Implementation notes: amd64, pluton1mn, crypto_hash/bmw256

Computer: pluton1mn
Architecture: amd64
CPU ID: GenuineIntel-00050671-bfebfbff
SUPERCOP version: 20160806
Operation: crypto_hash
Primitive: bmw256

Time	Implementation	Compiler	Benchmark date	SUPERCOP version
14518	`core_i7_45nm/x86-64/sse4_v1`	`icc -xMIC-AVX512 -O2 -fomit-frame-pointer`	20160806	20160731
14518	`core_i7_45nm/x86-64/sse4_v1`	`icc -xMIC-AVX512 -O3 -fomit-frame-pointer`	20160806	20160731
14616	`core_2_45nm/x86-64/sse4_v1`	`icc -xMIC-AVX512 -O2 -fomit-frame-pointer`	20160806	20160731
14616	`core_2_45nm/x86-64/sse4_v1`	`icc -xMIC-AVX512 -O3 -fomit-frame-pointer`	20160806	20160731
15302	`core_2_65nm/x86-64/ssse3_v1`	`icc -xMIC-AVX512 -O2 -fomit-frame-pointer`	20160806	20160731
15302	`core_2_65nm/x86-64/ssse3_v1`	`icc -xMIC-AVX512 -O3 -fomit-frame-pointer`	20160806	20160731
15316	`amd_k10/x86-64/sse2`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160806	20160731
15372	`amd_k10/x86-64/sse2`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160806	20160731
15372	`opt29_v5_ssse3_asm64`	`icc -xMIC-AVX512 -O3 -fomit-frame-pointer`	20160806	20160731
15386	`opt29_v5_ssse3_asm64`	`icc -xMIC-AVX512 -O2 -fomit-frame-pointer`	20160806	20160731
15442	`amd_k10/x86-64/sse2`	`icc -xMIC-AVX512 -O2 -fomit-frame-pointer`	20160806	20160731
15456	`amd_k10/x86-64/sse2`	`icc -xMIC-AVX512 -O3 -fomit-frame-pointer`	20160806	20160731
15484	`amd_k10/x86-64/sse2`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160806	20160731
15526	`amd_k10/x86-64/sse2`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160806	20160731
15932	`core_i7_45nm/x86-64/sse4_v1`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160806	20160731
15960	`core_i7_45nm/x86-64/sse4_v1`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160806	20160731
15960	`core_i7_45nm/x86-64/sse4_v1`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160806	20160731
16016	`core_i7_45nm/x86-64/sse4_v1`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160806	20160731
16282	`core_2_45nm/x86-64/sse4_v1`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160806	20160731
16310	`core_2_45nm/x86-64/sse4_v1`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160806	20160731
16366	`core_2_45nm/x86-64/sse4_v1`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160806	20160731
16548	`core_2_65nm/x86-64/ssse3_v1`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160806	20160731
16590	`core_2_65nm/x86-64/ssse3_v1`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160806	20160731
16618	`core_2_65nm/x86-64/ssse3_v1`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160806	20160731
16716	`core_2_65nm/x86-64/ssse3_v1`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160806	20160731
16744	`opt29_v5_ssse3_asm64`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160806	20160731
16744	`opt29_v5_ssse3_asm64`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160806	20160731
16758	`opt29_v5_ssse3_asm64`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160806	20160731
16828	`opt29_v5_ssse3_asm64`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160806	20160731
16926	`opt31ssse3_asm64`	`icc -xMIC-AVX512 -O2 -fomit-frame-pointer`	20160806	20160731
16926	`opt31ssse3_asm64`	`icc -xMIC-AVX512 -O3 -fomit-frame-pointer`	20160806	20160731
17066	`optabe`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160806	20160731
17080	`optbbe`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160806	20160731
17360	`optc03`	`icc -xMIC-AVX512 -O3 -fomit-frame-pointer`	20160806	20160731
17388	`optc03`	`icc -xMIC-AVX512 -O2 -fomit-frame-pointer`	20160806	20160731
17416	`optabe`	`icc -xMIC-AVX512 -O3 -fomit-frame-pointer`	20160806	20160731
17444	`optbbe`	`icc -xMIC-AVX512 -O3 -fomit-frame-pointer`	20160806	20160731
17458	`optc01`	`icc -xMIC-AVX512 -O2 -fomit-frame-pointer`	20160806	20160731
17486	`optc02`	`icc -xMIC-AVX512 -O3 -fomit-frame-pointer`	20160806	20160731
17486	`optc04`	`icc -xMIC-AVX512 -O3 -fomit-frame-pointer`	20160806	20160731
17570	`optc04`	`icc -xMIC-AVX512 -O2 -fomit-frame-pointer`	20160806	20160731
17598	`optc04`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160806	20160731
17696	`optc01`	`icc -xMIC-AVX512 -O3 -fomit-frame-pointer`	20160806	20160731
17724	`optc04`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160806	20160731
17780	`optc01`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160806	20160731
17864	`optc02`	`icc -xMIC-AVX512 -O2 -fomit-frame-pointer`	20160806	20160731
17892	`optabe`	`icc -xMIC-AVX512 -O2 -fomit-frame-pointer`	20160806	20160731
17920	`optbbe`	`icc -xMIC-AVX512 -O2 -fomit-frame-pointer`	20160806	20160731
17948	`optbbe`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160806	20160731
17962	`optc04`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160806	20160731
17976	`optbbe`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160806	20160731
17990	`optabe`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160806	20160731
17990	`optc02`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160806	20160731
18004	`opt31ssse3_asm64`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160806	20160731
18018	`opt31ssse3_asm64`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160806	20160731
18018	`opt31ssse3_asm64`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160806	20160731
18018	`optabe`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160806	20160731
18032	`optc03`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160806	20160731
18102	`optc02`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160806	20160731
18116	`optc02`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160806	20160731
18144	`opt31ssse3_asm64`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160806	20160731
18158	`optc01`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160806	20160731
18172	`optc01`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160806	20160731
18186	`optc03`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160806	20160731
18200	`optc03`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160806	20160731
18214	`optbbe`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160806	20160731
18214	`optc01`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160806	20160731
18228	`optabe`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160806	20160731
18284	`optbgbe`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160806	20160731
18368	`optbgbe`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160806	20160731
18620	`optbgbe`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160806	20160731
18928	`optbgbe`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160806	20160731
18928	`sphlib-small`	`icc -xMIC-AVX512 -O2 -fomit-frame-pointer`	20160806	20160731
18956	`optbgbe`	`icc -xMIC-AVX512 -O2 -fomit-frame-pointer`	20160806	20160731
19124	`optbgbe`	`icc -xMIC-AVX512 -O3 -fomit-frame-pointer`	20160806	20160731
19138	`sphlib-small`	`icc -xMIC-AVX512 -O3 -fomit-frame-pointer`	20160806	20160731
20132	`sphlib`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160806	20160731
20258	`sphlib`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160806	20160731
20356	`sphlib-small`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160806	20160731
20552	`sphlib`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160806	20160731
20678	`sphlib`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160806	20160731
20902	`sphlib-small`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160806	20160731
20902	`optc04`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160806	20160731
21098	`sphlib-small`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160806	20160731
21294	`optc03`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160806	20160731
21518	`sphlib-small`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160806	20160731
21938	`optc02`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160806	20160731
22120	`amd_k8/x86-64/noSSE`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160806	20160731
22120	`core_2_45nm/x86-64/sse4_v1`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160806	20160731
22120	`amd_k8/x86-64/noSSE`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160806	20160731
22134	`amd_k8/x86-64/noSSE`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160806	20160731
22134	`amd_k8/x86-64/noSSE`	`icc -xMIC-AVX512 -O2 -fomit-frame-pointer`	20160806	20160731
22134	`amd_k8/x86-64/noSSE`	`icc -xMIC-AVX512 -O3 -fomit-frame-pointer`	20160806	20160731
22204	`amd_k8/x86-64/noSSE`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160806	20160731
23968	`sphlib`	`icc -xMIC-AVX512 -O3 -fomit-frame-pointer`	20160806	20160731
24066	`sphlib`	`icc -xMIC-AVX512 -O2 -fomit-frame-pointer`	20160806	20160731
24108	`ref`	`icc -xMIC-AVX512 -O2 -fomit-frame-pointer`	20160806	20160731
24500	`ref`	`icc -xMIC-AVX512 -O3 -fomit-frame-pointer`	20160806	20160731
38290	`ref`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160806	20160731
38388	`ref`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160806	20160731
39074	`ref`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160806	20160731
78358	`ref`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160806	20160731

