Re: [PATCH v2 1/3] drm/i915: Drop racy markup of missed-irqs from idle-worker

[Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>]

On 21/07/16 11:10, Chris Wilson wrote:
> On Thu, Jul 21, 2016 at 10:58:05AM +0100, Tvrtko Ursulin wrote:
>>
>> On 21/07/16 07:57, Chris Wilson wrote:
>>> During the idle-worker we disable the hangcheck and so kick any waiters
>>> that should have been completed (since the GPU is now idle). Unlike the
>>> hangcheck, we do not take any care to avoid the race between the irq
>>> handler and ourselves, and so it is possible for us to declare a missed
>>> interrupt even as the bottom-half is being scheduled to run. Let's
>>> ignore this race to stop a potential false-positive error.
>>
>> If the bottom half is scheduled to run then then..
>>
>>> References: https://bugs.freedesktop.org/show_bug.cgi?id=96974
>>> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
>>> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
>>> Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
>>> ---
>>>   drivers/gpu/drm/i915/i915_gem.c | 7 +++----
>>>   1 file changed, 3 insertions(+), 4 deletions(-)
>>>
>>> diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
>>> index 40047eb48826..9e826585edb2 100644
>>> --- a/drivers/gpu/drm/i915/i915_gem.c
>>> +++ b/drivers/gpu/drm/i915/i915_gem.c
>>> @@ -2706,10 +2706,9 @@ i915_gem_idle_work_handler(struct work_struct *work)
>>>   	rearm_hangcheck = false;
>>>
>>>   	stuck_engines = intel_kick_waiters(dev_priv);
>>
>> ... this will not return a stucked engine since the there is a bh
>> task assigned all until the bh exits.
>
> It reports if it wakes up a waiter on any engine. If the bh is already
> running, we cannot know if it has missed the seqno update. If it isn't
> running yet, we cannot know if it is about to be run.

Oh I read the logic as completely opposite than what it is.

Since the idle worker runs 100ms after last retirement, that would mean 
a really slow waiter or what?

Regards,

Tvrtko

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx