minke.yu/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 12 Packages Projects Releases Wiki Activity

pr-test-extra: re-trigger on labeled event (#25732)

Browse Source
This commit is contained in:
Liangsheng Yin
2026-05-19 05:15:55 -07:00
committed by GitHub
parent 0e4d1b49d3
commit e0273dcd31
5 changed files with 76 additions and 14 deletions
Download Patch File Download Diff File Expand all files Collapse all files
.claude/skills/ci-workflow-guide/SKILL.md
+16 -2
View File
@@ -22,6 +22,7 @@ This skill covers the CI **infrastructure** layer — how tests are dispatched,
| File | Role |
|------|------|
| `.github/workflows/pr-test.yml` | Main workflow — all stages, jobs, conditions, matrix definitions |
| `.github/workflows/pr-test-extra.yml` | Extra workflow — gated by BOTH `run-ci` and `run-ci-extra` labels |
| `.github/workflows/pr-gate.yml` | PR gating: draft check, `run-ci` label, per-user rate limiting |
| `.github/actions/check-pr-test-health/action.yml` | Cross-job fast-fail: queries API for any failed job |
| `.github/actions/wait-for-jobs/action.yml` | Stage gating: polls API until stage jobs complete |
@@ -381,10 +382,23 @@ group: pr-test-{event_name}-{branch}-{pr_sha}-{stage}
| Command | Effect |
|---------|--------|
| `/tag-run-ci-label` | Adds `run-ci` label to PR |
| `/tag-run-ci-label extra` | Adds both `run-ci` and `run-ci-extra` labels |
| `/rerun-failed-ci` | Reruns failed jobs in the latest workflow run |
| `/tag-and-rerun-ci` | Adds label + reruns |
| `/rerun-stage <stage>` | Dispatches `pr-test.yml` with `target_stage=<stage>` |
| `/tag-and-rerun-ci` | Adds `run-ci` label + reruns failed |
| `/tag-and-rerun-ci extra` | Adds both `run-ci` and `run-ci-extra` labels + reruns failed |
| `/rerun-stage <stage>` | Deprecated; posts deprecation notice |
| `/rerun-test <test-file>` | Reruns a specific test file via `rerun-test.yml` |
| `/rerun-group <group> [<group> ...]` | Expands registered test groups, then reuses `/rerun-test` |
Handled by `scripts/ci/utils/slash_command_handler.py` → `.github/workflows/slash-command-handler.yml`.
### Label-gated workflow dispatch (pr-test, pr-test-extra)
`pr-test.yml` and `pr-test-extra.yml` both listen for `pull_request.labeled` (in addition to `opened`/`synchronize`/`reopened`). The `check-changes.if` gate has two clauses:
1. **For `labeled` events**: the just-added label must be one of the gating labels (`run-ci` for pr-test, `run-ci` or `run-ci-extra` for pr-test-extra) — otherwise every unrelated label addition would dispatch a full CI run.
2. **All events**: the PR must currently carry the required labels.
This is what lets `/tag-run-ci-label` (and the `extra` variant) trigger a fresh CI run without an extra push.
**Caveat — skipped runs cannot be un-skipped by `run.rerun()`:** GitHub's rerun API reuses the original event payload, so rerunning a `pull_request`-event run that was skipped because of missing labels will skip again (label set in the frozen payload doesn't update). The only way to recover a label-skipped run is to add the missing label, which fires a fresh `labeled` event with the current label set. `handle_rerun_failed_ci` in the slash handler is for rerunning failed/non-label-skipped runs; it cannot revive label-skipped ones.
.github/workflows/pr-test-extra.yml
+12
View File
@@ -12,6 +12,11 @@ name: PR Test Extra
on:
pull_request:
branches: [main]
# `labeled` lets the workflow re-fire when `run-ci-extra` (or `run-ci`)
# is added after the latest push — otherwise GHA leaves the stale
# skipped run in place and there's no way to enable extra without
# another push. See check-changes.if below for the matching guard.
types: [opened, synchronize, reopened, labeled]
workflow_dispatch:
inputs:
force_continue_on_error:
@@ -70,10 +75,17 @@ jobs:
