[Intel-gfx] [PATCH 1/6] drm/i915: hangcheck robustification

[Chris Wilson]

On Wed, 19 Oct 2011 08:02:57 -0700, Ben Widawsky <ben at bwidawsk.net> wrote:
> On Wed, 19 Oct 2011 12:32:25 +0100
> Chris Wilson <chris at chris-wilson.co.uk> wrote:
> > NAK: This failed to detect a hang, leaving my box frozen. I suspect that
> > the value of INSTDONE was fluctuating on the render ring even though we
> > had now requests pending and so could assume that it was idle.
> > -Chris
> > 
> How is that different than the previous behavior? We checked instdone on
> the render ring before this patch too.

As mentioned on irc, that is probably a wild goose chase. This is the
first false negative I have observed with hangcheck, so I blame the
robustification patch (as being a self-grandeous patch ;-). A false
negative has much worse consequences than a false positive - so we need
to be sure we can avoid such at any cost.
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre