Bug 156342 - eu-strip mangles separate debuginfo with relocation sections
Summary: eu-strip mangles separate debuginfo with relocation sections
Keywords:
Status: CLOSED ERRATA
Alias: None
Product: Red Hat Enterprise Linux 4
Classification: Red Hat
Component: elfutils
Version: 4.0
Hardware: All
OS: Linux
medium
medium
Target Milestone: ---
: ---
Assignee: Roland McGrath
QA Contact:
URL:
Whiteboard:
Depends On: 156341
Blocks: 156343 181409 189114
TreeView+ depends on / blocked
 
Reported: 2005-04-29 03:23 UTC by Roland McGrath
Modified: 2007-11-30 22:07 UTC (History)
2 users (show)

Fixed In Version: RHSA-2006-0354
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed: 2006-08-10 18:28:22 UTC
Target Upstream Version:
Embargoed:


Attachments (Terms of Use)


Links
System ID Private Priority Status Summary Last Updated
Red Hat Product Errata RHSA-2006:0354 0 normal SHIPPED_LIVE Low: elfutils security update 2006-08-10 04:00:00 UTC

Description Roland McGrath 2005-04-29 03:23:43 UTC
+++ This bug was initially created as a clone of Bug #156341 +++

Description of problem:

When eu-strip writes separate debuginfo from an ET_REL object such as a kernel
module (.ko file), it doesn't fix up the relocs referring to section symbols
removed from the symbol table.

Version-Release number of selected component (if applicable):
0.106-3

How reproducible:
100%

Steps to Reproduce:
1. Install kernel-debuginfo package.
2. Try to use foo.ko.debug.
3. See bad relocs.
  
Actual results:


Expected results:


Additional info:

This is a blocker for SystemTAP.

Comment 1 Roland McGrath 2005-05-09 08:23:20 UTC
This is fixed upstream/for Fedora (Bug #156341).
Needs backport fix or perhaps just upgrade of elfutils.

Comment 2 Roland McGrath 2005-05-26 01:24:52 UTC
The RHEL4 eu-strip actually just uses the binutils objcopy/strip.
The binutils strip-to-debug works differently and the debuginfo produced should
be fine for systemtap.

Comment 3 Jay Turner 2005-06-21 18:32:13 UTC
Removing from proposed list.

Comment 4 Roland McGrath 2005-08-02 01:40:29 UTC
Turns out I was wrong.  I confused myself trying to make sense of the insane
elfutils multi-stream errata situation in RHEL.  In fact, the elfutils being
used for RHEL4 builds is the real non-stub one, and eu-strip is the real one
with this bug rather than the binutils wrapper.

Comment 10 Roland McGrath 2005-10-10 21:03:56 UTC
This is an unrelated bug.  This error message is never the failure mode of the
eu-strip issue.

Comment 11 Roland McGrath 2005-10-10 21:12:42 UTC
As long as the buildroots don't change, this is not an issue for RHEL4.


Comment 17 Roland McGrath 2006-04-21 00:11:26 UTC
0.97.1-3 should fix this.

Comment 20 Red Hat Bugzilla 2006-08-10 18:28:22 UTC
An advisory has been issued which should help the problem
described in this bug report. This report is therefore being
closed with a resolution of ERRATA. For more information
on the solution and/or where to find the updated files,
please follow the link below. You may reopen this bug report
if the solution does not work for you.

http://rhn.redhat.com/errata/RHSA-2006-0354.html



Note You need to log in before you can comment on or make changes to this bug.