[Intel-gfx] [PATCH] drm/i915: Mark vm_free_page() as a potential sleeper agent

[Chris Wilson]

Quoting Daniel Vetter (2017-11-09 10:46:41)
> On Thu, Nov 09, 2017 at 10:23:44AM +0000, Chris Wilson wrote:
> > Quoting Chris Wilson (2017-11-09 10:21:41)
> > > vm_free_page() may call down into set_pages_array_wb() (which itself
> > > sleeps, on x86 at least) but only if on !llc and the caches overflow.
> > > Since this is unlikely, we only rarely trigger the error in practice,
> > > and so to make CI detection of this sleeping bug possible we want to
> > > mark the common vm_free_page() as a potential sleep.
> > > 
> > 
> > References: https://bugs.freedesktop.org/show_bug.cgi?id=103638
> > > Suggested-by: Tvrtko Ursulin <tvrtko.ursulin at intel.com>
> > > Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
> > > Cc: Tvrtko Ursulin <tvrtko.ursulin at intel.com>
> > > Cc: Joonas Lahtinen <joonas.lahtinen at linux.intel.com>
> > > Cc: Matthew Auld <matthew.william.auld at gmail.com>
> > 
> > The real test ofc is whether this reports an error from CI.
> 
> Reviewed-by: Daniel Vetter <daniel.vetter at ffwll.ch>
> 
> But if it indeed blows up on CI I think it'd be good to fix first, then
> apply the more aggressive debug check. Otherwise CI and Marta will be on
> fire :-)

It got a few more hits, but not across the board as I was hoping. The
next place is i915_vma_unbind(). That is always a potential sleeper due
to its implicit wait on the GPU.
-Chris