diff options
103 files changed, 1436 insertions, 1079 deletions
diff --git a/CMakeLists.txt b/CMakeLists.txt index 784c0bd..a642b38 100644 --- a/CMakeLists.txt +++ b/CMakeLists.txt @@ -141,7 +141,6 @@ if(CUDA_ENABLE) # set flags to create device code for the given architectures set(CLANG_BUILD_FLAGS "${CLANG_BUILD_FLAGS} --cuda-gpu-arch=sm_${CUDA_ARCH_ELEM}") endforeach() - elseif(CUDA_COMPILER STREQUAL "nvcc") # add c++11 for cuda if(NOT CMAKE_CXX_FLAGS MATCHES "-std=c\\+\\+11") @@ -149,7 +148,7 @@ if(CUDA_ENABLE) endif() # avoid that nvcc in CUDA 8 complains about sm_20 pending removal - if(CUDA_VERSION VERSION_EQUAL 8.0) + if(CUDA_VERSION VERSION_EQUAL 8.0) set(CUDA_NVCC_FLAGS "${CUDA_NVCC_FLAGS} -Wno-deprecated-gpu-targets") endif() @@ -180,14 +179,14 @@ if(CUDA_ENABLE) if(CUDA_KEEP_FILES) set(CUDA_NVCC_FLAGS "${CUDA_NVCC_FLAGS}" --keep --keep-dir "${PROJECT_BINARY_DIR}") endif(CUDA_KEEP_FILES) - + if(CUDA_VERSION VERSION_LESS 8.0) # for CUDA 7.5 fix compile error: https://github.com/fireice-uk/xmr-stak/issues/34 set(CUDA_NVCC_FLAGS "${CUDA_NVCC_FLAGS}" "-D_MWAITXINTRIN_H_INCLUDED") endif() if(CMAKE_CXX_COMPILER_ID MATCHES "MSVC" AND - (CUDA_VERSION VERSION_EQUAL 9.0 OR CUDA_VERSION VERSION_EQUAL 9.1) + (CUDA_VERSION VERSION_EQUAL 9.0 OR CUDA_VERSION VERSION_EQUAL 9.1 OR CUDA_VERSION VERSION_EQUAL 9.2) ) # workaround find_package(CUDA) is using the wrong path to the CXX host compiler # overwrite the CUDA host compiler variable with the used CXX MSVC @@ -232,7 +231,7 @@ if(OpenCL_ENABLE) "AMD APP/include") find_library(OpenCL_LIBRARY - NAMES + NAMES OpenCL OpenCL.lib NO_DEFAULT_PATH @@ -284,7 +283,7 @@ endif() # Find microhttpd ################################################################################ -option(MICROHTTPD_ENABLE "Enable or disable the requirement of microhttp (http deamon)" ON) +option(MICROHTTPD_ENABLE "Enable or disable the requirement of microhttp (http daemon)" ON) if(MICROHTTPD_ENABLE) find_path(MTHD_INCLUDE_DIR NAMES @@ -307,7 +306,7 @@ if(MICROHTTPD_ENABLE) PATH_SUFFIXES lib) if(MHTD STREQUAL "MHTD-NOTFOUND") - message(FATAL_ERROR "microhttpd NOT found: use `-DMICROHTTPD_ENABLE=OFF` to build without http deamon support") + message(FATAL_ERROR "microhttpd NOT found: use `-DMICROHTTPD_ENABLE=OFF` to build without http daemon support") else() set(LIBS ${LIBS} ${MHTD}) include_directories(AFTER ${MTHD_INCLUDE_DIR}) @@ -399,10 +398,10 @@ execute_process( ) if(NOT GIT_COMMIT_HASH STREQUAL "") - add_definitions("-DGIT_COMMIT_HASH=${GIT_COMMIT_HASH}") + add_definitions("-DGIT_COMMIT_HASH=${GIT_COMMIT_HASH}") endif() if(NOT GIT_BRANCH STREQUAL "") - add_definitions("-DGIT_BRANCH=${GIT_BRANCH}") + add_definitions("-DGIT_BRANCH=${GIT_BRANCH}") endif() # generate backend string @@ -426,7 +425,6 @@ else() # activate sse2 and aes-ni set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -msse2 -maes") set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -msse2 -maes") - endif() # add -Wall for debug builds with gcc @@ -478,7 +476,7 @@ target_link_libraries(xmr-stak-backend xmr-stak-c ${CMAKE_DL_LIBS}) # compile CUDA backend if(CUDA_FOUND) - file(GLOB CUDASRCFILES + file(GLOB CUDASRCFILES "xmrstak/backend/nvidia/nvcc_code/*.cu" "xmrstak/backend/nvidia/*.cpp") @@ -506,7 +504,7 @@ endif() # compile AMD backend if(OpenCL_FOUND) - file(GLOB OPENCLSRCFILES + file(GLOB OPENCLSRCFILES "xmrstak/backend/amd/amd_gpu/*.cpp" "xmrstak/backend/amd/*.cpp") add_library(xmrstak_opencl_backend @@ -536,7 +534,6 @@ target_link_libraries(xmr-stak ${LIBS} xmr-stak-c xmr-stak-backend) # Install ################################################################################ - # do not install the binary if the project and install are equal if( NOT CMAKE_INSTALL_PREFIX STREQUAL PROJECT_BINARY_DIR ) install(TARGETS xmr-stak @@ -1,7 +1,7 @@ ###### fireice-uk's and psychocrypt's # XMR-Stak - Monero/Aeon All-in-One Mining Software -**XMR-Stak is ready for the POW change of Monero-v7, Aeon-v7 and Sumukoin-v3** +**XMR-Stak is ready for the POW change of Monero-v7, Aeon-v7, stellite-v4 and Sumukoin-v3** XMR-Stak is a universal Stratum pool miner. This miner supports CPUs, AMD and NVIDIA gpus and can be used to mine the crypto currency Monero and Aeon. @@ -48,8 +48,11 @@ Besides [Monero](https://getmonero.org), following coins can be mined using this - [Graft](https://www.graft.network) - [Haven](https://havenprotocol.com) - [Intense](https://intensecoin.com) +- [IPBC](https://ipbc.io) - [Karbo](https://karbo.io) +- [Masari](https://getmasari.org) - [Sumokoin](https://www.sumokoin.org) +- [TurtleCoin](https://turtlecoin.lol) If your prefered coin is not listed, you can chose one of the following algorithms: diff --git a/THIRD-PARTY-LICENSES b/THIRD-PARTY-LICENSES index d3202f4..284b268 100644 --- a/THIRD-PARTY-LICENSES +++ b/THIRD-PARTY-LICENSES @@ -5,7 +5,7 @@ This application bundles the following third-party software in accordance with t Package: Original NVidia mining code Authors: tsiv and KlausT License: GNU GPLv3 -Notes: Improvements are (c) of Xmr-Stak team +Notes: Improvements are (c) of Xmr-Stak team and are covered by GNU GPLv3 ------------------------------------------------------------------------- @@ -27,3 +27,10 @@ Authors: okdshin License: MIT License ------------------------------------------------------------------------- + +Package: cpputil +Authors: Will Zhang +Source: https://github.com/willzhang4a58/cpputil +License: MIT License + +------------------------------------------------------------------------- @@ -44,10 +44,10 @@ Download and install this [runtime package](https://go.microsoft.com/fwlink/?Lin ## Error: MEMORY ALLOC FAILED: mmap failed -On Linux you will need to configure large page support and increase your ulimit -l. +On Linux you will need to configure large page support and increase your ulimit -l. To set large page support, add the following lines to /etc/sysctl.conf: - + vm.nr_hugepages=128 To increase the ulimit, add following lines to /etc/security/limits.conf: @@ -55,7 +55,7 @@ To increase the ulimit, add following lines to /etc/security/limits.conf: * soft memlock 262144 * hard memlock 262144 -You WILL need to log out and log back in for these settings to take affect on your user (no need to reboot, just relogin in your session). +You WILL need to log out and log back in for these settings to take effect on your user (no need to reboot, just relogin in your session). You can also do it Windows-style and simply run-as-root, but this is NOT recommended for security reasons. diff --git a/doc/compile_FreeBSD.md b/doc/compile_FreeBSD.md index 34db1ca..a4eb341 100644 --- a/doc/compile_FreeBSD.md +++ b/doc/compile_FreeBSD.md @@ -6,7 +6,7 @@ From the root shell, run the following commands: - pkg install git libmicrohttpd hwloc cmake + pkg install git libmicrohttpd hwloc cmake Type 'y' and hit enter to proceed with installing the packages. diff --git a/doc/compile_Linux.md b/doc/compile_Linux.md index 3eb4960..072402f 100644 --- a/doc/compile_Linux.md +++ b/doc/compile_Linux.md @@ -4,7 +4,8 @@ ### AMD APP SDK 3.0 (only needed to use AMD GPUs) -- download and install the latest version from [http://developer.amd.com/amd-accelerated-parallel-processing-app-sdk/](http://developer.amd.com/amd-accelerated-parallel-processing-app-sdk/) +- download and install the latest version from https://www.dropbox.com/sh/mpg882ekirnsfa7/AADWz5X-TgVdsmWt0QwMgTWLa/AMD-APP-SDKInstaller-v3.0.130.136-GA-linux64.tar.bz2?dl=0 + (do not wonder why it is a link to a dropbox but AMD has removed the SDK downloads, see https://community.amd.com/thread/228059) ### Cuda 8.0+ (only needed to use NVIDIA GPUs) @@ -64,7 +65,7 @@ cd xmr-stak/build cmake .. make install - + # TinyCore Linux 8.x # TinyCore is 32-bit only, but there is an x86-64 port, known as "Pure 64," # hosted on the TinyCore home page, and it works well. @@ -94,7 +95,7 @@ make install ``` -- g++ version 5.1 or higher is required for full C++11 support. +- g++ version 5.1 or higher is required for full C++11 support. If you want to compile the binary without installing libraries / compiler or just compile binary for some other distribution, please check the [build_xmr-stak_docker.sh script](scripts/build_xmr-stak_docker/build_xmr-stak_docker.sh). - Some newer gcc versions are not supported by CUDA (e.g. Ubuntu 17.10). It will require installing gcc 5 but you can avoid changing defaults. diff --git a/doc/compile_Windows.md b/doc/compile_Windows.md index 129596c..802d5c5 100644 --- a/doc/compile_Windows.md +++ b/doc/compile_Windows.md @@ -13,7 +13,7 @@ - During install choose following components: - `Desktop development with C++` (left side) - `VC++ 2015.3 v140 toolset for desktop` (right side - **NOT** needed for CUDA 9 or AMD GPU) - - Since release of VS2017 15.5 (12/04/17), require `VC++ 2017 version 15.4 v14.11 toolset` (under tab `Individual Components`, section `Compilers, build tools, and runtimes`), as CUDA 9.1 is not compatible with compiler 14.12.X + - Since release of VS2017 15.5 (12/04/17), require `VC++ 2017 version 15.4 v14.11 toolset` (under tab `Individual Components`, section `Compilers, build tools, and runtimes`), as CUDA 9.x is not compatible with compiler 14.12.X ### CMake for Win64 @@ -32,7 +32,8 @@ ### AMD APP SDK 3.0 (only needed for AMD GPUs) -- Download and install the latest version from http://developer.amd.com/amd-accelerated-parallel-processing-app-sdk/ +- Download and install the latest version from https://www.dropbox.com/s/gq8vqhelq0m6gj4/AMD-APP-SDKInstaller-v3.0.130.135-GA-windows-F-x64.exe + (do not wonder why it is a link to a dropbox but AMD has removed the SDK downloads, see https://community.amd.com/thread/222855) ### Dependencies OpenSSL/Hwloc and Microhttpd - For CUDA 8*: @@ -82,17 +83,17 @@ - Go to extracted source code directory (e.g. `cd C:\Users\USERNAME\xmr-stak-<version>`) - Execute the following commands (NOTE: path to Visual Studio Community 2017 can be different) ``` - # Execute next line only if compiling for Cuda 9.1 and using Visual Studio 2017 >= 15.5 (released 12/04/17) - "C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Auxiliary\Build\vcvarsall.bat" x64 -vcvars_ver=14.11 - + # Execute next line only if compiling for Cuda 9.x and using Visual Studio 2017 >= 15.5 (released 12/04/17) + "C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Auxiliary\Build\vcvarsall.bat" x64 -vcvars_ver=14.11 + "C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\Common7\Tools\VsMSBuildCmd.bat" ``` - Sometimes Windows will change the directory to `C:\Users\USERNAME\source\` instead of `C:\Users\USERNAME\xmr-stak-<version>\`. If that's the case execute `cd C:\Users\USERNAME\xmr-stak-<version>` followed by: ``` mkdir build - + cd build - + set CMAKE_PREFIX_PATH=C:\xmr-stak-dep\hwloc;C:\xmr-stak-dep\libmicrohttpd;C:\xmr-stak-dep\openssl ``` @@ -104,9 +105,9 @@ - Then execute ``` cmake --build . --config Release --target install - + cd bin\Release - + copy C:\xmr-stak-dep\openssl\bin\* . ``` - Miner is by default compiled for NVIDIA GPUs (if CUDA is installed), AMD GPUs (if the AMD APP SDK is installed) and CPUs. diff --git a/doc/tuning.md b/doc/tuning.md index 47ad0bb..6bf036e 100644 --- a/doc/tuning.md +++ b/doc/tuning.md @@ -82,10 +82,10 @@ If you are unsure of either GPU or platform index value, you can use `clinfo` to ``` "gpu_threads_conf" : [ - { "index" : 0, "intensity" : 1000, "worksize" : 8, "affine_to_cpu" : false, + { "index" : 0, "intensity" : 1000, "worksize" : 8, "affine_to_cpu" : false, "strided_index" : true, "mem_chunk" : 2, "comp_mode" : true }, - { "index" : 1, "intensity" : 1000, "worksize" : 8, "affine_to_cpu" : false, + { "index" : 1, "intensity" : 1000, "worksize" : 8, "affine_to_cpu" : false, "strided_index" : true, "mem_chunk" : 2, "comp_mode" : true }, ], diff --git a/xmrstak/backend/amd/amd_gpu/gpu.cpp b/xmrstak/backend/amd/amd_gpu/gpu.cpp index 03100d0..c664d53 100644 --- a/xmrstak/backend/amd/amd_gpu/gpu.cpp +++ b/xmrstak/backend/amd/amd_gpu/gpu.cpp @@ -26,7 +26,7 @@ #include <algorithm> #include <regex> #include <cassert> -#include <algorithm> +#include <algorithm> #include <fstream> #include <sstream> @@ -397,7 +397,7 @@ size_t InitOpenCLGpu(cl_context opencl_ctx, GpuContext* ctx, const char* source_ * used data: * - source code * - device name - * - compile paramater + * - compile parameter */ std::string src_str(source_code); src_str += options; @@ -476,14 +476,14 @@ size_t InitOpenCLGpu(cl_context opencl_ctx, GpuContext* ctx, const char* source_ } while(status == CL_BUILD_IN_PROGRESS); - std::vector<size_t> binary_sizes(num_devices); - clGetProgramInfo (ctx->Program[ii], CL_PROGRAM_BINARY_SIZES, sizeof(size_t) * binary_sizes.size(), binary_sizes.data(), NULL); - - std::vector<char*> all_programs(num_devices); - std::vector<std::vector<char>> program_storage; - if(xmrstak::params::inst().AMDCache) { + std::vector<size_t> binary_sizes(num_devices); + clGetProgramInfo (ctx->Program[ii], CL_PROGRAM_BINARY_SIZES, sizeof(size_t) * binary_sizes.size(), binary_sizes.data(), NULL); + + std::vector<char*> all_programs(num_devices); + std::vector<std::vector<char>> program_storage; + int p_id = 0; size_t mem_size = 0; // create memory structure to query all OpenCL program binaries @@ -935,7 +935,7 @@ size_t InitOpenCL(GpuContext* ctx, size_t num_gpus, size_t platform_idx) size_t XMRSetJob(GpuContext* ctx, uint8_t* input, size_t input_len, uint64_t target, xmrstak_algo miner_algo) { - // switch to the kernel storage + // switch to the kernel storage int kernel_storage = miner_algo == ::jconf::inst()->GetCurrentCoinSelection().GetDescription(1).GetMiningAlgo() ? 0 : 1; cl_int ret; @@ -1004,12 +1004,12 @@ size_t XMRSetJob(GpuContext* ctx, uint8_t* input, size_t input_len, uint64_t tar return(ERR_OCL_API); } - if(miner_algo == cryptonight_monero || miner_algo == cryptonight_aeon ) + if(miner_algo == cryptonight_monero || miner_algo == cryptonight_aeon || miner_algo == cryptonight_ipbc || miner_algo == cryptonight_stellite) { // Input if ((ret = clSetKernelArg(ctx->Kernels[kernel_storage][1], 3, sizeof(cl_mem), &ctx->InputBuffer)) != CL_SUCCESS) { - printer::inst()->print_msg(L1, "Error %s when calling clSetKernelArg for kernel 1, arugment 4(input buffer).", err_to_str(ret)); + printer::inst()->print_msg(L1, "Error %s when calling clSetKernelArg for kernel 1, argument 4(input buffer).", err_to_str(ret)); return ERR_OCL_API; } } @@ -1102,7 +1102,7 @@ size_t XMRRunJob(GpuContext* ctx, cl_uint* HashOutput, xmrstak_algo miner_algo) { // switch to the kernel storage int kernel_storage = miner_algo == ::jconf::inst()->GetCurrentCoinSelection().GetDescription(1).GetMiningAlgo() ? 0 : 1; - + cl_int ret; cl_uint zero = 0; size_t BranchNonces[4]; diff --git a/xmrstak/backend/amd/amd_gpu/opencl/cryptonight.cl b/xmrstak/backend/amd/amd_gpu/opencl/cryptonight.cl index d2ae1a7..c925c87 100644 --- a/xmrstak/backend/amd/amd_gpu/opencl/cryptonight.cl +++ b/xmrstak/backend/amd/amd_gpu/opencl/cryptonight.cl @@ -89,7 +89,7 @@ XMRSTAK_INCLUDE_BLAKE256 //#include "opencl/groestl256.cl" XMRSTAK_INCLUDE_GROESTL256 -static const __constant ulong keccakf_rndc[24] = +static const __constant ulong keccakf_rndc[24] = { 0x0000000000000001, 0x0000000000008082, 0x800000000000808a, 0x8000000080008000, 0x000000000000808b, 0x0000000080000001, @@ -101,7 +101,7 @@ static const __constant ulong keccakf_rndc[24] = 0x8000000000008080, 0x0000000080000001, 0x8000000080008008 }; -static const __constant uchar sbox[256] = +static const __constant uchar sbox[256] = { 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, @@ -124,7 +124,7 @@ static const __constant uchar sbox[256] = void keccakf1600(ulong *s) { - for(int i = 0; i < 24; ++i) + for(int i = 0; i < 24; ++i) { ulong bc[5], tmp1, tmp2; bc[0] = s[0] ^ s[5] ^ s[10] ^ s[15] ^ s[20] ^ rotate(s[2] ^ s[7] ^ s[12] ^ s[17] ^ s[22], 1UL); @@ -132,9 +132,9 @@ void keccakf1600(ulong *s) bc[2] = s[2] ^ s[7] ^ s[12] ^ s[17] ^ s[22] ^ rotate(s[4] ^ s[9] ^ s[14] ^ s[19] ^ s[24], 1UL); bc[3] = s[3] ^ s[8] ^ s[13] ^ s[18] ^ s[23] ^ rotate(s[0] ^ s[5] ^ s[10] ^ s[15] ^ s[20], 1UL); bc[4] = s[4] ^ s[9] ^ s[14] ^ s[19] ^ s[24] ^ rotate(s[1] ^ s[6] ^ s[11] ^ s[16] ^ s[21], 1UL); - + tmp1 = s[1] ^ bc[0]; - + s[0] ^= bc[4]; s[1] = rotate(s[6] ^ bc[0], 44UL); s[6] = rotate(s[9] ^ bc[3], 20UL); @@ -160,7 +160,7 @@ void keccakf1600(ulong *s) s[11] = rotate(s[7] ^ bc[1], 6UL); s[7] = rotate(s[10] ^ bc[4], 3UL); s[10] = rotate(tmp1, 1UL); - + tmp1 = s[0]; tmp2 = s[1]; s[0] = bitselect(s[0] ^ s[2], s[0], s[1]); s[1] = bitselect(s[1] ^ s[3], s[1], s[2]); s[2] = bitselect(s[2] ^ s[4], s[2], s[3]); s[3] = bitselect(s[3] ^ tmp1, s[3], s[4]); s[4] = bitselect(s[4] ^ tmp2, s[4], tmp1); tmp1 = s[5]; tmp2 = s[6]; s[5] = bitselect(s[5] ^ s[7], s[5], s[6]); s[6] = bitselect(s[6] ^ s[8], s[6], s[7]); s[7] = bitselect(s[7] ^ s[9], s[7], s[8]); s[8] = bitselect(s[8] ^ tmp1, s[8], s[9]); s[9] = bitselect(s[9] ^ tmp2, s[9], tmp1); tmp1 = s[10]; tmp2 = s[11]; s[10] = bitselect(s[10] ^ s[12], s[10], s[11]); s[11] = bitselect(s[11] ^ s[13], s[11], s[12]); s[12] = bitselect(s[12] ^ s[14], s[12], s[13]); s[13] = bitselect(s[13] ^ tmp1, s[13], s[14]); s[14] = bitselect(s[14] ^ tmp2, s[14], tmp1); @@ -170,23 +170,23 @@ void keccakf1600(ulong *s) } } -static const __constant uint keccakf_rotc[24] = +static const __constant uint keccakf_rotc[24] = { - 1, 3, 6, 10, 15, 21, 28, 36, 45, 55, 2, 14, + 1, 3, 6, 10, 15, 21, 28, 36, 45, 55, 2, 14, 27, 41, 56, 8, 25, 43, 62, 18, 39, 61, 20, 44 }; -static const __constant uint keccakf_piln[24] = +static const __constant uint keccakf_piln[24] = { - 10, 7, 11, 17, 18, 3, 5, 16, 8, 21, 24, 4, - 15, 23, 19, 13, 12, 2, 20, 14, 22, 9, 6, 1 + 10, 7, 11, 17, 18, 3, 5, 16, 8, 21, 24, 4, + 15, 23, 19, 13, 12, 2, 20, 14, 22, 9, 6, 1 }; void keccakf1600_1(ulong *st) { int i, round; ulong t, bc[5]; - + #pragma unroll 1 for(round = 0; round < 24; ++round) { @@ -197,7 +197,7 @@ void keccakf1600_1(ulong *st) bc[2] = st[2] ^ st[7] ^ st[12] ^ st[17] ^ st[22]; bc[3] = st[3] ^ st[8] ^ st[13] ^ st[18] ^ st[23]; bc[4] = st[4] ^ st[9] ^ st[14] ^ st[19] ^ st[24]; - + #pragma unroll 1 for (i = 0; i < 5; ++i) { t = bc[(i + 4) % 5] ^ rotate(bc[(i + 1) % 5], 1UL); @@ -222,20 +222,20 @@ void keccakf1600_1(ulong *st) //tmp1 = st[10]; tmp2 = st[11]; st[10] = bitselect(st[10] ^ st[12], st[10], st[11]); st[11] = bitselect(st[11] ^ st[13], st[11], st[12]); st[12] = bitselect(st[12] ^ st[14], st[12], st[13]); st[13] = bitselect(st[13] ^ tmp1, st[13], st[14]); st[14] = bitselect(st[14] ^ tmp2, st[14], tmp1); //tmp1 = st[15]; tmp2 = st[16]; st[15] = bitselect(st[15] ^ st[17], st[15], st[16]); st[16] = bitselect(st[16] ^ st[18], st[16], st[17]); st[17] = bitselect(st[17] ^ st[19], st[17], st[18]); st[18] = bitselect(st[18] ^ tmp1, st[18], st[19]); st[19] = bitselect(st[19] ^ tmp2, st[19], tmp1); //tmp1 = st[20]; tmp2 = st[21]; st[20] = bitselect(st[20] ^ st[22], st[20], st[21]); st[21] = bitselect(st[21] ^ st[23], st[21], st[22]); st[22] = bitselect(st[22] ^ st[24], st[22], st[23]); st[23] = bitselect(st[23] ^ tmp1, st[23], st[24]); st[24] = bitselect(st[24] ^ tmp2, st[24], tmp1); - + #pragma unroll 1 for(int i = 0; i < 25; i += 5) - { + { ulong tmp[5]; - + #pragma unroll 1 for(int x = 0; x < 5; ++x) tmp[x] = bitselect(st[i + x] ^ st[i + ((x + 2) % 5)], st[i + x], st[i + ((x + 1) % 5)]); - + #pragma unroll 1 for(int x = 0; x < 5; ++x) st[i + x] = tmp[x]; } - + // Iota st[0] ^= keccakf_rndc[round]; } @@ -246,7 +246,7 @@ void keccakf1600_2(ulong *st) { int i, round; ulong t, bc[5]; - + #pragma unroll 1 for(round = 0; round < 24; ++round) { @@ -257,7 +257,7 @@ void keccakf1600_2(ulong *st) //bc[2] = st[2] ^ st[7] ^ st[12] ^ st[17] ^ st[22]; //bc[3] = st[3] ^ st[8] ^ st[13] ^ st[18] ^ st[23]; //bc[4] = st[4] ^ st[9] ^ st[14] ^ st[19] ^ st[24]; - + /* #pragma unroll for (i = 0; i < 5; ++i) { @@ -269,43 +269,43 @@ void keccakf1600_2(ulong *st) st[i + 20] ^= t; } */ - + bc[0] = st[0] ^ st[5] ^ st[10] ^ st[15] ^ st[20] ^ rotate(st[2] ^ st[7] ^ st[12] ^ st[17] ^ st[22], 1UL); bc[1] = st[1] ^ st[6] ^ st[11] ^ st[16] ^ st[21] ^ rotate(st[3] ^ st[8] ^ st[13] ^ st[18] ^ st[23], 1UL); bc[2] = st[2] ^ st[7] ^ st[12] ^ st[17] ^ st[22] ^ rotate(st[4] ^ st[9] ^ st[14] ^ st[19] ^ st[24], 1UL); bc[3] = st[3] ^ st[8] ^ st[13] ^ st[18] ^ st[23] ^ rotate(st[0] ^ st[5] ^ st[10] ^ st[15] ^ st[20], 1UL); bc[4] = st[4] ^ st[9] ^ st[14] ^ st[19] ^ st[24] ^ rotate(st[1] ^ st[6] ^ st[11] ^ st[16] ^ st[21], 1UL); - + st[0] ^= bc[4]; st[5] ^= bc[4]; st[10] ^= bc[4]; st[15] ^= bc[4]; st[20] ^= bc[4]; - + st[1] ^= bc[0]; st[6] ^= bc[0]; st[11] ^= bc[0]; st[16] ^= bc[0]; st[21] ^= bc[0]; - + st[2] ^= bc[1]; st[7] ^= bc[1]; st[12] ^= bc[1]; st[17] ^= bc[1]; st[22] ^= bc[1]; - + st[3] ^= bc[2]; st[8] ^= bc[2]; st[13] ^= bc[2]; st[18] ^= bc[2]; st[23] ^= bc[2]; - + st[4] ^= bc[3]; st[9] ^= bc[3]; st[14] ^= bc[3]; st[19] ^= bc[3]; st[24] ^= bc[3]; - + // Rho Pi t = st[1]; #pragma unroll @@ -314,11 +314,11 @@ void keccakf1600_2(ulong *st) st[keccakf_piln[i]] = rotate(t, (ulong)keccakf_rotc[i]); t = bc[0]; } - - - + + + /*ulong tmp1 = st[1] ^ bc[0]; - + st[0] ^= bc[4]; st[1] = rotate(st[6] ^ bc[0], 44UL); st[6] = rotate(st[9] ^ bc[3], 20UL); @@ -345,26 +345,26 @@ void keccakf1600_2(ulong *st) st[7] = rotate(st[10] ^ bc[4], 3UL); st[10] = rotate(tmp1, 1UL); */ - - + + //ulong tmp1 = st[0]; ulong tmp2 = st[1]; st[0] = bitselect(st[0] ^ st[2], st[0], st[1]); st[1] = bitselect(st[1] ^ st[3], st[1], st[2]); st[2] = bitselect(st[2] ^ st[4], st[2], st[3]); st[3] = bitselect(st[3] ^ tmp1, st[3], st[4]); st[4] = bitselect(st[4] ^ tmp2, st[4], tmp1); //tmp1 = st[5]; tmp2 = st[6]; st[5] = bitselect(st[5] ^ st[7], st[5], st[6]); st[6] = bitselect(st[6] ^ st[8], st[6], st[7]); st[7] = bitselect(st[7] ^ st[9], st[7], st[8]); st[8] = bitselect(st[8] ^ tmp1, st[8], st[9]); st[9] = bitselect(st[9] ^ tmp2, st[9], tmp1); //tmp1 = st[10]; tmp2 = st[11]; st[10] = bitselect(st[10] ^ st[12], st[10], st[11]); st[11] = bitselect(st[11] ^ st[13], st[11], st[12]); st[12] = bitselect(st[12] ^ st[14], st[12], st[13]); st[13] = bitselect(st[13] ^ tmp1, st[13], st[14]); st[14] = bitselect(st[14] ^ tmp2, st[14], tmp1); //tmp1 = st[15]; tmp2 = st[16]; st[15] = bitselect(st[15] ^ st[17], st[15], st[16]); st[16] = bitselect(st[16] ^ st[18], st[16], st[17]); st[17] = bitselect(st[17] ^ st[19], st[17], st[18]); st[18] = bitselect(st[18] ^ tmp1, st[18], st[19]); st[19] = bitselect(st[19] ^ tmp2, st[19], tmp1); //tmp1 = st[20]; tmp2 = st[21]; st[20] = bitselect(st[20] ^ st[22], st[20], st[21]); st[21] = bitselect(st[21] ^ st[23], st[21], st[22]); st[22] = bitselect(st[22] ^ st[24], st[22], st[23]); st[23] = bitselect(st[23] ^ tmp1, st[23], st[24]); st[24] = bitselect(st[24] ^ tmp2, st[24], tmp1); - + #pragma unroll for(int i = 0; i < 25; i += 5) { ulong tmp1 = st[i], tmp2 = st[i + 1]; - + st[i] = bitselect(st[i] ^ st[i + 2], st[i], st[i + 1]); st[i + 1] = bitselect(st[i + 1] ^ st[i + 3], st[i + 1], st[i + 2]); st[i + 2] = bitselect(st[i + 2] ^ st[i + 4], st[i + 2], st[i + 3]); st[i + 3] = bitselect(st[i + 3] ^ tmp1, st[i + 3], st[i + 4]); st[i + 4] = bitselect(st[i + 4] ^ tmp2, st[i + 4], tmp1); } - + // Iota st[0] ^= keccakf_rndc[round]; } @@ -376,22 +376,22 @@ R"===( void CNKeccak(ulong *output, ulong *input) { ulong st[25]; - + // Copy 72 bytes for(int i = 0; i < 9; ++i) st[i] = input[i]; - + // Last four and '1' bit for padding //st[9] = as_ulong((uint2)(((uint *)input)[18], 0x00000001U)); - + st[9] = (input[9] & 0x00000000FFFFFFFFUL) | 0x0000000100000000UL; - + for(int i = 10; i < 25; ++i) st[i] = 0x00UL; - + // Last bit of padding st[16] = 0x8000000000000000UL; - + keccakf1600_1(st); - + for(int i = 0; i < 25; ++i) output[i] = st[i]; } @@ -408,7 +408,7 @@ void AESExpandKey256(uint *keybuf) { // For 256-bit keys, an sbox permutation is done every other 4th uint generated, AND every 8th uint t = ((!(c & 7)) || ((c & 7) == 4)) ? SubWord(keybuf[c - 1]) : keybuf[c - 1]; - + // If the uint we're generating has an index that is a multiple of 8, rotate and XOR with the round constant, // then XOR this with previously generated uint. If it's 4 after a multiple of 8, only the sbox permutation // is done, followed by the XOR. If neither are true, only the XOR with the previously generated uint is done. @@ -434,7 +434,7 @@ inline ulong getIdx() } #define mix_and_propagate(xin) (xin)[(get_local_id(1)) % 8][get_local_id(0)] ^ (xin)[(get_local_id(1) + 1) % 8][get_local_id(0)] - + #define JOIN_DO(x,y) x##y #define JOIN(x,y) JOIN_DO(x,y) @@ -460,7 +460,7 @@ __kernel void JOIN(cn0,ALGO)(__global ulong *input, __global uint4 *Scratchpad, } barrier(CLK_LOCAL_MEM_FENCE); - + #if(COMP_MODE==1) // do not use early return here if(gIdx < Threads) @@ -512,12 +512,12 @@ __kernel void JOIN(cn0,ALGO)(__global ulong *input, __global uint4 *Scratchpad, } mem_fence(CLK_LOCAL_MEM_FENCE); - + // cryptonight_heavy #if (ALGO == 4) __local uint4 xin[8][WORKSIZE]; - /* Also left over threads performe this loop. + /* Also left over threads perform this loop. * The left over thread results will be ignored */ for(size_t i=0; i < 16; i++) @@ -553,8 +553,8 @@ __kernel void JOIN(cn0,ALGO)(__global ulong *input, __global uint4 *Scratchpad, __attribute__((reqd_work_group_size(WORKSIZE, 1, 1))) __kernel void JOIN(cn1,ALGO) (__global uint4 *Scratchpad, __global ulong *states, ulong Threads -// cryptonight_monero || cryptonight_aeon -#if(ALGO == 3 || ALGO == 5) +// cryptonight_monero || cryptonight_aeon || cryptonight_ipbc || cryptonight_stellite +#if(ALGO == 3 || ALGO == 5 || ALGO == 6 || ALGO == 7) , __global ulong *input #endif ) @@ -574,7 +574,8 @@ __kernel void JOIN(cn1,ALGO) (__global uint4 *Scratchpad, __global ulong *states } barrier(CLK_LOCAL_MEM_FENCE); -#if(ALGO == 3 || ALGO == 5) +// cryptonight_monero || cryptonight_aeon || cryptonight_ipbc || cryptonight_stellite +#if(ALGO == 3 || ALGO == 5 || ALGO == 6 || ALGO == 7) uint2 tweak1_2; #endif uint4 b_x; @@ -598,7 +599,8 @@ __kernel void JOIN(cn1,ALGO) (__global uint4 *Scratchpad, __global ulong *states b[1] = states[3] ^ states[7]; b_x = ((uint4 *)b)[0]; -#if(ALGO == 3 || ALGO == 5) +// cryptonight_monero || cryptonight_aeon || cryptonight_ipbc || cryptonight_stellite +#if(ALGO == 3 || ALGO == 5 || ALGO == 6 || ALGO == 7) tweak1_2 = as_uint2(input[4]); tweak1_2.s0 >>= 24; tweak1_2.s0 |= tweak1_2.s1 << 8; @@ -625,9 +627,15 @@ __kernel void JOIN(cn1,ALGO) (__global uint4 *Scratchpad, __global ulong *states ((uint4 *)c)[0] = AES_Round(AES0, AES1, AES2, AES3, ((uint4 *)c)[0], ((uint4 *)a)[0]); b_x ^= ((uint4 *)c)[0]; -#if(ALGO == 3 || ALGO == 5) +// cryptonight_monero || cryptonight_aeon || cryptonight_ipbc || cryptonight_stellite +#if(ALGO == 3 || ALGO == 5 || ALGO == 6 || ALGO == 7) uint table = 0x75310U; +// cryptonight_stellite +# if(ALGO == 7) + uint index = ((b_x.s2 >> 27) & 12) | ((b_x.s2 >> 23) & 2); +# else uint index = ((b_x.s2 >> 26) & 12) | ((b_x.s2 >> 23) & 2); +# endif b_x.s2 ^= ((table >> index) & 0x30U) << 24; #endif Scratchpad[IDX((idx0 & MASK) >> 4)] = b_x; @@ -638,11 +646,20 @@ __kernel void JOIN(cn1,ALGO) (__global uint4 *Scratchpad, __global ulong *states a[1] += c[0] * as_ulong2(tmp).s0; a[0] += mul_hi(c[0], as_ulong2(tmp).s0); +// cryptonight_monero || cryptonight_aeon || cryptonight_ipbc || cryptonight_stellite +#if(ALGO == 3 || ALGO == 5 || ALGO == 6 || ALGO == 7) -#if(ALGO == 3 || ALGO == 5) +# if(ALGO == 6) + uint2 ipbc_tmp = tweak1_2 ^ ((uint2 *)&(a[0]))[0]; + ((uint2 *)&(a[1]))[0] ^= ipbc_tmp; + Scratchpad[IDX((c[0] & MASK) >> 4)] = ((uint4 *)a)[0]; + ((uint2 *)&(a[1]))[0] ^= ipbc_tmp; +# else ((uint2 *)&(a[1]))[0] ^= tweak1_2; Scratchpad[IDX((c[0] & MASK) >> 4)] = ((uint4 *)a)[0]; ((uint2 *)&(a[1]))[0] ^= tweak1_2; +# endif + #else Scratchpad[IDX((c[0] & MASK) >> 4)] = ((uint4 *)a)[0]; #endif @@ -671,7 +688,7 @@ __kernel void JOIN(cn2,ALGO) (__global uint4 *Scratchpad, __global ulong *states uint ExpandedKey2[40]; ulong State[25]; uint4 text; - + const ulong gIdx = getIdx(); for(int i = get_local_id(1) * WORKSIZE + get_local_id(0); @@ -759,7 +776,7 @@ __kernel void JOIN(cn2,ALGO) (__global uint4 *Scratchpad, __global ulong *states barrier(CLK_LOCAL_MEM_FENCE); text = mix_and_propagate(xin); } - + #else #pragma unroll 2 for(int i = 0; i < (MEMORY >> 7); ++i) @@ -835,7 +852,7 @@ R"===( __kernel void Skein(__global ulong *states, __global uint *BranchBuf, __global uint *output, ulong Target, ulong Threads) { const ulong idx = get_global_id(0) - get_global_offset(0); - + // do not use early return here if(idx < Threads) { @@ -887,7 +904,7 @@ __kernel void Skein(__global ulong *states, __global uint *BranchBuf, __global u output[outIdx] = BranchBuf[idx] + get_global_offset(0); } } - mem_fence(CLK_GLOBAL_MEM_FENCE); + mem_fence(CLK_GLOBAL_MEM_FENCE); } #define SWAP8(x) as_ulong(as_uchar8(x).s76543210) @@ -916,7 +933,7 @@ __kernel void Skein(__global ulong *states, __global uint *BranchBuf, __global u __kernel void JH(__global ulong *states, __global uint *BranchBuf, __global uint *output, ulong Target, ulong Threads) { const uint idx = get_global_id(0) - get_global_offset(0); - + // do not use early return here if(idx < Threads) { @@ -970,12 +987,12 @@ __kernel void JH(__global ulong *states, __global uint *BranchBuf, __global uint __kernel void Blake(__global ulong *states, __global uint *BranchBuf, __global uint *output, ulong Target, ulong Threads) { const uint idx = get_global_id(0) - get_global_offset(0); - + // do not use early return here if(idx < Threads) { states += 25 * BranchBuf[idx]; - + unsigned int m[16]; unsigned int v[16]; uint h[8]; @@ -1046,7 +1063,7 @@ __kernel void Blake(__global ulong *states, __global uint *BranchBuf, __global u __kernel void Groestl(__global ulong *states, __global uint *BranchBuf, __global uint *output, ulong Target, ulong Threads) { const uint idx = get_global_id(0) - get_global_offset(0); - + // do not use early return here if(idx < Threads) { diff --git a/xmrstak/backend/amd/amd_gpu/opencl/groestl256.cl b/xmrstak/backend/amd/amd_gpu/opencl/groestl256.cl index 1a7c96f..2260385 100644 --- a/xmrstak/backend/amd/amd_gpu/opencl/groestl256.cl +++ b/xmrstak/backend/amd/amd_gpu/opencl/groestl256.cl @@ -58,69 +58,69 @@ R"===( static const __constant ulong T0_G[] = { - 0xc6a597f4a5f432c6UL, 0xf884eb9784976ff8UL, 0xee99c7b099b05eeeUL, 0xf68df78c8d8c7af6UL, - 0xff0de5170d17e8ffUL, 0xd6bdb7dcbddc0ad6UL, 0xdeb1a7c8b1c816deUL, 0x915439fc54fc6d91UL, - 0x6050c0f050f09060UL, 0x0203040503050702UL, 0xcea987e0a9e02eceUL, 0x567dac877d87d156UL, - 0xe719d52b192bcce7UL, 0xb56271a662a613b5UL, 0x4de69a31e6317c4dUL, 0xec9ac3b59ab559ecUL, - 0x8f4505cf45cf408fUL, 0x1f9d3ebc9dbca31fUL, 0x894009c040c04989UL, 0xfa87ef92879268faUL, - 0xef15c53f153fd0efUL, 0xb2eb7f26eb2694b2UL, 0x8ec90740c940ce8eUL, 0xfb0bed1d0b1de6fbUL, - 0x41ec822fec2f6e41UL, 0xb3677da967a91ab3UL, 0x5ffdbe1cfd1c435fUL, 0x45ea8a25ea256045UL, - 0x23bf46dabfdaf923UL, 0x53f7a602f7025153UL, 0xe496d3a196a145e4UL, 0x9b5b2ded5bed769bUL, - 0x75c2ea5dc25d2875UL, 0xe11cd9241c24c5e1UL, 0x3dae7ae9aee9d43dUL, 0x4c6a98be6abef24cUL, - 0x6c5ad8ee5aee826cUL, 0x7e41fcc341c3bd7eUL, 0xf502f1060206f3f5UL, 0x834f1dd14fd15283UL, - 0x685cd0e45ce48c68UL, 0x51f4a207f4075651UL, 0xd134b95c345c8dd1UL, 0xf908e9180818e1f9UL, - 0xe293dfae93ae4ce2UL, 0xab734d9573953eabUL, 0x6253c4f553f59762UL, 0x2a3f54413f416b2aUL, - 0x080c10140c141c08UL, 0x955231f652f66395UL, 0x46658caf65afe946UL, 0x9d5e21e25ee27f9dUL, - 0x3028607828784830UL, 0x37a16ef8a1f8cf37UL, 0x0a0f14110f111b0aUL, 0x2fb55ec4b5c4eb2fUL, - 0x0e091c1b091b150eUL, 0x2436485a365a7e24UL, 0x1b9b36b69bb6ad1bUL, 0xdf3da5473d4798dfUL, - 0xcd26816a266aa7cdUL, 0x4e699cbb69bbf54eUL, 0x7fcdfe4ccd4c337fUL, 0xea9fcfba9fba50eaUL, - 0x121b242d1b2d3f12UL, 0x1d9e3ab99eb9a41dUL, 0x5874b09c749cc458UL, 0x342e68722e724634UL, - 0x362d6c772d774136UL, 0xdcb2a3cdb2cd11dcUL, 0xb4ee7329ee299db4UL, 0x5bfbb616fb164d5bUL, - 0xa4f65301f601a5a4UL, 0x764decd74dd7a176UL, 0xb76175a361a314b7UL, 0x7dcefa49ce49347dUL, - 0x527ba48d7b8ddf52UL, 0xdd3ea1423e429fddUL, 0x5e71bc937193cd5eUL, 0x139726a297a2b113UL, - 0xa6f55704f504a2a6UL, 0xb96869b868b801b9UL, 0x0000000000000000UL, 0xc12c99742c74b5c1UL, - 0x406080a060a0e040UL, 0xe31fdd211f21c2e3UL, 0x79c8f243c8433a79UL, 0xb6ed772ced2c9ab6UL, - 0xd4beb3d9bed90dd4UL, 0x8d4601ca46ca478dUL, 0x67d9ce70d9701767UL, 0x724be4dd4bddaf72UL, - 0x94de3379de79ed94UL, 0x98d42b67d467ff98UL, 0xb0e87b23e82393b0UL, 0x854a11de4ade5b85UL, - 0xbb6b6dbd6bbd06bbUL, 0xc52a917e2a7ebbc5UL, 0x4fe59e34e5347b4fUL, 0xed16c13a163ad7edUL, - 0x86c51754c554d286UL, 0x9ad72f62d762f89aUL, 0x6655ccff55ff9966UL, 0x119422a794a7b611UL, - 0x8acf0f4acf4ac08aUL, 0xe910c9301030d9e9UL, 0x0406080a060a0e04UL, 0xfe81e798819866feUL, - 0xa0f05b0bf00baba0UL, 0x7844f0cc44ccb478UL, 0x25ba4ad5bad5f025UL, 0x4be3963ee33e754bUL, - 0xa2f35f0ef30eaca2UL, 0x5dfeba19fe19445dUL, 0x80c01b5bc05bdb80UL, 0x058a0a858a858005UL, - 0x3fad7eecadecd33fUL, 0x21bc42dfbcdffe21UL, 0x7048e0d848d8a870UL, 0xf104f90c040cfdf1UL, - 0x63dfc67adf7a1963UL, 0x77c1ee58c1582f77UL, 0xaf75459f759f30afUL, 0x426384a563a5e742UL, - 0x2030405030507020UL, 0xe51ad12e1a2ecbe5UL, 0xfd0ee1120e12effdUL, 0xbf6d65b76db708bfUL, - 0x814c19d44cd45581UL, 0x1814303c143c2418UL, 0x26354c5f355f7926UL, 0xc32f9d712f71b2c3UL, - 0xbee16738e13886beUL, 0x35a26afda2fdc835UL, 0x88cc0b4fcc4fc788UL, 0x2e395c4b394b652eUL, - 0x93573df957f96a93UL, 0x55f2aa0df20d5855UL, 0xfc82e39d829d61fcUL, 0x7a47f4c947c9b37aUL, - 0xc8ac8befacef27c8UL, 0xbae76f32e73288baUL, 0x322b647d2b7d4f32UL, 0xe695d7a495a442e6UL, - 0xc0a09bfba0fb3bc0UL, 0x199832b398b3aa19UL, 0x9ed12768d168f69eUL, 0xa37f5d817f8122a3UL, - 0x446688aa66aaee44UL, 0x547ea8827e82d654UL, 0x3bab76e6abe6dd3bUL, 0x0b83169e839e950bUL, - 0x8cca0345ca45c98cUL, 0xc729957b297bbcc7UL, 0x6bd3d66ed36e056bUL, 0x283c50443c446c28UL, - 