On Fri, Nov 06, 2015 at 09:29:13AM +0000, Lee Jones wrote:
On Wed, 04 Nov 2015, Robert Jarzmik wrote:
Charles Keepax ckeepax@opensource.wolfsonmicro.com writes:
On Wed, Nov 04, 2015 at 06:12:44PM +0100, Robert Jarzmik wrote:
The Wolfson WM9713 provides 8 GPIOs. If the gpiolib is compiled in the kernel, declare a gpio chip.
Signed-off-by: Robert Jarzmik robert.jarzmik@free.fr
You should probably make a seperate driver within GPIO for this and then tie the two together, using an MFD. I appreciate that is more work but it is likely a nicer solution overall.
I'd like to first have a confirmation from :
- Mark (Brown)
- and Lee (Jones)
The confirmation I'm looking for states that :
- the wm9713 should have a part in the mfd tree
- the gpio part should be in drivers/gpio
- the sound soc codecs will remain as is
- if the future driver/mfd/wm9713.c is technically sound, it will be accepted
I remember at least one example where the MFD approach was rejected from mfd tree for pxa gpios, so I won't work unless I have a confirmation from both maintainers.
I have no idea what you're talking about. Context please?
Apologies Lee, we are discussing a patch that adds a GPIO driver into an AC97 CODEC. I had suggested that perhaps we should put the GPIO driver as a seperate driver under GPIO and link the two with an MFD. But Mark has already replied in the thread to say that he doesn't think that will be necessary. Although he did raise some concerns that perhaps it could be done more generally as it should apply to other AC97 CODECs as well.
So I think you can probably safely ignore this for now, sorry for the noise.
Thanks, Charles