[PATCH net-next 0/9] net: dsa: Plug in PHYLINK support

[PATCH net-next 0/9] net: dsa: Plug in PHYLINK support

[Florian Fainelli]

Hi all,

This patch series adds PHYLINK support to DSA which is necessary to support more
complex PHY and pluggable modules setups.

Patch series can be found here:

https://github.com/ffainelli/linux/commits/dsa-phylink

This was tested on:

- dsa-loop
- bcm_sf2
- mv88e6xxx
- b53

With a variety of test cases:
- internal & external MDIO PHYs
- MoCA with link notification through interrupt/MMIO register
- built-in PHYs
- ifconfig up/down for several cycles works
- bind/unbind of the drivers

This is technically v2 of what was posted back in March 2018, changes from last
time:

- fixed probe/remove of drivers
- fixed missing gpiod_put() for link GPIOs
- fixed polling of link GPIOs (Russell I would need your SoB on the patch you
  provided offline initially, added some modifications to it)
- tested across a wider set of platforms

And everything should still work as expected. Please be aware of the following:

- switch drivers (like bcm_sf2) which may have user-facing network ports using
  fixed links would need to implement phylink_mac_ops to remain functional.
  PHYLINK does not create a phy_device for fixed links, therefore our
  call to adjust_link() from phylink_mac_link_{up,down} would not be calling
  into the driver. This *should not* affect CPU/DSA ports which are configured
  through adjust_link() but have no network devices

- support for SFP/SFF is now possible, but switch drivers will still need some
  modifications to properly support those, including, but not limited to using
  the correct binding information. This will be submitted on top of this series

Please do test on your respective platforms/switches and let me know if you
find any issues, hopefully everything still works like before.

Thank you!

Florian Fainelli (7):
  net: phy: phylink: Use gpiod_get_value_cansleep()
  net: phy: phylink: Release link GPIO
  net: dsa: Add PHYLINK switch operations
  net: dsa: bcm_sf2: Implement phylink_mac_ops
  net: dsa: Eliminate dsa_slave_get_link()
  net: dsa: Plug in PHYLINK support
  net: dsa: bcm_sf2: Get rid of PHYLIB functions

Russell King (2):
  net: phy: phylink: Poll link GPIOs
  net: dsa: mv88e6xxx: add PHYLINK support

 drivers/net/dsa/bcm_sf2.c        | 191 ++++++++++++--------
 drivers/net/dsa/mv88e6xxx/chip.c |  81 +++++++++
 drivers/net/dsa/mv88e6xxx/port.c |  39 ++++
 drivers/net/dsa/mv88e6xxx/port.h |   3 +
 drivers/net/phy/phylink.c        |  20 ++-
 include/net/dsa.h                |  25 +++
 net/dsa/Kconfig                  |   2 +-
 net/dsa/dsa_priv.h               |   9 -
 net/dsa/slave.c                  | 293 ++++++++++++++++++-------------
 9 files changed, 456 insertions(+), 207 deletions(-)

-- 
2.17.0

[PATCH net-next 1/9] net: phy: phylink: Use gpiod_get_value_cansleep()

[Florian Fainelli]

The GPIO provider for the link GPIO line might require the use of the
_cansleep() API, utilize that. This is safe to do since we run in workqueue
context.

Fixes: 9525ae83959b ("phylink: add phylink infrastructure")
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
 drivers/net/phy/phylink.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
index c582b2d7546c..412d1cf4fa66 100644
--- a/drivers/net/phy/phylink.c
+++ b/drivers/net/phy/phylink.c
@@ -360,7 +360,7 @@ static void phylink_get_fixed_state(struct phylink *pl, struct phylink_link_stat
 	if (pl->get_fixed_state)
 		pl->get_fixed_state(pl->netdev, state);
 	else if (pl->link_gpio)
-		state->link = !!gpiod_get_value(pl->link_gpio);
+		state->link = !!gpiod_get_value_cansleep(pl->link_gpio);
 }
 
 /* Flow control is resolved according to our and the link partners
-- 
2.17.0

[PATCH net-next 2/9] net: phy: phylink: Release link GPIO

[Florian Fainelli]

We are not releasing the link GPIO descriptor with gpiod_put() which results in
subsequent probing to get -EBUSY when calling fwnode_get_named_gpiod(). Fix this
by doing the release in phylink_destroy().

Fixes: 9525ae83959b ("phylink: add phylink infrastructure")
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
 drivers/net/phy/phylink.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
index 412d1cf4fa66..6392b5248cf5 100644
--- a/drivers/net/phy/phylink.c
+++ b/drivers/net/phy/phylink.c
@@ -612,6 +612,8 @@ void phylink_destroy(struct phylink *pl)
 {
 	if (pl->sfp_bus)
 		sfp_unregister_upstream(pl->sfp_bus);
+	if (!IS_ERR(pl->link_gpio))
+		gpiod_put(pl->link_gpio);
 
 	cancel_work_sync(&pl->resolve);
 	kfree(pl);
-- 
2.17.0

[PATCH net-next 3/9] net: phy: phylink: Poll link GPIOs

[Florian Fainelli]

From: Russell King <rmk+kernel@armlinux.org.uk>

When using a fixed link with a link GPIO, we need to poll that GPIO to
determine link state changes. This is consistent with what fixed_phy.c does.

Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
 drivers/net/phy/phylink.c | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
index 6392b5248cf5..581ce93ecaf9 100644
--- a/drivers/net/phy/phylink.c
+++ b/drivers/net/phy/phylink.c
@@ -19,6 +19,7 @@
 #include <linux/phylink.h>
 #include <linux/rtnetlink.h>
 #include <linux/spinlock.h>
+#include <linux/timer.h>
 #include <linux/workqueue.h>
 
 #include "sfp.h"
@@ -54,6 +55,7 @@ struct phylink {
 	/* The link configuration settings */
 	struct phylink_link_state link_config;
 	struct gpio_desc *link_gpio;
+	struct timer_list link_poll;
 	void (*get_fixed_state)(struct net_device *dev,
 				struct phylink_link_state *s);
 
@@ -500,6 +502,15 @@ static void phylink_run_resolve(struct phylink *pl)
 		queue_work(system_power_efficient_wq, &pl->resolve);
 }
 
+static void phylink_fixed_poll(struct timer_list *t)
+{
+	struct phylink *pl = container_of(t, struct phylink, link_poll);
+
+	mod_timer(t, jiffies + HZ);
+
+	phylink_run_resolve(pl);
+}
+
 static const struct sfp_upstream_ops sfp_phylink_ops;
 
 static int phylink_register_sfp(struct phylink *pl,
@@ -572,6 +583,7 @@ struct phylink *phylink_create(struct net_device *ndev,
 	pl->link_config.an_enabled = true;
 	pl->ops = ops;
 	__set_bit(PHYLINK_DISABLE_STOPPED, &pl->phylink_disable_state);
+	timer_setup(&pl->link_poll, phylink_fixed_poll, 0);
 
 	bitmap_fill(pl->supported, __ETHTOOL_LINK_MODE_MASK_NBITS);
 	linkmode_copy(pl->link_config.advertising, pl->supported);
@@ -905,6 +917,8 @@ void phylink_start(struct phylink *pl)
 	clear_bit(PHYLINK_DISABLE_STOPPED, &pl->phylink_disable_state);
 	phylink_run_resolve(pl);
 
+	if (pl->link_an_mode == MLO_AN_FIXED && !IS_ERR(pl->link_gpio))
+		mod_timer(&pl->link_poll, jiffies + HZ);
 	if (pl->sfp_bus)
 		sfp_upstream_start(pl->sfp_bus);
 	if (pl->phydev)
@@ -929,6 +943,8 @@ void phylink_stop(struct phylink *pl)
 		phy_stop(pl->phydev);
 	if (pl->sfp_bus)
 		sfp_upstream_stop(pl->sfp_bus);
+	if (pl->link_an_mode == MLO_AN_FIXED && !IS_ERR(pl->link_gpio))
+		del_timer_sync(&pl->link_poll);
 
 	set_bit(PHYLINK_DISABLE_STOPPED, &pl->phylink_disable_state);
 	queue_work(system_power_efficient_wq, &pl->resolve);
-- 
2.17.0

[PATCH net-next 4/9] net: dsa: Add PHYLINK switch operations

[Florian Fainelli]

In preparation for adding support for PHYLINK within DSA, define a number of
operations that we will need and that switch drivers can start implementing.
Proper integration with PHYLINK will follow in subsequent patches.

