Implementation notes: amd64, gcc14, crypto_decode/256x2

Computer: gcc14
Architecture: amd64
CPU ID: GenuineIntel-00010676-bfebfbff
SUPERCOP version: 20220506
Operation: crypto_decode
Primitive: 256x2

Time	Object size	Test size	Implementation	Compiler	Benchmark date	SUPERCOP version
952	130 0 0	11347 820 880	`ref`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20210114	20210114
952	130 0 0	12283 820 880	`ref`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20210114	20210114
952	130 0 0	11347 820 880	`ref`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20210114	20210114
953	130 0 0	12155 820 880	`ref`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20210114	20210114
1038	91 0 0	10557 836 912	`ref`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20210114	20210114
1058	87 0 0	9649 820 912	`ref`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20210114	20210114
1069	94 0 0	10821 836 912	`ref`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20210114	20210114
1069	94 0 0	12798 844 912	`ref`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20210114	20210114
1071	87 0 0	9717 812 880	`ref`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20210114	20210114

Compiler output

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

decode.c: decode.c:16:17: error: always_inline function '_mm256_set1_epi32' requires target feature 'xsave', but would be inlined into function 'crypto_decode_256x2_avx_constbranchindex' that is compiled without support for 'xsave'
decode.c: __m256i x = _mm256_set1_epi32(*(int32_t *) s);
decode.c: ^
decode.c: decode.c:18:9: error: always_inline function '_mm256_shuffle_epi8' requires target feature 'xsave', but would be inlined into function 'crypto_decode_256x2_avx_constbranchindex' that is compiled without support for 'xsave'
decode.c: x = _mm256_shuffle_epi8(x,COPY);
decode.c: ^
decode.c: decode.c:18:31: error: always_inline function '_mm256_set_epi64x' requires target feature 'xsave', but would be inlined into function 'crypto_decode_256x2_avx_constbranchindex' that is compiled without support for 'xsave'
decode.c: x = _mm256_shuffle_epi8(x,COPY);
decode.c: ^
decode.c: decode.c:5:14: note: expanded from macro 'COPY'
decode.c: #define COPY _mm256_set_epi64x(0x0303030303030303,0x0202020202020202,0x0101010101010101,0x0000000000000000)
decode.c: ^
decode.c: decode.c:20:9: error: always_inline function '_mm256_andnot_si256' requires target feature 'xsave', but would be inlined into function 'crypto_decode_256x2_avx_constbranchindex' that is compiled without support for 'xsave'
decode.c: x = _mm256_andnot_si256(x,MASK);
decode.c: ^
decode.c: decode.c:20:31: error: always_inline function '_mm256_set1_epi64x' requires target feature 'xsave', but would be inlined into function 'crypto_decode_256x2_avx_constbranchindex' that is compiled without support for 'xsave'
decode.c: x = _mm256_andnot_si256(x,MASK);
decode.c: ^
decode.c: decode.c:6:14: note: expanded from macro 'MASK'
decode.c: #define MASK _mm256_set1_epi64x(0x8040201008040201)
decode.c: ^
decode.c: decode.c:21:9: error: always_inline function '_mm256_cmpeq_epi8' requires target feature 'xsave', but would be inlined into function 'crypto_decode_256x2_avx_constbranchindex' that is compiled without support for 'xsave'
decode.c: x = _mm256_cmpeq_epi8(x,_mm256_setzero_si256());
decode.c: ^
decode.c: decode.c:21:29: error: always_inline function '_mm256_setzero_si256' requires target feature 'xsave', but would be inlined into function 'crypto_decode_256x2_avx_constbranchindex' that is compiled without support for 'xsave'
decode.c: ...

