Re: [PATCH 04/62] drm/i915: Restore waitboost credit to the synchronous waiter

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Wed, Jun 08, 2016 at 11:04:57AM +0200, Daniel Vetter wrote:
> On Fri, Jun 03, 2016 at 05:36:29PM +0100, Chris Wilson wrote:
> > Ideally, we want to automagically have the GPU respond to the
> > instantaneous load by reclocking itself. However, reclocking occurs
> > relatively slowly, and to the client waiting for a result from the GPU,
> > too late. To compensate and reduce the client latency, we allow the
> > first wait from a client to boost the GPU clocks to maximum. This
> > overcomes the lag in autoreclocking, at the expense of forcing the GPU
> > clocks too high. So to offset the excessive power usage, we currently
> > allow a client to only boost the clocks once before we detect the GPU
> > is idle again. This works reasonably for say the first frame in a
> > benchmark, but for many more synchronous workloads (like OpenCL) we find
> > the GPU clocks remain too low. By noting a wait which would idle the GPU
> > (i.e. we just waited upon the last known request), we can give that
> > client the idle boost credit (for their next wait) without the 100ms
> > delay required for us to detect the GPU idle state. The intention is to
> > boost clients that are stalling in the process of feeding the GPU more
> > work (and who in doing so let the GPU idle), without granting boost
> > credits to clients that are throttling themselves (such as compositors).
> > 
> > Signed-off-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
> > Cc: "Zou, Nanhai" <nanhai.zou@xxxxxxxxx>
> > Cc: Jesse Barnes <jbarnes@xxxxxxxxxxxxxxxx>
> > Reviewed-by: Jesse Barnes <jbarnes@xxxxxxxxxxxxxxxx>
> 
> I wonder a bit what will happen here for workloads that flip-flop between
> engines, since you check for last request on a given engine. But maybe in
> the future we'll get clock domains per engine ;-)

We disable RPS boosting for inter ring synchronisation, so only if the
client does submit(RCS); submit(BCS); wait(RCS); wait(BCS) would it get
two bites at the cherry. That still falls under the notion of allowed
client behaviour as the second wait is presumably stalling the client
from submitting more work.

s/GPU idle/engine idle/ + tweaks
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/intel-gfx




[Index of Archives]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]
  Powered by Linux