Skip to contentSkip to conversation

phy: atheros: ar8035: Fix clock output calculation

PATCH Superseded Fabio Estevam 4 participants 10 messages Jun 18, 2020

Subsystems (4)

v1 of 3 Superseded Latest is v3 →

lore
Share & tools
lore

Fabio Estevam

author reviewer
#0 lore raw

PATCH 1/4 phy: atheros: ar8035: Fix clock output calculation

The clock ouput frequency is calculated incorrectly for AR8035 due to wrong masking of priv->clk_25m_reg and priv->clk_25m_mask.

This same issue has been already fixed in the kernel by:

commit b1f4c209d84057b6d40b939b6e4404854271d797

Author: Oleksij Rempel <[email protected]>
Date:   Wed Apr 1 11:57:32 2020 +0200

net: phy: at803x: fix clock sink configuration on ATH8030 and ATH8035

The masks in priv->clk_25m_reg and priv->clk_25m_mask are one-bits-set for the values that comprise the fields, not zero-bits-set.

This patch fixes the clock frequency configuration for ATH8030 and ATH8035 Atheros PHYs by removing the erroneous "~".

To reproduce this bug, configure the PHY with the device tree binding "qca,clk-out-frequency" and remove the machine specific PHY fixups.

Fixes 2f664823a47021 ("net: phy: at803x: add device tree binding")

Apply the same fix in the U-Boot driver.

Tested on a i.MX6 Hummingboard.

---
 drivers/net/phy/atheros.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)
drivers/net/phy/atheros.c +3 -4
index 13f7275d17..f922fecd6b 100644
@@ -275,11 +275,10 @@ static int ar803x_of_init(struct phy_device *phydev)
 		 * Fixup for the AR8035 which only has two bits. The two
 		 * remaining bits map to the same frequencies.
 		 */
-		if (phydev->drv->uid == AR8035_PHY_ID) {
-			u16 clear = AR803x_CLK_25M_MASK & AR8035_CLK_25M_MASK;
 
-			priv->clk_25m_mask &= ~clear;
-			priv->clk_25m_reg &= ~clear;
+		if (phydev->drv->uid == AR8035_PHY_ID) {
+			priv->clk_25m_reg &= AR8035_CLK_25M_MASK;
+			priv->clk_25m_mask &= AR8035_CLK_25M_MASK;
 		}
 	}
 
Signature · 1 line
2.17.1
author reviewer
#1 lore raw

PATCH 2/4 net: fec: Allow the PHY node to be retrieved

As we move towards driver model, it is required to let the FEC driver know how to properly deal with an Ethernet PHY subnode in the device tree.

For example:

&fec {
pinctrl-names = "default";

 	pinctrl-0 = <&pinctrl_microsom_enet_ar8035>;
	phy-handle = <&phy>;

phy-mode = "rgmii-id"; phy-reset-duration = <2>; phy-reset-gpios = <&gpio4 15 GPIO_ACTIVE_LOW>; status = "okay";

	mdio {
		#address-cells = <1>;
		#size-cells = <0>;

		phy: ethernet-phy at 0 {
			reg = <0>;
			qca,clk-out-frequency = <125000000>;
		};
	};

};

Currently the PHY node pointer is incorrectly associated with the Ethernel controller instead of the PHY node itself.

This causes the PHY properties, such as "qca,clk-out-frequency" in the example above to not get parsed.

Fix this problem by populating the phy_of_node node.

---
 drivers/net/fec_mxc.c | 7 +++++--
 drivers/net/fec_mxc.h | 1 +
 2 files changed, 6 insertions(+), 2 deletions(-)
drivers/net/fec_mxc.c +5 -2
index 9ae2db033e..992180df86 100644
@@ -1294,7 +1294,7 @@ static const struct eth_ops fecmxc_ops = {
 	.read_rom_hwaddr	= fecmxc_read_rom_hwaddr,
 };
 
-static int device_get_phy_addr(struct udevice *dev)
+static int device_get_phy_addr(struct fec_priv *priv, struct udevice *dev)
 {
 	struct ofnode_phandle_args phandle_args;
 	int reg;
@@ -1305,6 +1305,8 @@ static int device_get_phy_addr(struct udevice *dev)
 		return -ENODEV;
 	}
 
