lkml.org 
[lkml]   [2016]   [Nov]   [19]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
SubjectRE: [char-misc-next 4/4] mei: bus: enable non-blocking RX
Date

>
> On Fri, Nov 18, 2016 at 07:30:25PM +0000, Winkler, Tomas wrote:
> >
> > >
> > > On Thu, Nov 17, 2016 at 04:22:24PM +0000, Winkler, Tomas wrote:
> > > > > Just make a new function mei_cldev_recv_async() and then call a
> > > > > local, static function, that does the work with the correct flag
> > > > > set. That way the developer always knows exactly what is going on.
> > > >
> > > > We can do a wrapper, but _async() is not proper here maybe
> > > > _nonblock(),
> > >
> > > Yes, I just guessed at the name :)
> >
> > Understood, any how I believe that we should keep patch as is, there
> > is no one API _nonblock() function in the whole kernel unlike
> > _async(), nonblock is always passed in as a flag or Boolean.
>
> And I'll still argue that this is a horrible api and you can do better.
> No need to duplicate the errors of our childhood :)

Legacy is not always wrong it as is consistency, and the balance where to put change of behavior in arguments or function names is gentle one.
I'm not sure we want to end with Java function names, either.
bool doesShorterNameExistThatEquallyConvaysTheBehaviorOfTheMethod()

I'll respin the patch as it's not really worth the argument in this particular case, but I'm not convinced.

Thanks
Tomas



>
> thanks,
>
> greg k-h

\
 
 \ /
  Last update: 2016-11-19 11:43    [W:0.054 / U:0.316 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site