GNU bug report logs - #40139
25.2; SIGBUS

Previous Next

Full log

Message #11 received at 40139 <at> debbugs.gnu.org (full text, mbox):

From: Vadim Zaliva <lord <at> crocodile.org>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: 40139 <at> debbugs.gnu.org
Subject: Re: bug#40139: 25.2; SIGBUS
Date: Wed, 10 Jun 2020 13:50:33 -0700

[Message part 1 (text/plain, inline)]

I finally switched to 26.3 and the problem persists.
For example, it renders 'latex-preview-pane-mode` unusable.
Perhaps a more robust solution would be to spawn ImageMagic as
sub-process instead of using it as a library?

Vadim


On Fri, Mar 20, 2020 at 12:30 AM Eli Zaretskii <eliz <at> gnu.org> wrote:

> > From: Vadim Zaliva <lord <at> crocodile.org>
> > Date: Thu, 19 Mar 2020 15:36:01 -0700
> >
> > I am using 'latex-preview-pane-mode'. It renders PDF preview of the
> > document. It works first time. When I do any change to the document and
> > try to save it, it crashes. On console I see the following:
> > [...]
> > Thread 1 "emacs" received signal SIGBUS, Bus error.
> > __memmove_avx_unaligned_erms ()
> >     at ../sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S:306
> > 306 ../sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S: No such
> file or directory.
> > (gdb) bt full
> > #0  0x00007ffff0defb64 in __memmove_avx_unaligned_erms ()
> >     at ../sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S:306
> > #1  0x00007ffff36028ac in ReadBlob () at
> /usr/lib/x86_64-linux-gnu/libMagickCore-6.Q16.so.3
> > #2  0x00007fffe2b6843e in  ()
> >     at
> /usr/lib/x86_64-linux-gnu/ImageMagick-6.9.7/modules-Q16/coders/png.so
> > #3  0x00007ffff76dbb6f in  () at /usr/lib/x86_64-linux-gnu/libpng16.so.16
> > #4  0x00007ffff76dbc5f in  () at /usr/lib/x86_64-linux-gnu/libpng16.so.16
> > #5  0x00007ffff76e0322 in  () at /usr/lib/x86_64-linux-gnu/libpng16.so.16
> > #6  0x00007ffff76d2ae3 in png_read_row () at
> /usr/lib/x86_64-linux-gnu/libpng16.so.16
> > #7  0x00007fffe2b7a887 in  ()
> >     at
> /usr/lib/x86_64-linux-gnu/ImageMagick-6.9.7/modules-Q16/coders/png.so
> > #8  0x00007fffe2b7b4d6 in  ()
> >     at
> /usr/lib/x86_64-linux-gnu/ImageMagick-6.9.7/modules-Q16/coders/png.so
> > #9  0x00007ffff363a9df in ReadImage () at
> /usr/lib/x86_64-linux-gnu/libMagickCore-6.Q16.so.3
> > #10 0x00007ffff3b0f46a in MagickReadImage ()
> >     at /usr/lib/x86_64-linux-gnu/libMagickWand-6.Q16.so.3
> > [...]
> > It looks ImageMagick problem.
>
> Indeed.
>
> > Howevr if it crashes or otherwise
> > misbehaves it should not bring down my whole emacs process.
>
> A fatal signal in any C library used by Emacs will necessarily bring
> the whole Emacs process down.  Emacs tries to catch these signals and
> perform auto-saving, but after doing so it has no choice but shut
> down, because there's no way of continuing from such fatal signals on
> modern operating systems.
>
> ImageMagick has many stability problems, which is why Emacs is moving
> away of using it, by implementing some of its capabilities natively,
> and by replacing its other capabilities with other external libraries.
> My suggestion is to try the latest Emacs (version 26.3 is the latest
> official release; version 27.1 is in pretest and can be downloaded
> from alpha.gnu.org), and upgrade your ImageMagick to a more stable
> version if that is possible.
>
> Other than that, I don't see how this problem can be solved in Emacs.
>
> Thanks.
>


-- 
"Hated by fools, and fools to hate, be that my motto and my fate"
(Jonathan Swift)

[Message part 2 (text/html, inline)]

Previous Next

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