[PATCH v5 0/6] phy: realtek: usb2: support for RTL9607C USB2 PHY

[PATCH v5 0/6] phy: realtek: usb2: support for RTL9607C USB2 PHY

[Rustam Adilov]

This patch series for Realtek USB2 PHY driver adds support for RTL9607C
USB2 PHY.

RTL9607C is a big endian MIPS CPU which is quite far from RTD series SoCs
supported by realtek usb2 phy driver, but the phy initilization is found
to be very indentical in most areas.

Most of the code was based on the Realtek's usb driver from the GPL tarball
in [1] and adjusted to fit into the realtek usb2 phy driver code format.

The patch series was split into smaller patches that add/change something
in the driver that are not exactly related to RTL9607C and that also
helps for easier review. That also means, patch 5 depends on all the prior
patches that come before it.

USB2 PHY on RTL9607C is primarly used for its internal OHCI/EHCI controllers.

[1] - https://github.com/jameywine/GPL-for-GP3000/blob/main/linux-5.10.x/arch/mips/rtl9607c/usb.c

---
Changelog in v5:
Mostly addressing LLM review
- Patch 1
 - changed int to u32 type for new_reg_req and vstatus_busy data fields.
 - changed comments in rtk_phy_read/write from PHY_NEW_REG_REQ to phy_reg->new_reg_req.
- Patch 2
 - explained readl/writel native endianess issue in more detail.
 - explained why vstatus register doesn't need byte swapping.
- Patch 4
 - moved reset_control_deassert to rtk_phy_init function to keep it outside of for loop.
 - changed msleep(5) to usleep_range(5000, 6000).
 - explained why reset_control_assert is not needed.
- Patch 5
 - explained readl/writel native endianess issue here as well.
 - explained why FORCE_DISCONNECT_REG doesn't need byte swapping.
- Link to v4: https://lore.kernel.org/linux-phy/20260406181228.25892-1-adilov@disroot.org/

Changelog in v4:
- Patch 2
 - moved the le variations of read/write functions to Patch 5 where it is actually used because
   otherwise, it results in unused errors when only Patch 2 is applied.
 - updated the commit message to to point the reason for le32 wrappers around readl/writel.
- Patch 3
 - added "Reviewed by Krzysztof Kozlowski"
- Patch 5
 - updated the commit message to include the addition of little endian read/write functions from
   Patch 2.
- Link to v3: https://lore.kernel.org/linux-phy/20260402154414.196012-1-adilov@disroot.org/

Changelog in v3:
- Patch 2
 - renamed phy read and functions to "rtk_usb2phy" to not collide with networking API functions
 - fixed the sparse warnings by creating intermidiate "tmp" variable and then pass it to writel
 - sligtly adjusted commit message to instead use "default read" not "default phy_read"
- Patch 4
 - added the check for reset_control_deassert() just in case
 - changed mdelay(5) to msleep(5)
 - changed dev_err and return combo with one dev_err_probe for phy_rst
- Patch 5
 - changed mdelay(10) under force_host_disconnect to msleep(10)
 - removed struct fields with false like force_host_disconnect and more in rtl9607_phy_cfg
- Patch 6
 - updated the #endif commend to now include MACH_REALTEK_RTL to reflect if on top
- Link to v2: https://lore.kernel.org/linux-phy/20260327160638.15134-1-adilov@disroot.org/

Changelog in v2:
- Patch 3
 - removed the line about OHCI/EHCI controllers from description.
 - set the resets to false for RTD SoC devices and changed the
   commit message to reflect that.
- Link to v1: https://lore.kernel.org/linux-phy/20260326193419.48419-1-adilov@disroot.org/

Rustam Adilov (6):
  phy: realtek: usb2: introduce vstatus/new_reg_req variables to driver
    data
  phy: realtek: usb2: introduce read and write functions to driver data
  dt-bindings: phy: realtek,usb2phy.yaml: extend for resets and RTL9607C
    support
  phy: realtek: usb2: introduce reset controller struct
  phy: realtek: usb2: add support for RTL9607C USB2 PHY
  phy: realtek: Make configs available for MACH_REALTEK_RTL

 .../bindings/phy/realtek,usb2phy.yaml         |  25 ++-
 drivers/phy/realtek/Kconfig                   |   4 +-
 drivers/phy/realtek/phy-rtk-usb2.c            | 191 ++++++++++++++++--
 3 files changed, 195 insertions(+), 25 deletions(-)

-- 
2.53.0

[PATCH v5 1/6] phy: realtek: usb2: introduce vstatus/new_reg_req variables to driver data

[Rustam Adilov]

In RTL9607C SoC, the vstatus register is located at a certain offset from
the base and so introduce the vstatus_offset to handle it.

Busy bit of the vstatus and new_reg_req bit are also different and so
introduce these variables to the driver data as well.

Add these variables to the pre-existing phy cfg structs for RTD SoCs and
assign them the default values.

Co-developed-by: Michael Zavertkin <misha.zavertkin@mail.ru>
Signed-off-by: Michael Zavertkin <misha.zavertkin@mail.ru>
Signed-off-by: Rustam Adilov <adilov@disroot.org>
---
 drivers/phy/realtek/phy-rtk-usb2.c | 67 +++++++++++++++++++++++-------
 1 file changed, 52 insertions(+), 15 deletions(-)

diff --git a/drivers/phy/realtek/phy-rtk-usb2.c b/drivers/phy/realtek/phy-rtk-usb2.c
index 248550ef98ca..f5f943518d4f 100644
--- a/drivers/phy/realtek/phy-rtk-usb2.c
+++ b/drivers/phy/realtek/phy-rtk-usb2.c
@@ -64,6 +64,9 @@ struct phy_reg {
 	void __iomem *reg_wrap_vstatus;
 	void __iomem *reg_gusb2phyacc0;
 	int vstatus_index;
+	int vstatus_offset;
+	u32 vstatus_busy;
+	u32 new_reg_req;
 };
 
 struct phy_data {
@@ -96,6 +99,9 @@ struct phy_cfg {
 	bool do_toggle_driving;
 	bool use_default_parameter;
 	bool is_double_sensitivity_mode;
+	int vstatus_offset;
+	u32 vstatus_busy;
+	u32 new_reg_req;
 };
 
 struct phy_parameter {
@@ -162,21 +168,21 @@ static char rtk_phy_read(struct phy_reg *phy_reg, char addr)
 	addr -= OFFEST_PHY_READ;
 
 	/* polling until VBusy == 0 */
-	ret = utmi_wait_register(reg_gusb2phyacc0, PHY_VSTS_BUSY, 0);
+	ret = utmi_wait_register(reg_gusb2phyacc0, phy_reg->vstatus_busy, 0);
 	if (ret)
 		return (char)ret;
 
-	/* VCtrl = low nibble of addr, and set PHY_NEW_REG_REQ */
-	val = PHY_NEW_REG_REQ | (GET_LOW_NIBBLE(addr) << PHY_VCTRL_SHIFT);
+	/* VCtrl = low nibble of addr, and set phy_reg->new_reg_req */
+	val = phy_reg->new_reg_req | (GET_LOW_NIBBLE(addr) << PHY_VCTRL_SHIFT);
 	writel(val, reg_gusb2phyacc0);
-	ret = utmi_wait_register(reg_gusb2phyacc0, PHY_VSTS_BUSY, 0);
+	ret = utmi_wait_register(reg_gusb2phyacc0, phy_reg->vstatus_busy, 0);
 	if (ret)
 		return (char)ret;
 
-	/* VCtrl = high nibble of addr, and set PHY_NEW_REG_REQ */
-	val = PHY_NEW_REG_REQ | (GET_HIGH_NIBBLE(addr) << PHY_VCTRL_SHIFT);
+	/* VCtrl = high nibble of addr, and set phy_reg->new_reg_req */
+	val = phy_reg->new_reg_req | (GET_HIGH_NIBBLE(addr) << PHY_VCTRL_SHIFT);
 	writel(val, reg_gusb2phyacc0);
-	ret = utmi_wait_register(reg_gusb2phyacc0, PHY_VSTS_BUSY, 0);
+	ret = utmi_wait_register(reg_gusb2phyacc0, phy_reg->vstatus_busy, 0);
 	if (ret)
 		return (char)ret;
 
