NVIDIA / NVIDIA/AMGX

[Build]Compiling with Pre-Installed OpenMPI

Open
#328 2 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

build issues
Dominant language
Cuda
Stars
692
Forks
197
PR merge metrics
No merged PRs in 30d

Description

I would like to install AMGX on a remote computing cluster. This cluster has OpenMPI 4.1.5 available as a loadable module, but I also have OpenMPI 5.0.3 downloaded as an external package within another program.

I would like to have AMGX use OpenMPI 5.0.3. I have attempted to do this during the configuration phase by using code such as:

<$cmake -DCMAKE_PREFIX_PATH=[path to openmpi5.0.3] ../>
<$cmake -DMPI_C_HEADER_DIR=[path to openmpi5.0.3] ../>

However, despite trying these variations, OpenMPI is not seen and I get the output:

Is there a way to point AMGX to my current download/installation of OpenMPI 5.0.3?

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start with the CMake configuration and MPI detection path, reproducing the issue with the cluster's OpenMPI 5.0.3 paths and the provided CMAKE_PREFIX_PATH and MPI_C_HEADER_DIR values. Determine why configuration reports "This is a MPI build:FALSE"; done means AMGX detects the requested OpenMPI installation during configuration.

Written by the indexing model from the issue text.

Assessment

Tech stack
cmake
Domain
build-system, distributed-systems
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.