Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Thu, 5 Jan 2023 18:03:20 +0100
From:   Andrew Lunn <andrew@lunn.ch>
To:     Frank <Frank.Sae@motor-comm.com>
Cc:     Peter Geis <pgwipeout@gmail.com>,
        Heiner Kallweit <hkallweit1@gmail.com>,
        Russell King <linux@armlinux.org.uk>,
        "David S . Miller" <davem@davemloft.net>,
        Eric Dumazet <edumazet@google.com>,
        Jakub Kicinski <kuba@kernel.org>,
        Paolo Abeni <pabeni@redhat.com>,
        Rob Herring <robh+dt@kernel.org>,
        Krzysztof Kozlowski <krzysztof.kozlowski+dt@linaro.org>,
        xiaogang.fan@motor-comm.com, fei.zhang@motor-comm.com,
        hua.sun@motor-comm.com, netdev@vger.kernel.org,
        linux-kernel@vger.kernel.org, devicetree@vger.kernel.org
Subject: Re: [PATCH net-next v1 2/3] net: phy: Add dts support for Motorcomm
 yt8521/yt8531s gigabit ethernet phy
Message-ID: <Y7cC2MKYK4omdZKg@lunn.ch>
References: <20230105073024.8390-1-Frank.Sae@motor-comm.com>
 <20230105073024.8390-3-Frank.Sae@motor-comm.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20230105073024.8390-3-Frank.Sae@motor-comm.com>
Precedence: bulk

On Thu, Jan 05, 2023 at 03:30:23PM +0800, Frank wrote:
> Add dts support for yt8521 and yt8531s. This patch has
> been tested on AM335x platform which has one YT8531S interface
> card and passed all test cases.
> 
> Signed-off-by: Frank <Frank.Sae@motor-comm.com>
> ---
>  drivers/net/phy/motorcomm.c | 517 ++++++++++++++++++++++++++++++------
>  1 file changed, 434 insertions(+), 83 deletions(-)
> 
> diff --git a/drivers/net/phy/motorcomm.c b/drivers/net/phy/motorcomm.c
> index 685190db72de..7ebcca374a67 100644
> --- a/drivers/net/phy/motorcomm.c
> +++ b/drivers/net/phy/motorcomm.c
> @@ -10,10 +10,11 @@
>  #include <linux/kernel.h>
>  #include <linux/module.h>
>  #include <linux/phy.h>
> +#include <linux/of.h>
>  
>  #define PHY_ID_YT8511		0x0000010a
> -#define PHY_ID_YT8521		0x0000011A
> -#define PHY_ID_YT8531S		0x4F51E91A
> +#define PHY_ID_YT8521		0x0000011a
> +#define PHY_ID_YT8531S		0x4f51e91a

Please do the lower case conversion as a separate patch.

>  
>  /* YT8521/YT8531S Register Overview
>   *	UTP Register space	|	FIBER Register space
> @@ -144,6 +145,16 @@
>  #define YT8521_ESC1R_SLEEP_SW			BIT(15)
>  #define YT8521_ESC1R_PLLON_SLP			BIT(14)
>  
> +/* Phy Serdes analog cfg2 Register */
> +#define YTPHY_SERDES_ANALOG_CFG2_REG		0xA1
> +#define YTPHY_SAC2R_TX_AMPLITUDE_MASK		((0x7 << 13) | (0x7 << 1))
> +#define YT8521_SAC2R_TX_AMPLITUDE_LOW		((0x7 << 13) | (0x0 << 1))
> +#define YT8521_SAC2R_TX_AMPLITUDE_MIDDLE	((0x5 << 13) | (0x5 << 1))
> +#define YT8521_SAC2R_TX_AMPLITUDE_HIGH		((0x3 << 13) | (0x6 << 1))

So there are two values which control the amplitude? Buts 1-3, and bit
7-9?  Can they be used independently?  Also, 7, 5, 3 is also add. Does
bit 0 of this value have some special meaning? Please document this
fully.

> +#define YT8531S_SAC2R_TX_AMPLITUDE_LOW		((0x0 << 13) | (0x0 << 1))
> +#define YT8531S_SAC2R_TX_AMPLITUDE_MIDDLE	((0x0 << 13) | (0x1 << 1))
> +#define YT8531S_SAC2R_TX_AMPLITUDE_HIGH		((0x0 << 13) | (0x2 << 1))

This more sense, but why the 0 << 13? What do the bits 13-? mean?

