search

lorenzo-rovigatti / oxDNA

A new version of the code to simulate the oxDNA/oxRNA models, now equipped with Python bindings
https://dna.physics.ox.ac.uk/
GNU General Public License v3.0
42 stars 27 forks source link

[BUG] Installation, unable to build with make command #84

Closed sheneric88 closed 9 months ago

sheneric88 commented 10 months ago

Describe the bug I've managed to compile oxDNA into the build folder. However, when I try to build it, I get the following error/output:

Input: mingw32-make -j4 Output: mingw32-make: *** No targets specified and no makefile found. Stop.

I am using mingw32-make because my current configuration of conda/Python with multiple folders added to PATH still cannot find the make command.

I am not too familiar with version control, so I downloaded the latest 3.6.0 release and unzipped the folder to work in there.

Desktop (please complete the following information):

Additional context I used the following compile command: cmake .. -DCUDA=ON -DCUDA_COMMON_ARCH=OFF -DPython=ON -DOxpySystemInstall=On

ErikPoppleton commented 10 months ago

oxDNA is only made for Unix and Windows is not supported. I've heard rumors that a couple of people got it working, but you would be better off installing it under WSL. I've never done this, but I think @zoombya might have gotten it working with GPU support as well? Perhaps he has some advice.

lorenzo-rovigatti commented 10 months ago

As Erik noted, we don't explicitly support Windows, but the compilation should work regardless. Did you run cmake prior to running make? What's the output of the cmake command?

P.S. note that cmake does not compile anything. It simply prepares the code to be compiled on the current platform. make is what carries out compilation.

zoombya commented 10 months ago

Hi, as @lorenzo-rovigatti noted you need to run the cmakecommand to configure the installation before you can run the make. (which looks like the source of your current errors) I tried myself a similar setup about 4 years ago and remember running into type issues specific to MinGW. I was able to eventually compile the code, but had to hack quite a bit of code to make MinGW accept it. Therefore I would recommend @ErikPoppleton suggestion of using WSL 2 rather than the CygWin suite.
Its faster and allows for CUDA - GPU compilation in case your machine has an NVidia GPU. Also I'm more likely to help you with any potential issues you might run into as this is my current setup.

lorenzo-rovigatti commented 9 months ago

Did you manage to compile the code?