v2.4.5 RC

Update README.md

CHANGELOG.md

@@ -1,3 +1,14 @@
+# v2.4.5
+- [#324](https://github.com/xmrig/xmrig/pull/324) Fixed build without libmicrohttpd (CMake cache issue).
+- [#341](https://github.com/xmrig/xmrig/issues/341) Fixed wrong exit code and added command line option `--dry-run`.
+- [#385](https://github.com/xmrig/xmrig/pull/385) Up to 20% performance increase with for non-AES CPU and fixed Intel Core 2 cache detection.
+
+# v2.4.4
+ - Added libmicrohttpd version to --version output.
+ - Fixed bug in singal handler, in some cases miner wasn't shutdown properly.
+ - Fixed recent MSVC 2017 version detection.
+ - [#279](https://github.com/xmrig/xmrig/pull/279) Fixed build on some macOS versions.
+
 # v2.4.3
  - [#94](https://github.com/xmrig/xmrig/issues/94#issuecomment-342019257) [#216](https://github.com/xmrig/xmrig/issues/216) Added **ARMv8** and **ARMv7** support. Hardware AES supported, thanks [Imran Yusuff](https://github.com/imranyusuff).
  - [#157](https://github.com/xmrig/xmrig/issues/157) [#196](https://github.com/xmrig/xmrig/issues/196) Fixed Linux compile issues.

CMakeLists.txt

@@ -1,4 +1,4 @@
-cmake_minimum_required(VERSION 3.0)
+cmake_minimum_required(VERSION 2.8)
 project(xmrig)
 
 option(WITH_LIBCPUID "Use Libcpuid" ON)
@@ -195,6 +195,7 @@ if (WITH_HTTPD)
         message(FATAL_ERROR "microhttpd NOT found: use `-DWITH_HTTPD=OFF` to build without http deamon support")
     endif()
 else()
+    set(MHD_LIBRARY "")
     add_definitions(/DXMRIG_NO_HTTPD)
     add_definitions(/DXMRIG_NO_API)
 endif()

README.md

@@ -1,6 +1,13 @@
 # XMRig
-XMRig is high performance Monero (XMR) CPU miner, with the official full Windows support.
-Originally based on cpuminer-multi with heavy optimizations/rewrites and removing a lot of legacy code, since version 1.0.0 complete rewritten from scratch on C++.
+[![Github All Releases](https://img.shields.io/github/downloads/xmrig/xmrig/total.svg)](https://github.com/xmrig/xmrig/releases)
+[![GitHub release](https://img.shields.io/github/release/xmrig/xmrig/all.svg)](https://github.com/xmrig/xmrig/releases)
+[![GitHub Release Date](https://img.shields.io/github/release-date-pre/xmrig/xmrig.svg)](https://github.com/xmrig/xmrig/releases)
+[![GitHub license](https://img.shields.io/github/license/xmrig/xmrig.svg)](https://github.com/xmrig/xmrig/blob/master/LICENSE)
+[![GitHub stars](https://img.shields.io/github/stars/xmrig/xmrig.svg)](https://github.com/xmrig/xmrig/stargazers)
+[![GitHub forks](https://img.shields.io/github/forks/xmrig/xmrig.svg)](https://github.com/xmrig/xmrig/network)
+
+XMRig is a high performance Monero (XMR) CPU miner, with official support for Windows.
+Originally based on cpuminer-multi with heavy optimizations/rewrites and removing a lot of legacy code, since version 1.0.0 completely rewritten from scratch on C++.
 
 * This is the **CPU-mining** version, there is also a [NVIDIA GPU version](https://github.com/xmrig/xmrig-nvidia) and [AMD GPU version]( https://github.com/xmrig/xmrig-amd).
 * [Roadmap](https://github.com/xmrig/xmrig/issues/106) for next releases.
@@ -37,44 +44,38 @@ Originally based on cpuminer-multi with heavy optimizations/rewrites and removin
   * Clone with `git clone https://github.com/xmrig/xmrig.git` :hammer: [Build instructions](https://github.com/xmrig/xmrig/wiki/Build).
 
 ## Usage
-### Basic example
-```
-xmrig.exe -o pool.monero.hashvault.pro:5555 -u YOUR_WALLET -p x -k
-```
-
-### Failover
-```
-xmrig.exe -o pool.monero.hashvault.pro:5555 -u YOUR_WALLET1 -p x -k -o pool.supportxmr.com:5555 -u YOUR_WALLET2 -p x -k
-```
-For failover you can add multiple pools, maximum count not limited.
+Use [config.xmrig.com](https://config.xmrig.com/xmrig) to generate, edit or share configurations.
 
 ### Options
 ```
-  -a, --algo=ALGO       cryptonight (default) or cryptonight-lite
-  -o, --url=URL         URL of mining server
-  -O, --userpass=U:P    username:password pair for mining server
-  -u, --user=USERNAME   username for mining server
-  -p, --pass=PASSWORD   password for mining server
-  -t, --threads=N       number of miner threads
-  -v, --av=N            algorithm variation, 0 auto select
-  -k, --keepalive       send keepalived for prevent timeout (need pool support)
-  -r, --retries=N       number of times to retry before switch to backup server (default: 5)
-  -R, --retry-pause=N   time to pause between retries (default: 5)
-      --cpu-affinity    set process affinity to CPU core(s), mask 0x3 for cores 0 and 1
-      --cpu-priority    set process priority (0 idle, 2 normal to 5 highest)
-      --no-huge-pages   disable huge pages support
-      --no-color        disable colored output
-      --donate-level=N  donate level, default 5% (5 minutes in 100 minutes)
-      --user-agent      set custom user-agent string for pool
-  -B, --background      run the miner in the background
-  -c, --config=FILE     load a JSON-format configuration file
-  -l, --log-file=FILE   log all output to a file
-      --max-cpu-usage=N maximum CPU usage for automatic threads mode (default 75)
-      --safe            safe adjust threads and av settings for current CPU
-      --nicehash        enable nicehash support
-      --print-time=N    print hashrate report every N seconds
-  -h, --help            display this help and exit
-  -V, --version         output version information and exit
+  -a, --algo=ALGO          cryptonight (default) or cryptonight-lite
+  -o, --url=URL            URL of mining server
+  -O, --userpass=U:P       username:password pair for mining server
+  -u, --user=USERNAME      username for mining server
+  -p, --pass=PASSWORD      password for mining server
+  -t, --threads=N          number of miner threads
+  -v, --av=N               algorithm variation, 0 auto select
+  -k, --keepalive          send keepalived for prevent timeout (need pool support)
+  -r, --retries=N          number of times to retry before switch to backup server (default: 5)
+  -R, --retry-pause=N      time to pause between retries (default: 5)
+      --cpu-affinity       set process affinity to CPU core(s), mask 0x3 for cores 0 and 1
+      --cpu-priority       set process priority (0 idle, 2 normal to 5 highest)
+      --no-huge-pages      disable huge pages support
+      --no-color           disable colored output
+      --donate-level=N     donate level, default 5% (5 minutes in 100 minutes)
+      --user-agent         set custom user-agent string for pool
+  -B, --background         run the miner in the background
+  -c, --config=FILE        load a JSON-format configuration file
+  -l, --log-file=FILE      log all output to a file
+      --max-cpu-usage=N    maximum CPU usage for automatic threads mode (default 75)
+      --safe               safe adjust threads and av settings for current CPU
+      --nicehash           enable nicehash/xmrig-proxy support
+      --print-time=N       print hashrate report every N seconds
+      --api-port=N         port for the miner API
+      --api-access-token=T access token for API
+      --api-worker-id=ID   custom worker-id for API
+  -h, --help               display this help and exit
+  -V, --version            output version information and exit
 ```
 
 Also you can use configuration via config file, default **config.json**. You can load multiple config files and combine it with command line options.
@@ -89,7 +90,7 @@ Since version 0.8.0.
 ## Common Issues
 ### HUGE PAGES unavailable
 * Run XMRig as Administrator.
-* Since version 0.8.0 XMRig automatically enable SeLockMemoryPrivilege for current user, but reboot or sign out still required. [Manual instruction](https://msdn.microsoft.com/en-gb/library/ms190730.aspx).
+* Since version 0.8.0 XMRig automatically enables SeLockMemoryPrivilege for current user, but reboot or sign out still required. [Manual instruction](https://msdn.microsoft.com/en-gb/library/ms190730.aspx).
 
 ## Other information
 * No HTTP support, only stratum protocol support.
@@ -106,7 +107,7 @@ Please note performance is highly dependent on system load. The numbers above ar
 ### Maximum performance checklist
 * Idle operating system.
 * Do not exceed optimal thread count.
-* Use modern CPUs with AES-NI instructuon set.
+* Use modern CPUs with AES-NI instruction set.
 * Try setup optimal cpu affinity.
 * Enable fast memory (Large/Huge pages).
 

src/App.cpp

@@ -92,7 +92,9 @@ App::App(int argc, char **argv) :
 
     m_network = new Network(m_options);
 
-    uv_signal_init(uv_default_loop(), &m_signal);
+    uv_signal_init(uv_default_loop(), &m_sigHUP);
+    uv_signal_init(uv_default_loop(), &m_sigINT);
+    uv_signal_init(uv_default_loop(), &m_sigTERM);
 }
 
 
@@ -111,12 +113,12 @@ App::~App()
 int App::exec()
 {
     if (!m_options) {
-        return 0;
+        return 2;
     }
 
-    uv_signal_start(&m_signal, App::onSignal, SIGHUP);
-    uv_signal_start(&m_signal, App::onSignal, SIGTERM);
-    uv_signal_start(&m_signal, App::onSignal, SIGINT);
+    uv_signal_start(&m_sigHUP,  App::onSignal, SIGHUP);
+    uv_signal_start(&m_sigINT,  App::onSignal, SIGINT);
+    uv_signal_start(&m_sigTERM, App::onSignal, SIGTERM);
 
     background();
 
@@ -128,6 +130,13 @@ int App::exec()
     Mem::allocate(m_options->algo(), m_options->threads(), m_options->doubleHash(), m_options->hugePages());
     Summary::print();
 
+    if (m_options->dryRun()) {
+        LOG_NOTICE("OK");
+        release();
+
+        return 0;
+    }
+
 #   ifndef XMRIG_NO_API
     Api::start();
 #   endif
@@ -144,12 +153,7 @@ int App::exec()
     const int r = uv_run(uv_default_loop(), UV_RUN_DEFAULT);
     uv_loop_close(uv_default_loop());
 
-    delete m_network;
-
-    Options::release();
-    Mem::release();
-    Platform::release();
-
+    release();
     return r;
 }
 
@@ -164,8 +168,10 @@ void App::onConsoleCommand(char command)
 
     case 'p':
     case 'P':
-        LOG_INFO(m_options->colors() ? "\x1B[01;33mpaused\x1B[0m, press \x1B[01;35mr\x1B[0m to resume" : "paused, press 'r' to resume");
-        Workers::setEnabled(false);
+        if (Workers::isEnabled()) {
+            LOG_INFO(m_options->colors() ? "\x1B[01;33mpaused\x1B[0m, press \x1B[01;35mr\x1B[0m to resume" : "paused, press 'r' to resume");
+            Workers::setEnabled(false);
+        }
         break;
 
     case 'r':
@@ -196,6 +202,18 @@ void App::close()
 }
 
 
+void App::release()
+{
+    if (m_network) {
+        delete m_network;
+    }
+
+    Options::release();
+    Mem::release();
+    Platform::release();
+}
+
+
 void App::onSignal(uv_signal_t *handle, int signum)
 {
     switch (signum)

src/App.h

@@ -51,6 +51,7 @@ protected:
 private:
   void background();
   void close();
+  void release();
 
   static void onSignal(uv_signal_t *handle, int signum);
 
@@ -60,7 +61,9 @@ private:
   Httpd *m_httpd;
   Network *m_network;
   Options *m_options;
-  uv_signal_t m_signal;
+  uv_signal_t m_sigHUP;
+  uv_signal_t m_sigINT;
+  uv_signal_t m_sigTERM;
 };
 
 

src/Cpu.cpp

@@ -100,7 +100,13 @@ void Cpu::initCommon()
         m_l2_cache = data.l2_cache * (m_totalCores / 2) * m_sockets;
         m_l2_exclusive = true;
     }
-    else {
+    // Workaround for Intel Core Solo, Core Duo, Core 2 Duo, Core 2 Quad and their Xeon homologue
+    // These processors have L2 cache shared by 2 cores.
+    else if (data.vendor == VENDOR_INTEL && data.family == 0x06 && (data.model == 0x0E || data.model == 0x0F || data.model == 0x07)) {
+        int l2_count_per_socket = m_totalCores > 1 ? m_totalCores / 2 : 1;
+        m_l2_cache = data.l2_cache > 0 ? data.l2_cache * l2_count_per_socket * m_sockets : 0;
+    }
+    else{
         m_l2_cache = data.l2_cache > 0 ? data.l2_cache * m_totalCores * m_sockets : 0;
     }
 

