GNU bug report logs - #24790
Segmentation fault when changing default font on master

Previous Next

To add a comment to this bug, you must first unarchive it, by sending
a message to control AT debbugs.gnu.org, with unarchive 24790 in the body.
You can then email your comments to 24790 AT debbugs.gnu.org in the normal way.

Toggle the display of automated, internal messages from the tracker.

Message #5 received at submit <at> debbugs.gnu.org (full text, mbox):

From: Clément Pit--Claudel <clement.pitclaudel <at> live.com>
To: bug-gnu-emacs <at> gnu.org
Subject: Segmentation fault when changing default font on master
Date: Tue, 25 Oct 2016 10:33:32 -0400

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

Hi all,

When I select the attached font from the "Set default font" menu, nothing happens (the default font does not change).  When I select it again, emacs segfaults (see attached output of bt full).

My Emacs has patch af1a69f4d17a482c359d98c00ef86fac835b5fac reverted.  Could this make a difference?

Let me know what other information I can provide,
Cheers,
Clément.

--------

In GNU Emacs 26.0.50.3 (x86_64-pc-linux-gnu, GTK+ Version 3.18.9)
 of 2016-10-24 built on clem-w50-mint
Repository revision: bd0da7db4e4fdb50a3c17f6aeefe40cd720d2cc4
Windowing system distributor 'The X.Org Foundation', version 11.0.11804000
System Description:	Linux Mint 18 Sarah

Recent messages:
For information about GNU Emacs and the GNU system, type C-h C-a.

Configured using:
 'configure --without-toolkit-scroll-bars'

Configured features:
XPM JPEG TIFF GIF PNG RSVG IMAGEMAGICK SOUND GPM DBUS GCONF GSETTINGS
NOTIFY ACL LIBSELINUX GNUTLS LIBXML2 FREETYPE M17N_FLT LIBOTF XFT ZLIB
GTK3 X11 LIBSYSTEMD

Important settings:
  value of $LC_MONETARY: en_DK.UTF-8
  value of $LC_NUMERIC: en_DK.UTF-8
  value of $LC_TIME: en_DK.UTF-8
  value of $LANG: en_US.UTF-8
  value of $XMODIFIERS: @im=ibus
  locale-coding-system: utf-8-unix

Major mode: Lisp Interaction

Minor modes in effect:
  tooltip-mode: t
  global-eldoc-mode: t
  electric-indent-mode: t
  mouse-wheel-mode: t
  tool-bar-mode: t
  menu-bar-mode: t
  file-name-shadow-mode: t
  global-font-lock-mode: t
  font-lock-mode: t
  blink-cursor-mode: t
  auto-composition-mode: t
  auto-encryption-mode: t
  auto-compression-mode: t
  line-number-mode: t
  transient-mark-mode: t

Load-path shadows:
None found.

Features:
(shadow sort mail-extr emacsbug message subr-x puny seq byte-opt gv
bytecomp byte-compile cl-extra help-mode cconv cl-loaddefs pcase cl-lib
dired dired-loaddefs format-spec rfc822 mml easymenu mml-sec
password-cache epa derived epg epg-config gnus-util rmail rmail-loaddefs
mm-decode mm-bodies mm-encode mail-parse rfc2231 mailabbrev gmm-utils
mailheader sendmail rfc2047 rfc2045 ietf-drums mm-util mail-prsvr
mail-utils time-date mule-util tooltip eldoc electric uniquify
ediff-hook vc-hooks lisp-float-type mwheel term/x-win x-win
term/common-win x-dnd tool-bar dnd fontset image regexp-opt fringe
tabulated-list newcomment elisp-mode lisp-mode prog-mode register page
menu-bar rfn-eshadow timer select scroll-bar mouse jit-lock font-lock
syntax facemenu font-core term/tty-colors frame cl-generic cham georgian
utf-8-lang misc-lang vietnamese tibetan thai tai-viet lao korean
japanese eucjp-ms cp51932 hebrew greek romanian slovak czech european
ethiopic indian cyrillic chinese charscript case-table epa-hook
jka-cmpr-hook help simple abbrev obarray minibuffer cl-preloaded nadvice
loaddefs button faces cus-face macroexp files text-properties overlay
sha1 md5 base64 format env code-pages mule custom widget
hashtable-print-readable backquote dbusbind inotify dynamic-setting
system-font-setting font-render-setting move-toolbar gtk x-toolkit x
multi-tty make-network-process emacs)

Memory information:
((conses 16 101196 5435)
 (symbols 48 20853 0)
 (miscs 40 930 143)
 (strings 32 19171 5350)
 (string-bytes 1 625632)
 (vectors 16 14484)
 (vector-slots 8 475005 4725)
 (floats 8 183 4)
 (intervals 56 204 0)
 (buffers 976 11)
 (heap 1024 30982 1119))

[UbuntuMono-L.ttf (application/x-font-ttf, attachment)]

[bt-full (text/plain, attachment)]

[signature.asc (application/pgp-signature, attachment)]

Message #8 received at 24790 <at> debbugs.gnu.org (full text, mbox):

From: Clément Pit--Claudel <clement.pitclaudel <at> live.com>
To: 24790 <at> debbugs.gnu.org
Subject: Re: bug#24790: Acknowledgement (Segmentation fault when changing
 default font on master)
Date: Tue, 25 Oct 2016 12:05:35 -0400

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

The issue seems to be that the font has self-intersecting glyphs; still, this shouldn't crash Emacs.

On 2016-10-25 10:35, GNU bug Tracking System wrote:
> Thank you for filing a new bug report with debbugs.gnu.org.
> 
> This is an automatically generated reply to let you know your message
> has been received.
> 
> Your message is being forwarded to the package maintainers and other
> interested parties for their attention; they will reply in due course.
> 
> Your message has been sent to the package maintainer(s):
>  bug-gnu-emacs <at> gnu.org
> 
> If you wish to submit further information on this problem, please
> send it to 24790 <at> debbugs.gnu.org.
> 
> Please do not send mail to help-debbugs <at> gnu.org unless you wish
> to report a problem with the Bug-tracking system.
> 

[signature.asc (application/pgp-signature, attachment)]

Message #11 received at 24790 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Clément Pit--Claudel <clement.pitclaudel <at> live.com>
Cc: 24790 <at> debbugs.gnu.org
Subject: Re: bug#24790: Segmentation fault when changing default font on master
Date: Tue, 25 Oct 2016 19:43:50 +0300

> From: Clément Pit--Claudel <clement.pitclaudel <at> live.com>
> Date: Tue, 25 Oct 2016 10:33:32 -0400
> 
> When I select the attached font from the "Set default font" menu, nothing happens (the default font does not change).  When I select it again, emacs segfaults (see attached output of bt full).
> 
> My Emacs has patch af1a69f4d17a482c359d98c00ef86fac835b5fac reverted.  Could this make a difference?

With such a simple reproducer, it's much easier to re-apply the patch
and see if it makes the difference than answer your question by just
looking at your data.

Why do you still have that commit reverted, btw?  I thought the
problem caused by it is nowadays solved: you have a variable to avoid
it.

Message #14 received at 24790 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Clément Pit--Claudel <clement.pitclaudel <at> live.com>
Cc: 24790 <at> debbugs.gnu.org
Subject: Re: bug#24790: Acknowledgement (Segmentation fault when changing
 default font on master)
Date: Tue, 25 Oct 2016 20:09:57 +0300

> From: Clément Pit--Claudel <clement.pitclaudel <at> live.com>
> Date: Tue, 25 Oct 2016 12:05:35 -0400
> 
> The issue seems to be that the font has self-intersecting glyphs; still, this shouldn't crash Emacs.

What do you mean by self-intersecting glyphs?

The backtrace you posted has nothing to do with drawing, AFAICT, it
happens inside internal-set-lisp-face-attribute, which attempts to
open a font, and dies inside AREF during that.

So I don't necessarily see how the glyphs could be related, at least
not with the data you provided.

Message #17 received at 24790 <at> debbugs.gnu.org (full text, mbox):

From: Clément Pit--Claudel <clement.pitclaudel <at> live.com>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: 24790 <at> debbugs.gnu.org
Subject: Re: bug#24790: Acknowledgement (Segmentation fault when changing
 default font on master)
Date: Tue, 25 Oct 2016 13:50:18 -0400

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

On 2016-10-25 13:09, Eli Zaretskii wrote:
>> From: Clément Pit--Claudel <clement.pitclaudel <at> live.com>
>> Date: Tue, 25 Oct 2016 12:05:35 -0400
>>
>> The issue seems to be that the font has self-intersecting glyphs; still, this shouldn't crash Emacs.
> 
> What do you mean by self-intersecting glyphs?
> 
> The backtrace you posted has nothing to do with drawing, AFAICT, it
> happens inside internal-set-lisp-face-attribute, which attempts to
> open a font, and dies inside AREF during that.
> 
> So I don't necessarily see how the glyphs could be related, at least
> not with the data you provided.

The only thing that seems obviously wrong with the font that I posted is that some of the paths of its glyphs are self-intersecting (or at least FontForge says so).  I thought it could be relevant to mention this.

Let me know what data I could provide to help shed light on the issue (I'll remove my cache-related patch and send the results in a separate reply) :)

Clément.

[signature.asc (application/pgp-signature, attachment)]

Message #20 received at 24790 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Clément Pit--Claudel <clement.pitclaudel <at> live.com>
Cc: 24790 <at> debbugs.gnu.org
Subject: Re: bug#24790: Acknowledgement (Segmentation fault when changing
 default font on master)
Date: Tue, 25 Oct 2016 21:21:29 +0300

> Cc: 24790 <at> debbugs.gnu.org
> From: Clément Pit--Claudel <clement.pitclaudel <at> live.com>
> Date: Tue, 25 Oct 2016 13:50:18 -0400
> 
> Let me know what data I could provide to help shed light on the issue (I'll remove my cache-related patch and send the results in a separate reply) :)

Well, I would start by finding out why that AREF segfaults.  Is the
index wrong, or is the array bogus?

Thanks.

Message #23 received at 24790 <at> debbugs.gnu.org (full text, mbox):

From: Clément Pit--Claudel <clement.pitclaudel <at> live.com>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: 24790 <at> debbugs.gnu.org
Subject: Re: bug#24790: Segmentation fault when changing default font on master
Date: Tue, 25 Oct 2016 15:27:45 -0400

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

On 2016-10-25 12:43, Eli Zaretskii wrote:
>> From: Clément Pit--Claudel <clement.pitclaudel <at> live.com>
>> Date: Tue, 25 Oct 2016 10:33:32 -0400
>>
>> When I select the attached font from the "Set default font" menu, nothing happens (the default font does not change).  When I select it again, emacs segfaults (see attached output of bt full).
>>
>> My Emacs has patch af1a69f4d17a482c359d98c00ef86fac835b5fac reverted.  Could this make a difference?
> 
> With such a simple reproducer, it's much easier to re-apply the patch
> and see if it makes the difference than answer your question by just
> looking at your data.

Indeed, and it seems that my reversing this patch is what caused the issue (IOW, I can't crash Emacs any more after re-applying the patch).

Interestingly, if I use 'light as the weight, the right font is selected, and everything works fine (that is, the default frame font changes and Emacs doesn't crash).  But for some reason, x-select-font returns a font-spec in semi-light, and then set-frame-font crashes.

> Why do you still have that commit reverted, btw?  I thought the
> problem caused by it is nowadays solved: you have a variable to avoid
> it.

I never heard about that variable :) Which one is it?

Clément.

[signature.asc (application/pgp-signature, attachment)]

Message #26 received at 24790 <at> debbugs.gnu.org (full text, mbox):

From: Clément Pit--Claudel <clement.pitclaudel <at> live.com>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: 24790 <at> debbugs.gnu.org
Subject: Re: bug#24790: Acknowledgement (Segmentation fault when changing
 default font on master)
Date: Tue, 25 Oct 2016 15:31:22 -0400

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

On 2016-10-25 14:21, Eli Zaretskii wrote:
>> Cc: 24790 <at> debbugs.gnu.org
>> From: Clément Pit--Claudel <clement.pitclaudel <at> live.com>
>> Date: Tue, 25 Oct 2016 13:50:18 -0400
>>
>> Let me know what data I could provide to help shed light on the issue (I'll remove my cache-related patch and send the results in a separate reply) :)
> 
> Well, I would start by finding out why that AREF segfaults.  Is the
> index wrong, or is the array bogus?

Looks like the array is bogus.  I attached a new trace with full debugging information.

Some interesting data:

* This crashes:

(dotimes (_ 2)
  (set-frame-font (font-spec
                   :family "Ubuntu Mono"
                   :weight 'semi-light)))

This doesn't (it properly sets the font)

(dotimes (_ 2)
  (set-frame-font (font-spec
                   :family "Ubuntu Mono"
                   :weight 'light)))

The first one is what x-select-font returns.  I'm not sure why.

Cheers,
Clément.

[gdb.txt (text/plain, attachment)]

[crash.el (text/x-emacs-lisp, attachment)]

[signature.asc (application/pgp-signature, attachment)]

Message #29 received at 24790 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Clément Pit--Claudel <clement.pitclaudel <at> live.com>
Cc: 24790 <at> debbugs.gnu.org
Subject: Re: bug#24790: Segmentation fault when changing default font on master
Date: Tue, 25 Oct 2016 22:42:54 +0300

> Cc: 24790 <at> debbugs.gnu.org
> From: Clément Pit--Claudel <clement.pitclaudel <at> live.com>
> Date: Tue, 25 Oct 2016 15:27:45 -0400
> 
> >> My Emacs has patch af1a69f4d17a482c359d98c00ef86fac835b5fac reverted.  Could this make a difference?
> > 
> > With such a simple reproducer, it's much easier to re-apply the patch
> > and see if it makes the difference than answer your question by just
> > looking at your data.
> 
> Indeed, and it seems that my reversing this patch is what caused the issue (IOW, I can't crash Emacs any more after re-applying the patch).

So this report can be closed?

> > Why do you still have that commit reverted, btw?  I thought the
> > problem caused by it is nowadays solved: you have a variable to avoid
> > it.
> 
> I never heard about that variable :) Which one is it?

inhibit-compacting-font-caches, see NEWS on the emacs-25 branch.

Message #32 received at 24790 <at> debbugs.gnu.org (full text, mbox):

From: Clément Pit--Claudel <clement.pitclaudel <at> live.com>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: 24790 <at> debbugs.gnu.org
Subject: Re: bug#24790: Segmentation fault when changing default font on master
Date: Tue, 25 Oct 2016 16:18:18 -0400

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

On 2016-10-25 15:42, Eli Zaretskii wrote:
>> Cc: 24790 <at> debbugs.gnu.org
>> From: Clément Pit--Claudel <clement.pitclaudel <at> live.com>
>> Date: Tue, 25 Oct 2016 15:27:45 -0400
>>
>>>> My Emacs has patch af1a69f4d17a482c359d98c00ef86fac835b5fac reverted.  Could this make a difference?
>>>
>>> With such a simple reproducer, it's much easier to re-apply the patch
>>> and see if it makes the difference than answer your question by just
>>> looking at your data.
>>
>> Indeed, and it seems that my reversing this patch is what caused the issue (IOW, I can't crash Emacs any more after re-applying the patch).
> 
> So this report can be closed?

Maybe? See below. Emacs is unusable for me if I don't reverse this patch.

>>> Why do you still have that commit reverted, btw?  I thought the
>>> problem caused by it is nowadays solved: you have a variable to avoid
>>> it.
>>
>> I never heard about that variable :) Which one is it?
> 
> inhibit-compacting-font-caches, see NEWS on the emacs-25 branch.

Setting this to t doesn't make any visible difference to bug 21028.  Was it introduced to solve a different problem?

IOW, when I run (building on the example in the 21028 thread)

    src/emacs -Q -mm --eval '(setq inhibit-compacting-font-caches t)' -l 21028.el 21028 -f prettify-symbols-mode

I get the same speed issues (each redisplay takes > 5 seconds), regardless of the value of inhibit-compacting-font-caches.

Cheers,
Clément.

[signature.asc (application/pgp-signature, attachment)]

Message #35 received at 24790 <at> debbugs.gnu.org (full text, mbox):

From: Glenn Morris <rgm <at> gnu.org>
To: Clément Pit--Claudel <clement.pitclaudel <at> live.com>
Cc: 24790 <at> debbugs.gnu.org, Eli Zaretskii <eliz <at> gnu.org>
Subject: Re: bug#24790: Segmentation fault when changing default font on master
Date: Wed, 26 Oct 2016 12:43:43 -0400

So it seems to me that:

i) you should comment on bug #21028 that you tried reverting that
commit, but then Emacs crashes; and that inhibit-compacting-font-caches
doesn't make a difference.

ii) this report should be closed.

Previous Next

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