Skip to content

[CI][ROCm] Add NIXL PD e2e test - #218

Open
simondanielsson wants to merge 7 commits into
vllm-project:mainfrom
simondanielsson:ci/nixl-rocm-e2e
Open

[CI][ROCm] Add NIXL PD e2e test#218
simondanielsson wants to merge 7 commits into
vllm-project:mainfrom
simondanielsson:ci/nixl-rocm-e2e

Conversation

@simondanielsson

@simondanielsson simondanielsson commented Aug 17, 2026

Copy link
Copy Markdown
Contributor

Purpose

Fixes remainder of #142. Follow-up to #202.

  1. Add NIXL e2e test for ROCm
  2. Re-structure e2e PD tests into nvidia/ and rocm/ subfolders.

Note: will not run in fork PR's. Only runs in post-merge builds or nightly CI, similar to the MoRI tests added in #202.

Part of: vllm-project/vllm#44091

Scope

Currently tests only

  1. UCX backend
  2. 1p1d with TP1 on both P and D
  3. gfx942
  4. one NIC model

We can broaden this in follow-up PRs (e.g. other backends, xPyD, heterogenous TP, gfx950 with pensando AINIC ...)

Test Plan

CI and trusted ui build succeeds. (note trivy issue is fixed in #212)

Test Result

TODO


Essential Elements of an Effective PR Description Checklist
  • The purpose of the PR, such as "Fix some issue (link existing issues this PR will resolve)".
  • The test plan, such as providing test command.
  • The test results, such as pasting the results comparison before and after, or e2e results

Signed-off-by: simondanielsson <simon.danielsson99@hotmail.com>
@@ -1,279 +0,0 @@
#!/usr/bin/env bash

@simondanielsson simondanielsson Aug 17, 2026

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This one was

  1. moved to rocm/ folder
  2. modified to utilize shared utils from the _pd_rocm_common.sh (like pd_rocm_wait_for_health)

Point 2 was done as the ROCm NIXL tests also need these utils

Signed-off-by: simondanielsson <simon.danielsson99@hotmail.com>
Signed-off-by: simondanielsson <simon.danielsson99@hotmail.com>
Signed-off-by: simondanielsson <simon.danielsson99@hotmail.com>
Signed-off-by: simondanielsson <simon.danielsson99@hotmail.com>
@simondanielsson simondanielsson changed the title [CI][ROCm] Add NIXL e2e test [CI][ROCm] Add NIXL PD e2e test Aug 17, 2026
@simondanielsson
simondanielsson marked this pull request as ready for review August 17, 2026 10:02

@chatgpt-codex-connector chatgpt-codex-connector Bot left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: f667690536

ℹ️ About Codex in GitHub

Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".

Comment thread .buildkite/pipeline.yml
@simondanielsson

Copy link
Copy Markdown
Contributor Author

Failing Trivy scanner fixed here: #212

Signed-off-by: simondanielsson <simon.danielsson99@hotmail.com>
@simondanielsson

Copy link
Copy Markdown
Contributor Author

@codex review

@chatgpt-codex-connector

Copy link
Copy Markdown

Codex Review: Didn't find any major issues. Nice work!

Reviewed commit: 675d55ed47

ℹ️ About Codex in GitHub

Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".

@simondanielsson

simondanielsson commented Aug 17, 2026

Copy link
Copy Markdown
Contributor Author

@khluu Could I please ask you to start a trusted UI build for this PR's latest commit (cf5e62e49ab9bd30cee1416afc922076be13e281) on simondanielsson:ci/nixl-rocm-e2e with RUN_ROCM_NIXL=1? Once green and PR reviewed we can merge.

Similar to what we did for #202

Thanks 🙏

Signed-off-by: simondanielsson <simon.danielsson99@hotmail.com>
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

Successfully merging this pull request may close these issues.

1 participant