GNU bug report logs - #51688
ELC+ELN international/emoji.elc crashes

Previous Next

Full log

Message #134 received at 51688 <at> debbugs.gnu.org (full text, mbox):

From: Martin Liška <mliska <at> suse.cz>
To: Andrea Corallo <akrl <at> sdf.org>, Eli Zaretskii <eliz <at> gnu.org>
Cc: rpluim <at> gmail.com, 51688 <at> debbugs.gnu.org,
 Andreas Schwab <schwab <at> linux-m68k.org>
Subject: Re: bug#51688: ELC+ELN international/emoji.elc crashes
Date: Wed, 24 Nov 2021 16:23:53 +0100

On 11/24/21 16:14, Andrea Corallo wrote:
> Eli Zaretskii <eliz <at> gnu.org> writes:
> 
>>> From: Andreas Schwab <schwab <at> linux-m68k.org>
>>> Cc: rpluim <at> gmail.com,  akrl <at> sdf.org,  51688 <at> debbugs.gnu.org
>>> Date: Wed, 24 Nov 2021 14:06:04 +0100
>>>
>>> On Nov 24 2021, Eli Zaretskii wrote:
>>>
>>>> Do you happen to know which libgccjit versions are affected by the
>>>> bug?  We could tell people to avoid them.
>>>
>>> Most likely all of them.
>>
>> Too bad.
>>
>> Andrea, can anything be done to work around this somehow (except to
>> upgrade to a later GCC)?  Or is there nothing we can do except keep
>> fingers crossed?
> 
> I don't think there's much we can do to work around this on Emacs side
> (Martin please correct me if I'm wrong).

No, it's UBSAN in libgccjit library. Unfortunately, the issue is there
for quite a long time and it was exposed by a bad luck where an used
memory was reused (and not initialized) in libgccjit.

Martin

> 
>>>> Or did this bug exist since day one?  (But if it's an old bug, why
>>>> doesn't everyone see these crashes?)
>>>
>>> That's the effect of undefined behaviour.
>>
>> You mean, undefined behavior in libgccjit's own code, right?  Or in
>> our code?
> 
> In libgccjit's own code.
> 
> BR
> 
>    Andrea
> 

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.