popov/cpuminer-opt-gpu
1
0
Fork 0
mirror of https://github.com/JayDDee/cpuminer-opt.git synced 2025-09-17 23:44:27 +00:00
Code Issues Packages Projects Releases Wiki Activity

v3.12.1

Browse Source
This commit is contained in:
Jay D Dee
2020-02-07 20:18:20 -05:00
parent fc97ef174a
commit dc2f8d81d3
10 changed files with 110 additions and 109 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
RELEASE_NOTES
View File

@@ -65,6 +65,13 @@ If not what makes it happen or not happen?
Change Log Change Log
---------- ----------
v3.12.1
Fixed anime AVX2 low difficulty shares, git issue #236.
Periodic summary now reports lost hash rate due to rejected and stale shares,
displayed only when non-zero.
v3.12.0.1 v3.12.0.1
Fixed hodl rejects, git issue #237. Fixed hodl rejects, git issue #237.

21
algo/lyra2/allium.c
View File

@@ -76,37 +76,34 @@ int scanhash_allium( struct work *work, uint32_t max_nonce,
uint64_t *hashes_done, struct thr_info *mythr ) uint64_t *hashes_done, struct thr_info *mythr )
{ {
uint32_t _ALIGN(128) hash[8]; uint32_t _ALIGN(128) hash[8];
uint32_t _ALIGN(128) endiandata[20]; uint32_t _ALIGN(128) edata[20];
uint32_t *pdata = work->data; uint32_t *pdata = work->data;
uint32_t *ptarget = work->target; uint32_t *ptarget = work->target;
const uint32_t Htarg = ptarget[7];
const uint32_t first_nonce = pdata[19]; const uint32_t first_nonce = pdata[19];
uint32_t nonce = first_nonce; uint32_t nonce = first_nonce;
int thr_id = mythr->id; // thr_id arg is deprecated const int thr_id = mythr->id;
if ( opt_benchmark ) if ( opt_benchmark )
ptarget[7] = 0x3ffff; ptarget[7] = 0x3ffff;
for ( int i = 0; i < 19; i++ ) for ( int i = 0; i < 19; i++ )
be32enc( &endiandata[i], pdata[i] ); edata[i] = bswap_32( pdata[i] );
sph_blake256_init( &allium_ctx.blake ); sph_blake256_init( &allium_ctx.blake );
sph_blake256( &allium_ctx.blake, endiandata, 64 ); sph_blake256( &allium_ctx.blake, edata, 64 );
do { do {
be32enc( &endiandata[19], nonce ); edata[19] = nonce;
allium_hash( hash, endiandata ); allium_hash( hash, edata );
if ( hash[7] <= Htarg ) if ( valid_hash( hash, ptarget ) && !opt_benchmark )
if ( fulltest( hash, ptarget ) && !opt_benchmark )
{ {
pdata[19] = nonce; pdata[19] = bswap_32( nonce );
submit_solution( work, hash, mythr ); submit_solution( work, hash, mythr );
} }
nonce++; nonce++;
} while ( nonce < max_nonce && !work_restart[thr_id].restart ); } while ( nonce < max_nonce && !work_restart[thr_id].restart );
pdata[19] = nonce; pdata[19] = nonce;
*hashes_done = pdata[19] - first_nonce + 1; *hashes_done = pdata[19] - first_nonce;
return 0; return 0;
} }