We start selecting PHYLINK (which implies PHYLIB) in net/dsa/Kconfig
such that drivers can be guaranteed that this dependency is properly
taken care of and can start referencing PHYLINK helper functions without
requiring stubs or anything.

Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
 include/net/dsa.h | 24 ++++++++++++++++++++++++
 net/dsa/Kconfig   |  2 +-
 net/dsa/slave.c   |  5 +++++
 3 files changed, 30 insertions(+), 1 deletion(-)

diff --git a/include/net/dsa.h b/include/net/dsa.h
index 462e9741b210..ed64c1f3f117 100644
--- a/include/net/dsa.h
+++ b/include/net/dsa.h
@@ -20,12 +20,14 @@
 #include <linux/of.h>
 #include <linux/ethtool.h>
 #include <linux/net_tstamp.h>
+#include <linux/phy.h>
 #include <net/devlink.h>
 #include <net/switchdev.h>
 
 struct tc_action;
 struct phy_device;
 struct fixed_phy_status;
+struct phylink_link_state;
 
 enum dsa_tag_protocol {
 	DSA_TAG_PROTO_NONE = 0,
@@ -353,6 +355,27 @@ struct dsa_switch_ops {
 	void	(*fixed_link_update)(struct dsa_switch *ds, int port,
 				struct fixed_phy_status *st);
 
+	/*
+	 * PHYLINK integration
+	 */
+	void	(*phylink_validate)(struct dsa_switch *ds, int port,
+				    unsigned long *supported,
+				    struct phylink_link_state *state);
+	int	(*phylink_mac_link_state)(struct dsa_switch *ds, int port,
+					  struct phylink_link_state *state);
+	void	(*phylink_mac_config)(struct dsa_switch *ds, int port,
+				      unsigned int mode,
+				      const struct phylink_link_state *state);
+	void	(*phylink_mac_an_restart)(struct dsa_switch *ds, int port);
+	void	(*phylink_mac_link_down)(struct dsa_switch *ds, int port,
+					 unsigned int mode,
+					 phy_interface_t interface);
+	void	(*phylink_mac_link_up)(struct dsa_switch *ds, int port,
+				       unsigned int mode,
+				       phy_interface_t interface,
+				       struct phy_device *phydev);
+	void	(*phylink_fixed_state)(struct dsa_switch *ds, int port,
+				       struct phylink_link_state *state);
 	/*
 	 * ethtool hardware statistics.
 	 */
@@ -595,5 +618,6 @@ static inline int call_dsa_notifiers(unsigned long val, struct net_device *dev,
 int dsa_port_get_phy_strings(struct dsa_port *dp, uint8_t *data);
 int dsa_port_get_ethtool_phy_stats(struct dsa_port *dp, uint64_t *data);
 int dsa_port_get_phy_sset_count(struct dsa_port *dp);
+void dsa_port_phylink_mac_change(struct dsa_switch *ds, int port, bool up);
 
 #endif
diff --git a/net/dsa/Kconfig b/net/dsa/Kconfig
index bbf2c82cf7b2..4183e4ba27a5 100644
--- a/net/dsa/Kconfig
+++ b/net/dsa/Kconfig
@@ -9,7 +9,7 @@ config NET_DSA
 	depends on HAVE_NET_DSA && MAY_USE_DEVLINK
 	depends on BRIDGE || BRIDGE=n
 	select NET_SWITCHDEV
-	select PHYLIB
+	select PHYLINK
 	---help---
 	  Say Y if you want to enable support for the hardware switches supported
 	  by the Distributed Switch Architecture.
diff --git a/net/dsa/slave.c b/net/dsa/slave.c
index f3fb3a0880b1..e9be9458aa94 100644
--- a/net/dsa/slave.c
+++ b/net/dsa/slave.c
@@ -1119,6 +1119,11 @@ static int dsa_slave_phy_connect(struct net_device *slave_dev, int addr)
 				  dsa_slave_adjust_link, p->phy_interface);
 }
 
+void dsa_port_phylink_mac_change(struct dsa_switch *ds, int port, bool up)
+{
+}
+EXPORT_SYMBOL_GPL(dsa_port_phylink_mac_change);
+
 static int dsa_slave_phy_setup(struct net_device *slave_dev)
 {
 	struct dsa_port *dp = dsa_slave_to_port(slave_dev);
-- 
2.17.0

[PATCH net-next 5/9] net: dsa: bcm_sf2: Implement phylink_mac_ops

[Florian Fainelli]

Make the bcm_sf2 driver implement phylink_mac_ops since it needs to
support a wide variety of network interfaces: internal & external MDIO
PHYs, fixed PHYs, MoCA with MMIO link status.

A large amount of what needs to be done already exists under
bcm_sf2_sw_adjust_link() so we are essentially breaking this down into
the necessary operation for PHYLINK to work: mac_config, mac_link_up,
mac_link_down and validate. We can now entirely get rid of most of what
fixed_link_update() provided because only the link information is actually
necessary. We still have to force DUPLEX_FULL for legacy Device Tree bindings
that did not specify that before.

Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
 drivers/net/dsa/bcm_sf2.c | 214 ++++++++++++++++++++++++++++++++++++--
 1 file changed, 206 insertions(+), 8 deletions(-)

diff --git a/drivers/net/dsa/bcm_sf2.c b/drivers/net/dsa/bcm_sf2.c
index 97236cfcbae4..a20608b0329e 100644
--- a/drivers/net/dsa/bcm_sf2.c
+++ b/drivers/net/dsa/bcm_sf2.c
@@ -16,6 +16,7 @@
 #include <linux/platform_device.h>
 #include <linux/phy.h>
 #include <linux/phy_fixed.h>
+#include <linux/phylink.h>
 #include <linux/mii.h>
 #include <linux/of.h>
 #include <linux/of_irq.h>
@@ -306,7 +307,8 @@ static int bcm_sf2_sw_mdio_write(struct mii_bus *bus, int addr, int regnum,
 
 static irqreturn_t bcm_sf2_switch_0_isr(int irq, void *dev_id)
 {
-	struct bcm_sf2_priv *priv = dev_id;
+	struct dsa_switch *ds = dev_id;
+	struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds);
 
 	priv->irq0_stat = intrl2_0_readl(priv, INTRL2_CPU_STATUS) &
 				~priv->irq0_mask;
@@ -317,16 +319,21 @@ static irqreturn_t bcm_sf2_switch_0_isr(int irq, void *dev_id)
 
 static irqreturn_t bcm_sf2_switch_1_isr(int irq, void *dev_id)
 {
-	struct bcm_sf2_priv *priv = dev_id;
+	struct dsa_switch *ds = dev_id;
+	struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds);
 
 	priv->irq1_stat = intrl2_1_readl(priv, INTRL2_CPU_STATUS) &
 				~priv->irq1_mask;
 	intrl2_1_writel(priv, priv->irq1_stat, INTRL2_CPU_CLEAR);
 
-	if (priv->irq1_stat & P_LINK_UP_IRQ(P7_IRQ_OFF))
-		priv->port_sts[7].link = 1;
-	if (priv->irq1_stat & P_LINK_DOWN_IRQ(P7_IRQ_OFF))
-		priv->port_sts[7].link = 0;
+	if (priv->irq1_stat & P_LINK_UP_IRQ(P7_IRQ_OFF)) {
+		priv->port_sts[7].link = true;
+		dsa_port_phylink_mac_change(ds, 7, true);
+	}
+	if (priv->irq1_stat & P_LINK_DOWN_IRQ(P7_IRQ_OFF)) {
+		priv->port_sts[7].link = false;
+		dsa_port_phylink_mac_change(ds, 7, false);
+	}
 
 	return IRQ_HANDLED;
 }
@@ -620,6 +627,192 @@ static void bcm_sf2_sw_fixed_link_update(struct dsa_switch *ds, int port,
 		status->pause = 1;
 }
 
+static void bcm_sf2_sw_validate(struct dsa_switch *ds, int port,
+				unsigned long *supported,
+				struct phylink_link_state *state)
+{
+	__ETHTOOL_DECLARE_LINK_MODE_MASK(mask) = { 0, };
+
+	if (!phy_interface_mode_is_rgmii(state->interface) &&
+	    state->interface != PHY_INTERFACE_MODE_MII &&
+	    state->interface != PHY_INTERFACE_MODE_REVMII &&
+	    state->interface != PHY_INTERFACE_MODE_GMII &&
+	    state->interface != PHY_INTERFACE_MODE_INTERNAL &&
+	    state->interface != PHY_INTERFACE_MODE_MOCA) {
+		bitmap_zero(supported, __ETHTOOL_LINK_MODE_MASK_NBITS);
+		dev_err(ds->dev,
+			"Unsupported interface: %d\n", state->interface);
+		return;
+	}
+
+	/* Allow all the expected bits */
+	phylink_set(mask, Autoneg);
+	phylink_set_port_modes(mask);
+	phylink_set(mask, Pause);
+	phylink_set(mask, Asym_Pause);
+
+	/* With the exclusion of MII and Reverse MII, we support Gigabit,
+	 * including Half duplex
+	 */
+	if (state->interface != PHY_INTERFACE_MODE_MII &&
+	    state->interface != PHY_INTERFACE_MODE_REVMII) {
+		phylink_set(mask, 1000baseT_Full);
+		phylink_set(mask, 1000baseT_Half);
+	}
+
+	phylink_set(mask, 10baseT_Half);
+	phylink_set(mask, 10baseT_Full);
+	phylink_set(mask, 100baseT_Half);
+	phylink_set(mask, 100baseT_Full);
+
+	bitmap_and(supported, supported, mask,
+		   __ETHTOOL_LINK_MODE_MASK_NBITS);
+	bitmap_and(state->advertising, state->advertising, mask,
+		   __ETHTOOL_LINK_MODE_MASK_NBITS);
+}
+
+static void bcm_sf2_sw_mac_config(struct dsa_switch *ds, int port,
+				  unsigned int mode,
+				  const struct phylink_link_state *state)
+{
+	struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds);
+	u32 id_mode_dis = 0, port_mode;
+	u32 reg, offset;
+
+	if (priv->type == BCM7445_DEVICE_ID)
+		offset = CORE_STS_OVERRIDE_GMIIP_PORT(port);
+	else
+		offset = CORE_STS_OVERRIDE_GMIIP2_PORT(port);
+
+	switch (state->interface) {
+	case PHY_INTERFACE_MODE_RGMII:
+		id_mode_dis = 1;
+		/* fallthrough */
+	case PHY_INTERFACE_MODE_RGMII_TXID:
+		port_mode = EXT_GPHY;
+		break;
+	case PHY_INTERFACE_MODE_MII:
+		port_mode = EXT_EPHY;
+		break;
+	case PHY_INTERFACE_MODE_REVMII:
+		port_mode = EXT_REVMII;
+		break;
+	default:
+		/* all other PHYs: internal and MoCA */
+		goto force_link;
+	}
+
+	/* Clear id_mode_dis bit, and the existing port mode, let
+	 * RGMII_MODE_EN bet set by mac_link_{up,down}
+	 */
+	reg = reg_readl(priv, REG_RGMII_CNTRL_P(port));
+	reg &= ~ID_MODE_DIS;
+	reg &= ~(PORT_MODE_MASK << PORT_MODE_SHIFT);
+	reg &= ~(RX_PAUSE_EN | TX_PAUSE_EN);
+
+	reg |= port_mode;
+	if (id_mode_dis)
+		reg |= ID_MODE_DIS;
+
+	if (state->pause & MLO_PAUSE_TXRX_MASK) {
+		if (state->pause & MLO_PAUSE_TX)
+			reg |= TX_PAUSE_EN;
+		reg |= RX_PAUSE_EN;
+	}
+
+	reg_writel(priv, reg, REG_RGMII_CNTRL_P(port));
+
+force_link:
+	/* Force link settings detected from the PHY */
+	reg = SW_OVERRIDE;
+	switch (state->speed) {
+	case SPEED_1000:
+		reg |= SPDSTS_1000 << SPEED_SHIFT;
+		break;
+	case SPEED_100:
+		reg |= SPDSTS_100 << SPEED_SHIFT;
+		break;
+	}
+
+	if (state->link)
+		reg |= LINK_STS;
+	if (state->duplex == DUPLEX_FULL)
+		reg |= DUPLX_MODE;
+
+	core_writel(priv, reg, offset);
+}
+
+static void bcm_sf2_sw_mac_link_set(struct dsa_switch *ds, int port,
+				    phy_interface_t interface, bool link)
+{
+	struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds);
+	u32 reg;
+
+	if (!phy_interface_mode_is_rgmii(interface) &&
+	    interface != PHY_INTERFACE_MODE_MII &&
+	    interface != PHY_INTERFACE_MODE_REVMII)
+		return;
+
+	/* If the link is down, just disable the interface to conserve power */
+	reg = reg_readl(priv, REG_RGMII_CNTRL_P(port));
+	if (link)
+		reg |= RGMII_MODE_EN;
+	else
+		reg &= ~RGMII_MODE_EN;
+	reg_writel(priv, reg, REG_RGMII_CNTRL_P(port));
+}
+
+static void bcm_sf2_sw_mac_link_down(struct dsa_switch *ds, int port,
+				     unsigned int mode,
+				     phy_interface_t interface)
+{
+	bcm_sf2_sw_mac_link_set(ds, port, interface, false);
+}
+
+static void bcm_sf2_sw_mac_link_up(struct dsa_switch *ds, int port,
+				   unsigned int mode,
+				   phy_interface_t interface,
+				   struct phy_device *phydev)
+{
+	struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds);
+	struct ethtool_eee *p = &priv->dev->ports[port].eee;
+
+	bcm_sf2_sw_mac_link_set(ds, port, interface, true);
+
+	if (mode == MLO_AN_PHY && phydev)
+		p->eee_enabled = b53_eee_init(ds, port, phydev);
+}
+
+static void bcm_sf2_sw_fixed_state(struct dsa_switch *ds, int port,
+				   struct phylink_link_state *status)
+{
+	struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds);
+
+	status->link = false;
+
+	/* MoCA port is special as we do not get link status from CORE_LNKSTS,
+	 * which means that we need to force the link at the port override
+	 * level to get the data to flow. We do use what the interrupt handler
+	 * did determine before.
+	 *
+	 * For the other ports, we just force the link status, since this is
+	 * a fixed PHY device.
+	 */
+	if (port == priv->moca_port) {
+		status->link = priv->port_sts[port].link;
+		/* For MoCA interfaces, also force a link down notification
+		 * since some version of the user-space daemon (mocad) use
+		 * cmd->autoneg to force the link, which messes up the PHY
+		 * state machine and make it go in PHY_FORCING state instead.
+		 */
+		if (!status->link)
+			netif_carrier_off(ds->ports[port].slave);
+		status->duplex = DUPLEX_FULL;
+	} else {
+		status->link = true;
+	}
+}
+
 static void bcm_sf2_enable_acb(struct dsa_switch *ds)
 {
 	struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds);
@@ -863,6 +1056,11 @@ static const struct dsa_switch_ops bcm_sf2_ops = {
 	.get_phy_flags		= bcm_sf2_sw_get_phy_flags,
 	.adjust_link		= bcm_sf2_sw_adjust_link,
 	.fixed_link_update	= bcm_sf2_sw_fixed_link_update,
+	.phylink_validate	= bcm_sf2_sw_validate,
+	.phylink_mac_config	= bcm_sf2_sw_mac_config,
+	.phylink_mac_link_down	= bcm_sf2_sw_mac_link_down,
+	.phylink_mac_link_up	= bcm_sf2_sw_mac_link_up,
+	.phylink_fixed_state	= bcm_sf2_sw_fixed_state,
 	.suspend		= bcm_sf2_sw_suspend,
 	.resume			= bcm_sf2_sw_resume,
 	.get_wol		= bcm_sf2_sw_get_wol,
@@ -1065,14 +1263,14 @@ static int bcm_sf2_sw_probe(struct platform_device *pdev)
 	bcm_sf2_intr_disable(priv);
 
 	ret = devm_request_irq(&pdev->dev, priv->irq0, bcm_sf2_switch_0_isr, 0,
-			       "switch_0", priv);
+			       "switch_0", ds);
 	if (ret < 0) {
 		pr_err("failed to request switch_0 IRQ\n");
 		goto out_mdio;
 	}
 
 	ret = devm_request_irq(&pdev->dev, priv->irq1, bcm_sf2_switch_1_isr, 0,
-			       "switch_1", priv);
+			       "switch_1", ds);
 	if (ret < 0) {
 		pr_err("failed to request switch_1 IRQ\n");
 		goto out_mdio;
-- 
2.17.0

[PATCH net-next 6/9] net: dsa: Eliminate dsa_slave_get_link()

[Florian Fainelli]

Since we use PHYLIB to manage the per-port link indication, this will
also be reflected correctly in the network device's carrier state, so we
can use ethtool_op_get_link() instead.

Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
 net/dsa/slave.c | 12 +-----------
 1 file changed, 1 insertion(+), 11 deletions(-)

diff --git a/net/dsa/slave.c b/net/dsa/slave.c
index e9be9458aa94..e78894477598 100644
--- a/net/dsa/slave.c
+++ b/net/dsa/slave.c
@@ -498,16 +498,6 @@ dsa_slave_get_regs(struct net_device *dev, struct ethtool_regs *regs, void *_p)
 		ds->ops->get_regs(ds, dp->index, regs, _p);
 }
 
-static u32 dsa_slave_get_link(struct net_device *dev)
-{
-	if (!dev->phydev)
-		return -ENODEV;
-
-	genphy_update_link(dev->phydev);
-
-	return dev->phydev->link;
-}
-
 static int dsa_slave_get_eeprom_len(struct net_device *dev)
 {
 	struct dsa_port *dp = dsa_slave_to_port(dev);
@@ -982,7 +972,7 @@ static const struct ethtool_ops dsa_slave_ethtool_ops = {
 	.get_regs_len		= dsa_slave_get_regs_len,
 	.get_regs		= dsa_slave_get_regs,
 	.nway_reset		= phy_ethtool_nway_reset,
-	.get_link		= dsa_slave_get_link,
+	.get_link		= ethtool_op_get_link,
 	.get_eeprom_len		= dsa_slave_get_eeprom_len,
 	.get_eeprom		= dsa_slave_get_eeprom,
 	.set_eeprom		= dsa_slave_set_eeprom,
-- 
2.17.0

[PATCH net-next 7/9] net: dsa: mv88e6xxx: add PHYLINK support

[Florian Fainelli]

From: Russell King <rmk+kernel@armlinux.org.uk>

Add rudimentary phylink support to mv88e6xxx. This allows the driver
using user ports with fixed links to keep operating normally. User ports
with normal PHYs are not affected since the switch automatically manages
their link parameters. User facing ports which use a SFP/SFF with a
non-fixed link mode might require a call to phylink_mac_change() to
operate properly.

Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
[florian: expand commit message]
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
 drivers/net/dsa/mv88e6xxx/chip.c | 81 ++++++++++++++++++++++++++++++++
 drivers/net/dsa/mv88e6xxx/port.c | 39 +++++++++++++++
 drivers/net/dsa/mv88e6xxx/port.h |  3 ++
 3 files changed, 123 insertions(+)

diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c
index 9d62e4acc01b..f4e8a3469386 100644
--- a/drivers/net/dsa/mv88e6xxx/chip.c
+++ b/drivers/net/dsa/mv88e6xxx/chip.c
@@ -31,6 +31,7 @@
 #include <linux/netdevice.h>
 #include <linux/gpio/consumer.h>
 #include <linux/phy.h>
+#include <linux/phylink.h>
 #include <net/dsa.h>
 
 #include "chip.h"
@@ -580,6 +581,81 @@ static void mv88e6xxx_adjust_link(struct dsa_switch *ds, int port,
 		dev_err(ds->dev, "p%d: failed to configure MAC\n", port);
 }
 
+static void mv88e6xxx_validate(struct dsa_switch *ds, int port,
+			       unsigned long *supported,
+			       struct phylink_link_state *state)
+{
+}
+
+static int mv88e6xxx_link_state(struct dsa_switch *ds, int port,
+				struct phylink_link_state *state)
+{
+	struct mv88e6xxx_chip *chip = ds->priv;
+	int err;
+
+	mutex_lock(&chip->reg_lock);
+	err = mv88e6xxx_port_link_state(chip, port, state);
+	mutex_unlock(&chip->reg_lock);
+
+	return err;
+}
+
+static void mv88e6xxx_mac_config(struct dsa_switch *ds, int port,
+				 unsigned int mode,
+				 const struct phylink_link_state *state)
+{
+	struct mv88e6xxx_chip *chip = ds->priv;
+	int speed, duplex, err;
+
+	if (mode == MLO_AN_PHY)
+		return;
+
+	if (mode != MLO_AN_INBAND) {
+		speed = state->speed;
+		duplex = state->duplex;
+	} else {
+		speed = SPEED_UNFORCED;
+		duplex = DUPLEX_UNFORCED;
+	}
+
+	mutex_lock(&chip->reg_lock);
+	err = mv88e6xxx_port_setup_mac(chip, port, LINK_UNFORCED, speed, duplex,
+				       state->interface);
+	mutex_unlock(&chip->reg_lock);
+
+	if (err && err != -EOPNOTSUPP)
+		dev_err(ds->dev, "p%d: failed to configure MAC\n", port);
+}
+
+static void mv88e6xxx_mac_link_force(struct dsa_switch *ds, int port, int link)
+{
+	struct mv88e6xxx_chip *chip = ds->priv;
+	int err;
+
+	mutex_lock(&chip->reg_lock);
+	err = chip->info->ops->port_set_link(chip, port, link);
+	mutex_unlock(&chip->reg_lock);
+
+	if (err)
+		dev_err(chip->dev, "p%d: failed to force MAC link\n", port);
+}
+
+static void mv88e6xxx_mac_link_down(struct dsa_switch *ds, int port,
+				    unsigned int mode,
+				    phy_interface_t interface)
+{
+	if (mode == MLO_AN_FIXED)
+		mv88e6xxx_mac_link_force(ds, port, LINK_FORCED_DOWN);
+}
+
+static void mv88e6xxx_mac_link_up(struct dsa_switch *ds, int port,
+				  unsigned int mode, phy_interface_t interface,
+				  struct phy_device *phydev)
+{
+	if (mode == MLO_AN_FIXED)
+		mv88e6xxx_mac_link_force(ds, port, LINK_FORCED_UP);
+}
+
 static int mv88e6xxx_stats_snapshot(struct mv88e6xxx_chip *chip, int port)
 {
 	if (!chip->info->ops->stats_snapshot)
@@ -4138,6 +4214,11 @@ static const struct dsa_switch_ops mv88e6xxx_switch_ops = {
 	.get_tag_protocol	= mv88e6xxx_get_tag_protocol,
 	.setup			= mv88e6xxx_setup,
 	.adjust_link		= mv88e6xxx_adjust_link,
+	.phylink_validate	= mv88e6xxx_validate,
+	.phylink_mac_link_state	= mv88e6xxx_link_state,
+	.phylink_mac_config	= mv88e6xxx_mac_config,
+	.phylink_mac_link_down	= mv88e6xxx_mac_link_down,
+	.phylink_mac_link_up	= mv88e6xxx_mac_link_up,
 	.get_strings		= mv88e6xxx_get_strings,
 	.get_ethtool_stats	= mv88e6xxx_get_ethtool_stats,
 	.get_sset_count		= mv88e6xxx_get_sset_count,
diff --git a/drivers/net/dsa/mv88e6xxx/port.c b/drivers/net/dsa/mv88e6xxx/port.c
index 6315774d72b3..429d0ebcd5b1 100644
--- a/drivers/net/dsa/mv88e6xxx/port.c
+++ b/drivers/net/dsa/mv88e6xxx/port.c
@@ -15,6 +15,7 @@
 #include <linux/bitfield.h>
 #include <linux/if_bridge.h>
 #include <linux/phy.h>
+#include <linux/phylink.h>
 
 #include "chip.h"
 #include "port.h"
@@ -378,6 +379,44 @@ int mv88e6xxx_port_get_cmode(struct mv88e6xxx_chip *chip, int port, u8 *cmode)
 	return 0;
 }
 
