Fix GPU with cgroups devices #429
Open
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Both functions differ only in a call to `read_all_devices` and it was assumed that if was sufficient to call only one of them. This is not the case (because `read_all_devices` is required with GPU's and cgroups). The difference between both functions in now part of macro's.
Contributor
Author
|
@acvizi It would be nice if you could also merge this one. Without it we cannot get torque & GPUs to work. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
If you build torque with GPU support and cgroups, it doesn't work: the whitelisting of devices in cgroups is not done correctly.
The core of the problem is the difference between
initialize_hwloc_topologyandcg_initialize_hwloc_topology. The code assumes that it is sufficient to call just one of them which is not true: the functionread_all_devicesis only called from thecg_initialize_hwloc_topology. I've merged both functions together and the difference between both are handled by macro's. I'm not sure what the idea was to duplicate this function? Copy&pasting code like that is only going to give problems, as this issue shows.Please backport this to the 6.1.1.1.