36
algo/lyra2/lyra2rev3-4way.c
View File

@@ -130,15 +130,15 @@ int scanhash_lyra2rev3_16way( struct work *work, const uint32_t max_nonce,
{ {
uint32_t hash[8*16] __attribute__ ((aligned (128))); uint32_t hash[8*16] __attribute__ ((aligned (128)));
uint32_t vdata[20*16] __attribute__ ((aligned (64))); uint32_t vdata[20*16] __attribute__ ((aligned (64)));
uint32_t *hash7 = &hash[7<<4]; uint32_t *hash32 = &hash[7*16];
uint32_t lane_hash[8] __attribute__ ((aligned (64))); uint32_t lane_hash[8] __attribute__ ((aligned (64)));
uint32_t *pdata = work->data; uint32_t *pdata = work->data;
const uint32_t *ptarget = work->target; const uint32_t *ptarget = work->target;
const uint32_t first_nonce = pdata[19]; const uint32_t first_nonce = pdata[19];
uint32_t n = first_nonce; uint32_t n = first_nonce;
const uint32_t last_nonce = max_nonce - 16; const uint32_t last_nonce = max_nonce - 16;
const uint32_t Htarg = ptarget[7]; const uint32_t targ32 = ptarget[7];
__m512i *noncev = (__m512i*)vdata + 19; // aligned __m512i *noncev = (__m512i*)vdata + 19;
const int thr_id = mythr->id; const int thr_id = mythr->id;
if ( opt_benchmark ) ( (uint32_t*)ptarget )[7] = 0x0000ff; if ( opt_benchmark ) ( (uint32_t*)ptarget )[7] = 0x0000ff;
@@ -159,7 +159,7 @@ int scanhash_lyra2rev3_16way( struct work *work, const uint32_t max_nonce,
pdata[19] = n; pdata[19] = n;
for ( int lane = 0; lane < 16; lane++ ) for ( int lane = 0; lane < 16; lane++ )
if ( unlikely( hash7[lane] <= Htarg ) ) if ( unlikely( hash32[lane] <= targ32 ) )
{ {
extr_lane_16x32( lane_hash, hash, lane, 256 ); extr_lane_16x32( lane_hash, hash, lane, 256 );
if ( likely( fulltest( lane_hash, ptarget ) && !opt_benchmark ) ) if ( likely( fulltest( lane_hash, ptarget ) && !opt_benchmark ) )
@@ -252,15 +252,15 @@ int scanhash_lyra2rev3_8way( struct work *work, const uint32_t max_nonce,
{ {
uint32_t hash[8*8] __attribute__ ((aligned (64))); uint32_t hash[8*8] __attribute__ ((aligned (64)));
uint32_t vdata[20*8] __attribute__ ((aligned (64))); uint32_t vdata[20*8] __attribute__ ((aligned (64)));
uint32_t *hash7 = &hash[7<<3]; uint32_t *hash32 = &hash[7*8];
uint32_t lane_hash[8] __attribute__ ((aligned (32))); uint32_t lane_hash[8] __attribute__ ((aligned (32)));
uint32_t *pdata = work->data; uint32_t *pdata = work->data;
uint32_t *ptarget = work->target; uint32_t *ptarget = work->target;
const uint32_t first_nonce = pdata[19]; const uint32_t first_nonce = pdata[19];
const uint32_t last_nonce = max_nonce - 8; const uint32_t last_nonce = max_nonce - 8;
uint32_t n = first_nonce; uint32_t n = first_nonce;
const uint32_t Htarg = ptarget[7]; const uint32_t targ32 = ptarget[7];
__m256i *noncev = (__m256i*)vdata + 19; // aligned __m256i *noncev = (__m256i*)vdata + 19;
const int thr_id = mythr->id; const int thr_id = mythr->id;
const bool bench = opt_benchmark; const bool bench = opt_benchmark;