@@ -194,25 +200,25 @@ static int rtk_phy_write(struct phy_reg *phy_reg, char addr, char data)
 	int ret = 0;
 
 	/* write data to VStatusOut2 (data output to phy) */
-	writel((u32)data << shift_bits, reg_wrap_vstatus);
+	writel((u32)data << shift_bits, reg_wrap_vstatus + phy_reg->vstatus_offset);
 
-	ret = utmi_wait_register(reg_gusb2phyacc0, PHY_VSTS_BUSY, 0);
+	ret = utmi_wait_register(reg_gusb2phyacc0, phy_reg->vstatus_busy, 0);
 	if (ret)
 		return ret;
 
-	/* VCtrl = low nibble of addr, set PHY_NEW_REG_REQ */
-	val = PHY_NEW_REG_REQ | (GET_LOW_NIBBLE(addr) << PHY_VCTRL_SHIFT);
+	/* VCtrl = low nibble of addr, set phy_reg->new_reg_req */
+	val = phy_reg->new_reg_req | (GET_LOW_NIBBLE(addr) << PHY_VCTRL_SHIFT);
 
 	writel(val, reg_gusb2phyacc0);
-	ret = utmi_wait_register(reg_gusb2phyacc0, PHY_VSTS_BUSY, 0);
+	ret = utmi_wait_register(reg_gusb2phyacc0, phy_reg->vstatus_busy, 0);
 	if (ret)
 		return ret;
 
-	/* VCtrl = high nibble of addr, set PHY_NEW_REG_REQ */
-	val = PHY_NEW_REG_REQ | (GET_HIGH_NIBBLE(addr) << PHY_VCTRL_SHIFT);
+	/* VCtrl = high nibble of addr, set phy_reg->new_reg_req */
+	val = phy_reg->new_reg_req | (GET_HIGH_NIBBLE(addr) << PHY_VCTRL_SHIFT);
 
 	writel(val, reg_gusb2phyacc0);
-	ret = utmi_wait_register(reg_gusb2phyacc0, PHY_VSTS_BUSY, 0);
+	ret = utmi_wait_register(reg_gusb2phyacc0, phy_reg->vstatus_busy, 0);
 	if (ret)
 		return ret;
 
@@ -957,6 +963,7 @@ static int get_phy_data_by_efuse(struct rtk_phy *rtk_phy,
 
 static int parse_phy_data(struct rtk_phy *rtk_phy)
 {
+	struct phy_cfg *phy_cfg = rtk_phy->phy_cfg;
 	struct device *dev = rtk_phy->dev;
 	struct device_node *np = dev->of_node;
 	struct phy_parameter *phy_parameter;
@@ -974,6 +981,9 @@ static int parse_phy_data(struct rtk_phy *rtk_phy)
 		phy_parameter->phy_reg.reg_wrap_vstatus = of_iomap(np, 0);
 		phy_parameter->phy_reg.reg_gusb2phyacc0 = of_iomap(np, 1) + index;
 		phy_parameter->phy_reg.vstatus_index = index;
+		phy_parameter->phy_reg.vstatus_offset = phy_cfg->vstatus_offset;
+		phy_parameter->phy_reg.vstatus_busy = phy_cfg->vstatus_busy;
+		phy_parameter->phy_reg.new_reg_req = phy_cfg->new_reg_req;
 
 		if (of_property_read_bool(np, "realtek,inverse-hstx-sync-clock"))
 			phy_parameter->inverse_hstx_sync_clock = true;
@@ -1085,6 +1095,9 @@ static const struct phy_cfg rtd1295_phy_cfg = {
 	.driving_updated_for_dev_dis = 0xf,
 	.use_default_parameter = false,
 	.is_double_sensitivity_mode = false,
+	.vstatus_offset = 0,
+	.vstatus_busy = PHY_VSTS_BUSY,
+	.new_reg_req = PHY_NEW_REG_REQ,
 };
 
 static const struct phy_cfg rtd1395_phy_cfg = {
@@ -1109,6 +1122,9 @@ static const struct phy_cfg rtd1395_phy_cfg = {
 	.driving_updated_for_dev_dis = 0xf,
 	.use_default_parameter = false,
 	.is_double_sensitivity_mode = false,
+	.vstatus_offset = 0,
+	.vstatus_busy = PHY_VSTS_BUSY,
+	.new_reg_req = PHY_NEW_REG_REQ,
 };
 
 static const struct phy_cfg rtd1395_phy_cfg_2port = {
@@ -1133,6 +1149,9 @@ static const struct phy_cfg rtd1395_phy_cfg_2port = {
 	.driving_updated_for_dev_dis = 0xf,
 	.use_default_parameter = false,
 	.is_double_sensitivity_mode = false,
+	.vstatus_offset = 0,
+	.vstatus_busy = PHY_VSTS_BUSY,
+	.new_reg_req = PHY_NEW_REG_REQ,
 };
 
 static const struct phy_cfg rtd1619_phy_cfg = {
@@ -1155,6 +1174,9 @@ static const struct phy_cfg rtd1619_phy_cfg = {
 	.driving_updated_for_dev_dis = 0xf,
 	.use_default_parameter = false,
 	.is_double_sensitivity_mode = false,
+	.vstatus_offset = 0,
+	.vstatus_busy = PHY_VSTS_BUSY,
+	.new_reg_req = PHY_NEW_REG_REQ,
 };
 
 static const struct phy_cfg rtd1319_phy_cfg = {
@@ -1181,6 +1203,9 @@ static const struct phy_cfg rtd1319_phy_cfg = {
 	.driving_updated_for_dev_dis = 0xf,
 	.use_default_parameter = false,
 	.is_double_sensitivity_mode = true,
+	.vstatus_offset = 0,
+	.vstatus_busy = PHY_VSTS_BUSY,
+	.new_reg_req = PHY_NEW_REG_REQ,
 };
 
 static const struct phy_cfg rtd1312c_phy_cfg = {
@@ -1206,6 +1231,9 @@ static const struct phy_cfg rtd1312c_phy_cfg = {
 	.driving_updated_for_dev_dis = 0xf,
 	.use_default_parameter = false,
 	.is_double_sensitivity_mode = true,
+	.vstatus_offset = 0,
+	.vstatus_busy = PHY_VSTS_BUSY,
+	.new_reg_req = PHY_NEW_REG_REQ,
 };
 
 static const struct phy_cfg rtd1619b_phy_cfg = {
@@ -1231,6 +1259,9 @@ static const struct phy_cfg rtd1619b_phy_cfg = {
 	.driving_updated_for_dev_dis = 0x8,
 	.use_default_parameter = false,
 	.is_double_sensitivity_mode = true,
+	.vstatus_offset = 0,
+	.vstatus_busy = PHY_VSTS_BUSY,
+	.new_reg_req = PHY_NEW_REG_REQ,
 };
 
 static const struct phy_cfg rtd1319d_phy_cfg = {
@@ -1256,6 +1287,9 @@ static const struct phy_cfg rtd1319d_phy_cfg = {
 	.driving_updated_for_dev_dis = 0x8,
 	.use_default_parameter = false,
 	.is_double_sensitivity_mode = true,
+	.vstatus_offset = 0,
+	.vstatus_busy = PHY_VSTS_BUSY,
+	.new_reg_req = PHY_NEW_REG_REQ,
 };
 
 static const struct phy_cfg rtd1315e_phy_cfg = {
@@ -1282,6 +1316,9 @@ static const struct phy_cfg rtd1315e_phy_cfg = {
 	.driving_updated_for_dev_dis = 0x8,
 	.use_default_parameter = false,
 	.is_double_sensitivity_mode = true,
+	.vstatus_offset = 0,
+	.vstatus_busy = PHY_VSTS_BUSY,
+	.new_reg_req = PHY_NEW_REG_REQ,
 };
 
 static const struct of_device_id usbphy_rtk_dt_match[] = {
-- 
2.53.0

[PATCH v5 2/6] phy: realtek: usb2: introduce read and write functions to driver data

[Rustam Adilov]

RTL9607C is a big endian SoC but has little endian USB host controller and
thus, reads and writes to the reg_gusb2phyacc0 should go through
le32_to_cpu and cpu_to_le32 functions respectively. This doesn't apply to
vstatus register though as it is in the different register region compared
to USB host controller region.