+int mv88e6xxx_port_link_state(struct mv88e6xxx_chip *chip, int port,
+			      struct phylink_link_state *state)
+{
+	int err;
+	u16 reg;
+
+	err = mv88e6xxx_port_read(chip, port, MV88E6XXX_PORT_STS, &reg);
+	if (err)
+		return err;
+
+	switch (reg & MV88E6XXX_PORT_STS_SPEED_MASK) {
+	case MV88E6XXX_PORT_STS_SPEED_10:
+		state->speed = SPEED_10;
+		break;
+	case MV88E6XXX_PORT_STS_SPEED_100:
+		state->speed = SPEED_100;
+		break;
+	case MV88E6XXX_PORT_STS_SPEED_1000:
+		state->speed = SPEED_1000;
+		break;
+	case MV88E6XXX_PORT_STS_SPEED_10000:
+		if ((reg &MV88E6XXX_PORT_STS_CMODE_MASK) ==
+		    MV88E6XXX_PORT_STS_CMODE_2500BASEX)
+			state->speed = SPEED_2500;
+		else
+			state->speed = SPEED_10000;
+		break;
+	}
+
+	state->duplex = reg & MV88E6XXX_PORT_STS_DUPLEX ?
+			DUPLEX_FULL : DUPLEX_HALF;
+	state->link = !!(reg & MV88E6XXX_PORT_STS_LINK);
+	state->an_enabled = 1;
+	state->an_complete = state->link;
+
+	return 0;
+}
+
 /* Offset 0x02: Jamming Control
  *
  * Do not limit the period of time that this port can be paused for by
diff --git a/drivers/net/dsa/mv88e6xxx/port.h b/drivers/net/dsa/mv88e6xxx/port.h
index b16d5f0e6e9c..5e1db1b221ca 100644
--- a/drivers/net/dsa/mv88e6xxx/port.h
+++ b/drivers/net/dsa/mv88e6xxx/port.h
@@ -29,6 +29,7 @@
 #define MV88E6XXX_PORT_STS_SPEED_10		0x0000
 #define MV88E6XXX_PORT_STS_SPEED_100		0x0100
 #define MV88E6XXX_PORT_STS_SPEED_1000		0x0200
+#define MV88E6XXX_PORT_STS_SPEED_10000		0x0300
 #define MV88E6352_PORT_STS_EEE			0x0040
 #define MV88E6165_PORT_STS_AM_DIS		0x0040
 #define MV88E6185_PORT_STS_MGMII		0x0040
@@ -295,6 +296,8 @@ int mv88e6390_port_pause_limit(struct mv88e6xxx_chip *chip, int port, u8 in,
 int mv88e6390x_port_set_cmode(struct mv88e6xxx_chip *chip, int port,
 			      phy_interface_t mode);
 int mv88e6xxx_port_get_cmode(struct mv88e6xxx_chip *chip, int port, u8 *cmode);
+int mv88e6xxx_port_link_state(struct mv88e6xxx_chip *chip, int port,
+			      struct phylink_link_state *state);
 int mv88e6xxx_port_set_map_da(struct mv88e6xxx_chip *chip, int port);
 int mv88e6095_port_set_upstream_port(struct mv88e6xxx_chip *chip, int port,
 				     int upstream_port);
-- 
2.17.0

Re: [PATCH net-next 7/9] net: dsa: mv88e6xxx: add PHYLINK support

[Andrew Lunn]

On Sat, May 05, 2018 at 12:04:23PM -0700, Florian Fainelli wrote:
> From: Russell King <rmk+kernel@armlinux.org.uk>
> 
> Add rudimentary phylink support to mv88e6xxx. This allows the driver
> using user ports with fixed links to keep operating normally. User ports
> with normal PHYs are not affected since the switch automatically manages
> their link parameters. User facing ports which use a SFP/SFF with a
> non-fixed link mode might require a call to phylink_mac_change() to
> operate properly.

Hi Florian

I have a regression with this patch on ZII devel B, and i think a fix.
I'm running some more tests now. Once they pass, i will post a patch.

    Andrew

Re: [PATCH net-next 7/9] net: dsa: mv88e6xxx: add PHYLINK support

[Florian Fainelli]

On May 6, 2018 10:26:37 AM PDT, Andrew Lunn <andrew@lunn.ch> wrote:
>On Sat, May 05, 2018 at 12:04:23PM -0700, Florian Fainelli wrote:
>> From: Russell King <rmk+kernel@armlinux.org.uk>
>> 
>> Add rudimentary phylink support to mv88e6xxx. This allows the driver
>> using user ports with fixed links to keep operating normally. User
>ports
>> with normal PHYs are not affected since the switch automatically
>manages
>> their link parameters. User facing ports which use a SFP/SFF with a
>> non-fixed link mode might require a call to phylink_mac_change() to
>> operate properly.
>
>Hi Florian
>
>I have a regression with this patch on ZII devel B, and i think a fix.
>I'm running some more tests now. Once they pass, i will post a patch.

Thanks for giving this a spin, let me know what the results are. Things worked fine here with optical4 and a 1000basex media converter though that thing tends to be finicky...
Hi Andrew,
-- 
Florian

[PATCH net-next 8/9] net: dsa: Plug in PHYLINK support

[Florian Fainelli]

Add support for PHYLINK within the DSA subsystem in order to support more
complex devices such as pluggable (SFP) and non-pluggable (SFF) modules, 10G
PHYs, and traditional PHYs. Using PHYLINK allows us to drop some amount of
complexity we had while probing fixed and non-fixed PHYs using Device Tree.

Because PHYLINK separates the Ethernet MAC/port configuration into different
stages, we let switch drivers implement those, and for now, we maintain
functionality by calling dsa_slave_adjust_link() during
phylink_mac_link_{up,down} which provides semantically equivalent steps.

Drivers willing to take advantage of PHYLINK should implement the phylink_mac_*
operations that DSA wraps.

We cannot quite remove the adjust_link() callback just yet, because a number of
drivers rely on that for configuring their "CPU" and "DSA" ports, this is done
dsa_port_setup_phy_of() and dsa_port_fixed_link_register_of() still.

Drivers that utilize fixed links for user-facing ports (e.g: bcm_sf2) will need
to implement phylink_mac_ops from now on to preserve functionality, since PHYLINK
*does not* create a phy_device instance for fixed links.

Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
 include/net/dsa.h  |   1 +
 net/dsa/dsa_priv.h |   9 --
 net/dsa/slave.c    | 294 ++++++++++++++++++++++++++-------------------
 3 files changed, 172 insertions(+), 132 deletions(-)

diff --git a/include/net/dsa.h b/include/net/dsa.h
index ed64c1f3f117..fdbd6082945d 100644
--- a/include/net/dsa.h
+++ b/include/net/dsa.h
@@ -201,6 +201,7 @@ struct dsa_port {
 	u8			stp_state;
 	struct net_device	*bridge_dev;
 	struct devlink_port	devlink_port;
+	struct phylink		*pl;
 	/*
 	 * Original copy of the master netdev ethtool_ops
 	 */
diff --git a/net/dsa/dsa_priv.h b/net/dsa/dsa_priv.h
index 053731473c99..3964c6f7a7c0 100644
--- a/net/dsa/dsa_priv.h
+++ b/net/dsa/dsa_priv.h
@@ -75,15 +75,6 @@ struct dsa_slave_priv {
 	/* DSA port data, such as switch, port index, etc. */
 	struct dsa_port		*dp;
 
-	/*
-	 * The phylib phy_device pointer for the PHY connected
-	 * to this port.
-	 */
-	phy_interface_t		phy_interface;
-	int			old_link;
-	int			old_pause;
-	int			old_duplex;
-
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	struct netpoll		*netpoll;
 #endif
diff --git a/net/dsa/slave.c b/net/dsa/slave.c
index e78894477598..4a9abc66132f 100644
--- a/net/dsa/slave.c
+++ b/net/dsa/slave.c
@@ -13,6 +13,7 @@
 #include <linux/netdevice.h>
 #include <linux/phy.h>
 #include <linux/phy_fixed.h>
+#include <linux/phylink.h>
 #include <linux/of_net.h>
 #include <linux/of_mdio.h>
 #include <linux/mdio.h>
@@ -97,8 +98,7 @@ static int dsa_slave_open(struct net_device *dev)
 	if (err)
 		goto clear_promisc;
 
-	if (dev->phydev)
-		phy_start(dev->phydev);
+	phylink_start(dp->pl);
 
 	return 0;
 
@@ -120,8 +120,7 @@ static int dsa_slave_close(struct net_device *dev)
 	struct net_device *master = dsa_slave_to_master(dev);
 	struct dsa_port *dp = dsa_slave_to_port(dev);
 
-	if (dev->phydev)
-		phy_stop(dev->phydev);
+	phylink_stop(dp->pl);
 
 	dsa_port_disable(dp, dev->phydev);
 
@@ -272,10 +271,7 @@ static int dsa_slave_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 		break;
 	}
 
-	if (!dev->phydev)
-		return -ENODEV;
-
-	return phy_mii_ioctl(dev->phydev, ifr, cmd);
+	return phylink_mii_ioctl(p->dp->pl, ifr, cmd);
 }
 
 static int dsa_slave_port_attr_set(struct net_device *dev,
@@ -498,6 +494,13 @@ dsa_slave_get_regs(struct net_device *dev, struct ethtool_regs *regs, void *_p)
 		ds->ops->get_regs(ds, dp->index, regs, _p);
 }
 