@@ -277,7 +277,7 @@ int scanhash_lyra2rev3_8way( struct work *work, const uint32_t max_nonce,
pdata[19] = n; pdata[19] = n;
for ( int lane = 0; lane < 8; lane++ ) for ( int lane = 0; lane < 8; lane++ )
if ( unlikely( hash7[lane] <= Htarg ) ) if ( unlikely( hash32[lane] <= targ32 ) )
{ {
extr_lane_8x32( lane_hash, hash, lane, 256 ); extr_lane_8x32( lane_hash, hash, lane, 256 );
if ( likely( valid_hash( lane_hash, ptarget ) && !bench ) ) if ( likely( valid_hash( lane_hash, ptarget ) && !bench ) )
@@ -357,40 +357,38 @@ int scanhash_lyra2rev3_4way( struct work *work, const uint32_t max_nonce,
{ {
uint32_t hash[8*4] __attribute__ ((aligned (64))); uint32_t hash[8*4] __attribute__ ((aligned (64)));
uint32_t vdata[20*4] __attribute__ ((aligned (64))); uint32_t vdata[20*4] __attribute__ ((aligned (64)));
uint32_t *hash7 = &(hash[7<<2]); uint32_t *hash32 = &(hash[7*4]);
uint32_t lane_hash[8] __attribute__ ((aligned (32))); uint32_t lane_hash[8] __attribute__ ((aligned (32)));
uint32_t *pdata = work->data; uint32_t *pdata = work->data;
const uint32_t *ptarget = work->target; const uint32_t *ptarget = work->target;
const uint32_t first_nonce = pdata[19]; const uint32_t first_nonce = pdata[19];
uint32_t n = first_nonce; uint32_t n = first_nonce;
const uint32_t Htarg = ptarget[7]; const uint32_t targ32 = ptarget[7];
__m128i *noncev = (__m128i*)vdata + 19; // aligned __m128i *noncev = (__m128i*)vdata + 19;
const int thr_id = mythr->id; // thr_id arg is deprecated const int thr_id = mythr->id;
if ( opt_benchmark ) if ( opt_benchmark )
( (uint32_t*)ptarget )[7] = 0x0000ff; ( (uint32_t*)ptarget )[7] = 0x0000ff;
mm128_bswap32_intrlv80_4x32( vdata, pdata ); mm128_bswap32_intrlv80_4x32( vdata, pdata );
*noncev = _mm_set_epi32( n+3, n+2, n+1, n );
blake256_4way_init( &l2v3_4way_ctx.blake ); blake256_4way_init( &l2v3_4way_ctx.blake );
blake256_4way_update( &l2v3_4way_ctx.blake, vdata, 64 ); blake256_4way_update( &l2v3_4way_ctx.blake, vdata, 64 );
do do
{ {
*noncev = mm128_bswap_32( _mm_set_epi32( n+3, n+2, n+1, n ) );
lyra2rev3_4way_hash( hash, vdata ); lyra2rev3_4way_hash( hash, vdata );
pdata[19] = n; for ( int lane = 0; lane < 4; lane++ ) if ( hash32[lane] <= targ32 )
for ( int lane = 0; lane < 4; lane++ ) if ( hash7[lane] <= Htarg )
{ {
extr_lane_4x32( lane_hash, hash, lane, 256 ); extr_lane_4x32( lane_hash, hash, lane, 256 );
if ( fulltest( lane_hash, ptarget ) && !opt_benchmark ) if ( valid_hash( lane_hash, ptarget ) && !opt_benchmark )
{ {
pdata[19] = n + lane; pdata[19] = bswap_32( n + lane );
submit_lane_solution( work, lane_hash, mythr, lane ); submit_lane_solution( work, lane_hash, mythr, lane );
} }
} }
*noncev = _mm_add_epi32( *noncev, m128_const1_32( 4 ) );
n += 4; n += 4;
} while ( (n < max_nonce-4) && !work_restart[thr_id].restart); } while ( (n < max_nonce-4) && !work_restart[thr_id].restart);
*hashes_done = n - first_nonce + 1; *hashes_done = n - first_nonce + 1;

