[patch] Fix run-unstrip-n.sh regression on CentOS-5

Jan Kratochvil jan.kratochvil at redhat.com
Tue Jul 23 08:08:04 UTC 2013


On Tue, 23 Jul 2013 00:13:07 +0200, Mark Wielaard wrote:
> Looks OK to me, but I haven't tried it.

Thanks for the review!  Therefore going to check it in.


> On Mon, Jul 22, 2013 at 09:21:55AM +0200, Jan Kratochvil wrote:
> > tests/
> > 2013-07-22  Jan Kratochvil  <jan.kratochvil at redhat.com>
> > 
> > 	* run-unstrip-n.sh (test-core.*): Ignore libc.so.6 entry and order of
> > 	the entries.
> 
> I understand why you don't want to rely on the order of the entries.

BTW there could be maybe some sort-by-address postprocessing at the end of
dwfl_core_file_report in the future.


> But why ignore the libc entry?

Because libc entry in the core file has no build-id and when also your on-disk
libc entry has no build-id it matches.  But in such case the mapped length is
"random" - depending on your on-disk file.

Other possibility would be just to remove the mapped file length for
comparison.


Thanks,
Jan


More information about the elfutils-devel mailing list