Implementation notes: amd64, slide, crypto_hash/blake512

Computer: slide
Architecture: amd64
CPU ID: GenuineIntel-00040651-bfebfbff
SUPERCOP version: 20160806
Operation: crypto_hash
Primitive: blake512

Time	Implementation	Compiler	Benchmark date	SUPERCOP version
8904	`avxicc`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
9360	`vect128`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
9788	`sse41`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
9876	`sse41`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
9988	`avxicc`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
10024	`avxicc`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
10104	`sse41`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
10308	`avxicc`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
10432	`sse41`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
10520	`vect128`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
10592	`vect128`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
10600	`sphlib`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
10684	`vect128`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
10708	`sphlib`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
10852	`bswap`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
10868	`vect128-inplace`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
10900	`bswap`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
10912	`bswap`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
10920	`sandy`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
10960	`bswap`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
10964	`vect128-inplace`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
10984	`vect128-inplace`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
10984	`sphlib`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
11252	`vect128-inplace`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
11352	`sphlib`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
11580	`sandy`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
11608	`sandy`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
11760	`sandy`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
11952	`regs`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
12048	`regs`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
12072	`regs`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
12084	`sphlib-small`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
12148	`regs`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
12387	`sphlib-small`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
13252	`sse2`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
13312	`sse2`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
13344	`sse2`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
13608	`sphlib-small`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
13724	`sse2`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
13977	`ref`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
14232	`sphlib-small`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
14628	`ssse3`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
14712	`ssse3`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
14744	`ssse3`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
14756	`ssse3`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
14952	`ref`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
15420	`ref`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
15476	`sse2s`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160906	20160806
15504	`sse2s`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160906	20160806
15524	`sse2s`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806
15528	`sse2s`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160906	20160806
15800	`ref`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160906	20160806

Compiler output

Implementation: crypto_hash/blake512/xop-2
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv

hash.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/4.9/include/x86intrin.h:52:0,
hash.c: from hash.c:5:
hash.c: hash.c: In function 'blake512_compress':
hash.c: /usr/lib/gcc/x86_64-linux-gnu/4.9/include/xopintrin.h:212:1: error: inlining failed in call to always_inline '_mm_perm_epi8': target specific option mismatch
hash.c: _mm_perm_epi8(__m128i __A, __m128i __B, __m128i __C)
hash.c: ^
hash.c: In file included from hash.c:8:0:
hash.c: rounds.h:15:21: error: called from here
hash.c: #define BSWAP64(x) _mm_perm_epi8((x),(x),u8to64)
hash.c: ^
hash.c: hash.c:92:15: note: in expansion of macro 'BSWAP64'
hash.c: m.u128[0] = BSWAP64(m.u128[0]);
hash.c: ^
hash.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/4.9/include/x86intrin.h:52:0,
hash.c: from hash.c:5:
hash.c: /usr/lib/gcc/x86_64-linux-gnu/4.9/include/xopintrin.h:212:1: error: inlining failed in call to always_inline '_mm_perm_epi8': target specific option mismatch
hash.c: _mm_perm_epi8(__m128i __A, __m128i __B, __m128i __C)
hash.c: ^
hash.c: In file included from hash.c:8:0:
hash.c: rounds.h:15:21: error: called from here
hash.c: #define BSWAP64(x) _mm_perm_epi8((x),(x),u8to64)
hash.c: ^
hash.c: hash.c:93:15: note: in expansion of macro 'BSWAP64'
hash.c: m.u128[1] = BSWAP64(m.u128[1]);
hash.c: ^
hash.c: ...

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

Compiler	Implementations
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv	xop-2
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv	xop-2
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv	xop-2
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv	xop-2

Compiler output

Implementation: crypto_hash/blake512/xop
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv

hash.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/4.9/include/x86intrin.h:52:0,
hash.c: from hash.c:5:
hash.c: hash.c: In function 'blake512_compress':
hash.c: /usr/lib/gcc/x86_64-linux-gnu/4.9/include/xopintrin.h:212:1: error: inlining failed in call to always_inline '_mm_perm_epi8': target specific option mismatch
hash.c: _mm_perm_epi8(__m128i __A, __m128i __B, __m128i __C)
hash.c: ^
hash.c: hash.c:81:6: error: called from here
hash.c: m0 = BSWAP64(m0);
hash.c: ^
hash.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/4.9/include/x86intrin.h:52:0,
hash.c: from hash.c:5:
hash.c: /usr/lib/gcc/x86_64-linux-gnu/4.9/include/xopintrin.h:212:1: error: inlining failed in call to always_inline '_mm_perm_epi8': target specific option mismatch
hash.c: _mm_perm_epi8(__m128i __A, __m128i __B, __m128i __C)
hash.c: ^
hash.c: hash.c:82:6: error: called from here
hash.c: m1 = BSWAP64(m1);
hash.c: ^
hash.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/4.9/include/x86intrin.h:52:0,
hash.c: from hash.c:5:
hash.c: /usr/lib/gcc/x86_64-linux-gnu/4.9/include/xopintrin.h:212:1: error: inlining failed in call to always_inline '_mm_perm_epi8': target specific option mismatch
hash.c: _mm_perm_epi8(__m128i __A, __m128i __B, __m128i __C)
hash.c: ^
hash.c: hash.c:83:6: error: called from here
hash.c: m2 = BSWAP64(m2);
hash.c: ^
hash.c: ...

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

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

Compiler output

Implementation: crypto_hash/blake512/vect128-xop
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv

vector.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/4.9/include/x86intrin.h:52:0,
vector.c: from vector.h:29,
vector.c: from vector.c:7:
vector.c: vector.c: In function 'round512':
vector.c: /usr/lib/gcc/x86_64-linux-gnu/4.9/include/xopintrin.h:212:1: error: inlining failed in call to always_inline '_mm_perm_epi8': target specific option mismatch
vector.c: _mm_perm_epi8(__m128i __A, __m128i __B, __m128i __C)
vector.c: ^
vector.c: vector.c:646:7: error: called from here
vector.c: v64 mm0 = v64_lswap(MM[0]), mm1 = v64_lswap(MM[1]);
vector.c: ^
vector.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/4.9/include/x86intrin.h:52:0,
vector.c: from vector.h:29,
vector.c: from vector.c:7:
vector.c: /usr/lib/gcc/x86_64-linux-gnu/4.9/include/xopintrin.h:212:1: error: inlining failed in call to always_inline '_mm_perm_epi8': target specific option mismatch
vector.c: _mm_perm_epi8(__m128i __A, __m128i __B, __m128i __C)
vector.c: ^
vector.c: vector.c:646:31: error: called from here
vector.c: v64 mm0 = v64_lswap(MM[0]), mm1 = v64_lswap(MM[1]);
vector.c: ^
vector.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/4.9/include/x86intrin.h:52:0,
vector.c: from vector.h:29,
vector.c: from vector.c:7:
vector.c: /usr/lib/gcc/x86_64-linux-gnu/4.9/include/xopintrin.h:212:1: error: inlining failed in call to always_inline '_mm_perm_epi8': target specific option mismatch
vector.c: _mm_perm_epi8(__m128i __A, __m128i __B, __m128i __C)
vector.c: ^
vector.c: ...

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

Compiler	Implementations
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv	vect128-xop
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv	vect128-xop
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv	vect128-xop
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv	vect128-xop