0xa779558b798b2ca7UL, 0xbce2633de23d81bcUL, 0x161d2c271d273116UL, 0xad76419a769a37adUL, - 0xdb3bad4d3b4d96dbUL, 0x6456c8fa56fa9e64UL, 0x744ee8d24ed2a674UL, 0x141e28221e223614UL, - 0x92db3f76db76e492UL, 0x0c0a181e0a1e120cUL, 0x486c90b46cb4fc48UL, 0xb8e46b37e4378fb8UL, - 0x9f5d25e75de7789fUL, 0xbd6e61b26eb20fbdUL, 0x43ef862aef2a6943UL, 0xc4a693f1a6f135c4UL, - 0x39a872e3a8e3da39UL, 0x31a462f7a4f7c631UL, 0xd337bd5937598ad3UL, 0xf28bff868b8674f2UL, - 0xd532b156325683d5UL, 0x8b430dc543c54e8bUL, 0x6e59dceb59eb856eUL, 0xdab7afc2b7c218daUL, - 0x018c028f8c8f8e01UL, 0xb16479ac64ac1db1UL, 0x9cd2236dd26df19cUL, 0x49e0923be03b7249UL, - 0xd8b4abc7b4c71fd8UL, 0xacfa4315fa15b9acUL, 0xf307fd090709faf3UL, 0xcf25856f256fa0cfUL, - 0xcaaf8feaafea20caUL, 0xf48ef3898e897df4UL, 0x47e98e20e9206747UL, 0x1018202818283810UL, - 0x6fd5de64d5640b6fUL, 0xf088fb83888373f0UL, 0x4a6f94b16fb1fb4aUL, 0x5c72b8967296ca5cUL, - 0x3824706c246c5438UL, 0x57f1ae08f1085f57UL, 0x73c7e652c7522173UL, 0x975135f351f36497UL, - 0xcb238d652365aecbUL, 0xa17c59847c8425a1UL, 0xe89ccbbf9cbf57e8UL, 0x3e217c6321635d3eUL, - 0x96dd377cdd7cea96UL, 0x61dcc27fdc7f1e61UL, 0x0d861a9186919c0dUL, 0x0f851e9485949b0fUL, - 0xe090dbab90ab4be0UL, 0x7c42f8c642c6ba7cUL, 0x71c4e257c4572671UL, 0xccaa83e5aae529ccUL, - 0x90d83b73d873e390UL, 0x06050c0f050f0906UL, 0xf701f5030103f4f7UL, 0x1c12383612362a1cUL, - 0xc2a39ffea3fe3cc2UL, 0x6a5fd4e15fe18b6aUL, 0xaef94710f910beaeUL, 0x69d0d26bd06b0269UL, - 0x17912ea891a8bf17UL, 0x995829e858e87199UL, 0x3a2774692769533aUL, 0x27b94ed0b9d0f727UL, - 0xd938a948384891d9UL, 0xeb13cd351335deebUL, 0x2bb356ceb3cee52bUL, 0x2233445533557722UL, - 0xd2bbbfd6bbd604d2UL, 0xa9704990709039a9UL, 0x07890e8089808707UL, 0x33a766f2a7f2c133UL, - 0x2db65ac1b6c1ec2dUL, 0x3c22786622665a3cUL, 0x15922aad92adb815UL, 0xc92089602060a9c9UL, - 0x874915db49db5c87UL, 0xaaff4f1aff1ab0aaUL, 0x5078a0887888d850UL, 0xa57a518e7a8e2ba5UL, - 0x038f068a8f8a8903UL, 0x59f8b213f8134a59UL, 0x0980129b809b9209UL, 0x1a1734391739231aUL, - 0x65daca75da751065UL, 0xd731b553315384d7UL, 0x84c61351c651d584UL, 0xd0b8bbd3b8d303d0UL, - 0x82c31f5ec35edc82UL, 0x29b052cbb0cbe229UL, 0x5a77b4997799c35aUL, 0x1e113c3311332d1eUL, + 0xc6a597f4a5f432c6UL, 0xf884eb9784976ff8UL, 0xee99c7b099b05eeeUL, 0xf68df78c8d8c7af6UL, + 0xff0de5170d17e8ffUL, 0xd6bdb7dcbddc0ad6UL, 0xdeb1a7c8b1c816deUL, 0x915439fc54fc6d91UL, + 0x6050c0f050f09060UL, 0x0203040503050702UL, 0xcea987e0a9e02eceUL, 0x567dac877d87d156UL, + 0xe719d52b192bcce7UL, 0xb56271a662a613b5UL, 0x4de69a31e6317c4dUL, 0xec9ac3b59ab559ecUL, + 0x8f4505cf45cf408fUL, 0x1f9d3ebc9dbca31fUL, 0x894009c040c04989UL, 0xfa87ef92879268faUL, + 0xef15c53f153fd0efUL, 0xb2eb7f26eb2694b2UL, 0x8ec90740c940ce8eUL, 0xfb0bed1d0b1de6fbUL, + 0x41ec822fec2f6e41UL, 0xb3677da967a91ab3UL, 0x5ffdbe1cfd1c435fUL, 0x45ea8a25ea256045UL, + 0x23bf46dabfdaf923UL, 0x53f7a602f7025153UL, 0xe496d3a196a145e4UL, 0x9b5b2ded5bed769bUL, + 0x75c2ea5dc25d2875UL, 0xe11cd9241c24c5e1UL, 0x3dae7ae9aee9d43dUL, 0x4c6a98be6abef24cUL, + 0x6c5ad8ee5aee826cUL, 0x7e41fcc341c3bd7eUL, 0xf502f1060206f3f5UL, 0x834f1dd14fd15283UL, + 0x685cd0e45ce48c68UL, 0x51f4a207f4075651UL, 0xd134b95c345c8dd1UL, 0xf908e9180818e1f9UL, + 0xe293dfae93ae4ce2UL, 0xab734d9573953eabUL, 0x6253c4f553f59762UL, 0x2a3f54413f416b2aUL, + 0x080c10140c141c08UL, 0x955231f652f66395UL, 0x46658caf65afe946UL, 0x9d5e21e25ee27f9dUL, + 0x3028607828784830UL, 0x37a16ef8a1f8cf37UL, 0x0a0f14110f111b0aUL, 0x2fb55ec4b5c4eb2fUL, + 0x0e091c1b091b150eUL, 0x2436485a365a7e24UL, 0x1b9b36b69bb6ad1bUL, 0xdf3da5473d4798dfUL, + 0xcd26816a266aa7cdUL, 0x4e699cbb69bbf54eUL, 0x7fcdfe4ccd4c337fUL, 0xea9fcfba9fba50eaUL, + 0x121b242d1b2d3f12UL, 0x1d9e3ab99eb9a41dUL, 0x5874b09c749cc458UL, 0x342e68722e724634UL, + 0x362d6c772d774136UL, 0xdcb2a3cdb2cd11dcUL, 0xb4ee7329ee299db4UL, 0x5bfbb616fb164d5bUL, + 0xa4f65301f601a5a4UL, 0x764decd74dd7a176UL, 0xb76175a361a314b7UL, 0x7dcefa49ce49347dUL, + 0x527ba48d7b8ddf52UL, 0xdd3ea1423e429fddUL, 0x5e71bc937193cd5eUL, 0x139726a297a2b113UL, + 0xa6f55704f504a2a6UL, 0xb96869b868b801b9UL, 0x0000000000000000UL, 0xc12c99742c74b5c1UL, + 0x406080a060a0e040UL, 0xe31fdd211f21c2e3UL, 0x79c8f243c8433a79UL, 0xb6ed772ced2c9ab6UL, + 0xd4beb3d9bed90dd4UL, 0x8d4601ca46ca478dUL, 0x67d9ce70d9701767UL, 0x724be4dd4bddaf72UL, + 0x94de3379de79ed94UL, 0x98d42b67d467ff98UL, 0xb0e87b23e82393b0UL, 0x854a11de4ade5b85UL, + 0xbb6b6dbd6bbd06bbUL, 0xc52a917e2a7ebbc5UL, 0x4fe59e34e5347b4fUL, 0xed16c13a163ad7edUL, + 0x86c51754c554d286UL, 0x9ad72f62d762f89aUL, 0x6655ccff55ff9966UL, 0x119422a794a7b611UL, + 0x8acf0f4acf4ac08aUL, 0xe910c9301030d9e9UL, 0x0406080a060a0e04UL, 0xfe81e798819866feUL, + 0xa0f05b0bf00baba0UL, 0x7844f0cc44ccb478UL, 0x25ba4ad5bad5f025UL, 0x4be3963ee33e754bUL, + 0xa2f35f0ef30eaca2UL, 0x5dfeba19fe19445dUL, 0x80c01b5bc05bdb80UL, 0x058a0a858a858005UL, + 0x3fad7eecadecd33fUL, 0x21bc42dfbcdffe21UL, 0x7048e0d848d8a870UL, 0xf104f90c040cfdf1UL, + 0x63dfc67adf7a1963UL, 0x77c1ee58c1582f77UL, 0xaf75459f759f30afUL, 0x426384a563a5e742UL, + 0x2030405030507020UL, 0xe51ad12e1a2ecbe5UL, 0xfd0ee1120e12effdUL, 0xbf6d65b76db708bfUL, + 0x814c19d44cd45581UL, 0x1814303c143c2418UL, 0x26354c5f355f7926UL, 0xc32f9d712f71b2c3UL, + 0xbee16738e13886beUL, 0x35a26afda2fdc835UL, 0x88cc0b4fcc4fc788UL, 0x2e395c4b394b652eUL, + 0x93573df957f96a93UL, 0x55f2aa0df20d5855UL, 0xfc82e39d829d61fcUL, 0x7a47f4c947c9b37aUL, + 0xc8ac8befacef27c8UL, 0xbae76f32e73288baUL, 0x322b647d2b7d4f32UL, 0xe695d7a495a442e6UL, + 0xc0a09bfba0fb3bc0UL, 0x199832b398b3aa19UL, 0x9ed12768d168f69eUL, 0xa37f5d817f8122a3UL, + 0x446688aa66aaee44UL, 0x547ea8827e82d654UL, 0x3bab76e6abe6dd3bUL, 0x0b83169e839e950bUL, + 0x8cca0345ca45c98cUL, 0xc729957b297bbcc7UL, 0x6bd3d66ed36e056bUL, 0x283c50443c446c28UL, + 0xa779558b798b2ca7UL, 0xbce2633de23d81bcUL, 0x161d2c271d273116UL, 0xad76419a769a37adUL, + 0xdb3bad4d3b4d96dbUL, 0x6456c8fa56fa9e64UL, 0x744ee8d24ed2a674UL, 0x141e28221e223614UL, + 0x92db3f76db76e492UL, 0x0c0a181e0a1e120cUL, 0x486c90b46cb4fc48UL, 0xb8e46b37e4378fb8UL, + 0x9f5d25e75de7789fUL, 0xbd6e61b26eb20fbdUL, 0x43ef862aef2a6943UL, 0xc4a693f1a6f135c4UL, + 0x39a872e3a8e3da39UL, 0x31a462f7a4f7c631UL, 0xd337bd5937598ad3UL, 0xf28bff868b8674f2UL, + 0xd532b156325683d5UL, 0x8b430dc543c54e8bUL, 0x6e59dceb59eb856eUL, 0xdab7afc2b7c218daUL, + 0x018c028f8c8f8e01UL, 0xb16479ac64ac1db1UL, 0x9cd2236dd26df19cUL, 0x49e0923be03b7249UL, + 0xd8b4abc7b4c71fd8UL, 0xacfa4315fa15b9acUL, 0xf307fd090709faf3UL, 0xcf25856f256fa0cfUL, + 0xcaaf8feaafea20caUL, 0xf48ef3898e897df4UL, 0x47e98e20e9206747UL, 0x1018202818283810UL, + 0x6fd5de64d5640b6fUL, 0xf088fb83888373f0UL, 0x4a6f94b16fb1fb4aUL, 0x5c72b8967296ca5cUL, + 0x3824706c246c5438UL, 0x57f1ae08f1085f57UL, 0x73c7e652c7522173UL, 0x975135f351f36497UL, + 0xcb238d652365aecbUL, 0xa17c59847c8425a1UL, 0xe89ccbbf9cbf57e8UL, 0x3e217c6321635d3eUL, + 0x96dd377cdd7cea96UL, 0x61dcc27fdc7f1e61UL, 0x0d861a9186919c0dUL, 0x0f851e9485949b0fUL, + 0xe090dbab90ab4be0UL, 0x7c42f8c642c6ba7cUL, 0x71c4e257c4572671UL, 0xccaa83e5aae529ccUL, + 0x90d83b73d873e390UL, 0x06050c0f050f0906UL, 0xf701f5030103f4f7UL, 0x1c12383612362a1cUL, + 0xc2a39ffea3fe3cc2UL, 0x6a5fd4e15fe18b6aUL, 0xaef94710f910beaeUL, 0x69d0d26bd06b0269UL, + 0x17912ea891a8bf17UL, 0x995829e858e87199UL, 0x3a2774692769533aUL, 0x27b94ed0b9d0f727UL, + 0xd938a948384891d9UL, 0xeb13cd351335deebUL, 0x2bb356ceb3cee52bUL, 0x2233445533557722UL, + 0xd2bbbfd6bbd604d2UL, 0xa9704990709039a9UL, 0x07890e8089808707UL, 0x33a766f2a7f2c133UL, + 0x2db65ac1b6c1ec2dUL, 0x3c22786622665a3cUL, 0x15922aad92adb815UL, 0xc92089602060a9c9UL, + 0x874915db49db5c87UL, 0xaaff4f1aff1ab0aaUL, 0x5078a0887888d850UL, 0xa57a518e7a8e2ba5UL, + 0x038f068a8f8a8903UL, 0x59f8b213f8134a59UL, 0x0980129b809b9209UL, 0x1a1734391739231aUL, + 0x65daca75da751065UL, 0xd731b553315384d7UL, 0x84c61351c651d584UL, 0xd0b8bbd3b8d303d0UL, + 0x82c31f5ec35edc82UL, 0x29b052cbb0cbe229UL, 0x5a77b4997799c35aUL, 0x1e113c3311332d1eUL, 0x7bcbf646cb463d7bUL, 0xa8fc4b1ffc1fb7a8UL, 0x6dd6da61d6610c6dUL, 0x2c3a584e3a4e622cUL }; @@ -129,69 +129,69 @@ R"===( static const __constant ulong T4_G[] = { - 0xA5F432C6C6A597F4UL, 0x84976FF8F884EB97UL, 0x99B05EEEEE99C7B0UL, 0x8D8C7AF6F68DF78CUL, - 0x0D17E8FFFF0DE517UL, 0xBDDC0AD6D6BDB7DCUL, 0xB1C816DEDEB1A7C8UL, 0x54FC6D91915439FCUL, - 0x50F090606050C0F0UL, 0x0305070202030405UL, 0xA9E02ECECEA987E0UL, 0x7D87D156567DAC87UL, - 0x192BCCE7E719D52BUL, 0x62A613B5B56271A6UL, 0xE6317C4D4DE69A31UL, 0x9AB559ECEC9AC3B5UL, - 0x45CF408F8F4505CFUL, 0x9DBCA31F1F9D3EBCUL, 0x40C04989894009C0UL, 0x879268FAFA87EF92UL, - 0x153FD0EFEF15C53FUL, 0xEB2694B2B2EB7F26UL, 0xC940CE8E8EC90740UL, 0x0B1DE6FBFB0BED1DUL, - 0xEC2F6E4141EC822FUL, 0x67A91AB3B3677DA9UL, 0xFD1C435F5FFDBE1CUL, 0xEA25604545EA8A25UL, - 0xBFDAF92323BF46DAUL, 0xF702515353F7A602UL, 0x96A145E4E496D3A1UL, 0x5BED769B9B5B2DEDUL, - 0xC25D287575C2EA5DUL, 0x1C24C5E1E11CD924UL, 0xAEE9D43D3DAE7AE9UL, 0x6ABEF24C4C6A98BEUL, - 0x5AEE826C6C5AD8EEUL, 0x41C3BD7E7E41FCC3UL, 0x0206F3F5F502F106UL, 0x4FD15283834F1DD1UL, - 0x5CE48C68685CD0E4UL, 0xF407565151F4A207UL, 0x345C8DD1D134B95CUL, 0x0818E1F9F908E918UL, - 0x93AE4CE2E293DFAEUL, 0x73953EABAB734D95UL, 0x53F597626253C4F5UL, 0x3F416B2A2A3F5441UL, - 0x0C141C08080C1014UL, 0x52F66395955231F6UL, 0x65AFE94646658CAFUL, 0x5EE27F9D9D5E21E2UL, - 0x2878483030286078UL, 0xA1F8CF3737A16EF8UL, 0x0F111B0A0A0F1411UL, 0xB5C4EB2F2FB55EC4UL, - 0x091B150E0E091C1BUL, 0x365A7E242436485AUL, 0x9BB6AD1B1B9B36B6UL, 0x3D4798DFDF3DA547UL, - 0x266AA7CDCD26816AUL, 0x69BBF54E4E699CBBUL, 0xCD4C337F7FCDFE4CUL, 0x9FBA50EAEA9FCFBAUL, - 0x1B2D3F12121B242DUL, 0x9EB9A41D1D9E3AB9UL, 0x749CC4585874B09CUL, 0x2E724634342E6872UL, - 0x2D774136362D6C77UL, 0xB2CD11DCDCB2A3CDUL, 0xEE299DB4B4EE7329UL, 0xFB164D5B5BFBB616UL, - 0xF601A5A4A4F65301UL, 0x4DD7A176764DECD7UL, 0x61A314B7B76175A3UL, 0xCE49347D7DCEFA49UL, - 0x7B8DDF52527BA48DUL, 0x3E429FDDDD3EA142UL, 0x7193CD5E5E71BC93UL, 0x97A2B113139726A2UL, - 0xF504A2A6A6F55704UL, 0x68B801B9B96869B8UL, 0x0000000000000000UL, 0x2C74B5C1C12C9974UL, - 0x60A0E040406080A0UL, 0x1F21C2E3E31FDD21UL, 0xC8433A7979C8F243UL, 0xED2C9AB6B6ED772CUL, - 0xBED90DD4D4BEB3D9UL, 0x46CA478D8D4601CAUL, 0xD970176767D9CE70UL, 0x4BDDAF72724BE4DDUL, - 0xDE79ED9494DE3379UL, 0xD467FF9898D42B67UL, 0xE82393B0B0E87B23UL, 0x4ADE5B85854A11DEUL, - 0x6BBD06BBBB6B6DBDUL, 0x2A7EBBC5C52A917EUL, 0xE5347B4F4FE59E34UL, 0x163AD7EDED16C13AUL, - 0xC554D28686C51754UL, 0xD762F89A9AD72F62UL, 0x55FF99666655CCFFUL, 0x94A7B611119422A7UL, - 0xCF4AC08A8ACF0F4AUL, 0x1030D9E9E910C930UL, 0x060A0E040406080AUL, 0x819866FEFE81E798UL, - 0xF00BABA0A0F05B0BUL, 0x44CCB4787844F0CCUL, 0xBAD5F02525BA4AD5UL, 0xE33E754B4BE3963EUL, - 0xF30EACA2A2F35F0EUL, 0xFE19445D5DFEBA19UL, 0xC05BDB8080C01B5BUL, 0x8A858005058A0A85UL, - 0xADECD33F3FAD7EECUL, 0xBCDFFE2121BC42DFUL, 0x48D8A8707048E0D8UL, 0x040CFDF1F104F90CUL, - 0xDF7A196363DFC67AUL, 0xC1582F7777C1EE58UL, 0x759F30AFAF75459FUL, 0x63A5E742426384A5UL, - 0x3050702020304050UL, 0x1A2ECBE5E51AD12EUL, 0x0E12EFFDFD0EE112UL, 0x6DB708BFBF6D65B7UL, - 0x4CD45581814C19D4UL, 0x143C24181814303CUL, 0x355F792626354C5FUL, 0x2F71B2C3C32F9D71UL, - 0xE13886BEBEE16738UL, 0xA2FDC83535A26AFDUL, 0xCC4FC78888CC0B4FUL, 0x394B652E2E395C4BUL, - 0x57F96A9393573DF9UL, 0xF20D585555F2AA0DUL, 0x829D61FCFC82E39DUL, 0x47C9B37A7A47F4C9UL, - 0xACEF27C8C8AC8BEFUL, 0xE73288BABAE76F32UL, 0x2B7D4F32322B647DUL, 0x95A442E6E695D7A4UL, - 0xA0FB3BC0C0A09BFBUL, 0x98B3AA19199832B3UL, 0xD168F69E9ED12768UL, 0x7F8122A3A37F5D81UL, - 0x66AAEE44446688AAUL, 0x7E82D654547EA882UL, 0xABE6DD3B3BAB76E6UL, 0x839E950B0B83169EUL, - 0xCA45C98C8CCA0345UL, 0x297BBCC7C729957BUL, 0xD36E056B6BD3D66EUL, 0x3C446C28283C5044UL, - 0x798B2CA7A779558BUL, 0xE23D81BCBCE2633DUL, 0x1D273116161D2C27UL, 0x769A37ADAD76419AUL, - 0x3B4D96DBDB3BAD4DUL, 0x56FA9E646456C8FAUL, 0x4ED2A674744EE8D2UL, 0x1E223614141E2822UL, - 0xDB76E49292DB3F76UL, 0x0A1E120C0C0A181EUL, 0x6CB4FC48486C90B4UL, 0xE4378FB8B8E46B37UL, - 0x5DE7789F9F5D25E7UL, 0x6EB20FBDBD6E61B2UL, 0xEF2A694343EF862AUL, 0xA6F135C4C4A693F1UL, - 0xA8E3DA3939A872E3UL, 0xA4F7C63131A462F7UL, 0x37598AD3D337BD59UL, 0x8B8674F2F28BFF86UL, - 0x325683D5D532B156UL, 0x43C54E8B8B430DC5UL, 0x59EB856E6E59DCEBUL, 0xB7C218DADAB7AFC2UL, - 0x8C8F8E01018C028FUL, 0x64AC1DB1B16479ACUL, 0xD26DF19C9CD2236DUL, 0xE03B724949E0923BUL, - 0xB4C71FD8D8B4ABC7UL, 0xFA15B9ACACFA4315UL, 0x0709FAF3F307FD09UL, 0x256FA0CFCF25856FUL, - 0xAFEA20CACAAF8FEAUL, 0x8E897DF4F48EF389UL, 0xE920674747E98E20UL, 0x1828381010182028UL, - 0xD5640B6F6FD5DE64UL, 0x888373F0F088FB83UL, 0x6FB1FB4A4A6F94B1UL, 0x7296CA5C5C72B896UL, - 0x246C54383824706CUL, 0xF1085F5757F1AE08UL, 0xC752217373C7E652UL, 0x51F36497975135F3UL, - 0x2365AECBCB238D65UL, 0x7C8425A1A17C5984UL, 0x9CBF57E8E89CCBBFUL, 0x21635D3E3E217C63UL, - 0xDD7CEA9696DD377CUL, 0xDC7F1E6161DCC27FUL, 0x86919C0D0D861A91UL, 0x85949B0F0F851E94UL, - 0x90AB4BE0E090DBABUL, 0x42C6BA7C7C42F8C6UL, 0xC457267171C4E257UL, 0xAAE529CCCCAA83E5UL, - 0xD873E39090D83B73UL, 0x050F090606050C0FUL, 0x0103F4F7F701F503UL, 0x12362A1C1C123836UL, - 0xA3FE3CC2C2A39FFEUL, 0x5FE18B6A6A5FD4E1UL, 0xF910BEAEAEF94710UL, 0xD06B026969D0D26BUL, - 0x91A8BF1717912EA8UL, 0x58E87199995829E8UL, 0x2769533A3A277469UL, 0xB9D0F72727B94ED0UL, - 0x384891D9D938A948UL, 0x1335DEEBEB13CD35UL, 0xB3CEE52B2BB356CEUL, 0x3355772222334455UL, - 0xBBD604D2D2BBBFD6UL, 0x709039A9A9704990UL, 0x8980870707890E80UL, 0xA7F2C13333A766F2UL, - 0xB6C1EC2D2DB65AC1UL, 0x22665A3C3C227866UL, 0x92ADB81515922AADUL, 0x2060A9C9C9208960UL, - 0x49DB5C87874915DBUL, 0xFF1AB0AAAAFF4F1AUL, 0x7888D8505078A088UL, 0x7A8E2BA5A57A518EUL, - 0x8F8A8903038F068AUL, 0xF8134A5959F8B213UL, 0x809B92090980129BUL, 0x1739231A1A173439UL, - 0xDA75106565DACA75UL, 0x315384D7D731B553UL, 0xC651D58484C61351UL, 0xB8D303D0D0B8BBD3UL, - 0xC35EDC8282C31F5EUL, 0xB0CBE22929B052CBUL, 0x7799C35A5A77B499UL, 0x11332D1E1E113C33UL, + 0xA5F432C6C6A597F4UL, 0x84976FF8F884EB97UL, 0x99B05EEEEE99C7B0UL, 0x8D8C7AF6F68DF78CUL, + 0x0D17E8FFFF0DE517UL, 0xBDDC0AD6D6BDB7DCUL, 0xB1C816DEDEB1A7C8UL, 0x54FC6D91915439FCUL, + 0x50F090606050C0F0UL, 0x0305070202030405UL, 0xA9E02ECECEA987E0UL, 0x7D87D156567DAC87UL, + 0x192BCCE7E719D52BUL, 0x62A613B5B56271A6UL, 0xE6317C4D4DE69A31UL, 0x9AB559ECEC9AC3B5UL, + 0x45CF408F8F4505CFUL, 0x9DBCA31F1F9D3EBCUL, 0x40C04989894009C0UL, 0x879268FAFA87EF92UL, + 0x153FD0EFEF15C53FUL, 0xEB2694B2B2EB7F26UL, 0xC940CE8E8EC90740UL, 0x0B1DE6FBFB0BED1DUL, + 0xEC2F6E4141EC822FUL, 0x67A91AB3B3677DA9UL, 0xFD1C435F5FFDBE1CUL, 0xEA25604545EA8A25UL, + 0xBFDAF92323BF46DAUL, 0xF702515353F7A602UL, 0x96A145E4E496D3A1UL, 0x5BED769B9B5B2DEDUL, + 0xC25D287575C2EA5DUL, 0x1C24C5E1E11CD924UL, 0xAEE9D43D3DAE7AE9UL, 0x6ABEF24C4C6A98BEUL, + 0x5AEE826C6C5AD8EEUL, 0x41C3BD7E7E41FCC3UL, 0x0206F3F5F502F106UL, 0x4FD15283834F1DD1UL, + 0x5CE48C68685CD0E4UL, 0xF407565151F4A207UL, 0x345C8DD1D134B95CUL, 0x0818E1F9F908E918UL, + 0x93AE4CE2E293DFAEUL, 0x73953EABAB734D95UL, 0x53F597626253C4F5UL, 0x3F416B2A2A3F5441UL, + 0x0C141C08080C1014UL, 0x52F66395955231F6UL, 0x65AFE94646658CAFUL, 0x5EE27F9D9D5E21E2UL, + 0x2878483030286078UL, 0xA1F8CF3737A16EF8UL, 0x0F111B0A0A0F1411UL, 0xB5C4EB2F2FB55EC4UL, + 0x091B150E0E091C1BUL, 0x365A7E242436485AUL, 0x9BB6AD1B1B9B36B6UL, 0x3D4798DFDF3DA547UL, + 0x266AA7CDCD26816AUL, 0x69BBF54E4E699CBBUL, 0xCD4C337F7FCDFE4CUL, 0x9FBA50EAEA9FCFBAUL, + 0x1B2D3F12121B242DUL, 0x9EB9A41D1D9E3AB9UL, 0x749CC4585874B09CUL, 0x2E724634342E6872UL, + 0x2D774136362D6C77UL, 0xB2CD11DCDCB2A3CDUL, 0xEE299DB4B4EE7329UL, 0xFB164D5B5BFBB616UL, + 0xF601A5A4A4F65301UL, 0x4DD7A176764DECD7UL, 0x61A314B7B76175A3UL, 0xCE49347D7DCEFA49UL, + 0x7B8DDF52527BA48DUL, 0x3E429FDDDD3EA142UL, 0x7193CD5E5E71BC93UL, 0x97A2B113139726A2UL, + 0xF504A2A6A6F55704UL, 0x68B801B9B96869B8UL, 0x0000000000000000UL, 0x2C74B5C1C12C9974UL, + 0x60A0E040406080A0UL, 0x1F21C2E3E31FDD21UL, 0xC8433A7979C8F243UL, 0xED2C9AB6B6ED772CUL, + 0xBED90DD4D4BEB3D9UL, 0x46CA478D8D4601CAUL, 0xD970176767D9CE70UL, 0x4BDDAF72724BE4DDUL, + 0xDE79ED9494DE3379UL, 0xD467FF9898D42B67UL, 0xE82393B0B0E87B23UL, 0x4ADE5B85854A11DEUL, + 0x6BBD06BBBB6B6DBDUL, 0x2A7EBBC5C52A917EUL, 0xE5347B4F4FE59E34UL, 0x163AD7EDED16C13AUL, + 0xC554D28686C51754UL, 0xD762F89A9AD72F62UL, 0x55FF99666655CCFFUL, 0x94A7B611119422A7UL, + 0xCF4AC08A8ACF0F4AUL, 0x1030D9E9E910C930UL, 0x060A0E040406080AUL, 0x819866FEFE81E798UL, + 0xF00BABA0A0F05B0BUL, 0x44CCB4787844F0CCUL, 0xBAD5F02525BA4AD5UL, 0xE33E754B4BE3963EUL, + 0xF30EACA2A2F35F0EUL, 0xFE19445D5DFEBA19UL, 0xC05BDB8080C01B5BUL, 0x8A858005058A0A85UL, + 0xADECD33F3FAD7EECUL, 0xBCDFFE2121BC42DFUL, 0x48D8A8707048E0D8UL, 0x040CFDF1F104F90CUL, + 0xDF7A196363DFC67AUL, 0xC1582F7777C1EE58UL, 0x759F30AFAF75459FUL, 0x63A5E742426384A5UL, + 0x3050702020304050UL, 0x1A2ECBE5E51AD12EUL, 0x0E12EFFDFD0EE112UL, 0x6DB708BFBF6D65B7UL, + 0x4CD45581814C19D4UL, 0x143C24181814303CUL, 0x355F792626354C5FUL, 0x2F71B2C3C32F9D71UL, + 0xE13886BEBEE16738UL, 0xA2FDC83535A26AFDUL, 0xCC4FC78888CC0B4FUL, 0x394B652E2E395C4BUL, + 0x57F96A9393573DF9UL, 0xF20D585555F2AA0DUL, 0x829D61FCFC82E39DUL, 0x47C9B37A7A47F4C9UL, + 0xACEF27C8C8AC8BEFUL, 0xE73288BABAE76F32UL, 0x2B7D4F32322B647DUL, 0x95A442E6E695D7A4UL, + 0xA0FB3BC0C0A09BFBUL, 0x98B3AA19199832B3UL, 0xD168F69E9ED12768UL, 0x7F8122A3A37F5D81UL, + 0x66AAEE44446688AAUL, 0x7E82D654547EA882UL, 0xABE6DD3B3BAB76E6UL, 0x839E950B0B83169EUL, + 0xCA45C98C8CCA0345UL, 0x297BBCC7C729957BUL, 0xD36E056B6BD3D66EUL, 0x3C446C28283C5044UL, + 0x798B2CA7A779558BUL, 0xE23D81BCBCE2633DUL, 0x1D273116161D2C27UL, 0x769A37ADAD76419AUL, + 0x3B4D96DBDB3BAD4DUL, 0x56FA9E646456C8FAUL, 0x4ED2A674744EE8D2UL, 0x1E223614141E2822UL, + 0xDB76E49292DB3F76UL, 0x0A1E120C0C0A181EUL, 0x6CB4FC48486C90B4UL, 0xE4378FB8B8E46B37UL, + 0x5DE7789F9F5D25E7UL, 0x6EB20FBDBD6E61B2UL, 0xEF2A694343EF862AUL, 0xA6F135C4C4A693F1UL, + 0xA8E3DA3939A872E3UL, 0xA4F7C63131A462F7UL, 0x37598AD3D337BD59UL, 0x8B8674F2F28BFF86UL, + 0x325683D5D532B156UL, 0x43C54E8B8B430DC5UL, 0x59EB856E6E59DCEBUL, 0xB7C218DADAB7AFC2UL, + 0x8C8F8E01018C028FUL, 0x64AC1DB1B16479ACUL, 0xD26DF19C9CD2236DUL, 0xE03B724949E0923BUL, + 0xB4C71FD8D8B4ABC7UL, 0xFA15B9ACACFA4315UL, 0x0709FAF3F307FD09UL, 0x256FA0CFCF25856FUL, + 0xAFEA20CACAAF8FEAUL, 0x8E897DF4F48EF389UL, 0xE920674747E98E20UL, 0x1828381010182028UL, + 0xD5640B6F6FD5DE64UL, 0x888373F0F088FB83UL, 0x6FB1FB4A4A6F94B1UL, 0x7296CA5C5C72B896UL, + 0x246C54383824706CUL, 0xF1085F5757F1AE08UL, 0xC752217373C7E652UL, 0x51F36497975135F3UL, + 0x2365AECBCB238D65UL, 0x7C8425A1A17C5984UL, 0x9CBF57E8E89CCBBFUL, 0x21635D3E3E217C63UL, + 0xDD7CEA9696DD377CUL, 0xDC7F1E6161DCC27FUL, 0x86919C0D0D861A91UL, 0x85949B0F0F851E94UL, + 0x90AB4BE0E090DBABUL, 0x42C6BA7C7C42F8C6UL, 0xC457267171C4E257UL, 0xAAE529CCCCAA83E5UL, + 0xD873E39090D83B73UL, 0x050F090606050C0FUL, 0x0103F4F7F701F503UL, 0x12362A1C1C123836UL, + 0xA3FE3CC2C2A39FFEUL, 0x5FE18B6A6A5FD4E1UL, 0xF910BEAEAEF94710UL, 0xD06B026969D0D26BUL, + 0x91A8BF1717912EA8UL, 0x58E87199995829E8UL, 0x2769533A3A277469UL, 0xB9D0F72727B94ED0UL, + 0x384891D9D938A948UL, 0x1335DEEBEB13CD35UL, 0xB3CEE52B2BB356CEUL, 0x3355772222334455UL, + 0xBBD604D2D2BBBFD6UL, 0x709039A9A9704990UL, 0x8980870707890E80UL, 0xA7F2C13333A766F2UL, + 0xB6C1EC2D2DB65AC1UL, 0x22665A3C3C227866UL, 0x92ADB81515922AADUL, 0x2060A9C9C9208960UL, + 0x49DB5C87874915DBUL, 0xFF1AB0AAAAFF4F1AUL, 0x7888D8505078A088UL, 0x7A8E2BA5A57A518EUL, + 0x8F8A8903038F068AUL, 0xF8134A5959F8B213UL, 0x809B92090980129BUL, 0x1739231A1A173439UL, + 0xDA75106565DACA75UL, 0x315384D7D731B553UL, 0xC651D58484C61351UL, 0xB8D303D0D0B8BBD3UL, + 0xC35EDC8282C31F5EUL, 0xB0CBE22929B052CBUL, 0x7799C35A5A77B499UL, 0x11332D1E1E113C33UL, 0xCB463D7B7BCBF646UL, 0xFC1FB7A8A8FC4B1FUL, 0xD6610C6D6DD6DA61UL, 0x3A4E622C2C3A584EUL }; @@ -292,4 +292,4 @@ static const __constant ulong T4_G[] = } while (0) )===" -
\ No newline at end of file + diff --git a/xmrstak/backend/amd/amd_gpu/opencl/jh.cl b/xmrstak/backend/amd/amd_gpu/opencl/jh.cl index fe70ea3..486d232 100644 --- a/xmrstak/backend/amd/amd_gpu/opencl/jh.cl +++ b/xmrstak/backend/amd/amd_gpu/opencl/jh.cl @@ -6,7 +6,7 @@ R"===( * ==========================(LICENSE BEGIN)============================ * * Copyright (c) 2007-2010 Projet RNRT SAPHIR - * + * * Permission is hereby granted, free of charge, to any person obtaining * a copy of this software and associated documentation files (the * "Software"), to deal in the Software without restriction, including @@ -14,10 +14,10 @@ R"===( * distribute, sublicense, and/or sell copies of the Software, and to * permit persons to whom the Software is furnished to do so, subject to * the following conditions: - * + * * The above copyright notice and this permission notice shall be * included in all copies or substantial portions of the Software. - * + * * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. @@ -106,47 +106,47 @@ typedef ulong sph_u64; static const __constant ulong C[] = { - 0x67F815DFA2DED572UL, 0x571523B70A15847BUL, 0xF6875A4D90D6AB81UL, 0x402BD1C3C54F9F4EUL, - 0x9CFA455CE03A98EAUL, 0x9A99B26699D2C503UL, 0x8A53BBF2B4960266UL, 0x31A2DB881A1456B5UL, - 0xDB0E199A5C5AA303UL, 0x1044C1870AB23F40UL, 0x1D959E848019051CUL, 0xDCCDE75EADEB336FUL, - 0x416BBF029213BA10UL, 0xD027BBF7156578DCUL, 0x5078AA3739812C0AUL, 0xD3910041D2BF1A3FUL, - 0x907ECCF60D5A2D42UL, 0xCE97C0929C9F62DDUL, 0xAC442BC70BA75C18UL, 0x23FCC663D665DFD1UL, - 0x1AB8E09E036C6E97UL, 0xA8EC6C447E450521UL, 0xFA618E5DBB03F1EEUL, 0x97818394B29796FDUL, - 0x2F3003DB37858E4AUL, 0x956A9FFB2D8D672AUL, 0x6C69B8F88173FE8AUL, 0x14427FC04672C78AUL, - 0xC45EC7BD8F15F4C5UL, 0x80BB118FA76F4475UL, 0xBC88E4AEB775DE52UL, 0xF4A3A6981E00B882UL, - 0x1563A3A9338FF48EUL, 0x89F9B7D524565FAAUL, 0xFDE05A7C20EDF1B6UL, 0x362C42065AE9CA36UL, - 0x3D98FE4E433529CEUL, 0xA74B9A7374F93A53UL, 0x86814E6F591FF5D0UL, 0x9F5AD8AF81AD9D0EUL, - 0x6A6234EE670605A7UL, 0x2717B96EBE280B8BUL, 0x3F1080C626077447UL, 0x7B487EC66F7EA0E0UL, - 0xC0A4F84AA50A550DUL, 0x9EF18E979FE7E391UL, 0xD48D605081727686UL, 0x62B0E5F3415A9E7EUL, - 0x7A205440EC1F9FFCUL, 0x84C9F4CE001AE4E3UL, 0xD895FA9DF594D74FUL, 0xA554C324117E2E55UL, - 0x286EFEBD2872DF5BUL, 0xB2C4A50FE27FF578UL, 0x2ED349EEEF7C8905UL, 0x7F5928EB85937E44UL, - 0x4A3124B337695F70UL, 0x65E4D61DF128865EUL, 0xE720B95104771BC7UL, 0x8A87D423E843FE74UL, - 0xF2947692A3E8297DUL, 0xC1D9309B097ACBDDUL, 0xE01BDC5BFB301B1DUL, 0xBF829CF24F4924DAUL, - 0xFFBF70B431BAE7A4UL, 0x48BCF8DE0544320DUL, 0x39D3BB5332FCAE3BUL, 0xA08B29E0C1C39F45UL, - 0x0F09AEF7FD05C9E5UL, 0x34F1904212347094UL, 0x95ED44E301B771A2UL, 0x4A982F4F368E3BE9UL, - 0x15F66CA0631D4088UL, 0xFFAF52874B44C147UL, 0x30C60AE2F14ABB7EUL, 0xE68C6ECCC5B67046UL, - 0x00CA4FBD56A4D5A4UL, 0xAE183EC84B849DDAUL, 0xADD1643045CE5773UL, 0x67255C1468CEA6E8UL, - 0x16E10ECBF28CDAA3UL, 0x9A99949A5806E933UL, 0x7B846FC220B2601FUL, 0x1885D1A07FACCED1UL, - 0xD319DD8DA15B5932UL, 0x46B4A5AAC01C9A50UL, 0xBA6B04E467633D9FUL, 0x7EEE560BAB19CAF6UL, - 0x742128A9EA79B11FUL, 0xEE51363B35F7BDE9UL, 0x76D350755AAC571DUL, 0x01707DA3FEC2463AUL, - 0x42D8A498AFC135F7UL, 0x79676B9E20ECED78UL, 0xA8DB3AEA15638341UL, 0x832C83324D3BC3FAUL, - 0xF347271C1F3B40A7UL, 0x9A762DB734F04059UL, 0xFD4F21D26C4E3EE7UL, 0xEF5957DC398DFDB8UL, - 0xDAEB492B490C9B8DUL, 0x0D70F36849D7A25BUL, 0x84558D7AD0AE3B7DUL, 0x658EF8E4F0E9A5F5UL, - 0x533B1036F4A2B8A0UL, 0x5AEC3E759E07A80CUL, 0x4F88E85692946891UL, 0x4CBCBAF8555CB05BUL, - 0x7B9487F3993BBBE3UL, 0x5D1C6B72D6F4DA75UL, 0x6DB334DC28ACAE64UL, 0x71DB28B850A5346CUL, - 0x2A518D10F2E261F8UL, 0xFC75DD593364DBE3UL, 0xA23FCE43F1BCAC1CUL, 0xB043E8023CD1BB67UL, - 0x75A12988CA5B0A33UL, 0x5C5316B44D19347FUL, 0x1E4D790EC3943B92UL, 0x3FAFEEB6D7757479UL, - 0x21391ABEF7D4A8EAUL, 0x5127234C097EF45CUL, 0xD23C32BA5324A326UL, 0xADD5A66D4A17A344UL, - 0x08C9F2AFA63E1DB5UL, 0x563C6B91983D5983UL, 0x4D608672A17CF84CUL, 0xF6C76E08CC3EE246UL, - 0x5E76BCB1B333982FUL, 0x2AE6C4EFA566D62BUL, 0x36D4C1BEE8B6F406UL, 0x6321EFBC1582EE74UL, - 0x69C953F40D4EC1FDUL, 0x26585806C45A7DA7UL, 0x16FAE0061614C17EUL, 0x3F9D63283DAF907EUL, - 0x0CD29B00E3F2C9D2UL, 0x300CD4B730CEAA5FUL, 0x9832E0F216512A74UL, 0x9AF8CEE3D830EB0DUL, - 0x9279F1B57B9EC54BUL, 0xD36886046EE651FFUL, 0x316796E6574D239BUL, 0x05750A17F3A6E6CCUL, - 0xCE6C3213D98176B1UL, 0x62A205F88452173CUL, 0x47154778B3CB2BF4UL, 0x486A9323825446FFUL, - 0x65655E4E0758DF38UL, 0x8E5086FC897CFCF2UL, 0x86CA0BD0442E7031UL, 0x4E477830A20940F0UL, - 0x8338F7D139EEA065UL, 0xBD3A2CE437E95EF7UL, 0x6FF8130126B29721UL, 0xE7DE9FEFD1ED44A3UL, - 0xD992257615DFA08BUL, 0xBE42DC12F6F7853CUL, 0x7EB027AB7CECA7D8UL, 0xDEA83EAADA7D8D53UL, - 0xD86902BD93CE25AAUL, 0xF908731AFD43F65AUL, 0xA5194A17DAEF5FC0UL, 0x6A21FD4C33664D97UL, + 0x67F815DFA2DED572UL, 0x571523B70A15847BUL, 0xF6875A4D90D6AB81UL, 0x402BD1C3C54F9F4EUL, + 0x9CFA455CE03A98EAUL, 0x9A99B26699D2C503UL, 0x8A53BBF2B4960266UL, 0x31A2DB881A1456B5UL, + 0xDB0E199A5C5AA303UL, 0x1044C1870AB23F40UL, 0x1D959E848019051CUL, 0xDCCDE75EADEB336FUL, + 0x416BBF029213BA10UL, 0xD027BBF7156578DCUL, 0x5078AA3739812C0AUL, 0xD3910041D2BF1A3FUL, + 0x907ECCF60D5A2D42UL, 0xCE97C0929C9F62DDUL, 0xAC442BC70BA75C18UL, 0x23FCC663D665DFD1UL, + 0x1AB8E09E036C6E97UL, 0xA8EC6C447E450521UL, 0xFA618E5DBB03F1EEUL, 0x97818394B29796FDUL, + 0x2F3003DB37858E4AUL, 0x956A9FFB2D8D672AUL, 0x6C69B8F88173FE8AUL, 0x14427FC04672C78AUL, + 0xC45EC7BD8F15F4C5UL, 0x80BB118FA76F4475UL, 0xBC88E4AEB775DE52UL, 0xF4A3A6981E00B882UL, + 0x1563A3A9338FF48EUL, 0x89F9B7D524565FAAUL, 0xFDE05A7C20EDF1B6UL, 0x362C42065AE9CA36UL, + 0x3D98FE4E433529CEUL, 0xA74B9A7374F93A53UL, 0x86814E6F591FF5D0UL, 0x9F5AD8AF81AD9D0EUL, + 0x6A6234EE670605A7UL, 0x2717B96EBE280B8BUL, 0x3F1080C626077447UL, 0x7B487EC66F7EA0E0UL, + 0xC0A4F84AA50A550DUL, 0x9EF18E979FE7E391UL, 0xD48D605081727686UL, 0x62B0E5F3415A9E7EUL, + 0x7A205440EC1F9FFCUL, 0x84C9F4CE001AE4E3UL, 0xD895FA9DF594D74FUL, 0xA554C324117E2E55UL, + 0x286EFEBD2872DF5BUL, 0xB2C4A50FE27FF578UL, 0x2ED349EEEF7C8905UL, 0x7F5928EB85937E44UL, + 0x4A3124B337695F70UL, 0x65E4D61DF128865EUL, 0xE720B95104771BC7UL, 0x8A87D423E843FE74UL, + 0xF2947692A3E8297DUL, 0xC1D9309B097ACBDDUL, 0xE01BDC5BFB301B1DUL, 0xBF829CF24F4924DAUL, + 0xFFBF70B431BAE7A4UL, 0x48BCF8DE0544320DUL, 0x39D3BB5332FCAE3BUL, 0xA08B29E0C1C39F45UL, + 0x0F09AEF7FD05C9E5UL, 0x34F1904212347094UL, 0x95ED44E301B771A2UL, 0x4A982F4F368E3BE9UL, + 0x15F66CA0631D4088UL, 0xFFAF52874B44C147UL, 0x30C60AE2F14ABB7EUL, 0xE68C6ECCC5B67046UL, + 0x00CA4FBD56A4D5A4UL, 0xAE183EC84B849DDAUL, 0xADD1643045CE5773UL, 0x67255C1468CEA6E8UL, + 0x16E10ECBF28CDAA3UL, 0x9A99949A5806E933UL, 0x7B846FC220B2601FUL, 0x1885D1A07FACCED1UL, + 0xD319DD8DA15B5932UL, 0x46B4A5AAC01C9A50UL, 0xBA6B04E467633D9FUL, 0x7EEE560BAB19CAF6UL, + 0x742128A9EA79B11FUL, 0xEE51363B35F7BDE9UL, 0x76D350755AAC571DUL, 0x01707DA3FEC2463AUL, + 0x42D8A498AFC135F7UL, 0x79676B9E20ECED78UL, 0xA8DB3AEA15638341UL, 0x832C83324D3BC3FAUL, + 0xF347271C1F3B40A7UL, 0x9A762DB734F04059UL, 0xFD4F21D26C4E3EE7UL, 0xEF5957DC398DFDB8UL, + 0xDAEB492B490C9B8DUL, 0x0D70F36849D7A25BUL, 0x84558D7AD0AE3B7DUL, 0x658EF8E4F0E9A5F5UL, + 0x533B1036F4A2B8A0UL, 0x5AEC3E759E07A80CUL, 0x4F88E85692946891UL, 0x4CBCBAF8555CB05BUL, + 0x7B9487F3993BBBE3UL, 0x5D1C6B72D6F4DA75UL, 0x6DB334DC28ACAE64UL, 0x71DB28B850A5346CUL, + 0x2A518D10F2E261F8UL, 0xFC75DD593364DBE3UL, 0xA23FCE43F1BCAC1CUL, 0xB043E8023CD1BB67UL, + 0x75A12988CA5B0A33UL, 0x5C5316B44D19347FUL, 0x1E4D790EC3943B92UL, 0x3FAFEEB6D7757479UL, + 0x21391ABEF7D4A8EAUL, 0x5127234C097EF45CUL, 0xD23C32BA5324A326UL, 0xADD5A66D4A17A344UL, + 0x08C9F2AFA63E1DB5UL, 0x563C6B91983D5983UL, 0x4D608672A17CF84CUL, 0xF6C76E08CC3EE246UL, + 0x5E76BCB1B333982FUL, 0x2AE6C4EFA566D62BUL, 0x36D4C1BEE8B6F406UL, 0x6321EFBC1582EE74UL, + 0x69C953F40D4EC1FDUL, 0x26585806C45A7DA7UL, 0x16FAE0061614C17EUL, 0x3F9D63283DAF907EUL, + 0x0CD29B00E3F2C9D2UL, 0x300CD4B730CEAA5FUL, 0x9832E0F216512A74UL, 0x9AF8CEE3D830EB0DUL, + 0x9279F1B57B9EC54BUL, 0xD36886046EE651FFUL, 0x316796E6574D239BUL, 0x05750A17F3A6E6CCUL, + 0xCE6C3213D98176B1UL, 0x62A205F88452173CUL, 0x47154778B3CB2BF4UL, 0x486A9323825446FFUL, + 0x65655E4E0758DF38UL, 0x8E5086FC897CFCF2UL, 0x86CA0BD0442E7031UL, 0x4E477830A20940F0UL, + 0x8338F7D139EEA065UL, 0xBD3A2CE437E95EF7UL, 0x6FF8130126B29721UL, 0xE7DE9FEFD1ED44A3UL, + 0xD992257615DFA08BUL, 0xBE42DC12F6F7853CUL, 0x7EB027AB7CECA7D8UL, 0xDEA83EAADA7D8D53UL, + 0xD86902BD93CE25AAUL, 0xF908731AFD43F65AUL, 0xA5194A17DAEF5FC0UL, 0x6A21FD4C33664D97UL, 0x701541DB3198B435UL, 0x9B54CDEDBB0F1EEAUL, 0x72409751A163D09AUL, 0xE26F4791BF9D75F6UL }; diff --git a/xmrstak/backend/amd/amd_gpu/opencl/wolf-skein.cl b/xmrstak/backend/amd/amd_gpu/opencl/wolf-skein.cl index 279b652..73ef908 100644 --- a/xmrstak/backend/amd/amd_gpu/opencl/wolf-skein.cl +++ b/xmrstak/backend/amd/amd_gpu/opencl/wolf-skein.cl @@ -50,19 +50,19 @@ ulong8 SkeinEvenRound(ulong8 p, const ulong8 h, const ulong *t, const uint s, co { SKEIN_INJECT_KEY(p, s, q); ulong4 pv0 = p.even, pv1 = p.odd; - + SkeinMix8(&pv0, &pv1, (ulong4)(46, 36, 19, 37)); pv0 = shuffle(pv0, (ulong4)(1, 2, 3, 0)); pv1 = shuffle(pv1, (ulong4)(0, 3, 2, 1)); - + SkeinMix8(&pv0, &pv1, (ulong4)(33, 27, 14, 42)); pv0 = shuffle(pv0, (ulong4)(1, 2, 3, 0)); pv1 = shuffle(pv1, (ulong4)(0, 3, 2, 1)); - + SkeinMix8(&pv0, &pv1, (ulong4)(17, 49, 36, 39)); pv0 = shuffle(pv0, (ulong4)(1, 2, 3, 0)); pv1 = shuffle(pv1, (ulong4)(0, 3, 2, 1)); - + SkeinMix8(&pv0, &pv1, (ulong4)(44, 9, 54, 56)); return(shuffle2(pv0, pv1, (ulong8)(1, 4, 2, 7, 3, 6, 0, 5))); } @@ -71,19 +71,19 @@ ulong8 SkeinOddRound(ulong8 p, const ulong8 h, const ulong *t, const uint s, con { SKEIN_INJECT_KEY(p, s, q); ulong4 pv0 = p.even, pv1 = p.odd; - + SkeinMix8(&pv0, &pv1, (ulong4)(39, 30, 34, 24)); pv0 = shuffle(pv0, (ulong4)(1, 2, 3, 0)); pv1 = shuffle(pv1, (ulong4)(0, 3, 2, 1)); - + SkeinMix8(&pv0, &pv1, (ulong4)(13, 50, 10, 17)); pv0 = shuffle(pv0, (ulong4)(1, 2, 3, 0)); pv1 = shuffle(pv1, (ulong4)(0, 3, 2, 1)); - + SkeinMix8(&pv0, &pv1, (ulong4)(25, 29, 39, 43)); pv0 = shuffle(pv0, (ulong4)(1, 2, 3, 0)); pv1 = shuffle(pv1, (ulong4)(0, 3, 2, 1)); - + SkeinMix8(&pv0, &pv1, (ulong4)(8, 35, 56, 22)); return(shuffle2(pv0, pv1, (ulong8)(1, 4, 2, 7, 3, 6, 0, 5))); } @@ -129,7 +129,7 @@ ulong8 Skein512Block(ulong8 p, ulong8 h, ulong h8, const ulong *t) h.s7 = h8; h8 = tmp; } - + p += h; p.s5 += t[0]; p.s6 += t[1]; diff --git a/xmrstak/backend/amd/autoAdjust.hpp b/xmrstak/backend/amd/autoAdjust.hpp index 685890b..d6acec9 100644 --- a/xmrstak/backend/amd/autoAdjust.hpp +++ b/xmrstak/backend/amd/autoAdjust.hpp @@ -31,7 +31,7 @@ namespace amd { class autoAdjust -{ +{ public: autoAdjust() diff --git a/xmrstak/backend/amd/minethd.cpp b/xmrstak/backend/amd/minethd.cpp index 4353e3d..88431cc 100644 --- a/xmrstak/backend/amd/minethd.cpp +++ b/xmrstak/backend/amd/minethd.cpp @@ -73,7 +73,7 @@ minethd::minethd(miner_work& pWork, size_t iNo, GpuContext* ctx, const jconf::th extern "C" { #ifdef WIN32 -__declspec(dllexport) +__declspec(dllexport) #endif std::vector<iBackend*>* xmrstak_start_backend(uint32_t threadOffset, miner_work& pWork, environment& env) { @@ -122,7 +122,7 @@ std::vector<iBackend*>* minethd::thread_starter(uint32_t threadOffset, miner_wor win_exit(); } - // \ todo get device count and exit if no opencl device + // \ todo get device count and exit if no opencl device if(!init_gpus()) { @@ -139,7 +139,7 @@ std::vector<iBackend*>* minethd::thread_starter(uint32_t threadOffset, miner_wor jconf::inst()->GetThreadConfig(i, cfg); const std::string backendName = xmrstak::params::inst().openCLVendor; - + if(cfg.cpu_aff >= 0) { #if defined(__APPLE__) @@ -158,27 +158,6 @@ std::vector<iBackend*>* minethd::thread_starter(uint32_t threadOffset, miner_wor return pvThreads; } -void minethd::switch_work(miner_work& pWork) -{ - // iConsumeCnt is a basic lock-like polling mechanism just in case we happen to push work - // faster than threads can consume them. This should never happen in real life. - // Pool cant physically send jobs faster than every 250ms or so due to net latency. - - while (globalStates::inst().iConsumeCnt.load(std::memory_order_seq_cst) < globalStates::inst().iThreadCount) - std::this_thread::sleep_for(std::chrono::milliseconds(100)); - - globalStates::inst().oGlobalWork = pWork; - globalStates::inst().iConsumeCnt.store(0, std::memory_order_seq_cst); - globalStates::inst().iGlobalJobNo++; -} - -void minethd::consume_work() -{ - memcpy(&oWork, &globalStates::inst().oGlobalWork, sizeof(miner_work)); - iJobNo++; - globalStates::inst().iConsumeCnt++; - -} void minethd::work_main() { @@ -193,13 +172,11 @@ void minethd::work_main() uint64_t iCount = 0; cryptonight_ctx* cpu_ctx; cpu_ctx = cpu::minethd::minethd_alloc_ctx(); - + // start with root algorithm and switch later if fork version is reached auto miner_algo = ::jconf::inst()->GetCurrentCoinSelection().GetDescription(1).GetMiningAlgoRoot(); cn_hash_fun hash_fun = cpu::minethd::func_selector(::jconf::inst()->HaveHardwareAes(), true /*bNoPrefetch*/, miner_algo); - globalStates::inst().iConsumeCnt++; - uint8_t version = 0; size_t lastPoolId = 0; @@ -215,7 +192,7 @@ void minethd::work_main() while (globalStates::inst().iGlobalJobNo.load(std::memory_order_relaxed) == iJobNo) std::this_thread::sleep_for(std::chrono::milliseconds(100)); - consume_work(); + globalStates::inst().consume_work(oWork, iJobNo); continue; } @@ -242,7 +219,7 @@ void minethd::work_main() assert(sizeof(job_result::sJobID) == sizeof(pool_job::sJobID)); uint64_t target = oWork.iTarget; - + XMRSetJob(pGpuCtx, oWork.bWorkBlob, oWork.iWorkSize, target, miner_algo); if(oWork.bNiceHash) @@ -254,7 +231,11 @@ void minethd::work_main() if((round_ctr++ & 0xF) == 0) { globalStates::inst().calc_start_nonce(pGpuCtx->Nonce, oWork.bNiceHash, h_per_round * 16); + // check if the job is still valid, there is a small possibility that the job is switched + if(globalStates::inst().iGlobalJobNo.load(std::memory_order_relaxed) != iJobNo) + break; } + cl_uint results[0x100]; memset(results,0,sizeof(cl_uint)*(0x100)); @@ -285,7 +266,7 @@ void minethd::work_main() std::this_thread::yield(); } - consume_work(); + globalStates::inst().consume_work(oWork, iJobNo); } } diff --git a/xmrstak/backend/amd/minethd.hpp b/xmrstak/backend/amd/minethd.hpp