Bug 1261196
| Summary: | test-metrics fails on aarch64 | ||
|---|---|---|---|
| Product: | [Fedora] Fedora | Reporter: | Peter Robinson <pbrobinson> |
| Component: | cockpit | Assignee: | Marius Vollmer <mvollmer> |
| Status: | CLOSED ERRATA | QA Contact: | Fedora Extras Quality Assurance <extras-qa> |
| Severity: | unspecified | Docs Contact: | |
| Priority: | unspecified | ||
| Version: | 23 | CC: | mjuszkie, sgallagh, stefw |
| Target Milestone: | --- | ||
| Target Release: | --- | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| Whiteboard: | |||
| Fixed In Version: | cockpit-0.78-1.fc23 | Doc Type: | Bug Fix |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2015-10-02 13:25:42 UTC | Type: | Bug |
| 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: | 922257 | ||
|
Description
Peter Robinson
2015-09-08 21:01:08 UTC
Still broken in 0.75 http://arm.koji.fedoraproject.org/koji/buildinfo?buildID=315087 I do not see the 'Inappropriate ioctl for device' in the build log anymore. Are you talking about a separate problem? This bug has gone into commit messages as fixing exactly the problem you reported. If you wish to report a separate problem, please do so. (In reply to Stef Walter from comment #3) > I do not see the 'Inappropriate ioctl for device' in the build log anymore. > Are you talking about a separate problem? Maybe, the build still fails. Spilt the hair how ever you like... ./test-metrics terminated with SIGABRT FAIL: test-metrics 5 /metrics/interpolate So it looks like it's building successfully but failing in the test suite now. Since we're using one bugzilla bug for multiple bugs, renaming the title to the next one. In the future please report one issue per bugzilla bug. Please provide Marius with shell access to the machine that this problem is happening on. He's the one familiar with this code but I doubt he has a machine of this architecture. > Please provide Marius with shell access to the machine that this problem is
> happening on. He's the one familiar with this code but I doubt he has a
> machine of this architecture.
They are provision-able through beaker
cockpit 0.77 built fine for aarch64 with "-ffp-context=off" given for compiler. (In reply to Marcin Juszkiewicz from comment #8) > cockpit 0.77 built fine for aarch64 with "-ffp-context=off" given for > compiler. The test is fundamentally broken in that it compares results of floating point calculations for exact equality: cockpit-bridge:ERROR:src/bridge/test-metrics.c:392:test_interpolate: [[19.999999999999996,99.999999999999957]] != [[20,100]] Trying to make it pass by finding the right combination of compiler flags would be wrong, we just need to fix that test. A acceptable quick workaround is to just skip the test. The actual Cockpit code seems correct. (Yeah, I knew this when writing the test, but... well, sorry for the pain.) (In reply to Marius Vollmer from comment #9) The code is fine. But (as with rest of CERN software) the issue is related to FMA instructions as they give more precise results while test data is from build without FMA. We had that in HepMC, cockpit and alglib so far. (In reply to Marcin Juszkiewicz from comment #10) > (In reply to Marius Vollmer from comment #9) > > The code is fine. But (as with rest of CERN software) the issue is related > to FMA instructions as they give more precise results while test data is > from build without FMA. > > We had that in HepMC, cockpit and alglib so far. cockpit isn't CERN software. It's server management console stuff Sorry. too much discussions about CERN recently D: Marius will fix the test-interpolate test here for the 0.78 release, and I think that'll fix the issue. Marcin, since we do continuous deployment, the Fedora cockpit master and branched koji branches both get their spec file from upstream: https://github.com/cockpit-project/cockpit/blob/master/tools/cockpit.spec I'll put a note in the cockpit.spec file: that any changes get clobbered regularly, unless they make their way upstream. cockpit-0.78-1.fc23 has been submitted as an update to Fedora 23. https://bodhi.fedoraproject.org/updates/FEDORA-2015-977ba13a92 cockpit-0.78-1.fc23 has been pushed to the Fedora 23 testing repository. If problems still persist, please make note of it in this bug report. If you want to test the update, you can install it with $ su -c 'dnf --enablerepo=updates-testing update cockpit' You can provide feedback for this update here: https://bodhi.fedoraproject.org/updates/FEDORA-2015-977ba13a92 cockpit-0.78-1.fc23 has been pushed to the Fedora 23 testing repository. If problems still persist, please make note of it in this bug report. If you want to test the update, you can install it with $ su -c 'dnf --enablerepo=updates-testing update cockpit' You can provide feedback for this update here: https://bodhi.fedoraproject.org/updates/FEDORA-2015-977ba13a92 cockpit-0.78-1.fc23 has been pushed to the Fedora 23 stable repository. If problems still persist, please make note of it in this bug report. |