+	priv->phy_of_node = phandle_args.node;
+
 	reg = ofnode_read_u32_default(phandle_args.node, "reg", 0);
 
 	return reg;
@@ -1315,7 +1317,7 @@ static int fec_phy_init(struct fec_priv *priv, struct udevice *dev)
 	struct phy_device *phydev;
 	int addr;
 
-	addr = device_get_phy_addr(dev);
+	addr = device_get_phy_addr(priv, dev);
 #ifdef CONFIG_FEC_MXC_PHYADDR
 	addr = CONFIG_FEC_MXC_PHYADDR;
 #endif
@@ -1325,6 +1327,7 @@ static int fec_phy_init(struct fec_priv *priv, struct udevice *dev)
 		return -ENODEV;
 
 	priv->phydev = phydev;
+	priv->phydev->node = priv->phy_of_node;
 	phy_config(phydev);
 
 	return 0;
drivers/net/fec_mxc.h +1 -0
index 0e8f08a51a..659d62646f 100644
@@ -250,6 +250,7 @@ struct fec_priv {
 	struct mii_dev *bus;
 #ifdef CONFIG_PHYLIB
 	struct phy_device *phydev;
+	ofnode phy_of_node;
 #else
 	int phy_id;
 	int (*mii_postcall)(int);
Signature · 1 line
2.17.1
#2 lore raw

PATCH 2/4 net: fec: Allow the PHY node to be retrieved

Quoted 28 lines · Fabio Estevam

As we move towards driver model, it is required to let the FEC driver know how to properly deal with an Ethernet PHY subnode in the device tree.

For example:

&fec {

(quoted diff) +0 -0
 	pinctrl-names = "default";
 	pinctrl-0 = <&pinctrl_microsom_enet_ar8035>;
	phy-handle = <&phy>;
(quoted diff) +0 -0
 	phy-mode = "rgmii-id";
 	phy-reset-duration = <2>;
 	phy-reset-gpios = <&gpio4 15 GPIO_ACTIVE_LOW>;
 	status = "okay";
	mdio {
		#address-cells = <1>;
		#size-cells = <0>;

		phy: ethernet-phy at 0 {
			reg = <0>;
			qca,clk-out-frequency = <125000000>;
		};
	};

};

Currently the PHY node pointer is incorrectly associated with the Ethernel controller instead of the PHY node itself.

Quoted 7 lines

This causes the PHY properties, such as "qca,clk-out-frequency" in the example above to not get parsed.

Fix this problem by populating the phy_of_node node.

Signature · 8 lines
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 659 bytes
Desc: not available
URL: <https://lists.denx.de/pipermail/u-boot/attachments/20200618/f0c80191/attachment.sig>
author reviewer
#3 lore raw

PATCH 3/4 ARM: dts: imx6qdl-sr-som: Sync with kernel 5.8-rc1

In reply to Fabio Estevam

Sync the device tree with 5.8-rc1.

It basically contains the following extra kernel commit:

commit 86b08bd5b99480b79a25343f24c1b8c4ddcb5c09

Author: Russell King <[email protected]>
Date:   Wed Apr 15 16:44:17 2020 +0100

ARM: dts: imx6-sr-som: add ethernet PHY configuration

Add ethernet PHY configuration ahead of removing the quirk that configures the clocking mode for the PHY. The RGMII delay is already set correctly.

, which passes the 'qca,clk-out-frequency' property and it is important to specify the correct frequency generated by the AR8031.

---
 arch/arm/dts/imx6qdl-sr-som.dtsi | 11 +++++++++++
 1 file changed, 11 insertions(+)
arch/arm/dts/imx6qdl-sr-som.dtsi +11 -0
index 6d7f6b9035..b06577808f 100644
@@ -53,10 +53,21 @@
 &fec {
 	pinctrl-names = "default";
 	pinctrl-0 = <&pinctrl_microsom_enet_ar8035>;
+	phy-handle = <&phy>;
 	phy-mode = "rgmii-id";
 	phy-reset-duration = <2>;
 	phy-reset-gpios = <&gpio4 15 GPIO_ACTIVE_LOW>;
 	status = "okay";
+
+	mdio {
+		#address-cells = <1>;
+		#size-cells = <0>;
+
+		phy: ethernet-phy at 0 {
+			reg = <0>;
+			qca,clk-out-frequency = <125000000>;
+		};
+	};
 };
 
 &iomuxc {
Signature · 1 line
2.17.1
#4 lore raw

PATCH 3/4 ARM: dts: imx6qdl-sr-som: Sync with kernel 5.8-rc1

Quoted 14 lines · Fabio Estevam

Sync the device tree with 5.8-rc1.

