On Thu, Dec 03, 2009 at 09:27:18PM +0000, Matthew Garrett wrote: > On Thu, Dec 03, 2009 at 11:20:12AM +0100, Jerome Glisse wrote: > > > Why do you need that ? PM should happen in a callback from a timer > > (we might want to use the HW timer for that) ie from some function > > called by the IRQ/MSI handler. In the PM function we should take > > all lock to assure that we have exclusive hw access, thus we need > > to take cs lock but also the lock which is protecting modesetting > > (iirc there is one). > > If a modeset results in a second display being activated, we need to > clock back up. vblank-driven reclocking doesn't work so well with > multihead. >
I did take this into account in my description on modeset you recompute minimal valid clock and schedule the reclock before doing the modeset. Missing part is computing minimal GPU/Memory clock. Cheers, Jerome ------------------------------------------------------------------------------ Join us December 9, 2009 for the Red Hat Virtual Experience, a free event focused on virtualization and cloud computing. Attend in-depth sessions from your desk. Your couch. Anywhere. http://p.sf.net/sfu/redhat-sfdev2dev -- _______________________________________________ Dri-devel mailing list Dri-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
