Implementation notes: amd64, par, crypto_hash/bmw512

Computer: par
Architecture: amd64
CPU ID: GenuineIntel-000406c3-bfebfbff
SUPERCOP version: 20161026
Operation: crypto_hash
Primitive: bmw512

Time	Implementation	Compiler	Benchmark date	SUPERCOP version
13040	`opt008_asm64`	`gcc -march=native -mcpu=native -O3`	20161214	20161026
13060	`opt008_asm64`	`gcc -march=native -mcpu=native -O2`	20161214	20161026
13080	`opt008_asm64`	`gcc -funroll-loops -march=native -mcpu=native -O2`	20161214	20161026
13080	`opt008_asm64`	`gcc -funroll-loops -march=native -mcpu=native -O3`	20161214	20161026
13280	`opt008_asm64`	`gcc -march=native -mcpu=native -Os`	20161214	20161026
13300	`opt008_asm64`	`gcc -funroll-loops -march=native -mcpu=native -Os`	20161214	20161026
13360	`optc01`	`gcc -funroll-loops -march=native -mcpu=native -Os`	20161214	20161026
13400	`optc02`	`gcc -funroll-loops -march=native -mcpu=native -Os`	20161214	20161026
13600	`optbbe`	`gcc -funroll-loops -march=native -mcpu=native -Os`	20161214	20161026
13660	`optabe`	`gcc -funroll-loops -march=native -mcpu=native -Os`	20161214	20161026
13660	`optc02`	`gcc -march=native -mcpu=native -Os`	20161214	20161026
13680	`optc04`	`gcc -funroll-loops -march=native -mcpu=native -O2`	20161214	20161026
13680	`optc04`	`gcc -funroll-loops -march=native -mcpu=native -O3`	20161214	20161026
13740	`optc01`	`gcc -march=native -mcpu=native -Os`	20161214	20161026
13780	`optabe`	`gcc -march=native -mcpu=native -Os`	20161214	20161026
13780	`optbbe`	`gcc -march=native -mcpu=native -Os`	20161214	20161026
13860	`optc03`	`gcc -funroll-loops -march=native -mcpu=native -O2`	20161214	20161026
13980	`optc03`	`gcc -funroll-loops -march=native -mcpu=native -O3`	20161214	20161026
14000	`optc02`	`gcc -funroll-loops -march=native -mcpu=native -O3`	20161214	20161026
14020	`optc01`	`gcc -funroll-loops -march=native -mcpu=native -O3`	20161214	20161026
14080	`optc04`	`gcc -march=native -mcpu=native -O2`	20161214	20161026
14120	`optc02`	`gcc -funroll-loops -march=native -mcpu=native -O2`	20161214	20161026
14120	`optc04`	`gcc -march=native -mcpu=native -O3`	20161214	20161026
14160	`optc01`	`gcc -funroll-loops -march=native -mcpu=native -O2`	20161214	20161026
14220	`optabe`	`gcc -funroll-loops -march=native -mcpu=native -O2`	20161214	20161026
14220	`optbbe`	`gcc -funroll-loops -march=native -mcpu=native -O2`	20161214	20161026
14220	`optabe`	`gcc -funroll-loops -march=native -mcpu=native -O3`	20161214	20161026
14220	`optbbe`	`gcc -funroll-loops -march=native -mcpu=native -O3`	20161214	20161026
14300	`optc03`	`gcc -march=native -mcpu=native -O2`	20161214	20161026
14320	`optc03`	`gcc -march=native -mcpu=native -O3`	20161214	20161026
14560	`optc02`	`gcc -march=native -mcpu=native -O3`	20161214	20161026
14580	`optbbe`	`gcc -march=native -mcpu=native -O2`	20161214	20161026
14580	`optc01`	`gcc -march=native -mcpu=native -O2`	20161214	20161026
14580	`optc01`	`gcc -march=native -mcpu=native -O3`	20161214	20161026
14600	`optc02`	`gcc -march=native -mcpu=native -O2`	20161214	20161026
14620	`optabe`	`gcc -march=native -mcpu=native -O2`	20161214	20161026
14620	`optabe`	`gcc -march=native -mcpu=native -O3`	20161214	20161026
14660	`optbgbe`	`gcc -funroll-loops -march=native -mcpu=native -O2`	20161214	20161026
14660	`optbgbe`	`gcc -funroll-loops -march=native -mcpu=native -O3`	20161214	20161026
14660	`optbbe`	`gcc -march=native -mcpu=native -O3`	20161214	20161026
14940	`optbgbe`	`gcc -march=native -mcpu=native -O3`	20161214	20161026
15020	`optbgbe`	`gcc -march=native -mcpu=native -O2`	20161214	20161026
15080	`optbgbe`	`gcc -march=native -mcpu=native -Os`	20161214	20161026
15240	`sphlib`	`gcc -funroll-loops -march=native -mcpu=native -Os`	20161214	20161026
15300	`optbgbe`	`gcc -funroll-loops -march=native -mcpu=native -Os`	20161214	20161026
15320	`sphlib`	`gcc -funroll-loops -march=native -mcpu=native -O2`	20161214	20161026
15320	`sphlib`	`gcc -funroll-loops -march=native -mcpu=native -O3`	20161214	20161026
15740	`optc03`	`gcc -funroll-loops -march=native -mcpu=native -Os`	20161214	20161026
15740	`optc03`	`gcc -march=native -mcpu=native -Os`	20161214	20161026
15800	`sphlib`	`gcc -march=native -mcpu=native -Os`	20161214	20161026
15960	`sphlib`	`gcc -march=native -mcpu=native -O3`	20161214	20161026
16000	`optc04`	`gcc -funroll-loops -march=native -mcpu=native -Os`	20161214	20161026
16020	`sphlib`	`gcc -march=native -mcpu=native -O2`	20161214	20161026
16400	`optc04`	`gcc -march=native -mcpu=native -Os`	20161214	20161026
20600	`sphlib-small`	`gcc -funroll-loops -march=native -mcpu=native -Os`	20161214	20161026
21120	`sphlib-small`	`gcc -funroll-loops -march=native -mcpu=native -O2`	20161214	20161026
21360	`sphlib-small`	`gcc -march=native -mcpu=native -O2`	20161214	20161026
21680	`sphlib-small`	`gcc -march=native -mcpu=native -Os`	20161214	20161026
24320	`sphlib-small`	`gcc -march=native -mcpu=native -O3`	20161214	20161026
24800	`sphlib-small`	`gcc -funroll-loops -march=native -mcpu=native -O3`	20161214	20161026
26860	`ref`	`gcc -funroll-loops -march=native -mcpu=native -O3`	20161214	20161026
26940	`ref`	`gcc -funroll-loops -march=native -mcpu=native -O2`	20161214	20161026
27900	`ref`	`gcc -march=native -mcpu=native -O3`	20161214	20161026
28240	`ref`	`gcc -march=native -mcpu=native -O2`	20161214	20161026
53360	`ref`	`gcc -funroll-loops -march=native -mcpu=native -Os`	20161214	20161026
55100	`ref`	`gcc -march=native -mcpu=native -Os`	20161214	20161026

