GNU bug report logs - #61960
30.0.50; Unexec build reliably crashes during loadup

Previous Next

Full log

View this message in rfc822 format

From: Eli Zaretskii <eliz <at> gnu.org>
To: Po Lu <luangruo <at> yahoo.com>
Cc: hi-angel <at> yandex.ru, 61960 <at> debbugs.gnu.org, akrl <at> sdf.org
Subject: bug#61960: 30.0.50; Unexec build reliably crashes during loadup
Date: Thu, 16 Mar 2023 08:25:28 +0200

> From: Po Lu <luangruo <at> yahoo.com>
> Cc: akrl <at> sdf.org,  61960 <at> debbugs.gnu.org,  hi-angel <at> yandex.ru
> Date: Thu, 16 Mar 2023 08:30:00 +0800
> 
> Eli Zaretskii <eliz <at> gnu.org> writes:
> 
> > Why astonishing?  AFAIR, it was supposed to work, so unless it
> > bit-rotted, it should still build and work.
> 
> For example: Emacs with pdumper is supposed to be able to write into
> objects in pure space.  But that will crash an unexec build without the
> necessary protections with a write into program text, and AFAIK those
> protections are disabled if HAVE_PDUMPER.

They are disabled, but no sane Lisp should write into pure space,
ever.

And it isn't really correct that they are disabled: the recent
discussions about changing names of primitives clearly shows that at
least some areas in the pure space are write-protected even in pdumper
builds.  At least that's what I see on my system.

Anyway, I think you are mistaken in how you look at this (somewhat
weird) configuration: all it was supposed to allow is decide how to
dump Emacs at run time rather than at configure time.  That's all.

Previous Next

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