10
algo/quark/anime-4way.c
View File

@@ -90,7 +90,7 @@ void anime_4way_hash( void *state, const void *input )
intrlv_4x64( vhashA, hash0, hash1, hash2, hash3, 512 ); intrlv_4x64( vhashA, hash0, hash1, hash2, hash3, 512 );
if ( mm256_anybits0( vh_mask ) ) if ( mm256_anybits1( vh_mask ) )
{ {
skein512_4way_update( &ctx.skein, vhash, 64 ); skein512_4way_update( &ctx.skein, vhash, 64 );
skein512_4way_close( &ctx.skein, vhashB ); skein512_4way_close( &ctx.skein, vhashB );
@@ -116,13 +116,13 @@ void anime_4way_hash( void *state, const void *input )
vh_mask = _mm256_cmpeq_epi64( _mm256_and_si256( vh[0], bit3_mask ), zero ); vh_mask = _mm256_cmpeq_epi64( _mm256_and_si256( vh[0], bit3_mask ), zero );
if ( mm256_anybits1( vh_mask ) ) if ( mm256_anybits0( vh_mask ) )
{ {
blake512_4way_init( &ctx.blake ); blake512_4way_init( &ctx.blake );
blake512_4way_update( &ctx.blake, vhash, 64 ); blake512_4way_update( &ctx.blake, vhash, 64 );
blake512_4way_close( &ctx.blake, vhashA ); blake512_4way_close( &ctx.blake, vhashA );
} }
if ( mm256_anybits0( vh_mask ) ) if ( mm256_anybits1( vh_mask ) )
{ {
bmw512_4way_init( &ctx.bmw ); bmw512_4way_init( &ctx.bmw );
bmw512_4way_update( &ctx.bmw, vhash, 64 ); bmw512_4way_update( &ctx.bmw, vhash, 64 );
@@ -140,13 +140,13 @@ void anime_4way_hash( void *state, const void *input )
vh_mask = _mm256_cmpeq_epi64( _mm256_and_si256( vh[0], bit3_mask ), zero ); vh_mask = _mm256_cmpeq_epi64( _mm256_and_si256( vh[0], bit3_mask ), zero );
if ( mm256_anybits1( vh_mask ) ) if ( mm256_anybits0( vh_mask ) )
{ {
keccak512_4way_init( &ctx.keccak ); keccak512_4way_init( &ctx.keccak );
keccak512_4way_update( &ctx.keccak, vhash, 64 ); keccak512_4way_update( &ctx.keccak, vhash, 64 );
keccak512_4way_close( &ctx.keccak, vhashA ); keccak512_4way_close( &ctx.keccak, vhashA );
} }
if ( mm256_anybits0( vh_mask ) ) if ( mm256_anybits1( vh_mask ) )
{ {
jh512_4way_init( &ctx.jh ); jh512_4way_init( &ctx.jh );
jh512_4way_update( &ctx.jh, vhash, 64 ); jh512_4way_update( &ctx.jh, vhash, 64 );

