Openembedded Core Discussions
 help / color / mirror / Atom feed
From: Alexander Kanavin <alexander.kanavin@linux.intel.com>
To: Mark Hatle <mark.hatle@windriver.com>,
	Mark Wielaard <mark@klomp.org>, Neal Gompa <ngompa13@gmail.com>
Cc: OpenEmbedded Core Mailing List
	<openembedded-core@lists.openembedded.org>
Subject: Re: [PATCH] rpm: update to 4.14.0
Date: Thu, 9 Nov 2017 15:19:32 +0200	[thread overview]
Message-ID: <bec31b67-6192-4a9d-7d8b-eaaeb3f02051@linux.intel.com> (raw)
In-Reply-To: <26ddd9fc-7d5d-0dd7-fc48-c7e3ca5d4128@windriver.com>

On 11/08/2017 05:31 PM, Mark Hatle wrote:

>> So a couple of questions.
>>
>> - What exactly breaks currently when you let debugedit rewrite the
>>    debuginfo source paths?
> 
> Alexander has more info then I do here, but I believe running it against grub
> modules breaks the modules.

Here's how you can reproduce the breakage:


1) download 
https://drive.google.com/open?id=1yKav0llpfrLjuMZOCj6ZyfVWGbxZQzIA

(a binary module from a grub build)

2) run
objcopy --only-keep-debug ./ntldr.module ntldr.module.debug

to establish that objcopy is happy with the download

3) run

debugedit -l debug.source ./ntldr.module


4) run the same objcopy command again ---> kaboom

objcopy: ./ntldr.module: invalid relocation type 112
objcopy: ./ntldr.module(rel.debug_info): relocation 556 has invalid 
symbol index 7497472
objcopy: ./ntldr.module: invalid relocation type 98
objcopy: ./ntldr.module(rel.debug_info): relocation 557 has invalid 
symbol index 6648415
objcopy: ./ntldr.module: invalid relocation type 108
objcopy: ./ntldr.module(rel.debug_info): relocation 558 has invalid 
symbol index 7495522
objcopy: ./ntldr.module: invalid relocation type 117
objcopy: ./ntldr.module(rel.debug_info): relocation 559 has invalid 
symbol index 7499636
objcopy: ./ntldr.module: invalid relocation type 97
objcopy: ./ntldr.module(rel.debug_info): relocation 560 has invalid 
symbol index 25697
objcopy: ./ntldr.module: invalid relocation type 111
.....
Segmentation fault (core dumped)


Like Mark said, this isn't really the main issue for us; we would want 
to not have the binary file modified at all. Currently debugedit doesn't 
seem to provide that (and that's totally fine - a replacement is not 
difficult to find).

Alex


  reply	other threads:[~2017-11-09 13:19 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-10-18 12:14 [PATCH] rpm: update to 4.14.0 Alexander Kanavin
2017-11-03 10:14 ` Alexander Kanavin
2017-11-03 16:21   ` Khem Raj
2017-11-03 17:58   ` Mark Hatle
2017-11-06 16:00     ` Alexander Kanavin
2017-11-06 16:54       ` Mark Hatle
2017-11-07 14:53         ` Alexander Kanavin
2017-11-07 15:50           ` Otavio Salvador
2017-11-07 16:41           ` Mark Hatle
2017-11-07 21:17       ` Neal Gompa
2017-11-08 13:14         ` Alexander Kanavin
     [not found]           ` <1510153768.10879.80.camel@klomp.org>
2017-11-08 15:31             ` Mark Hatle
2017-11-09 13:19               ` Alexander Kanavin [this message]
2017-11-09 16:49               ` Alexander Kanavin
  -- strict thread matches above, loose matches on Subject: below --
2017-12-18 16:00 Alexander Kanavin

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=bec31b67-6192-4a9d-7d8b-eaaeb3f02051@linux.intel.com \
    --to=alexander.kanavin@linux.intel.com \
    --cc=mark.hatle@windriver.com \
    --cc=mark@klomp.org \
    --cc=ngompa13@gmail.com \
    --cc=openembedded-core@lists.openembedded.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox