GNU bug report logs - #22305
24.5; permanent 100% CPU load for no apparent reason

Previous Next

Full log

Message #8 received at 22305 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: "Roland Winkler" <winkler <at> gnu.org>
Cc: 22305 <at> debbugs.gnu.org
Subject: Re: bug#22305: 24.5; permanent 100% CPU load for no apparent reason
Date: Mon, 04 Jan 2016 19:16:46 +0200

> Date: Mon, 4 Jan 2016 10:48:02 -0600
> From: "Roland Winkler" <winkler <at> gnu.org>
> 
> Lately emacs gives me some strange problem, where according to `top'
> the CPU load due to the emacs process goes up to a permanent value
> of 100% (and I notice this because the laptop fan starts spinning
> loudly), whereas according to `ps' the emacs process continues to
> give rise to only a marginal CPU load.  While this happens, emacs
> responds normally, so it appears that emacs is not trying to do
> anything unusual in the background.

Not necessarily.  AFAIR, 'top' shows the CPU usage of each program in
terms of processor execution units.  IOW, 100% means Emacs is pegging
a single execution unit.  How many execution units (cores) do you have
on that system?

> What can I do to pin this down further?  So far, I have not noticed
> anything that triggers this behavior reproducibly and my workaround
> has been to restart emacs.  But that's not nice for a recurring
> problem (about once per day).

When this happens, attach a debugger and show us the backtrace,
including the Lisp backtrace.

Previous Next

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