[PATCH net-next v2 1/3] net: phy: marvell: add basic support of 88E308X/88E609X family

[PATCH net-next v2 1/3] net: phy: marvell: add basic support of 88E308X/88E609X family

[Pawel Dembicki]

This patch implements only basic support.

It covers PHY used in multiple IC:
PHY: 88E3082, 88E3083
Switch: 88E6096, 88E6097

Signed-off-by: Pawel Dembicki <paweldembicki@gmail.com>

---
v2:
  - resend only

 drivers/net/phy/marvell.c   | 13 +++++++++++++
 include/linux/marvell_phy.h |  1 +
 2 files changed, 14 insertions(+)

diff --git a/drivers/net/phy/marvell.c b/drivers/net/phy/marvell.c
index 42ed013385bf..fae7eb57ee2c 100644
--- a/drivers/net/phy/marvell.c
+++ b/drivers/net/phy/marvell.c
@@ -3289,6 +3289,18 @@ static struct phy_driver marvell_drivers[] = {
 		.get_strings = marvell_get_strings,
 		.get_stats = marvell_get_stats,
 	},
+	{
+		.phy_id = MARVELL_PHY_ID_88E3082,
+		.phy_id_mask = MARVELL_PHY_ID_MASK,
+		.name = "Marvell 88E308X/88E609X Family",
+		/* PHY_BASIC_FEATURES */
+		.probe = marvell_probe,
+		.config_init = marvell_config_init,
+		.aneg_done = marvell_aneg_done,
+		.read_status = marvell_read_status,
+		.resume = genphy_resume,
+		.suspend = genphy_suspend,
+	},
 	{
 		.phy_id = MARVELL_PHY_ID_88E1112,
 		.phy_id_mask = MARVELL_PHY_ID_MASK,
@@ -3742,6 +3754,7 @@ module_phy_driver(marvell_drivers);
 
 static struct mdio_device_id __maybe_unused marvell_tbl[] = {
 	{ MARVELL_PHY_ID_88E1101, MARVELL_PHY_ID_MASK },
+	{ MARVELL_PHY_ID_88E3082, MARVELL_PHY_ID_MASK },
 	{ MARVELL_PHY_ID_88E1112, MARVELL_PHY_ID_MASK },
 	{ MARVELL_PHY_ID_88E1111, MARVELL_PHY_ID_MASK },
 	{ MARVELL_PHY_ID_88E1111_FINISAR, MARVELL_PHY_ID_MASK },
diff --git a/include/linux/marvell_phy.h b/include/linux/marvell_phy.h
index 693eba9869e4..88254f9aec2b 100644
--- a/include/linux/marvell_phy.h
+++ b/include/linux/marvell_phy.h
@@ -7,6 +7,7 @@
 
 /* Known PHY IDs */
 #define MARVELL_PHY_ID_88E1101		0x01410c60
+#define MARVELL_PHY_ID_88E3082		0x01410c80
 #define MARVELL_PHY_ID_88E1112		0x01410c90
 #define MARVELL_PHY_ID_88E1111		0x01410cc0
 #define MARVELL_PHY_ID_88E1118		0x01410e10
-- 
2.25.1

[PATCH net-next v2 2/3] net: ethtool: Add impedance mismatch result code to cable test

[Pawel Dembicki]

Some PHYs can recognize during a cable test if the impedance in the cable
is okay.

This commit introduces a new result code:
ETHTOOL_A_CABLE_RESULT_CODE_IMPEDANCE_MISMATCH
which represents the results of a cable test indicating abnormal impedance.

Signed-off-by: Pawel Dembicki <paweldembicki@gmail.com>
---
v2:
  - introduce patch

 include/uapi/linux/ethtool_netlink.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/include/uapi/linux/ethtool_netlink.h b/include/uapi/linux/ethtool_netlink.h
index 3f89074aa06c..ecc020bd47d1 100644
--- a/include/uapi/linux/ethtool_netlink.h
+++ b/include/uapi/linux/ethtool_netlink.h
@@ -515,6 +515,7 @@ enum {
 	ETHTOOL_A_CABLE_RESULT_CODE_OPEN,
 	ETHTOOL_A_CABLE_RESULT_CODE_SAME_SHORT,
 	ETHTOOL_A_CABLE_RESULT_CODE_CROSS_SHORT,
+	ETHTOOL_A_CABLE_RESULT_CODE_IMPEDANCE_MISMATCH,
 };
 
 enum {
-- 
2.25.1

[PATCH net-next v2 3/3] net: phy: marvell: implement cable-test for 88E308X/88E609X family

[Pawel Dembicki]

This commit implements VCT in 88E308X/88E609X Family.