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 [-Wincompatible-pointer-types]
Compress256_sse_31.c: _mm_store_si128(&p256[00], copyTemp);
Compress256_sse_31.c: ^
Compress256_sse_31.c: In file included from /usr/local/gcc-6.1.0/lib/gcc/x86_64-pc-linux-gnu/6.1.0/include/pmmintrin.h:31:0,
Compress256_sse_31.c: from /usr/local/gcc-6.1.0/lib/gcc/x86_64-pc-linux-gnu/6.1.0/include/tmmintrin.h:31,
Compress256_sse_31.c: from Compress256_sse_31.c:12:
Compress256_sse_31.c: /usr/local/gcc-6.1.0/lib/gcc/x86_64-pc-linux-gnu/6.1.0/include/emmintrin.h:708:1: note: expected '__vector(2) long long int *' but argument is of type 'u_int32_t * {aka unsigned int *}'
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 [-Wincompatible-pointer-types]
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/local/gcc-6.1.0/lib/gcc/x86_64-pc-linux-gnu/6.1.0/include/pmmintrin.h:31:0,
Compress256_sse_31.c: from /usr/local/gcc-6.1.0/lib/gcc/x86_64-pc-linux-gnu/6.1.0/include/tmmintrin.h:31,
Compress256_sse_31.c: from Compress256_sse_31.c:12:
Compress256_sse_31.c: /usr/local/gcc-6.1.0/lib/gcc/x86_64-pc-linux-gnu/6.1.0/include/emmintrin.h:690:1: note: expected 'const __vector(2) long long int *' but argument is of type 'u_int32_t * {aka unsigned int *}'
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 [-Wincompatible-pointer-types]
Compress256_sse_31.c: _mm_store_si128(&p256[04], copyTemp);
Compress256_sse_31.c: ^
Compress256_sse_31.c: In file included from /usr/local/gcc-6.1.0/lib/gcc/x86_64-pc-linux-gnu/6.1.0/include/pmmintrin.h:31:0,
Compress256_sse_31.c: from /usr/local/gcc-6.1.0/lib/gcc/x86_64-pc-linux-gnu/6.1.0/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:39:8: warning: type defaults to 'int' in declaration of 'Compress256' [-Wimplicit-int]
wrapper.c: extern Compress256(u_int32_t *data32, u_int32_t *data32_end, hashState *state);
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 [-Wincompatible-pointer-types]
wrapper.c: Compress256(data32, data32 +16, CONST32final-4); // TODO: This is a HACK...
wrapper.c: ^~~~~~~~~~~~
wrapper.c: wrapper.c:39:8: note: expected 'hashState * {aka struct gt; *}' but argument is of type 'u_int32_t * {aka unsigned int *}'
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

