Implementation notes: amd64, nucnuc, crypto_core/scale3sntrup653

Computer: nucnuc
Microarchitecture: amd64; Airmont (406c3)
Architecture: amd64
CPU ID: GenuineIntel-000406c3-bfebfbff
SUPERCOP version: 20230530
Operation: crypto_core
Primitive: scale3sntrup653

Time	Object size	Test size	Implementation	Compiler	Benchmark date	SUPERCOP version
1446	1151 0 0	15000 812 920	`ref`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20221010	20221005
1468	1144 0 0	15448 812 920	`ref`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20221010	20221005
1510	443 0 0	13096 812 920	`ref`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20221010	20221005
1516	457 0 0	14024 780 984	`ref`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20221010	20221005
6414	198 0 0	12592 780 984	`ref`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20221010	20221005
6840	186 0 0	12111 772 984	`ref`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20221010	20221005
7835	188 0 0	11798 804 920	`ref`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20221010	20221005
7836	179 0 0	11302 804 920	`ref`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20221010	20221005
8777	188 0 0	10927 764 952	`ref`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20221010	20221005

Compiler output

Implementation: avx
Security model: constbranchindex
Compiler: clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE

core.c: core.c:20:18: error: always_inline function '_mm256_loadu_si256' requires target feature 'avx', but would be inlined into function 'crypto_core_scale3sntrup653_avx_constbranchindex' that is compiled without support for 'avx'
core.c: __m256i save = _mm256_loadu_si256((__m256i *) (inbytes+2*i));
core.c: ^
core.c: core.c:20:18: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
core.c: core.c:25:19: error: always_inline function '_mm256_loadu_si256' requires target feature 'avx', but would be inlined into function 'crypto_core_scale3sntrup653_avx_constbranchindex' that is compiled without support for 'avx'
core.c: __m256i x = _mm256_loadu_si256((__m256i *) inbytes);
core.c: ^
core.c: core.c:25:19: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
core.c: core.c:27:32: error: always_inline function '_mm256_set1_epi16' requires target feature 'avx', but would be inlined into function 'crypto_core_scale3sntrup653_avx_constbranchindex' that is compiled without support for 'avx'
core.c: x = _mm256_mullo_epi16(x,_mm256_set1_epi16(3));
core.c: ^
core.c: core.c:27:32: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
core.c: core.c:27:11: error: always_inline function '_mm256_mullo_epi16' requires target feature 'avx2', but would be inlined into function 'crypto_core_scale3sntrup653_avx_constbranchindex' that is compiled without support for 'avx2'
core.c: x = _mm256_mullo_epi16(x,_mm256_set1_epi16(3));
core.c: ^
core.c: core.c:27:11: error: AVX vector argument of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
core.c: core.c:28:30: error: always_inline function '_mm256_set1_epi16' requires target feature 'avx', but would be inlined into function 'crypto_core_scale3sntrup653_avx_constbranchindex' that is compiled without support for 'avx'
core.c: x = _mm256_sub_epi16(x,_mm256_set1_epi16((q+1)/2));
core.c: ^
core.c: core.c:28:30: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
core.c: core.c:28:11: error: always_inline function '_mm256_sub_epi16' requires target feature 'avx2', but would be inlined into function 'crypto_core_scale3sntrup653_avx_constbranchindex' that is compiled without support for 'avx2'
core.c: x = _mm256_sub_epi16(x,_mm256_set1_epi16((q+1)/2));
core.c: ^
core.c: core.c:28:11: error: AVX vector argument of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
core.c: core.c:29:14: error: always_inline function '_mm256_srai_epi16' requires target feature 'avx2', but would be inlined into function 'crypto_core_scale3sntrup653_avx_constbranchindex' that is compiled without support for 'avx2'
core.c: ...

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

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

Compiler output

Implementation: avx
Security model: constbranchindex
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE

core.c: core.c: In function 'crypto_core_scale3sntrup653_avx_constbranchindex':
core.c: core.c:20:11: warning: AVX vector return without AVX enabled changes the ABI [-Wpsabi]
core.c: 20 | __m256i save = _mm256_loadu_si256((__m256i *) (inbytes+2*i));
core.c: | ^~~~
core.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/10/include/immintrin.h:51,
core.c: from core.c:1:
core.c: /usr/lib/gcc/x86_64-linux-gnu/10/include/avxintrin.h:920:1: error: inlining failed in call to 'always_inline' '_mm256_loadu_si256': target specific option mismatch
core.c: 920 | _mm256_loadu_si256 (__m256i_u const *__P)
core.c: | ^~~~~~~~~~~~~~~~~~
core.c: core.c:20:18: note: called from here
core.c: 20 | __m256i save = _mm256_loadu_si256((__m256i *) (inbytes+2*i));
core.c: | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
core.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/10/include/immintrin.h:51,
core.c: from core.c:1:
core.c: /usr/lib/gcc/x86_64-linux-gnu/10/include/avxintrin.h:926:1: error: inlining failed in call to 'always_inline' '_mm256_storeu_si256': target specific option mismatch
core.c: 926 | _mm256_storeu_si256 (__m256i_u *__P, __m256i __A)
core.c: | ^~~~~~~~~~~~~~~~~~~
core.c: core.c:34:7: note: called from here
core.c: 34 | _mm256_storeu_si256((__m256i *) outbytes,x);
core.c: | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
core.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/10/include/immintrin.h:53,
core.c: from core.c:1:
core.c: /usr/lib/gcc/x86_64-linux-gnu/10/include/avx2intrin.h:808:1: error: inlining failed in call to 'always_inline' '_mm256_sub_epi16': target specific option mismatch
core.c: 808 | _mm256_sub_epi16 (__m256i __A, __m256i __B)
core.c: | ^~~~~~~~~~~~~~~~
core.c: ...

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

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