The reason is readl/writel functions, despite the supposed little endian
byte swap, still operate with native endian and this is because RTL9607C
is MIPS and here readl/writel are native unless SWAP_IO_SPACE is enabled.
Enabling SWAP_IO_SPACE results in boot hangs so that is not an option.
And so wrapping them around le32 makes a proper byte swap from big endian
to little endian.

To handle this situation, introduce read and write functions to the driver
data and create a default variation of read and write function for the
current RTD SoCs.

Adjust all instances of utmi_wait_register function to now include the read
function as one of its arguments.

Assign the existing phy configuration for RTD SoCs to the default read
and write functions.

Co-developed-by: Michael Zavertkin <misha.zavertkin@mail.ru>
Signed-off-by: Michael Zavertkin <misha.zavertkin@mail.ru>
Signed-off-by: Rustam Adilov <adilov@disroot.org>
---
 drivers/phy/realtek/phy-rtk-usb2.c | 63 ++++++++++++++++++++++++------
 1 file changed, 50 insertions(+), 13 deletions(-)

diff --git a/drivers/phy/realtek/phy-rtk-usb2.c b/drivers/phy/realtek/phy-rtk-usb2.c
index f5f943518d4f..19297a42a841 100644
--- a/drivers/phy/realtek/phy-rtk-usb2.c
+++ b/drivers/phy/realtek/phy-rtk-usb2.c
@@ -67,6 +67,9 @@ struct phy_reg {
 	int vstatus_offset;
 	u32 vstatus_busy;
 	u32 new_reg_req;
+
+	u32 (*read)(void __iomem *reg);
+	void (*write)(u32 val, void __iomem *reg);
 };
 
 struct phy_data {
@@ -102,6 +105,9 @@ struct phy_cfg {
 	int vstatus_offset;
 	u32 vstatus_busy;
 	u32 new_reg_req;
+
+	u32 (*read)(void __iomem *reg);
+	void (*write)(u32 val, void __iomem *reg);
 };
 
 struct phy_parameter {
@@ -128,6 +134,16 @@ struct rtk_phy {
 	struct dentry *debug_dir;
 };
 
+static u32 rtk_usb2phy_read(void __iomem *reg)
+{
+	return readl(reg);
+}
+
+static void rtk_usb2phy_write(u32 val, void __iomem *reg)
+{
+	writel(val, reg);
+}
+
 /* mapping 0xE0 to 0 ... 0xE7 to 7, 0xF0 to 8 ,,, 0xF7 to 15 */
 static inline int page_addr_to_array_index(u8 addr)
 {
@@ -144,12 +160,13 @@ static inline u8 array_index_to_page_addr(int index)
 #define PHY_IO_TIMEOUT_USEC		(50000)
 #define PHY_IO_DELAY_US			(100)
 
-static inline int utmi_wait_register(void __iomem *reg, u32 mask, u32 result)
+static inline int utmi_wait_register(u32 (*read)(void __iomem *reg), void __iomem *reg, u32 mask,
+				     u32 result)
 {
 	int ret;
 	unsigned int val;
 
-	ret = read_poll_timeout(readl, val, ((val & mask) == result),
+	ret = read_poll_timeout(read, val, ((val & mask) == result),
 				PHY_IO_DELAY_US, PHY_IO_TIMEOUT_USEC, false, reg);
 	if (ret) {
 		pr_err("%s can't program USB phy\n", __func__);
@@ -168,25 +185,25 @@ static char rtk_phy_read(struct phy_reg *phy_reg, char addr)
 	addr -= OFFEST_PHY_READ;
 
