Le Mon, 10 Feb 2014 09:31:48 +0100,
Sascha Hauer <s.hauer@xxxxxxxxxxxxxx> a écrit :
> On Mon, Feb 10, 2014 at 09:28:54AM +0100, Eric Bénard wrote:
> > Hi Sascha,
> >
> > Le Mon, 10 Feb 2014 09:16:17 +0100,
> > Sascha Hauer <s.hauer@xxxxxxxxxxxxxx> a écrit :
> > > On Sat, Feb 08, 2014 at 03:16:10PM +0100, Eric Bénard wrote:
> > > > &usbotg {
> > > > - status = "okay";
> > > > + phy_type = "utmi";
> > > > + dr_mode = "host";
> > > > + status = "okay";
> > > > };
> > >
> > > Do we need a barebox specific option to specify dr_mode? I mean barebox
> > > cannot do otg, so we need to specify dr_mode to host or device, but with
> > > the above change Linux is limited to host mode aswell.
> > >
> > in chipidea-imx.c there is :
> >
> > mode = of_usb_get_dr_mode(ci->dev->device_node, NULL);
> >
> > switch (mode) {
> > case USB_DR_MODE_HOST:
> > default:
> > ci->mode = IMX_USB_MODE_HOST;
> > break;
> > case USB_DR_MODE_PERIPHERAL:
> > ci->mode = IMX_USB_MODE_DEVICE;
> > break;
> > }
> >
> > so we could omit the dr_mode=host as it will default to host.
>
> Then let's do this. The default-to-host beahaviour in barebox may not
> be desirable in all cases, but at least we do not propagate this
> behaviour to the kernel.
>
do you want a v2 or will you remove the line when applying ?
Eric
_______________________________________________
barebox mailing list
barebox@xxxxxxxxxxxxxxxxxxx
http://lists.infradead.org/mailman/listinfo/barebox
