v3.8.8.1

AUTHORS

@@ -29,3 +29,5 @@ Wolf0
 Optiminer
 
 Jay D Dee
+
+xcouiz@gmail.com

README.md

@@ -59,7 +59,8 @@ Supported Algorithms
                           bmw           BMW 256
                           c11           Chaincoin
                           cryptolight   Cryptonight-light
-                          cryptonight  cryptonote, Monero (XMR)
+                          cryptonight  
+                          cryptonightv7 Monero (XMR)
                           decred
                           deep          Deepcoin (DCN)
                           dmd-gr        Diamond-Groestl

README.txt

@@ -27,7 +27,8 @@ Changes in v3.8.4 may have improved compatibility with some of these CPUs.
 Exe name                Compile flags            Arch name
 
 cpuminer-sse2.exe      "-msse2"                  Core2, Nehalem   
-cpuminer-aes-sse42.exe "-maes -msse4.2"          Westmere, Sandy-Ivybridge
+cpuminer-aes-sse42.exe "-march=westmere"         Westmere, Sandy-Ivybridge
+cpuminer-avx.exe       "-march=corei7-avx"       Sandy-Ivybridge
 cpuminer-avx2.exe      "-march=core-avx2"        Haswell, Sky-Kaby-Coffeelake
 cpuminer-avx2-sha.exe  "-march=core-avx2 -msha"  Ryzen
 

RELEASE_NOTES

@@ -50,8 +50,7 @@ will give a clue as to the missing package.
 The following command should install everything you need on Debian based
 distributions such as Ubuntu:
 
-sudo apt-get install build-essential libssl-dev libcurl4-openssl-dev libjansson-dev libgmp-dev automake
+sudo apt-get install build-essential libssl-dev libcurl4-openssl-dev libjansson-dev libgmp-dev automake zlib1g-dev
-
 
 build-essential  (for Ubuntu, Development Tools package group on Fedora)
 automake
@@ -86,7 +85,7 @@ make
 
 Additional optional compile flags, add the following to CFLAGS to activate:
 
--DUSE_SPH_SHA
+-DUSE_SPH_SHA (deprecated)
 
 SPH may give slightly better performance on algos that use sha256 when using
 openssl 1.0.1 or older. Openssl 1.0.2 adds AVX2 and 1.1 adds SHA and perform
@@ -160,6 +159,21 @@ Support for even older x86_64 without AES_NI or SSE2 is not availble.
 Change Log
 ----------
 
+v3.8.8.1
+
+Fixed x16r.
+Removed cryptonight variant check due to false positives.
+API displays hashrate before shares are submitted.
+
+v3.8.8
+
+Added cryptonightv7 for Monero.
+
+v3.8.7.2
+
+Fixed argon2d-dyn regression in v3.8.7.1.
+Changed compile options for aes-sse42 Windows build to -march=westmere
+
 v3.8.7.1
 
 Fixed argon2d-uis low difficulty rejects.

aclocal.m4

@@ -1,6 +1,6 @@
-# generated automatically by aclocal 1.14.1 -*- Autoconf -*-
+# generated automatically by aclocal 1.15.1 -*- Autoconf -*-
 
-# Copyright (C) 1996-2013 Free Software Foundation, Inc.
+# Copyright (C) 1996-2017 Free Software Foundation, Inc.
 
 # This file is free software; the Free Software Foundation
 # gives unlimited permission to copy and/or distribute it,
@@ -20,7 +20,7 @@ You have another version of autoconf.  It may work, but is not guaranteed to.
 If you have problems, you may need to regenerate the build system entirely.
 To do so, use the procedure documented by the package, typically 'autoreconf'.])])
 
-# Copyright (C) 2002-2013 Free Software Foundation, Inc.
+# Copyright (C) 2002-2017 Free Software Foundation, Inc.
 #
 # This file is free software; the Free Software Foundation
 # gives unlimited permission to copy and/or distribute it,
@@ -32,10 +32,10 @@ To do so, use the procedure documented by the package, typically 'autoreconf'.])
 # generated from the m4 files accompanying Automake X.Y.
 # (This private macro should not be called outside this file.)
 AC_DEFUN([AM_AUTOMAKE_VERSION],
-[am__api_version='1.14'
+[am__api_version='1.15'
 dnl Some users find AM_AUTOMAKE_VERSION and mistake it for a way to
 dnl require some minimum version.  Point them to the right macro.
-m4_if([$1], [1.14.1], [],
+m4_if([$1], [1.15.1], [],
       [AC_FATAL([Do not call $0, use AM_INIT_AUTOMAKE([$1]).])])dnl
 ])
 
@@ -51,14 +51,14 @@ m4_define([_AM_AUTOCONF_VERSION], [])
 # Call AM_AUTOMAKE_VERSION and AM_AUTOMAKE_VERSION so they can be traced.
 # This function is AC_REQUIREd by AM_INIT_AUTOMAKE.
 AC_DEFUN([AM_SET_CURRENT_AUTOMAKE_VERSION],
-[AM_AUTOMAKE_VERSION([1.14.1])dnl
+[AM_AUTOMAKE_VERSION([1.15.1])dnl
 m4_ifndef([AC_AUTOCONF_VERSION],
   [m4_copy([m4_PACKAGE_VERSION], [AC_AUTOCONF_VERSION])])dnl
 _AM_AUTOCONF_VERSION(m4_defn([AC_AUTOCONF_VERSION]))])
 
 # Figure out how to run the assembler.                      -*- Autoconf -*-
 
-# Copyright (C) 2001-2013 Free Software Foundation, Inc.
+# Copyright (C) 2001-2017 Free Software Foundation, Inc.
 #
 # This file is free software; the Free Software Foundation
 # gives unlimited permission to copy and/or distribute it,
@@ -78,7 +78,7 @@ _AM_IF_OPTION([no-dependencies],, [_AM_DEPENDENCIES([CCAS])])dnl
 
 # AM_AUX_DIR_EXPAND                                         -*- Autoconf -*-
 
-# Copyright (C) 2001-2013 Free Software Foundation, Inc.
+# Copyright (C) 2001-2017 Free Software Foundation, Inc.
 #
 # This file is free software; the Free Software Foundation
 # gives unlimited permission to copy and/or distribute it,
@@ -123,15 +123,14 @@ _AM_IF_OPTION([no-dependencies],, [_AM_DEPENDENCIES([CCAS])])dnl
 # configured tree to be moved without reconfiguration.
 
 AC_DEFUN([AM_AUX_DIR_EXPAND],
-[dnl Rely on autoconf to set up CDPATH properly.
+[AC_REQUIRE([AC_CONFIG_AUX_DIR_DEFAULT])dnl
-AC_PREREQ([2.50])dnl
+# Expand $ac_aux_dir to an absolute path.
-# expand $ac_aux_dir to an absolute path
+am_aux_dir=`cd "$ac_aux_dir" && pwd`
-am_aux_dir=`cd $ac_aux_dir && pwd`
 ])
 
 # AM_CONDITIONAL                                            -*- Autoconf -*-
 
-# Copyright (C) 1997-2013 Free Software Foundation, Inc.
+# Copyright (C) 1997-2017 Free Software Foundation, Inc.
 #
 # This file is free software; the Free Software Foundation
 # gives unlimited permission to copy and/or distribute it,
@@ -162,7 +161,7 @@ AC_CONFIG_COMMANDS_PRE(
 Usually this means the macro was only invoked conditionally.]])
 fi])])
 
-# Copyright (C) 1999-2013 Free Software Foundation, Inc.
+# Copyright (C) 1999-2017 Free Software Foundation, Inc.
 #
 # This file is free software; the Free Software Foundation
 # gives unlimited permission to copy and/or distribute it,
@@ -353,7 +352,7 @@ _AM_SUBST_NOTMAKE([am__nodep])dnl
 
 # Generate code to set up dependency tracking.              -*- Autoconf -*-
 
-# Copyright (C) 1999-2013 Free Software Foundation, Inc.
+# Copyright (C) 1999-2017 Free Software Foundation, Inc.
 #
 # This file is free software; the Free Software Foundation
 # gives unlimited permission to copy and/or distribute it,
@@ -429,7 +428,7 @@ AC_DEFUN([AM_OUTPUT_DEPENDENCY_COMMANDS],
 
 # Do all the work for Automake.                             -*- Autoconf -*-
 
-# Copyright (C) 1996-2013 Free Software Foundation, Inc.
+# Copyright (C) 1996-2017 Free Software Foundation, Inc.
 #
 # This file is free software; the Free Software Foundation
 # gives unlimited permission to copy and/or distribute it,
@@ -519,8 +518,8 @@ AC_REQUIRE([AC_PROG_MKDIR_P])dnl
 # <http://lists.gnu.org/archive/html/automake/2012-07/msg00001.html>
 # <http://lists.gnu.org/archive/html/automake/2012-07/msg00014.html>
 AC_SUBST([mkdir_p], ['$(MKDIR_P)'])
-# We need awk for the "check" target.  The system "awk" is bad on
+# We need awk for the "check" target (and possibly the TAP driver).  The
-# some platforms.
+# system "awk" is bad on some platforms.
 AC_REQUIRE([AC_PROG_AWK])dnl
 AC_REQUIRE([AC_PROG_MAKE_SET])dnl
 AC_REQUIRE([AM_SET_LEADING_DOT])dnl
@@ -593,7 +592,11 @@ to "yes", and re-run configure.
 END
     AC_MSG_ERROR([Your 'rm' program is bad, sorry.])
   fi
-fi])
+fi
+dnl The trailing newline in this macro's definition is deliberate, for
+dnl backward compatibility and to allow trailing 'dnl'-style comments
+dnl after the AM_INIT_AUTOMAKE invocation. See automake bug#16841.
+])
 
 dnl Hook into '_AC_COMPILER_EXEEXT' early to learn its expansion.  Do not
 dnl add the conditional right here, as _AC_COMPILER_EXEEXT may be further
@@ -622,7 +625,7 @@ for _am_header in $config_headers :; do
 done
 echo "timestamp for $_am_arg" >`AS_DIRNAME(["$_am_arg"])`/stamp-h[]$_am_stamp_count])
 