 	/* polling until VBusy == 0 */
-	ret = utmi_wait_register(reg_gusb2phyacc0, phy_reg->vstatus_busy, 0);
+	ret = utmi_wait_register(phy_reg->read, reg_gusb2phyacc0, phy_reg->vstatus_busy, 0);
 	if (ret)
 		return (char)ret;
 
 	/* VCtrl = low nibble of addr, and set phy_reg->new_reg_req */
 	val = phy_reg->new_reg_req | (GET_LOW_NIBBLE(addr) << PHY_VCTRL_SHIFT);
-	writel(val, reg_gusb2phyacc0);
-	ret = utmi_wait_register(reg_gusb2phyacc0, phy_reg->vstatus_busy, 0);
+	phy_reg->write(val, reg_gusb2phyacc0);
+	ret = utmi_wait_register(phy_reg->read, reg_gusb2phyacc0, phy_reg->vstatus_busy, 0);
 	if (ret)
 		return (char)ret;
 
 	/* VCtrl = high nibble of addr, and set phy_reg->new_reg_req */
 	val = phy_reg->new_reg_req | (GET_HIGH_NIBBLE(addr) << PHY_VCTRL_SHIFT);
-	writel(val, reg_gusb2phyacc0);
-	ret = utmi_wait_register(reg_gusb2phyacc0, phy_reg->vstatus_busy, 0);
+	phy_reg->write(val, reg_gusb2phyacc0);
+	ret = utmi_wait_register(phy_reg->read, reg_gusb2phyacc0, phy_reg->vstatus_busy, 0);
 	if (ret)
 		return (char)ret;
 
-	val = readl(reg_gusb2phyacc0);
+	val = phy_reg->read(reg_gusb2phyacc0);
 
 	return (char)(val & PHY_REG_DATA_MASK);
 }
@@ -202,23 +219,23 @@ static int rtk_phy_write(struct phy_reg *phy_reg, char addr, char data)
 	/* write data to VStatusOut2 (data output to phy) */
 	writel((u32)data << shift_bits, reg_wrap_vstatus + phy_reg->vstatus_offset);
 
-	ret = utmi_wait_register(reg_gusb2phyacc0, phy_reg->vstatus_busy, 0);
+	ret = utmi_wait_register(phy_reg->read, reg_gusb2phyacc0, phy_reg->vstatus_busy, 0);
 	if (ret)
 		return ret;
 
 	/* VCtrl = low nibble of addr, set phy_reg->new_reg_req */
 	val = phy_reg->new_reg_req | (GET_LOW_NIBBLE(addr) << PHY_VCTRL_SHIFT);
 
-	writel(val, reg_gusb2phyacc0);
-	ret = utmi_wait_register(reg_gusb2phyacc0, phy_reg->vstatus_busy, 0);
+	phy_reg->write(val, reg_gusb2phyacc0);
+	ret = utmi_wait_register(phy_reg->read, reg_gusb2phyacc0, phy_reg->vstatus_busy, 0);
 	if (ret)
 		return ret;
 
 	/* VCtrl = high nibble of addr, set phy_reg->new_reg_req */
 	val = phy_reg->new_reg_req | (GET_HIGH_NIBBLE(addr) << PHY_VCTRL_SHIFT);
 
-	writel(val, reg_gusb2phyacc0);
-	ret = utmi_wait_register(reg_gusb2phyacc0, phy_reg->vstatus_busy, 0);
+	phy_reg->write(val, reg_gusb2phyacc0);
+	ret = utmi_wait_register(phy_reg->read, reg_gusb2phyacc0, phy_reg->vstatus_busy, 0);
 	if (ret)
 		return ret;
 
@@ -984,6 +1001,8 @@ static int parse_phy_data(struct rtk_phy *rtk_phy)
 		phy_parameter->phy_reg.vstatus_offset = phy_cfg->vstatus_offset;
 		phy_parameter->phy_reg.vstatus_busy = phy_cfg->vstatus_busy;
 		phy_parameter->phy_reg.new_reg_req = phy_cfg->new_reg_req;
+		phy_parameter->phy_reg.read = phy_cfg->read;
+		phy_parameter->phy_reg.write = phy_cfg->write;
 
 		if (of_property_read_bool(np, "realtek,inverse-hstx-sync-clock"))
 			phy_parameter->inverse_hstx_sync_clock = true;
@@ -1098,6 +1117,8 @@ static const struct phy_cfg rtd1295_phy_cfg = {
 	.vstatus_offset = 0,
 	.vstatus_busy = PHY_VSTS_BUSY,
 	.new_reg_req = PHY_NEW_REG_REQ,
+	.read = rtk_usb2phy_read,
+	.write = rtk_usb2phy_write,
 };
 
 static const struct phy_cfg rtd1395_phy_cfg = {
@@ -1125,6 +1146,8 @@ static const struct phy_cfg rtd1395_phy_cfg = {
 	.vstatus_offset = 0,
 	.vstatus_busy = PHY_VSTS_BUSY,
 	.new_reg_req = PHY_NEW_REG_REQ,
+	.read = rtk_usb2phy_read,
+	.write = rtk_usb2phy_write,
 };
 
 static const struct phy_cfg rtd1395_phy_cfg_2port = {
@@ -1152,6 +1175,8 @@ static const struct phy_cfg rtd1395_phy_cfg_2port = {
 	.vstatus_offset = 0,
 	.vstatus_busy = PHY_VSTS_BUSY,
 	.new_reg_req = PHY_NEW_REG_REQ,
+	.read = rtk_usb2phy_read,
+	.write = rtk_usb2phy_write,
 };
 
 static const struct phy_cfg rtd1619_phy_cfg = {
@@ -1177,6 +1202,8 @@ static const struct phy_cfg rtd1619_phy_cfg = {
 	.vstatus_offset = 0,
 	.vstatus_busy = PHY_VSTS_BUSY,
 	.new_reg_req = PHY_NEW_REG_REQ,
+	.read = rtk_usb2phy_read,
+	.write = rtk_usb2phy_write,
 };
 
 static const struct phy_cfg rtd1319_phy_cfg = {
@@ -1206,6 +1233,8 @@ static const struct phy_cfg rtd1319_phy_cfg = {
 	.vstatus_offset = 0,
 	.vstatus_busy = PHY_VSTS_BUSY,
 	.new_reg_req = PHY_NEW_REG_REQ,
+	.read = rtk_usb2phy_read,
+	.write = rtk_usb2phy_write,
 };
 
 static const struct phy_cfg rtd1312c_phy_cfg = {
@@ -1234,6 +1263,8 @@ static const struct phy_cfg rtd1312c_phy_cfg = {
 	.vstatus_offset = 0,
 	.vstatus_busy = PHY_VSTS_BUSY,
 	.new_reg_req = PHY_NEW_REG_REQ,
+	.read = rtk_usb2phy_read,
+	.write = rtk_usb2phy_write,
 };
 
 static const struct phy_cfg rtd1619b_phy_cfg = {
@@ -1262,6 +1293,8 @@ static const struct phy_cfg rtd1619b_phy_cfg = {
 	.vstatus_offset = 0,
 	.vstatus_busy = PHY_VSTS_BUSY,
 	.new_reg_req = PHY_NEW_REG_REQ,
+	.read = rtk_usb2phy_read,
+	.write = rtk_usb2phy_write,
 };
 
 static const struct phy_cfg rtd1319d_phy_cfg = {
@@ -1290,6 +1323,8 @@ static const struct phy_cfg rtd1319d_phy_cfg = {
 	.vstatus_offset = 0,
 	.vstatus_busy = PHY_VSTS_BUSY,
 	.new_reg_req = PHY_NEW_REG_REQ,
+	.read = rtk_usb2phy_read,
+	.write = rtk_usb2phy_write,
 };
 
 static const struct phy_cfg rtd1315e_phy_cfg = {
@@ -1319,6 +1354,8 @@ static const struct phy_cfg rtd1315e_phy_cfg = {
 	.vstatus_offset = 0,
 	.vstatus_busy = PHY_VSTS_BUSY,
 	.new_reg_req = PHY_NEW_REG_REQ,
+	.read = rtk_usb2phy_read,
+	.write = rtk_usb2phy_write,
 };
 
 static const struct of_device_id usbphy_rtk_dt_match[] = {
-- 
2.53.0

[PATCH v5 3/6] dt-bindings: phy: realtek,usb2phy.yaml: extend for resets and RTL9607C support