45
algo/x16/x16rv2.c
View File

@@ -34,7 +34,6 @@
#endif #endif
static __thread uint32_t s_ntime = UINT32_MAX; static __thread uint32_t s_ntime = UINT32_MAX;
static __thread char hashOrder[X16R_HASH_FUNC_COUNT + 1] = { 0 };
union _x16rv2_context_overlay union _x16rv2_context_overlay
{ {
@@ -74,16 +73,10 @@ void x16rv2_hash( void* output, const void* input )
x16rv2_context_overlay ctx; x16rv2_context_overlay ctx;
void *in = (void*) input; void *in = (void*) input;
int size = 80; int size = 80;
/*
if ( s_ntime == UINT32_MAX )
{
const uint8_t* in8 = (uint8_t*) input;
x16_r_s_getAlgoString( &in8[4], hashOrder );
}
*/
for ( int i = 0; i < 16; i++ ) for ( int i = 0; i < 16; i++ )
{ {
const char elem = hashOrder[i]; const char elem = x16r_hash_order[i];
const uint8_t algo = elem >= 'A' ? elem - 'A' + 10 : elem - '0'; const uint8_t algo = elem >= 'A' ? elem - 'A' + 10 : elem - '0';
switch ( algo ) switch ( algo )
@@ -203,42 +196,42 @@ int scanhash_x16rv2( struct work *work, uint32_t max_nonce,
uint64_t *hashes_done, struct thr_info *mythr ) uint64_t *hashes_done, struct thr_info *mythr )
{ {
uint32_t _ALIGN(128) hash32[8]; uint32_t _ALIGN(128) hash32[8];
uint32_t _ALIGN(128) endiandata[20]; uint32_t _ALIGN(128) edata[20];
uint32_t *pdata = work->data; uint32_t *pdata = work->data;
uint32_t *ptarget = work->target; uint32_t *ptarget = work->target;
const uint32_t Htarg = ptarget[7];
const uint32_t first_nonce = pdata[19]; const uint32_t first_nonce = pdata[19];
int thr_id = mythr->id; // thr_id arg is deprecated const int thr_id = mythr->id;
uint32_t nonce = first_nonce; uint32_t nonce = first_nonce;
volatile uint8_t *restart = &(work_restart[thr_id].restart); volatile uint8_t *restart = &(work_restart[thr_id].restart);
const bool bench = opt_benchmark;
casti_m128i( endiandata, 0 ) = mm128_bswap_32( casti_m128i( pdata, 0 ) ); casti_m128i( edata, 0 ) = mm128_bswap_32( casti_m128i( pdata, 0 ) );
casti_m128i( endiandata, 1 ) = mm128_bswap_32( casti_m128i( pdata, 1 ) ); casti_m128i( edata, 1 ) = mm128_bswap_32( casti_m128i( pdata, 1 ) );
casti_m128i( endiandata, 2 ) = mm128_bswap_32( casti_m128i( pdata, 2 ) ); casti_m128i( edata, 2 ) = mm128_bswap_32( casti_m128i( pdata, 2 ) );
casti_m128i( endiandata, 3 ) = mm128_bswap_32( casti_m128i( pdata, 3 ) ); casti_m128i( edata, 3 ) = mm128_bswap_32( casti_m128i( pdata, 3 ) );
casti_m128i( endiandata, 4 ) = mm128_bswap_32( casti_m128i( pdata, 4 ) ); casti_m128i( edata, 4 ) = mm128_bswap_32( casti_m128i( pdata, 4 ) );
static __thread uint32_t s_ntime = UINT32_MAX;
if ( s_ntime != pdata[17] ) if ( s_ntime != pdata[17] )
{ {
uint32_t ntime = swab32(pdata[17]); uint32_t ntime = swab32(pdata[17]);
x16_r_s_getAlgoString( (const uint8_t*) (&endiandata[1]), hashOrder ); x16_r_s_getAlgoString( (const uint8_t*) (&edata[1]), x16r_hash_order );
s_ntime = ntime; s_ntime = ntime;
if ( opt_debug && !thr_id ) if ( opt_debug && !thr_id )
applog( LOG_DEBUG, "hash order %s (%08x)", hashOrder, ntime ); applog( LOG_DEBUG, "hash order %s (%08x)",
x16r_hash_order, ntime );
} }
if ( opt_benchmark ) if ( bench ) ptarget[7] = 0x0cff;
ptarget[7] = 0x0cff;
do do
{ {
be32enc( &endiandata[19], nonce ); edata[19] = nonce;
x16rv2_hash( hash32, endiandata ); x16rv2_hash( hash32, edata );
if ( hash32[7] <= Htarg ) if ( unlikely( valid_hash( hash32, ptarget ) && !bench ) )
if (fulltest( hash32, ptarget ) && !opt_benchmark )
{ {
pdata[19] = nonce; pdata[19] = bswap_32( nonce );
submit_solution( work, hash32, mythr ); submit_solution( work, hash32, mythr );
} }
nonce++; nonce++;

