public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Jason Merrill <jason@redhat.com>
To: Thomas Neumann <neumann@in.tum.de>, gcc-patches@gcc.gnu.org
Cc: Stephan Bergmann <sbergman@redhat.com>
Subject: Re: [PATCH] Avoid depending on destructor order
Date: Thu, 22 Sep 2022 18:22:38 -0400	[thread overview]
Message-ID: <b63484f7-df05-1ac2-2834-4b49f4b0b832@redhat.com> (raw)
In-Reply-To: <5aba5553-ed71-50a8-e934-6e94f3a1057e@in.tum.de>

On 9/19/22 12:20, Thomas Neumann wrote:
> In some scenarios (e.g., when mixing gcc and clang code), it can
> happen that frames are deregistered after the lookup structure
> has already been destroyed. That in itself would be fine, but
> it triggers an assert in __deregister_frame_info_bases that
> expects to find the frame.
> 
> To avoid that, we now remember that the btree as already been
> destroyed and disable the assert in that case.

OK.

> libgcc/ChangeLog:
> 
>      * unwind-dw2-fde.c: (release_register_frames) Remember
>      when the btree has been destroyed.
>      (__deregister_frame_info_bases) Disable the assert when
>      shutting down.
> ---
>   libgcc/unwind-dw2-fde.c | 4 +++-
>   1 file changed, 3 insertions(+), 1 deletion(-)
> 
> diff --git a/libgcc/unwind-dw2-fde.c b/libgcc/unwind-dw2-fde.c
> index 919abfe0664..d237179f4ea 100644
> --- a/libgcc/unwind-dw2-fde.c
> +++ b/libgcc/unwind-dw2-fde.c
> @@ -48,6 +48,7 @@ typedef __UINTPTR_TYPE__ uintptr_type;
>   #include "unwind-dw2-btree.h"
> 
>   static struct btree registered_frames;
> +static bool in_shutdown;
> 
>   static void
>   release_registered_frames (void) __attribute__ ((destructor (110)));
> @@ -57,6 +58,7 @@ release_registered_frames (void)
>     /* Release the b-tree and all frames. Frame releases that happen 
> later are
>      * silently ignored */
>     btree_destroy (&registered_frames);
> +  in_shutdown = true;
>   }
> 
>   static void
> @@ -282,7 +284,7 @@ __deregister_frame_info_bases (const void *begin)
>     __gthread_mutex_unlock (&object_mutex);
>   #endif
> 
> -  gcc_assert (ob);
> +  gcc_assert (in_shutdown || ob);
>     return (void *) ob;
>   }
> 


  reply	other threads:[~2022-09-22 22:22 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-09-19 16:20 Thomas Neumann
2022-09-22 22:22 ` Jason Merrill [this message]
2022-09-26 11:46 ` Claudiu Zissulescu Ianculescu
2022-09-26 11:49   ` Thomas Neumann
2022-09-26 11:50     ` Claudiu Zissulescu Ianculescu
2022-09-26 12:53     ` Iain Sandoe
2022-09-26 13:14       ` Thomas Neumann
2022-09-23 13:25 David Edelsohn
2022-09-23 13:38 ` Thomas Neumann
2022-09-23 14:01   ` David Edelsohn
2022-09-23 14:12     ` Thomas Neumann
2022-09-23 14:30       ` David Edelsohn
2022-09-25  6:29         ` Iain Sandoe
2022-09-25 15:29           ` Jeff Law
2022-09-26  7:55             ` Rainer Orth
2022-09-27  0:17       ` Jason Merrill
2022-09-23 14:11   ` David Edelsohn

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=b63484f7-df05-1ac2-2834-4b49f4b0b832@redhat.com \
    --to=jason@redhat.com \
    --cc=gcc-patches@gcc.gnu.org \
    --cc=neumann@in.tum.de \
    --cc=sbergman@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).