Bug 2278709
| Summary: | Review Request: rust-buddy-alloc - Memory allocator for no-std Rust, used for embedded environments | ||
|---|---|---|---|
| Product: | [Fedora] Fedora | Reporter: | Jordan Rome <jordan> |
| Component: | Package Review | Assignee: | Fabio Valentini <decathorpe> |
| Status: | CLOSED NOTABUG | QA Contact: | Fedora Extras Quality Assurance <extras-qa> |
| Severity: | unspecified | Docs Contact: | |
| Priority: | unspecified | ||
| Version: | rawhide | CC: | decathorpe, package-review |
| Target Milestone: | --- | ||
| Target Release: | --- | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| URL: | https://crates.io/crates/buddy-alloc | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | If docs needed, set a value | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2025-04-29 00:45:30 UTC | Type: | --- |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
| Bug Depends On: | |||
| Bug Blocks: | 201449 | ||
|
Description
Jordan Rome
2024-05-02 19:56:56 UTC
This package built on koji: https://koji.fedoraproject.org/koji/taskinfo?taskID=117164443 Copr build: https://copr.fedorainfracloud.org/coprs/build/7399587 (succeeded) Review template: https://download.copr.fedorainfracloud.org/results/@fedora-review/fedora-review-2278709-rust-buddy-alloc/fedora-rawhide-x86_64/07399587-rust-buddy-alloc/fedora-review/review.txt Found issues: - License file LICENSE is not marked as %license Read more: https://docs.fedoraproject.org/en-US/packaging-guidelines/LicensingGuidelines/#_license_text Please know that there can be false-positives. --- This comment was created by the fedora-review-service https://github.com/FrostyX/fedora-review-service If you want to trigger a new Copr build, add a comment containing new Spec and SRPM URLs or [fedora-review-service-build] string. This fails to build for me:
> File not found: /builddir/build/BUILDROOT/rust-buddy-alloc-0.5.1-1.fc41.x86_64/usr/share/cargo/registry/buddy-alloc-0.5.1/LICENSE-MIT.txt
> File not found: /builddir/build/BUILDROOT/rust-buddy-alloc-0.5.1-1.fc41.x86_64/usr/share/cargo/registry/buddy-alloc-0.5.1/Readme.md
Sorry for the delay. Do you mind giving it another try? Package builds now, thanks!
Two small issues:
1. Please don't include the rpmlintrc file in the spec / SRPM though. It's unrelated to the RPM build.
You can include it in the dist-git repo after importing the package though.
2. I cannot reproduce the test failures.
> # fast_alloc tests end in a segfault on x86
> %cargo_test -- -- --skip failing_tests --skip tests::fast_alloc
I can run "cargo test" and "cargo test --release" in the sources on my x86 machine just fine.
> I can run "cargo test" and "cargo test --release" in the sources on my x86 machine just fine. The comment in the spec might just be wrong. I ran `mock -r fedora-rawhide-x86_64 --sources . --spec rust-buddy-alloc.spec --postinstall ` locally on my aarch64 (Mac) with Parallels and I got that failure (maybe because of the architecture mismatch?). https://pastebin.com/NYejgAYp If I run `cargo test` from the buddy-alloc repo directly it works just fine. Test failure is still happening with `mock -r fedora-rawhide-aarch64 --sources . --spec rust-buddy-alloc.spec --postinstall` - maybe it has something to do with being in a VM. > Please don't include the rpmlintrc file in the spec / SRPM though. It's unrelated to the RPM build. You can include it in the dist-git repo after importing the package though.
Done - removed it from the rpm and spec files.
> Test failure is still happening with `mock -r fedora-rawhide-aarch64 --sources . --spec rust-buddy-alloc.spec --postinstall` - maybe it has something to do with being in a VM.
OK, if you want to skip tests because koji builds happen either on "bare metal" or in VMs, and they will fail if they're scheduled to run in a VM, please document that. Right now there is no documented reason why some tests are skipped.
> please document that
Makes sense. I updated the spec file with a comment.
Sorry, the new comment is only more confusing to me. I don't see how Fedora 38 is relevant, it is EOL. Additionally, the files behind the spec and SRPM links now mismatch. I ran a rawhide scratch build of your package (without skipped tests): https://koji.fedoraproject.org/koji/taskinfo?taskID=118963473 Tests crashed on *all* architectures with SIGSEGV. So it looks like something is seriously wrong with the code of this crate ... No worries, I updated the comment and also synced the uploaded spec file and SRPM.
> Tests crashed on *all* architectures with SIGSEGV. So it looks like something is seriously wrong with the code of this crate ...
What's strange is that when I run `cargo test` on this package on my same machine (where the mock build fails), they pass as normal.
No worries, I updated the comment and also synced the uploaded spec file and SRPM.
> Tests crashed on *all* architectures with SIGSEGV. So it looks like something is seriously wrong with the code of this crate ...
What's strange is that when I run `cargo test` on this package on my same machine (where the mock build fails), they pass as normal.
Can you try running "cargo test --release"? Tests are run and built in "release" mode during package builds. If the presence of "--release" makes a difference, then it sounds like a soundness bug that is exposed by turning on optimizations. > Can you try running "cargo test --release"?
Ah ha. Now I see the failure. I guess this is something we need to fix upstream before packaging this?
If this is indeed a soundness issue (which is what it looks like), then yes, I would prefer if it was fixed before we package this for Fedora. Can you file bug against upstream for this issue? Looks like the issues have been fixed upstream! Can you update this review for the new version that was published? This is an automatic action taken by review-stats script. The ticket submitter failed to clear the NEEDINFO flag in a month. As per https://fedoraproject.org/wiki/Policy_for_stalled_package_reviews we consider this ticket as DEADREVIEW and proceed to close it. |