On Mon, Jan 11, 2016 at 06:42:49PM +0000, john.c.harri...@intel.com wrote: > From: John Harrison <john.c.harri...@intel.com> > > When requesting that all GPU work is completed, it is now necessary to > get the scheduler involved in order to flush out work that queued and > not yet submitted.
But why is this needed over and above waiting on the request? Why do we actually need to flush the work as the scheduler will get to the request eventually (one hopes!)? Why the priority bump? That would not be intended for either the idle (wait until everybody has finished) or throttling. -Chris -- Chris Wilson, Intel Open Source Technology Centre _______________________________________________ Intel-gfx mailing list Intel-gfx@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/intel-gfx
