Repository navigation
Conversation
The default job count only accounts for CPU threads and host memory. Callers whose workers share another limited resource, such as GPU memory, had no way to express it short of forcing the count with --jobs, which bypasses the memory check, or rebuilding ParsedArgs. max_default_jobs caps the automatic default and, like memory_per_worker, is ignored when --jobs or PTR_NUM_JOBS is set.
Collaborator
Just realised this broke Oceananigans too... |
This was referenced Oct 7, 2026
Collaborator
|
I think this should be in |
christiangnrd
added this pull request to stack #213
October 7, 2026 17:34
This branch has not been deployed
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Adds a
max_default_jobskeyword torunteststhat bounds the default number of jobs, for resources PTR doesn't account for itself, such as GPU memory shared by all workers.Precedence is
--jobs>PTR_NUM_JOBS>min(default_njobs(; memory_per_worker), max_default_jobs), so explicit settings still bypass it, likememory_per_worker. Invalid values (non-integer,Bool,< 1) throw anArgumentError.CUDA.jl currently caps the job count by free GPU memory by rebuilding
ParsedArgswith the positional constructor, which broke in 2.9.2 whenskippassedwas added. With this keyword it can pass the cap directly and stop depending on the struct layout.