public inbox for binutils@sourceware.org
 help / color / mirror / Atom feed
From: Nick Clifton <nickc@redhat.com>
To: Deepesh <deepeshvrajan0774@gmail.com>
Cc: Binutils <binutils@sourceware.org>
Subject: Re: Bug 29925 - Unable to reproduce
Date: Wed, 5 Jun 2024 11:12:56 +0100	[thread overview]
Message-ID: <110f5b21-9d97-468a-9cc8-6b1c76c4e05f@redhat.com> (raw)
In-Reply-To: <FEF7C7E4-7599-4E19-996C-2B8C3846D9E8@hxcore.ol>

Hi Deepesh,

> _but while doing make getting the below errors :_
> 
> /==489551==WARNING: invalid path to external symbolizer!/
> /==489551==WARNING: Failed to use and restart external symbolizer!/

Hmm, this is obviously some kind of issue with the sanitizer.  It may
be related to the fact that the test you are running is using the nm
program to examine a fuzzed input file.


> We’ve not seen the above errors by using the binutils tar file.
> And, the same errors were seen with latest binutils upstream sources.
> 
> 1. Why we are seeing these errors only with upstream git sources.

I *suspect* that it is because the release tarballs disable experimental
and development features.  (See the file bfd/development.sh - it has
different contents for release tarballs vs git sources).  In particular
releases do not define the ENABLE_CHECKING preprocessor define, which
in turn affects how some of the functions defined in bfd/elf-bfd.h work.

> 2. Is this issue applicable for older versions on binutils (2.36)?

Probably.  The code changed by the patch to fix the bug is also present
in the 2.35 sources.  I have not actually checked by building a set of
the binutils using these sources however.


> 3. Is this issue with any specific version of clang (we are using 14.0).

I could not say.  But you might try using a different version of clang,
or gcc instead of clang, and see if anything changes.


> 4. The bug Target Milestone 2.40, says fixed in 2.40 but we could still reproduce the issue on 2.40 

Are you sure ?  I have just checked and my 2.40 build does not show the
problem.  (I checked compiling both with Clang and with gcc).

I checked using both the official 2.40 release tarball and the binutils-2_40-branch
sources from the git repository.  Both work.

I also confirmed that building with the 2.39 sources does create a version
of nm that has the problem.

Cheers
   Nick


  parent reply	other threads:[~2024-06-05 10:13 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <FEF7C7E4-7599-4E19-996C-2B8C3846D9E8@hxcore.ol>
2024-06-05  4:50 ` FW: " Deepesh
2024-06-05 10:12 ` Nick Clifton [this message]
2024-05-29 13:25 Deepesh Rajan

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=110f5b21-9d97-468a-9cc8-6b1c76c4e05f@redhat.com \
    --to=nickc@redhat.com \
    --cc=binutils@sourceware.org \
    --cc=deepeshvrajan0774@gmail.com \
    /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;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).