It require two workarounds with some magic configuration.
Regular use require only one register configuration. But Open Circuit
require second workaround.
It cause implementation two phases for fault length measuring.

Fast Ethernet PHY have implemented very simple version of VCT. It's
complitley different than vct5 or vct7.

Signed-off-by: Pawel Dembicki <paweldembicki@gmail.com>

---
v2:
  - change 'm88e3082_vct_distrfln_2_cm' to static
  - simplify 'm88e3082_vct_cable_test_get_status'
  - replace magic numbers in MII_BMCR configuration
  - remove unnecessary backup of MII_BMCR register
  - use ETHTOOL_A_CABLE_RESULT_CODE_IMPEDANCE_MISMATCH for impedance
    mismatch

 drivers/net/phy/marvell.c | 208 ++++++++++++++++++++++++++++++++++++++
 1 file changed, 208 insertions(+)

diff --git a/drivers/net/phy/marvell.c b/drivers/net/phy/marvell.c
index fae7eb57ee2c..7c00f47e4ded 100644
--- a/drivers/net/phy/marvell.c
+++ b/drivers/net/phy/marvell.c
@@ -279,6 +279,23 @@
 #define MII_VCT7_CTRL_METERS			BIT(10)
 #define MII_VCT7_CTRL_CENTIMETERS		0
 
+#define MII_VCT_TXPINS			0x1A
+#define MII_VCT_RXPINS			0x1B
+#define MII_VCT_TXPINS_ENVCT		BIT(15)
+#define MII_VCT_TXRXPINS_VCTTST		GENMASK(14, 13)
+#define MII_VCT_TXRXPINS_VCTTST_SHIFT	13
+#define MII_VCT_TXRXPINS_VCTTST_OK	0
+#define MII_VCT_TXRXPINS_VCTTST_SHORT	1
+#define MII_VCT_TXRXPINS_VCTTST_OPEN	2
+#define MII_VCT_TXRXPINS_VCTTST_FAIL	3
+#define MII_VCT_TXRXPINS_AMPRFLN	GENMASK(12, 8)
+#define MII_VCT_TXRXPINS_AMPRFLN_SHIFT	8
+#define MII_VCT_TXRXPINS_DISTRFLN	GENMASK(7, 0)
+#define MII_VCT_TXRXPINS_DISTRFLN_MAX	0xff
+
+#define M88E3082_PAIR_A		BIT(0)
+#define M88E3082_PAIR_B		BIT(1)
+
 #define LPA_PAUSE_FIBER		0x180
 #define LPA_PAUSE_ASYM_FIBER	0x100
 
@@ -301,6 +318,12 @@ static struct marvell_hw_stat marvell_hw_stats[] = {
 	{ "phy_receive_errors_fiber", 1, 21, 16},
 };
 
+enum {
+	M88E3082_VCT_OFF,
+	M88E3082_VCT_PHASE1,
+	M88E3082_VCT_PHASE2,
+};
+
 struct marvell_priv {
 	u64 stats[ARRAY_SIZE(marvell_hw_stats)];
 	char *hwmon_name;
@@ -310,6 +333,7 @@ struct marvell_priv {
 	u32 last;
 	u32 step;
 	s8 pair;
+	u8 vct_phase;
 };
 
 static int marvell_read_page(struct phy_device *phydev)
@@ -2417,6 +2441,188 @@ static int marvell_vct7_cable_test_get_status(struct phy_device *phydev,
 	return 0;
 }
 
