Search found 8 matches
- Thu Oct 23, 2014 6:17 pm
- Forum: OpenMM
- Topic: Getting both OpenCL and CUDA working with OpenMM 6.1
- Replies: 11
- Views: 759
Re: Getting both OpenCL and CUDA working with OpenMM 6.1
In any case, for practical purposes, I can just disable everything OpenCL and worry about it later or never. I just thought that it might indicate a possible problem, but now it more likely looks like a problem with my system.
- Thu Oct 23, 2014 6:12 pm
- Forum: OpenMM
- Topic: Getting both OpenCL and CUDA working with OpenMM 6.1
- Replies: 11
- Views: 759
Re: Getting both OpenCL and CUDA working with OpenMM 6.1
Ah, I see. It could be that it is not packaged well and there is a mismatch between headers and library. OPENCL_INCLUDE_DIR=/usr/include If these are the correct files, then, I can see that /usr/include/CL/cl.h and company are provided by the package opencl-headers, rather than anything from nvidia,...
- Thu Oct 23, 2014 12:10 pm
- Forum: OpenMM
- Topic: Getting both OpenCL and CUDA working with OpenMM 6.1
- Replies: 11
- Views: 759
Re: Getting both OpenCL and CUDA working with OpenMM 6.1
Lee-Ping, that is exactly the case and because of the potential pain that you mention, I try to stick to the stable version from the repo, when possible. For other reason, I have decided to update to 14.10 anyway, so I might get a sufficiently new version from the repo anyway.
- Thu Oct 23, 2014 12:07 pm
- Forum: OpenMM
- Topic: Getting both OpenCL and CUDA working with OpenMM 6.1
- Replies: 11
- Views: 759
Re: Getting both OpenCL and CUDA working with OpenMM 6.1
Hi Lee-Ping, I should have clarified that the type of issue and concepts involved are not a problem. I was interested specifically in these undefined symbols - perhaps it is a known issue or someone has encountered it before. The implicit question (which I should have made explicit) was: "Why does l...
- Thu Oct 23, 2014 10:55 am
- Forum: OpenMM
- Topic: Getting both OpenCL and CUDA working with OpenMM 6.1
- Replies: 11
- Views: 759
Re: Getting both OpenCL and CUDA working with OpenMM 6.1
Ah, sorry about that, must have missed the version bump. I remembered that OpenMM 6.0 was compiled against CUDA 5.5.
- Thu Oct 23, 2014 10:53 am
- Forum: OpenMM
- Topic: User install location of build from source - Python API
- Replies: 2
- Views: 158
Re: User install location of build from source - Python API
Thanks, sounds good. I found that before I found the PythonInstall target, but was not sure if there perhaps were side effects of the target that would make it preferable. Would you consider adding this to the manual? I would expect this to be a fairly common scenario - you are installing on a clust...
- Wed Oct 22, 2014 7:41 pm
- Forum: OpenMM
- Topic: Getting both OpenCL and CUDA working with OpenMM 6.1
- Replies: 11
- Views: 759
Getting both OpenCL and CUDA working with OpenMM 6.1
I am trying to install OpenMM 6.1 on Ubuntu 12.04 with CUDA 5.5 from the repository. When I install the pre-built binary, the CUDA platform is not found when I run the installation test script. With otherwise identical settings, it is found when I build from source and install. However, when I build...
- Wed Oct 22, 2014 7:13 pm
- Forum: OpenMM
- Topic: User install location of build from source - Python API
- Replies: 2
- Views: 158
User install location of build from source - Python API
I want to build OpenMM 6.1 from source and install it in a user directory. I also want to install the Python API. As far as I can tell, the PythonInstall target does not provide a mechanism to specify the location of the Python bindings. If system Python (e.g. /usr/bin/python) is used, it will try t...