GNU bug report logs - #22739
25.0.50; Hangs on revertig buffer

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 22739 in the body.
You can then email your comments to 22739 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: Óscar Fuentes <ofv <at> wanadoo.es>
To: bug-gnu-emacs <at> gnu.org
Subject: 25.0.50; Hangs on revertig buffer
Date: Fri, 19 Feb 2016 04:33:03 +0100

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

While executing revert-buffer on a file, Emacs hanged using 100% CPU.
I've executed this operation thousands of times on this file on the past
without problem. The file contains pure ASCII text.

I was unable to reproduce the bug on a fresh session.

Full backtrace attached.

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

[Message part 3 (text/plain, inline)]

In GNU Emacs 25.0.50.29 (x86_64-unknown-linux-gnu, X toolkit)
 of 2016-01-21 built on qcore
Repository revision: 5293d1bdb3e665a565af032a163004c976328cc2
Windowing system distributor 'The X.Org Foundation', version 11.0.11702000
System Description:	Ubuntu 15.10

Configured using:
 'configure --without-toolkit-scroll-bars --with-x-toolkit=lucid'

Configured features:
XAW3D XPM JPEG TIFF GIF PNG SOUND GSETTINGS NOTIFY GNUTLS LIBXML2
FREETYPE XFT ZLIB LUCID X11

Important settings:
  value of $LANG: es_ES.UTF-8
  locale-coding-system: utf-8-unix

Major mode: Outline

Minor modes in effect:
  TeX-PDF-mode: t
  show-paren-mode: t
  diff-auto-refine-mode: t
  global-git-commit-mode: t
  shell-dirtrack-mode: t
  ido-grid-mode: t
  flx-ido-mode: t
  ido-hacks-mode: t
  ido-everywhere: t
  buffer-flip-mode: t
  evil-leader-mode: t
  evil-mode: t
  evil-local-mode: t
  global-anzu-mode: t
  anzu-mode: t
  tooltip-mode: t
  global-eldoc-mode: t
  electric-indent-mode: t
  mouse-wheel-mode: t
  tool-bar-mode: t
  file-name-shadow-mode: t
  global-font-lock-mode: t
  font-lock-mode: t
  auto-composition-mode: t
  auto-encryption-mode: t
  auto-compression-mode: t
  column-number-mode: t
  line-number-mode: t
  auto-fill-function: do-auto-fill

Recent messages:
Loading /home/oscar/elisp/ofv-lp0.el (source)...done
Loading /home/oscar/.emacs.d/elpa/auctex-11.89/preview.el (source)...done
Estableciendo variables para AucTex
Loading /home/oscar/elisp/ofv-wordnet.el (source)...done
Loading /home/oscar/elisp/tkidb.el (source)...done
Loading /home/oscar/elisp/ofv-org.el (source)...done
Loading /home/oscar/elisp/ofv-issues.el (source)...done
Loading /home/oscar/elisp/emacs.el (source)...done
For information about GNU Emacs and the GNU system, type C-h C-a.
Mark saved where search started [2 times]
Quit

