Implementation notes: amd64, waldorf, crypto_hash/bmw256

Computer: waldorf
Architecture: amd64
CPU ID: GenuineIntel-000106e5-bfebfbff
SUPERCOP version: 20160715
Operation: crypto_hash
Primitive: bmw256

Time	Implementation	Compiler	Benchmark date	SUPERCOP version
14704	`opt29_v5_ssse3_asm64`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
15440	`opt29_v5_ssse3_asm64`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
15516	`core_i7_45nm/x86-64/sse4_v1`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
15764	`core_2_45nm/x86-64/sse4_v1`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
15832	`core_i7_45nm/x86-64/sse4_v1`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
15832	`core_i7_45nm/x86-64/sse4_v1`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
15912	`core_i7_45nm/x86-64/sse4_v1`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
16268	`opt29_v5_ssse3_asm64`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
16376	`amd_k10/x86-64/sse2`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
16456	`core_2_65nm/x86-64/ssse3_v1`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
16480	`opt29_v5_ssse3_asm64`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
16568	`core_2_45nm/x86-64/sse4_v1`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
16772	`opt31ssse3_asm64`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
16864	`core_2_65nm/x86-64/ssse3_v1`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
16896	`core_2_45nm/x86-64/sse4_v1`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
16928	`core_2_45nm/x86-64/sse4_v1`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
16988	`core_i7_45nm/x86-64/sse4_v1`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
17272	`amd_k10/x86-64/sse2`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
17304	`core_2_65nm/x86-64/ssse3_v1`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
17316	`amd_k10/x86-64/sse2`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
17532	`core_2_65nm/x86-64/ssse3_v1`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
17584	`amd_k10/x86-64/sse2`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
17588	`core_2_65nm/x86-64/ssse3_v1`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
17636	`amd_k10/x86-64/sse2`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
17648	`core_2_45nm/x86-64/sse4_v1`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
17876	`opt29_v5_ssse3_asm64`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
18328	`opt31ssse3_asm64`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
19280	`opt31ssse3_asm64`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
19752	`opt31ssse3_asm64`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
20252	`opt31ssse3_asm64`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
23580	`amd_k8/x86-64/noSSE`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
24140	`amd_k8/x86-64/noSSE`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
25080	`amd_k8/x86-64/noSSE`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
26268	`amd_k8/x86-64/noSSE`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
26436	`amd_k8/x86-64/noSSE`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
26704	`optc03`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
26960	`optc04`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
27376	`optbbe`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
27404	`optc01`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
27532	`optabe`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
27600	`optc01`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
27796	`optc04`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
28188	`optc02`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
28324	`optabe`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
28612	`optabe`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
28692	`optbbe`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
29268	`optc02`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
29300	`optabe`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
29540	`optabe`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
29696	`optbbe`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
29968	`optbbe`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
30204	`optc02`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
30316	`sphlib`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
30360	`optc03`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
30572	`sphlib`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
30772	`optc01`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
30852	`optc04`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
30920	`optbgbe`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
30924	`optbbe`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
30936	`optc03`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
31000	`optc03`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
31276	`optc04`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
31344	`optc01`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
31508	`optc03`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
31660	`sphlib`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
31912	`optbgbe`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
31992	`optbgbe`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
32000	`optc02`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
32004	`sphlib-small`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
32404	`sphlib-small`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
32456	`optbgbe`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
32576	`optbgbe`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
33040	`sphlib-small`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
33136	`optc02`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
33544	`sphlib`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
34796	`sphlib`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
34988	`optc01`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
35632	`optc04`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
35956	`sphlib-small`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
38240	`sphlib-small`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
48420	`ref`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
48568	`ref`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
51760	`ref`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
55192	`ref`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
61696	`ref`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715

Compiler output

Implementation: crypto_hash/bmw256/optc31ssse3
Compiler: clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments

Compress256_sse_31.c: Compress256_sse_31.c:374:18: warning: incompatible pointer types passing 'u_int32_t *' (aka 'unsigned int *') to parameter of type '__m128i *' [-Wincompatible-pointer-types]
Compress256_sse_31.c: _mm_store_si128(&p256[00], copyTemp);
Compress256_sse_31.c: ^~~~~~~~~
Compress256_sse_31.c: /usr/include/clang/3.5.0/include/emmintrin.h:1177:26: note: passing argument to parameter '__p' here
Compress256_sse_31.c: _mm_store_si128(__m128i *__p, __m128i __b)
Compress256_sse_31.c: ^
Compress256_sse_31.c: Compress256_sse_31.c:377:28: warning: incompatible pointer types passing 'u_int32_t *' (aka 'unsigned int *') to parameter of type 'const __m128i *' [-Wincompatible-pointer-types]
Compress256_sse_31.c: copyTemp = _mm_load_si128(&hashState256(state)->gt;DoublePipe[ 4]);
Compress256_sse_31.c: ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Compress256_sse_31.c: /usr/include/clang/3.5.0/include/emmintrin.h:1063:31: note: passing argument to parameter '__p' here
Compress256_sse_31.c: _mm_load_si128(__m128i const *__p)
Compress256_sse_31.c: ^
Compress256_sse_31.c: Compress256_sse_31.c:378:18: warning: incompatible pointer types passing 'u_int32_t *' (aka 'unsigned int *') to parameter of type '__m128i *' [-Wincompatible-pointer-types]
Compress256_sse_31.c: _mm_store_si128(&p256[04], copyTemp);
Compress256_sse_31.c: ^~~~~~~~~
Compress256_sse_31.c: /usr/include/clang/3.5.0/include/emmintrin.h:1177:26: note: passing argument to parameter '__p' here
Compress256_sse_31.c: _mm_store_si128(__m128i *__p, __m128i __b)
Compress256_sse_31.c: ^
Compress256_sse_31.c: Compress256_sse_31.c:379:28: warning: incompatible pointer types passing 'u_int32_t *' (aka 'unsigned int *') to parameter of type 'const __m128i *' [-Wincompatible-pointer-types]
Compress256_sse_31.c: copyTemp = _mm_load_si128(&hashState256(state)->gt;DoublePipe[ 8]);
Compress256_sse_31.c: ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Compress256_sse_31.c: /usr/include/clang/3.5.0/include/emmintrin.h:1063:31: note: passing argument to parameter '__p' here
Compress256_sse_31.c: _mm_load_si128(__m128i const *__p)
Compress256_sse_31.c: ^
Compress256_sse_31.c: Compress256_sse_31.c:380:18: warning: incompatible pointer types passing 'u_int32_t *' (aka 'unsigned int *') to parameter of type '__m128i *' [-Wincompatible-pointer-types]
Compress256_sse_31.c: ...

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

Compiler	Implementations
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments	optc31ssse3

Compiler output

Implementation: crypto_hash/bmw256/opt31ssse3_asm64
Compiler: clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments

wrapper.c: wrapper.c:37:9: warning: 'crypto_hash_BYTES' macro redefined [-Wmacro-redefined]
wrapper.c: #define crypto_hash_BYTES 32
wrapper.c: ^
wrapper.c: ./crypto_hash.h:7:9: note: previous definition is here
wrapper.c: #define crypto_hash_BYTES crypto_hash_bmw256_BYTES
wrapper.c: ^
wrapper.c: wrapper.c:39:8: warning: type specifier missing, defaults to 'int' [-Wimplicit-int]
wrapper.c: extern Compress256(u_int32_t *data32, u_int32_t *data32_end, hashState *state);
wrapper.c: ~~~~~~ ^
wrapper.c: wrapper.c:238:34: warning: incompatible pointer types passing 'u_int32_t *' (aka 'unsigned int *') to parameter of type 'hashState *' [-Wincompatible-pointer-types]
wrapper.c: Compress256(data32, data32 +16, CONST32final-4); // TODO: This is a HACK...
wrapper.c: ^~~~~~~~~~~~~~
wrapper.c: wrapper.c:39:73: note: passing argument to parameter 'state' here
wrapper.c: extern Compress256(u_int32_t *data32, u_int32_t *data32_end, hashState *state);
wrapper.c: ^
wrapper.c: 3 warnings generated.

