Skip to content

[CI]init intel ci dispatch in buildkite folder#1721

Merged
gcanlin merged 10 commits into
vllm-project:mainfrom
xuechendi:xpu_ci_dispatch
Mar 19, 2026
Merged

[CI]init intel ci dispatch in buildkite folder#1721
gcanlin merged 10 commits into
vllm-project:mainfrom
xuechendi:xpu_ci_dispatch

Conversation

@xuechendi
Copy link
Copy Markdown
Contributor

@xuechendi xuechendi commented Mar 7, 2026

PLEASE FILL IN THE PR DESCRIPTION HERE ENSURING ALL CHECKLIST ITEMS (AT THE BOTTOM) HAVE BEEN CONSIDERED.

Purpose

Add CI dispatch

[x] add 'intel-gpu-omni' queue in buildkite CI cluster => Done
[x] register buildkite agent to 'intel-gpu-omni'
image

Current design is:

.buildkite/pipeline.yaml
->
.buildkite/bootstrap-intel-omni.sh
->
pipeline-intel.yml
->
bash .buildkite/scripts/hardware_ci/run-xpu-test.sh

Note:
bootstrap-intel-omni.sh : used to check if we need to skip CI for pipeline-intel.yaml
pipeline-intel.yaml : actual steps for CI
- now, single step to trigger an all-in-one file;
- Future, multiple steps to build docker in AWS and test in Intel-gpu-omni agents)

Test Plan

Test Result


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. Please provide the test scripts & test commands. Please state the reasons if your codes don't require additional test scripts. For test file guidelines, please check the test style doc
  • The test results. Please paste the results comparison before and after, or the e2e results.
  • (Optional) The necessary documentation update, such as updating supported_models.md and examples for a new model. Please run mkdocs serve to sync the documentation editions to ./docs.
  • (Optional) Release notes update. If your change is user-facing, please update the release notes draft.

BEFORE SUBMITTING, PLEASE READ https://github.com/vllm-project/vllm-omni/blob/main/CONTRIBUTING.md (anything written below this line will be removed by GitHub Actions)

Signed-off-by: Chendi Xue <chendi.xue@intel.com>
@xuechendi
Copy link
Copy Markdown
Contributor Author

@tjtanaa , need help here.
I am trying to prepare the CI dispatch codes corresponding to #1340
Wondering how this works. In main repo, we are doing that by adding a dispatch in ci-infra - https://github.com/vllm-project/ci-infra/blob/5a7d5a3daecce9de45aa709f1422a0a425869a3c/buildkite/test-template-ci.j2#L651-L656

But not sure how it works in OMNI project.

Signed-off-by: Chendi Xue <chendi.xue@intel.com>
Signed-off-by: Chendi Xue <chendi.xue@intel.com>
@xuechendi xuechendi marked this pull request as ready for review March 9, 2026 22:13
@xuechendi xuechendi requested a review from hsliuustc0106 as a code owner March 9, 2026 22:13
@xuechendi
Copy link
Copy Markdown
Contributor Author

@khluu @hsliuustc0106 , I think the PR should be working now.
I set CI as soft-fail with a fake test for test out.

CI node is registered to buildkite agent queue. Wondering if we can get this PR merged and test out?

@gcanlin
Copy link
Copy Markdown
Collaborator

gcanlin commented Mar 10, 2026

Should this PR wait for #1340?

Signed-off-by: Chendi Xue <chendi.xue@intel.com>
@congw729
Copy link
Copy Markdown
Collaborator

Please let me know if you need an extra Buildkite pipeline for intel CI tests.

Comment thread .buildkite/pipeline.yml Outdated
agents:
queue: "cpu_queue_premerge"

# L5 Test
Copy link
Copy Markdown
Collaborator

@congw729 congw729 Mar 11, 2026

Choose a reason for hiding this comment

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

L5 tests mean stability & reliability tests as mentioned in the table in https://docs.vllm.ai/projects/vllm-omni/en/latest/contributing/ci/CI_5levels/#document-overview. Comment like # Intel tests is fine for your case.

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.

Got it, I have removed changes in pipeline.yml for now. I think with a separate buildkite pipeline makes more sense.

@congw729
Copy link
Copy Markdown
Collaborator

Will this PR not introduce any tests run on the XPU?
According to my understanding, this PR only handles the dispatch for Intel CI tests, right?

@xuechendi
Copy link
Copy Markdown
Contributor Author

xuechendi commented Mar 11, 2026

Will this PR not introduce any tests run on the XPU? According to my understanding, this PR only handles the dispatch for Intel CI tests, right?

yes, this PR is to confirm dispatch works, and xpu CI is added with: #1340

@congw729
Copy link
Copy Markdown
Collaborator

#1340

Good. I saw this PR yesterday and left some comments about pytest marks. Please consider it. If you prefer instant communication, you can find me in the WeChat group.

@xuechendi
Copy link
Copy Markdown
Contributor Author

@congw729 thanks. For this PR, I would like to get it in firstly, so we can test with #1340
Will check with @ywang96 if we are good to go with adding a new Pipeline for "Intel CI OMNI"

@Gaohan123 Gaohan123 added this to the v0.18.0 milestone Mar 13, 2026
Signed-off-by: Chendi Xue <chendi.xue@intel.com>
Signed-off-by: Chendi Xue <chendi.xue@intel.com>
@xuechendi
Copy link
Copy Markdown
Contributor Author

Hi, @congw729 , have confirmed with @ywang96 , please help to

  1. add vllm-omni INTEL CI pipeline
  2. help to trigger .buildkite/bootstrap-intel-omni.sh in this PR

I have added 3 agents
image

@congw729
Copy link
Copy Markdown
Collaborator

congw729 commented Mar 18, 2026

Pipeline created. Need the admin to make it public and set the cluster.
image

@xuechendi
Copy link
Copy Markdown
Contributor Author

@congw729 , Thanks,
may you also help to merge this PR? Otherwise the .buildkite/bootstrap-intel-omni.sh is not in repo.

@gcanlin gcanlin added the ready label to trigger buildkite CI label Mar 18, 2026
@xuechendi
Copy link
Copy Markdown
Contributor Author

xuechendi commented Mar 18, 2026

Thanks, @gcanlin
seems the test didn't get dispatched to https://buildkite.com/vllm/vllm-omni-intel-ci
does that mean only when bootstrap-intel-omni.sh gets landed in main, the vllm-omni-intel-ci will pop out?

@gcanlin
Copy link
Copy Markdown
Collaborator

gcanlin commented Mar 18, 2026

Thanks, @gcanlin seems the test didn't get dispatched to https://buildkite.com/vllm/vllm-omni-intel-ci does that mean only when bootstrap-intel-omni.sh gets landed in main, the vllm-omni-intel-ci will pop out?

Not really sure. @tjtanaa Do you know anything about it?

@tjtanaa
Copy link
Copy Markdown
Contributor

tjtanaa commented Mar 18, 2026

Thanks, @gcanlin seems the test didn't get dispatched to https://buildkite.com/vllm/vllm-omni-intel-ci does that mean only when bootstrap-intel-omni.sh gets landed in main, the vllm-omni-intel-ci will pop out?

Not really sure. @tjtanaa Do you know anything about it?

Let me check, if the buildkite commands are setup correctly, it should be able to invoke the bash script from this PR branch. We don't have to wait until the bash script landed on main

@congw729
Copy link
Copy Markdown
Collaborator

This PR has to be merged before triggering the INTEL CI pipeline, please take a look @hsliuustc0106 @Gaohan123

@tjtanaa
Copy link
Copy Markdown
Contributor

tjtanaa commented Mar 18, 2026

@xuechendi @congw729 I helped trigger the build to validate this PR https://buildkite.com/vllm/vllm-omni-intel-ci/builds/8/steps/canvas

hsliuustc0106 and others added 3 commits March 18, 2026 23:11
Signed-off-by: Chendi Xue <chendi.xue@intel.com>
Signed-off-by: Chendi Xue <chendi.xue@intel.com>
@xuechendi
Copy link
Copy Markdown
Contributor Author

xuechendi commented Mar 18, 2026

@Gaohan123 @hsliuustc0106 ,
@tjtanaa helped me to manual trigger the CI and now CI passed
https://buildkite.com/vllm/vllm-omni-intel-ci/builds/10/steps/canvas

image

Copy link
Copy Markdown
Collaborator

@congw729 congw729 left a comment

Choose a reason for hiding this comment

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

LGTM. This PR can be merged @hsliuustc0106

@gcanlin gcanlin merged commit e1ac182 into vllm-project:main Mar 19, 2026
4 of 5 checks passed
fhfuih pushed a commit to fhfuih/vllm-omni that referenced this pull request Mar 19, 2026
Signed-off-by: Chendi Xue <chendi.xue@intel.com>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>
yiliu30 pushed a commit to yiliu30/vllm-omni-fork that referenced this pull request Mar 20, 2026
Signed-off-by: Chendi Xue <chendi.xue@intel.com>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>

Signed-off-by: yiliu30 <yi4.liu@intel.com>
clodaghwalsh17 pushed a commit to clodaghwalsh17/nm-vllm-omni-ent that referenced this pull request May 12, 2026
Signed-off-by: Chendi Xue <chendi.xue@intel.com>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

ready label to trigger buildkite CI

Projects

None yet

Development

Successfully merging this pull request may close these issues.

7 participants