[PATCH 3/9] phy: qcom-m31: Introduce qcom,m31 USB phy driver

[Dmitry Baryshkov]

Two minor nits on top of the review:

On 07/06/2023 14:54, Konrad Dybcio wrote:
> On 7.06.2023 12:56, Varadarajan Narayanan wrote:
>> Add the M31 USB2 phy driver
>>
>> Signed-off-by: Varadarajan Narayanan <quic_varada at quicinc.com>
>> ---
>>   drivers/phy/qualcomm/phy-qcom-m31.c | 360 ++++++++++++++++++++++++++++++++++++
>>   1 file changed, 360 insertions(+)
>>   create mode 100644 drivers/phy/qualcomm/phy-qcom-m31.c
>>
>> diff --git a/drivers/phy/qualcomm/phy-qcom-m31.c b/drivers/phy/qualcomm/phy-qcom-m31.c
>> new file mode 100644
>> index 0000000..d29a91e
>> --- /dev/null
>> +++ b/drivers/phy/qualcomm/phy-qcom-m31.c
>> @@ -0,0 +1,360 @@
>> +// SPDX-License-Identifier: GPL-2.0+
>> +/*
>> + * Copyright (c) 2014-2016, 2020, The Linux Foundation. All rights reserved.
>> + */
>> +
>> +#include <linux/module.h>
>> +#include <linux/kernel.h>
>> +#include <linux/err.h>
>> +#include <linux/slab.h>
>> +#include <linux/clk.h>
>> +#include <linux/delay.h>
>> +#include <linux/io.h>
>> +#include <linux/of.h>
>> +#include <linux/platform_device.h>
>> +#include <linux/usb/phy.h>
>> +#include <linux/reset.h>
>> +#include <linux/of_device.h>
> Please sort these
> 
>> +
>> +enum clk_reset_action {
>> +	CLK_RESET_DEASSERT	= 0,
>> +	CLK_RESET_ASSERT	= 1
>> +};
>> +
>> +#define USB2PHY_PORT_POWERDOWN		0xA4
>> +#define POWER_UP			BIT(0)
>> +#define POWER_DOWN			0
>> +
>> +#define USB2PHY_PORT_UTMI_CTRL1	0x40
>> +
>> +#define USB2PHY_PORT_UTMI_CTRL2	0x44
>> +#define UTMI_ULPI_SEL			BIT(7)
>> +#define UTMI_TEST_MUX_SEL		BIT(6)
>> +
>> +#define HS_PHY_CTRL_REG			0x10
>> +#define UTMI_OTG_VBUS_VALID             BIT(20)
>> +#define SW_SESSVLD_SEL                  BIT(28)
>> +
>> +#define USB_PHY_CFG0			0x94
>> +#define USB_PHY_UTMI_CTRL5		0x50
>> +#define USB_PHY_FSEL_SEL		0xB8
>> +#define USB_PHY_HS_PHY_CTRL_COMMON0	0x54
>> +#define USB_PHY_REFCLK_CTRL		0xA0
>> +#define USB_PHY_HS_PHY_CTRL2		0x64
>> +#define USB_PHY_UTMI_CTRL0		0x3c
>> +#define USB2PHY_USB_PHY_M31_XCFGI_1	0xBC
>> +#define USB2PHY_USB_PHY_M31_XCFGI_4	0xC8
>> +#define USB2PHY_USB_PHY_M31_XCFGI_5	0xCC
>> +#define USB2PHY_USB_PHY_M31_XCFGI_11	0xE4
> Could you sort them address-wise?

... and lowercase the hex values, please.

> 
>> +
>> +#define USB2_0_TX_ENABLE		BIT(2)
>> +#define HSTX_SLEW_RATE_565PS		3
>> +#define PLL_CHARGING_PUMP_CURRENT_35UA	(3 << 3)
>> +#define ODT_VALUE_38_02_OHM		(3 << 6)
>> +#define ODT_VALUE_45_02_OHM		BIT(2)
>> +#define HSTX_PRE_EMPHASIS_LEVEL_0_55MA	(1)
> Weird mix of values, bits, bitfields.. perhaps BIT(n) and
> GENMASK() (+ FIELD_PREP) would be more suitable?
> 
>> +
>> +#define UTMI_PHY_OVERRIDE_EN		BIT(1)
>> +#define POR_EN				BIT(1)
> Please associate these with their registers, like
> 
> #define FOO_REG		0xf00
>   #define POR_EN		BIT(1)
> 
>> +#define FREQ_SEL			BIT(0)
>> +#define COMMONONN			BIT(7)
>> +#define FSEL				BIT(4)
>> +#define RETENABLEN			BIT(3)
>> +#define USB2_SUSPEND_N_SEL		BIT(3)
>> +#define USB2_SUSPEND_N			BIT(2)
>> +#define USB2_UTMI_CLK_EN		BIT(1)
>> +#define CLKCORE				BIT(1)
>> +#define ATERESET			~BIT(0)
>> +#define FREQ_24MHZ			(5 << 4)
>> +#define XCFG_COARSE_TUNE_NUM		(2 << 0)
>> +#define XCFG_FINE_TUNE_NUM		(1 << 3)
> same comment
> 
>> +
>> +static void m31usb_write_readback(void *base, u32 offset,
>> +					const u32 mask, u32 val);
> We don't need this forward-definition, just move the function up.
> 
>> +
>> +struct m31usb_phy {
>> +	struct usb_phy		phy;
>> +	void __iomem		*base;
>> +	void __iomem		*qscratch_base;
>> +
>> +	struct reset_control	*phy_reset;
>> +
>> +	bool			cable_connected;
>> +	bool			suspended;
>> +	bool			ulpi_mode;
>> +};
>> +
>> +static void m31usb_reset(struct m31usb_phy *qphy, u32 action)
>> +{
>> +	if (action == CLK_RESET_ASSERT)
>> +		reset_control_assert(qphy->phy_reset);
>> +	else
>> +		reset_control_deassert(qphy->phy_reset);
>> +	wmb(); /* ensure data is written to hw register */
> Please move the comment above the call.
> 
>> +}

Or even better just inline the function. I was never a fan of such 
multiplexers.

Also does wmb() make sense here? Doesn't regmap (which is used by reset 
controller) remove the need for it?

>> +
>> +static void m31usb_phy_enable_clock(struct m31usb_phy *qphy)
>> +{
>> +	/* Enable override ctrl */
>> +	writel(UTMI_PHY_OVERRIDE_EN, qphy->base + USB_PHY_CFG0);
> Some of the comments are missing a space before '*/'
> 
> Also, please consider adding some newlines to logically split the
> actions.


-- 
With best wishes
Dmitry