Bug 4818

Summary: no version displayed in /etc/version
Product: [Build System, Metadata & Runtime] OE-Core Reporter: Alexandru Georgescu <alexandru.c.georgescu>
Component: coreAssignee: Paul Eggleton <bluelightning>
Status: VERIFIED FIXED QA Contact:
Severity: normal    
Priority: Medium CC: bluelightning, ionutx.chisanovici, meta.mr.watcher, meta.watcher, richard.purdie, yp.bsp.watcher, yp.watcher
Version: 1.5   
Target Milestone: 1.5 M5   
Hardware: x86   
OS: Multiple   
Whiteboard:
OS type for building Yocto: --- Type of Regression: Regression (Used to work)
Verified: Documentation change: No (bug/feature does not impact docs)

Description Alexandru Georgescu 2013-07-05 08:51:56 UTC 
poky: eaa5df34af42b6a37f6506847d0f3ef6ba0d298a
meta-intel: e0d6134ed2e2687ff9f2ee77701666447842bf33

crownbay image: http://autobuilder03.jf.intel.com/pub/nightly-meta-intel/20130703-1/machines/crownbay-noemgd/core-image-sato-sdk-crownbay-noemgd.hddimg


SUMMARY
Usually the version (date) of the BSP is displayed in the /etc/version file.

For this build, no version is displayed.
Tested with crownbay-noemgd and fri2 core-image-sato-sdk images.

STEPS TO REPRODUCE
Install the image to the hardware device. boot it and check the version in /etc/version file.

root@crownbay-noemgd:/# cat /etc/version

root@crownbay-noemgd:/#
Comment 1 Paul Eggleton 2013-07-05 09:38:20 UTC 
Reproduced with qemux86 as well. This would appear to be a core issue.
Comment 2 Paul Eggleton 2013-07-05 10:30:27 UTC 
Just checked, this worked in dylan, so it's a regression in master. I suspect that it has something to do with the recent bitbake refactoring since bitbake is setting a value for BUILDNAME if none is set in the metadata. Interestingly, buildhistory includes ${BUILDNAME} in its commit messages and seems to be getting the value correctly, so I wonder if it's just an issue of the scoping of the datastore in which the value is being set.
Comment 3 Saul Wold 2013-07-09 21:00:58 UTC 
Another data point, I added a bb.note in the anonymous python of rootfs_rpm to 
bring BUILDNAME and it is showing blank, but as you note it does work 
buildhistory, I am not sure how the scoping works.
Comment 4 Richard Purdie 2013-08-22 15:47:21 UTC 
Quick Observation: This Looks like a problem with the bitbake-worker split. We need to ensure that BUILDNAME gets set in the worker contexts. In _start_worker() in runqueue.py we probably need to transfer that variable
Comment 5 Paul Eggleton 2013-09-05 09:03:35 UTC 
Fix merged to master:

http://git.yoctoproject.org/cgit/cgit.cgi/poky/commit/?id=e63c5775732055a32b417c01fb68711f33ff4ac3

Note in case it is unclear, this was a regression in master; released versions are not affected.
Comment 6 Ionut Chisanovici 2013-10-01 07:01:30 UTC 
tested with NUC.