It basically contains the following extra kernel commit:

commit 86b08bd5b99480b79a25343f24c1b8c4ddcb5c09

Author: Russell King <[email protected]>
Date:   Wed Apr 15 16:44:17 2020 +0100

ARM: dts: imx6-sr-som: add ethernet PHY configuration

Add ethernet PHY configuration ahead of removing the quirk that configures the clocking mode for the PHY. The RGMII delay is already set correctly.

Quoted 8 lines

, which passes the 'qca,clk-out-frequency' property and it is important to specify the correct frequency generated by the AR8031.

Signature · 8 lines
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 659 bytes
Desc: not available
URL: <https://lists.denx.de/pipermail/u-boot/attachments/20200618/0541d562/attachment.sig>
author reviewer
#5 lore raw

PATCH 4/4 mx6cuboxi: Convert to DM_ETH

In reply to Fabio Estevam

Migration to DM_ETH is mandatory, so convert mx6cuboxi to Ethernet Driver Model.

This also brings the benefit of restoring Ethernet functionality.

---
 board/solidrun/mx6cuboxi/mx6cuboxi.c | 123 ++++-----------------------
 configs/mx6cuboxi_defconfig          |   4 +
 include/configs/mx6cuboxi.h          |   6 --
 3 files changed, 21 insertions(+), 112 deletions(-)
board/solidrun/mx6cuboxi/mx6cuboxi.c +17 -106
index 94707bccb2..225fea2cb2 100644
@@ -17,7 +17,6 @@
 #include <image.h>
 #include <init.h>
 #include <log.h>
-#include <net.h>
 #include <asm/arch/clock.h>
 #include <asm/arch/imx-regs.h>
 #include <asm/arch/iomux.h>
@@ -34,7 +33,6 @@
 #include <fsl_esdhc_imx.h>
 #include <malloc.h>
 #include <miiphy.h>
-#include <netdev.h>
 #include <asm/arch/crm_regs.h>
 #include <asm/io.h>
 #include <asm/arch/sys_proto.h>
@@ -237,110 +235,6 @@ int board_mmc_init(bd_t *bis)
 	return 0;
 }
 
-static iomux_v3_cfg_t const enet_pads[] = {
-	IOMUX_PADS(PAD_ENET_MDIO__ENET_MDIO | MUX_PAD_CTRL(ENET_PAD_CTRL)),
-	IOMUX_PADS(PAD_ENET_MDC__ENET_MDC | MUX_PAD_CTRL(ENET_PAD_CTRL)),
-	/* AR8035 reset */
-	IOMUX_PADS(PAD_KEY_ROW4__GPIO4_IO15 | MUX_PAD_CTRL(ENET_PAD_CTRL_PD)),
-	/* AR8035 interrupt */
-	IOMUX_PADS(PAD_DI0_PIN2__GPIO4_IO18 | MUX_PAD_CTRL(NO_PAD_CTRL)),
-	/* GPIO16 -> AR8035 25MHz */
-	IOMUX_PADS(PAD_GPIO_16__ENET_REF_CLK	  | MUX_PAD_CTRL(NO_PAD_CTRL)),
-	IOMUX_PADS(PAD_RGMII_TXC__RGMII_TXC	  | MUX_PAD_CTRL(NO_PAD_CTRL)),
-	IOMUX_PADS(PAD_RGMII_TD0__RGMII_TD0 | MUX_PAD_CTRL(ENET_PAD_CTRL)),
-	IOMUX_PADS(PAD_RGMII_TD1__RGMII_TD1 | MUX_PAD_CTRL(ENET_PAD_CTRL)),
-	IOMUX_PADS(PAD_RGMII_TD2__RGMII_TD2 | MUX_PAD_CTRL(ENET_PAD_CTRL)),
-	IOMUX_PADS(PAD_RGMII_TD3__RGMII_TD3 | MUX_PAD_CTRL(ENET_PAD_CTRL)),
-	IOMUX_PADS(PAD_RGMII_TX_CTL__RGMII_TX_CTL | MUX_PAD_CTRL(ENET_PAD_CTRL)),
-	/* AR8035 CLK_25M --> ENET_REF_CLK (V22) */
-	IOMUX_PADS(PAD_ENET_REF_CLK__ENET_TX_CLK | MUX_PAD_CTRL(ENET_PAD_CTRL_CLK)),
-	IOMUX_PADS(PAD_RGMII_RXC__RGMII_RXC | MUX_PAD_CTRL(ENET_PAD_CTRL)),
-	IOMUX_PADS(PAD_RGMII_RD0__RGMII_RD0 | MUX_PAD_CTRL(ENET_PAD_CTRL_PD)),
-	IOMUX_PADS(PAD_RGMII_RD1__RGMII_RD1 | MUX_PAD_CTRL(ENET_PAD_CTRL_PD)),
-	IOMUX_PADS(PAD_RGMII_RD2__RGMII_RD2 | MUX_PAD_CTRL(ENET_PAD_CTRL)),
-	IOMUX_PADS(PAD_RGMII_RD3__RGMII_RD3 | MUX_PAD_CTRL(ENET_PAD_CTRL)),
-	IOMUX_PADS(PAD_RGMII_RX_CTL__RGMII_RX_CTL | MUX_PAD_CTRL(ENET_PAD_CTRL_PD)),
-	IOMUX_PADS(PAD_ENET_RXD0__GPIO1_IO27 | MUX_PAD_CTRL(ENET_PAD_CTRL_PD)),
-	IOMUX_PADS(PAD_ENET_RXD1__GPIO1_IO26 | MUX_PAD_CTRL(ENET_PAD_CTRL_PD)),
-};
-
-static void setup_iomux_enet(void)
-{
-	struct gpio_desc desc;
-	int ret;
-
-	SETUP_IOMUX_PADS(enet_pads);
-
-	ret = dm_gpio_lookup_name("GPIO4_15", &desc);
-	if (ret) {
-		printf("%s: phy reset lookup failed\n", __func__);
-		return;
-	}
-
-	ret = dm_gpio_request(&desc, "phy-reset");
-	if (ret) {
-		printf("%s: phy reset request failed\n", __func__);
-		return;
-	}
-
-	gpio_direction_output(ETH_PHY_RESET, 0);
-	mdelay(10);
-	gpio_set_value(ETH_PHY_RESET, 1);
-	udelay(100);
-
-	gpio_free_list_nodev(&desc, 1);
-}
-
-int board_phy_config(struct phy_device *phydev)
-{
-	if (phydev->drv->config)
-		phydev->drv->config(phydev);
-
-	return 0;
-}
-
-/* On Cuboxi Ethernet PHY can be located at addresses 0x0 or 0x4 */
-#define ETH_PHY_MASK	((1 << 0x0) | (1 << 0x4))
-
-int board_eth_init(bd_t *bis)
-{
-	struct iomuxc *const iomuxc_regs = (struct iomuxc *)IOMUXC_BASE_ADDR;
-	struct mii_dev *bus;
-	struct phy_device *phydev;
-
-	int ret = enable_fec_anatop_clock(0, ENET_25MHZ);
-	if (ret)
-		return ret;
-
-	/* set gpr1[ENET_CLK_SEL] */
-	setbits_le32(&iomuxc_regs->gpr[1], IOMUXC_GPR1_ENET_CLK_SEL_MASK);
-
-	setup_iomux_enet();
-
-	bus = fec_get_miibus(IMX_FEC_BASE, -1);
-	if (!bus)
-		return -EINVAL;
-
-	phydev = phy_find_by_mask(bus, ETH_PHY_MASK, PHY_INTERFACE_MODE_RGMII);
-	if (!phydev) {
-		ret = -EINVAL;
-		goto free_bus;
-	}
-
-	debug("using phy at address %d\n", phydev->addr);
-	ret = fec_probe(bis, -1, IMX_FEC_BASE, bus, phydev);
-	if (ret)
-		goto free_phydev;
-
-	return 0;
-
-free_phydev:
-	free(phydev);
-free_bus:
-	free(bus);
-	return ret;
-}
-
 #ifdef CONFIG_VIDEO_IPUV3
 static void do_enable_hdmi(struct display_info_t const *dev)
 {
@@ -433,6 +327,21 @@ static int setup_display(void)
 }
 #endif /* CONFIG_VIDEO_IPUV3 */
 
+static int setup_fec(void)
+{
+	struct iomuxc *const iomuxc_regs = (struct iomuxc *)IOMUXC_BASE_ADDR;
+	int ret;
+
+	ret = enable_fec_anatop_clock(0, ENET_25MHZ);
+	if (ret)
+		return ret;
+
+	/* set gpr1[ENET_CLK_SEL] */
+	setbits_le32(&iomuxc_regs->gpr[1], IOMUXC_GPR1_ENET_CLK_SEL_MASK);
+
+	return 0;
+}
+
 int board_early_init_f(void)
 {
 	setup_iomux_uart();
@@ -440,6 +349,8 @@ int board_early_init_f(void)
 #ifdef CONFIG_CMD_SATA
 	setup_sata();
 #endif
+	setup_fec();
+
 	return 0;
 }
 
configs/mx6cuboxi_defconfig +4 -0
index df7e4611a0..3f43fb7ad0 100644
@@ -50,6 +50,10 @@ CONFIG_DM_MMC=y
 CONFIG_FSL_USDHC=y
 CONFIG_PHYLIB=y
 CONFIG_PHY_ATHEROS=y
+CONFIG_DM_ETH=y
+CONFIG_DM_MDIO=y
+CONFIG_FEC_MXC=y
+CONFIG_RGMII=y
 CONFIG_MII=y
 CONFIG_PINCTRL=y
 CONFIG_PINCTRL_IMX6=y
include/configs/mx6cuboxi.h +0 -6
index 2ccf44e573..96f79e6b58 100644
@@ -29,12 +29,6 @@
 #define CONFIG_LBA48
 #endif
 
-/* Ethernet Configuration */
-#define CONFIG_FEC_MXC
-#define IMX_FEC_BASE			ENET_BASE_ADDR
-#define CONFIG_FEC_XCV_TYPE		RGMII
-#define CONFIG_FEC_MXC_PHYADDR		0
-
 /* Framebuffer */
 #define CONFIG_VIDEO_BMP_RLE8
 #define CONFIG_SPLASH_SCREEN
Signature · 1 line
2.17.1
#6 lore raw

PATCH 4/4 mx6cuboxi: Convert to DM_ETH

Hi Fabio,

Thanks for your debug and fix. Small nits below.

Quoted 26 lines · Fabio Estevam

Migration to DM_ETH is mandatory, so convert mx6cuboxi to Ethernet Driver Model.

This also brings the benefit of restoring Ethernet functionality.

---
 board/solidrun/mx6cuboxi/mx6cuboxi.c | 123 ++++-----------------------
 configs/mx6cuboxi_defconfig          |   4 +
 include/configs/mx6cuboxi.h          |   6 --
 3 files changed, 21 insertions(+), 112 deletions(-)
board/solidrun/mx6cuboxi/mx6cuboxi.c +0 -1
index 94707bccb2..225fea2cb2 100644
@@ -17,7 +17,6 @@
 #include <image.h>
 #include <init.h>
 #include <log.h>
-#include <net.h>
 #include <asm/arch/clock.h>
 #include <asm/arch/imx-regs.h>
 #include <asm/arch/iomux.h>

board/solidrun/mx6cuboxi/mx6cuboxi.c line 35

Quoted 3 lines
board/solidrun/mx6cuboxi/mx6cuboxi.c +0 -0
@@ -34,7 +33,6 @@
 #include <fsl_esdhc_imx.h>
 #include <malloc.h>
 #include <miiphy.h>

Do we still need this header?

board/solidrun/mx6cuboxi/mx6cuboxi.c int board_mmc_init(bd_t *bis) · line 241

Quoted 10 lines
board/solidrun/mx6cuboxi/mx6cuboxi.c +0 -3
-#include <netdev.h>
 #include <asm/arch/crm_regs.h>
 #include <asm/io.h>
 #include <asm/arch/sys_proto.h>
@@ -237,110 +235,6 @@ int board_mmc_init(bd_t *bis)
 	return 0;
 }
 
-static iomux_v3_cfg_t const enet_pads[] = {
-	IOMUX_PADS(PAD_ENET_MDIO__ENET_MDIO | MUX_PAD_CTRL(ENET_PAD_CTRL)),

The definitions of these ENET_PAD_CTRL* macros can go away as well.

baruch

Quoted 5 lines
board/solidrun/mx6cuboxi/mx6cuboxi.c +0 -5
-	IOMUX_PADS(PAD_ENET_MDC__ENET_MDC | MUX_PAD_CTRL(ENET_PAD_CTRL)),
-	/* AR8035 reset */
-	IOMUX_PADS(PAD_KEY_ROW4__GPIO4_IO15 | MUX_PAD_CTRL(ENET_PAD_CTRL_PD)),
-	/* AR8035 interrupt */
-	IOMUX_PADS(PAD_DI0_PIN2__GPIO4_IO18 | MUX_PAD_CTRL(NO_PAD_CTRL)),

...