> +
>  /* Phy fiber Link timer cfg2 Register */
>  #define YT8521_LINK_TIMER_CFG2_REG		0xA5
>  #define YT8521_LTCR_EN_AUTOSEN			BIT(15)
> @@ -161,6 +172,7 @@
>  
>  #define YT8521_CHIP_CONFIG_REG			0xA001
>  #define YT8521_CCR_SW_RST			BIT(15)
> +#define YT8521_CCR_RXC_DLY_EN			BIT(8)
>  
>  #define YT8521_CCR_MODE_SEL_MASK		(BIT(2) | BIT(1) | BIT(0))
>  #define YT8521_CCR_MODE_UTP_TO_RGMII		0
> @@ -178,22 +190,27 @@
>  #define YT8521_MODE_POLL			0x3
>  
>  #define YT8521_RGMII_CONFIG1_REG		0xA003
> -
> +#define YT8521_RC1R_TX_CLK_SEL_MASK		BIT(14)
> +#define YT8521_RC1R_TX_CLK_SEL_ORIGINAL		(0x0 << 14)
> +#define YT8521_RC1R_TX_CLK_SEL_INVERTED		(0x1 << 14)

Please use the BIT macro.


>  /* TX Gig-E Delay is bits 3:0, default 0x1
>   * TX Fast-E Delay is bits 7:4, default 0xf
>   * RX Delay is bits 13:10, default 0x0
>   * Delay = 150ps * N
>   * On = 2250ps, off = 0ps
>   */
> -#define YT8521_RC1R_RX_DELAY_MASK		(0xF << 10)
> -#define YT8521_RC1R_RX_DELAY_EN			(0xF << 10)
> -#define YT8521_RC1R_RX_DELAY_DIS		(0x0 << 10)
> -#define YT8521_RC1R_FE_TX_DELAY_MASK		(0xF << 4)
> -#define YT8521_RC1R_FE_TX_DELAY_EN		(0xF << 4)
> -#define YT8521_RC1R_FE_TX_DELAY_DIS		(0x0 << 4)
> -#define YT8521_RC1R_GE_TX_DELAY_MASK		(0xF << 0)
> -#define YT8521_RC1R_GE_TX_DELAY_EN		(0xF << 0)
> -#define YT8521_RC1R_GE_TX_DELAY_DIS		(0x0 << 0)
> +#define YT8521_RC1R_GE_TX_DELAY_BIT		(0)
> +#define YT8521_RC1R_FE_TX_DELAY_BIT		(4)
> +#define YT8521_RC1R_RX_DELAY_BIT		(10)
> +#define YT8521_RC1R_RX_DELAY_MASK		(0xF << YT8521_RC1R_RX_DELAY_BIT)
> +#define YT8521_RC1R_RX_DELAY_EN			(0xF << YT8521_RC1R_RX_DELAY_BIT)
> +#define YT8521_RC1R_RX_DELAY_DIS		(0x0 << YT8521_RC1R_RX_DELAY_BIT)
> +#define YT8521_RC1R_FE_TX_DELAY_MASK		(0xF << YT8521_RC1R_FE_TX_DELAY_BIT)
> +#define YT8521_RC1R_FE_TX_DELAY_EN		(0xF << YT8521_RC1R_FE_TX_DELAY_BIT)
> +#define YT8521_RC1R_FE_TX_DELAY_DIS		(0x0 << YT8521_RC1R_FE_TX_DELAY_BIT)
> +#define YT8521_RC1R_GE_TX_DELAY_MASK		(0xF << YT8521_RC1R_GE_TX_DELAY_BIT)
> +#define YT8521_RC1R_GE_TX_DELAY_EN		(0xF << YT8521_RC1R_GE_TX_DELAY_BIT)
> +#define YT8521_RC1R_GE_TX_DELAY_DIS		(0x0 << YT8521_RC1R_GE_TX_DELAY_BIT)
>  
>  #define YTPHY_MISC_CONFIG_REG			0xA006
>  #define YTPHY_MCR_FIBER_SPEED_MASK		BIT(0)
> @@ -222,11 +239,33 @@
>   */
>  #define YTPHY_WCR_TYPE_PULSE			BIT(0)
>  
> -#define YT8531S_SYNCE_CFG_REG			0xA012
> -#define YT8531S_SCR_SYNCE_ENABLE		BIT(6)
> +#define YTPHY_SYNCE_CFG_REG			0xA012
> +#define YT8521_SCR_CLK_SRC_MASK			(BIT(2) | BIT(1))
> +#define YT8521_SCR_CLK_SRC_PLL_125M		(0x0 << 1)
> +#define YT8521_SCR_CLK_SRC_REF_25M		(0x3 << 1)
> +#define YT8521_SCR_SYNCE_ENABLE			BIT(5)
> +#define YT8521_SCR_CLK_FRE_SEL_MASK		BIT(3)
> +#define YT8521_SCR_CLK_FRE_SEL_125M		(0x1 << 3)
> +#define YT8521_SCR_CLK_FRE_SEL_25M		(0x0 << 3)

Whenever it is a single bit, please use the BIT macro.

> +#define YT8531_SCR_CLK_SRC_MASK			(BIT(3) | BIT(2) | BIT(1))
> +#define YT8531_SCR_CLK_SRC_PLL_125M		(0x0 << 1)
> +#define YT8531_SCR_CLK_SRC_REF_25M		(0x4 << 1)
> +#define YT8531_SCR_SYNCE_ENABLE			BIT(6)
> +#define YT8531_SCR_CLK_FRE_SEL_MASK		BIT(4)
> +#define YT8531_SCR_CLK_FRE_SEL_125M		(0x1 << 4)
> +#define YT8531_SCR_CLK_FRE_SEL_25M		(0x0 << 4)
>  
>  /* Extended Register  end */
>  
> +#define YTPHY_DTS_MAX_TX_AMPLITUDE		0x2
> +#define YTPHY_DTS_MAX_DELAY_VAL			2250
> +#define YTPHY_DTS_STEP_DELAY_VAL		150
> +#define YTPHY_DTS_INVAL_VAL			0xFF
> +
> +#define YTPHY_DTS_OUTPUT_CLK_DIS		0
> +#define YTPHY_DTS_OUTPUT_CLK_25M		25000000
> +#define YTPHY_DTS_OUTPUT_CLK_125M		125000000
> +
>  struct yt8521_priv {
>  	/* combo_advertising is used for case of YT8521 in combo mode,
>  	 * this means that yt8521 may work in utp or fiber mode which depends
> @@ -243,6 +282,30 @@ struct yt8521_priv {
>  	 * YT8521_RSSR_TO_BE_ARBITRATED
>  	 */
>  	u8 reg_page;
> +
> +	/* The following parameters are from dts */
> +	/* rx delay = rx_delay_basic + rx_delay_additional
> +	 * basic delay is ~2ns, 0 = off, 1 = on
> +	 * rx_delay_additional,delay time = 150ps * val
> +	 */
> +	u8 rx_delay_basic;
> +	u8 rx_delay_additional;
> +
> +	/* tx_delay_ge is tx_delay for 1000Mbps
> +	 * tx_delay_fe is tx_delay for 100Mbps or 10Mbps
> +	 * delay time = 150ps * val
> +	 */
> +	u8 tx_delay_ge;
> +	u8 tx_delay_fe;
> +	u8 sds_tx_amplitude;
> +	bool keep_pll_enabled;
> +	bool auto_sleep_disabled;
> +	bool clock_ouput;	/* output clock ctl: 0=off, 1=on */
> +	bool clock_freq_125M;	/* output clock freq selcect: 0=25M, 1=125M */
> +	bool tx_clk_adj_enabled;/* tx clk adj ctl: 0=off, 1=on */
> +	bool tx_clk_10_inverted;
> +	bool tx_clk_100_inverted;
> +	bool tx_clk_1000_inverted;

Do you need to store all these values? In general, PHY drivers parse
DT, and program the hardware directly. If these values are lost on
reset, and you need to perform a reset for normal operation, then yes,
it makes sense to store them. But in general, that is not how hardware
works.

> +static int ytphy_parse_dt(struct phy_device *phydev)
> +{
> +	struct device_node *node = phydev->mdio.dev.of_node;
> +	struct yt8521_priv *priv = phydev->priv;
> +	u32 freq, val;
> +	int ret;
> +
> +	priv->rx_delay_additional = YTPHY_DTS_INVAL_VAL;
> +	priv->sds_tx_amplitude = YTPHY_DTS_INVAL_VAL;
> +	priv->rx_delay_basic = YTPHY_DTS_INVAL_VAL;
> +	priv->tx_delay_ge = YTPHY_DTS_INVAL_VAL;
> +	priv->tx_delay_fe = YTPHY_DTS_INVAL_VAL;
> +
> +	if (!IS_ENABLED(CONFIG_OF_MDIO)) {

No other PHY driver does this. Why is this here?

As a general rule, if you do something which no other driver does, you
are doing something wrong. All PHY drivers should basically look the
same, follow the same structure, etc. So it is a good idea to review 5
other drivers, and make your driver look similar.

