v3.9.8.1

Makefile.am

@@ -18,7 +18,6 @@ dist_man_MANS	= cpuminer.1
 cpuminer_SOURCES = \
   cpu-miner.c \
   util.c \
-  uint256.cpp \
   api.c \
   sysinfos.c \
   algo-gate-api.c\
@@ -263,6 +262,8 @@ cpuminer_SOURCES = \
   algo/x16/x16r-gate.c \
   algo/x16/x16r.c \
   algo/x16/x16r-4way.c \
+  algo/x16/x16rv2.c \
+  algo/x16/x16rv2-4way.c \
   algo/x16/x16rt.c \
   algo/x16/x16rt-4way.c \
   algo/x16/hex.c \

README.md

@@ -24,7 +24,7 @@ Requirements
 
 1. A x86_64 architecture CPU with a minimum of SSE2 support. This includes
 Intel Core2 and newer and AMD equivalents. In order to take advantage of AES_NI
-optimizations a CPU with AES_NI is required. This includes Intel Westbridge
+optimizations a CPU with AES_NI is required. This includes Intel Westmere
 and newer and AMD equivalents. Further optimizations are available on some
 algoritms for CPUs with AVX and AVX2, Sandybridge and Haswell respectively.
 
@@ -87,8 +87,9 @@ Supported Algorithms
                           neoscrypt     NeoScrypt(128, 2, 1)
                           nist5         Nist5
                           pentablake    Pentablake
-                          phi1612       phi, LUX coin (original algo)
-                          phi2          LUX coin (new algo)
+                          phi1612       phi
+                          phi2          Luxcoin (LUX)
+                          phi2-lux      identical to phi2
                           pluck         Pluck:128 (Supcoin)
                           polytimos     Ninja
                           quark         Quark
@@ -120,7 +121,8 @@ Supported Algorithms
                           x13sm3        hsr (Hshare)
                           x14           X14
                           x15           X15
-                          x16r          Ravencoin (RVN)
+                          x16r          Ravencoin (RVN) (original algo)
+                          x16rv2        Ravencoin (RVN) (new algo)
                           x16rt         Gincoin (GIN)
                           x16rt_veil    Veil (VEIL)
                           x16s          Pigeoncoin (PGN)
@@ -155,14 +157,15 @@ Benchmark testing does not work for x11evo.
 Bugs
 ----
 
-Users are encouraged to post their bug reports on the Bitcoin Talk
-forum at:
+Users are encouraged to post their bug reports using git issues or on the
+Bitcoin Talk forum at:
 
 https://bitcointalk.org/index.php?topic=1326803.0
 
-All problem reports must be accompanied by a proper definition.
+All problem reports must be accompanied by a proper problem definition.
 This should include how the problem occurred, the command line and
-output from the miner showing the startup and any errors.
+output from the miner showing the startup messages and any errors.
+A history is also useful, ie did it work before.
 
 Donations
 ---------

RELEASE_NOTES

@@ -38,6 +38,46 @@ supported.
 Change Log
 ----------
 
+v3.9.8.1
+
+Summary log report will be generated on stratum diff change or after 5 minutes,
+whichever comes first, to prevent incorrect data in the report.
+
+Removed phi2-lux alias (introduced in v3.9.8) due to Luxcoin's planned fork
+to a new algo. The new Luxcoin algo is not supported by cpuminer-opt.
+Until the fork Luxcoin can be mined using phi2 algo.
+
+--hide-diff option is deprecated and has no effect. It will be removed in a
+future release.
+
+v3.9.8
+
+Changes to log output to provide data more relevant to actual mining
+performance.
+phi2 can now handle pools with a mix of coins that use and don't use roots.
+phi2-lux added as an alias for phi2 as they are identical except for roots.
+Add x16rv2 algo for Ravencoin fork.
+
+v3.9.7
+
+Command line option changes:
+
+"-R" is no longer used as a shortcut for "--retry-pause", users must
+use the long option.
+
+New options:
+
+-N, --param-n: set the N parameter for yescrypt, yespower or scrypt algos
+-R, --param-r: set the R parameter for yescrypt or yespower algos, scrypt is
+     hardcoded with R=1
+-K, --param-key: set the client key/pers parameter for yescrypt/yespower algos.
+
+These options can be used to mine yescrypt or yespower variations using
+the generic yescrypt or yespower algo name and specifying the parameters
+manually. They can even be used to mine variations that aren't formally
+supported by a unique algo name. Existing algos can continue to to be mined
+using their original name without parameters.
+
 v3.9.6.2
 
 New algo blake2b.

algo-gate-api.c

@@ -122,7 +122,6 @@ void init_algo_gate( algo_gate_t* gate )
    gate->stratum_gen_work        = (void*)&std_stratum_gen_work;
    gate->build_stratum_request   = (void*)&std_le_build_stratum_request;
    gate->malloc_txs_request      = (void*)&std_malloc_txs_request;
-   gate->set_target              = (void*)&std_set_target;
    gate->submit_getwork_result   = (void*)&std_le_submit_getwork_result;
    gate->build_block_header      = (void*)&std_build_block_header;
    gate->build_extraheader       = (void*)&std_build_extraheader;
@@ -234,6 +233,7 @@ bool register_algo_gate( int algo, algo_gate_t *gate )
     case ALGO_X14:           register_x14_algo           ( gate ); break;
     case ALGO_X15:           register_x15_algo           ( gate ); break;
     case ALGO_X16R:          register_x16r_algo          ( gate ); break;
+    case ALGO_X16RV2:        register_x16rv2_algo        ( gate ); break;
     case ALGO_X16RT:         register_x16rt_algo         ( gate ); break;
     case ALGO_X16RT_VEIL:    register_x16rt_veil_algo    ( gate ); break;
     case ALGO_X16S:          register_x16s_algo          ( gate ); break;
@@ -337,7 +337,6 @@ const char* const algo_alias_map[][2] =
   { "myriad",            "myr-gr"       },
   { "neo",               "neoscrypt"    },
   { "phi",               "phi1612"      },
-//  { "sia",               "blake2b"      },
   { "sib",               "x11gost"      },
   { "timetravel8",       "timetravel"   },
   { "veil",              "x16rt-veil"   },
@@ -365,40 +364,3 @@ void get_algo_alias( char** algo_or_alias )
 #undef ALIAS
 #undef PROPER
 
-bool submit_solution( struct work *work, void *hash,
-                      struct thr_info *thr )
-{
-     work_set_target_ratio( work, hash );
-     if ( submit_work( thr, work ) )
-     {
-         if ( !opt_quiet )
-            applog( LOG_BLUE, "Share %d submitted by thread %d, job %s.",
-                    accepted_share_count + rejected_share_count + 1,
-                    thr->id, work->job_id );
-         return true;
-     }
-     else
-          applog( LOG_WARNING, "Failed to submit share." );
-     return false;
-}
-
-bool submit_lane_solution( struct work *work, void *hash,
-                           struct thr_info *thr, int lane )
-{
-     work_set_target_ratio( work, hash );
-     if ( submit_work( thr, work ) )
-     {
-         if ( !opt_quiet )
-//            applog( LOG_BLUE, "Share %d submitted by thread %d, lane %d.",
-//                    accepted_share_count + rejected_share_count + 1,
-//                    thr->id, lane );
-            applog( LOG_BLUE, "Share %d submitted by thread %d, lane %d, job %s.",
-                    accepted_share_count + rejected_share_count + 1, thr->id,
-                    lane, work->job_id );
-         return true;
-     }
-     else
-          applog( LOG_WARNING, "Failed to submit share." );
-     return false;
-}
-

algo-gate-api.h

@@ -132,7 +132,6 @@ void ( *decode_extra_data )      ( struct work*, uint64_t* );
 void ( *wait_for_diff )          ( struct stratum_ctx* );
 int64_t ( *get_max64 )           ();
 bool ( *work_decode )            ( const json_t*, struct work* );
-void ( *set_target)              ( struct work*, double );
 bool ( *submit_getwork_result )  ( CURL*, struct work* );
 void ( *gen_merkle_root )        ( char*, struct stratum_ctx* );
 void ( *build_extraheader )      ( struct work*, struct stratum_ctx* );
@@ -193,15 +192,6 @@ void four_way_not_tested();
 // allways returns failure
 int null_scanhash();
 
-// Allow algos to submit from scanhash loop.
-bool submit_solution( struct work *work, void *hash,
-                      struct thr_info *thr );
-bool submit_lane_solution( struct work *work, void *hash,
-                          struct thr_info *thr, int lane );
-
- 
-bool submit_work( struct thr_info *thr, const struct work *work_in );
-
 // displays warning
 void null_hash    ();
 void null_hash_suw();
@@ -232,10 +222,6 @@ int64_t get_max64_0x3fffffLL();
 int64_t get_max64_0x1ffff();
 int64_t get_max64_0xffffLL();
 
-void std_set_target(    struct work *work, double job_diff );
-void alt_set_target(    struct work* work, double job_diff );
-void scrypt_set_target( struct work *work, double job_diff );
-
 bool std_le_work_decode( const json_t *val, struct work *work );
 bool std_be_work_decode( const json_t *val, struct work *work );
 bool jr2_work_decode( const json_t *val, struct work *work );

algo/argon2/argon2a/argon2a.c

@@ -85,8 +85,9 @@ bool register_argon2_algo( algo_gate_t* gate )
   gate->scanhash        = (void*)&scanhash_argon2;
   gate->hash            = (void*)&argon2hash;
   gate->gen_merkle_root = (void*)&SHA256_gen_merkle_root;
-  gate->set_target      = (void*)&scrypt_set_target;
   gate->get_max64       = (void*)&argon2_get_max64;
+  opt_target_factor = 65536.0;
+
   return true;
 };
 

algo/argon2/argon2d/argon2d-gate.c

@@ -67,8 +67,8 @@ bool register_argon2d_crds_algo( algo_gate_t* gate )
 {
         gate->scanhash = (void*)&scanhash_argon2d_crds;
         gate->hash = (void*)&argon2d_crds_hash;
-        gate->set_target = (void*)&scrypt_set_target;
         gate->optimizations = SSE2_OPT | AVX2_OPT | AVX512_OPT;
+        opt_target_factor = 65536.0;
         return true;
 }
 
@@ -135,8 +135,8 @@ bool register_argon2d_dyn_algo( algo_gate_t* gate )
 {
         gate->scanhash = (void*)&scanhash_argon2d_dyn;
         gate->hash = (void*)&argon2d_dyn_hash;
-        gate->set_target = (void*)&scrypt_set_target;
         gate->optimizations = SSE2_OPT | AVX2_OPT | AVX512_OPT;
+        opt_target_factor = 65536.0;
         return true;
 }
 
@@ -184,9 +184,9 @@ int64_t get_max64_0x1ff() { return 0x1ff; }
 bool register_argon2d4096_algo( algo_gate_t* gate )
 {
         gate->scanhash = (void*)&scanhash_argon2d4096;
-        gate->set_target = (void*)&scrypt_set_target;
         gate->get_max64  = (void*)&get_max64_0x1ff;
         gate->optimizations = SSE2_OPT | AVX2_OPT | AVX512_OPT;
+        opt_target_factor = 65536.0;
         return true;
 }
 

algo/bmw/bmw512-gate.c

@@ -5,8 +5,8 @@ int64_t bmw512_get_max64() { return 0x7ffffLL; }
 bool register_bmw512_algo( algo_gate_t* gate )
 {
   gate->optimizations = AVX2_OPT;
-  gate->set_target      = (void*)&alt_set_target;
   gate->get_max64       = (void*)&bmw512_get_max64;
+  opt_target_factor = 256.0;
 #if defined (BMW512_4WAY)
   gate->scanhash  = (void*)&scanhash_bmw512_4way;
   gate->hash      = (void*)&bmw512hash_4way;

algo/groestl/groestl.c

@@ -94,19 +94,14 @@ int scanhash_groestl( struct work *work, uint32_t max_nonce,
 	return 0;
 }
 
-void groestl_set_target( struct work* work, double job_diff )
-{
- work_set_target( work, job_diff / (256.0 * opt_diff_factor) );
-}
-
 bool register_dmd_gr_algo( algo_gate_t* gate )
 {
     init_groestl_ctx();
     gate->optimizations   = SSE2_OPT | AES_OPT;
     gate->scanhash        = (void*)&scanhash_groestl;
     gate->hash            = (void*)&groestlhash;
-    gate->set_target      = (void*)&groestl_set_target;
     gate->get_max64       = (void*)&get_max64_0x3ffff;
+    opt_target_factor = 256.0;
     return true;
 };
 

algo/hodl/hodl-gate.c

@@ -15,11 +15,6 @@ pthread_barrier_t hodl_barrier;
 // need to be passed.
 unsigned char *hodl_scratchbuf = NULL;
 
-void hodl_set_target( struct work* work, double diff )
-{
-     diff_to_target(work->target, diff / 8388608.0 );
-}
-
 void hodl_le_build_stratum_request( char* req, struct work* work,
                                     struct stratum_ctx *sctx ) 
 {
@@ -170,7 +165,6 @@ bool register_hodl_algo( algo_gate_t* gate )
   gate->scanhash              = (void*)&hodl_scanhash;
   gate->get_new_work          = (void*)&hodl_get_new_work;
   gate->longpoll_rpc_call     = (void*)&hodl_longpoll_rpc_call;
-  gate->set_target            = (void*)&hodl_set_target;
   gate->build_stratum_request = (void*)&hodl_le_build_stratum_request;
   gate->malloc_txs_request    = (void*)&hodl_malloc_txs_request;
   gate->build_block_header    = (void*)&hodl_build_block_header;
@@ -179,6 +173,7 @@ bool register_hodl_algo( algo_gate_t* gate )
   gate->work_cmp_size         = 76;
   hodl_scratchbuf = (unsigned char*)malloc( 1 << 30 );
   allow_getwork = false;
+  opt_target_factor = 8388608.0;
   return ( hodl_scratchbuf != NULL );
 }
 

algo/jh/jha-gate.c

@@ -12,7 +12,7 @@ bool register_jha_algo( algo_gate_t* gate )
   gate->hash             = (void*)&jha_hash;
 #endif
   gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT;
-  gate->set_target       = (void*)&scrypt_set_target;
+  opt_target_factor = 65536.0;
   return true;
 };
 

algo/keccak/keccak-gate.c

@@ -1,18 +1,13 @@
 #include "keccak-gate.h"
 
-void keccak_set_target( struct work* work, double job_diff )
-{
-  work_set_target( work, job_diff / (128.0 * opt_diff_factor) );
-}
-
 int64_t keccak_get_max64() { return 0x7ffffLL; }
 
 bool register_keccak_algo( algo_gate_t* gate )
 {
   gate->optimizations = AVX2_OPT;
   gate->gen_merkle_root = (void*)&SHA256_gen_merkle_root;
-  gate->set_target      = (void*)&keccak_set_target;
   gate->get_max64       = (void*)&keccak_get_max64;
+  opt_target_factor = 128.0;
 #if defined (KECCAK_4WAY)
   gate->scanhash  = (void*)&scanhash_keccak_4way;
   gate->hash      = (void*)&keccakhash_4way;
@@ -23,17 +18,12 @@ bool register_keccak_algo( algo_gate_t* gate )
   return true;
 };
 
-void keccakc_set_target( struct work* work, double job_diff )
-{
-  work_set_target( work, job_diff / (256.0 * opt_diff_factor) );
-}
-
 bool register_keccakc_algo( algo_gate_t* gate )
 {
   gate->optimizations = AVX2_OPT;
   gate->gen_merkle_root = (void*)&sha256d_gen_merkle_root;
-  gate->set_target      = (void*)&keccakc_set_target;
   gate->get_max64       = (void*)&keccak_get_max64;
+  opt_target_factor = 256.0;
 #if defined (KECCAK_4WAY)
   gate->scanhash  = (void*)&scanhash_keccak_4way;
   gate->hash      = (void*)&keccakhash_4way;

algo/lyra2/lyra2-gate.c

@@ -71,7 +71,7 @@ bool register_lyra2rev3_algo( algo_gate_t* gate )
 #endif
   gate->optimizations = SSE2_OPT | SSE42_OPT | AVX2_OPT;
   gate->miner_thread_init = (void*)&lyra2rev3_thread_init;
-  gate->set_target        = (void*)&alt_set_target;
+  opt_target_factor = 256.0;
   return true;
 };
 
@@ -105,7 +105,7 @@ bool register_lyra2rev2_algo( algo_gate_t* gate )
 #endif
   gate->optimizations = SSE2_OPT | AES_OPT | SSE42_OPT | AVX2_OPT;
   gate->miner_thread_init = (void*)&lyra2rev2_thread_init;
-  gate->set_target        = (void*)&alt_set_target;
+  opt_target_factor = 256.0;
   return true;
 };
 
@@ -128,7 +128,7 @@ bool register_lyra2z_algo( algo_gate_t* gate )
 #endif
   gate->optimizations = SSE42_OPT | AVX2_OPT;
   gate->get_max64  = (void*)&get_max64_0xffffLL;
-  gate->set_target = (void*)&alt_set_target;
+  opt_target_factor = 256.0;
   return true;
 };
 
@@ -148,7 +148,7 @@ bool register_lyra2h_algo( algo_gate_t* gate )
 #endif
   gate->optimizations = SSE42_OPT | AVX2_OPT;
   gate->get_max64  = (void*)&get_max64_0xffffLL;
-  gate->set_target = (void*)&alt_set_target;
+  opt_target_factor = 256.0;
   return true;
 };
 
@@ -168,8 +168,8 @@ bool register_allium_algo( algo_gate_t* gate )
   gate->hash      = (void*)&allium_hash;
 #endif
   gate->optimizations = SSE2_OPT | AES_OPT | SSE42_OPT | AVX2_OPT;
-  gate->set_target        = (void*)&alt_set_target;
   gate->get_max64         = (void*)&allium_get_max64_0xFFFFLL;
+  opt_target_factor = 256.0;
   return true;
 };
 
@@ -182,6 +182,7 @@ int phi2_get_work_data_size() { return phi2_use_roots ? 144 : 128; }
 
 void phi2_decode_extra_data( struct work *work )
 {
+   phi2_use_roots = false;
    if ( work->data[0] & ( 1<<30 ) ) phi2_use_roots = true;
    else for ( int i = 20; i < 36; i++ )
    {
@@ -213,8 +214,8 @@ bool register_phi2_algo( algo_gate_t* gate )
    gate->get_work_data_size = (void*)&phi2_get_work_data_size;
    gate->decode_extra_data  = (void*)&phi2_decode_extra_data;
    gate->build_extraheader  = (void*)&phi2_build_extraheader;
-   gate->set_target         = (void*)&alt_set_target; 
    gate->get_max64          = (void*)&get_max64_0xffffLL;
+   opt_target_factor = 256.0;
 #if defined(PHI2_4WAY)
    gate->scanhash           = (void*)&scanhash_phi2_4way;
 #else

algo/lyra2/lyra2re.c

@@ -118,11 +118,6 @@ int64_t lyra2re_get_max64 ()
   return 0xffffLL;
 }
 
-void lyra2re_set_target ( struct work* work, double job_diff )
-{
-   work_set_target(work, job_diff / (128.0 * opt_diff_factor) );
-}
-
 bool register_lyra2re_algo( algo_gate_t* gate )
 {
   init_lyra2re_ctx();
@@ -130,7 +125,7 @@ bool register_lyra2re_algo( algo_gate_t* gate )
   gate->scanhash   = (void*)&scanhash_lyra2re;
   gate->hash       = (void*)&lyra2re_hash;
   gate->get_max64  = (void*)&lyra2re_get_max64;
-  gate->set_target = (void*)&lyra2re_set_target;
+  opt_target_factor = 128.0;
   return true;
 };
 

algo/lyra2/lyra2z330.c

@@ -53,11 +53,6 @@ int scanhash_lyra2z330( struct work *work, uint32_t max_nonce,
    return 0;
 }
 
-void lyra2z330_set_target( struct work* work, double job_diff )
-{
- work_set_target( work, job_diff / (256.0 * opt_diff_factor) );
-}
-
 bool lyra2z330_thread_init()
 {
    const int64_t ROW_LEN_INT64 = BLOCK_LEN_INT64 * 256; // nCols
@@ -76,7 +71,7 @@ bool register_lyra2z330_algo( algo_gate_t* gate )
   gate->scanhash   = (void*)&scanhash_lyra2z330;
   gate->hash       = (void*)&lyra2z330_hash;
   gate->get_max64  = (void*)&get_max64_0xffffLL;
-  gate->set_target = (void*)&lyra2z330_set_target;
+  opt_target_factor = 256.0;
   return true;
 };
 

algo/m7m.c

@@ -323,9 +323,9 @@ bool register_m7m_algo( algo_gate_t *gate )
   gate->build_stratum_request = (void*)&std_be_build_stratum_request;
   gate->work_decode           = (void*)&std_be_work_decode;
   gate->submit_getwork_result = (void*)&std_be_submit_getwork_result;
-  gate->set_target            = (void*)&scrypt_set_target;
   gate->get_max64             = (void*)&get_max64_0x1ffff;
   gate->set_work_data_endian  = (void*)&set_work_data_big_endian;
+  opt_target_factor = 65536.0;
   return true;
 }
 

algo/quark/hmq1725-gate.c

@@ -10,8 +10,8 @@ bool register_hmq1725_algo( algo_gate_t* gate )
   gate->scanhash  = (void*)&scanhash_hmq1725;
   gate->hash      = (void*)&hmq1725hash;
 #endif
-  gate->set_target       = (void*)&scrypt_set_target;
   gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT;
+  opt_target_factor = 65536.0;
   return true;
 };
 

algo/quark/hmq1725.c

@@ -409,14 +409,3 @@ int scanhash_hmq1725( struct work *work, uint32_t max_nonce,
 	pdata[19] = n;
 	return 0;
 }
-/*
-bool register_hmq1725_algo( algo_gate_t* gate )
-{
-  init_hmq1725_ctx();
-  gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT;
-  gate->set_target       = (void*)&scrypt_set_target;
-  gate->scanhash         = (void*)&scanhash_hmq1725;
-  gate->hash             = (void*)&hmq1725hash;
-  return true;
-};
-*/

algo/ripemd/lbry-gate.c

@@ -41,6 +41,7 @@ void lbry_le_build_stratum_request( char *req, struct work *work,
    free(xnonce2str);
 }
 
+/*
 void lbry_build_block_header( struct work* g_work, uint32_t version,
                              uint32_t *prevhash, uint32_t *merkle_root,
                              uint32_t ntime, uint32_t nbits )
@@ -63,6 +64,7 @@ void lbry_build_block_header( struct work* g_work, uint32_t version,
    g_work->data[ LBRY_NBITS_INDEX ] = nbits;
    g_work->data[28] = 0x80000000;
 }
+*/
 
 void lbry_build_extraheader( struct work* g_work, struct stratum_ctx* sctx )
 {
@@ -92,11 +94,6 @@ void lbry_build_extraheader( struct work* g_work, struct stratum_ctx* sctx )
    g_work->data[28] = 0x80000000;
 }
 
-void lbry_set_target( struct work* work, double job_diff )
-{
- work_set_target( work, job_diff / (256.0 * opt_diff_factor) );
-}
-
 int64_t lbry_get_max64() { return 0x1ffffLL; }
 
 int lbry_get_work_data_size() { return LBRY_WORK_DATA_SIZE; }
@@ -119,11 +116,11 @@ bool register_lbry_algo( algo_gate_t* gate )
   gate->build_stratum_request = (void*)&lbry_le_build_stratum_request;
 //  gate->build_block_header    = (void*)&build_block_header;
   gate->build_extraheader     = (void*)&lbry_build_extraheader;
-  gate->set_target            = (void*)&lbry_set_target;
   gate->ntime_index           = LBRY_NTIME_INDEX;
   gate->nbits_index           = LBRY_NBITS_INDEX;
   gate->nonce_index           = LBRY_NONCE_INDEX;
   gate->get_work_data_size    = (void*)&lbry_get_work_data_size;
+  opt_target_factor = 256.0;
   return true;
 }
 

algo/scrypt/neoscrypt.c

@@ -1089,13 +1089,13 @@ bool register_neoscrypt_algo( algo_gate_t* gate )
   gate->scanhash              = (void*)&scanhash_neoscrypt;
   gate->hash                  = (void*)&neoscrypt;
   gate->get_max64             = (void*)&get_neoscrypt_max64;
-  gate->set_target            = (void*)&scrypt_set_target;
   gate->wait_for_diff         = (void*)&neoscrypt_wait_for_diff;
   gate->build_stratum_request = (void*)&std_be_build_stratum_request;
   gate->work_decode           = (void*)&std_be_work_decode;
   gate->submit_getwork_result = (void*)&std_be_submit_getwork_result;
   gate->set_work_data_endian  = (void*)&set_work_data_big_endian;
   gate->get_work_data_size    = (void*)&neoscrypt_get_work_data_size;
+  opt_target_factor = 65536.0;
   return true;
 };
 

algo/scrypt/pluck.c

@@ -503,8 +503,8 @@ bool register_pluck_algo( algo_gate_t* gate )
   gate->miner_thread_init = (void*)&pluck_miner_thread_init;
   gate->scanhash         = (void*)&scanhash_pluck;
   gate->hash             = (void*)&pluck_hash;
-  gate->set_target       = (void*)&scrypt_set_target;
   gate->get_max64        = (void*)&pluck_get_max64;
+  opt_target_factor = 65536.0;
   return true;
 };
 

algo/scrypt/scrypt.c

@@ -698,8 +698,8 @@ static void scrypt_1024_1_1_256_24way(const uint32_t *input,
 extern int scanhash_scrypt( struct work *work, uint32_t max_nonce,
                             uint64_t *hashes_done, struct thr_info *mythr )
 {
-        uint32_t *pdata = work->data;
-        uint32_t *ptarget = work->target;
+   uint32_t *pdata = work->data;
+   uint32_t *ptarget = work->target;
 	uint32_t data[SCRYPT_MAX_WAYS * 20], hash[SCRYPT_MAX_WAYS * 8];
 	uint32_t midstate[8];
 	uint32_t n = pdata[19] - 1;
@@ -783,13 +783,18 @@ bool register_scrypt_algo( algo_gate_t* gate )
   gate->miner_thread_init =(void*)&scrypt_miner_thread_init;
   gate->scanhash         = (void*)&scanhash_scrypt;
 //  gate->hash             = (void*)&scrypt_1024_1_1_256_24way;
-  gate->set_target       = (void*)&scrypt_set_target;
   gate->get_max64        = (void*)&scrypt_get_max64;
+  opt_target_factor = 65536.0;
 
-  if ( !opt_scrypt_n )
+
+  if ( !opt_param_n )
+  {
+     opt_param_n = 1024;
      scratchbuf_size = 1024;
+  }
   else
-     scratchbuf_size = opt_scrypt_n;
+     scratchbuf_size = opt_param_n;
+  applog(LOG_INFO,"Scrypt paramaters: N= %d, R= 1.", opt_param_n );
   return true;
 };
 

algo/scryptjane/scrypt-jane-chacha.h

@@ -55,6 +55,7 @@ typedef uint32_t scrypt_mix_word_t;
 	#include "scrypt-jane-romix-template.h"
 #endif
 
+
 /* cpu agnostic */
 #define SCRYPT_ROMIX_FN scrypt_ROMix_basic
 #define SCRYPT_MIX_FN chacha_core_basic

algo/scryptjane/scrypt-jane-romix-template.h

@@ -1,9 +1,11 @@
 #if !defined(SCRYPT_CHOOSE_COMPILETIME) || !defined(SCRYPT_HAVE_ROMIX)
 
+/*
 #if defined(SCRYPT_CHOOSE_COMPILETIME)
 #undef SCRYPT_ROMIX_FN
 #define SCRYPT_ROMIX_FN scrypt_ROMix
 #endif
+*/
 
 #undef SCRYPT_HAVE_ROMIX
 #define SCRYPT_HAVE_ROMIX

algo/scryptjane/scrypt-jane.c

@@ -240,24 +240,24 @@ bool register_scryptjane_algo( algo_gate_t* gate )
 {
     gate->scanhash   = (void*)&scanhash_scryptjane;
     gate->hash       = (void*)&scryptjanehash;
-    gate->set_target = (void*)&scrypt_set_target;
     gate->get_max64  = (void*)&get_max64_0x40LL;
+    opt_target_factor = 65536.0;
 
     // figure out if arg in N or Nfactor
-    if ( !opt_scrypt_n )
+    if ( !opt_param_n )
     {
       applog( LOG_ERR, "The N factor must be specified in the form algo:nf");
       return false;
     }
-    else if ( opt_scrypt_n < 32 )
+    else if ( opt_param_n < 32 )
     {
       // arg is Nfactor, calculate N
-      sj_N = 1 << ( opt_scrypt_n + 1 );
+      sj_N = 1 << ( opt_param_n + 1 );
     }
     else
     {
       // arg is N
-      sj_N = opt_scrypt_n;
+      sj_N = opt_param_n;
     }
     return true;
 }

algo/x11/fresh.c

@@ -120,19 +120,13 @@ int scanhash_fresh( struct work *work,
 	return 0;
 }
 
-void fresh_set_target( struct work* work, double job_diff )
-{
- work_set_target( work, job_diff / (256.0 * opt_diff_factor) );
-}
-
-
 bool register_fresh_algo( algo_gate_t* gate )
 {
     algo_not_tested();
     gate->scanhash   = (void*)&scanhash_fresh;
     gate->hash       = (void*)&freshhash;
-    gate->set_target = (void*)&fresh_set_target;
     gate->get_max64  = (void*)&get_max64_0x3ffff;
+    opt_target_factor = 256.0;
     return true;
 };
 

algo/x11/timetravel-gate.c

@@ -1,10 +1,5 @@
 #include "timetravel-gate.h"
 
-void tt8_set_target( struct work* work, double job_diff )
-{
- work_set_target( work, job_diff / (256.0 * opt_diff_factor) );
-}
-
 bool register_timetravel_algo( algo_gate_t* gate )
 {
 #ifdef TIMETRAVEL_4WAY
@@ -16,9 +11,9 @@ bool register_timetravel_algo( algo_gate_t* gate )
   gate->scanhash   = (void*)&scanhash_timetravel;
   gate->hash       = (void*)&timetravel_hash;
 #endif
-  gate->set_target = (void*)&tt8_set_target;
   gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT;
   gate->get_max64  = (void*)&get_max64_0xffffLL;
+  opt_target_factor = 256.0;
   return true;
 };
 

algo/x11/timetravel10-gate.c

@@ -1,10 +1,5 @@
 #include "timetravel10-gate.h"
 
-void tt10_set_target( struct work* work, double job_diff )
-{
- work_set_target( work, job_diff / (256.0 * opt_diff_factor) );
-}
-
 bool register_timetravel10_algo( algo_gate_t* gate )
 {
 #ifdef TIMETRAVEL10_4WAY
@@ -16,9 +11,9 @@ bool register_timetravel10_algo( algo_gate_t* gate )
   gate->scanhash   = (void*)&scanhash_timetravel10;
   gate->hash       = (void*)&timetravel10_hash;
 #endif
-  gate->set_target = (void*)&tt10_set_target;
   gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT;
   gate->get_max64  = (void*)&get_max64_0xffffLL;
+  opt_target_factor = 256.0;
   return true;
 };
 

algo/x13/drop.c

@@ -249,7 +249,6 @@ bool register_drop_algo( algo_gate_t* gate )
     gate->scanhash              = (void*)&scanhash_drop;
     gate->hash                  = (void*)&droplp_hash_pok;
     gate->get_new_work          = (void*)&drop_get_new_work;
-    gate->set_target            = (void*)&scrypt_set_target;
     gate->build_stratum_request = (void*)&std_be_build_stratum_request;
     gate->work_decode           = (void*)&std_be_work_decode;
     gate->submit_getwork_result = (void*)&std_be_submit_getwork_result;
@@ -257,6 +256,7 @@ bool register_drop_algo( algo_gate_t* gate )
     gate->decode_extra_data     = (void*)&drop_display_pok;
     gate->get_work_data_size    = (void*)&drop_get_work_data_size;
     gate->work_cmp_size         = 72;
+    opt_target_factor = 65536.0;
     return true;
 };
 

algo/x16/x16r-4way.c

@@ -68,21 +68,7 @@ void x16r_4way_hash( void* output, const void* input )
    int size = 80;
 
    dintrlv_4x64( hash0, hash1, hash2, hash3, input, 640 );
-/* 
-   if ( s_ntime == UINT32_MAX )
-   {
-      const uint8_t* tmp = (uint8_t*) in0;
-      x16_r_s_getAlgoString( &tmp[4], hashOrder );
-   }
-*/
-   // Input data is both 64 bit interleaved (input)
-   // and deinterleaved in inp0-3.
-   // If First function uses 64 bit data it is not required to interleave inp
-   // first. It may use the inerleaved data dmost convenient, ie 4way 64 bit.
-   // All other functions assume data is deinterleaved in hash0-3
-   // All functions must exit with data deinterleaved in hash0-3.
-   // Alias in0-3 points to either inp0-3 or hash0-3 according to
-   // its hashOrder position. Size is also set accordingly.
+
    for ( int i = 0; i < 16; i++ )
    {
       const char elem = hashOrder[i];

algo/x16/x16r-gate.c

@@ -42,8 +42,23 @@ bool register_x16r_algo( algo_gate_t* gate )
   gate->hash      = (void*)&x16r_hash;
 #endif
   gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT;
-  gate->set_target = (void*)&alt_set_target;
   x16_r_s_getAlgoString = (void*)&x16r_getAlgoString;
+  opt_target_factor = 256.0;
+  return true;
+};
+
+bool register_x16rv2_algo( algo_gate_t* gate )
+{
+#if defined (X16R_4WAY)
+  gate->scanhash  = (void*)&scanhash_x16rv2_4way;
+  gate->hash      = (void*)&x16rv2_4way_hash;
+#else
+  gate->scanhash  = (void*)&scanhash_x16rv2;
+  gate->hash      = (void*)&x16rv2_hash;
+#endif
+  gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT;
+  x16_r_s_getAlgoString = (void*)&x16r_getAlgoString;
+  opt_target_factor = 256.0;
   return true;
 };
 
@@ -57,8 +72,8 @@ bool register_x16s_algo( algo_gate_t* gate )
   gate->hash      = (void*)&x16r_hash;
 #endif
   gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT;
-  gate->set_target = (void*)&alt_set_target;
   x16_r_s_getAlgoString = (void*)&x16s_getAlgoString;
+  opt_target_factor = 256.0;
   return true;
 };
 
@@ -189,7 +204,7 @@ bool register_x16rt_algo( algo_gate_t* gate )
   gate->hash      = (void*)&x16rt_hash;
 #endif
   gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT;
-  gate->set_target = (void*)&alt_set_target;
+  opt_target_factor = 256.0;
   return true;
 };
 
@@ -203,8 +218,8 @@ bool register_x16rt_veil_algo( algo_gate_t* gate )
   gate->hash      = (void*)&x16rt_hash;
 #endif
   gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT;
-  gate->set_target = (void*)&alt_set_target;
   gate->build_extraheader = (void*)&veil_build_extraheader;
+  opt_target_factor = 256.0;
   return true;
 };
 
@@ -212,19 +227,13 @@ bool register_x16rt_veil_algo( algo_gate_t* gate )
 //
 //    HEX
 
-
-void hex_set_target( struct work* work, double job_diff )
-{
-   work_set_target( work, job_diff / (128.0 * opt_diff_factor) );
-}
-
 bool register_hex_algo( algo_gate_t* gate )
 {
   gate->scanhash        = (void*)&scanhash_hex;
   gate->hash            = (void*)&hex_hash;
   gate->optimizations   = SSE2_OPT | AES_OPT | AVX2_OPT;
   gate->gen_merkle_root = (void*)&SHA256_gen_merkle_root;
-  gate->set_target      = (void*)&hex_set_target;
+  opt_target_factor = 128.0;
   return true;
 };
 
@@ -244,8 +253,8 @@ bool register_x21s_algo( algo_gate_t* gate )
   gate->miner_thread_init = (void*)&x21s_thread_init;
 #endif
   gate->optimizations     = SSE2_OPT | AES_OPT | AVX2_OPT | SHA_OPT;
-  gate->set_target        = (void*)&alt_set_target;
   x16_r_s_getAlgoString   = (void*)&x16s_getAlgoString;
+  opt_target_factor = 256.0;
   return true;
 };
 

algo/x16/x16r-gate.h

@@ -38,6 +38,7 @@ void x16rt_getAlgoString( const uint32_t *timeHash, char *output );
 void x16rt_getTimeHash( const uint32_t timeStamp, void* timeHash );
 
 bool register_x16r_algo( algo_gate_t* gate );
+bool register_x16rv2_algo( algo_gate_t* gate );
 bool register_x16s_algo( algo_gate_t* gate );
 bool register_x16rt_algo( algo_gate_t* gate );
 bool register_hex__algo( algo_gate_t* gate );
@@ -49,6 +50,10 @@ void x16r_4way_hash( void *state, const void *input );
 int scanhash_x16r_4way( struct work *work, uint32_t max_nonce,
                         uint64_t *hashes_done, struct thr_info *mythr );
 
+void x16rv2_4way_hash( void *state, const void *input );
+int scanhash_x16rv2_4way( struct work *work, uint32_t max_nonce,
+                        uint64_t *hashes_done, struct thr_info *mythr );
+
 void x16rt_4way_hash( void *state, const void *input );
 int scanhash_x16rt_4way( struct work *work, uint32_t max_nonce,
                         uint64_t *hashes_done, struct thr_info *mythr );
@@ -64,6 +69,10 @@ void x16r_hash( void *state, const void *input );
 int scanhash_x16r( struct work *work, uint32_t max_nonce,
                    uint64_t *hashes_done, struct thr_info *mythr );
 
+void x16rv2_hash( void *state, const void *input );
+int scanhash_x16rv2( struct work *work, uint32_t max_nonce,
+                   uint64_t *hashes_done, struct thr_info *mythr );
+
 void x16rt_hash( void *state, const void *input );
 int scanhash_x16rt( struct work *work, uint32_t max_nonce,
                    uint64_t *hashes_done, struct thr_info *mythr );