+static int m88e3082_vct_cable_test_start(struct phy_device *phydev)
+{
+	struct marvell_priv *priv = phydev->priv;
+	int ret;
+
+	/* It needs some magic workarounds described in VCT manual for this PHY.
+	 */
+	ret = phy_write(phydev, 29, 0x0003);
+	if (ret < 0)
+		return ret;
+
+	ret = phy_write(phydev, 30, 0x6440);
+	if (ret < 0)
+		return ret;
+
+	if (priv->vct_phase == M88E3082_VCT_PHASE1) {
+		ret = phy_write(phydev, 29, 0x000a);
+		if (ret < 0)
+			return ret;
+
+		ret = phy_write(phydev, 30, 0x0002);
+		if (ret < 0)
+			return ret;
+	}
+
+	ret = phy_write(phydev, MII_BMCR,
+			BMCR_RESET | BMCR_SPEED100 | BMCR_FULLDPLX);
+	if (ret < 0)
+		return ret;
+
+	ret = phy_write(phydev, MII_VCT_TXPINS, MII_VCT_TXPINS_ENVCT);
+	if (ret < 0)
+		return ret;
+
+	ret = phy_write(phydev, 29, 0x0003);
+	if (ret < 0)
+		return ret;
+
+	ret = phy_write(phydev, 30, 0x0);
+	if (ret < 0)
+		return ret;
+
+	if (priv->vct_phase == M88E3082_VCT_OFF) {
+		priv->vct_phase = M88E3082_VCT_PHASE1;
+		priv->pair = 0;
+
+		return 0;
+	}
+
+	ret = phy_write(phydev, 29, 0x000a);
+	if (ret < 0)
+		return ret;
+
+	ret = phy_write(phydev, 30, 0x0);
+	if (ret < 0)
+		return ret;
+
+	priv->vct_phase = M88E3082_VCT_PHASE2;
+
+	return 0;
+}
+
+static int m88e3082_vct_cable_test_report_trans(int result, u8 distance)
+{
+	switch (result) {
+	case MII_VCT_TXRXPINS_VCTTST_OK:
+		if (distance == MII_VCT_TXRXPINS_DISTRFLN_MAX)
+			return ETHTOOL_A_CABLE_RESULT_CODE_OK;
+		return ETHTOOL_A_CABLE_RESULT_CODE_IMPEDANCE_MISMATCH;
+	case MII_VCT_TXRXPINS_VCTTST_SHORT:
+		return ETHTOOL_A_CABLE_RESULT_CODE_SAME_SHORT;
+	case MII_VCT_TXRXPINS_VCTTST_OPEN:
+		return ETHTOOL_A_CABLE_RESULT_CODE_OPEN;
+	default:
+		return ETHTOOL_A_CABLE_RESULT_CODE_UNSPEC;
+	}
+}
+
+static u32 m88e3082_vct_distrfln_2_cm(u8 distrfln)
+{
+	if (distrfln < 24)
+		return 0;
+
+	/* Original function for meters: y = 0.7861x - 18.862 */
+	return (7861 * distrfln - 188620) / 100;
+}
+
+static int m88e3082_vct_cable_test_get_status(struct phy_device *phydev,
+					      bool *finished)
+{
+	u8 tx_vcttst_res, rx_vcttst_res, tx_distrfln, rx_distrfln;
+	struct marvell_priv *priv = phydev->priv;
+	int ret, tx_result, rx_result;
+	bool done_phase = true;
+
+	*finished = false;
+
+	ret = phy_read(phydev, MII_VCT_TXPINS);
+	if (ret < 0)
+		return ret;
+	else if (ret & MII_VCT_TXPINS_ENVCT)
+		return 0;
+
+	tx_distrfln = ret & MII_VCT_TXRXPINS_DISTRFLN;
+	tx_vcttst_res = (ret & MII_VCT_TXRXPINS_VCTTST) >>
+			MII_VCT_TXRXPINS_VCTTST_SHIFT;
+
+	ret = phy_read(phydev, MII_VCT_RXPINS);
+	if (ret < 0)
+		return ret;
+
+	rx_distrfln = ret & MII_VCT_TXRXPINS_DISTRFLN;
+	rx_vcttst_res = (ret & MII_VCT_TXRXPINS_VCTTST) >>
+			MII_VCT_TXRXPINS_VCTTST_SHIFT;
+
+	*finished = true;
+
+	switch (priv->vct_phase) {
+	case M88E3082_VCT_PHASE1:
+		tx_result = m88e3082_vct_cable_test_report_trans(tx_vcttst_res,
+								 tx_distrfln);
+		rx_result = m88e3082_vct_cable_test_report_trans(rx_vcttst_res,
+								 rx_distrfln);
+
+		ethnl_cable_test_result(phydev, ETHTOOL_A_CABLE_PAIR_A,
+					tx_result);
+		ethnl_cable_test_result(phydev, ETHTOOL_A_CABLE_PAIR_B,
+					rx_result);
+
+		if (tx_vcttst_res == MII_VCT_TXRXPINS_VCTTST_OPEN) {
+			done_phase = false;
+			priv->pair |= M88E3082_PAIR_A;
+		} else if (tx_distrfln < MII_VCT_TXRXPINS_DISTRFLN_MAX) {
+			u8 pair = ETHTOOL_A_CABLE_PAIR_A;
+			u32 cm = m88e3082_vct_distrfln_2_cm(tx_distrfln);
+
+			ethnl_cable_test_fault_length(phydev, pair, cm);
+		}
+
+		if (rx_vcttst_res == MII_VCT_TXRXPINS_VCTTST_OPEN) {
+			done_phase = false;
+			priv->pair |= M88E3082_PAIR_B;
+		} else if (rx_distrfln < MII_VCT_TXRXPINS_DISTRFLN_MAX) {
+			u8 pair = ETHTOOL_A_CABLE_PAIR_B;
+			u32 cm = m88e3082_vct_distrfln_2_cm(rx_distrfln);
+
+			ethnl_cable_test_fault_length(phydev, pair, cm);
+		}
+
+		break;
+	case M88E3082_VCT_PHASE2:
+		if (priv->pair & M88E3082_PAIR_A &&
+		    tx_vcttst_res == MII_VCT_TXRXPINS_VCTTST_OPEN &&
+		    tx_distrfln < MII_VCT_TXRXPINS_DISTRFLN_MAX) {
+			u8 pair = ETHTOOL_A_CABLE_PAIR_A;
+			u32 cm = m88e3082_vct_distrfln_2_cm(tx_distrfln);
+
+			ethnl_cable_test_fault_length(phydev, pair, cm);
+		}
+		if (priv->pair & M88E3082_PAIR_B &&
+		    rx_vcttst_res == MII_VCT_TXRXPINS_VCTTST_OPEN &&
+		    rx_distrfln < MII_VCT_TXRXPINS_DISTRFLN_MAX) {
+			u8 pair = ETHTOOL_A_CABLE_PAIR_B;
+			u32 cm = m88e3082_vct_distrfln_2_cm(rx_distrfln);
+
+			ethnl_cable_test_fault_length(phydev, pair, cm);
+		}
+
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	if (!done_phase) {
+		*finished = false;
+		return m88e3082_vct_cable_test_start(phydev);
+	}
+	if (*finished)
+		priv->vct_phase = M88E3082_VCT_OFF;
+	return 0;
+}
+
 #ifdef CONFIG_HWMON
 struct marvell_hwmon_ops {
 	int (*config)(struct phy_device *phydev);
@@ -3300,6 +3506,8 @@ static struct phy_driver marvell_drivers[] = {
 		.read_status = marvell_read_status,
 		.resume = genphy_resume,
 		.suspend = genphy_suspend,
+		.cable_test_start = m88e3082_vct_cable_test_start,
+		.cable_test_get_status = m88e3082_vct_cable_test_get_status,
 	},
 	{
 		.phy_id = MARVELL_PHY_ID_88E1112,
-- 
2.25.1

Re: [PATCH net-next v2 2/3] net: ethtool: Add impedance mismatch result code to cable test

[Jakub Kicinski]

On Wed, 27 Mar 2024 17:29:13 +0100 Pawel Dembicki wrote:
> This commit introduces a new result code:
> ETHTOOL_A_CABLE_RESULT_CODE_IMPEDANCE_MISMATCH
> which represents the results of a cable test indicating abnormal impedance.

I'm not a cable expert but going purely by the language
abnormal != mismatch. Mismatch indicates there are two
values we are comparing.

Re: [PATCH net-next v2 2/3] net: ethtool: Add impedance mismatch result code to cable test

[Paweł Dembicki]

pt., 29 mar 2024 o 03:02 Jakub Kicinski <kuba@kernel.org> napisał(a):
>
> On Wed, 27 Mar 2024 17:29:13 +0100 Pawel Dembicki wrote:
> > This commit introduces a new result code:
> > ETHTOOL_A_CABLE_RESULT_CODE_IMPEDANCE_MISMATCH
> > which represents the results of a cable test indicating abnormal impedance.
>
> I'm not a cable expert but going purely by the language
> abnormal != mismatch. Mismatch indicates there are two
> values we are comparing.

Impedance mismatch can be detected because some parts of the cable may
have different (abnormal) impedance, causing reflections at those
points. Ethernet cables should have a characteristic impedance of 100
Ohms, so any mismatch from this value is considered abnormal. I can
provide a rephrased version if the commit description was not detailed
enough.

Re: [PATCH net-next v2 2/3] net: ethtool: Add impedance mismatch result code to cable test

[Jakub Kicinski]

On Fri, 29 Mar 2024 10:31:10 +0100 Paweł Dembicki wrote:
> > I'm not a cable expert but going purely by the language
> > abnormal != mismatch. Mismatch indicates there are two
> > values we are comparing.  
> 
> Impedance mismatch can be detected because some parts of the cable may
> have different (abnormal) impedance, causing reflections at those
> points. Ethernet cables should have a characteristic impedance of 100
> Ohms, so any mismatch from this value is considered abnormal.

I see, makes sense.

> I can provide a rephrased version if the commit description was not
> detailed enough.

That'd be great. Or maybe even better a short comment above the enum
entry?