On Tue, Jan 30, 2018 at 01:22:18PM +0200, Andy Shevchenko wrote: > On Tue, 2018-01-30 at 16:14 +1100, Johan Hovold wrote: > > On Thu, Jan 25, 2018 at 02:08:35PM +0200, Andy Shevchenko wrote: > > > On Wed, 2018-01-24 at 21:35 +0000, David Howells wrote: > > > > Andy Shevchenko <andriy.shevchenko@xxxxxxxxxxxxxxx> wrote: > > > > > Not tested at all. > > > > Your patch doesn't say why or whether this causes userspace API > > > > breakage. Can > > > > you please include a patch description? > > > > > > This feature is considered deprecated for many years already. As you > > > can > > > see by the change done here, there are only few drivers that still > > > support that. > > > > You still need to provide a proper commit message. > > It needs more, of course, that's why it's marked as WIP. > > I stopped at this point to gather comments like yours, which is very > useful. I meant that you should provide a proper commit message also for an RFC. So that people have something to comment on without forcing every reviewer to try and figure out the intention, impact, etc, for themselves. Perhaps I should have refrained from doing just that... > > And every serial driver currently supports this through serial core. > > What about as first approximation to get rid of custom handling in some > drivers? Not sure I understand that sentence; do you suggest removing the SPD_ASYNC handling from the non-serial-core drivers as a first step? > Also there are few drivers where some leftover definitions or alike can > be removed. I'm sure there a lot of bit rot in this area. > > Also it looks like you are changing the default speed for a couple of > > ftdi devices, which would need to be addressed. > > I *think* I got it and changed code accordingly (see if (custom_divisor) > condition). Of course it needs to be tested properly. Yeah, you're right, you seem to have taken care of that. But this really needs to be broken out in its own patch with a mention in the commit message etc. Johan -- 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
