Hi Andy. Thanks for the review and sorry for the late reply. > -----Original Message----- > From: Andy Shevchenko <andriy.shevchenko@xxxxxxxxxxxxxxx> > Sent: Monday, November 9, 2020 7:41 PM > To: Wan Mohamad, Wan Ahmad Zainie > <wan.ahmad.zainie.wan.mohamad@xxxxxxxxx> > Cc: kishon@xxxxxx; vkoul@xxxxxxxxxx; robh+dt@xxxxxxxxxx; linux- > kernel@xxxxxxxxxxxxxxx; devicetree@xxxxxxxxxxxxxxx; > mgross@xxxxxxxxxxxxxxx; Raja Subramanian, Lakshmi Bai > <lakshmi.bai.raja.subramanian@xxxxxxxxx> > Subject: Re: [PATCH v2 2/2] phy: intel: Add Keem Bay USB PHY support > > On Mon, Nov 09, 2020 at 11:16:54AM +0800, Wan Ahmad Zainie wrote: > > Add support for USB PHY on Intel Keem Bay SoC. > > ... > > > +config PHY_INTEL_KEEMBAY_USB > > + tristate "Intel Keem Bay USB PHY driver" > > + depends on ARCH_KEEMBAY || (ARM64 && COMPILE_TEST) > > > + depends on OF && HAS_IOMEM > > Do you really need dependency to OF (yes, I see that it will be not functional, > but still can be compile tested)? No, I can drop OF. I will remove in v3. > > > + select GENERIC_PHY > > + select REGMAP_MMIO > > + help > > + Choose this option if you have an Intel Keem Bay SoC. > > + > > + To compile this driver as a module, choose M here: the module > > + will be called phy-keembay-usb.ko. > > ... > > > +#include <linux/bitfield.h> > > +#include <linux/bits.h> > > +#include <linux/clk.h> > > +#include <linux/delay.h> > > +#include <linux/module.h> > > > +#include <linux/of.h> > > No evidence of anything being used in this code. > mod_devicetable.h is missed, though. I will fix in v3. Remove of.h and add mod_devicetable.h. > > > +#include <linux/phy/phy.h> > > +#include <linux/platform_device.h> > > +#include <linux/regmap.h> > > ... > > > + usleep_range(30, 50); > > Why 30-50? I take this value from boot firmware. There is a delay of 30us after clearing IDDQ_enable bit. I believe the purpose is to ensure all analog blocks are powered up. > > ... > > > + usleep_range(20, 50); > > Why these numbers? In Keem Bay data book, under USB initialization section, there is step that there must be a minimum 20us wait after clock enable, before bringing PHYs out of reset. 50 is the value that I picked randomly. Is usleep_range(20, 20) Better? > > ... > > > + usleep_range(2, 10); > > Ditto. Under the same section above, there is a step for 2us wait. I believe it is for register write to go through. > > ... > > > + usleep_range(20, 50); > > Ditto. Under the same section above, there is a step to wait 20us after setting SRAM load bit, before release the controller reset. I will add comment for those 4 delay above. Before I proceed with v3, I would like to know if I should use udelay(), instead of usleep_range()? I refer to https://www.kernel.org/doc/Documentation/timers/timers-howto.txt. > > > ... > > > + struct device_node *np = dev->of_node; > > It's being used only once and it doesn't bring any benefit. I will fix in v3. Use dev->of_node directly. > > -- > With Best Regards, > Andy Shevchenko > Best regards, Zainie
