[oe] numerous "strip command failed" ERRORs in building beagle image

[Scott Garman]

On 05/14/2010 03:00 AM, Robert P. J. Day wrote:
>
>    obviously not fatal since the build kept chugging along, but there
> were a number of "ERROR"s of the following form:
>
>
> NOTE: Running task 3873 of 11130 (ID: 4231,
> /home/rpjday/oe/openembedded/recipes/xcb/libpthread-stubs_0.2.bb,
> do_patch)
> ERROR: runstrip: ''arm-angstrom-linux-gnueabi-strip'
> --remove-section=.comment --remove-section=.note
> '/home/rpjday/oe/angstrom-dev/work/beagleboard-angstrom-linux-gnueabi/linux-omap-psp-2.6.32-r75+gitr627293ad28604b22612f9a4a318f64cfab241e22/package/kernel/scripts/pnmtologo''
> strip command failed
> ERROR: runstrip: ''arm-angstrom-linux-gnueabi-strip'
> --remove-section=.comment --remove-section=.note
> '/home/rpjday/oe/angstrom-dev/work/beagleboard-angstrom-linux-gnueabi/linux-omap-psp-2.6.32-r75+gitr627293ad28604b22612f9a4a318f64cfab241e22/package/kernel/scripts/bin2c''
> strip command failed
> ... etc etc ...
>
>    normally, i don't care but i typically take anything labelled as an
> "ERROR" seriously so if it's genuinely not an error, perhaps it should
> be relabelled.  (suggested from a position of complete ignorance.)

This happens in the do_package task of many of the kernel builds. I was 
looking into it just the other night as it triggers BuildBot to think 
that the build failed (because it sees a non-zero exit status from this 
section).

Is it even necessary to stage the kernel scripts/ directory? That's the 
final thing happening in the do_install task in kernel.bbclass. Removing 
that line would be a one-liner patch fix to this issue.

Scott

-- 
Scott Garman
sgarman at zenlinux dot com