> > > -----Original Message----- > > From: Felipe Balbi [mailto:balbi@xxxxxx] > > Sent: Saturday, March 01, 2014 12:51 AM > > To: Chanho Park > > Cc: linux-usb@xxxxxxxxxxxxxxx; Felipe Balbi; Greg Kroah-Hartman > > Subject: Re: [PATCH] usb: phy: move phy driver up for early > > initialization > > > > On Fri, Feb 28, 2014 at 02:50:07PM +0900, Chanho Park wrote: > > > This patch moves initialization order of the phy driver ahead of the > > > dwc3 driver. The dwc3 driver needs phy driver when probing. > > > Of cource, the driver can be initialized later after the phy driver > > > is initialized. For efficiency, however, it would be better to > > > change the initialization order. > > > > > > Cc: Felipe Balbi <balbi@xxxxxx> > > > Cc: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx> > > > Signed-off-by: Chanho Park <chanho61.park@xxxxxxxxxxx> > > > > NAK, every now and again we have a patch like this and it's just > > nonsense. All drivers should already be using EPROBE_DEFER by now. > > Please assume the case we select a gadget driver in compile time not as > module. > In that case, probing of the gadget driver is failed and cannot try it > again. > (Of course, it is not a problem if we use the gadget driver as module.) > How can I address this problem? > We have discussed before, but without a good solution, in fact, not only phy, not else regulator, etc, all will be possible to be deferred by probe. http://marc.info/?l=linux-usb&m=137991612311893&w=2 Peter -- 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
