| Summary: | Behavior change: projects setup with bitbake-setup no longer respects MACHINE=xxx env settings | ||
|---|---|---|---|
| Product: | [Build System, Metadata & Runtime] BitBake | Reporter: | Chen Qi <Qi.Chen> |
| Component: | bitbake | Assignee: | Chen Qi <Qi.Chen> |
| Status: | RESOLVED FIXED | QA Contact: | |
| Severity: | normal | ||
| Priority: | Medium+ | CC: | alex.kanavin, ccasciato, poky.bs.watcher, poky.watcher, randy.macleod, richard.purdie |
| Version: | unspecified | ||
| Target Milestone: | 5.3 M4 | ||
| Hardware: | x86 | ||
| OS: | Multiple | ||
| Whiteboard: | |||
| OS type for building Yocto: | --- | Type of Regression: | --- |
| Verified: | Documentation change: | Yes (doc changes required) | |
|
Description
Chen Qi
2025-11-12 05:48:24 UTC
I submitted a patch: https://lists.openembedded.org/g/bitbake-devel/message/18343 We think the best way to handle this is to error if a fragment is set and there is also a value for the variable already set (from the environment or the conf files). We could recommend the user unset the fragment with: bitbake-config-build disable-fragment @Alex & @Richard, The above patch has two things that I don't feel good: 1) The env pass-through mechanism is used to avoid 'bitbake-config-build' error out. 2) The '??=' is OK, but '?=' still triggers an error. If you have some suggestion, please let me know. V3 patchset: https://lists.openembedded.org/g/bitbake-devel/message/18449 https://lists.openembedded.org/g/bitbake-devel/message/18450 Patch V4 (fix an issue with resident bitbake server): https://lists.openembedded.org/g/bitbake-devel/message/18466 https://lists.openembedded.org/g/bitbake-devel/message/18467 https://lists.openembedded.org/g/bitbake-devel/message/18468 |