index 29ddb74..3142117 100644 --- a/xmrstak/backend/amd/minethd.hpp +++ b/xmrstak/backend/amd/minethd.hpp @@ -20,7 +20,6 @@ class minethd : public iBackend { public: - static void switch_work(miner_work& pWork); static std::vector<iBackend*>* thread_starter(uint32_t threadOffset, miner_work& pWork); static bool init_gpus(); @@ -30,11 +29,9 @@ private: minethd(miner_work& pWork, size_t iNo, GpuContext* ctx, const jconf::thd_cfg cfg); void work_main(); - void consume_work(); uint64_t iJobNo; - - static miner_work oGlobalWork; + miner_work oWork; std::promise<void> order_fix; diff --git a/xmrstak/backend/backendConnector.cpp b/xmrstak/backend/backendConnector.cpp index 6f80a0f..525413f 100644 --- a/xmrstak/backend/backendConnector.cpp +++ b/xmrstak/backend/backendConnector.cpp @@ -57,9 +57,6 @@ bool BackendConnector::self_test() std::vector<iBackend*>* BackendConnector::thread_starter(miner_work& pWork) { - globalStates::inst().iGlobalJobNo = 0; - globalStates::inst().iConsumeCnt = 0; - std::vector<iBackend*>* pvThreads = new std::vector<iBackend*>; diff --git a/xmrstak/backend/cpu/autoAdjust.hpp b/xmrstak/backend/cpu/autoAdjust.hpp index 518721a..57dbef0 100644 --- a/xmrstak/backend/cpu/autoAdjust.hpp +++ b/xmrstak/backend/cpu/autoAdjust.hpp @@ -52,7 +52,7 @@ public: std::string conf; - + if(!detectL3Size() || L3KB_size < halfHashMemSizeKB || L3KB_size > (halfHashMemSizeKB * 2048u)) { if(L3KB_size < halfHashMemSizeKB || L3KB_size > (halfHashMemSizeKB * 2048)) @@ -127,7 +127,7 @@ private: if(get_masked(cpu_info[0], 7, 5) != 3) { - printer::inst()->print_msg(L0, "Autoconf failed: Couln't find L3 cache page."); + printer::inst()->print_msg(L0, "Autoconf failed: Couldn't find L3 cache page."); return false; } diff --git a/xmrstak/backend/cpu/autoAdjustHwloc.hpp b/xmrstak/backend/cpu/autoAdjustHwloc.hpp index b1f3914..01d2280 100644 --- a/xmrstak/backend/cpu/autoAdjustHwloc.hpp +++ b/xmrstak/backend/cpu/autoAdjustHwloc.hpp @@ -37,7 +37,7 @@ public: bool printConfig() { - + hwloc_topology_t topology; hwloc_topology_init(&topology); hwloc_topology_load(topology); @@ -64,8 +64,8 @@ public: throw(std::runtime_error("The CPU doesn't seem to have a cache.")); for(hwloc_obj_t obj : tlcs) - proccessTopLevelCache(obj); - + processTopLevelCache(obj); + for(uint32_t id : results) { conf += std::string(" { \"low_power_mode\" : "); @@ -138,7 +138,7 @@ private: // Top level cache isn't shared with other cores on the same package // This will usually be 1 x L3, but can be 2 x L2 per package - void proccessTopLevelCache(hwloc_obj_t obj) + void processTopLevelCache(hwloc_obj_t obj) { if(obj->attr == nullptr) throw(std::runtime_error("Cache object hasn't got attributes.")); @@ -158,7 +158,7 @@ private: //Try our luck with lower level caches for(size_t i=0; i < obj->arity; i++) - proccessTopLevelCache(obj->children[i]); + processTopLevelCache(obj->children[i]); return; } diff --git a/xmrstak/backend/cpu/config.tpl b/xmrstak/backend/cpu/config.tpl index cb4b950..2fc9a47 100644 --- a/xmrstak/backend/cpu/config.tpl +++ b/xmrstak/backend/cpu/config.tpl @@ -2,25 +2,25 @@ R"===( /* * Thread configuration for each thread. Make sure it matches the number above. * low_power_mode - This can either be a boolean (true or false), or a number between 1 to 5. When set to true, - * this mode will double the cache usage, and double the single thread performance. It will - * consume much less power (as less cores are working), but will max out at around 80-85% of + * this mode will double the cache usage, and double the single thread performance. It will + * consume much less power (as less cores are working), but will max out at around 80-85% of * the maximum performance. When set to a number N greater than 1, this mode will increase the * cache usage and single thread performance by N times. * - * no_prefetch - Some sytems can gain up to extra 5% here, but sometimes it will have no difference or make + * no_prefetch - Some systems can gain up to extra 5% here, but sometimes it will have no difference or make * things slower. * - * affine_to_cpu - This can be either false (no affinity), or the CPU core number. Note that on hyperthreading - * systems it is better to assign threads to physical cores. On Windows this usually means selecting - * even or odd numbered cpu numbers. For Linux it will be usually the lower CPU numbers, so for a 4 + * affine_to_cpu - This can be either false (no affinity), or the CPU core number. Note that on hyperthreading + * systems it is better to assign threads to physical cores. On Windows this usually means selecting + * even or odd numbered cpu numbers. For Linux it will be usually the lower CPU numbers, so for a 4 * physical core CPU you should select cpu numbers 0-3. * * On the first run the miner will look at your system and suggest a basic configuration that will work, * you can try to tweak it from there to get the best performance. - * + * * A filled out configuration should look like this: * "cpu_threads_conf" : - * [ + * [ * { "low_power_mode" : false, "no_prefetch" : true, "affine_to_cpu" : 0 }, * { "low_power_mode" : false, "no_prefetch" : true, "affine_to_cpu" : 1 }, * ], diff --git a/xmrstak/backend/cpu/crypto/c_groestl.c b/xmrstak/backend/cpu/crypto/c_groestl.c index 1318d5a..5b3523e 100644 --- a/xmrstak/backend/cpu/crypto/c_groestl.c +++ b/xmrstak/backend/cpu/crypto/c_groestl.c @@ -4,7 +4,7 @@ * * This work is based on the implementation of * Soeren S. Thomsen and Krystian Matusiewicz - * + * * */ @@ -22,7 +22,7 @@ const uint8_t indices_cyclic[15] = {0,1,2,3,4,5,6,7,0,1,2,3,4,5,6}; #define ROTATE_COLUMN_DOWN(v1, v2, amount_bytes, temp_var) {temp_var = (v1<<(8*amount_bytes))|(v2>>(8*(4-amount_bytes))); \ v2 = (v2<<(8*amount_bytes))|(v1>>(8*(4-amount_bytes))); \ v1 = temp_var;} - + #define COLUMN(x,y,i,c0,c1,c2,c3,c4,c5,c6,c7,tv1,tv2,tu,tl,t) \ tu = T[2*(uint32_t)x[4*c0+0]]; \ @@ -161,11 +161,11 @@ static void F512(uint32_t *h, const uint32_t *m) { /* digest up to msglen bytes of input (full blocks only) */ static void Transform(groestlHashState *ctx, - const uint8_t *input, + const uint8_t *input, int msglen) { /* digest message, one block at a time */ - for (; msglen >= SIZE512; + for (; msglen >= SIZE512; msglen -= SIZE512, input += SIZE512) { F512(ctx->chaining,(uint32_t*)input); @@ -199,7 +199,7 @@ static void OutputTransformation(groestlHashState *ctx) { RND512P((uint8_t*)y, temp, 0x00000009); for (j = 0; j < 2*COLS512; j++) { ctx->chaining[j] ^= temp[j]; - } + } } /* initialise context */ @@ -313,7 +313,7 @@ static void Final(groestlHashState* ctx, ctx->block_counter2 >>= 8; } /* digest final padding block */ - Transform(ctx, ctx->buffer, SIZE512); + Transform(ctx, ctx->buffer, SIZE512); /* perform output transformation */ OutputTransformation(ctx); @@ -332,7 +332,7 @@ static void Final(groestlHashState* ctx, } /* hash bit sequence */ -void groestl(const BitSequence* data, +void groestl(const BitSequence* data, DataLength databitlen, BitSequence* hashval) { diff --git a/xmrstak/backend/cpu/crypto/c_groestl.h b/xmrstak/backend/cpu/crypto/c_groestl.h index 2b51339..47044b4 100644 --- a/xmrstak/backend/cpu/crypto/c_groestl.h +++ b/xmrstak/backend/cpu/crypto/c_groestl.h @@ -4,10 +4,10 @@ #include "crypto_uint8.h" #include "crypto_uint32.h" #include "crypto_uint64.h" -#include "crypto_hash.h" +#include "crypto_hash.h" -typedef crypto_uint8 uint8_t; -typedef crypto_uint32 uint32_t; +typedef crypto_uint8 uint8_t; +typedef crypto_uint32 uint32_t; typedef crypto_uint64 uint64_t; */ #include <stdint.h> @@ -42,7 +42,7 @@ typedef struct { BitSequence buffer[SIZE512]; /* data buffer */ int buf_ptr; /* data buffer pointer */ int bits_in_last_byte; /* no. of message bits in last byte of - data buffer */ + data buffer */ } groestlHashState; /*void Init(hashState*); @@ -53,8 +53,8 @@ void groestl(const BitSequence*, DataLength, BitSequence*); /* int crypto_hash(unsigned char *out, - const unsigned char *in, - unsigned long long len); + const unsigned char *in, + unsigned long long len); */ #endif /* __hash_h */ diff --git a/xmrstak/backend/cpu/crypto/c_jh.c b/xmrstak/backend/cpu/crypto/c_jh.c index 9d685a0..0256a0f 100644 --- a/xmrstak/backend/cpu/crypto/c_jh.c +++ b/xmrstak/backend/cpu/crypto/c_jh.c @@ -234,7 +234,7 @@ static HashReturn Init(hashState *state, int hashbitlen) /*initialize the initial hash value of JH*/ state->hashbitlen = hashbitlen; - /*load the intital hash value into state*/ + /*load the initial hash value into state*/ switch (hashbitlen) { case 224: memcpy(state->x,JH224_H0,128); break; diff --git a/xmrstak/backend/cpu/crypto/c_keccak.c b/xmrstak/backend/cpu/crypto/c_keccak.c index eadb85b..63c1614 100644 --- a/xmrstak/backend/cpu/crypto/c_keccak.c +++ b/xmrstak/backend/cpu/crypto/c_keccak.c @@ -12,14 +12,14 @@ #define ROTL64(x, y) (((x) << (y)) | ((x) >> (64 - (y)))) #endif -const uint64_t keccakf_rndc[24] = +const uint64_t keccakf_rndc[24] = { 0x0000000000000001, 0x0000000000008082, 0x800000000000808a, 0x8000000080008000, 0x000000000000808b, 0x0000000080000001, 0x8000000080008081, 0x8000000000008009, 0x000000000000008a, 0x0000000000000088, 0x0000000080008009, 0x000000008000000a, 0x000000008000808b, 0x800000000000008b, 0x8000000000008089, - 0x8000000000008003, 0x8000000000008002, 0x8000000000000080, + 0x8000000000008003, 0x8000000000008002, 0x8000000000000080, 0x000000000000800a, 0x800000008000000a, 0x8000000080008081, 0x8000000000008080, 0x0000000080000001, 0x8000000080008008 }; @@ -130,7 +130,7 @@ void keccakf(uint64_t st[25], int rounds) st[j + 2] ^= (~bc[3]) & bc[4]; st[j + 3] ^= (~bc[4]) & bc[0]; st[j + 4] ^= (~bc[0]) & bc[1]; - + // Iota st[0] ^= keccakf_rndc[round]; } @@ -147,7 +147,7 @@ void keccak(const uint8_t *in, int inlen, uint8_t *md, int mdlen) rsiz = sizeof(state_t) == mdlen ? HASH_DATA_AREA : 200 - 2 * mdlen; rsizw = rsiz / 8; - + memset(st, 0, sizeof(st)); for ( ; inlen >= rsiz; inlen -= rsiz, in += rsiz) { @@ -155,7 +155,7 @@ void keccak(const uint8_t *in, int inlen, uint8_t *md, int mdlen) st[i] ^= ((uint64_t *) in)[i]; keccakf(st, KECCAK_ROUNDS); } - + // last block and padding memcpy(temp, in, inlen); temp[inlen++] = 1; @@ -173,4 +173,4 @@ void keccak(const uint8_t *in, int inlen, uint8_t *md, int mdlen) void keccak1600(const uint8_t *in, int inlen, uint8_t *md) { keccak(in, inlen, md, sizeof(state_t)); -}
\ No newline at end of file +} diff --git a/xmrstak/backend/cpu/crypto/c_skein.c b/xmrstak/backend/cpu/crypto/c_skein.c index 2453713..e2d5442 100644 --- a/xmrstak/backend/cpu/crypto/c_skein.c +++ b/xmrstak/backend/cpu/crypto/c_skein.c @@ -5,7 +5,7 @@ ** Source code author: Doug Whiting, 2008. ** ** This algorithm and source code is released to the public domain. -** +** ************************************************************************/ #define SKEIN_PORT_CODE /* instantiate any code in skein_port.h */ @@ -96,12 +96,12 @@ static int Skein1024_Final (Skein1024_Ctxt_t *ctx, u08b_t * hashVal); ** After an InitExt() call, just use Update/Final calls as with Init(). ** ** Notes: Same parameters as _Init() calls, plus treeInfo/key/keyBytes. -** When keyBytes == 0 and treeInfo == SKEIN_SEQUENTIAL, +** When keyBytes == 0 and treeInfo == SKEIN_SEQUENTIAL, ** the results of InitExt() are identical to calling Init(). ** The function Init() may be called once to "precompute" the IV for ** a given hashBitLen value, then by saving a copy of the context ** the IV computation may be avoided in later calls. -** Similarly, the function InitExt() may be called once per MAC key +** Similarly, the function InitExt() may be called once per MAC key ** to precompute the MAC IV, then a copy of the context saved and ** reused for each new MAC computation. **/ @@ -135,7 +135,7 @@ static int Skein1024_Output (Skein1024_Ctxt_t *ctx, u08b_t * hashVal); /***************************************************************** ** "Internal" Skein definitions -** -- not needed for sequential hashing API, but will be +** -- not needed for sequential hashing API, but will be ** helpful for other uses of Skein (e.g., tree hash mode). ** -- included here so that they can be shared between ** reference and optimized code. @@ -257,11 +257,11 @@ static int Skein1024_Output (Skein1024_Ctxt_t *ctx, u08b_t * hashVal); #define Skein_Assert(x,retCode)/* default: ignore all Asserts, for performance */ #define Skein_assert(x) #elif defined(SKEIN_ASSERT) -#include <assert.h> -#define Skein_Assert(x,retCode) assert(x) -#define Skein_assert(x) assert(x) +#include <assert.h> +#define Skein_Assert(x,retCode) assert(x) +#define Skein_assert(x) assert(x) #else -#include <assert.h> +#include <assert.h> #define Skein_Assert(x,retCode) { if (!(x)) return retCode; } /* caller error */ #define Skein_assert(x) assert(x) /* internal error */ #endif @@ -269,8 +269,8 @@ static int Skein1024_Output (Skein1024_Ctxt_t *ctx, u08b_t * hashVal); /***************************************************************** ** Skein block function constants (shared across Ref and Opt code) ******************************************************************/ -enum -{ +enum +{ /* Skein_256 round rotation constants */ R_256_0_0=14, R_256_0_1=16, R_256_1_0=52, R_256_1_1=57, @@ -518,7 +518,7 @@ const u64b_t SKEIN1024_IV_1024[] = #define BLK_BITS (WCNT*64) /* some useful definitions for code here */ #define KW_TWK_BASE (0) #define KW_KEY_BASE (3) -#define ks (kw + KW_KEY_BASE) +#define ks (kw + KW_KEY_BASE) #define ts (kw + KW_TWK_BASE) #ifdef SKEIN_DEBUG @@ -567,7 +567,7 @@ static void Skein_256_Process_Block(Skein_256_Ctxt_t *ctx,const u08b_t *blkPtr,s ts[0] += byteCntAdd; /* update processed length */ /* precompute the key schedule for this block */ - ks[0] = ctx->X[0]; + ks[0] = ctx->X[0]; ks[1] = ctx->X[1]; ks[2] = ctx->X[2]; ks[3] = ctx->X[3]; @@ -594,7 +594,7 @@ static void Skein_256_Process_Block(Skein_256_Ctxt_t *ctx,const u08b_t *blkPtr,s X##p0 += X##p1; X##p1 = RotL_64(X##p1,ROT##_0); X##p1 ^= X##p0; \ X##p2 += X##p3; X##p3 = RotL_64(X##p3,ROT##_1); X##p3 ^= X##p2; \ -#if SKEIN_UNROLL_256 == 0 +#if SKEIN_UNROLL_256 == 0 #define R256(p0,p1,p2,p3,ROT,rNum) /* fully unrolled */ \ Round256(p0,p1,p2,p3,ROT,rNum) \ Skein_Show_R_Ptr(BLK_BITS,&ctx->h,rNum,Xptr); @@ -620,8 +620,8 @@ static void Skein_256_Process_Block(Skein_256_Ctxt_t *ctx,const u08b_t *blkPtr,s Skein_Show_R_Ptr(BLK_BITS,&ctx->h,SKEIN_RND_KEY_INJECT,Xptr); for (r=1;r < 2*RCNT;r+=2*SKEIN_UNROLL_256) /* loop thru it */ -#endif - { +#endif + { #define R256_8_rounds(R) \ R256(0,1,2,3,R_256_0,8*(R) + 1); \ R256(0,3,2,1,R_256_1,8*(R) + 2); \ @@ -762,7 +762,7 @@ static void Skein_512_Process_Block(Skein_512_Ctxt_t *ctx,const u08b_t *blkPtr,s ks[5] = ctx->X[5]; ks[6] = ctx->X[6]; ks[7] = ctx->X[7]; - ks[8] = ks[0] ^ ks[1] ^ ks[2] ^ ks[3] ^ + ks[8] = ks[0] ^ ks[1] ^ ks[2] ^ ks[3] ^ ks[4] ^ ks[5] ^ ks[6] ^ ks[7] ^ SKEIN_KS_PARITY; ts[2] = ts[0] ^ ts[1]; @@ -790,7 +790,7 @@ static void Skein_512_Process_Block(Skein_512_Ctxt_t *ctx,const u08b_t *blkPtr,s X##p4 += X##p5; X##p5 = RotL_64(X##p5,ROT##_2); X##p5 ^= X##p4; \ X##p6 += X##p7; X##p7 = RotL_64(X##p7,ROT##_3); X##p7 ^= X##p6; \ -#if SKEIN_UNROLL_512 == 0 +#if SKEIN_UNROLL_512 == 0 #define R512(p0,p1,p2,p3,p4,p5,p6,p7,ROT,rNum) /* unrolled */ \ Round512(p0,p1,p2,p3,p4,p5,p6,p7,ROT,rNum) \ Skein_Show_R_Ptr(BLK_BITS,&ctx->h,rNum,Xptr); @@ -1022,7 +1022,7 @@ static void Skein1024_Process_Block(Skein1024_Ctxt_t *ctx,const u08b_t *blkPtr,s X##pC += X##pD; X##pD = RotL_64(X##pD,ROT##_6); X##pD ^= X##pC; \ X##pE += X##pF; X##pF = RotL_64(X##pF,ROT##_7); X##pF ^= X##pE; \ -#if SKEIN_UNROLL_1024 == 0 +#if SKEIN_UNROLL_1024 == 0 #define R1024(p0,p1,p2,p3,p4,p5,p6,p7,p8,p9,pA,pB,pC,pD,pE,pF,ROT,rn) \ Round1024(p0,p1,p2,p3,p4,p5,p6,p7,p8,p9,pA,pB,pC,pD,pE,pF,ROT,rn) \ Skein_Show_R_Ptr(BLK_BITS,&ctx->h,rn,Xptr); @@ -1044,7 +1044,7 @@ static void Skein1024_Process_Block(Skein1024_Ctxt_t *ctx,const u08b_t *blkPtr,s X13 += ks[((R)+14) % 17] + ts[((R)+1) % 3]; \ X14 += ks[((R)+15) % 17] + ts[((R)+2) % 3]; \ X15 += ks[((R)+16) % 17] + (R)+1; \ - Skein_Show_R_Ptr(BLK_BITS,&ctx->h,SKEIN_RND_KEY_INJECT,Xptr); + Skein_Show_R_Ptr(BLK_BITS,&ctx->h,SKEIN_RND_KEY_INJECT,Xptr); #else /* looping version */ #define R1024(p0,p1,p2,p3,p4,p5,p6,p7,p8,p9,pA,pB,pC,pD,pE,pF,ROT,rn) \ Round1024(p0,p1,p2,p3,p4,p5,p6,p7,p8,p9,pA,pB,pC,pD,pE,pF,ROT,rn) \ @@ -1072,7 +1072,7 @@ static void Skein1024_Process_Block(Skein1024_Ctxt_t *ctx,const u08b_t *blkPtr,s Skein_Show_R_Ptr(BLK_BITS,&ctx->h,SKEIN_RND_KEY_INJECT,Xptr); for (r=1;r <= 2*RCNT;r+=2*SKEIN_UNROLL_1024) /* loop thru it */ -#endif +#endif { #define R1024_8_rounds(R) /* do 8 full rounds */ \ R1024(00,01,02,03,04,05,06,07,08,09,10,11,12,13,14,15,R1024_0,8*(R) + 1); \ @@ -1156,7 +1156,7 @@ static void Skein1024_Process_Block(Skein1024_Ctxt_t *ctx,const u08b_t *blkPtr,s ctx->X[15] = X15 ^ w[15]; Skein_Show_Round(BLK_BITS,&ctx->h,SKEIN_RND_FEED_FWD,ctx->X); - + ts[1] &= ~SKEIN_T1_FLAG_FIRST; blkPtr += SKEIN1024_BLOCK_BYTES; } @@ -1193,7 +1193,7 @@ static int Skein_256_Init(Skein_256_Ctxt_t *ctx, size_t hashBitLen) u08b_t b[SKEIN_256_STATE_BYTES]; u64b_t w[SKEIN_256_STATE_WORDS]; } cfg; /* config block */ - + Skein_Assert(hashBitLen > 0,SKEIN_BAD_HASHLEN); ctx->h.hashBitLen = hashBitLen; /* output hash bit count */ @@ -1237,13 +1237,13 @@ static int Skein_256_InitExt(Skein_256_Ctxt_t *ctx,size_t hashBitLen,u64b_t tree u08b_t b[SKEIN_256_STATE_BYTES]; u64b_t w[SKEIN_256_STATE_WORDS]; } cfg; /* config block */ - + Skein_Assert(hashBitLen > 0,SKEIN_BAD_HASHLEN); Skein_Assert(keyBytes == 0 || key != NULL,SKEIN_FAIL); /* compute the initial chaining values ctx->X[], based on key */ if (keyBytes == 0) /* is there a key? */ - { + { memset(ctx->X,0,sizeof(ctx->X)); /* no key: use all zeroes as key for config block */ } else /* here to pre-process a key */ @@ -1282,7 +1282,7 @@ static int Skein_256_InitExt(Skein_256_Ctxt_t *ctx,size_t hashBitLen,u64b_t tree /* Set up to process the data message portion of the hash (default) */ ctx->h.bCnt = 0; /* buffer b[] starts out empty */ Skein_Start_New_Type(ctx,MSG); - + return SKEIN_SUCCESS; } #endif @@ -1334,7 +1334,7 @@ static int Skein_256_Update(Skein_256_Ctxt_t *ctx, const u08b_t *msg, size_t msg return SKEIN_SUCCESS; } - + /*++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++*/ /* finalize the hash computation and output the result */ static int Skein_256_Final(Skein_256_Ctxt_t *ctx, u08b_t *hashVal) @@ -1348,7 +1348,7 @@ static int Skein_256_Final(Skein_256_Ctxt_t *ctx, u08b_t *hashVal) memset(&ctx->b[ctx->h.bCnt],0,SKEIN_256_BLOCK_BYTES - ctx->h.bCnt); Skein_256_Process_Block(ctx,ctx->b,1,ctx->h.bCnt); /* process the final block */ - + /* now output the result */ byteCnt = (ctx->h.hashBitLen + 7) >> 3; /* total number of output bytes */ @@ -1391,7 +1391,7 @@ static int Skein_512_Init(Skein_512_Ctxt_t *ctx, size_t hashBitLen) u08b_t b[SKEIN_512_STATE_BYTES]; u64b_t w[SKEIN_512_STATE_WORDS]; } cfg; /* config block */ - + Skein_Assert(hashBitLen > 0,SKEIN_BAD_HASHLEN); ctx->h.hashBitLen = hashBitLen; /* output hash bit count */ @@ -1437,13 +1437,13 @@ static int Skein_512_InitExt(Skein_512_Ctxt_t *ctx,size_t hashBitLen,u64b_t tree u08b_t b[SKEIN_512_STATE_BYTES]; u64b_t w[SKEIN_512_STATE_WORDS]; } cfg; /* config block */ - + Skein_Assert(hashBitLen > 0,SKEIN_BAD_HASHLEN); Skein_Assert(keyBytes == 0 || key != NULL,SKEIN_FAIL); /* compute the initial chaining values ctx->X[], based on key */ if (keyBytes == 0) /* is there a key? */ - { + { memset(ctx->X,0,sizeof(ctx->X)); /* no key: use all zeroes as key for config block */ } else /* here to pre-process a key */ @@ -1482,7 +1482,7 @@ static int Skein_512_InitExt(Skein_512_Ctxt_t *ctx,size_t hashBitLen,u64b_t tree /* Set up to process the data message portion of the hash (default) */ ctx->h.bCnt = 0; /* buffer b[] starts out empty */ Skein_Start_New_Type(ctx,MSG); - + return SKEIN_SUCCESS; } #endif @@ -1534,7 +1534,7 @@ static int Skein_512_Update(Skein_512_Ctxt_t *ctx, const u08b_t *msg, size_t msg return SKEIN_SUCCESS; } - + /*++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++*/ /* finalize the hash computation and output the result */ static int Skein_512_Final(Skein_512_Ctxt_t *ctx, u08b_t *hashVal) @@ -1548,7 +1548,7 @@ static int Skein_512_Final(Skein_512_Ctxt_t *ctx, u08b_t *hashVal) memset(&ctx->b[ctx->h.bCnt],0,SKEIN_512_BLOCK_BYTES - ctx->h.bCnt); Skein_512_Process_Block(ctx,ctx->b,1,ctx->h.bCnt); /* process the final block */ - + /* now output the result */ byteCnt = (ctx->h.hashBitLen + 7) >> 3; /* total number of output bytes */ @@ -1590,7 +1590,7 @@ static int Skein1024_Init(Skein1024_Ctxt_t *ctx, size_t hashBitLen) u08b_t b[SKEIN1024_STATE_BYTES]; u64b_t w[SKEIN1024_STATE_WORDS]; } cfg; /* config block */ - + Skein_Assert(hashBitLen > 0,SKEIN_BAD_HASHLEN); ctx->h.hashBitLen = hashBitLen; /* output hash bit count */ @@ -1635,13 +1635,13 @@ static int Skein1024_InitExt(Skein1024_Ctxt_t *ctx,size_t hashBitLen,u64b_t tree u08b_t b[SKEIN1024_STATE_BYTES]; u64b_t w[SKEIN1024_STATE_WORDS]; } cfg; /* config block */ - + Skein_Assert(hashBitLen > 0,SKEIN_BAD_HASHLEN); Skein_Assert(keyBytes == 0 || key != NULL,SKEIN_FAIL); /* compute the initial chaining values ctx->X[], based on key */ if (keyBytes == 0) /* is there a key? */ - { + { memset(ctx->X,0,sizeof(ctx->X)); /* no key: use all zeroes as key for config block */ } else /* here to pre-process a key */ @@ -1680,7 +1680,7 @@ static int Skein1024_InitExt(Skein1024_Ctxt_t *ctx,size_t hashBitLen,u64b_t tree /* Set up to process the data message portion of the hash (default) */ ctx->h.bCnt = 0; /* buffer b[] starts out empty */ Skein_Start_New_Type(ctx,MSG); - + return SKEIN_SUCCESS; } #endif @@ -1732,7 +1732,7 @@ static int Skein1024_Update(Skein1024_Ctxt_t *ctx, const u08b_t *msg, size_t msg return SKEIN_SUCCESS; } - + /*++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++*/ /* finalize the hash computation and output the result */ static int Skein1024_Final(Skein1024_Ctxt_t *ctx, u08b_t *hashVal) @@ -1746,7 +1746,7 @@ static int Skein1024_Final(Skein1024_Ctxt_t *ctx, u08b_t *hashVal) memset(&ctx->b[ctx->h.bCnt],0,SKEIN1024_BLOCK_BYTES - ctx->h.bCnt); Skein1024_Process_Block(ctx,ctx->b,1,ctx->h.bCnt); /* process the final block */ - + /* now output the result */ byteCnt = (ctx->h.hashBitLen + 7) >> 3; /* total number of output bytes */ @@ -1790,9 +1790,9 @@ static int Skein_256_Final_Pad(Skein_256_Ctxt_t *ctx, u08b_t *hashVal) if (ctx->h.bCnt < SKEIN_256_BLOCK_BYTES) /* zero pad b[] if necessary */ memset(&ctx->b[ctx->h.bCnt],0,SKEIN_256_BLOCK_BYTES - ctx->h.bCnt); Skein_256_Process_Block(ctx,ctx->b,1,ctx->h.bCnt); /* process the final block */ - + Skein_Put64_LSB_First(hashVal,ctx->X,SKEIN_256_BLOCK_BYTES); /* "output" the state bytes */ - + return SKEIN_SUCCESS; } @@ -1806,9 +1806,9 @@ static int Skein_512_Final_Pad(Skein_512_Ctxt_t *ctx, u08b_t *hashVal) if (ctx->h.bCnt < SKEIN_512_BLOCK_BYTES) /* zero pad b[] if necessary */ memset(&ctx->b[ctx->h.bCnt],0,SKEIN_512_BLOCK_BYTES - ctx->h.bCnt); Skein_512_Process_Block(ctx,ctx->b,1,ctx->h.bCnt); /* process the final block */ - + Skein_Put64_LSB_First(hashVal,ctx->X,SKEIN_512_BLOCK_BYTES); /* "output" the state bytes */ - + return SKEIN_SUCCESS; } @@ -1822,9 +1822,9 @@ static int Skein1024_Final_Pad(Skein1024_Ctxt_t *ctx, u08b_t *hashVal) if (ctx->h.bCnt < SKEIN1024_BLOCK_BYTES) /* zero pad b[] if necessary */ memset(&ctx->b[ctx->h.bCnt],0,SKEIN1024_BLOCK_BYTES - ctx->h.bCnt); Skein1024_Process_Block(ctx,ctx->b,1,ctx->h.bCnt); /* process the final block */ - + Skein_Put64_LSB_First(hashVal,ctx->X,SKEIN1024_BLOCK_BYTES); /* "output" the state bytes */ - + return SKEIN_SUCCESS; } diff --git a/xmrstak/backend/cpu/crypto/c_skein.h b/xmrstak/backend/cpu/crypto/c_skein.h index 86dbc08..1aa11de 100644 --- a/xmrstak/backend/cpu/crypto/c_skein.h +++ b/xmrstak/backend/cpu/crypto/c_skein.h @@ -9,7 +9,7 @@ ** This algorithm and source code is released to the public domain. ** *************************************************************************** -** +** ** The following compile-time switches may be defined to control some ** tradeoffs between speed, code size, error checking, and security. ** @@ -20,8 +20,8 @@ ** [default: no callouts (no overhead)] ** ** SKEIN_ERR_CHECK -- how error checking is handled inside Skein -** code. If not defined, most error checking -** is disabled (for performance). Otherwise, +** code. If not defined, most error checking +** is disabled (for performance). Otherwise, ** the switch value is interpreted as: ** 0: use assert() to flag errors ** 1: return SKEIN_FAIL to flag errors @@ -42,6 +42,6 @@ typedef u08b_t SkeinBitSequence; /* bit stream type */ /* "all-in-one" call */ SkeinHashReturn skein_hash(int hashbitlen, const SkeinBitSequence *data, - SkeinDataLength databitlen, SkeinBitSequence *hashval); + SkeinDataLength databitlen, SkeinBitSequence *hashval); #endif /* ifndef _SKEIN_H_ */ diff --git a/xmrstak/backend/cpu/crypto/cryptonight_aesni.h b/xmrstak/backend/cpu/crypto/cryptonight_aesni.h index 7562de1..e15c474 100644 --- a/xmrstak/backend/cpu/crypto/cryptonight_aesni.h +++ b/xmrstak/backend/cpu/crypto/cryptonight_aesni.h @@ -28,7 +28,7 @@ static inline uint64_t _umul128(uint64_t a, uint64_t b, uint64_t* hi) *hi = r >> 64; return (uint64_t)r; } -#define _mm256_set_m128i(v0, v1) _mm256_insertf128_si256(_mm256_castsi128_si256(v1), (v0), 1) + #else #include <intrin.h> #endif // __GNUC__ @@ -422,6 +422,7 @@ void cn_implode_scratchpad(const __m128i* input, __m128i* output) _mm_store_si128(output + 11, xout7); } +template<xmrstak_algo ALGO> inline void cryptonight_monero_tweak(uint64_t* mem_out, __m128i tmp) { mem_out[0] = _mm_cvtsi128_si64(tmp); @@ -431,10 +432,21 @@ inline void cryptonight_monero_tweak(uint64_t* mem_out, __m128i tmp) uint8_t x = static_cast<uint8_t>(vh >> 24); static const uint16_t table = 0x7531; - const uint8_t index = (((x >> 3) & 6) | (x & 1)) << 1; - vh ^= ((table >> index) & 0x3) << 28; + if(ALGO == cryptonight_monero || ALGO == cryptonight_aeon || ALGO == cryptonight_ipbc) + { + const uint8_t index = (((x >> 3) & 6) | (x & 1)) << 1; + vh ^= ((table >> index) & 0x3) << 28; + + mem_out[1] = vh; + } + else if(ALGO == cryptonight_stellite) + { + const uint8_t index = (((x >> 4) & 6) | (x & 1)) << 1; + vh ^= ((table >> index) & 0x3) << 28; + + mem_out[1] = vh; + } - mem_out[1] = vh; } template<xmrstak_algo ALGO, bool SOFT_AES, bool PREFETCH> @@ -444,7 +456,7 @@ void cryptonight_hash(const void* input, size_t len, void* output, cryptonight_c constexpr size_t ITERATIONS = cn_select_iter<ALGO>(); constexpr size_t MEM = cn_select_memory<ALGO>(); - if((ALGO == cryptonight_monero || ALGO == cryptonight_aeon) && len < 43) + if((ALGO == cryptonight_monero || ALGO == cryptonight_aeon || ALGO == cryptonight_ipbc || ALGO == cryptonight_stellite) && len < 43) { memset(output, 0, 32); return; @@ -453,7 +465,7 @@ void cryptonight_hash(const void* input, size_t len, void* output, cryptonight_c keccak((const uint8_t *)input, len, ctx0->hash_state, 200); uint64_t monero_const; - if(ALGO == cryptonight_monero || ALGO == cryptonight_aeon) + if(ALGO == cryptonight_monero || ALGO == cryptonight_aeon || ALGO == cryptonight_ipbc || ALGO == cryptonight_stellite) { monero_const = *reinterpret_cast<const uint64_t*>(reinterpret_cast<const uint8_t*>(input) + 35); monero_const ^= *(reinterpret_cast<const uint64_t*>(ctx0->hash_state) + 24); @@ -482,8 +494,8 @@ void cryptonight_hash(const void* input, size_t len, void* output, cryptonight_c else cx = _mm_aesenc_si128(cx, _mm_set_epi64x(ah0, al0)); - if(ALGO == cryptonight_monero || ALGO == cryptonight_aeon) - cryptonight_monero_tweak((uint64_t*)&l0[idx0 & MASK], _mm_xor_si128(bx0, cx)); + if(ALGO == cryptonight_monero || ALGO == cryptonight_aeon || ALGO == cryptonight_ipbc || ALGO == cryptonight_stellite) + cryptonight_monero_tweak<ALGO>((uint64_t*)&l0[idx0 & MASK], _mm_xor_si128(bx0, cx)); else _mm_store_si128((__m128i *)&l0[idx0 & MASK], _mm_xor_si128(bx0, cx)); @@ -506,8 +518,13 @@ void cryptonight_hash(const void* input, size_t len, void* output, cryptonight_c _mm_prefetch((const char*)&l0[al0 & MASK], _MM_HINT_T0); ah0 += lo; - if(ALGO == cryptonight_monero || ALGO == cryptonight_aeon) - ((uint64_t*)&l0[idx0 & MASK])[1] = ah0 ^ monero_const; + if(ALGO == cryptonight_monero || ALGO == cryptonight_aeon || ALGO == cryptonight_ipbc || ALGO == cryptonight_stellite) + { + if(ALGO == cryptonight_ipbc) + ((uint64_t*)&l0[idx0 & MASK])[1] = ah0 ^ monero_const ^ ((uint64_t*)&l0[idx0 & MASK])[0]; + else + ((uint64_t*)&l0[idx0 & MASK])[1] = ah0 ^ monero_const; + } else ((uint64_t*)&l0[idx0 & MASK])[1] = ah0; ah0 ^= ch; @@ -544,7 +561,7 @@ void cryptonight_double_hash(const void* input, size_t len, void* output, crypto constexpr size_t ITERATIONS = cn_select_iter<ALGO>(); constexpr size_t MEM = cn_select_memory<ALGO>(); - if((ALGO == cryptonight_monero || ALGO == cryptonight_aeon) && len < 43) + if((ALGO == cryptonight_monero || ALGO == cryptonight_aeon || ALGO == cryptonight_ipbc || ALGO == cryptonight_stellite) && len < 43) { memset(output, 0, 64); return; @@ -554,7 +571,7 @@ void cryptonight_double_hash(const void* input, size_t len, void* output, crypto keccak((const uint8_t *)input+len, len, ctx[1]->hash_state, 200); uint64_t monero_const_0, monero_const_1; - if(ALGO == cryptonight_monero || ALGO == cryptonight_aeon) + if(ALGO == cryptonight_monero || ALGO == cryptonight_aeon || ALGO == cryptonight_ipbc || ALGO == cryptonight_stellite) { monero_const_0 = *reinterpret_cast<const uint64_t*>(reinterpret_cast<const uint8_t*>(input) + 35); monero_const_0 ^= *(reinterpret_cast<const uint64_t*>(ctx[0]->hash_state) + 24); @@ -592,8 +609,8 @@ void cryptonight_double_hash(const void* input, size_t len, void* output, crypto else cx = _mm_aesenc_si128(cx, _mm_set_epi64x(axh0, axl0)); - if(ALGO == cryptonight_monero || ALGO == cryptonight_aeon) - cryptonight_monero_tweak((uint64_t*)&l0[idx0 & MASK], _mm_xor_si128(bx0, cx)); + if(ALGO == cryptonight_monero || ALGO == cryptonight_aeon || ALGO == cryptonight_ipbc || ALGO == cryptonight_stellite) + cryptonight_monero_tweak<ALGO>((uint64_t*)&l0[idx0 & MASK], _mm_xor_si128(bx0, cx)); else _mm_store_si128((__m128i *)&l0[idx0 & MASK], _mm_xor_si128(bx0, cx)); @@ -610,8 +627,8 @@ void cryptonight_double_hash(const void* input, size_t len, void* output, crypto else cx = _mm_aesenc_si128(cx, _mm_set_epi64x(axh1, axl1)); - if(ALGO == cryptonight_monero || ALGO == cryptonight_aeon) - cryptonight_monero_tweak((uint64_t*)&l1[idx1 & MASK], _mm_xor_si128(bx1, cx)); + if(ALGO == cryptonight_monero || ALGO == cryptonight_aeon || ALGO == cryptonight_ipbc || ALGO == cryptonight_stellite) + cryptonight_monero_tweak<ALGO>((uint64_t*)&l1[idx1 & MASK], _mm_xor_si128(bx1, cx)); else _mm_store_si128((__m128i *)&l1[idx1 & MASK], _mm_xor_si128(bx1, cx)); @@ -631,8 +648,13 @@ void cryptonight_double_hash(const void* input, size_t len, void* output, crypto axh0 += lo; ((uint64_t*)&l0[idx0 & MASK])[0] = axl0; - if(ALGO == cryptonight_monero || ALGO == cryptonight_aeon) - ((uint64_t*)&l0[idx0 & MASK])[1] = axh0 ^ monero_const_0; + if(ALGO == cryptonight_monero || ALGO == cryptonight_aeon || ALGO == cryptonight_ipbc || ALGO == cryptonight_stellite) + { + if(ALGO == cryptonight_ipbc) + ((uint64_t*)&l0[idx0 & MASK])[1] = axh0 ^ monero_const_0 ^ ((uint64_t*)&l0[idx0 & MASK])[0]; + else + ((uint64_t*)&l0[idx0 & MASK])[1] = axh0 ^ monero_const_0; + } else ((uint64_t*)&l0[idx0 & MASK])[1] = axh0; @@ -662,8 +684,13 @@ void cryptonight_double_hash(const void* input, size_t len, void* output, crypto axh1 += lo; ((uint64_t*)&l1[idx1 & MASK])[0] = axl1; - if(ALGO == cryptonight_monero || ALGO == cryptonight_aeon) - ((uint64_t*)&l1[idx1 & MASK])[1] = axh1 ^ monero_const_1; + if(ALGO == cryptonight_monero || ALGO == cryptonight_aeon || ALGO == cryptonight_ipbc || ALGO == cryptonight_stellite) + { + if(ALGO == cryptonight_ipbc) + ((uint64_t*)&l1[idx1 & MASK])[1] = axh1 ^ monero_const_1 ^ ((uint64_t*)&l1[idx1 & MASK])[0]; + else + ((uint64_t*)&l1[idx1 & MASK])[1] = axh1 ^ monero_const_1; + } else ((uint64_t*)&l1[idx1 & MASK])[1] = axh1; @@ -701,7 +728,7 @@ void cryptonight_double_hash(const void* input, size_t len, void* output, crypto ptr = (__m128i *)&l[idx & MASK]; \ if(PREFETCH) \ _mm_prefetch((const char*)ptr, _MM_HINT_T0); \ - c = _mm_load_si128(ptr); + c = _mm_load_si128(ptr); #define CN_STEP2(a, b, c, l, ptr, idx) \ if(SOFT_AES) \ @@ -709,8 +736,8 @@ void cryptonight_double_hash(const void* input, size_t len, void* output, crypto else \ c = _mm_aesenc_si128(c, a); \ b = _mm_xor_si128(b, c); \ - if(ALGO == cryptonight_monero || ALGO == cryptonight_aeon) \ - cryptonight_monero_tweak((uint64_t*)ptr, b); \ + if(ALGO == cryptonight_monero || ALGO == cryptonight_aeon || ALGO == cryptonight_ipbc || ALGO == cryptonight_stellite) \ + cryptonight_monero_tweak<ALGO>((uint64_t*)ptr, b); \ else \ _mm_store_si128(ptr, b);\ @@ -724,8 +751,12 @@ void cryptonight_double_hash(const void* input, size_t len, void* output, crypto #define CN_STEP4(a, b, c, l, mc, ptr, idx) \ lo = _umul128(idx, _mm_cvtsi128_si64(b), &hi); \ a = _mm_add_epi64(a, _mm_set_epi64x(lo, hi)); \ - if(ALGO == cryptonight_monero || ALGO == cryptonight_aeon) \ + if(ALGO == cryptonight_monero || ALGO == cryptonight_aeon || ALGO == cryptonight_ipbc || ALGO == cryptonight_stellite) \ + { \ _mm_store_si128(ptr, _mm_xor_si128(a, mc)); \ + if (ALGO == cryptonight_ipbc) \ + ((uint64_t*)ptr)[1] ^= ((uint64_t*)ptr)[0];\ + } \ else \ _mm_store_si128(ptr, a);\ a = _mm_xor_si128(a, b); \ @@ -751,7 +782,7 @@ void cryptonight_triple_hash(const void* input, size_t len, void* output, crypto constexpr size_t ITERATIONS = cn_select_iter<ALGO>(); constexpr size_t MEM = cn_select_memory<ALGO>(); - if((ALGO == cryptonight_monero || ALGO == cryptonight_aeon) && len < 43) + if((ALGO == cryptonight_monero || ALGO == cryptonight_aeon || ALGO == cryptonight_ipbc || ALGO == cryptonight_stellite) && len < 43) { memset(output, 0, 32 * 3); return; @@ -845,7 +876,7 @@ void cryptonight_quad_hash(const void* input, size_t len, void* output, cryptoni constexpr size_t ITERATIONS = cn_select_iter<ALGO>(); constexpr size_t MEM = cn_select_memory<ALGO>(); - if((ALGO == cryptonight_monero || ALGO == cryptonight_aeon) && len < 43) + if((ALGO == cryptonight_monero || ALGO == cryptonight_aeon || ALGO == cryptonight_ipbc || ALGO == cryptonight_stellite) && len < 43) { memset(output, 0, 32 * 4); return; @@ -883,13 +914,13 @@ void cryptonight_quad_hash(const void* input, size_t len, void* output, cryptoni __m128i cx1 = _mm_set_epi64x(0, 0); __m128i cx2 = _mm_set_epi64x(0, 0); __m128i cx3 = _mm_set_epi64x(0, 0); - + uint64_t idx0, idx1, idx2, idx3; idx0 = _mm_cvtsi128_si64(ax0); idx1 = _mm_cvtsi128_si64(ax1); idx2 = _mm_cvtsi128_si64(ax2); idx3 = _mm_cvtsi128_si64(ax3); - + for (size_t i = 0; i < ITERATIONS/2; i++) { uint64_t hi, lo; @@ -954,7 +985,7 @@ void cryptonight_penta_hash(const void* input, size_t len, void* output, crypton constexpr size_t ITERATIONS = cn_select_iter<ALGO>(); constexpr size_t MEM = cn_select_memory<ALGO>(); - if((ALGO == cryptonight_monero || ALGO == cryptonight_aeon) && len < 43) + if((ALGO == cryptonight_monero || ALGO == cryptonight_aeon || ALGO == cryptonight_ipbc || ALGO == cryptonight_stellite) && len < 43) { memset(output, 0, 32 * 5); return; diff --git a/xmrstak/backend/cpu/crypto/cryptonight_common.cpp b/xmrstak/backend/cpu/crypto/cryptonight_common.cpp index ee3b663..a478c9b 100644 --- a/xmrstak/backend/cpu/crypto/cryptonight_common.cpp +++ b/xmrstak/backend/cpu/crypto/cryptonight_common.cpp @@ -118,7 +118,7 @@ BOOL AddLargePageRights() DWORD size = 0; GetTokenInformation(hToken, TokenUser, NULL, 0, &size); - + if (size > 0 && bIsElevated) { user = (PTOKEN_USER)LocalAlloc(LPTR, size); @@ -136,7 +136,7 @@ BOOL AddLargePageRights() ZeroMemory(&attributes, sizeof(attributes)); BOOL result = FALSE; - if (LsaOpenPolicy(NULL, &attributes, POLICY_ALL_ACCESS, &handle) == 0) + if (LsaOpenPolicy(NULL, &attributes, POLICY_ALL_ACCESS, &handle) == 0) { LSA_UNICODE_STRING lockmem; lockmem.Buffer = L"SeLockMemoryPrivilege"; diff --git a/xmrstak/backend/cpu/crypto/soft_aes.hpp b/xmrstak/backend/cpu/crypto/soft_aes.hpp index d3f4637..9b4ae0a 100644 --- a/xmrstak/backend/cpu/crypto/soft_aes.hpp +++ b/xmrstak/backend/cpu/crypto/soft_aes.hpp @@ -104,9 +104,9 @@ static inline __m128i soft_aesenc(__m128i in, __m128i 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 ) | + return (saes_sbox[key >> 24 ] << 24) | + (saes_sbox[(key >> 16) & 0xff] << 16 ) | + (saes_sbox[(key >> 8) & 0xff] << 8 ) | saes_sbox[key & 0xff]; } diff --git a/xmrstak/backend/cpu/minethd.cpp b/xmrstak/backend/cpu/minethd.cpp index f8f70f9..482c085 100644 --- a/xmrstak/backend/cpu/minethd.cpp +++ b/xmrstak/backend/cpu/minethd.cpp @@ -95,6 +95,7 @@ bool minethd::thd_setaffinity(std::thread::native_handle_type h, uint64_t cpu_id return pthread_setaffinity_np(h, sizeof(cpuset_t), &mn) == 0; #elif defined(__OpenBSD__) printer::inst()->print_msg(L0,"WARNING: thread pinning is not supported under OPENBSD."); + return true; #else cpu_set_t mn; CPU_ZERO(&mn); @@ -285,7 +286,12 @@ bool minethd::self_test() else if(::jconf::inst()->GetCurrentCoinSelection().GetDescription(1).GetMiningAlgo() == cryptonight_aeon) { } - + else if(::jconf::inst()->GetCurrentCoinSelection().GetDescription(1).GetMiningAlgo() == cryptonight_ipbc) + { + } + else if(::jconf::inst()->GetCurrentCoinSelection().GetDescription(1).GetMiningAlgo() == cryptonight_stellite) + { + } for (int i = 0; i < MAX_N; i++) cryptonight_free_ctx(ctx[i]); @@ -333,7 +339,7 @@ std::vector<iBackend*> minethd::thread_starter(uint32_t threadOffset, miner_work } else printer::inst()->print_msg(L1, "Starting %dx thread, no affinity.", cfg.iMultiway); - + minethd* thd = new minethd(pWork, i + threadOffset, cfg.iMultiway, cfg.bNoPrefetch, cfg.iCpuAff); pvThreads.push_back(thd); } @@ -341,13 +347,6 @@ std::vector<iBackend*> minethd::thread_starter(uint32_t threadOffset, miner_work return pvThreads; } -void minethd::consume_work() -{ - memcpy(&oWork, &globalStates::inst().inst().oGlobalWork, sizeof(miner_work)); - iJobNo++; - globalStates::inst().inst().iConsumeCnt++; -} - minethd::cn_hash_fun minethd::func_selector(bool bHaveAes, bool bNoPrefetch, xmrstak_algo algo) { // We have two independent flag bits in the functions @@ -372,6 +371,12 @@ minethd::cn_hash_fun minethd::func_selector(bool bHaveAes, bool bNoPrefetch, xmr case cryptonight_aeon: algv = 4; break; + case cryptonight_ipbc: + algv = 5; + break; + case cryptonight_stellite: + algv = 6; + break; default: algv = 2; break; @@ -397,7 +402,15 @@ minethd::cn_hash_fun minethd::func_selector(bool bHaveAes, bool bNoPrefetch, xmr cryptonight_hash<cryptonight_aeon, false, false>, cryptonight_hash<cryptonight_aeon, true, false>, cryptonight_hash<cryptonight_aeon, false, true>, - cryptonight_hash<cryptonight_aeon, true, true> + cryptonight_hash<cryptonight_aeon, true, true>, + cryptonight_hash<cryptonight_ipbc, false, false>, + cryptonight_hash<cryptonight_ipbc, true, false>, + cryptonight_hash<cryptonight_ipbc, false, true>, + cryptonight_hash<cryptonight_ipbc, true, true>, + cryptonight_hash<cryptonight_stellite, false, false>, + cryptonight_hash<cryptonight_stellite, true, false>, + cryptonight_hash<cryptonight_stellite, false, true>, + cryptonight_hash<cryptonight_stellite, true, true> }; std::bitset<2> digit; @@ -430,7 +443,6 @@ void minethd::work_main() piHashVal = (uint64_t*)(result.bResult + 24); piNonce = (uint32_t*)(oWork.bWorkBlob + 39); - globalStates::inst().inst().iConsumeCnt++; result.iThreadId = iThreadNo; uint8_t version = 0; @@ -448,7 +460,7 @@ void minethd::work_main() while (globalStates::inst().iGlobalJobNo.load(std::memory_order_relaxed) == iJobNo) std::this_thread::sleep_for(std::chrono::milliseconds(100)); - consume_work(); + globalStates::inst().consume_work(oWork, iJobNo); continue; } @@ -491,6 +503,9 @@ void minethd::work_main() if((nonce_ctr++ & (nonce_chunk-1)) == 0) { globalStates::inst().calc_start_nonce(result.iNonce, oWork.bNiceHash, nonce_chunk); + // check if the job is still valid, there is a small posibility that the job is switched + if(globalStates::inst().iGlobalJobNo.load(std::memory_order_relaxed) != iJobNo) + break; } *piNonce = result.iNonce; @@ -504,7 +519,7 @@ void minethd::work_main() std::this_thread::yield(); } - consume_work(); + globalStates::inst().consume_work(oWork, iJobNo); } cryptonight_free_ctx(ctx); @@ -534,6 +549,12 @@ minethd::cn_hash_fun_multi minethd::func_multi_selector(size_t N, bool bHaveAes, case cryptonight_aeon: algv = 4; break; + case cryptonight_ipbc: + algv = 5; + break; + case cryptonight_stellite: + algv = 6; + break; default: algv = 2; break; @@ -573,7 +594,7 @@ minethd::cn_hash_fun_multi minethd::func_multi_selector(size_t N, bool bHaveAes, cryptonight_penta_hash<cryptonight_lite, true, false>, cryptonight_penta_hash<cryptonight_lite, false, true>, cryptonight_penta_hash<cryptonight_lite, true, true>, - + cryptonight_double_hash<cryptonight, false, false>, cryptonight_double_hash<cryptonight, true, false>, cryptonight_double_hash<cryptonight, false, true>, @@ -623,13 +644,47 @@ minethd::cn_hash_fun_multi minethd::func_multi_selector(size_t N, bool bHaveAes, cryptonight_penta_hash<cryptonight_aeon, false, false>, cryptonight_penta_hash<cryptonight_aeon, true, false>, cryptonight_penta_hash<cryptonight_aeon, false, true>, - cryptonight_penta_hash<cryptonight_aeon, true, true> + cryptonight_penta_hash<cryptonight_aeon, true, true>, + + cryptonight_double_hash<cryptonight_ipbc, false, false>, + cryptonight_double_hash<cryptonight_ipbc, true, false>, + cryptonight_double_hash<cryptonight_ipbc, false, true>, + cryptonight_double_hash<cryptonight_ipbc, true, true>, + cryptonight_triple_hash<cryptonight_ipbc, false, false>, + cryptonight_triple_hash<cryptonight_ipbc, true, false>, + cryptonight_triple_hash<cryptonight_ipbc, false, true>, + cryptonight_triple_hash<cryptonight_ipbc, true, true>, + cryptonight_quad_hash<cryptonight_ipbc, false, false>, + cryptonight_quad_hash<cryptonight_ipbc, true, false>, + cryptonight_quad_hash<cryptonight_ipbc, false, true>, + cryptonight_quad_hash<cryptonight_ipbc, true, true>, + cryptonight_penta_hash<cryptonight_ipbc, false, false>, + cryptonight_penta_hash<cryptonight_ipbc, true, false>, + cryptonight_penta_hash<cryptonight_ipbc, false, true>, + cryptonight_penta_hash<cryptonight_ipbc, true, true>, + + cryptonight_double_hash<cryptonight_stellite, false, false>, + cryptonight_double_hash<cryptonight_stellite, true, false>, + cryptonight_double_hash<cryptonight_stellite, false, true>, + cryptonight_double_hash<cryptonight_stellite, true, true>, + cryptonight_triple_hash<cryptonight_stellite, false, false>, + cryptonight_triple_hash<cryptonight_stellite, true, false>, + cryptonight_triple_hash<cryptonight_stellite, false, true>, + cryptonight_triple_hash<cryptonight_stellite, true, true>, + cryptonight_quad_hash<cryptonight_stellite, false, false>, + cryptonight_quad_hash<cryptonight_stellite, true, false>, + cryptonight_quad_hash<cryptonight_stellite, false, true>, + cryptonight_quad_hash<cryptonight_stellite, true, true>, + cryptonight_penta_hash<cryptonight_stellite, false, false>, + cryptonight_penta_hash<cryptonight_stellite, true, false>, + cryptonight_penta_hash<cryptonight_stellite, false, true>, + cryptonight_penta_hash<cryptonight_stellite, true, true>, }; std::bitset<2> digit; digit.set(0, !bHaveAes); digit.set(1, !bNoPrefetch); - + return func_table[algv << 4 | (N-2) << 2 | digit.to_ulong()]; } @@ -713,7 +768,7 @@ void minethd::multiway_work_main() while (globalStates::inst().iGlobalJobNo.load(std::memory_order_relaxed) == iJobNo) std::this_thread::sleep_for(std::chrono::milliseconds(100)); - consume_work(); + globalStates::inst().consume_work(oWork, iJobNo); prep_multiway_work<N>(bWorkBlob, piNonce); continue; } @@ -758,6 +813,9 @@ void minethd::multiway_work_main() { globalStates::inst().calc_start_nonce(iNonce, oWork.bNiceHash, nonce_chunk); nonce_ctr = nonce_chunk; + // check if the job is still valid, there is a small posibility that the job is switched + if(globalStates::inst().iGlobalJobNo.load(std::memory_order_relaxed) != iJobNo) + break; } for (size_t i = 0; i < N; i++) @@ -776,7 +834,7 @@ void minethd::multiway_work_main() std::this_thread::yield(); } - consume_work(); + globalStates::inst().consume_work(oWork, iJobNo); prep_multiway_work<N>(bWorkBlob, piNonce); } diff --git a/xmrstak/backend/cpu/minethd.hpp b/xmrstak/backend/cpu/minethd.hpp index 85a95d1..2d40ce3 100644 --- a/xmrstak/backend/cpu/minethd.hpp +++ b/xmrstak/backend/cpu/minethd.hpp @@ -47,11 +47,8 @@ private: void quad_work_main(); void penta_work_main(); - void consume_work(); - uint64_t iJobNo; - static miner_work oGlobalWork; miner_work oWork; std::promise<void> order_fix; diff --git a/xmrstak/backend/cryptonight.hpp b/xmrstak/backend/cryptonight.hpp index 8a8e259..633ddf4 100644 --- a/xmrstak/backend/cryptonight.hpp +++ b/xmrstak/backend/cryptonight.hpp @@ -3,14 +3,16 @@ #include <inttypes.h> #include <type_traits> -enum xmrstak_algo +enum xmrstak_algo { invalid_algo = 0, cryptonight = 1, cryptonight_lite = 2, cryptonight_monero = 3, cryptonight_heavy = 4, - cryptonight_aeon = 5 + cryptonight_aeon = 5, + cryptonight_ipbc = 6, // equal to cryptonight_aeon with a small tweak in the miner code + cryptonight_stellite = 7 //equal to cryptonight_monero but with one tiny change }; // define aeon settings @@ -44,14 +46,22 @@ inline constexpr size_t cn_select_memory<cryptonight_heavy>() { return CRYPTONIG template<> inline constexpr size_t cn_select_memory<cryptonight_aeon>() { return CRYPTONIGHT_LITE_MEMORY; } +template<> +inline constexpr size_t cn_select_memory<cryptonight_ipbc>() { return CRYPTONIGHT_LITE_MEMORY; } + +template<> +inline constexpr size_t cn_select_memory<cryptonight_stellite>() { return CRYPTONIGHT_MEMORY; } + inline size_t cn_select_memory(xmrstak_algo algo) { switch(algo) { + case cryptonight_stellite: case cryptonight_monero: case cryptonight: - return CRYPTONIGHT_MEMORY; + return CRYPTONIGHT_MEMORY; + case cryptonight_ipbc: case cryptonight_aeon: case cryptonight_lite: return CRYPTONIGHT_LITE_MEMORY; @@ -80,13 +90,21 @@ inline constexpr uint32_t cn_select_mask<cryptonight_heavy>() { return CRYPTONIG template<> inline constexpr uint32_t cn_select_mask<cryptonight_aeon>() { return CRYPTONIGHT_LITE_MASK; } +template<> +inline constexpr uint32_t cn_select_mask<cryptonight_ipbc>() { return CRYPTONIGHT_LITE_MASK; } + +template<> +inline constexpr uint32_t cn_select_mask<cryptonight_stellite>() { return CRYPTONIGHT_MASK; } + inline size_t cn_select_mask(xmrstak_algo algo) { switch(algo) { + case cryptonight_stellite: case cryptonight_monero: case cryptonight: return CRYPTONIGHT_MASK; + case cryptonight_ipbc: case cryptonight_aeon: case cryptonight_lite: return CRYPTONIGHT_LITE_MASK; @@ -115,13 +133,21 @@ inline constexpr uint32_t cn_select_iter<cryptonight_heavy>() { return CRYPTONIG template<> inline constexpr uint32_t cn_select_iter<cryptonight_aeon>() { return CRYPTONIGHT_LITE_ITER; } +template<> +inline constexpr uint32_t cn_select_iter<cryptonight_ipbc>() { return CRYPTONIGHT_LITE_ITER; } + +template<> +inline constexpr uint32_t cn_select_iter<cryptonight_stellite>() { return CRYPTONIGHT_ITER; } + inline size_t cn_select_iter(xmrstak_algo algo) { switch(algo) { + case cryptonight_stellite: case cryptonight_monero: case cryptonight: return CRYPTONIGHT_ITER; + case cryptonight_ipbc: case cryptonight_aeon: case cryptonight_lite: return CRYPTONIGHT_LITE_ITER; diff --git a/xmrstak/backend/globalStates.cpp b/xmrstak/backend/globalStates.cpp index 1ec7983..3bd7d0e 100644 --- a/xmrstak/backend/globalStates.cpp +++ b/xmrstak/backend/globalStates.cpp @@ -33,24 +33,37 @@ namespace xmrstak { +void globalStates::consume_work( miner_work& threadWork, uint64_t& currentJobId) +{ + jobLock.ReadLock(); + + threadWork = oGlobalWork; + currentJobId = iGlobalJobNo.load(std::memory_order_relaxed); + + jobLock.UnLock(); +} void globalStates::switch_work(miner_work& pWork, pool_data& dat) { - // iConsumeCnt is a basic lock-like polling mechanism just in case we happen to push work - // faster than threads can consume them. This should never happen in real life. - // Pool cant physically send jobs faster than every 250ms or so due to net latency. - - while (iConsumeCnt.load(std::memory_order_seq_cst) < iThreadCount) - std::this_thread::sleep_for(std::chrono::milliseconds(100)); + jobLock.WriteLock(); + /* This notifies all threads that the job has changed. + * To avoid duplicated shared this must be done before the nonce is exchanged. + */ + iGlobalJobNo++; + size_t xid = dat.pool_id; dat.pool_id = pool_id; pool_id = xid; - dat.iSavedNonce = iGlobalNonce.exchange(dat.iSavedNonce, std::memory_order_seq_cst); + /* Maybe a worker thread is updating the nonce while we read it. + * To avoid duplicated share calculations the job ID is checked in the worker thread + * after the nonce is read. + */ + dat.iSavedNonce = iGlobalNonce.exchange(dat.iSavedNonce, std::memory_order_relaxed); oGlobalWork = pWork; - iConsumeCnt.store(0, std::memory_order_seq_cst); - iGlobalJobNo++; + + jobLock.UnLock(); } } // namespace xmrstak diff --git a/xmrstak/backend/globalStates.hpp b/xmrstak/backend/globalStates.hpp index fafd232..c8d6917 100644 --- a/xmrstak/backend/globalStates.hpp +++ b/xmrstak/backend/globalStates.hpp @@ -1,26 +1,16 @@ #pragma once -#include "miner_work.hpp" +#include "xmrstak/backend/miner_work.hpp" #include "xmrstak/misc/environment.hpp" #include "xmrstak/misc/console.hpp" +#include "xmrstak/backend/pool_data.hpp" +#include "xmrstak/cpputil/read_write_lock.h" #include <atomic> -constexpr static size_t invalid_pool_id = (-1); - namespace xmrstak { -struct pool_data -{ - uint32_t iSavedNonce; - size_t pool_id; - - pool_data() : iSavedNonce(0), pool_id(invalid_pool_id) - { - } -}; - struct globalStates { static inline globalStates& inst() @@ -42,6 +32,8 @@ struct globalStates nonce = iGlobalNonce.fetch_add(reserve_count); } + void consume_work( miner_work& threadWork, uint64_t& currentJobId); + miner_work oGlobalWork; std::atomic<uint64_t> iGlobalJobNo; std::atomic<uint64_t> iConsumeCnt; @@ -50,9 +42,11 @@ struct globalStates size_t pool_id = invalid_pool_id; private: - globalStates() : iThreadCount(0) + globalStates() : iThreadCount(0), iGlobalJobNo(0), iConsumeCnt(0) { } + + ::cpputil::RWLock jobLock; }; } // namespace xmrstak diff --git a/xmrstak/backend/iBackend.hpp b/xmrstak/backend/iBackend.hpp index fdc647e..18411b7 100644 --- a/xmrstak/backend/iBackend.hpp +++ b/xmrstak/backend/iBackend.hpp @@ -20,7 +20,7 @@ namespace xmrstak { enum BackendType : uint32_t { UNKNOWN = 0u, CPU = 1u, AMD = 2u, NVIDIA = 3u }; - + static const char* getName(const BackendType type) { const char* backendNames[] = { diff --git a/xmrstak/backend/miner_work.hpp b/xmrstak/backend/miner_work.hpp index 438ec0d..b6456f0 100644 --- a/xmrstak/backend/miner_work.hpp +++ b/xmrstak/backend/miner_work.hpp @@ -1,5 +1,7 @@ #pragma once +#include "xmrstak/backend/pool_data.hpp" + #include <thread> #include <atomic> #include <mutex> @@ -20,7 +22,7 @@ namespace xmrstak bool bStall; size_t iPoolId; - miner_work() : iWorkSize(0), bNiceHash(false), bStall(true), iPoolId(0) { } + miner_work() : iWorkSize(0), bNiceHash(false), bStall(true), iPoolId(invalid_pool_id) { } miner_work(const char* sJobID, const uint8_t* bWork, uint32_t iWorkSize, uint64_t iTarget, bool bNiceHash, size_t iPoolId) : iWorkSize(iWorkSize), diff --git a/xmrstak/backend/nvidia/autoAdjust.hpp b/xmrstak/backend/nvidia/autoAdjust.hpp index d8bb621..1246809 100644 --- a/xmrstak/backend/nvidia/autoAdjust.hpp +++ b/xmrstak/backend/nvidia/autoAdjust.hpp @@ -21,7 +21,7 @@ namespace nvidia { class autoAdjust -{ +{ public: autoAdjust() @@ -42,7 +42,7 @@ public: // evaluate config parameter for if auto adjustment is needed for(int i = 0; i < deviceCount; i++) { - + nvid_ctx ctx; ctx.device_id = i; diff --git a/xmrstak/backend/nvidia/config.tpl b/xmrstak/backend/nvidia/config.tpl index f489956..2aa68dc 100644 --- a/xmrstak/backend/nvidia/config.tpl +++ b/xmrstak/backend/nvidia/config.tpl @@ -22,7 +22,7 @@ R"===( * A filled out configuration should look like this: * "gpu_threads_conf" : * [ - * { "index" : 0, "threads" : 17, "blocks" : 60, "bfactor" : 0, "bsleep" : 0, + * { "index" : 0, "threads" : 17, "blocks" : 60, "bfactor" : 0, "bsleep" : 0, * "affine_to_cpu" : false, "sync_mode" : 3, * }, * ], diff --git a/xmrstak/backend/nvidia/minethd.cpp b/xmrstak/backend/nvidia/minethd.cpp index 92f5f78..16171e1 100644 --- a/xmrstak/backend/nvidia/minethd.cpp +++ b/xmrstak/backend/nvidia/minethd.cpp @@ -181,7 +181,7 @@ std::vector<iBackend*>* minethd::thread_starter(uint32_t threadOffset, miner_wor } else printer::inst()->print_msg(L1, "Starting NVIDIA GPU thread %d, no affinity.", i); - + minethd* thd = new minethd(pWork, i + threadOffset, cfg); pvThreads->push_back(thd); @@ -195,27 +195,6 @@ std::vector<iBackend*>* minethd::thread_starter(uint32_t threadOffset, miner_wor return pvThreads; } -void minethd::switch_work(miner_work& pWork) -{ - // iConsumeCnt is a basic lock-like polling mechanism just in case we happen to push work - // faster than threads can consume them. This should never happen in real life. - // Pool cant physically send jobs faster than every 250ms or so due to net latency. - - while (globalStates::inst().iConsumeCnt.load(std::memory_order_seq_cst) < globalStates::inst().iThreadCount) - std::this_thread::sleep_for(std::chrono::milliseconds(100)); - - globalStates::inst().oGlobalWork = pWork; - globalStates::inst().iConsumeCnt.store(0, std::memory_order_seq_cst); - globalStates::inst().iGlobalJobNo++; -} - -void minethd::consume_work() -{ - memcpy(&oWork, &globalStates::inst().oGlobalWork, sizeof(miner_work)); - iJobNo++; - globalStates::inst().iConsumeCnt++; -} - void minethd::work_main() { if(affinity >= 0) //-1 means no affinity @@ -223,7 +202,7 @@ void minethd::work_main() if(cuda_get_deviceinfo(&ctx) != 0 || cryptonight_extra_cpu_init(&ctx) != 1) { - printer::inst()->print_msg(L0, "Setup failed for GPU %d. Exitting.\n", (int)iThreadNo); + printer::inst()->print_msg(L0, "Setup failed for GPU %d. Exiting.\n", (int)iThreadNo); std::exit(0); } @@ -237,14 +216,12 @@ void minethd::work_main() uint64_t iCount = 0; cryptonight_ctx* cpu_ctx; cpu_ctx = cpu::minethd::minethd_alloc_ctx(); - + // start with root algorithm and switch later if fork version is reached auto miner_algo = ::jconf::inst()->GetCurrentCoinSelection().GetDescription(1).GetMiningAlgoRoot(); cn_hash_fun hash_fun = cpu::minethd::func_selector(::jconf::inst()->HaveHardwareAes(), true /*bNoPrefetch*/, miner_algo); - - uint32_t iNonce; - globalStates::inst().iConsumeCnt++; + uint32_t iNonce; uint8_t version = 0; size_t lastPoolId = 0; @@ -261,7 +238,7 @@ void minethd::work_main() while (globalStates::inst().iGlobalJobNo.load(std::memory_order_relaxed) == iJobNo) std::this_thread::sleep_for(std::chrono::milliseconds(100)); - consume_work(); + globalStates::inst().consume_work(oWork, iJobNo); continue; } uint8_t new_version = oWork.getVersion(); @@ -298,8 +275,11 @@ void minethd::work_main() if((round_ctr++ & 0xF) == 0) { globalStates::inst().calc_start_nonce(iNonce, oWork.bNiceHash, h_per_round * 16); + // check if the job is still valid, there is a small posibility that the job is switched + if(globalStates::inst().iGlobalJobNo.load(std::memory_order_relaxed) != iJobNo) + break; } - + uint32_t foundNonce[10]; uint32_t foundCount; @@ -337,7 +317,7 @@ void minethd::work_main() std::this_thread::yield(); } - consume_work(); + globalStates::inst().consume_work(oWork, iJobNo); } } diff --git a/xmrstak/backend/nvidia/minethd.hpp b/xmrstak/backend/nvidia/minethd.hpp index 89c2944..d4ae038 100644 --- a/xmrstak/backend/nvidia/minethd.hpp +++ b/xmrstak/backend/nvidia/minethd.hpp @@ -24,7 +24,6 @@ class minethd : public iBackend { public: - static void switch_work(miner_work& pWork); static std::vector<iBackend*>* thread_starter(uint32_t threadOffset, miner_work& pWork); static bool self_test(); @@ -33,16 +32,14 @@ private: minethd(miner_work& pWork, size_t iNo, const jconf::thd_cfg& cfg); void start_mining(); - + void work_main(); - void consume_work(); static std::atomic<uint64_t> iGlobalJobNo; static std::atomic<uint64_t> iConsumeCnt; static uint64_t iThreadCount; uint64_t iJobNo; - static miner_work oGlobalWork; miner_work oWork; std::promise<void> numa_promise; diff --git a/xmrstak/backend/nvidia/nvcc_code/cryptonight.hpp b/xmrstak/backend/nvidia/nvcc_code/cryptonight.hpp index c1e31b9..d588641 100644 --- a/xmrstak/backend/nvidia/nvcc_code/cryptonight.hpp +++ b/xmrstak/backend/nvidia/nvcc_code/cryptonight.hpp @@ -14,7 +14,7 @@ typedef struct { int device_blocks; int device_threads; int device_bfactor; - int device_bsleep; + int device_bsleep; int syncMode; uint32_t *d_input; diff --git a/xmrstak/backend/nvidia/nvcc_code/cuda_blake.hpp b/xmrstak/backend/nvidia/nvcc_code/cuda_blake.hpp index 340174c..611fe1c 100644 --- a/xmrstak/backend/nvidia/nvcc_code/cuda_blake.hpp +++ b/xmrstak/backend/nvidia/nvcc_code/cuda_blake.hpp @@ -95,7 +95,7 @@ __device__ void cn_blake_update(blake_state * S, const uint8_t * data, uint64_ uint32_t left = S->buflen >> 3; uint32_t fill = 64 - left; - if (left && (((datalen >> 3) & 0x3F) >= fill)) + if (left && (((datalen >> 3) & 0x3F) >= fill)) { memcpy((void *) (S->buf + left), (void *) data, fill); S->t[0] += 512; @@ -106,7 +106,7 @@ __device__ void cn_blake_update(blake_state * S, const uint8_t * data, uint64_ left = 0; } - while (datalen >= 512) + while (datalen >= 512) { S->t[0] += 512; if (S->t[0] == 0) S->t[1]++; @@ -115,12 +115,12 @@ __device__ void cn_blake_update(blake_state * S, const uint8_t * data, uint64_ datalen -= 512; } - if (datalen > 0) + if (datalen > 0) { memcpy((void *) (S->buf + left), (void *) data, datalen >> 3); S->buflen = (left << 3) + datalen; } - else + else { S->buflen = 0; } @@ -128,7 +128,7 @@ __device__ void cn_blake_update(blake_state * S, const uint8_t * data, uint64_ __device__ void cn_blake_final(blake_state * S, uint8_t * digest) { - const uint8_t padding[] = + const uint8_t padding[] = { 0x80,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 @@ -141,20 +141,20 @@ __device__ void cn_blake_final(blake_state * S, uint8_t * digest) U32TO8(msglen + 0, hi); U32TO8(msglen + 4, lo); - if (S->buflen == 440) + if (S->buflen == 440) { S->t[0] -= 8; cn_blake_update(S, &pa, 8); - } - else + } + else { - if (S->buflen < 440) + if (S->buflen < 440) { if (S->buflen == 0) S->nullt = 1; S->t[0] -= 440 - S->buflen; cn_blake_update(S, padding, 440 - S->buflen); } - else + else { S->t[0] -= 512 - S->buflen; cn_blake_update(S, padding, 512 - S->buflen); diff --git a/xmrstak/backend/nvidia/nvcc_code/cuda_core.cu b/xmrstak/backend/nvidia/nvcc_code/cuda_core.cu index 43740d2..57b6ad0 100644 --- a/xmrstak/backend/nvidia/nvcc_code/cuda_core.cu +++ b/xmrstak/backend/nvidia/nvcc_code/cuda_core.cu @@ -170,7 +170,7 @@ __forceinline__ __device__ void unusedVar( const T& ) * - for <sm_30 shared memory is needed * * group_n - must be a power of 2! - * + * * @param ptr pointer to shared memory, size must be `threadIdx.x * sizeof(uint32_t)` * value can be NULL for compute architecture >=sm_30 * @param sub thread number within the group, range [0:group_n] @@ -187,7 +187,7 @@ __forceinline__ __device__ uint32_t shuffle(volatile uint32_t* ptr,const uint32_ unusedVar( ptr ); unusedVar( sub ); # if(__CUDACC_VER_MAJOR__ >= 9) - return __shfl_sync(0xFFFFFFFF, val, src, group_n ); + return __shfl_sync(__activemask(), val, src, group_n ); # else return __shfl( val, src, group_n ); # endif @@ -231,7 +231,7 @@ __global__ void cryptonight_core_gpu_phase2( int threads, int bfactor, int parti uint32_t t1[2], t2[2], res; uint32_t tweak1_2[2]; - if (ALGO == cryptonight_monero || ALGO == cryptonight_aeon) + if (ALGO == cryptonight_monero || ALGO == cryptonight_aeon || ALGO == cryptonight_ipbc || ALGO == cryptonight_stellite) { uint32_t * state = d_ctx_state + thread * 50; tweak1_2[0] = (d_input[8] >> 24) | (d_input[9] << 8); @@ -275,16 +275,25 @@ __global__ void cryptonight_core_gpu_phase2( int threads, int bfactor, int parti t1[0] = shuffle<4>(sPtr,sub, d[x], 0); const uint32_t z = d[0] ^ d[1]; - if(ALGO == cryptonight_monero || ALGO == cryptonight_aeon) + if(ALGO == cryptonight_monero || ALGO == cryptonight_aeon || ALGO == cryptonight_ipbc || ALGO == cryptonight_stellite) { const uint32_t table = 0x75310U; - const uint32_t index = ((z >> 26) & 12) | ((z >> 23) & 2); - const uint32_t fork_7 = z ^ ((table >> index) & 0x30U) << 24; - storeGlobal32( long_state + j, sub == 2 ? fork_7 : z ); + if(ALGO == cryptonight_monero || ALGO == cryptonight_aeon || ALGO == cryptonight_ipbc) + { + const uint32_t index = ((z >> 26) & 12) | ((z >> 23) & 2); + const uint32_t fork_7 = z ^ ((table >> index) & 0x30U) << 24; + storeGlobal32( long_state + j, sub == 2 ? fork_7 : z ); + } + else if(ALGO == cryptonight_stellite) + { + const uint32_t index = ((z >> 27) & 12) | ((z >> 23) & 2); + const uint32_t fork_7 = z ^ ((table >> index) & 0x30U) << 24; + storeGlobal32( long_state + j, sub == 2 ? fork_7 : z ); + } } else storeGlobal32( long_state + j, z ); - + //MUL_SUM_XOR_DST(c, a, &long_state[((uint32_t *)c)[0] & MASK]); j = ( ( *t1 & MASK ) >> 2 ) + sub; @@ -303,16 +312,22 @@ __global__ void cryptonight_core_gpu_phase2( int threads, int bfactor, int parti res = *( (uint64_t *) t2 ) >> ( sub & 1 ? 32 : 0 ); - - if(ALGO == cryptonight_monero || ALGO == cryptonight_aeon) + if(ALGO == cryptonight_monero || ALGO == cryptonight_aeon || ALGO == cryptonight_ipbc || ALGO == cryptonight_stellite) { const uint32_t tweaked_res = tweak1_2[sub & 1] ^ res; - const uint32_t long_state_update = sub2 ? tweaked_res : res; + uint32_t long_state_update = sub2 ? tweaked_res : res; + + if (ALGO == cryptonight_ipbc) + { + uint32_t value = shuffle<4>(sPtr,sub, long_state_update, sub & 1) ^ long_state_update; + long_state_update = sub >= 2 ? value : long_state_update; + } + storeGlobal32( long_state + j, long_state_update ); } else storeGlobal32( long_state + j, res ); - + a = ( sub & 1 ? yy[1] : yy[0] ) ^ res; idx0 = shuffle<4>(sPtr,sub, a, 0); if(ALGO == cryptonight_heavy) @@ -363,7 +378,7 @@ __global__ void cryptonight_core_gpu_phase3( int threads, int bfactor, int parti MEMCPY8( text, d_ctx_state + thread * 50 + sub + 16, 2 ); __syncthreads( ); - + #if( __CUDA_ARCH__ < 300 ) extern __shared__ uint32_t shuffleMem[]; volatile uint32_t* sPtr = (volatile uint32_t*)(shuffleMem + (threadIdx.x& 0xFFFFFFF8)); @@ -378,7 +393,7 @@ __global__ void cryptonight_core_gpu_phase3( int threads, int bfactor, int parti text[j] ^= long_state[((IndexType) thread * MEMORY) + ( sub + i + j)]; cn_aes_pseudo_round_mut( sharedMemory, text, key ); - + if(ALGO == cryptonight_heavy) { #pragma unroll @@ -415,7 +430,7 @@ void cryptonight_core_gpu_hash(nvid_ctx* ctx, uint32_t nonce) { CUDA_CHECK_KERNEL(ctx->device_id, cryptonight_core_gpu_phase1<ITERATIONS,MEMORY><<< grid, block8 >>>( ctx->device_blocks*ctx->device_threads, bfactorOneThree, i, - ctx->d_long_state, + ctx->d_long_state, (ALGO == cryptonight_heavy ? ctx->d_ctx_state2 : ctx->d_ctx_state), ctx->d_ctx_key1 )); @@ -458,7 +473,7 @@ void cryptonight_core_gpu_hash(nvid_ctx* ctx, uint32_t nonce) for ( int i = 0; i < roundsPhase3; i++ ) { - CUDA_CHECK_KERNEL(ctx->device_id, cryptonight_core_gpu_phase3<ITERATIONS,MEMORY, ALGO><<< + CUDA_CHECK_KERNEL(ctx->device_id, cryptonight_core_gpu_phase3<ITERATIONS,MEMORY, ALGO><<< grid, block8, block8.x * sizeof(uint32_t) * static_cast< int >( ctx->device_arch[0] < 3 ) @@ -492,5 +507,13 @@ void cryptonight_core_cpu_hash(nvid_ctx* ctx, xmrstak_algo miner_algo, uint32_t { cryptonight_core_gpu_hash<CRYPTONIGHT_LITE_ITER, CRYPTONIGHT_LITE_MASK, CRYPTONIGHT_LITE_MEMORY/4, cryptonight_aeon>(ctx, startNonce); } + else if(miner_algo == cryptonight_ipbc) + { + cryptonight_core_gpu_hash<CRYPTONIGHT_LITE_ITER, CRYPTONIGHT_LITE_MASK, CRYPTONIGHT_LITE_MEMORY/4, cryptonight_ipbc>(ctx, startNonce); + } + else if(miner_algo == cryptonight_stellite) + { + cryptonight_core_gpu_hash<CRYPTONIGHT_ITER, CRYPTONIGHT_MASK, CRYPTONIGHT_MEMORY/4, cryptonight_stellite>(ctx, startNonce); + } } diff --git a/xmrstak/backend/nvidia/nvcc_code/cuda_extra.cu b/xmrstak/backend/nvidia/nvcc_code/cuda_extra.cu index f192f01..304997e 100644 --- a/xmrstak/backend/nvidia/nvcc_code/cuda_extra.cu +++ b/xmrstak/backend/nvidia/nvcc_code/cuda_extra.cu @@ -184,7 +184,7 @@ __global__ void cryptonight_extra_gpu_final( int threads, uint64_t target, uint3 uint32_t * __restrict__ ctx_state = d_ctx_state + thread * 50; uint64_t hash[4]; uint32_t state[50]; - + #pragma unroll for ( i = 0; i < 50; i++ ) state[i] = ctx_state[i]; @@ -296,7 +296,7 @@ extern "C" int cryptonight_extra_cpu_init(nvid_ctx* ctx) } else ctx->d_ctx_state2 = ctx->d_ctx_state; - + CUDA_CHECK(ctx->device_id, cudaMalloc(&ctx->d_ctx_key1, 40 * sizeof(uint32_t) * wsize)); CUDA_CHECK(ctx->device_id, cudaMalloc(&ctx->d_ctx_key2, 40 * sizeof(uint32_t) * wsize)); CUDA_CHECK(ctx->device_id, cudaMalloc(&ctx->d_ctx_text, 32 * sizeof(uint32_t) * wsize)); @@ -472,7 +472,7 @@ extern "C" int cuda_get_deviceinfo(nvid_ctx* ctx) std::vector<int>::iterator it = std::find(arch.begin(), arch.end(), 20); if(it == arch.end()) { - printf("WARNING: NVIDIA GPU %d: miner not compiled for the gpu architecture %d.\n", ctx->device_id, gpuArch); + printf("WARNING: NVIDIA GPU %d: miner not compiled for CUDA architecture %d.\n", ctx->device_id, gpuArch); return 5; } } @@ -490,7 +490,7 @@ extern "C" int cuda_get_deviceinfo(nvid_ctx* ctx) minSupportedArch = arch[i]; if(minSupportedArch < 30 || gpuArch < minSupportedArch) { - printf("WARNING: NVIDIA GPU %d: miner not compiled for the gpu architecture %d.\n", ctx->device_id, gpuArch); + printf("WARNING: NVIDIA GPU %d: miner not compiled for CUDA architecture %d.