Compiler output

Implementation: crypto_hash/bmw512/optx86sse2
Compiler: gcc -funroll-loops -march=native -mcpu=native -O2

BlueMidnightWish.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
BlueMidnightWish.c: BlueMidnightWish.c: In function 'Compress512':
BlueMidnightWish.c: BlueMidnightWish.c:390:44: warning: passing argument 1 of '_mm_load_si128' from incompatible pointer type [-Wincompatible-pointer-types]
BlueMidnightWish.c: _mm_store_si128(&p512[00], _mm_load_si128(&hashState512(state)->gt;DoublePipe[0]));
BlueMidnightWish.c: ^
BlueMidnightWish.c: In file included from BlueMidnightWish.c:12:0:
BlueMidnightWish.c: /usr/lib/gcc/x86_64-pc-linux-gnu/6.2.1/include/emmintrin.h:690:1: note: expected 'const __vector(2) long long int *' but argument is of type 'u_int64_t * {aka long unsigned int *}'
BlueMidnightWish.c: _mm_load_si128 (__m128i const *__P)
BlueMidnightWish.c: ^~~~~~~~~~~~~~
BlueMidnightWish.c: BlueMidnightWish.c:390:18: warning: passing argument 1 of '_mm_store_si128' from incompatible pointer type [-Wincompatible-pointer-types]
BlueMidnightWish.c: _mm_store_si128(&p512[00], _mm_load_si128(&hashState512(state)->gt;DoublePipe[0]));
BlueMidnightWish.c: ^
BlueMidnightWish.c: In file included from BlueMidnightWish.c:12:0:
BlueMidnightWish.c: /usr/lib/gcc/x86_64-pc-linux-gnu/6.2.1/include/emmintrin.h:708:1: note: expected '__vector(2) long long int *' but argument is of type 'u_int64_t * {aka long unsigned int *}'
BlueMidnightWish.c: _mm_store_si128 (__m128i *__P, __m128i __B)
BlueMidnightWish.c: ^~~~~~~~~~~~~~~
BlueMidnightWish.c: BlueMidnightWish.c:391:44: warning: passing argument 1 of '_mm_load_si128' from incompatible pointer type [-Wincompatible-pointer-types]
BlueMidnightWish.c: _mm_store_si128(&p512[02], _mm_load_si128(&hashState512(state)->gt;DoublePipe[2]));
BlueMidnightWish.c: ^
BlueMidnightWish.c: In file included from BlueMidnightWish.c:12:0:
BlueMidnightWish.c: /usr/lib/gcc/x86_64-pc-linux-gnu/6.2.1/include/emmintrin.h:690:1: note: expected 'const __vector(2) long long int *' but argument is of type 'u_int64_t * {aka long unsigned int *}'
BlueMidnightWish.c: _mm_load_si128 (__m128i const *__P)
BlueMidnightWish.c: ^~~~~~~~~~~~~~
BlueMidnightWish.c: BlueMidnightWish.c:391:18: warning: passing argument 1 of '_mm_store_si128' from incompatible pointer type [-Wincompatible-pointer-types]
BlueMidnightWish.c: _mm_store_si128(&p512[02], _mm_load_si128(&hashState512(state)->gt;DoublePipe[2]));
BlueMidnightWish.c: ...