algo/x16/x16rv2-4way.c

@@ -0,0 +1,384 @@
+/**
+ * x16r algo implementation
+ *
+ * Implementation by tpruvot@github Jan 2018
+ * Optimized by JayDDee@github Jan 2018
+ */
+#include "x16r-gate.h"
+
+#if defined (X16R_4WAY)
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include "algo/blake/blake-hash-4way.h"
+#include "algo/bmw/bmw-hash-4way.h"
+#include "algo/groestl/aes_ni/hash-groestl.h"
+#include "algo/groestl/aes_ni/hash-groestl.h"
+#include "algo/skein/skein-hash-4way.h"
+#include "algo/jh/jh-hash-4way.h"
+#include "algo/keccak/keccak-hash-4way.h"
+#include "algo/shavite/sph_shavite.h"
+#include "algo/luffa/luffa-hash-2way.h"
+#include "algo/cubehash/cubehash_sse2.h"
+#include "algo/simd/simd-hash-2way.h"
+#include "algo/echo/aes_ni/hash_api.h"
+#include "algo/hamsi/hamsi-hash-4way.h"
+#include "algo/fugue/sph_fugue.h"
+#include "algo/shabal/shabal-hash-4way.h"
+#include "algo/whirlpool/sph_whirlpool.h"
+#include "algo/sha/sha-hash-4way.h"
+#include "algo/tiger/sph_tiger.h"
+
+static __thread uint32_t s_ntime = UINT32_MAX;
+static __thread char hashOrder[X16R_HASH_FUNC_COUNT + 1] = { 0 };
+
+union _x16rv2_4way_context_overlay
+{
+    blake512_4way_context   blake;
+    bmw512_4way_context     bmw;
+    hashState_echo          echo;
+    hashState_groestl       groestl;
+    skein512_4way_context   skein;
+    jh512_4way_context      jh;
+    keccak512_4way_context  keccak;
+    luffa_2way_context      luffa;
+    cubehashParam           cube;
+    sph_shavite512_context  shavite;
+    simd_2way_context       simd;
+    hamsi512_4way_context   hamsi;
+    sph_fugue512_context    fugue;
+    shabal512_4way_context  shabal;
+    sph_whirlpool_context   whirlpool;
+    sha512_4way_context     sha512;
+    sph_tiger_context       tiger;
+};
+typedef union _x16rv2_4way_context_overlay x16rv2_4way_context_overlay;
+
+// Pad the 24 bytes tiger hash to 64 bytes
+inline void padtiger512( uint32_t* hash )
+{
+  for ( int i = 6; i < 16; i++ ) hash[i] = 0;
+}
+
+void x16rv2_4way_hash( void* output, const void* input )
+{
+   uint32_t hash0[24] __attribute__ ((aligned (64)));
+   uint32_t hash1[24] __attribute__ ((aligned (64)));
+   uint32_t hash2[24] __attribute__ ((aligned (64)));
+   uint32_t hash3[24] __attribute__ ((aligned (64)));
+   uint32_t vhash[24*4] __attribute__ ((aligned (64)));
+   x16rv2_4way_context_overlay ctx;
+   void *in0 = (void*) hash0;
+   void *in1 = (void*) hash1;
+   void *in2 = (void*) hash2;
+   void *in3 = (void*) hash3;
+   int size = 80;
+
+   dintrlv_4x64( hash0, hash1, hash2, hash3, input, 640 );
+
+   for ( int i = 0; i < 16; i++ )
+   {
+      const char elem = hashOrder[i];
+      const uint8_t algo = elem >= 'A' ? elem - 'A' + 10 : elem - '0';
+
+      switch ( algo )
+      {
+         case BLAKE:
+            blake512_4way_init( &ctx.blake );
+            if ( i == 0 )
+               blake512_4way( &ctx.blake, input, size );
+            else
+            {
+               intrlv_4x64( vhash, in0, in1, in2, in3, size<<3 );
+               blake512_4way( &ctx.blake, vhash, size );
+            }
+            blake512_4way_close( &ctx.blake, vhash );
+            dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
+         break;
+         case BMW:
+            bmw512_4way_init( &ctx.bmw );
+            if ( i == 0 )
+               bmw512_4way( &ctx.bmw, input, size );
+            else
+            {
+               intrlv_4x64( vhash, in0, in1, in2, in3, size<<3 );
+               bmw512_4way( &ctx.bmw, vhash, size );
+            }
+            bmw512_4way_close( &ctx.bmw, vhash );
+            dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
+         break;
+         case GROESTL:
+               init_groestl( &ctx.groestl, 64 );
+               update_and_final_groestl( &ctx.groestl, (char*)hash0,
+                                                 (const char*)in0, size<<3 );
+               init_groestl( &ctx.groestl, 64 );
+               update_and_final_groestl( &ctx.groestl, (char*)hash1,
+                                                 (const char*)in1, size<<3 );
+               init_groestl( &ctx.groestl, 64 );
+               update_and_final_groestl( &ctx.groestl, (char*)hash2,
+                                                 (const char*)in2, size<<3 );
+               init_groestl( &ctx.groestl, 64 );
+               update_and_final_groestl( &ctx.groestl, (char*)hash3,
+                                                 (const char*)in3, size<<3 );
+         break;
+         case SKEIN:
+            skein512_4way_init( &ctx.skein );
+            if ( i == 0 )
+               skein512_4way( &ctx.skein, input, size );
+            else
+            {
+               intrlv_4x64( vhash, in0, in1, in2, in3, size<<3 );
+               skein512_4way( &ctx.skein, vhash, size );
+            }
+            skein512_4way_close( &ctx.skein, vhash );
+            dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
+         break;
+         case JH:
+            jh512_4way_init( &ctx.jh );
+            if ( i == 0 )
+               jh512_4way( &ctx.jh, input, size );
+            else
+            {
+               intrlv_4x64( vhash, in0, in1, in2, in3, size<<3 );
+               jh512_4way( &ctx.jh, vhash, size );
+            }
+            jh512_4way_close( &ctx.jh, vhash );
+            dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
+         break;
+         case KECCAK:
+             sph_tiger_init( &ctx.tiger );
+			    sph_tiger( &ctx.tiger, in0, size );
+			    sph_tiger_close( &ctx.tiger, hash0 );
+             sph_tiger_init( &ctx.tiger );
+             sph_tiger( &ctx.tiger, in1, size );
+             sph_tiger_close( &ctx.tiger, hash1 );
+             sph_tiger_init( &ctx.tiger );
+             sph_tiger( &ctx.tiger, in2, size );
+             sph_tiger_close( &ctx.tiger, hash2 );
+             sph_tiger_init( &ctx.tiger );
+             sph_tiger( &ctx.tiger, in3, size );
+             sph_tiger_close( &ctx.tiger, hash3 );
+
+             for ( int i = (24/4); i < (64/4); i++ )
+                hash0[i] = hash1[i] = hash2[i] = hash3[i] = 0;
+
+             intrlv_4x64( vhash, hash0, hash1, hash2, hash3, 512 );
+             keccak512_4way_init( &ctx.keccak );
+             keccak512_4way( &ctx.keccak, vhash, 64 );
+             keccak512_4way_close( &ctx.keccak, vhash );
+             dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
+         break;
+         case LUFFA:
+             sph_tiger_init( &ctx.tiger );
+             sph_tiger( &ctx.tiger, in0, size );
+             sph_tiger_close( &ctx.tiger, hash0 );
+             sph_tiger_init( &ctx.tiger );
+             sph_tiger( &ctx.tiger, in1, size );
+             sph_tiger_close( &ctx.tiger, hash1 );
+
+             for ( int i = (24/4); i < (64/4); i++ )
+                hash0[i] = hash1[i] = 0;
+
+             intrlv_2x128( vhash, hash0, hash1, 512 );
+             luffa_2way_init( &ctx.luffa, 512 );
+             luffa_2way_update_close( &ctx.luffa, vhash, vhash, 64 );
+             dintrlv_2x128( hash0, hash1, vhash, 512 );
+
+             sph_tiger_init( &ctx.tiger );
+             sph_tiger( &ctx.tiger, in2, size );
+             sph_tiger_close( &ctx.tiger, hash2 );
+             sph_tiger_init( &ctx.tiger );
+             sph_tiger( &ctx.tiger, in3, size );
+             sph_tiger_close( &ctx.tiger, hash3 );
+
+             for ( int i = (24/4); i < (64/4); i++ )
+                hash2[i] = hash3[i] = 0;
+             
+             intrlv_2x128( vhash, hash2, hash3, 512 );
+             luffa_2way_init( &ctx.luffa, 512 );
+             luffa_2way_update_close( &ctx.luffa, vhash, vhash, 64 );
+             dintrlv_2x128( hash2, hash3, vhash, 512 );
+         break;
+         case CUBEHASH:
+             cubehashInit( &ctx.cube, 512, 16, 32 );
+             cubehashUpdateDigest( &ctx.cube, (byte*) hash0,
+                                   (const byte*)in0, size );
+             cubehashInit( &ctx.cube, 512, 16, 32 );
+             cubehashUpdateDigest( &ctx.cube, (byte*) hash1,
+                                   (const byte*)in1, size );
+             cubehashInit( &ctx.cube, 512, 16, 32 );
+             cubehashUpdateDigest( &ctx.cube, (byte*) hash2,
+                                   (const byte*)in2, size );
+             cubehashInit( &ctx.cube, 512, 16, 32 );
+             cubehashUpdateDigest( &ctx.cube, (byte*) hash3,
+                                   (const byte*)in3, size );
+         break;
+         case SHAVITE:
+             sph_shavite512_init( &ctx.shavite );
+             sph_shavite512( &ctx.shavite, in0, size );
+             sph_shavite512_close( &ctx.shavite, hash0 );
+             sph_shavite512_init( &ctx.shavite );
+             sph_shavite512( &ctx.shavite, in1, size );
+             sph_shavite512_close( &ctx.shavite, hash1 );
+             sph_shavite512_init( &ctx.shavite );
+             sph_shavite512( &ctx.shavite, in2, size );
+             sph_shavite512_close( &ctx.shavite, hash2 );
+             sph_shavite512_init( &ctx.shavite );
+             sph_shavite512( &ctx.shavite, in3, size );
+             sph_shavite512_close( &ctx.shavite, hash3 );
+         break;
+         case SIMD:
+             intrlv_2x128( vhash, in0, in1, size<<3 );
+             simd_2way_init( &ctx.simd, 512 );
+             simd_2way_update_close( &ctx.simd, vhash, vhash, size<<3 );
+             dintrlv_2x128( hash0, hash1, vhash, 512 );
+             intrlv_2x128( vhash, in2, in3, size<<3 );
+             simd_2way_init( &ctx.simd, 512 );
+             simd_2way_update_close( &ctx.simd, vhash, vhash, size<<3 );
+             dintrlv_2x128( hash2, hash3, vhash, 512 );
+         break;
+         case ECHO:
+             init_echo( &ctx.echo, 512 );
+             update_final_echo ( &ctx.echo, (BitSequence *)hash0,
+                                (const BitSequence*)in0, size<<3 );
+             init_echo( &ctx.echo, 512 );
+             update_final_echo ( &ctx.echo, (BitSequence *)hash1,
+                                (const BitSequence*)in1, size<<3 );
+             init_echo( &ctx.echo, 512 );
+             update_final_echo ( &ctx.echo, (BitSequence *)hash2,
+                                (const BitSequence*)in2, size<<3 );
+             init_echo( &ctx.echo, 512 );
+             update_final_echo ( &ctx.echo, (BitSequence *)hash3,
+                                (const BitSequence*)in3, size<<3 );
+         break;
+         case HAMSI:
+             intrlv_4x64( vhash, in0, in1, in2, in3, size<<3 );
+             hamsi512_4way_init( &ctx.hamsi );
+             hamsi512_4way( &ctx.hamsi, vhash, size );
+             hamsi512_4way_close( &ctx.hamsi, vhash );
+             dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
+         break;
+         case FUGUE:
+             sph_fugue512_init( &ctx.fugue );
+             sph_fugue512( &ctx.fugue, in0, size );
+             sph_fugue512_close( &ctx.fugue, hash0 );
+             sph_fugue512_init( &ctx.fugue );
+             sph_fugue512( &ctx.fugue, in1, size );
+             sph_fugue512_close( &ctx.fugue, hash1 );
+             sph_fugue512_init( &ctx.fugue );
+             sph_fugue512( &ctx.fugue, in2, size );
+             sph_fugue512_close( &ctx.fugue, hash2 );
+             sph_fugue512_init( &ctx.fugue );
+             sph_fugue512( &ctx.fugue, in3, size );
+             sph_fugue512_close( &ctx.fugue, hash3 );
+         break;
+         case SHABAL:
+             intrlv_4x32( vhash, in0, in1, in2, in3, size<<3 );
+             shabal512_4way_init( &ctx.shabal );
+             shabal512_4way( &ctx.shabal, vhash, size );
+             shabal512_4way_close( &ctx.shabal, vhash );
+             dintrlv_4x32( hash0, hash1, hash2, hash3, vhash, 512 );
+         break;
+         case WHIRLPOOL:
+             sph_whirlpool_init( &ctx.whirlpool );
+             sph_whirlpool( &ctx.whirlpool, in0, size );
+             sph_whirlpool_close( &ctx.whirlpool, hash0 );
+             sph_whirlpool_init( &ctx.whirlpool );
+             sph_whirlpool( &ctx.whirlpool, in1, size );
+             sph_whirlpool_close( &ctx.whirlpool, hash1 );
+             sph_whirlpool_init( &ctx.whirlpool );
+             sph_whirlpool( &ctx.whirlpool, in2, size );
+             sph_whirlpool_close( &ctx.whirlpool, hash2 );
+             sph_whirlpool_init( &ctx.whirlpool );
+             sph_whirlpool( &ctx.whirlpool, in3, size );
+             sph_whirlpool_close( &ctx.whirlpool, hash3 );
+         break;
+         case SHA_512:
+             sph_tiger_init( &ctx.tiger );
+             sph_tiger( &ctx.tiger, in0, size );
+             sph_tiger_close( &ctx.tiger, hash0 );
+             sph_tiger_init( &ctx.tiger );
+             sph_tiger( &ctx.tiger, in1, size );
+             sph_tiger_close( &ctx.tiger, hash1 );
+             sph_tiger_init( &ctx.tiger );
+             sph_tiger( &ctx.tiger, in2, size );
+             sph_tiger_close( &ctx.tiger, hash2 );
+             sph_tiger_init( &ctx.tiger );
+             sph_tiger( &ctx.tiger, in3, size );
+             sph_tiger_close( &ctx.tiger, hash3 );
+
+             for ( int i = (24/4); i < (64/4); i++ )
+                hash0[i] = hash1[i] = hash2[i] = hash3[i] = 0;
+ 
+             intrlv_4x64( vhash, hash0, hash1, hash2, hash3, 512 );
+             sha512_4way_init( &ctx.sha512 );
+             sha512_4way( &ctx.sha512, vhash, 64 );
+             sha512_4way_close( &ctx.sha512, vhash );
+             dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
+         break;
+      }
+      size = 64;
+   }
+   memcpy( output,    hash0, 32 );
+   memcpy( output+32, hash1, 32 );
+   memcpy( output+64, hash2, 32 );
+   memcpy( output+96, hash3, 32 );
+}
+
+int scanhash_x16rv2_4way( struct work *work, uint32_t max_nonce,
+                        uint64_t *hashes_done, struct thr_info *mythr)
+{
+   uint32_t hash[4*16] __attribute__ ((aligned (64)));
+   uint32_t vdata[24*4] __attribute__ ((aligned (64)));
+   uint32_t endiandata[20] __attribute__((aligned(64)));
+   uint32_t *pdata = work->data;
+   uint32_t *ptarget = work->target;
+   const uint32_t Htarg = ptarget[7];
+   const uint32_t first_nonce = pdata[19];
+   uint32_t n = first_nonce;
+   int thr_id = mythr->id;  // thr_id arg is deprecated
+    __m256i  *noncev = (__m256i*)vdata + 9;   // aligned
+   volatile uint8_t *restart = &(work_restart[thr_id].restart);
+
+   casti_m256i( endiandata, 0 ) = mm256_bswap_32( casti_m256i( pdata, 0 ) );
+   casti_m256i( endiandata, 1 ) = mm256_bswap_32( casti_m256i( pdata, 1 ) );
+   casti_m128i( endiandata, 4 ) = mm128_bswap_32( casti_m128i( pdata, 4 ) );
+
+   if ( s_ntime != endiandata[17] )
+   {
+      uint32_t ntime = swab32(pdata[17]);
+      x16_r_s_getAlgoString( (const uint8_t*) (&endiandata[1]), hashOrder );
+      s_ntime = ntime;
+      if ( opt_debug && !thr_id )
+              applog( LOG_DEBUG, "hash order %s (%08x)", hashOrder, ntime );
+   }
+
+   if ( opt_benchmark )
+      ptarget[7] = 0x0cff;
+
+   uint64_t *edata = (uint64_t*)endiandata;
+   intrlv_4x64( (uint64_t*)vdata, edata, edata, edata, edata, 640 );
+
+   do
+   {
+      *noncev = mm256_intrlv_blend_32( mm256_bswap_32(
+               _mm256_set_epi32( n+3, 0, n+2, 0, n+1, 0, n, 0 ) ), *noncev );
+
+      x16rv2_4way_hash( hash, vdata );
+      pdata[19] = n;
+
+      for ( int i = 0; i < 4; i++ )  if ( (hash+(i<<3))[7] <= Htarg )
+      if( fulltest( hash+(i<<3), ptarget ) && !opt_benchmark )
+      {
+         pdata[19] = n+i;
+         submit_lane_solution( work, hash+(i<<3), mythr, i );
+      }
+      n += 4;
+   } while ( (  n < max_nonce ) && !(*restart) );
+
+   *hashes_done = n - first_nonce + 1;
+   return 0;
+}
+
+#endif

algo/x16/x16rv2.c

@@ -0,0 +1,247 @@
+/**
+ * x16r algo implementation
+ *
+ * Implementation by tpruvot@github Jan 2018
+ * Optimized by JayDDee@github Jan 2018
+ */
+#include "x16r-gate.h"
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include "algo/blake/sph_blake.h"
+#include "algo/bmw/sph_bmw.h"
+#include "algo/groestl/sph_groestl.h"
+#include "algo/jh/sph_jh.h"
+#include "algo/keccak/sph_keccak.h"
+#include "algo/skein/sph_skein.h"
+#include "algo/shavite/sph_shavite.h"
+#include "algo/luffa/luffa_for_sse2.h"
+#include "algo/cubehash/cubehash_sse2.h"
+#include "algo/simd/nist.h"
+#include "algo/echo/sph_echo.h"
+#include "algo/hamsi/sph_hamsi.h"
+#include "algo/fugue/sph_fugue.h"
+#include "algo/shabal/sph_shabal.h"
+#include "algo/whirlpool/sph_whirlpool.h"
+#include <openssl/sha.h>
+#include "algo/tiger/sph_tiger.h"
+#if defined(__AES__)
+  #include "algo/echo/aes_ni/hash_api.h"
+  #include "algo/groestl/aes_ni/hash-groestl.h"
+#endif
+
+static __thread uint32_t s_ntime = UINT32_MAX;
+static __thread char hashOrder[X16R_HASH_FUNC_COUNT + 1] = { 0 };
+
+union _x16rv2_context_overlay
+{
+#if defined(__AES__)
+        hashState_echo          echo;
+        hashState_groestl       groestl;
+#else
+        sph_groestl512_context   groestl;
+        sph_echo512_context      echo;
+#endif
+        sph_blake512_context    blake;
+        sph_bmw512_context      bmw;
+        sph_skein512_context    skein;
+        sph_jh512_context       jh;
+        sph_keccak512_context   keccak;
+        hashState_luffa         luffa;
+        cubehashParam           cube;
+        sph_shavite512_context  shavite;
+        hashState_sd            simd;
+        sph_hamsi512_context    hamsi;
+        sph_fugue512_context    fugue;
+        sph_shabal512_context   shabal;
+        sph_whirlpool_context   whirlpool;
+        SHA512_CTX              sha512;
+        sph_tiger_context       tiger;
+};
+typedef union _x16rv2_context_overlay x16rv2_context_overlay;
+
+// Pad the 24 bytes tiger hash to 64 bytes
+inline void padtiger512(uint32_t* hash) {
+   for (int i = (24/4); i < (64/4); i++) hash[i] = 0;
+}
+
+void x16rv2_hash( void* output, const void* input )
+{
+   uint32_t _ALIGN(128) hash[16];
+   x16rv2_context_overlay ctx;
+   void *in = (void*) input;
+   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++ )
+   {
+      const char elem = hashOrder[i];
+      const uint8_t algo = elem >= 'A' ? elem - 'A' + 10 : elem - '0';
+
+      switch ( algo )
+      {
+         case BLAKE:
+            sph_blake512_init( &ctx.blake );
+            sph_blake512( &ctx.blake, in, size );
+            sph_blake512_close( &ctx.blake, hash );
+         break;
+         case BMW:
+            sph_bmw512_init( &ctx.bmw );
+            sph_bmw512(&ctx.bmw, in, size);
+            sph_bmw512_close(&ctx.bmw, hash);
+         break;
+         case GROESTL:
+#if defined(__AES__)
+            init_groestl( &ctx.groestl, 64 );
+            update_and_final_groestl( &ctx.groestl, (char*)hash,
+                                      (const char*)in, size<<3 );
+#else
+            sph_groestl512_init( &ctx.groestl );
+            sph_groestl512( &ctx.groestl, in, size );
+            sph_groestl512_close(&ctx.groestl, hash);
+#endif
+         break;
+         case SKEIN:
+            sph_skein512_init( &ctx.skein );
+            sph_skein512( &ctx.skein, in, size );
+            sph_skein512_close( &ctx.skein, hash );
+         break;
+         case JH:
+            sph_jh512_init( &ctx.jh );
+            sph_jh512(&ctx.jh, in, size );
+            sph_jh512_close(&ctx.jh, hash );
+         break;
+         case KECCAK:
+            sph_tiger_init( &ctx.tiger );
+            sph_tiger( &ctx.tiger, in, size );
+            sph_tiger_close( &ctx.tiger, hash );
+            padtiger512( hash );
+            sph_keccak512_init( &ctx.keccak );
+            sph_keccak512( &ctx.keccak, hash, 64 );
+            sph_keccak512_close( &ctx.keccak, hash );
+         break;
+         case LUFFA:
+            sph_tiger_init( &ctx.tiger );
+            sph_tiger( &ctx.tiger, in, size );
+            sph_tiger_close( &ctx.tiger, hash );
+            padtiger512( hash );
+            init_luffa( &ctx.luffa, 512 );
+            update_and_final_luffa( &ctx.luffa, (BitSequence*)hash,
+                                    (const BitSequence*)hash, 64 );
+         break;
+         case CUBEHASH:
+            cubehashInit( &ctx.cube, 512, 16, 32 );
+            cubehashUpdateDigest( &ctx.cube, (byte*) hash,
+                                  (const byte*)in, size );
+         break;
+         case SHAVITE:
+            sph_shavite512_init( &ctx.shavite );
+            sph_shavite512( &ctx.shavite, in, size );
+            sph_shavite512_close( &ctx.shavite, hash );
+         break;
+         case SIMD:
+             init_sd( &ctx.simd, 512 );
+             update_final_sd( &ctx.simd, (BitSequence *)hash,
+                              (const BitSequence*)in, size<<3 );
+         break;
+         case ECHO:
+#if defined(__AES__)
+             init_echo( &ctx.echo, 512 );
+             update_final_echo ( &ctx.echo, (BitSequence *)hash,
+                                (const BitSequence*)in, size<<3 );
+#else
+             sph_echo512_init( &ctx.echo );
+             sph_echo512( &ctx.echo, in, size );
+             sph_echo512_close( &ctx.echo, hash );
+#endif
+         break;
+         case HAMSI:
+             sph_hamsi512_init( &ctx.hamsi );
+             sph_hamsi512( &ctx.hamsi, in, size );
+             sph_hamsi512_close( &ctx.hamsi, hash );
+         break;
+         case FUGUE:
+             sph_fugue512_init( &ctx.fugue );
+             sph_fugue512( &ctx.fugue, in, size );
+             sph_fugue512_close( &ctx.fugue, hash );
+         break;
+         case SHABAL:
+             sph_shabal512_init( &ctx.shabal );
+             sph_shabal512( &ctx.shabal, in, size );
+             sph_shabal512_close( &ctx.shabal, hash );
+         break;
+         case WHIRLPOOL:
+             sph_whirlpool_init( &ctx.whirlpool );
+             sph_whirlpool( &ctx.whirlpool, in, size );
+             sph_whirlpool_close( &ctx.whirlpool, hash );
+         break;
+         case SHA_512:
+             sph_tiger_init( &ctx.tiger );
+             sph_tiger( &ctx.tiger, in, size );
+             sph_tiger_close( &ctx.tiger, hash );
+             padtiger512( hash );
+             SHA512_Init( &ctx.sha512 );
+             SHA512_Update( &ctx.sha512, hash, 64 );
+             SHA512_Final( (unsigned char*) hash, &ctx.sha512 );
+         break;
+      }
+      in = (void*) hash;
+      size = 64;
+   }
+   memcpy(output, hash, 32);
+}
+
+int scanhash_x16rv2( struct work *work, uint32_t max_nonce,
+                   uint64_t *hashes_done, struct thr_info *mythr )
+{
+   uint32_t _ALIGN(128) hash32[8];
+   uint32_t _ALIGN(128) endiandata[20];
+   uint32_t *pdata = work->data;
+   uint32_t *ptarget = work->target;
+   const uint32_t Htarg = ptarget[7];
+   const uint32_t first_nonce = pdata[19];
+   int thr_id = mythr->id;  // thr_id arg is deprecated
+   uint32_t nonce = first_nonce;
+   volatile uint8_t *restart = &(work_restart[thr_id].restart);
+
+   casti_m128i( endiandata, 0 ) = mm128_bswap_32( casti_m128i( pdata, 0 ) );
+   casti_m128i( endiandata, 1 ) = mm128_bswap_32( casti_m128i( pdata, 1 ) );
+   casti_m128i( endiandata, 2 ) = mm128_bswap_32( casti_m128i( pdata, 2 ) );
+   casti_m128i( endiandata, 3 ) = mm128_bswap_32( casti_m128i( pdata, 3 ) );
+   casti_m128i( endiandata, 4 ) = mm128_bswap_32( casti_m128i( pdata, 4 ) );
+
+   if ( s_ntime != pdata[17] )
+   {
+      uint32_t ntime = swab32(pdata[17]);
+      x16_r_s_getAlgoString( (const uint8_t*) (&endiandata[1]), hashOrder );
+      s_ntime = ntime;
+      if ( opt_debug && !thr_id )
+              applog( LOG_DEBUG, "hash order %s (%08x)", hashOrder, ntime );
+   }
+
+   if ( opt_benchmark )
+      ptarget[7] = 0x0cff;
+
+   do
+   {
+      be32enc( &endiandata[19], nonce );
+      x16rv2_hash( hash32, endiandata );
+
+      if ( hash32[7] <= Htarg )
+      if (fulltest( hash32, ptarget ) && !opt_benchmark )
+      {
+         pdata[19] = nonce;
+         submit_solution( work, hash32, mythr );
+      }
+      nonce++;
+   } while ( nonce < max_nonce && !(*restart) );
+   pdata[19] = nonce;
+   *hashes_done = pdata[19] - first_nonce + 1;
+   return 0;
+}

algo/x17/xevan-gate.c

@@ -1,10 +1,5 @@
 #include "xevan-gate.h"
 
