Repository navigation
Revert MSI publishing from the release publish script (#8617) - #8629
Merged
Merged
Conversation
PR #8097 added publish_native_windows_packages() and called it unconditionally for --platform=windows, but nothing builds MSIs during a release yet (multi_arch_build_native_windows_packages.yml is not wired into the release pipeline), so the {run_id}-windows/packages/msi prefix is always empty and the publish job failed every nightly run. Rather than paper over the empty prefix, revert MSI publishing from the release path entirely until the build workflow is actually wired in: - Restore publish_rocm_to_release_buckets.py and its test to their pre-#8097 state: drop publish_native_windows_packages(), the windows dispatch branch, the docstring/checklist entry, and the "Windows msi" wording on --skip-native-packages (back to Linux deb/rpm only). Windows still publishes tarballs and python packages as before. - Remove the --skip-native-packages workaround #8618 added to multi_arch_release_windows.yml; with no windows native-package publish, it is a no-op there. MSI publishing will be reintroduced together with wiring the build workflow into the release pipeline (#1987). Co-Authored-By: Claude Opus 4 (1M context) <noreply@anthropic.com>
✅ All Checks Passed — Ready for Review
📖 Need help? See the Policy FAQ for details on every check and how to fix failures. |
|
🎉 All checks passed! This PR is ready for review. |
ScottTodd
approved these changes
Sep 30, 2026
This was referenced Oct 1, 2026
idass1990
added a commit
that referenced
this pull request
Oct 15, 2026
Re-adds publish_native_windows_packages() to the release publish script, previously reverted in #8629 to fix the #8617 regression. The original #8097 implementation published to core/packages/msi/, which does not match RFC0012: per the repo-structure RFC, Windows installer artifacts (.exe/.msi) belong under core/windows-installers/, not core/packages/. Addresses marbre's post-merge review comment on #8097 (discussion_r4145328177). The current MSI is a single OS-agnostic installer, so it is published to a flat windows-installers/ prefix (dated subdir for dev/nightly, stable path for prerelease) rather than the per-OS subfolders (win11/, server2022/, server2025/) RFC0012 reserves for OS-specific installers — our MSI payload is identical across Windows versions. NOTE: do not merge until the MSI build workflow (multi_arch_build_native_windows_packages.yml) is wired into the release pipeline (Milestone 4, #1987). Publishing hard-fails on zero MSI files; merging this alone, while the build has no callers, reintroduces #8617. Co-Authored-By: Claude Opus 4 (1M context) <noreply@anthropic.com>
3 tasks done
idass1990
added a commit
that referenced
this pull request
Oct 20, 2026
Part 2 of 2 of the Milestone 4 MSI wiring (#1987), stacked on the CI-wiring PR. With the MSI build already wired into the release workflow's build graph, this adds the release-only pieces: - publish_rocm_to_release_buckets.py: re-add publish_native_windows_packages() (reverted in #8629 because the build wasn't wired in — the #8617 regression). MSIs now copy to v5/rocm/core/windows-installers/{date}-{run_id} per RFC0012. - multi_arch_release_windows.yml: add build_native_msi_packages, make publish_to_release_buckets depend on it (the needs-edge that guarantees the MSI prefix is non-empty before publish — the fix for #8617), add trigger_test_native_windows_install (async install-test dispatch), and thread quartz_tracking_id to both new edges. - test_native_windows_packages_install.yml: declare quartz_tracking_id so the release graph's propagation check stays green. Together with the CI-wiring PR this reconstructs the original combined change exactly (verified: identical trees). Co-Authored-By: Claude Opus 4 (1M context) <noreply@anthropic.com>
idass1990
added a commit
that referenced
this pull request
Oct 26, 2026
Part 2 of 2 of the Milestone 4 MSI wiring (#1987), stacked on the CI-wiring PR. With the MSI build already wired into the release workflow's build graph, this adds the release-only pieces: - publish_rocm_to_release_buckets.py: re-add publish_native_windows_packages() (reverted in #8629 because the build wasn't wired in — the #8617 regression). MSIs now copy to v5/rocm/core/windows-installers/{date}-{run_id} per RFC0012. - multi_arch_release_windows.yml: add build_native_msi_packages, make publish_to_release_buckets depend on it (the needs-edge that guarantees the MSI prefix is non-empty before publish — the fix for #8617), add trigger_test_native_windows_install (async install-test dispatch), and thread quartz_tracking_id to both new edges. - test_native_windows_packages_install.yml: declare quartz_tracking_id so the release graph's propagation check stays green. Together with the CI-wiring PR this reconstructs the original combined change exactly (verified: identical trees). Co-Authored-By: Claude Opus 4 (1M context) <noreply@anthropic.com>
idass1990
added a commit
that referenced
this pull request
Nov 2, 2026
Part 2 of 2 of the Milestone 4 MSI wiring (#1987), stacked on the CI-wiring PR. With the MSI build already wired into the release workflow's build graph, this adds the release-only pieces: - publish_rocm_to_release_buckets.py: re-add publish_native_windows_packages() (reverted in #8629 because the build wasn't wired in — the #8617 regression). MSIs now copy to v5/rocm/core/windows-installers/{date}-{run_id} per RFC0012. - multi_arch_release_windows.yml: add build_native_msi_packages, make publish_to_release_buckets depend on it (the needs-edge that guarantees the MSI prefix is non-empty before publish — the fix for #8617), add trigger_test_native_windows_install (async install-test dispatch), and thread quartz_tracking_id to both new edges. - test_native_windows_packages_install.yml: declare quartz_tracking_id so the release graph's propagation check stays green. Together with the CI-wiring PR this reconstructs the original combined change exactly (verified: identical trees). Co-Authored-By: Claude Opus 4 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Fixes #8617 — the Windows
Publish to Release Bucketsjob fails every nightly run withFileNotFoundError: No msi packages found.PR #8097 added
publish_native_windows_packages()and called it unconditionally for--platform=windows. But nothing builds MSIs during a release yet:multi_arch_build_native_windows_packages.ymlis not wired into the release (or CI) pipeline, so{run_id}-windows/packages/msi/is always empty and the publish hard-fails.Rather than paper over the empty prefix, this reverts MSI publishing from the release path until the build workflow is actually wired in.
Changes
publish_rocm_to_release_buckets.pyand its test to their pre-ci: Windows MSI build workflow + dev-bucket upload (Milestone 2) #8097 state:publish_native_windows_packages()and theplatform == "windows"dispatch branch.--skip-native-packageshelp text to "native Linux packages (deb/rpm)".--skip-native-packagesworkaround that fix(bug): skip native MSI publish in Windows release until build workflow is wired in #8618 added tomulti_arch_release_windows.yml; with no windows native-package publish, it is a no-op there.MSI publishing will be reintroduced together with wiring the build workflow into the release pipeline (#1987).
Relationship to #8618
#8618 worked around this by passing
--skip-native-packagesin the windows release step. That reuses the Linux skip flag (couples deb/rpm and msi skipping), is a manual flag someone must remember to remove, and leaves the unused MSI publish code in the script. This PR removes the code path instead, so there is nothing to skip and nothing to remember.Test plan
pytest build_tools/github_actions/tests/publish_rocm_to_release_buckets_test.py— 14 tests pass (the pre-ci: Windows MSI build workflow + dev-bucket upload (Milestone 2) #8097 set; the windows case now asserts tarballs + python only, no msi).pre-commit(black + actionlint) on all changed files.