Re: [i915] BUG: Bad page state in process Xorg

[Thomas Meyer <thomas@m3y3r.de>]

Am Samstag, den 30.11.2013, 08:55 -0500 schrieb Rob Clark:
> On Sat, Nov 30, 2013 at 3:33 AM, Thomas Meyer <thomas@m3y3r.de> wrote:
> > Am Montag, den 25.11.2013, 08:23 -0500 schrieb Rob Clark:
> >> oh, hmm.. are you importing buffers from i915?  It looks like this part:
> >
> > My computer has an i915 graphic card and I use an USB docking station,
> > that has an unconnected UDL device inside.
> >
> >>
> >> - if (obj->base.import_attach) {
> >> - drm_free_large(obj->pages);
> >> - obj->pages = NULL;
> >> - return;
> >> - }
> >>
> >> should not have been removed from udl_gem_put_pages()..
> >
> > Do you want me to test this? Can this please be fixed in the stable
> > tree?
> > I removed the UDL driver from my config for now, as I don't use it.
> 
> It would be nice if you could test that (I can send you a patch if you
> prefer)..  I don't actually have any UDL device to confirm.  If this
> fixes the problem, send the patch to stable as well should not be a
> problem.

I did a quick test and above code makes the bad page bug disappear.

> 
> BR,
> -R
> 
> 
> >       *
> > best regards
> > thomas
> >
> >>
> >> BR,
> >> -R
> >>
> >> On Mon, Nov 25, 2013 at 8:17 AM,  <thomas@m3y3r.de> wrote:
> >> > Hi,
> >> >
> >> > It turns out that this seems to be a bug in udl DRM driver.
> >> >
> >> > I bisected the problem to this patch:
> >> >
> >> > https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/drivers/gpu/drm/udl?id=5dc9e1e87229cb786a5bb58ddd0d60fee6eb4641
> >> >
> >> > With kind regards
> >> > Thomas
> >> >
> >> > Am 22.11.2013 17:18 schrieb Daniel Vetter <daniel@ffwll.ch>:
> >> >>
> >> >> On Fri, Nov 22, 2013 at 4:54 PM, Thomas Meyer <thomas@m3y3r.de> wrote:
> >> >> >> Am 22.11.2013 um 11:55 schrieb Daniel Vetter <daniel@ffwll.ch>:
> >> >> >>
> >> >> >> On Fri, Nov 22, 2013 at 11:36 AM, Dave Airlie <airlied@linux.ie> wrote:
> >> >> >>>> Hi,
> >> >> >>>
> >> >> >>> cc'ing mailing list,
> >> >> >>>
> >> >> >>> Daniel any ideas?
> >> >> >>
> >> >> >> Nope, not really :( And no ideas how to triage this further - if it
> >> >> >> takes 9 days to hit it eventually we'll have a real hard time. Or does
> >> >> >> this happen even after just a short X run?
> >> >> >
> >> >> > Seems to happen every time while stopping the x server. Also after a short run time.
> >> >> >
> >> >> > The current fedora 3.11 kernel doesn't show this bug. I'm using fedora 19, with a self compiled kernel.
> >> >> >
> >> >> > I did turn on config-debug-pagealloc but this didn't show any wrongness.
> >> >>
> >> >> In that case I think the bisect is the fastest way to insight - atm
> >> >> I'm really at loss what could be wrong here.
> >> >> -Daniel
> >> >> --
> >> >> Daniel Vetter
> >> >> Software Engineer, Intel Corporation
> >> >> +41 (0) 79 365 57 48 - http://blog.ffwll.ch
> >
> >
> >