Ecosystem/xmrig
Ecosystem/xmrig
1
0
Fork 0
mirror of https://github.com/xmrig/xmrig.git synced 2024-11-17 16:27:44 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Merge branch 'master' into dev

Browse source
This commit is contained in:
XMRig 2018-02-20 22:48:39 +07:00
commit 71759e3752
34 changed files with 2619 additions and 404 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

24
CHANGELOG.md
View file

@ -1,3 +1,27 @@
# 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 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.
- [#184](https://github.com/xmrig/xmrig/issues/184) Fixed cache size detection for CPUs with disabled Hyper-Threading.
- [#200](https://github.com/xmrig/xmrig/issues/200) In some cases miner was doesn't write log to stdout.
# v2.4.2
- [#60](https://github.com/xmrig/xmrig/issues/60) Added FreeBSD support, thanks [vcambur](https://github.com/vcambur).
- [#153](https://github.com/xmrig/xmrig/issues/153) Fixed issues with dwarfpool.com.
# v2.4.1
- [#147](https://github.com/xmrig/xmrig/issues/147) Fixed comparability with monero-stratum.
# v2.4.0 # v2.4.0
- Added [HTTP API](https://github.com/xmrig/xmrig/wiki/API). - Added [HTTP API](https://github.com/xmrig/xmrig/wiki/API).
- Added comments support in config file. - Added comments support in config file.

86
CMakeLists.txt
View file

@ -1,4 +1,4 @@
cmake_minimum_required(VERSION 3.0) cmake_minimum_required(VERSION 2.8)
project(xmrig) project(xmrig)
option(WITH_LIBCPUID "Use Libcpuid" ON) option(WITH_LIBCPUID "Use Libcpuid" ON)
@ -6,6 +6,7 @@ option(WITH_AEON "CryptoNight-Lite support" ON)
option(WITH_HTTPD "HTTP REST API" ON) option(WITH_HTTPD "HTTP REST API" ON)
include (CheckIncludeFile) include (CheckIncludeFile)
include (cmake/cpu.cmake)
set(HEADERS set(HEADERS
@ -56,13 +57,19 @@ set(HEADERS_CRYPTO
src/crypto/c_keccak.h src/crypto/c_keccak.h
src/crypto/c_skein.h src/crypto/c_skein.h
src/crypto/CryptoNight.h src/crypto/CryptoNight.h
src/crypto/CryptoNight_p.h
src/crypto/CryptoNight_test.h src/crypto/CryptoNight_test.h
src/crypto/groestl_tables.h src/crypto/groestl_tables.h
src/crypto/hash.h src/crypto/hash.h
src/crypto/skein_port.h src/crypto/skein_port.h
src/crypto/soft_aes.h
) )
if (XMRIG_ARM)
set(HEADERS_CRYPTO "${HEADERS_CRYPTO}" src/crypto/CryptoNight_arm.h)
else()
set(HEADERS_CRYPTO "${HEADERS_CRYPTO}" src/crypto/CryptoNight_x86.h)
endif()
set(SOURCES set(SOURCES
src/api/Api.cpp src/api/Api.cpp
src/api/ApiState.cpp src/api/ApiState.cpp
@ -99,8 +106,6 @@ set(SOURCES_CRYPTO
src/crypto/c_blake256.c src/crypto/c_blake256.c
src/crypto/c_jh.c src/crypto/c_jh.c
src/crypto/c_skein.c src/crypto/c_skein.c
src/crypto/soft_aes.c
src/crypto/soft_aes.c
src/crypto/CryptoNight.cpp src/crypto/CryptoNight.cpp
) )
@ -130,68 +135,29 @@ else()
src/Platform_unix.cpp src/Platform_unix.cpp
) )
set(EXTRA_LIBS pthread) set(EXTRA_LIBS pthread rt)
endif()
if (CMAKE_SYSTEM_NAME STREQUAL FreeBSD)
set(EXTRA_LIBS ${EXTRA_LIBS} kvm)
endif()
if (CMAKE_SYSTEM_NAME MATCHES "Linux")
EXECUTE_PROCESS(COMMAND uname -o COMMAND tr -d '\n' OUTPUT_VARIABLE OPERATING_SYSTEM)
if (OPERATING_SYSTEM MATCHES "Android")
set(EXTRA_LIBS ${EXTRA_LIBS} log)
endif()
endif() endif()
add_definitions(/D__STDC_FORMAT_MACROS) add_definitions(/D__STDC_FORMAT_MACROS)
add_definitions(/DUNICODE) add_definitions(/DUNICODE)
add_definitions(/DRAPIDJSON_SSE2)
#add_definitions(/DAPP_DEBUG) #add_definitions(/DAPP_DEBUG)
set(CMAKE_MODULE_PATH ${CMAKE_MODULE_PATH} "${CMAKE_SOURCE_DIR}/cmake") set(CMAKE_MODULE_PATH ${CMAKE_MODULE_PATH} "${CMAKE_SOURCE_DIR}/cmake")
find_package(UV REQUIRED) find_package(UV REQUIRED)
if ("${CMAKE_BUILD_TYPE}" STREQUAL "") include(cmake/flags.cmake)
set(CMAKE_BUILD_TYPE Release)
endif()
set(CMAKE_CXX_STANDARD_REQUIRED ON)
set(CMAKE_CXX_EXTENSIONS OFF)
set(CMAKE_CXX_STANDARD 11)
# https://cmake.org/cmake/help/latest/variable/CMAKE_LANG_COMPILER_ID.html
if (CMAKE_CXX_COMPILER_ID MATCHES GNU)
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -maes -Wall -Wno-strict-aliasing")
set(CMAKE_C_FLAGS_RELEASE "${CMAKE_C_FLAGS_RELEASE} -Ofast -funroll-loops -fvariable-expansion-in-unroller -ftree-loop-if-convert-stores -fmerge-all-constants -fbranch-target-load-optimize2")
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -maes -Wall -fno-exceptions -fno-rtti")
set(CMAKE_CXX_FLAGS_RELEASE "${CMAKE_CXX_FLAGS_RELEASE} -Ofast -s -funroll-loops -fvariable-expansion-in-unroller -ftree-loop-if-convert-stores -fmerge-all-constants -fbranch-target-load-optimize2")
if (WIN32)
set(CMAKE_EXE_LINKER_FLAGS "${CMAKE_EXE_LINKER_FLAGS} -static")
else()
set(CMAKE_EXE_LINKER_FLAGS "${CMAKE_EXE_LINKER_FLAGS} -static-libgcc -static-libstdc++")
endif()
add_definitions(/D_GNU_SOURCE)
if (${CMAKE_VERSION} VERSION_LESS "3.1.0")
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -std=c++11")
endif()
#set(CMAKE_C_FLAGS_RELEASE "${CMAKE_CXX_FLAGS_RELEASE} -gdwarf-2")
elseif (CMAKE_CXX_COMPILER_ID MATCHES MSVC)
set(CMAKE_C_FLAGS_RELEASE "${CMAKE_C_FLAGS_RELEASE} /Ox /Ot /Oi /MT /GL")
set(CMAKE_CXX_FLAGS_RELEASE "${CMAKE_CXX_FLAGS_RELEASE} /Ox /Ot /Oi /MT /GL")
add_definitions(/D_CRT_SECURE_NO_WARNINGS)
add_definitions(/D_CRT_NONSTDC_NO_WARNINGS)
add_definitions(/DNOMINMAX)
elseif (CMAKE_CXX_COMPILER_ID MATCHES Clang)
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -maes -Wall")
set(CMAKE_C_FLAGS_RELEASE "${CMAKE_C_FLAGS_RELEASE} -Ofast -funroll-loops -fmerge-all-constants")
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -maes -Wall -fno-exceptions -fno-rtti")
set(CMAKE_CXX_FLAGS_RELEASE "${CMAKE_CXX_FLAGS_RELEASE} -Ofast -funroll-loops -fmerge-all-constants")
endif()
if (WITH_LIBCPUID) if (WITH_LIBCPUID)
add_subdirectory(src/3rdparty/libcpuid) add_subdirectory(src/3rdparty/libcpuid)
@ -201,7 +167,12 @@ if (WITH_LIBCPUID)
set(SOURCES_CPUID src/Cpu.cpp) set(SOURCES_CPUID src/Cpu.cpp)
else() else()
add_definitions(/DXMRIG_NO_LIBCPUID) add_definitions(/DXMRIG_NO_LIBCPUID)
set(SOURCES_CPUID src/Cpu_stub.cpp)
if (XMRIG_ARM)
set(SOURCES_CPUID src/Cpu_arm.cpp)
else()
set(SOURCES_CPUID src/Cpu_stub.cpp)
endif()
endif() endif()
CHECK_INCLUDE_FILE (syslog.h HAVE_SYSLOG_H) CHECK_INCLUDE_FILE (syslog.h HAVE_SYSLOG_H)
@ -224,6 +195,7 @@ if (WITH_HTTPD)
message(FATAL_ERROR "microhttpd NOT found: use `-DWITH_HTTPD=OFF` to build without http deamon support") message(FATAL_ERROR "microhttpd NOT found: use `-DWITH_HTTPD=OFF` to build without http deamon support")
endif() endif()
else() else()
set(MHD_LIBRARY "")
add_definitions(/DXMRIG_NO_HTTPD) add_definitions(/DXMRIG_NO_HTTPD)
add_definitions(/DXMRIG_NO_API) add_definitions(/DXMRIG_NO_API)
endif() endif()

82
README.md
View file

@ -1,8 +1,16 @@
# XMRig # XMRig
XMRig is high performance Monero (XMR) CPU miner, with the official full Windows support. [![Github All Releases](https://img.shields.io/github/downloads/xmrig/xmrig/total.svg)](https://github.com/xmrig/xmrig/releases)
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 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)
* This is the CPU-mining version, there is also a [NVIDIA GPU version](https://github.com/xmrig/xmrig-nvidia). 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.
<img src="http://i.imgur.com/OKZRVDh.png" width="619" > <img src="http://i.imgur.com/OKZRVDh.png" width="619" >
@ -36,44 +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). * Clone with `git clone https://github.com/xmrig/xmrig.git` :hammer: [Build instructions](https://github.com/xmrig/xmrig/wiki/Build).
## Usage ## Usage
### Basic example Use [config.xmrig.com](https://config.xmrig.com/xmrig) to generate, edit or share configurations.
```
xmrig.exe -o pool.minemonero.pro:5555 -u YOUR_WALLET -p x -k
```
### Failover
```
xmrig.exe -o pool.minemonero.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.
### Options ### Options
``` ```
-a, --algo=ALGO cryptonight (default) or cryptonight-lite -a, --algo=ALGO cryptonight (default) or cryptonight-lite
-o, --url=URL URL of mining server -o, --url=URL URL of mining server
-O, --userpass=U:P username:password pair for mining server -O, --userpass=U:P username:password pair for mining server
-u, --user=USERNAME username for mining server -u, --user=USERNAME username for mining server
-p, --pass=PASSWORD password for mining server -p, --pass=PASSWORD password for mining server
-t, --threads=N number of miner threads -t, --threads=N number of miner threads
-v, --av=N algorithm variation, 0 auto select -v, --av=N algorithm variation, 0 auto select
-k, --keepalive send keepalived for prevent timeout (need pool support) -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, --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) -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-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) --cpu-priority set process priority (0 idle, 2 normal to 5 highest)
--no-huge-pages disable huge pages support --no-huge-pages disable huge pages support
--no-color disable colored output --no-color disable colored output
--donate-level=N donate level, default 5% (5 minutes in 100 minutes) --donate-level=N donate level, default 5% (5 minutes in 100 minutes)
--user-agent set custom user-agent string for pool --user-agent set custom user-agent string for pool
-B, --background run the miner in the background -B, --background run the miner in the background
-c, --config=FILE load a JSON-format configuration file -c, --config=FILE load a JSON-format configuration file
-l, --log-file=FILE log all output to a file -l, --log-file=FILE log all output to a file
--max-cpu-usage=N maximum CPU usage for automatic threads mode (default 75) --max-cpu-usage=N maximum CPU usage for automatic threads mode (default 75)
--safe safe adjust threads and av settings for current CPU --safe safe adjust threads and av settings for current CPU
--nicehash enable nicehash support --nicehash enable nicehash/xmrig-proxy support
--print-time=N print hashrate report every N seconds --print-time=N print hashrate report every N seconds
-h, --help display this help and exit --api-port=N port for the miner API
-V, --version output version information and exit --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. Also you can use configuration via config file, default **config.json**. You can load multiple config files and combine it with command line options.
@ -88,7 +90,7 @@ Since version 0.8.0.
## Common Issues ## Common Issues
### HUGE PAGES unavailable ### HUGE PAGES unavailable
* Run XMRig as Administrator. * 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 ## Other information
* No HTTP support, only stratum protocol support. * No HTTP support, only stratum protocol support.
@ -105,7 +107,7 @@ Please note performance is highly dependent on system load. The numbers above ar
### Maximum performance checklist ### Maximum performance checklist
* Idle operating system. * Idle operating system.
* Do not exceed optimal thread count. * 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. * Try setup optimal cpu affinity.
* Enable fast memory (Large/Huge pages). * Enable fast memory (Large/Huge pages).

25
cmake/cpu.cmake Normal file
View file

@ -0,0 +1,25 @@
if (NOT CMAKE_SYSTEM_PROCESSOR)
message(WARNING "CMAKE_SYSTEM_PROCESSOR not defined")
endif()
if (CMAKE_SYSTEM_PROCESSOR MATCHES "^(x86_64|AMD64)$")
add_definitions(/DRAPIDJSON_SSE2)
endif()
if (CMAKE_SYSTEM_PROCESSOR MATCHES "^(aarch64)$")
set(XMRIG_ARM ON)
set(XMRIG_ARMv8 ON)
set(WITH_LIBCPUID OFF)
add_definitions(/DXMRIG_ARM)
add_definitions(/DXMRIG_ARMv8)
elseif (CMAKE_SYSTEM_PROCESSOR MATCHES "^(armv7|armv7f|armv7s|armv7k|armv7-a|armv7l)$")
set(XMRIG_ARM ON)
set(XMRIG_ARMv7 ON)
set(WITH_LIBCPUID OFF)
add_definitions(/DXMRIG_ARM)
add_definitions(/DXMRIG_ARMv7)
endif()

69
cmake/flags.cmake Normal file
View file

@ -0,0 +1,69 @@
set(CMAKE_CXX_STANDARD_REQUIRED ON)
set(CMAKE_CXX_EXTENSIONS OFF)
set(CMAKE_CXX_STANDARD 11)
if ("${CMAKE_BUILD_TYPE}" STREQUAL "")
set(CMAKE_BUILD_TYPE Release)
endif()
if (CMAKE_CXX_COMPILER_ID MATCHES GNU)
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -Wall -Wno-strict-aliasing")
set(CMAKE_C_FLAGS_RELEASE "${CMAKE_C_FLAGS_RELEASE} -Ofast -funroll-loops -fvariable-expansion-in-unroller -ftree-loop-if-convert-stores -fmerge-all-constants -fbranch-target-load-optimize2")
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wall -fno-exceptions -fno-rtti")
set(CMAKE_CXX_FLAGS_RELEASE "${CMAKE_CXX_FLAGS_RELEASE} -Ofast -s -funroll-loops -fvariable-expansion-in-unroller -ftree-loop-if-convert-stores -fmerge-all-constants -fbranch-target-load-optimize2")
if (XMRIG_ARMv8)
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -march=armv8-a+crypto")
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -march=armv8-a+crypto -flax-vector-conversions")
elseif (XMRIG_ARMv7)
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -mfpu=neon")
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -mfpu=neon -flax-vector-conversions")
else()
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -maes")
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -maes")
endif()
if (WIN32)
set(CMAKE_EXE_LINKER_FLAGS "${CMAKE_EXE_LINKER_FLAGS} -static")
else()
set(CMAKE_EXE_LINKER_FLAGS "${CMAKE_EXE_LINKER_FLAGS} -static-libgcc -static-libstdc++")
endif()
add_definitions(/D_GNU_SOURCE)
if (${CMAKE_VERSION} VERSION_LESS "3.1.0")
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -std=c++11")
endif()
#set(CMAKE_C_FLAGS_RELEASE "${CMAKE_CXX_FLAGS_RELEASE} -gdwarf-2")
elseif (CMAKE_CXX_COMPILER_ID MATCHES MSVC)
set(CMAKE_C_FLAGS_RELEASE "${CMAKE_C_FLAGS_RELEASE} /Ox /Ot /Oi /MT /GL")
set(CMAKE_CXX_FLAGS_RELEASE "${CMAKE_CXX_FLAGS_RELEASE} /Ox /Ot /Oi /MT /GL")
add_definitions(/D_CRT_SECURE_NO_WARNINGS)
add_definitions(/D_CRT_NONSTDC_NO_WARNINGS)
add_definitions(/DNOMINMAX)
elseif (CMAKE_CXX_COMPILER_ID MATCHES Clang)
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -Wall")
set(CMAKE_C_FLAGS_RELEASE "${CMAKE_C_FLAGS_RELEASE} -Ofast -funroll-loops -fmerge-all-constants")
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wall -fno-exceptions -fno-rtti -Wno-missing-braces")
set(CMAKE_CXX_FLAGS_RELEASE "${CMAKE_CXX_FLAGS_RELEASE} -Ofast -funroll-loops -fmerge-all-constants")
if (XMRIG_ARMv8)
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -march=armv8-a+crypto")
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -march=armv8-a+crypto")
elseif (XMRIG_ARMv7)
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -mfpu=neon -march=${CMAKE_SYSTEM_PROCESSOR}")
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -mfpu=neon -march=${CMAKE_SYSTEM_PROCESSOR}")
else()
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -maes")
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -maes")
endif()
endif()

65
src/3rdparty/aligned_malloc.h vendored Normal file
View file

@ -0,0 +1,65 @@
/* XMRig
* Copyright 2010 Jeff Garzik <jgarzik@pobox.com>
* Copyright 2012-2014 pooler <pooler@litecoinpool.org>
* 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>
*
*
* This program is free software: you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program. If not, see <http://www.gnu.org/licenses/>.
*/
#ifndef __ALIGNED_MALLOC_H__
#define __ALIGNED_MALLOC_H__
#include <stdlib.h>
#ifndef __cplusplus
extern int posix_memalign(void **__memptr, size_t __alignment, size_t __size);
#else
// Some systems (e.g. those with GNU libc) declare posix_memalign with an
// exception specifier. Via an "egregious workaround" in
// Sema::CheckEquivalentExceptionSpec, Clang accepts the following as a valid
// redeclaration of glibc's declaration.
extern "C" int posix_memalign(void **__memptr, size_t __alignment, size_t __size);
#endif
static __inline__ void *__attribute__((__always_inline__, __malloc__)) _mm_malloc(size_t __size, size_t __align)
{
if (__align == 1) {
return malloc(__size);
}
if (!(__align & (__align - 1)) && __align < sizeof(void *))
__align = sizeof(void *);
void *__mallocedMemory;
if (posix_memalign(&__mallocedMemory, __align, __size)) {
return 0;
}
return __mallocedMemory;
}
static __inline__ void __attribute__((__always_inline__)) _mm_free(void *__p)
{
free(__p);
}
#endif /* __ALIGNED_MALLOC_H__ */

44
src/App.cpp
View file

@ -92,7 +92,9 @@ App::App(int argc, char **argv) :
m_network = new Network(m_options); 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() int App::exec()
{ {
if (!m_options) { if (!m_options) {
return 0; return 2;
} }
uv_signal_start(&m_signal, App::onSignal, SIGHUP); uv_signal_start(&m_sigHUP, App::onSignal, SIGHUP);
uv_signal_start(&m_signal, App::onSignal, SIGTERM); uv_signal_start(&m_sigINT, App::onSignal, SIGINT);
uv_signal_start(&m_signal, App::onSignal, SIGINT); uv_signal_start(&m_sigTERM, App::onSignal, SIGTERM);
background(); background();
@ -128,6 +130,13 @@ int App::exec()
Mem::allocate(m_options->algo(), m_options->threads(), m_options->doubleHash(), m_options->hugePages()); Mem::allocate(m_options->algo(), m_options->threads(), m_options->doubleHash(), m_options->hugePages());
Summary::print(); Summary::print();
if (m_options->dryRun()) {
LOG_NOTICE("OK");
release();
return 0;
}
# ifndef XMRIG_NO_API # ifndef XMRIG_NO_API
Api::start(); Api::start();
# endif # endif
@ -144,12 +153,7 @@ int App::exec()
const int r = uv_run(uv_default_loop(), UV_RUN_DEFAULT); const int r = uv_run(uv_default_loop(), UV_RUN_DEFAULT);
uv_loop_close(uv_default_loop()); uv_loop_close(uv_default_loop());
delete m_network; release();
Options::release();
Mem::release();
Platform::release();
return r; return r;
} }
@ -164,8 +168,10 @@ void App::onConsoleCommand(char command)
case 'p': case 'p':
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"); if (Workers::isEnabled()) {
Workers::setEnabled(false); 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; break;
case 'r': 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) void App::onSignal(uv_signal_t *handle, int signum)
{ {
switch (signum) switch (signum)

5
src/App.h
View file

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

16
src/Cpu.cpp
View file

@ -86,9 +86,13 @@ void Cpu::initCommon()
strncpy(m_brand, data.brand_str, sizeof(m_brand) - 1); strncpy(m_brand, data.brand_str, sizeof(m_brand) - 1);
m_totalThreads = data.total_logical_cpus; m_totalThreads = data.total_logical_cpus;
m_sockets = m_totalThreads / data.num_logical_cpus; m_sockets = m_totalThreads / data.num_logical_cpus;
m_totalCores = data.num_cores *m_sockets;
if (m_sockets == 0) {
m_sockets = 1;
}
m_totalCores = data.num_cores * m_sockets;
m_l3_cache = data.l3_cache > 0 ? data.l3_cache * m_sockets : 0; m_l3_cache = data.l3_cache > 0 ? data.l3_cache * m_sockets : 0;
// Workaround for AMD CPUs https://github.com/anrieff/libcpuid/issues/97 // Workaround for AMD CPUs https://github.com/anrieff/libcpuid/issues/97
@ -96,7 +100,13 @@ void Cpu::initCommon()
m_l2_cache = data.l2_cache * (m_totalCores / 2) * m_sockets; m_l2_cache = data.l2_cache * (m_totalCores / 2) * m_sockets;
m_l2_exclusive = true; 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; m_l2_cache = data.l2_cache > 0 ? data.l2_cache * m_totalCores * m_sockets : 0;
} }

54
src/Cpu_arm.cpp Normal file
View file

@ -0,0 +1,54 @@
/* XMRig
* Copyright 2010 Jeff Garzik <jgarzik@pobox.com>
* Copyright 2012-2014 pooler <pooler@litecoinpool.org>
* 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>
*
*
* This program is free software: you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program. If not, see <http://www.gnu.org/licenses/>.
*/
#include <string.h>
#include "Cpu.h"
char Cpu::m_brand[64] = { 0 };
int Cpu::m_flags = 0;
int Cpu::m_l2_cache = 0;
int Cpu::m_l3_cache = 0;
int Cpu::m_sockets = 1;
int Cpu::m_totalCores = 0;
int Cpu::m_totalThreads = 0;
int Cpu::optimalThreadsCount(int algo, bool doubleHash, int maxCpuUsage)
{
return m_totalThreads;
}
void Cpu::initCommon()
{
memcpy(m_brand, "Unknown", 7);
# if defined(XMRIG_ARMv8)
m_flags |= X86_64;
m_flags |= AES;
# endif
}

19
src/Cpu_unix.cpp
View file

@ -22,6 +22,14 @@
*/ */
#ifdef __FreeBSD__
# include <sys/types.h>
# include <sys/param.h>
# include <sys/cpuset.h>
# include <pthread_np.h>
#endif
#include <pthread.h> #include <pthread.h>
#include <sched.h> #include <sched.h>
#include <unistd.h> #include <unistd.h>
@ -31,6 +39,11 @@
#include "Cpu.h" #include "Cpu.h"
#ifdef __FreeBSD__
typedef cpuset_t cpu_set_t;
#endif
void Cpu::init() void Cpu::init()
{ {
# ifdef XMRIG_NO_LIBCPUID # ifdef XMRIG_NO_LIBCPUID
@ -53,8 +66,14 @@ void Cpu::setAffinity(int id, uint64_t mask)
} }
if (id == -1) { if (id == -1) {
# ifndef __FreeBSD__
sched_setaffinity(0, sizeof(&set), &set); sched_setaffinity(0, sizeof(&set), &set);
# endif
} else { } else {
# ifndef __ANDROID__
pthread_setaffinity_np(pthread_self(), sizeof(&set), &set); pthread_setaffinity_np(pthread_self(), sizeof(&set), &set);
# else
sched_setaffinity(gettid(), sizeof(&set), &set);
# endif
} }
} }

11
src/Mem_unix.cpp
View file

@ -23,10 +23,16 @@
#include <stdlib.h> #include <stdlib.h>
#include <mm_malloc.h>
#include <sys/mman.h> #include <sys/mman.h>
#if defined(XMRIG_ARM) && !defined(__clang__)
# include "aligned_malloc.h"
#else
# include <mm_malloc.h>
#endif
#include "crypto/CryptoNight.h" #include "crypto/CryptoNight.h"
#include "log/Log.h" #include "log/Log.h"
#include "Mem.h" #include "Mem.h"
@ -51,10 +57,11 @@ bool Mem::allocate(int algo, int threads, bool doubleHash, bool enabled)
# if defined(__APPLE__) # if defined(__APPLE__)
m_memory = static_cast<uint8_t*>(mmap(0, size, PROT_READ | PROT_WRITE, MAP_PRIVATE | MAP_ANON, VM_FLAGS_SUPERPAGE_SIZE_2MB, 0)); m_memory = static_cast<uint8_t*>(mmap(0, size, PROT_READ | PROT_WRITE, MAP_PRIVATE | MAP_ANON, VM_FLAGS_SUPERPAGE_SIZE_2MB, 0));
# elif defined(__FreeBSD__)
m_memory = static_cast<uint8_t*>(mmap(0, size, PROT_READ | PROT_WRITE, MAP_PRIVATE | MAP_ANONYMOUS | MAP_ALIGNED_SUPER | MAP_PREFAULT_READ, -1, 0));
# else # else
m_memory = static_cast<uint8_t*>(mmap(0, size, PROT_READ | PROT_WRITE, MAP_PRIVATE | MAP_ANONYMOUS | MAP_HUGETLB | MAP_POPULATE, 0, 0)); m_memory = static_cast<uint8_t*>(mmap(0, size, PROT_READ | PROT_WRITE, MAP_PRIVATE | MAP_ANONYMOUS | MAP_HUGETLB | MAP_POPULATE, 0, 0));
# endif # endif
if (m_memory == MAP_FAILED) { if (m_memory == MAP_FAILED) {
m_memory = static_cast<uint8_t*>(_mm_malloc(size, 16)); m_memory = static_cast<uint8_t*>(_mm_malloc(size, 16));
return true; return true;

23
src/Options.cpp
View file

@ -33,6 +33,11 @@
#endif #endif
#ifndef XMRIG_NO_HTTPD
# include <microhttpd.h>
#endif
#include "Cpu.h" #include "Cpu.h"
#include "donate.h" #include "donate.h"
#include "net/Url.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[] = { static struct option const options[] = {
{ "algo", 1, nullptr, 'a' }, { "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' }, { "av", 1, nullptr, 'v' },
{ "background", 0, nullptr, 'B' }, { "background", 0, nullptr, 'B' },
{ "config", 1, nullptr, 'c' }, { "config", 1, nullptr, 'c' },
{ "cpu-affinity", 1, nullptr, 1020 }, { "cpu-affinity", 1, nullptr, 1020 },
{ "cpu-priority", 1, nullptr, 1021 }, { "cpu-priority", 1, nullptr, 1021 },
{ "donate-level", 1, nullptr, 1003 }, { "donate-level", 1, nullptr, 1003 },
{ "dry-run", 0, nullptr, 5000 },
{ "help", 0, nullptr, 'h' }, { "help", 0, nullptr, 'h' },
{ "keepalive", 0, nullptr ,'k' }, { "keepalive", 0, nullptr ,'k' },
{ "log-file", 1, nullptr, 'l' }, { "log-file", 1, nullptr, 'l' },
@ -121,9 +130,6 @@ static struct option const options[] = {
{ "user-agent", 1, nullptr, 1008 }, { "user-agent", 1, nullptr, 1008 },
{ "userpass", 1, nullptr, 'O' }, { "userpass", 1, nullptr, 'O' },
{ "version", 0, nullptr, 'V' }, { "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 } { 0, 0, 0, 0 }
}; };
@ -136,6 +142,7 @@ static struct option const config_options[] = {
{ "cpu-affinity", 1, nullptr, 1020 }, { "cpu-affinity", 1, nullptr, 1020 },
{ "cpu-priority", 1, nullptr, 1021 }, { "cpu-priority", 1, nullptr, 1021 },
{ "donate-level", 1, nullptr, 1003 }, { "donate-level", 1, nullptr, 1003 },
{ "dry-run", 0, nullptr, 5000 },
{ "huge-pages", 0, nullptr, 1009 }, { "huge-pages", 0, nullptr, 1009 },
{ "log-file", 1, nullptr, 'l' }, { "log-file", 1, nullptr, 'l' },
{ "max-cpu-usage", 1, nullptr, 1004 }, { "max-cpu-usage", 1, nullptr, 1004 },
@ -200,6 +207,7 @@ Options::Options(int argc, char **argv) :
m_background(false), m_background(false),
m_colors(true), m_colors(true),
m_doubleHash(false), m_doubleHash(false),
m_dryRun(false),
m_hugePages(true), m_hugePages(true),
m_ready(false), m_ready(false),
m_safe(false), m_safe(false),
@ -379,6 +387,7 @@ bool Options::parseArg(int key, const char *arg)
case 'S': /* --syslog */ case 'S': /* --syslog */
case 1005: /* --safe */ case 1005: /* --safe */
case 1006: /* --nicehash */ case 1006: /* --nicehash */
case 5000: /* --dry-run */
return parseBoolean(key, true); return parseBoolean(key, true);
case 1002: /* --no-color */ case 1002: /* --no-color */
@ -552,6 +561,10 @@ bool Options::parseBoolean(int key, bool enable)
m_colors = enable; m_colors = enable;
break; break;
case 5000: /* --dry-run */
m_dryRun = enable;
break;
default: default:
break; break;
} }
@ -665,6 +678,10 @@ void Options::showVersion()
"\n"); "\n");
printf("\nlibuv/%s\n", uv_version_string()); printf("\nlibuv/%s\n", uv_version_string());
# ifndef XMRIG_NO_HTTPD
printf("libmicrohttpd/%s\n", MHD_get_version());
# endif
} }

3
src/Options.h
View file

@ -59,6 +59,7 @@ public:
inline bool background() const { return m_background; } inline bool background() const { return m_background; }
inline bool colors() const { return m_colors; } inline bool colors() const { return m_colors; }
inline bool doubleHash() const { return m_doubleHash; } inline bool doubleHash() const { return m_doubleHash; }
inline bool dryRun() const { return m_dryRun; }
inline bool hugePages() const { return m_hugePages; } inline bool hugePages() const { return m_hugePages; }
inline bool syslog() const { return m_syslog; } inline bool syslog() const { return m_syslog; }
inline const char *apiToken() const { return m_apiToken; } inline const char *apiToken() const { return m_apiToken; }
@ -76,6 +77,7 @@ public:
inline int retryPause() const { return m_retryPause; } inline int retryPause() const { return m_retryPause; }
inline int threads() const { return m_threads; } inline int threads() const { return m_threads; }
inline int64_t affinity() const { return m_affinity; } inline int64_t affinity() const { return m_affinity; }
inline void setColors(bool colors) { m_colors = colors; }
inline static void release() { delete m_self; } inline static void release() { delete m_self; }
@ -109,6 +111,7 @@ private:
bool m_background; bool m_background;
bool m_colors; bool m_colors;
bool m_doubleHash; bool m_doubleHash;
bool m_dryRun;
bool m_hugePages; bool m_hugePages;
bool m_ready; bool m_ready;
bool m_safe; bool m_safe;

1
src/Platform_mac.cpp
View file

@ -22,6 +22,7 @@
*/ */
#include <stdio.h>
#include <stdlib.h> #include <stdlib.h>
#include <sys/resource.h> #include <sys/resource.h>
#include <uv.h> #include <uv.h>

4
src/Platform_unix.cpp
View file

@ -21,8 +21,8 @@
* along with this program. If not, see <http://www.gnu.org/licenses/>. * along with this program. If not, see <http://www.gnu.org/licenses/>.
*/ */
#include <sched.h> #include <sched.h>
#include <stdio.h>
#include <stdlib.h> #include <stdlib.h>
#include <string.h> #include <string.h>
#include <sys/resource.h> #include <sys/resource.h>
@ -116,6 +116,7 @@ void Platform::setThreadPriority(int priority)
setpriority(PRIO_PROCESS, 0, prio); setpriority(PRIO_PROCESS, 0, prio);
# ifdef SCHED_IDLE
if (priority == 0) { if (priority == 0) {
sched_param param; sched_param param;
param.sched_priority = 0; param.sched_priority = 0;
@ -124,4 +125,5 @@ void Platform::setThreadPriority(int priority)
sched_setscheduler(0, SCHED_BATCH, &param); sched_setscheduler(0, SCHED_BATCH, &param);
} }
} }
# endif
} }

1
src/Summary.cpp
View file

@ -23,6 +23,7 @@
#include <inttypes.h> #include <inttypes.h>
#include <stdio.h>
#include <uv.h> #include <uv.h>

6
src/api/ApiState.cpp
View file

@ -21,7 +21,7 @@
* along with this program. If not, see <http://www.gnu.org/licenses/>. * along with this program. If not, see <http://www.gnu.org/licenses/>.
*/ */
#include <cmath> #include <math.h>
#include <string.h> #include <string.h>
#include <uv.h> #include <uv.h>
@ -53,11 +53,11 @@ extern "C"
static inline double normalize(double d) static inline double normalize(double d)
{ {
if (!std::isnormal(d)) { if (!isnormal(d)) {
return 0.0; return 0.0;
} }
return std::floor(d * 100.0) / 100.0; return floor(d * 100.0) / 100.0;
} }

1
src/api/ApiState.h
View file

@ -57,7 +57,6 @@ private:
double m_highestHashrate; double m_highestHashrate;
double m_totalHashrate[3]; double m_totalHashrate[3];
int m_threads; int m_threads;
mutable char m_buf[4096];
NetworkState m_network; NetworkState m_network;
}; };

4
src/api/Httpd.cpp
View file

@ -31,10 +31,6 @@
#include "log/Log.h" #include "log/Log.h"
static const char k500 [] = "{\"error\":\"INTERNAL_SERVER_ERROR\"}";
static const size_t k500Size = sizeof(k500) - 1;
Httpd::Httpd(int port, const char *accessToken) : Httpd::Httpd(int port, const char *accessToken) :
m_accessToken(accessToken), m_accessToken(accessToken),
m_port(port), m_port(port),

5
src/api/NetworkState.cpp
View file

@ -23,6 +23,7 @@
#include <algorithm> #include <algorithm>
#include <stdio.h>
#include <string.h> #include <string.h>
#include <uv.h> #include <uv.h>
@ -45,7 +46,7 @@ NetworkState::NetworkState() :
int NetworkState::connectionTime() const 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;
} }
@ -55,7 +56,7 @@ uint32_t NetworkState::avgTime() const
return 0; return 0;
} }
return (uint32_t) connectionTime() / m_latency.size(); return connectionTime() / (uint32_t)m_latency.size();
} }

16
src/crypto/CryptoNight.cpp
View file

@ -23,7 +23,13 @@
#include "crypto/CryptoNight.h" #include "crypto/CryptoNight.h"
#include "crypto/CryptoNight_p.h"
#if defined(XMRIG_ARM)
# include "crypto/CryptoNight_arm.h"
#else
# include "crypto/CryptoNight_x86.h"
#endif
#include "crypto/CryptoNight_test.h" #include "crypto/CryptoNight_test.h"
#include "net/Job.h" #include "net/Job.h"
#include "net/JobResult.h" #include "net/JobResult.h"
@ -34,12 +40,16 @@ void (*cryptonight_hash_ctx)(const void *input, size_t size, void *output, crypt
static void cryptonight_av1_aesni(const void *input, size_t size, void *output, struct cryptonight_ctx *ctx) { static void cryptonight_av1_aesni(const void *input, size_t size, void *output, struct cryptonight_ctx *ctx) {
# if !defined(XMRIG_ARMv7)
cryptonight_hash<0x80000, MEMORY, 0x1FFFF0, false>(input, size, output, ctx); cryptonight_hash<0x80000, MEMORY, 0x1FFFF0, false>(input, size, output, ctx);
# endif
} }
static void cryptonight_av2_aesni_double(const void *input, size_t size, void *output, cryptonight_ctx *ctx) { static void cryptonight_av2_aesni_double(const void *input, size_t size, void *output, cryptonight_ctx *ctx) {
# if !defined(XMRIG_ARMv7)
cryptonight_double_hash<0x80000, MEMORY, 0x1FFFF0, false>(input, size, output, ctx); cryptonight_double_hash<0x80000, MEMORY, 0x1FFFF0, false>(input, size, output, ctx);
# endif
} }
@ -55,12 +65,16 @@ static void cryptonight_av4_softaes_double(const void *input, size_t size, void
#ifndef XMRIG_NO_AEON #ifndef XMRIG_NO_AEON
static void cryptonight_lite_av1_aesni(const void *input, size_t size, void *output, cryptonight_ctx *ctx) { static void cryptonight_lite_av1_aesni(const void *input, size_t size, void *output, cryptonight_ctx *ctx) {
# if !defined(XMRIG_ARMv7)
cryptonight_hash<0x40000, MEMORY_LITE, 0xFFFF0, false>(input, size, output, ctx); cryptonight_hash<0x40000, MEMORY_LITE, 0xFFFF0, false>(input, size, output, ctx);
#endif
} }
static void cryptonight_lite_av2_aesni_double(const void *input, size_t size, void *output, cryptonight_ctx *ctx) { static void cryptonight_lite_av2_aesni_double(const void *input, size_t size, void *output, cryptonight_ctx *ctx) {
# if !defined(XMRIG_ARMv7)
cryptonight_double_hash<0x40000, MEMORY_LITE, 0xFFFF0, false>(input, size, output, ctx); cryptonight_double_hash<0x40000, MEMORY_LITE, 0xFFFF0, false>(input, size, output, ctx);
# endif
} }

493
src/crypto/CryptoNight_arm.h Normal file
View file

@ -0,0 +1,493 @@
/* XMRig
* Copyright 2010 Jeff Garzik <jgarzik@pobox.com>
* Copyright 2012-2014 pooler <pooler@litecoinpool.org>
* 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 Imran Yusuff <https://github.com/imranyusuff>
* Copyright 2016-2017 XMRig <support@xmrig.com>
*
*
* This program is free software: you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program. If not, see <http://www.gnu.org/licenses/>.
*/
#ifndef __CRYPTONIGHT_ARM_H__
#define __CRYPTONIGHT_ARM_H__
#if defined(XMRIG_ARM) && !defined(__clang__)
# include "aligned_malloc.h"
#else
# include <mm_malloc.h>
#endif
#include "crypto/CryptoNight.h"
#include "crypto/soft_aes.h"
extern "C"
{
#include "crypto/c_keccak.h"
#include "crypto/c_groestl.h"
#include "crypto/c_blake256.h"
#include "crypto/c_jh.h"
#include "crypto/c_skein.h"
}
static inline void do_blake_hash(const void* input, size_t len, char* output) {
blake256_hash(reinterpret_cast<uint8_t*>(output), static_cast<const uint8_t*>(input), len);
}
static inline void do_groestl_hash(const void* input, size_t len, char* output) {
groestl(static_cast<const uint8_t*>(input), len * 8, reinterpret_cast<uint8_t*>(output));
}
static inline void do_jh_hash(const void* input, size_t len, char* output) {
jh_hash(32 * 8, static_cast<const uint8_t*>(input), 8 * len, reinterpret_cast<uint8_t*>(output));
}
static inline void do_skein_hash(const void* input, size_t len, char* output) {
xmr_skein(static_cast<const uint8_t*>(input), reinterpret_cast<uint8_t*>(output));
}
void (* const extra_hashes[4])(const void *, size_t, char *) = {do_blake_hash, do_groestl_hash, do_jh_hash, do_skein_hash};
static inline __attribute__((always_inline)) __m128i _mm_set_epi64x(const uint64_t a, const uint64_t b)
{
return vcombine_u64(vcreate_u64(b), vcreate_u64(a));
}
/* this one was not implemented yet so here it is */
static inline __attribute__((always_inline)) uint64_t _mm_cvtsi128_si64(__m128i a)
{
return vgetq_lane_u64(a, 0);
}
#define EXTRACT64(X) _mm_cvtsi128_si64(X)
#if defined(XMRIG_ARMv8)
static inline uint64_t __umul128(uint64_t a, uint64_t b, uint64_t* hi)
{
unsigned __int128 r = (unsigned __int128) a * (unsigned __int128) b;
*hi = r >> 64;
return (uint64_t) r;
}
#else
static inline uint64_t __umul128(uint64_t multiplier, uint64_t multiplicand, uint64_t *product_hi) {
// multiplier = ab = a * 2^32 + b
// multiplicand = cd = c * 2^32 + d
// ab * cd = a * c * 2^64 + (a * d + b * c) * 2^32 + b * d
uint64_t a = multiplier >> 32;
uint64_t b = multiplier & 0xFFFFFFFF;
uint64_t c = multiplicand >> 32;
uint64_t d = multiplicand & 0xFFFFFFFF;
//uint64_t ac = a * c;
uint64_t ad = a * d;
//uint64_t bc = b * c;
uint64_t bd = b * d;
uint64_t adbc = ad + (b * c);
uint64_t adbc_carry = adbc < ad ? 1 : 0;
// multiplier * multiplicand = product_hi * 2^64 + product_lo
uint64_t product_lo = bd + (adbc << 32);
uint64_t product_lo_carry = product_lo < bd ? 1 : 0;
*product_hi = (a * c) + (adbc >> 32) + (adbc_carry << 32) + product_lo_carry;
return product_lo;
}
#endif
// This will shift and xor tmp1 into itself as 4 32-bit vals such as
// sl_xor(a1 a2 a3 a4) = a1 (a2^a1) (a3^a2^a1) (a4^a3^a2^a1)
static inline __m128i sl_xor(__m128i tmp1)
{
__m128i tmp4;
tmp4 = _mm_slli_si128(tmp1, 0x04);
tmp1 = _mm_xor_si128(tmp1, tmp4);
tmp4 = _mm_slli_si128(tmp4, 0x04);
tmp1 = _mm_xor_si128(tmp1, tmp4);
tmp4 = _mm_slli_si128(tmp4, 0x04);
tmp1 = _mm_xor_si128(tmp1, tmp4);
return tmp1;
}
template<uint8_t rcon>
static inline void aes_genkey_sub(__m128i* xout0, __m128i* xout2)
{
// __m128i xout1 = _mm_aeskeygenassist_si128(*xout2, rcon);
// xout1 = _mm_shuffle_epi32(xout1, 0xFF); // see PSHUFD, set all elems to 4th elem
// *xout0 = sl_xor(*xout0);
// *xout0 = _mm_xor_si128(*xout0, xout1);
// xout1 = _mm_aeskeygenassist_si128(*xout0, 0x00);
// xout1 = _mm_shuffle_epi32(xout1, 0xAA); // see PSHUFD, set all elems to 3rd elem
// *xout2 = sl_xor(*xout2);
// *xout2 = _mm_xor_si128(*xout2, xout1);
}
template<uint8_t rcon>
static inline void soft_aes_genkey_sub(__m128i* xout0, __m128i* xout2)
{
__m128i xout1 = soft_aeskeygenassist<rcon>(*xout2);
xout1 = _mm_shuffle_epi32(xout1, 0xFF); // see PSHUFD, set all elems to 4th elem
*xout0 = sl_xor(*xout0);
*xout0 = _mm_xor_si128(*xout0, xout1);
xout1 = soft_aeskeygenassist<0x00>(*xout0);
xout1 = _mm_shuffle_epi32(xout1, 0xAA); // see PSHUFD, set all elems to 3rd elem
*xout2 = sl_xor(*xout2);
*xout2 = _mm_xor_si128(*xout2, xout1);
}
template<bool SOFT_AES>
static inline void aes_genkey(const __m128i* memory, __m128i* k0, __m128i* k1, __m128i* k2, __m128i* k3, __m128i* k4, __m128i* k5, __m128i* k6, __m128i* k7, __m128i* k8, __m128i* k9)
{
__m128i xout0 = _mm_load_si128(memory);
__m128i xout2 = _mm_load_si128(memory + 1);
*k0 = xout0;
*k1 = xout2;
SOFT_AES ? soft_aes_genkey_sub<0x01>(&xout0, &xout2) : soft_aes_genkey_sub<0x01>(&xout0, &xout2);
*k2 = xout0;
*k3 = xout2;
SOFT_AES ? soft_aes_genkey_sub<0x02>(&xout0, &xout2) : soft_aes_genkey_sub<0x02>(&xout0, &xout2);
*k4 = xout0;
*k5 = xout2;
SOFT_AES ? soft_aes_genkey_sub<0x04>(&xout0, &xout2) : soft_aes_genkey_sub<0x04>(&xout0, &xout2);
*k6 = xout0;
*k7 = xout2;
SOFT_AES ? soft_aes_genkey_sub<0x08>(&xout0, &xout2) : soft_aes_genkey_sub<0x08>(&xout0, &xout2);
*k8 = xout0;
*k9 = xout2;
}
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((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 {
*x0 = vaesmcq_u8(vaeseq_u8(*((uint8x16_t *) x0), key));
*x1 = vaesmcq_u8(vaeseq_u8(*((uint8x16_t *) x1), key));
*x2 = vaesmcq_u8(vaeseq_u8(*((uint8x16_t *) x2), key));
*x3 = vaesmcq_u8(vaeseq_u8(*((uint8x16_t *) x3), key));
*x4 = vaesmcq_u8(vaeseq_u8(*((uint8x16_t *) x4), key));
*x5 = vaesmcq_u8(vaeseq_u8(*((uint8x16_t *) x5), key));
*x6 = vaesmcq_u8(vaeseq_u8(*((uint8x16_t *) x6), key));
*x7 = vaesmcq_u8(vaeseq_u8(*((uint8x16_t *) x7), key));
}
# endif
}
template<size_t MEM, bool SOFT_AES>
static inline void cn_explode_scratchpad(const __m128i *input, __m128i *output)
{
__m128i xin0, xin1, xin2, xin3, xin4, xin5, xin6, xin7;
__m128i k0, k1, k2, k3, k4, k5, k6, k7, k8, k9;
aes_genkey<SOFT_AES>(input, &k0, &k1, &k2, &k3, &k4, &k5, &k6, &k7, &k8, &k9);
xin0 = _mm_load_si128(input + 4);
xin1 = _mm_load_si128(input + 5);
xin2 = _mm_load_si128(input + 6);
xin3 = _mm_load_si128(input + 7);
xin4 = _mm_load_si128(input + 8);
xin5 = _mm_load_si128(input + 9);
xin6 = _mm_load_si128(input + 10);
xin7 = _mm_load_si128(input + 11);
for (size_t i = 0; i < MEM / sizeof(__m128i); i += 8) {
if (!SOFT_AES) {
aes_round<SOFT_AES>(_mm_setzero_si128(), &xin0, &xin1, &xin2, &xin3, &xin4, &xin5, &xin6, &xin7);
}
aes_round<SOFT_AES>(k0, &xin0, &xin1, &xin2, &xin3, &xin4, &xin5, &xin6, &xin7);
aes_round<SOFT_AES>(k1, &xin0, &xin1, &xin2, &xin3, &xin4, &xin5, &xin6, &xin7);
aes_round<SOFT_AES>(k2, &xin0, &xin1, &xin2, &xin3, &xin4, &xin5, &xin6, &xin7);
aes_round<SOFT_AES>(k3, &xin0, &xin1, &xin2, &xin3, &xin4, &xin5, &xin6, &xin7);
aes_round<SOFT_AES>(k4, &xin0, &xin1, &xin2, &xin3, &xin4, &xin5, &xin6, &xin7);
aes_round<SOFT_AES>(k5, &xin0, &xin1, &xin2, &xin3, &xin4, &xin5, &xin6, &xin7);
aes_round<SOFT_AES>(k6, &xin0, &xin1, &xin2, &xin3, &xin4, &xin5, &xin6, &xin7);
aes_round<SOFT_AES>(k7, &xin0, &xin1, &xin2, &xin3, &xin4, &xin5, &xin6, &xin7);
aes_round<SOFT_AES>(k8, &xin0, &xin1, &xin2, &xin3, &xin4, &xin5, &xin6, &xin7);
if (!SOFT_AES) {
xin0 ^= k9;
xin1 ^= k9;
xin2 ^= k9;
xin3 ^= k9;
xin4 ^= k9;
xin5 ^= k9;
xin6 ^= k9;
xin7 ^= k9;
}
else {
aes_round<SOFT_AES>(k9, &xin0, &xin1, &xin2, &xin3, &xin4, &xin5, &xin6, &xin7);
}
_mm_store_si128(output + i + 0, xin0);
_mm_store_si128(output + i + 1, xin1);
_mm_store_si128(output + i + 2, xin2);
_mm_store_si128(output + i + 3, xin3);
_mm_store_si128(output + i + 4, xin4);
_mm_store_si128(output + i + 5, xin5);
_mm_store_si128(output + i + 6, xin6);
_mm_store_si128(output + i + 7, xin7);
}
}
template<size_t MEM, bool SOFT_AES>
static inline void cn_implode_scratchpad(const __m128i *input, __m128i *output)
{
__m128i xout0, xout1, xout2, xout3, xout4, xout5, xout6, xout7;
__m128i k0, k1, k2, k3, k4, k5, k6, k7, k8, k9;
aes_genkey<SOFT_AES>(output + 2, &k0, &k1, &k2, &k3, &k4, &k5, &k6, &k7, &k8, &k9);
xout0 = _mm_load_si128(output + 4);
xout1 = _mm_load_si128(output + 5);
xout2 = _mm_load_si128(output + 6);
xout3 = _mm_load_si128(output + 7);
xout4 = _mm_load_si128(output + 8);
xout5 = _mm_load_si128(output + 9);
xout6 = _mm_load_si128(output + 10);
xout7 = _mm_load_si128(output + 11);
for (size_t i = 0; i < MEM / sizeof(__m128i); i += 8)
{
xout0 = _mm_xor_si128(_mm_load_si128(input + i + 0), xout0);
xout1 = _mm_xor_si128(_mm_load_si128(input + i + 1), xout1);
xout2 = _mm_xor_si128(_mm_load_si128(input + i + 2), xout2);
xout3 = _mm_xor_si128(_mm_load_si128(input + i + 3), xout3);
xout4 = _mm_xor_si128(_mm_load_si128(input + i + 4), xout4);
xout5 = _mm_xor_si128(_mm_load_si128(input + i + 5), xout5);
xout6 = _mm_xor_si128(_mm_load_si128(input + i + 6), xout6);
xout7 = _mm_xor_si128(_mm_load_si128(input + i + 7), xout7);
if (!SOFT_AES) {
aes_round<SOFT_AES>(_mm_setzero_si128(), &xout0, &xout1, &xout2, &xout3, &xout4, &xout5, &xout6, &xout7);
}
aes_round<SOFT_AES>(k0, &xout0, &xout1, &xout2, &xout3, &xout4, &xout5, &xout6, &xout7);
aes_round<SOFT_AES>(k1, &xout0, &xout1, &xout2, &xout3, &xout4, &xout5, &xout6, &xout7);
aes_round<SOFT_AES>(k2, &xout0, &xout1, &xout2, &xout3, &xout4, &xout5, &xout6, &xout7);
aes_round<SOFT_AES>(k3, &xout0, &xout1, &xout2, &xout3, &xout4, &xout5, &xout6, &xout7);
aes_round<SOFT_AES>(k4, &xout0, &xout1, &xout2, &xout3, &xout4, &xout5, &xout6, &xout7);
aes_round<SOFT_AES>(k5, &xout0, &xout1, &xout2, &xout3, &xout4, &xout5, &xout6, &xout7);
aes_round<SOFT_AES>(k6, &xout0, &xout1, &xout2, &xout3, &xout4, &xout5, &xout6, &xout7);
aes_round<SOFT_AES>(k7, &xout0, &xout1, &xout2, &xout3, &xout4, &xout5, &xout6, &xout7);
aes_round<SOFT_AES>(k8, &xout0, &xout1, &xout2, &xout3, &xout4, &xout5, &xout6, &xout7);
if (!SOFT_AES) {
xout0 ^= k9;
xout1 ^= k9;
xout2 ^= k9;
xout3 ^= k9;
xout4 ^= k9;
xout5 ^= k9;
xout6 ^= k9;
xout7 ^= k9;
}
else {
aes_round<SOFT_AES>(k9, &xout0, &xout1, &xout2, &xout3, &xout4, &xout5, &xout6, &xout7);
}
}
_mm_store_si128(output + 4, xout0);
_mm_store_si128(output + 5, xout1);
_mm_store_si128(output + 6, xout2);
_mm_store_si128(output + 7, xout3);
_mm_store_si128(output + 8, xout4);
_mm_store_si128(output + 9, xout5);
_mm_store_si128(output + 10, xout6);
_mm_store_si128(output + 11, xout7);
}
template<size_t ITERATIONS, size_t MEM, size_t MASK, bool SOFT_AES>
inline void cryptonight_hash(const void *__restrict__ input, size_t size, void *__restrict__ output, cryptonight_ctx *__restrict__ ctx)
{
keccak(static_cast<const uint8_t*>(input), (int) size, ctx->state0, 200);
cn_explode_scratchpad<MEM, SOFT_AES>((__m128i*) ctx->state0, (__m128i*) ctx->memory);
const uint8_t* l0 = ctx->memory;
uint64_t* h0 = reinterpret_cast<uint64_t*>(ctx->state0);
uint64_t al0 = h0[0] ^ h0[4];
uint64_t ah0 = h0[1] ^ h0[5];
__m128i bx0 = _mm_set_epi64x(h0[3] ^ h0[7], h0[2] ^ h0[6]);
uint64_t idx0 = h0[0] ^ h0[4];
for (size_t i = 0; i < ITERATIONS; i++) {
__m128i cx;
if (SOFT_AES) {
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
}
_mm_store_si128((__m128i *) &l0[idx0 & MASK], _mm_xor_si128(bx0, cx));
idx0 = EXTRACT64(cx);
bx0 = cx;
uint64_t hi, lo, cl, ch;
cl = ((uint64_t*) &l0[idx0 & MASK])[0];
ch = ((uint64_t*) &l0[idx0 & MASK])[1];
lo = __umul128(idx0, cl, &hi);
al0 += hi;
ah0 += lo;
((uint64_t*)&l0[idx0 & MASK])[0] = al0;
((uint64_t*)&l0[idx0 & MASK])[1] = ah0;
ah0 ^= ch;
al0 ^= cl;
idx0 = al0;
}
cn_implode_scratchpad<MEM, SOFT_AES>((__m128i*) ctx->memory, (__m128i*) ctx->state0);
keccakf(h0, 24);
extra_hashes[ctx->state0[0] & 3](ctx->state0, 200, static_cast<char*>(output));
}
template<size_t ITERATIONS, size_t MEM, size_t MASK, bool SOFT_AES>
inline void cryptonight_double_hash(const void *__restrict__ input, size_t size, void *__restrict__ output, struct cryptonight_ctx *__restrict__ ctx)
{
keccak((const uint8_t *) input, (int) size, ctx->state0, 200);
keccak((const uint8_t *) input + size, (int) size, ctx->state1, 200);
const uint8_t* l0 = ctx->memory;
const uint8_t* l1 = ctx->memory + MEM;
uint64_t* h0 = reinterpret_cast<uint64_t*>(ctx->state0);
uint64_t* h1 = reinterpret_cast<uint64_t*>(ctx->state1);
cn_explode_scratchpad<MEM, SOFT_AES>((__m128i*) h0, (__m128i*) l0);
cn_explode_scratchpad<MEM, SOFT_AES>((__m128i*) h1, (__m128i*) l1);
uint64_t al0 = h0[0] ^ h0[4];
uint64_t al1 = h1[0] ^ h1[4];
uint64_t ah0 = h0[1] ^ h0[5];
uint64_t ah1 = h1[1] ^ h1[5];
__m128i bx0 = _mm_set_epi64x(h0[3] ^ h0[7], h0[2] ^ h0[6]);
__m128i bx1 = _mm_set_epi64x(h1[3] ^ h1[7], h1[2] ^ h1[6]);
uint64_t idx0 = h0[0] ^ h0[4];
uint64_t idx1 = h1[0] ^ h1[4];
for (size_t i = 0; i < ITERATIONS; i++) {
__m128i cx0, cx1;
if (SOFT_AES) {
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);
# endif
}
_mm_store_si128((__m128i *) &l0[idx0 & MASK], _mm_xor_si128(bx0, cx0));
_mm_store_si128((__m128i *) &l1[idx1 & MASK], _mm_xor_si128(bx1, cx1));
idx0 = EXTRACT64(cx0);
idx1 = EXTRACT64(cx1);
bx0 = cx0;
bx1 = cx1;
uint64_t hi, lo, cl, ch;
cl = ((uint64_t*) &l0[idx0 & MASK])[0];
ch = ((uint64_t*) &l0[idx0 & MASK])[1];
lo = __umul128(idx0, cl, &hi);
al0 += hi;
ah0 += lo;
((uint64_t*) &l0[idx0 & MASK])[0] = al0;
((uint64_t*) &l0[idx0 & MASK])[1] = ah0;
ah0 ^= ch;
al0 ^= cl;
idx0 = al0;
cl = ((uint64_t*) &l1[idx1 & MASK])[0];
ch = ((uint64_t*) &l1[idx1 & MASK])[1];
lo = __umul128(idx1, cl, &hi);
al1 += hi;
ah1 += lo;
((uint64_t*) &l1[idx1 & MASK])[0] = al1;
((uint64_t*) &l1[idx1 & MASK])[1] = ah1;
ah1 ^= ch;
al1 ^= cl;
idx1 = al1;
}
cn_implode_scratchpad<MEM, SOFT_AES>((__m128i*) l0, (__m128i*) h0);
cn_implode_scratchpad<MEM, SOFT_AES>((__m128i*) l1, (__m128i*) h1);
keccakf(h0, 24);
keccakf(h1, 24);
extra_hashes[ctx->state0[0] & 3](ctx->state0, 200, static_cast<char*>(output));
extra_hashes[ctx->state1[0] & 3](ctx->state1, 200, static_cast<char*>(output) + 32);
}
#endif /* __CRYPTONIGHT_ARM_H__ */

57
src/crypto/CryptoNight_p.h → src/crypto/CryptoNight_x86.h
View file

@ -21,8 +21,8 @@
* along with this program. If not, see <http://www.gnu.org/licenses/>. * along with this program. If not, see <http://www.gnu.org/licenses/>.
*/ */
#ifndef __CRYPTONIGHT_P_H__ #ifndef __CRYPTONIGHT_X86_H__
#define __CRYPTONIGHT_P_H__ #define __CRYPTONIGHT_X86_H__
#ifdef __GNUC__ #ifdef __GNUC__
@ -34,6 +34,7 @@
#include "crypto/CryptoNight.h" #include "crypto/CryptoNight.h"
#include "crypto/soft_aes.h"
extern "C" extern "C"
@ -43,9 +44,6 @@ extern "C"
#include "crypto/c_blake256.h" #include "crypto/c_blake256.h"
#include "crypto/c_jh.h" #include "crypto/c_jh.h"
#include "crypto/c_skein.h" #include "crypto/c_skein.h"
__m128i soft_aesenc(__m128i in, __m128i key);
__m128i soft_aeskeygenassist(__m128i key, uint8_t rcon);
} }
@ -151,13 +149,14 @@ static inline void aes_genkey_sub(__m128i* xout0, __m128i* xout2)
} }
static inline void soft_aes_genkey_sub(__m128i* xout0, __m128i* xout2, uint8_t rcon) template<uint8_t rcon>
static inline void soft_aes_genkey_sub(__m128i* xout0, __m128i* xout2)
{ {
__m128i xout1 = soft_aeskeygenassist(*xout2, rcon); __m128i xout1 = soft_aeskeygenassist<rcon>(*xout2);
xout1 = _mm_shuffle_epi32(xout1, 0xFF); // see PSHUFD, set all elems to 4th elem xout1 = _mm_shuffle_epi32(xout1, 0xFF); // see PSHUFD, set all elems to 4th elem
*xout0 = sl_xor(*xout0); *xout0 = sl_xor(*xout0);
*xout0 = _mm_xor_si128(*xout0, xout1); *xout0 = _mm_xor_si128(*xout0, xout1);
xout1 = soft_aeskeygenassist(*xout0, 0x00); xout1 = soft_aeskeygenassist<0x00>(*xout0);
xout1 = _mm_shuffle_epi32(xout1, 0xAA); // see PSHUFD, set all elems to 3rd elem xout1 = _mm_shuffle_epi32(xout1, 0xAA); // see PSHUFD, set all elems to 3rd elem
*xout2 = sl_xor(*xout2); *xout2 = sl_xor(*xout2);
*xout2 = _mm_xor_si128(*xout2, xout1); *xout2 = _mm_xor_si128(*xout2, xout1);
@ -168,23 +167,23 @@ template<bool SOFT_AES>
static inline void aes_genkey(const __m128i* memory, __m128i* k0, __m128i* k1, __m128i* k2, __m128i* k3, __m128i* k4, __m128i* k5, __m128i* k6, __m128i* k7, __m128i* k8, __m128i* k9) static inline void aes_genkey(const __m128i* memory, __m128i* k0, __m128i* k1, __m128i* k2, __m128i* k3, __m128i* k4, __m128i* k5, __m128i* k6, __m128i* k7, __m128i* k8, __m128i* k9)
{ {
__m128i xout0 = _mm_load_si128(memory); __m128i xout0 = _mm_load_si128(memory);
__m128i xout2 = _mm_load_si128(memory +1 ); __m128i xout2 = _mm_load_si128(memory + 1);
*k0 = xout0; *k0 = xout0;
*k1 = xout2; *k1 = xout2;
SOFT_AES ? soft_aes_genkey_sub(&xout0, &xout2, 0x01) : aes_genkey_sub<0x01>(&xout0, &xout2); SOFT_AES ? soft_aes_genkey_sub<0x01>(&xout0, &xout2) : aes_genkey_sub<0x01>(&xout0, &xout2);
*k2 = xout0; *k2 = xout0;
*k3 = xout2; *k3 = xout2;
SOFT_AES ? soft_aes_genkey_sub(&xout0, &xout2, 0x02) : aes_genkey_sub<0x02>(&xout0, &xout2); SOFT_AES ? soft_aes_genkey_sub<0x02>(&xout0, &xout2) : aes_genkey_sub<0x02>(&xout0, &xout2);
*k4 = xout0; *k4 = xout0;
*k5 = xout2; *k5 = xout2;
SOFT_AES ? soft_aes_genkey_sub(&xout0, &xout2, 0x04) : aes_genkey_sub<0x04>(&xout0, &xout2); SOFT_AES ? soft_aes_genkey_sub<0x04>(&xout0, &xout2) : aes_genkey_sub<0x04>(&xout0, &xout2);
*k6 = xout0; *k6 = xout0;
*k7 = xout2; *k7 = xout2;
SOFT_AES ? soft_aes_genkey_sub(&xout0, &xout2, 0x08) : aes_genkey_sub<0x08>(&xout0, &xout2); SOFT_AES ? soft_aes_genkey_sub<0x08>(&xout0, &xout2) : aes_genkey_sub<0x08>(&xout0, &xout2);
*k8 = xout0; *k8 = xout0;
*k9 = xout2; *k9 = xout2;
} }
@ -194,14 +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) 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) { if (SOFT_AES) {
*x0 = soft_aesenc(*x0, key); *x0 = soft_aesenc((uint32_t*)x0, key);
*x1 = soft_aesenc(*x1, key); *x1 = soft_aesenc((uint32_t*)x1, key);
*x2 = soft_aesenc(*x2, key); *x2 = soft_aesenc((uint32_t*)x2, key);
*x3 = soft_aesenc(*x3, key); *x3 = soft_aesenc((uint32_t*)x3, key);
*x4 = soft_aesenc(*x4, key); *x4 = soft_aesenc((uint32_t*)x4, key);
*x5 = soft_aesenc(*x5, key); *x5 = soft_aesenc((uint32_t*)x5, key);
*x6 = soft_aesenc(*x6, key); *x6 = soft_aesenc((uint32_t*)x6, key);
*x7 = soft_aesenc(*x7, key); *x7 = soft_aesenc((uint32_t*)x7, key);
} }
else { else {
*x0 = _mm_aesenc_si128(*x0, key); *x0 = _mm_aesenc_si128(*x0, key);
@ -326,15 +325,14 @@ inline void cryptonight_hash(const void *__restrict__ input, size_t size, void *
for (size_t i = 0; i < ITERATIONS; i++) { for (size_t i = 0; i < ITERATIONS; i++) {
__m128i cx; __m128i cx;
cx = _mm_load_si128((__m128i *) &l0[idx0 & MASK]);
if (SOFT_AES) { 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)); cx = _mm_aesenc_si128(cx, _mm_set_epi64x(ah0, al0));
} }
_mm_store_si128((__m128i *) &l0[idx0 & MASK], _mm_xor_si128(bx0, cx)); _mm_store_si128((__m128i *) &l0[idx0 & MASK], _mm_xor_si128(bx0, cx));
idx0 = EXTRACT64(cx); idx0 = EXTRACT64(cx);
bx0 = cx; bx0 = cx;
@ -388,14 +386,15 @@ inline void cryptonight_double_hash(const void *__restrict__ input, size_t size,
uint64_t idx1 = h1[0] ^ h1[4]; uint64_t idx1 = h1[0] ^ h1[4];
for (size_t i = 0; i < ITERATIONS; i++) { for (size_t i = 0; i < ITERATIONS; i++) {
__m128i cx0 = _mm_load_si128((__m128i *) &l0[idx0 & MASK]); __m128i cx0, cx1;
__m128i cx1 = _mm_load_si128((__m128i *) &l1[idx1 & MASK]);
if (SOFT_AES) { if (SOFT_AES) {
cx0 = soft_aesenc(cx0, _mm_set_epi64x(ah0, al0)); cx0 = soft_aesenc((uint32_t*)&l0[idx0 & MASK], _mm_set_epi64x(ah0, al0));
cx1 = soft_aesenc(cx1, _mm_set_epi64x(ah1, al1)); cx1 = soft_aesenc((uint32_t*)&l1[idx1 & MASK], _mm_set_epi64x(ah1, al1));
} }
else { 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)); cx0 = _mm_aesenc_si128(cx0, _mm_set_epi64x(ah0, al0));
cx1 = _mm_aesenc_si128(cx1, _mm_set_epi64x(ah1, al1)); cx1 = _mm_aesenc_si128(cx1, _mm_set_epi64x(ah1, al1));
} }
@ -449,4 +448,4 @@ inline void cryptonight_double_hash(const void *__restrict__ input, size_t size,
extra_hashes[ctx->state1[0] & 3](ctx->state1, 200, static_cast<char*>(output) + 32); extra_hashes[ctx->state1[0] & 3](ctx->state1, 200, static_cast<char*>(output) + 32);
} }
#endif /* __CRYPTONIGHT_P_H__ */ #endif /* __CRYPTONIGHT_X86_H__ */

1497
src/crypto/SSE2NEON.h Normal file
View file

File diff suppressed because it is too large Load diff

212
src/crypto/soft_aes.c
View file

@ -1,212 +0,0 @@
/*
* This program is free software: you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program. If not, see <http://www.gnu.org/licenses/>.
*
* Additional permission under GNU GPL version 3 section 7
*
* If you modify this Program, or any covered work, by linking or combining
* it with OpenSSL (or a modified version of that library), containing parts
* covered by the terms of OpenSSL License and SSLeay License, the licensors
* of this Program grant you additional permission to convey the resulting work.
*
*/
/*
* The orginal author of this AES implementation is Karl Malbrain.
*/
#ifdef __GNUC__
#include <x86intrin.h>
#else
#include <intrin.h>
#endif // __GNUC__
#include <inttypes.h>
#define TABLE_ALIGN 32
#define WPOLY 0x011b
#define N_COLS 4
#define AES_BLOCK_SIZE 16
#define RC_LENGTH (5 * (AES_BLOCK_SIZE / 4 - 2))
#if defined(_MSC_VER)
#define ALIGN __declspec(align(TABLE_ALIGN))
#elif defined(__GNUC__)
#define ALIGN __attribute__ ((aligned(16)))
#else
#define ALIGN
#endif
#define rf1(r,c) (r)
#define word_in(x,c) (*((uint32_t*)(x)+(c)))
#define word_out(x,c,v) (*((uint32_t*)(x)+(c)) = (v))
#define s(x,c) x[c]
#define si(y,x,c) (s(y,c) = word_in(x, c))
#define so(y,x,c) word_out(y, c, s(x,c))
#define state_in(y,x) si(y,x,0); si(y,x,1); si(y,x,2); si(y,x,3)
#define state_out(y,x) so(y,x,0); so(y,x,1); so(y,x,2); so(y,x,3)
#define round(y,x,k) \
y[0] = (k)[0] ^ (t_fn[0][x[0] & 0xff] ^ t_fn[1][(x[1] >> 8) & 0xff] ^ t_fn[2][(x[2] >> 16) & 0xff] ^ t_fn[3][x[3] >> 24]); \
y[1] = (k)[1] ^ (t_fn[0][x[1] & 0xff] ^ t_fn[1][(x[2] >> 8) & 0xff] ^ t_fn[2][(x[3] >> 16) & 0xff] ^ t_fn[3][x[0] >> 24]); \
y[2] = (k)[2] ^ (t_fn[0][x[2] & 0xff] ^ t_fn[1][(x[3] >> 8) & 0xff] ^ t_fn[2][(x[0] >> 16) & 0xff] ^ t_fn[3][x[1] >> 24]); \
y[3] = (k)[3] ^ (t_fn[0][x[3] & 0xff] ^ t_fn[1][(x[0] >> 8) & 0xff] ^ t_fn[2][(x[1] >> 16) & 0xff] ^ t_fn[3][x[2] >> 24]);
#define to_byte(x) ((x) & 0xff)
#define bval(x,n) to_byte((x) >> (8 * (n)))
#define fwd_var(x,r,c)\
( r == 0 ? ( c == 0 ? s(x,0) : c == 1 ? s(x,1) : c == 2 ? s(x,2) : s(x,3))\
: r == 1 ? ( c == 0 ? s(x,1) : c == 1 ? s(x,2) : c == 2 ? s(x,3) : s(x,0))\
: r == 2 ? ( c == 0 ? s(x,2) : c == 1 ? s(x,3) : c == 2 ? s(x,0) : s(x,1))\
: ( c == 0 ? s(x,3) : c == 1 ? s(x,0) : c == 2 ? s(x,1) : s(x,2)))
#define fwd_rnd(y,x,k,c) (s(y,c) = (k)[c] ^ four_tables(x,t_use(f,n),fwd_var,rf1,c))
#define sb_data(w) {\
w(0x63), w(0x7c), w(0x77), w(0x7b), w(0xf2), w(0x6b), w(0x6f), w(0xc5),\
w(0x30), w(0x01), w(0x67), w(0x2b), w(0xfe), w(0xd7), w(0xab), w(0x76),\
w(0xca), w(0x82), w(0xc9), w(0x7d), w(0xfa), w(0x59), w(0x47), w(0xf0),\
w(0xad), w(0xd4), w(0xa2), w(0xaf), w(0x9c), w(0xa4), w(0x72), w(0xc0),\
w(0xb7), w(0xfd), w(0x93), w(0x26), w(0x36), w(0x3f), w(0xf7), w(0xcc),\
w(0x34), w(0xa5), w(0xe5), w(0xf1), w(0x71), w(0xd8), w(0x31), w(0x15),\
w(0x04), w(0xc7), w(0x23), w(0xc3), w(0x18), w(0x96), w(0x05), w(0x9a),\
w(0x07), w(0x12), w(0x80), w(0xe2), w(0xeb), w(0x27), w(0xb2), w(0x75),\
w(0x09), w(0x83), w(0x2c), w(0x1a), w(0x1b), w(0x6e), w(0x5a), w(0xa0),\
w(0x52), w(0x3b), w(0xd6), w(0xb3), w(0x29), w(0xe3), w(0x2f), w(0x84),\
w(0x53), w(0xd1), w(0x00), w(0xed), w(0x20), w(0xfc), w(0xb1), w(0x5b),\
w(0x6a), w(0xcb), w(0xbe), w(0x39), w(0x4a), w(0x4c), w(0x58), w(0xcf),\
w(0xd0), w(0xef), w(0xaa), w(0xfb), w(0x43), w(0x4d), w(0x33), w(0x85),\
w(0x45), w(0xf9), w(0x02), w(0x7f), w(0x50), w(0x3c), w(0x9f), w(0xa8),\
w(0x51), w(0xa3), w(0x40), w(0x8f), w(0x92), w(0x9d), w(0x38), w(0xf5),\
w(0xbc), w(0xb6), w(0xda), w(0x21), w(0x10), w(0xff), w(0xf3), w(0xd2),\
w(0xcd), w(0x0c), w(0x13), w(0xec), w(0x5f), w(0x97), w(0x44), w(0x17),\
w(0xc4), w(0xa7), w(0x7e), w(0x3d), w(0x64), w(0x5d), w(0x19), w(0x73),\
w(0x60), w(0x81), w(0x4f), w(0xdc), w(0x22), w(0x2a), w(0x90), w(0x88),\
w(0x46), w(0xee), w(0xb8), w(0x14), w(0xde), w(0x5e), w(0x0b), w(0xdb),\
w(0xe0), w(0x32), w(0x3a), w(0x0a), w(0x49), w(0x06), w(0x24), w(0x5c),\
w(0xc2), w(0xd3), w(0xac), w(0x62), w(0x91), w(0x95), w(0xe4), w(0x79),\
w(0xe7), w(0xc8), w(0x37), w(0x6d), w(0x8d), w(0xd5), w(0x4e), w(0xa9),\
w(0x6c), w(0x56), w(0xf4), w(0xea), w(0x65), w(0x7a), w(0xae), w(0x08),\
w(0xba), w(0x78), w(0x25), w(0x2e), w(0x1c), w(0xa6), w(0xb4), w(0xc6),\
w(0xe8), w(0xdd), w(0x74), w(0x1f), w(0x4b), w(0xbd), w(0x8b), w(0x8a),\
w(0x70), w(0x3e), w(0xb5), w(0x66), w(0x48), w(0x03), w(0xf6), w(0x0e),\
w(0x61), w(0x35), w(0x57), w(0xb9), w(0x86), w(0xc1), w(0x1d), w(0x9e),\
w(0xe1), w(0xf8), w(0x98), w(0x11), w(0x69), w(0xd9), w(0x8e), w(0x94),\
w(0x9b), w(0x1e), w(0x87), w(0xe9), w(0xce), w(0x55), w(0x28), w(0xdf),\
w(0x8c), w(0xa1), w(0x89), w(0x0d), w(0xbf), w(0xe6), w(0x42), w(0x68),\
w(0x41), w(0x99), w(0x2d), w(0x0f), w(0xb0), w(0x54), w(0xbb), w(0x16) }
#define rc_data(w) {\
w(0x01), w(0x02), w(0x04), w(0x08), w(0x10),w(0x20), w(0x40), w(0x80),\
w(0x1b), w(0x36) }
#define bytes2word(b0, b1, b2, b3) (((uint32_t)(b3) << 24) | \
((uint32_t)(b2) << 16) | ((uint32_t)(b1) << 8) | (b0))
#define h0(x) (x)
#define w0(p) bytes2word(p, 0, 0, 0)
#define w1(p) bytes2word(0, p, 0, 0)
#define w2(p) bytes2word(0, 0, p, 0)
#define w3(p) bytes2word(0, 0, 0, p)
#define u0(p) bytes2word(f2(p), p, p, f3(p))
#define u1(p) bytes2word(f3(p), f2(p), p, p)
#define u2(p) bytes2word(p, f3(p), f2(p), p)
#define u3(p) bytes2word(p, p, f3(p), f2(p))
#define v0(p) bytes2word(fe(p), f9(p), fd(p), fb(p))
#define v1(p) bytes2word(fb(p), fe(p), f9(p), fd(p))
#define v2(p) bytes2word(fd(p), fb(p), fe(p), f9(p))
#define v3(p) bytes2word(f9(p), fd(p), fb(p), fe(p))
#define f2(x) ((x<<1) ^ (((x>>7) & 1) * WPOLY))
#define f4(x) ((x<<2) ^ (((x>>6) & 1) * WPOLY) ^ (((x>>6) & 2) * WPOLY))
#define f8(x) ((x<<3) ^ (((x>>5) & 1) * WPOLY) ^ (((x>>5) & 2) * WPOLY) ^ (((x>>5) & 4) * WPOLY))
#define f3(x) (f2(x) ^ x)
#define f9(x) (f8(x) ^ x)
#define fb(x) (f8(x) ^ f2(x) ^ x)
#define fd(x) (f8(x) ^ f4(x) ^ x)
#define fe(x) (f8(x) ^ f4(x) ^ f2(x))
#define t_dec(m,n) t_##m##n
#define t_set(m,n) t_##m##n
#define t_use(m,n) t_##m##n
#define d_4(t,n,b,e,f,g,h) ALIGN const t n[4][256] = { b(e), b(f), b(g), b(h) }
#define four_tables(x,tab,vf,rf,c) \
(tab[0][bval(vf(x,0,c),rf(0,c))] \
^ tab[1][bval(vf(x,1,c),rf(1,c))] \
^ tab[2][bval(vf(x,2,c),rf(2,c))] \
^ tab[3][bval(vf(x,3,c),rf(3,c))])
d_4(uint32_t, t_dec(f,n), sb_data, u0, u1, u2, u3);
__m128i soft_aesenc(__m128i in, __m128i key)
{
uint32_t x0, x1, x2, x3;
x0 = _mm_cvtsi128_si32(in);
x1 = _mm_cvtsi128_si32(_mm_shuffle_epi32(in, 0x55));
x2 = _mm_cvtsi128_si32(_mm_shuffle_epi32(in, 0xAA));
x3 = _mm_cvtsi128_si32(_mm_shuffle_epi32(in, 0xFF));
__m128i out = _mm_set_epi32(
(t_fn[0][x3 & 0xff] ^ t_fn[1][(x0 >> 8) & 0xff] ^ t_fn[2][(x1 >> 16) & 0xff] ^ t_fn[3][x2 >> 24]),
(t_fn[0][x2 & 0xff] ^ t_fn[1][(x3 >> 8) & 0xff] ^ t_fn[2][(x0 >> 16) & 0xff] ^ t_fn[3][x1 >> 24]),
(t_fn[0][x1 & 0xff] ^ t_fn[1][(x2 >> 8) & 0xff] ^ t_fn[2][(x3 >> 16) & 0xff] ^ t_fn[3][x0 >> 24]),
(t_fn[0][x0 & 0xff] ^ t_fn[1][(x1 >> 8) & 0xff] ^ t_fn[2][(x2 >> 16) & 0xff] ^ t_fn[3][x3 >> 24]));
return _mm_xor_si128(out, key);
}
uint8_t Sbox[256] = { // forward s-box
0x63, 0x7c, 0x77, 0x7b, 0xf2, 0x6b, 0x6f, 0xc5, 0x30, 0x01, 0x67, 0x2b, 0xfe, 0xd7, 0xab, 0x76,
0xca, 0x82, 0xc9, 0x7d, 0xfa, 0x59, 0x47, 0xf0, 0xad, 0xd4, 0xa2, 0xaf, 0x9c, 0xa4, 0x72, 0xc0,
0xb7, 0xfd, 0x93, 0x26, 0x36, 0x3f, 0xf7, 0xcc, 0x34, 0xa5, 0xe5, 0xf1, 0x71, 0xd8, 0x31, 0x15,
0x04, 0xc7, 0x23, 0xc3, 0x18, 0x96, 0x05, 0x9a, 0x07, 0x12, 0x80, 0xe2, 0xeb, 0x27, 0xb2, 0x75,
0x09, 0x83, 0x2c, 0x1a, 0x1b, 0x6e, 0x5a, 0xa0, 0x52, 0x3b, 0xd6, 0xb3, 0x29, 0xe3, 0x2f, 0x84,
0x53, 0xd1, 0x00, 0xed, 0x20, 0xfc, 0xb1, 0x5b, 0x6a, 0xcb, 0xbe, 0x39, 0x4a, 0x4c, 0x58, 0xcf,
0xd0, 0xef, 0xaa, 0xfb, 0x43, 0x4d, 0x33, 0x85, 0x45, 0xf9, 0x02, 0x7f, 0x50, 0x3c, 0x9f, 0xa8,
0x51, 0xa3, 0x40, 0x8f, 0x92, 0x9d, 0x38, 0xf5, 0xbc, 0xb6, 0xda, 0x21, 0x10, 0xff, 0xf3, 0xd2,
0xcd, 0x0c, 0x13, 0xec, 0x5f, 0x97, 0x44, 0x17, 0xc4, 0xa7, 0x7e, 0x3d, 0x64, 0x5d, 0x19, 0x73,
0x60, 0x81, 0x4f, 0xdc, 0x22, 0x2a, 0x90, 0x88, 0x46, 0xee, 0xb8, 0x14, 0xde, 0x5e, 0x0b, 0xdb,
0xe0, 0x32, 0x3a, 0x0a, 0x49, 0x06, 0x24, 0x5c, 0xc2, 0xd3, 0xac, 0x62, 0x91, 0x95, 0xe4, 0x79,
0xe7, 0xc8, 0x37, 0x6d, 0x8d, 0xd5, 0x4e, 0xa9, 0x6c, 0x56, 0xf4, 0xea, 0x65, 0x7a, 0xae, 0x08,
0xba, 0x78, 0x25, 0x2e, 0x1c, 0xa6, 0xb4, 0xc6, 0xe8, 0xdd, 0x74, 0x1f, 0x4b, 0xbd, 0x8b, 0x8a,
0x70, 0x3e, 0xb5, 0x66, 0x48, 0x03, 0xf6, 0x0e, 0x61, 0x35, 0x57, 0xb9, 0x86, 0xc1, 0x1d, 0x9e,
0xe1, 0xf8, 0x98, 0x11, 0x69, 0xd9, 0x8e, 0x94, 0x9b, 0x1e, 0x87, 0xe9, 0xce, 0x55, 0x28, 0xdf,
0x8c, 0xa1, 0x89, 0x0d, 0xbf, 0xe6, 0x42, 0x68, 0x41, 0x99, 0x2d, 0x0f, 0xb0, 0x54, 0xbb, 0x16};
static inline void sub_word(uint8_t* key)
{
key[0] = Sbox[key[0]];
key[1] = Sbox[key[1]];
key[2] = Sbox[key[2]];
key[3] = Sbox[key[3]];
}
#ifdef __clang__
uint32_t _rotr(uint32_t value, uint32_t amount)
{
return (value >> amount) | (value << ((32 - amount) & 31));
}
#endif
__m128i soft_aeskeygenassist(__m128i key, uint8_t rcon)
{
uint32_t X1 = _mm_cvtsi128_si32(_mm_shuffle_epi32(key, 0x55));
uint32_t X3 = _mm_cvtsi128_si32(_mm_shuffle_epi32(key, 0xFF));
sub_word((uint8_t*)&X1);
sub_word((uint8_t*)&X3);
return _mm_set_epi32(_rotr(X3, 8) ^ rcon, X3,_rotr(X1, 8) ^ rcon, X1);
}

129
src/crypto/soft_aes.h Normal file
View file

@ -0,0 +1,129 @@
/*
* This program is free software: you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program. If not, see <http://www.gnu.org/licenses/>.
*
* Additional permission under GNU GPL version 3 section 7
*
* If you modify this Program, or any covered work, by linking or combining
* it with OpenSSL (or a modified version of that library), containing parts
* covered by the terms of OpenSSL License and SSLeay License, the licensors
* of this Program grant you additional permission to convey the resulting work.
*
*/
/*
* Parts of this file are originally copyright (c) 2014-2017, The Monero Project
*/
#pragma once
#if defined(XMRIG_ARM)
# include "crypto/SSE2NEON.h"
#elif defined(__GNUC__)
# include <x86intrin.h>
#else
# include <intrin.h>
#endif
#include <inttypes.h>
#define saes_data(w) {\
w(0x63), w(0x7c), w(0x77), w(0x7b), w(0xf2), w(0x6b), w(0x6f), w(0xc5),\
w(0x30), w(0x01), w(0x67), w(0x2b), w(0xfe), w(0xd7), w(0xab), w(0x76),\
w(0xca), w(0x82), w(0xc9), w(0x7d), w(0xfa), w(0x59), w(0x47), w(0xf0),\
w(0xad), w(0xd4), w(0xa2), w(0xaf), w(0x9c), w(0xa4), w(0x72), w(0xc0),\
w(0xb7), w(0xfd), w(0x93), w(0x26), w(0x36), w(0x3f), w(0xf7), w(0xcc),\
w(0x34), w(0xa5), w(0xe5), w(0xf1), w(0x71), w(0xd8), w(0x31), w(0x15),\
w(0x04), w(0xc7), w(0x23), w(0xc3), w(0x18), w(0x96), w(0x05), w(0x9a),\
w(0x07), w(0x12), w(0x80), w(0xe2), w(0xeb), w(0x27), w(0xb2), w(0x75),\
w(0x09), w(0x83), w(0x2c), w(0x1a), w(0x1b), w(0x6e), w(0x5a), w(0xa0),\
w(0x52), w(0x3b), w(0xd6), w(0xb3), w(0x29), w(0xe3), w(0x2f), w(0x84),\
w(0x53), w(0xd1), w(0x00), w(0xed), w(0x20), w(0xfc), w(0xb1), w(0x5b),\
w(0x6a), w(0xcb), w(0xbe), w(0x39), w(0x4a), w(0x4c), w(0x58), w(0xcf),\
w(0xd0), w(0xef), w(0xaa), w(0xfb), w(0x43), w(0x4d), w(0x33), w(0x85),\
w(0x45), w(0xf9), w(0x02), w(0x7f), w(0x50), w(0x3c), w(0x9f), w(0xa8),\
w(0x51), w(0xa3), w(0x40), w(0x8f), w(0x92), w(0x9d), w(0x38), w(0xf5),\
w(0xbc), w(0xb6), w(0xda), w(0x21), w(0x10), w(0xff), w(0xf3), w(0xd2),\
w(0xcd), w(0x0c), w(0x13), w(0xec), w(0x5f), w(0x97), w(0x44), w(0x17),\
w(0xc4), w(0xa7), w(0x7e), w(0x3d), w(0x64), w(0x5d), w(0x19), w(0x73),\
w(0x60), w(0x81), w(0x4f), w(0xdc), w(0x22), w(0x2a), w(0x90), w(0x88),\
w(0x46), w(0xee), w(0xb8), w(0x14), w(0xde), w(0x5e), w(0x0b), w(0xdb),\
w(0xe0), w(0x32), w(0x3a), w(0x0a), w(0x49), w(0x06), w(0x24), w(0x5c),\
w(0xc2), w(0xd3), w(0xac), w(0x62), w(0x91), w(0x95), w(0xe4), w(0x79),\
w(0xe7), w(0xc8), w(0x37), w(0x6d), w(0x8d), w(0xd5), w(0x4e), w(0xa9),\
w(0x6c), w(0x56), w(0xf4), w(0xea), w(0x65), w(0x7a), w(0xae), w(0x08),\
w(0xba), w(0x78), w(0x25), w(0x2e), w(0x1c), w(0xa6), w(0xb4), w(0xc6),\
w(0xe8), w(0xdd), w(0x74), w(0x1f), w(0x4b), w(0xbd), w(0x8b), w(0x8a),\
w(0x70), w(0x3e), w(0xb5), w(0x66), w(0x48), w(0x03), w(0xf6), w(0x0e),\
w(0x61), w(0x35), w(0x57), w(0xb9), w(0x86), w(0xc1), w(0x1d), w(0x9e),\
w(0xe1), w(0xf8), w(0x98), w(0x11), w(0x69), w(0xd9), w(0x8e), w(0x94),\
w(0x9b), w(0x1e), w(0x87), w(0xe9), w(0xce), w(0x55), w(0x28), w(0xdf),\
w(0x8c), w(0xa1), w(0x89), w(0x0d), w(0xbf), w(0xe6), w(0x42), w(0x68),\
w(0x41), w(0x99), w(0x2d), w(0x0f), w(0xb0), w(0x54), w(0xbb), w(0x16) }
#define SAES_WPOLY 0x011b
#define saes_b2w(b0, b1, b2, b3) (((uint32_t)(b3) << 24) | \
((uint32_t)(b2) << 16) | ((uint32_t)(b1) << 8) | (b0))
#define saes_f2(x) ((x<<1) ^ (((x>>7) & 1) * SAES_WPOLY))
#define saes_f3(x) (saes_f2(x) ^ x)
#define saes_h0(x) (x)
#define saes_u0(p) saes_b2w(saes_f2(p), p, p, saes_f3(p))
#define saes_u1(p) saes_b2w(saes_f3(p), saes_f2(p), p, p)
#define saes_u2(p) saes_b2w( p, saes_f3(p), saes_f2(p), p)
#define saes_u3(p) saes_b2w( p, p, saes_f3(p), saes_f2(p))
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(const uint32_t* in, __m128i key)
{
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]));
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];
}
#if defined(__clang__) || defined(XMRIG_ARM)
static inline uint32_t _rotr(uint32_t value, uint32_t amount)
{
return (value >> amount) | (value << ((32 - amount) & 31));
}
#endif
template<uint8_t rcon>
static inline __m128i soft_aeskeygenassist(__m128i key)
{
const uint32_t X1 = sub_word(_mm_cvtsi128_si32(_mm_shuffle_epi32(key, 0x55)));
const uint32_t X3 = sub_word(_mm_cvtsi128_si32(_mm_shuffle_epi32(key, 0xFF)));
return _mm_set_epi32(_rotr(X3, 8) ^ rcon, X3, _rotr(X1, 8) ^ rcon, X1);
}

19
src/log/ConsoleLog.cpp
View file

@ -36,6 +36,7 @@
#include "log/ConsoleLog.h" #include "log/ConsoleLog.h"
#include "log/Log.h" #include "log/Log.h"
#include "Options.h"
ConsoleLog::ConsoleLog(bool colors) : ConsoleLog::ConsoleLog(bool colors) :
@ -43,6 +44,8 @@ ConsoleLog::ConsoleLog(bool colors) :
m_stream(nullptr) m_stream(nullptr)
{ {
if (uv_tty_init(uv_default_loop(), &m_tty, 1, 0) < 0) { if (uv_tty_init(uv_default_loop(), &m_tty, 1, 0) < 0) {
Options::i()->setColors(false);
m_colors = false;
return; return;
} }
@ -65,10 +68,6 @@ ConsoleLog::ConsoleLog(bool colors) :
void ConsoleLog::message(int level, const char* fmt, va_list args) void ConsoleLog::message(int level, const char* fmt, va_list args)
{ {
if (!isWritable()) {
return;
}
time_t now = time(nullptr); time_t now = time(nullptr);
tm stime; tm stime;
@ -121,10 +120,6 @@ void ConsoleLog::message(int level, const char* fmt, va_list args)
void ConsoleLog::text(const char* fmt, va_list args) void ConsoleLog::text(const char* fmt, va_list args)
{ {
if (!isWritable()) {
return;
}
snprintf(m_fmt, sizeof(m_fmt) - 1, "%s%s\n", fmt, m_colors ? Log::kCL_N : ""); snprintf(m_fmt, sizeof(m_fmt) - 1, "%s%s\n", fmt, m_colors ? Log::kCL_N : "");
print(args); print(args);
@ -149,5 +144,11 @@ void ConsoleLog::print(va_list args)
return; return;
} }
uv_try_write(m_stream, &m_uvBuf, 1); if (!isWritable()) {
fputs(m_buf, stdout);
fflush(stdout);
}
else {
uv_try_write(m_stream, &m_uvBuf, 1);
}
} }

14
src/net/Client.cpp
View file

@ -220,7 +220,11 @@ bool Client::parseJob(const rapidjson::Value &params, int *code)
} }
if (m_job == job) { if (m_job == job) {
LOG_WARN("[%s:%u] duplicate job received, ignore", m_url.host(), m_url.port()); if (!m_quiet) {
LOG_WARN("[%s:%u] duplicate job received, reconnect", m_url.host(), m_url.port());
}
close();
return false; return false;
} }
@ -528,7 +532,7 @@ void Client::onAllocBuffer(uv_handle_t *handle, size_t suggested_size, uv_buf_t
auto client = getClient(handle->data); auto client = getClient(handle->data);
buf->base = &client->m_recvBuf.base[client->m_recvBufPos]; 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;
} }
@ -578,11 +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)); 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)) { if ((size_t) nread > (sizeof(m_buf) - 8 - client->m_recvBufPos)) {
return client->close();; return client->close();
} }
client->m_recvBufPos += nread; client->m_recvBufPos += nread;
@ -619,7 +623,7 @@ void Client::onResolved(uv_getaddrinfo_t *req, int status, struct addrinfo *res)
auto client = getClient(req->data); auto client = getClient(req->data);
if (status < 0) { if (status < 0) {
LOG_ERR("[%s:%u] DNS error: \"%s\"", client->m_url.host(), client->m_url.port(), uv_strerror(status)); 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; addrinfo *ptr = res;

2
src/net/Job.cpp
View file

@ -171,5 +171,5 @@ void Job::toHex(const unsigned char* in, unsigned int len, char* out)
bool Job::operator==(const Job &other) const bool Job::operator==(const Job &other) const
{ {
return m_id == other.m_id && memcmp(m_blob, other.m_blob, sizeof(m_blob) == 0); return m_id == other.m_id && memcmp(m_blob, other.m_blob, sizeof(m_blob)) == 0;
} }

3
src/net/Job.h
View file

@ -70,6 +70,8 @@ public:
bool operator==(const Job &other) const; bool operator==(const Job &other) const;
private: private:
VAR_ALIGN(16, uint8_t m_blob[84]); // Max blob size is 84 (75 fixed + 9 variable), aligned to 96. https://github.com/xmrig/xmrig/issues/1 Thanks fireice-uk.
bool m_nicehash; bool m_nicehash;
int m_poolId; int m_poolId;
int m_threadId; int m_threadId;
@ -77,7 +79,6 @@ private:
size_t m_size; size_t m_size;
uint64_t m_diff; uint64_t m_diff;
uint64_t m_target; uint64_t m_target;
VAR_ALIGN(16, uint8_t m_blob[84]); // Max blob size is 84 (75 fixed + 9 variable), aligned to 96. https://github.com/xmrig/xmrig/issues/1 Thanks fireice-uk.
# ifdef XMRIG_PROXY_PROJECT # ifdef XMRIG_PROXY_PROJECT
VAR_ALIGN(16, char m_rawBlob[169]); VAR_ALIGN(16, char m_rawBlob[169]);

2
src/net/JobId.h
View file

@ -63,7 +63,7 @@ public:
} }
const size_t size = strlen(id); const size_t size = strlen(id);
if (size < 4 || size >= sizeof(m_data)) { if (size >= sizeof(m_data)) {
return false; return false;
} }

10
src/version.h
View file

@ -4,7 +4,7 @@
* Copyright 2014 Lucas Jones <https://github.com/lucasjones> * Copyright 2014 Lucas Jones <https://github.com/lucasjones>
* Copyright 2014-2016 Wolf9466 <https://github.com/OhGodAPet> * Copyright 2014-2016 Wolf9466 <https://github.com/OhGodAPet>
* Copyright 2016 Jay D Dee <jayddee246@gmail.com> * 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 * This program is free software: you can redistribute it and/or modify
@ -27,19 +27,19 @@
#define APP_ID "xmrig" #define APP_ID "xmrig"
#define APP_NAME "XMRig" #define APP_NAME "XMRig"
#define APP_DESC "XMRig CPU miner" #define APP_DESC "XMRig CPU miner"
#define APP_VERSION "2.4.0-dev" #define APP_VERSION "2.4.5"
#define APP_DOMAIN "xmrig.com" #define APP_DOMAIN "xmrig.com"
#define APP_SITE "www.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_KIND "cpu"
#define APP_VER_MAJOR 2 #define APP_VER_MAJOR 2
#define APP_VER_MINOR 4 #define APP_VER_MINOR 4
#define APP_VER_BUILD 0 #define APP_VER_BUILD 5
#define APP_VER_REV 0 #define APP_VER_REV 0
#ifdef _MSC_VER #ifdef _MSC_VER
# if (_MSC_VER == 1910 || _MSC_VER == 1911) # if (_MSC_VER >= 1910)
# define MSVC_VERSION 2017 # define MSVC_VERSION 2017
# elif _MSC_VER == 1900 # elif _MSC_VER == 1900
# define MSVC_VERSION 2015 # define MSVC_VERSION 2015

9
src/workers/Hashrate.cpp
View file

@ -23,8 +23,9 @@
#include <chrono> #include <chrono>
#include <cmath> #include <math.h>
#include <memory.h> #include <memory.h>
#include <stdio.h>
#include "log/Log.h" #include "log/Log.h"
#include "Options.h" #include "Options.h"
@ -33,7 +34,7 @@
inline const char *format(double h, char* buf, size_t size) inline const char *format(double h, char* buf, size_t size)
{ {
if (std::isnormal(h)) { if (isnormal(h)) {
snprintf(buf, size, "%03.1f", h); snprintf(buf, size, "%03.1f", h);
return buf; return buf;
} }
@ -77,7 +78,7 @@ double Hashrate::calc(size_t ms) const
for (int i = 0; i < m_threads; ++i) { for (int i = 0; i < m_threads; ++i) {
data = calc(i, ms); data = calc(i, ms);
if (std::isnormal(data)) { if (isnormal(data)) {
result += data; result += data;
} }
} }
@ -170,7 +171,7 @@ void Hashrate::stop()
void Hashrate::updateHighest() void Hashrate::updateHighest()
{ {
double highest = calc(ShortInterval); double highest = calc(ShortInterval);
if (std::isnormal(highest) && highest > m_highest) { if (isnormal(highest) && highest > m_highest) {
m_highest = highest; m_highest = highest;
} }
} }