src/Options.cpp

@@ -33,6 +33,11 @@
 #endif
 
 
+#ifndef XMRIG_NO_HTTPD
+#   include <microhttpd.h>
+#endif
+
+
 #include "Cpu.h"
 #include "donate.h"
 #include "net/Url.h"
@@ -96,12 +101,16 @@ static char const short_options[] = "a:c:khBp:Px:r:R:s:t:T:o:u:O:v:Vl:S";
 
 static struct option const options[] = {
     { "algo",             1, nullptr, 'a'  },
+    { "api-access-token", 1, nullptr, 4001 },
+    { "api-port",         1, nullptr, 4000 },
+    { "api-worker-id",    1, nullptr, 4002 },
     { "av",               1, nullptr, 'v'  },
     { "background",       0, nullptr, 'B'  },
     { "config",           1, nullptr, 'c'  },
     { "cpu-affinity",     1, nullptr, 1020 },
     { "cpu-priority",     1, nullptr, 1021 },
     { "donate-level",     1, nullptr, 1003 },
+    { "dry-run",          0, nullptr, 5000 },
     { "help",             0, nullptr, 'h'  },
     { "keepalive",        0, nullptr ,'k'  },
     { "log-file",         1, nullptr, 'l'  },
@@ -121,9 +130,6 @@ static struct option const options[] = {
     { "user-agent",       1, nullptr, 1008 },
     { "userpass",         1, nullptr, 'O'  },
     { "version",          0, nullptr, 'V'  },
-    { "api-port",         1, nullptr, 4000 },
-    { "api-access-token", 1, nullptr, 4001 },
-    { "api-worker-id",    1, nullptr, 4002 },
     { 0, 0, 0, 0 }
 };
 
@@ -136,6 +142,7 @@ static struct option const config_options[] = {
     { "cpu-affinity",  1, nullptr, 1020 },
     { "cpu-priority",  1, nullptr, 1021 },
     { "donate-level",  1, nullptr, 1003 },
+    { "dry-run",       0, nullptr, 5000 },
     { "huge-pages",    0, nullptr, 1009 },
     { "log-file",      1, nullptr, 'l'  },
     { "max-cpu-usage", 1, nullptr, 1004 },
@@ -200,6 +207,7 @@ Options::Options(int argc, char **argv) :
     m_background(false),
     m_colors(true),
     m_doubleHash(false),
+    m_dryRun(false),
     m_hugePages(true),
     m_ready(false),
     m_safe(false),
@@ -379,6 +387,7 @@ bool Options::parseArg(int key, const char *arg)
     case 'S':  /* --syslog */
     case 1005: /* --safe */
     case 1006: /* --nicehash */
+    case 5000: /* --dry-run */
         return parseBoolean(key, true);
 
     case 1002: /* --no-color */
@@ -552,6 +561,10 @@ bool Options::parseBoolean(int key, bool enable)
         m_colors = enable;
         break;
 
+    case 5000: /* --dry-run */
+        m_dryRun = enable;
+        break;
+
     default:
         break;
     }
@@ -665,6 +678,10 @@ void Options::showVersion()
     "\n");
 
     printf("\nlibuv/%s\n", uv_version_string());
+
+#   ifndef XMRIG_NO_HTTPD
+    printf("libmicrohttpd/%s\n", MHD_get_version());
+#   endif
 }
 
 

src/Options.h

@@ -59,6 +59,7 @@ public:
     inline bool background() const                { return m_background; }
     inline bool colors() const                    { return m_colors; }
     inline bool doubleHash() const                { return m_doubleHash; }
+    inline bool dryRun() const                    { return m_dryRun; }
     inline bool hugePages() const                 { return m_hugePages; }
     inline bool syslog() const                    { return m_syslog; }
     inline const char *apiToken() const           { return m_apiToken; }
@@ -110,6 +111,7 @@ private:
     bool m_background;
     bool m_colors;
     bool m_doubleHash;
+    bool m_dryRun;
     bool m_hugePages;
     bool m_ready;
     bool m_safe;

src/Platform_mac.cpp

@@ -22,6 +22,7 @@
  */
 
 
+#include <stdio.h>
 #include <stdlib.h>
 #include <sys/resource.h>
 #include <uv.h>
@@ -41,11 +42,11 @@ static inline char *createUserAgent()
 
     char *buf = new char[max];
 
-#   ifdef XMRIG_NVIDIA_PROJECT
-    const int cudaVersion = cuda_get_runtime_version();
-    snprintf(buf, max, "%s/%s (Macintosh; Intel Mac OS X) libuv/%s CUDA/%d.%d clang/%d.%d.%d", APP_NAME, APP_VERSION, uv_version_string(), cudaVersion / 1000, cudaVersion % 100, __clang_major__, __clang_minor__, __clang_patchlevel__);
-#   else
-    snprintf(buf, max, "%s/%s (Macintosh; Intel Mac OS X) libuv/%s clang/%d.%d.%d", APP_NAME, APP_VERSION, uv_version_string(), __clang_major__, __clang_minor__, __clang_patchlevel__);
+#   ifdef XMRIG_NVIDIA_PROJECT
+    const int cudaVersion = cuda_get_runtime_version();
+    snprintf(buf, max, "%s/%s (Macintosh; Intel Mac OS X) libuv/%s CUDA/%d.%d clang/%d.%d.%d", APP_NAME, APP_VERSION, uv_version_string(), cudaVersion / 1000, cudaVersion % 100, __clang_major__, __clang_minor__, __clang_patchlevel__);
+#   else
+    snprintf(buf, max, "%s/%s (Macintosh; Intel Mac OS X) libuv/%s clang/%d.%d.%d", APP_NAME, APP_VERSION, uv_version_string(), __clang_major__, __clang_minor__, __clang_patchlevel__);
 #   endif
 
     return buf;

src/api/NetworkState.cpp

@@ -46,7 +46,7 @@ NetworkState::NetworkState() :
 
 int NetworkState::connectionTime() const
 {
-    return m_active ? ((uv_now(uv_default_loop()) - m_connectionTime) / 1000) : 0;
+    return m_active ? (int)((uv_now(uv_default_loop()) - m_connectionTime) / 1000) : 0;
 }
 
 
@@ -56,7 +56,7 @@ uint32_t NetworkState::avgTime() const
         return 0;
     }
 
-    return (uint32_t) connectionTime() / m_latency.size();
+    return connectionTime() / (uint32_t)m_latency.size();
 }
 
 

src/crypto/CryptoNight_arm.h

@@ -194,14 +194,14 @@ template<bool SOFT_AES>
 static inline void aes_round(__m128i key, __m128i* x0, __m128i* x1, __m128i* x2, __m128i* x3, __m128i* x4, __m128i* x5, __m128i* x6, __m128i* x7)
 {
     if (SOFT_AES) {
-        *x0 = soft_aesenc(*x0, key);
-        *x1 = soft_aesenc(*x1, key);
-        *x2 = soft_aesenc(*x2, key);
-        *x3 = soft_aesenc(*x3, key);
-        *x4 = soft_aesenc(*x4, key);
-        *x5 = soft_aesenc(*x5, key);
-        *x6 = soft_aesenc(*x6, key);
-        *x7 = soft_aesenc(*x7, key);
+        *x0 = soft_aesenc((uint32_t*)x0, key);
+        *x1 = soft_aesenc((uint32_t*)x1, key);
+        *x2 = soft_aesenc((uint32_t*)x2, key);
+        *x3 = soft_aesenc((uint32_t*)x3, key);
+        *x4 = soft_aesenc((uint32_t*)x4, key);
+        *x5 = soft_aesenc((uint32_t*)x5, key);
+        *x6 = soft_aesenc((uint32_t*)x6, key);
+        *x7 = soft_aesenc((uint32_t*)x7, key);
     }
 #   ifndef XMRIG_ARMv7
     else {
@@ -361,12 +361,13 @@ inline void cryptonight_hash(const void *__restrict__ input, size_t size, void *
     uint64_t idx0 = h0[0] ^ h0[4];
 
     for (size_t i = 0; i < ITERATIONS; i++) {
-        __m128i cx = _mm_load_si128((__m128i *) &l0[idx0 & MASK]);
+        __m128i cx;
 
         if (SOFT_AES) {
-            cx = soft_aesenc(cx, _mm_set_epi64x(ah0, al0));
+            cx = soft_aesenc((uint32_t*)&l0[idx0 & MASK], _mm_set_epi64x(ah0, al0));
         }
         else {
+            cx = _mm_load_si128((__m128i *) &l0[idx0 & MASK]);
 #           ifndef XMRIG_ARMv7
             cx = vreinterpretq_m128i_u8(vaesmcq_u8(vaeseq_u8(cx, vdupq_n_u8(0)))) ^ _mm_set_epi64x(ah0, al0);
 #           endif
@@ -425,14 +426,15 @@ inline void cryptonight_double_hash(const void *__restrict__ input, size_t size,
     uint64_t idx1 = h1[0] ^ h1[4];
 
     for (size_t i = 0; i < ITERATIONS; i++) {
-        __m128i cx0 = _mm_load_si128((__m128i *) &l0[idx0 & MASK]);
-        __m128i cx1 = _mm_load_si128((__m128i *) &l1[idx1 & MASK]);
+        __m128i cx0, cx1;
 
         if (SOFT_AES) {
-            cx0 = soft_aesenc(cx0, _mm_set_epi64x(ah0, al0));
-            cx1 = soft_aesenc(cx1, _mm_set_epi64x(ah1, al1));
+            cx0 = soft_aesenc((uint32_t*)&l0[idx0 & MASK], _mm_set_epi64x(ah0, al0));
+            cx1 = soft_aesenc((uint32_t*)&l1[idx1 & MASK], _mm_set_epi64x(ah1, al1));
         }
         else {
+            cx0 = _mm_load_si128((__m128i *) &l0[idx0 & MASK]);
+            cx1 = _mm_load_si128((__m128i *) &l1[idx1 & MASK]);
 #           ifndef XMRIG_ARMv7
             cx0 = vreinterpretq_m128i_u8(vaesmcq_u8(vaeseq_u8(cx0, vdupq_n_u8(0)))) ^ _mm_set_epi64x(ah0, al0);
             cx1 = vreinterpretq_m128i_u8(vaesmcq_u8(vaeseq_u8(cx1, vdupq_n_u8(0)))) ^ _mm_set_epi64x(ah1, al1);

src/crypto/CryptoNight_x86.h

@@ -193,14 +193,14 @@ template<bool SOFT_AES>
 static inline void aes_round(__m128i key, __m128i* x0, __m128i* x1, __m128i* x2, __m128i* x3, __m128i* x4, __m128i* x5, __m128i* x6, __m128i* x7)
 {
     if (SOFT_AES) {
-        *x0 = soft_aesenc(*x0, key);
-        *x1 = soft_aesenc(*x1, key);
-        *x2 = soft_aesenc(*x2, key);
-        *x3 = soft_aesenc(*x3, key);
-        *x4 = soft_aesenc(*x4, key);
-        *x5 = soft_aesenc(*x5, key);
-        *x6 = soft_aesenc(*x6, key);
-        *x7 = soft_aesenc(*x7, key);
+        *x0 = soft_aesenc((uint32_t*)x0, key);
+        *x1 = soft_aesenc((uint32_t*)x1, key);
+        *x2 = soft_aesenc((uint32_t*)x2, key);
+        *x3 = soft_aesenc((uint32_t*)x3, key);
+        *x4 = soft_aesenc((uint32_t*)x4, key);
+        *x5 = soft_aesenc((uint32_t*)x5, key);
+        *x6 = soft_aesenc((uint32_t*)x6, key);
+        *x7 = soft_aesenc((uint32_t*)x7, key);
     }
     else {
         *x0 = _mm_aesenc_si128(*x0, key);
@@ -325,15 +325,14 @@ inline void cryptonight_hash(const void *__restrict__ input, size_t size, void *
 
     for (size_t i = 0; i < ITERATIONS; i++) {
         __m128i cx;
-        cx = _mm_load_si128((__m128i *) &l0[idx0 & MASK]);
 
         if (SOFT_AES) {
-            cx = soft_aesenc(cx, _mm_set_epi64x(ah0, al0));
+            cx = soft_aesenc((uint32_t*)&l0[idx0 & MASK], _mm_set_epi64x(ah0, al0));
         }
-        else {
+        else {  
+            cx = _mm_load_si128((__m128i *) &l0[idx0 & MASK]);
             cx = _mm_aesenc_si128(cx, _mm_set_epi64x(ah0, al0));
         }
-
         _mm_store_si128((__m128i *) &l0[idx0 & MASK], _mm_xor_si128(bx0, cx));
         idx0 = EXTRACT64(cx);
         bx0 = cx;
@@ -387,14 +386,15 @@ inline void cryptonight_double_hash(const void *__restrict__ input, size_t size,
     uint64_t idx1 = h1[0] ^ h1[4];
 