# (workflow_dispatch / workflow_call) always run. When this job is
# skipped by the gate, every downstream caller stub naturally skips
# because its needs do not resolve.
#
# For `labeled` events we additionally require the just-added label to be
# one of the two gating labels — otherwise every unrelated label addition
# would dispatch a full CI run.
check-changes:
if: |
github.event_name != 'pull_request' ||
(
(github.event.action != 'labeled' ||
github.event.label.name == 'run-ci' ||
github.event.label.name == 'run-ci-extra') &&
contains(github.event.pull_request.labels.*.name, 'run-ci') &&
contains(github.event.pull_request.labels.*.name, 'run-ci-extra')
)
docs_new/docs/developer_guide/contribution_guide.mdx
+2 -2
View File
@@ -114,9 +114,9 @@ Users with permission are listed in the [CI_PERMISSIONS.json](https://github.com
For CI to run on a pull request, it must have the "run-ci" label. Authorized users can add the label or rerun failed tests by commenting on the PR with one of these commands:
- `/tag-run-ci-label`: Adds the "run-ci" label. Every future commit will trigger CI.
- `/tag-run-ci-label`: Adds the "run-ci" label. Every future commit will trigger CI. Add the `extra` argument (`/tag-run-ci-label extra`) to additionally apply the "run-ci-extra" label, opting the PR into the extra test workflow (`pr-test-extra.yml`).
- `/rerun-failed-ci`: Reruns the failed or flaky tests from the most recent commit.
- `/tag-and-rerun-ci`: A single command that performs both `/tag-run-ci-label` and `/rerun-failed-ci`.
- `/tag-and-rerun-ci`: A single command that performs both `/tag-run-ci-label` and `/rerun-failed-ci`. Accepts the same `extra` argument (`/tag-and-rerun-ci extra`).
- `/rerun-stage <stage-name>`: Reruns a specific test stage without waiting for its dependencies. This is useful when you want to quickly validate a fix for a specific test failure instead of waiting ~30 minutes for preceding stages to complete.
- `/rerun-test <test-spec> [<test-spec> ...]`: Reruns one or more specific tests directly, bypassing stage boundaries. Each `<test-spec>` is pytest-style `<file>::<TestClass>[.<test_method>]` (the `::TestClass` and `.<test_method>` parts are optional). The handler resolves each spec, groups specs by their registered runner-label, and dispatches one [Rerun Test workflow](https://github.com/sgl-project/sglang/actions/workflows/rerun-test.yml) per group. Examples: `/rerun-test test_srt_endpoint.py`, `/rerun-test registered/core/test_srt_endpoint.py::TestSRTEndpoint.test_simple_decode`, `/rerun-test test_a.py test_b.py` (multiple at once).
docs_new/docs/hardware-platforms/ascend-npus/ascend_contribution_guide.mdx
+2 -2
View File
@@ -94,9 +94,9 @@ Users with permission are listed in the [CI_PERMISSIONS.json](https://github.com
For CI to run on a pull request, it must have the "run-ci" label. Authorized users can add the label or rerun failed tests by commenting on the PR with one of these commands:
- `/tag-run-ci-label`: Adds the "run-ci" label. Every future commit will trigger CI.
- `/tag-run-ci-label`: Adds the "run-ci" label. Every future commit will trigger CI. Add the `extra` argument (`/tag-run-ci-label extra`) to additionally apply the "run-ci-extra" label, opting the PR into the extra test workflow (`pr-test-extra.yml`).
- `/rerun-failed-ci`: Reruns the failed or flaky tests from the most recent commit.
- `/tag-and-rerun-ci`: A single command that performs both `/tag-run-ci-label` and `/rerun-failed-ci`.
- `/tag-and-rerun-ci`: A single command that performs both `/tag-run-ci-label` and `/rerun-failed-ci`. Accepts the same `extra` argument (`/tag-and-rerun-ci extra`).
- `/rerun-stage <stage-name>`: Reruns a specific test stage without waiting for its dependencies. This is useful when you want to quickly validate a fix for a specific test failure instead of waiting ~30 minutes for preceding stages to complete.