-void xevan_set_target( struct work* work, double job_diff )
-{
- work_set_target( work, job_diff / (256.0 * opt_diff_factor) );
-}
-
 bool register_xevan_algo( algo_gate_t* gate )
 {
 #if defined (XEVAN_4WAY)
@@ -17,8 +12,8 @@ bool register_xevan_algo( algo_gate_t* gate )
   gate->hash      = (void*)&xevan_hash;
 #endif
   gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT;
-  gate->set_target = (void*)&xevan_set_target;
   gate->get_max64  = (void*)&get_max64_0xffffLL;
+  opt_target_factor = 256.0;
   return true;
 };
 

algo/x20/x20r-gate.c

@@ -26,9 +26,9 @@ bool register_x20r_algo( algo_gate_t* gate )
   gate->scanhash  = (void*)&scanhash_x20r;
   gate->hash      = (void*)&x20r_hash;
 #endif
-  gate->set_target = (void*)&alt_set_target;
   gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT;
   x20_r_s_getAlgoString = (void*)&x20r_getAlgoString;
+  opt_target_factor = 256.;
   return true;
 };
 

algo/yescrypt/yescrypt.c

@@ -431,18 +431,38 @@ void yescrypt_gate_base(algo_gate_t *gate )
    gate->optimizations = SSE2_OPT | SHA_OPT;
    gate->scanhash   = (void*)&scanhash_yescrypt;
    gate->hash       = (void*)&yescrypt_hash;
-   gate->set_target = (void*)&scrypt_set_target;
+   opt_target_factor = 65536.0;
 }
 
 bool register_yescrypt_algo( algo_gate_t* gate )
 {
    yescrypt_gate_base( gate );
    gate->get_max64  = (void*)&yescrypt_get_max64;
-   yescrypt_client_key = NULL;
-   yescrypt_client_key_len = 0;
-   YESCRYPT_N = 2048;
-   YESCRYPT_R = 8;
+
+   if ( opt_param_n )  YESCRYPT_N = opt_param_n;
+   else                YESCRYPT_N = 2048;
+
+   if ( opt_param_r )  YESCRYPT_R = opt_param_r;
+   else                YESCRYPT_R = 8;
+ 
+   if ( opt_param_key ) 
+   {   
+     yescrypt_client_key = opt_param_key;
+     yescrypt_client_key_len = strlen( opt_param_key );
+   }
+   else
+   {   
+     yescrypt_client_key = NULL;
+     yescrypt_client_key_len = 0;
+   }
+
    YESCRYPT_P = 1;
+
+   applog( LOG_NOTICE,"Yescrypt parameters: N= %d, R= %d.", YESCRYPT_N,
+                                                            YESCRYPT_R );
+   if ( yescrypt_client_key )
+     applog( LOG_NOTICE,"Key= \"%s\"\n", yescrypt_client_key );
+
    return true;
 }
 

algo/yespower/yespower.c

@@ -78,15 +78,34 @@ int64_t yespower_get_max64()
 bool register_yespower_algo( algo_gate_t* gate )
 {
   yespower_params.version = YESPOWER_1_0;
-  yespower_params.N       = 2048;
-  yespower_params.r       = 32;
-  yespower_params.pers    = NULL;
-  yespower_params.perslen = 0;
+
+  if ( opt_param_n )  yespower_params.N = opt_param_n;
+  else                yespower_params.N = 2048;
+
+  if ( opt_param_r )  yespower_params.r = opt_param_r;
+  else                yespower_params.r = 32;
+
+  if ( opt_param_key )
+  {
+     yespower_params.pers = opt_param_key;
+     yespower_params.perslen = strlen( opt_param_key );
+  }
+  else
+  {
+     yespower_params.pers    = NULL;
+     yespower_params.perslen = 0;
+  }
+
+  applog( LOG_NOTICE,"Yespower parameters: N= %d, R= %d.", yespower_params.N,
+                                                           yespower_params.r );
+  if ( yespower_params.pers )
+     applog( LOG_NOTICE,"Key= \"%s\"\n", yespower_params.pers );
+
   gate->optimizations = SSE2_OPT;
   gate->get_max64     = (void*)&yespower_get_max64;
   gate->scanhash      = (void*)&scanhash_yespower;
   gate->hash          = (void*)&yespower_hash;
-  gate->set_target    = (void*)&scrypt_set_target;
+  opt_target_factor = 65536.0;
   return true;
 };
 
@@ -101,7 +120,7 @@ bool register_yespowerr16_algo( algo_gate_t* gate )
   gate->get_max64     = (void*)&yespower_get_max64;
   gate->scanhash      = (void*)&scanhash_yespower;
   gate->hash          = (void*)&yespower_hash;
-  gate->set_target    = (void*)&scrypt_set_target;
+  opt_target_factor = 65536.0;
   return true;
  };
 
@@ -120,13 +139,13 @@ bool register_yescrypt_05_algo( algo_gate_t* gate )
 {
    gate->optimizations = SSE2_OPT | SHA_OPT;
    gate->scanhash   = (void*)&scanhash_yespower;
-   gate->set_target = (void*)&scrypt_set_target;
    gate->get_max64  = (void*)&yescrypt_05_get_max64;
    yespower_params.version = YESPOWER_0_5;
    yespower_params.N       = 2048;
    yespower_params.r       = 8;
    yespower_params.pers    = NULL;
    yespower_params.perslen = 0;
+   opt_target_factor = 65536.0;
    return true;
 }
 
@@ -134,13 +153,13 @@ bool register_yescryptr8_05_algo( algo_gate_t* gate )
 {
    gate->optimizations = SSE2_OPT | SHA_OPT;
    gate->scanhash   = (void*)&scanhash_yespower;
-   gate->set_target = (void*)&scrypt_set_target;
    gate->get_max64  = (void*)&yescrypt_05_get_max64;
    yespower_params.version = YESPOWER_0_5;
    yespower_params.N       = 2048;
    yespower_params.r       = 8;
    yespower_params.pers    = "Client Key";
    yespower_params.perslen = 10;
+   opt_target_factor = 65536.0;
    return true;
 }
 
@@ -148,13 +167,13 @@ bool register_yescryptr16_05_algo( algo_gate_t* gate )
 {
    gate->optimizations = SSE2_OPT | SHA_OPT;
    gate->scanhash   = (void*)&scanhash_yespower;
-   gate->set_target = (void*)&scrypt_set_target;
    gate->get_max64  = (void*)&yescryptr16_05_get_max64;
    yespower_params.version = YESPOWER_0_5;
    yespower_params.N       = 4096;
    yespower_params.r       = 16;
    yespower_params.pers    = NULL;
    yespower_params.perslen = 0;
+   opt_target_factor = 65536.0;
    return true;
 }
 
@@ -162,13 +181,13 @@ bool register_yescryptr32_05_algo( algo_gate_t* gate )
 {
    gate->optimizations = SSE2_OPT | SHA_OPT;
    gate->scanhash   = (void*)&scanhash_yespower;
-   gate->set_target = (void*)&scrypt_set_target;
    gate->get_max64  = (void*)&yescryptr16_05_get_max64;
    yespower_params.version = YESPOWER_0_5;
    yespower_params.N       = 4096;
    yespower_params.r       = 32;
    yespower_params.pers    = "WaviBanana";
    yespower_params.perslen = 10;
+   opt_target_factor = 65536.0;
    return true;
 }
 

configure

@@ -1,6 +1,6 @@
 #! /bin/sh
 # Guess values for system-dependent variables and create Makefiles.
-# Generated by GNU Autoconf 2.69 for cpuminer-opt 3.9.6.2.
+# Generated by GNU Autoconf 2.69 for cpuminer-opt 3.9.8.1.
 #
 #
 # Copyright (C) 1992-1996, 1998-2012 Free Software Foundation, Inc.
@@ -577,8 +577,8 @@ MAKEFLAGS=
 # Identity of this package.
 PACKAGE_NAME='cpuminer-opt'
 PACKAGE_TARNAME='cpuminer-opt'
-PACKAGE_VERSION='3.9.6.2'
-PACKAGE_STRING='cpuminer-opt 3.9.6.2'
+PACKAGE_VERSION='3.9.8.1'
+PACKAGE_STRING='cpuminer-opt 3.9.8.1'
 PACKAGE_BUGREPORT=''
 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.
   # This message is too long to be a string in the A/UX 3.1 sh.
   cat <<_ACEOF
-\`configure' configures cpuminer-opt 3.9.6.2 to adapt to many kinds of systems.
+\`configure' configures cpuminer-opt 3.9.8.1 to adapt to many kinds of systems.
 
 Usage: $0 [OPTION]... [VAR=VALUE]...
 
@@ -1404,7 +1404,7 @@ fi
 
 if test -n "$ac_init_help"; then
   case $ac_init_help in
-     short | recursive ) echo "Configuration of cpuminer-opt 3.9.6.2:";;
+     short | recursive ) echo "Configuration of cpuminer-opt 3.9.8.1:";;
    esac
   cat <<\_ACEOF
 
@@ -1509,7 +1509,7 @@ fi
 test -n "$ac_init_help" && exit $ac_status
 if $ac_init_version; then
   cat <<\_ACEOF
-cpuminer-opt configure 3.9.6.2
+cpuminer-opt configure 3.9.8.1
 generated by GNU Autoconf 2.69
 
 Copyright (C) 2012 Free Software Foundation, Inc.
@@ -2012,7 +2012,7 @@ cat >config.log <<_ACEOF
 This file contains any messages produced by compilers while
 running configure, to aid debugging if configure makes a mistake.
 
-It was created by cpuminer-opt $as_me 3.9.6.2, which was
+It was created by cpuminer-opt $as_me 3.9.8.1, which was
 generated by GNU Autoconf 2.69.  Invocation command line was
 
   $ $0 $@
@@ -2993,7 +2993,7 @@ fi
 
 # Define the identity of the package.
  PACKAGE='cpuminer-opt'
- VERSION='3.9.6.2'
+ VERSION='3.9.8.1'
 
 
 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
 # values after options handling.
 ac_log="
-This file was extended by cpuminer-opt $as_me 3.9.6.2, which was
+This file was extended by cpuminer-opt $as_me 3.9.8.1, which was
 generated by GNU Autoconf 2.69.  Invocation command line was
 
   CONFIG_FILES    = $CONFIG_FILES
@@ -6756,7 +6756,7 @@ _ACEOF
 cat >>$CONFIG_STATUS <<_ACEOF || ac_write_fail=1
 ac_cs_config="`$as_echo "$ac_configure_args" | sed 's/^ //; s/[\\""\`\$]/\\\\&/g'`"
 ac_cs_version="\\
-cpuminer-opt config.status 3.9.6.2
+cpuminer-opt config.status 3.9.8.1
 configured by $0, generated by GNU Autoconf 2.69,
   with options \\"\$ac_cs_config\\"
 

configure.ac

@@ -1,4 +1,4 @@
-AC_INIT([cpuminer-opt], [3.9.6.2])
+AC_INIT([cpuminer-opt], [3.9.8.1])
 
 AC_PREREQ([2.59c])
 AC_CANONICAL_SYSTEM

miner.h

@@ -310,6 +310,7 @@ struct thr_api {
 #define CL_WHT  "\x1B[01;37m" /* white */
 
 void   applog(int prio, const char *fmt, ...);
+void   applog2(int prio, const char *fmt, ...);
 void   restart_threads(void);
 extern json_t *json_rpc_call( CURL *curl, const char *url, const char *userpass,
                 	const char *rpc_req, int *curl_err, int flags );
@@ -331,6 +332,24 @@ extern void diff_to_target(uint32_t *target, double diff);
 double hash_target_ratio( uint32_t* hash, uint32_t* target );
 void   work_set_target_ratio( struct work* work, uint32_t* hash );
 
+struct thr_info {
+        int id;
+        pthread_t pth;
+        pthread_attr_t attr;
+        struct thread_q *q;
+        struct cpu_info cpu;
+};
+
+//struct thr_info *thr_info;
+
+bool   submit_solution( struct work *work, void *hash,
+                        struct thr_info *thr );
+bool   submit_lane_solution( struct work *work, void *hash,
+                             struct thr_info *thr, int lane );
+
+
+//bool submit_work( struct thr_info *thr, const struct work *work_in );
+
 
 void   get_currentalgo( char* buf, int sz );
 bool   has_sha();
@@ -355,7 +374,7 @@ struct work {
 	uint32_t target[8];
 
 	double targetdiff;
-	double shareratio;
+//	double shareratio;
 	double sharediff;
 
 	int height;
@@ -471,6 +490,9 @@ void print_hash_tests(void);
 
 void scale_hash_for_display ( double* hashrate, char* units );
 
+void report_summary_log( bool force );
+
+/*
 struct thr_info {
         int id;
         pthread_t pth;
@@ -478,6 +500,7 @@ struct thr_info {
         struct thread_q *q;
         struct cpu_info cpu;
 };
+*/
 
 struct work_restart {
         volatile uint8_t restart;
@@ -578,6 +601,7 @@ enum algos {
         ALGO_X14,        
         ALGO_X15,       
         ALGO_X16R,
+        ALGO_X16RV2,
         ALGO_X16RT,
         ALGO_X16RT_VEIL,
         ALGO_X16S,
@@ -672,6 +696,7 @@ static const char* const algo_names[] = {
         "x14",
         "x15",
         "x16r",
+        "x16rv2",
         "x16rt",
         "x16rt-veil",
         "x16s",
@@ -729,8 +754,11 @@ extern double stratum_diff;
 extern double net_diff;
 extern double net_hashrate;
 extern int opt_pluck_n;
-extern int opt_scrypt_n;
+extern int opt_param_n;
+extern int opt_param_r;
+extern char* opt_param_key;
 extern double opt_diff_factor;
+extern double opt_target_factor;
 extern bool opt_randomize;
 extern bool allow_mininginfo;
 extern time_t g_work_time;
@@ -796,8 +824,8 @@ Options:\n\
                           neoscrypt     NeoScrypt(128, 2, 1)\n\
                           nist5         Nist5\n\
                           pentablake    5 x blake512\n\
-                          phi1612       phi, LUX coin (original algo)\n\
-                          phi2          LUX (new algo)\n\
+                          phi1612       phi\n\
+                          phi2\n\
 			                 pluck         Pluck:128 (Supcoin)\n\
                           polytimos\n\
                           quark         Quark\n\
@@ -829,7 +857,8 @@ Options:\n\
                           x13sm3        hsr (Hshare)\n\
                           x14           X14\n\
                           x15           X15\n\
-                          x16r          Ravencoin (RVN)\n\
+                          x16r\n\
+                          x16rv2        Ravencoin (RVN)\n\
                           x16rt         Gincoin (GIN)\n\
                           x16rt-veil    Veil (VEIL)\n\
                           x16s          Pigeoncoin (PGN)\n\
@@ -843,6 +872,9 @@ Options:\n\
                           yespower      Cryply\n\
                           yespowerr16   Yenten (YTN)\n\
                           zr5           Ziftr\n\
+  -N, --param-n         N parameter for scrypt based algos\n\
+  -R, --patam-r         R parameter for scrypt based algos\n\
+  -K, --param-key       Key parameter for algos that use it\n\
   -o, --url=URL         URL of mining server\n\
   -O, --userpass=U:P    username:password pair for mining server\n\
   -u, --user=USERNAME   username for mining server\n\
@@ -852,7 +884,7 @@ Options:\n\
   -t, --threads=N       number of miner threads (default: number of processors)\n\
   -r, --retries=N       number of times to retry if a network call fails\n\
                           (default: retry indefinitely)\n\
-  -R, --retry-pause=N   time to pause between retries, in seconds (default: 30)\n\
+      --retry-pause=N   time to pause between retries, in seconds (default: 30)\n\
       --time-limit=N    maximum time [s] to mine before exiting the program.\n\
   -T, --timeout=N       timeout for long poll and stratum (default: 300 seconds)\n\
   -s, --scantime=N      upper bound on time spent scanning current work when\n\
@@ -927,6 +959,7 @@ static struct option const options[] = {
         { "hash-meter", 0, NULL, 1014 },
         { "hide-diff", 0, NULL, 1013 },
         { "help", 0, NULL, 'h' },
+        { "key", 1, NULL, 'K' },
         { "no-gbt", 0, NULL, 1011 },
         { "no-getwork", 0, NULL, 1010 },
         { "no-longpoll", 0, NULL, 1003 },
@@ -936,13 +969,16 @@ static struct option const options[] = {
         { "max-temp", 1, NULL, 1060 },
         { "max-diff", 1, NULL, 1061 },
         { "max-rate", 1, NULL, 1062 },
+        { "param-key", 1, NULL, 'K' },
+        { "param-n", 1, NULL, 'N' },
+        { "param-r", 1, NULL, 'R' },
         { "pass", 1, NULL, 'p' },
         { "protocol", 0, NULL, 'P' },
         { "protocol-dump", 0, NULL, 'P' },
         { "proxy", 1, NULL, 'x' },
         { "quiet", 0, NULL, 'q' },
         { "retries", 1, NULL, 'r' },
-        { "retry-pause", 1, NULL, 'R' },
+        { "retry-pause", 1, NULL, 1025 },
         { "randomize", 0, NULL, 1024 },
         { "scantime", 1, NULL, 's' },
 #ifdef HAVE_SYSLOG_H

simd-utils.h

@@ -175,7 +175,6 @@
 
 // 64 bit vectors
 #include "simd-utils/simd-64.h"
-//#include "simd-utils/intrlv-mmx.h"
 
 #if defined(__SSE2__)
 
@@ -189,6 +188,8 @@
 
 #if defined(__AVX2__)
 
