GNU bug report logs - #65843
28.2; Too many iconified frames in .emacs.desktop -> crash (macOS)

Previous Next

Package: emacs;

Reported by: tanzer <at> gg32.com

Date: Sat, 9 Sep 2023 16:37:02 UTC

Severity: normal

Found in version 28.2

Done: Alan Third <alan <at> idiocy.org>

Bug is archived. No further changes may be made.

Full log


View this message in rfc822 format

From: Gerd Möllmann <gerd.moellmann <at> gmail.com>
To: Alan Third <alan <at> idiocy.org>
Cc: 65843 <at> debbugs.gnu.org, tanzer <at> gg32.com
Subject: bug#65843: 28.2; Too many iconified frames in .emacs.desktop -> crash (macOS)
Date: Sun, 10 Sep 2023 06:20:05 +0200
Alan Third <alan <at> idiocy.org> writes:

> On Sat, Sep 09, 2023 at 04:35:56PM -0000, Christian Tanzer via Bug reports for GNU Emacs, the Swiss army knife of text editors wrote:
>> 
>> =============================================================================
>> In Emacs 28 and Emacs 29, desktop-load crashes when there are too many
>> iconified frames in the .emacs.desktop file.
>> 
>> The two attached .emacs.desktop files show the smallest test case that
>> I could come up with:
>> 
>> - .emacs.desktop.works with 3 non-iconified and 4 iconified frames
>>   loads without problem in Emacs 28
>> 
>>   + desktop-load of this file crashes in Emacs 29.1
>> 
>>   + a .emacs.desktop with 3 non-iconified and 3 iconified frames works
>>     in Emacs 29.1
>> 
>> - .emacs.desktop.crashes with 3 non-iconified and 5 iconified frames
>>   leads to a crash in Emacs 28 (see attached problem report generated
>>   by macOS)
>> 
>> - I tried .emacs.desktop with up to 17 non-iconified frames in Emacs
>>   28 without problems
>> 
>> - Up to Emacs 27, I've never seen a problem with desktop-load crashing
>> =============================================================================
>
> Can you please try the master branch with the two attached patches
> applied? The v2 patch should be applied first.
>
> The apple crash report is enough to see that the crash is happening
> somewhere deep inside the AppKit code on frame creation. No idea why,
> but maybe we'll be lucky and the fix will be the same as the last
> frame-creation crash fix...

Alas, I cannot reproduce the abort here, with an unpatched master or
emacs-29, and neither on Ventura/arm64 nor Big Sur/x86_64 :-(.

Christian, can you reproduce this with emacs -Q?  Or does it perhaps
depend on your init.el?  (This was the case for me in the child frane
case that Alan fixed.)




This bug report was last modified 1 year and 248 days ago.

Previous Next


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