| Summary: | qemux86-64 packages packaged as qemux86 | ||
|---|---|---|---|
| Product: | [Build System, Metadata & Runtime] OE-Core | Reporter: | Qing He <qing.he> |
| Component: | core | Assignee: | Mark Hatle <mark.hatle> |
| Status: | RESOLVED FIXED | QA Contact: | |
| Severity: | major | ||
| Priority: | High | CC: | jiajun.xu, kevin.tian, mark.hatle, meta.mr.watcher, meta.watcher, richard.purdie |
| Version: | 1.0 | ||
| Target Milestone: | 1.0 M4 | ||
| Hardware: | x86 | ||
| OS: | x86_64 | ||
| Whiteboard: | |||
| OS type for building Yocto: | --- | Type of Regression: | --- |
| Verified: | Documentation change: | --- | |
|
Description
Qing He
2011-03-11 01:35:19 UTC
Some findings after a brief investigation:
${PACKAGE_ARCH} is qemux86-64 and this is passed to rpmbuild as
--target qemux86-64-poky-linux-gnu
That's where it goes wrong.
I don't know the reason to use qemux86-64 instead of qemux86_64 at the very beginning, but guess it's to avoid things like FOO_qemux86_64 in bitbake?
Can we use MACHINE_ARCH_qemux86-64 = "qemux86_64" and not break anything?
I agree. We need to resolve this issue by updating / translating the architecture.. I believe Qing's patch resolved this. Please re-verify. Fix: http://git.pokylinux.org/cgit.cgi/poky/commit/?h=bernard&id=8797e389c6bd84a20da4cafce1df8568213d3b85 |