+static int dsa_slave_nway_reset(struct net_device *dev)
+{
+	struct dsa_port *dp = dsa_slave_to_port(dev);
+
+	return phylink_ethtool_nway_reset(dp->pl);
+}
+
 static int dsa_slave_get_eeprom_len(struct net_device *dev)
 {
 	struct dsa_port *dp = dsa_slave_to_port(dev);
@@ -609,6 +612,8 @@ static void dsa_slave_get_wol(struct net_device *dev, struct ethtool_wolinfo *w)
 	struct dsa_port *dp = dsa_slave_to_port(dev);
 	struct dsa_switch *ds = dp->ds;
 
+	phylink_ethtool_get_wol(dp->pl, w);
+
 	if (ds->ops->get_wol)
 		ds->ops->get_wol(ds, dp->index, w);
 }
@@ -619,6 +624,8 @@ static int dsa_slave_set_wol(struct net_device *dev, struct ethtool_wolinfo *w)
 	struct dsa_switch *ds = dp->ds;
 	int ret = -EOPNOTSUPP;
 
+	phylink_ethtool_set_wol(dp->pl, w);
+
 	if (ds->ops->set_wol)
 		ret = ds->ops->set_wol(ds, dp->index, w);
 
@@ -642,13 +649,7 @@ static int dsa_slave_set_eee(struct net_device *dev, struct ethtool_eee *e)
 	if (ret)
 		return ret;
 
-	if (e->eee_enabled) {
-		ret = phy_init_eee(dev->phydev, 0);
-		if (ret)
-			return ret;
-	}
-
-	return phy_ethtool_set_eee(dev->phydev, e);
+	return phylink_ethtool_set_eee(dp->pl, e);
 }
 
 static int dsa_slave_get_eee(struct net_device *dev, struct ethtool_eee *e)
@@ -668,7 +669,23 @@ static int dsa_slave_get_eee(struct net_device *dev, struct ethtool_eee *e)
 	if (ret)
 		return ret;
 
-	return phy_ethtool_get_eee(dev->phydev, e);
+	return phylink_ethtool_get_eee(dp->pl, e);
+}
+
+static int dsa_slave_get_link_ksettings(struct net_device *dev,
+					struct ethtool_link_ksettings *cmd)
+{
+	struct dsa_port *dp = dsa_slave_to_port(dev);
+
+	return phylink_ethtool_ksettings_get(dp->pl, cmd);
+}
+
+static int dsa_slave_set_link_ksettings(struct net_device *dev,
+					const struct ethtool_link_ksettings *cmd)
+{
+	struct dsa_port *dp = dsa_slave_to_port(dev);
+
+	return phylink_ethtool_ksettings_set(dp->pl, cmd);
 }
 
 #ifdef CONFIG_NET_POLL_CONTROLLER
@@ -971,7 +988,7 @@ static const struct ethtool_ops dsa_slave_ethtool_ops = {
 	.get_drvinfo		= dsa_slave_get_drvinfo,
 	.get_regs_len		= dsa_slave_get_regs_len,
 	.get_regs		= dsa_slave_get_regs,
-	.nway_reset		= phy_ethtool_nway_reset,
+	.nway_reset		= dsa_slave_nway_reset,
 	.get_link		= ethtool_op_get_link,
 	.get_eeprom_len		= dsa_slave_get_eeprom_len,
 	.get_eeprom		= dsa_slave_get_eeprom,
@@ -983,8 +1000,8 @@ static const struct ethtool_ops dsa_slave_ethtool_ops = {
 	.get_wol		= dsa_slave_get_wol,
 	.set_eee		= dsa_slave_set_eee,
 	.get_eee		= dsa_slave_get_eee,
-	.get_link_ksettings	= phy_ethtool_get_link_ksettings,
-	.set_link_ksettings	= phy_ethtool_set_link_ksettings,
+	.get_link_ksettings	= dsa_slave_get_link_ksettings,
+	.set_link_ksettings	= dsa_slave_set_link_ksettings,
 	.get_rxnfc		= dsa_slave_get_rxnfc,
 	.set_rxnfc		= dsa_slave_set_rxnfc,
 	.get_ts_info		= dsa_slave_get_ts_info,
@@ -1043,56 +1060,122 @@ static struct device_type dsa_type = {
 	.name	= "dsa",
 };
 
-static void dsa_slave_adjust_link(struct net_device *dev)
+static void dsa_slave_phylink_validate(struct net_device *dev,
+				       unsigned long *supported,
+				       struct phylink_link_state *state)
 {
 	struct dsa_port *dp = dsa_slave_to_port(dev);
-	struct dsa_slave_priv *p = netdev_priv(dev);
 	struct dsa_switch *ds = dp->ds;
-	unsigned int status_changed = 0;
 
-	if (p->old_link != dev->phydev->link) {
-		status_changed = 1;
-		p->old_link = dev->phydev->link;
-	}
+	if (!ds->ops->phylink_validate)
+		return;
 
-	if (p->old_duplex != dev->phydev->duplex) {
-		status_changed = 1;
-		p->old_duplex = dev->phydev->duplex;
-	}
+	ds->ops->phylink_validate(ds, dp->index, supported, state);
+}
 
-	if (p->old_pause != dev->phydev->pause) {
-		status_changed = 1;
-		p->old_pause = dev->phydev->pause;
-	}
+static int dsa_slave_phylink_mac_link_state(struct net_device *dev,
+					    struct phylink_link_state *state)
+{
+	struct dsa_port *dp = dsa_slave_to_port(dev);
+	struct dsa_switch *ds = dp->ds;
 
-	if (ds->ops->adjust_link && status_changed)
-		ds->ops->adjust_link(ds, dp->index, dev->phydev);
+	/* Only called for SGMII and 802.3z */
+	if (!ds->ops->phylink_mac_link_state)
+		return -EOPNOTSUPP;
 
-	if (status_changed)
-		phy_print_status(dev->phydev);
+	return ds->ops->phylink_mac_link_state(ds, dp->index, state);
 }
 
-static int dsa_slave_fixed_link_update(struct net_device *dev,
-				       struct fixed_phy_status *status)
+static void dsa_slave_phylink_mac_config(struct net_device *dev,
+					 unsigned int mode,
+					 const struct phylink_link_state *state)
 {
-	struct dsa_switch *ds;
-	struct dsa_port *dp;
+	struct dsa_port *dp = dsa_slave_to_port(dev);
+	struct dsa_switch *ds = dp->ds;
+
+	if (!ds->ops->phylink_mac_config)
+		return;
 
-	if (dev) {
-		dp = dsa_slave_to_port(dev);
-		ds = dp->ds;
-		if (ds->ops->fixed_link_update)
-			ds->ops->fixed_link_update(ds, dp->index, status);
+	ds->ops->phylink_mac_config(ds, dp->index, mode, state);
+}
+
+static void dsa_slave_phylink_mac_an_restart(struct net_device *dev)
+{
+	struct dsa_port *dp = dsa_slave_to_port(dev);
+	struct dsa_switch *ds = dp->ds;
+
+	if (!ds->ops->phylink_mac_an_restart)
+		return;
+
+	ds->ops->phylink_mac_an_restart(ds, dp->index);
+}
+
+static void dsa_slave_phylink_mac_link_down(struct net_device *dev,
+					    unsigned int mode,
+					    phy_interface_t interface)
+{
+	struct dsa_port *dp = dsa_slave_to_port(dev);
+	struct dsa_switch *ds = dp->ds;
+
+	if (!ds->ops->phylink_mac_link_down) {
+		if (ds->ops->adjust_link && dev->phydev)
+			ds->ops->adjust_link(ds, dp->index, dev->phydev);
+		return;
 	}
 
-	return 0;
+	ds->ops->phylink_mac_link_down(ds, dp->index, mode, interface);
+}
+
+static void dsa_slave_phylink_mac_link_up(struct net_device *dev,
+					  unsigned int mode,
+					  phy_interface_t interface,
+					  struct phy_device *phydev)
+{
+	struct dsa_port *dp = dsa_slave_to_port(dev);
+	struct dsa_switch *ds = dp->ds;
+
+	if (!ds->ops->phylink_mac_link_up) {
+		if (ds->ops->adjust_link && dev->phydev)
+			ds->ops->adjust_link(ds, dp->index, dev->phydev);
+		return;
+	}
+
+	ds->ops->phylink_mac_link_up(ds, dp->index, mode, interface, phydev);
+}
+
+static const struct phylink_mac_ops dsa_slave_phylink_mac_ops = {
+	.validate = dsa_slave_phylink_validate,
+	.mac_link_state = dsa_slave_phylink_mac_link_state,
+	.mac_config = dsa_slave_phylink_mac_config,
+	.mac_an_restart = dsa_slave_phylink_mac_an_restart,
+	.mac_link_down = dsa_slave_phylink_mac_link_down,
+	.mac_link_up = dsa_slave_phylink_mac_link_up,
+};
+
+void dsa_port_phylink_mac_change(struct dsa_switch *ds, int port, bool up)
+{
+	const struct dsa_port *dp = dsa_to_port(ds, port);
+
+	phylink_mac_change(dp->pl, up);
+}
+EXPORT_SYMBOL_GPL(dsa_port_phylink_mac_change);
+
+static void dsa_slave_phylink_fixed_state(struct net_device *dev,
+					  struct phylink_link_state *state)
+{
+	struct dsa_port *dp = dsa_slave_to_port(dev);
+	struct dsa_switch *ds = dp->ds;
+
+	/* No need to check that this operation is valid, the callback would
+	 * not be called if it was not.
+	 */
+	ds->ops->phylink_fixed_state(ds, dp->index, state);
 }
 
