Re: [PATCH 3/6] ir-kbd-i2c: Switch to the new-style device binding model

[Janne Grunau <j@jannau.net>]

On Sun, Apr 05, 2009 at 01:39:33PM -0400, Andy Walls wrote:
> On Sun, 2009-04-05 at 16:37 +0200, Janne Grunau wrote:
> > 
> > I would guess that it won't work. There is an effort to merge lirc. It's
> > currently stalled though.
> 
> Perhaps you and Jarrod and Christopher have already discussed this,
> but...
> 
> Instead of trying to push all of the LIRC kernel components through in
> one big patch set, perhaps it would be easier to just get the lirc_dev
> and any other needed infrastructure components in first.
> 
> If one focuses on satisfying the LKML comments to lirc_dev and the
> Makefile to get that kernel module in the kernel, then, at least for
> video card hosted IR devices, there is an infrastructure to which to
> hook new or rewritten i2c IR driver modules.

I guess lkml would NAK patches adding infrastructure only bits but we
will probably for the next patchset concentrate on a few lirc drivers.
Christopher doesn't participate in the merge attempt.

> >  A git tree is available at
> > 
> > git://git.wilsonet.com/linux-2.6-lirc.git
> > 
> > Jared Wilson and I were working on it (mainly last september). Since the
> > IR on the HD PVR is also driven by the same zilog chip as on other
> > hauppauge devices I'll take of lirc_zilog. Help converting the i2c
> > drivers to the new i2c model is welcome. General cleanup of lirc to make
> > it ready for mainline is of course wellcome too.
> 
> I can help with this.  I'm mainly concerned with lirc_dev, lirc_i2c (for
> Rx only use of the zilog at 0x71), lirc_zilog, and lirc_mceusb2.  That's
> because, of course, I have devices that use those modules. :)

I have devices for lirc_zilog (which should probably be merged with
lirc_i2c) and lirc serial. Jarod has at least mce usb and imon devices.
That are probably the devices we'll concentrate on the next submission.

> lirc_dev and the API header would be my first priority, if you need
> help.  Did anyone consolidate all the comments from the LKML on Jarrod's
> patch submission?

no and I lost track which comments were already handled.

Janne