     for (size_t i = 0; i < ITERATIONS; i++) {
-        __m128i cx0 = _mm_load_si128((__m128i *) &l0[idx0 & MASK]);
-        __m128i cx1 = _mm_load_si128((__m128i *) &l1[idx1 & MASK]);
+        __m128i cx0, cx1;
 
         if (SOFT_AES) {
-            cx0 = soft_aesenc(cx0, _mm_set_epi64x(ah0, al0));
-            cx1 = soft_aesenc(cx1, _mm_set_epi64x(ah1, al1));
+            cx0 = soft_aesenc((uint32_t*)&l0[idx0 & MASK], _mm_set_epi64x(ah0, al0));
+            cx1 = soft_aesenc((uint32_t*)&l1[idx1 & MASK], _mm_set_epi64x(ah1, al1));
         }
         else {
+            cx0 = _mm_load_si128((__m128i *) &l0[idx0 & MASK]);
+            cx1 = _mm_load_si128((__m128i *) &l1[idx1 & MASK]);
             cx0 = _mm_aesenc_si128(cx0, _mm_set_epi64x(ah0, al0));
             cx1 = _mm_aesenc_si128(cx1, _mm_set_epi64x(ah1, al1));
         }

src/crypto/soft_aes.h

@@ -89,34 +89,34 @@
 alignas(16) const uint32_t saes_table[4][256] = { saes_data(saes_u0), saes_data(saes_u1), saes_data(saes_u2), saes_data(saes_u3) };
 alignas(16) const uint8_t  saes_sbox[256] = saes_data(saes_h0);
 
-static inline __m128i soft_aesenc(__m128i in, __m128i key)
+static inline __m128i soft_aesenc(const uint32_t* in, __m128i key)
 {
-    const uint32_t x0 = _mm_cvtsi128_si32(in);
-    const uint32_t x1 = _mm_cvtsi128_si32(_mm_shuffle_epi32(in, 0x55));
-    const uint32_t x2 = _mm_cvtsi128_si32(_mm_shuffle_epi32(in, 0xAA));
-    const uint32_t x3 = _mm_cvtsi128_si32(_mm_shuffle_epi32(in, 0xFF));
+    const uint32_t x0 = in[0];
+    const uint32_t x1 = in[1];
+    const uint32_t x2 = in[2];
+    const uint32_t x3 = in[3];
 
-	__m128i out = _mm_set_epi32(
-		(saes_table[0][x3 & 0xff] ^ saes_table[1][(x0 >> 8) & 0xff] ^ saes_table[2][(x1 >> 16) & 0xff] ^ saes_table[3][x2 >> 24]),
-		(saes_table[0][x2 & 0xff] ^ saes_table[1][(x3 >> 8) & 0xff] ^ saes_table[2][(x0 >> 16) & 0xff] ^ saes_table[3][x1 >> 24]),
-		(saes_table[0][x1 & 0xff] ^ saes_table[1][(x2 >> 8) & 0xff] ^ saes_table[2][(x3 >> 16) & 0xff] ^ saes_table[3][x0 >> 24]),
-		(saes_table[0][x0 & 0xff] ^ saes_table[1][(x1 >> 8) & 0xff] ^ saes_table[2][(x2 >> 16) & 0xff] ^ saes_table[3][x3 >> 24]));
+    __m128i out = _mm_set_epi32(
+        (saes_table[0][x3 & 0xff] ^ saes_table[1][(x0 >> 8) & 0xff] ^ saes_table[2][(x1 >> 16) & 0xff] ^ saes_table[3][x2 >> 24]),
+        (saes_table[0][x2 & 0xff] ^ saes_table[1][(x3 >> 8) & 0xff] ^ saes_table[2][(x0 >> 16) & 0xff] ^ saes_table[3][x1 >> 24]),
+        (saes_table[0][x1 & 0xff] ^ saes_table[1][(x2 >> 8) & 0xff] ^ saes_table[2][(x3 >> 16) & 0xff] ^ saes_table[3][x0 >> 24]),
+        (saes_table[0][x0 & 0xff] ^ saes_table[1][(x1 >> 8) & 0xff] ^ saes_table[2][(x2 >> 16) & 0xff] ^ saes_table[3][x3 >> 24]));
 
