Re: [PATCH v2 2/4] phy: exynos5-usbdrd: Add pipe-clk and utmi-clk support

[Kishon Vijay Abraham I <kishon@ti.com>]

Hi,

On Tuesday 07 October 2014 03:49 PM, Vivek Gautam wrote:
> Exynos7 SoC has now separate gate control for 125MHz pipe3 phy
> clock, as well as 60MHz utmi phy clock.
> So get the same and control in the phy-exynos5-usbdrd driver.
> 
> Signed-off-by: Vivek Gautam <gautam.vivek@samsung.com>
> ---
>  .../devicetree/bindings/phy/samsung-phy.txt        |    4 ++++
>  drivers/phy/phy-exynos5-usbdrd.c                   |   22 ++++++++++++++++++++
>  2 files changed, 26 insertions(+)
> 
> diff --git a/Documentation/devicetree/bindings/phy/samsung-phy.txt b/Documentation/devicetree/bindings/phy/samsung-phy.txt
> index 15e0f2c..c2bc9dc 100644
> --- a/Documentation/devicetree/bindings/phy/samsung-phy.txt
> +++ b/Documentation/devicetree/bindings/phy/samsung-phy.txt
> @@ -138,6 +138,10 @@ Required properties:
>  	       PHY operations, associated by phy name. It is used to
>  	       determine bit values for clock settings register.
>  	       For Exynos5420 this is given as 'sclk_usbphy30' in CMU.
> +	- optional clocks: Exynos7 SoC has now following additional
> +			   gate clocks available:
> +			   - phy_pipe: for PIPE3 phy
> +			   - phy_utmi: for UTMI+ phy
>  - samsung,pmu-syscon: phandle for PMU system controller interface, used to
>  		      control pmu registers for power isolation.
>  - #phy-cells : from the generic PHY bindings, must be 1;
> diff --git a/drivers/phy/phy-exynos5-usbdrd.c b/drivers/phy/phy-exynos5-usbdrd.c
> index f756aca..013ee84 100644
> --- a/drivers/phy/phy-exynos5-usbdrd.c
> +++ b/drivers/phy/phy-exynos5-usbdrd.c
> @@ -148,6 +148,8 @@ struct exynos5_usbdrd_phy_drvdata {
>   * @dev: pointer to device instance of this platform device
>   * @reg_phy: usb phy controller register memory base
>   * @clk: phy clock for register access
> + * @pipeclk: clock for pipe3 phy
> + * @utmiclk: clock for utmi+ phy
>   * @drv_data: pointer to SoC level driver data structure
>   * @phys[]: array for 'EXYNOS5_DRDPHYS_NUM' number of PHY
>   *	    instances each with its 'phy' and 'phy_cfg'.
> @@ -161,6 +163,8 @@ struct exynos5_usbdrd_phy {
>  	struct device *dev;
>  	void __iomem *reg_phy;
>  	struct clk *clk;
> +	struct clk *pipeclk;
> +	struct clk *utmiclk;
>  	const struct exynos5_usbdrd_phy_drvdata *drv_data;
>  	struct phy_usb_instance {
>  		struct phy *phy;
> @@ -446,6 +450,8 @@ static int exynos5_usbdrd_phy_power_on(struct phy *phy)
>  
>  	dev_dbg(phy_drd->dev, "Request to power_on usbdrd_phy phy\n");
>  
> +	clk_prepare_enable(phy_drd->utmiclk);
> +	clk_prepare_enable(phy_drd->pipeclk);

We can have a separate function for powering on/off usb3 phy and usb2 phy
independently. That way if the USB controller is operating only on high speed
mode, the pipe clock need not be enabled at all.

I think we should create separate PHYs (phy_create) for usb2 phy and usb3 phy
here. Please refer how Lee Jones did that for miphy365x modelling each PHY as a
child node to the PHY provider and doing phy_create for each child node.

Thanks
Kishon