1600	`avx2`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200713	20200702
1856686	35863 0 0	57002 784 1600	`ref`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200713	20200702
2070796	50392 0 0	71893 784 1632	`ref`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200713	20200702
2109328	36294 0 0	57418 784 1600	`ref`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200713	20200702
2125032	30434 0 0	51682 784 1600	`ref`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200713	20200702
2155790	30434 0 0	51682 784 1600	`ref`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200713	20200702
2203708	20065 0 0	38492 776 1600	`ref`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200713	20200702
2402634	22227 0 0	42237 784 1632	`ref`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200713	20200702
2459418	21244 0 0	41085 784 1632	`ref`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200713	20200702
2636634	19754 0 0	38525 776 1600	`ref`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200713	20200702

Compiler output

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

fips202x4.c: fips202x4.c:32:12: error: always_inline function '_mm256_xor_si256' requires target feature 'avx2', but would be inlined into function 'keccakx4_absorb' that is compiled without support for 'avx2'
fips202x4.c: s[i] = _mm256_xor_si256(s[i], s[i]);
fips202x4.c: ^
fips202x4.c: fips202x4.c:34:9: error: always_inline function '_mm256_set_epi64x' requires target feature 'avx', but would be inlined into function 'keccakx4_absorb' that is compiled without support for 'avx'
fips202x4.c: idx = _mm256_set_epi64x((long long)in3, (long long)in2, (long long)in1, (long long)in0);
fips202x4.c: ^
fips202x4.c: fips202x4.c:37:11: error: '__builtin_ia32_gatherq_q256' needs target feature avx2
fips202x4.c: t = _mm256_i64gather_epi64((long long *)pos, idx, 1);
fips202x4.c: ^
fips202x4.c: /usr/lib/llvm-10/lib/clang/10.0.0/include/avx2intrin.h:1138:12: note: expanded from macro '_mm256_i64gather_epi64'
fips202x4.c: (__m256i)__builtin_ia32_gatherq_q256((__v4di)_mm256_undefined_si256(), \
fips202x4.c: ^
fips202x4.c: fips202x4.c:37:11: error: always_inline function '_mm256_undefined_si256' requires target feature 'avx', but would be inlined into function 'keccakx4_absorb' that is compiled without support for 'avx'
fips202x4.c: /usr/lib/llvm-10/lib/clang/10.0.0/include/avx2intrin.h:1138:48: note: expanded from macro '_mm256_i64gather_epi64'
fips202x4.c: (__m256i)__builtin_ia32_gatherq_q256((__v4di)_mm256_undefined_si256(), \
fips202x4.c: ^
fips202x4.c: fips202x4.c:37:11: error: always_inline function '_mm256_set1_epi64x' requires target feature 'avx', but would be inlined into function 'keccakx4_absorb' that is compiled without support for 'avx'
fips202x4.c: /usr/lib/llvm-10/lib/clang/10.0.0/include/avx2intrin.h:1141:48: note: expanded from macro '_mm256_i64gather_epi64'
fips202x4.c: (__v4di)_mm256_set1_epi64x(-1), (s))
fips202x4.c: ^
fips202x4.c: fips202x4.c:38:14: error: always_inline function '_mm256_xor_si256' requires target feature 'avx2', but would be inlined into function 'keccakx4_absorb' that is compiled without support for 'avx2'
fips202x4.c: s[i] = _mm256_xor_si256(s[i], t);
fips202x4.c: ^
fips202x4.c: fips202x4.c:48:9: error: '__builtin_ia32_gatherq_q256' needs target feature avx2
fips202x4.c: t = _mm256_i64gather_epi64((long long *)pos, idx, 1);
fips202x4.c: ...

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

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