Signature · 3 lines
                                                     ~. .~   Tk Open Systems
=}------------------------------------------------ooO--U--Ooo------------{=
   - baruch at tkos.co.il - tel: +972.52.368.4656, http://www.tkos.co.il -
#7 lore raw

PATCH 4/4 mx6cuboxi: Convert to DM_ETH

In reply to Fabio Estevam
Quoted 7 lines · Fabio Estevam

Migration to DM_ETH is mandatory, so convert mx6cuboxi to Ethernet Driver Model.

This also brings the benefit of restoring Ethernet functionality.

Signature · 8 lines
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 659 bytes
Desc: not available
URL: <https://lists.denx.de/pipermail/u-boot/attachments/20200618/1456212f/attachment.sig>
#8 lore raw

PATCH 4/4 mx6cuboxi: Convert to DM_ETH

In reply to Fabio Estevam

Hi Fabio,

Quoted 181 lines · Fabio Estevam

Migration to DM_ETH is mandatory, so convert mx6cuboxi to Ethernet Driver Model.

This also brings the benefit of restoring Ethernet functionality.

---
 board/solidrun/mx6cuboxi/mx6cuboxi.c | 123 ++++-----------------------
 configs/mx6cuboxi_defconfig          |   4 +
 include/configs/mx6cuboxi.h          |   6 --
 3 files changed, 21 insertions(+), 112 deletions(-)
board/solidrun/mx6cuboxi/mx6cuboxi.c +17 -106
index 94707bccb2..225fea2cb2 100644
@@ -17,7 +17,6 @@
 #include <image.h>
 #include <init.h>
 #include <log.h>
-#include <net.h>
 #include <asm/arch/clock.h>
 #include <asm/arch/imx-regs.h>
 #include <asm/arch/iomux.h>
@@ -34,7 +33,6 @@
 #include <fsl_esdhc_imx.h>
 #include <malloc.h>
 #include <miiphy.h>
-#include <netdev.h>
 #include <asm/arch/crm_regs.h>
 #include <asm/io.h>
 #include <asm/arch/sys_proto.h>
@@ -237,110 +235,6 @@ int board_mmc_init(bd_t *bis)
        return 0;
 }
 
