Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

cgroups memory limit does not cover memory allocated by ROCm hipMalloc() #5949

Open
garlick opened this issue May 8, 2024 · 0 comments
Open

Comments

@garlick
Copy link
Member

garlick commented May 8, 2024

Problem: Memory allocated by ROCm hipMalloc() is not subject to memory limits.

When flux runs with

[systemd]
enable = true
[exec]
service = "sdexec"
[exec.sdexec-properties]
MemoryMax = "95%"

We observed that memory allocated by hipMalloc() is not accounted for in the job's cgroup, and the user is able to exceed MemoryMax and run the kernel out of memory.

An issue was opened with AMD (though not public at this time).

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant