On Tue, Jul 23, 2019 at 07:48:49PM +0200, Evgeny Kolesnikov wrote: > On 23/07/2019 03:56, Andrew Lunn wrote: > >On Mon, Jul 22, 2019 at 09:53:00PM +0200, Evgeny Kolesnikov wrote: > >> > >>The difference between uart-poweroff and qnap-poweroff is small, but important: > >>uart-poweroff is able to send to an MCU a command of arbitrary length, and the command > >>itself is defined in a DTS file for a specific device/board, thus making this driver > >>applicable to wider range of devices. > > > >There is a lot of replicated code here, and in the original > >qnap-poweroff.c driver. Please consolidate it by extending the current > >driver. It should be easy to add a new compatible string, and turn > >power_off_cfg.cmd into an array. > > Hi, Andrew. > > I've considered extending qnap driver, but I have some doubts about this > approach. > > First of all there is only a poweroff counterpart. As there is no > qnap-restart driver, what should I do with uart-restart? Is it OK to have > xxx-restart-poweroff driver (never saw anything like that)? Hi Evgeny There are a few options. You can refactor all the code into a library and small drivers which wrap around the library. Or you can make the driver handle both, using the compatible string to determine which it should do. > While I can add cmd as a parameter to qnap driver (having it converted > into an array) it should be optional as original qnap relies on two > hardcoded values for its devices. That is not what i meant. You can make the current code more generic by changing the single byte in power_off_cfg to an array. DT should describe the hardware, not bytes you poke into registers. So it is perfectly valid to have the bytes hard coded in the driver. Andrew
