Quoting Uwe Kleine-König (2021-11-01 15:02:09) > On Mon, Nov 01, 2021 at 02:22:33PM +0000, Kieran Bingham wrote: > > Quoting Uwe Kleine-K�nig (2021-10-26 20:40:10) > > > If disabling the hardware fails the driver propagates the error code to > > > the i2c core. However this only results in a generic error message; the > > > device still disappears. > > > > > > So instead emit a message that better describes the actual problem than > > > the i2c core's "remove failed (ESOMETHING), will be ignored" and return > > > 0 to suppress the generic message. > > > > You almost caught me out. I was going to say, this changes the > > behaviour of the return code. But It's intentional ;-) > > > > It's still a bit concerning, /not/ returning an error on an error - but > > as it's not going to prevent removal, and it won't add further (helpful) > > diagnosticts), maybe it makes sense. > > > > My only concern would be if it might be better to keep the return code, > > so that the core frameworks know about the issue at least. > > The long term goal is to make the remove callback return void. For that > change I want all implementations to remove 0 to make the change easy to > review. Thanks for the update, that makes it much clearer indeed. And will make it clear that there's no use for a return code. -- Kieran > Best regards and thanks for your thoughts, > Uwe > > -- > Pengutronix e.K. | Uwe Kleine-K�nig | > Industrial Linux Solutions | https://www.pengutronix.de/ |