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

Compiler	Implementations
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments	opt31ssse3_asm64

Compiler output

Implementation: crypto_hash/bmw256/optc31ssse3
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv

Compress256_sse_31.c: Compress256_sse_31.c: In function 'Compress256':
Compress256_sse_31.c: Compress256_sse_31.c:374:18: warning: passing argument 1 of '_mm_store_si128' from incompatible pointer type
Compress256_sse_31.c: _mm_store_si128(&p256[00], copyTemp);
Compress256_sse_31.c: ^
Compress256_sse_31.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/4.9/include/pmmintrin.h:31:0,
Compress256_sse_31.c: from /usr/lib/gcc/x86_64-linux-gnu/4.9/include/tmmintrin.h:31,
Compress256_sse_31.c: from Compress256_sse_31.c:12:
Compress256_sse_31.c: /usr/lib/gcc/x86_64-linux-gnu/4.9/include/emmintrin.h:704:1: note: expected '__vector(2) long long int *' but argument is of type 'u_int32_t *'
Compress256_sse_31.c: _mm_store_si128 (__m128i *__P, __m128i __B)
Compress256_sse_31.c: ^
Compress256_sse_31.c: Compress256_sse_31.c:377:28: warning: passing argument 1 of '_mm_load_si128' from incompatible pointer type
Compress256_sse_31.c: copyTemp = _mm_load_si128(&hashState256(state)->gt;DoublePipe[ 4]);
Compress256_sse_31.c: ^
Compress256_sse_31.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/4.9/include/pmmintrin.h:31:0,
Compress256_sse_31.c: from /usr/lib/gcc/x86_64-linux-gnu/4.9/include/tmmintrin.h:31,
Compress256_sse_31.c: from Compress256_sse_31.c:12:
Compress256_sse_31.c: /usr/lib/gcc/x86_64-linux-gnu/4.9/include/emmintrin.h:686:1: note: expected 'const __vector(2) long long int *' but argument is of type 'u_int32_t *'
Compress256_sse_31.c: _mm_load_si128 (__m128i const *__P)
Compress256_sse_31.c: ^
Compress256_sse_31.c: Compress256_sse_31.c:378:18: warning: passing argument 1 of '_mm_store_si128' from incompatible pointer type
Compress256_sse_31.c: _mm_store_si128(&p256[04], copyTemp);
Compress256_sse_31.c: ^
Compress256_sse_31.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/4.9/include/pmmintrin.h:31:0,
Compress256_sse_31.c: from /usr/lib/gcc/x86_64-linux-gnu/4.9/include/tmmintrin.h:31,
Compress256_sse_31.c: from Compress256_sse_31.c:12:
Compress256_sse_31.c: ...

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

Compiler	Implementations
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv	optc31ssse3
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv	optc31ssse3
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv	optc31ssse3
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv	optc31ssse3

Compiler output

Implementation: crypto_hash/bmw256/opt31ssse3_asm64
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv

wrapper.c: wrapper.c:37:0: warning: "crypto_hash_BYTES" redefined
wrapper.c: #define crypto_hash_BYTES 32
wrapper.c: ^
wrapper.c: In file included from wrapper.c:20:0:
wrapper.c: crypto_hash.h:7:0: note: this is the location of the previous definition
wrapper.c: #define crypto_hash_BYTES crypto_hash_bmw256_BYTES
wrapper.c: ^
wrapper.c: wrapper.c: In function 'crypto_hash_bmw256_opt31ssse3_asm64':
wrapper.c: wrapper.c:238:34: warning: passing argument 3 of 'Compress256' from incompatible pointer type
wrapper.c: Compress256(data32, data32 +16, CONST32final-4); // TODO: This is a HACK...
wrapper.c: ^
wrapper.c: wrapper.c:39:8: note: expected 'struct hashState *' but argument is of type 'u_int32_t *'
wrapper.c: extern Compress256(u_int32_t *data32, u_int32_t *data32_end, hashState *state);
wrapper.c: ^

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

Compiler	Implementations
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv	opt31ssse3_asm64
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv	opt31ssse3_asm64
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv	opt31ssse3_asm64
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv	opt31ssse3_asm64