-	return _mm_xor_si128(out, key);
+    return _mm_xor_si128(out, key);
 }
 
 static inline uint32_t sub_word(uint32_t key)
 {
-	return (saes_sbox[key >> 24 ] << 24)   | 
-		(saes_sbox[(key >> 16) & 0xff] << 16 ) | 
-		(saes_sbox[(key >> 8)  & 0xff] << 8  ) | 
-		 saes_sbox[key & 0xff];
+    return (saes_sbox[key >> 24 ] << 24)   | 
+        (saes_sbox[(key >> 16) & 0xff] << 16 ) | 
+        (saes_sbox[(key >> 8)  & 0xff] << 8  ) | 
+         saes_sbox[key & 0xff];
 }
 
 #if defined(__clang__) || defined(XMRIG_ARM)
 static inline uint32_t _rotr(uint32_t value, uint32_t amount)
 {
-	return (value >> amount) | (value << ((32 - amount) & 31));
+    return (value >> amount) | (value << ((32 - amount) & 31));
 }
 #endif
 

src/net/Client.cpp

@@ -532,7 +532,7 @@ void Client::onAllocBuffer(uv_handle_t *handle, size_t suggested_size, uv_buf_t
     auto client = getClient(handle->data);
 
     buf->base = &client->m_recvBuf.base[client->m_recvBufPos];
-    buf->len  = client->m_recvBuf.len - client->m_recvBufPos;
+    buf->len  = client->m_recvBuf.len - (unsigned long)client->m_recvBufPos;
 }
 
 
@@ -582,11 +582,11 @@ void Client::onRead(uv_stream_t *stream, ssize_t nread, const uv_buf_t *buf)
             LOG_ERR("[%s:%u] read error: \"%s\"", client->m_url.host(), client->m_url.port(), uv_strerror((int) nread));
         }
 
-        return client->close();;
+        return client->close();
     }
 
     if ((size_t) nread > (sizeof(m_buf) - 8 - client->m_recvBufPos)) {
-        return client->close();;
+        return client->close();
     }
 
     client->m_recvBufPos += nread;
@@ -623,7 +623,7 @@ void Client::onResolved(uv_getaddrinfo_t *req, int status, struct addrinfo *res)
     auto client = getClient(req->data);
     if (status < 0) {
         LOG_ERR("[%s:%u] DNS error: \"%s\"", client->m_url.host(), client->m_url.port(), uv_strerror(status));
-        return client->reconnect();;
+        return client->reconnect();
     }
 
     addrinfo *ptr = res;

src/version.h

@@ -4,7 +4,7 @@
  * Copyright 2014      Lucas Jones <https://github.com/lucasjones>
  * Copyright 2014-2016 Wolf9466    <https://github.com/OhGodAPet>
  * Copyright 2016      Jay D Dee   <jayddee246@gmail.com>
- * Copyright 2016-2017 XMRig       <support@xmrig.com>
+ * Copyright 2016-2018 XMRig       <support@xmrig.com>
  *
  *
  *   This program is free software: you can redistribute it and/or modify
@@ -27,19 +27,19 @@
 #define APP_ID        "xmrig"
 #define APP_NAME      "XMRig"
 #define APP_DESC      "XMRig CPU miner"
-#define APP_VERSION   "2.4.3"
+#define APP_VERSION   "2.4.5"
 #define APP_DOMAIN    "xmrig.com"
 #define APP_SITE      "www.xmrig.com"
-#define APP_COPYRIGHT "Copyright (C) 2016-2017 xmrig.com"
+#define APP_COPYRIGHT "Copyright (C) 2016-2018 xmrig.com"
 #define APP_KIND      "cpu"
 
 #define APP_VER_MAJOR  2
 #define APP_VER_MINOR  4
-#define APP_VER_BUILD  3
+#define APP_VER_BUILD  5
 #define APP_VER_REV    0
 
 #ifdef _MSC_VER
-#   if (_MSC_VER == 1910 || _MSC_VER == 1911)
+#   if (_MSC_VER >= 1910)
 #       define MSVC_VERSION 2017
 #   elif _MSC_VER == 1900
 #       define MSVC_VERSION 2015