+// Utilities that require AVX2 are defined in simd-256.h.
+
 // Skylake-X has all these
 #if defined(__AVX512VL__) && defined(__AVX512DQ__) && defined(__AVX512BW__)
 

simd-utils/intrlv.h

@@ -477,13 +477,15 @@ static inline void mm256_bswap32_intrlv80_8x32( void *d, void *src )
    __m256i s0 = mm256_bswap_32( casti_m256i( src,0 ) );
    __m256i s1 = mm256_bswap_32( casti_m256i( src,1 ) );
    __m128i s2 = mm128_bswap_32( casti_m128i( src,4 ) );
-  const __m256i zero  = m256_zero;
+//  const __m256i zero  = m256_zero;
   const __m256i one   = m256_one_32;
   const __m256i two   = _mm256_add_epi32( one, one );
   const __m256i three = _mm256_add_epi32( two, one );
   const __m256i four  = _mm256_add_epi32( two, two );
 
-  casti_m256i( d, 0 ) = _mm256_permutevar8x32_epi32( s0, zero  );
+  casti_m256i( d, 0 ) = _mm256_broadcastd_epi32(
+                             _mm256_castsi256_si128( s0 ) );
+//  casti_m256i( d, 0 ) = _mm256_permutevar8x32_epi32( s0, m256_zero  );
   casti_m256i( d, 1 ) = _mm256_permutevar8x32_epi32( s0, one   );
   casti_m256i( d, 2 ) = _mm256_permutevar8x32_epi32( s0, two   );
   casti_m256i( d, 3 ) = _mm256_permutevar8x32_epi32( s0, three );
@@ -494,7 +496,9 @@ static inline void mm256_bswap32_intrlv80_8x32( void *d, void *src )
                                        _mm256_add_epi32( four, two   ) );
   casti_m256i( d, 7 ) = _mm256_permutevar8x32_epi32( s0,
                                        _mm256_add_epi32( four, three ) );
-  casti_m256i( d, 8 ) = _mm256_permutevar8x32_epi32( s1, zero  );
+  casti_m256i( d, 8 ) = _mm256_broadcastd_epi32(
+                             _mm256_castsi256_si128( s1 ) );
+//  casti_m256i( d, 8 ) = _mm256_permutevar8x32_epi32( s1, m256_zero  );
   casti_m256i( d, 9 ) = _mm256_permutevar8x32_epi32( s1, one   );
   casti_m256i( d,10 ) = _mm256_permutevar8x32_epi32( s1, two   );
   casti_m256i( d,11 ) = _mm256_permutevar8x32_epi32( s1, three );
@@ -505,8 +509,9 @@ static inline void mm256_bswap32_intrlv80_8x32( void *d, void *src )
                                        _mm256_add_epi32( four, two   ) );
   casti_m256i( d,15 ) = _mm256_permutevar8x32_epi32( s1,
                                        _mm256_add_epi32( four, three ) );
-  casti_m256i( d,16 ) = _mm256_permutevar8x32_epi32(
-                             _mm256_castsi128_si256( s2 ), zero  );
+  casti_m256i( d,16 ) = _mm256_broadcastd_epi32(     s2 );
+//  casti_m256i( d,16 ) = _mm256_permutevar8x32_epi32(
+//                             _mm256_castsi128_si256( s2 ), m256_zero  );
   casti_m256i( d,17 ) = _mm256_permutevar8x32_epi32(
                              _mm256_castsi128_si256( s2 ), one   );
   casti_m256i( d,18 ) = _mm256_permutevar8x32_epi32(
@@ -682,7 +687,9 @@ static inline void mm512_bswap32_intrlv80_16x32( void *d, void *src )
   const __m512i three = _mm512_add_epi32( two,   one   );
         __m512i     x = _mm512_add_epi32( three, three );
 
-  casti_m512i( d, 0 ) = _mm512_permutexvar_epi32( s0, m512_zero );
+  casti_m512i( d, 0 ) = _mm512_broadcastd_epi32(
+                          _mm512_castsi512_si128( s0 ) );
+//  casti_m512i( d, 0 ) = _mm512_permutexvar_epi32( s0, m512_zero );
   casti_m512i( d, 1 ) = _mm512_permutexvar_epi32( s0, one   );
   casti_m512i( d, 2 ) = _mm512_permutexvar_epi32( s0, two   );
   casti_m512i( d, 3 ) = _mm512_permutexvar_epi32( s0, three );
@@ -709,8 +716,9 @@ static inline void mm512_bswap32_intrlv80_16x32( void *d, void *src )
                                     _mm512_add_epi32( x,      two   ) );
   casti_m512i( d,15 ) = _mm512_permutexvar_epi32( s0,
                                     _mm512_add_epi32( x,      three ) );
-  casti_m512i( d,16 ) = _mm512_permutexvar_epi32(
-                          _mm512_castsi128_si512( s1 ), m512_zero );
+  casti_m512i( d,16 ) = _mm512_broadcastd_epi32(  s1 );
+//  casti_m512i( d,16 ) = _mm512_permutexvar_epi32(
+//                          _mm512_castsi128_si512( s1 ), m512_zero );
   casti_m512i( d,17 ) = _mm512_permutexvar_epi32(
                           _mm512_castsi128_si512( s1 ), one  );
   casti_m512i( d,18 ) = _mm512_permutexvar_epi32(
@@ -874,17 +882,6 @@ static inline void extr_lane_4x64( void *d, const void *s,
    ((uint64_t*)d)[ 5] = ((uint64_t*)s)[ lane+20 ];
    ((uint64_t*)d)[ 6] = ((uint64_t*)s)[ lane+24 ];
    ((uint64_t*)d)[ 7] = ((uint64_t*)s)[ lane+28 ];
-/*
-   if ( bit_len <= 256 ) return;
-   ((uint64_t*)d)[ 8] = ((uint64_t*)s)[ lane+32 ];
-   ((uint64_t*)d)[ 9] = ((uint64_t*)s)[ lane+36 ];
-   ((uint64_t*)d)[10] = ((uint64_t*)s)[ lane+40 ];
-   ((uint64_t*)d)[11] = ((uint64_t*)s)[ lane+44 ];
-   ((uint64_t*)d)[12] = ((uint64_t*)s)[ lane+48 ];
-   ((uint64_t*)d)[13] = ((uint64_t*)s)[ lane+52 ];
-   ((uint64_t*)d)[14] = ((uint64_t*)s)[ lane+56 ];
-   ((uint64_t*)d)[15] = ((uint64_t*)s)[ lane+60 ];
-*/
 }
 
 #if defined(__AVX2__)
@@ -991,17 +988,6 @@ static inline void extr_lane_8x64( void *d, const void *s,
    ((uint64_t*)d)[ 5] = ((uint64_t*)s)[ lane+ 40 ];
    ((uint64_t*)d)[ 6] = ((uint64_t*)s)[ lane+ 48 ];
    ((uint64_t*)d)[ 7] = ((uint64_t*)s)[ lane+ 56 ];
-/*
-   if ( bit_len <= 256 ) return;
-   ((uint64_t*)d)[ 8] = ((uint64_t*)s)[ lane+ 64 ];
-   ((uint64_t*)d)[ 9] = ((uint64_t*)s)[ lane+ 72 ];
-   ((uint64_t*)d)[10] = ((uint64_t*)s)[ lane+ 80 ];
-   ((uint64_t*)d)[11] = ((uint64_t*)s)[ lane+ 88 ];
-   ((uint64_t*)d)[12] = ((uint64_t*)s)[ lane+ 96 ];
-   ((uint64_t*)d)[13] = ((uint64_t*)s)[ lane+104 ];
-   ((uint64_t*)d)[14] = ((uint64_t*)s)[ lane+112 ];
-   ((uint64_t*)d)[15] = ((uint64_t*)s)[ lane+120 ];
-*/
 }
 
 #if defined(__AVX512F__) && defined(__AVX512VL__)
@@ -1009,15 +995,17 @@ static inline void extr_lane_8x64( void *d, const void *s,
 static inline void mm512_bswap32_intrlv80_8x64( void *dst, void *src )
 {
    __m512i *d = (__m512i*)dst;
-   __m512i s0 = mm512_bswap_32( casti_m512i(src, 0 ) );
-   __m128i s1 = mm128_bswap_32( casti_m128i(src, 4 ) );
+   __m512i s0 = mm512_bswap_32( casti_m512i( src, 0 ) );
+   __m128i s1 = mm128_bswap_32( casti_m128i( src, 4 ) );
 //  const __m512i zero   = m512_zero;
   const __m512i one    = m512_one_64;
   const __m512i two    = _mm512_add_epi64( one, one );
   const __m512i three  = _mm512_add_epi64( two, one );
   const __m512i four   = _mm512_add_epi64( two, two );
 
-  d[0] = _mm512_permutexvar_epi64( s0, m512_zero );
+  d[0] = _mm512_broadcastq_epi64(
+           _mm512_castsi512_si128( s0 ) );
+//  d[0] = _mm512_permutexvar_epi64( s0, m512_zero );
   d[1] = _mm512_permutexvar_epi64( s0, one  );
   d[2] = _mm512_permutexvar_epi64( s0, two  );
   d[3] = _mm512_permutexvar_epi64( s0, three  );
@@ -1025,8 +1013,9 @@ static inline void mm512_bswap32_intrlv80_8x64( void *dst, void *src )
   d[5] = _mm512_permutexvar_epi64( s0, _mm512_add_epi64( four, one   ) );
   d[6] = _mm512_permutexvar_epi64( s0, _mm512_add_epi64( four, two   ) );
   d[7] = _mm512_permutexvar_epi64( s0, _mm512_add_epi64( four, three ) );
-  d[8] = _mm512_permutexvar_epi64(
-           _mm512_castsi128_si512( s1 ), m512_zero );
+  d[8] = _mm512_broadcastq_epi64(  s1 );
+//  d[8] = _mm512_permutexvar_epi64(
+//           _mm512_castsi128_si512( s1 ), m512_zero );
   d[9] = _mm512_permutexvar_epi64(
            _mm512_castsi128_si512( s1 ), one  );
 }

simd-utils/simd-128.h

@@ -42,9 +42,11 @@
 
 static inline __m128i m128_one_128_fn()
 {
+   register uint64_t one = 1;
    register __m128i a;
-   asm( "movq $1, %0\n\t"
-        : "=x"(a) );
+   asm( "movq %1, %0\n\t"
+        : "=x"(a)
+        : "r" (one) );
    return a;
 }
 #define m128_one_128    m128_one_128_fn()
@@ -54,9 +56,9 @@ static inline __m128i m128_one_64_fn()
   register uint64_t one = 1;
   register __m128i a;
   asm( "movq %1, %0\n\t"
-       : "=x"(a)
-       : "r"(one) );
-  return _mm_shuffle_epi32( a, 0x04 );
+       : "=x" (a)
+       : "r"  (one) );
+  return _mm_shuffle_epi32( a, 0x44 );
 }
 #define m128_one_64    m128_one_64_fn()
 
@@ -65,8 +67,8 @@ static inline __m128i m128_one_32_fn()
   register uint32_t one = 1;
   register __m128i a;
   asm( "movd %1, %0\n\t"
-       : "=x"(a)
-       : "r"(one) );
+       : "=x" (a)
+       : "r"  (one) );
   return _mm_shuffle_epi32( a, 0x00 );
 }
 #define m128_one_32    m128_one_32_fn()
@@ -76,8 +78,8 @@ static inline __m128i m128_one_16_fn()
   register uint32_t one = 0x00010001;
   register __m128i a;
   asm( "movd %1, %0\n\t"
-       : "=x"(a)
-       : "r"(one) );
+       : "=x" (a)
+       : "r"  (one) );
   return _mm_shuffle_epi32( a, 0x00 );
 }
 #define m128_one_16    m128_one_16_fn()
@@ -87,8 +89,8 @@ static inline __m128i m128_one_8_fn()
   register uint32_t one = 0x01010101;
   register __m128i a;
   asm( "movd %1, %0\n\t"
-       : "=x"(a)
-       : "r"(one) );
+       : "=x" (a)
+       : "r"  (one) );
   return _mm_shuffle_epi32( a, 0x00 );
 }
 #define m128_one_8    m128_one_8_fn()
@@ -97,7 +99,7 @@ static inline __m128i m128_neg1_fn()
 {
    __m128i a;
    asm( "pcmpeqd %0, %0\n\t"
-        : "=x"(a) );
+        : "=x" (a) );
    return a;
 }
 #define m128_neg1    m128_neg1_fn()
@@ -108,7 +110,7 @@ static inline __m128i mm128_mov64_128( uint64_t n )
   register __m128i a;
   asm( "movq %1, %0\n\t"
        : "=x" (a)
-       : "r" (n) );
+       : "r"  (n) );
   return  a;
 }
 
@@ -117,7 +119,7 @@ static inline __m128i mm128_mov32_128( uint32_t n )
   register __m128i a;
   asm( "movd %1, %0\n\t"
        : "=x" (a)
-       : "r" (n) );
+       : "r"  (n) );
   return  a;
 }
 
@@ -126,7 +128,7 @@ static inline uint64_t mm128_mov128_64( __m128i a )
   register uint64_t n;
   asm( "movq %1, %0\n\t"
        : "=x" (n)
-       : "r" (a) );
+       : "r"  (a) );
   return  n;
 }
 
@@ -135,10 +137,28 @@ static inline uint32_t mm128_mov128_32( __m128i a )
   register uint32_t n;
   asm( "movd %1, %0\n\t"
        : "=x" (n)
-       : "r" (a) );
+       : "r"  (a) );
   return  n;
 }
 
+static inline __m128i m128_const1_64( const uint64_t n )
+{
+  register __m128i a;
+  asm( "movq %1, %0\n\t"
+       : "=x" (a)
+       : "r"  (n) );
+  return _mm_shuffle_epi32( a, 0x44 );
+}
+
+static inline __m128i m128_const1_32( const uint32_t n )
+{
+  register __m128i a;
+  asm( "movd %1, %0\n\t"
+       : "=x" (a)
+       : "r"  (n) );
+  return _mm_shuffle_epi32( a, 0x00 );
+}
+
 #if defined(__SSE41__)
 
 // alternative to _mm_set_epi64x, doesn't use mem,
@@ -148,11 +168,12 @@ static inline __m128i m128_const_64( const uint64_t hi, const uint64_t lo )
    register __m128i a;
    asm( "movq %2, %0\n\t"
         "pinsrq $1, %1, %0\n\t"
-        : "=x"(a)
-        : "r"(hi), "r"(lo) );
+        : "=x" (a)
+        : "r"  (hi), "r" (lo) );
    return a;
 }
 
+/*
 static inline __m128i m128_const1_64( const uint64_t n )
 {
    register __m128i a;
@@ -162,13 +183,13 @@ static inline __m128i m128_const1_64( const uint64_t n )
         : "r"(n) );
    return a;
 }
-
+*/
 #else
 
 // #define m128_one_128   _mm_set_epi64x( 0ULL, 1ULL )
 
 #define m128_const_64  _mm_set_epi64x
-#define m128_const1_64 _mm_set1_epi64x
+// #define m128_const1_64 _mm_set1_epi64x
 
 #endif
 
@@ -263,46 +284,6 @@ do { \
 #endif
 
 
-// Gather and scatter data.
-// Surprise, they don't use vector instructions. Several reasons why.
-// Since scalar data elements are being manipulated scalar instructions
-// are most appropriate and can bypass vector registers. They are faster
-// and more efficient on a per instruction basis due to the higher clock
-// speed and greater avaiability of execution resources. It's good for
-// interleaving data buffers for parallel processing.
-// May suffer overhead if data is already in a vector register. This can
-// usually be easilly avoided by the coder. Sometimes _mm_set is simply better.
-// These macros are likely to be used when transposing matrices rather than
-// conversions of a single vector.
-
-// Gather data elements into contiguous memory for vector use.
-// Source args are appropriately sized value integers, destination arg  is a
-// type agnostic pointer.
-// Vector alignment is not required, though likely. Appropriate integer
-// alignment satisfies these macros.
-
-// rewrite using insert
-#define mm128_gather_64( d, s0, s1 ) \
-    ((uint64_t*)d)[0] = (uint64_t)s0; \
-    ((uint64_t*)d)[1] = (uint64_t)s1;
-
-#define mm128_gather_32( d, s0, s1, s2, s3 ) \
-    ((uint32_t*)d)[0] = (uint32_t)s0; \
-    ((uint32_t*)d)[1] = (uint32_t)s1; \
-    ((uint32_t*)d)[2] = (uint32_t)s2; \
-    ((uint32_t*)d)[3] = (uint32_t)s3;
-
-// Scatter data from contiguous memory.
-#define mm128_scatter_64( d0, d1, s ) \
-   *( (uint64_t*)d0) = ((uint64_t*)s)[0]; \
-   *( (uint64_t*)d1) = ((uint64_t*)s)[1]; 
-
-#define mm128_scatter_32( d0, d1, d2, d3, s ) \
-   *( (uint32_t*)d0) = ((uint32_t*)s)[0]; \
-   *( (uint32_t*)d1) = ((uint32_t*)s)[1]; \
-   *( (uint32_t*)d2) = ((uint32_t*)s)[2]; \
-   *( (uint32_t*)d3) = ((uint32_t*)s)[3];
-
 // Memory functions
 // Mostly for convenience, avoids calculating bytes.
 // Assumes data is alinged and integral.
@@ -565,57 +546,73 @@ do { \
 
 #define mm128_ror1x64_256( v1, v2 ) \
 do { \
-   __m128i t  = _mm_srli_si128( v1, 8 ) | _mm_slli_si128( v2, 8 ); \
-           v2 = _mm_srli_si128( v2, 8 ) | _mm_slli_si128( v1, 8 ); \
+   __m128i t  = _mm_or_si128( _mm_srli_si128( v1, 8 ), \
+                              _mm_slli_si128( v2, 8 ) ); \
+           v2 = _mm_or_si128( _mm_srli_si128( v2, 8 ), \
+                              _mm_slli_si128( v1, 8 ) ); \
            v1 = t; \
 } while(0)
 
 #define mm128_rol1x64_256( v1, v2 ) \
 do { \
-   __m128i t  = _mm_slli_si128( v1, 8 ) | _mm_srli_si128( v2, 8 ); \
-           v2 = _mm_slli_si128( v2, 8 ) | _mm_srli_si128( v1, 8 ); \
+   __m128i t  = _mm_or_si128( _mm_slli_si128( v1, 8 ), \
+                              _mm_srli_si128( v2, 8 ) ); \
+           v2 = _mm_or_si128( _mm_slli_si128( v2, 8 ), \
+                              _mm_srli_si128( v1, 8 ) ); \
            v1 = t; \
 } while(0)
 
 #define mm128_ror1x32_256( v1, v2 ) \
 do { \
-   __m128i t  = _mm_srli_si128( v1, 4 ) | _mm_slli_si128( v2, 12 ); \
-           v2 = _mm_srli_si128( v2, 4 ) | _mm_slli_si128( v1, 12 ); \
+   __m128i t  = _mm_or_si128( _mm_srli_si128( v1, 4 ), \
+                              _mm_slli_si128( v2, 12 ) ); \
+           v2 = _mm_or_si128( _mm_srli_si128( v2, 4 ), \
+                              _mm_slli_si128( v1, 12 ) ); \
            v1 = t; \
 } while(0)
 
 #define mm128_rol1x32_256( v1, v2 ) \
 do { \
-   __m128i t  = _mm_slli_si128( v1, 4 ) | _mm_srli_si128( v2, 12 ); \
-           v2 = _mm_slli_si128( v2, 4 ) | _mm_srli_si128( v1, 12 ); \
+   __m128i t  = _mm_or_si128( _mm_slli_si128( v1, 4 ), \
+                              _mm_srli_si128( v2, 12 ) ); \
+           v2 = _mm_or_si128( _mm_slli_si128( v2, 4 ), \
+                              _mm_srli_si128( v1, 12 ) ); \
            v1 = t; \
 } while(0)
 
 #define mm128_ror1x16_256( v1, v2 ) \
 do { \
-   __m128i t  = _mm_srli_si128( v1, 2 ) | _mm_slli_si128( v2, 14 ); \
-           v2 = _mm_srli_si128( v2, 2 ) | _mm_slli_si128( v1, 14 ); \
+   __m128i t  = _mm_or_si128( _mm_srli_si128( v1, 2 ), \
+                              _mm_slli_si128( v2, 14 ) ); \
+           v2 = _mm_or_si128( _mm_srli_si128( v2, 2 ), \
+                              _mm_slli_si128( v1, 14 ) ); \
            v1 = t; \
 } while(0)
 
 #define mm128_rol1x16_256( v1, v2 ) \
 do { \
-   __m128i t  = _mm_slli_si128( v1, 2 ) | _mm_srli_si128( v2, 14 ); \
-           v2 = _mm_slli_si128( v2, 2 ) | _mm_srli_si128( v1, 14 ); \
+   __m128i t  = _mm_or_si128( _mm_slli_si128( v1, 2 ), \
+                              _mm_srli_si128( v2, 14 ) ); \
+           v2 = _mm_or_si128( _mm_slli_si128( v2, 2 ), \
+                              _mm_srli_si128( v1, 14 ) ); \
            v1 = t; \
 } while(0)
 
 #define mm128_ror1x8_256( v1, v2 ) \
 do { \
-   __m128i t  = _mm_srli_si128( v1, 1 ) | _mm_slli_si128( v2, 15 ); \
-           v2 = _mm_srli_si128( v2, 1 ) | _mm_slli_si128( v1, 15 ); \
+   __m128i t  = _mm_or_si128( _mm_srli_si128( v1, 1 ), \
+                              _mm_slli_si128( v2, 15 ) ); \
+           v2 = _mm_or_si128( _mm_srli_si128( v2, 1 ), \
+                              _mm_slli_si128( v1, 15 ) ); \
            v1 = t; \
 } while(0)
 
 #define mm128_rol1x8_256( v1, v2 ) \
 do { \
-   __m128i t  = _mm_slli_si128( v1, 1 ) | _mm_srli_si128( v2, 15 ); \
-           v2 = _mm_slli_si128( v2, 1 ) | _mm_srli_si128( v1, 15 ); \
+   __m128i t  = _mm_or_si128( _mm_slli_si128( v1, 1 ), \
+                              _mm_srli_si128( v2, 15 ) ); \
+           v2 = _mm_or_si128( _mm_slli_si128( v2, 1 ), \
+                              _mm_srli_si128( v1, 15 ) ); \
            v1 = t; \
 } while(0)
 

simd-utils/simd-256.h

@@ -39,11 +39,20 @@ static inline __m256i m256_const1_64( uint64_t i )
 {
    register __m128i a;
    asm( "movq %1, %0\n\t"
-        : "=x"(a)
-        : "r"(i) );
+      : "=x" (a)
+      : "r"  (i) );
    return _mm256_broadcastq_epi64( a );
 }
 
+static inline __m256i m256_const1_32( uint32_t i )
+{
+   register __m128i a;
+   asm( "movd %1, %0\n\t"
+      : "=x" (a)
+      : "r"  (i) );
+   return _mm256_broadcastd_epi32( a );
+}
+
 #if defined(__AVX2__)
 
 // Don't call the frunction directly, use the macro to make appear like
@@ -142,7 +151,7 @@ do { \
   __m128i hi = _mm256_extracti128_si256( src, 1 ); \
   a0 = mm256_mov256_64( src ); \
   a1 = _mm_extract_epi64( _mm256_castsi256_si128( src ), 1 ); \
-  a2 = _mm_extract_epi64( hi, 0 ); \
+  a2 = mm128_mov128_64( hi ); \
   a3 = _mm_extract_epi64( hi, 1 ); \
 } while(0)
 
@@ -246,44 +255,6 @@ static inline __m256i mm256_mov32_256( uint32_t n )
 #define casto_m256i(p,o) (((__m256i*)(p))+(o))
 
 
-// Gather scatter
-
-#define mm256_gather_64( d, s0, s1, s2, s3 ) \
-    ((uint64_t*)(d))[0] = (uint64_t)(s0); \
-    ((uint64_t*)(d))[1] = (uint64_t)(s1); \
-    ((uint64_t*)(d))[2] = (uint64_t)(s2); \
-    ((uint64_t*)(d))[3] = (uint64_t)(s3);
-
-#define mm256_gather_32( d, s0, s1, s2, s3, s4, s5, s6, s7 ) \
-    ((uint32_t*)(d))[0] = (uint32_t)(s0); \
-    ((uint32_t*)(d))[1] = (uint32_t)(s1); \
-    ((uint32_t*)(d))[2] = (uint32_t)(s2); \
-    ((uint32_t*)(d))[3] = (uint32_t)(s3); \
-    ((uint32_t*)(d))[4] = (uint32_t)(s4); \
-    ((uint32_t*)(d))[5] = (uint32_t)(s5); \
-    ((uint32_t*)(d))[6] = (uint32_t)(s6); \
-    ((uint32_t*)(d))[7] = (uint32_t)(s7);
-
-
-// Scatter data from contiguous memory.
-// All arguments are pointers
-#define mm256_scatter_64( d0, d1, d2, d3, s ) \
-   *((uint64_t*)(d0)) = ((uint64_t*)(s))[0]; \
-   *((uint64_t*)(d1)) = ((uint64_t*)(s))[1]; \
-   *((uint64_t*)(d2)) = ((uint64_t*)(s))[2]; \
-   *((uint64_t*)(d3)) = ((uint64_t*)(s))[3];
-
-#define mm256_scatter_32( d0, d1, d2, d3, d4, d5, d6, d7, s ) \
-   *((uint32_t*)(d0)) = ((uint32_t*)(s))[0]; \
-   *((uint32_t*)(d1)) = ((uint32_t*)(s))[1]; \
-   *((uint32_t*)(d2)) = ((uint32_t*)(s))[2]; \
-   *((uint32_t*)(d3)) = ((uint32_t*)(s))[3]; \
-   *((uint32_t*)(d4)) = ((uint32_t*)(s))[4]; \
-   *((uint32_t*)(d5)) = ((uint32_t*)(s))[5]; \
-   *((uint32_t*)(d6)) = ((uint32_t*)(s))[6]; \
-   *((uint32_t*)(d7)) = ((uint32_t*)(s))[7];
-
-
 //
 // Memory functions
 // n = number of 256 bit (32 byte) vectors

simd-utils/simd-512.h

@@ -47,7 +47,6 @@
 // _mm512_setzero_si512 uses xor instruction. If needed frequently
 // in a function is it better to define a register variable (const?)
 // initialized to zero.
-// It isn't clear to me yet how set or set1 actually work.
 
 #define m512_zero       _mm512_setzero_si512()
 #define m512_one_512    _mm512_set_epi64(  0ULL, 0ULL, 0ULL, 0ULL, \
@@ -60,10 +59,18 @@
 //#define m512_one_8      _mm512_set1_epi8(  1U )
 //#define m512_neg1       _mm512_set1_epi64( 0xFFFFFFFFFFFFFFFFULL )
 
-#define mi512_const_64( i7, i6, i5, i4, i3, i2, i1, i0 ) \
-  _mm512_inserti64x4( _mm512_castsi512_si256( m256_const_64( i3.i2,i1,i0 ) ), \
+#define m512_const_64( i7, i6, i5, i4, i3, i2, i1, i0 ) \
+  _mm512_inserti64x4( _mm512_castsi256_si512( m256_const_64( i3,i2,i1,i0 ) ), \
                       m256_const_64( i7,i6,i5,i4 ), 1 )
-#define m512_const1_64( i ) m256_const_64( i, i, i, i, i, i, i, i )
+
+static inline __m512i m512_const1_64( uint64_t i )
+{
+   register __m128i a;
+   asm( "movq %1, %0\n\t"
+        : "=x"(a)
+        : "r"(i) );
+   return _mm512_broadcastq_epi64( a );
+}
 
 static inline __m512i m512_one_64_fn()
 {
@@ -135,15 +142,12 @@ static inline __m512i m512_neg1_fn()
 #define mm512_negate_32( x ) _mm512_sub_epi32( m512_zero, x )  
 #define mm512_negate_16( x ) _mm512_sub_epi16( m512_zero, x )  
 
-
-
 #define mm256_extr_lo256_512( a ) _mm512_castsi512_si256( a )
 #define mm256_extr_hi256_512( a ) _mm512_extracti64x4_epi64( a, 1 )
 
 #define mm128_extr_lo128_512( a ) _mm512_castsi512_si256( a )
 
 
-
 //
 // Pointer casting
 
@@ -163,73 +167,9 @@ static inline __m512i m512_neg1_fn()
 // returns p+o as pointer to vector
 #define casto_m512i(p,o) (((__m512i*)(p))+(o))
 
-// Gather scatter
-
-#define mm512_gather_64( d, s0, s1, s2, s3, s4, s5, s6, s7 ) \
-    ((uint64_t*)(d))[0] = (uint64_t)(s0); \
-    ((uint64_t*)(d))[1] = (uint64_t)(s1); \
-    ((uint64_t*)(d))[2] = (uint64_t)(s2); \
-    ((uint64_t*)(d))[3] = (uint64_t)(s3); \
-    ((uint64_t*)(d))[4] = (uint64_t)(s4); \
-    ((uint64_t*)(d))[5] = (uint64_t)(s5); \
-    ((uint64_t*)(d))[6] = (uint64_t)(s6); \
-    ((uint64_t*)(d))[7] = (uint64_t)(s7); 
-
-
-#define mm512_gather_32( d, s00, s01, s02, s03, s04, s05, s06, s07, \
-                   s08, s09, s10, s11, s12, s13, s14, s15 ) \
-    ((uint32_t*)(d))[ 0] = (uint32_t)(s00); \
-    ((uint32_t*)(d))[ 1] = (uint32_t)(s01); \
-    ((uint32_t*)(d))[ 2] = (uint32_t)(s02); \
-    ((uint32_t*)(d))[ 3] = (uint32_t)(s03); \
-    ((uint32_t*)(d))[ 4] = (uint32_t)(s04); \
-    ((uint32_t*)(d))[ 5] = (uint32_t)(s05); \
-    ((uint32_t*)(d))[ 6] = (uint32_t)(s06); \
-    ((uint32_t*)(d))[ 7] = (uint32_t)(s07); \
-    ((uint32_t*)(d))[ 8] = (uint32_t)(s08); \
-    ((uint32_t*)(d))[ 9] = (uint32_t)(s09); \
-    ((uint32_t*)(d))[10] = (uint32_t)(s10); \
-    ((uint32_t*)(d))[11] = (uint32_t)(s11); \
-    ((uint32_t*)(d))[12] = (uint32_t)(s12); \
-    ((uint32_t*)(d))[13] = (uint32_t)(s13); \
-    ((uint32_t*)(d))[13] = (uint32_t)(s14); \
-    ((uint32_t*)(d))[15] = (uint32_t)(s15);
-
-// Scatter data from contiguous memory.
-// All arguments are pointers
-#define mm512_scatter_64( d0, d1, d2, d3, d4, d5, d6, d7, s ) \
-   *((uint64_t*)(d0)) = ((uint64_t*)(s))[0]; \
-   *((uint64_t*)(d1)) = ((uint64_t*)(s))[1]; \
-   *((uint64_t*)(d2)) = ((uint64_t*)(s))[2]; \
-   *((uint64_t*)(d3)) = ((uint64_t*)(s))[3]; \
-   *((uint64_t*)(d4)) = ((uint64_t*)(s))[4]; \
-   *((uint64_t*)(d5)) = ((uint64_t*)(s))[5]; \
-   *((uint64_t*)(d6)) = ((uint64_t*)(s))[6]; \
-   *((uint64_t*)(d7)) = ((uint64_t*)(s))[7];
-
-
-#define mm512_scatter_32( d00, d01, d02, d03, d04, d05, d06, d07, \
-                     d08, d09, d10, d11, d12, d13, d14, d15, s ) \
-   *((uint32_t*)(d00)) = ((uint32_t*)(s))[ 0]; \
-   *((uint32_t*)(d01)) = ((uint32_t*)(s))[ 1]; \
-   *((uint32_t*)(d02)) = ((uint32_t*)(s))[ 2]; \
-   *((uint32_t*)(d03)) = ((uint32_t*)(s))[ 3]; \
-   *((uint32_t*)(d04)) = ((uint32_t*)(s))[ 4]; \
-   *((uint32_t*)(d05)) = ((uint32_t*)(s))[ 5]; \
-   *((uint32_t*)(d06)) = ((uint32_t*)(s))[ 6]; \
-   *((uint32_t*)(d07)) = ((uint32_t*)(s))[ 7]; \
-   *((uint32_t*)(d00)) = ((uint32_t*)(s))[ 8]; \
-   *((uint32_t*)(d01)) = ((uint32_t*)(s))[ 9]; \
-   *((uint32_t*)(d02)) = ((uint32_t*)(s))[10]; \
-   *((uint32_t*)(d03)) = ((uint32_t*)(s))[11]; \
-   *((uint32_t*)(d04)) = ((uint32_t*)(s))[12]; \
-   *((uint32_t*)(d05)) = ((uint32_t*)(s))[13]; \
-   *((uint32_t*)(d06)) = ((uint32_t*)(s))[14]; \
-   *((uint32_t*)(d07)) = ((uint32_t*)(s))[15];
 
 // Add 4 values, fewer dependencies than sequential addition.
 
-
 #define mm512_add4_64( a, b, c, d ) \
    _mm512_add_epi64( _mm512_add_epi64( a, b ), _mm512_add_epi64( c, d ) )
 
@@ -246,7 +186,6 @@ static inline __m512i m512_neg1_fn()
    _mm512_xor_si512( _mm512_xor_si256( a, b ), _mm512_xor_si256( c, d ) )
 
 
-
 //
 // Bit rotations.
 
@@ -321,23 +260,26 @@ static inline __m512i m512_neg1_fn()
    _mm512_permutex_epi64( v, m512_const_64( 0,1,2,3,4,5,6,7 ) )
 
 #define mm512_invert_32( v ) \
-   _mm512_permutexvar_epi32( v, _mm512_set_epi32( \
-                      0, 1, 2, 3, 4, 5, 6, 7,   8, 9,10,11,12,13,14,15 ) )
+   _mm512_permutexvar_epi32( v, m512_const_64( \
+                      0x0000000000000001,0x0000000200000003, \
+                      0x0000000400000005,0x0000000600000007, \
+                      0x0000000800000009,0x0000000a0000000b, \
+                      0x0000000c0000000d,0x0000000e0000000f ) )
 
 
 #define mm512_invert_16( v ) \
-   _mm512_permutexvar_epi16( v, _mm512_set_epi32( \
-                       0x00000001, 0x00020003, 0x00040005, 0x00060007, \
-                       0x00080009, 0x000A000B, 0x000C000D, 0x000E000F, \
-                       0x00100011, 0x00120013, 0x00140015, 0x00160017, \
-                       0x00180019, 0x001A001B, 0x001C001D, 0x001E001F ) )
+   _mm512_permutexvar_epi16( v, m512_const_64( \
+                       0x0000000100020003, 0x0004000500060007, \
+                       0x00080009000A000B, 0x000C000D000E000F, \
+                       0x0010001100120013, 0x0014001500160017, \
+                       0x00180019001A001B, 0x001C001D001E001F ) )
 
 #define mm512_invert_8(  v ) \
-   _mm512_permutexvar_epi8( v, _mm512_set_epi32( \
-                       0x00010203, 0x04050607, 0x08090A0B, 0x0C0D0E0F, \
-                       0x10111213, 0x14151617, 0x18191A1B, 0x1C1D1E1F, \
-                       0x20212223, 0x24252627, 0x28292A2B, 0x2C2D2E2F, \
-                       0x30313233, 0x34353637, 0x38393A3B, 0x3C3D3E3F ) )
+   _mm512_permutexvar_epi8( v, m512_const_64( \
+                       0x0001020304050607, 0x08090A0B0C0D0E0F, \
+                       0x1011121314151617, 0x18191A1B1C1D1E1F, \
+                       0x2021222324252627, 0x28292A2B2C2D2E2F, \
+                       0x3031323334353637, 0x38393A3B3C3D3E3F ) )
 
 //
 // Rotate elements within 256 bit lanes of 512 bit vector.
@@ -351,38 +293,46 @@ static inline __m512i m512_neg1_fn()
 
 // Rotate 256 bit lanes by one 32 bit element
 #define mm512_ror1x32_256( v ) \
-   _mm512_permutexvar_epi32( v, _mm512_set_epi32( \
-                      8,15,14,13,12,11,10, 9,   0, 7, 6, 5, 4, 3, 2, 1 ) )
+   _mm512_permutexvar_epi32( v, m512_const_64( \
+                      0x000000080000000f, 0x0000000e0000000d, \
+                      0x0000000c0000000b, 0x0000000a00000009, \
+                      0x0000000000000007, 0x0000000600000005, \
+                      0x0000000400000003, 0x0000000200000001 ) )
+            
 #define mm512_rol1x32_256( v ) \
-   _mm512_permutexvar_epi32( v, _mm512_set_epi32( \
-                     14,13,12,11,10, 9, 8,15,   6, 5, 4, 3, 2, 1, 0, 7 ) )
+   _mm512_permutexvar_epi32( v, m512_const_64( \
+                      0x0000000e0000000d, 0x0000000c0000000b, \
+                      0x0000000a00000009, 0x000000080000000f, \
+                      0x0000000600000005, 0x0000000400000003, \
+                      0x0000000200000001, 0x0000000000000007 ) )
+
 #define mm512_ror1x16_256( v ) \
-   _mm512_permutexvar_epi16( v, _mm512_set_epi32( \
-                     0x0010001F, 0x001E001D, 0x001C001B, 0x001A0019, \
-                     0x00180017, 0x00160015, 0x00140013, 0x00120011, \
-                     0x0000000F, 0x000E000D, 0x000C000B, 0x000A0009, \
-                     0x00080007, 0x00060005, 0x00040003, 0x00020001 ) )
+   _mm512_permutexvar_epi16( v, m512_const_64( \
+                     0x0010001F001E001D, 0x001C001B001A0019, \
+                     0x0018001700160015, 0x0014001300120011, \
+                     0x0000000F000E000D, 0x000C000B000A0009, \
+                     0x0008000700060005, 0x0004000300020001 ) )
 
 #define mm512_rol1x16_256( v ) \
-   _mm512_permutexvar_epi16( v, _mm512_set_epi32( \
-                     0x001E001D, 0x001C001B, 0x001A0019, 0x00180017, \
-                     0x00160015, 0x00140013, 0x00120011, 0x0000000F, \
-                     0x000E000D, 0x000C000B, 0x000A0009, 0x00080007, \
-                     0x00060005, 0x00040003, 0x00020001, 0x0000001F ) )
+   _mm512_permutexvar_epi16( v, m512_const_64( \
+                     0x001E001D001C001B, 0x001A001900180017, \
+                     0x0016001500140013, 0x001200110000000F, \
+                     0x000E000D000C000B, 0x000A000900080007, \
+                     0x0006000500040003, 0x000200010000001F ) )
 
 #define mm512_ror1x8_256( v ) \
-   _mm512_permutexvar_epi8( v, _mm512_set_epi32( \
-                     0x203F3E3D, 0x3C3B3A39, 0x38373635, 0x34333231, \
-                     0x302F2E2D, 0x2C2B2A29, 0x28272625, 0x24232221, \
-                     0x001F1E1D, 0x1C1B1A19, 0x18171615, 0x14131211, \
-                     0x100F0E0D, 0x0C0B0A09, 0x08070605, 0x04030201 ) )
+   _mm512_permutexvar_epi8( v, m512_const_64( \
+                     0x203F3E3D3C3B3A39, 0x3837363534333231, \
+                     0x302F2E2D2C2B2A29, 0x2827262524232221, \
+                     0x001F1E1D1C1B1A19, 0x1817161514131211, \
+                     0x100F0E0D0C0B0A09, 0x0807060504030201 ) )
 
 #define mm512_rol1x8_256( v ) \
-    _mm512_permutexvar_epi8( v, _mm512_set_epi32( \
-                     0x3E3D3C3B, 0x3A393837, 0x36353433, 0x3231302F, \
-                     0x2E2D2C2B, 0x2A292827, 0x26252423, 0x2221203F, \
-                     0x1E1D1C1B, 0x1A191817, 0x16151413, 0x1211100F, \
-                     0x0E0D0C0B, 0x0A090807, 0x06050403, 0x0201001F ) )
+    _mm512_permutexvar_epi8( v, m512_const_64( \
+                     0x3E3D3C3B3A393837, 0x363534333231302F, \
+                     0x2E2D2C2B2A292827, 0x262524232221203F, \
+                     0x1E1D1C1B1A191817, 0x161514131211100F, \
+                     0x0E0D0C0B0A090807, 0x060504030201001F ) )
 
 //
 // Rotate elements within 128 bit lanes of 512 bit vector.
@@ -441,80 +391,80 @@ static inline __m512i m512_neg1_fn()
 
 // Rotate each 64 bit lane by one 16 bit element.
 #define mm512_ror1x16_64( v ) \
-    _mm512_permutexvar_epi8( v, _mm512_set_epi32( \
-                      0x39383F3E, 0x3D3C3B3A,   0x31303736, 0x35343332, \
-                      0x29282F2E, 0x2D2C2B2A,   0x21202726, 0x25242322, \
-                      0x19181F1E, 0x1D1C1B1A,   0x11101716, 0x15141312, \
-                      0x09080F0E, 0x0D0C0B0A,   0x01000706, 0x05040302 ) )
+    _mm512_permutexvar_epi8( v, m512_const_64( \
+                      0x39383F3E3D3C3B3A, 0x3130373635343332, \
+                      0x29282F2E2D2C2B2A, 0x2120272625242322, \
+                      0x19181F1E1D1C1B1A, 0x1110171615141312, \
+                      0x09080F0E0D0C0B0A, 0x0100070605040302 ) )
 
 #define mm512_rol1x16_64( v ) \
-    _mm512_permutexvar_epi8( v, _mm512_set_epi32( \
-                      0x3D3C3B3A, 0x39383F3E,   0x35343332, 0x31303736 \
-                      0x2D2C2B2A, 0x29282F2E,   0x25242322, 0x21202726 \
-                      0x1D1C1B1A, 0x19181F1E,   0x15141312, 0x11101716 \
-                      0x0D0C0B0A, 0x09080F0E,   0x05040302, 0x01000706 ) ) 
+    _mm512_permutexvar_epi8( v, m512_const_64( \
+                      0x3D3C3B3A39383F3E, 0x3534333231303736 \
+                      0x2D2C2B2A29282F2E, 0x2524232221202726 \
+                      0x1D1C1B1A19181F1E, 0x1514131211101716 \
+                      0x0D0C0B0A09080F0E, 0x0504030201000706 ) ) 
 
 // Rotate each 64 bit lane by one byte.
 #define mm512_ror1x8_64( v ) \
-    _mm512_permutexvar_epi8( v, _mm512_set_epi32( \
-                      0x383F3E3D, 0x3C3B3A39,   0x30373635, 0x34333231, \
-                      0x282F2E2D, 0x2C2B2A29,   0x20272625, 0x24232221, \
-                      0x181F1E1D, 0x1C1B1A19,   0x10171615, 0x14131211, \
-                      0x080F0E0D, 0x0C0B0A09,   0x00070605, 0x0403020 )
+    _mm512_permutexvar_epi8( v, m512_const_64( \
+                      0x383F3E3D3C3B3A39, 0x3037363534333231, \
+                      0x282F2E2D2C2B2A29, 0x2027262524232221, \
+                      0x181F1E1D1C1B1A19, 0x1017161514131211, \
+                      0x080F0E0D0C0B0A09, 0x0007060504030201 ) )
 #define mm512_rol1x8_64( v ) \
-    _mm512_permutexvar_epi8( v, _mm512_set_epi32( \
-                       0x3E3D3C3B, 0x3A39383F,   0x36353433, 0x32313037, \
-                       0x2E2D2C2B, 0x2A29282F,   0x26252423, 0x22212027, \
-                       0x1E1D1C1B, 0x1A19181F,   0x16151413, 0x12111017, \
-                       0x0E0D0C0B, 0x0A09080F,   0x06050403, 0x02010007 )
+    _mm512_permutexvar_epi8( v, m512_const_64( \
+                       0x3E3D3C3B3A39383F, 0x3635343332313037, \
+                       0x2E2D2C2B2A29282F, 0x2625242322212027, \
+                       0x1E1D1C1B1A19181F, 0x1615141312111017, \
+                       0x0E0D0C0B0A09080F, 0x0605040302010007 ) )
 
 //
 // Rotate elements within 32 bit lanes.
 