-# Copyright (C) 2001-2013 Free Software Foundation, Inc.
+# Copyright (C) 2001-2017 Free Software Foundation, Inc.
 #
 # This file is free software; the Free Software Foundation
 # gives unlimited permission to copy and/or distribute it,
@@ -633,7 +636,7 @@ echo "timestamp for $_am_arg" >`AS_DIRNAME(["$_am_arg"])`/stamp-h[]$_am_stamp_co
 # Define $install_sh.
 AC_DEFUN([AM_PROG_INSTALL_SH],
 [AC_REQUIRE([AM_AUX_DIR_EXPAND])dnl
-if test x"${install_sh}" != xset; then
+if test x"${install_sh+set}" != xset; then
   case $am_aux_dir in
   *\ * | *\	*)
     install_sh="\${SHELL} '$am_aux_dir/install-sh'" ;;
@@ -643,7 +646,7 @@ if test x"${install_sh}" != xset; then
 fi
 AC_SUBST([install_sh])])
 
-# Copyright (C) 2003-2013 Free Software Foundation, Inc.
+# Copyright (C) 2003-2017 Free Software Foundation, Inc.
 #
 # This file is free software; the Free Software Foundation
 # gives unlimited permission to copy and/or distribute it,
@@ -665,7 +668,7 @@ AC_SUBST([am__leading_dot])])
 # Add --enable-maintainer-mode option to configure.         -*- Autoconf -*-
 # From Jim Meyering
 
-# Copyright (C) 1996-2013 Free Software Foundation, Inc.
+# Copyright (C) 1996-2017 Free Software Foundation, Inc.
 #
 # This file is free software; the Free Software Foundation
 # gives unlimited permission to copy and/or distribute it,
@@ -700,7 +703,7 @@ AC_MSG_CHECKING([whether to enable maintainer-specific portions of Makefiles])
 
 # Check to see how 'make' treats includes.	            -*- Autoconf -*-
 
-# Copyright (C) 2001-2013 Free Software Foundation, Inc.
+# Copyright (C) 2001-2017 Free Software Foundation, Inc.
 #
 # This file is free software; the Free Software Foundation
 # gives unlimited permission to copy and/or distribute it,
@@ -750,7 +753,7 @@ rm -f confinc confmf
 
 # Fake the existence of programs that GNU maintainers use.  -*- Autoconf -*-
 
-# Copyright (C) 1997-2013 Free Software Foundation, Inc.
+# Copyright (C) 1997-2017 Free Software Foundation, Inc.
 #
 # This file is free software; the Free Software Foundation
 # gives unlimited permission to copy and/or distribute it,
@@ -789,7 +792,7 @@ fi
 
 # Helper functions for option handling.                     -*- Autoconf -*-
 
-# Copyright (C) 2001-2013 Free Software Foundation, Inc.
+# Copyright (C) 2001-2017 Free Software Foundation, Inc.
 #
 # This file is free software; the Free Software Foundation
 # gives unlimited permission to copy and/or distribute it,
@@ -818,7 +821,7 @@ AC_DEFUN([_AM_SET_OPTIONS],
 AC_DEFUN([_AM_IF_OPTION],
 [m4_ifset(_AM_MANGLE_OPTION([$1]), [$2], [$3])])
 
-# Copyright (C) 1999-2013 Free Software Foundation, Inc.
+# Copyright (C) 1999-2017 Free Software Foundation, Inc.
 #
 # This file is free software; the Free Software Foundation
 # gives unlimited permission to copy and/or distribute it,
@@ -865,7 +868,7 @@ AC_LANG_POP([C])])
 # For backward compatibility.
 AC_DEFUN_ONCE([AM_PROG_CC_C_O], [AC_REQUIRE([AC_PROG_CC])])
 
-# Copyright (C) 2001-2013 Free Software Foundation, Inc.
+# Copyright (C) 2001-2017 Free Software Foundation, Inc.
 #
 # This file is free software; the Free Software Foundation
 # gives unlimited permission to copy and/or distribute it,
@@ -884,7 +887,7 @@ AC_DEFUN([AM_RUN_LOG],
 
 # Check to make sure that the build environment is sane.    -*- Autoconf -*-
 
-# Copyright (C) 1996-2013 Free Software Foundation, Inc.
+# Copyright (C) 1996-2017 Free Software Foundation, Inc.
 #
 # This file is free software; the Free Software Foundation
 # gives unlimited permission to copy and/or distribute it,
@@ -965,7 +968,7 @@ AC_CONFIG_COMMANDS_PRE(
 rm -f conftest.file
 ])
 
-# Copyright (C) 2009-2013 Free Software Foundation, Inc.
+# Copyright (C) 2009-2017 Free Software Foundation, Inc.
 #
 # This file is free software; the Free Software Foundation
 # gives unlimited permission to copy and/or distribute it,
@@ -1025,7 +1028,7 @@ AC_SUBST([AM_BACKSLASH])dnl
 _AM_SUBST_NOTMAKE([AM_BACKSLASH])dnl
 ])
 
-# Copyright (C) 2001-2013 Free Software Foundation, Inc.
+# Copyright (C) 2001-2017 Free Software Foundation, Inc.
 #
 # This file is free software; the Free Software Foundation
 # gives unlimited permission to copy and/or distribute it,
@@ -1053,7 +1056,7 @@ fi
 INSTALL_STRIP_PROGRAM="\$(install_sh) -c -s"
 AC_SUBST([INSTALL_STRIP_PROGRAM])])
 
-# Copyright (C) 2006-2013 Free Software Foundation, Inc.
+# Copyright (C) 2006-2017 Free Software Foundation, Inc.
 #
 # This file is free software; the Free Software Foundation
 # gives unlimited permission to copy and/or distribute it,
@@ -1072,7 +1075,7 @@ AC_DEFUN([AM_SUBST_NOTMAKE], [_AM_SUBST_NOTMAKE($@)])
 
 # Check how to create a tarball.                            -*- Autoconf -*-
 
-# Copyright (C) 2004-2013 Free Software Foundation, Inc.
+# Copyright (C) 2004-2017 Free Software Foundation, Inc.
 #
 # This file is free software; the Free Software Foundation
 # gives unlimited permission to copy and/or distribute it,

algo-gate-api.c

@@ -160,7 +160,7 @@ bool register_algo_gate( int algo, algo_gate_t *gate )
      case ALGO_ALLIUM:       register_allium_algo       ( gate ); break;
      case ALGO_ANIME:        register_anime_algo        ( gate ); break;
      case ALGO_ARGON2:       register_argon2_algo       ( gate ); break;
-     case ALGO_ARGON2D250:   register_argon2d_crds_algo( gate ); break;
+     case ALGO_ARGON2D250:   register_argon2d_crds_algo ( gate ); break;
      case ALGO_ARGON2D500:   register_argon2d_dyn_algo  ( gate ); break;
      case ALGO_ARGON2D4096:  register_argon2d4096_algo  ( gate ); break;
      case ALGO_AXIOM:        register_axiom_algo        ( gate ); break;
@@ -172,6 +172,7 @@ bool register_algo_gate( int algo, algo_gate_t *gate )
      case ALGO_C11:          register_c11_algo          ( gate ); break;
      case ALGO_CRYPTOLIGHT:  register_cryptolight_algo  ( gate ); break;
      case ALGO_CRYPTONIGHT:  register_cryptonight_algo  ( gate ); break;
+     case ALGO_CRYPTONIGHTV7:register_cryptonightv7_algo( gate ); break;
      case ALGO_DECRED:       register_decred_algo       ( gate ); break;
      case ALGO_DEEP:         register_deep_algo         ( gate ); break;
      case ALGO_DMD_GR:       register_dmd_gr_algo       ( gate ); break;
@@ -210,7 +211,7 @@ bool register_algo_gate( int algo, algo_gate_t *gate )
      case ALGO_SKEIN2:       register_skein2_algo       ( gate ); break;
      case ALGO_SKUNK:        register_skunk_algo        ( gate ); break;
      case ALGO_TIMETRAVEL:   register_timetravel_algo   ( gate ); break;
-     case ALGO_TIMETRAVEL10: register_timetravel10_algo( gate ); break;
+     case ALGO_TIMETRAVEL10: register_timetravel10_algo ( gate ); break;
      case ALGO_TRIBUS:       register_tribus_algo       ( gate ); break;
      case ALGO_VANILLA:      register_vanilla_algo      ( gate ); break;
      case ALGO_VELTOR:       register_veltor_algo       ( gate ); break;

algo/argon2/argon2d/argon2d-gate.c

@@ -28,6 +28,7 @@ void argon2d_crds_hash( void *output, const void *input )
 	context.lanes = 4;    // Degree of Parallelism
 	context.threads = 1;  // Threads
 	context.t_cost = 1;   // Iterations
+        context.version = ARGON2_VERSION_10;
 
 	argon2_ctx( &context, Argon2_d );
 }
@@ -71,7 +72,6 @@ bool register_argon2d_crds_algo( algo_gate_t* gate )
         gate->hash = (void*)&argon2d_crds_hash;
         gate->set_target = (void*)&scrypt_set_target;
         gate->optimizations = SSE2_OPT | AVX2_OPT | AVX512_OPT;
-        ARGON2_VERSION_NUMBER = ARGON2_VERSION_10;
         return true;
 }
 
@@ -98,6 +98,7 @@ void argon2d_dyn_hash( void *output, const void *input )
     context.lanes = 8;     // Degree of Parallelism
     context.threads = 1;   // Threads
     context.t_cost = 2;    // Iterations
+    context.version = ARGON2_VERSION_10;
 
     argon2_ctx( &context, Argon2_d );
 }
@@ -141,10 +142,11 @@ bool register_argon2d_dyn_algo( algo_gate_t* gate )
         gate->hash = (void*)&argon2d_dyn_hash;
         gate->set_target = (void*)&scrypt_set_target;
         gate->optimizations = SSE2_OPT | AVX2_OPT | AVX512_OPT;
-        ARGON2_VERSION_NUMBER = ARGON2_VERSION_10;
         return true;
 }
 
+// Unitus
+
 int scanhash_argon2d4096( int thr_id, struct work *work, uint32_t max_nonce,
                            uint64_t *hashes_done)
 {
@@ -166,7 +168,7 @@ int scanhash_argon2d4096( int thr_id, struct work *work, uint32_t max_nonce,
    do {
       be32enc( &endiandata[19], n );
       argon2d_hash_raw( t_cost, m_cost, parallelism, (char*) endiandata, 80,
-                        (char*) endiandata, 80, (char*) vhash, 32 );
+                 (char*) endiandata, 80, (char*) vhash, 32, ARGON2_VERSION_13 );
       if ( vhash[7] < Htarg && fulltest( vhash, ptarget ) )
       {
          *hashes_done = n - first_nonce + 1;
@@ -191,7 +193,6 @@ bool register_argon2d4096_algo( algo_gate_t* gate )
         gate->set_target = (void*)&scrypt_set_target;
         gate->get_max64  = (void*)&get_max64_0x1ff;
         gate->optimizations = SSE2_OPT | AVX2_OPT | AVX512_OPT;
-        ARGON2_VERSION_NUMBER = ARGON2_VERSION_13;
         return true;
 }
 

algo/argon2/argon2d/argon2d-gate.h

@@ -4,7 +4,7 @@
 #include "algo-gate-api.h"
 #include <stdint.h>
 
-// Credits
+// Credits: version = 0x10, m_cost = 250.
 bool register_argon2d_crds_algo( algo_gate_t* gate );
 
 void argon2d_crds_hash( void *state, const void *input );
@@ -12,7 +12,7 @@ void argon2d_crds_hash( void *state, const void *input );
 int scanhash_argon2d_crds( int thr_id, struct work *work, uint32_t max_nonce,
                     uint64_t *hashes_done );
 
-// Dynamic
+// Dynamic: version = 0x10, m_cost = 500.
 bool register_argon2d_dyn_algo( algo_gate_t* gate );
 
 void argon2d_dyn_hash( void *state, const void *input );
@@ -21,5 +21,11 @@ int scanhash_argon2d_dyn( int thr_id, struct work *work, uint32_t max_nonce,
                     uint64_t *hashes_done );
 
 
+// Unitus: version = 0x13, m_cost = 4096.
+bool register_argon2d4096_algo( algo_gate_t* gate );
+
+int scanhash_argon2d4096( int thr_id, struct work *work, uint32_t max_nonce,
+                    uint64_t *hashes_done );
+
 #endif
 

algo/argon2/argon2d/argon2d/argon2.c

@@ -180,60 +180,65 @@ int argon2i_hash_encoded(const uint32_t t_cost, const uint32_t m_cost,
                          const uint32_t parallelism, const void *pwd,
                          const size_t pwdlen, const void *salt,
                          const size_t saltlen, const size_t hashlen,
-                         char *encoded, const size_t encodedlen) {
+                         char *encoded, const size_t encodedlen,
+                         const uint32_t version) {
 
     return argon2_hash(t_cost, m_cost, parallelism, pwd, pwdlen, salt, saltlen,
                        NULL, hashlen, encoded, encodedlen, Argon2_i,
-                       ARGON2_VERSION_NUMBER);
+                       version );
 }
 
 int argon2i_hash_raw(const uint32_t t_cost, const uint32_t m_cost,
                      const uint32_t parallelism, const void *pwd,
                      const size_t pwdlen, const void *salt,
-                     const size_t saltlen, void *hash, const size_t hashlen) {
+                     const size_t saltlen, void *hash, const size_t hashlen,
+                     const uint32_t version ) {
 
     return argon2_hash(t_cost, m_cost, parallelism, pwd, pwdlen, salt, saltlen,
-                       hash, hashlen, NULL, 0, Argon2_i, ARGON2_VERSION_NUMBER);
+                       hash, hashlen, NULL, 0, Argon2_i, version );
 }
 
 int argon2d_hash_encoded(const uint32_t t_cost, const uint32_t m_cost,
                          const uint32_t parallelism, const void *pwd,
                          const size_t pwdlen, const void *salt,
                          const size_t saltlen, const size_t hashlen,
-                         char *encoded, const size_t encodedlen) {
+                         char *encoded, const size_t encodedlen,
+                         const uint32_t version ) {
 
     return argon2_hash(t_cost, m_cost, parallelism, pwd, pwdlen, salt, saltlen,
                        NULL, hashlen, encoded, encodedlen, Argon2_d,
-                       ARGON2_VERSION_NUMBER);
+                       version );
 }
 
 int argon2d_hash_raw(const uint32_t t_cost, const uint32_t m_cost,
                      const uint32_t parallelism, const void *pwd,
                      const size_t pwdlen, const void *salt,
-                     const size_t saltlen, void *hash, const size_t hashlen) {
+                     const size_t saltlen, void *hash, const size_t hashlen,
+                     const uint32_t version ) {
 
     return argon2_hash(t_cost, m_cost, parallelism, pwd, pwdlen, salt, saltlen,
-                       hash, hashlen, NULL, 0, Argon2_d, ARGON2_VERSION_NUMBER);
+                       hash, hashlen, NULL, 0, Argon2_d, version );
 }
 
 int argon2id_hash_encoded(const uint32_t t_cost, const uint32_t m_cost,
                           const uint32_t parallelism, const void *pwd,
                           const size_t pwdlen, const void *salt,
                           const size_t saltlen, const size_t hashlen,
-                          char *encoded, const size_t encodedlen) {
+                          char *encoded, const size_t encodedlen,
+                          const uint32_t version ) {
 
     return argon2_hash(t_cost, m_cost, parallelism, pwd, pwdlen, salt, saltlen,
                        NULL, hashlen, encoded, encodedlen, Argon2_id,
-                       ARGON2_VERSION_NUMBER);
+                       version);
 }
 
 int argon2id_hash_raw(const uint32_t t_cost, const uint32_t m_cost,
                       const uint32_t parallelism, const void *pwd,
                       const size_t pwdlen, const void *salt,
-                      const size_t saltlen, void *hash, const size_t hashlen) {
+                      const size_t saltlen, void *hash, const size_t hashlen,
+                      const uint32_t version ) {
     return argon2_hash(t_cost, m_cost, parallelism, pwd, pwdlen, salt, saltlen,
-                       hash, hashlen, NULL, 0, Argon2_id,
+                       hash, hashlen, NULL, 0, Argon2_id, version );
-                       ARGON2_VERSION_NUMBER);
 }
 
 static int argon2_compare(const uint8_t *b1, const uint8_t *b2, size_t len) {
@@ -443,10 +448,11 @@ const char *argon2_error_message(int error_code) {
         return "Unknown error code";
     }
 }
-
+/*
 size_t argon2_encodedlen(uint32_t t_cost, uint32_t m_cost, uint32_t parallelism,
                          uint32_t saltlen, uint32_t hashlen, argon2_type type) {
   return strlen("$$v=$m=,t=,p=$$") + strlen(argon2_type2string(type, 0)) +
          numlen(t_cost) + numlen(m_cost) + numlen(parallelism) +
          b64len(saltlen) + b64len(hashlen) + numlen(ARGON2_VERSION_NUMBER) + 1;
 }
+*/

algo/argon2/argon2d/argon2d/argon2.h

@@ -224,22 +224,9 @@ typedef enum Argon2_type {
   Argon2_id = 2
 } argon2_type;
 
-// Version implementation sucks, needs to be changed.
-// It should be set in the ctx before calling the argon2_ctx.
-// It should be a function arg of *hash_raw & *hash_encoded.
-// Then remove initialization from gate register function.
 /* Version of the algorithm */
-/*
-typedef enum Argon2_version {
-    ARGON2_VERSION_10 = 0x10,
-    ARGON2_VERSION_13 = 0x13,
-    ARGON2_VERSION_NUMBER = ARGON2_VERSION_10
-} argon2_version;
-*/
-
 #define ARGON2_VERSION_10 0x10
 #define ARGON2_VERSION_13 0x13
-uint32_t ARGON2_VERSION_NUMBER;
 
 /*
  * Function that gives the string representation of an argon2_type.
@@ -277,7 +264,8 @@ ARGON2_PUBLIC int argon2i_hash_encoded(const uint32_t t_cost,
                                        const void *pwd, const size_t pwdlen,
                                        const void *salt, const size_t saltlen,
                                        const size_t hashlen, char *encoded,
-                                       const size_t encodedlen);
+                                       const size_t encodedlen,
+                                       const uint32_t version );
 
 /**
  * Hashes a password with Argon2i, producing a raw hash at @hash
@@ -297,7 +285,8 @@ ARGON2_PUBLIC int argon2i_hash_raw(const uint32_t t_cost, const uint32_t m_cost,
                                    const uint32_t parallelism, const void *pwd,
                                    const size_t pwdlen, const void *salt,
                                    const size_t saltlen, void *hash,
-                                   const size_t hashlen);
+                                   const size_t hashlen,
+                                   const uint32_t version );
 
 ARGON2_PUBLIC int argon2d_hash_encoded(const uint32_t t_cost,
                                        const uint32_t m_cost,
@@ -305,13 +294,15 @@ ARGON2_PUBLIC int argon2d_hash_encoded(const uint32_t t_cost,
                                        const void *pwd, const size_t pwdlen,
                                        const void *salt, const size_t saltlen,
                                        const size_t hashlen, char *encoded,
-                                       const size_t encodedlen);
+                                       const size_t encodedlen,
+                                       const uint32_t version );
 
 ARGON2_PUBLIC int argon2d_hash_raw(const uint32_t t_cost, const uint32_t m_cost,
                                    const uint32_t parallelism, const void *pwd,
                                    const size_t pwdlen, const void *salt,
                                    const size_t saltlen, void *hash,
-                                   const size_t hashlen);
+                                   const size_t hashlen,
+                                   const uint32_t version );
 
 ARGON2_PUBLIC int argon2id_hash_encoded(const uint32_t t_cost,
                                         const uint32_t m_cost,
@@ -319,14 +310,16 @@ ARGON2_PUBLIC int argon2id_hash_encoded(const uint32_t t_cost,
                                         const void *pwd, const size_t pwdlen,
                                         const void *salt, const size_t saltlen,
                                         const size_t hashlen, char *encoded,
-                                        const size_t encodedlen);
+                                        const size_t encodedlen,
+                                        const uint32_t version );
 
 ARGON2_PUBLIC int argon2id_hash_raw(const uint32_t t_cost,
                                     const uint32_t m_cost,
                                     const uint32_t parallelism, const void *pwd,
                                     const size_t pwdlen, const void *salt,
                                     const size_t saltlen, void *hash,
-                                    const size_t hashlen);
+                                    const size_t hashlen,
+                                    const uint32_t version );
 
 /* generic function underlying the above ones */
 ARGON2_PUBLIC int argon2_hash(const uint32_t t_cost, const uint32_t m_cost,
@@ -335,7 +328,7 @@ ARGON2_PUBLIC int argon2_hash(const uint32_t t_cost, const uint32_t m_cost,
                               const size_t saltlen, void *hash,
                               const size_t hashlen, char *encoded,
                               const size_t encodedlen, argon2_type type,
-                              const uint32_t version);
+                              const uint32_t version );
 
 /**
  * Verifies a password against an encoded string

algo/argon2/argon2d/argon2d/core.c

@@ -544,7 +544,8 @@ void initial_hash(uint8_t *blockhash, argon2_context *context,
     store32(&value, context->t_cost);
     blake2b_update(&BlakeHash, (const uint8_t *)&value, sizeof(value));
 
-    store32(&value, ARGON2_VERSION_NUMBER);
+//    store32(&value, ARGON2_VERSION_NUMBER);
+    store32(&value, context->version);
     blake2b_update(&BlakeHash, (const uint8_t *)&value, sizeof(value));
 
     store32(&value, (uint32_t)type);

algo/cryptonight/cryptolight.c

@@ -325,7 +325,7 @@ int scanhash_cryptolight(int thr_id, struct work *work,
 
 	struct cryptonight_ctx *ctx = (struct cryptonight_ctx*)malloc(sizeof(struct cryptonight_ctx));
 
-#ifndef NO_AES_NI
+#if defined(__AES__)
 		do {
 			*nonceptr = ++n;
 			cryptolight_hash_ctx_aes_ni(hash, pdata, 76, ctx);

algo/cryptonight/cryptonight-aesni.c

@@ -1,14 +1,11 @@
+#if defined(__AES__)
+
 #include <x86intrin.h>
 #include <memory.h>
 #include "cryptonight.h"
 #include "miner.h"
 #include "crypto/c_keccak.h"
 #include <immintrin.h>
-//#include "avxdefs.h"
-
-void aesni_parallel_noxor(uint8_t *long_state, uint8_t *text, uint8_t *ExpandedKey);
-void aesni_parallel_xor(uint8_t *text, uint8_t *ExpandedKey, uint8_t *long_state);
-void that_fucking_loop(uint8_t a[16], uint8_t b[16], uint8_t *long_state);
 
 static inline void ExpandAESKey256_sub1(__m128i *tmp1, __m128i *tmp2)
 {
@@ -25,7 +22,6 @@ static inline void ExpandAESKey256_sub1(__m128i *tmp1, __m128i *tmp2)
 
 static inline void ExpandAESKey256_sub2(__m128i *tmp1, __m128i *tmp3)
 {
-#ifndef NO_AES_NI
 	__m128i tmp2, tmp4;
 	
 	tmp4 = _mm_aeskeygenassist_si128(*tmp1, 0x00);
@@ -37,14 +33,12 @@ static inline void ExpandAESKey256_sub2(__m128i *tmp1, __m128i *tmp3)
 	tmp4 = _mm_slli_si128(tmp4, 0x04);
 	*tmp3 = _mm_xor_si128(*tmp3, tmp4);
 	*tmp3 = _mm_xor_si128(*tmp3, tmp2);
-#endif
 }
 
 // Special thanks to Intel for helping me
 // with ExpandAESKey256() and its subroutines
 static inline void ExpandAESKey256(char *keybuf)
 {
-#ifndef NO_AES_NI
 	__m128i tmp1, tmp2, tmp3, *keys;
 	
 	keys = (__m128i *)keybuf;
@@ -91,7 +85,6 @@ static inline void ExpandAESKey256(char *keybuf)
 	tmp2 = _mm_aeskeygenassist_si128(tmp3, 0x40);
 	ExpandAESKey256_sub1(&tmp1, &tmp2);
 	keys[14] = tmp1;
-#endif
 }
 
 // align to 64 byte cache line
@@ -109,13 +102,19 @@ static __thread cryptonight_ctx ctx;
 
 void cryptonight_hash_aes( void *restrict output, const void *input, int len )
 {
-#ifndef NO_AES_NI
-
     uint8_t ExpandedKey[256] __attribute__((aligned(64)));
     __m128i *longoutput, *expkey, *xmminput;
     size_t i, j;
     
     keccak( (const uint8_t*)input, 76, (char*)&ctx.state.hs.b, 200 );
+
+    if ( cryptonightV7 && len < 43 )
+      return;
+
+    const uint64_t tweak = cryptonightV7 
+                         ? *((const uint64_t*) (((const uint8_t*)input) + 35))
+                           ^ ctx.state.hs.w[24] : 0; 
+
     memcpy( ExpandedKey, ctx.state.hs.b, AES_KEY_SIZE );
     ExpandAESKey256( ExpandedKey );
     memcpy( ctx.text, ctx.state.init, INIT_SIZE_BYTE );
@@ -215,6 +214,14 @@ void cryptonight_hash_aes( void *restrict output, const void *input, int len )
         b_x = _mm_xor_si128( b_x, c_x );
         nextblock = (uint64_t *)&ctx.long_state[c[0] & 0x1FFFF0];
         _mm_store_si128( lsa, b_x );
+
+        if ( cryptonightV7 )
+        {
+           const uint8_t tmp = ( (const uint8_t*)(lsa) )[11];
+           const uint8_t index = ( ( (tmp >> 3) & 6 ) | (tmp & 1) ) << 1;
+           ((uint8_t*)(lsa))[11] = tmp ^ ( ( 0x75310 >> index) & 0x30 );
+        } 
+
 	b[0] = nextblock[0];
 	b[1] = nextblock[1];
 
@@ -227,10 +234,14 @@ void cryptonight_hash_aes( void *restrict output, const void *input, int len )
 		 : "cc" );
 
         b_x = c_x;
-        nextblock[0] = a[0] + hi;
+
-        nextblock[1] = a[1] + lo;
+        a[0] += hi;
-        a[0] = b[0] ^ nextblock[0];
+        a[1] += lo;
-        a[1] = b[1] ^ nextblock[1];
+        nextblock[0] = a[0];
+        nextblock[1] = cryptonightV7 ? a[1] ^ tweak : a[1];
+        a[0] ^= b[0];
+        a[1] ^= b[1];
+
         lsa = (__m128i*)&ctx.long_state[ a[0] & 0x1FFFF0 ];
         a_x = _mm_load_si128( (__m128i*)a );
         c_x = _mm_load_si128( lsa );
@@ -241,6 +252,14 @@ void cryptonight_hash_aes( void *restrict output, const void *input, int len )
     b_x = _mm_xor_si128( b_x, c_x );
     nextblock = (uint64_t *)&ctx.long_state[c[0] & 0x1FFFF0];
     _mm_store_si128( lsa, b_x );
+
+    if ( cryptonightV7 )
+    {
+       const uint8_t tmp = ( (const uint8_t*)(lsa) )[11];
+       const uint8_t index = ( ( (tmp >> 3) & 6 ) | (tmp & 1) ) << 1;
+       ((uint8_t*)(lsa))[11] = tmp ^ ( ( 0x75310 >> index) & 0x30 );
+    }
+
     b[0] = nextblock[0];
     b[1] = nextblock[1];
 
@@ -251,8 +270,12 @@ void cryptonight_hash_aes( void *restrict output, const void *input, int len )
                "rm" ( b[0] )
              : "cc" );
 
-    nextblock[0] = a[0] + hi;
+    a[0] += hi;
-    nextblock[1] = a[1] + lo;
+    a[1] += lo;
+    nextblock[0] = a[0];
+    nextblock[1] = cryptonightV7 ? a[1] ^ tweak : a[1];
+    a[0] ^= b[0];
+    a[1] ^= b[1];
 
     memcpy( ExpandedKey, &ctx.state.hs.b[32], AES_KEY_SIZE );
     ExpandAESKey256( ExpandedKey );
@@ -330,5 +353,5 @@ void cryptonight_hash_aes( void *restrict output, const void *input, int len )
     keccakf( (uint64_t*)&ctx.state.hs.w, 24 );
     extra_hashes[ctx.state.hs.b[0] & 3](&ctx.state, 200, output);
 
-#endif
 }
+#endif

algo/cryptonight/cryptonight-common.c

@@ -7,11 +7,11 @@
 #include "cpuminer-config.h"
 #include "algo-gate-api.h"
 
-#ifndef NO_AES_NI
+#if defined(__AES__)
   #include "algo/groestl/aes_ni/hash-groestl256.h"
-#endif
+#else
-
 #include "crypto/c_groestl.h"
+#endif
 #include "crypto/c_blake256.h"
 #include "crypto/c_jh.h"
 #include "crypto/c_skein.h"
@@ -30,12 +30,12 @@ void do_blake_hash(const void* input, size_t len, char* output) {
 }
 
 void do_groestl_hash(const void* input, size_t len, char* output) {
-#ifdef NO_AES_NI
+#if defined(__AES__)
-    groestl(input, len * 8, (uint8_t*)output);
-#else
     hashState_groestl256 ctx;
     init_groestl256( &ctx, 32 );
     update_and_final_groestl256( &ctx, output, input, len * 8 );
+#else
+    groestl(input, len * 8, (uint8_t*)output);
 #endif
 }
 
@@ -52,23 +52,24 @@ void (* const extra_hashes[4])( const void *, size_t, char *) =
 
 void cryptonight_hash( void *restrict output, const void *input, int len )
 {
-
+#if defined(__AES__)
-#ifdef NO_AES_NI
-  cryptonight_hash_ctx ( output, input, len );
-#else 
   cryptonight_hash_aes( output, input, len );
+#else
+  cryptonight_hash_ctx ( output, input, len );
 #endif
 }
 
 void cryptonight_hash_suw( void *restrict output, const void *input )
 {
-#ifdef NO_AES_NI
+#if defined(__AES__)
-  cryptonight_hash_ctx ( output, input, 76 );
-#else
   cryptonight_hash_aes( output, input, 76 );
+#else
+  cryptonight_hash_ctx ( output, input, 76 );
 #endif
 }
 
+bool cryptonightV7 = false;
+
 int scanhash_cryptonight( int thr_id, struct work *work, uint32_t max_nonce,
                    uint64_t *hashes_done )
  {
@@ -80,6 +81,11 @@ int scanhash_cryptonight( int thr_id, struct work *work, uint32_t max_nonce,
     const uint32_t first_nonce = n + 1;
     const uint32_t Htarg = ptarget[7];
     uint32_t hash[32 / 4] __attribute__((aligned(32)));
+
+//    if (  (  cryptonightV7 && ( *(uint8_t*)pdata <  7 ) )
+//       || ( !cryptonightV7 && ( *(uint8_t*)pdata == 7 ) ) )
+//          applog(LOG_WARNING,"Cryptonight variant mismatch, shares may be rejected.");
+
     do
     {
        *nonceptr = ++n;
@@ -87,6 +93,7 @@ int scanhash_cryptonight( int thr_id, struct work *work, uint32_t max_nonce,
        if (unlikely( hash[7] < Htarg ))
        {
            *hashes_done = n - first_nonce + 1;
+//           work_set_target_ratio( work, hash );
 	   return true;
        }
     } while (likely((n <= max_nonce && !work_restart[thr_id].restart)));
@@ -97,6 +104,19 @@ int scanhash_cryptonight( int thr_id, struct work *work, uint32_t max_nonce,
 
 bool register_cryptonight_algo( algo_gate_t* gate )
 {
+  cryptonightV7 = false;
+  register_json_rpc2( gate );
+  gate->optimizations = SSE2_OPT | AES_OPT;
+  gate->scanhash         = (void*)&scanhash_cryptonight;
+  gate->hash             = (void*)&cryptonight_hash;
+  gate->hash_suw         = (void*)&cryptonight_hash_suw;  
+  gate->get_max64        = (void*)&get_max64_0x40LL;
+  return true;
+};
+
+bool register_cryptonightv7_algo( algo_gate_t* gate )
+{
+  cryptonightV7 = true;
   register_json_rpc2( gate );
   gate->optimizations = SSE2_OPT | AES_OPT;
   gate->scanhash      = (void*)&scanhash_cryptonight;

algo/cryptonight/cryptonight.c

@@ -20,8 +20,8 @@
 #include "crypto/c_jh.h"
 #include "crypto/c_skein.h"
 #include "crypto/int-util.h"
-#include "crypto/hash-ops.h"
+//#include "crypto/hash-ops.h"
-//#include "cryptonight.h"
+#include "cryptonight.h"
 
 #if USE_INT128
 
@@ -51,6 +51,7 @@ typedef __uint128_t uint128_t;
 #define INIT_SIZE_BLK   8
 #define INIT_SIZE_BYTE (INIT_SIZE_BLK * AES_BLOCK_SIZE)
 
+/*
 #pragma pack(push, 1)
 union cn_slow_hash_state {
 	union hash_state hs;
@@ -78,6 +79,7 @@ static void do_skein_hash(const void* input, size_t len, char* output) {
 	int r = skein_hash(8 * HASH_SIZE, input, 8 * len, (uint8_t*)output);
 	assert(likely(SKEIN_SUCCESS == r));
 }
+*/
 
 extern int aesb_single_round(const uint8_t *in, uint8_t*out, const uint8_t *expandedKey);
 extern int aesb_pseudo_round_mut(uint8_t *val, uint8_t *expandedKey);
@@ -120,9 +122,11 @@ static uint64_t mul128(uint64_t multiplier, uint64_t multiplicand, uint64_t* pro
 extern uint64_t mul128(uint64_t multiplier, uint64_t multiplicand, uint64_t* product_hi);
 #endif
 
+/*
 static void (* const extra_hashes[4])(const void *, size_t, char *) = {
 		do_blake_hash, do_groestl_hash, do_jh_hash, do_skein_hash
 };
+*/
 
 static inline size_t e2i(const uint8_t* a) {
 #if !LITE
@@ -132,14 +136,16 @@ static inline size_t e2i(const uint8_t* a) {
 #endif
 }
 
-static inline void mul_sum_xor_dst(const uint8_t* a, uint8_t* c, uint8_t* dst) {
+static inline void mul_sum_xor_dst( const uint8_t* a, uint8_t* c, uint8_t* dst, 
+         const uint64_t tweak )
+{
 	uint64_t hi, lo = mul128(((uint64_t*) a)[0], ((uint64_t*) dst)[0], &hi) + ((uint64_t*) c)[1];
 	hi += ((uint64_t*) c)[0];
 
 	((uint64_t*) c)[0] = ((uint64_t*) dst)[0] ^ hi;
 	((uint64_t*) c)[1] = ((uint64_t*) dst)[1] ^ lo;
 	((uint64_t*) dst)[0] = hi;
-	((uint64_t*) dst)[1] = lo;
+	((uint64_t*) dst)[1] = cryptonightV7 ? lo ^ tweak : lo;
 }
 
 static inline void xor_blocks(uint8_t* a, const uint8_t* b) {
@@ -174,7 +180,15 @@ static __thread cryptonight_ctx ctx;
 
 void cryptonight_hash_ctx(void* output, const void* input, int len)
 {
-	hash_process(&ctx.state.hs, (const uint8_t*) input, len);
+//    hash_process(&ctx.state.hs, (const uint8_t*) input, len);
+    keccak( (const uint8_t*)input, 76, (char*)&ctx.state.hs.b, 200 );
+
+    if ( cryptonightV7 && len < 43 )
+      return;
+    const uint64_t tweak = cryptonightV7
+                         ? *((const uint64_t*) (((const uint8_t*)input) + 35))
+                           ^ ctx.state.hs.w[24] : 0;
+
     ctx.aes_ctx = (oaes_ctx*) oaes_alloc();
 
     __builtin_prefetch( ctx.text,             0, 3 );
@@ -211,7 +225,8 @@ void cryptonight_hash_ctx(void* output, const void* input, int len)
 	xor_blocks_dst(&ctx.state.k[0], &ctx.state.k[32], ctx.a);
 	xor_blocks_dst(&ctx.state.k[16], &ctx.state.k[48], ctx.b);
 
-	for (i = 0; likely(i < ITER / 4); ++i) {
+	for (i = 0; likely(i < ITER / 4); ++i)
+        {
            /* Dependency chain: address -> read value ------+
             * written value <-+ hard function (AES or MUL) <+
             * next address  <-+
@@ -220,14 +235,34 @@ void cryptonight_hash_ctx(void* output, const void* input, int len)
            j = e2i(ctx.a);
            aesb_single_round(&ctx.long_state[j], ctx.c, ctx.a);
            xor_blocks_dst(ctx.c, ctx.b, &ctx.long_state[j]);
+
+           if ( cryptonightV7 )
+           {
+              uint8_t *lsa = (uint8_t*)&ctx.long_state[((uint64_t *)(ctx.a))[0] & 0x1FFFF0];
+              const uint8_t tmp = lsa[11];
+              const uint8_t index = ( ( (tmp >> 3) & 6 ) | (tmp & 1) ) << 1;
+              lsa[11] = tmp ^ ( ( 0x75310 >> index) & 0x30 );
+           }
+
            /* Iteration 2 */
-		mul_sum_xor_dst(ctx.c, ctx.a, &ctx.long_state[e2i(ctx.c)]);
+           mul_sum_xor_dst(ctx.c, ctx.a, &ctx.long_state[e2i(ctx.c)], tweak );
+
            /* Iteration 3 */
            j = e2i(ctx.a);
            aesb_single_round(&ctx.long_state[j], ctx.b, ctx.a);
            xor_blocks_dst(ctx.b, ctx.c, &ctx.long_state[j]);
+
+           if ( cryptonightV7 )
+           {
+              uint8_t *lsa = (uint8_t*)&ctx.long_state[((uint64_t *)(ctx.a))[0] & 0x1FFFF0];
+              const uint8_t tmp = lsa[11];
+              const uint8_t index = ( ( (tmp >> 3) & 6 ) | (tmp & 1) ) << 1;
+              lsa[11] = tmp ^ ( ( 0x75310 >> index) & 0x30 );
+           }
+
            /* Iteration 4 */
-		mul_sum_xor_dst(ctx.b, ctx.a, &ctx.long_state[e2i(ctx.b)]);
+           mul_sum_xor_dst(ctx.b, ctx.a, &ctx.long_state[e2i(ctx.b)], tweak );
+
 	}
 
     __builtin_prefetch( ctx.text,             0, 3 );
@@ -266,7 +301,8 @@ void cryptonight_hash_ctx(void* output, const void* input, int len)
 		aesb_pseudo_round_mut(&ctx.text[7 * AES_BLOCK_SIZE], ctx.aes_ctx->key->exp_data);
 	}
 	memcpy(ctx.state.init, ctx.text, INIT_SIZE_BYTE);
-	hash_permutation(&ctx.state.hs);
+//	hash_permutation(&ctx.state.hs);
+        keccakf( (uint64_t*)&ctx.state.hs.w, 24 );
 	/*memcpy(hash, &state, 32);*/
 	extra_hashes[ctx.state.hs.b[0] & 3](&ctx.state, 200, output);
 	oaes_free((OAES_CTX **) &ctx.aes_ctx);

algo/cryptonight/cryptonight.h

@@ -45,5 +45,7 @@ int scanhash_cryptonight( int thr_id, struct work *work, uint32_t max_nonce,
 
 void cryptonight_hash_aes( void *restrict output, const void *input, int len );
 
+extern bool cryptonightV7;
+
 #endif
 

algo/lyra2/lyra2h-4way.c

@@ -67,7 +67,7 @@ int scanhash_lyra2h_4way( int thr_id, struct work *work, uint32_t max_nonce,
    if ( opt_benchmark )
       ptarget[7] = 0x0000ff;
 
-   for ( int i=0; i < 19; i++ )
+   for ( int i=0; i < 20; i++ )
       be32enc( &edata[i], pdata[i] );
 
    mm_interleave_4x32( vdata, edata, edata, edata, edata, 640 );

algo/lyra2/lyra2z-4way.c

@@ -67,7 +67,7 @@ int scanhash_lyra2z_4way( int thr_id, struct work *work, uint32_t max_nonce,
    if ( opt_benchmark )
       ptarget[7] = 0x0000ff;
 
-   for ( int i=0; i < 19; i++ )
+   for ( int i=0; i < 20; i++ )
       be32enc( &edata[i], pdata[i] );
 
    mm_interleave_4x32( vdata, edata, edata, edata, edata, 640 );

algo/sha/sha256t-4way.c

@@ -60,7 +60,7 @@ int scanhash_sha256t_8way( int thr_id, struct work *work,
                                0xFFFF0000,
                                         0 };
 
-   for ( int k = 0; k < 19; k++ )
+   for ( int k = 0; k < 20; k++ )
       be32enc( &edata[k], pdata[k] );
 
    mm256_interleave_8x32( vdata, edata, edata, edata, edata,

algo/x11/x11evo-gate.c

@@ -34,8 +34,10 @@ int nextPerm( uint8_t n[], uint32_t count )
    tail = i;
 
    if ( tail > 0 )
+   {
       for ( j = count - 1; j>tail && n[j] <= n[tail - 1]; j-- );
       evo_swap( &n[tail - 1], &n[j] );
+   }
 
    for ( i = tail, j = count - 1; i<j; i++, j-- )
       evo_swap( &n[i], &n[j] );

algo/x17/x16r-4way.c

@@ -68,13 +68,7 @@ void x16r_4way_hash( void* output, const void* input )
    uint32_t hash2[24] __attribute__ ((aligned (64)));
    uint32_t hash3[24] __attribute__ ((aligned (64)));
    uint32_t vhash[24*4] __attribute__ ((aligned (64)));
-//   uint32_t inp0[24] __attribute__ ((aligned (64)));
-//   uint32_t inp1[24] __attribute__ ((aligned (64)));
-//   uint32_t inp2[24] __attribute__ ((aligned (64)));
-//   uint32_t inp3[24] __attribute__ ((aligned (64)));
-
    x16r_4way_ctx_holder ctx;
-   
    void *in0 = (void*) hash0;
    void *in1 = (void*) hash1;
    void *in2 = (void*) hash2;
@@ -290,10 +284,6 @@ void x16r_4way_hash( void* output, const void* input )
              mm256_deinterleave_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
          break;
       }
-//      in0 = (void*) hash0;
-//      in1 = (void*) hash1;
-//      in2 = (void*) hash2;
-//      in3 = (void*) hash3;
       size = 64;
    }
    memcpy( output,    hash0, 32 );
@@ -325,7 +315,7 @@ int scanhash_x16r_4way( int thr_id, struct work *work, uint32_t max_nonce,
    if ( s_ntime != endiandata[17] )
    {
       uint32_t ntime = swab32(pdata[17]);
-      x16_r_s_getAlgoString( (const char*) (&endiandata[1]), hashOrder );
+      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 );

algo/x17/x16r-gate.c

@@ -1,6 +1,6 @@
 #include "x16r-gate.h"
 
-void x16r_getAlgoString( const char* prevblock, char *output )
+void x16r_getAlgoString( const uint8_t* prevblock, char *output )
 {
    char *sptr = output;
    for ( int j = 0; j < X16R_HASH_FUNC_COUNT; j++ )
@@ -16,14 +16,13 @@ void x16r_getAlgoString( const char* prevblock, char *output )
    *sptr = '\0';
 }
 
-void x16s_getAlgoString( const char* prevblock, char *output )
+void x16s_getAlgoString( const uint8_t* prevblock, char *output )
 {
-   uint8_t* data = (uint8_t*)prevblock;
    strcpy( output, "0123456789ABCDEF" );
    for ( int i = 0; i < 16; i++ )
    {
       uint8_t b = (15 - i) >> 1; // 16 ascii hex chars, reversed
-      uint8_t algoDigit = (i & 1) ? data[b] & 0xF : data[b] >> 4;
+      uint8_t algoDigit = (i & 1) ? prevblock[b] & 0xF : prevblock[b] >> 4;
       int offset = algoDigit;
       // insert the nth character at the front
       char oldVal = output[offset];

algo/x17/x16r-gate.h

@@ -29,9 +29,9 @@ enum x16r_Algo {
         X16R_HASH_FUNC_COUNT
 };
 
-bool (*x16_r_s_getAlgoString) ( const char*, char* );
+void (*x16_r_s_getAlgoString) ( const uint8_t*, char* );
-void x16r_getAlgoString( const char* prevblock, char *output );
+void x16r_getAlgoString( const uint8_t* prevblock, char *output );
-void x16s_getAlgoString( const char* prevblock, char *output );
+void x16s_getAlgoString( const uint8_t* prevblock, char *output );
 
 bool register_x16r_algo( algo_gate_t* gate );
 bool register_x16s_algo( algo_gate_t* gate );

algo/x17/x16r.c

@@ -205,7 +205,7 @@ int scanhash_x16r( int thr_id, struct work *work, uint32_t max_nonce,
    if ( s_ntime != pdata[17] )
    {
       uint32_t ntime = swab32(pdata[17]);
-      x16_r_s_getAlgoString( (const char*) (&endiandata[1]), hashOrder );
+      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 );

avxdefs.h

@@ -156,7 +156,7 @@ static inline __m128i foo()
 // _mm_setzero_si128 uses pxor instruction, it's unclear what _mm_set_epi does.
 // If a pseudo constant is used repeatedly in a function it may be worthwhile
 // to define a register variable to represent that constant.
-// register __m128i zero = mm_zero;
+// register __m128i zero = mm_setzero_si128();
 
 // Constant zero
 #define m128_zero      _mm_setzero_si128()
@@ -1018,7 +1018,8 @@ inline __m256i mm256_aesenc_nokey_2x128_obs( __m256i x )
 
 //////////////////////////////////////////////////////////////
 
-#if defined(__AVX512F__) && defined(__AVX512DQ__) && defined(__AVX512BW__) && defined(__AVX512VBMI__)
+#if defined(__AVX512F__)
+//#if defined(__AVX512F__) && defined(__AVX512DQ__) && defined(__AVX512BW__) && defined(__AVX512VBMI__)
 
 // Experimental, not tested.
 

build-allarch.sh

@@ -3,10 +3,30 @@
 make distclean || echo clean
 rm -f config.status
 ./autogen.sh || echo done
+CFLAGS="-O3 -march=core-avx2 -msha -Wall" ./configure --with-curl
+make -j 4
+strip -s cpuminer.exe
+mv cpuminer.exe cpuminer-avx2-sha.exe
+strip -s cpuminer
+mv cpuminer cpuminer-avx2-sha
+
+make clean || echo clean
+rm -f config.status
+./autogen.sh || echo done
+CFLAGS="-O3 -march=skylake-avx512 -Wall" ./configure --with-curl
+make -j 4
+strip -s cpuminer.exe
+mv cpuminer.exe cpuminer-avx512.exe
+strip -s cpuminer
+mv cpuminer cpuminer-avx512
+
+make clean || echo clean
+rm -f config.status
+./autogen.sh || echo done
 CFLAGS="-O3 -march=core-avx2 -Wall" ./configure --with-curl
 make -j 4
 strip -s cpuminer.exe
-mv cpuminer.exe cpuminer-aes-avx2.exe
+mv cpuminer.exe cpuminer-avx2.exe
 strip -s cpuminer
 mv cpuminer cpuminer-avx2
 

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 3.8.7.1.
+# Generated by GNU Autoconf 2.69 for cpuminer-opt 3.8.8.1.
 #
 #
 # 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='3.8.7.1'
+PACKAGE_VERSION='3.8.8.1'
-PACKAGE_STRING='cpuminer-opt 3.8.7.1'
+PACKAGE_STRING='cpuminer-opt 3.8.8.1'
 PACKAGE_BUGREPORT=''
 PACKAGE_URL=''
 
@@ -738,6 +738,7 @@ infodir
 docdir
 oldincludedir
 includedir
+runstatedir
 localstatedir
 sharedstatedir
 sysconfdir
@@ -819,6 +820,7 @@ datadir='${datarootdir}'
 sysconfdir='${prefix}/etc'
 sharedstatedir='${prefix}/com'
 localstatedir='${prefix}/var'
+runstatedir='${localstatedir}/run'
 includedir='${prefix}/include'
 oldincludedir='/usr/include'
 docdir='${datarootdir}/doc/${PACKAGE_TARNAME}'
@@ -1071,6 +1073,15 @@ do
   | -silent | --silent | --silen | --sile | --sil)
     silent=yes ;;
 
+  -runstatedir | --runstatedir | --runstatedi | --runstated \
+  | --runstate | --runstat | --runsta | --runst | --runs \
+  | --run | --ru | --r)
+    ac_prev=runstatedir ;;
+  -runstatedir=* | --runstatedir=* | --runstatedi=* | --runstated=* \
+  | --runstate=* | --runstat=* | --runsta=* | --runst=* | --runs=* \
+  | --run=* | --ru=* | --r=*)
+    runstatedir=$ac_optarg ;;
+
   -sbindir | --sbindir | --sbindi | --sbind | --sbin | --sbi | --sb)
     ac_prev=sbindir ;;
   -sbindir=* | --sbindir=* | --sbindi=* | --sbind=* | --sbin=* \
@@ -1208,7 +1219,7 @@ fi
 for ac_var in	exec_prefix prefix bindir sbindir libexecdir datarootdir \
 		datadir sysconfdir sharedstatedir localstatedir includedir \
 		oldincludedir docdir infodir htmldir dvidir pdfdir psdir \
-		libdir localedir mandir
+		libdir localedir mandir runstatedir
 do
   eval ac_val=\$$ac_var
   # Remove trailing slashes.
@@ -1321,7 +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 3.8.7.1 to adapt to many kinds of systems.
+\`configure' configures cpuminer-opt 3.8.8.1 to adapt to many kinds of systems.
 
 Usage: $0 [OPTION]... [VAR=VALUE]...
 
@@ -1361,6 +1372,7 @@ Fine tuning of the installation directories:
   --sysconfdir=DIR        read-only single-machine data [PREFIX/etc]
   --sharedstatedir=DIR    modifiable architecture-independent data [PREFIX/com]
   --localstatedir=DIR     modifiable single-machine data [PREFIX/var]
+  --runstatedir=DIR       modifiable per-process data [LOCALSTATEDIR/run]
   --libdir=DIR            object code libraries [EPREFIX/lib]
   --includedir=DIR        C header files [PREFIX/include]
   --oldincludedir=DIR     C header files for non-gcc [/usr/include]
@@ -1392,7 +1404,7 @@ fi
 
 if test -n "$ac_init_help"; then
   case $ac_init_help in
-     short | recursive ) echo "Configuration of cpuminer-opt 3.8.7.1:";;
+     short | recursive ) echo "Configuration of cpuminer-opt 3.8.8.1:";;
    esac
   cat <<\_ACEOF
 
@@ -1497,7 +1509,7 @@ fi
 test -n "$ac_init_help" && exit $ac_status
 if $ac_init_version; then
   cat <<\_ACEOF
-cpuminer-opt configure 3.8.7.1
+cpuminer-opt configure 3.8.8.1
 generated by GNU Autoconf 2.69
 
 Copyright (C) 2012 Free Software Foundation, Inc.
@@ -2000,7 +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 3.8.7.1, which was
+It was created by cpuminer-opt $as_me 3.8.8.1, which was
 generated by GNU Autoconf 2.69.  Invocation command line was
 
   $ $0 $@
@@ -2495,7 +2507,7 @@ ac_config_headers="$ac_config_headers cpuminer-config.h"
 
 
 
-am__api_version='1.14'
+am__api_version='1.15'
 
 # Find a good install program.  We prefer a C program (faster),
 # so one script is as good as another.  But avoid the broken or
@@ -2667,8 +2679,8 @@ test "$program_suffix" != NONE &&
 ac_script='s/[\\$]/&&/g;s/;s,x,x,$//'
 program_transform_name=`$as_echo "$program_transform_name" | sed "$ac_script"`
 
-# expand $ac_aux_dir to an absolute path
+# Expand $ac_aux_dir to an absolute path.
-am_aux_dir=`cd $ac_aux_dir && pwd`
+am_aux_dir=`cd "$ac_aux_dir" && pwd`
 
 if test x"${MISSING+set}" != xset; then
   case $am_aux_dir in
@@ -2687,7 +2699,7 @@ else
 $as_echo "$as_me: WARNING: 'missing' script is too old or missing" >&2;}
 fi
 
-if test x"${install_sh}" != xset; then
+if test x"${install_sh+set}" != xset; then
   case $am_aux_dir in
   *\ * | *\	*)
     install_sh="\${SHELL} '$am_aux_dir/install-sh'" ;;
@@ -2981,7 +2993,7 @@ fi
 
 # Define the identity of the package.
  PACKAGE='cpuminer-opt'
- VERSION='3.8.7.1'
+ VERSION='3.8.8.1'
 
 
 cat >>confdefs.h <<_ACEOF
@@ -3015,8 +3027,8 @@ MAKEINFO=${MAKEINFO-"${am_missing_run}makeinfo"}
 # <http://lists.gnu.org/archive/html/automake/2012-07/msg00014.html>
 mkdir_p='$(MKDIR_P)'
 
-# We need awk for the "check" target.  The system "awk" is bad on
+# We need awk for the "check" target (and possibly the TAP driver).  The
-# some platforms.
+# system "awk" is bad on some platforms.
 # Always define AMTAR for backward compatibility.  Yes, it's still used
 # in the wild :-(  We should find a proper way to deprecate it ...
 AMTAR='$${TAR-tar}'
@@ -3075,6 +3087,7 @@ END
 fi
 
 
+
 { $as_echo "$as_me:${as_lineno-$LINENO}: checking whether to enable maintainer-specific portions of Makefiles" >&5
 $as_echo_n "checking whether to enable maintainer-specific portions of Makefiles... " >&6; }
     # Check whether --enable-maintainer-mode was given.
@@ -6677,7 +6690,7 @@ cat >>$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 3.8.7.1, which was
+This file was extended by cpuminer-opt $as_me 3.8.8.1, which was
 generated by GNU Autoconf 2.69.  Invocation command line was
 
   CONFIG_FILES    = $CONFIG_FILES
@@ -6743,7 +6756,7 @@ _ACEOF
 cat >>$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 3.8.7.1
+cpuminer-opt config.status 3.8.8.1
 configured by $0, generated by GNU Autoconf 2.69,
   with options \\"\$ac_cs_config\\"
 

configure.ac

@@ -1,4 +1,4 @@
-AC_INIT([cpuminer-opt], [3.8.7.1])
+AC_INIT([cpuminer-opt], [3.8.8.1])
 
 AC_PREREQ([2.59c])
 AC_CANONICAL_SYSTEM

cpu-miner.c

@@ -98,7 +98,7 @@ static int opt_fail_pause = 10;
 static int opt_time_limit = 0;
 int opt_timeout = 300;
 static int opt_scantime = 5;
-static const bool opt_time = true;
+//static const bool opt_time = true;
 enum algos opt_algo = ALGO_NULL;
 int opt_scrypt_n = 0;
 int opt_pluck_n = 128;
@@ -1988,7 +1988,9 @@ static void *miner_thread( void *userdata )
           }
        }
        // Display benchmark total
-       if ( opt_benchmark && thr_id == opt_n_threads - 1 )
+       // Update hashrate for API if no shares accepted yet.
+       if ( ( opt_benchmark || !accepted_count ) 
+            && thr_id == opt_n_threads - 1 )
        {
           double hashrate  = 0.;
           double hashcount = 0.;
@@ -1998,13 +2000,15 @@ static void *miner_thread( void *userdata )
               hashcount += thr_hashcount[i];
           }
           if ( hashcount )
+          {
+             global_hashcount = hashcount;
+             global_hashrate  = hashrate;
+             if ( opt_benchmark )
              {
                 char hc[16];
                 char hc_units[2] = {0,0};
                 char hr[16];
                 char hr_units[2] = {0,0};
-             global_hashcount = hashcount;
-             global_hashrate  = hashrate;
 	        scale_hash_for_display( &hashcount, hc_units );
                 scale_hash_for_display( &hashrate,  hr_units );
                 if ( hc_units[0] )
@@ -2021,6 +2025,7 @@ static void *miner_thread( void *userdata )
 #endif
              }
 	  }
+       }
    }  // miner_thread loop
 
 out:
@@ -3002,11 +3007,13 @@ bool check_cpu_capability ()
      bool cpu_has_sse2   = has_sse2();
      bool cpu_has_aes    = has_aes_ni();
      bool cpu_has_sse42  = has_sse42();
+     bool cpu_has_avx    = has_avx1();
      bool cpu_has_avx2   = has_avx2();
      bool cpu_has_sha    = has_sha();
      bool cpu_has_avx512 = has_avx512f();
      bool sw_has_aes    = false;
      bool sw_has_sse42  = false;
+     bool sw_has_avx    = false;
      bool sw_has_avx2   = false;
      bool sw_has_avx512 = false;
      bool sw_has_sha    = false;
@@ -3031,6 +3038,9 @@ bool check_cpu_capability ()
      #ifdef __SSE4_2__
          sw_has_sse42 = true;
      #endif
+     #ifdef __AVX__
+         sw_has_avx = true;
+     #endif
      #ifdef __AVX2__
          sw_has_avx2 = true;
      #endif
@@ -3062,6 +3072,7 @@ bool check_cpu_capability ()
      if ( cpu_has_sse2   )    printf( " SSE2"   );
      if ( cpu_has_aes    )    printf( " AES"    );
      if ( cpu_has_sse42  )    printf( " SSE4.2" );
+     if ( cpu_has_avx    )    printf( " AVX"    );
      if ( cpu_has_avx2   )    printf( " AVX2"   );
      if ( cpu_has_avx512 )    printf( " AVX512" );
      if ( cpu_has_sha    )    printf( " SHA"    );
@@ -3069,6 +3080,7 @@ bool check_cpu_capability ()
      printf(".\nSW features: SSE2");
      if ( sw_has_aes    )     printf( " AES"    );
      if ( sw_has_sse42  )     printf( " SSE4.2" );
+     if ( sw_has_avx    )     printf( " AVX"    );
      if ( sw_has_avx2   )     printf( " AVX2"   );
      if ( sw_has_avx512 )     printf( " AVX512" );
      if ( sw_has_sha    )     printf( " SHA"    );

interleave.h

@@ -789,8 +789,8 @@ static inline void mm_reinterleave_4x32( void *dst, void *src, int  bit_len )
 