> +		priv->auto_sleep_disabled = true;
> +		priv->keep_pll_enabled = true;
> +		return 0;
> +	}
> +
> +	ret = of_property_read_u32(node, "motorcomm,clk-out-frequency", &freq);
> +	if (ret < 0)
> +		freq = YTPHY_DTS_OUTPUT_CLK_DIS;/* default value as dts*/
> +
> +	switch (freq) {
> +	case YTPHY_DTS_OUTPUT_CLK_DIS:
> +		priv->clock_ouput = false;
> +		break;
> +	case YTPHY_DTS_OUTPUT_CLK_25M:
> +		priv->clock_freq_125M = false;
> +		priv->clock_ouput = true;
> +		break;
> +	case YTPHY_DTS_OUTPUT_CLK_125M:
> +		priv->clock_freq_125M = true;
> +		priv->clock_ouput = true;
> +		break;
> +	default:
> +		phydev_err(phydev, "invalid motorcomm,clk-out-frequency\n");
> +		return -EINVAL;
> +	}
> +
> +	if (!of_property_read_u32(node, "motorcomm,rx-delay-basic", &val)) {
> +		if (val > 1) {
> +			phydev_err(phydev,
> +				   "invalid motorcomm,rx-delay-basic\n");
> +			return -EINVAL;
> +		}
> +		priv->rx_delay_basic = val;
> +	}
> +
> +	if (!of_property_read_u32(node, "motorcomm,rx-delay-additional-ps", &val)) {
> +		if (val > YTPHY_DTS_MAX_DELAY_VAL) {
> +			phydev_err(phydev, "invalid motorcomm,rx-delay-additional-ps\n");
> +			return -EINVAL;
> +		}

Please check the value is also one of the supported values. Please do
that for all your delays.

> +	if (of_property_read_bool(node, "motorcomm,keep-pll-enabled"))
> +		priv->keep_pll_enabled = true;

I think this only makes sense when priv->clock_output is true? Please
test for that.


> +static int ytphy_rgmii_clk_delay_config(struct phy_device *phydev)
> +{
> +	struct yt8521_priv *priv = phydev->priv;
> +	u16 mask = 0;
> +	u16 val = 0;
> +	int ret;
> +
> +	/* rx delay basic controlled by dts.*/
> +	if (priv->rx_delay_basic != YTPHY_DTS_INVAL_VAL) {
> +		if (priv->rx_delay_basic)
> +			val = YT8521_CCR_RXC_DLY_EN;
> +		ret = ytphy_modify_ext(phydev, YT8521_CHIP_CONFIG_REG,
> +				       YT8521_CCR_RXC_DLY_EN, val);
> +		if (ret < 0)
> +			return ret;
> +	}
> +
> +	val = 0;
> +	/* If rx_delay_additional and tx_delay_* are all not be seted in dts,
> +	 * then used the fixed *_DELAY_DIS or *_DELAY_EN. Otherwise, use the
> +	 * value set by rx_delay_additional, tx_delay_ge and tx_delay_fe.
> +	 */

So what you should be doing here is always respecting
phydev->interface. You can then fine tune the delays using
rx-internal-delay-ps and tx-internal-delay-ps.

> +static int ytphy_probe_helper(struct phy_device *phydev)
> +{
> +	struct device *dev = &phydev->mdio.dev;
> +	struct yt8521_priv *priv;
> +	int ret;
> +
> +	priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL);
> +	if (!priv)
> +		return -ENOMEM;
> +
> +	phydev->priv = priv;
> +
> +	ret = ytphy_parse_dt(phydev);
> +	if (ret < 0)
> +		return ret;
> +
> +	phy_lock_mdio_bus(phydev);
> +	ret = ytphy_clk_out_config(phydev);
> +	phy_unlock_mdio_bus(phydev);
> +	return ret;
> +}
> +
>  /**
>   * yt8521_probe() - read chip config then set suitable polling_mode
>   * @phydev: a pointer to a &struct phy_device
> @@ -601,16 +983,15 @@ static int yt8521_write_page(struct phy_device *phydev, int page)
>   */
>  static int yt8521_probe(struct phy_device *phydev)
>  {
> -	struct device *dev = &phydev->mdio.dev;
>  	struct yt8521_priv *priv;
>  	int chip_config;
>  	int ret;
>  
> -	priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL);
> -	if (!priv)
> -		return -ENOMEM;
> +	ret = ytphy_probe_helper(phydev);
> +	if (ret < 0)
> +		return ret;
>  
> -	phydev->priv = priv;
> +	priv = phydev->priv;

I don't see why you added this probe helper.

Is this to make the driver look more like the vendor driver? Please
seperate refactoring patches from new functionality. We want to see
lots of simple, obviously correct patches which are easy to review.

     Andrew