public inbox for gdb-patches@sourceware.org
 help / color / mirror / Atom feed
From: Tom de Vries <tdevries@suse.de>
To: Aditya Kamath1 <Aditya.Kamath1@ibm.com>,
	Ulrich Weigand <Ulrich.Weigand@de.ibm.com>,
	Aditya Kamath1 via Gdb-patches <gdb-patches@sourceware.org>
Cc: Sangamesh Mallayya <sangamesh.swamy@in.ibm.com>,
	Tom Tromey <tom@tromey.com>
Subject: Re: [PATCH] Fix AIX build break
Date: Fri, 12 Jan 2024 10:50:04 +0100	[thread overview]
Message-ID: <da839d76-c075-48aa-8841-df6f921e687e@suse.de> (raw)
In-Reply-To: <CH2PR15MB3544BC85F0820D6A2ACB7E91D66F2@CH2PR15MB3544.namprd15.prod.outlook.com>

On 1/12/24 10:20, Aditya Kamath1 wrote:
> Respected community members,
> 
> Hi,
> 
> Our CI in AIX for GDB has been broken for the last two days with the 
> following error.
> 
> CXXLD  gdb
> 
> ld: 0711-317 ERROR: Undefined symbol: _ZTH23deprecated_warning_hook
> 
> ld: 0711-317 ERROR: Undefined symbol: ._ZTH23deprecated_warning_hook
> 
> ld: 0711-345 Use the -bloadmap or -bnoquiet option to obtain more 
> information.
> 
> collect2: error: ld returned 8 exit status
> 
> gmake: *** [Makefile:2184: gdb] Error 1
> 
> These symbols came due to the commit 
> https://sourceware.org/git/?p=binutils-gdb.git;a=commitdiff;h=54b815ddb428944a70694e3767a0fadbdd9ca9ea <https://sourceware.org/git/?p=binutils-gdb.git;a=commitdiff;h=54b815ddb428944a70694e3767a0fadbdd9ca9ea>
> 

I think this is triggered by using thread-local with extern.

I found this commit:
...
commit fece451c2aca57b095e7e4063e342781cf74aa75
Author: Christian Biesinger <cbiesinger@google.com>
Date:   Tue Mar 9 08:16:23 2021 -0600

     Use RAII to set the per-thread SIGSEGV handler

     This avoids using a thread-local extern variable, which causes link 
errors
     on some platforms, notably Cygwin.  But I think this is a better 
pattern
     even outside of working around linker bugs because it encapsulates 
direct
     access to the variable inside the class, instead of having a global 
extern
     variable.
...
which seems related, given that it mentions link errors.  Perhaps we can 
do something similar here.

Thanks,
- Tom

> The problem in AIX is marking a variable for thread storage as 
> thread_local that makes that symbol a weak symbol. The dump output of 
> the complaints.o shows this as pasted here:-
> 
> # dump -tov -X64 complaints.o | grep _ZTH23deprecated_warning_hook
> 
> [32]    m   0x00000000     undef     1    weak                    
> _ZTH23deprecated_warning_hook
> 
> [36]    m   0x00000000     undef     1    weak                    
> ._ZTH23deprecated_warning_hook
> 
> [3209]  m   0x000203d8     .data     1  unamex                    
> _ZTH23deprecated_warning_hook
> 
> Hence in the final stage of compilation while creating the binary gdb, 
> the ld error came and the compilation is unsuccessful since the symbol 
> is not visible though complaints.c file was successfully compiled.
> 
> This patch is a fix for the same where instead of thread_local we used 
> __thread and GDB now compiles in AIX. Please find attached the patch. 
> (See: 0001-Fix-AIX-build-break.patch).
> 
> I want to know your opinion about this. Also, I do not know the impact 
> of this change on other targets. If we can do this better, kindly let me 
> know.
> 
> Have a nice day ahead.
> 
> Thanks and regards,
> Aditya.
> 


  reply	other threads:[~2024-01-12  9:49 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-01-12  9:20 Aditya Kamath1
2024-01-12  9:50 ` Tom de Vries [this message]
2024-01-12 15:56   ` Tom Tromey
2024-01-15  9:57     ` [RFC] " Aditya Kamath1
2024-01-16  9:42     ` [PATCH] " Aditya Kamath1
2024-01-17 15:09       ` Tom Tromey
  -- strict thread matches above, loose matches on Subject: below --
2024-06-09  5:38 Aditya Vidyadhar Kamath
2024-06-09 14:12 ` Simon Marchi
2024-01-17 18:28 Aditya Vidyadhar Kamath
2024-01-17 19:25 ` Tom Tromey
2024-01-19 11:58   ` Aditya Kamath1
2024-01-19 16:25     ` Tom Tromey
2024-01-20 15:03       ` Aditya Kamath1
2024-01-24  0:53         ` Aditya Kamath1
2024-01-25 14:45           ` Aditya Kamath1
2024-01-25 19:53             ` Tom Tromey
2024-01-26  8:37               ` Aditya Kamath1
2024-01-31  0:40                 ` Tom Tromey
2024-01-31  9:15                   ` Aditya Kamath1
2024-02-01  0:37                     ` Tom Tromey
2024-02-01  3:03                       ` Aditya Kamath1
2024-02-06 12:36                         ` Ciaran Woodward
2024-02-06 14:49                           ` Tom Tromey
2024-02-06 15:04                             ` Ciaran Woodward
2024-02-06 15:54                           ` Aditya Kamath1
2023-08-11  8:49 Aditya Kamath1
2023-08-11  9:43 ` Ulrich Weigand
2023-08-11  9:53   ` Aditya Kamath1

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=da839d76-c075-48aa-8841-df6f921e687e@suse.de \
    --to=tdevries@suse.de \
    --cc=Aditya.Kamath1@ibm.com \
    --cc=Ulrich.Weigand@de.ibm.com \
    --cc=gdb-patches@sourceware.org \
    --cc=sangamesh.swamy@in.ibm.com \
    --cc=tom@tromey.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).