\n", ctx->device_id, gpuArch); return 5; } } @@ -517,7 +517,7 @@ extern "C" int cuda_get_deviceinfo(nvid_ctx* ctx) */ ctx->device_threads = 64; constexpr size_t byteToMiB = 1024u * 1024u; - + // no limit by default 1TiB size_t maxMemUsage = byteToMiB * byteToMiB; if(props.major == 6) @@ -575,7 +575,7 @@ extern "C" int cuda_get_deviceinfo(nvid_ctx* ctx) CUDA_CHECK(ctx->device_id, cudaFree(tmp)); // delete created context on the gpu CUDA_CHECK(ctx->device_id, cudaDeviceReset()); - + ctx->total_device_memory = totalMemory; ctx->free_device_memory = freeMemory; @@ -614,7 +614,7 @@ extern "C" int cuda_get_deviceinfo(nvid_ctx* ctx) size_t perThread = hashMemSize + 16192u + 680u; if(cryptonight_heavy == ::jconf::inst()->GetCurrentCoinSelection().GetDescription(1).GetMiningAlgo()) perThread += 50 * 4; // state double buffer - + size_t max_intensity = limitedMemory / perThread; ctx->device_threads = max_intensity / ctx->device_blocks; // use only odd number of threads diff --git a/xmrstak/backend/nvidia/nvcc_code/cuda_extra.hpp b/xmrstak/backend/nvidia/nvcc_code/cuda_extra.hpp index 055a8bd..4d369f8 100644 --- a/xmrstak/backend/nvidia/nvcc_code/cuda_extra.hpp +++ b/xmrstak/backend/nvidia/nvcc_code/cuda_extra.hpp @@ -36,8 +36,8 @@ __forceinline__ __device__ uint64_t cuda_ROTL64(const uint64_t value, const int { asm("shf.l.wrap.b32 %0, %1, %2, %3;" : "=r"(result.x) : "r"(__double2loint(__longlong_as_double(value))), "r"(__double2hiint(__longlong_as_double(value))), "r"(offset)); asm("shf.l.wrap.b32 %0, %1, %2, %3;" : "=r"(result.y) : "r"(__double2hiint(__longlong_as_double(value))), "r"(__double2loint(__longlong_as_double(value))), "r"(offset)); - } - else + } + else { asm("shf.l.wrap.b32 %0, %1, %2, %3;" : "=r"(result.x) : "r"(__double2hiint(__longlong_as_double(value))), "r"(__double2loint(__longlong_as_double(value))), "r"(offset)); asm("shf.l.wrap.b32 %0, %1, %2, %3;" : "=r"(result.y) : "r"(__double2loint(__longlong_as_double(value))), "r"(__double2hiint(__longlong_as_double(value))), "r"(offset)); diff --git a/xmrstak/backend/nvidia/nvcc_code/cuda_groestl.hpp b/xmrstak/backend/nvidia/nvcc_code/cuda_groestl.hpp index a37934c..d5a98b7 100644 --- a/xmrstak/backend/nvidia/nvcc_code/cuda_groestl.hpp +++ b/xmrstak/backend/nvidia/nvcc_code/cuda_groestl.hpp @@ -167,7 +167,7 @@ __device__ void cn_groestl_F512(uint32_t * __restrict__ h, const uint32_t * __re uint32_t y[2*GROESTL_COLS512]; uint32_t z[2*GROESTL_COLS512]; - for (i = 0; i < 2*GROESTL_COLS512; i++) + for (i = 0; i < 2*GROESTL_COLS512; i++) { z[i] = m[i]; Ptmp[i] = h[i]^m[i]; @@ -227,23 +227,23 @@ __device__ void cn_groestl_outputtransformation(groestlHashState *ctx) __device__ void cn_groestl_transform(groestlHashState * __restrict__ ctx, const uint8_t * __restrict__ input, int msglen) { - for (; msglen >= GROESTL_SIZE512; msglen -= GROESTL_SIZE512, input += GROESTL_SIZE512) + for (; msglen >= GROESTL_SIZE512; msglen -= GROESTL_SIZE512, input += GROESTL_SIZE512) { cn_groestl_F512(ctx->chaining,(uint32_t*)input); ctx->block_counter1++; - if (ctx->block_counter1 == 0) + if (ctx->block_counter1 == 0) ctx->block_counter2++; } } -__device__ void cn_groestl_final(groestlHashState* __restrict__ ctx, +__device__ void cn_groestl_final(groestlHashState* __restrict__ ctx, BitSequence* __restrict__ output) { int i, j = 0, hashbytelen = GROESTL_HASH_BIT_LEN/8; uint8_t *s = (BitSequence*)ctx->chaining; - if (ctx->bits_in_last_byte) + if (ctx->bits_in_last_byte) { ctx->buffer[(int)ctx->buf_ptr-1] &= ((1<<ctx->bits_in_last_byte)-1)<<(8-ctx->bits_in_last_byte); ctx->buffer[(int)ctx->buf_ptr-1] ^= 0x1<<(7-ctx->bits_in_last_byte); @@ -254,9 +254,9 @@ __device__ void cn_groestl_final(groestlHashState* __restrict__ ctx, ctx->buffer[(int)ctx->buf_ptr++] = 0x80; } - if (ctx->buf_ptr > GROESTL_SIZE512-GROESTL_LENGTHFIELDLEN) + if (ctx->buf_ptr > GROESTL_SIZE512-GROESTL_LENGTHFIELDLEN) { - while (ctx->buf_ptr < GROESTL_SIZE512) + while (ctx->buf_ptr < GROESTL_SIZE512) ctx->buffer[(int)ctx->buf_ptr++] = 0; cn_groestl_transform(ctx, ctx->buffer, GROESTL_SIZE512); @@ -300,14 +300,14 @@ __device__ void cn_groestl_update(groestlHashState* __restrict__ ctx, int msglen = (int)(databitlen/8); int rem = (int)(databitlen%8); - if (ctx->buf_ptr) + if (ctx->buf_ptr) { while (ctx->buf_ptr < GROESTL_SIZE512 && index < msglen) ctx->buffer[(int)ctx->buf_ptr++] = input[index++]; - if (ctx->buf_ptr < GROESTL_SIZE512) + if (ctx->buf_ptr < GROESTL_SIZE512) { - if (rem) + if (rem) { ctx->bits_in_last_byte = rem; ctx->buffer[(int)ctx->buf_ptr++] = input[index]; diff --git a/xmrstak/backend/nvidia/nvcc_code/cuda_jh.hpp b/xmrstak/backend/nvidia/nvcc_code/cuda_jh.hpp index 679046e..284039f 100644 --- a/xmrstak/backend/nvidia/nvcc_code/cuda_jh.hpp +++ b/xmrstak/backend/nvidia/nvcc_code/cuda_jh.hpp @@ -111,7 +111,7 @@ __device__ void cn_jh_E8(jhHashState *state) { uint64_t i,roundnumber,temp0,temp1; - for (roundnumber = 0; roundnumber < 42; roundnumber = roundnumber+7) + for (roundnumber = 0; roundnumber < 42; roundnumber = roundnumber+7) { for (i = 0; i < 2; i++) { @@ -155,13 +155,13 @@ __device__ void cn_jh_E8(jhHashState *state) JH_SWAP32(state->x[1][i]); JH_SWAP32(state->x[3][i]); JH_SWAP32(state->x[5][i]); JH_SWAP32(state->x[7][i]); } - for (i = 0; i < 2; i++) + for (i = 0; i < 2; i++) { JH_SS(state->x[0][i],state->x[2][i],state->x[4][i],state->x[6][i],state->x[1][i],state->x[3][i],state->x[5][i],state->x[7][i],((uint64_t *)d_E8_rc[roundnumber+6])[i],((uint64_t *)d_E8_rc[roundnumber+6])[i+2] ); JH_L(state->x[0][i],state->x[2][i],state->x[4][i],state->x[6][i],state->x[1][i],state->x[3][i],state->x[5][i],state->x[7][i]); } - for (i = 1; i < 8; i = i+2) + for (i = 1; i < 8; i = i+2) { temp0 = state->x[i][0]; state->x[i][0] = state->x[i][1]; @@ -190,7 +190,7 @@ __device__ void cn_jh_update(jhHashState * __restrict__ state, const BitSequence state->databitlen += databitlen; index = 0; - if ( (state->datasize_in_buffer > 0 ) && (( state->datasize_in_buffer + databitlen) < 512) ) + if ( (state->datasize_in_buffer > 0 ) && (( state->datasize_in_buffer + databitlen) < 512) ) { if ( (databitlen & 7) == 0 ) memcpy(state->buffer + (state->datasize_in_buffer >> 3), data, 64-(state->datasize_in_buffer >> 3)); @@ -215,7 +215,7 @@ __device__ void cn_jh_update(jhHashState * __restrict__ state, const BitSequence cn_jh_F8(state); } - if ( databitlen > 0) + if ( databitlen > 0) { if ((databitlen & 7) == 0) memcpy(state->buffer, data+index, (databitlen & 0x1ff) >> 3); @@ -247,7 +247,7 @@ __device__ void cn_jh_final(jhHashState * __restrict__ state, BitSequence * __re state->buffer[56] = (state->databitlen >> 56) & 0xff; cn_jh_F8(state); } - else + else { /*set the rest of the bytes in the buffer to 0*/ if ( (state->datasize_in_buffer & 7) == 0) diff --git a/xmrstak/backend/nvidia/nvcc_code/cuda_skein.hpp b/xmrstak/backend/nvidia/nvcc_code/cuda_skein.hpp index 041a593..fc45db1 100644 --- a/xmrstak/backend/nvidia/nvcc_code/cuda_skein.hpp +++ b/xmrstak/backend/nvidia/nvcc_code/cuda_skein.hpp @@ -221,7 +221,7 @@ __device__ void cn_skein512_processblock(Skein_512_Ctxt_t * __restrict__ ctx, co ctx->X[7] = X7 ^ w[7]; ts[1] &= ~SKEIN_T1_FLAG_FIRST; - } + } while (--blkCnt); ctx->h.T[0] = ts[0]; @@ -239,7 +239,7 @@ __device__ void cn_skein_final(skeinHashState * __restrict__ state, uint8_t * __ ctx->h.T[1] |= SKEIN_T1_FLAG_FINAL; - if (ctx->h.bCnt < SKEIN_512_BLOCK_BYTES) + if (ctx->h.bCnt < SKEIN_512_BLOCK_BYTES) { memset(&ctx->b[ctx->h.bCnt],0,SKEIN_512_BLOCK_BYTES - ctx->h.bCnt); //p8 = &ctx->b[ctx->h.bCnt]; @@ -258,7 +258,7 @@ __device__ void cn_skein_final(skeinHashState * __restrict__ state, uint8_t * __ memcpy(X,ctx->X,sizeof(X)); - for (i=0;i*SKEIN_512_BLOCK_BYTES < byteCnt;i++) + for (i=0;i*SKEIN_512_BLOCK_BYTES < byteCnt;i++) { ((uint64_t *)ctx->b)[0]= (uint64_t)i; Skein_Start_New_Type(ctx,OUT_FINAL); @@ -275,15 +275,15 @@ __device__ void cn_skein512_update(Skein_512_Ctxt_t * __restrict__ ctx, const ui { size_t n; - if (msgByteCnt + ctx->h.bCnt > SKEIN_512_BLOCK_BYTES) + if (msgByteCnt + ctx->h.bCnt > SKEIN_512_BLOCK_BYTES) { - if (ctx->h.bCnt) + if (ctx->h.bCnt) { n = SKEIN_512_BLOCK_BYTES - ctx->h.bCnt; - if (n) + if (n) { memcpy(&ctx->b[ctx->h.bCnt],msg,n); msgByteCnt -= n; @@ -295,7 +295,7 @@ __device__ void cn_skein512_update(Skein_512_Ctxt_t * __restrict__ ctx, const ui ctx->h.bCnt = 0; } - if (msgByteCnt > SKEIN_512_BLOCK_BYTES) + if (msgByteCnt > SKEIN_512_BLOCK_BYTES) { n = (msgByteCnt-1) / SKEIN_512_BLOCK_BYTES; cn_skein512_processblock(ctx,msg,n,SKEIN_512_BLOCK_BYTES); @@ -304,7 +304,7 @@ __device__ void cn_skein512_update(Skein_512_Ctxt_t * __restrict__ ctx, const ui } } - if (msgByteCnt) + if (msgByteCnt) { memcpy(&ctx->b[ctx->h.bCnt],msg,msgByteCnt); ctx->h.bCnt += msgByteCnt; @@ -313,11 +313,11 @@ __device__ void cn_skein512_update(Skein_512_Ctxt_t * __restrict__ ctx, const ui __device__ void cn_skein_update(skeinHashState * __restrict__ state, const BitSequence * __restrict__ data, DataLength databitlen) { - if ((databitlen & 7) == 0) + if ((databitlen & 7) == 0) { cn_skein512_update(&state->u.ctx_512,data,databitlen >> 3); } - else + else { size_t bCnt = (databitlen >> 3) + 1; diff --git a/xmrstak/backend/plugin.hpp b/xmrstak/backend/plugin.hpp index 2610db8..1811af2 100644 --- a/xmrstak/backend/plugin.hpp +++ b/xmrstak/backend/plugin.hpp @@ -27,7 +27,7 @@ namespace xmrstak struct plugin { - plugin(const std::string backendName, const std::string libName) : fn_starterBackend(nullptr), m_backendName(backendName) + plugin(const std::string backendName, const std::string libName) : fn_startBackend(nullptr), m_backendName(backendName) { #ifdef WIN32 libBackend = LoadLibrary(TEXT((libName + ".dll").c_str())); @@ -59,15 +59,15 @@ struct plugin #endif #ifdef WIN32 - fn_starterBackend = (starterBackend_t) GetProcAddress(libBackend, "xmrstak_start_backend"); - if (!fn_starterBackend) + fn_startBackend = (startBackend_t) GetProcAddress(libBackend, "xmrstak_start_backend"); + if (!fn_startBackend) { std::cerr << "WARNING: backend plugin " << libName << " contains no entry 'xmrstak_start_backend': " <<GetLastError()<< std::endl; } #else // reset last error dlerror(); - fn_starterBackend = (starterBackend_t) dlsym(libBackend, "xmrstak_start_backend"); + fn_startBackend = (startBackend_t) dlsym(libBackend, "xmrstak_start_backend"); const char* dlsym_error = dlerror(); if(dlsym_error) { @@ -78,21 +78,21 @@ struct plugin std::vector<iBackend*>* startBackend(uint32_t threadOffset, miner_work& pWork, environment& env) { - if(fn_starterBackend == nullptr) + if(fn_startBackend == nullptr) { std::vector<iBackend*>* pvThreads = new std::vector<iBackend*>(); std::cerr << "WARNING: " << m_backendName << " Backend disabled"<< std::endl; return pvThreads; } - return fn_starterBackend(threadOffset, pWork, env); + return fn_startBackend(threadOffset, pWork, env); } std::string m_backendName; - typedef std::vector<iBackend*>* (*starterBackend_t)(uint32_t threadOffset, miner_work& pWork, environment& env); + typedef std::vector<iBackend*>* (*startBackend_t)(uint32_t threadOffset, miner_work& pWork, environment& env); - starterBackend_t fn_starterBackend; + startBackend_t fn_startBackend; #ifdef WIN32 HINSTANCE libBackend; @@ -100,7 +100,7 @@ struct plugin void *libBackend; #endif -/* \todo add unload to destructor and change usage of plugin that libs keeped open until the miner endss +/* \todo add unload to destructor and change usage of plugin that libs kept open until the miner ends #ifdef WIN32 FreeLibrary(libBackend); #else diff --git a/xmrstak/backend/pool_data.hpp b/xmrstak/backend/pool_data.hpp new file mode 100644 index 0000000..4e92359 --- /dev/null +++ b/xmrstak/backend/pool_data.hpp @@ -0,0 +1,21 @@ +#pragma once + +#include <cstdint> +#include <string> + +constexpr static size_t invalid_pool_id = (-1); + +namespace xmrstak +{ + +struct pool_data +{ + uint32_t iSavedNonce; + size_t pool_id; + + pool_data() : iSavedNonce(0), pool_id(invalid_pool_id) + { + } +}; + +} // namespace xmrstak diff --git a/xmrstak/cli/cli-miner.cpp b/xmrstak/cli/cli-miner.cpp index 6118682..ae39d25 100644 --- a/xmrstak/cli/cli-miner.cpp +++ b/xmrstak/cli/cli-miner.cpp @@ -55,7 +55,7 @@ # include "xmrstak/misc/uac.hpp" #endif // _WIN32 -int do_benchmark(int block_version); +int do_benchmark(int block_version, int wait_sec, int work_sec); void help() { @@ -72,7 +72,9 @@ void help() #ifdef _WIN32 cout<<" --noUAC disable the UAC dialog"<<endl; #endif - cout<<" --benchmark BLOCKVERSION ONLY do a 60-second benchmark and exit"<<endl; + cout<<" --benchmark BLOCKVERSION ONLY do a benchmark and exit"<<endl; + cout<<" --benchwait WAIT_SEC ... benchmark wait time"<<endl; + cout<<" --benchwork WORK_SEC ... benchmark work time"<<endl; #ifndef CONF_NO_CPU cout<<" --noCPU disable the CPU miner backend"<<endl; cout<<" --cpu FILE CPU backend miner config file"<<endl; @@ -110,7 +112,7 @@ void help() #endif std::string algos; jconf::GetAlgoList(algos); - cout<< "Supported coin opitons: " << endl << algos << endl; + cout<< "Supported coin options: " << endl << algos << endl; cout<< "Version: " << get_version_str_short() << endl; cout<<"Brought to by fireice_uk and psychocrypt under GPLv3."<<endl; } @@ -150,7 +152,7 @@ std::string get_multipool_entry(bool& final) std::cin.clear(); std::cin.ignore(INT_MAX,'\n'); std::cout<<"- Password (mostly empty or x):"<<std::endl; getline(std::cin, passwd); - + std::string rigid; std::cout<<"- Rig identifier for pool-side statistics (needs pool support). Can be empty:"<<std::endl; getline(std::cin, rigid); @@ -174,7 +176,7 @@ std::string get_multipool_entry(bool& final) final = !read_yes_no("- Do you want to add another pool? (y/n)"); return "\t{\"pool_address\" : \"" + pool +"\", \"wallet_address\" : \"" + userName + "\", \"rig_id\" : \"" + rigid + - "\", \"pool_password\" : \"" + passwd + "\", \"use_nicehash\" : " + bool_to_str(nicehash) + ", \"use_tls\" : " + + "\", \"pool_password\" : \"" + passwd + "\", \"use_nicehash\" : " + bool_to_str(nicehash) + ", \"use_tls\" : " + bool_to_str(tls) + ", \"tls_fingerprint\" : \"\", \"pool_weight\" : " + std::to_string(pool_weight) + " },\n"; } @@ -213,7 +215,7 @@ void do_guided_pool_config() std::cout << "- Please enter the currency that you want to mine: "<<std::endl; std::cout << list << std::endl; std::cin >> tmp; - } + } currency = tmp; } @@ -301,7 +303,7 @@ void do_guided_pool_config() std::cout << "Miner will mine mostly at the pool with the highest weight, unless the pool fails." << std::endl; std::cout << "Weight must be an integer larger than 0." << std::endl; std::cout << "- Please enter a weight for this pool: "<<std::endl; - + while(!(std::cin >> pool_weight) || pool_weight <= 0) { std::cin.clear(); @@ -314,7 +316,7 @@ void do_guided_pool_config() std::string pool_table; pool_table += "\t{\"pool_address\" : \"" + pool +"\", \"wallet_address\" : \"" + userName + "\", \"rig_id\" : \"" + rigid + - "\", \"pool_password\" : \"" + passwd + "\", \"use_nicehash\" : " + bool_to_str(nicehash) + ", \"use_tls\" : " + + "\", \"pool_password\" : \"" + passwd + "\", \"use_nicehash\" : " + bool_to_str(nicehash) + ", \"use_tls\" : " + bool_to_str(tls) + ", \"tls_fingerprint\" : \"\", \"pool_weight\" : " + std::to_string(pool_weight) + " },\n"; if(multipool) @@ -391,20 +393,20 @@ int main(int argc, char *argv[]) using namespace xmrstak; std::string pathWithName(argv[0]); - std::string seperator("/"); - auto pos = pathWithName.rfind(seperator); + std::string separator("/"); + auto pos = pathWithName.rfind(separator); if(pos == std::string::npos) { // try windows "\" - seperator = "\\"; - pos = pathWithName.rfind(seperator); + separator = "\\"; + pos = pathWithName.rfind(separator); } params::inst().binaryName = std::string(pathWithName, pos + 1, std::string::npos); if(params::inst().binaryName.compare(pathWithName) != 0) { params::inst().executablePrefix = std::string(pathWithName, 0, pos); - params::inst().executablePrefix += seperator; + params::inst().executablePrefix += separator; } params::inst().minerArg0 = argv[0]; @@ -599,7 +601,7 @@ int main(int argc, char *argv[]) win_exit(); return 1; } - + params::inst().userSetRigid = true; params::inst().poolRigid = argv[i]; } @@ -674,6 +676,44 @@ int main(int argc, char *argv[]) } params::inst().benchmark_block_version = bversion; } + else if(opName.compare("--benchwait") == 0) + { + ++i; + if( i >= argc ) + { + printer::inst()->print_msg(L0, "No argument for parameter '--benchwait' given"); + win_exit(); + return 1; + } + char* wait_sec = nullptr; + long int waitsec = strtol(argv[i], &wait_sec, 10); + + if(waitsec < 0 || waitsec >= 300) + { + printer::inst()->print_msg(L0, "Benchmark wait seconds must be in the range [0,300]"); + return 1; + } + params::inst().benchmark_wait_sec = waitsec; + } + else if(opName.compare("--benchwork") == 0) + { + ++i; + if( i >= argc ) + { + printer::inst()->print_msg(L0, "No argument for parameter '--benchwork' given"); + win_exit(); + return 1; + } + char* work_sec = nullptr; + long int worksec = strtol(argv[i], &work_sec, 10); + + if(worksec < 10 || worksec >= 300) + { + printer::inst()->print_msg(L0, "Benchmark work seconds must be in the range [10,300]"); + return 1; + } + params::inst().benchmark_work_sec = worksec; + } else { printer::inst()->print_msg(L0, "Parameter unknown '%s'",argv[i]); @@ -752,9 +792,9 @@ int main(int argc, char *argv[]) if(params::inst().benchmark_block_version >= 0) { printer::inst()->print_str("!!!! Doing only a benchmark and exiting. To mine, remove the '--benchmark' option. !!!!\n"); - return do_benchmark(params::inst().benchmark_block_version); + return do_benchmark(params::inst().benchmark_block_version, params::inst().benchmark_wait_sec, params::inst().benchmark_work_sec); } - + executor::inst()->ex_start(jconf::inst()->DaemonMode()); uint64_t lastTime = get_timestamp_ms(); @@ -789,7 +829,7 @@ int main(int argc, char *argv[]) return 0; } -int do_benchmark(int block_version) +int do_benchmark(int block_version, int wait_sec, int work_sec) { using namespace std::chrono; std::vector<xmrstak::iBackend*>* pvThreads; @@ -805,18 +845,18 @@ int do_benchmark(int block_version) xmrstak::miner_work oWork = xmrstak::miner_work(); pvThreads = xmrstak::BackendConnector::thread_starter(oWork); - printer::inst()->print_msg(L0, "Wait 30 sec until all backends are initialized"); - std::this_thread::sleep_for(std::chrono::seconds(30)); + printer::inst()->print_msg(L0, "Wait %d sec until all backends are initialized",wait_sec); + std::this_thread::sleep_for(std::chrono::seconds(wait_sec)); /* AMD and NVIDIA is currently only supporting work sizes up to 84byte * \todo fix this issue */ xmrstak::miner_work benchWork = xmrstak::miner_work("", work, 84, 0, false, 0); - printer::inst()->print_msg(L0, "Start a 60 second benchmark..."); + printer::inst()->print_msg(L0, "Start a %d second benchmark...",work_sec); xmrstak::globalStates::inst().switch_work(benchWork, dat); uint64_t iStartStamp = get_timestamp_ms(); - std::this_thread::sleep_for(std::chrono::seconds(60)); + std::this_thread::sleep_for(std::chrono::seconds(work_sec)); xmrstak::globalStates::inst().switch_work(oWork, dat); double fTotalHps = 0.0; diff --git a/xmrstak/cli/xmr-stak.manifest b/xmrstak/cli/xmr-stak.manifest index ed65c97..5735693 100644 --- a/xmrstak/cli/xmr-stak.manifest +++ b/xmrstak/cli/xmr-stak.manifest @@ -13,22 +13,22 @@ <requestedExecutionLevel level="asInvoker" uiAccess="false" - /> + /> </requestedPrivileges> </security> </trustInfo> - <compatibility xmlns="urn:schemas-microsoft-com:compatibility.v1"> - <application> - <!-- Windows 10 --> + <compatibility xmlns="urn:schemas-microsoft-com:compatibility.v1"> + <application> + <!-- Windows 10 --> <supportedOS Id="{8e0f7a12-bfb3-4fe8-b9a5-48fd50a15a9a}"/> <!-- Windows 8.1 --> <supportedOS Id="{1f676c76-80e1-4239-95bb-83d0f6d0da78}"/> <!-- Windows Vista --> - <supportedOS Id="{e2011457-1546-43c5-a5fe-008deee3d3f0}"/> + <supportedOS Id="{e2011457-1546-43c5-a5fe-008deee3d3f0}"/> <!-- Windows 7 --> <supportedOS Id="{35138b9a-5d96-4fbd-8e2d-a2440225f93a}"/> <!-- Windows 8 --> <supportedOS Id="{4a2f28e3-53b9-4441-ba9c-d69d4a4a6e38}"/> - </application> + </application> </compatibility> </assembly> diff --git a/xmrstak/config.tpl b/xmrstak/config.tpl index c95d142..34f7656 100644 --- a/xmrstak/config.tpl +++ b/xmrstak/config.tpl @@ -1,8 +1,8 @@ R"===( /* * Network timeouts. - * Because of the way this client is written it doesn't need to constantly talk (keep-alive) to the server to make - * sure it is there. We detect a buggy / overloaded server by the call timeout. The default values will be ok for + * Because of the way this client is written it doesn't need to constantly talk (keep-alive) to the server to make + * sure it is there. We detect a buggy / overloaded server by the call timeout. The default values will be ok for * nearly all cases. If they aren't the pool has most likely overload issues. Low call timeout values are preferable - * long timeouts mean that we waste hashes on potentially stale jobs. Connection report will tell you how long the * server usually takes to process our calls. @@ -148,10 +148,10 @@ R"===( * * http_login - Login. Empty login disables authentication. * http_pass - Password. - */ + */ "http_login" : "", "http_pass" : "", - + /* * prefer_ipv4 - IPv6 preference. If the host is available on both IPv4 and IPv6 net, which one should be choose? * This setting will only be needed in 2020's. No need to worry about it now. @@ -159,4 +159,4 @@ R"===( "prefer_ipv4" : true, )===" - + diff --git a/xmrstak/cpputil/LICENSE.txt b/xmrstak/cpputil/LICENSE.txt new file mode 100644 index 0000000..7615431 --- /dev/null +++ b/xmrstak/cpputil/LICENSE.txt @@ -0,0 +1,21 @@ +MIT License + +Copyright (c) 2018 Will Zhang + +Permission is hereby granted, free of charge, to any person obtaining a copy +of this software and associated documentation files (the "Software"), to deal +in the Software without restriction, including without limitation the rights +to use, copy, modify, merge, publish, distribute, sublicense, and/or sell +copies of the Software, and to permit persons to whom the Software is +furnished to do so, subject to the following conditions: + +The above copyright notice and this permission notice shall be included in all +copies or substantial portions of the Software. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, +OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE +SOFTWARE. diff --git a/xmrstak/cpputil/read_write_lock.h b/xmrstak/cpputil/read_write_lock.h new file mode 100644 index 0000000..9139dfd --- /dev/null +++ b/xmrstak/cpputil/read_write_lock.h @@ -0,0 +1,96 @@ +/* MIT License + * + * Copyright (c) 2018 Will Zhang + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#pragma once + +#include <mutex> +#include <condition_variable> + +namespace cpputil +{ + +class RWLock +{ + public: + RWLock() : status_(0), waiting_readers_(0), waiting_writers_(0) {} + RWLock(const RWLock&) = delete; + RWLock(RWLock&&) = delete; + RWLock& operator = (const RWLock&) = delete; + RWLock& operator = (RWLock&&) = delete; + + void ReadLock() + { + std::unique_lock<std::mutex> lck(mtx_); + waiting_readers_ += 1; + read_cv_.wait(lck, [&]() { return waiting_writers_ == 0 && status_ >= 0; }); + waiting_readers_ -= 1; + status_ += 1; + } + + void WriteLock() + { + std::unique_lock<std::mutex> lck(mtx_); + waiting_writers_ += 1; + write_cv_.wait(lck, [&]() { return status_ == 0; }); + waiting_writers_ -= 1; + status_ = -1; + } + + void UnLock() + { + std::unique_lock<std::mutex> lck(mtx_); + if (status_ == -1) { + status_ = 0; + } + else + { + status_ -= 1; + } + if (waiting_writers_ > 0) + { + if (status_ == 0) + { + write_cv_.notify_one(); + } + } + else + { + read_cv_.notify_all(); + } + } + + private: + /** status of the lock + * -1 : one writer + * 0 : no reader and no writer + * n > 0 : n reader + */ + int32_t status_; + int32_t waiting_readers_; + int32_t waiting_writers_; + std::mutex mtx_; + std::condition_variable read_cv_; + std::condition_variable write_cv_; +}; + +} // namespace cpputil diff --git a/xmrstak/jconf.cpp b/xmrstak/jconf.cpp index 3af4028..b85ddd3 100644 --- a/xmrstak/jconf.cpp +++ b/xmrstak/jconf.cpp @@ -51,8 +51,8 @@ using namespace rapidjson; * This enum needs to match index in oConfigValues, otherwise we will get a runtime error */ enum configEnum { - aPoolList, sCurrency, bTlsSecureAlgo, iCallTimeout, iNetRetry, iGiveUpLimit, iVerboseLevel, bPrintMotd, iAutohashTime, - bFlushStdout, bDaemonMode, sOutputFile, iHttpdPort, sHttpLogin, sHttpPass, bPreferIpv4, bAesOverride, sUseSlowMem + aPoolList, sCurrency, bTlsSecureAlgo, iCallTimeout, iNetRetry, iGiveUpLimit, iVerboseLevel, bPrintMotd, iAutohashTime, + bFlushStdout, bDaemonMode, sOutputFile, iHttpdPort, sHttpLogin, sHttpPass, bPreferIpv4, bAesOverride, sUseSlowMem }; struct configVal { @@ -95,19 +95,24 @@ xmrstak::coin_selection coins[] = { { "cryptonight_heavy", {cryptonight_heavy, cryptonight_heavy, 0u}, {cryptonight_heavy, cryptonight_heavy, 0u}, nullptr }, { "cryptonight_lite", {cryptonight_aeon, cryptonight_lite, 255u}, {cryptonight_aeon, cryptonight_lite, 7u}, nullptr }, { "cryptonight_lite_v7", {cryptonight_lite, cryptonight_aeon, 255u}, {cryptonight_aeon, cryptonight_lite, 7u}, nullptr }, + { "cryptonight_lite_v7_xor", {cryptonight_aeon, cryptonight_ipbc, 255u}, {cryptonight_aeon, cryptonight_aeon, 255u}, nullptr }, { "cryptonight_v7", {cryptonight_monero, cryptonight_monero, 0u}, {cryptonight_monero, cryptonight_monero, 0u}, nullptr }, + { "cryptonight_v7_stellite", {cryptonight_monero, cryptonight_stellite, 255u}, {cryptonight_monero, cryptonight_monero, 255u}, nullptr }, { "edollar", {cryptonight_monero, cryptonight, 255u}, {cryptonight_monero, cryptonight_monero, 0u}, nullptr }, { "electroneum", {cryptonight_monero, cryptonight, 255u}, {cryptonight_monero, cryptonight_monero, 0u}, nullptr }, { "graft", {cryptonight_monero, cryptonight, 8u}, {cryptonight_monero, cryptonight_monero, 0u}, nullptr }, { "haven", {cryptonight_heavy, cryptonight, 2u}, {cryptonight_heavy, cryptonight_heavy, 0u}, nullptr }, { "intense", {cryptonight_monero, cryptonight, 4u}, {cryptonight_monero, cryptonight_monero, 0u}, nullptr }, + { "ipbc", {cryptonight_aeon, cryptonight_ipbc, 255u}, {cryptonight_aeon, cryptonight_aeon, 255u}, nullptr }, { "karbo", {cryptonight_monero, cryptonight, 255u}, {cryptonight_monero, cryptonight_monero, 0u}, nullptr }, + { "masari", {cryptonight_monero, cryptonight, 5u}, {cryptonight_monero, cryptonight_monero, 0u}, nullptr }, { "monero7", {cryptonight_monero, cryptonight_monero, 0u}, {cryptonight_monero, cryptonight_monero, 0u}, "pool.usxmrpool.com:3333" }, - { "stellite", {cryptonight_monero, cryptonight, 3u}, {cryptonight_monero, cryptonight_monero, 0u}, nullptr }, - { "sumokoin", {cryptonight_heavy, cryptonight_heavy, 0u}, {cryptonight_heavy, cryptonight_heavy, 0u}, nullptr } + { "stellite", {cryptonight_stellite, cryptonight_monero, 4u}, {cryptonight_monero, cryptonight_monero, 0u}, nullptr }, + { "sumokoin", {cryptonight_heavy, cryptonight_heavy, 0u}, {cryptonight_heavy, cryptonight_heavy, 0u}, nullptr }, + { "turtlecoin", {cryptonight_lite, cryptonight_aeon, 255u}, {cryptonight_aeon, cryptonight_lite, 7u}, nullptr } }; -constexpr size_t coin_alogo_size = (sizeof(coins)/sizeof(coins[0])); +constexpr size_t coin_algo_size = (sizeof(coins)/sizeof(coins[0])); inline bool checkType(Type have, Type want) { @@ -308,7 +313,7 @@ std::string jconf::GetMiningCoin() void jconf::GetAlgoList(std::string& list) { list.reserve(256); - for(size_t i=0; i < coin_alogo_size; i++) + for(size_t i=0; i < coin_algo_size; i++) { list += "\t- "; list += coins[i].coin_name; @@ -319,7 +324,7 @@ void jconf::GetAlgoList(std::string& list) bool jconf::IsOnAlgoList(std::string& needle) { std::transform(needle.begin(), needle.end(), needle.begin(), ::tolower); - + if(needle == "monero") { printer::inst()->print_msg(L0, "You entered Monero as coin name. Monero will hard-fork the PoW.\nThis means it will stop being compatible with other cryptonight coins.\n" @@ -327,7 +332,7 @@ bool jconf::IsOnAlgoList(std::string& needle) return false; } - for(size_t i=0; i < coin_alogo_size; i++) + for(size_t i=0; i < coin_algo_size; i++) { if(needle == coins[i].coin_name) return true; @@ -338,8 +343,8 @@ bool jconf::IsOnAlgoList(std::string& needle) const char* jconf::GetDefaultPool(const char* needle) { const char* default_example = "pool.example.com:3333"; - - for(size_t i=0; i < coin_alogo_size; i++) + + for(size_t i=0; i < coin_algo_size; i++) { if(strcmp(needle, coins[i].coin_name) == 0) { @@ -510,7 +515,7 @@ bool jconf::parse_config(const char* sFilename, const char* sFilenamePools) for(uint32_t i=0; i < pool_cnt; i++) { const Value& oThdConf = prv->configValues[aPoolList]->GetArray()[i]; - + if(!oThdConf.IsObject()) { printer::inst()->print_msg(L0, "Invalid config file. pool_list must contain objects."); @@ -619,7 +624,7 @@ bool jconf::parse_config(const char* sFilename, const char* sFilenamePools) return false; } - for(size_t i=0; i < coin_alogo_size; i++) + for(size_t i=0; i < coin_algo_size; i++) { if(ctmp == "monero") { diff --git a/xmrstak/misc/configEditor.hpp b/xmrstak/misc/configEditor.hpp index 8a81ad6..d95ea6b 100644 --- a/xmrstak/misc/configEditor.hpp +++ b/xmrstak/misc/configEditor.hpp @@ -14,7 +14,7 @@ struct configEditor { std::string m_fileContent; - configEditor() + configEditor() { } diff --git a/xmrstak/misc/console.cpp b/xmrstak/misc/console.cpp index d961b71..7b14b4f 100644 --- a/xmrstak/misc/console.cpp +++ b/xmrstak/misc/console.cpp @@ -239,7 +239,7 @@ void win_exit(int code) #else void win_exit(int code) -{ +{ std::exit(code); } #endif // _WIN32 diff --git a/xmrstak/misc/executor.cpp b/xmrstak/misc/executor.cpp index 8c454eb..e4b8504 100644 --- a/xmrstak/misc/executor.cpp +++ b/xmrstak/misc/executor.cpp @@ -132,7 +132,7 @@ bool executor::get_live_pools(std::vector<jpsock*>& eval_pools, bool is_dev) if(over_limit == pool_count) { - printer::inst()->print_msg(L0, "All pools are over give up limit. Exitting."); + printer::inst()->print_msg(L0, "All pools are over give up limit. Exiting."); exit(0); } @@ -232,7 +232,7 @@ void executor::eval_pool_choice() else { /* All is good - but check if we can do better */ - std::sort(eval_pools.begin(), eval_pools.end(), [](jpsock* a, jpsock* b) { return b->get_pool_weight(false) < a->get_pool_weight(false); }); + std::sort(eval_pools.begin(), eval_pools.end(), [](jpsock* a, jpsock* b) { return b->get_pool_weight(false) < a->get_pool_weight(false); }); jpsock* goal2 = eval_pools[0]; if(goal->get_pool_id() != goal2->get_pool_id()) @@ -334,7 +334,7 @@ void executor::on_sock_ready(size_t pool_id) std::string str = "Login error: " + pool->get_call_error(); log_socket_error(pool, std::move(str)); } - + if(!pool->have_sock_error()) pool->disconnect(); } @@ -421,7 +421,7 @@ void executor::on_miner_result(size_t pool_id, job_result& oResult) { //Ignore errors silently if(pool->is_running() && pool->is_logged_in()) - pool->cmd_submit(oResult.sJobID, oResult.iNonce, oResult.bResult, backend_name, + pool->cmd_submit(oResult.sJobID, oResult.iNonce, oResult.bResult, backend_name, backend_hashcount, total_hashcount, jconf::inst()->GetCurrentCoinSelection().GetDescription(0).GetMiningAlgo() ); return; @@ -434,7 +434,7 @@ void executor::on_miner_result(size_t pool_id, job_result& oResult) } size_t t_start = get_timestamp_ms(); - bool bResult = pool->cmd_submit(oResult.sJobID, oResult.iNonce, oResult.bResult, + bool bResult = pool->cmd_submit(oResult.sJobID, oResult.iNonce, oResult.bResult, backend_name, backend_hashcount, total_hashcount, jconf::inst()->GetCurrentCoinSelection().GetDescription(1).GetMiningAlgo() ); size_t t_len = get_timestamp_ms() - t_start; @@ -528,12 +528,12 @@ void executor::ex_main() { auto& params = xmrstak::params::inst(); already_have_cli_pool = true; - + const char* wallet = params.poolUsername.empty() ? cfg.sWalletAddr : params.poolUsername.c_str(); const char* rigid = params.userSetRigid ? params.poolRigid.c_str() : cfg.sRigId; const char* pwd = params.userSetPwd ? params.poolPasswd.c_str() : cfg.sPasswd; bool nicehash = cfg.nicehash || params.nicehashMode; - + pools.emplace_back(i+1, cfg.sPoolAddr, wallet, rigid, pwd, 9.9, false, params.poolUseTls, cfg.tls_fingerprint, nicehash); } else @@ -548,7 +548,7 @@ void executor::ex_main() printer::inst()->print_msg(L1, "ERROR: You didn't specify the username / wallet address for %s", xmrstak::params::inst().poolURL.c_str()); win_exit(); } - + pools.emplace_back(i+1, params.poolURL.c_str(), params.poolUsername.c_str(), params.poolRigid.c_str(), params.poolPasswd.c_str(), 9.9, false, params.poolUseTls, "", params.nicehashMode); } @@ -560,13 +560,14 @@ void executor::ex_main() else pools.emplace_front(0, "donate.xmr-stak.net:5555", "", "", "", 0.0, true, false, "", true); break; - + case cryptonight_monero: if(dev_tls) pools.emplace_front(0, "donate.xmr-stak.net:8800", "", "", "", 0.0, true, true, "", false); else pools.emplace_front(0, "donate.xmr-stak.net:5500", "", "", "", 0.0, true, false, "", false); break; + case cryptonight_ipbc: case cryptonight_aeon: case cryptonight_lite: if(dev_tls) @@ -785,7 +786,7 @@ void executor::hashrate_report(std::string& out) auto bType = static_cast<xmrstak::iBackend::BackendType>(b); std::string name(xmrstak::iBackend::getName(bType)); std::transform(name.begin(), name.end(), name.begin(), ::toupper); - + out.append("HASHRATE REPORT - ").append(name).append("\n"); out.append("| ID | 10s | 60s | 15m |"); if(nthd != 1) @@ -808,11 +809,11 @@ void executor::hashrate_report(std::string& out) out.append(hps_format(fHps[0], num, sizeof(num))).append(" |"); out.append(hps_format(fHps[1], num, sizeof(num))).append(" |"); out.append(hps_format(fHps[2], num, sizeof(num))).append(1, ' '); - + fTotal[0] += (std::isnormal(fHps[0])) ? fHps[0] : 0.0; fTotal[1] += (std::isnormal(fHps[1])) ? fHps[1] : 0.0; fTotal[2] += (std::isnormal(fHps[2])) ? fHps[2] : 0.0; - + fTotalCur[0] += (std::isnormal(fHps[0])) ? fHps[0] : 0.0; fTotalCur[1] += (std::isnormal(fHps[1])) ? fHps[1] : 0.0; fTotalCur[2] += (std::isnormal(fHps[2])) ? fHps[2] : 0.0; @@ -823,13 +824,13 @@ void executor::hashrate_report(std::string& out) if((i & 0x1) == 1) //We had odd number of threads out.append("|\n"); - + out.append("Totals (").append(name).append("): "); out.append(hps_format(fTotalCur[0], num, sizeof(num))); out.append(hps_format(fTotalCur[1], num, sizeof(num))); out.append(hps_format(fTotalCur[2], num, sizeof(num))); out.append(" H/s\n"); - + out.append("-----------------------------------------------------------------\n"); } } @@ -1021,7 +1022,7 @@ void executor::http_hashrate_report(std::string& out) out.append(sHtmlMotdBoxStart); have_motd = true; } - + snprintf(buffer, sizeof(buffer), sHtmlMotdEntry, pool.get_pool_addr(), motd.c_str()); out.append(buffer); } diff --git a/xmrstak/misc/telemetry.cpp b/xmrstak/misc/telemetry.cpp index a3a2a12..5642f6b 100644 --- a/xmrstak/misc/telemetry.cpp +++ b/xmrstak/misc/telemetry.cpp @@ -47,14 +47,14 @@ telemetry::telemetry(size_t iThd) } } -double telemetry::calc_telemetry_data(size_t iLastMilisec, size_t iThread) +double telemetry::calc_telemetry_data(size_t iLastMillisec, size_t iThread) { uint64_t iTimeNow = get_timestamp_ms(); uint64_t iEarliestHashCnt = 0; uint64_t iEarliestStamp = 0; - uint64_t iLastestStamp = 0; - uint64_t iLastestHashCnt = 0; + uint64_t iLatestStamp = 0; + uint64_t iLatestHashCnt = 0; bool bHaveFullSet = false; //Start at 1, buckettop points to next empty @@ -65,13 +65,13 @@ double telemetry::calc_telemetry_data(size_t iLastMilisec, size_t iThread) if (ppTimestamps[iThread][idx] == 0) break; //That means we don't have the data yet - if (iLastestStamp == 0) + if (iLatestStamp == 0) { - iLastestStamp = ppTimestamps[iThread][idx]; - iLastestHashCnt = ppHashCounts[iThread][idx]; + iLatestStamp = ppTimestamps[iThread][idx]; + iLatestHashCnt = ppHashCounts[iThread][idx]; } - if (iTimeNow - ppTimestamps[iThread][idx] > iLastMilisec) + if (iTimeNow - ppTimestamps[iThread][idx] > iLastMillisec) { bHaveFullSet = true; break; //We are out of the requested time period @@ -81,16 +81,16 @@ double telemetry::calc_telemetry_data(size_t iLastMilisec, size_t iThread) iEarliestHashCnt = ppHashCounts[iThread][idx]; } - if (!bHaveFullSet || iEarliestStamp == 0 || iLastestStamp == 0) + if (!bHaveFullSet || iEarliestStamp == 0 || iLatestStamp == 0) return nan(""); //Don't think that can happen, but just in case - if (iLastestStamp - iEarliestStamp == 0) + if (iLatestStamp - iEarliestStamp == 0) return nan(""); double fHashes, fTime; - fHashes = static_cast<double>(iLastestHashCnt - iEarliestHashCnt); - fTime = static_cast<double>(iLastestStamp - iEarliestStamp); + fHashes = static_cast<double>(iLatestHashCnt - iEarliestHashCnt); + fTime = static_cast<double>(iLatestStamp - iEarliestStamp); fTime /= 1000.0; return fHashes / fTime; diff --git a/xmrstak/misc/telemetry.hpp b/xmrstak/misc/telemetry.hpp index 2f84dfa..309fd6d 100644 --- a/xmrstak/misc/telemetry.hpp +++ b/xmrstak/misc/telemetry.hpp @@ -11,7 +11,7 @@ class telemetry public: telemetry(size_t iThd); void push_perf_value(size_t iThd, uint64_t iHashCount, uint64_t iTimestamp); - double calc_telemetry_data(size_t iLastMilisec, size_t iThread); + double calc_telemetry_data(size_t iLastMillisec, size_t iThread); private: constexpr static size_t iBucketSize = 2 << 11; //Power of 2 to simplify calculations diff --git a/xmrstak/misc/uac.cpp b/xmrstak/misc/uac.cpp index ad9d394..9f94093 100644 --- a/xmrstak/misc/uac.cpp +++ b/xmrstak/misc/uac.cpp @@ -5,11 +5,11 @@ #include <string> #include <windows.h> -BOOL IsElevated() +BOOL IsElevated() { BOOL fRet = FALSE; HANDLE hToken = NULL; - if (OpenProcessToken(GetCurrentProcess(), TOKEN_QUERY, &hToken)) + if (OpenProcessToken(GetCurrentProcess(), TOKEN_QUERY, &hToken)) { TOKEN_ELEVATION Elevation; DWORD cbSize = sizeof(TOKEN_ELEVATION); diff --git a/xmrstak/net/jpsock.cpp b/xmrstak/net/jpsock.cpp index 74d1c26..6c41f2b 100644 --- a/xmrstak/net/jpsock.cpp +++ b/xmrstak/net/jpsock.cpp @@ -43,8 +43,9 @@ struct jpsock::call_rsp uint64_t iCallId; Value* pCallData; std::string sCallErr; + uint64_t iMessageId; - call_rsp(Value* val) : pCallData(val) + call_rsp(Value* val) : pCallData(val), iMessageId(0) { bHaveResponse = false; iCallId = 0; @@ -215,6 +216,7 @@ void jpsock::jpsock_thread() prv->oCallRsp.bHaveResponse = true; prv->oCallRsp.iCallId = 0; prv->oCallRsp.pCallData = nullptr; + prv->oCallRsp.iMessageId = 0; bCallWaiting = true; } mlock.unlock(); @@ -286,6 +288,7 @@ bool jpsock::process_line(char* line, size_t len) prv->jsonDoc.SetNull(); prv->parseAllocator.Clear(); prv->callAllocator.Clear(); + ++iMessageCnt; /*NULL terminate the line instead of '\n', parsing will add some more NULLs*/ line[len-1] = '\0'; @@ -320,7 +323,7 @@ bool jpsock::process_line(char* line, size_t len) return set_socket_error("PARSE error: Protocol error 2"); opq_json_val v(mt); - return process_pool_job(&v); + return process_pool_job(&v, iMessageCnt); } else { @@ -334,7 +337,7 @@ bool jpsock::process_line(char* line, size_t len) mt = GetObjectMember(prv->jsonDoc, "error"); const char* sError = nullptr; - size_t iErrorLn = 0; + size_t iErrorLen = 0; if (mt == nullptr || mt->IsNull()) { /* If there was no error we need a result */ @@ -351,7 +354,7 @@ bool jpsock::process_line(char* line, size_t len) if(msg == nullptr || !msg->IsString()) return set_socket_error("PARSE error: Protocol error 6"); - iErrorLn = msg->GetStringLength(); + iErrorLen = msg->GetStringLength(); sError = msg->GetString(); } @@ -365,11 +368,12 @@ bool jpsock::process_line(char* line, size_t len) prv->oCallRsp.bHaveResponse = true; prv->oCallRsp.iCallId = iCallId; + prv->oCallRsp.iMessageId = iMessageCnt; if(sError != nullptr) { prv->oCallRsp.pCallData = nullptr; - prv->oCallRsp.sCallErr.assign(sError, iErrorLn); + prv->oCallRsp.sCallErr.assign(sError, iErrorLen); call_error = true; } else @@ -382,8 +386,20 @@ bool jpsock::process_line(char* line, size_t len) } } -bool jpsock::process_pool_job(const opq_json_val* params) +bool jpsock::process_pool_job(const opq_json_val* params, const uint64_t messageId) { + std::unique_lock<std::mutex> mlock(job_mutex); + if(messageId < iLastMessageId) + { + /* In the case where the processed job message id is lesser than the last + * processed job message id we skip the processing to avoid mining old jobs + */ + return true; + } + iLastMessageId = messageId; + + mlock.unlock(); + if (!params->val->IsObject()) return set_socket_error("PARSE error: Job error 1"); @@ -399,18 +415,45 @@ bool jpsock::process_pool_job(const opq_json_val* params) return set_socket_error("PARSE error: Job error 2"); } + if(motd != nullptr && motd->IsString() && (motd->GetStringLength() & 0x01) == 0) + { + std::unique_lock<std::mutex> lck(motd_mutex); + if(motd->GetStringLength() > 0) + { + pool_motd.resize(motd->GetStringLength()/2 + 1); + if(!hex2bin(motd->GetString(), motd->GetStringLength(), (unsigned char*)&pool_motd.front())) + pool_motd.clear(); + } + else + pool_motd.clear(); + } + if (jobid->GetStringLength() >= sizeof(pool_job::sJobID)) // Note >= return set_socket_error("PARSE error: Job error 3"); - uint32_t iWorkLn = blob->GetStringLength() / 2; - if (iWorkLn > sizeof(pool_job::bWorkBlob)) - return set_socket_error("PARSE error: Invalid job legth. Are you sure you are mining the correct coin?"); - pool_job oPoolJob; - if (!hex2bin(blob->GetString(), iWorkLn * 2, oPoolJob.bWorkBlob)) + + const uint32_t iWorkLen = blob->GetStringLength() / 2; + oPoolJob.iWorkLen = iWorkLen; + + if (iWorkLen > sizeof(pool_job::bWorkBlob)) + return set_socket_error("PARSE error: Invalid job length. Are you sure you are mining the correct coin?"); + + if (!hex2bin(blob->GetString(), iWorkLen * 2, oPoolJob.bWorkBlob)) return set_socket_error("PARSE error: Job error 4"); - oPoolJob.iWorkLen = iWorkLn; + // lock reading of oCurrentJob + std::unique_lock<std::mutex> jobIdLock(job_mutex); + // compare possible non equal length job id's + if(iWorkLen == oCurrentJob.iWorkLen && + memcmp(oPoolJob.bWorkBlob, oCurrentJob.bWorkBlob, iWorkLen) == 0 && + strcmp(jobid->GetString(), oCurrentJob.sJobID) == 0 + ) + { + return set_socket_error("Duplicate equal job detected! Please contact your pool admin."); + } + jobIdLock.unlock(); + memset(oPoolJob.sJobID, 0, sizeof(pool_job::sJobID)); memcpy(oPoolJob.sJobID, jobid->GetString(), jobid->GetStringLength()); //Bounds checking at proto error 3 @@ -423,7 +466,7 @@ bool jpsock::process_pool_job(const opq_json_val* params) if(!hex2bin(sTempStr, 8, (unsigned char*)&iTempInt) || iTempInt == 0) return set_socket_error("PARSE error: Invalid target"); - + oPoolJob.iTarget = t32_to_t64(iTempInt); } else if(target_slen <= 16) @@ -437,25 +480,14 @@ bool jpsock::process_pool_job(const opq_json_val* params) else return set_socket_error("PARSE error: Job error 5"); - if(motd != nullptr && motd->IsString() && (motd->GetStringLength() & 0x01) == 0) - { - std::unique_lock<std::mutex> lck(motd_mutex); - if(motd->GetStringLength() > 0) - { - pool_motd.resize(motd->GetStringLength()/2 + 1); - if(!hex2bin(motd->GetString(), motd->GetStringLength(), (unsigned char*)&pool_motd.front())) - pool_motd.clear(); - } - else - pool_motd.clear(); - } - iJobDiff = t64_to_diff(oPoolJob.iTarget); - executor::inst()->push_event(ex_event(oPoolJob, pool_id)); - std::unique_lock<std::mutex> lck(job_mutex); oCurrentJob = oPoolJob; + lck.unlock(); + // send event after current job data are updated + executor::inst()->push_event(ex_event(oPoolJob, pool_id)); + return true; } @@ -498,7 +530,7 @@ void jpsock::disconnect(bool quiet) quiet_close = false; } -bool jpsock::cmd_ret_wait(const char* sPacket, opq_json_val& poResult) +bool jpsock::cmd_ret_wait(const char* sPacket, opq_json_val& poResult, uint64_t& messageId) { //printf("SEND: %s\n", sPacket); @@ -538,8 +570,10 @@ bool jpsock::cmd_ret_wait(const char* sPacket, opq_json_val& poResult) } if(bSuccess) + { poResult.val = &prv->oCallValue; - + messageId = prv->oCallRsp.iMessageId; + } return bSuccess; } @@ -551,9 +585,10 @@ bool jpsock::cmd_login() usr_login.c_str(), usr_pass.c_str(), usr_rigid.c_str(), get_version_str().c_str()); opq_json_val oResult(nullptr); + uint64_t messageId = 0; /*Normal error conditions (failed login etc..) will end here*/ - if (!cmd_ret_wait(cmd_buffer, oResult)) + if (!cmd_ret_wait(cmd_buffer, oResult, messageId)) return false; if (!oResult.val->IsObject()) @@ -589,7 +624,7 @@ bool jpsock::cmd_login() for(size_t i=0; i < ext->Size(); i++) { const Value& jextname = ext->GetArray()[i]; - + if(!jextname.IsString()) continue; @@ -608,7 +643,7 @@ bool jpsock::cmd_login() } opq_json_val v(job); - if(!process_pool_job(&v)) + if(!process_pool_job(&v, messageId)) { disconnect(); return false; @@ -645,16 +680,22 @@ bool jpsock::cmd_submit(const char* sJobId, uint32_t iNonce, const uint8_t* bRes algo_name = "cryptonight"; break; case cryptonight_lite: - algo_name = "cryptonight-lite"; + algo_name = "cryptonight_lite"; break; case cryptonight_monero: - algo_name = "cryptonight-monerov7"; + algo_name = "cryptonight_v7"; break; case cryptonight_aeon: - algo_name = "cryptonight-aeonv7"; + algo_name = "cryptonight_lite_v7"; + break; + case cryptonight_stellite: + algo_name = "cryptonight_v7_stellite"; + break; + case cryptonight_ipbc: + algo_name = "cryptonight_lite_v7_xor"; break; case cryptonight_heavy: - algo_name = "cryptonight-heavy"; + algo_name = "cryptonight_heavy"; break; default: algo_name = "unknown"; @@ -673,8 +714,9 @@ bool jpsock::cmd_submit(const char* sJobId, uint32_t iNonce, const uint8_t* bRes snprintf(cmd_buffer, sizeof(cmd_buffer), "{\"method\":\"submit\",\"params\":{\"id\":\"%s\",\"job_id\":\"%s\",\"nonce\":\"%s\",\"result\":\"%s\"%s%s%s},\"id\":1}\n", sMinerId, sJobId, sNonce, sResult, sBackend, sHashcount, sAlgo); + uint64_t messageId = 0; opq_json_val oResult(nullptr); - return cmd_ret_wait(cmd_buffer, oResult); + return cmd_ret_wait(cmd_buffer, oResult, messageId); } void jpsock::save_nonce(uint32_t nonce) @@ -696,7 +738,7 @@ bool jpsock::get_current_job(pool_job& job) bool jpsock::get_pool_motd(std::string& strin) { - if(!ext_motd) + if(!ext_motd) return false; std::unique_lock<std::mutex> lck(motd_mutex); diff --git a/xmrstak/net/jpsock.hpp b/xmrstak/net/jpsock.hpp index 2ddeeee..ad34f6c 100644 --- a/xmrstak/net/jpsock.hpp +++ b/xmrstak/net/jpsock.hpp @@ -15,7 +15,7 @@ - Parsing or connection error Those are fatal errors (we drop the connection if we encounter them). After they are constructed from const char* strings from various places. - (can be from read-only mem), we passs them in an exectutor message + (can be from read-only mem), we pass them in an executor message once the recv thread expires. - Call error This error happens when the "server says no". Usually because the job was @@ -40,9 +40,9 @@ public: static bool hex2bin(const char* in, unsigned int len, unsigned char* out); static void bin2hex(const unsigned char* in, unsigned int len, char* out); - inline double get_pool_weight(bool gross_weight) - { - double ret = pool_weight; + inline double get_pool_weight(bool gross_weight) + { + double ret = pool_weight; if(gross_weight && bRunning) ret += 10.0; if(gross_weight && bLoggedIn) @@ -124,8 +124,8 @@ private: void jpsock_thread(); bool jpsock_thd_main(); bool process_line(char* line, size_t len); - bool process_pool_job(const opq_json_val* params); - bool cmd_ret_wait(const char* sPacket, opq_json_val& poResult); + bool process_pool_job(const opq_json_val* params, const uint64_t messageId); + bool cmd_ret_wait(const char* sPacket, opq_json_val& poResult, uint64_t& messageId); char sMinerId[64]; std::atomic<uint64_t> iJobDiff; @@ -142,5 +142,8 @@ private: opaque_private* prv; base_socket* sck; + + uint64_t iMessageCnt = 0; + uint64_t iLastMessageId = 0; }; diff --git a/xmrstak/net/msgstruct.hpp b/xmrstak/net/msgstruct.hpp index e401f59..20092fe 100644 --- a/xmrstak/net/msgstruct.hpp +++ b/xmrstak/net/msgstruct.hpp @@ -5,7 +5,7 @@ #include <assert.h> // Structures that we use to pass info between threads constructors are here just to make -// the stack allocation take up less space, heap is a shared resouce that needs locks too of course +// the stack allocation take up less space, heap is a shared resource that needs locks too of course struct pool_job { @@ -72,15 +72,15 @@ struct gpu_res_err }; enum ex_event_name { EV_INVALID_VAL, EV_SOCK_READY, EV_SOCK_ERROR, EV_GPU_RES_ERROR, - EV_POOL_HAVE_JOB, EV_MINER_HAVE_RESULT, EV_PERF_TICK, EV_EVAL_POOL_CHOICE, - EV_USR_HASHRATE, EV_USR_RESULTS, EV_USR_CONNSTAT, EV_HASHRATE_LOOP, + EV_POOL_HAVE_JOB, EV_MINER_HAVE_RESULT, EV_PERF_TICK, EV_EVAL_POOL_CHOICE, + EV_USR_HASHRATE, EV_USR_RESULTS, EV_USR_CONNSTAT, EV_HASHRATE_LOOP, EV_HTML_HASHRATE, EV_HTML_RESULTS, EV_HTML_CONNSTAT, EV_HTML_JSON }; /* This is how I learned to stop worrying and love c++11 =). Ghosts of endless heap allocations have finally been exorcised. Thanks to the nifty magic of move semantics, string will only be allocated - once on the heap. Considering that it makes a jorney across stack, + once on the heap. Considering that it makes a journey across stack, heap alloced queue, to another stack before being finally processed I think it is kind of nifty, don't you? Also note that for non-arg events we only copy two qwords diff --git a/xmrstak/net/socket.cpp b/xmrstak/net/socket.cpp index 9bc608f..6fcb454 100644 --- a/xmrstak/net/socket.cpp +++ b/xmrstak/net/socket.cpp @@ -339,7 +339,7 @@ bool tls_socket::connect() { if(!pCallback->is_dev_pool()) { - printer::inst()->print_msg(L0, "FINGERPRINT FAILED CHECK [%s] %.*s was given, %s was configured", + printer::inst()->print_msg(L0, "FINGERPRINT FAILED CHECK [%s] %.*s was given, %s was configured", pCallback->get_pool_addr(), (int)b64_len, b64_md, conf_md); } diff --git a/xmrstak/params.hpp b/xmrstak/params.hpp index 6e676cf..16089ae 100644 --- a/xmrstak/params.hpp +++ b/xmrstak/params.hpp @@ -54,6 +54,8 @@ struct params // block_version >= 0 enable benchmark int benchmark_block_version = -1; + int benchmark_wait_sec = 30; + int benchmark_work_sec = 60; params() : binaryName("xmr-stak"), diff --git a/xmrstak/pools.tpl b/xmrstak/pools.tpl index 58ca8b5..50299cd 100644 --- a/xmrstak/pools.tpl +++ b/xmrstak/pools.tpl @@ -7,12 +7,12 @@ R"===( * use_nicehash - Limit the nonce to 3 bytes as required by nicehash. * use_tls - This option will make us connect using Transport Layer Security. * tls_fingerprint - Server's SHA256 fingerprint. If this string is non-empty then we will check the server's cert against it. - * pool_weight - Pool weight is a number telling the miner how important the pool is. Miner will mine mostly at the pool + * pool_weight - Pool weight is a number telling the miner how important the pool is. Miner will mine mostly at the pool * with the highest weight, unless the pool fails. Weight must be an integer larger than 0. * * We feature pools up to 1MH/s. For a more complete list see M5M400's pool list at www.moneropools.com */ - + "pool_list" : [ POOLCONF], @@ -28,15 +28,19 @@ POOLCONF], * graft * haven * intense + * ipbc * karbo + * masari * monero7 (use this for Monero's new PoW) * sumokoin (automatic switch with block version 3 to cryptonight_heavy) + * turtlecoin * * Native algorithms which not depends on any block versions: * * # 1MiB scratchpad memory * cryptonight_lite * cryptonight_lite_v7 + * cryptonight_lite_v7_xor (algorithm used by ipbc) * # 2MiB scratchpad memory * cryptonight * cryptonight_v7 @@ -47,4 +51,4 @@ POOLCONF], "currency" : "CURRENCY", )===" - + diff --git a/xmrstak/rapidjson/allocators.h b/xmrstak/rapidjson/allocators.h index 98affe0..cf5f536 100644 --- a/xmrstak/rapidjson/allocators.h +++ b/xmrstak/rapidjson/allocators.h @@ -1,5 +1,5 @@ // Tencent is pleased to support the open source community by making RapidJSON available. -// +// // Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. All rights reserved. // // Licensed under the MIT License (the "License"); you may not use this file except @@ -7,9 +7,9 @@ // // http://opensource.org/licenses/MIT // -// Unless required by applicable law or agreed to in writing, software distributed -// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR -// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the // specific language governing permissions and limitations under the License. #ifndef RAPIDJSON_ALLOCATORS_H_ @@ -24,10 +24,10 @@ RAPIDJSON_NAMESPACE_BEGIN /*! \class rapidjson::Allocator \brief Concept for allocating, resizing and freeing memory block. - + Note that Malloc() and Realloc() are non-static but Free() is static. - - So if an allocator need to support Free(), it needs to put its pointer in + + So if an allocator need to support Free(), it needs to put its pointer in the header of memory block. \code @@ -62,7 +62,7 @@ concept Allocator { class CrtAllocator { public: static const bool kNeedFree = true; - void* Malloc(size_t size) { + void* Malloc(size_t size) { if (size) // behavior of malloc(0) is implementation defined. return std::malloc(size); else @@ -83,7 +83,7 @@ public: // MemoryPoolAllocator //! Default memory allocator used by the parser and DOM. -/*! This allocator allocate memory blocks from pre-allocated memory chunks. +/*! This allocator allocate memory blocks from pre-allocated memory chunks. It does not free memory blocks. And Realloc() only allocate new memory. @@ -107,7 +107,7 @@ public: /*! \param chunkSize The size of memory chunk. The default is kDefaultChunkSize. \param baseAllocator The allocator for allocating memory chunks. */ - MemoryPoolAllocator(size_t chunkSize = kDefaultChunkCapacity, BaseAllocator* baseAllocator = 0) : + MemoryPoolAllocator(size_t chunkSize = kDefaultChunkCapacity, BaseAllocator* baseAllocator = 0) : chunkHead_(0), chunk_capacity_(chunkSize), userBuffer_(0), baseAllocator_(baseAllocator), ownBaseAllocator_(0) { } diff --git a/xmrstak/rapidjson/document.h b/xmrstak/rapidjson/document.h index 895af88..48ff019 100644 --- a/xmrstak/rapidjson/document.h +++ b/xmrstak/rapidjson/document.h @@ -1,5 +1,5 @@ // Tencent is pleased to support the open source community by making RapidJSON available. -// +// // Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. All rights reserved. // // Licensed under the MIT License (the "License"); you may not use this file except @@ -7,9 +7,9 @@ // // http://opensource.org/licenses/MIT // -// Unless required by applicable law or agreed to in writing, software distributed -// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR -// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the // specific language governing permissions and limitations under the License. #ifndef RAPIDJSON_DOCUMENT_H_ @@ -75,8 +75,8 @@ class GenericDocument; But a compiler (IBM XL C/C++ for AIX) have reported to have problem with that so it moved as a namespace scope struct. https://code.google.com/p/rapidjson/issues/detail?id=64 */ -template <typename Encoding, typename Allocator> -struct GenericMember { +template <typename Encoding, typename Allocator> +struct GenericMember { GenericValue<Encoding, Allocator> name; //!< name of member (must be a string) GenericValue<Encoding, Allocator> value; //!< value of member. }; @@ -418,7 +418,7 @@ namespace internal { template <typename ValueType, typename T> struct TypeHelper {}; -template<typename ValueType> +template<typename ValueType> struct TypeHelper<ValueType, bool> { static bool Is(const ValueType& v) { return v.IsBool(); } static bool Get(const ValueType& v) { return v.GetBool(); } @@ -426,7 +426,7 @@ struct TypeHelper<ValueType, bool> { static ValueType& Set(ValueType& v, bool data, typename ValueType::AllocatorType&) { return v.SetBool(data); } }; -template<typename ValueType> +template<typename ValueType> struct TypeHelper<ValueType, int> { static bool Is(const ValueType& v) { return v.IsInt(); } static int Get(const ValueType& v) { return v.GetInt(); } @@ -434,7 +434,7 @@ struct TypeHelper<ValueType, int> { static ValueType& Set(ValueType& v, int data, typename ValueType::AllocatorType&) { return v.SetInt(data); } }; -template<typename ValueType> +template<typename ValueType> struct TypeHelper<ValueType, unsigned> { static bool Is(const ValueType& v) { return v.IsUint(); } static unsigned Get(const ValueType& v) { return v.GetUint(); } @@ -442,7 +442,7 @@ struct TypeHelper<ValueType, unsigned> { static ValueType& Set(ValueType& v, unsigned data, typename ValueType::AllocatorType&) { return v.SetUint(data); } }; -template<typename ValueType> +template<typename ValueType> struct TypeHelper<ValueType, int64_t> { static bool Is(const ValueType& v) { return v.IsInt64(); } static int64_t Get(const ValueType& v) { return v.GetInt64(); } @@ -450,7 +450,7 @@ struct TypeHelper<ValueType, int64_t> { static ValueType& Set(ValueType& v, int64_t data, typename ValueType::AllocatorType&) { return v.SetInt64(data); } }; -template<typename ValueType> +template<typename ValueType> struct TypeHelper<ValueType, uint64_t> { static bool Is(const ValueType& v) { return v.IsUint64(); } static uint64_t Get(const ValueType& v) { return v.GetUint64(); } @@ -458,7 +458,7 @@ struct TypeHelper<ValueType, uint64_t> { static ValueType& Set(ValueType& v, uint64_t data, typename ValueType::AllocatorType&) { return v.SetUint64(data); } }; -template<typename ValueType> +template<typename ValueType> struct TypeHelper<ValueType, double> { static bool Is(const ValueType& v) { return v.IsDouble(); } static double Get(const ValueType& v) { return v.GetDouble(); } @@ -466,7 +466,7 @@ struct TypeHelper<ValueType, double> { static ValueType& Set(ValueType& v, double data, typename ValueType::AllocatorType&) { return v.SetDouble(data); } }; -template<typename ValueType> +template<typename ValueType> struct TypeHelper<ValueType, float> { static bool Is(const ValueType& v) { return v.IsFloat(); } static float Get(const ValueType& v) { return v.GetFloat(); } @@ -474,7 +474,7 @@ struct TypeHelper<ValueType, float> { static ValueType& Set(ValueType& v, float data, typename ValueType::AllocatorType&) { return v.SetFloat(data); } }; -template<typename ValueType> +template<typename ValueType> struct TypeHelper<ValueType, const typename ValueType::Ch*> { typedef const typename ValueType::Ch* StringType; static bool Is(const ValueType& v) { return v.IsString(); } @@ -484,7 +484,7 @@ struct TypeHelper<ValueType, const typename ValueType::Ch*> { }; #if RAPIDJSON_HAS_STDSTRING -template<typename ValueType> +template<typename ValueType> struct TypeHelper<ValueType, std::basic_string<typename ValueType::Ch> > { typedef std::basic_string<typename ValueType::Ch> StringType; static bool Is(const ValueType& v) { return v.IsString(); } @@ -493,7 +493,7 @@ struct TypeHelper<ValueType, std::basic_string<typename ValueType::Ch> > { }; #endif -template<typename ValueType> +template<typename ValueType> struct TypeHelper<ValueType, typename ValueType::Array> { typedef typename ValueType::Array ArrayType; static bool Is(const ValueType& v) { return v.IsArray(); } @@ -502,14 +502,14 @@ struct TypeHelper<ValueType, typename ValueType::Array> { static ValueType& Set(ValueType& v, ArrayType data, typename ValueType::AllocatorType&) { return v = data; } }; -template<typename ValueType> +template<typename ValueType> struct TypeHelper<ValueType, typename ValueType::ConstArray> { typedef typename ValueType::ConstArray ArrayType; static bool Is(const ValueType& v) { return v.IsArray(); } static ArrayType Get(const ValueType& v) { return v.GetArray(); } }; -template<typename ValueType> +template<typename ValueType> struct TypeHelper<ValueType, typename ValueType::Object> { typedef typename ValueType::Object ObjectType; static bool Is(const ValueType& v) { return v.IsObject(); } @@ -518,7 +518,7 @@ struct TypeHelper<ValueType, typename ValueType::Object> { static ValueType& Set(ValueType& v, ObjectType data, typename ValueType::AllocatorType&) { v = data; } }; -template<typename ValueType> +template<typename ValueType> struct TypeHelper<ValueType, typename ValueType::ConstObject> { typedef typename ValueType::ConstObject ObjectType; static bool Is(const ValueType& v) { return v.IsObject(); } @@ -544,7 +544,7 @@ template <bool, typename> class GenericObject; \tparam Encoding Encoding of the value. (Even non-string values need to have the same encoding in a document) \tparam Allocator Allocator type for allocating memory of object, array and string. */ -template <typename Encoding, typename Allocator = MemoryPoolAllocator<> > +template <typename Encoding, typename Allocator = MemoryPoolAllocator<> > class GenericValue { public: //! Name-value pair in an object. @@ -685,7 +685,7 @@ public: //! Constructor for unsigned value. explicit GenericValue(unsigned u) RAPIDJSON_NOEXCEPT : data_() { - data_.n.u64 = u; + data_.n.u64 = u; data_.f.flags = (u & 0x80000000) ? kNumberUintFlag : (kNumberUintFlag | kIntFlag | kInt64Flag); } @@ -907,14 +907,14 @@ public: switch (GetType()) { case kObjectType: // Warning: O(n^2) inner-loop if (data_.o.size != rhs.data_.o.size) - return false; + return false; for (ConstMemberIterator lhsMemberItr = MemberBegin(); lhsMemberItr != MemberEnd(); ++lhsMemberItr) { typename RhsType::ConstMemberIterator rhsMemberItr = rhs.FindMember(lhsMemberItr->name); if (rhsMemberItr == rhs.MemberEnd() || lhsMemberItr->value != rhsMemberItr->value) return false; } return true; - + case kArrayType: if (data_.a.size != rhs.data_.a.size) return false; @@ -1385,7 +1385,7 @@ public: \note Linear time complexity. */ void RemoveAllMembers() { - RAPIDJSON_ASSERT(IsObject()); + RAPIDJSON_ASSERT(IsObject()); for (MemberIterator m = MemberBegin(); m != MemberEnd(); ++m) m->~Member(); data_.o.size = 0; @@ -1531,7 +1531,7 @@ public: \note Linear time complexity. */ void Clear() { - RAPIDJSON_ASSERT(IsArray()); + RAPIDJSON_ASSERT(IsArray()); GenericValue* e = GetElementsPointer(); for (GenericValue* v = e; v != e + data_.a.size; ++v) v->~GenericValue(); @@ -1678,7 +1678,7 @@ public: RAPIDJSON_ASSERT(last <= End()); ValueIterator pos = Begin() + (first - Begin()); for (ValueIterator itr = pos; itr != last; ++itr) - itr->~GenericValue(); + itr->~GenericValue(); std::memmove(pos, last, static_cast<size_t>(End() - last) * sizeof(GenericValue)); data_.a.size -= static_cast<SizeType>(last - first); return pos; @@ -1737,7 +1737,7 @@ public: //! Set this value as a string without copying source string. /*! This version has better performance with supplied length, and also support string containing null character. - \param s source string pointer. + \param s source string pointer. \param length The length of source string, excluding the trailing null terminator. \return The value itself for fluent API. \post IsString() == true && GetString() == s && GetStringLength() == length @@ -1754,7 +1754,7 @@ public: //! Set this value as a string by copying from source string. /*! This version has better performance with supplied length, and also support string containing null character. - \param s source string. + \param s source string. \param length The length of source string, excluding the trailing null terminator. \param allocator Allocator for allocating copied buffer. Commonly use GenericDocument::GetAllocator(). \return The value itself for fluent API. @@ -1763,7 +1763,7 @@ public: GenericValue& SetString(const Ch* s, SizeType length, Allocator& allocator) { this->~GenericValue(); SetStringRaw(StringRef(s, length), allocator); return *this; } //! Set this value as a string by copying from source string. - /*! \param s source string. + /*! \param s source string. \param allocator Allocator for allocating copied buffer. Commonly use GenericDocument::GetAllocator(). \return The value itself for fluent API. \post IsString() == true && GetString() != s && strcmp(GetString(),s) == 0 && GetStringLength() == length @@ -1840,10 +1840,10 @@ public: if (RAPIDJSON_UNLIKELY(!v->Accept(handler))) return false; return handler.EndArray(data_.a.size); - + case kStringType: return handler.String(GetString(), GetStringLength(), (data_.f.flags & kCopyFlag) != 0); - + default: RAPIDJSON_ASSERT(GetType() == kNumberType); if (IsDouble()) return handler.Double(data_.n.d); @@ -2060,7 +2060,7 @@ private: typedef GenericValue<UTF8<> > Value; /////////////////////////////////////////////////////////////////////////////// -// GenericDocument +// GenericDocument //! A document for parsing JSON text as DOM. /*! @@ -2092,12 +2092,12 @@ public: } //! Constructor - /*! Creates an empty document which type is Null. + /*! Creates an empty document which type is Null. \param allocator Optional allocator for allocating memory. \param stackCapacity Optional initial capacity of stack in bytes. \param stackAllocator Optional allocator for allocating memory for stack. */ - GenericDocument(Allocator* allocator = 0, size_t stackCapacity = kDefaultStackCapacity, StackAllocator* stackAllocator = 0) : + GenericDocument(Allocator* allocator = 0, size_t stackCapacity = kDefaultStackCapacity, StackAllocator* stackAllocator = 0) : allocator_(allocator), ownAllocator_(0), stack_(stackAllocator, stackCapacity), parseResult_() { if (!allocator_) @@ -2303,7 +2303,7 @@ public: GenericDocument& Parse(const Ch* str, size_t length) { return Parse<parseFlags, Encoding>(str, length); } - + GenericDocument& Parse(const Ch* str, size_t length) { return Parse<kParseDefaultFlags>(str, length); } @@ -2323,7 +2323,7 @@ public: GenericDocument& Parse(const std::basic_string<Ch>& str) { return Parse<kParseDefaultFlags>(str); } -#endif // RAPIDJSON_HAS_STDSTRING +#endif // RAPIDJSON_HAS_STDSTRING //!@} @@ -2388,16 +2388,16 @@ public: bool Uint64(uint64_t i) { new (stack_.template Push<ValueType>()) ValueType(i); return true; } bool Double(double d) { new (stack_.template Push<ValueType>()) ValueType(d); return true; } - bool RawNumber(const Ch* str, SizeType length, bool copy) { - if (copy) + bool RawNumber(const Ch* str, SizeType length, bool copy) { + if (copy) new (stack_.template Push<ValueType>()) ValueType(str, length, GetAllocator()); else new (stack_.template Push<ValueType>()) ValueType(str, length); return true; } - bool String(const Ch* str, SizeType length, bool copy) { - if (copy) + bool String(const Ch* str, SizeType length, bool copy) { + if (copy) new (stack_.template Push<ValueType>()) ValueType(str, length, GetAllocator()); else new (stack_.template Push<ValueType>()) ValueType(str, length); @@ -2405,7 +2405,7 @@ public: } bool StartObject() { new (stack_.template Push<ValueType>()) ValueType(kObjectType); return true; } - + bool Key(const Ch* str, SizeType length, bool copy) { return String(str, length, copy); } bool EndObject(SizeType memberCount) { @@ -2415,7 +2415,7 @@ public: } bool StartArray() { new (stack_.template Push<ValueType>()) ValueType(kArrayType); return true; } - + bool EndArray(SizeType elementCount) { ValueType* elements = stack_.template Pop<ValueType>(elementCount); stack_.template Top<ValueType>()->SetArrayRaw(elements, elementCount, GetAllocator()); diff --git a/xmrstak/rapidjson/encodedstream.h b/xmrstak/rapidjson/encodedstream.h index 1450683..a294617 100644 --- a/xmrstak/rapidjson/encodedstream.h +++ b/xmrstak/rapidjson/encodedstream.h @@ -1,5 +1,5 @@ // Tencent is pleased to support the open source community by making RapidJSON available. -// +// // Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. All rights reserved. // // Licensed under the MIT License (the "License"); you may not use this file except @@ -7,9 +7,9 @@ // // http://opensource.org/licenses/MIT // -// Unless required by applicable law or agreed to in writing, software distributed -// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR -// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the // specific language governing permissions and limitations under the License. #ifndef RAPIDJSON_ENCODEDSTREAM_H_ @@ -41,7 +41,7 @@ class EncodedInputStream { public: typedef typename Encoding::Ch Ch; - EncodedInputStream(InputByteStream& is) : is_(is) { + EncodedInputStream(InputByteStream& is) : is_(is) { current_ = Encoding::TakeBOM(is_); } @@ -51,7 +51,7 @@ public: // Not implemented void Put(Ch) { RAPIDJSON_ASSERT(false); } - void Flush() { RAPIDJSON_ASSERT(false); } + void Flush() { RAPIDJSON_ASSERT(false); } Ch* PutBegin() { RAPIDJSON_ASSERT(false); return 0; } size_t PutEnd(Ch*) { RAPIDJSON_ASSERT(false); return 0; } @@ -80,7 +80,7 @@ public: // Not implemented void Put(Ch) {} - void Flush() {} + void Flush() {} Ch* PutBegin() { return 0; } size_t PutEnd(Ch*) { return 0; } @@ -102,7 +102,7 @@ class EncodedOutputStream { public: typedef typename Encoding::Ch Ch; - EncodedOutputStream(OutputByteStream& os, bool putBOM = true) : os_(os) { + EncodedOutputStream(OutputByteStream& os, bool putBOM = true) : os_(os) { if (putBOM) Encoding::PutBOM(os_); } @@ -143,7 +143,7 @@ public: \param type UTF encoding type if it is not detected from the stream. */ AutoUTFInputStream(InputByteStream& is, UTFType type = kUTF8) : is_(&is), type_(type), hasBOM_(false) { - RAPIDJSON_ASSERT(type >= kUTF8 && type <= kUTF32BE); + RAPIDJSON_ASSERT(type >= kUTF8 && type <= kUTF32BE); DetectType(); static const TakeFunc f[] = { RAPIDJSON_ENCODINGS_FUNC(Take) }; takeFunc_ = f[type_]; @@ -159,7 +159,7 @@ public: // Not implemented void Put(Ch) { RAPIDJSON_ASSERT(false); } - void Flush() { RAPIDJSON_ASSERT(false); } + void Flush() { RAPIDJSON_ASSERT(false); } Ch* PutBegin() { RAPIDJSON_ASSERT(false); return 0; } size_t PutEnd(Ch*) { RAPIDJSON_ASSERT(false); return 0; } @@ -258,7 +258,7 @@ public: UTFType GetType() const { return type_; } void Put(Ch c) { putFunc_(*os_, c); } - void Flush() { os_->Flush(); } + void Flush() { os_->Flush(); } // Not implemented Ch Peek() const { RAPIDJSON_ASSERT(false); return 0;} @@ -271,7 +271,7 @@ private: AutoUTFOutputStream(const AutoUTFOutputStream&); AutoUTFOutputStream& operator=(const AutoUTFOutputStream&); - void PutBOM() { + void PutBOM() { typedef void (*PutBOMFunc)(OutputByteStream&); static const PutBOMFunc f[] = { RAPIDJSON_ENCODINGS_FUNC(PutBOM) }; f[type_](*os_); diff --git a/xmrstak/rapidjson/encodings.h b/xmrstak/rapidjson/encodings.h index baa7c2b..9b4da1d 100644 --- a/xmrstak/rapidjson/encodings.h +++ b/xmrstak/rapidjson/encodings.h @@ -1,5 +1,5 @@ // Tencent is pleased to support the open source community by making RapidJSON available. -// +// // Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. All rights reserved. // // Licensed under the MIT License (the "License"); you may not use this file except @@ -7,9 +7,9 @@ // // http://opensource.org/licenses/MIT // -// Unless required by applicable law or agreed to in writing, software distributed -// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR -// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the // specific language governing permissions and limitations under the License. #ifndef RAPIDJSON_ENCODINGS_H_ @@ -100,7 +100,7 @@ struct UTF8 { template<typename OutputStream> static void Encode(OutputStream& os, unsigned codepoint) { - if (codepoint <= 0x7F) + if (codepoint <= 0x7F) os.Put(static_cast<Ch>(codepoint & 0xFF)); else if (codepoint <= 0x7FF) { os.Put(static_cast<Ch>(0xC0 | ((codepoint >> 6) & 0xFF))); @@ -122,7 +122,7 @@ struct UTF8 { template<typename OutputStream> static void EncodeUnsafe(OutputStream& os, unsigned codepoint) { - if (codepoint <= 0x7F) + if (codepoint <= 0x7F) PutUnsafe(os, static_cast<Ch>(codepoint & 0xFF)); else if (codepoint <= 0x7FF) { PutUnsafe(os, static_cast<Ch>(0xC0 | ((codepoint >> 6) & 0xFF))); @@ -276,7 +276,7 @@ struct UTF16 { static void Encode(OutputStream& os, unsigned codepoint) { RAPIDJSON_STATIC_ASSERT(sizeof(typename OutputStream::Ch) >= 2); if (codepoint <= 0xFFFF) { - RAPIDJSON_ASSERT(codepoint < 0xD800 || codepoint > 0xDFFF); // Code point itself cannot be surrogate pair + RAPIDJSON_ASSERT(codepoint < 0xD800 || codepoint > 0xDFFF); // Code point itself cannot be surrogate pair os.Put(static_cast<typename OutputStream::Ch>(codepoint)); } else { @@ -292,7 +292,7 @@ struct UTF16 { static void EncodeUnsafe(OutputStream& os, unsigned codepoint) { RAPIDJSON_STATIC_ASSERT(sizeof(typename OutputStream::Ch) >= 2); if (codepoint <= 0xFFFF) { - RAPIDJSON_ASSERT(codepoint < 0xD800 || codepoint > 0xDFFF); // Code point itself cannot be surrogate pair + RAPIDJSON_ASSERT(codepoint < 0xD800 || codepoint > 0xDFFF); // Code point itself cannot be surrogate pair PutUnsafe(os, static_cast<typename OutputStream::Ch>(codepoint)); } else { @@ -406,7 +406,7 @@ struct UTF16BE : UTF16<CharType> { /////////////////////////////////////////////////////////////////////////////// // UTF32 -//! UTF-32 encoding. +//! UTF-32 encoding. /*! http://en.wikipedia.org/wiki/UTF-32 \tparam CharType Type for storing 32-bit UTF-32 data. Default is unsigned. C++11 may use char32_t instead. \note implements Encoding concept @@ -498,7 +498,7 @@ struct UTF32BE : UTF32<CharType> { static CharType TakeBOM(InputByteStream& is) { RAPIDJSON_STATIC_ASSERT(sizeof(typename InputByteStream::Ch) == 1); CharType c = Take(is); - return static_cast<uint32_t>(c) == 0x0000FEFFu ? Take(is) : c; + return static_cast<uint32_t>(c) == 0x0000FEFFu ? Take(is) : c; } template <typename InputByteStream> @@ -694,13 +694,13 @@ struct Transcoder<Encoding, Encoding> { os.Put(is.Take()); // Just copy one code unit. This semantic is different from primary template class. return true; } - + template<typename InputStream, typename OutputStream> RAPIDJSON_FORCEINLINE static bool TranscodeUnsafe(InputStream& is, OutputStream& os) { PutUnsafe(os, is.Take()); // Just copy one code unit. This semantic is different from primary template class. return true; } - + template<typename InputStream, typename OutputStream> RAPIDJSON_FORCEINLINE static bool Validate(InputStream& is, OutputStream& os) { return Encoding::Validate(is, os); // source/target encoding are the same diff --git a/xmrstak/rapidjson/error/en.h b/xmrstak/rapidjson/error/en.h index 2db838b..3f87159 100644 --- a/xmrstak/rapidjson/error/en.h +++ b/xmrstak/rapidjson/error/en.h @@ -1,5 +1,5 @@ // Tencent is pleased to support the open source community by making RapidJSON available. -// +// // Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. All rights reserved. // // Licensed under the MIT License (the "License"); you may not use this file except @@ -7,9 +7,9 @@ // // http://opensource.org/licenses/MIT // -// Unless required by applicable law or agreed to in writing, software distributed -// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR -// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the // specific language governing permissions and limitations under the License. #ifndef RAPIDJSON_ERROR_EN_H_ @@ -39,13 +39,13 @@ inline const RAPIDJSON_ERROR_CHARTYPE* GetParseError_En(ParseErrorCode parseErro case kParseErrorDocumentEmpty: return RAPIDJSON_ERROR_STRING("The document is empty."); case kParseErrorDocumentRootNotSingular: return RAPIDJSON_ERROR_STRING("The document root must not be followed by other values."); - + case kParseErrorValueInvalid: return RAPIDJSON_ERROR_STRING("Invalid value."); - + case kParseErrorObjectMissName: return RAPIDJSON_ERROR_STRING("Missing a name for object member."); case kParseErrorObjectMissColon: return RAPIDJSON_ERROR_STRING("Missing a colon after a name of object member."); case kParseErrorObjectMissCommaOrCurlyBracket: return RAPIDJSON_ERROR_STRING("Missing a comma or '}' after an object member."); - + case kParseErrorArrayMissCommaOrSquareBracket: return RAPIDJSON_ERROR_STRING("Missing a comma or ']' after an array element."); case kParseErrorStringUnicodeEscapeInvalidHex: return RAPIDJSON_ERROR_STRING("Incorrect hex digit after \\u escape in string."); diff --git a/xmrstak/rapidjson/error/error.h b/xmrstak/rapidjson/error/error.h index 95cb31a..e4c5714 100644 --- a/xmrstak/rapidjson/error/error.h +++ b/xmrstak/rapidjson/error/error.h @@ -1,5 +1,5 @@ // Tencent is pleased to support the open source community by making RapidJSON available. -// +// // Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. All rights reserved. // // Licensed under the MIT License (the "License"); you may not use this file except @@ -7,9 +7,9 @@ // // http://opensource.org/licenses/MIT // -// Unless required by applicable law or agreed to in writing, software distributed -// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR -// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the // specific language governing permissions and limitations under the License. #ifndef RAPIDJSON_ERROR_ERROR_H_ diff --git a/xmrstak/rapidjson/filereadstream.h b/xmrstak/rapidjson/filereadstream.h index b56ea13..793c5de 100644 --- a/xmrstak/rapidjson/filereadstream.h +++ b/xmrstak/rapidjson/filereadstream.h @@ -1,5 +1,5 @@ // Tencent is pleased to support the open source community by making RapidJSON available. -// +// // Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. All rights reserved. // // Licensed under the MIT License (the "License"); you may not use this file except @@ -7,9 +7,9 @@ // // http://opensource.org/licenses/MIT // -// Unless required by applicable law or agreed to in writing, software distributed -// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR -// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the // specific language governing permissions and limitations under the License. #ifndef RAPIDJSON_FILEREADSTREAM_H_ @@ -41,7 +41,7 @@ public: \param buffer user-supplied buffer. \param bufferSize size of buffer in bytes. Must >=4 bytes. */ - FileReadStream(std::FILE* fp, char* buffer, size_t bufferSize) : fp_(fp), buffer_(buffer), bufferSize_(bufferSize), bufferLast_(0), current_(buffer_), readCount_(0), count_(0), eof_(false) { + FileReadStream(std::FILE* fp, char* buffer, size_t bufferSize) : fp_(fp), buffer_(buffer), bufferSize_(bufferSize), bufferLast_(0), current_(buffer_), readCount_(0), count_(0), eof_(false) { RAPIDJSON_ASSERT(fp_ != 0); RAPIDJSON_ASSERT(bufferSize >= 4); Read(); @@ -53,7 +53,7 @@ public: // Not implemented void Put(Ch) { RAPIDJSON_ASSERT(false); } - void Flush() { RAPIDJSON_ASSERT(false); } + void Flush() { RAPIDJSON_ASSERT(false); } Ch* PutBegin() { RAPIDJSON_ASSERT(false); return 0; } size_t PutEnd(Ch*) { RAPIDJSON_ASSERT(false); return 0; } diff --git a/xmrstak/rapidjson/filewritestream.h b/xmrstak/rapidjson/filewritestream.h index 6378dd6..2d3dbab 100644 --- a/xmrstak/rapidjson/filewritestream.h +++ b/xmrstak/rapidjson/filewritestream.h @@ -1,5 +1,5 @@ // Tencent is pleased to support the open source community by making RapidJSON available. -// +// // Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. All rights reserved. // // Licensed under the MIT License (the "License"); you may not use this file except @@ -7,9 +7,9 @@ // // http://opensource.org/licenses/MIT // -// Unless required by applicable law or agreed to in writing, software distributed -// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR -// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the // specific language governing permissions and limitations under the License. #ifndef RAPIDJSON_FILEWRITESTREAM_H_ @@ -33,11 +33,11 @@ class FileWriteStream { public: typedef char Ch; //!< Character type. Only support char. - FileWriteStream(std::FILE* fp, char* buffer, size_t bufferSize) : fp_(fp), buffer_(buffer), bufferEnd_(buffer + bufferSize), current_(buffer_) { + FileWriteStream(std::FILE* fp, char* buffer, size_t bufferSize) : fp_(fp), buffer_(buffer), bufferEnd_(buffer + bufferSize), current_(buffer_) { RAPIDJSON_ASSERT(fp_ != 0); } - void Put(char c) { + void Put(char c) { if (current_ >= bufferEnd_) Flush(); diff --git a/xmrstak/rapidjson/fwd.h b/xmrstak/rapidjson/fwd.h index e8104e8..54480df 100644 --- a/xmrstak/rapidjson/fwd.h +++ b/xmrstak/rapidjson/fwd.h @@ -1,5 +1,5 @@ // Tencent is pleased to support the open source community by making RapidJSON available. -// +// // Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. All rights reserved. // // Licensed under the MIT License (the "License"); you may not use this file except @@ -7,9 +7,9 @@ // // http://opensource.org/licenses/MIT // -// Unless required by applicable law or agreed to in writing, software distributed -// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR -// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the // specific language governing permissions and limitations under the License. #ifndef RAPIDJSON_FWD_H_ @@ -101,7 +101,7 @@ class PrettyWriter; // document.h -template <typename Encoding, typename Allocator> +template <typename Encoding, typename Allocator> struct GenericMember; template <bool Const, typename Encoding, typename Allocator> @@ -110,7 +110,7 @@ class GenericMemberIterator; template<typename CharType> struct GenericStringRef; -template <typename Encoding, typename Allocator> +template <typename Encoding, typename Allocator> class GenericValue; typedef GenericValue<UTF8<char>, MemoryPoolAllocator<CrtAllocator> > Value; diff --git a/xmrstak/rapidjson/internal/biginteger.h b/xmrstak/rapidjson/internal/biginteger.h index 9d3e88c..f095992 100644 --- a/xmrstak/rapidjson/internal/biginteger.h +++ b/xmrstak/rapidjson/internal/biginteger.h @@ -1,5 +1,5 @@ // Tencent is pleased to support the open source community by making RapidJSON available. -// +// // Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. All rights reserved. // // Licensed under the MIT License (the "License"); you may not use this file except @@ -7,9 +7,9 @@ // // http://opensource.org/licenses/MIT // -// Unless required by applicable law or agreed to in writing, software distributed -// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR -// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the // specific language governing permissions and limitations under the License. #ifndef RAPIDJSON_BIGINTEGER_H_ @@ -51,7 +51,7 @@ public: if (length > 0) AppendDecimal64(decimals + i, decimals + i + length); } - + BigInteger& operator=(const BigInteger &rhs) { if (this != &rhs) { @@ -60,9 +60,9 @@ public: } return *this; } - + BigInteger& operator=(uint64_t u) { - digits_[0] = u; + digits_[0] = u; count_ = 1; return *this; } @@ -95,7 +95,7 @@ public: digits_[i] = MulAdd64(digits_[i], u, k, &hi); k = hi; } - + if (k > 0) PushBack(k); @@ -118,7 +118,7 @@ public: digits_[i] = (p0 & 0xFFFFFFFF) | (p1 << 32); k = p1 >> 32; } - + if (k > 0) PushBack(k); diff --git a/xmrstak/rapidjson/internal/diyfp.h b/xmrstak/rapidjson/internal/diyfp.h index c9fefdc..71a6d02 100644 --- a/xmrstak/rapidjson/internal/diyfp.h +++ b/xmrstak/rapidjson/internal/diyfp.h @@ -1,5 +1,5 @@ // Tencent is pleased to support the open source community by making RapidJSON available. -// +// // Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. All rights reserved. // // Licensed under the MIT License (the "License"); you may not use this file except @@ -7,9 +7,9 @@ // // http://opensource.org/licenses/MIT // -// Unless required by applicable law or agreed to in writing, software distributed -// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR -// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the // specific language governing permissions and limitations under the License. // This is a C++ header-only implementation of Grisu2 algorithm from the publication: @@ -56,7 +56,7 @@ struct DiyFp { if (biased_e != 0) { f = significand + kDpHiddenBit; e = biased_e - kDpExponentBias; - } + } else { f = significand; e = kDpMinExponent + 1; @@ -141,7 +141,7 @@ struct DiyFp { double d; uint64_t u64; }u; - const uint64_t be = (e == kDpDenormalExponent && (f & kDpHiddenBit) == 0) ? 0 : + const uint64_t be = (e == kDpDenormalExponent && (f & kDpHiddenBit) == 0) ? 0 : static_cast<uint64_t>(e + kDpExponentBias); u.u64 = (f & kDpSignificandMask) | (be << kDpSignificandSize); return u.d; @@ -222,7 +222,7 @@ inline DiyFp GetCachedPowerByIndex(size_t index) { }; return DiyFp(kCachedPowers_F[index], kCachedPowers_E[index]); } - + inline DiyFp GetCachedPower(int e, int* K) { //int k = static_cast<int>(ceil((-61 - e) * 0.30102999566398114)) + 374; diff --git a/xmrstak/rapidjson/internal/dtoa.h b/xmrstak/rapidjson/internal/dtoa.h index 8d6350e..1edc39b 100644 --- a/xmrstak/rapidjson/internal/dtoa.h +++ b/xmrstak/rapidjson/internal/dtoa.h @@ -1,5 +1,5 @@ // Tencent is pleased to support the open source community by making RapidJSON available. -// +// // Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. All rights reserved. // // Licensed under the MIT License (the "License"); you may not use this file except @@ -7,9 +7,9 @@ // // http://opensource.org/licenses/MIT // -// Unless required by applicable law or agreed to in writing, software distributed -// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR -// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the // specific language governing permissions and limitations under the License. // This is a C++ header-only implementation of Grisu2 algorithm from the publication: diff --git a/xmrstak/rapidjson/internal/ieee754.h b/xmrstak/rapidjson/internal/ieee754.h index 82bb0b9..32fdd21 100644 --- a/xmrstak/rapidjson/internal/ieee754.h +++ b/xmrstak/rapidjson/internal/ieee754.h @@ -1,5 +1,5 @@ // Tencent is pleased to support the open source community by making RapidJSON available. -// +// // Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. All rights reserved. // // Licensed under the MIT License (the "License"); you may not use this file except @@ -7,9 +7,9 @@ // // http://opensource.org/licenses/MIT // -// Unless required by applicable law or agreed to in writing, software distributed -// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR -// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the // specific language governing permissions and limitations under the License. #ifndef RAPIDJSON_IEEE754_ diff --git a/xmrstak/rapidjson/internal/itoa.h b/xmrstak/rapidjson/internal/itoa.h index 01a4e7e..f2304a7 100644 --- a/xmrstak/rapidjson/internal/itoa.h +++ b/xmrstak/rapidjson/internal/itoa.h @@ -1,5 +1,5 @@ // Tencent is pleased to support the open source community by making RapidJSON available. -// +// // Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. All rights reserved. // // Licensed under the MIT License (the "License"); you may not use this file except @@ -7,9 +7,9 @@ // // http://opensource.org/licenses/MIT // -// Unless required by applicable law or agreed to in writing, software distributed -// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR -// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the // specific language governing permissions and limitations under the License. #ifndef RAPIDJSON_ITOA_ @@ -42,7 +42,7 @@ inline char* u32toa(uint32_t value, char* buffer) { if (value < 10000) { const uint32_t d1 = (value / 100) << 1; const uint32_t d2 = (value % 100) << 1; - + if (value >= 1000) *buffer++ = cDigitsLut[d1]; if (value >= 100) @@ -55,13 +55,13 @@ inline char* u32toa(uint32_t value, char* buffer) { // value = bbbbcccc const uint32_t b = value / 10000; const uint32_t c = value % 10000; - + const uint32_t d1 = (b / 100) << 1; const uint32_t d2 = (b % 100) << 1; - + const uint32_t d3 = (c / 100) << 1; const uint32_t d4 = (c % 100) << 1; - + if (value >= 10000000) *buffer++ = cDigitsLut[d1]; if (value >= 1000000) @@ -69,7 +69,7 @@ inline char* u32toa(uint32_t value, char* buffer) { if (value >= 100000) *buffer++ = cDigitsLut[d2]; *buffer++ = cDigitsLut[d2 + 1]; - + *buffer++ = cDigitsLut[d3]; *buffer++ = cDigitsLut[d3 + 1]; *buffer++ = cDigitsLut[d4]; @@ -77,10 +77,10 @@ inline char* u32toa(uint32_t value, char* buffer) { } else { // value = aabbbbcccc in decimal - + const uint32_t a = value / 100000000; // 1 to 42 value %= 100000000; - + if (a >= 10) { const unsigned i = a << 1; *buffer++ = cDigitsLut[i]; @@ -91,13 +91,13 @@ inline char* u32toa(uint32_t value, char* buffer) { const uint32_t b = value / 10000; // 0 to 9999 const uint32_t c = value % 10000; // 0 to 9999 - + const uint32_t d1 = (b / 100) << 1; const uint32_t d2 = (b % 100) << 1; - + const uint32_t d3 = (c / 100) << 1; const uint32_t d4 = (c % 100) << 1; - + *buffer++ = cDigitsLut[d1]; *buffer++ = cDigitsLut[d1 + 1]; *buffer++ = cDigitsLut[d2]; @@ -131,13 +131,13 @@ inline char* u64toa(uint64_t value, char* buffer) { const uint64_t kTen14 = kTen8 * 1000000; const uint64_t kTen15 = kTen8 * 10000000; const uint64_t kTen16 = kTen8 * kTen8; - + if (value < kTen8) { uint32_t v = static_cast<uint32_t>(value); if (v < 10000) { const uint32_t d1 = (v / 100) << 1; const uint32_t d2 = (v % 100) << 1; - + if (v >= 1000) *buffer++ = cDigitsLut[d1]; if (v >= 100) @@ -150,13 +150,13 @@ inline char* u64toa(uint64_t value, char* buffer) { // value = bbbbcccc const uint32_t b = v / 10000; const uint32_t c = v % 10000; - + const uint32_t d1 = (b / 100) << 1; const uint32_t d2 = (b % 100) << 1; - + const uint32_t d3 = (c / 100) << 1; const uint32_t d4 = (c % 100) << 1; - + if (value >= 10000000) *buffer++ = cDigitsLut[d1]; if (value >= 1000000) @@ -164,7 +164,7 @@ inline char* u64toa(uint64_t value, char* buffer) { if (value >= 100000) *buffer++ = cDigitsLut[d2]; *buffer++ = cDigitsLut[d2 + 1]; - + *buffer++ = cDigitsLut[d3]; *buffer++ = cDigitsLut[d3 + 1]; *buffer++ = cDigitsLut[d4]; @@ -174,22 +174,22 @@ inline char* u64toa(uint64_t value, char* buffer) { else if (value < kTen16) { const uint32_t v0 = static_cast<uint32_t>(value / kTen8); const uint32_t v1 = static_cast<uint32_t>(value % kTen8); - + const uint32_t b0 = v0 / 10000; const uint32_t c0 = v0 % 10000; - + const uint32_t d1 = (b0 / 100) << 1; const uint32_t d2 = (b0 % 100) << 1; - + const uint32_t d3 = (c0 / 100) << 1; const uint32_t d4 = (c0 % 100) << 1; const uint32_t b1 = v1 / 10000; const uint32_t c1 = v1 % 10000; - + const uint32_t d5 = (b1 / 100) << 1; const uint32_t d6 = (b1 % 100) << 1; - + const uint32_t d7 = (c1 / 100) << 1; const uint32_t d8 = (c1 % 100) << 1; @@ -209,7 +209,7 @@ inline char* u64toa(uint64_t value, char* buffer) { *buffer++ = cDigitsLut[d4]; if (value >= kTen8) *buffer++ = cDigitsLut[d4 + 1]; - + *buffer++ = cDigitsLut[d5]; *buffer++ = cDigitsLut[d5 + 1]; *buffer++ = cDigitsLut[d6]; @@ -222,7 +222,7 @@ inline char* u64toa(uint64_t value, char* buffer) { else { const uint32_t a = static_cast<uint32_t>(value / kTen16); // 1 to 1844 value %= kTen16; - + if (a < 10) *buffer++ = static_cast<char>('0' + static_cast<char>(a)); else if (a < 100) { @@ -232,7 +232,7 @@ inline char* u64toa(uint64_t value, char* buffer) { } else if (a < 1000) { *buffer++ = static_cast<char>('0' + static_cast<char>(a / 100)); - + const uint32_t i = (a % 100) << 1; *buffer++ = cDigitsLut[i]; *buffer++ = cDigitsLut[i + 1]; @@ -245,28 +245,28 @@ inline char* u64toa(uint64_t value, char* buffer) { *buffer++ = cDigitsLut[j]; *buffer++ = cDigitsLut[j + 1]; } - + const uint32_t v0 = static_cast<uint32_t>(value / kTen8); const uint32_t v1 = static_cast<uint32_t>(value % kTen8); - + const uint32_t b0 = v0 / 10000; const uint32_t c0 = v0 % 10000; - + const uint32_t d1 = (b0 / 100) << 1; const uint32_t d2 = (b0 % 100) << 1; - + const uint32_t d3 = (c0 / 100) << 1; const uint32_t d4 = (c0 % 100) << 1; - + const uint32_t b1 = v1 / 10000; const uint32_t c1 = v1 % 10000; - + const uint32_t d5 = (b1 / 100) << 1; const uint32_t d6 = (b1 % 100) << 1; - + const uint32_t d7 = (c1 / 100) << 1; const uint32_t d8 = (c1 % 100) << 1; - + *buffer++ = cDigitsLut[d1]; *buffer++ = cDigitsLut[d1 + 1]; *buffer++ = cDigitsLut[d2]; @@ -284,7 +284,7 @@ inline char* u64toa(uint64_t value, char* buffer) { *buffer++ = cDigitsLut[d8]; *buffer++ = cDigitsLut[d8 + 1]; } - + return buffer; } diff --git a/xmrstak/rapidjson/internal/meta.h b/xmrstak/rapidjson/internal/meta.h index 5a9aaa4..6c74b31 100644 --- a/xmrstak/rapidjson/internal/meta.h +++ b/xmrstak/rapidjson/internal/meta.h @@ -1,5 +1,5 @@ // Tencent is pleased to support the open source community by making RapidJSON available. -// +// // Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. All rights reserved. // // Licensed under the MIT License (the "License"); you may not use this file except @@ -7,9 +7,9 @@ // // http://opensource.org/licenses/MIT // -// Unless required by applicable law or agreed to in writing, software distributed -// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR -// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the // specific language governing permissions and limitations under the License. #ifndef RAPIDJSON_INTERNAL_META_H_ diff --git a/xmrstak/rapidjson/internal/pow10.h b/xmrstak/rapidjson/internal/pow10.h index 02f475d..b593982 100644 --- a/xmrstak/rapidjson/internal/pow10.h +++ b/xmrstak/rapidjson/internal/pow10.h @@ -1,5 +1,5 @@ // Tencent is pleased to support the open source community by making RapidJSON available. -// +// // Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. All rights reserved. // // Licensed under the MIT License (the "License"); you may not use this file except @@ -7,9 +7,9 @@ // // http://opensource.org/licenses/MIT // -// Unless required by applicable law or agreed to in writing, software distributed -// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR -// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the // specific language governing permissions and limitations under the License. #ifndef RAPIDJSON_POW10_ @@ -27,8 +27,8 @@ namespace internal { */ inline double Pow10(int n) { static const double e[] = { // 1e-0...1e308: 309 * 8 bytes = 2472 bytes - 1e+0, - 1e+1, 1e+2, 1e+3, 1e+4, 1e+5, 1e+6, 1e+7, 1e+8, 1e+9, 1e+10, 1e+11, 1e+12, 1e+13, 1e+14, 1e+15, 1e+16, 1e+17, 1e+18, 1e+19, 1e+20, + 1e+0, + 1e+1, 1e+2, 1e+3, 1e+4, 1e+5, 1e+6, 1e+7, 1e+8, 1e+9, 1e+10, 1e+11, 1e+12, 1e+13, 1e+14, 1e+15, 1e+16, 1e+17, 1e+18, 1e+19, 1e+20, 1e+21, 1e+22, 1e+23, 1e+24, 1e+25, 1e+26, 1e+27, 1e+28, 1e+29, 1e+30, 1e+31, 1e+32, 1e+33, 1e+34, 1e+35, 1e+36, 1e+37, 1e+38, 1e+39, 1e+40, 1e+41, 1e+42, 1e+43, 1e+44, 1e+45, 1e+46, 1e+47, 1e+48, 1e+49, 1e+50, 1e+51, 1e+52, 1e+53, 1e+54, 1e+55, 1e+56, 1e+57, 1e+58, 1e+59, 1e+60, 1e+61, 1e+62, 1e+63, 1e+64, 1e+65, 1e+66, 1e+67, 1e+68, 1e+69, 1e+70, 1e+71, 1e+72, 1e+73, 1e+74, 1e+75, 1e+76, 1e+77, 1e+78, 1e+79, 1e+80, diff --git a/xmrstak/rapidjson/internal/regex.h b/xmrstak/rapidjson/internal/regex.h index 8530cd7..fd15122 100644 --- a/xmrstak/rapidjson/internal/regex.h +++ b/xmrstak/rapidjson/internal/regex.h @@ -1,5 +1,5 @@ // Tencent is pleased to support the open source community by making RapidJSON available. -// +// // Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. All rights reserved. // // Licensed under the MIT License (the "License"); you may not use this file except @@ -7,9 +7,9 @@ // // http://opensource.org/licenses/MIT // -// Unless required by applicable law or agreed to in writing, software distributed -// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR -// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the // specific language governing permissions and limitations under the License. #ifndef RAPIDJSON_INTERNAL_REGEX_H_ @@ -105,9 +105,9 @@ class GenericRegexSearch; - \c \\t Tab (U+0009) - \c \\v Vertical tab (U+000B) - \note This is a Thompson NFA engine, implemented with reference to - Cox, Russ. "Regular Expression Matching Can Be Simple And Fast (but is slow in Java, Perl, PHP, Python, Ruby,...).", - https://swtch.com/~rsc/regexp/regexp1.html + \note This is a Thompson NFA engine, implemented with reference to + Cox, Russ. "Regular Expression Matching Can Be Simple And Fast (but is slow in Java, Perl, PHP, Python, Ruby,...).", + https://swtch.com/~rsc/regexp/regexp1.html */ template <typename Encoding, typename Allocator = CrtAllocator> class GenericRegex { @@ -116,8 +116,8 @@ public: typedef typename Encoding::Ch Ch; template <typename, typename> friend class GenericRegexSearch; - GenericRegex(const Ch* source, Allocator* allocator = 0) : - states_(allocator, 256), ranges_(allocator, 256), root_(kRegexInvalidState), stateCount_(), rangeCount_(), + GenericRegex(const Ch* source, Allocator* allocator = 0) : + states_(allocator, 256), ranges_(allocator, 256), root_(kRegexInvalidState), stateCount_(), rangeCount_(), anchorBegin_(), anchorEnd_() { GenericStringStream<Encoding> ss(source); @@ -146,7 +146,7 @@ private: static const unsigned kRangeNegationFlag = 0x80000000; struct Range { - unsigned start; // + unsigned start; // unsigned end; SizeType next; }; @@ -391,7 +391,7 @@ private: } return false; - default: + default: RAPIDJSON_ASSERT(op == kOneOrMore); if (operandStack.GetSize() >= sizeof(Frag)) { Frag e = *operandStack.template Pop<Frag>(1); @@ -544,7 +544,7 @@ private: } return false; } - + SizeType NewRange(unsigned codepoint) { Range* r = ranges_.template Push<Range>(); r->start = r->end = codepoint; @@ -600,7 +600,7 @@ public: typedef typename RegexType::EncodingType Encoding; typedef typename Encoding::Ch Ch; - GenericRegexSearch(const RegexType& regex, Allocator* allocator = 0) : + GenericRegexSearch(const RegexType& regex, Allocator* allocator = 0) : regex_(regex), allocator_(allocator), ownAllocator_(0), state0_(allocator, 0), state1_(allocator, 0), stateSet_() { @@ -659,7 +659,7 @@ private: for (const SizeType* s = current->template Bottom<SizeType>(); s != current->template End<SizeType>(); ++s) { const State& sr = regex_.GetState(*s); if (sr.codepoint == codepoint || - sr.codepoint == RegexType::kAnyCharacterClass || + sr.codepoint == RegexType::kAnyCharacterClass || (sr.codepoint == RegexType::kRangeCharacterClass && MatchRange(sr.rangeStart, codepoint))) { matched = AddState(*next, sr.out) || matched; diff --git a/xmrstak/rapidjson/internal/stack.h b/xmrstak/rapidjson/internal/stack.h index 022c9aa..5cb9d9c 100644 --- a/xmrstak/rapidjson/internal/stack.h +++ b/xmrstak/rapidjson/internal/stack.h @@ -1,5 +1,5 @@ // Tencent is pleased to support the open source community by making RapidJSON available. -// +// // Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. All rights reserved. // // Licensed under the MIT License (the "License"); you may not use this file except @@ -7,9 +7,9 @@ // // http://opensource.org/licenses/MIT // -// Unless required by applicable law or agreed to in writing, software distributed -// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR -// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the // specific language governing permissions and limitations under the License. #ifndef RAPIDJSON_INTERNAL_STACK_H_ @@ -97,7 +97,7 @@ public: void Clear() { stackTop_ = stack_; } - void ShrinkToFit() { + void ShrinkToFit() { if (Empty()) { // If the stack is empty, completely deallocate the memory. Allocator::Free(stack_); @@ -140,7 +140,7 @@ public: } template<typename T> - T* Top() { + T* Top() { RAPIDJSON_ASSERT(GetSize() >= sizeof(T)); return reinterpret_cast<T*>(stackTop_ - sizeof(T)); } diff --git a/xmrstak/rapidjson/internal/strfunc.h b/xmrstak/rapidjson/internal/strfunc.h index de41d8f..becf6cd 100644 --- a/xmrstak/rapidjson/internal/strfunc.h +++ b/xmrstak/rapidjson/internal/strfunc.h @@ -1,5 +1,5 @@ // Tencent is pleased to support the open source community by making RapidJSON available. -// +// // Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. All rights reserved. // // Licensed under the MIT License (the "License"); you may not use this file except @@ -7,9 +7,9 @@ // // http://opensource.org/licenses/MIT // -// Unless required by applicable law or agreed to in writing, software distributed -// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR -// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the // specific language governing permissions and limitations under the License. #ifndef RAPIDJSON_INTERNAL_STRFUNC_H_ @@ -23,7 +23,7 @@ namespace internal { //! Custom strlen() which works on different character types. /*! \tparam Ch Character type (e.g. char, wchar_t, short) \param s Null-terminated input string. - \return Number of characters in the string. + \return Number of characters in the string. \note This has the same semantics as strlen(), the return value is not number of Unicode codepoints. */ template <typename Ch> diff --git a/xmrstak/rapidjson/internal/strtod.h b/xmrstak/rapidjson/internal/strtod.h index 289c413..8b010eb 100644 --- a/xmrstak/rapidjson/internal/strtod.h +++ b/xmrstak/rapidjson/internal/strtod.h @@ -1,5 +1,5 @@ // Tencent is pleased to support the open source community by making RapidJSON available. -// +// // Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. All rights reserved. // // Licensed under the MIT License (the "License"); you may not use this file except @@ -7,9 +7,9 @@ // // http://opensource.org/licenses/MIT // -// Unless required by applicable law or agreed to in writing, software distributed -// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR -// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the // specific language governing permissions and limitations under the License. #ifndef RAPIDJSON_STRTOD_ @@ -128,14 +128,14 @@ inline bool StrtodFast(double d, int p, double* result) { // Compute an approximation and see if it is within 1/2 ULP inline bool StrtodDiyFp(const char* decimals, size_t length, size_t decimalPosition, int exp, double* result) { uint64_t significand = 0; - size_t i = 0; // 2^64 - 1 = 18446744073709551615, 1844674407370955161 = 0x1999999999999999 + size_t i = 0; // 2^64 - 1 = 18446744073709551615, 1844674407370955161 = 0x1999999999999999 for (; i < length; i++) { if (significand > RAPIDJSON_UINT64_C2(0x19999999, 0x99999999) || (significand == RAPIDJSON_UINT64_C2(0x19999999, 0x99999999) && decimals[i] > '5')) break; significand = significand * 10u + static_cast<unsigned>(decimals[i] - '0'); } - + if (i < length && decimals[i] >= '5') // Rounding significand++; @@ -182,7 +182,7 @@ inline bool StrtodDiyFp(const char* decimals, size_t length, size_t decimalPosit if (precisionSize + kUlpShift >= 64) { unsigned scaleExp = (precisionSize + kUlpShift) - 63; v.f >>= scaleExp; - v.e += scaleExp; + v.e += scaleExp; error = (error >> scaleExp) + 1 + static_cast<int>(kUlp); precisionSize -= scaleExp; } diff --git a/xmrstak/rapidjson/istreamwrapper.h b/xmrstak/rapidjson/istreamwrapper.h index f5fe289..8007480 100644 --- a/xmrstak/rapidjson/istreamwrapper.h +++ b/xmrstak/rapidjson/istreamwrapper.h @@ -1,5 +1,5 @@ // Tencent is pleased to support the open source community by making RapidJSON available. -// +// // Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. All rights reserved. // // Licensed under the MIT License (the "License"); you may not use this file except @@ -7,9 +7,9 @@ // // http://opensource.org/licenses/MIT // -// Unless required by applicable law or agreed to in writing, software distributed -// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR -// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the // specific language governing permissions and limitations under the License. #ifndef RAPIDJSON_ISTREAMWRAPPER_H_ @@ -45,19 +45,19 @@ RAPIDJSON_NAMESPACE_BEGIN \tparam StreamType Class derived from \c std::basic_istream. */ - + template <typename StreamType> class BasicIStreamWrapper { public: typedef typename StreamType::char_type Ch; BasicIStreamWrapper(StreamType& stream) : stream_(stream), count_(), peekBuffer_() {} - Ch Peek() const { + Ch Peek() const { typename StreamType::int_type c = stream_.peek(); return RAPIDJSON_LIKELY(c != StreamType::traits_type::eof()) ? static_cast<Ch>(c) : '\0'; } - Ch Take() { + Ch Take() { typename StreamType::int_type c = stream_.get(); if (RAPIDJSON_LIKELY(c != StreamType::traits_type::eof())) { count_++; diff --git a/xmrstak/rapidjson/memorybuffer.h b/xmrstak/rapidjson/memorybuffer.h index 39bee1d..8e8d6c3 100644 --- a/xmrstak/rapidjson/memorybuffer.h +++ b/xmrstak/rapidjson/memorybuffer.h @@ -1,5 +1,5 @@ // Tencent is pleased to support the open source community by making RapidJSON available. -// +// // Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. All rights reserved. // // Licensed under the MIT License (the "License"); you may not use this file except @@ -7,9 +7,9 @@ // // http://opensource.org/licenses/MIT // -// Unless required by applicable law or agreed to in writing, software distributed -// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR -// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the // specific language governing permissions and limitations under the License. #ifndef RAPIDJSON_MEMORYBUFFER_H_ @@ -27,7 +27,7 @@ RAPIDJSON_NAMESPACE_BEGIN It is similar to FileWriteBuffer but the destination is an in-memory buffer instead of a file. Differences between MemoryBuffer and StringBuffer: - 1. StringBuffer has Encoding but MemoryBuffer is only a byte buffer. + 1. StringBuffer has Encoding but MemoryBuffer is only a byte buffer. 2. StringBuffer::GetString() returns a null-terminated string. MemoryBuffer::GetBuffer() returns a buffer without terminator. \tparam Allocator type for allocating memory buffer. diff --git a/xmrstak/rapidjson/memorystream.h b/xmrstak/rapidjson/memorystream.h index 1d71d8a..300a6e6 100644 --- a/xmrstak/rapidjson/memorystream.h +++ b/xmrstak/rapidjson/memorystream.h @@ -1,5 +1,5 @@ // Tencent is pleased to support the open source community by making RapidJSON available. -// +// // Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. All rights reserved. // // Licensed under the MIT License (the "License"); you may not use this file except @@ -7,9 +7,9 @@ // // http://opensource.org/licenses/MIT // -// Unless required by applicable law or agreed to in writing, software distributed -// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR -// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the // specific language governing permissions and limitations under the License. #ifndef RAPIDJSON_MEMORYSTREAM_H_ diff --git a/xmrstak/rapidjson/msinttypes/inttypes.h b/xmrstak/rapidjson/msinttypes/inttypes.h index 1811128..1620402 100644 --- a/xmrstak/rapidjson/msinttypes/inttypes.h +++ b/xmrstak/rapidjson/msinttypes/inttypes.h @@ -1,37 +1,37 @@ // ISO C9x compliant inttypes.h for Microsoft Visual Studio -// Based on ISO/IEC 9899:TC2 Committee draft (May 6, 2005) WG14/N1124 -// +// Based on ISO/IEC 9899:TC2 Committee draft (May 6, 2005) WG14/N1124 +// // Copyright (c) 2006-2013 Alexander Chemeris -// +// // Redistribution and use in source and binary forms, with or without // modification, are permitted provided that the following conditions are met: -// +// // 1. Redistributions of source code must retain the above copyright notice, // this list of conditions and the following disclaimer. -// +// // 2. Redistributions in binary form must reproduce the above copyright // notice, this list of conditions and the following disclaimer in the // documentation and/or other materials provided with the distribution. -// +// // 3. Neither the name of the product nor the names of its contributors may // be used to endorse or promote products derived from this software // without specific prior written permission. -// +// // THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR IMPLIED // WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF // MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO // EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, // SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, // PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; -// OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, +// OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, // WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR // OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF // ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. -// +// /////////////////////////////////////////////////////////////////////////////// -// The above software in this distribution may have been modified by -// THL A29 Limited ("Tencent Modifications"). +// The above software in this distribution may have been modified by +// THL A29 Limited ("Tencent Modifications"). // All Tencent Modifications are Copyright (C) 2015 THL A29 Limited. #ifndef _MSC_VER // [ diff --git a/xmrstak/rapidjson/msinttypes/stdint.h b/xmrstak/rapidjson/msinttypes/stdint.h index 3d4477b..1c266ec 100644 --- a/xmrstak/rapidjson/msinttypes/stdint.h +++ b/xmrstak/rapidjson/msinttypes/stdint.h @@ -1,37 +1,37 @@ // ISO C9x compliant stdint.h for Microsoft Visual Studio -// Based on ISO/IEC 9899:TC2 Committee draft (May 6, 2005) WG14/N1124 -// +// Based on ISO/IEC 9899:TC2 Committee draft (May 6, 2005) WG14/N1124 +// // Copyright (c) 2006-2013 Alexander Chemeris -// +// // Redistribution and use in source and binary forms, with or without // modification, are permitted provided that the following conditions are met: -// +// // 1. Redistributions of source code must retain the above copyright notice, // this list of conditions and the following disclaimer. -// +// // 2. Redistributions in binary form must reproduce the above copyright // notice, this list of conditions and the following disclaimer in the // documentation and/or other materials provided with the distribution. -// +// // 3. Neither the name of the product nor the names of its contributors may // be used to endorse or promote products derived from this software // without specific prior written permission. -// +// // THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR IMPLIED // WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF // MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO // EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, // SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, // PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; -// OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, +// OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, // WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR // OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF // ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. -// +// /////////////////////////////////////////////////////////////////////////////// -// The above software in this distribution may have been modified by -// THL A29 Limited ("Tencent Modifications"). +// The above software in this distribution may have been modified by +// THL A29 Limited ("Tencent Modifications"). // All Tencent Modifications are Copyright (C) 2015 THL A29 Limited. #ifndef _MSC_VER // [ diff --git a/xmrstak/rapidjson/ostreamwrapper.h b/xmrstak/rapidjson/ostreamwrapper.h index 6f4667c..896e772 100644 --- a/xmrstak/rapidjson/ostreamwrapper.h +++ b/xmrstak/rapidjson/ostreamwrapper.h @@ -1,5 +1,5 @@ // Tencent is pleased to support the open source community by making RapidJSON available. -// +// // Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. All rights reserved. // // Licensed under the MIT License (the "License"); you may not use this file except @@ -7,9 +7,9 @@ // // http://opensource.org/licenses/MIT // -// Unless required by applicable law or agreed to in writing, software distributed -// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR -// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the // specific language governing permissions and limitations under the License. #ifndef RAPIDJSON_OSTREAMWRAPPER_H_ @@ -40,7 +40,7 @@ RAPIDJSON_NAMESPACE_BEGIN \tparam StreamType Class derived from \c std::basic_ostream. */ - + template <typename StreamType> class BasicOStreamWrapper { public: diff --git a/xmrstak/rapidjson/pointer.h b/xmrstak/rapidjson/pointer.h index 0206ac1..0ecf6c4 100644 --- a/xmrstak/rapidjson/pointer.h +++ b/xmrstak/rapidjson/pointer.h @@ -1,5 +1,5 @@ // Tencent is pleased to support the open source community by making RapidJSON available. -// +// // Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. All rights reserved. // // Licensed under the MIT License (the "License"); you may not use this file except @@ -7,9 +7,9 @@ // // http://opensource.org/licenses/MIT // -// Unless required by applicable law or agreed to in writing, software distributed -// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR -// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the // specific language governing permissions and limitations under the License. #ifndef RAPIDJSON_POINTER_H_ @@ -50,16 +50,16 @@ enum PointerParseErrorCode { //! Represents a JSON Pointer. Use Pointer for UTF8 encoding and default allocator. /*! - This class implements RFC 6901 "JavaScript Object Notation (JSON) Pointer" + This class implements RFC 6901 "JavaScript Object Notation (JSON) Pointer" (https://tools.ietf.org/html/rfc6901). A JSON pointer is for identifying a specific value in a JSON document (GenericDocument). It can simplify coding of DOM tree manipulation, because it can access multiple-level depth of DOM tree with single API call. - After it parses a string representation (e.g. "/foo/0" or URI fragment + After it parses a string representation (e.g. "/foo/0" or URI fragment representation (e.g. "#/foo/0") into its internal representation (tokens), - it can be used to resolve a specific value in multiple documents, or sub-tree + it can be used to resolve a specific value in multiple documents, or sub-tree of documents. Contrary to GenericValue, Pointer can be copy constructed and copy assigned. @@ -70,10 +70,10 @@ enum PointerParseErrorCode { supplied tokens eliminates these. GenericPointer depends on GenericDocument and GenericValue. - + \tparam ValueType The value type of the DOM tree. E.g. GenericValue<UTF8<> > \tparam Allocator The allocator type for allocating memory for internal representation. - + \note GenericPointer uses same encoding of ValueType. However, Allocator of GenericPointer is independent of Allocator of Value. */ @@ -85,7 +85,7 @@ public: //! A token is the basic units of internal representation. /*! - A JSON pointer string representation "/foo/123" is parsed to two tokens: + A JSON pointer string representation "/foo/123" is parsed to two tokens: "foo" and 123. 123 will be represented in both numeric form and string form. They are resolved according to the actual value type (object or array). @@ -93,7 +93,7 @@ public: (greater than limits of SizeType), they are only treated as string form (i.e. the token's index will be equal to kPointerInvalidIndex). - This struct is public so that user can create a Pointer without parsing and + This struct is public so that user can create a Pointer without parsing and allocation, using a special constructor. */ struct Token { @@ -337,7 +337,7 @@ public: for (size_t i = 0; i < tokenCount_; i++) { if (tokens_[i].index != rhs.tokens_[i].index || - tokens_[i].length != rhs.tokens_[i].length || + tokens_[i].length != rhs.tokens_[i].length || (tokens_[i].length != 0 && std::memcmp(tokens_[i].name, rhs.tokens_[i].name, sizeof(Ch)* tokens_[i].length) != 0)) { return false; @@ -388,9 +388,9 @@ public: If the value is not exist, it creates all parent values and a JSON Null value. So it always succeed and return the newly created or existing value. - Remind that it may change types of parents according to tokens, so it - potentially removes previously stored values. For example, if a document - was an array, and "/foo" is used to create a value, then the document + Remind that it may change types of parents according to tokens, so it + potentially removes previously stored values. For example, if a document + was an array, and "/foo" is used to create a value, then the document will be changed to an object, and all existing array elements are lost. \param root Root value of a DOM subtree to be resolved. It can be any value other than document root. @@ -511,7 +511,7 @@ public: \param root Root value of a DOM sub-tree to be resolved. It can be any value other than document root. \return Pointer to the value if it can be resolved. Otherwise null. */ - const ValueType* Get(const ValueType& root, size_t* unresolvedTokenIndex = 0) const { + const ValueType* Get(const ValueType& root, size_t* unresolvedTokenIndex = 0) const { return Get(const_cast<ValueType&>(root), unresolvedTokenIndex); } @@ -573,7 +573,7 @@ public: ValueType& GetWithDefault(GenericDocument<EncodingType, typename ValueType::AllocatorType, stackAllocator>& document, const Ch* defaultValue) const { return GetWithDefault(document, defaultValue, document.GetAllocator()); } - + #if RAPIDJSON_HAS_STDSTRING //! Query a value in a document with default std::basic_string. template <typename stackAllocator> @@ -810,7 +810,7 @@ private: // Count number of '/' as tokenCount tokenCount_ = 0; - for (const Ch* s = source; s != source + length; s++) + for (const Ch* s = source; s != source + length; s++) if (*s == '/') tokenCount_++; @@ -867,7 +867,7 @@ private: } i++; - + // Escaping "~0" -> '~', "~1" -> '/' if (c == '~') { if (i < length) { @@ -956,7 +956,7 @@ private: os.Put('~'); os.Put('1'); } - else if (uriFragment && NeedPercentEncode(c)) { + else if (uriFragment && NeedPercentEncode(c)) { // Transcode to UTF8 sequence GenericStringStream<typename ValueType::EncodingType> source(&t->name[j]); PercentEncodeStream<OutputStream> target(os); @@ -974,7 +974,7 @@ private: //! A helper stream for decoding a percent-encoded sequence into code unit. /*! This stream decodes %XY triplet into code unit (0-255). - If it encounters invalid characters, it sets output code unit as 0 and + If it encounters invalid characters, it sets output code unit as 0 and mark invalid, and to be checked by IsValid(). */ class PercentDecodeStream { diff --git a/xmrstak/rapidjson/prettywriter.h b/xmrstak/rapidjson/prettywriter.h index abd964f..af99630 100644 --- a/xmrstak/rapidjson/prettywriter.h +++ b/xmrstak/rapidjson/prettywriter.h @@ -1,5 +1,5 @@ // Tencent is pleased to support the open source community by making RapidJSON available. -// +// // Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. All rights reserved. // // Licensed under the MIT License (the "License"); you may not use this file except @@ -7,9 +7,9 @@ // // http://opensource.org/licenses/MIT // -// Unless required by applicable law or agreed to in writing, software distributed -// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR -// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the // specific language governing permissions and limitations under the License. #ifndef RAPIDJSON_PRETTYWRITER_H_ @@ -55,11 +55,11 @@ public: \param allocator User supplied allocator. If it is null, it will create a private one. \param levelDepth Initial capacity of stack. */ - explicit PrettyWriter(OutputStream& os, StackAllocator* allocator = 0, size_t levelDepth = Base::kDefaultLevelDepth) : + explicit PrettyWriter(OutputStream& os, StackAllocator* allocator = 0, size_t levelDepth = Base::kDefaultLevelDepth) : Base(os, allocator, levelDepth), indentChar_(' '), indentCharCount_(4), formatOptions_(kFormatDefault) {} - explicit PrettyWriter(StackAllocator* allocator = 0, size_t levelDepth = Base::kDefaultLevelDepth) : + explicit PrettyWriter(StackAllocator* allocator = 0, size_t levelDepth = Base::kDefaultLevelDepth) : Base(allocator, levelDepth), indentChar_(' '), indentCharCount_(4) {} #if RAPIDJSON_HAS_CXX11_RVALUE_REFS @@ -133,7 +133,7 @@ public: return Key(str.data(), SizeType(str.size())); } #endif - + bool EndObject(SizeType memberCount = 0) { (void)memberCount; RAPIDJSON_ASSERT(Base::level_stack_.GetSize() >= sizeof(typename Base::Level)); diff --git a/xmrstak/rapidjson/rapidjson.h b/xmrstak/rapidjson/rapidjson.h index 053b2ce..11777d2 100644 --- a/xmrstak/rapidjson/rapidjson.h +++ b/xmrstak/rapidjson/rapidjson.h @@ -1,5 +1,5 @@ // Tencent is pleased to support the open source community by making RapidJSON available. -// +// // Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. All rights reserved. // // Licensed under the MIT License (the "License"); you may not use this file except @@ -7,9 +7,9 @@ // // http://opensource.org/licenses/MIT // -// Unless required by applicable law or agreed to in writing, software distributed -// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR -// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the // specific language governing permissions and limitations under the License. #ifndef RAPIDJSON_RAPIDJSON_H_ @@ -17,7 +17,7 @@ /*!\file rapidjson.h \brief common definitions and configuration - + \see RAPIDJSON_CONFIG */ @@ -241,7 +241,7 @@ # elif defined(RAPIDJSON_DOXYGEN_RUNNING) # define RAPIDJSON_ENDIAN # else -# error Unknown machine endianess detected. User needs to define RAPIDJSON_ENDIAN. +# error Unknown machine endianess detected. User needs to define RAPIDJSON_ENDIAN. # endif #endif // RAPIDJSON_ENDIAN @@ -423,7 +423,7 @@ RAPIDJSON_NAMESPACE_END #if defined(__GNUC__) #define RAPIDJSON_STATIC_ASSERT_UNUSED_ATTRIBUTE __attribute__((unused)) #else -#define RAPIDJSON_STATIC_ASSERT_UNUSED_ATTRIBUTE +#define RAPIDJSON_STATIC_ASSERT_UNUSED_ATTRIBUTE #endif #ifndef __clang__ //!@endcond @@ -474,7 +474,7 @@ RAPIDJSON_NAMESPACE_END //!@cond RAPIDJSON_HIDDEN_FROM_DOXYGEN -#define RAPIDJSON_MULTILINEMACRO_BEGIN do { +#define RAPIDJSON_MULTILINEMACRO_BEGIN do { #define RAPIDJSON_MULTILINEMACRO_END \ } while((void)0, 0) @@ -605,7 +605,7 @@ enum Type { kFalseType = 1, //!< false kTrueType = 2, //!< true kObjectType = 3, //!< object - kArrayType = 4, //!< array + kArrayType = 4, //!< array kStringType = 5, //!< string kNumberType = 6 //!< number }; diff --git a/xmrstak/rapidjson/reader.h b/xmrstak/rapidjson/reader.h index 71916c0..318fc9a 100644 --- a/xmrstak/rapidjson/reader.h +++ b/xmrstak/rapidjson/reader.h @@ -1065,7 +1065,7 @@ private: RAPIDJSON_FORCEINLINE Ch Peek() const { return is.Peek(); } RAPIDJSON_FORCEINLINE Ch TakePush() { return is.Take(); } RAPIDJSON_FORCEINLINE Ch Take() { return is.Take(); } - RAPIDJSON_FORCEINLINE void Push(char) {} + RAPIDJSON_FORCEINLINE void Push(char) {} size_t Tell() { return is.Tell(); } size_t Length() { return 0; } diff --git a/xmrstak/rapidjson/schema.h b/xmrstak/rapidjson/schema.h index e7af3cf..aca3561 100644 --- a/xmrstak/rapidjson/schema.h +++ b/xmrstak/rapidjson/schema.h @@ -1,5 +1,5 @@ // Tencent is pleased to support the open source community by making RapidJSON available-> -// +// // Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip-> All rights reserved-> // // Licensed under the MIT License (the "License"); you may not use this file except @@ -7,9 +7,9 @@ // // http://opensource->org/licenses/MIT // -// Unless required by applicable law or agreed to in writing, software distributed -// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR -// CONDITIONS OF ANY KIND, either express or implied-> See the License for the +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied-> See the License for the // specific language governing permissions and limitations under the License-> #ifndef RAPIDJSON_SCHEMA_H_ @@ -174,10 +174,10 @@ public: bool Uint(unsigned u) { Number n; n.u.u = u; n.d = static_cast<double>(u); return WriteNumber(n); } bool Int64(int64_t i) { Number n; n.u.i = i; n.d = static_cast<double>(i); return WriteNumber(n); } bool Uint64(uint64_t u) { Number n; n.u.u = u; n.d = static_cast<double>(u); return WriteNumber(n); } - bool Double(double d) { - Number n; + bool Double(double d) { + Number n; if (d < 0) n.u.i = static_cast<int64_t>(d); - else n.u.u = static_cast<uint64_t>(d); + else n.u.u = static_cast<uint64_t>(d); n.d = d; return WriteNumber(n); } @@ -194,7 +194,7 @@ public: bool StartObject() { return true; } bool Key(const Ch* str, SizeType len, bool copy) { return String(str, len, copy); } - bool EndObject(SizeType memberCount) { + bool EndObject(SizeType memberCount) { uint64_t h = Hash(0, kObjectType); uint64_t* kv = stack_.template Pop<uint64_t>(memberCount * 2); for (SizeType i = 0; i < memberCount; i++) @@ -202,9 +202,9 @@ public: *stack_.template Push<uint64_t>() = h; return true; } - + bool StartArray() { return true; } - bool EndArray(SizeType elementCount) { + bool EndArray(SizeType elementCount) { uint64_t h = Hash(0, kArrayType); uint64_t* e = stack_.template Pop<uint64_t>(elementCount); for (SizeType i = 0; i < elementCount; i++) @@ -231,9 +231,9 @@ private: }; bool WriteType(Type type) { return WriteBuffer(type, 0, 0); } - + bool WriteNumber(const Number& n) { return WriteBuffer(kNumberType, &n, sizeof(n)); } - + bool WriteBuffer(Type type, const void* data, size_t len) { // FNV-1a from http://isthe.com/chongo/tech/comp/fnv/ uint64_t h = Hash(RAPIDJSON_UINT64_C2(0x84222325, 0xcbf29ce4), type); @@ -433,7 +433,7 @@ public: if (properties && properties->IsObject()) for (ConstMemberIterator itr = properties->MemberBegin(); itr != properties->MemberEnd(); ++itr) AddUniqueElement(allProperties, itr->name); - + if (required && required->IsArray()) for (ConstValueIterator itr = required->Begin(); itr != required->End(); ++itr) if (itr->IsString()) @@ -660,7 +660,7 @@ public: for (SizeType i = allOf_.begin; i < allOf_.begin + allOf_.count; i++) if (!context.validators[i]->IsValid()) RAPIDJSON_INVALID_KEYWORD_RETURN(GetAllOfString()); - + if (anyOf_.schemas) { for (SizeType i = anyOf_.begin; i < anyOf_.begin + anyOf_.count; i++) if (context.validators[i]->IsValid()) @@ -688,13 +688,13 @@ public: return true; } - bool Null(Context& context) const { + bool Null(Context& context) const { if (!(type_ & (1 << kNullSchemaType))) RAPIDJSON_INVALID_KEYWORD_RETURN(GetTypeString()); return CreateParallelValidator(context); } - - bool Bool(Context& context, bool) const { + + bool Bool(Context& context, bool) const { if (!(type_ & (1 << kBooleanSchemaType))) RAPIDJSON_INVALID_KEYWORD_RETURN(GetTypeString()); return CreateParallelValidator(context); @@ -733,13 +733,13 @@ public: if (!maximum_.IsNull() && !CheckDoubleMaximum(context, d)) return false; - + if (!multipleOf_.IsNull() && !CheckDoubleMultipleOf(context, d)) return false; - + return CreateParallelValidator(context); } - + bool String(Context& context, const Ch* str, SizeType length, bool) const { if (!(type_ & (1 << kStringSchemaType))) RAPIDJSON_INVALID_KEYWORD_RETURN(GetTypeString()); @@ -760,7 +760,7 @@ public: return CreateParallelValidator(context); } - bool StartObject(Context& context) const { + bool StartObject(Context& context) const { if (!(type_ & (1 << kObjectSchemaType))) RAPIDJSON_INVALID_KEYWORD_RETURN(GetTypeString()); @@ -778,7 +778,7 @@ public: return CreateParallelValidator(context); } - + bool Key(Context& context, const Ch* str, SizeType len, bool) const { if (patternProperties_) { context.patternPropertiesSchemaCount = 0; @@ -854,7 +854,7 @@ public: return true; } - bool StartArray(Context& context) const { + bool StartArray(Context& context) const { if (!(type_ & (1 << kArraySchemaType))) RAPIDJSON_INVALID_KEYWORD_RETURN(GetTypeString()); @@ -864,12 +864,12 @@ public: return CreateParallelValidator(context); } - bool EndArray(Context& context, SizeType elementCount) const { + bool EndArray(Context& context, SizeType elementCount) const { context.inArray = false; - + if (elementCount < minItems_) RAPIDJSON_INVALID_KEYWORD_RETURN(GetMinItemsString()); - + if (elementCount > maxItems_) RAPIDJSON_INVALID_KEYWORD_RETURN(GetMaxItemsString()); @@ -1055,13 +1055,13 @@ private: if (anyOf_.schemas) CreateSchemaValidators(context, anyOf_); - + if (oneOf_.schemas) CreateSchemaValidators(context, oneOf_); - + if (not_) context.validators[notValidatorIndex_] = context.factory.CreateSchemaValidator(*not_); - + if (hasSchemaDependencies_) { for (SizeType i = 0; i < propertyCount_; i++) if (properties_[i].dependenciesSchema) @@ -1082,7 +1082,7 @@ private: SizeType len = name.GetStringLength(); const Ch* str = name.GetString(); for (SizeType index = 0; index < propertyCount_; index++) - if (properties_[index].name.GetStringLength() == len && + if (properties_[index].name.GetStringLength() == len && (std::memcmp(properties_[index].name.GetString(), str, sizeof(Ch) * len) == 0)) { *outIndex = index; @@ -1202,7 +1202,7 @@ private: struct PatternProperty { PatternProperty() : schema(), pattern() {} - ~PatternProperty() { + ~PatternProperty() { if (pattern) { pattern->~RegexType(); AllocatorType::Free(pattern); @@ -1564,7 +1564,7 @@ template < typename OutputHandler = BaseReaderHandler<typename SchemaDocumentType::SchemaType::EncodingType>, typename StateAllocator = CrtAllocator> class GenericSchemaValidator : - public internal::ISchemaStateFactory<typename SchemaDocumentType::SchemaType>, + public internal::ISchemaStateFactory<typename SchemaDocumentType::SchemaType>, public internal::ISchemaValidator { public: @@ -1582,7 +1582,7 @@ public: */ GenericSchemaValidator( const SchemaDocumentType& schemaDocument, - StateAllocator* allocator = 0, + StateAllocator* allocator = 0, size_t schemaStackCapacity = kDefaultSchemaStackCapacity, size_t documentStackCapacity = kDefaultDocumentStackCapacity) : @@ -1610,7 +1610,7 @@ public: GenericSchemaValidator( const SchemaDocumentType& schemaDocument, OutputHandler& outputHandler, - StateAllocator* allocator = 0, + StateAllocator* allocator = 0, size_t schemaStackCapacity = kDefaultSchemaStackCapacity, size_t documentStackCapacity = kDefaultDocumentStackCapacity) : @@ -1721,7 +1721,7 @@ RAPIDJSON_MULTILINEMACRO_END RAPIDJSON_SCHEMA_HANDLE_PARALLEL_(StartObject, ()); return valid_ = outputHandler_.StartObject(); } - + bool Key(const Ch* str, SizeType len, bool copy) { if (!valid_) return false; AppendToken(str, len); @@ -1729,8 +1729,8 @@ RAPIDJSON_MULTILINEMACRO_END RAPIDJSON_SCHEMA_HANDLE_PARALLEL_(Key, (str, len, copy)); return valid_ = outputHandler_.Key(str, len, copy); } - - bool EndObject(SizeType memberCount) { + + bool EndObject(SizeType memberCount) { if (!valid_) return false; RAPIDJSON_SCHEMA_HANDLE_PARALLEL_(EndObject, (memberCount)); if (!CurrentSchema().EndObject(CurrentContext(), memberCount)) return valid_ = false; @@ -1742,7 +1742,7 @@ RAPIDJSON_MULTILINEMACRO_END RAPIDJSON_SCHEMA_HANDLE_PARALLEL_(StartArray, ()); return valid_ = outputHandler_.StartArray(); } - + bool EndArray(SizeType elementCount) { if (!valid_) return false; RAPIDJSON_SCHEMA_HANDLE_PARALLEL_(EndArray, (elementCount)); @@ -1797,7 +1797,7 @@ private: typedef GenericValue<UTF8<>, StateAllocator> HashCodeArray; typedef internal::Hasher<EncodingType, StateAllocator> HasherType; - GenericSchemaValidator( + GenericSchemaValidator( const SchemaDocumentType& schemaDocument, const SchemaType& root, #if RAPIDJSON_SCHEMA_VERBOSE @@ -1872,7 +1872,7 @@ private: #endif uint64_t h = CurrentContext().arrayUniqueness ? static_cast<HasherType*>(CurrentContext().hasher)->GetHashCode() : 0; - + PopSchema(); if (!schemaStack_.Empty()) { @@ -1913,7 +1913,7 @@ private: } RAPIDJSON_FORCEINLINE void PushSchema(const SchemaType& schema) { new (schemaStack_.template Push<Context>()) Context(*this, &schema); } - + RAPIDJSON_FORCEINLINE void PopSchema() { Context* c = schemaStack_.template Pop<Context>(1); if (HashCodeArray* a = static_cast<HashCodeArray*>(c->arrayElementHashCodes)) { diff --git a/xmrstak/rapidjson/stream.h b/xmrstak/rapidjson/stream.h index fef82c2..608e9c5 100644 --- a/xmrstak/rapidjson/stream.h +++ b/xmrstak/rapidjson/stream.h @@ -1,5 +1,5 @@ // Tencent is pleased to support the open source community by making RapidJSON available. -// +// // Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. All rights reserved. // // Licensed under the MIT License (the "License"); you may not use this file except @@ -7,9 +7,9 @@ // // http://opensource.org/licenses/MIT // -// Unless required by applicable law or agreed to in writing, software distributed -// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR -// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the // specific language governing permissions and limitations under the License. #include "rapidjson.h" diff --git a/xmrstak/rapidjson/stringbuffer.h b/xmrstak/rapidjson/stringbuffer.h index 4e38b82..88dc15a 100644 --- a/xmrstak/rapidjson/stringbuffer.h +++ b/xmrstak/rapidjson/stringbuffer.h @@ -1,5 +1,5 @@ // Tencent is pleased to support the open source community by making RapidJSON available. -// +// // Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. All rights reserved. // // Licensed under the MIT License (the "License"); you may not use this file except @@ -7,9 +7,9 @@ // // http://opensource.org/licenses/MIT // -// Unless required by applicable law or agreed to in writing, software distributed -// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR -// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the // specific language governing permissions and limitations under the License. #ifndef RAPIDJSON_STRINGBUFFER_H_ diff --git a/xmrstak/rapidjson/writer.h b/xmrstak/rapidjson/writer.h index 8f6e174..9cfcfdb 100644 --- a/xmrstak/rapidjson/writer.h +++ b/xmrstak/rapidjson/writer.h @@ -1,5 +1,5 @@ // Tencent is pleased to support the open source community by making RapidJSON available. -// +// // Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. All rights reserved. // // Licensed under the MIT License (the "License"); you may not use this file except @@ -7,9 +7,9 @@ // // http://opensource.org/licenses/MIT // -// Unless required by applicable law or agreed to in writing, software distributed -// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR -// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the // specific language governing permissions and limitations under the License. #ifndef RAPIDJSON_WRITER_H_ @@ -50,7 +50,7 @@ RAPIDJSON_NAMESPACE_BEGIN /////////////////////////////////////////////////////////////////////////////// // WriteFlag -/*! \def RAPIDJSON_WRITE_DEFAULT_FLAGS +/*! \def RAPIDJSON_WRITE_DEFAULT_FLAGS \ingroup RAPIDJSON_CONFIG \brief User-defined kWriteDefaultFlags definition. @@ -74,7 +74,7 @@ enum WriteFlag { User may programmatically calls the functions of a writer to generate JSON text. - On the other side, a writer can also be passed to objects that generates events, + On the other side, a writer can also be passed to objects that generates events, for example Reader::Parse() and Document::Accept(). @@ -97,7 +97,7 @@ public: \param levelDepth Initial capacity of stack. */ explicit - Writer(OutputStream& os, StackAllocator* stackAllocator = 0, size_t levelDepth = kDefaultLevelDepth) : + Writer(OutputStream& os, StackAllocator* stackAllocator = 0, size_t levelDepth = kDefaultLevelDepth) : os_(&os), level_stack_(stackAllocator, levelDepth * sizeof(Level)), maxDecimalPlaces_(kDefaultMaxDecimalPlaces), hasRoot_(false) {} explicit @@ -151,7 +151,7 @@ public: /*! This setting truncates the output with specified number of decimal places. - For example, + For example, \code writer.SetMaxDecimalPlaces(3); @@ -407,7 +407,7 @@ protected: PutUnsafe(*os_, hexDigits[(trail >> 12) & 15]); PutUnsafe(*os_, hexDigits[(trail >> 8) & 15]); PutUnsafe(*os_, hexDigits[(trail >> 4) & 15]); - PutUnsafe(*os_, hexDigits[(trail ) & 15]); + PutUnsafe(*os_, hexDigits[(trail ) & 15]); } } else if ((sizeof(Ch) == 1 || static_cast<unsigned>(c) < 256) && RAPIDJSON_UNLIKELY(escape[static_cast<unsigned char>(c)])) { @@ -421,7 +421,7 @@ protected: PutUnsafe(*os_, hexDigits[static_cast<unsigned char>(c) & 0xF]); } } - else if (RAPIDJSON_UNLIKELY(!(writeFlags & kWriteValidateEncodingFlag ? + else if (RAPIDJSON_UNLIKELY(!(writeFlags & kWriteValidateEncodingFlag ? Transcoder<SourceEncoding, TargetEncoding>::Validate(is, *os_) : Transcoder<SourceEncoding, TargetEncoding>::TranscodeUnsafe(is, *os_)))) return false; @@ -453,7 +453,7 @@ protected: if (RAPIDJSON_LIKELY(level_stack_.GetSize() != 0)) { // this value is not at root Level* level = level_stack_.template Top<Level>(); if (level->valueCount > 0) { - if (level->inArray) + if (level->inArray) os_->Put(','); // add comma if it is not the first element in array else // in object os_->Put((level->valueCount % 2 == 0) ? ',' : ':'); @@ -541,7 +541,7 @@ inline bool Writer<StringBuffer>::WriteDouble(double d) { PutUnsafe(*os_, 'i'); PutUnsafe(*os_, 'n'); PutUnsafe(*os_, 'i'); PutUnsafe(*os_, 't'); PutUnsafe(*os_, 'y'); return true; } - + char *buffer = os_->Push(25); char* end = internal::dtoa(d, buffer, maxDecimalPlaces_); os_->Pop(static_cast<size_t>(25 - (end - buffer))); diff --git a/xmrstak/version.cpp b/xmrstak/version.cpp index f46060c..a680f44 100644 --- a/xmrstak/version.cpp +++ b/xmrstak/version.cpp @@ -18,7 +18,7 @@ #endif #define XMR_STAK_NAME "xmr-stak" -#define XMR_STAK_VERSION "2.4.3" +#define XMR_STAK_VERSION "2.4.4" #if defined(_WIN32) #define OS_TYPE "win" |
