public inbox for annobin@sourceware.org
 help / color / mirror / Atom feed
From: <grant@hcubed.com>
To: "'Nick Clifton'" <nickc@redhat.com>
Cc: <annobin@sourceware.org>,
	"'Tulio Magno Quites Machado Filho'" <tuliom@redhat.com>
Subject: RE: global-file-syms for clang
Date: Mon, 25 Mar 2024 10:38:52 -0700	[thread overview]
Message-ID: <008301da7edb$4db9c9d0$e92d5d70$@hcubed.com> (raw)
In-Reply-To: <fd527e32-fb39-4a70-8258-e4fa5d8961ca@redhat.com>

Nick,

Thanks! Couple of questions:
1) Will the new annobin plugin/annocheck work with clang version16.0.6-1.el9)? Specifically, will the interaction between annocheck & the compile cause any issues/false positives or negatives?
2) I noticed that annocheck has a considerable number of errors around gaps. This is after regenerating all the source (except standard libraries) with annobin plugins. Is this expected. For reference, LTO is enabled in these cases.

Thanks again,
Grant Haidinyak

-----Original Message-----
From: Nick Clifton <nickc@redhat.com> 
Sent: Monday, March 25, 2024 10:19 AM
To: grant@hcubed.com
Cc: annobin@sourceware.org; Tulio Magno Quites Machado Filho <tuliom@redhat.com>
Subject: Re: global-file-syms for clang

Hi Grant,

> Thanks. As a data point, when I use the GCC toolchain, I didn't need 
> to set the global-file-syms flag. I believe the issue is an 
> interaction between the LLVM gold plugin and the annobin plugin and linking in LTO archives that have objects that have the same filename for the source. The LLVM gold plugin tries to bring in all the fine annobin symbols, rather than just bringing the annobin symbols from the first file that satisfies the symbols for linking.

Hmm, strange.  It sounds like an issue with the LLVM plugin, but I doubt if it is likely to be fixed...

> I was just going to use the global-file-syms to overcome the LLVM gold plugin/LTO issue.

Fair enough.  Please try the 12.46 sources, available from the annobin repository.
This has Tulio's patch to implement the global-file-syms option for the Clang and LLVM plugins.  (For the LLVM plugin you have to pass the options in the ANNOBIN environment variable, since LLVM does not support passing options to plugins).

Cheers
   Nick


  reply	other threads:[~2024-03-25 17:38 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-03-20 17:28 grant
2024-03-21  8:35 ` Nick Clifton
2024-03-21 15:32   ` grant
2024-03-21 20:47     ` Tulio Magno Quites Machado Filho
2024-03-22 20:17     ` [PATCH] Add support for global-file-syms to the clang and llvm plugins Tulio Magno Quites Machado Filho
2024-03-25 16:00       ` Nick Clifton
2024-03-25 17:19     ` global-file-syms for clang Nick Clifton
2024-03-25 17:38       ` grant [this message]
2024-03-26 11:31         ` Nick Clifton

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='008301da7edb$4db9c9d0$e92d5d70$@hcubed.com' \
    --to=grant@hcubed.com \
    --cc=annobin@sourceware.org \
    --cc=nickc@redhat.com \
    --cc=tuliom@redhat.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).