Implementation notes: x86, thoth, crypto_aead/deoxyseq128128v1

Computer: thoth
Architecture: x86
CPU ID: AuthenticAMD-00000622-0183f9ff
SUPERCOP version: 20160806
Operation: crypto_aead
Primitive: deoxyseq128128v1

Time	Implementation	Compiler	Benchmark date	SUPERCOP version
3244185	`ref`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160726	20160724
3456777	`ref`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160726	20160724
4978075	`ref`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160726	20160724
7439292	`ref`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160726	20160724
7604298	`ref`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160726	20160724

Compiler output

Implementation: crypto_aead/deoxyseq128128v1/opt
Compiler: clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments

deoxys.c: deoxys.c:77:34: error: always_inline function '_mm_set_epi8' requires target feature 'sse2', but would be inlined into function 'XLS' that is compiled without support for 'sse2'
deoxys.c: const __m128i STAGE_MASK = constant8( 0x0f ,0xff,0xff,0xff, 0xff,0xff,0xff,0xff, 0xff,0xff,0xff,0xff, 0xff,0xff,0xff,0xff );
deoxys.c: ^
deoxys.c: ./tweakable-cipher.macros:12:76: note: expanded from macro 'constant8'
deoxys.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.c: ^
deoxys.c: deoxys.c:78:34: error: always_inline function '_mm_set_epi8' requires target feature 'sse2', but would be inlined into function 'XLS' that is compiled without support for 'sse2'
deoxys.c: const __m128i BLOCK_MASK = constant8( 0xff ,0xff,0xff,0xff, 0xff,0xff,0xff,0xff, 0xf0,0x00,0x00,0x00, 0x00,0x00,0x00,0x00 );
deoxys.c: ^
deoxys.c: ./tweakable-cipher.macros:12:76: note: expanded from macro 'constant8'
deoxys.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.c: ^
deoxys.c: deoxys.c:79:30: error: always_inline function '_mm_set_epi8' requires target feature 'sse2', but would be inlined into function 'XLS' that is compiled without support for 'sse2'
deoxys.c: const __m128i MSB_XLS1 = constant8( (0x8 deoxys.c: ^
deoxys.c: ./tweakable-cipher.macros:12:76: note: expanded from macro 'constant8'
deoxys.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.c: ^
deoxys.c: deoxys.c:80:30: error: always_inline function '_mm_set_epi8' requires target feature 'sse2', but would be inlined into function 'XLS' that is compiled without support for 'sse2'
deoxys.c: const __m128i MSB_XLS2 = constant8( (0x9 deoxys.c: ^
deoxys.c: ./tweakable-cipher.macros:12:76: note: expanded from macro 'constant8'
deoxys.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.c: ^
deoxys.c: deoxys.c:81:36: error: always_inline function '_mm_set_epi8' requires target feature 'sse2', but would be inlined into function 'XLS' that is compiled without support for 'sse2'
deoxys.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/deoxyseq128128v1/opt
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv

deoxys.c: deoxys.c: In function 'XLS':
deoxys.c: deoxys.c:77:17: warning: SSE vector return without SSE enabled changes the ABI [-Wpsabi]
deoxys.c: const __m128i STAGE_MASK = constant8( 0x0f ,0xff,0xff,0xff, 0xff,0xff,0xff,0xff, 0xff,0xff,0xff,0xff, 0xff,0xff,0xff,0xff );
deoxys.c: ^
deoxys.c: In file included from /usr/lib/gcc/i686-linux-gnu/5/include/xmmintrin.h:1249:0,
deoxys.c: from deoxys.c:26:
deoxys.c: /usr/lib/gcc/i686-linux-gnu/5/include/emmintrin.h:690:1: error: inlining failed in call to always_inline '_mm_load_si128': target specific option mismatch
deoxys.c: _mm_load_si128 (__m128i const *__P)
deoxys.c: ^
deoxys.c: deoxys.c:95:9: error: called from here
deoxys.c: State = load( (__m128i *) ( M1 ) );
deoxys.c: ^
deoxys.c: In file included from deoxys.c:28:0:
deoxys.c: /usr/lib/gcc/i686-linux-gnu/5/include/wmmintrin.h:77:1: error: inlining failed in call to always_inline '_mm_aesimc_si128': target specific option mismatch
deoxys.c: _mm_aesimc_si128 (__m128i __X)
deoxys.c: ^
deoxys.c: deoxys.c:94:10: error: called from here
deoxys.c: dTweak = mc_inv(Tweak);
deoxys.c: ^
deoxys.c: In file included from /usr/lib/gcc/i686-linux-gnu/5/include/xmmintrin.h:1249:0,
deoxys.c: from deoxys.c:26:
deoxys.c: /usr/lib/gcc/i686-linux-gnu/5/include/emmintrin.h:1286:1: error: inlining failed in call to always_inline '_mm_xor_si128': target specific option mismatch
deoxys.c: _mm_xor_si128 (__m128i __A, __m128i __B)
deoxys.c: ^
deoxys.c: In file included from deoxys.c:31:0:
deoxys.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