[PATCH net-next v2 00/14] net: mvpp2: comphy configuration

[PATCH net-next v2 00/14] net: mvpp2: comphy configuration

[Antoine Tenart]

Hi all,

This series, following up the one one the GoP/MAC configuration, aims at
stopping to depend on the firmware/bootloader configuration when using
the PPv2 engine. With this series the PPv2 driver does not need to rely
on a previous configuration, and dynamic reconfiguration while the
kernel is running can be done (i.e. switch one port from SGMII to 10GKR,
or the opposite). A port can now be configured in a different mode than
what's done in the firmware/bootloader as well.

The series first contain patches in the generic PHY framework to support
what is called the comphy (common PHYs), which is an h/w block providing
PHYs that can be configured in various modes ranging from SGMII, 10GKR
to SATA and others. As of now only the SGMII and 10GKR modes are
supported by the comphy driver.

Then patches are modifying the PPv2 driver to first add the comphy
initialization sequence (i.e. calls to the generic PHY framework) and to
then take advantage of this to allow dynamic reconfiguration (i.e.
configuring the mode of a port given what's connected, between sgmii and
10G). Note the use of the comphy in the PPv2 driver is kept optional
(i.e. if not described in dt the driver still as before an relies on the
firmware/bootloader configuration).

Finally there are dt/defconfig patches to describe and take advantage of
this.

This was tested on a range of devices: 8040-db, 8040-mcbin and 7040-db.

Thanks!
Antoine

Since v1:
  - Updated the mode settings variable name in the comphy driver to
    have 'cp110' in it.
  - Documented the PHY cell argument in the dt documentation.
  - New patch adding comphy phandles for the 7040-db board.
  - Checked if the carrier_on/off functions were needed. They are.
  - s/PHY/generic PHY/ in commit log of patch 1.
  - Rebased on the latest net-next/master.

Antoine Tenart (13):
  phy: add sgmii and 10gkr modes to the phy_mode enum
  phy: add the mvebu cp110 comphy driver
  Documentation/bindings: phy: document the Marvell comphy driver
  net: mvpp2: initialize the comphy
  net: mvpp2: do not force the link mode
  net: mvpp2: simplify the link_event function
  net: mvpp2: improve the link management function
  net: mvpp2: check the netif is running in the link_event function
  net: mvpp2: dynamic reconfiguration of the PHY mode
  arm64: dts: marvell: extend the cp110 syscon register area length
  arm64: dts: marvell: add comphy nodes on cp110 master and slave
  arm64: dts: marvell: mcbin: add comphy references to Ethernet ports
  arm64: dts: marvell: 7040-db: add comphy references to Ethernet ports

Miquel Raynal (1):
  arm64: defconfig: enable Marvell CP110 comphy

 .../devicetree/bindings/phy/phy-mvebu-comphy.txt   |  43 ++
 arch/arm64/boot/dts/marvell/armada-7040-db.dts     |   1 +
 arch/arm64/boot/dts/marvell/armada-8040-mcbin.dts  |   3 +
 .../boot/dts/marvell/armada-cp110-master.dtsi      |  40 +-
 .../arm64/boot/dts/marvell/armada-cp110-slave.dtsi |  40 +-
 arch/arm64/configs/defconfig                       |   1 +
 drivers/net/ethernet/marvell/mvpp2.c               | 111 ++--
 drivers/phy/marvell/Kconfig                        |  10 +
 drivers/phy/marvell/Makefile                       |   1 +
 drivers/phy/marvell/phy-mvebu-cp110-comphy.c       | 656 +++++++++++++++++++++
 include/linux/phy/phy.h                            |   2 +
 11 files changed, 873 insertions(+), 35 deletions(-)
 create mode 100644 Documentation/devicetree/bindings/phy/phy-mvebu-comphy.txt
 create mode 100644 drivers/phy/marvell/phy-mvebu-cp110-comphy.c

-- 
2.13.5

[PATCH net-next v2 01/14] phy: add sgmii and 10gkr modes to the phy_mode enum

[Antoine Tenart]

This patch adds more generic PHY modes to the phy_mode enum, to
allow configuring generic PHYs to the SGMII and/or the 10GKR mode
by using the set_mode callback.

Signed-off-by: Antoine Tenart <antoine.tenart@free-electrons.com>
---
 include/linux/phy/phy.h | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/include/linux/phy/phy.h b/include/linux/phy/phy.h
index 78bb0d7f6b11..e694d4008c4a 100644
--- a/include/linux/phy/phy.h
+++ b/include/linux/phy/phy.h
@@ -27,6 +27,8 @@ enum phy_mode {
 	PHY_MODE_USB_HOST,
 	PHY_MODE_USB_DEVICE,
 	PHY_MODE_USB_OTG,
+	PHY_MODE_SGMII,
+	PHY_MODE_10GKR,
 };
 
 /**
-- 
2.13.5

[PATCH net-next v2 02/14] phy: add the mvebu cp110 comphy driver

[Antoine Tenart]

On the CP110 unit, which can be found on various Marvell platforms such
as the 7k and 8k (currently), a comphy (common PHYs) hardware block can
be found. This block provides a number of PHYs which can be used in
various modes by other controllers (network, SATA ...). These common
PHYs must be configured for the controllers using them to work correctly
either at boot time, or when the system runs to switch the mode used.
This patch adds a driver for this comphy hardware block, providing
callbacks for the its PHYs so that consumers can configure the modes
used.

As of this commit, two modes are supported by the comphy driver: sgmii
and 10gkr.

Signed-off-by: Antoine Tenart <antoine.tenart@free-electrons.com>
---
 drivers/phy/marvell/Kconfig                  |  10 +
 drivers/phy/marvell/Makefile                 |   1 +
 drivers/phy/marvell/phy-mvebu-cp110-comphy.c | 656 +++++++++++++++++++++++++++
 3 files changed, 667 insertions(+)
 create mode 100644 drivers/phy/marvell/phy-mvebu-cp110-comphy.c

diff --git a/drivers/phy/marvell/Kconfig b/drivers/phy/marvell/Kconfig
index 048d8893bc2e..26755f3d1a9a 100644
--- a/drivers/phy/marvell/Kconfig
+++ b/drivers/phy/marvell/Kconfig
@@ -21,6 +21,16 @@ config PHY_BERLIN_USB
 	help
 	  Enable this to support the USB PHY on Marvell Berlin SoCs.
 
+config PHY_MVEBU_CP110_COMPHY
+	tristate "Marvell CP110 comphy driver"
+	depends on ARCH_MVEBU && OF
+	select GENERIC_PHY
+	help
+	  This driver allows to control the comphy, an hardware block providing
+	  shared serdes PHYs on Marvell Armada 7k/8k (in the CP110). Its serdes
+	  lanes can be used by various controllers (Ethernet, sata, usb,
+	  PCIe...).
+
 config PHY_MVEBU_SATA
 	def_bool y
 	depends on ARCH_DOVE || MACH_DOVE || MACH_KIRKWOOD
diff --git a/drivers/phy/marvell/Makefile b/drivers/phy/marvell/Makefile
index 3fc188f59118..0cf6a7cbaf9f 100644
--- a/drivers/phy/marvell/Makefile
+++ b/drivers/phy/marvell/Makefile
@@ -1,6 +1,7 @@
 obj-$(CONFIG_ARMADA375_USBCLUSTER_PHY)	+= phy-armada375-usb2.o
 obj-$(CONFIG_PHY_BERLIN_SATA)		+= phy-berlin-sata.o
 obj-$(CONFIG_PHY_BERLIN_USB)		+= phy-berlin-usb.o
+obj-$(CONFIG_PHY_MVEBU_CP110_COMPHY)	+= phy-mvebu-cp110-comphy.o
 obj-$(CONFIG_PHY_MVEBU_SATA)		+= phy-mvebu-sata.o
 obj-$(CONFIG_PHY_PXA_28NM_HSIC)		+= phy-pxa-28nm-hsic.o
 obj-$(CONFIG_PHY_PXA_28NM_USB2)		+= phy-pxa-28nm-usb2.o
diff --git a/drivers/phy/marvell/phy-mvebu-cp110-comphy.c b/drivers/phy/marvell/phy-mvebu-cp110-comphy.c
new file mode 100644
index 000000000000..41556e790856
--- /dev/null
+++ b/drivers/phy/marvell/phy-mvebu-cp110-comphy.c
@@ -0,0 +1,656 @@
+/*
+ * Copyright (C) 2017 Marvell
+ *
+ * Antoine Tenart <antoine.tenart@free-electrons.com>
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2. This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#include <linux/io.h>
+#include <linux/iopoll.h>
+#include <linux/mfd/syscon.h>
+#include <linux/module.h>
+#include <linux/phy/phy.h>
+#include <linux/platform_device.h>
+#include <linux/regmap.h>
+
+/* Relative to priv->base */
+#define MVEBU_COMPHY_SERDES_CFG0(n)		(0x0 + (n) * 0x1000)
+#define     MVEBU_COMPHY_SERDES_CFG0_PU_PLL	BIT(1)
+#define     MVEBU_COMPHY_SERDES_CFG0_GEN_RX(n)	((n) << 3)
+#define     MVEBU_COMPHY_SERDES_CFG0_GEN_TX(n)	((n) << 7)
+#define     MVEBU_COMPHY_SERDES_CFG0_PU_RX	BIT(11)
+#define     MVEBU_COMPHY_SERDES_CFG0_PU_TX	BIT(12)
+#define     MVEBU_COMPHY_SERDES_CFG0_HALF_BUS	BIT(14)
+#define MVEBU_COMPHY_SERDES_CFG1(n)		(0x4 + (n) * 0x1000)
+#define     MVEBU_COMPHY_SERDES_CFG1_RESET	BIT(3)
+#define     MVEBU_COMPHY_SERDES_CFG1_RX_INIT	BIT(4)
+#define     MVEBU_COMPHY_SERDES_CFG1_CORE_RESET	BIT(5)
+#define     MVEBU_COMPHY_SERDES_CFG1_RF_RESET	BIT(6)
+#define MVEBU_COMPHY_SERDES_CFG2(n)		(0x8 + (n) * 0x1000)
+#define     MVEBU_COMPHY_SERDES_CFG2_DFE_EN	BIT(4)
+#define MVEBU_COMPHY_SERDES_STATUS0(n)		(0x18 + (n) * 0x1000)
+#define     MVEBU_COMPHY_SERDES_STATUS0_TX_PLL_RDY	BIT(2)
+#define     MVEBU_COMPHY_SERDES_STATUS0_RX_PLL_RDY	BIT(3)
+#define     MVEBU_COMPHY_SERDES_STATUS0_RX_INIT		BIT(4)
+#define MVEBU_COMPHY_PWRPLL_CTRL(n)		(0x804 + (n) * 0x1000)
+#define     MVEBU_COMPHY_PWRPLL_CTRL_RFREQ(n)	((n) << 0)
+#define     MVEBU_COMPHY_PWRPLL_PHY_MODE(n)	((n) << 5)
+#define MVEBU_COMPHY_IMP_CAL(n)			(0x80c + (n) * 0x1000)
+#define     MVEBU_COMPHY_IMP_CAL_TX_EXT(n)	((n) << 10)
+#define     MVEBU_COMPHY_IMP_CAL_TX_EXT_EN	BIT(15)
+#define MVEBU_COMPHY_DFE_RES(n)			(0x81c + (n) * 0x1000)
+#define     MVEBU_COMPHY_DFE_RES_FORCE_GEN_TBL	BIT(15)
+#define MVEBU_COMPHY_COEF(n)			(0x828 + (n) * 0x1000)
+#define     MVEBU_COMPHY_COEF_DFE_EN		BIT(14)
+#define     MVEBU_COMPHY_COEF_DFE_CTRL		BIT(15)
+#define MVEBU_COMPHY_GEN1_S0(n)			(0x834 + (n) * 0x1000)
+#define     MVEBU_COMPHY_GEN1_S0_TX_AMP(n)	((n) << 1)
+#define     MVEBU_COMPHY_GEN1_S0_TX_EMPH(n)	((n) << 7)
+#define MVEBU_COMPHY_GEN1_S1(n)			(0x838 + (n) * 0x1000)
+#define     MVEBU_COMPHY_GEN1_S1_RX_MUL_PI(n)	((n) << 0)
+#define     MVEBU_COMPHY_GEN1_S1_RX_MUL_PF(n)	((n) << 3)
+#define     MVEBU_COMPHY_GEN1_S1_RX_MUL_FI(n)	((n) << 6)
+#define     MVEBU_COMPHY_GEN1_S1_RX_MUL_FF(n)	((n) << 8)
+#define     MVEBU_COMPHY_GEN1_S1_RX_DFE_EN	BIT(10)
+#define     MVEBU_COMPHY_GEN1_S1_RX_DIV(n)	((n) << 11)
+#define MVEBU_COMPHY_GEN1_S2(n)			(0x8f4 + (n) * 0x1000)
+#define     MVEBU_COMPHY_GEN1_S2_TX_EMPH(n)	((n) << 0)
+#define     MVEBU_COMPHY_GEN1_S2_TX_EMPH_EN	BIT(4)
+#define MVEBU_COMPHY_LOOPBACK(n)		(0x88c + (n) * 0x1000)
+#define     MVEBU_COMPHY_LOOPBACK_DBUS_WIDTH(n)	((n) << 1)
+#define MVEBU_COMPHY_VDD_CAL0(n)		(0x908 + (n) * 0x1000)
+#define     MVEBU_COMPHY_VDD_CAL0_CONT_MODE	BIT(15)
+#define MVEBU_COMPHY_EXT_SELV(n)		(0x914 + (n) * 0x1000)
+#define     MVEBU_COMPHY_EXT_SELV_RX_SAMPL(n)	((n) << 5)
+#define MVEBU_COMPHY_MISC_CTRL0(n)		(0x93c + (n) * 0x1000)
+#define     MVEBU_COMPHY_MISC_CTRL0_ICP_FORCE	BIT(5)
+#define     MVEBU_COMPHY_MISC_CTRL0_REFCLK_SEL	BIT(10)
+#define MVEBU_COMPHY_RX_CTRL1(n)		(0x940 + (n) * 0x1000)
+#define     MVEBU_COMPHY_RX_CTRL1_RXCLK2X_SEL	BIT(11)
+#define     MVEBU_COMPHY_RX_CTRL1_CLK8T_EN	BIT(12)
+#define MVEBU_COMPHY_SPEED_DIV(n)		(0x954 + (n) * 0x1000)
+#define     MVEBU_COMPHY_SPEED_DIV_TX_FORCE	BIT(7)
+#define MVEBU_SP_CALIB(n)			(0x96c + (n) * 0x1000)
+#define     MVEBU_SP_CALIB_SAMPLER(n)		((n) << 8)
+#define     MVEBU_SP_CALIB_SAMPLER_EN		BIT(12)
+#define MVEBU_COMPHY_TX_SLEW_RATE(n)		(0x974 + (n) * 0x1000)
+#define     MVEBU_COMPHY_TX_SLEW_RATE_EMPH(n)	((n) << 5)
+#define     MVEBU_COMPHY_TX_SLEW_RATE_SLC(n)	((n) << 10)
+#define MVEBU_COMPHY_DLT_CTRL(n)		(0x984 + (n) * 0x1000)
+#define     MVEBU_COMPHY_DLT_CTRL_DTL_FLOOP_EN	BIT(2)
+#define MVEBU_COMPHY_FRAME_DETECT0(n)		(0xa14 + (n) * 0x1000)
+#define     MVEBU_COMPHY_FRAME_DETECT0_PATN(n)	((n) << 7)
+#define MVEBU_COMPHY_FRAME_DETECT3(n)		(0xa20 + (n) * 0x1000)
+#define     MVEBU_COMPHY_FRAME_DETECT3_LOST_TIMEOUT_EN	BIT(12)
+#define MVEBU_COMPHY_DME(n)			(0xa28 + (n) * 0x1000)
+#define     MVEBU_COMPHY_DME_ETH_MODE		BIT(7)
+#define MVEBU_COMPHY_TRAINING0(n)		(0xa68 + (n) * 0x1000)
+#define     MVEBU_COMPHY_TRAINING0_P2P_HOLD	BIT(15)
+#define MVEBU_COMPHY_TRAINING5(n)		(0xaa4 + (n) * 0x1000)
+#define	    MVEBU_COMPHY_TRAINING5_RX_TIMER(n)	((n) << 0)
+#define MVEBU_COMPHY_TX_TRAIN_PRESET(n)		(0xb1c + (n) * 0x1000)
+#define     MVEBU_COMPHY_TX_TRAIN_PRESET_16B_AUTO_EN	BIT(8)
+#define     MVEBU_COMPHY_TX_TRAIN_PRESET_PRBS11		BIT(9)
+#define MVEBU_COMPHY_GEN1_S3(n)			(0xc40 + (n) * 0x1000)
+#define     MVEBU_COMPHY_GEN1_S3_FBCK_SEL	BIT(9)
+#define MVEBU_COMPHY_GEN1_S4(n)			(0xc44 + (n) * 0x1000)
+#define	    MVEBU_COMPHY_GEN1_S4_DFE_RES(n)	((n) << 8)
+#define MVEBU_COMPHY_TX_PRESET(n)		(0xc68 + (n) * 0x1000)
+#define     MVEBU_COMPHY_TX_PRESET_INDEX(n)	((n) << 0)
+#define MVEBU_COMPHY_GEN1_S5(n)			(0xd38 + (n) * 0x1000)
+#define     MVEBU_COMPHY_GEN1_S5_ICP(n)		((n) << 0)
+
+/* Relative to priv->regmap */
+#define MVEBU_COMPHY_CONF1(n)			(0x1000 + (n) * 0x28)
+#define     MVEBU_COMPHY_CONF1_PWRUP		BIT(1)
+#define     MVEBU_COMPHY_CONF1_USB_PCIE		BIT(2)	/* 0: Ethernet/SATA */
+#define MVEBU_COMPHY_CONF6(n)			(0x1014 + (n) * 0x28)
+#define     MVEBU_COMPHY_CONF6_40B		BIT(18)
+#define MVEBU_COMPHY_SELECTOR			0x1140
+#define     MVEBU_COMPHY_SELECTOR_PHY(n)	((n) * 0x4)
+
+#define MVEBU_COMPHY_LANES	6
+#define MVEBU_COMPHY_PORTS	3
+
+struct mvebu_comhy_conf {
+	enum phy_mode mode;
+	unsigned lane;
+	unsigned port;
+	u32 mux;
+};
+
+#define MVEBU_COMPHY_CONF(_lane, _port, _mode, _mux)	\
+	{						\
+		.lane = _lane,				\
+		.port = _port,				\
+		.mode = _mode,				\
+		.mux = _mux,				\
+	}
+
+static const struct mvebu_comhy_conf mvebu_comphy_cp110_modes[] = {
+	/* lane 0 */
+	MVEBU_COMPHY_CONF(0, 1, PHY_MODE_SGMII, 0x1),
+	/* lane 1 */
+	MVEBU_COMPHY_CONF(1, 2, PHY_MODE_SGMII, 0x1),
+	/* lane 2 */
+	MVEBU_COMPHY_CONF(2, 0, PHY_MODE_SGMII, 0x1),
+	MVEBU_COMPHY_CONF(2, 0, PHY_MODE_10GKR, 0x1),
+	/* lane 3 */
+	MVEBU_COMPHY_CONF(3, 1, PHY_MODE_SGMII, 0x2),
+	/* lane 4 */
+	MVEBU_COMPHY_CONF(4, 0, PHY_MODE_SGMII, 0x2),
+	MVEBU_COMPHY_CONF(4, 0, PHY_MODE_10GKR, 0x2),
+	MVEBU_COMPHY_CONF(4, 1, PHY_MODE_SGMII, 0x1),
+	/* lane 5 */
+	MVEBU_COMPHY_CONF(5, 2, PHY_MODE_SGMII, 0x1),
+};
+
+struct mvebu_comphy_priv {
+	void __iomem *base;
+	struct regmap *regmap;
+	struct device *dev;
+	struct phy *phys[MVEBU_COMPHY_LANES];
+	int modes[MVEBU_COMPHY_LANES];
+};
+
+struct mvebu_comphy_lane {
+	struct mvebu_comphy_priv *priv;
+	struct device_node *of_node;
+	unsigned id;
+	enum phy_mode mode;
+	int port;
+};
+
+static int mvebu_comphy_get_mux(int lane, int port, enum phy_mode mode)
+{
+	int i, n = ARRAY_SIZE(mvebu_comphy_cp110_modes);
+
+	/* Unused PHY mux value is 0x0 */
+	if (mode == PHY_MODE_INVALID)
+		return 0;
+
+	for (i = 0; i < n; i++) {
+		if (mvebu_comphy_cp110_modes[i].lane == lane &&
+		    mvebu_comphy_cp110_modes[i].port == port &&
+		    mvebu_comphy_cp110_modes[i].mode == mode)
+			break;
+	}
+
+	if (i == n)
+		return -EINVAL;
+
+	return mvebu_comphy_cp110_modes[i].mux;
+}
+
+static void mvebu_comphy_ethernet_init_reset(struct mvebu_comphy_lane *lane,
+					     enum phy_mode mode)
+{
+	struct mvebu_comphy_priv *priv = lane->priv;
+	u32 val;
+
+	regmap_read(priv->regmap, MVEBU_COMPHY_CONF1(lane->id), &val);
+	val &= ~MVEBU_COMPHY_CONF1_USB_PCIE;
+	val |= MVEBU_COMPHY_CONF1_PWRUP;
+	regmap_write(priv->regmap, MVEBU_COMPHY_CONF1(lane->id), val);
+
+	/* Select baud rates and PLLs */
+	val = readl(priv->base + MVEBU_COMPHY_SERDES_CFG0(lane->id));
+	val &= ~(MVEBU_COMPHY_SERDES_CFG0_PU_PLL |
+		 MVEBU_COMPHY_SERDES_CFG0_PU_RX |
+		 MVEBU_COMPHY_SERDES_CFG0_PU_TX |
+		 MVEBU_COMPHY_SERDES_CFG0_HALF_BUS |
+		 MVEBU_COMPHY_SERDES_CFG0_GEN_RX(0xf) |
+		 MVEBU_COMPHY_SERDES_CFG0_GEN_TX(0xf));
+	if (mode == PHY_MODE_10GKR)
+		val |= MVEBU_COMPHY_SERDES_CFG0_GEN_RX(0xe) |
+		       MVEBU_COMPHY_SERDES_CFG0_GEN_TX(0xe);
+	else if (mode == PHY_MODE_SGMII)
+		val |= MVEBU_COMPHY_SERDES_CFG0_GEN_RX(0x6) |
+		       MVEBU_COMPHY_SERDES_CFG0_GEN_TX(0x6) |
+		       MVEBU_COMPHY_SERDES_CFG0_HALF_BUS;
+	writel(val, priv->base + MVEBU_COMPHY_SERDES_CFG0(lane->id));
+
+	/* reset */
+	val = readl(priv->base + MVEBU_COMPHY_SERDES_CFG1(lane->id));
+	val &= ~(MVEBU_COMPHY_SERDES_CFG1_RESET |
+		 MVEBU_COMPHY_SERDES_CFG1_CORE_RESET |
+		 MVEBU_COMPHY_SERDES_CFG1_RF_RESET);
+	writel(val, priv->base + MVEBU_COMPHY_SERDES_CFG1(lane->id));
+
+	/* de-assert reset */
+	val = readl(priv->base + MVEBU_COMPHY_SERDES_CFG1(lane->id));
+	val |= MVEBU_COMPHY_SERDES_CFG1_RESET |
+	       MVEBU_COMPHY_SERDES_CFG1_CORE_RESET;
+	writel(val, priv->base + MVEBU_COMPHY_SERDES_CFG1(lane->id));
+
+	/* wait until clocks are ready */
+	mdelay(1);
+
+	/* exlicitly disable 40B, the bits isn't clear on reset */
+	regmap_read(priv->regmap, MVEBU_COMPHY_CONF6(lane->id), &val);
+	val &= ~MVEBU_COMPHY_CONF6_40B;
+	regmap_write(priv->regmap, MVEBU_COMPHY_CONF6(lane->id), val);
+
+	/* refclk selection */
+	val = readl(priv->base + MVEBU_COMPHY_MISC_CTRL0(lane->id));
+	val &= ~MVEBU_COMPHY_MISC_CTRL0_REFCLK_SEL;
+	if (mode == PHY_MODE_10GKR)
+		val |= MVEBU_COMPHY_MISC_CTRL0_ICP_FORCE;
+	writel(val, priv->base + MVEBU_COMPHY_MISC_CTRL0(lane->id));
+
+	/* power and pll selection */
+	val = readl(priv->base + MVEBU_COMPHY_PWRPLL_CTRL(lane->id));
+	val &= ~(MVEBU_COMPHY_PWRPLL_CTRL_RFREQ(0x1f) |
+		 MVEBU_COMPHY_PWRPLL_PHY_MODE(0x7));
+	val |= MVEBU_COMPHY_PWRPLL_CTRL_RFREQ(0x1) |
+	       MVEBU_COMPHY_PWRPLL_PHY_MODE(0x4);
+	writel(val, priv->base + MVEBU_COMPHY_PWRPLL_CTRL(lane->id));
+
+	val = readl(priv->base + MVEBU_COMPHY_LOOPBACK(lane->id));
+	val &= ~MVEBU_COMPHY_LOOPBACK_DBUS_WIDTH(0x7);
+	val |= MVEBU_COMPHY_LOOPBACK_DBUS_WIDTH(0x1);
+	writel(val, priv->base + MVEBU_COMPHY_LOOPBACK(lane->id));
+}
+
+static int mvebu_comphy_init_plls(struct mvebu_comphy_lane *lane,
+				  enum phy_mode mode)
+{
+	struct mvebu_comphy_priv *priv = lane->priv;
+	u32 val;
+
+	/* SERDES external config */
+	val = readl(priv->base + MVEBU_COMPHY_SERDES_CFG0(lane->id));
+	val |= MVEBU_COMPHY_SERDES_CFG0_PU_PLL |
+	       MVEBU_COMPHY_SERDES_CFG0_PU_RX |
+	       MVEBU_COMPHY_SERDES_CFG0_PU_TX;
+	writel(val, priv->base + MVEBU_COMPHY_SERDES_CFG0(lane->id));
+
+	/* check rx/tx pll */
+	readl_poll_timeout(priv->base + MVEBU_COMPHY_SERDES_STATUS0(lane->id),
+			   val,
+			   val & (MVEBU_COMPHY_SERDES_STATUS0_RX_PLL_RDY |
+				  MVEBU_COMPHY_SERDES_STATUS0_TX_PLL_RDY),
+			   1000, 150000);
+	if (!(val & (MVEBU_COMPHY_SERDES_STATUS0_RX_PLL_RDY |
+		     MVEBU_COMPHY_SERDES_STATUS0_TX_PLL_RDY)))
+		return -ETIMEDOUT;
+
+	/* rx init */
+	val = readl(priv->base + MVEBU_COMPHY_SERDES_CFG1(lane->id));
+	val |= MVEBU_COMPHY_SERDES_CFG1_RX_INIT;
+	writel(val, priv->base + MVEBU_COMPHY_SERDES_CFG1(lane->id));
+
+	/* check rx */
+	readl_poll_timeout(priv->base + MVEBU_COMPHY_SERDES_STATUS0(lane->id),
+			   val, val & MVEBU_COMPHY_SERDES_STATUS0_RX_INIT,
+			   1000, 10000);
+	if (!(val & MVEBU_COMPHY_SERDES_STATUS0_RX_INIT))
+		return -ETIMEDOUT;
+
+	val = readl(priv->base + MVEBU_COMPHY_SERDES_CFG1(lane->id));
+	val &= ~MVEBU_COMPHY_SERDES_CFG1_RX_INIT;
+	writel(val, priv->base + MVEBU_COMPHY_SERDES_CFG1(lane->id));
+
+	return 0;
+}
+
+static int mvebu_comphy_set_mode_sgmii(struct phy *phy, enum phy_mode mode)
+{
+	struct mvebu_comphy_lane *lane = phy_get_drvdata(phy);
+	struct mvebu_comphy_priv *priv = lane->priv;
+	u32 val;
+
+	mvebu_comphy_ethernet_init_reset(lane, mode);
+
+	val = readl(priv->base + MVEBU_COMPHY_RX_CTRL1(lane->id));
+	val &= ~MVEBU_COMPHY_RX_CTRL1_CLK8T_EN;
+	val |= MVEBU_COMPHY_RX_CTRL1_RXCLK2X_SEL;
+	writel(val, priv->base + MVEBU_COMPHY_RX_CTRL1(lane->id));
+
+	val = readl(priv->base + MVEBU_COMPHY_DLT_CTRL(lane->id));
+	val &= ~MVEBU_COMPHY_DLT_CTRL_DTL_FLOOP_EN;
+	writel(val, priv->base + MVEBU_COMPHY_DLT_CTRL(lane->id));
+
+	regmap_read(priv->regmap, MVEBU_COMPHY_CONF1(lane->id), &val);
+	val &= ~MVEBU_COMPHY_CONF1_USB_PCIE;
+	val |= MVEBU_COMPHY_CONF1_PWRUP;
+	regmap_write(priv->regmap, MVEBU_COMPHY_CONF1(lane->id), val);
+
+	val = readl(priv->base + MVEBU_COMPHY_GEN1_S0(lane->id));
+	val &= ~MVEBU_COMPHY_GEN1_S0_TX_EMPH(0xf);
+	val |= MVEBU_COMPHY_GEN1_S0_TX_EMPH(0x1);
+	writel(val, priv->base + MVEBU_COMPHY_GEN1_S0(lane->id));
+
+	return mvebu_comphy_init_plls(lane, mode);
+}
+
+static int mvebu_comphy_set_mode_10gkr(struct phy *phy, enum phy_mode mode)
+{
+	struct mvebu_comphy_lane *lane = phy_get_drvdata(phy);
+	struct mvebu_comphy_priv *priv = lane->priv;
+	u32 val;
+
+	mvebu_comphy_ethernet_init_reset(lane, mode);
+
+	val = readl(priv->base + MVEBU_COMPHY_RX_CTRL1(lane->id));
+	val |= MVEBU_COMPHY_RX_CTRL1_RXCLK2X_SEL |
+	       MVEBU_COMPHY_RX_CTRL1_CLK8T_EN;
+	writel(val, priv->base + MVEBU_COMPHY_RX_CTRL1(lane->id));
+
+	val = readl(priv->base + MVEBU_COMPHY_DLT_CTRL(lane->id));
+	val |= MVEBU_COMPHY_DLT_CTRL_DTL_FLOOP_EN;
+	writel(val, priv->base + MVEBU_COMPHY_DLT_CTRL(lane->id));
+
+	/* Speed divider */
+	val = readl(priv->base + MVEBU_COMPHY_SPEED_DIV(lane->id));
+	val |= MVEBU_COMPHY_SPEED_DIV_TX_FORCE;
+	writel(val, priv->base + MVEBU_COMPHY_SPEED_DIV(lane->id));
+
+	val = readl(priv->base + MVEBU_COMPHY_SERDES_CFG2(lane->id));
+	val |= MVEBU_COMPHY_SERDES_CFG2_DFE_EN;
+	writel(val, priv->base + MVEBU_COMPHY_SERDES_CFG2(lane->id));
+
+	/* DFE resolution */
+	val = readl(priv->base + MVEBU_COMPHY_DFE_RES(lane->id));
+	val |= MVEBU_COMPHY_DFE_RES_FORCE_GEN_TBL;
+	writel(val, priv->base + MVEBU_COMPHY_DFE_RES(lane->id));
+
+	val = readl(priv->base + MVEBU_COMPHY_GEN1_S0(lane->id));
+	val &= ~(MVEBU_COMPHY_GEN1_S0_TX_AMP(0x1f) |
+		 MVEBU_COMPHY_GEN1_S0_TX_EMPH(0xf));
+	val |= MVEBU_COMPHY_GEN1_S0_TX_AMP(0x1c) |
+	       MVEBU_COMPHY_GEN1_S0_TX_EMPH(0xe);
+	writel(val, priv->base + MVEBU_COMPHY_GEN1_S0(lane->id));
+
+	val = readl(priv->base + MVEBU_COMPHY_GEN1_S2(lane->id));
+	val &= ~MVEBU_COMPHY_GEN1_S2_TX_EMPH(0xf);
+	val |= MVEBU_COMPHY_GEN1_S2_TX_EMPH_EN;
+	writel(val, priv->base + MVEBU_COMPHY_GEN1_S2(lane->id));
+
+	val = readl(priv->base + MVEBU_COMPHY_TX_SLEW_RATE(lane->id));
+	val |= MVEBU_COMPHY_TX_SLEW_RATE_EMPH(0x3) |
+	       MVEBU_COMPHY_TX_SLEW_RATE_SLC(0x3f);
+	writel(val, priv->base + MVEBU_COMPHY_TX_SLEW_RATE(lane->id));
+
+	/* Impedance calibration */
+	val = readl(priv->base + MVEBU_COMPHY_IMP_CAL(lane->id));
+	val &= ~MVEBU_COMPHY_IMP_CAL_TX_EXT(0x1f);
+	val |= MVEBU_COMPHY_IMP_CAL_TX_EXT(0xe) |
+	       MVEBU_COMPHY_IMP_CAL_TX_EXT_EN;
+	writel(val, priv->base + MVEBU_COMPHY_IMP_CAL(lane->id));
+
+	val = readl(priv->base + MVEBU_COMPHY_GEN1_S5(lane->id));
+	val &= ~MVEBU_COMPHY_GEN1_S5_ICP(0xf);
+	writel(val, priv->base + MVEBU_COMPHY_GEN1_S5(lane->id));
+
+	val = readl(priv->base + MVEBU_COMPHY_GEN1_S1(lane->id));
+	val &= ~(MVEBU_COMPHY_GEN1_S1_RX_MUL_PI(0x7) |
+		 MVEBU_COMPHY_GEN1_S1_RX_MUL_PF(0x7) |
+		 MVEBU_COMPHY_GEN1_S1_RX_MUL_FI(0x3) |
+		 MVEBU_COMPHY_GEN1_S1_RX_MUL_FF(0x3));
+	val |= MVEBU_COMPHY_GEN1_S1_RX_DFE_EN |
+	       MVEBU_COMPHY_GEN1_S1_RX_MUL_PI(0x2) |
+	       MVEBU_COMPHY_GEN1_S1_RX_MUL_PF(0x2) |
+	       MVEBU_COMPHY_GEN1_S1_RX_MUL_FF(0x1) |
+	       MVEBU_COMPHY_GEN1_S1_RX_DIV(0x3);
+	writel(val, priv->base + MVEBU_COMPHY_GEN1_S1(lane->id));
+
+	val = readl(priv->base + MVEBU_COMPHY_COEF(lane->id));
+	val &= ~(MVEBU_COMPHY_COEF_DFE_EN | MVEBU_COMPHY_COEF_DFE_CTRL);
+	writel(val, priv->base + MVEBU_COMPHY_COEF(lane->id));
+
+	val = readl(priv->base + MVEBU_COMPHY_GEN1_S4(lane->id));
+	val &= ~MVEBU_COMPHY_GEN1_S4_DFE_RES(0x3);
+	val |= MVEBU_COMPHY_GEN1_S4_DFE_RES(0x1);
+	writel(val, priv->base + MVEBU_COMPHY_GEN1_S4(lane->id));
+
+	val = readl(priv->base + MVEBU_COMPHY_GEN1_S3(lane->id));
+	val |= MVEBU_COMPHY_GEN1_S3_FBCK_SEL;
+	writel(val, priv->base + MVEBU_COMPHY_GEN1_S3(lane->id));
+
+	/* rx training timer */
+	val = readl(priv->base + MVEBU_COMPHY_TRAINING5(lane->id));
+	val &= ~MVEBU_COMPHY_TRAINING5_RX_TIMER(0x3ff);
+	val |= MVEBU_COMPHY_TRAINING5_RX_TIMER(0x13);
+	writel(val, priv->base + MVEBU_COMPHY_TRAINING5(lane->id));
+
+	/* tx train peak to peak hold */
+	val = readl(priv->base + MVEBU_COMPHY_TRAINING0(lane->id));
+	val |= MVEBU_COMPHY_TRAINING0_P2P_HOLD;
+	writel(val, priv->base + MVEBU_COMPHY_TRAINING0(lane->id));
+
+	val = readl(priv->base + MVEBU_COMPHY_TX_PRESET(lane->id));
+	val &= ~MVEBU_COMPHY_TX_PRESET_INDEX(0xf);
+	val |= MVEBU_COMPHY_TX_PRESET_INDEX(0x2);	/* preset coeff */
+	writel(val, priv->base + MVEBU_COMPHY_TX_PRESET(lane->id));
+
+	val = readl(priv->base + MVEBU_COMPHY_FRAME_DETECT3(lane->id));
+	val &= ~MVEBU_COMPHY_FRAME_DETECT3_LOST_TIMEOUT_EN;
+	writel(val, priv->base + MVEBU_COMPHY_FRAME_DETECT3(lane->id));
+
+	val = readl(priv->base + MVEBU_COMPHY_TX_TRAIN_PRESET(lane->id));
+	val |= MVEBU_COMPHY_TX_TRAIN_PRESET_16B_AUTO_EN |
+	       MVEBU_COMPHY_TX_TRAIN_PRESET_PRBS11;
+	writel(val, priv->base + MVEBU_COMPHY_TX_TRAIN_PRESET(lane->id));
+
+	val = readl(priv->base + MVEBU_COMPHY_FRAME_DETECT0(lane->id));
+	val &= ~MVEBU_COMPHY_FRAME_DETECT0_PATN(0x1ff);
+	val |= MVEBU_COMPHY_FRAME_DETECT0_PATN(0x88);
+	writel(val, priv->base + MVEBU_COMPHY_FRAME_DETECT0(lane->id));
+
+	val = readl(priv->base + MVEBU_COMPHY_DME(lane->id));
+	val |= MVEBU_COMPHY_DME_ETH_MODE;
+	writel(val, priv->base + MVEBU_COMPHY_DME(lane->id));
+
+	val = readl(priv->base + MVEBU_COMPHY_VDD_CAL0(lane->id));
+	val |= MVEBU_COMPHY_VDD_CAL0_CONT_MODE;
+	writel(val, priv->base + MVEBU_COMPHY_VDD_CAL0(lane->id));
+
+	val = readl(priv->base + MVEBU_SP_CALIB(lane->id));
+	val &= ~MVEBU_SP_CALIB_SAMPLER(0x3);
+	val |= MVEBU_SP_CALIB_SAMPLER(0x3) |
+	       MVEBU_SP_CALIB_SAMPLER_EN;
+	writel(val, priv->base + MVEBU_SP_CALIB(lane->id));
+	val &= ~MVEBU_SP_CALIB_SAMPLER_EN;
+	writel(val, priv->base + MVEBU_SP_CALIB(lane->id));
+
+	/* External rx regulator */
+	val = readl(priv->base + MVEBU_COMPHY_EXT_SELV(lane->id));
+	val &= ~MVEBU_COMPHY_EXT_SELV_RX_SAMPL(0x1f);
+	val |= MVEBU_COMPHY_EXT_SELV_RX_SAMPL(0x1a);
+	writel(val, priv->base + MVEBU_COMPHY_EXT_SELV(lane->id));
+
+	return mvebu_comphy_init_plls(lane, mode);
+}
+
+static int mvebu_comphy_power_on(struct phy *phy)
+{
+	struct mvebu_comphy_lane *lane = phy_get_drvdata(phy);
+	struct mvebu_comphy_priv *priv = lane->priv;
+	int ret;
+	u32 mux, val;
+
+	mux = mvebu_comphy_get_mux(lane->id, lane->port, lane->mode);
+	if (mux < 0)
+		return -ENOTSUPP;
+
+	regmap_read(priv->regmap, MVEBU_COMPHY_SELECTOR, &val);
+	val &= ~(0xf << MVEBU_COMPHY_SELECTOR_PHY(lane->id));
+	val |= mux << MVEBU_COMPHY_SELECTOR_PHY(lane->id);
+	regmap_write(priv->regmap, MVEBU_COMPHY_SELECTOR, val);
+
+	switch (lane->mode) {
+	case PHY_MODE_SGMII:
+		ret = mvebu_comphy_set_mode_sgmii(phy, lane->mode);
+		break;
+	case PHY_MODE_10GKR:
+		ret = mvebu_comphy_set_mode_10gkr(phy, lane->mode);
+		break;
+	default:
+		return -ENOTSUPP;
+	}
+
+	/* digital reset */
+	val = readl(priv->base + MVEBU_COMPHY_SERDES_CFG1(lane->id));
+	val |= MVEBU_COMPHY_SERDES_CFG1_RF_RESET;
+	writel(val, priv->base + MVEBU_COMPHY_SERDES_CFG1(lane->id));
+
+	return ret;
+}
+
+static int mvebu_comphy_set_mode(struct phy *phy, enum phy_mode mode)
+{
+	struct mvebu_comphy_lane *lane = phy_get_drvdata(phy);
+
+	if (mvebu_comphy_get_mux(lane->id, lane->port, mode) < 0)
+		return -EINVAL;
+
+	lane->mode = mode;
+	return 0;
+}
+
+static int mvebu_comphy_power_off(struct phy *phy)
+{
+	struct mvebu_comphy_lane *lane = phy_get_drvdata(phy);
+	struct mvebu_comphy_priv *priv = lane->priv;
+	u32 val;
+
+	val = readl(priv->base + MVEBU_COMPHY_SERDES_CFG1(lane->id));
+	val &= ~(MVEBU_COMPHY_SERDES_CFG1_RESET |
+		 MVEBU_COMPHY_SERDES_CFG1_CORE_RESET |
+		 MVEBU_COMPHY_SERDES_CFG1_RF_RESET);
+	writel(val, priv->base + MVEBU_COMPHY_SERDES_CFG1(lane->id));
+
+	regmap_read(priv->regmap, MVEBU_COMPHY_SELECTOR, &val);
+	val &= ~(0xf << MVEBU_COMPHY_SELECTOR_PHY(lane->id));
+	regmap_write(priv->regmap, MVEBU_COMPHY_SELECTOR, val);
+
+	return 0;
+}
+
+static const struct phy_ops mvebu_comphy_ops = {
+	.power_on	= mvebu_comphy_power_on,
+	.power_off	= mvebu_comphy_power_off,
+	.set_mode	= mvebu_comphy_set_mode,
+};
+
+static struct phy *mvebu_comphy_xlate(struct device *dev,
+				      struct of_phandle_args *args)
+{
+	struct mvebu_comphy_priv *priv = dev_get_drvdata(dev);
+	struct mvebu_comphy_lane *lane;
+	int i;
+
+	if (WARN_ON(args->args[0] >= MVEBU_COMPHY_PORTS))
+		return ERR_PTR(-EINVAL);
+
+	for (i = 0; i < MVEBU_COMPHY_LANES; i++) {
+		if (!priv->phys[i])
+			continue;
+
+		lane = phy_get_drvdata(priv->phys[i]);
+		if (priv->phys[i] && args->np == lane->of_node)
+			break;
+	}
+
+	if (i == MVEBU_COMPHY_LANES)
+		return ERR_PTR(-ENODEV);
+
+	if (lane->port >= 0)
+		return ERR_PTR(-EBUSY);
+
+	lane->port = args->args[0];
+	return priv->phys[i];
+}
+
+static int mvebu_comphy_probe(struct platform_device *pdev)
+{
+	struct mvebu_comphy_priv *priv;
+	struct phy_provider *provider;
+	struct device_node *child;
+	struct resource *res;
+
+	priv = devm_kzalloc(&pdev->dev, sizeof(*priv), GFP_KERNEL);
+	if (!priv)
+		return -ENOMEM;
+
+	priv->dev = &pdev->dev;
+	priv->regmap =
+		syscon_regmap_lookup_by_phandle(pdev->dev.of_node,
+						"marvell,system-controller");
+	if (IS_ERR(priv->regmap))
+		return PTR_ERR(priv->regmap);
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	priv->base = devm_ioremap_resource(&pdev->dev, res);
+	if (!priv->base)
+		return -ENOMEM;
+
+	for_each_available_child_of_node(pdev->dev.of_node, child) {
+		struct mvebu_comphy_lane *lane;
+		struct phy *phy;
+		int ret;
+		u32 val;
+
+		ret = of_property_read_u32(child, "reg", &val);
+		if (ret < 0) {
+			dev_err(&pdev->dev, "missing 'reg' property (%d)\n",
+				ret);
+			continue;
+		}
+
+		if (val >= MVEBU_COMPHY_LANES) {
+			dev_err(&pdev->dev, "invalid 'reg' property\n");
+			continue;
+		}
+
+		lane = devm_kzalloc(&pdev->dev, sizeof(*lane), GFP_KERNEL);
+		if (!lane)
+			return -ENOMEM;
+
+		phy = devm_phy_create(&pdev->dev, NULL, &mvebu_comphy_ops);
+		if (IS_ERR(phy))
+			return PTR_ERR(phy);
+
+		lane->priv = priv;
+		lane->of_node = child;
+		lane->mode = PHY_MODE_INVALID;
+		lane->id = val;
+		lane->port = -1;
+		phy_set_drvdata(phy, lane);
+
+		priv->phys[val] = phy;
+
+		/*
+		 * Once all modes are supported in this driver we should call
+		 * mvebu_comphy_power_off(phy) here to avoid relying on the
+		 * bootloader/firmware configuration.
+		 */
+	}
+
+	dev_set_drvdata(&pdev->dev, priv);
+	provider = devm_of_phy_provider_register(&pdev->dev,
+						 mvebu_comphy_xlate);
+	return PTR_ERR_OR_ZERO(provider);
+}
+
+static const struct of_device_id mvebu_comphy_of_match_table[] = {
+	{ .compatible = "marvell,comphy-cp110" },
+	{ },
+};
+MODULE_DEVICE_TABLE(of, mvebu_comphy_of_match_table);
+
+static struct platform_driver mvebu_comphy_driver = {
+	.probe	= mvebu_comphy_probe,
+	.driver	= {
+		.name = "mvebu-comphy",
+		.of_match_table = mvebu_comphy_of_match_table,
+	},
+};
+module_platform_driver(mvebu_comphy_driver);
+
+MODULE_AUTHOR("Antoine Tenart <antoine.tenart@free-electrons.com>");
+MODULE_DESCRIPTION("Common PHY driver for mvebu SoCs");
+MODULE_LICENSE("GPL v2");
-- 
2.13.5

[PATCH net-next v2 03/14] Documentation/bindings: phy: document the Marvell comphy driver

[Antoine Tenart]

The Marvell Armada 7K/8K SoCs contains an hardware block called COMPHY
that provides a number of shared PHYs used by various interfaces in the
SoC: network, SATA, PCIe, etc. This Device Tree binding allows to
describe this COMPHY hardware block.

Signed-off-by: Antoine Tenart <antoine.tenart@free-electrons.com>
---
 .../devicetree/bindings/phy/phy-mvebu-comphy.txt   | 43 ++++++++++++++++++++++
 1 file changed, 43 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/phy/phy-mvebu-comphy.txt

diff --git a/Documentation/devicetree/bindings/phy/phy-mvebu-comphy.txt b/Documentation/devicetree/bindings/phy/phy-mvebu-comphy.txt
new file mode 100644
index 000000000000..bfcf80341657
--- /dev/null
+++ b/Documentation/devicetree/bindings/phy/phy-mvebu-comphy.txt
@@ -0,0 +1,43 @@
+mvebu comphy driver
+-------------------
+
+A comphy controller can be found on Marvell Armada 7k/8k on the CP110. It
+provides a number of shared PHYs used by various interfaces (network, sata,
+usb, PCIe...).
+
+Required properties:
+
+- compatible: should be "marvell,comphy-cp110"
+- reg: should contain the comphy register location and length.
+- marvell,system-controller: should contain a phandle to the
+                             system controller node.
+- #address-cells: should be 1.
+- #size-cells: should be 0.
+
+A sub-node is required for each comphy lane provided by the comphy.
+
+Required properties (child nodes):
+
+- reg: comphy lane number.
+- #phy-cells : from the generic phy bindings, must be 1. Defines the
+               input port to use for a given comphy lane.
+
+Example:
+
+	cpm_comphy: phy@120000 {
+		compatible = "marvell,comphy-cp110";
+		reg = <0x120000 0x6000>;
+		marvell,system-controller = <&cpm_syscon0>;
+		#address-cells = <1>;
+		#size-cells = <0>;
+
+		cpm_comphy0: phy@0 {
+			reg = <0>;
+			#phy-cells = <1>;
+		};
+
+		cpm_comphy1: phy@1 {
+			reg = <1>;
+			#phy-cells = <1>;
+		};
+	};
-- 
2.13.5

[PATCH net-next v2 04/14] net: mvpp2: initialize the comphy

[Antoine Tenart]

On some platforms, the comphy is between the MAC GoP and the PHYs. The
mvpp2 driver currently relies on the firmware/bootloader to configure
the comphy. As a comphy driver was added to the generic PHY framework,
this patch uses it in the mvpp2 driver to configure the comphy at boot
time to avoid relying on the bootloader.

Signed-off-by: Antoine Tenart <antoine.tenart@free-electrons.com>
---
 drivers/net/ethernet/marvell/mvpp2.c | 44 +++++++++++++++++++++++++++++++++++-
 1 file changed, 43 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c
index e312dfc3555b..fab231858a41 100644
--- a/drivers/net/ethernet/marvell/mvpp2.c
+++ b/drivers/net/ethernet/marvell/mvpp2.c
@@ -28,6 +28,7 @@
 #include <linux/of_address.h>
 #include <linux/of_device.h>
 #include <linux/phy.h>
+#include <linux/phy/phy.h>
 #include <linux/clk.h>
 #include <linux/hrtimer.h>
 #include <linux/ktime.h>
@@ -861,6 +862,7 @@ struct mvpp2_port {
 
 	phy_interface_t phy_interface;
 	struct device_node *phy_node;
+	struct phy *comphy;
 	unsigned int link;
 	unsigned int duplex;
 	unsigned int speed;
@@ -4420,6 +4422,32 @@ static int mvpp22_gop_init(struct mvpp2_port *port)
 	return -EINVAL;
 }
 
+static int mvpp22_comphy_init(struct mvpp2_port *port)
+{
+	enum phy_mode mode;
+	int ret;
+
+	if (!port->comphy)
+		return 0;
+
+	switch (port->phy_interface) {
+	case PHY_INTERFACE_MODE_SGMII:
+		mode = PHY_MODE_SGMII;
+		break;
+	case PHY_INTERFACE_MODE_10GKR:
+		mode = PHY_MODE_10GKR;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	ret = phy_set_mode(port->comphy, mode);
+	if (ret)
+		return ret;
+
+	return phy_power_on(port->comphy);
+}
+
 static void mvpp2_port_mii_gmac_configure_mode(struct mvpp2_port *port)
 {
 	u32 val;
@@ -6404,8 +6432,10 @@ static void mvpp2_start_dev(struct mvpp2_port *port)
 	/* Enable interrupts on all CPUs */
 	mvpp2_interrupts_enable(port);
 
-	if (port->priv->hw_version == MVPP22)
+	if (port->priv->hw_version == MVPP22) {
+		mvpp22_comphy_init(port);
 		mvpp22_gop_init(port);
+	}
 
 	mvpp2_port_mii_set(port);
 	mvpp2_port_enable(port);
@@ -6436,6 +6466,7 @@ static void mvpp2_stop_dev(struct mvpp2_port *port)
 	mvpp2_egress_disable(port);
 	mvpp2_port_disable(port);
 	phy_stop(ndev->phydev);
+	phy_power_off(port->comphy);
 }
 
 static int mvpp2_check_ringparam_valid(struct net_device *dev,
@@ -7270,6 +7301,7 @@ static int mvpp2_port_probe(struct platform_device *pdev,
 			    struct mvpp2 *priv)
 {
 	struct device_node *phy_node;
+	struct phy *comphy;
 	struct mvpp2_port *port;
 	struct mvpp2_port_pcpu *port_pcpu;
 	struct net_device *dev;
@@ -7311,6 +7343,15 @@ static int mvpp2_port_probe(struct platform_device *pdev,
 		goto err_free_netdev;
 	}
 
+	comphy = devm_of_phy_get(&pdev->dev, port_node, NULL);
+	if (IS_ERR(comphy)) {
+		if (PTR_ERR(comphy) == -EPROBE_DEFER) {
+			err = -EPROBE_DEFER;
+			goto err_free_netdev;
+		}
+		comphy = NULL;
+	}
+
 	if (of_property_read_u32(port_node, "port-id", &id)) {
 		err = -EINVAL;
 		dev_err(&pdev->dev, "missing port-id value\n");
@@ -7344,6 +7385,7 @@ static int mvpp2_port_probe(struct platform_device *pdev,
 
 	port->phy_node = phy_node;
 	port->phy_interface = phy_mode;
+	port->comphy = comphy;
 
 	if (priv->hw_version == MVPP21) {
 		res = platform_get_resource(pdev, IORESOURCE_MEM, 2 + id);
-- 
2.13.5

[PATCH net-next v2 05/14] net: mvpp2: do not force the link mode

[Antoine Tenart]

The link mode (speed, duplex) was forced based on what the phylib
returns. This should not be the case, and only forced by ethtool
functions manually. This patch removes the link mode enforcement from
the phylib link_event callback.

Signed-off-by: Antoine Tenart <antoine.tenart@free-electrons.com>
---
 drivers/net/ethernet/marvell/mvpp2.c | 24 ------------------------
 1 file changed, 24 deletions(-)

diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c
index fab231858a41..498a4969dc58 100644
--- a/drivers/net/ethernet/marvell/mvpp2.c
+++ b/drivers/net/ethernet/marvell/mvpp2.c
@@ -5741,30 +5741,10 @@ static void mvpp2_link_event(struct net_device *dev)
 	struct mvpp2_port *port = netdev_priv(dev);
 	struct phy_device *phydev = dev->phydev;
 	int status_change = 0;
-	u32 val;
 
 	if (phydev->link) {
 		if ((port->speed != phydev->speed) ||
 		    (port->duplex != phydev->duplex)) {
-			u32 val;
-
-			val = readl(port->base + MVPP2_GMAC_AUTONEG_CONFIG);
-			val &= ~(MVPP2_GMAC_CONFIG_MII_SPEED |
-				 MVPP2_GMAC_CONFIG_GMII_SPEED |
-				 MVPP2_GMAC_CONFIG_FULL_DUPLEX |
-				 MVPP2_GMAC_AN_SPEED_EN |
-				 MVPP2_GMAC_AN_DUPLEX_EN);
-
-			if (phydev->duplex)
-				val |= MVPP2_GMAC_CONFIG_FULL_DUPLEX;
-
-			if (phydev->speed == SPEED_1000)
-				val |= MVPP2_GMAC_CONFIG_GMII_SPEED;
-			else if (phydev->speed == SPEED_100)
-				val |= MVPP2_GMAC_CONFIG_MII_SPEED;
-
-			writel(val, port->base + MVPP2_GMAC_AUTONEG_CONFIG);
-
 			port->duplex = phydev->duplex;
 			port->speed  = phydev->speed;
 		}
@@ -5782,10 +5762,6 @@ static void mvpp2_link_event(struct net_device *dev)
 
 	if (status_change) {
 		if (phydev->link) {
-			val = readl(port->base + MVPP2_GMAC_AUTONEG_CONFIG);
-			val |= (MVPP2_GMAC_FORCE_LINK_PASS |
-				MVPP2_GMAC_FORCE_LINK_DOWN);
-			writel(val, port->base + MVPP2_GMAC_AUTONEG_CONFIG);
 			mvpp2_egress_enable(port);
 			mvpp2_ingress_enable(port);
 		} else {
-- 
2.13.5

[PATCH net-next v2 06/14] net: mvpp2: simplify the link_event function

[Antoine Tenart]

The link_event function is somewhat complicated. This cosmetic patch
simplifies it.

Signed-off-by: Antoine Tenart <antoine.tenart@free-electrons.com>
---
 drivers/net/ethernet/marvell/mvpp2.c | 13 ++++---------
 1 file changed, 4 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c
index 498a4969dc58..a9dea9b344de 100644
--- a/drivers/net/ethernet/marvell/mvpp2.c
+++ b/drivers/net/ethernet/marvell/mvpp2.c
@@ -5740,7 +5740,6 @@ static void mvpp2_link_event(struct net_device *dev)
 {
 	struct mvpp2_port *port = netdev_priv(dev);
 	struct phy_device *phydev = dev->phydev;
-	int status_change = 0;
 
 	if (phydev->link) {
 		if ((port->speed != phydev->speed) ||
@@ -5751,23 +5750,19 @@ static void mvpp2_link_event(struct net_device *dev)
 	}
 
 	if (phydev->link != port->link) {
-		if (!phydev->link) {
-			port->duplex = -1;
-			port->speed = 0;
-		}
-
 		port->link = phydev->link;
-		status_change = 1;
-	}
 
-	if (status_change) {
 		if (phydev->link) {
 			mvpp2_egress_enable(port);
 			mvpp2_ingress_enable(port);
 		} else {
+			port->duplex = -1;
+			port->speed = 0;
+
 			mvpp2_ingress_disable(port);
 			mvpp2_egress_disable(port);
 		}
+
 		phy_print_status(phydev);
 	}
 }
-- 
2.13.5

[PATCH net-next v2 07/14] net: mvpp2: improve the link management function

[Antoine Tenart]

When the link status changes, the phylib calls the link_event function
in the mvpp2 driver. Before this patch only the egress/ingress transmit
was enabled/disabled. This patch adds more functionality to the link
status management code by enabling/disabling the port per-cpu
interrupts, and the port itself. The queues are now stopped as well, and
the netif carrier helpers are called.

Signed-off-by: Antoine Tenart <antoine.tenart@free-electrons.com>
---
 drivers/net/ethernet/marvell/mvpp2.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c
index a9dea9b344de..1b26f5ed994f 100644
--- a/drivers/net/ethernet/marvell/mvpp2.c
+++ b/drivers/net/ethernet/marvell/mvpp2.c
@@ -5753,14 +5753,24 @@ static void mvpp2_link_event(struct net_device *dev)
 		port->link = phydev->link;
 
 		if (phydev->link) {
+			mvpp2_interrupts_enable(port);
+			mvpp2_port_enable(port);
+
 			mvpp2_egress_enable(port);
 			mvpp2_ingress_enable(port);
+			netif_carrier_on(dev);
+			netif_tx_wake_all_queues(dev);
 		} else {
 			port->duplex = -1;
 			port->speed = 0;
 
+			netif_tx_stop_all_queues(dev);
+			netif_carrier_off(dev);
 			mvpp2_ingress_disable(port);
 			mvpp2_egress_disable(port);
+
+			mvpp2_port_disable(port);
+			mvpp2_interrupts_disable(port);
 		}
 
 		phy_print_status(phydev);
-- 
2.13.5

[PATCH net-next v2 08/14] net: mvpp2: check the netif is running in the link_event function

[Antoine Tenart]

This patch adds an extra check when the link_event function is called,
so that it won't do anything when the netif isn't running.

Signed-off-by: Antoine Tenart <antoine.tenart@free-electrons.com>
---
 drivers/net/ethernet/marvell/mvpp2.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c
index 1b26f5ed994f..49a6789a4142 100644
--- a/drivers/net/ethernet/marvell/mvpp2.c
+++ b/drivers/net/ethernet/marvell/mvpp2.c
@@ -5741,6 +5741,9 @@ static void mvpp2_link_event(struct net_device *dev)
 	struct mvpp2_port *port = netdev_priv(dev);
 	struct phy_device *phydev = dev->phydev;
 
+	if (!netif_running(dev))
+		return;
+
 	if (phydev->link) {
 		if ((port->speed != phydev->speed) ||
 		    (port->duplex != phydev->duplex)) {
-- 
2.13.5

[PATCH net-next v2 09/14] net: mvpp2: dynamic reconfiguration of the PHY mode

[Antoine Tenart]

This patch adds logic to reconfigure the comphy/gop when the link status
change at runtime. This is very useful on boards such as the mcbin which
have SFP and Ethernet ports connected to the same MAC port: depending on
what the user connects the driver will automatically reconfigure the
link mode.

Signed-off-by: Antoine Tenart <antoine.tenart@free-electrons.com>
---
 drivers/net/ethernet/marvell/mvpp2.c | 21 ++++++++++++++++++++-
 1 file changed, 20 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c
index 49a6789a4142..04e0c8ab7b51 100644
--- a/drivers/net/ethernet/marvell/mvpp2.c
+++ b/drivers/net/ethernet/marvell/mvpp2.c
@@ -5740,6 +5740,7 @@ static void mvpp2_link_event(struct net_device *dev)
 {
 	struct mvpp2_port *port = netdev_priv(dev);
 	struct phy_device *phydev = dev->phydev;
+	bool link_reconfigured = false;
 
 	if (!netif_running(dev))
 		return;
@@ -5750,9 +5751,27 @@ static void mvpp2_link_event(struct net_device *dev)
 			port->duplex = phydev->duplex;
 			port->speed  = phydev->speed;
 		}
+
+		if (port->phy_interface != phydev->interface && port->comphy) {
+	                /* disable current port for reconfiguration */
+	                mvpp2_interrupts_disable(port);
+	                netif_carrier_off(port->dev);
+	                mvpp2_port_disable(port);
+			phy_power_off(port->comphy);
+
+	                /* comphy reconfiguration */
+	                port->phy_interface = phydev->interface;
+	                mvpp22_comphy_init(port);
+
+	                /* gop/mac reconfiguration */
+	                mvpp22_gop_init(port);
+	                mvpp2_port_mii_set(port);
+
+	                link_reconfigured = true;
+		}
 	}
 
-	if (phydev->link != port->link) {
+	if (phydev->link != port->link || link_reconfigured) {
 		port->link = phydev->link;
 
 		if (phydev->link) {
-- 
2.13.5

[PATCH net-next v2 10/14] arm64: dts: marvell: extend the cp110 syscon register area length

[Antoine Tenart]

This patch extends on both cp110 the system register area length to
include some of the comphy registers as well.

Signed-off-by: Antoine Tenart <antoine.tenart@free-electrons.com>
---
 arch/arm64/boot/dts/marvell/armada-cp110-master.dtsi | 2 +-
 arch/arm64/boot/dts/marvell/armada-cp110-slave.dtsi  | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/arm64/boot/dts/marvell/armada-cp110-master.dtsi b/arch/arm64/boot/dts/marvell/armada-cp110-master.dtsi
index 18299e164cb7..9b2581473183 100644
--- a/arch/arm64/boot/dts/marvell/armada-cp110-master.dtsi
+++ b/arch/arm64/boot/dts/marvell/armada-cp110-master.dtsi
@@ -118,7 +118,7 @@
 
 			cpm_syscon0: system-controller@440000 {
 				compatible = "syscon", "simple-mfd";
-				reg = <0x440000 0x1000>;
+				reg = <0x440000 0x2000>;
 
 				cpm_clk: clock {
 					compatible = "marvell,cp110-clock";
diff --git a/arch/arm64/boot/dts/marvell/armada-cp110-slave.dtsi b/arch/arm64/boot/dts/marvell/armada-cp110-slave.dtsi
index 5ae8fa575859..d3902f218c46 100644
--- a/arch/arm64/boot/dts/marvell/armada-cp110-slave.dtsi
+++ b/arch/arm64/boot/dts/marvell/armada-cp110-slave.dtsi
@@ -125,7 +125,7 @@
 
 			cps_syscon0: system-controller@440000 {
 				compatible = "syscon", "simple-mfd";
-				reg = <0x440000 0x1000>;
+				reg = <0x440000 0x2000>;
 
 				cps_clk: clock {
 					compatible = "marvell,cp110-clock";
-- 
2.13.5

[PATCH net-next v2 11/14] arm64: dts: marvell: add comphy nodes on cp110 master and slave

[Antoine Tenart]

Now that the comphy driver is available, this patch adds the
corresponding nodes in the cp110 master and slave device trees.

Signed-off-by: Antoine Tenart <antoine.tenart@free-electrons.com>
---
 .../boot/dts/marvell/armada-cp110-master.dtsi      | 38 ++++++++++++++++++++++
 .../arm64/boot/dts/marvell/armada-cp110-slave.dtsi | 38 ++++++++++++++++++++++
 2 files changed, 76 insertions(+)

diff --git a/arch/arm64/boot/dts/marvell/armada-cp110-master.dtsi b/arch/arm64/boot/dts/marvell/armada-cp110-master.dtsi
index 9b2581473183..f2a50552bad4 100644
--- a/arch/arm64/boot/dts/marvell/armada-cp110-master.dtsi
+++ b/arch/arm64/boot/dts/marvell/armada-cp110-master.dtsi
@@ -91,6 +91,44 @@
 				};
 			};
 
+			cpm_comphy: phy@120000 {
+				compatible = "marvell,comphy-cp110";
+				reg = <0x120000 0x6000>;
+				marvell,system-controller = <&cpm_syscon0>;
+				#address-cells = <1>;
+				#size-cells = <0>;
+
+				cpm_comphy0: phy@0 {
+					reg = <0>;
+					#phy-cells = <1>;
+				};
+
+				cpm_comphy1: phy@1 {
+					reg = <1>;
+					#phy-cells = <1>;
+				};
+
+				cpm_comphy2: phy@2 {
+					reg = <2>;
+					#phy-cells = <1>;
+				};
+
+				cpm_comphy3: phy@3 {
+					reg = <3>;
+					#phy-cells = <1>;
+				};
+
+				cpm_comphy4: phy@4 {
+					reg = <4>;
+					#phy-cells = <1>;
+				};
+
+				cpm_comphy5: phy@5 {
+					reg = <5>;
+					#phy-cells = <1>;
+				};
+			};
+
 			cpm_mdio: mdio@12a200 {
 				#address-cells = <1>;
 				#size-cells = <0>;
diff --git a/arch/arm64/boot/dts/marvell/armada-cp110-slave.dtsi b/arch/arm64/boot/dts/marvell/armada-cp110-slave.dtsi
index d3902f218c46..bd7f7d0e6de9 100644
--- a/arch/arm64/boot/dts/marvell/armada-cp110-slave.dtsi
+++ b/arch/arm64/boot/dts/marvell/armada-cp110-slave.dtsi
@@ -98,6 +98,44 @@
 				};
 			};
 
+			cps_comphy: phy@120000 {
+				compatible = "marvell,comphy-cp110";
+				reg = <0x120000 0x6000>;
+				marvell,system-controller = <&cps_syscon0>;
+				#address-cells = <1>;
+				#size-cells = <0>;
+
+				cps_comphy0: phy@0 {
+					reg = <0>;
+					#phy-cells = <1>;
+				};
+
+				cps_comphy1: phy@1 {
+					reg = <1>;
+					#phy-cells = <1>;
+				};
+
+				cps_comphy2: phy@2 {
+					reg = <2>;
+					#phy-cells = <1>;
+				};
+
+				cps_comphy3: phy@3 {
+					reg = <3>;
+					#phy-cells = <1>;
+				};
+
+				cps_comphy4: phy@4 {
+					reg = <4>;
+					#phy-cells = <1>;
+				};
+
+				cps_comphy5: phy@5 {
+					reg = <5>;
+					#phy-cells = <1>;
+				};
+			};
+
 			cps_mdio: mdio@12a200 {
 				#address-cells = <1>;
 				#size-cells = <0>;
-- 
2.13.5

[PATCH net-next v2 12/14] arm64: dts: marvell: mcbin: add comphy references to Ethernet ports

[Antoine Tenart]

This patch adds comphy phandles to the Ethernet ports in the mcbin
device tree. The comphy is used to configure the serdes PHYs used by
these ports.

Signed-off-by: Antoine Tenart <antoine.tenart@free-electrons.com>
---
 arch/arm64/boot/dts/marvell/armada-8040-mcbin.dts | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/arch/arm64/boot/dts/marvell/armada-8040-mcbin.dts b/arch/arm64/boot/dts/marvell/armada-8040-mcbin.dts
index 9f0a00802452..970081ca197b 100644
--- a/arch/arm64/boot/dts/marvell/armada-8040-mcbin.dts
+++ b/arch/arm64/boot/dts/marvell/armada-8040-mcbin.dts
@@ -148,6 +148,7 @@
 &cpm_eth0 {
 	status = "okay";
 	phy = <&phy0>;
+	phys = <&cpm_comphy4 0>;
 	phy-mode = "10gbase-kr";
 };
 
@@ -181,6 +182,7 @@
 &cps_eth0 {
 	status = "okay";
 	phy = <&phy8>;
+	phys = <&cps_comphy4 0>;
 	phy-mode = "10gbase-kr";
 };
 
@@ -189,6 +191,7 @@
 	status = "okay";
 	phy = <&ge_phy>;
 	phy-mode = "sgmii";
+	phys = <&cps_comphy0 1>;
 };
 
 &cps_sata0 {
-- 
2.13.5

[PATCH net-next v2 13/14] arm64: dts: marvell: 7040-db: add comphy references to Ethernet ports

[Antoine Tenart]

This patch adds comphy phandles to the Ethernet ports in the 7040-db
device tree. The comphy is used to configure the serdes PHYs used by
these ports.

Signed-off-by: Antoine Tenart <antoine.tenart@free-electrons.com>
---
 arch/arm64/boot/dts/marvell/armada-7040-db.dts | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/arm64/boot/dts/marvell/armada-7040-db.dts b/arch/arm64/boot/dts/marvell/armada-7040-db.dts
index 92c761c380d3..03d1c42d7c47 100644
--- a/arch/arm64/boot/dts/marvell/armada-7040-db.dts
+++ b/arch/arm64/boot/dts/marvell/armada-7040-db.dts
@@ -180,6 +180,7 @@
 	status = "okay";
 	phy = <&phy0>;
 	phy-mode = "sgmii";
+	phys = <&cpm_comphy0 1>;
 };
 
 &cpm_eth2 {
-- 
2.13.5

[PATCH net-next v2 14/14] arm64: defconfig: enable Marvell CP110 comphy

[Antoine Tenart]

From: Miquel Raynal <miquel.raynal@free-electrons.com>

The comphy is an hardware block giving access to common PHYs that can be
used by various other engines (Network, SATA, ...). This is used on
Marvell 7k/8k platforms for now. Enable the corresponding driver.

Signed-off-by: Miquel Raynal <miquel.raynal@free-electrons.com>
Signed-off-by: Antoine Tenart <antoine.tenart@free-electrons.com>
---
 arch/arm64/configs/defconfig | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/arm64/configs/defconfig b/arch/arm64/configs/defconfig
index cdde4f56a281..e671e37b30af 100644
--- a/arch/arm64/configs/defconfig
+++ b/arch/arm64/configs/defconfig
@@ -509,6 +509,7 @@ CONFIG_PWM_TEGRA=m
 CONFIG_PHY_RCAR_GEN3_USB2=y
 CONFIG_PHY_HI6220_USB=y
 CONFIG_PHY_SUN4I_USB=y
+CONFIG_PHY_MVEBU_CP110_COMPHY=y
 CONFIG_PHY_ROCKCHIP_INNO_USB2=y
 CONFIG_PHY_ROCKCHIP_EMMC=y
 CONFIG_PHY_ROCKCHIP_PCIE=m
-- 
2.13.5

Re: [PATCH net-next v2 00/14] net: mvpp2: comphy configuration

[Andrew Lunn]

>   - Checked if the carrier_on/off functions were needed. They are.

Hi Antoine

Could you explain the situations they are needed in?

Quite a few drivers do this, so i'm not saying it is wrong.  But it
would be nice to understand if we are missing something in phylib.

      Andrew

Re: [PATCH net-next v2 00/14] net: mvpp2: comphy configuration

[Antoine Tenart]

[-- Attachment #1: Type: text/plain, Size: 574 bytes --]

Hi Andrew,

On Fri, Aug 25, 2017 at 05:51:11PM +0200, Andrew Lunn wrote:
> >   - Checked if the carrier_on/off functions were needed. They are.
> 
> Could you explain the situations they are needed in?
> 
> Quite a few drivers do this, so i'm not saying it is wrong.  But it
> would be nice to understand if we are missing something in phylib.

I know it's not working without these calls, but I can't tell why (yet).
I can try to dig into this.

Antoine

-- 
Antoine Ténart, Free Electrons
Embedded Linux and Kernel engineering
http://free-electrons.com

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: [PATCH net-next v2 08/14] net: mvpp2: check the netif is running in the link_event function

[Florian Fainelli]

On 08/25/2017 07:48 AM, Antoine Tenart wrote:
> This patch adds an extra check when the link_event function is called,
> so that it won't do anything when the netif isn't running.

Why is this needed? Are you possibly starting the PHY state machine
earlier than your ndo_open() call? Looking quickly through the driver
does not suggest this is going on since you properly connect to the PHY
in mvpp2_open() and start the PHY there.

> 
> Signed-off-by: Antoine Tenart <antoine.tenart@free-electrons.com>
> ---
>  drivers/net/ethernet/marvell/mvpp2.c | 3 +++
>  1 file changed, 3 insertions(+)
> 
> diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c
> index 1b26f5ed994f..49a6789a4142 100644
> --- a/drivers/net/ethernet/marvell/mvpp2.c
> +++ b/drivers/net/ethernet/marvell/mvpp2.c
> @@ -5741,6 +5741,9 @@ static void mvpp2_link_event(struct net_device *dev)
>  	struct mvpp2_port *port = netdev_priv(dev);
>  	struct phy_device *phydev = dev->phydev;
>  
> +	if (!netif_running(dev))
> +		return;
> +
>  	if (phydev->link) {
>  		if ((port->speed != phydev->speed) ||
>  		    (port->duplex != phydev->duplex)) {
> 


-- 
Florian

Re: [PATCH net-next v2 08/14] net: mvpp2: check the netif is running in the link_event function

[Antoine Tenart]

[-- Attachment #1: Type: text/plain, Size: 821 bytes --]

Hi Florian,

On Fri, Aug 25, 2017 at 09:49:15AM -0700, Florian Fainelli wrote:
> On 08/25/2017 07:48 AM, Antoine Tenart wrote:
> > This patch adds an extra check when the link_event function is called,
> > so that it won't do anything when the netif isn't running.
> 
> Why is this needed? Are you possibly starting the PHY state machine
> earlier than your ndo_open() call? Looking quickly through the driver
> does not suggest this is going on since you properly connect to the PHY
> in mvpp2_open() and start the PHY there.

I added some checks while working on this, and kept this one. But I
looked at the driver again and I assume you're right and the patch could
be dropped.

Thanks!
Antoine

-- 
Antoine Ténart, Free Electrons
Embedded Linux and Kernel engineering
http://free-electrons.com

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: [PATCH net-next v2 05/14] net: mvpp2: do not force the link mode

[Russell King - ARM Linux]

On Fri, Aug 25, 2017 at 04:48:12PM +0200, Antoine Tenart wrote:
> The link mode (speed, duplex) was forced based on what the phylib
> returns. This should not be the case, and only forced by ethtool
> functions manually. This patch removes the link mode enforcement from
> the phylib link_event callback.

So how does RGMII work (which has no in-band signalling between the PHY
and MAC)?

phylib expects the network driver to configure it according to the PHY
state at link_event time - I think you need to explain more why you
think that this is not necessary.

> 
> Signed-off-by: Antoine Tenart <antoine.tenart@free-electrons.com>
> ---
>  drivers/net/ethernet/marvell/mvpp2.c | 24 ------------------------
>  1 file changed, 24 deletions(-)
> 
> diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c
> index fab231858a41..498a4969dc58 100644
> --- a/drivers/net/ethernet/marvell/mvpp2.c
> +++ b/drivers/net/ethernet/marvell/mvpp2.c
> @@ -5741,30 +5741,10 @@ static void mvpp2_link_event(struct net_device *dev)
>  	struct mvpp2_port *port = netdev_priv(dev);
>  	struct phy_device *phydev = dev->phydev;
>  	int status_change = 0;
> -	u32 val;
>  
>  	if (phydev->link) {
>  		if ((port->speed != phydev->speed) ||
>  		    (port->duplex != phydev->duplex)) {
> -			u32 val;
> -
> -			val = readl(port->base + MVPP2_GMAC_AUTONEG_CONFIG);
> -			val &= ~(MVPP2_GMAC_CONFIG_MII_SPEED |
> -				 MVPP2_GMAC_CONFIG_GMII_SPEED |
> -				 MVPP2_GMAC_CONFIG_FULL_DUPLEX |
> -				 MVPP2_GMAC_AN_SPEED_EN |
> -				 MVPP2_GMAC_AN_DUPLEX_EN);
> -
> -			if (phydev->duplex)
> -				val |= MVPP2_GMAC_CONFIG_FULL_DUPLEX;
> -
> -			if (phydev->speed == SPEED_1000)
> -				val |= MVPP2_GMAC_CONFIG_GMII_SPEED;
> -			else if (phydev->speed == SPEED_100)
> -				val |= MVPP2_GMAC_CONFIG_MII_SPEED;
> -
> -			writel(val, port->base + MVPP2_GMAC_AUTONEG_CONFIG);
> -
>  			port->duplex = phydev->duplex;
>  			port->speed  = phydev->speed;
>  		}
> @@ -5782,10 +5762,6 @@ static void mvpp2_link_event(struct net_device *dev)
>  
>  	if (status_change) {
>  		if (phydev->link) {
> -			val = readl(port->base + MVPP2_GMAC_AUTONEG_CONFIG);
> -			val |= (MVPP2_GMAC_FORCE_LINK_PASS |
> -				MVPP2_GMAC_FORCE_LINK_DOWN);
> -			writel(val, port->base + MVPP2_GMAC_AUTONEG_CONFIG);
>  			mvpp2_egress_enable(port);
>  			mvpp2_ingress_enable(port);
>  		} else {
> -- 
> 2.13.5
> 

-- 
RMK's Patch system: http://www.armlinux.org.uk/developer/patches/
FTTC broadband for 0.8mile line in suburbia: sync at 8.8Mbps down 630kbps up
According to speedtest.net: 8.21Mbps down 510kbps up

Re: [PATCH net-next v2 09/14] net: mvpp2: dynamic reconfiguration of the PHY mode

[Russell King - ARM Linux]

On Fri, Aug 25, 2017 at 04:48:16PM +0200, Antoine Tenart wrote:
> This patch adds logic to reconfigure the comphy/gop when the link status
> change at runtime. This is very useful on boards such as the mcbin which
> have SFP and Ethernet ports connected to the same MAC port: depending on
> what the user connects the driver will automatically reconfigure the
> link mode.

This commit commentry needs updating - as I've already pointed out in
the previous round, the need to reconfigure things has *nothing* to do
with there being SFP and "Ethernet" ports present.  Hence, your commit
message is entirely misleading.

> 
> Signed-off-by: Antoine Tenart <antoine.tenart@free-electrons.com>
> ---
>  drivers/net/ethernet/marvell/mvpp2.c | 21 ++++++++++++++++++++-
>  1 file changed, 20 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c
> index 49a6789a4142..04e0c8ab7b51 100644
> --- a/drivers/net/ethernet/marvell/mvpp2.c
> +++ b/drivers/net/ethernet/marvell/mvpp2.c
> @@ -5740,6 +5740,7 @@ static void mvpp2_link_event(struct net_device *dev)
>  {
>  	struct mvpp2_port *port = netdev_priv(dev);
>  	struct phy_device *phydev = dev->phydev;
> +	bool link_reconfigured = false;
>  
>  	if (!netif_running(dev))
>  		return;
> @@ -5750,9 +5751,27 @@ static void mvpp2_link_event(struct net_device *dev)
>  			port->duplex = phydev->duplex;
>  			port->speed  = phydev->speed;
>  		}
> +
> +		if (port->phy_interface != phydev->interface && port->comphy) {
> +	                /* disable current port for reconfiguration */
> +	                mvpp2_interrupts_disable(port);
> +	                netif_carrier_off(port->dev);
> +	                mvpp2_port_disable(port);
> +			phy_power_off(port->comphy);
> +
> +	                /* comphy reconfiguration */
> +	                port->phy_interface = phydev->interface;
> +	                mvpp22_comphy_init(port);
> +
> +	                /* gop/mac reconfiguration */
> +	                mvpp22_gop_init(port);
> +	                mvpp2_port_mii_set(port);
> +
> +	                link_reconfigured = true;
> +		}
>  	}
>  
> -	if (phydev->link != port->link) {
> +	if (phydev->link != port->link || link_reconfigured) {
>  		port->link = phydev->link;
>  
>  		if (phydev->link) {
> -- 
> 2.13.5
> 

-- 
RMK's Patch system: http://www.armlinux.org.uk/developer/patches/
FTTC broadband for 0.8mile line in suburbia: sync at 8.8Mbps down 630kbps up
According to speedtest.net: 8.21Mbps down 510kbps up

Re: [PATCH net-next v2 09/14] net: mvpp2: dynamic reconfiguration of the PHY mode

[Antoine Tenart]

[-- Attachment #1: Type: text/plain, Size: 915 bytes --]

Hi Russell,

On Fri, Aug 25, 2017 at 11:46:16PM +0100, Russell King - ARM Linux wrote:
> On Fri, Aug 25, 2017 at 04:48:16PM +0200, Antoine Tenart wrote:
> > This patch adds logic to reconfigure the comphy/gop when the link status
> > change at runtime. This is very useful on boards such as the mcbin which
> > have SFP and Ethernet ports connected to the same MAC port: depending on
> > what the user connects the driver will automatically reconfigure the
> > link mode.
> 
> This commit commentry needs updating - as I've already pointed out in
> the previous round, the need to reconfigure things has *nothing* to do
> with there being SFP and "Ethernet" ports present.  Hence, your commit
> message is entirely misleading.

That's right. I'll update the commit message.

Thanks!
Antoine

-- 
Antoine Ténart, Free Electrons
Embedded Linux and Kernel engineering
http://free-electrons.com

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: [PATCH net-next v2 05/14] net: mvpp2: do not force the link mode

[Antoine Tenart]

[-- Attachment #1: Type: text/plain, Size: 1086 bytes --]

Hi Russell,

On Fri, Aug 25, 2017 at 11:43:13PM +0100, Russell King - ARM Linux wrote:
> On Fri, Aug 25, 2017 at 04:48:12PM +0200, Antoine Tenart wrote:
> > The link mode (speed, duplex) was forced based on what the phylib
> > returns. This should not be the case, and only forced by ethtool
> > functions manually. This patch removes the link mode enforcement from
> > the phylib link_event callback.
> 
> So how does RGMII work (which has no in-band signalling between the PHY
> and MAC)?
> 
> phylib expects the network driver to configure it according to the PHY
> state at link_event time - I think you need to explain more why you
> think that this is not necessary.

Good catch, this won't work properly with RGMII. This could be done
out-of-band according to the spec, but that would use PHY polling and we
do not want that (the same concern was raised by Andrew on another
patch).

I'll keep this mode enforcement for RGMII then.

Thanks!
Antoine

-- 
Antoine Ténart, Free Electrons
Embedded Linux and Kernel engineering
http://free-electrons.com

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: [PATCH net-next v2 05/14] net: mvpp2: do not force the link mode

[Marcin Wojtas]

Hi Antoine,

2017-08-28 8:55 GMT+02:00 Antoine Tenart <antoine.tenart@free-electrons.com>:
> Hi Russell,
>
> On Fri, Aug 25, 2017 at 11:43:13PM +0100, Russell King - ARM Linux wrote:
>> On Fri, Aug 25, 2017 at 04:48:12PM +0200, Antoine Tenart wrote:
>> > The link mode (speed, duplex) was forced based on what the phylib
>> > returns. This should not be the case, and only forced by ethtool
>> > functions manually. This patch removes the link mode enforcement from
>> > the phylib link_event callback.
>>
>> So how does RGMII work (which has no in-band signalling between the PHY
>> and MAC)?
>>
>> phylib expects the network driver to configure it according to the PHY
>> state at link_event time - I think you need to explain more why you
>> think that this is not necessary.
>
> Good catch, this won't work properly with RGMII. This could be done
> out-of-band according to the spec, but that would use PHY polling and we
> do not want that (the same concern was raised by Andrew on another
> patch).
>
> I'll keep this mode enforcement for RGMII then.
>

Can you be 100% sure that when using SGMII with PHY's (like Marvell
Alaska 88E1xxx series), is in-band link information always available?
I'd be very cautious with such assumption and use in-band management
only when set in the DT, like mvneta. I think phylib can properly can
do its work when MDIO connection is provided on the board.

Did you check the change also on A375?

Best regards,
Marcin

Re: [PATCH net-next v2 05/14] net: mvpp2: do not force the link mode

[Russell King - ARM Linux]

On Mon, Aug 28, 2017 at 10:38:37AM +0200, Marcin Wojtas wrote:
> Hi Antoine,
>
> Can you be 100% sure that when using SGMII with PHY's (like Marvell
> Alaska 88E1xxx series), is in-band link information always available?
> I'd be very cautious with such assumption and use in-band management
> only when set in the DT, like mvneta. I think phylib can properly can
> do its work when MDIO connection is provided on the board.

There is another issue to be aware of: if you're wanting to use flow
control autonegotiation, that is not carried across SGMII's in-band
signalling.  If you want to use SGMII's in-band signalling for the
duplex and speed information, you still need phylib's notification
to properly set the flow control.

<plug>
Switching mvpp2 to use phylink (which is needed for the 1G SFP slot on
mcbin) will handle all this for you - dealing with both in-band and
out-of-band negotiation methods, and combining them in the appropriate
manner for the selected operation mode.
</plug>

-- 
RMK's Patch system: http://www.armlinux.org.uk/developer/patches/
FTTC broadband for 0.8mile line in suburbia: sync at 8.8Mbps down 630kbps up
According to speedtest.net: 8.21Mbps down 510kbps up

Re: [PATCH net-next v2 05/14] net: mvpp2: do not force the link mode

[Antoine Tenart]

[-- Attachment #1: Type: text/plain, Size: 1356 bytes --]

On Mon, Aug 28, 2017 at 09:51:52AM +0100, Russell King - ARM Linux wrote:
> On Mon, Aug 28, 2017 at 10:38:37AM +0200, Marcin Wojtas wrote:
> >
> > Can you be 100% sure that when using SGMII with PHY's (like Marvell
> > Alaska 88E1xxx series), is in-band link information always available?
> > I'd be very cautious with such assumption and use in-band management
> > only when set in the DT, like mvneta. I think phylib can properly can
> > do its work when MDIO connection is provided on the board.
> 
> There is another issue to be aware of: if you're wanting to use flow
> control autonegotiation, that is not carried across SGMII's in-band
> signalling.  If you want to use SGMII's in-band signalling for the
> duplex and speed information, you still need phylib's notification
> to properly set the flow control.
> 
> <plug>
> Switching mvpp2 to use phylink (which is needed for the 1G SFP slot on
> mcbin) will handle all this for you - dealing with both in-band and
> out-of-band negotiation methods, and combining them in the appropriate
> manner for the selected operation mode.
> </plug>

So probably the best move here is to remove this patch, and wait for the
phylink support in the PPv2 driver.

Thanks!
Antoine

-- 
Antoine Ténart, Free Electrons
Embedded Linux and Kernel engineering
http://free-electrons.com

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: [PATCH net-next v2 05/14] net: mvpp2: do not force the link mode

[Russell King - ARM Linux]

On Mon, Aug 28, 2017 at 11:40:51AM +0200, Antoine Tenart wrote:
> On Mon, Aug 28, 2017 at 09:51:52AM +0100, Russell King - ARM Linux wrote:
> > On Mon, Aug 28, 2017 at 10:38:37AM +0200, Marcin Wojtas wrote:
> > >
> > > Can you be 100% sure that when using SGMII with PHY's (like Marvell
> > > Alaska 88E1xxx series), is in-band link information always available?
> > > I'd be very cautious with such assumption and use in-band management
> > > only when set in the DT, like mvneta. I think phylib can properly can
> > > do its work when MDIO connection is provided on the board.
> > 
> > There is another issue to be aware of: if you're wanting to use flow
> > control autonegotiation, that is not carried across SGMII's in-band
> > signalling.  If you want to use SGMII's in-band signalling for the
> > duplex and speed information, you still need phylib's notification
> > to properly set the flow control.
> > 
> > <plug>
> > Switching mvpp2 to use phylink (which is needed for the 1G SFP slot on
> > mcbin) will handle all this for you - dealing with both in-band and
> > out-of-band negotiation methods, and combining them in the appropriate
> > manner for the selected operation mode.
> > </plug>
> 
> So probably the best move here is to remove this patch, and wait for the
> phylink support in the PPv2 driver.

I've nothing on that specifically for the mvpp2 driver - what I have is
for mvneta and the Marvell mvpp2x driver, with GMAC support extracted
from mvneta (that last bit is rather dirty at the moment so not
published anywhere, and doesn't cater for PP v2.1 at all.)

I ought to have posted the mvneta part of the phylink patches, but I
didn't get around to it early enough in this cycle - there are probably
quite a number of conflicts with net-next now, so I think it's too late
to submit it for mainline.

I know Andrew has already looked at them in my git tree as part of the
review of phylink when that was merged - which should be adequate to
give an example of how to implement it for the mainline PP v2 driver.

-- 
RMK's Patch system: http://www.armlinux.org.uk/developer/patches/
FTTC broadband for 0.8mile line in suburbia: sync at 8.8Mbps down 630kbps up
According to speedtest.net: 8.21Mbps down 510kbps up

Re: [PATCH net-next v2 05/14] net: mvpp2: do not force the link mode

[Antoine Tenart]

[-- Attachment #1: Type: text/plain, Size: 2557 bytes --]

On Mon, Aug 28, 2017 at 12:06:24PM +0100, Russell King - ARM Linux wrote:
> On Mon, Aug 28, 2017 at 11:40:51AM +0200, Antoine Tenart wrote:
> > On Mon, Aug 28, 2017 at 09:51:52AM +0100, Russell King - ARM Linux wrote:
> > > On Mon, Aug 28, 2017 at 10:38:37AM +0200, Marcin Wojtas wrote:
> > > >
> > > > Can you be 100% sure that when using SGMII with PHY's (like Marvell
> > > > Alaska 88E1xxx series), is in-band link information always available?
> > > > I'd be very cautious with such assumption and use in-band management
> > > > only when set in the DT, like mvneta. I think phylib can properly can
> > > > do its work when MDIO connection is provided on the board.
> > > 
> > > There is another issue to be aware of: if you're wanting to use flow
> > > control autonegotiation, that is not carried across SGMII's in-band
> > > signalling.  If you want to use SGMII's in-band signalling for the
> > > duplex and speed information, you still need phylib's notification
> > > to properly set the flow control.
> > > 
> > > <plug>
> > > Switching mvpp2 to use phylink (which is needed for the 1G SFP slot on
> > > mcbin) will handle all this for you - dealing with both in-band and
> > > out-of-band negotiation methods, and combining them in the appropriate
> > > manner for the selected operation mode.
> > > </plug>
> > 
> > So probably the best move here is to remove this patch, and wait for the
> > phylink support in the PPv2 driver.
> 
> I've nothing on that specifically for the mvpp2 driver - what I have is
> for mvneta and the Marvell mvpp2x driver, with GMAC support extracted
> from mvneta (that last bit is rather dirty at the moment so not
> published anywhere, and doesn't cater for PP v2.1 at all.)
> 
> I ought to have posted the mvneta part of the phylink patches, but I
> didn't get around to it early enough in this cycle - there are probably
> quite a number of conflicts with net-next now, so I think it's too late
> to submit it for mainline.
> 
> I know Andrew has already looked at them in my git tree as part of the
> review of phylink when that was merged - which should be adequate to
> give an example of how to implement it for the mainline PP v2 driver.

OK, good. When looking at the phylink support in the PPv2 driver we'll
look into what you did for mvneta (I saw it as well on your tree). In
the meantime I'll respin this series without this patch.

Thanks!
Antoine

-- 
Antoine Ténart, Free Electrons
Embedded Linux and Kernel engineering
http://free-electrons.com

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]