| Summary: | 4 tests of ECM 5.11.0 fail | ||
|---|---|---|---|
| Product: | [Frameworks and Libraries] extra-cmake-modules | Reporter: | Heiko Becker <heiko.becker> |
| Component: | general | Assignee: | Alex Merry <alex.merry> |
| Status: | RESOLVED FIXED | ||
| Severity: | normal | CC: | ecm-bugs-null |
| Priority: | NOR | ||
| Version First Reported In: | unspecified | ||
| Target Milestone: | --- | ||
| Platform: | Exherbo | ||
| OS: | Linux | ||
| Latest Commit: | Version Fixed/Implemented In: | ||
| Sentry Crash Report: | |||
| Attachments: | Full output of failing tests | ||
Created attachment 93121 [details]
Full output of failing tests
Is this report still valid with recent ECM versions ? (In reply to Christophe Giboudeaux from comment #2) > Is this report still valid with recent ECM versions ? Apparently no. |
The following tests FAILED: 15 - ecm_add_tests-single_tests (Failed) 16 - ecm_add_tests_did_run-single_tests (Failed) 17 - ecm_add_tests-multi_tests (Failed) 18 - ecm_add_tests_did_run-multi_tests (Failed) I'll attach the full output of those tests. 16 and 18 only seem to fail because 15 and 17 didn't succeed. 15 and 17 fail due to ar and ranlib being prefixed with the build architecture. Normally we pass CMAKE_AR and CMAKE_RANLIB to cmake builds but that doesn't get picked up when called from the add_check macro (as well as env variables, it seems). What I don't know is how to fix this in way that would make it configureable or doesn't break with "normal" ar and ranlib. Reproducible: Always Steps to Reproduce: 1.cmake ... -DCMAKE_AR:PATH=x86_64-pc-linux-gnu-ar -DCMAKE_RANLIB:PATH=x86_64-pc-linux-gnu-ranlib ... -DBUILD_TESTING:BOOL=TRUE 2. make 3. make ARGS="--verbose" test