Hi, > From: Rob Herring > Sent: Wednesday, June 29, 2016 5:57 AM > > On Mon, Jun 27, 2016 at 09:09:19PM +0900, Yoshihiro Shimoda wrote: > > This patch uses devm_usb_get_phy_by_phandle() instead of usb_get_phy() > > for device tree environment. This change is not compabile with the > > previous code, but it is no problem because nobody calls usb_bind_phy() > > for this driver now. > > > > Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@xxxxxxxxxxx> > > --- > > Documentation/devicetree/bindings/usb/renesas_usbhs.txt | 3 ++- > > drivers/usb/renesas_usbhs/mod_gadget.c | 7 ++++++- > > 2 files changed, 8 insertions(+), 2 deletions(-) > > > > diff --git a/Documentation/devicetree/bindings/usb/renesas_usbhs.txt > b/Documentation/devicetree/bindings/usb/renesas_usbhs.txt > > index b604056..a419aea 100644 > > --- a/Documentation/devicetree/bindings/usb/renesas_usbhs.txt > > +++ b/Documentation/devicetree/bindings/usb/renesas_usbhs.txt > > @@ -24,8 +24,9 @@ Optional properties: > > - renesas,buswait: Integer to use BUSWAIT register > > - renesas,enable-gpio: A gpio specifier to check GPIO determining if USB > > function should be enabled > > - - phys: phandle + phy specifier pair > > + - phys: phandle of *Generic PHY* + phy specifier pair > > - phy-names: must be "usb" > > + - usb-phy: phandle of usb phy > > No. The binding should not change based on which kernel subsystem > handles this. The USB-PHY code should learn to parse 'phys' if you want > to use that subsystem. Thank you for the review. The "phys" is for Generic PHY even if this patch is applied. So, I think that I should make a patch to revise the explanation at first? I would to write overview of renesas_usbhs driver and a question below: < phys > - This is a generic phy property. - The renesas_usbhs driver will use it to turn on/off the PHY - Almost all platforms use this generic phy instance. < usb-phy > - This is a USB-PHY property. - The renesas_usbhs driver will use it to call an OTG related function. (otg_set_peripheral) - The renesas_usbhs driver will NOT use it to turn the power on/off. - Some platforms don't use this USB-PHY instance. - In other words, some platforms use both a generic phy and USB-PHY. - Is it not suitable from the device tree point of view? Best regards, Yoshihiro Shimoda > > - dmas: Must contain a list of references to DMA specifiers. > > - dma-names : named "ch%d", where %d is the channel number ranging from zero > > to the number of channels (DnFIFOs) minus one.
