bug-gnu-emacs
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

bug#21738: 25.0.50; eww freezes/crashes at times


From: Kaushal Modi
Subject: bug#21738: 25.0.50; eww freezes/crashes at times
Date: Fri, 23 Oct 2015 16:35:47 -0400

> You can type "kill" at the GDB prompt.

Thanks!

> The file etc/DEBUG has a section that explains how to find where Emacs
> is inflooping.  Can you try that technique and report the results?

Alright. I will work on that next week and post here when I get results.

By the way, I am now using emacs built without imagemagick. Now emacs
does not freeze in "emacs -Q" sessions.
But I got it to freeze (in a C-g quittable way) in my emacs using my config.

fci (fill column indicator mode is the culplrit!) -
http://www.emacswiki.org/emacs/fill-column-indicator.el

The bug is nasty.. probably in the package or my config (need to yet
figure that out).
If I have only the window with eww buffer in the whole frame then
everything is fine.
But (1) if I have two windows (or more) in a frame and
      (2) if one of the windows has a buffer with a major mode where I
have fci-mode enabled, and
     (3) I am loading eww in the other buffer, my eww freezes, luckily
doing C-g exits that state.

I took your advice and did

(setq debug-on-quit t)

On doing C-g, I get the below.

The bug, I mentioned, is nasty, because I don't enable fci-mode
globally; only in few prog modes, and definitely not in eww-mode. But
some sort of window walking is freezing emacs when probably fci tried
to do something in the eww buffer window. If I don't load the fci
package at all, eww loads http://www.braveclojure.com/basic-emacs fine
in my emacs with my config too! :)


Debugger entered--Lisp error: (quit)
  window-end(#<window 3 on *Backtrace*> updated)
  #[(w) "\301 !\302 \303\"B\207" [w window-start window-end updated]
4](#<window 3 on *Backtrace*>)
  mapcar(#[(w) "\301 !\302 \303\"B\207" [w window-start window-end
updated] 4] (#<window 3 on *Backtrace*>))
  fci-delete-unneeded()
  fci-redraw-frame()
  set-window-buffer(nil #<buffer  *temp*-559543>)
  shr-render-td-1((td ((class . "Body") (shr-td-cache-natural . 214)
(shr-td-cache-10-nil 209 214 1 (#("M->" 0 1 (face (variable-pitch
bold) shr-indentation nil) 1 3 (face (variable-pitch bold)))) 1 nil
nil)) "\n      " (strong nil "M->") "\n    ") 213 t)
  shr-render-td((td ((class . "Body") (shr-td-cache-natural . 214)
(shr-td-cache-10-nil 209 214 1 (#("M->" 0 1 (face (variable-pitch
bold) shr-indentation nil) 1 3 (face (variable-pitch bold)))) 1 nil
nil)) "\n      " (strong nil "M->") "\n    ") 213 t)


So, inspite of not yet finding a clean solution, I at least got that
web page loading in my regular emacs setup.

- no imagemagick
- no fci-mode





reply via email to

[Prev in Thread] Current Thread [Next in Thread]