Compiler output

Implementation: crypto_hash/bmw256/optc31ssse3
Compiler: icc -xMIC-AVX512 -O2 -fomit-frame-pointer

Compress256_sse_31.c: Compress256_sse_31.c(374): warning #167: argument of type "u_int32_t={unsigned int} *" is incompatible with parameter of type "union __m128i *"
Compress256_sse_31.c: _mm_store_si128(&p256[00], copyTemp);
Compress256_sse_31.c: ^
Compress256_sse_31.c:
Compress256_sse_31.c: Compress256_sse_31.c(377): warning #167: argument of type "u_int32_t={unsigned int} *" is incompatible with parameter of type "const union __m128i *"
Compress256_sse_31.c: copyTemp = _mm_load_si128(&hashState256(state)->gt;DoublePipe[ 4]);
Compress256_sse_31.c: ^
Compress256_sse_31.c:
Compress256_sse_31.c: Compress256_sse_31.c(378): warning #167: argument of type "u_int32_t={unsigned int} *" is incompatible with parameter of type "union __m128i *"
Compress256_sse_31.c: _mm_store_si128(&p256[04], copyTemp);
Compress256_sse_31.c: ^
Compress256_sse_31.c:
Compress256_sse_31.c: Compress256_sse_31.c(379): warning #167: argument of type "u_int32_t={unsigned int} *" is incompatible with parameter of type "const union __m128i *"
Compress256_sse_31.c: copyTemp = _mm_load_si128(&hashState256(state)->gt;DoublePipe[ 8]);
Compress256_sse_31.c: ^
Compress256_sse_31.c:
Compress256_sse_31.c: Compress256_sse_31.c(380): warning #167: argument of type "u_int32_t={unsigned int} *" is incompatible with parameter of type "union __m128i *"
Compress256_sse_31.c: _mm_store_si128(&p256[ 8], copyTemp);
Compress256_sse_31.c: ^
Compress256_sse_31.c:
Compress256_sse_31.c: Compress256_sse_31.c(381): warning #167: argument of type "u_int32_t={unsigned int} *" is incompatible with parameter of type "const union __m128i *"
Compress256_sse_31.c: copyTemp = _mm_load_si128(&hashState256(state)->gt;DoublePipe[12]);
Compress256_sse_31.c: ^
Compress256_sse_31.c:
Compress256_sse_31.c: Compress256_sse_31.c(382): warning #167: argument of type "u_int32_t={unsigned int} *" is incompatible with parameter of type "union __m128i *"
Compress256_sse_31.c: ...

Number of similar (compiler,implementation) pairs: 2, namely:

Compiler	Implementations
icc -xMIC-AVX512 -O2 -fomit-frame-pointer	optc31ssse3
icc -xMIC-AVX512 -O3 -fomit-frame-pointer	optc31ssse3

Compiler output

Implementation: crypto_hash/bmw256/opt31ssse3_asm64
Compiler: icc -xMIC-AVX512 -O2 -fomit-frame-pointer

wrapper.c: wrapper.c(37): warning #47: incompatible redefinition of macro "crypto_hash_BYTES" (declared at line 7 of "crypto_hash.h")
wrapper.c: #define crypto_hash_BYTES 32
wrapper.c: ^
wrapper.c:
wrapper.c: wrapper.c(238): warning #167: argument of type "u_int32_t={unsigned int} *" is incompatible with parameter of type "hashState *"
wrapper.c: Compress256(data32, data32 +16, CONST32final-4); // TODO: This is a HACK...
wrapper.c: ^
wrapper.c:

Number of similar (compiler,implementation) pairs: 2, namely:

Compiler	Implementations
icc -xMIC-AVX512 -O2 -fomit-frame-pointer	opt31ssse3_asm64
icc -xMIC-AVX512 -O3 -fomit-frame-pointer	opt31ssse3_asm64