On Tue, Dec 22, 2015 at 11:58:17AM +0200, Andy Shevchenko wrote: > On Tue, Dec 22, 2015 at 6:27 AM, Sudip Mukherjee > <sudipm.mukherjee@xxxxxxxxx> wrote: > > On Mon, Dec 21, 2015 at 08:28:51PM +0200, Andy Shevchenko wrote: > >> On Mon, Dec 21, 2015 at 5:19 PM, Sudip Mukherjee > >> <sudipm.mukherjee@xxxxxxxxx> wrote: > >> > > >> There are at least two approaches: > >> - use 8250_pci, etc as a library (see example: 8250_mid.c) > >> - force 8250_pci to use external libraries in some cases (seems your approach) > > Third one btw is to blow up the 8250_pci. (This actually was the main > reason why we chose separate driver approach in our case). > > >> Of course better to gather maintainer's opinion first. > > > > Greg, can you please give some idea here about the best way to approach... > > I personally think, having it as a module with the minor changes that > > Alan and Andy has suggested is the best approach. > > > The only downside is > > that the module gets loaded even if the device is not there. > > How is that? Alan explained that in https://lkml.org/lkml/2015/12/20/103 Quoting from his mail "you reference the methods in it so it will always be dragged in". And I wanted to verify that so I tested today morning after removing the card from my local system and after booting I saw having 8250_gpi loaded. regards sudip -- To unsubscribe from this list: send the line "unsubscribe linux-serial" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
