Implementation notes: amd64, margaux, crypto_kem/ntskem13136

Computer: margaux
Microarchitecture: amd64; Core 2 65nm (6fb)
Architecture: amd64
CPU ID: GenuineIntel-000006fb-bfebfbff
SUPERCOP version: 20240107
Operation: crypto_kem
Primitive: ntskem13136

Time	Object size	Test size	Implementation	Compiler	Benchmark date	SUPERCOP version
3301880	179383 84 16	198213 980 1728	`T:sse2`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231229	20231212
3355089	97123 84 16	114845 980 1728	`T:sse2`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231229	20231212
3517688	77489 84 16	93271 972 1728	`T:sse2`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231229	20231212
3596694	103531 84 16	120141 980 1728	`T:sse2`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231229	20231212
5062790	165147 84 16	185461 980 1728	`T:opt`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231229	20231212
5135555	165809 84 16	185845 980 1728	`T:opt`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231229	20231212
5270049	95144 84 16	114381 980 1728	`T:opt`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231229	20231212
5365870	71529 84 16	88855 972 1728	`T:opt`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231229	20231212
5506236	97793 84 16	115965 980 1728	`T:opt`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231229	20231212
5685369	78125 84 16	97007 956 1792	`T:opt`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231229	20231212
5760784	105581 84 16	125567 956 1792	`T:opt`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231229	20231212
6011247	75321 84 16	93535 956 1792	`T:opt`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231229	20231212
8365077	70487 84 16	87462 948 1760	`T:opt`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231229	20231212
38786696	29886 76 16	47997 964 1728	`T:ref`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231229	20231212
41034165	33973 76 16	53911 924 1792	`T:ref`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231229	20231212
44074893	26786 76 16	45623 924 1792	`T:ref`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231229	20231212
44268044	24066 76 16	42207 924 1792	`T:ref`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231229	20231212
45282055	38521 76 16	57989 964 1728	`T:ref`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231229	20231212
45526703	38704 76 16	59269 964 1728	`T:ref`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231229	20231212
46886831	38672 76 16	59013 964 1728	`T:ref`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231229	20231212
47341096	22621 76 16	39943 956 1728	`T:ref`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231229	20231212
48626033	21176 76 16	38158 916 1760	`T:ref`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231229	20231212

Compiler output

Implementation: T:avx2
Security model: timingleaks
Compiler: clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE

bitslice_fft_256.c: bitslice_fft_256.c:103:29: error: always_inline function '_mm256_setr_epi32' requires target feature 'avx', but would be inlined into function 'bitslice_butterflies13_256' that is compiled without support for 'avx'
bitslice_fft_256.c: const __m256i mask256 = _mm256_setr_epi32(0, 2, 4, 6, 1, 3, 5, 7);
bitslice_fft_256.c: ^
bitslice_fft_256.c: bitslice_fft_256.c:103:29: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
bitslice_fft_256.c: bitslice_fft_256.c:127:25: error: always_inline function '_mm256_set_epi32' requires target feature 'avx', but would be inlined into function 'bitslice_butterflies13_256' that is compiled without support for 'avx'
bitslice_fft_256.c: out[j][i] = _mm256_set_epi32(-u32[7], -u32[6], -u32[5], -u32[4],
bitslice_fft_256.c: ^
bitslice_fft_256.c: bitslice_fft_256.c:127:25: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
bitslice_fft_256.c: bitslice_fft_256.c:140:20: error: always_inline function '_mm256_permutevar8x32_epi32' requires target feature 'avx2', but would be inlined into function 'bitslice_butterflies13_256' that is compiled without support for 'avx2'
bitslice_fft_256.c: t[0] = _mm256_permutevar8x32_epi32(out[ j ][b], mask256);
bitslice_fft_256.c: ^
bitslice_fft_256.c: bitslice_fft_256.c:140:20: error: AVX vector argument of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
bitslice_fft_256.c: bitslice_fft_256.c:141:20: error: always_inline function '_mm256_permutevar8x32_epi32' requires target feature 'avx2', but would be inlined into function 'bitslice_butterflies13_256' that is compiled without support for 'avx2'
bitslice_fft_256.c: t[1] = _mm256_permutevar8x32_epi32(out[j+1][b], mask256);
bitslice_fft_256.c: ^
bitslice_fft_256.c: bitslice_fft_256.c:141:20: error: AVX vector argument of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
bitslice_fft_256.c: bitslice_fft_256.c:142:20: error: '__builtin_ia32_permti256' needs target feature avx2
bitslice_fft_256.c: v[b] = _mm256_permute2x128_si256(t[0], t[1], 0x20);
bitslice_fft_256.c: ^
bitslice_fft_256.c: /usr/lib/llvm-14/lib/clang/14.0.0/include/avx2intrin.h:821:13: note: expanded from macro '_mm256_permute2x128_si256'
bitslice_fft_256.c: ((__m256i)__builtin_ia32_permti256((__m256i)(V1), (__m256i)(V2), (int)(M)))
bitslice_fft_256.c: ^
bitslice_fft_256.c: bitslice_fft_256.c:143:20: error: '__builtin_ia32_permdi256' needs target feature avx2
bitslice_fft_256.c: v[b] = _mm256_permute4x64_epi64(v[b], 0xD8);
bitslice_fft_256.c: ^
bitslice_fft_256.c: ...