-static iomux_v3_cfg_t const enet_pads[] = {
-       IOMUX_PADS(PAD_ENET_MDIO__ENET_MDIO | MUX_PAD_CTRL(ENET_PAD_CTRL)),
-       IOMUX_PADS(PAD_ENET_MDC__ENET_MDC | MUX_PAD_CTRL(ENET_PAD_CTRL)),
-       /* AR8035 reset */
-       IOMUX_PADS(PAD_KEY_ROW4__GPIO4_IO15 | MUX_PAD_CTRL(ENET_PAD_CTRL_PD)),
-       /* AR8035 interrupt */
-       IOMUX_PADS(PAD_DI0_PIN2__GPIO4_IO18 | MUX_PAD_CTRL(NO_PAD_CTRL)),
-       /* GPIO16 -> AR8035 25MHz */
-       IOMUX_PADS(PAD_GPIO_16__ENET_REF_CLK      | MUX_PAD_CTRL(NO_PAD_CTRL)),
-       IOMUX_PADS(PAD_RGMII_TXC__RGMII_TXC       | MUX_PAD_CTRL(NO_PAD_CTRL)),
-       IOMUX_PADS(PAD_RGMII_TD0__RGMII_TD0 | MUX_PAD_CTRL(ENET_PAD_CTRL)),
-       IOMUX_PADS(PAD_RGMII_TD1__RGMII_TD1 | MUX_PAD_CTRL(ENET_PAD_CTRL)),
-       IOMUX_PADS(PAD_RGMII_TD2__RGMII_TD2 | MUX_PAD_CTRL(ENET_PAD_CTRL)),
-       IOMUX_PADS(PAD_RGMII_TD3__RGMII_TD3 | MUX_PAD_CTRL(ENET_PAD_CTRL)),
-       IOMUX_PADS(PAD_RGMII_TX_CTL__RGMII_TX_CTL | MUX_PAD_CTRL(ENET_PAD_CTRL)),
-       /* AR8035 CLK_25M --> ENET_REF_CLK (V22) */
-       IOMUX_PADS(PAD_ENET_REF_CLK__ENET_TX_CLK | MUX_PAD_CTRL(ENET_PAD_CTRL_CLK)),
-       IOMUX_PADS(PAD_RGMII_RXC__RGMII_RXC | MUX_PAD_CTRL(ENET_PAD_CTRL)),
-       IOMUX_PADS(PAD_RGMII_RD0__RGMII_RD0 | MUX_PAD_CTRL(ENET_PAD_CTRL_PD)),
-       IOMUX_PADS(PAD_RGMII_RD1__RGMII_RD1 | MUX_PAD_CTRL(ENET_PAD_CTRL_PD)),
-       IOMUX_PADS(PAD_RGMII_RD2__RGMII_RD2 | MUX_PAD_CTRL(ENET_PAD_CTRL)),
-       IOMUX_PADS(PAD_RGMII_RD3__RGMII_RD3 | MUX_PAD_CTRL(ENET_PAD_CTRL)),
-       IOMUX_PADS(PAD_RGMII_RX_CTL__RGMII_RX_CTL | MUX_PAD_CTRL(ENET_PAD_CTRL_PD)),
-       IOMUX_PADS(PAD_ENET_RXD0__GPIO1_IO27 | MUX_PAD_CTRL(ENET_PAD_CTRL_PD)),
-       IOMUX_PADS(PAD_ENET_RXD1__GPIO1_IO26 | MUX_PAD_CTRL(ENET_PAD_CTRL_PD)),
-};
-
-static void setup_iomux_enet(void)
-{
-       struct gpio_desc desc;
-       int ret;
-
-       SETUP_IOMUX_PADS(enet_pads);
-
-       ret = dm_gpio_lookup_name("GPIO4_15", &desc);
-       if (ret) {
-               printf("%s: phy reset lookup failed\n", __func__);
-               return;
-       }
-
-       ret = dm_gpio_request(&desc, "phy-reset");
-       if (ret) {
-               printf("%s: phy reset request failed\n", __func__);
-               return;
-       }
-
-       gpio_direction_output(ETH_PHY_RESET, 0);
-       mdelay(10);
-       gpio_set_value(ETH_PHY_RESET, 1);
-       udelay(100);
-
-       gpio_free_list_nodev(&desc, 1);
-}
-
-int board_phy_config(struct phy_device *phydev)
-{
-       if (phydev->drv->config)
-               phydev->drv->config(phydev);
-
-       return 0;
-}
-
-/* On Cuboxi Ethernet PHY can be located at addresses 0x0 or 0x4 */
-#define ETH_PHY_MASK   ((1 << 0x0) | (1 << 0x4))
-
-int board_eth_init(bd_t *bis)
-{
-       struct iomuxc *const iomuxc_regs = (struct iomuxc *)IOMUXC_BASE_ADDR;
-       struct mii_dev *bus;
-       struct phy_device *phydev;
-
-       int ret = enable_fec_anatop_clock(0, ENET_25MHZ);
-       if (ret)
-               return ret;
-
-       /* set gpr1[ENET_CLK_SEL] */
-       setbits_le32(&iomuxc_regs->gpr[1], IOMUXC_GPR1_ENET_CLK_SEL_MASK);
-
-       setup_iomux_enet();
-
-       bus = fec_get_miibus(IMX_FEC_BASE, -1);
-       if (!bus)
-               return -EINVAL;
-
-       phydev = phy_find_by_mask(bus, ETH_PHY_MASK, PHY_INTERFACE_MODE_RGMII);
-       if (!phydev) {
-               ret = -EINVAL;
-               goto free_bus;
-       }
-
-       debug("using phy at address %d\n", phydev->addr);
-       ret = fec_probe(bis, -1, IMX_FEC_BASE, bus, phydev);
-       if (ret)
-               goto free_phydev;
-
-       return 0;
-
-free_phydev:
-       free(phydev);
-free_bus:
-       free(bus);
-       return ret;
-}
-
 #ifdef CONFIG_VIDEO_IPUV3
 static void do_enable_hdmi(struct display_info_t const *dev)
 {
@@ -433,6 +327,21 @@ static int setup_display(void)
 }
 #endif /* CONFIG_VIDEO_IPUV3 */
 