Number of similar (compiler,implementation) pairs: 6, namely:

Compiler	Implementations
gcc -funroll-loops -march=native -mcpu=native -O2	optx86sse2
gcc -funroll-loops -march=native -mcpu=native -O3	optx86sse2
gcc -funroll-loops -march=native -mcpu=native -Os	optx86sse2
gcc -march=native -mcpu=native -O2	optx86sse2
gcc -march=native -mcpu=native -O3	optx86sse2
gcc -march=native -mcpu=native -Os	optx86sse2

Compiler output

Implementation: crypto_hash/bmw512/optc01
Compiler: gcc -funroll-loops -march=native -mcpu=native -O2

BlueMidnightWish.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
hash.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
try.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
try.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
measure.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead

Number of similar (compiler,implementation) pairs: 30, namely:

Compiler	Implementations
gcc -funroll-loops -march=native -mcpu=native -O2	optc01 optc02 optc03 optc04 ref
gcc -funroll-loops -march=native -mcpu=native -O3	optc01 optc02 optc03 optc04 ref
gcc -funroll-loops -march=native -mcpu=native -Os	optc01 optc02 optc03 optc04 ref
gcc -march=native -mcpu=native -O2	optc01 optc02 optc03 optc04 ref
gcc -march=native -mcpu=native -O3	optc01 optc02 optc03 optc04 ref
gcc -march=native -mcpu=native -Os	optc01 optc02 optc03 optc04 ref

Compiler output

Implementation: crypto_hash/bmw512/sphlib
Compiler: gcc -funroll-loops -march=native -mcpu=native -O2

bmw.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
hash.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
try.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
try.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
measure.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead

Number of similar (compiler,implementation) pairs: 12, namely:

Compiler	Implementations
gcc -funroll-loops -march=native -mcpu=native -O2	sphlib sphlib-small
gcc -funroll-loops -march=native -mcpu=native -O3	sphlib sphlib-small
gcc -funroll-loops -march=native -mcpu=native -Os	sphlib sphlib-small
gcc -march=native -mcpu=native -O2	sphlib sphlib-small
gcc -march=native -mcpu=native -O3	sphlib sphlib-small
gcc -march=native -mcpu=native -Os	sphlib sphlib-small

Compiler output

Implementation: crypto_hash/bmw512/optbgbe
Compiler: gcc -funroll-loops -march=native -mcpu=native -O2

bmw_large_speed.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
hash.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
memxor.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
try.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
try.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
measure.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead

Number of similar (compiler,implementation) pairs: 6, namely:

Compiler	Implementations
gcc -funroll-loops -march=native -mcpu=native -O2	optbgbe
gcc -funroll-loops -march=native -mcpu=native -O3	optbgbe
gcc -funroll-loops -march=native -mcpu=native -Os	optbgbe
gcc -march=native -mcpu=native -O2	optbgbe
gcc -march=native -mcpu=native -O3	optbgbe
gcc -march=native -mcpu=native -Os	optbgbe

Compiler output

Implementation: crypto_hash/bmw512/optabe
Compiler: gcc -funroll-loops -march=native -mcpu=native -O2

hash.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
try.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
try.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
measure.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead

Number of similar (compiler,implementation) pairs: 12, namely:

Compiler	Implementations
gcc -funroll-loops -march=native -mcpu=native -O2	optabe optbbe
gcc -funroll-loops -march=native -mcpu=native -O3	optabe optbbe
gcc -funroll-loops -march=native -mcpu=native -Os	optabe optbbe
gcc -march=native -mcpu=native -O2	optabe optbbe
gcc -march=native -mcpu=native -O3	optabe optbbe
gcc -march=native -mcpu=native -Os	optabe optbbe

Compiler output

Implementation: crypto_hash/bmw512/opt008_asm64
Compiler: gcc -funroll-loops -march=native -mcpu=native -O2

hash.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
update.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
Compress.s: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
try.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
try.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
measure.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead

Number of similar (compiler,implementation) pairs: 6, namely:

Compiler	Implementations
gcc -funroll-loops -march=native -mcpu=native -O2	opt008_asm64
gcc -funroll-loops -march=native -mcpu=native -O3	opt008_asm64
gcc -funroll-loops -march=native -mcpu=native -Os	opt008_asm64
gcc -march=native -mcpu=native -O2	opt008_asm64
gcc -march=native -mcpu=native -O3	opt008_asm64
gcc -march=native -mcpu=native -Os	opt008_asm64