On Wed, 21 May 2014 20:54:03 +0300 Ville Syrjälä <ville.syrjala@xxxxxxxxxxxxxxx> wrote: > + if (power_well_id == PUNIT_POWER_WELL_DPIO_CMN_BC && enable) > + deassert cmnreset > } > > > If those measures aren't enough, then I think this patch would be OK. > But right now I have this nagging feeling that we may have shot > ourselves in the foot and we're about to pile workarounds on top of our > own bug. I'll add these changes, with the additon that we need to do an actual toggle at boot and init. Simply asserting in suspend or gate and then de-asserting isn't enough to re-calibrate, since we tried that on the Chromebooks across suspend/resume and it wasn't enough to get things going again after the chip had lost power. Across simple power well transitions it ought to be enough though, and that's easy enough to verify in testing. -- Jesse Barnes, Intel Open Source Technology Center _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx http://lists.freedesktop.org/mailman/listinfo/intel-gfx
