I am trying to install the MPI FFTW2 libraries (fftw2.1.5) as I need them for a program that only works with that version. The specific library I need is "drfftw_mpi.h" so I try the followed:
./configure --enable-mpi --enable-type-prefix
But it always gives me the following error:
configure: error: couldn't find mpi library for --enable-mpi
In a previous question they solve it by running sudo ldconfig before, but for me it does not work giving the same error. For other mpi software they solve it in this post by passing mpicc as the right compiler to ./configure, with the CC flag CC=mpicc, something that does not work for me.
I have installed Open MPI version 4.0.3 on a Ubuntu 20.04 LTS, and when --enable-mpi is not used I was able to install "dfftw.h" "drfftw.h" "fftw.h" and "rfftw.h", do you have an idea of what I am doing wrong or what should I do to install "drfftw_mpi.h"?
Related
Hi, All
I am currently installing the openmpi-4.1.1 on ubuntu18.04 from the tar.gz file.
However, when I use the nvcc (CUDA 11.2.2) compiler with -lmpi_cxx, it reports that this linking option does not exist.
is there anything wrong when I am building and installing the openmpi?
I use the following commands when building openmpi with CUDA-aware capability.
./configure --with-cuda
make -j8 install
I try to remove -lmpi_cxx and only keep -lmpi, the compiler reports errors like
undefined reference to `MPI::Comm::Comm()'
Thanks a lot!
I just figure this out by myself.
I need to enable the c++ binding of the MPI when building the openmpi.
Here are the commands
./configure --enable-mpi-cxx --with-cuda
make all install
I'm trying to install this project: https://github.com/drufat/triangle. Unfortunatly, I get the following error:
error: Cygwin gcc cannot be used with --compiler=mingw32
I'm using the lastest version of gcc for Cygwin on Windows 7 64 bit and I'm trying to install that project for python 3.7.1. It's seems to come from my distutil configuration. My distutils.cfg file contains:
[build]
compiler=mingw32
I tried that solution but it doens't helped:
https://stackoverflow.com/a/16740123/5075502
The gcc options to build for MinGW64 is no longer supported by the regular cygqin compiler gcc, aka x86_64-pc-cygwin-gcc. Instead, you should also install the mingw64-x86_64-gcc-core package which provides the GCC for Win64 toolchain (C, OpenMP). This will also install other packages (through dependencies) which provide compatible binutils, headers, libraries, and runtime.
In your Makefile or compiler command line, replace gcc with x86_64-w64-mingw32-gcc. This will allow you to compile and link executables which do not depend on the cygwin1.dll.
A similar problem occured. I fixed it by changing mingw32 by gcc like #phd said.
I have a problem using Paho MQTT client in C.
I downloaded the pre built binaries for my system (Windows 10 64) from their projects page. I unpacked the zip file to a folder in the documents folder, where I also created a .c file with the example at the bottom of the Paho product page. My editor is atom and my compiler is gcc.
When I tried to compile it in Atom, I got this error:
undefined reference to MQTTClient_create'
So I went searching and found plenty of topics, but I still couldn't figure out, how to resolve this issue. From this stackoverflow topic I gather that it's a linker problem and that I need to link the files during compile, so here's what I tried:
gcc MQTT.c -L "C:\Users\Pete\Documents\MQTT on C\Examples\Paho\lib" -l paho-mqtt3c
Which still gives me the same undefined reference error. When I try to link to the dll of the same name, the compiler does not find the file.
Can anyone point me in the right direction, please?
Any help is appreciated!
I'd be interested to know if you have registered the Paho MQTT dll in Windows?
You should have paho-mqtt3c.dll as part of the download.
I don't know if this will work for you, but I have the same issue (undefined reference to MQTTClient_create)...
Copy the dll file into c:/windows/system32
Run CMD Prompt as Administrator and type:
regsvr32 i/ paho-mqtt3c.dll
Worth a try.
The problem was that I was compiling the program with the 32-bit gcc compiler for the 64-bit library. Installing and using MinGW64 worked.
I had the same issue in OS X. This is how I resolved it
I compiled the paho-mqtt library in Linux/EC2 instance.
Installed all the dependencies:
sudo yum install install build-essential gcc make cmake cmake-gui cmake-curses-gui
sudo yum install cmake
sudo yum install doxygen graphviz
cmake -DPAHO_WITH_SSL=TRUE -DPAHO_BUILD_DOCUMENTATION=FALSE -DPAHO_BUILD_STATIC=TRUE -DPAHO_BUILD_SHARED=FALSE -DPAHO_MQTT_C_PATH=../paho.mqtt.c/ -DPAHO_MQTT_C_LIB=../paho.mqtt.c/src/libpaho.mqtt3as-static.a
make
make html
make install
LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/home/ec2-user/paho.mqtt.c/src
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH
sudo ldconfig
gcc test1.c -o test1 -lpaho-mqtt3c
Compiled successfully..
./test1
This solution works for OS X as well. Replace step 2 with
brew install cmake doxygen openssl-devel
Trying to install fftw to use GADGET2, and after running the following command:
./configure --enable-mpi --enable-type-prefix --enable-float
I get the message:
checking for mpicc... mpicc
checking for MPI_Init... no
checking for MPI_Init in -lmpi... no
checking for MPI_Init in -lmpich... no
configure: error: couldn't find mpi library for --enable-mpi
Now I have installed openMPI already, so it seems to be the case that it just can't find it. There was a similar question posted a while ago with different mpi software. I think the problem is that my openMPI and other compilers might be in different folders? This brings me to a more general question (I have tried googling this but can't find anything that explains it well):
If I extract a tar and use ./configure without any prefix, where will the program install to? And is this is an issue that my openMPI has for some reason installed to a different place?
Thanks!
edit: found a solution for this from this question, where just running
sudo ldconfig
seemed to fix it and now it compiles fine.
I'm trying to install the software TinyOWS on a Linux computer. The source code is written in C programming language. I don't have much experience with installing C or C++ programs from source code on Linux. I read installation instructions, and it says that I must follow 3 steps after I unzip the source code folder:
./configure
make
make install
When I run ./configure, I get the following error:
checking for xml2-config... /usr/bin/xml2-config
checking for libxml2 version... 2.7.6
configure: error: You need at least libxml 2.8.0
This means that the software being installed requires version 2.8.0 of a library called libxml2 but my computer only has the 2.7.6 version. Unfortunately I found out that I cannot easily install the libxml 2.8.0 on my Linux distribution. The reason why the software TinyOWS that I'm trying to install requires libxml2.8.0 of the library is because libxml2.7.6 has a small bug. But most of the functionality of the software is not affected by the bug.
So, here is my question:
What files do I have to edit or what command line options do I need to use If I want the ./configure command to use the library libxml2.7.6 instead of libxml2.8.0?
Get the source, in the file configure.in replace the line:
AC_MSG_ERROR([You need at least libxml 2.8.0])
By
AC_MSG_NOTICE([You need at least libxml 2.8.0])
Then, do:
autoreconf -f -i -s
./configure
make
The configure step will not stop. But i think if the developpers have restricted this version of libxml2 there must be a good reason.