From mboxrd@z Thu Jan  1 00:00:00 1970
From: cpollock@embarqmail.com (Chris)
Date: Mon, 15 Dec 2014 15:17:32 -0600
Subject: kernel: [154858.820009] [drm:i915_hangcheck_elapsed] *ERROR*
	Hangcheck timer elapsed.
In-Reply-To: <241246.1418677915@turing-police.cc.vt.edu>
References: <1418652133.24586.11.camel@embarqmail.com>
	<20141215165827.GA2155@kroah.com>
	<1418666367.2852.1.camel@embarqmail.com>
	<20141215180154.GA4962@kroah.com>
	<233694.1418668997@turing-police.cc.vt.edu>
	<1418674772.2852.5.camel@embarqmail.com>
	<241246.1418677915@turing-police.cc.vt.edu>
Message-ID: <1418678252.2852.7.camel@embarqmail.com>
To: kernelnewbies@lists.kernelnewbies.org
List-Id: kernelnewbies.lists.kernelnewbies.org

On Mon, 2014-12-15 at 16:11 -0500, Valdis.Kletnieks at vt.edu wrote:
> On Mon, 15 Dec 2014 14:19:32 -0600, Chris said:
> > On Mon, 2014-12-15 at 13:43 -0500, Valdis.Kletnieks at vt.edu wrote:
> 
> > > I'm pretty sure that's a timer for the i915 driver that detects if the
> > > GPU has gone into an infinite loop due to (usually) buggy programming from
> > > the operating system....
> > 
> > Thank you Valdis as far as the OS we're talking about Ubuntu not the
> > bios correct?
> 
> More specifically, the i915 is actually a fully programmable CPU in its own
> right, but it's a crazy morass of interrupts and circular buggers
> and race conditions.  The operating system (Ubuntu or what have you) actually
> send the i915 a program of what to display, and the i915 goes off and does it.
> Of course, missing a single interrupt or an off-by-one error in a circular
> buffer will cause the i915 to go off the rails and usually get hung in a hard
> loop.
> 
> (It doesn't help that the i915 has a ton of wonky restrictions of the form "you
> can do this during a vertical retrace period, but only during the first half
> of a horizontal retrace".  And yes, I'm pretty sure it actually enforces
> vertical and horizontal retrace timings on LCDs.  Silly, huh? ;)

I believe I understand Valdis, what kind of actions would need to be
taken to track this down, if it's possible, or to simply fix the problem
once and for all?

-- 
Chris
KeyID 0xE372A7DA98E6705C
31.11?N 97.89?W (Elev. 1092 ft)
15:16:18 up 3:21, 3 users, load average: 0.17, 0.19, 0.18
Ubuntu 14.04.1 LTS, kernel 3.13.0-43-generic