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

Compiler output

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

decode.c: decode.c:16:17: error: always_inline function '_mm256_set1_epi32' requires target feature 'sse4.2', but would be inlined into function 'crypto_decode_256x2_avx_constbranchindex' that is compiled without support for 'sse4.2'
decode.c: __m256i x = _mm256_set1_epi32(*(int32_t *) s);
decode.c: ^
decode.c: decode.c:18:9: error: always_inline function '_mm256_shuffle_epi8' requires target feature 'sse4.2', but would be inlined into function 'crypto_decode_256x2_avx_constbranchindex' that is compiled without support for 'sse4.2'
decode.c: x = _mm256_shuffle_epi8(x,COPY);
decode.c: ^
decode.c: decode.c:18:31: error: always_inline function '_mm256_set_epi64x' requires target feature 'sse4.2', but would be inlined into function 'crypto_decode_256x2_avx_constbranchindex' that is compiled without support for 'sse4.2'
decode.c: x = _mm256_shuffle_epi8(x,COPY);
decode.c: ^
decode.c: decode.c:5:14: note: expanded from macro 'COPY'
decode.c: #define COPY _mm256_set_epi64x(0x0303030303030303,0x0202020202020202,0x0101010101010101,0x0000000000000000)
decode.c: ^
decode.c: decode.c:20:9: error: always_inline function '_mm256_andnot_si256' requires target feature 'sse4.2', but would be inlined into function 'crypto_decode_256x2_avx_constbranchindex' that is compiled without support for 'sse4.2'
decode.c: x = _mm256_andnot_si256(x,MASK);
decode.c: ^
decode.c: decode.c:20:31: error: always_inline function '_mm256_set1_epi64x' requires target feature 'sse4.2', but would be inlined into function 'crypto_decode_256x2_avx_constbranchindex' that is compiled without support for 'sse4.2'
decode.c: x = _mm256_andnot_si256(x,MASK);
decode.c: ^
decode.c: decode.c:6:14: note: expanded from macro 'MASK'
decode.c: #define MASK _mm256_set1_epi64x(0x8040201008040201)
decode.c: ^
decode.c: decode.c:21:9: error: always_inline function '_mm256_cmpeq_epi8' requires target feature 'sse4.2', but would be inlined into function 'crypto_decode_256x2_avx_constbranchindex' that is compiled without support for 'sse4.2'
decode.c: x = _mm256_cmpeq_epi8(x,_mm256_setzero_si256());
decode.c: ^
decode.c: decode.c:21:29: error: always_inline function '_mm256_setzero_si256' requires target feature 'sse4.2', but would be inlined into function 'crypto_decode_256x2_avx_constbranchindex' that is compiled without support for 'sse4.2'
decode.c: ...

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

Compiler	Implementations
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

decode.c: decode.c: In function 'crypto_decode_256x2_avx_constbranchindex':
decode.c: decode.c:16:13: warning: AVX vector return without AVX enabled changes the ABI [-Wpsabi]
decode.c: __m256i x = _mm256_set1_epi32(*(int32_t *) s);
decode.c: ^
decode.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/6/include/immintrin.h:41:0,
decode.c: from decode.c:1:
decode.c: /usr/lib/gcc/x86_64-linux-gnu/6/include/avxintrin.h:900:1: error: inlining failed in call to always_inline '_mm256_storeu_si256': target specific option mismatch
decode.c: _mm256_storeu_si256 (__m256i *__P, __m256i __A)
decode.c: ^~~~~~~~~~~~~~~~~~~
decode.c: decode.c:23:5: note: called from here
decode.c: _mm256_storeu_si256(r,x);
decode.c: ^~~~~~~~~~~~~~~~~~~~~~~~
decode.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/6/include/immintrin.h:41:0,
decode.c: from decode.c:1:
decode.c: /usr/lib/gcc/x86_64-linux-gnu/6/include/avxintrin.h:1316:1: error: inlining failed in call to always_inline '_mm256_set1_epi64x': target specific option mismatch
decode.c: _mm256_set1_epi64x (long long __A)
decode.c: ^~~~~~~~~~~~~~~~~~
decode.c: decode.c:7:15: note: called from here
decode.c: #define MASK2 _mm256_set1_epi64x(0x0101010101010101)
decode.c: ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
decode.c: decode.c:22:10: note: in expansion of macro 'MASK2'
decode.c: x &= MASK2;
decode.c: ^~~~~
decode.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/6/include/immintrin.h:43:0,
decode.c: from decode.c:1:
decode.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