== Series Details ==
Series: drm/i915/execlists: Avoid false matching of preemption when not enabled
(rev2)
URL : https://patchwork.freedesktop.org/series/37232/
State : failure
== Summary ==
Series 37232v2 drm/i915/execlists: Avoid false matching of preemption when not
enabled
https://patch
If we remove some hardcoded assumptions about the preempt context having
a fixed id, reserved from use by normal user contexts, we may only
allocate the i915_gem_context when required. Then the subsequent
decisions on using preemption reduce to having the preempt context
available.
Signed-off-by:
Quoting Chris Wilson (2018-01-27 18:12:00)
> If we haven't enabled preemption for execlists, the PREEMPT_ID may be
> assigned to an ordinary user context, for which we should not falsely
> declare preemption complete!
Actually we are always allocating the preempt context. I had thought I
had follo
== Series Details ==
Series: drm/i915/execlists: Avoid false matching of preemption when not enabled
URL : https://patchwork.freedesktop.org/series/37232/
State : failure
== Summary ==
Series 37232v1 drm/i915/execlists: Avoid false matching of preemption when not
enabled
https://patchwork.fre
If we haven't enabled preemption for execlists, the PREEMPT_ID may be
assigned to an ordinary user context, for which we should not falsely
declare preemption complete!
If we record the id assigned to the preempt context (or an invalid one
if preemption is disabled), then we can forgo hardcoding a
Hi,
On 26-01-18 17:38, Ville Syrjälä wrote:
On Fri, Jan 26, 2018 at 08:52:07AM +0100, Hans de Goede wrote:
So far models of the Dell Venue 8 Pro, with a panel with MIPI panel
index = 3, one of which has been kindly provided to me by Jan Brummer,
where not working with the i915 driver, giving a
On Fri, 26 Jan 2018, Dhinakaran Pandiyan wrote:
> Add a macro to check for a bit offset in a DPCD reg, use this macro to
> eliminate three functions and a local.
IMO less readable with this change.
BR,
Jani.
>
> Cc: Rodrigo Vivi
> Signed-off-by: Dhinakaran Pandiyan
> ---
> drivers/gpu/drm/i
On Fri, 26 Jan 2018, Chris Wilson wrote:
> Quoting David Weinehall (2018-01-26 15:53:53)
>> On Fri, Jan 26, 2018 at 04:48:05PM +0200, Jani Nikula wrote:
>> > Currently pch_id gets assigned also when there's no pch. It doesn't look
>> > like it makes a difference, but do the right thing anyway.
>>
On Fri, 26 Jan 2018, Rodrigo Vivi wrote:
> On Fri, Jan 26, 2018 at 10:12:00AM +, Jani Nikula wrote:
>> On Thu, 25 Jan 2018, Rodrigo Vivi wrote:
>> > The only difference is that this SKUs has the full
>> > Port A/E split named as Port F.
>> >
>> > But since SKUs differences don't matter on the