If you have permission, the [Slash Command Handler](https://github.com/sgl-project/sglang/actions/workflows/slash-command-handler.yml) will run your command and react with a 👍 to your comment. It may take up to a few minutes for the reaction to appear. Here’s a usage [example](https://github.com/sgl-project/sglang/pull/14253#issuecomment-3599509302).
scripts/ci/utils/slash_command_handler.py
+44 -8
View File
@@ -282,23 +282,46 @@ def has_sgl_kernel_changes(pr):
return False
def handle_tag_run_ci(gh_repo, pr, comment, user_perms, react_on_success=True):
def handle_tag_run_ci(
gh_repo, pr, comment, user_perms, react_on_success=True, tag_extra=False
):
"""
Handles the /tag-run-ci-label command.
When tag_extra is True (triggered by the `extra` argument), also adds the
`run-ci-extra` label. pr-test-extra.yml gates on BOTH `run-ci` and
`run-ci-extra`, so both must be present for the extra workflow to run —
we always add `run-ci` alongside `run-ci-extra`. Reuses the same
`can_tag_run_ci_label` permission.
How fresh runs get dispatched: pr-test.yml and pr-test-extra.yml both
include `labeled` in `on.pull_request.types`, so adding a label fires a
new `pull_request.labeled` event with the up-to-date label set in its
payload, which spawns a fresh workflow run that satisfies the
`check-changes.if` gate. Note that this is the ONLY way to "un-skip" a
label-gated run — `run.rerun()` on a previously-skipped pull_request run
reuses the original event payload (frozen labels), so it would skip
again. handle_rerun_failed_ci can't recover label-skipped runs; the
labeled event is the recovery mechanism.
Returns True if action was taken, False otherwise.
"""
if not user_perms.get("can_tag_run_ci_label", False):
print("Permission denied: can_tag_run_ci_label is false.")
return False
print("Permission granted. Adding 'run-ci' label.")
pr.add_to_labels("run-ci")
labels = ["run-ci"]
if tag_extra:
labels.append("run-ci-extra")
print(f"Permission granted. Adding labels: {labels}.")
for label in labels:
pr.add_to_labels(label)
if react_on_success:
comment.create_reaction("+1")
print("Label added and comment reacted.")
print("Labels added and comment reacted.")
else:
print("Label added (reaction suppressed).")
print("Labels added (reaction suppressed).")
return True
@@ -366,6 +389,14 @@ def handle_rerun_failed_ci(gh_repo, pr, comment, user_perms, react_on_success=Tr
# core.setFailed(...) so their conclusion is "failure" and are covered.
# - skipped: the entire run was skipped (no jobs ran), so there are no
# failed jobs for rerun_failed_jobs() to target. Use run.rerun().
#
# Caveat: GitHub's `run.rerun()` reuses the original event payload, so
# reruns of `pull_request`-event runs that were skipped because their
# `if` evaluated to false (e.g. missing label) will skip again — the
# label set in the frozen payload doesn't update. To un-skip a
# label-gated workflow, add the missing label (the `labeled` event
# dispatches a fresh run with the current label set); this function
# cannot recover those by rerun alone.
# - kernel wheel escape: if the PR touches sgl-kernel and not all wheel
# builds are success yet, full-rerun failure runs too — Build Wheel
# lives in pr-test-sgl-kernel.yml, consumers in pr-test.yml, and
@@ -1071,19 +1102,24 @@ def main():
# 4. Parse Command and Execute
first_line = comment_body.split("\n")[0].strip()
# `extra` argument opts in to also tagging `run-ci-extra`. Both
# `/tag-run-ci-label extra` and `/tag-and-rerun-ci extra` share this
# parser so the surface is symmetric.
tokens = first_line.split()
tag_extra = len(tokens) > 1 and "extra" in tokens[1:]
if first_line.startswith("/tag-run-ci-label"):
handle_tag_run_ci(repo, pr, comment, user_perms)
handle_tag_run_ci(repo, pr, comment, user_perms, tag_extra=tag_extra)
elif first_line.startswith("/rerun-failed-ci"):
handle_rerun_failed_ci(repo, pr, comment, user_perms)
elif first_line.startswith("/tag-and-rerun-ci"):
# Perform both actions, but suppress individual reactions
print("Processing combined command: /tag-and-rerun-ci")
print(f"Processing combined command: /tag-and-rerun-ci (tag_extra={tag_extra})")
tagged = handle_tag_run_ci(
repo, pr, comment, user_perms, react_on_success=False
repo, pr, comment, user_perms, react_on_success=False, tag_extra=tag_extra
)
# Wait for the label to propagate before triggering rerun