Skip to content

[GPU Driver Container Resiliency] Implement userspace-only fast path when kernel modules already loaded with matching config #534

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.

Sign up for GitHub

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

Jump to bottom
Open
karthikvetrivel wants to merge 1 commit into
base: main
Choose a base branch
from
Open

[GPU Driver Container Resiliency] Implement userspace-only fast path when kernel modules already loaded with matching config #534

karthikvetrivel wants to merge 1 commit into from
+552 −227

Conversation

@karthikvetrivel
Copy link
Member

@karthikvetrivel karthikvetrivel commented Jan 8, 2026 •
edited
Loading

Relevant PRs:

Description

  • Implements userspace-only fast path install when kernel modules are already loaded with matching configuration
  • Checks if NVIDIA modules are loaded (/sys/module/nvidia/refcnt) and compares configuration digest
  • When match detected, performs userspace-only install (libraries/binaries) while skipping kernel module build/load
  • Stores digest after successful installation for future comparisons

@karthikvetrivel karthikvetrivel marked this pull request as draft January 8, 2026 19:29
@karthikvetrivel karthikvetrivel force-pushed the feat/operator-digest-driver-config branch 2 times, most recently from 7bf634d to f959234 Compare January 8, 2026 19:59
This was referenced Jan 8, 2026
Open
Open
@karthikvetrivel karthikvetrivel force-pushed the feat/operator-digest-driver-config branch from f959234 to a9bde48 Compare January 8, 2026 20:11
@karthikvetrivel karthikvetrivel marked this pull request as ready for review January 9, 2026 19:36
cdesiniotis
cdesiniotis reviewed Jan 12, 2026
View reviewed changes
@karthikvetrivel karthikvetrivel force-pushed the feat/operator-digest-driver-config branch from a9bde48 to 82123af Compare January 12, 2026 20:36
tariq1890
tariq1890 reviewed Jan 12, 2026
View reviewed changes
cdesiniotis
cdesiniotis reviewed Jan 13, 2026
View reviewed changes
@karthikvetrivel karthikvetrivel force-pushed the feat/operator-digest-driver-config branch from a7c51cb to a83583e Compare January 13, 2026 16:35
tariq1890
tariq1890 reviewed Jan 13, 2026
View reviewed changes
@karthikvetrivel karthikvetrivel mentioned this pull request Jan 13, 2026
Draft
cdesiniotis
cdesiniotis reviewed Jan 13, 2026
View reviewed changes
@karthikvetrivel karthikvetrivel force-pushed the feat/operator-digest-driver-config branch 2 times, most recently from d612e96 to d684c7c Compare January 14, 2026 21:48
cdesiniotis
cdesiniotis reviewed Jan 14, 2026
View reviewed changes
@karthikvetrivel karthikvetrivel force-pushed the feat/operator-digest-driver-config branch from d684c7c to a577c50 Compare January 14, 2026 23:46
@karthikvetrivel karthikvetrivel force-pushed the feat/operator-digest-driver-config branch from a577c50 to d0f4547 Compare January 15, 2026 00:12
@tariq1890
Copy link
Contributor

tariq1890 commented Jan 15, 2026

@shivakunv Please note the changes added by @karthikvetrivel in this PR. These changes have to be added to the rhel10 directory you're working on

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@tariq1890 tariq1890 tariq1890 approved these changes

@cdesiniotis cdesiniotis cdesiniotis approved these changes

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

3 participants

@karthikvetrivel @tariq1890 @cdesiniotis