Implementation notes: amd64, margaux, crypto_kem/lotus128

Computer: margaux
Microarchitecture: amd64; Core 2 65nm (6fb)
Architecture: amd64
CPU ID: GenuineIntel-000006fb-bfebfbff
SUPERCOP version: 20240425
Operation: crypto_kem
Primitive: lotus128

Time	Object size	Test size	Implementation	Compiler	Benchmark date	SUPERCOP version
1507462	42044 0 8	62114 884 1792	`T:opt`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240511	20240425
2075621	32905 0 8	52898 884 1792	`T:ref`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240511	20240425
2426822	36484 0 8	55746 908 1728	`T:opt`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240511	20240425
2547645	46900 0 8	67106 908 1728	`T:opt`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240511	20240425
2580330	46916 0 8	67362 908 1728	`T:opt`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240511	20240425
2645472	23810 0 8	43026 908 1728	`T:ref`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240511	20240425
2782769	30841 0 8	51210 908 1728	`T:ref`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240511	20240425
2816647	31032 0 8	51146 908 1728	`T:ref`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240511	20240425
4149994	11413 0 8	28636 900 1728	`T:opt`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240511	20240425
5303353	13845 0 8	31922 908 1728	`T:opt`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240511	20240425
6047293	9510 0 8	27490 908 1728	`T:ref`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240511	20240425
6293508	8253 0 8	25412 900 1728	`T:ref`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240511	20240425
10970654	13283 0 8	32274 884 1792	`T:opt`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240511	20240425
11465959	9993 0 8	28898 884 1792	`T:ref`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240511	20240425
12115813	8375 0 8	25578 876 1760	`T:ref`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240511	20240425
12649095	9035 0 8	27418 884 1792	`T:ref`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240511	20240425
12709218	11346 0 8	29834 884 1792	`T:opt`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240511	20240425
13275155	10639 0 8	27970 876 1760	`T:opt`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240511	20240425

Compiler output

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

lwe-arithmetics_avx2.c: lwe-arithmetics_avx2.c:158:12: error: always_inline function '_mm256_setzero_si256' requires target feature 'avx', but would be inlined into function 'submat_negmul' that is compiled without support for 'avx'
lwe-arithmetics_avx2.c: c[0] = _mm256_setzero_si256();
lwe-arithmetics_avx2.c: ^
lwe-arithmetics_avx2.c: lwe-arithmetics_avx2.c:158:12: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
lwe-arithmetics_avx2.c: lwe-arithmetics_avx2.c:159:12: error: always_inline function '_mm256_setzero_si256' requires target feature 'avx', but would be inlined into function 'submat_negmul' that is compiled without support for 'avx'
lwe-arithmetics_avx2.c: c[1] = _mm256_setzero_si256();
lwe-arithmetics_avx2.c: ^
lwe-arithmetics_avx2.c: lwe-arithmetics_avx2.c:159:12: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
lwe-arithmetics_avx2.c: lwe-arithmetics_avx2.c:160:12: error: always_inline function '_mm256_setzero_si256' requires target feature 'avx', but would be inlined into function 'submat_negmul' that is compiled without support for 'avx'
lwe-arithmetics_avx2.c: c[2] = _mm256_setzero_si256();
lwe-arithmetics_avx2.c: ^
lwe-arithmetics_avx2.c: lwe-arithmetics_avx2.c:160:12: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
lwe-arithmetics_avx2.c: lwe-arithmetics_avx2.c:161:12: error: always_inline function '_mm256_setzero_si256' requires target feature 'avx', but would be inlined into function 'submat_negmul' that is compiled without support for 'avx'
lwe-arithmetics_avx2.c: c[3] = _mm256_setzero_si256();
lwe-arithmetics_avx2.c: ^
lwe-arithmetics_avx2.c: lwe-arithmetics_avx2.c:161:12: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
lwe-arithmetics_avx2.c: lwe-arithmetics_avx2.c:164:11: error: always_inline function '_mm256_set1_epi16' requires target feature 'avx', but would be inlined into function 'submat_negmul' that is compiled without support for 'avx'
lwe-arithmetics_avx2.c: a = _mm256_set1_epi16(A[j]);
lwe-arithmetics_avx2.c: ^
lwe-arithmetics_avx2.c: lwe-arithmetics_avx2.c:164:11: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
lwe-arithmetics_avx2.c: lwe-arithmetics_avx2.c:165:14: error: always_inline function '_mm256_loadu_si256' requires target feature 'avx', but would be inlined into function 'submat_negmul' that is compiled without support for 'avx'
lwe-arithmetics_avx2.c: b[0] = _mm256_loadu_si256((__m256i*)(p));
lwe-arithmetics_avx2.c: ^
lwe-arithmetics_avx2.c: lwe-arithmetics_avx2.c:165:14: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
lwe-arithmetics_avx2.c: lwe-arithmetics_avx2.c:166:14: error: always_inline function '_mm256_loadu_si256' requires target feature 'avx', but would be inlined into function 'submat_negmul' that is compiled without support for 'avx'
lwe-arithmetics_avx2.c: ...

Number of similar (compiler,implementation) pairs: 5, 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
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

lwe-arithmetics_avx2.c: lwe-arithmetics_avx2.c: In function 'submat_negmul':
lwe-arithmetics_avx2.c: lwe-arithmetics_avx2.c:158:10: warning: AVX vector return without AVX enabled changes the ABI [-Wpsabi]
lwe-arithmetics_avx2.c: 158 | c[0] = _mm256_setzero_si256();
lwe-arithmetics_avx2.c: | ~~~~~^~~~~~~~~~~~~~~~~~~~~~~~
lwe-arithmetics_avx2.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/11/include/immintrin.h:43,
lwe-arithmetics_avx2.c: from lwe-arithmetics_avx2.c:10:
lwe-arithmetics_avx2.c: /usr/lib/gcc/x86_64-linux-gnu/11/include/avxintrin.h:1242:1: error: inlining failed in call to 'always_inline' '_mm256_setzero_si256': target specific option mismatch
lwe-arithmetics_avx2.c: 1242 | _mm256_setzero_si256 (void)
lwe-arithmetics_avx2.c: | ^~~~~~~~~~~~~~~~~~~~
lwe-arithmetics_avx2.c: lwe-arithmetics_avx2.c:161:12: note: called from here
lwe-arithmetics_avx2.c: 161 | c[3] = _mm256_setzero_si256();
lwe-arithmetics_avx2.c: | ^~~~~~~~~~~~~~~~~~~~~~
lwe-arithmetics_avx2.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/11/include/immintrin.h:43,
lwe-arithmetics_avx2.c: from lwe-arithmetics_avx2.c:10:
lwe-arithmetics_avx2.c: /usr/lib/gcc/x86_64-linux-gnu/11/include/avxintrin.h:1242:1: error: inlining failed in call to 'always_inline' '_mm256_setzero_si256': target specific option mismatch
lwe-arithmetics_avx2.c: 1242 | _mm256_setzero_si256 (void)
lwe-arithmetics_avx2.c: | ^~~~~~~~~~~~~~~~~~~~
lwe-arithmetics_avx2.c: lwe-arithmetics_avx2.c:160:12: note: called from here
lwe-arithmetics_avx2.c: 160 | c[2] = _mm256_setzero_si256();
lwe-arithmetics_avx2.c: | ^~~~~~~~~~~~~~~~~~~~~~
lwe-arithmetics_avx2.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/11/include/immintrin.h:43,
lwe-arithmetics_avx2.c: from lwe-arithmetics_avx2.c:10:
lwe-arithmetics_avx2.c: /usr/lib/gcc/x86_64-linux-gnu/11/include/avxintrin.h:1242:1: error: inlining failed in call to 'always_inline' '_mm256_setzero_si256': target specific option mismatch
lwe-arithmetics_avx2.c: 1242 | _mm256_setzero_si256 (void)
lwe-arithmetics_avx2.c: | ^~~~~~~~~~~~~~~~~~~~
lwe-arithmetics_avx2.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