[Intel-gfx] [PATCH] tests/gem_eio: Resilience against "hanging too fast"

[Daniel Vetter]

On Wed, Nov 25, 2015 at 04:28:02PM +0000, Chris Wilson wrote:
> On Wed, Nov 25, 2015 at 05:19:24PM +0100, Daniel Vetter wrote:
> > Since $debugfs/i915_wedged restores a wedged gpu by using a normal gpu
> > hang we need to be careful to not run into the "hanging too fast
> > check":
> 
> That's not how it works. It restores the GPU by triggering a manual
> reset.

Maybe in your tree, but here it goes through the full hang recovery. And
the gem_eio default context _does_ get banned without this. Maybe it's a
bug in the ban logic, but it didn't look like that. After all
debugfs/i915_wedged is a hack for developers - by that time userspace has
already either died or fallen back to sw rendering anyway. So I figured no
point in fixing this in the kernel with special cases.
-Daniel
-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch