documentation: add instructions on how to use external opengl (#40987)

This commit is contained in:
Billae 2023-12-05 12:59:41 +01:00 committed by GitHub
parent b9f4d9f6fc
commit eb9c63541a
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23

View File

@ -111,3 +111,28 @@ CUDA is split into fewer components and is simpler to specify:
prefix: /opt/cuda/cuda-11.0.2/
where ``/opt/cuda/cuda-11.0.2/lib/`` contains ``libcudart.so``.
-----------------------------------
Using an External OpenGL API
-----------------------------------
Depending on whether we have a graphics card or not, we may choose to use OSMesa or GLX to implement the OpenGL API.
If a graphics card is unavailable, OSMesa is recommended and can typically be built with Spack.
However, if we prefer to utilize the system GLX tailored to our graphics card, we need to declare it as an external. Here's how to do it:
.. code-block:: yaml
packages:
libglx:
require: [opengl]
opengl:
buildable: false
externals:
- prefix: /usr/
spec: opengl@4.6
Note that prefix has to be the root of both the libraries and the headers, using is /usr not the path the the lib.
To know which spec for opengl is available use ``cd /usr/include/GL && grep -Ri gl_version``.