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

Compiler	Implementations
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

bitslice_bma_128.c: bitslice_bma_128.c:315:21: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'bitslice_bma' that is compiled without support for 'ssse3'
bitslice_bma_128.c: out[1][j] = _mm_shuffle_epi8(out[0][j], mask);
bitslice_bma_128.c: ^
bitslice_bma_128.c: bitslice_bma_128.c:316:21: error: '__builtin_ia32_palignr128' needs target feature ssse3
bitslice_bma_128.c: out[0][j] = _mm_alignr_epi8(out[0][j], psi[0][j], 15);
bitslice_bma_128.c: ^
bitslice_bma_128.c: /usr/lib/llvm-14/lib/clang/14.0.0/include/tmmintrin.h:152:13: note: expanded from macro '_mm_alignr_epi8'
bitslice_bma_128.c: ((__m128i)__builtin_ia32_palignr128((__v16qi)(__m128i)(a), \
bitslice_bma_128.c: ^
bitslice_bma_128.c: 2 errors generated.

Number of similar (compiler,implementation) pairs: 2, namely:

Compiler	Implementations
clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:avx2 T:sse2

Compiler output

Implementation: T:avx2
Security model: timingleaks
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE

bitslice_fft_256.c: bitslice_fft_256.c: In function 'bitslice_butterflies13_256':
bitslice_fft_256.c: bitslice_fft_256.c:103:19: warning: AVX vector return without AVX enabled changes the ABI [-Wpsabi]
bitslice_fft_256.c: 103 | const __m256i mask256 = _mm256_setr_epi32(0, 2, 4, 6, 1, 3, 5, 7);
bitslice_fft_256.c: | ^~~~~~~
bitslice_fft_256.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/11/include/immintrin.h:43,
bitslice_fft_256.c: from bitslice_fft_256.h:21,
bitslice_fft_256.c: from bitslice_fft_256.c:19:
bitslice_fft_256.c: /usr/lib/gcc/x86_64-linux-gnu/11/include/avxintrin.h:1371:1: error: inlining failed in call to 'always_inline' '_mm256_setr_epi32': target specific option mismatch
bitslice_fft_256.c: 1371 | _mm256_setr_epi32 (int __A, int __B, int __C, int __D,
bitslice_fft_256.c: | ^~~~~~~~~~~~~~~~~
bitslice_fft_256.c: bitslice_fft_256.c:103:29: note: called from here
bitslice_fft_256.c: 103 | const __m256i mask256 = _mm256_setr_epi32(0, 2, 4, 6, 1, 3, 5, 7);
bitslice_fft_256.c: | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
bitslice_fft_256.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/11/include/immintrin.h:43,
bitslice_fft_256.c: from bitslice_fft_256.h:21,
bitslice_fft_256.c: from bitslice_fft_256.c:19:
bitslice_fft_256.c: /usr/lib/gcc/x86_64-linux-gnu/11/include/avxintrin.h:1265:1: error: inlining failed in call to 'always_inline' '_mm256_set_epi32': target specific option mismatch
bitslice_fft_256.c: 1265 | _mm256_set_epi32 (int __A, int __B, int __C, int __D,
bitslice_fft_256.c: | ^~~~~~~~~~~~~~~~
bitslice_fft_256.c: bitslice_fft_256.c:127:25: note: called from here
bitslice_fft_256.c: 127 | out[j][i] = _mm256_set_epi32(-u32[7], -u32[6], -u32[5], -u32[4],
bitslice_fft_256.c: | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
bitslice_fft_256.c: 128 | -u32[3], -u32[2], -u32[1], -u32[0]);
bitslice_fft_256.c: | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
bitslice_fft_256.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/11/include/immintrin.h:47,
bitslice_fft_256.c: ...

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

Compiler	Implementations
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:sse2
Security model: timingleaks
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE

nts_kem.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/11/include/immintrin.h:39,
nts_kem.c: from ff.h:18,
nts_kem.c: from nts_kem.c:18:
nts_kem.c: nts_kem.c: In function 'create_matrix_G':
nts_kem.c: /usr/lib/gcc/x86_64-linux-gnu/11/include/smmintrin.h:455:1: error: inlining failed in call to 'always_inline' '_mm_extract_epi64': target specific option mismatch
nts_kem.c: 455 | _mm_extract_epi64 (__m128i __X, const int __N)
nts_kem.c: | ^~~~~~~~~~~~~~~~~
nts_kem.c: nts_kem.c:860:57: note: called from here
nts_kem.c: 860 | v[i >> 6] |= (((_mm_extract_epi64(vh[l >> 7][j], 1) & (1ULL << (l & 63))) >> (l & 63)) << (i & 63));
nts_kem.c: | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
nts_kem.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/11/include/immintrin.h:39,
nts_kem.c: from ff.h:18,
nts_kem.c: from nts_kem.c:18:
nts_kem.c: /usr/lib/gcc/x86_64-linux-gnu/11/include/smmintrin.h:455:1: error: inlining failed in call to 'always_inline' '_mm_extract_epi64': target specific option mismatch
nts_kem.c: 455 | _mm_extract_epi64 (__m128i __X, const int __N)
nts_kem.c: | ^~~~~~~~~~~~~~~~~
nts_kem.c: nts_kem.c:863:57: note: called from here
nts_kem.c: 863 | v[i >> 6] |= (((_mm_extract_epi64(vh[l >> 7][j], 0) & (1ULL << (l & 63))) >> (l & 63)) << (i & 63));
nts_kem.c: | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
nts_kem.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/11/include/immintrin.h:39,
nts_kem.c: from ff.h:18,
nts_kem.c: from nts_kem.c:18:
nts_kem.c: /usr/lib/gcc/x86_64-linux-gnu/11/include/smmintrin.h:455:1: error: inlining failed in call to 'always_inline' '_mm_extract_epi64': target specific option mismatch
nts_kem.c: 455 | _mm_extract_epi64 (__m128i __X, const int __N)
nts_kem.c: | ^~~~~~~~~~~~~~~~~
nts_kem.c: ...

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

Compiler	Implementations
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:sse2
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:sse2
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:sse2
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:sse2