k-Wave User Forum » Topic: C++ source compilation on Apple M1 Pro

k-Wave User Forum » Topic: C++ source compilation on Apple M1 Pro http://www.k-wave.org/forum/topic/c-source-compilation-on-apple-m1-pro Support for the k-Wave MATLAB toolbox en-US Tue, 12 May 2026 23:36:52 +0000 http://bbpress.org/?v=1.0.2 <![CDATA[Search]]> q http://www.k-wave.org/forum/search.php wsimson on "C++ source compilation on Apple M1 Pro" http://www.k-wave.org/forum/topic/c-source-compilation-on-apple-m1-pro#post-9110 Thu, 11 Jul 2024 17:38:34 +0000 wsimson 9110@http://www.k-wave.org/forum/ Hey mmj, I have compiled OMP for MacOS here: <a href="https://github.com/waltsims/k-wave-omp-darwin" rel="nofollow">https://github.com/waltsims/k-wave-omp-darwin</a> Let me know if this works for you. This is the version of the binaries we use for k-wave-python. Hope this helps! Walter mmj on "C++ source compilation on Apple M1 Pro" http://www.k-wave.org/forum/topic/c-source-compilation-on-apple-m1-pro#post-9107 Wed, 03 Jul 2024 18:19:54 +0000 mmj 9107@http://www.k-wave.org/forum/ Hi everyone, I am new to this, so I already apologize for my naivety here. I am trying to compile C++ source codes for kspaceFirstOrder-OMP binaries on my MacOS, Sonoma 14.5. I downloaded the source code "k-wave-toolbox-version-1.3-cpp-linux-source" from <a href="http://www.k-wave.org/download.php" rel="nofollow">http://www.k-wave.org/download.php</a>. I am getting the following error after attempting to make: g++ -m64 -march=native -mtune=native -O3 -ffast-math -fassociative-math -Wall -Xpreprocessor -fopenmp -lomp -std=c++11 -D__KWAVE_GIT_HASH__=\"0ba023063e3f29685e1e346f56883378d961f9f1\" -I/opt/homebrew/opt/hdf5/include -I/opt/homebrew/opt/fftw/include -I. -o main.o -c main.cpp clang: warning: -lomp: 'linker' input unused [-Wunused-command-line-argument] main.cpp:798:12: fatal error: 'omp.h' file not found #include <omp.h> ^~~~~~~ 1 error generated. make: *** [Makefile:313: main.o] Error 1. -------------------------------------------- If use "make -j" to run multiple tasks concurrently, I am getting like 20 errors including: In file included from MatrixClasses/RealMatrix.cpp:34: In file included from ./Logger/Logger.h:39: ./Logger/OutputMessages.h:47:1: error: unknown type name 'OutputMessage' OutputMessage kOutFmtDone ^ ./Logger/OutputMessages.h:48:18: error: use of undeclared identifier 'kOutFmtEol' = " Done " + kOutFmtEol; . . . ------------------------------------------- I used "homebrew" to install required packages and libraries. The "Makefile" that I am using is copied below. Has anyone tried compiling OMP source codes on M1 processor facing the same issue? I was just wondering if this is relevant to OpenMP parallelization on M1 Pro or whether the error is related to my libraries/packages. Thank you in advance for your help. -Morteza # /** # * @file Makefile # * @author Jiri Jaros # * Faculty of Information Technology # * Brno University of Technology # * @email <a href="mailto:jarosjir@fit.vutbr.cz">jarosjir@fit.vutbr.cz</a> # * @comments Linux makefile for Release version on Ubuntu 18.04 # * # * @tool kspaceFirstOrder 2.17 # * @created 04 September 2017, 12:50 # * @lastModif 26 February 2020, 16:20 # * # * @copyright Copyright (C) 2017 - 2020 SC\@FIT Research Group, # * Brno University of Technology, Brno, CZ. # * # * This file is part of the C++ extension of the k-Wave Toolbox # * (<a href="http://www.k-wave.org)" rel="nofollow">http://www.k-wave.org)</a>. # * # * This file is part of the k-Wave. k-Wave is free software: you can # * redistribute it and/or modify it under the terms of the GNU Lesser General # * Public License as published by the Free Software Foundation, either version # * 3 of the License, or (at your option) any later version. # * # * k-Wave is distributed in the hope that it will be useful, but WITHOUT ANY # * WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS # * FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for # * more details. # * # * You should have received a copy of the GNU Lesser General Public License # * along with k-Wave. # * If not, see [http://www.gnu.org/licenses/](http://www.gnu.org/licenses/). # */ ################################################################################ # # # The source codes can be compiled ONLY under Linux x64 by GNU g++ 6.0 and # # newer, or Intel Compiler icpc 2018 and newer. The newer the compiler, the # # more advanced instruction set can be used. # # We recommend compilation with g++ 8.3 or icpc 2019. # # # # This makefile uses the GNU compiler and static linking by default. # # # # Necessary libraries: # # - FFTW 3.3.x and newer, or Intel MKL 2018 and newer # # - HDF5 version 1.8.x and newer # # # # How to compile libraries # # - FFTW : download from "http://www.fftw.org/" # # run configure script with following parameters: # # --enable-float --enable-avx --enable-openmp # # or # # --enable-float --enable-avx2 --enable-openmp # # - MKL : Only if not using FFTW. Dowload from # # "http://software.intel.com/en-us/intel-mkl" # # - HDF5 : download from # # "https://www.hdfgroup.org/downloads/hdf5/source-code/" # # run configure script with these parameters: # # --enable-hl --enable-static --enable-shared # # # # How to compile the code # # 1. Select the compiler (GNU + FFTW or Intel + MKL). # # 2. Select one of the linking possibilities. # # 3. Select CPU architecture (native binary is default). # # 4. Load necessary software modules, or set the library paths manually. # # 5. make -j # # # ################################################################################ ################################################################################ # Set following flags based on your compiler and library paths # ################################################################################ # Select compiler. GNU is default but Intel may be faster. COMPILER = GNU #COMPILER = Intel # Static lining is default LINKING = STATIC #LINKING = DYNAMIC # Set up paths: If using modules, the paths are set up automatically, ## otherwise, set paths manually #MKL_DIR = $(EBROOTMKL) #FFT_DIR = $(EBROOTFFTW) #HDF5_DIR = $(EBROOTHDF5) #ZLIB_DIR = $(EBROOTZLIB) #SZIP_DIR = $(EBROOTSZIP) ## homebrew installations, set paths manually FFT_DIR = /opt/homebrew/opt/fftw HDF5_DIR = /opt/homebrew/opt/hdf5 ZLIB_DIR = /opt/homebrew/opt/zlib SZIP_DIR = /opt/homebrew/opt/szip # Select CPU architecture (what instruction set to be used). # The libraries such as FFTW, HDF5 and MKL are to be compiled under the same # architecture, e.g., if you want to use AVX in k-Wave, compile FFTW with # --enable-avx # The native architecture will compile and optimize the code for the underlying # processor. # Fat binary, available only for Intel, includes all AVX, AVX2 and AVX512 sets. CPU_ARCH = native #CPU_ARCH = AVX #CPU_ARCH = AVX2 #CPU_ARCH = AVX512 #CPU_ARCH = FAT_BIN ############################### Common flags ################################### # Git hash of release 1.3 GIT_HASH = -D__KWAVE_GIT_HASH__=\"0ba023063e3f29685e1e346f56883378d961f9f1\" # Replace tabs by spaces .RECIPEPREFIX += ################################ GNU g++ + FFTW ################################ ifeq ($(COMPILER), GNU) # Compiler name CXX = g++ # C++ standard CPP_STD = -std=c++11 # Enable OpenMP # OPENMP = -fopenmp OPENMP = -Xpreprocessor -fopenmp -lomp # Set CPU architecture # Sandy Bridge, Ivy Bridge ifeq ($(CPU_ARCH), AVX) CPU_FLAGS = -m64 -mavx # Haswell, Broadwell else ifeq ($(CPU_ARCH), AVX2) CPU_FLAGS = -m64 -mavx2 # Skylake-X, Ice Lake, Cannon Lake else ifeq ($(CPU_ARCH), AVX512) CPU_FLAGS = -m64 -mavx512f # Defautl is native - the max performance for this CPU else CPU_FLAGS = -m64 -march=native -mtune=native endif # Use maximum optimization OPT = -O3 -ffast-math -fassociative-math # Debug flags DEBUG = # Profile flags PROFILE = # C++ warning flags WARNING = -Wall # Add include directories INCLUDES = -I$(HDF5_DIR)/include -I$(FFT_DIR)/include -I. # Add library directories LIB_PATHS = -L$(HDF5_DIR)/lib -L$(FFT_DIR)/lib # Set compiler flags and header files directories CXXFLAGS = $(CPU_FLAGS) $(OPT) $(DEBUG) $(WARNING) $(PROFILE) \ $(OPENMP) $(CPP_STD) \ $(GIT_HASH) \ $(INCLUDES) # Set linker flags and library files directories ifeq ($(LINKING), STATIC) # Static link LDFLAGS = $(CPU_FLAGS) $(DEBUG) $(WARNING) $(PROFILE) \ $(OPENMP) $(CPP_STD) \ $(LIB_PATHS) \ -static LDLIBS = $(FFT_DIR)/lib/libfftw3f.a \ $(FFT_DIR)/lib/libfftw3f_omp.a \ $(HDF5_DIR)/lib/libhdf5_hl.a \ $(HDF5_DIR)/lib/libhdf5.a \ $(ZLIB_DIR)/lib/libz.a \ $(SZIP_DIR)/lib/libsz.a \ -lm -lmvec \ -ldl # We need this for HDF5-1.8.11 and newer else # Dynamic link with runtime paths LDFLAGS = $(CPU_FLAGS) $(OPT) $(DEBUG) $(WARNING) $(PROFILE) \ $(OPENMP) $(CPP_STD) \ $(LIB_PATHS) \ -Wl,-rpath,$(HDF5_DIR)/lib:$(FFT_DIR)/lib LDLIBS = -lfftw3f -lfftw3f_omp -lhdf5 -lhdf5_hl -lm -lz endif endif ########################### Intel Compiler + MKL ############################### ifeq ($(COMPILER), Intel) # Compiler name CXX = icpc # C++ standard CPP_STD = -std=c++11 # Enable OpenMP OPENMP = -qopenmp # Set CPU architecture # Sandy Bridge, Ivy Bridge ifeq ($(CPU_ARCH), AVX) CPU_FLAGS = -m64 -xAVX # Haswell, Broadwell else ifeq ($(CPU_ARCH), AVX2) CPU_FLAGS = -m64 -xCORE-AVX2 # Skylake-X, Ice Lake, Cannon Lake else ifeq ($(CPU_ARCH), AVX512) CPU_FLAGS = -m64 -xCORE-AVX512 # AMD-AVX, Intel's AVX, AVX2 and AVX-512 else ifeq ($(CPU_ARCH), FAT_BIN) CPU_FLAGS = -m64 -mavx -axCORE-AVX2,CORE-AVX512 # Defautl is native - the max performance for this CPU else CPU_FLAGS = -m64 -xhost endif # Use maximum optimization OPT = -Ofast -ipo # Debug flags DEBUG = # Profile flags PROFILE = # C++ warning flags WARNING = -Wall # Add include directories INCLUDES = -I$(HDF5_DIR)/include -I. # Add library directories LIB_PATHS = -L$(HDF5_DIR)/lib # Set compiler flags and header files directories CXXFLAGS = $(CPU_FLAGS) $(OPT) $(DEBUG) $(WARNING) $(PROFILE) \ $(OPENMP) $(CPP_STD) \ $(GIT_HASH) \ $(INCLUDES) # Set linker flags ifeq ($(LINKING), STATIC) LDFLAGS = $(CPU_FLAGS) $(OPT) $(DEBUG) $(WARNING) $(PROFILE) \ $(OPENMP) $(CPP_STD) \ $(LIB_PATHS) \ -mkl=parallel -static-intel -qopenmp-link=static LDLIBS = $(HDF5_DIR)/lib/libhdf5_hl.a \ $(HDF5_DIR)/lib/libhdf5.a \ $(ZLIB_DIR)/lib/libz.a \ $(SZIP_DIR)/lib/libsz.a \ -lm -ldl else # Dynamic link with runtime paths LDFLAGS = $(CPU_FLAGS) $(OPT) $(DEBUG) $(WARNING) $(PROFILE) \ $(OPENMP) $(CPP_STD) \ $(LIB_PATHS) \ -Wl,-rpath,$(HDF5_DIR)/lib \ -mkl=parallel LDLIBS = -lhdf5 -lhdf5_hl -lz -lsz endif endif ################################### Build ###################################### # Target binary name TARGET = kspaceFirstOrder-OMP # Units to be compiled DEPENDENCIES = main.o \ Containers/MatrixContainer.o \ Containers/OutputStreamContainer.o \ Hdf5/Hdf5File.o \ Hdf5/Hdf5FileHeader.o \ KSpaceSolver/KSpaceFirstOrderSolver.o \ Logger/Logger.o \ MatrixClasses/BaseFloatMatrix.o \ MatrixClasses/BaseIndexMatrix.o \ MatrixClasses/ComplexMatrix.o \ MatrixClasses/FftwComplexMatrix.o \ MatrixClasses/FftwRealMatrix.o \ MatrixClasses/IndexMatrix.o \ MatrixClasses/RealMatrix.o \ OutputStreams/BaseOutputStream.o \ OutputStreams/IndexOutputStream.o \ OutputStreams/CuboidOutputStream.o \ OutputStreams/WholeDomainOutputStream.o \ Parameters/CommandLineParameters.o \ Parameters/Parameters.o # Build target all: $(TARGET) # Link target $(TARGET): $(DEPENDENCIES) $(CXX) $(LDFLAGS) $(DEPENDENCIES) $(LDLIBS) -o $@ # Compile units %.o: %.cpp $(CXX) $(CXXFLAGS) -o $@ -c $< # Clean repository .PHONY: clean clean: rm -f $(DEPENDENCIES) $(TARGET)