bug-gnulib@gnu.org mirror (unofficial)
 help / color / mirror / Atom feed
From: Bruno Haible <bruno@clisp.org>
To: bug-gnulib@gnu.org
Cc: Florian Weimer <fweimer@redhat.com>,
	libc-alpha@sourceware.org, Andreas Schwab <schwab@linux-m68k.org>,
	binutils@sourceware.org
Subject: Re: Undefined use of weak symbols in gnulib
Date: Wed, 28 Apr 2021 02:09:19 +0200	[thread overview]
Message-ID: <1680226.UWtE2gOZdF@omega> (raw)
In-Reply-To: <87eeewnfzw.fsf@oldenburg.str.redhat.com>

Hi Florian,

> Here's a fairly representative test case, I think.
> 
> #include <pthread.h>
> #include <stdio.h>
> 
> extern __typeof (pthread_key_create) __pthread_key_create __attribute__ ((weak));
> extern __typeof (pthread_once) pthread_once __attribute__ ((weak));
> 
> void
> f1 (void)
> {
>   puts ("f1 called");
> }
> 
> pthread_once_t once_var;
> 
> void __attribute__ ((weak))
> f2 (void)
> {
>   if (__pthread_key_create != NULL)
>     pthread_once (&once_var, f1);
> }
> 
> int
> main (void)
> {
>   f2 ();
> }
> 
> Building it with “gcc -O2 -fpie -pie” and linking with binutils 2.30
> does not result in a crash with LD_PRELOAD=libpthread.so.0.

Thank you for the test case. It helps the understanding.

But I don't understand
  - why anyone would redeclare 'pthread_once', when it's a standard POSIX
    function,
  - why f2 is declared weak,
  - why the program skips its initializations in single-threaded mode,
  - why libpthread would be loaded through LD_PRELOAD or dlopen, given
    that the long-term statement has been that declaring a symbol weak
    has no effect on the dynamic linker [1][2][3][4]?

How about the following test case instead?

=====================================================================
#include <pthread.h>
#include <stdio.h>

#pragma weak pthread_key_create
#pragma weak pthread_once

void
do_init (void)
{
  puts ("initialization code");
}

pthread_once_t once_var;

void
init (void)
{
  if (pthread_key_create != NULL)
    {
      puts ("multi-threaded initialization");
      pthread_once (&once_var, do_init);
    }
  else
    do_init ();
}

int
main (void)
{
  init ();
}
=====================================================================

$ gcc -Wall -fpie -pie foo.c ; ./a.out 
initialization code

$ gcc -Wall -fpie -pie foo.c -Wl,--no-as-needed -lpthread ; ./a.out
multi-threaded initialization
initialization code

What will change for this program with glibc 2.34?

Bruno

[1] https://sourceware.org/legacy-ml/libc-hacker/2000-06/msg00029.html
[2] https://www.akkadia.org/drepper/dsohowto.pdf page 6
[3] https://stackoverflow.com/questions/21092601/is-pthread-in-glibc-so-implemented-by-weak-symbol-to-provide-pthread-stub-functi/21103255
[4] https://stackoverflow.com/questions/20658809/dynamic-loading-and-weak-symbol-resolution



  reply	other threads:[~2021-04-28  0:15 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-04-27  5:53 Undefined use of weak symbols in gnulib Florian Weimer
2021-04-27  6:50 ` Paul Eggert
2021-04-27  6:58   ` Florian Weimer
2021-04-27  7:13     ` Paul Eggert
2021-04-27  7:24 ` Andreas Schwab
2021-04-27 11:06   ` Florian Weimer
2021-04-28  0:09     ` Bruno Haible [this message]
2021-04-28  2:10       ` H.J. Lu
2021-04-28  2:13         ` H.J. Lu
2021-05-05 20:31           ` Fangrui Song
2021-04-28  8:35         ` Florian Weimer
2021-04-28 13:15           ` Michael Matz
2021-04-28  7:44       ` Florian Weimer
2021-04-28 14:48         ` Bruno Haible
2021-04-28 17:44           ` Florian Weimer
2021-07-17 14:38         ` Bruno Haible
2021-07-17 14:55           ` Florian Weimer
2021-07-17 16:39             ` Bruno Haible
2021-07-27 20:02           ` Joseph Myers
2021-07-27 20:19             ` Florian Weimer
2021-07-27 23:38               ` Paul Eggert
2021-07-17 16:21         ` Bruno Haible
2021-04-27 23:22   ` Bruno Haible
2021-04-27 23:47 ` Bruno Haible
2021-04-28  7:57   ` Florian Weimer
2021-04-28 14:40     ` Bruno Haible
2021-04-28 17:43       ` Florian Weimer
2021-04-29 15:15         ` Bruno Haible
2021-04-30  9:55           ` Florian Weimer
2021-04-29  6:33       ` Ben Pfaff
2021-05-03  1:44 ` Alan Modra
2021-07-12 10:04 ` Michael Hudson-Doyle
2021-07-12 15:03   ` Florian Weimer
2021-07-12 15:30     ` Matthias Klose
2021-07-12 15:37       ` Florian Weimer
2021-07-13  0:22         ` Michael Hudson-Doyle

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

  List information: https://lists.gnu.org/mailman/listinfo/bug-gnulib

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

  git send-email \
    --in-reply-to=1680226.UWtE2gOZdF@omega \
    --to=bruno@clisp.org \
    --cc=binutils@sourceware.org \
    --cc=bug-gnulib@gnu.org \
    --cc=fweimer@redhat.com \
    --cc=libc-alpha@sourceware.org \
    --cc=schwab@linux-m68k.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;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).