+static int setup_fec(void)
+{
+       struct iomuxc *const iomuxc_regs = (struct iomuxc *)IOMUXC_BASE_ADDR;
+       int ret;
+
+       ret = enable_fec_anatop_clock(0, ENET_25MHZ);
+       if (ret)
+               return ret;
+
+       /* set gpr1[ENET_CLK_SEL] */
+       setbits_le32(&iomuxc_regs->gpr[1], IOMUXC_GPR1_ENET_CLK_SEL_MASK);
+
+       return 0;
+}
+
 int board_early_init_f(void)
 {
        setup_iomux_uart();
@@ -440,6 +349,8 @@ int board_early_init_f(void)
 #ifdef CONFIG_CMD_SATA
        setup_sata();
 #endif
+       setup_fec();
+
        return 0;
 }
 
configs/mx6cuboxi_defconfig +0 -0
index df7e4611a0..3f43fb7ad0 100644

configs/mx6cuboxi_defconfig CONFIG_DM_MMC=y · line 54

Quoted 5 lines
configs/mx6cuboxi_defconfig +2 -0
@@ -50,6 +50,10 @@ CONFIG_DM_MMC=y
 CONFIG_FSL_USDHC=y
 CONFIG_PHYLIB=y
 CONFIG_PHY_ATHEROS=y
+CONFIG_DM_ETH=y
+CONFIG_DM_MDIO=y

I wasn't going to nitpick, but since Baruch started giving some more relevant feedback, I figured I could as well:
Does this board make any use of DM_MDIO? Doesn't look like it.

Quoted 23 lines
configs/mx6cuboxi_defconfig +2 -0
+CONFIG_FEC_MXC=y
+CONFIG_RGMII=y
 CONFIG_MII=y
 CONFIG_PINCTRL=y
 CONFIG_PINCTRL_IMX6=y
include/configs/mx6cuboxi.h +0 -7
index 2ccf44e573..96f79e6b58 100644
@@ -29,12 +29,6 @@
 #define CONFIG_LBA48
 #endif
 
-/* Ethernet Configuration */
-#define CONFIG_FEC_MXC
-#define IMX_FEC_BASE                   ENET_BASE_ADDR
-#define CONFIG_FEC_XCV_TYPE            RGMII
-#define CONFIG_FEC_MXC_PHYADDR         0
-
 /* Framebuffer */
 #define CONFIG_VIDEO_BMP_RLE8
 #define CONFIG_SPLASH_SCREEN
--
Quoted 2 lines

2.17.1

Thanks,
-Vladimir

#9 lore raw

PATCH 1/4 phy: atheros: ar8035: Fix clock output calculation

In reply to Fabio Estevam
Quoted 27 lines · Fabio Estevam

The clock ouput frequency is calculated incorrectly for AR8035 due to wrong masking of priv->clk_25m_reg and priv->clk_25m_mask.

This same issue has been already fixed in the kernel by:

commit b1f4c209d84057b6d40b939b6e4404854271d797

Author: Oleksij Rempel <[email protected]>
Date:   Wed Apr 1 11:57:32 2020 +0200

net: phy: at803x: fix clock sink configuration on ATH8030 and ATH8035

The masks in priv->clk_25m_reg and priv->clk_25m_mask are one-bits-set for the values that comprise the fields, not zero-bits-set.

This patch fixes the clock frequency configuration for ATH8030 and ATH8035 Atheros PHYs by removing the erroneous "~".

To reproduce this bug, configure the PHY with the device tree binding "qca,clk-out-frequency" and remove the machine specific PHY fixups.

Fixes 2f664823a47021 ("net: phy: at803x: add device tree binding")
Quoted 6 lines

Apply the same fix in the U-Boot driver.

Tested on a i.MX6 Hummingboard.

Signature · 8 lines
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 659 bytes
Desc: not available
URL: <https://lists.denx.de/pipermail/u-boot/attachments/20200618/afa99e64/attachment.sig>