[PATCH 3/3] mx51: Move OTG initialisation for all boards to a single file
s.hauer at pengutronix.de
Mon Oct 11 09:10:09 EDT 2010
On Mon, Oct 11, 2010 at 01:34:30PM +0300, Amit Kucheria wrote:
> On Thu, Oct 7, 2010 at 10:48 AM, Sascha Hauer <s.hauer at pengutronix.de> wrote:
> > Added Daniel to Cc as he might say something here aswell.
> > I have recently looked at the way the USB phy settings are handled on
> > i.MX and it's coming to its limits. Currently the phy settings are coded
> > into generic flags, passed to the usb driver which then calls
> > mxc_initialize_usb_hw() which dispatches the different SoCs and translates
> > the generic flags back into SoC specific ones. While this was doable
> > until now it fails badly on i.MX51 as we see here.
> > The phy settings are totally static to a given board, so we could simply
> > let the board call a mx51_init_usb_phy() which gets a pointer to a
> > struct with all relevant register settings. And we could pass this
> > function SoC specific register settings without having to encode them
> > into generic flags and decode them again later.
> > Sascha
> Do you think something like this will be better? It looks even more
> scattered to me, the usb configuration information is scattered across
> platform_data and phy_data.
My problem is not that it's scattered around platform_data and phy_data,
but that it's not possible to add proper i.MX51 support without
introducing a initialize_otg_port which wont' be flexible enough for
> Or did you want to get rid of platform_data->init too from the driver?
No, I think we can't do this without breaking at least some boards. I
found out that on some boards we have to configure the iomux pins and
the rest of the USB hardware in a short time frame, otherwise we get
Pengutronix e.K. | |
Industrial Linux Solutions | http://www.pengutronix.de/ |
Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0 |
Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 |
More information about the linux-arm-kernel