On Thu, Oct 03, 2019 at 03:12:11PM +0100, Chris Wilson wrote: > Quoting Ville Syrjala (2019-10-03 15:02:31) > > From: Ville Syrjälä <ville.syrjala@xxxxxxxxxxxxxxx> > > > > The current "disable C3+" workaround for the delayed vblank > > irqs on i945gm no longer works. I'm not sure what changed, but > > now I need to also disable C2. I also got my hands on a i915gm > > machine that suffers from the same issue. > > > > After some furious poking of registers I managed to find a > > better workaround: The "Do not Turn off Core Render Clock in C > > states" bit. With that I no longer have to disable any C-states, > > and as a nice bonus the power cost is only ~1/4 of the > > "disable C3+" method (which mind you doesn't even work anymore, > > and so would have an even higher power cost if we made it work > > by also disabling C2). > > > > So let's throw out all the cpuidle/qos crap and just toggle > > the magic bit as needed. And we extend the workaround to cover > > i915gm as well. > > Nice discovery. ScratchPad0 suggests that it may have been a late > addition, there might be some chips out there that don't have the magic > bit. Working on most is better than broken on all. Yeah. And at least 100% of *my* broken machines now work ;) > > > Cc: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx> > > Signed-off-by: Ville Syrjälä <ville.syrjala@xxxxxxxxxxxxxxx> > > Empirical results rule, and I for one am much happier without needing > qos dma_latency, > Acked-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx> > -Chris -- Ville Syrjälä Intel _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/intel-gfx
