Quoting Bjorn Andersson (2019-12-26 18:26:52) > On Mon 23 Dec 18:16 PST 2019, Stephen Boyd wrote: > > > Quoting Niklas Cassel (2019-12-20 09:56:16) > > > On Wed, Dec 18, 2019 at 10:23:39PM -0800, Stephen Boyd wrote: > > > > This is odd. The clks could be registered with of_clk_hw_register() but > > > > then we lose the device provider information. Maybe we should search up > > > > one level to the parent node and if that has a DT node but the > > > > clk controller device doesn't we should use that instead? > > > > > > Hello Stephen, > > > > > > Having this in the clk core is totally fine with me, > > > since it solves my problem. > > > > > > Will you cook up a patch, or do you want me to do it? > > > > > > > Can you try the patch I appended to my previous mail? I can write > > something up more proper later this week. > > > > Unfortunately we have clocks with no dev, so this fail as below. Adding > a second check for dev != NULL to your oneliner works fine though. > > I.e. this ugly hack works fine: > core->of_node = np ? : (dev ? dev_of_node(dev->parent) : NULL); > Ok, thanks for testing!
