On Wed, Dec 08, 2010 at 06:56:45PM +0300, Sergei Shtylyov wrote:
Hello.
Mian Yousaf Kaukab wrote:
You probably copied this from omap2430.c?
Yes.
This is incorrect way of inplemeting this method.
Ok, I was letting the the connector chose the mode.
You are explicitly passed the 'mode' argument, you can't choose it
at connector's discretion.
Ideally that's how OTG should work. First define roles based on ID
state, then later based on user interaction aka HNP.
The set_mode() is mostly used for debugging where you can forcefully put
MUSB in a defined mode.
It looks like da8xx is overriding the comparators and so. Is it what
enforcing meant here?
Overriding VBUS comparator is a DA8xx specific side effect of
overriding the ID pin -- you just can't do it otherwise.
Or the driver should only allow the
desired mode, but should wait for the proper conditions as well. For
example, if desired mode is MUSB_PERIPHERAL,
wait till VBUS is valid and in the mean time ignore ID ground event
if you get it.
No, you should not wait for anything as far as I understand. Let
Felipe clarify this if he wants...
No, don't wait anything. You can also send the glue layer without the
set_mode method for now and implement it later if you want. It's not a
problem.
--
balbi
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