 /* slave device setup *******************************************************/
 static int dsa_slave_phy_connect(struct net_device *slave_dev, int addr)
 {
 	struct dsa_port *dp = dsa_slave_to_port(slave_dev);
-	struct dsa_slave_priv *p = netdev_priv(slave_dev);
 	struct dsa_switch *ds = dp->ds;
 
 	slave_dev->phydev = mdiobus_get_phy(ds->slave_mii_bus, addr);
@@ -1101,80 +1184,54 @@ static int dsa_slave_phy_connect(struct net_device *slave_dev, int addr)
 		return -ENODEV;
 	}
 
-	/* Use already configured phy mode */
-	if (p->phy_interface == PHY_INTERFACE_MODE_NA)
-		p->phy_interface = slave_dev->phydev->interface;
-
-	return phy_connect_direct(slave_dev, slave_dev->phydev,
-				  dsa_slave_adjust_link, p->phy_interface);
-}
-
-void dsa_port_phylink_mac_change(struct dsa_switch *ds, int port, bool up)
-{
+	return phylink_connect_phy(dp->pl, slave_dev->phydev);
 }
-EXPORT_SYMBOL_GPL(dsa_port_phylink_mac_change);
 
 static int dsa_slave_phy_setup(struct net_device *slave_dev)
 {
 	struct dsa_port *dp = dsa_slave_to_port(slave_dev);
-	struct dsa_slave_priv *p = netdev_priv(slave_dev);
 	struct device_node *port_dn = dp->dn;
 	struct dsa_switch *ds = dp->ds;
-	struct device_node *phy_dn;
-	bool phy_is_fixed = false;
 	u32 phy_flags = 0;
 	int mode, ret;
 
 	mode = of_get_phy_mode(port_dn);
 	if (mode < 0)
 		mode = PHY_INTERFACE_MODE_NA;
-	p->phy_interface = mode;
 
-	phy_dn = of_parse_phandle(port_dn, "phy-handle", 0);
-	if (!phy_dn && of_phy_is_fixed_link(port_dn)) {
-		/* In the case of a fixed PHY, the DT node associated
-		 * to the fixed PHY is the Port DT node
-		 */
-		ret = of_phy_register_fixed_link(port_dn);
-		if (ret) {
-			netdev_err(slave_dev, "failed to register fixed PHY: %d\n", ret);
-			return ret;
-		}
-		phy_is_fixed = true;
-		phy_dn = of_node_get(port_dn);
+	dp->pl = phylink_create(slave_dev, of_fwnode_handle(port_dn), mode,
+				&dsa_slave_phylink_mac_ops);
+	if (IS_ERR(dp->pl)) {
+		netdev_err(slave_dev,
+			   "error creating PHYLINK: %ld\n", PTR_ERR(dp->pl));
+		return PTR_ERR(dp->pl);
 	}
 
+	/* Register only if the switch provides such a callback, since this
+	 * callback takes precedence over polling the link GPIO in PHYLINK
+	 * (see phylink_get_fixed_state).
+	 */
+	if (ds->ops->phylink_fixed_state)
+		phylink_fixed_state_cb(dp->pl, dsa_slave_phylink_fixed_state);
+
 	if (ds->ops->get_phy_flags)
 		phy_flags = ds->ops->get_phy_flags(ds, dp->index);
 
-	if (phy_dn) {
-		slave_dev->phydev = of_phy_connect(slave_dev, phy_dn,
-						   dsa_slave_adjust_link,
-						   phy_flags,
-						   p->phy_interface);
-		of_node_put(phy_dn);
-	}
-
-	if (slave_dev->phydev && phy_is_fixed)
-		fixed_phy_set_link_update(slave_dev->phydev,
-					  dsa_slave_fixed_link_update);
-
-	/* We could not connect to a designated PHY, so use the switch internal
-	 * MDIO bus instead
-	 */
-	if (!slave_dev->phydev) {
+	ret = phylink_of_phy_connect(dp->pl, port_dn, phy_flags);
+	if (ret == -ENODEV) {
+		/* We could not connect to a designated PHY or SFP, so use the
+		 * switch internal MDIO bus instead
+		 */
 		ret = dsa_slave_phy_connect(slave_dev, dp->index);
 		if (ret) {
-			netdev_err(slave_dev, "failed to connect to port %d: %d\n",
+			netdev_err(slave_dev,
+				   "failed to connect to port %d: %d\n",
 				   dp->index, ret);
-			if (phy_is_fixed)
-				of_phy_deregister_fixed_link(port_dn);
+			phylink_destroy(dp->pl);
 			return ret;
 		}
 	}
 
-	phy_attached_info(slave_dev->phydev);
-
 	return 0;
 }
 
@@ -1189,29 +1246,26 @@ static void dsa_slave_set_lockdep_class_one(struct net_device *dev,
 
 int dsa_slave_suspend(struct net_device *slave_dev)
 {
-	struct dsa_slave_priv *p = netdev_priv(slave_dev);
+	struct dsa_port *dp = dsa_slave_to_port(slave_dev);
 
 	netif_device_detach(slave_dev);
 
-	if (slave_dev->phydev) {
-		phy_stop(slave_dev->phydev);
-		p->old_pause = -1;
-		p->old_link = -1;
-		p->old_duplex = -1;
-		phy_suspend(slave_dev->phydev);
-	}
+	rtnl_lock();
+	phylink_stop(dp->pl);
+	rtnl_unlock();
 
 	return 0;
 }
 
 int dsa_slave_resume(struct net_device *slave_dev)
 {
+	struct dsa_port *dp = dsa_slave_to_port(slave_dev);
+
 	netif_device_attach(slave_dev);
 
-	if (slave_dev->phydev) {
-		phy_resume(slave_dev->phydev);
-		phy_start(slave_dev->phydev);
-	}
+	rtnl_lock();
+	phylink_start(dp->pl);
+	rtnl_unlock();
 
 	return 0;
 }
@@ -1276,11 +1330,6 @@ int dsa_slave_create(struct dsa_port *port)
 	p->dp = port;
 	INIT_LIST_HEAD(&p->mall_tc_list);
 	p->xmit = cpu_dp->tag_ops->xmit;
-
-	p->old_pause = -1;
-	p->old_link = -1;
-	p->old_duplex = -1;
-
 	port->slave = slave_dev;
 
 	netif_carrier_off(slave_dev);
@@ -1303,9 +1352,10 @@ int dsa_slave_create(struct dsa_port *port)
 	return 0;
 
 out_phy:
-	phy_disconnect(slave_dev->phydev);
-	if (of_phy_is_fixed_link(port->dn))
-		of_phy_deregister_fixed_link(port->dn);
+	rtnl_lock();
+	phylink_disconnect_phy(p->dp->pl);
+	rtnl_unlock();
+	phylink_destroy(p->dp->pl);
 out_free:
 	free_percpu(p->stats64);
 	free_netdev(slave_dev);
@@ -1317,17 +1367,15 @@ void dsa_slave_destroy(struct net_device *slave_dev)
 {
 	struct dsa_port *dp = dsa_slave_to_port(slave_dev);
 	struct dsa_slave_priv *p = netdev_priv(slave_dev);
-	struct device_node *port_dn = dp->dn;
 
 	netif_carrier_off(slave_dev);
-	if (slave_dev->phydev) {
-		phy_disconnect(slave_dev->phydev);
+	rtnl_lock();
+	phylink_disconnect_phy(dp->pl);
+	rtnl_unlock();
 
-		if (of_phy_is_fixed_link(port_dn))
-			of_phy_deregister_fixed_link(port_dn);
-	}
 	dsa_slave_notify(slave_dev, DSA_PORT_UNREGISTER);
 	unregister_netdev(slave_dev);
+	phylink_destroy(dp->pl);
 	free_percpu(p->stats64);
 	free_netdev(slave_dev);
 }
-- 
2.17.0

[PATCH net-next 9/9] net: dsa: bcm_sf2: Get rid of PHYLIB functions

[Florian Fainelli]

Now that we have converted the bcm_sf2 driver to implement PHYLINK MAC
operations, we can remove the PHYLIB callbacks: adjust_link() and
fixed_link_update() which are no longer called by DSA.

Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
 drivers/net/dsa/bcm_sf2.c | 149 --------------------------------------
 1 file changed, 149 deletions(-)

diff --git a/drivers/net/dsa/bcm_sf2.c b/drivers/net/dsa/bcm_sf2.c
index a20608b0329e..ac621f44237a 100644
--- a/drivers/net/dsa/bcm_sf2.c
+++ b/drivers/net/dsa/bcm_sf2.c
@@ -480,153 +480,6 @@ static u32 bcm_sf2_sw_get_phy_flags(struct dsa_switch *ds, int port)
 	return priv->hw_params.gphy_rev;
 }
 
-static void bcm_sf2_sw_adjust_link(struct dsa_switch *ds, int port,
-				   struct phy_device *phydev)
-{
-	struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds);
-	struct ethtool_eee *p = &priv->dev->ports[port].eee;
-	u32 id_mode_dis = 0, port_mode;
-	const char *str = NULL;
-	u32 reg, offset;
-
-	if (priv->type == BCM7445_DEVICE_ID)
-		offset = CORE_STS_OVERRIDE_GMIIP_PORT(port);
-	else
-		offset = CORE_STS_OVERRIDE_GMIIP2_PORT(port);
-
-	switch (phydev->interface) {
-	case PHY_INTERFACE_MODE_RGMII:
-		str = "RGMII (no delay)";
-		id_mode_dis = 1;
-	case PHY_INTERFACE_MODE_RGMII_TXID:
-		if (!str)
-			str = "RGMII (TX delay)";
-		port_mode = EXT_GPHY;
-		break;
-	case PHY_INTERFACE_MODE_MII:
-		str = "MII";
-		port_mode = EXT_EPHY;
-		break;
-	case PHY_INTERFACE_MODE_REVMII:
-		str = "Reverse MII";
-		port_mode = EXT_REVMII;
-		break;
-	default:
-		/* All other PHYs: internal and MoCA */
-		goto force_link;
-	}
-
-	/* If the link is down, just disable the interface to conserve power */
-	if (!phydev->link) {
-		reg = reg_readl(priv, REG_RGMII_CNTRL_P(port));
-		reg &= ~RGMII_MODE_EN;
-		reg_writel(priv, reg, REG_RGMII_CNTRL_P(port));
-		goto force_link;
-	}
-
-	/* Clear id_mode_dis bit, and the existing port mode, but
-	 * make sure we enable the RGMII block for data to pass
-	 */
-	reg = reg_readl(priv, REG_RGMII_CNTRL_P(port));
-	reg &= ~ID_MODE_DIS;
-	reg &= ~(PORT_MODE_MASK << PORT_MODE_SHIFT);
-	reg &= ~(RX_PAUSE_EN | TX_PAUSE_EN);
-
-	reg |= port_mode | RGMII_MODE_EN;
-	if (id_mode_dis)
-		reg |= ID_MODE_DIS;
-
-	if (phydev->pause) {
-		if (phydev->asym_pause)
-			reg |= TX_PAUSE_EN;
-		reg |= RX_PAUSE_EN;
-	}
-
-	reg_writel(priv, reg, REG_RGMII_CNTRL_P(port));
-
-	pr_info("Port %d configured for %s\n", port, str);
-
-force_link:
-	/* Force link settings detected from the PHY */
-	reg = SW_OVERRIDE;
-	switch (phydev->speed) {
-	case SPEED_1000:
-		reg |= SPDSTS_1000 << SPEED_SHIFT;
-		break;
-	case SPEED_100:
-		reg |= SPDSTS_100 << SPEED_SHIFT;
-		break;
-	}
-
-	if (phydev->link)
-		reg |= LINK_STS;
-	if (phydev->duplex == DUPLEX_FULL)
-		reg |= DUPLX_MODE;
-
-	core_writel(priv, reg, offset);
-
-	if (!phydev->is_pseudo_fixed_link)
-		p->eee_enabled = b53_eee_init(ds, port, phydev);
-}
-
-static void bcm_sf2_sw_fixed_link_update(struct dsa_switch *ds, int port,
-					 struct fixed_phy_status *status)
-{
-	struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds);
-	u32 duplex, pause, offset;
-	u32 reg;
-
-	if (priv->type == BCM7445_DEVICE_ID)
-		offset = CORE_STS_OVERRIDE_GMIIP_PORT(port);
-	else
-		offset = CORE_STS_OVERRIDE_GMIIP2_PORT(port);
-
-	duplex = core_readl(priv, CORE_DUPSTS);
-	pause = core_readl(priv, CORE_PAUSESTS);
-
-	status->link = 0;
-
-	/* MoCA port is special as we do not get link status from CORE_LNKSTS,
-	 * which means that we need to force the link at the port override
-	 * level to get the data to flow. We do use what the interrupt handler
-	 * did determine before.
-	 *
-	 * For the other ports, we just force the link status, since this is
-	 * a fixed PHY device.
-	 */
-	if (port == priv->moca_port) {
-		status->link = priv->port_sts[port].link;
-		/* For MoCA interfaces, also force a link down notification
-		 * since some version of the user-space daemon (mocad) use
-		 * cmd->autoneg to force the link, which messes up the PHY
-		 * state machine and make it go in PHY_FORCING state instead.
-		 */
-		if (!status->link)
-			netif_carrier_off(ds->ports[port].slave);
-		status->duplex = 1;
-	} else {
-		status->link = 1;
-		status->duplex = !!(duplex & (1 << port));
-	}
-
-	reg = core_readl(priv, offset);
-	reg |= SW_OVERRIDE;
-	if (status->link)
-		reg |= LINK_STS;
-	else
-		reg &= ~LINK_STS;
-	core_writel(priv, reg, offset);
-
-	if ((pause & (1 << port)) &&
-	    (pause & (1 << (port + PAUSESTS_TX_PAUSE_SHIFT)))) {
-		status->asym_pause = 1;
-		status->pause = 1;
-	}
-
-	if (pause & (1 << port))
-		status->pause = 1;
-}
-
 static void bcm_sf2_sw_validate(struct dsa_switch *ds, int port,
 				unsigned long *supported,
 				struct phylink_link_state *state)
@@ -1054,8 +907,6 @@ static const struct dsa_switch_ops bcm_sf2_ops = {
 	.get_sset_count		= b53_get_sset_count,
 	.get_ethtool_phy_stats	= b53_get_ethtool_phy_stats,
 	.get_phy_flags		= bcm_sf2_sw_get_phy_flags,
-	.adjust_link		= bcm_sf2_sw_adjust_link,
-	.fixed_link_update	= bcm_sf2_sw_fixed_link_update,
 	.phylink_validate	= bcm_sf2_sw_validate,
 	.phylink_mac_config	= bcm_sf2_sw_mac_config,
 	.phylink_mac_link_down	= bcm_sf2_sw_mac_link_down,
-- 
2.17.0

Re: [PATCH net-next 0/9] net: dsa: Plug in PHYLINK support

[David Miller]

From: Florian Fainelli <f.fainelli@gmail.com>
Date: Sat,  5 May 2018 12:04:16 -0700

> This patch series adds PHYLINK support to DSA which is necessary to
> support more complex PHY and pluggable modules setups.

The regression Andrew reported with patch #7 is holding up my
applying this series.

How would you like this to be handled?  Are you going to respin
once Andrew sends you the fix?

Thanks.

Re: [PATCH net-next 0/9] net: dsa: Plug in PHYLINK support

[Florian Fainelli]

Le 05/08/18 à 17:17, David Miller a écrit :
> From: Florian Fainelli <f.fainelli@gmail.com>
> Date: Sat,  5 May 2018 12:04:16 -0700
> 
>> This patch series adds PHYLINK support to DSA which is necessary to
>> support more complex PHY and pluggable modules setups.
> 
> The regression Andrew reported with patch #7 is holding up my
> applying this series.
> 
> How would you like this to be handled?  Are you going to respin
> once Andrew sends you the fix?

Andrew sent me the fix and I squashed it in patch 7, waiting for Russell
to give me his SoB for patch 3 since he originally sent that to me
privately and I changed it a bit.

Expect a v2 soon, should have made that clearer. Thanks!
-- 
Florian