 #define mm512_swap16_32( v ) \
-   _mm512_permutexvar_epi8( v, _mm512_set_epi32( \
-                       0x001D001C, 0x001F001E, 0x00190018, 0x001B001A, \
-                       0x00150014, 0x00170016, 0x00110010, 0x00130012, \
-                       0x000D000C, 0x000F000E, 0x00190008, 0x000B000A, \
-                       0x00050004, 0x00070006, 0x00110000, 0x00030002 )
+   _mm512_permutexvar_epi8( v, m512_const_64( \
+                       0x001D001C001F001E, 0x00190018001B001A, \
+                       0x0015001400170016, 0x0011001000130012, \
+                       0x000D000C000F000E, 0x00190008000B000A, \
+                       0x0005000400070006, 0x0011000000030002 ) )
 
 #define mm512_ror1x8_32( v ) \
-   _mm512_permutexvar_epi8( v, _mm512_set_epi32( \
-                       0x3C3F3E3D, 0x383B3A39, 0x34373635, 0x30333231, \
-                       0x2C2F2E2D, 0x282B2A29, 0x24272625, 0x20232221, \
-                       0x1C1F1E1D, 0x181B1A19, 0x14171615, 0x10131211, \
-                       0x0C0F0E0D, 0x080B0A09, 0x04070605, 0x00030201 ) )
+   _mm512_permutexvar_epi8( v, m512_const_64( \
+                       0x3C3F3E3D383B3A39, 0x3437363530333231, \
+                       0x2C2F2E2D282B2A29, 0x2427262520232221, \
+                       0x1C1F1E1D181B1A19, 0x1417161510131211, \
+                       0x0C0F0E0D080B0A09, 0x0407060500030201 ) )
 
 #define mm512_rol1x8_32( v ) \
-   _mm512_permutexvar_epi8( v, _mm512_set_epi32( \
-                       0x3E3D3C3F, 0x3A39383B, 0x36353437, 0x32313033, \
-                       0x2E2D2C2F, 0x2A29282B, 0x26252427, 0x22212023, \
-                       0x1E1D1C1F, 0x1A19181B, 0x16151417, 0x12111013, \
-                       0x0E0D0C0F, 0x0A09080B, 0x06050407, 0x02010003 ) )
+   _mm512_permutexvar_epi8( v, m512_const_64( \
+                       0x3E3D3C3F3A39383B, 0x3635343732313033, \
+                       0x2E2D2C2F2A29282B, 0x2625242722212023, \
+                       0x1E1D1C1F1A19181B, 0x1615141712111013, \
+                       0x0E0D0C0F0A09080B, 0x0605040702010003 ) )
 
 //
 // Swap bytes in vector elements, vectorized bswap.
 
 #define mm512_bswap_64( v ) \
-   _mm512_permutexvar_epi8( v, _mm512_set_epi32( \
-                       0x38393A3B, 0x3C3D3E3F,   0x20313233, 0x34353637, \
-                       0x28292A2B, 0x2C2D2E2F,   0x20212223, 0x34353637, \
-                       0x18191A1B, 0x1C1D1E1F,   0x10111213, 0x14151617, \
-                       0x08090A0B, 0x0C0D0E0F,   0x00010203, 0x04050607 ) )
+   _mm512_permutexvar_epi8( v, m512_const_64( \
+                       0x38393A3B3C3D3E3F, 0x2031323334353637, \
+                       0x28292A2B2C2D2E2F, 0x2021222334353637, \
+                       0x18191A1B1C1D1E1F, 0x1011121314151617, \
+                       0x08090A0B0C0D0E0F, 0x0001020304050607 ) )
 
 #define mm512_bswap_32( v ) \
-   _mm512_permutexvar_epi8( v, _mm512_set_epi32( \
-                       0x3C3D3E3F, 0x38393A3B, 0x34353637, 0x30313233, \
-                       0x3C3D3E3F, 0x38393A3B, 0x34353637, 0x30313233, \
-                       0x3C3D3E3F, 0x38393A3B, 0x34353637, 0x30313233, \
-                       0x3C3D3E3F, 0x38393A3B, 0x34353637, 0x30313233 ) )
+   _mm512_permutexvar_epi8( v, m512_const_64( \
+                       0x3C3D3E3F38393A3B, 0x3435363730313233, \
+                       0x3C3D3E3F38393A3B, 0x3435363730313233, \
+                       0x3C3D3E3F38393A3B, 0x3435363730313233, \
+                       0x3C3D3E3F38393A3B, 0x3435363730313233 ) )
 
 #define mm512_bswap_16( v ) \
-   _mm512_permutexvar_epi8( v, _mm512_set_epi32( \
-                       0x3E3F3C3D, 0x3A3B3839, 0x36373435, 0x32333031, \
-                       0x2E2F2C2D, 0x2A2B2829, 0x26272425, 0x22232021, \
-                       0x1E1F1C1D, 0x1A1B1819, 0x16171415, 0x12131011, \
-                       0x0E0F0C0D, 0x0A0B0809, 0x06070405, 0x02030001 ) )
+   _mm512_permutexvar_epi8( v, m512_const_64( \
+                       0x3E3F3C3D3A3B3839, 0x3637343532333031, \
+                       0x2E2F2C2D2A2B2829, 0x2627242522232021, \
+                       0x1E1F1C1D1A1B1819, 0x1617141512131011, \
+                       0x0E0F0C0D0A0B0809, 0x0607040502030001 ) )
 
 //
 //  Rotate elements from 2 512 bit vectors in place, source arguments

simd-utils/simd-64.h

@@ -33,7 +33,8 @@
 // cast all arguments as the're likely to be uint64_t
 
 // Bitwise not: ~(a)
-#define mm64_not( a ) _mm_xor_si64( (__m64)a, m64_neg1 )
+//#define mm64_not( a ) _mm_xor_si64( (__m64)a, m64_neg1 )
+#define mm64_not( a ) ( (__m64)( ~( (uint64_t)(a) ) )
 
 // Unary negate elements
 #define mm64_negate_32( v ) _mm_sub_pi32( m64_zero, (__m64)v )

simd-utils/simd-int.h

@@ -34,7 +34,7 @@
       (uint32_t)( ( (uint32_t)(x) << (c) ) | ( (uint32_t)(x) >> (32-(c)) ) )
 #define u16_ror_16( x, c ) \
       (uint16_t)( ( (uint16_t)(x) >> (c) ) | ( (uint16_t)(x) << (16-(c)) ) )
-#define u16rol_16( x, c ) \
+#define u16_rol_16( x, c ) \
       (uint16_t)( ( (uint16_t)(x) << (c) ) | ( (uint16_t)(x) >> (16-(c)) ) )
 #define u8_ror_8( x, c ) \
       (uint8_t) ( ( (uint8_t) (x) >> (c) ) | ( (uint8_t) (x) << ( 8-(c)) ) )

uint256.cpp

@@ -1,40 +0,0 @@
-#include "uint256.h"
-
-#ifdef __cplusplus
-extern "C"{
-#endif
-
-#include "miner.h"
-
-// compute the diff ratio between a found hash and the target
-double hash_target_ratio(uint32_t* hash, uint32_t* target)
-{
-	uint256 h, t;
-	double dhash;
-
-	if (!opt_showdiff)
-		return 0.0;
-
-	memcpy(&t, (void*) target, 32);
-	memcpy(&h, (void*) hash, 32);
-
-	dhash = h.getdouble();
-	if (dhash > 0.)
-		return t.getdouble() / dhash;
-	else
-		return dhash;
-}
-
-// store ratio in work struct
-void work_set_target_ratio( struct work* work, uint32_t* hash )
-{
-	// only if the option is enabled (to reduce cpu usage)
-	if (opt_showdiff) {
-		work->shareratio = hash_target_ratio(hash, work->target);
-		work->sharediff = work->targetdiff * work->shareratio;
-	}
-}
-
-#ifdef __cplusplus
-}
-#endif