22
algo/x17/x17-4way.c
View File

@@ -287,10 +287,10 @@ void x17_8way_hash( void *state, const void *input )
int scanhash_x17_8way( struct work *work, uint32_t max_nonce, int scanhash_x17_8way( struct work *work, uint32_t max_nonce,
uint64_t *hashes_done, struct thr_info *mythr ) uint64_t *hashes_done, struct thr_info *mythr )
{ {
uint32_t hash[8*16] __attribute__ ((aligned (128))); uint32_t hash[8*8] __attribute__ ((aligned (128)));
uint32_t vdata[24*8] __attribute__ ((aligned (64))); uint32_t vdata[20*8] __attribute__ ((aligned (64)));
uint32_t lane_hash[8] __attribute__ ((aligned (64))); uint32_t lane_hash[8] __attribute__ ((aligned (64)));
uint32_t *hash7 = &(hash[7<<3]); uint32_t *hash32 = &(hash[7*8]);
uint32_t *pdata = work->data; uint32_t *pdata = work->data;
const uint32_t *ptarget = work->target; const uint32_t *ptarget = work->target;
const uint32_t first_nonce = pdata[19]; const uint32_t first_nonce = pdata[19];
@@ -298,7 +298,7 @@ int scanhash_x17_8way( struct work *work, uint32_t max_nonce,
__m512i *noncev = (__m512i*)vdata + 9; // aligned __m512i *noncev = (__m512i*)vdata + 9; // aligned
uint32_t n = first_nonce; uint32_t n = first_nonce;
const int thr_id = mythr->id; const int thr_id = mythr->id;
const uint32_t Htarg = ptarget[7]; const uint32_t targ32 = ptarget[7];
const bool bench = opt_benchmark; const bool bench = opt_benchmark;
mm512_bswap32_intrlv80_8x64( vdata, pdata ); mm512_bswap32_intrlv80_8x64( vdata, pdata );
@@ -310,7 +310,7 @@ int scanhash_x17_8way( struct work *work, uint32_t max_nonce,
x17_8way_hash( hash, vdata ); x17_8way_hash( hash, vdata );
for ( int lane = 0; lane < 8; lane++ ) for ( int lane = 0; lane < 8; lane++ )
if ( unlikely( ( hash7[ lane ] <= Htarg ) && !bench ) ) if ( unlikely( ( hash32[ lane ] <= targ32 ) && !bench ) )
{ {
extr_lane_8x32( lane_hash, hash, lane, 256 ); extr_lane_8x32( lane_hash, hash, lane, 256 );
if ( likely( valid_hash( lane_hash, ptarget ) ) ) if ( likely( valid_hash( lane_hash, ptarget ) ) )
@@ -474,18 +474,18 @@ void x17_4way_hash( void *state, const void *input )
int scanhash_x17_4way( struct work *work, uint32_t max_nonce, int scanhash_x17_4way( struct work *work, uint32_t max_nonce,
uint64_t *hashes_done, struct thr_info *mythr ) uint64_t *hashes_done, struct thr_info *mythr )
{ {
uint32_t hash[16*4] __attribute__ ((aligned (64))); uint32_t hash[8*4] __attribute__ ((aligned (64)));
uint32_t vdata[20*4] __attribute__ ((aligned (64))); uint32_t vdata[20*4] __attribute__ ((aligned (64)));
uint32_t lane_hash[8] __attribute__ ((aligned (64))); uint32_t lane_hash[8] __attribute__ ((aligned (64)));
uint32_t *hash7 = &(hash[7<<2]); uint32_t *hash32 = &(hash[ 7*4 ]);
uint32_t *pdata = work->data; uint32_t *pdata = work->data;
const uint32_t *ptarget = work->target; const uint32_t *ptarget = work->target;
const uint32_t first_nonce = pdata[19]; const uint32_t first_nonce = pdata[19];
const uint32_t last_nonce = max_nonce -4; const uint32_t last_nonce = max_nonce - 4;
__m256i *noncev = (__m256i*)vdata + 9; // aligned __m256i *noncev = (__m256i*)vdata + 9;
uint32_t n = first_nonce; uint32_t n = first_nonce;
const int thr_id = mythr->id; const int thr_id = mythr->id;
const uint32_t Htarg = ptarget[7]; const uint32_t targ32 = ptarget[7];
const bool bench = opt_benchmark; const bool bench = opt_benchmark;
mm256_bswap32_intrlv80_4x64( vdata, pdata ); mm256_bswap32_intrlv80_4x64( vdata, pdata );
@@ -496,7 +496,7 @@ int scanhash_x17_4way( struct work *work, uint32_t max_nonce,
x17_4way_hash( hash, vdata ); x17_4way_hash( hash, vdata );
for ( int lane = 0; lane < 4; lane++ ) for ( int lane = 0; lane < 4; lane++ )
if ( unlikely( hash7[ lane ] <= Htarg && !bench ) ) if ( unlikely( hash32[ lane ] <= targ32 && !bench ) )
{ {
extr_lane_4x32( lane_hash, hash, lane, 256 ); extr_lane_4x32( lane_hash, hash, lane, 256 );
if ( valid_hash( lane_hash, ptarget ) ) if ( valid_hash( lane_hash, ptarget ) )

20
configure vendored
View File

@@ -1,6 +1,6 @@
#! /bin/sh #! /bin/sh
# Guess values for system-dependent variables and create Makefiles. # Guess values for system-dependent variables and create Makefiles.
# Generated by GNU Autoconf 2.69 for cpuminer-opt 3.12.0.1. # Generated by GNU Autoconf 2.69 for cpuminer-opt 3.12.1.
# #
# #
# Copyright (C) 1992-1996, 1998-2012 Free Software Foundation, Inc. # Copyright (C) 1992-1996, 1998-2012 Free Software Foundation, Inc.
@@ -577,8 +577,8 @@ MAKEFLAGS=
# Identity of this package. # Identity of this package.
PACKAGE_NAME='cpuminer-opt' PACKAGE_NAME='cpuminer-opt'
PACKAGE_TARNAME='cpuminer-opt' PACKAGE_TARNAME='cpuminer-opt'
PACKAGE_VERSION='3.12.0.1' PACKAGE_VERSION='3.12.1'
PACKAGE_STRING='cpuminer-opt 3.12.0.1' PACKAGE_STRING='cpuminer-opt 3.12.1'
PACKAGE_BUGREPORT='' PACKAGE_BUGREPORT=''
PACKAGE_URL='' PACKAGE_URL=''
@@ -1332,7 +1332,7 @@ if test "$ac_init_help" = "long"; then
# Omit some internal or obsolete options to make the list less imposing. # Omit some internal or obsolete options to make the list less imposing.
# This message is too long to be a string in the A/UX 3.1 sh. # This message is too long to be a string in the A/UX 3.1 sh.
cat <<_ACEOF cat <<_ACEOF
\`configure' configures cpuminer-opt 3.12.0.1 to adapt to many kinds of systems. \`configure' configures cpuminer-opt 3.12.1 to adapt to many kinds of systems.
Usage: $0 [OPTION]... [VAR=VALUE]... Usage: $0 [OPTION]... [VAR=VALUE]...
@@ -1404,7 +1404,7 @@ fi
if test -n "$ac_init_help"; then if test -n "$ac_init_help"; then
case $ac_init_help in case $ac_init_help in
short | recursive ) echo "Configuration of cpuminer-opt 3.12.0.1:";; short | recursive ) echo "Configuration of cpuminer-opt 3.12.1:";;
esac esac
cat <<\_ACEOF cat <<\_ACEOF
@@ -1509,7 +1509,7 @@ fi
test -n "$ac_init_help" && exit $ac_status test -n "$ac_init_help" && exit $ac_status
if $ac_init_version; then if $ac_init_version; then
cat <<\_ACEOF cat <<\_ACEOF
cpuminer-opt configure 3.12.0.1 cpuminer-opt configure 3.12.1
generated by GNU Autoconf 2.69 generated by GNU Autoconf 2.69
Copyright (C) 2012 Free Software Foundation, Inc. Copyright (C) 2012 Free Software Foundation, Inc.
@@ -2012,7 +2012,7 @@ cat >config.log <<_ACEOF
This file contains any messages produced by compilers while This file contains any messages produced by compilers while
running configure, to aid debugging if configure makes a mistake. running configure, to aid debugging if configure makes a mistake.
It was created by cpuminer-opt $as_me 3.12.0.1, which was It was created by cpuminer-opt $as_me 3.12.1, which was
generated by GNU Autoconf 2.69. Invocation command line was generated by GNU Autoconf 2.69. Invocation command line was
$ $0 $@ $ $0 $@
@@ -2993,7 +2993,7 @@ fi
# Define the identity of the package. # Define the identity of the package.
PACKAGE='cpuminer-opt' PACKAGE='cpuminer-opt'
VERSION='3.12.0.1' VERSION='3.12.1'
cat >>confdefs.h <<_ACEOF cat >>confdefs.h <<_ACEOF
@@ -6690,7 +6690,7 @@ cat >>$CONFIG_STATUS <<\_ACEOF || ac_write_fail=1
# report actual input values of CONFIG_FILES etc. instead of their # report actual input values of CONFIG_FILES etc. instead of their
# values after options handling. # values after options handling.
ac_log=" ac_log="
This file was extended by cpuminer-opt $as_me 3.12.0.1, which was This file was extended by cpuminer-opt $as_me 3.12.1, which was
generated by GNU Autoconf 2.69. Invocation command line was generated by GNU Autoconf 2.69. Invocation command line was
CONFIG_FILES = $CONFIG_FILES CONFIG_FILES = $CONFIG_FILES
@@ -6756,7 +6756,7 @@ _ACEOF
cat >>$CONFIG_STATUS <<_ACEOF || ac_write_fail=1 cat >>$CONFIG_STATUS <<_ACEOF || ac_write_fail=1
ac_cs_config="`$as_echo "$ac_configure_args" | sed 's/^ //; s/[\\""\`\$]/\\\\&/g'`" ac_cs_config="`$as_echo "$ac_configure_args" | sed 's/^ //; s/[\\""\`\$]/\\\\&/g'`"
ac_cs_version="\\ ac_cs_version="\\
cpuminer-opt config.status 3.12.0.1 cpuminer-opt config.status 3.12.1
configured by $0, generated by GNU Autoconf 2.69, configured by $0, generated by GNU Autoconf 2.69,
with options \\"\$ac_cs_config\\" with options \\"\$ac_cs_config\\"