Load-path shadows:
/home/oscar/elisp/magit/lisp/with-editor hides ~/elisp/singles/with-editor
/home/oscar/.emacs.d/elpa/org-20160208/ob-ref hides /usr/local/share/emacs/25.0.50/lisp/org/ob-ref
/home/oscar/.emacs.d/elpa/org-20160208/ob-dot hides /usr/local/share/emacs/25.0.50/lisp/org/ob-dot
/home/oscar/.emacs.d/elpa/org-20160208/ob-python hides /usr/local/share/emacs/25.0.50/lisp/org/ob-python
/home/oscar/.emacs.d/elpa/org-20160208/ob-fortran hides /usr/local/share/emacs/25.0.50/lisp/org/ob-fortran
/home/oscar/.emacs.d/elpa/org-20160208/org-timer hides /usr/local/share/emacs/25.0.50/lisp/org/org-timer
/home/oscar/.emacs.d/elpa/org-20160208/org-compat hides /usr/local/share/emacs/25.0.50/lisp/org/org-compat
/home/oscar/.emacs.d/elpa/org-20160208/org-info hides /usr/local/share/emacs/25.0.50/lisp/org/org-info
/home/oscar/.emacs.d/elpa/org-20160208/org-docview hides /usr/local/share/emacs/25.0.50/lisp/org/org-docview
/home/oscar/.emacs.d/elpa/org-20160208/ob-haskell hides /usr/local/share/emacs/25.0.50/lisp/org/ob-haskell
/home/oscar/.emacs.d/elpa/org-20160208/ob-lisp hides /usr/local/share/emacs/25.0.50/lisp/org/ob-lisp
/home/oscar/.emacs.d/elpa/org-20160208/ob-css hides /usr/local/share/emacs/25.0.50/lisp/org/ob-css
/home/oscar/.emacs.d/elpa/org-20160208/ob-sass hides /usr/local/share/emacs/25.0.50/lisp/org/ob-sass
/home/oscar/.emacs.d/elpa/org-20160208/ob-core hides /usr/local/share/emacs/25.0.50/lisp/org/ob-core
/home/oscar/.emacs.d/elpa/org-20160208/ox-html hides /usr/local/share/emacs/25.0.50/lisp/org/ox-html
/home/oscar/.emacs.d/elpa/org-20160208/ob-table hides /usr/local/share/emacs/25.0.50/lisp/org/ob-table
/home/oscar/.emacs.d/elpa/org-20160208/org-irc hides /usr/local/share/emacs/25.0.50/lisp/org/org-irc
/home/oscar/.emacs.d/elpa/org-20160208/org-loaddefs hides /usr/local/share/emacs/25.0.50/lisp/org/org-loaddefs
/home/oscar/.emacs.d/elpa/org-20160208/ob-C hides /usr/local/share/emacs/25.0.50/lisp/org/ob-C
/home/oscar/.emacs.d/elpa/org-20160208/org-bibtex hides /usr/local/share/emacs/25.0.50/lisp/org/org-bibtex
/home/oscar/.emacs.d/elpa/org-20160208/ox-publish hides /usr/local/share/emacs/25.0.50/lisp/org/ox-publish
/home/oscar/.emacs.d/elpa/org-20160208/ob-maxima hides /usr/local/share/emacs/25.0.50/lisp/org/ob-maxima
/home/oscar/.emacs.d/elpa/org-20160208/ob-awk hides /usr/local/share/emacs/25.0.50/lisp/org/ob-awk
/home/oscar/.emacs.d/elpa/org-20160208/org-mouse hides /usr/local/share/emacs/25.0.50/lisp/org/org-mouse
/home/oscar/.emacs.d/elpa/org-20160208/ob-eval hides /usr/local/share/emacs/25.0.50/lisp/org/ob-eval
/home/oscar/.emacs.d/elpa/org-20160208/ob-plantuml hides /usr/local/share/emacs/25.0.50/lisp/org/ob-plantuml
/home/oscar/.emacs.d/elpa/org-20160208/org-id hides /usr/local/share/emacs/25.0.50/lisp/org/org-id
/home/oscar/.emacs.d/elpa/org-20160208/org-list hides /usr/local/share/emacs/25.0.50/lisp/org/org-list
/home/oscar/.emacs.d/elpa/org-20160208/ox-beamer hides /usr/local/share/emacs/25.0.50/lisp/org/ox-beamer
/home/oscar/.emacs.d/elpa/org-20160208/ob-picolisp hides /usr/local/share/emacs/25.0.50/lisp/org/ob-picolisp
/home/oscar/.emacs.d/elpa/org-20160208/ob-lilypond hides /usr/local/share/emacs/25.0.50/lisp/org/ob-lilypond
/home/oscar/.emacs.d/elpa/org-20160208/ob-calc hides /usr/local/share/emacs/25.0.50/lisp/org/ob-calc
/home/oscar/.emacs.d/elpa/org-20160208/ob-js hides /usr/local/share/emacs/25.0.50/lisp/org/ob-js
/home/oscar/.emacs.d/elpa/org-20160208/org-colview hides /usr/local/share/emacs/25.0.50/lisp/org/org-colview
/home/oscar/.emacs.d/elpa/org-20160208/org-faces hides /usr/local/share/emacs/25.0.50/lisp/org/org-faces
/home/oscar/.emacs.d/elpa/org-20160208/org-feed hides /usr/local/share/emacs/25.0.50/lisp/org/org-feed
/home/oscar/.emacs.d/elpa/org-20160208/org-plot hides /usr/local/share/emacs/25.0.50/lisp/org/org-plot
/home/oscar/.emacs.d/elpa/org-20160208/ob-clojure hides /usr/local/share/emacs/25.0.50/lisp/org/ob-clojure
/home/oscar/.emacs.d/elpa/org-20160208/org-archive hides /usr/local/share/emacs/25.0.50/lisp/org/org-archive
/home/oscar/.emacs.d/elpa/org-20160208/ob-octave hides /usr/local/share/emacs/25.0.50/lisp/org/ob-octave
/home/oscar/.emacs.d/elpa/org-20160208/ob-scala hides /usr/local/share/emacs/25.0.50/lisp/org/ob-scala
/home/oscar/.emacs.d/elpa/org-20160208/org-crypt hides /usr/local/share/emacs/25.0.50/lisp/org/org-crypt
/home/oscar/.emacs.d/elpa/org-20160208/ob-screen hides /usr/local/share/emacs/25.0.50/lisp/org/ob-screen
/home/oscar/.emacs.d/elpa/org-20160208/org-ctags hides /usr/local/share/emacs/25.0.50/lisp/org/org-ctags
/home/oscar/.emacs.d/elpa/org-20160208/ob-java hides /usr/local/share/emacs/25.0.50/lisp/org/ob-java
/home/oscar/.emacs.d/elpa/org-20160208/ob-io hides /usr/local/share/emacs/25.0.50/lisp/org/ob-io
/home/oscar/.emacs.d/elpa/org-20160208/ob-latex hides /usr/local/share/emacs/25.0.50/lisp/org/ob-latex
/home/oscar/.emacs.d/elpa/org-20160208/ob-ocaml hides /usr/local/share/emacs/25.0.50/lisp/org/ob-ocaml
/home/oscar/.emacs.d/elpa/org-20160208/org-clock hides /usr/local/share/emacs/25.0.50/lisp/org/org-clock
/home/oscar/.emacs.d/elpa/org-20160208/ox-md hides /usr/local/share/emacs/25.0.50/lisp/org/ox-md
/home/oscar/.emacs.d/elpa/org-20160208/ob-org hides /usr/local/share/emacs/25.0.50/lisp/org/ob-org
/home/oscar/.emacs.d/elpa/org-20160208/org-gnus hides /usr/local/share/emacs/25.0.50/lisp/org/org-gnus
/home/oscar/.emacs.d/elpa/org-20160208/ob-R hides /usr/local/share/emacs/25.0.50/lisp/org/ob-R
/home/oscar/.emacs.d/elpa/org-20160208/ob-matlab hides /usr/local/share/emacs/25.0.50/lisp/org/ob-matlab
/home/oscar/.emacs.d/elpa/org-20160208/org-entities hides /usr/local/share/emacs/25.0.50/lisp/org/org-entities
/home/oscar/.emacs.d/elpa/org-20160208/ob-exp hides /usr/local/share/emacs/25.0.50/lisp/org/ob-exp
/home/oscar/.emacs.d/elpa/org-20160208/org-mhe hides /usr/local/share/emacs/25.0.50/lisp/org/org-mhe
/home/oscar/.emacs.d/elpa/org-20160208/ox-texinfo hides /usr/local/share/emacs/25.0.50/lisp/org/ox-texinfo
/home/oscar/.emacs.d/elpa/org-20160208/ox-odt hides /usr/local/share/emacs/25.0.50/lisp/org/ox-odt
/home/oscar/.emacs.d/elpa/org-20160208/org-footnote hides /usr/local/share/emacs/25.0.50/lisp/org/org-footnote
/home/oscar/.emacs.d/elpa/org-20160208/org-agenda hides /usr/local/share/emacs/25.0.50/lisp/org/org-agenda
/home/oscar/.emacs.d/elpa/org-20160208/org-element hides /usr/local/share/emacs/25.0.50/lisp/org/org-element
/home/oscar/.emacs.d/elpa/org-20160208/ox-ascii hides /usr/local/share/emacs/25.0.50/lisp/org/ox-ascii
/home/oscar/.emacs.d/elpa/org-20160208/org-inlinetask hides /usr/local/share/emacs/25.0.50/lisp/org/org-inlinetask
/home/oscar/.emacs.d/elpa/org-20160208/org-attach hides /usr/local/share/emacs/25.0.50/lisp/org/org-attach
/home/oscar/.emacs.d/elpa/org-20160208/ob-ditaa hides /usr/local/share/emacs/25.0.50/lisp/org/ob-ditaa
/home/oscar/.emacs.d/elpa/org-20160208/org-indent hides /usr/local/share/emacs/25.0.50/lisp/org/org-indent
/home/oscar/.emacs.d/elpa/org-20160208/ob-lob hides /usr/local/share/emacs/25.0.50/lisp/org/ob-lob
/home/oscar/.emacs.d/elpa/org-20160208/org-capture hides /usr/local/share/emacs/25.0.50/lisp/org/org-capture
/home/oscar/.emacs.d/elpa/org-20160208/ob-ruby hides /usr/local/share/emacs/25.0.50/lisp/org/ob-ruby
/home/oscar/.emacs.d/elpa/org-20160208/ob-mscgen hides /usr/local/share/emacs/25.0.50/lisp/org/ob-mscgen
/home/oscar/.emacs.d/elpa/org-20160208/ob-shen hides /usr/local/share/emacs/25.0.50/lisp/org/ob-shen
/home/oscar/.emacs.d/elpa/org-20160208/org-protocol hides /usr/local/share/emacs/25.0.50/lisp/org/org-protocol
/home/oscar/.emacs.d/elpa/org-20160208/org-version hides /usr/local/share/emacs/25.0.50/lisp/org/org-version
/home/oscar/.emacs.d/elpa/org-20160208/org-bbdb hides /usr/local/share/emacs/25.0.50/lisp/org/org-bbdb
/home/oscar/.emacs.d/elpa/org-20160208/ox-latex hides /usr/local/share/emacs/25.0.50/lisp/org/ox-latex
/home/oscar/.emacs.d/elpa/org-20160208/org-mobile hides /usr/local/share/emacs/25.0.50/lisp/org/org-mobile
/home/oscar/.emacs.d/elpa/org-20160208/ob-comint hides /usr/local/share/emacs/25.0.50/lisp/org/ob-comint
/home/oscar/.emacs.d/elpa/org-20160208/ob-gnuplot hides /usr/local/share/emacs/25.0.50/lisp/org/ob-gnuplot
/home/oscar/.emacs.d/elpa/org-20160208/ox-icalendar hides /usr/local/share/emacs/25.0.50/lisp/org/ox-icalendar
/home/oscar/.emacs.d/elpa/org-20160208/org-eshell hides /usr/local/share/emacs/25.0.50/lisp/org/org-eshell
/home/oscar/.emacs.d/elpa/org-20160208/org-src hides /usr/local/share/emacs/25.0.50/lisp/org/org-src
/home/oscar/.emacs.d/elpa/org-20160208/ob-scheme hides /usr/local/share/emacs/25.0.50/lisp/org/ob-scheme
/home/oscar/.emacs.d/elpa/org-20160208/org-install hides /usr/local/share/emacs/25.0.50/lisp/org/org-install
/home/oscar/.emacs.d/elpa/org-20160208/ox-org hides /usr/local/share/emacs/25.0.50/lisp/org/ox-org
/home/oscar/.emacs.d/elpa/org-20160208/ob-emacs-lisp hides /usr/local/share/emacs/25.0.50/lisp/org/ob-emacs-lisp
/home/oscar/.emacs.d/elpa/org-20160208/ob-makefile hides /usr/local/share/emacs/25.0.50/lisp/org/ob-makefile
/home/oscar/.emacs.d/elpa/org-20160208/org-datetree hides /usr/local/share/emacs/25.0.50/lisp/org/org-datetree
/home/oscar/.emacs.d/elpa/org-20160208/org-rmail hides /usr/local/share/emacs/25.0.50/lisp/org/org-rmail
/home/oscar/.emacs.d/elpa/org-20160208/ob-keys hides /usr/local/share/emacs/25.0.50/lisp/org/ob-keys
/home/oscar/.emacs.d/elpa/org-20160208/org-macro hides /usr/local/share/emacs/25.0.50/lisp/org/org-macro
/home/oscar/.emacs.d/elpa/org-20160208/ob hides /usr/local/share/emacs/25.0.50/lisp/org/ob
/home/oscar/.emacs.d/elpa/org-20160208/ob-ledger hides /usr/local/share/emacs/25.0.50/lisp/org/ob-ledger
/home/oscar/.emacs.d/elpa/org-20160208/ox hides /usr/local/share/emacs/25.0.50/lisp/org/ox
/home/oscar/.emacs.d/elpa/org-20160208/org-habit hides /usr/local/share/emacs/25.0.50/lisp/org/org-habit
/home/oscar/.emacs.d/elpa/org-20160208/ob-sql hides /usr/local/share/emacs/25.0.50/lisp/org/ob-sql
/home/oscar/.emacs.d/elpa/org-20160208/org-macs hides /usr/local/share/emacs/25.0.50/lisp/org/org-macs
/home/oscar/.emacs.d/elpa/org-20160208/org-pcomplete hides /usr/local/share/emacs/25.0.50/lisp/org/org-pcomplete
/home/oscar/.emacs.d/elpa/org-20160208/org-table hides /usr/local/share/emacs/25.0.50/lisp/org/org-table
/home/oscar/.emacs.d/elpa/org-20160208/ox-man hides /usr/local/share/emacs/25.0.50/lisp/org/ox-man
/home/oscar/.emacs.d/elpa/org-20160208/org hides /usr/local/share/emacs/25.0.50/lisp/org/org
/home/oscar/.emacs.d/elpa/org-20160208/ob-asymptote hides /usr/local/share/emacs/25.0.50/lisp/org/ob-asymptote
/home/oscar/.emacs.d/elpa/org-20160208/ob-tangle hides /usr/local/share/emacs/25.0.50/lisp/org/ob-tangle
/home/oscar/.emacs.d/elpa/org-20160208/ob-perl hides /usr/local/share/emacs/25.0.50/lisp/org/ob-perl
/home/oscar/.emacs.d/elpa/org-20160208/org-w3m hides /usr/local/share/emacs/25.0.50/lisp/org/org-w3m
/home/oscar/.emacs.d/elpa/org-20160208/ob-sqlite hides /usr/local/share/emacs/25.0.50/lisp/org/ob-sqlite