uint256.h

@@ -1,784 +0,0 @@
-// Copyright (c) 2009-2010 Satoshi Nakamoto
-// Copyright (c) 2009-2012 The Bitcoin developers
-// Distributed under the MIT/X11 software license, see the accompanying
-// file COPYING or http://www.opensource.org/licenses/mit-license.php.
-#ifndef BITCOIN_UINT256_H
-#define BITCOIN_UINT256_H
-
-#include <limits.h>
-#include <stdio.h>
-#include <string.h>
-#include <inttypes.h>
-#include <string>
-#include <vector>
-
-typedef long long  int64;
-typedef unsigned long long  uint64;
-
-
-inline int Testuint256AdHoc(std::vector<std::string> vArg);
-
-
-
-/** Base class without constructors for uint256 and uint160.
- * This makes the compiler let you use it in a union.
- */
-template<unsigned int BITS>
-class base_uint
-{
-protected:
-    enum { WIDTH=BITS/32 };
-    uint32_t pn[WIDTH];
-public:
-
-    bool operator!() const
-    {
-        for (int i = 0; i < WIDTH; i++)
-            if (pn[i] != 0)
-                return false;
-        return true;
-    }
-
-    const base_uint operator~() const
-    {
-        base_uint ret;
-        for (int i = 0; i < WIDTH; i++)
-            ret.pn[i] = ~pn[i];
-        return ret;
-    }
-
-    const base_uint operator-() const
-    {
-        base_uint ret;
-        for (int i = 0; i < WIDTH; i++)
-            ret.pn[i] = ~pn[i];
-        ret++;
-        return ret;
-    }
-
-    double getdouble() const
-    {
-        double ret = 0.0;
-        double fact = 1.0;
-        for (int i = 0; i < WIDTH; i++) {
-            ret += fact * pn[i];
-            fact *= 4294967296.0;
-        }
-        return ret;
-    }
-
-    base_uint& operator=(uint64 b)
-    {
-        pn[0] = (unsigned int)b;
-        pn[1] = (unsigned int)(b >> 32);
-        for (int i = 2; i < WIDTH; i++)
-            pn[i] = 0;
-        return *this;
-    }
-
-    base_uint& operator^=(const base_uint& b)
-    {
-        for (int i = 0; i < WIDTH; i++)
-            pn[i] ^= b.pn[i];
-        return *this;
-    }
-
-    base_uint& operator&=(const base_uint& b)
-    {
-        for (int i = 0; i < WIDTH; i++)
-            pn[i] &= b.pn[i];
-        return *this;
-    }
-
-    base_uint& operator|=(const base_uint& b)
-    {
-        for (int i = 0; i < WIDTH; i++)
-            pn[i] |= b.pn[i];
-        return *this;
-    }
-
-    base_uint& operator^=(uint64 b)
-    {
-        pn[0] ^= (unsigned int)b;
-        pn[1] ^= (unsigned int)(b >> 32);
-        return *this;
-    }
-
-    base_uint& operator|=(uint64 b)
-    {
-        pn[0] |= (unsigned int)b;
-        pn[1] |= (unsigned int)(b >> 32);
-        return *this;
-    }
-
-    base_uint& operator<<=(unsigned int shift)
-    {
-        base_uint a(*this);
-        for (int i = 0; i < WIDTH; i++)
-            pn[i] = 0;
-        int k = shift / 32;
-        shift = shift % 32;
-        for (int i = 0; i < WIDTH; i++)
-        {
-            if (i+k+1 < WIDTH && shift != 0)
-                pn[i+k+1] |= (a.pn[i] >> (32-shift));
-            if (i+k < WIDTH)
-                pn[i+k] |= (a.pn[i] << shift);
-        }
-        return *this;
-    }
-
-    base_uint& operator>>=(unsigned int shift)
-    {
-        base_uint a(*this);
-        for (int i = 0; i < WIDTH; i++)
-            pn[i] = 0;
-        int k = shift / 32;
-        shift = shift % 32;
-        for (int i = 0; i < WIDTH; i++)
-        {
-            if (i-k-1 >= 0 && shift != 0)
-                pn[i-k-1] |= (a.pn[i] << (32-shift));
-            if (i-k >= 0)
-                pn[i-k] |= (a.pn[i] >> shift);
-        }
-        return *this;
-    }
-
-    base_uint& operator+=(const base_uint& b)
-    {
-        uint64 carry = 0;
-        for (int i = 0; i < WIDTH; i++)
-        {
-            uint64 n = carry + pn[i] + b.pn[i];
-            pn[i] = n & 0xffffffff;
-            carry = n >> 32;
-        }
-        return *this;
-    }
-
-    base_uint& operator-=(const base_uint& b)
-    {
-        *this += -b;
-        return *this;
-    }
-
-    base_uint& operator+=(uint64 b64)
-    {
-        base_uint b;
-        b = b64;
-        *this += b;
-        return *this;
-    }
-
-    base_uint& operator-=(uint64 b64)
-    {
-        base_uint b;
-        b = b64;
-        *this += -b;
-        return *this;
-    }
-
-
-    base_uint& operator++()
-    {
-        // prefix operator
-        int i = 0;
-        while (++pn[i] == 0 && i < WIDTH-1)
-            i++;
-        return *this;
-    }
-
-    const base_uint operator++(int)
-    {
-        // postfix operator
-        const base_uint ret = *this;
-        ++(*this);
-        return ret;
-    }
-
-    base_uint& operator--()
-    {
-        // prefix operator
-        int i = 0;
-        while (--pn[i] == -1 && i < WIDTH-1)
-            i++;
-        return *this;
-    }
-
-    const base_uint operator--(int)
-    {
-        // postfix operator
-        const base_uint ret = *this;
-        --(*this);
-        return ret;
-    }
-
-
-    friend inline bool operator<(const base_uint& a, const base_uint& b)
-    {
-        for (int i = base_uint::WIDTH-1; i >= 0; i--)
-        {
-            if (a.pn[i] < b.pn[i])
-                return true;
-            else if (a.pn[i] > b.pn[i])
-                return false;
-        }
-        return false;
-    }
-
-    friend inline bool operator<=(const base_uint& a, const base_uint& b)
-    {
-        for (int i = base_uint::WIDTH-1; i >= 0; i--)
-        {
-            if (a.pn[i] < b.pn[i])
-                return true;
-            else if (a.pn[i] > b.pn[i])
-                return false;
-        }
-        return true;
-    }
-
-    friend inline bool operator>(const base_uint& a, const base_uint& b)
-    {
-        for (int i = base_uint::WIDTH-1; i >= 0; i--)
-        {
-            if (a.pn[i] > b.pn[i])
-                return true;
-            else if (a.pn[i] < b.pn[i])
-                return false;
-        }
-        return false;
-    }
-
-    friend inline bool operator>=(const base_uint& a, const base_uint& b)
-    {
-        for (int i = base_uint::WIDTH-1; i >= 0; i--)
-        {
-            if (a.pn[i] > b.pn[i])
-                return true;
-            else if (a.pn[i] < b.pn[i])
-                return false;
-        }
-        return true;
-    }
-
-    friend inline bool operator==(const base_uint& a, const base_uint& b)
-    {
-        for (int i = 0; i < base_uint::WIDTH; i++)
-            if (a.pn[i] != b.pn[i])
-                return false;
-        return true;
-    }
-
-    friend inline bool operator==(const base_uint& a, uint64 b)
-    {
-        if (a.pn[0] != (unsigned int)b)
-            return false;
-        if (a.pn[1] != (unsigned int)(b >> 32))
-            return false;
-        for (int i = 2; i < base_uint::WIDTH; i++)
-            if (a.pn[i] != 0)
-                return false;
-        return true;
-    }
-
-    friend inline bool operator!=(const base_uint& a, const base_uint& b)
-    {
-        return (!(a == b));
-    }
-
-    friend inline bool operator!=(const base_uint& a, uint64 b)
-    {
-        return (!(a == b));
-    }
-
-
-
-    std::string GetHex() const
-    {
-        char psz[sizeof(pn)*2 + 1];
-        for (unsigned int i = 0; i < sizeof(pn); i++)
-            sprintf(psz + i*2, "%02x", ((unsigned char*)pn)[sizeof(pn) - i - 1]);
-        return std::string(psz, psz + sizeof(pn)*2);
-    }
-
-    void SetHex(const char* psz)
-    {
-        for (int i = 0; i < WIDTH; i++)
-            pn[i] = 0;
-
-        // skip leading spaces
-        while (isspace(*psz))
-            psz++;
-
-        // skip 0x
-        if (psz[0] == '0' && tolower(psz[1]) == 'x')
-            psz += 2;
-
-        // hex string to uint
-        static const unsigned char phexdigit[256] = { 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,1,2,3,4,5,6,7,8,9,0,0,0,0,0,0, 0,0xa,0xb,0xc,0xd,0xe,0xf,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0xa,0xb,0xc,0xd,0xe,0xf,0,0,0,0,0,0,0,0,0 };
-        const char* pbegin = psz;
-        while (phexdigit[(unsigned char)*psz] || *psz == '0')
-            psz++;
-        psz--;
-        unsigned char* p1 = (unsigned char*)pn;
-        unsigned char* pend = p1 + WIDTH * 4;
-        while (psz >= pbegin && p1 < pend)
-        {
-            *p1 = phexdigit[(unsigned char)*psz--];
-            if (psz >= pbegin)
-            {
-                *p1 |= (phexdigit[(unsigned char)*psz--] << 4);
-                p1++;
-            }
-        }
-    }
-
-    void SetHex(const std::string& str)
-    {
-        SetHex(str.c_str());
-    }
-
-    std::string ToString() const
-    {
-        return (GetHex());
-    }
-
-    unsigned char* begin()
-    {
-        return (unsigned char*)&pn[0];
-    }
-
-    unsigned char* end()
-    {
-        return (unsigned char*)&pn[WIDTH];
-    }
-
-    const unsigned char* begin() const
-    {
-        return (unsigned char*)&pn[0];
-    }
-
-    const unsigned char* end() const
-    {
-        return (unsigned char*)&pn[WIDTH];
-    }
-
-    unsigned int size() const
-    {
-        return sizeof(pn);
-    }
-
-    uint64 Get64(int n=0) const
-    {
-        return pn[2*n] | (uint64)pn[2*n+1] << 32;
-    }
-
-//    unsigned int GetSerializeSize(int nType=0, int nVersion=PROTOCOL_VERSION) const
-    unsigned int GetSerializeSize(int nType, int nVersion) const
-    {
-        return sizeof(pn);
-    }
-
-    template<typename Stream>
-//    void Serialize(Stream& s, int nType=0, int nVersion=PROTOCOL_VERSION) const
-    void Serialize(Stream& s, int nType, int nVersion) const
-    {
-        s.write((char*)pn, sizeof(pn));
-    }
-
-    template<typename Stream>
-//    void Unserialize(Stream& s, int nType=0, int nVersion=PROTOCOL_VERSION)
-    void Unserialize(Stream& s, int nType, int nVersion)
-    {
-        s.read((char*)pn, sizeof(pn));
-    }
-
-
-    friend class uint160;
-    friend class uint256;
-    friend inline int Testuint256AdHoc(std::vector<std::string> vArg);
-};
-
-typedef base_uint<160> base_uint160;
-typedef base_uint<256> base_uint256;
-
-
-
-//
-// uint160 and uint256 could be implemented as templates, but to keep
-// compile errors and debugging cleaner, they're copy and pasted.
-//
-
-
-
-//////////////////////////////////////////////////////////////////////////////
-//
-// uint160
-//
-
-/** 160-bit unsigned integer */
-class uint160 : public base_uint160
-{
-public:
-    typedef base_uint160 basetype;
-
-    uint160()
-    {
-        for (int i = 0; i < WIDTH; i++)
-            pn[i] = 0;
-    }
-
-    uint160(const basetype& b)
-    {
-        for (int i = 0; i < WIDTH; i++)
-            pn[i] = b.pn[i];
-    }
-
-    uint160& operator=(const basetype& b)
-    {
-        for (int i = 0; i < WIDTH; i++)
-            pn[i] = b.pn[i];
-        return *this;
-    }
-
-    uint160(uint64 b)
-    {
-        pn[0] = (unsigned int)b;
-        pn[1] = (unsigned int)(b >> 32);
-        for (int i = 2; i < WIDTH; i++)
-            pn[i] = 0;
-    }
-
-    uint160& operator=(uint64 b)
-    {
-        pn[0] = (unsigned int)b;
-        pn[1] = (unsigned int)(b >> 32);
-        for (int i = 2; i < WIDTH; i++)
-            pn[i] = 0;
-        return *this;
-    }
-
-    explicit uint160(const std::string& str)
-    {
-        SetHex(str);
-    }
-
-    explicit uint160(const std::vector<unsigned char>& vch)
-    {
-        if (vch.size() == sizeof(pn))
-            memcpy(pn, &vch[0], sizeof(pn));
-        else
-            *this = 0;
-    }
-};
-
-inline bool operator==(const uint160& a, uint64 b)                           { return (base_uint160)a == b; }
-inline bool operator!=(const uint160& a, uint64 b)                           { return (base_uint160)a != b; }
-inline const uint160 operator<<(const base_uint160& a, unsigned int shift)   { return uint160(a) <<= shift; }
-inline const uint160 operator>>(const base_uint160& a, unsigned int shift)   { return uint160(a) >>= shift; }
-inline const uint160 operator<<(const uint160& a, unsigned int shift)        { return uint160(a) <<= shift; }
-inline const uint160 operator>>(const uint160& a, unsigned int shift)        { return uint160(a) >>= shift; }
-
-inline const uint160 operator^(const base_uint160& a, const base_uint160& b) { return uint160(a) ^= b; }
-inline const uint160 operator&(const base_uint160& a, const base_uint160& b) { return uint160(a) &= b; }
-inline const uint160 operator|(const base_uint160& a, const base_uint160& b) { return uint160(a) |= b; }
-inline const uint160 operator+(const base_uint160& a, const base_uint160& b) { return uint160(a) += b; }
-inline const uint160 operator-(const base_uint160& a, const base_uint160& b) { return uint160(a) -= b; }
-
-inline bool operator<(const base_uint160& a, const uint160& b)          { return (base_uint160)a <  (base_uint160)b; }
-inline bool operator<=(const base_uint160& a, const uint160& b)         { return (base_uint160)a <= (base_uint160)b; }
-inline bool operator>(const base_uint160& a, const uint160& b)          { return (base_uint160)a >  (base_uint160)b; }
-inline bool operator>=(const base_uint160& a, const uint160& b)         { return (base_uint160)a >= (base_uint160)b; }
-inline bool operator==(const base_uint160& a, const uint160& b)         { return (base_uint160)a == (base_uint160)b; }
-inline bool operator!=(const base_uint160& a, const uint160& b)         { return (base_uint160)a != (base_uint160)b; }
-inline const uint160 operator^(const base_uint160& a, const uint160& b) { return (base_uint160)a ^  (base_uint160)b; }
-inline const uint160 operator&(const base_uint160& a, const uint160& b) { return (base_uint160)a &  (base_uint160)b; }
-inline const uint160 operator|(const base_uint160& a, const uint160& b) { return (base_uint160)a |  (base_uint160)b; }
-inline const uint160 operator+(const base_uint160& a, const uint160& b) { return (base_uint160)a +  (base_uint160)b; }
-inline const uint160 operator-(const base_uint160& a, const uint160& b) { return (base_uint160)a -  (base_uint160)b; }
-
-inline bool operator<(const uint160& a, const base_uint160& b)          { return (base_uint160)a <  (base_uint160)b; }
-inline bool operator<=(const uint160& a, const base_uint160& b)         { return (base_uint160)a <= (base_uint160)b; }
-inline bool operator>(const uint160& a, const base_uint160& b)          { return (base_uint160)a >  (base_uint160)b; }
-inline bool operator>=(const uint160& a, const base_uint160& b)         { return (base_uint160)a >= (base_uint160)b; }
-inline bool operator==(const uint160& a, const base_uint160& b)         { return (base_uint160)a == (base_uint160)b; }
-inline bool operator!=(const uint160& a, const base_uint160& b)         { return (base_uint160)a != (base_uint160)b; }
-inline const uint160 operator^(const uint160& a, const base_uint160& b) { return (base_uint160)a ^  (base_uint160)b; }
-inline const uint160 operator&(const uint160& a, const base_uint160& b) { return (base_uint160)a &  (base_uint160)b; }
-inline const uint160 operator|(const uint160& a, const base_uint160& b) { return (base_uint160)a |  (base_uint160)b; }
-inline const uint160 operator+(const uint160& a, const base_uint160& b) { return (base_uint160)a +  (base_uint160)b; }
-inline const uint160 operator-(const uint160& a, const base_uint160& b) { return (base_uint160)a -  (base_uint160)b; }
-
-inline bool operator<(const uint160& a, const uint160& b)               { return (base_uint160)a <  (base_uint160)b; }
-inline bool operator<=(const uint160& a, const uint160& b)              { return (base_uint160)a <= (base_uint160)b; }
-inline bool operator>(const uint160& a, const uint160& b)               { return (base_uint160)a >  (base_uint160)b; }
-inline bool operator>=(const uint160& a, const uint160& b)              { return (base_uint160)a >= (base_uint160)b; }
-inline bool operator==(const uint160& a, const uint160& b)              { return (base_uint160)a == (base_uint160)b; }
-inline bool operator!=(const uint160& a, const uint160& b)              { return (base_uint160)a != (base_uint160)b; }
-inline const uint160 operator^(const uint160& a, const uint160& b)      { return (base_uint160)a ^  (base_uint160)b; }
-inline const uint160 operator&(const uint160& a, const uint160& b)      { return (base_uint160)a &  (base_uint160)b; }
-inline const uint160 operator|(const uint160& a, const uint160& b)      { return (base_uint160)a |  (base_uint160)b; }
-inline const uint160 operator+(const uint160& a, const uint160& b)      { return (base_uint160)a +  (base_uint160)b; }
-inline const uint160 operator-(const uint160& a, const uint160& b)      { return (base_uint160)a -  (base_uint160)b; }
-
-
-
-
-
-
-//////////////////////////////////////////////////////////////////////////////
-//
-// uint256
-//
-
-/** 256-bit unsigned integer */
-class uint256 : public base_uint256
-{
-public:
-    typedef base_uint256 basetype;
-
-    uint256()
-    {
-        for (int i = 0; i < WIDTH; i++)
-            pn[i] = 0;
-    }
-
-    uint256(const basetype& b)
-    {
-        for (int i = 0; i < WIDTH; i++)
-            pn[i] = b.pn[i];
-    }
-
-    uint256& operator=(const basetype& b)
-    {
-        for (int i = 0; i < WIDTH; i++)
-            pn[i] = b.pn[i];
-        return *this;
-    }
-
-    uint256(uint64 b)
-    {
-        pn[0] = (unsigned int)b;
-        pn[1] = (unsigned int)(b >> 32);
-        for (int i = 2; i < WIDTH; i++)
-            pn[i] = 0;
-    }
-
-    uint256& operator=(uint64 b)
-    {
-        pn[0] = (unsigned int)b;
-        pn[1] = (unsigned int)(b >> 32);
-        for (int i = 2; i < WIDTH; i++)
-            pn[i] = 0;
-        return *this;
-    }
-
-    explicit uint256(const std::string& str)
-    {
-        SetHex(str);
-    }
-
-    explicit uint256(const std::vector<unsigned char>& vch)
-    {
-        if (vch.size() == sizeof(pn))
-            memcpy(pn, &vch[0], sizeof(pn));
-        else
-            *this = 0;
-    }
-};
-
-inline bool operator==(const uint256& a, uint64 b)                           { return (base_uint256)a == b; }
-inline bool operator!=(const uint256& a, uint64 b)                           { return (base_uint256)a != b; }
-inline const uint256 operator<<(const base_uint256& a, unsigned int shift)   { return uint256(a) <<= shift; }
-inline const uint256 operator>>(const base_uint256& a, unsigned int shift)   { return uint256(a) >>= shift; }
-inline const uint256 operator<<(const uint256& a, unsigned int shift)        { return uint256(a) <<= shift; }
-inline const uint256 operator>>(const uint256& a, unsigned int shift)        { return uint256(a) >>= shift; }
-
-inline const uint256 operator^(const base_uint256& a, const base_uint256& b) { return uint256(a) ^= b; }
-inline const uint256 operator&(const base_uint256& a, const base_uint256& b) { return uint256(a) &= b; }
-inline const uint256 operator|(const base_uint256& a, const base_uint256& b) { return uint256(a) |= b; }
-inline const uint256 operator+(const base_uint256& a, const base_uint256& b) { return uint256(a) += b; }
-inline const uint256 operator-(const base_uint256& a, const base_uint256& b) { return uint256(a) -= b; }
-
-inline bool operator<(const base_uint256& a, const uint256& b)          { return (base_uint256)a <  (base_uint256)b; }
-inline bool operator<=(const base_uint256& a, const uint256& b)         { return (base_uint256)a <= (base_uint256)b; }
-inline bool operator>(const base_uint256& a, const uint256& b)          { return (base_uint256)a >  (base_uint256)b; }
-inline bool operator>=(const base_uint256& a, const uint256& b)         { return (base_uint256)a >= (base_uint256)b; }
-inline bool operator==(const base_uint256& a, const uint256& b)         { return (base_uint256)a == (base_uint256)b; }
-inline bool operator!=(const base_uint256& a, const uint256& b)         { return (base_uint256)a != (base_uint256)b; }
-inline const uint256 operator^(const base_uint256& a, const uint256& b) { return (base_uint256)a ^  (base_uint256)b; }
-inline const uint256 operator&(const base_uint256& a, const uint256& b) { return (base_uint256)a &  (base_uint256)b; }
-inline const uint256 operator|(const base_uint256& a, const uint256& b) { return (base_uint256)a |  (base_uint256)b; }
-inline const uint256 operator+(const base_uint256& a, const uint256& b) { return (base_uint256)a +  (base_uint256)b; }
-inline const uint256 operator-(const base_uint256& a, const uint256& b) { return (base_uint256)a -  (base_uint256)b; }
-
-inline bool operator<(const uint256& a, const base_uint256& b)          { return (base_uint256)a <  (base_uint256)b; }
-inline bool operator<=(const uint256& a, const base_uint256& b)         { return (base_uint256)a <= (base_uint256)b; }
-inline bool operator>(const uint256& a, const base_uint256& b)          { return (base_uint256)a >  (base_uint256)b; }
-inline bool operator>=(const uint256& a, const base_uint256& b)         { return (base_uint256)a >= (base_uint256)b; }
-inline bool operator==(const uint256& a, const base_uint256& b)         { return (base_uint256)a == (base_uint256)b; }
-inline bool operator!=(const uint256& a, const base_uint256& b)         { return (base_uint256)a != (base_uint256)b; }
-inline const uint256 operator^(const uint256& a, const base_uint256& b) { return (base_uint256)a ^  (base_uint256)b; }
-inline const uint256 operator&(const uint256& a, const base_uint256& b) { return (base_uint256)a &  (base_uint256)b; }
-inline const uint256 operator|(const uint256& a, const base_uint256& b) { return (base_uint256)a |  (base_uint256)b; }
-inline const uint256 operator+(const uint256& a, const base_uint256& b) { return (base_uint256)a +  (base_uint256)b; }
-inline const uint256 operator-(const uint256& a, const base_uint256& b) { return (base_uint256)a -  (base_uint256)b; }
-
-inline bool operator<(const uint256& a, const uint256& b)               { return (base_uint256)a <  (base_uint256)b; }
-inline bool operator<=(const uint256& a, const uint256& b)              { return (base_uint256)a <= (base_uint256)b; }
-inline bool operator>(const uint256& a, const uint256& b)               { return (base_uint256)a >  (base_uint256)b; }
-inline bool operator>=(const uint256& a, const uint256& b)              { return (base_uint256)a >= (base_uint256)b; }
-inline bool operator==(const uint256& a, const uint256& b)              { return (base_uint256)a == (base_uint256)b; }
-inline bool operator!=(const uint256& a, const uint256& b)              { return (base_uint256)a != (base_uint256)b; }
-inline const uint256 operator^(const uint256& a, const uint256& b)      { return (base_uint256)a ^  (base_uint256)b; }
-inline const uint256 operator&(const uint256& a, const uint256& b)      { return (base_uint256)a &  (base_uint256)b; }
-inline const uint256 operator|(const uint256& a, const uint256& b)      { return (base_uint256)a |  (base_uint256)b; }
-inline const uint256 operator+(const uint256& a, const uint256& b)      { return (base_uint256)a +  (base_uint256)b; }
-inline const uint256 operator-(const uint256& a, const uint256& b)      { return (base_uint256)a -  (base_uint256)b; }
-
-
-
-
-
-
-
-
-
-
-#ifdef TEST_UINT256
-
-inline int Testuint256AdHoc(std::vector<std::string> vArg)
-{
-    uint256 g(0);
-
-
-    printf("%s\n", g.ToString().c_str());
-    g--;  printf("g--\n");
-    printf("%s\n", g.ToString().c_str());
-    g--;  printf("g--\n");
-    printf("%s\n", g.ToString().c_str());
-    g++;  printf("g++\n");
-    printf("%s\n", g.ToString().c_str());
-    g++;  printf("g++\n");
-    printf("%s\n", g.ToString().c_str());
-    g++;  printf("g++\n");
-    printf("%s\n", g.ToString().c_str());
-    g++;  printf("g++\n");
-    printf("%s\n", g.ToString().c_str());
-
-
-
-    uint256 a(7);
-    printf("a=7\n");
-    printf("%s\n", a.ToString().c_str());
-
-    uint256 b;
-    printf("b undefined\n");
-    printf("%s\n", b.ToString().c_str());
-    int c = 3;
-
-    a = c;
-    a.pn[3] = 15;
-    printf("%s\n", a.ToString().c_str());
-    uint256 k(c);
-
-    a = 5;
-    a.pn[3] = 15;
-    printf("%s\n", a.ToString().c_str());
-    b = 1;
-    b <<= 52;
-
-    a |= b;
-
-    a ^= 0x500;
-
-    printf("a %s\n", a.ToString().c_str());
-
-    a = a | b | (uint256)0x1000;
-
-
-    printf("a %s\n", a.ToString().c_str());
-    printf("b %s\n", b.ToString().c_str());
-
-    a = 0xfffffffe;
-    a.pn[4] = 9;
-
-    printf("%s\n", a.ToString().c_str());
-    a++;
-    printf("%s\n", a.ToString().c_str());
-    a++;
-    printf("%s\n", a.ToString().c_str());
-    a++;
-    printf("%s\n", a.ToString().c_str());
-    a++;
-    printf("%s\n", a.ToString().c_str());
-
-    a--;
-    printf("%s\n", a.ToString().c_str());
-    a--;
-    printf("%s\n", a.ToString().c_str());
-    a--;
-    printf("%s\n", a.ToString().c_str());
-    uint256 d = a--;
-    printf("%s\n", d.ToString().c_str());
-    printf("%s\n", a.ToString().c_str());
-    a--;
-    printf("%s\n", a.ToString().c_str());
-    a--;
-    printf("%s\n", a.ToString().c_str());
-
-    d = a;
-
-    printf("%s\n", d.ToString().c_str());
-    for (int i = uint256::WIDTH-1; i >= 0; i--) printf("%08x", d.pn[i]); printf("\n");
-
-    uint256 neg = d;
-    neg = ~neg;
-    printf("%s\n", neg.ToString().c_str());
-
-
-    uint256 e = uint256("0xABCDEF123abcdef12345678909832180000011111111");
-    printf("\n");
-    printf("%s\n", e.ToString().c_str());
-
-
-    printf("\n");
-    uint256 x1 = uint256("0xABCDEF123abcdef12345678909832180000011111111");
-    uint256 x2;
-    printf("%s\n", x1.ToString().c_str());
-    for (int i = 0; i < 270; i += 4)
-    {
-        x2 = x1 << i;
-        printf("%s\n", x2.ToString().c_str());
-    }
-
-    printf("\n");
-    printf("%s\n", x1.ToString().c_str());
-    for (int i = 0; i < 270; i += 4)
-    {
-        x2 = x1;
-        x2 >>= i;
-        printf("%s\n", x2.ToString().c_str());
-    }
-
-
-    for (int i = 0; i < 100; i++)
-    {
-        uint256 k = (~uint256(0) >> i);
-        printf("%s\n", k.ToString().c_str());
-    }
-
-    for (int i = 0; i < 100; i++)
-    {
-        uint256 k = (~uint256(0) << i);
-        printf("%s\n", k.ToString().c_str());
-    }
-
-    return (0);
-}
-
-#endif
-
-#endif

util.c

@@ -80,6 +80,86 @@ struct thread_q {
 	pthread_cond_t		cond;
 };
 
+
+void applog2( int prio, const char *fmt, ... )
+{
+   va_list ap;
+
+   va_start(ap, fmt);
+
+#ifdef HAVE_SYSLOG_H
+   if (use_syslog) {
+      va_list ap2;
+      char *buf;
+      int len;
+
+      /* custom colors to syslog prio */
+      if (prio > LOG_DEBUG) {
+         switch (prio) {
+            case LOG_BLUE: prio = LOG_NOTICE; break;
+         }
+      }
+
+      va_copy(ap2, ap);
+      len = vsnprintf(NULL, 0, fmt, ap2) + 1;
+      va_end(ap2);
+      buf = alloca(len);
+      if (vsnprintf(buf, len, fmt, ap) >= 0)
+         syslog(prio, "%s", buf);
+   }
+#else
+   if (0) {}
+#endif
+   else {
+      const char* color = "";
+      char *f;
+      int len;
+//    struct tm tm;
+//    time_t now = time(NULL);
+
+//    localtime_r(&now, &tm);
+
+      switch (prio) {
+         case LOG_ERR:     color = CL_RED; break;
+         case LOG_WARNING: color = CL_YLW; break;
+         case LOG_NOTICE:  color = CL_WHT; break;
+         case LOG_INFO:    color = ""; break;
+         case LOG_DEBUG:   color = CL_GRY; break;
+
+         case LOG_BLUE:
+            prio = LOG_NOTICE;
+            color = CL_CYN;
+            break;
+      }
+      if (!use_colors)
+         color = "";
+
+      len = 64 + (int) strlen(fmt) + 2;
+      f = (char*) malloc(len);
+      sprintf(f, "                     %s %s%s\n",
+//      sprintf(f, "[%d-%02d-%02d %02d:%02d:%02d]%s %s%s\n",
+//         tm.tm_year + 1900,
+//         tm.tm_mon + 1,
+//         tm.tm_mday,
+//         tm.tm_hour,
+//         tm.tm_min,
+//         tm.tm_sec,
+         color,
+         fmt,
+         use_colors ? CL_N : ""
+      );
+      pthread_mutex_lock(&applog_lock);
+      vfprintf(stdout, f, ap);   /* atomic write to stdout */
+      fflush(stdout);
+      free(f);
+      pthread_mutex_unlock(&applog_lock);
+   }
+   va_end(ap);
+}
+
+
+
+
 void applog(int prio, const char *fmt, ...)
 {
 	va_list ap;
@@ -1853,13 +1933,6 @@ static bool stratum_set_difficulty(struct stratum_ctx *sctx, json_t *params)
 	pthread_mutex_lock(&sctx->work_lock);
 	sctx->next_diff = diff;
 	pthread_mutex_unlock(&sctx->work_lock);
-
-	/* store for api stats */
-	stratum_diff = diff;
-
-   if ( !opt_quiet )
-	     applog(LOG_BLUE, "Stratum difficulty set to %g", diff);
-
 	return true;
 }