On Fri, Nov 07, 2014 at 04:14:23PM -0800, Doug Anderson wrote: > Russell, > I guess I'm still confused. My patch continues to be about orphans > and I don't see the bug you are pointing to. Ah, in which case, the question changes: how can an orphaned clock be succesfully prepared and enabled? Drivers expect that a clock for which clk_enable() has returned successfully _will_ at that point be supplying the clock. If we don't yet know it's parent, how do we know that it will be supplying that clock? What about a driver calling clk_set_rate() on an orphaned clock? >From what I can see (__clk_reparent will re-set the child's clock when reparenting) having a driver able to claim an orphaned clock, let alone prepare and enable it, looks rather buggy to me. -- FTTC broadband for 0.8mile line: currently at 9.5Mbps down 400kbps up according to speedtest.net.