 #endif // __AVX2__
 
-//#if defined(__AVX512F__)
+#if defined(__AVX512F__)
-#if 0
+//#if 0
 
 // Macro functions returning vector.
 // Abstracted typecasting, avoid temp pointers.
@@ -804,7 +804,7 @@ static inline void mm_reinterleave_4x32( void *dst, void *src, int  bit_len )
 
 #define mm512_put_32( s00, s01, s02, s03, s04, s05, s06, s07, \
                       s08, s09, s10, s11, s12, s13, s14, s15 ) \
-  _mm512_set_epi64( *((const uint32_t*)(s15)), *((const uint32_t*)(s14)), \
+  _mm512_set_epi32( *((const uint32_t*)(s15)), *((const uint32_t*)(s14)), \
                     *((const uint32_t*)(s13)), *((const uint32_t*)(s12)), \
                     *((const uint32_t*)(s11)), *((const uint32_t*)(s10)), \
                     *((const uint32_t*)(s09)), *((const uint32_t*)(s08)), \
@@ -931,9 +931,8 @@ static inline void mm256_deinterleave_8x64x256( void *d0, void *d1, void *d2,
 }
 
 // 8 lanes of 512 bits using 64 bit interleaving (typical intermediate hash) 
-static inline void mm512_deinterleave_8x64x512( void *d00, void *d01,
+static inline void mm512_deinterleave_8x64x512( void *d0, void *d1, void *d2,
-               void *d02, void *d03, void *d04, void *d05, void *d06,
+             void *d3, void *d4, void *d5, void *d6, void *d7, const void *s )
-               void *d07, const void *s )
 {
    cast_m512i( d0 ) = mm512_get_64( s, 56, 48, 40, 32, 24, 16,  8,  0 );
    cast_m512i( d1 ) = mm512_get_64( s, 57, 49, 41, 33, 25, 17,  9,  1 );
@@ -941,7 +940,7 @@ static inline void mm512_deinterleave_8x64x512( void *d00, void *d01,
    cast_m512i( d3 ) = mm512_get_64( s, 59, 51, 43, 35, 27, 19, 11,  3 );
    cast_m512i( d4 ) = mm512_get_64( s, 60, 52, 44, 36, 28, 20, 12,  4 );
    cast_m512i( d5 ) = mm512_get_64( s, 61, 53, 45, 37, 29, 21, 13,  5 );
-   cast_m512i( d7 ) = mm512_get_64( s, 62, 54, 46, 38, 30, 22, 14,  6 );
+   cast_m512i( d6 ) = mm512_get_64( s, 62, 54, 46, 38, 30, 22, 14,  6 );
    cast_m512i( d7 ) = mm512_get_64( s, 63, 55, 47, 39, 31, 23, 15,  7 );
 }
 
@@ -953,7 +952,7 @@ static inline void mm512_deinterleave_8x64( void *dst0, void *dst1, void *dst2,
    {
       mm256_deinterleave_8x64x256( dst0, dst1, dst2, dst3,
                                    dst4, dst5, dst6, dst7, src );
-      return
+      return;
    }
 
    mm512_deinterleave_8x64x512( dst0, dst1, dst2, dst3,
@@ -976,7 +975,6 @@ static inline void mm512_deinterleave_8x64( void *dst0, void *dst1, void *dst2,
 static inline void mm512_extract_lane_8x64( void *dst, const void *src,
                                             const int lane, const int bit_len )
 {
-  const uint64_t *s = (const uint64_t*)src;
 
   if ( bit_len <= 256 )
   {
@@ -992,7 +990,7 @@ static inline void mm512_extract_lane_8x64( void *dst, const void *src,
 }
 
 
-static inline void mm512_interleave_16x32( void *dst, const void *s00,
+static inline void mm512_interleave_16x32( void *d, const void *s00,
     const void *s01, const void *s02, const void *s03, const void *s04,
     const void *s05, const void *s06, const void *s07, const void *s08,
     const void *s09, const void *s10, const void *s11, const void *s12,
@@ -1233,7 +1231,7 @@ static inline void mm512_deinterleave_16x32( void *d00, void *d01, void *d02,
       mm256_deinterleave_16x32x256( d00, d01, d02, d03, d04, d05, d06, d07,
                                     d08, d09, d10, d11, d12, d13, d14, d15,
                                     src );
-      return
+      return;
    }
    mm512_deinterleave_16x32x512( d00, d01, d02, d03, d04, d05, d06, d07,
                                  d08, d09, d10, d11, d12, d13, d14, d15, 
@@ -1303,7 +1301,7 @@ static inline void mm512_interleave_4x128( void *d, const void *s0,
 }
 
 static inline void mm512_deinterleave_4x128x128( void *d0, void *d1, void *d2,
-                                void *d3, const void *src, const int bit_len )
+                                void *d3, const void *s )
 {
    cast_m128i( d0 ) = mm_get_64( s, 1, 0 );
    cast_m128i( d1 ) = mm_get_64( s, 3, 2 );
@@ -1312,7 +1310,7 @@ static inline void mm512_deinterleave_4x128x128( void *d0, void *d1, void *d2,
 }
 
 static inline void mm512_deinterleave_4x128x256( void *d0, void *d1, void *d2,
-                                void *d3, const void *src, const int bit_len )
+                                void *d3, const void *s )
 {
    cast_m256i( d0 ) = mm256_get_64( s,  9,  8, 1, 0 );
    cast_m256i( d1 ) = mm256_get_64( s, 11, 10, 3, 2 );
@@ -1321,7 +1319,7 @@ static inline void mm512_deinterleave_4x128x256( void *d0, void *d1, void *d2,
 }
 
 static inline void mm512_deinterleave_4x128x512( void *d0, void *d1, void *d2,
-                                void *d3, const void *src, const int bit_len )
+                                void *d3, const void *s )
 {
    cast_m512i( d0 ) = mm512_get_64( s, 25, 24, 17, 16,  9,  8, 1, 0 );
    cast_m512i( d1 ) = mm512_get_64( s, 27, 26, 19, 18, 11, 10, 3, 2 );
@@ -1334,8 +1332,8 @@ static inline void mm512_deinterleave_4x128( void *dst0, void *dst1, void *dst2,
 {
    if ( bit_len <= 256 )
    {
-      mm256_deinterleave_4x128x256( dst0, dst1, dst2, dst3, src );
+      mm512_deinterleave_4x128x256( dst0, dst1, dst2, dst3, src );
-      return
+      return;
    }
 
    mm512_deinterleave_4x128x512( dst0, dst1, dst2, dst3, src );
@@ -1343,7 +1341,7 @@ static inline void mm512_deinterleave_4x128( void *dst0, void *dst1, void *dst2,
 
    if ( bit_len <= 640 )
    {
-      mm_deinterleave_4x128x128( dst0+128, dst1+128, dst2+128, dst3+128,
+      mm512_deinterleave_4x128x128( dst0+128, dst1+128, dst2+128, dst3+128,
                                  src+512 );
       return;
    }

miner.h

@@ -499,6 +499,7 @@ enum algos {
         ALGO_C11,         
         ALGO_CRYPTOLIGHT, 
         ALGO_CRYPTONIGHT,
+        ALGO_CRYPTONIGHTV7, 
         ALGO_DECRED,
         ALGO_DEEP,
         ALGO_DMD_GR,
@@ -580,6 +581,7 @@ static const char* const algo_names[] = {
         "c11",
         "cryptolight",
         "cryptonight",
+        "cryptonightv7",
         "decred",
         "deep",
         "dmd-gr",
@@ -718,7 +720,8 @@ Options:\n\
                           bmw           BMW 256\n\
                           c11           Chaincoin\n\
                           cryptolight   Cryptonight-light\n\
-                          cryptonight  cryptonote, Monero (XMR)\n\
+                          cryptonight   Cryptonote legacy\n\
+                          cryptonightv7 variant 7, Monero (XMR)\n\
                           decred        Blake256r14dcr\n\
                           deep          Deepcoin (DCN)\n\
                           dmd-gr        Diamond\n\

winbuild-cross.sh

@@ -46,16 +46,18 @@ mv cpuminer.exe release/cpuminer-avx2.exe
 #mv cpuminer.exe release/cpuminer-aes-sha.exe
 
 
-#make clean || echo clean
-#rm -f config.status
-#CFLAGS="-O3 -march=corei7-avx -Wall" ./configure $F 
-#make 
-#strip -s cpuminer.exe
-#mv cpuminer.exe release/cpuminer-aes-avx.exe
-
 make clean || echo clean
 rm -f config.status
-CFLAGS="-O3 -maes -msse4.2 -Wall" ./configure $F
+CFLAGS="-O3 -march=corei7-avx -Wall" ./configure $F 
+make 
+strip -s cpuminer.exe
+mv cpuminer.exe release/cpuminer-avx.exe
+
+# -march=westmere is supported in gcc5
+make clean || echo clean
+rm -f config.status
+CFLAGS="-O3 -march=westmere -Wall" ./configure $F
+#CFLAGS="-O3 -maes -msse4.2 -Wall" ./configure $F
 make 
 strip -s cpuminer.exe
 mv cpuminer.exe release/cpuminer-aes-sse42.exe