Hi! On 08/17/2012 03:14 PM, Sebastian Andrzej Siewior wrote: > On 08/16/2012 11:31 AM, Roland Stigge wrote: >> USB_ISP1301 already depends on I2C. >> >> Maybe we should rather let USB_LPC32XX "depends on USB_ISP1301" instead >> of "select USB_ISP1301"? > > And your OHCI select has the same problem. So we make both things > depend on the transceiver instead of selecting it right? > > But now that I've seen the transceiver driver it is probably the > smallest one. > > Further questions: > - why is there "normal_i2c" ? > - why is that thing not doing anything at all? Good catch, it is a leftover from factoring out the phy driver. > Couldn't you just move the transiver logic out of the UDC? And fail in > probe if it does not find one and get probed later again? I think we do > something like that in the device case where the interrupt controller > is not yet probed but the currently probed device requires it. > > Felipe, do we have a phy framework yet or is it still in work? When I created the isp1301 driver to be used by ohci and udc, I made a very minimal API, no complex init functions since the drivers do it differently. However, as soon as we have a phy framework, I can adjust this. When Felipe decides how to proceed with the UDC's controller struct (see my proposal), I can provide an updated patch series, including the normal_i2c cleanup. Thanks, Roland -- 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