2
configure.ac
View File

@@ -1,4 +1,4 @@
AC_INIT([cpuminer-opt], [3.12.0.1]) AC_INIT([cpuminer-opt], [3.12.1])
AC_PREREQ([2.59c]) AC_PREREQ([2.59c])
AC_CANONICAL_SYSTEM AC_CANONICAL_SYSTEM

51
cpu-miner.c
View File

@@ -883,8 +883,6 @@ static double norm_diff_sum = 0.;
static uint32_t last_block_height = 0; static uint32_t last_block_height = 0;
//static bool new_job = false; //static bool new_job = false;
static double last_targetdiff = 0.; static double last_targetdiff = 0.;
static double ref_rate_hi = 0.;
static double ref_rate_lo = 1e100;
#if !(defined(__WINDOWS__) || defined(_WIN64) || defined(_WIN32)) #if !(defined(__WINDOWS__) || defined(_WIN64) || defined(_WIN32))
static uint32_t hi_temp = 0; static uint32_t hi_temp = 0;
#endif #endif
@@ -932,7 +930,6 @@ void report_summary_log( bool force )
uint64_t accepts = accept_sum; accept_sum = 0; uint64_t accepts = accept_sum; accept_sum = 0;
uint64_t rejects = reject_sum; reject_sum = 0; uint64_t rejects = reject_sum; reject_sum = 0;
uint64_t stales = stale_sum; stale_sum = 0; uint64_t stales = stale_sum; stale_sum = 0;
// int latency = latency_sum; latency_sum = 0;
memcpy( &start_time, &five_min_start, sizeof start_time ); memcpy( &start_time, &five_min_start, sizeof start_time );
memcpy( &five_min_start, &now, sizeof now ); memcpy( &five_min_start, &now, sizeof now );
@@ -943,34 +940,21 @@ void report_summary_log( bool force )
double share_time = (double)et.tv_sec + (double)et.tv_usec / 1e6; double share_time = (double)et.tv_sec + (double)et.tv_usec / 1e6;
double ghrate = global_hashrate; double ghrate = global_hashrate;
double scaled_ghrate = ghrate;
double shrate = share_time == 0. ? 0. : diff_to_hash * last_targetdiff double shrate = share_time == 0. ? 0. : diff_to_hash * last_targetdiff
* (double)(accepts) / share_time; * (double)(accepts) / share_time;
double sess_hrate = uptime.tv_sec == 0. ? 0. : diff_to_hash * norm_diff_sum double sess_hrate = uptime.tv_sec == 0. ? 0. : diff_to_hash * norm_diff_sum
/ (double)uptime.tv_sec; / (double)uptime.tv_sec;
double scaled_shrate = shrate;
// int avg_latency = 0; double submit_rate = share_time == 0. ? 0. : (double)submits*60. / share_time;
// double latency_pc = 0.;
double submit_rate = 0.;
char shr_units[4] = {0}; char shr_units[4] = {0};
char ghr_units[4] = {0}; char ghr_units[4] = {0};
char sess_hr_units[4] = {0}; char sess_hr_units[4] = {0};
char et_str[24]; char et_str[24];
char upt_str[24]; char upt_str[24];
// if ( submits ) avg_latency = latency / submits; scale_hash_for_display( &shrate, shr_units );
scale_hash_for_display( &ghrate, ghr_units );
if ( share_time != 0. )
{
submit_rate = (double)submits*60. / share_time;
// latency_pc = (double)latency / (share_time * 10.);
}
if ( ghrate > ref_rate_hi ) ref_rate_hi = ghrate;
if ( ghrate < ref_rate_lo ) ref_rate_lo = ghrate;
scale_hash_for_display( &scaled_shrate, shr_units );
scale_hash_for_display( &scaled_ghrate, ghr_units );
scale_hash_for_display( &sess_hrate, sess_hr_units ); scale_hash_for_display( &sess_hrate, sess_hr_units );
sprintf_et( et_str, et.tv_sec ); sprintf_et( et_str, et.tv_sec );
@@ -981,8 +965,29 @@ void report_summary_log( bool force )
submit_rate, (double)submitted_share_count*60. / submit_rate, (double)submitted_share_count*60. /
( (double)uptime.tv_sec + (double)uptime.tv_usec / 1e6 ) ); ( (double)uptime.tv_sec + (double)uptime.tv_usec / 1e6 ) );
applog2( LOG_INFO, "Hash rate %7.2f%sh/s %7.2f%sh/s (%.2f%sh/s)", applog2( LOG_INFO, "Hash rate %7.2f%sh/s %7.2f%sh/s (%.2f%sh/s)",
scaled_shrate, shr_units, sess_hrate, sess_hr_units, shrate, shr_units, sess_hrate, sess_hr_units,
scaled_ghrate, ghr_units ); ghrate, ghr_units );
if ( accepted_share_count < submitted_share_count )
{
double lost_ghrate = uptime.tv_sec == 0. ? 0.
: diff_to_hash * last_targetdiff
* (double)(submitted_share_count - accepted_share_count )
/ (double)uptime.tv_sec;
double shrate = share_time == 0. ? 0. : diff_to_hash * last_targetdiff
* (double)(accepts) / share_time;
double lost_shrate = share_time == 0. ? 0.
: diff_to_hash * last_targetdiff * (double)(submits - accepts )
/ share_time;
char lshr_units[4] = {0};
char lghr_units[4] = {0};
scale_hash_for_display( &lost_shrate, lshr_units );
scale_hash_for_display( &lost_ghrate, lghr_units );
applog2( LOG_INFO, "Lost hash rate %7.2f%sh/s %7.2f%sh/s",
lost_shrate, lshr_units, lost_ghrate, lghr_units );
}
applog2( LOG_INFO,"Submitted %6d %6d", applog2( LOG_INFO,"Submitted %6d %6d",
submits, submitted_share_count ); submits, submitted_share_count );
applog2( LOG_INFO,"Accepted %6d %6d", applog2( LOG_INFO,"Accepted %6d %6d",

5
simd-utils/simd-256.h
View File

@@ -121,12 +121,13 @@ do { \
// Horizontal vector testing // Horizontal vector testing
// needs a proper test, seems to be working in the code but polarity appears
// reversed.
#define mm256_allbits0( a ) _mm256_testz_si256( a, a ) #define mm256_allbits0( a ) _mm256_testz_si256( a, a )
#define mm256_allbits1( a ) _mm256_testc_si256( a, m256_neg1 ) #define mm256_allbits1( a ) _mm256_testc_si256( a, m256_neg1 )
//broken //broken
//#define mm256_allbitsne( a ) _mm256_testnzc_si256( a, m256_neg1 ) //#define mm256_allbitsne( a ) _mm256_testnzc_si256( a, m256_neg1 )
#define mm256_anybits0( a ) !mm256_allbits1( a ) #define mm256_anybits0( a ) !mm256_allbits1( a )
#define mm256_anybits1( a ) !mm256_allbits0( a ) #define mm256_anybits1( a ) !mm256_allbits0( a )