Implementation notes: x86, thoth, crypto_aead/deoxysneq256128v1

Computer: thoth
Architecture: x86
CPU ID: AuthenticAMD-00000622-0183f9ff
SUPERCOP version: 20160806
Operation: crypto_aead
Primitive: deoxysneq256128v1

Time	Implementation	Compiler	Benchmark date	SUPERCOP version
3891721	`ref`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160726	20160724
4032078	`ref`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160726	20160724
5975322	`ref`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160726	20160724
8537085	`ref`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160726	20160724
8582084	`ref`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160726	20160724

Compiler output

Implementation: crypto_aead/deoxysneq256128v1/opt
Compiler: clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments

deoxys-256.c: deoxys-256.c:42:29: error: always_inline function '_mm_set_epi8' requires target feature 'sse2', but would be inlined into function 'deoxys_aead_encrypt' that is compiled without support for 'sse2'
deoxys-256.c: const __m128i EIGHT = constant8( 0,0,0,0, 0,0,0,0, 0,0,0,0, 0,0,0,8);
deoxys-256.c: ^
deoxys-256.c: ./tweakable-cipher.macros:12:76: note: expanded from macro 'constant8'
deoxys-256.c: #define constant8(b15,b14,b13,b12,b11,b10,b9,b8,b7,b6,b5,b4,b3,b2,b1,b0) _mm_set_epi8 ((b0),(b1),(b2),(b3),(b4),(b5),(b6),(b7),(b8),(b9),(b10),(b11),(b12),(b13),(b14),(b15))
deoxys-256.c: ^
deoxys-256.c: deoxys-256.c:43:29: error: always_inline function '_mm_set_epi8' requires target feature 'sse2', but would be inlined into function 'deoxys_aead_encrypt' that is compiled without support for 'sse2'
deoxys-256.c: const __m128i SEIGHT = constant8( 0,0,0,0, 0,0,0,0, 8,0,0,0, 0,0,0,0);
deoxys-256.c: ^
deoxys-256.c: ./tweakable-cipher.macros:12:76: note: expanded from macro 'constant8'
deoxys-256.c: #define constant8(b15,b14,b13,b12,b11,b10,b9,b8,b7,b6,b5,b4,b3,b2,b1,b0) _mm_set_epi8 ((b0),(b1),(b2),(b3),(b4),(b5),(b6),(b7),(b8),(b9),(b10),(b11),(b12),(b13),(b14),(b15))
deoxys-256.c: ^
deoxys-256.c: deoxys-256.c:44:36: error: always_inline function '_mm_set_epi8' requires target feature 'sse2', but would be inlined into function 'deoxys_aead_encrypt' that is compiled without support for 'sse2'
deoxys-256.c: const __m128i STAGE_MASK = constant8( 0x1f ,0xff,0xff,0xff, 0xff,0xff,0xff,0xff, 0xff,0xff,0xff,0xff, 0xff,0xff,0xff,0xff );
deoxys-256.c: ^
deoxys-256.c: ./tweakable-cipher.macros:12:76: note: expanded from macro 'constant8'
deoxys-256.c: #define constant8(b15,b14,b13,b12,b11,b10,b9,b8,b7,b6,b5,b4,b3,b2,b1,b0) _mm_set_epi8 ((b0),(b1),(b2),(b3),(b4),(b5),(b6),(b7),(b8),(b9),(b10),(b11),(b12),(b13),(b14),(b15))
deoxys-256.c: ^
deoxys-256.c: deoxys-256.c:45:36: error: always_inline function '_mm_set_epi8' requires target feature 'sse2', but would be inlined into function 'deoxys_aead_encrypt' that is compiled without support for 'sse2'
deoxys-256.c: const __m128i BLOCK_MASK = constant8( 0xff ,0xff,0xff,0xff, 0xff,0xff,0xff,0xff, 0xe0,0x00,0x00,0x00, 0x00,0x00,0x00,0x00 );
deoxys-256.c: ^
deoxys-256.c: ./tweakable-cipher.macros:12:76: note: expanded from macro 'constant8'
deoxys-256.c: #define constant8(b15,b14,b13,b12,b11,b10,b9,b8,b7,b6,b5,b4,b3,b2,b1,b0) _mm_set_epi8 ((b0),(b1),(b2),(b3),(b4),(b5),(b6),(b7),(b8),(b9),(b10),(b11),(b12),(b13),(b14),(b15))
deoxys-256.c: ^
deoxys-256.c: deoxys-256.c:46:32: error: always_inline function '_mm_set_epi8' requires target feature 'sse2', but would be inlined into function 'deoxys_aead_encrypt' that is compiled without support for 'sse2'
deoxys-256.c: ...

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

Compiler	Implementations
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments	opt

Compiler output

Implementation: crypto_aead/deoxysneq256128v1/opt
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv

deoxys-256.c: deoxys-256.c: In function 'deoxys_aead_encrypt':
deoxys-256.c: deoxys-256.c:42:19: warning: SSE vector return without SSE enabled changes the ABI [-Wpsabi]
deoxys-256.c: const __m128i EIGHT = constant8( 0,0,0,0, 0,0,0,0, 0,0,0,0, 0,0,0,8);
deoxys-256.c: ^
deoxys-256.c: In file included from /usr/lib/gcc/i686-linux-gnu/5/include/xmmintrin.h:1249:0,
deoxys-256.c: from deoxys-256.c:26:
deoxys-256.c: /usr/lib/gcc/i686-linux-gnu/5/include/emmintrin.h:611:1: error: inlining failed in call to always_inline '_mm_set_epi8': target specific option mismatch
deoxys-256.c: _mm_set_epi8 (char __q15, char __q14, char __q13, char __q12,
deoxys-256.c: ^
deoxys-256.c: deoxys-256.c:79:13: error: called from here
deoxys-256.c: __m128i kill_shift = constant8( 0xfe,0xfe,0xfe,0xfe,0xfe,0xfe,0xfe,0xfe,0xfe,0xfe,0xfe,0xfe,0xfe,0xfe,0xfe,0xfe);
deoxys-256.c: ^
deoxys-256.c: In file included from /usr/lib/gcc/i686-linux-gnu/5/include/xmmintrin.h:1249:0,
deoxys-256.c: from deoxys-256.c:26:
deoxys-256.c: /usr/lib/gcc/i686-linux-gnu/5/include/emmintrin.h:611:1: error: inlining failed in call to always_inline '_mm_set_epi8': target specific option mismatch
deoxys-256.c: _mm_set_epi8 (char __q15, char __q14, char __q13, char __q12,
deoxys-256.c: ^
deoxys-256.c: deoxys-256.c:78:13: error: called from here
deoxys-256.c: __m128i simply_1b = constant8( 0x1b,0x1b,0x1b,0x1b,0x1b,0x1b,0x1b,0x1b,0x1b,0x1b,0x1b,0x1b,0x1b,0x1b,0x1b,0x1b);
deoxys-256.c: ^
deoxys-256.c: In file included from /usr/lib/gcc/i686-linux-gnu/5/include/xmmintrin.h:1249:0,
deoxys-256.c: from deoxys-256.c:26:
deoxys-256.c: /usr/lib/gcc/i686-linux-gnu/5/include/emmintrin.h:611:1: error: inlining failed in call to always_inline '_mm_set_epi8': target specific option mismatch
deoxys-256.c: _mm_set_epi8 (char __q15, char __q14, char __q13, char __q12,
deoxys-256.c: ^
deoxys-256.c: ...

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

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