[Bug 95472] [i915] Feature request: Add support for fencing for PRIME setups

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=95472

--- Comment #12 from Chris Wilson <chris at chris-wilson.co.uk> ---
(In reply to Mike Lothian from comment #11)
> I've bisected back to:
> 
> 6e934d43fe55156c68a2e55f1d51f182003b3b0b is the first bad commit
> commit 6e934d43fe55156c68a2e55f1d51f182003b3b0b
> Author: Chris Wilson <chris at chris-wilson.co.uk>
> Date:   Wed Jul 6 14:21:07 2016 +0100
> 
>     nonblocking-dispatch

Of course that's the key one for being able to wait on fences in the execbuf!

I've split that patch up and pushed the branch out to ~ickle/linux-2.6 #fence

Can you please test

commit c9941749bae4030bbadbe268916c1aa43bf0c130
Author: Chris Wilson <chris at chris-wilson.co.uk>
Date:   Fri Aug 5 09:01:21 2016 +0100

    drm/i915: Drive request submission through fence callbacks

which introduces the callback mechanics

and then

commit 9759f6e63e8fecbe5c1d251586cdfe713d20dce0
Author: Chris Wilson <chris at chris-wilson.co.uk>
Date:   Fri Aug 5 09:02:16 2016 +0100

    drm/i915: Nonblocking request submission

which is the first use in anger (and presumably where you start to see the
issue).

If you have your phone logged in at the time of the hang, can you get any
diagnostics? For convenience, you probably want to set panic-on-oops and
reboot-on-panic (with a timeout of 1 or 2s) in the kconfig.

-- 
You are receiving this mail because:
You are the assignee for the bug.
You are on the CC list for the bug.
You are the QA Contact for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/intel-gfx-bugs/attachments/20160805/f2330dec/attachment-0001.html>