Features:
(shadow sort mail-extr emacsbug sendmail character-fold misearch
multi-isearch vc vc-dispatcher map wgrep grep ag vc-svn find-dired s
etags-select etags xref project org-crypt preview prv-emacs reporter
desktop frameset latex tex-style tex-buf tex lp0-mode aggressive-indent
lisp-mnt compile paren paredit-menu yasnippet psvn wid-edit pp elp
ediff-merg ediff-wind ediff-diff ediff-mult ediff-help ediff-init
ediff-util vc-git org-magit org-element avl-tree rx org org-macro
org-footnote org-pcomplete org-list org-faces org-entities noutline
outline org-version ob-emacs-lisp ob ob-tangle ob-ref ob-lob ob-table
ob-exp org-src ob-keys ob-comint ob-core ob-eval org-compat org-macs
org-loaddefs find-func cal-menu calendar cal-loaddefs magit-blame
magit-stash magit-bisect magit-remote magit-commit magit-sequence magit
magit-apply magit-wip magit-log magit-diff smerge-mode diff-mode
magit-core magit-autorevert autorevert filenotify magit-process
magit-popup magit-mode magit-git crm magit-section magit-utils
git-commit log-edit message dired rfc822 mml mml-sec epg mm-decode
mm-bodies mm-encode mail-parse rfc2231 rfc2047 rfc2045 ietf-drums
mailabbrev mail-utils gmm-utils mailheader pcvs-util add-log with-editor
tramp-sh tramp tramp-compat tramp-loaddefs trampver ucs-normalize shell
pcomplete comint ansi-color format-spec server column-overflow
flx-isearch ido-grid-mode flx-ido flx ido-hacks ido seq buffer-flip
key-chord evil-leader evil-paredit paredit evil-anzu evil
evil-integration evil-maps evil-commands ffap url-parse auth-source
eieio byte-opt bytecomp byte-compile cl-extra cconv eieio-core gnus-util
mm-util help-fns help-mode mail-prsvr password-cache url-vars
evil-command-window derived evil-types evil-search evil-ex evil-macros
evil-repeat evil-states evil-core evil-common windmove rect
evil-digraphs evil-vars goto-chg edmacro kmacro avy ring ws-butler
diminish cl anzu cl-seq easy-mmode cl-macs thingatpt cl-loaddefs pcase
cl-lib avoid solarized-light-theme solarized-dark-theme solarized dash
gv finder-inf tex-site info package easymenu epg-config advice time-date
mule-util tooltip eldoc electric uniquify ediff-hook vc-hooks
lisp-float-type mwheel 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 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 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 inotify dynamic-setting
system-font-setting font-render-setting x-toolkit x multi-tty
make-network-process emacs)

Message #8 received at 22739 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Óscar Fuentes <ofv <at> wanadoo.es>
Cc: 22739 <at> debbugs.gnu.org
Subject: Re: bug#22739: 25.0.50; Hangs on revertig buffer
Date: Fri, 19 Feb 2016 10:16:50 +0200

> From: Óscar Fuentes <ofv <at> wanadoo.es>
> Date: Fri, 19 Feb 2016 04:33:03 +0100
> 
> While executing revert-buffer on a file, Emacs hanged using 100% CPU.
> I've executed this operation thousands of times on this file on the past
> without problem. The file contains pure ASCII text.
> 
> I was unable to reproduce the bug on a fresh session.
> 
> Full backtrace attached.

It sounds like Emacs was inflooping somewhere.  When there's a
suspicion that Emacs infloops, the single most important information
is where it loops.  The file etc/DEBUG describes a technique to find
that out (search for "If the symptom of the bug is that Emacs fails to
respond").  So next time this happens, please use that technique and
post the findings.

Just looking at the backtrace, it's hard to say anything intelligent,
as backtraces taken from a looping program tend to be random.  The
below is a feeble attempt to do something with the backtrace
information.

What is in the buffer that visits this file between positions 282624
and 330908?  Are there any display strings or overlays in that range,
or some other properties that affect the display (like invisible
text), and if so, can you describe them?

Thanks.

Message #11 received at 22739 <at> debbugs.gnu.org (full text, mbox):

From: Óscar Fuentes <ofv <at> wanadoo.es>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: 22739 <at> debbugs.gnu.org
Subject: Re: bug#22739: 25.0.50; Hangs on revertig buffer
Date: Fri, 19 Feb 2016 13:44:41 +0100

Eli Zaretskii <eliz <at> gnu.org> writes:

> It sounds like Emacs was inflooping somewhere.  When there's a
> suspicion that Emacs infloops, the single most important information
> is where it loops.  The file etc/DEBUG describes a technique to find
> that out (search for "If the symptom of the bug is that Emacs fails to
> respond").  So next time this happens, please use that technique and
> post the findings.

I still have the gdb session running. Will try the `finish' trick later.

> What is in the buffer that visits this file between positions 282624
> and 330908?  Are there any display strings or overlays in that range,
> or some other properties that affect the display (like invisible
> text), and if so, can you describe them?

The buffer is visiting a pure ASCII file in Fundamental mode. There is
nothing special about it. The longest line is about 105 chars wide.

Message #14 received at 22739 <at> debbugs.gnu.org (full text, mbox):

From: Óscar Fuentes <ofv <at> wanadoo.es>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: 22739 <at> debbugs.gnu.org
Subject: Re: bug#22739: 25.0.50; Hangs on revertig buffer
Date: Fri, 19 Feb 2016 14:04:34 +0100

(gdb) finish
Run till exit from #0  bidi_resolve_weak (bidi_it=bidi_it <at> entry=0x7ffead896e00)
    at ../../emacs/src/bidi.c:2346
bidi_resolve_brackets (bidi_it=bidi_it <at> entry=0x7ffead896e00)
    at ../../emacs/src/bidi.c:2809
2809          if (type == NEUTRAL_ON)
Value returned is $1 = WEAK_BN
(gdb) 
Run till exit from #0  bidi_resolve_brackets (bidi_it=bidi_it <at> entry=0x7ffead896e00)
    at ../../emacs/src/bidi.c:2809
0x00000000004a2563 in bidi_resolve_neutral (bidi_it=bidi_it <at> entry=0x7ffead896e00)
    at ../../emacs/src/bidi.c:3026
3026                type = bidi_resolve_brackets (bidi_it);
Value returned is $2 = WEAK_BN
(gdb) 
Run till exit from #0  0x00000000004a2563 in bidi_resolve_neutral (
    bidi_it=bidi_it <at> entry=0x7ffead896e00) at ../../emacs/src/bidi.c:3026




Just as an illustration, this is part of the contents of the buffer at
the displayed position. The point is on the first `!', on the center of
the screen. I'll mention too that the frame is in fullscreen mode and
emacs is daemonized.


Testeando ./redef/redef015.lp0
Cmd: /home/oscar/dev/idb/lp0/build/remove-isws/clang/release/lp0 -cs -p ../lib -t -f ./redef/redef015.lp0
Terminado ./redef/redef015.lp0
Ok
Success en result.
Se obtuvo: Primera definicion de bar
Segunda definicion de bar.

Testeando ./redef/redef016.lp0
Cmd: /home/oscar/dev/idb/lp0/build/remove-isws/clang/release/lp0 -cs -p ../lib -t -f ./redef/redef016.lp0
Terminado ./redef/redef016.lp0
!!!!Failure
Complexity: 11
Result:
Segunda definicion de bar
Segunda definicion de bar

Testeando ./redef/redef017.lp0
Cmd: /home/oscar/dev/idb/lp0/build/remove-isws/clang/release/lp0 -cs -p ../lib -t -f ./redef/redef017.lp0
Terminado ./redef/redef017.lp0
Ok
Success en result.
Se obtuvo: Primera definicion de bar
Segunda definicion de bar.

Message #17 received at 22739 <at> debbugs.gnu.org (full text, mbox):

From: Óscar Fuentes <ofv <at> wanadoo.es>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: 22739 <at> debbugs.gnu.org
Subject: Re: bug#22739: 25.0.50; Hangs on revertig buffer
Date: Fri, 19 Feb 2016 14:21:48 +0100

Sorry, for some reason `finish' was taking a long time to return from
the last frame I mentioned on my previous message. This is the complete
sequence:


(gdb) finish
Run till exit from #0  bidi_resolve_weak (bidi_it=bidi_it <at> entry=0x7ffead896e00)
    at ../../emacs/src/bidi.c:2346
bidi_resolve_brackets (bidi_it=bidi_it <at> entry=0x7ffead896e00)
    at ../../emacs/src/bidi.c:2809
2809          if (type == NEUTRAL_ON)
Value returned is $1 = WEAK_BN
(gdb) 
Run till exit from #0  bidi_resolve_brackets (bidi_it=bidi_it <at> entry=0x7ffead896e00)
    at ../../emacs/src/bidi.c:2809
0x00000000004a2563 in bidi_resolve_neutral (bidi_it=bidi_it <at> entry=0x7ffead896e00)
    at ../../emacs/src/bidi.c:3026
3026                type = bidi_resolve_brackets (bidi_it);
Value returned is $2 = WEAK_BN
(gdb) 
Run till exit from #0  0x00000000004a2563 in bidi_resolve_neutral (
    bidi_it=bidi_it <at> entry=0x7ffead896e00) at ../../emacs/src/bidi.c:3026
bidi_level_of_next_char (bidi_it=bidi_it <at> entry=0x7ffead896e00)
    at ../../emacs/src/bidi.c:3201
3201      if (type == NEUTRAL_B)
Value returned is $3 = STRONG_L
(gdb) finish
Run till exit from #0  bidi_level_of_next_char (bidi_it=bidi_it <at> entry=0x7ffead896e00)
    at ../../emacs/src/bidi.c:3201
bidi_move_to_visually_next (bidi_it=bidi_it <at> entry=0x7ffead896e00)
    at ../../emacs/src/bidi.c:3394
3394      if (new_level != old_level)
Value returned is $4 = 0
(gdb) 
Run till exit from #0  bidi_move_to_visually_next (bidi_it=bidi_it <at> entry=0x7ffead896e00)
    at ../../emacs/src/bidi.c:3394
set_iterator_to_next (it=it <at> entry=0x7ffead896460, reseat_p=reseat_p <at> entry=true)
    at ../../emacs/src/xdisp.c:7398
7398                  if (prev_scan_dir != it->bidi_it.scan_dir)
(gdb) 
Run till exit from #0  set_iterator_to_next (it=it <at> entry=0x7ffead896460, 
    reseat_p=reseat_p <at> entry=true) at ../../emacs/src/xdisp.c:7398
move_it_in_display_line_to (it=it <at> entry=0x7ffead896460, 
    to_charpos=to_charpos <at> entry=330908, to_x=to_x <at> entry=-1, 
    op=op <at> entry=(MOVE_TO_VPOS | MOVE_TO_POS)) at ../../emacs/src/xdisp.c:8980
8980          if (IT_CHARPOS (*it) < CHARPOS (this_line_min_pos))
(gdb) 
Run till exit from #0  move_it_in_display_line_to (it=it <at> entry=0x7ffead896460, 
    to_charpos=to_charpos <at> entry=330908, to_x=to_x <at> entry=-1, 
    op=op <at> entry=(MOVE_TO_VPOS | MOVE_TO_POS)) at ../../emacs/src/xdisp.c:8980
move_it_to (it=it <at> entry=0x7ffead896460, to_charpos=to_charpos <at> entry=330908, 
    to_x=to_x <at> entry=-1, to_y=to_y <at> entry=-1, to_vpos=1, op=op <at> entry=12)
    at ../../emacs/src/xdisp.c:9161
9161                  if (skip == MOVE_POS_MATCH_OR_ZV || it->vpos == to_vpos)
Value returned is $5 = MOVE_LINE_CONTINUED
(gdb) 
Run till exit from #0  move_it_to (it=it <at> entry=0x7ffead896460, 
    to_charpos=to_charpos <at> entry=330908, to_x=to_x <at> entry=-1, to_y=to_y <at> entry=-1, 
    to_vpos=1, op=op <at> entry=12) at ../../emacs/src/xdisp.c:9161
move_it_vertically_backward (it=it <at> entry=0x7ffead898b20, dy=dy <at> entry=500)
    at ../../emacs/src/xdisp.c:9489
9489                   && SREF (it2.string, IT_STRING_BYTEPOS (it2) - 1) == '\n')));
Value returned is $6 = 812
(gdb) 
Run till exit from #0  move_it_vertically_backward (it=it <at> entry=0x7ffead898b20, 
    dy=dy <at> entry=500) at ../../emacs/src/xdisp.c:9489
0x000000000054ae08 in internal_condition_case (
    bfun=bfun <at> entry=0x4e80d0 <command_loop_1>, handlers=handlers <at> entry=17856, 
    hfun=hfun <at> entry=0x4de9c0 <cmd_error>) at ../../emacs/src/eval.c:1300
1300      if (sys_setjmp (c->jmp))
(gdb) 
Run till exit from #0  0x000000000054ae08 in internal_condition_case (
    bfun=bfun <at> entry=0x4e80d0 <command_loop_1>, handlers=handlers <at> entry=17856, 
    hfun=hfun <at> entry=0x4de9c0 <cmd_error>) at ../../emacs/src/eval.c:1300
command_loop_2 (ignore=ignore <at> entry=0) at ../../emacs/src/keyboard.c:1090
1090      while (!NILP (val));
Value returned is $7 = 2
(gdb) 
Run till exit from #0  command_loop_2 (ignore=ignore <at> entry=0)
    at ../../emacs/src/keyboard.c:1090

Message #20 received at 22739 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Óscar Fuentes <ofv <at> wanadoo.es>
Cc: 22739 <at> debbugs.gnu.org
Subject: Re: bug#22739: 25.0.50; Hangs on revertig buffer
Date: Fri, 19 Feb 2016 17:28:14 +0200

> From: Óscar Fuentes <ofv <at> wanadoo.es>
> Cc: 22739 <at> debbugs.gnu.org
> Date: Fri, 19 Feb 2016 13:44:41 +0100
> 
> > What is in the buffer that visits this file between positions 282624
> > and 330908?  Are there any display strings or overlays in that range,
> > or some other properties that affect the display (like invisible
> > text), and if so, can you describe them?
> 
> The buffer is visiting a pure ASCII file in Fundamental mode. There is
> nothing special about it. The longest line is about 105 chars wide.

OK, thanks.  I had a theory about a possible place in the code where
it could infloop, but that theory now eats dust, as it cannot happen
with plain text.

Message #23 received at 22739 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Óscar Fuentes <ofv <at> wanadoo.es>
Cc: 22739 <at> debbugs.gnu.org
Subject: Re: bug#22739: 25.0.50; Hangs on revertig buffer
Date: Fri, 19 Feb 2016 17:45:58 +0200

> From: Óscar Fuentes <ofv <at> wanadoo.es>
> Cc: 22739 <at> debbugs.gnu.org
> Date: Fri, 19 Feb 2016 14:21:48 +0100
> 
> move_it_vertically_backward (it=it <at> entry=0x7ffead898b20, dy=dy <at> entry=500)
>     at ../../emacs/src/xdisp.c:9489
> 9489                   && SREF (it2.string, IT_STRING_BYTEPOS (it2) - 1) == '\n')));
> Value returned is $6 = 812
> (gdb) 
> Run till exit from #0  move_it_vertically_backward (it=it <at> entry=0x7ffead898b20, 
>     dy=dy <at> entry=500) at ../../emacs/src/xdisp.c:9489
> 0x000000000054ae08 in internal_condition_case (
>     bfun=bfun <at> entry=0x4e80d0 <command_loop_1>, handlers=handlers <at> entry=17856, 
>     hfun=hfun <at> entry=0x4de9c0 <cmd_error>) at ../../emacs/src/eval.c:1300
> 1300      if (sys_setjmp (c->jmp))
> (gdb) 

This seems to say that the display engine signaled an error.  If so,
the error message should be in *Messages*.  One way of accessing the
error messages is to use the facilities in etc/emacs-buffer.gdb (after
you instruct GDB to load that file with the "source" command) to
access the text of *Messages*.  An alternative is to put a breakpoint
in Fsignal, and look at the error symbol and data when the breakpoint
breaks, using the commands in src/.gdbinit that display Lisp objects.
(But do NOT use "pp", because it invokes a function inside Emacs.)

Can you do any of these?  If you need more detailed instructions, just
ask.

Thanks.

Message #26 received at 22739 <at> debbugs.gnu.org (full text, mbox):

From: Óscar Fuentes <ofv <at> wanadoo.es>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: 22739 <at> debbugs.gnu.org
Subject: Re: bug#22739: 25.0.50; Hangs on revertig buffer
Date: Fri, 19 Feb 2016 23:09:02 +0100

Eli Zaretskii <eliz <at> gnu.org> writes:

> This seems to say that the display engine signaled an error.  If so,
> the error message should be in *Messages*.  One way of accessing the
> error messages is to use the facilities in etc/emacs-buffer.gdb (after
> you instruct GDB to load that file with the "source" command) to
> access the text of *Messages*.

After using ybuffer-save for recovering *Messages*, the last lines are

reverting...
reverted
Quit

The "reverting... reverted" lines are the output of a wrapper function I
wrote for executing revert-buffer:

(defun ofv-revert-buffer ()
  (interactive)
  (message "reverting...")
  (revert-buffer nil (not (buffer-modified-p)))
  (message "reverted"))


It is interesting that the last line of *Messages* is "Quit", but the
minibuffer on the frozen Emacs session shows "reverted". From where
comes "Quit"? Why it was not displayed? When Emacs froze, I tried C-g,
but that was after a few seconds.

> An alternative is to put a breakpoint
> in Fsignal, and look at the error symbol and data when the breakpoint
> breaks, using the commands in src/.gdbinit that display Lisp objects.
> (But do NOT use "pp", because it invokes a function inside Emacs.)

I'm afraid that the Emacs instance is not resumable anymore. After using
the "finish" trick on the previous message, I was unable to interrup
Emacs either with C-c on the gdb console (as I did the first time after
attaching gdb) nor C-z on the Emacs window (as DEBUG suggests.) I sent a
SIGSTOP signal to the Emacs process to gain access to the gdb command
line and now it can't resume.

Message #29 received at 22739 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Óscar Fuentes <ofv <at> wanadoo.es>
Cc: 22739 <at> debbugs.gnu.org
Subject: Re: bug#22739: 25.0.50; Hangs on revertig buffer
Date: Sat, 20 Feb 2016 10:08:08 +0200

> From: Óscar Fuentes <ofv <at> wanadoo.es>
> Cc: 22739 <at> debbugs.gnu.org
> Date: Fri, 19 Feb 2016 23:09:02 +0100
> 
> Eli Zaretskii <eliz <at> gnu.org> writes:
> 
> > This seems to say that the display engine signaled an error.  If so,
> > the error message should be in *Messages*.  One way of accessing the
> > error messages is to use the facilities in etc/emacs-buffer.gdb (after
> > you instruct GDB to load that file with the "source" command) to
> > access the text of *Messages*.
> 
> After using ybuffer-save for recovering *Messages*, the last lines are
> 
> reverting...
> reverted
> Quit
> 
> The "reverting... reverted" lines are the output of a wrapper function I
> wrote for executing revert-buffer:
> 
> (defun ofv-revert-buffer ()
>   (interactive)
>   (message "reverting...")
>   (revert-buffer nil (not (buffer-modified-p)))
>   (message "reverted"))
> 
> 
> It is interesting that the last line of *Messages* is "Quit", but the
> minibuffer on the frozen Emacs session shows "reverted". From where
> comes "Quit"? Why it was not displayed? When Emacs froze, I tried C-g,
> but that was after a few seconds.

It was probably not displayed because Emacs was busy doing something,
so it didn't enter redisplay.

In any case, that C-g is most probably the reason for what you saw
using "finish" -- it caused Emacs to throw to top level.

IOW, the data you collected is probably not relevant to the root cause
of the problem.

> I'm afraid that the Emacs instance is not resumable anymore. After using
> the "finish" trick on the previous message, I was unable to interrup
> Emacs either with C-c on the gdb console (as I did the first time after
> attaching gdb)

Are you saying that just attaching GDB didn't stop Emacs, and you
needed C-c to do that?

> nor C-z on the Emacs window (as DEBUG suggests.) I sent a SIGSTOP
> signal to the Emacs process to gain access to the gdb command line
> and now it can't resume.

What do you mean by "cannot resume"?  What happens if you try?
What happens if you kill the GDB process, and then attach another GDB
to Emacs?

Thanks.

Message #32 received at 22739 <at> debbugs.gnu.org (full text, mbox):

From: Óscar Fuentes <ofv <at> wanadoo.es>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: 22739 <at> debbugs.gnu.org
Subject: Re: bug#22739: 25.0.50; Hangs on revertig buffer
Date: Sat, 20 Feb 2016 17:40:35 +0100

Eli Zaretskii <eliz <at> gnu.org> writes:

>> I'm afraid that the Emacs instance is not resumable anymore. After using
>> the "finish" trick on the previous message, I was unable to interrup
>> Emacs either with C-c on the gdb console (as I did the first time after
>> attaching gdb)
>
> Are you saying that just attaching GDB didn't stop Emacs, and you
> needed C-c to do that?

gdb attached correctly and used C-c on the gdb console to interrupt it
(or it was interrupted automatically when gdb attached to it, I can not
remember for sure), then I obtained the backtrace posted on this thread
and used `finish' to locate the frame that infloops. Obviously, at that
point Emacs was inflooping again and C-c was ignored. Then I sent a
SIGSTOP signal to emacs, which interrupted the process and was able to
obtain the contents of *Messages* from gdb.

>> nor C-z on the Emacs window (as DEBUG suggests.) I sent a SIGSTOP
>> signal to the Emacs process to gain access to the gdb command line
>> and now it can't resume.
>
> What do you mean by "cannot resume"?  What happens if you try?

After executing `c', gdb showed the signal handler. Same when repeated.

> What happens if you kill the GDB process, and then attach another GDB
> to Emacs?

Sorry, the Emacs session is gone now.

Next time I'll try to not use C-g and attach gdb right away.

Message #35 received at 22739 <at> debbugs.gnu.org (full text, mbox):

From: Óscar Fuentes <ofv <at> wanadoo.es>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: 22739 <at> debbugs.gnu.org
Subject: Re: bug#22739: 25.0.50; Hangs on revertig buffer
Date: Sun, 06 Mar 2016 02:28:29 +0100

It happened again, but this time the file was corrupt and contained 230
KB of garbage (all nulls except for the last few characters, which are
ASCII.)

The infloop was on bidi functions and it was necessary to SIGKILL the
process. SIGINT didn't work.

It is repeatable with emacs -Q.

In GNU/Linux you can create a file full of nulls with

$ truncate -s 100000 kk
$ emacs -Q kk

It seems that it is some type of quadratic (or worse) time complexity
with long lines, because with smaller sizes (100, 1000, ...) Emacs comes
back.

So it is the old problem with very long lines. We should try to bail out
when processing this type of lines takes too much time. From certain
point, simply refusing to display the line (or the entire file) is
preferable to freezing Emacs.

Message #38 received at 22739 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Óscar Fuentes <ofv <at> wanadoo.es>
Cc: 22739 <at> debbugs.gnu.org
Subject: Re: bug#22739: 25.0.50; Hangs on revertig buffer
Date: Sun, 06 Mar 2016 05:41:36 +0200

> From: Óscar Fuentes <ofv <at> wanadoo.es>
> Cc: 22739 <at> debbugs.gnu.org
> Date: Sun, 06 Mar 2016 02:28:29 +0100
> 
> So it is the old problem with very long lines. We should try to bail out
> when processing this type of lines takes too much time. From certain
> point, simply refusing to display the line (or the entire file) is
> preferable to freezing Emacs.

We don't have any mechanism in the display engine for "refusing to
display a line", AFAIK.

Message #41 received at 22739 <at> debbugs.gnu.org (full text, mbox):

From: Óscar Fuentes <ofv <at> wanadoo.es>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: 22739 <at> debbugs.gnu.org
Subject: Re: bug#22739: 25.0.50; Hangs on revertig buffer
Date: Sun, 06 Mar 2016 05:39:24 +0100

Eli Zaretskii <eliz <at> gnu.org> writes:

> We don't have any mechanism in the display engine for "refusing to
> display a line", AFAIK.

What about scanning the contents before entering display?

Message #44 received at 22739 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Óscar Fuentes <ofv <at> wanadoo.es>
Cc: 22739 <at> debbugs.gnu.org
Subject: Re: bug#22739: 25.0.50; Hangs on revertig buffer
Date: Sun, 06 Mar 2016 18:22:34 +0200

> From: Óscar Fuentes <ofv <at> wanadoo.es>
> Cc: 22739 <at> debbugs.gnu.org
> Date: Sun, 06 Mar 2016 05:39:24 +0100
> 
> Eli Zaretskii <eliz <at> gnu.org> writes:
> 
> > We don't have any mechanism in the display engine for "refusing to
> > display a line", AFAIK.
> 
> What about scanning the contents before entering display?

The problem is not with detecting a long line, the problem is what we
should do when we detect it.  If you just refrain from displaying it,
2 things will happen: (a) Emacs will start re-entering redisplay like
crazy, and (b) users will file bug reports about incorrect display.

So we'd need to display _something_ instead of that line which will
indicate the problem to the user.  This solves (b) above, but (a) is
still unsolved, and I don't know how to solve it in general.  Ideas
welcome.

Anyway, it turned out this is a regression wrt Emacs 24.5: a redisplay
optimization that targeted this particular use case was disabled as
part of fixing another display bug related to control characters.  I
have now partially re-enabled the optimization, which speeds up your
case by about 2 orders of magnitude.  Please take a look at the latest
emacs-25 branch.

Message #49 received at 22739-done <at> debbugs.gnu.org (full text, mbox):

From: Óscar Fuentes <ofv <at> wanadoo.es>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: 22739-done <at> debbugs.gnu.org
Subject: Re: bug#22739: 25.0.50; Hangs on revertig buffer
Date: Mon, 07 Mar 2016 01:29:57 +0100

Eli Zaretskii <eliz <at> gnu.org> writes:

I tried files of 3MB of NULLs and now Emacs handles them just fine.
Thank you.

I don't know what happened with the case that prompted this bug report.
I looked at the file and it was the usual ASCII text. Let's suppose that
what I saw was not the same content that Emacs read (that file is a log
and sometimes, by mistake, more than one process instance is launched,
each of them writing to the same log opened with w+.)

Since I can no longer freeze Emacs with files containing "reasonably"
long lines, I think that the severity of the issue decreased enough to
close this bug report.

Message #52 received at 22739 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Óscar Fuentes <ofv <at> wanadoo.es>
Cc: 22739 <at> debbugs.gnu.org
Subject: Re: bug#22739: 25.0.50; Hangs on revertig buffer
Date: Mon, 07 Mar 2016 18:32:13 +0200

> From: Óscar Fuentes <ofv <at> wanadoo.es>
> Cc: 22739-done <at> debbugs.gnu.org
> Date: Mon, 07 Mar 2016 01:29:57 +0100
> 
> Eli Zaretskii <eliz <at> gnu.org> writes:
> 
> I tried files of 3MB of NULLs and now Emacs handles them just fine.
> Thank you.

Thanks for testing.

> I don't know what happened with the case that prompted this bug report.
> I looked at the file and it was the usual ASCII text. Let's suppose that
> what I saw was not the same content that Emacs read (that file is a log
> and sometimes, by mistake, more than one process instance is launched,
> each of them writing to the same log opened with w+.)
> 
> Since I can no longer freeze Emacs with files containing "reasonably"
> long lines, I think that the severity of the issue decreased enough to
> close this bug report.

OK, thanks.  If this ever comes back, feel free to reopen, or file a
new bug.

Previous Next

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