Implementation notes: amd64, margaux, crypto_encode/1277x2627round

Computer: margaux
Microarchitecture: amd64; Core 2 65nm (6fb)
Architecture: amd64
CPU ID: GenuineIntel-000006fb-bfebfbff
SUPERCOP version: 20240107
Operation: crypto_encode
Primitive: 1277x2627round

Time	Object size	Test size	Implementation	Compiler	Benchmark date	SUPERCOP version
7284	479 0 0	14387 844 896	`ref`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231213	20231212
7341	495 0 0	14611 844 896	`ref`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231213	20231212
7345	495 0 0	13499 844 896	`ref`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231213	20231212
8124	541 0 0	13973 804 960	`ref`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
9406	201 0 0	12100 796 960	`ref`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
9414	207 0 0	12509 804 960	`ref`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212
9442	117 0 0	11851 844 896	`ref`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231213	20231212
9477	110 0 0	11045 836 896	`ref`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231213	20231212
10543	204 0 0	11136 780 928	`ref`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231213	20231212

Compiler output

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

encode.c: encode.c:34:9: error: always_inline function '_mm256_loadu_si256' requires target feature 'avx', but would be inlined into function 'crypto_encode_1277x2627round_avx_constbranchindex' that is compiled without support for 'avx'
encode.c: x = _mm256_loadu_si256((__m256i *) (reading+0));
encode.c: ^
encode.c: encode.c:34:9: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
encode.c: encode.c:35:10: error: always_inline function '_mm256_loadu_si256' requires target feature 'avx', but would be inlined into function 'crypto_encode_1277x2627round_avx_constbranchindex' that is compiled without support for 'avx'
encode.c: x2 = _mm256_loadu_si256((__m256i *) (reading+16));
encode.c: ^
encode.c: encode.c:35:10: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
encode.c: encode.c:36:31: error: always_inline function '_mm256_set1_epi16' requires target feature 'avx', but would be inlined into function 'crypto_encode_1277x2627round_avx_constbranchindex' that is compiled without support for 'avx'
encode.c: x = _mm256_mulhrs_epi16(x,_mm256_set1_epi16(10923));
encode.c: ^
encode.c: encode.c:36:31: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
encode.c: encode.c:36:9: error: always_inline function '_mm256_mulhrs_epi16' requires target feature 'avx2', but would be inlined into function 'crypto_encode_1277x2627round_avx_constbranchindex' that is compiled without support for 'avx2'
encode.c: x = _mm256_mulhrs_epi16(x,_mm256_set1_epi16(10923));
encode.c: ^
encode.c: encode.c:36:9: error: AVX vector argument of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
encode.c: encode.c:37:33: error: always_inline function '_mm256_set1_epi16' requires target feature 'avx', but would be inlined into function 'crypto_encode_1277x2627round_avx_constbranchindex' that is compiled without support for 'avx'
encode.c: x2 = _mm256_mulhrs_epi16(x2,_mm256_set1_epi16(10923));
encode.c: ^
encode.c: encode.c:37:33: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
encode.c: encode.c:37:10: error: always_inline function '_mm256_mulhrs_epi16' requires target feature 'avx2', but would be inlined into function 'crypto_encode_1277x2627round_avx_constbranchindex' that is compiled without support for 'avx2'
encode.c: x2 = _mm256_mulhrs_epi16(x2,_mm256_set1_epi16(10923));
encode.c: ^
encode.c: encode.c:37:10: error: AVX vector argument of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
encode.c: encode.c:38:28: error: always_inline function '_mm256_add_epi16' requires target feature 'avx2', but would be inlined into function 'crypto_encode_1277x2627round_avx_constbranchindex' that is compiled without support for 'avx2'
encode.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

encode.c: encode.c: In function 'crypto_encode_1277x2627round_avx_constbranchindex':
encode.c: encode.c:34:7: warning: AVX vector return without AVX enabled changes the ABI [-Wpsabi]
encode.c: 34 | x = _mm256_loadu_si256((__m256i *) (reading+0));
encode.c: | ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
encode.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/11/include/immintrin.h:43,
encode.c: from encode.c:3:
encode.c: /usr/lib/gcc/x86_64-linux-gnu/11/include/avxintrin.h:933:1: error: inlining failed in call to 'always_inline' '_mm256_storeu_si256': target specific option mismatch
encode.c: 933 | _mm256_storeu_si256 (__m256i_u *__P, __m256i __A)
encode.c: | ^~~~~~~~~~~~~~~~~~~
encode.c: encode.c:65:5: note: called from here
encode.c: 65 | _mm256_storeu_si256((__m256i *) out,_mm256_permute2f128_si256(x,x2,0x20));
encode.c: | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
encode.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/11/include/immintrin.h:43,
encode.c: from encode.c:3:
encode.c: /usr/lib/gcc/x86_64-linux-gnu/11/include/avxintrin.h:704:1: error: inlining failed in call to 'always_inline' '_mm256_permute2f128_si256': target specific option mismatch
encode.c: 704 | _mm256_permute2f128_si256 (__m256i __X, __m256i __Y, const int __C)
encode.c: | ^~~~~~~~~~~~~~~~~~~~~~~~~
encode.c: encode.c:65:5: note: called from here
encode.c: 65 | _mm256_storeu_si256((__m256i *) out,_mm256_permute2f128_si256(x,x2,0x20));
encode.c: | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
encode.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/11/include/immintrin.h:43,
encode.c: from encode.c:3:
encode.c: /usr/lib/gcc/x86_64-linux-gnu/11/include/avxintrin.h:933:1: error: inlining failed in call to 'always_inline' '_mm256_storeu_si256': target specific option mismatch
encode.c: 933 | _mm256_storeu_si256 (__m256i_u *__P, __m256i __A)
encode.c: | ^~~~~~~~~~~~~~~~~~~
encode.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