[Rustam Adilov]

Add the "realtek,rtl9607-usb2phy" compatible for USB2 PHY on the RTL9607C
SoC series.

Add a resets property to properties to describe the usb2phy reset line.

In RTL9607C, USB2 PHY reset line is from "IP Enable controller" which is
multipurpose and handle activating various SoC peripherals.

It is unclear whether RTD SoCs have something similar to that so set
the resets to false for these devices.

RTL9607C requires the "resets" to be specified so add the corresponding
if check for the "realtek,rtl9607-usb2phy" compatible.

Signed-off-by: Rustam Adilov <adilov@disroot.org>
---
 .../bindings/phy/realtek,usb2phy.yaml         | 25 ++++++++++++++++++-
 1 file changed, 24 insertions(+), 1 deletion(-)

diff --git a/Documentation/devicetree/bindings/phy/realtek,usb2phy.yaml b/Documentation/devicetree/bindings/phy/realtek,usb2phy.yaml
index 9911ada39ee7..7b50833c8e19 100644
--- a/Documentation/devicetree/bindings/phy/realtek,usb2phy.yaml
+++ b/Documentation/devicetree/bindings/phy/realtek,usb2phy.yaml
@@ -11,7 +11,8 @@ maintainers:
   - Stanley Chang <stanley_chang@realtek.com>
 
 description: |
-  Realtek USB 2.0 PHY support the digital home center (DHC) RTD series SoCs.
+  Realtek USB 2.0 PHY support the digital home center (DHC) RTD and
+  RTL9607C series SoCs.
   The USB 2.0 PHY driver is designed to support the XHCI controller. The SoCs
   support multiple XHCI controllers. One PHY device node maps to one XHCI
   controller.
@@ -57,6 +58,12 @@ description: |
   XHCI controller#1 -- usb2phy -- phy#0
   XHCI controller#2 -- usb2phy -- phy#0
 
+  RTL9607C SoCs USB
+  The USB architecture includes OHCI and EHCI controllers.
+  Both of them map to one USB2.0 PHY.
+  OHCI controller#0 -- usb2phy -- phy#0
+  EHCI controller#0 -- usb2phy -- phy#0
+
 properties:
   compatible:
     enum:
@@ -69,6 +76,7 @@ properties:
       - realtek,rtd1395-usb2phy-2port
       - realtek,rtd1619-usb2phy
       - realtek,rtd1619b-usb2phy
+      - realtek,rtl9607-usb2phy
 
   reg:
     items:
@@ -130,6 +138,9 @@ properties:
     minimum: -8
     maximum: 8
 
+  resets:
+    maxItems: 1
+
 required:
   - compatible
   - reg
@@ -157,6 +168,18 @@ allOf:
     then:
       properties:
         realtek,driving-level-compensate: false
+  - if:
+      properties:
+        compatible:
+          contains:
+            enum:
+              - realtek,rtl9607-usb2phy
+    then:
+      required:
+        - resets
+    else:
+      properties:
+        resets: false
 
 additionalProperties: false
 
-- 
2.53.0

Re: [PATCH v5 3/6] dt-bindings: phy: realtek,usb2phy.yaml: extend for resets and RTL9607C support

[Krzysztof Kozlowski]

On Tue, Apr 21, 2026 at 12:19:38AM +0500, Rustam Adilov wrote:
> Add the "realtek,rtl9607-usb2phy" compatible for USB2 PHY on the RTL9607C
> SoC series.
> 
> Add a resets property to properties to describe the usb2phy reset line.
> 
> In RTL9607C, USB2 PHY reset line is from "IP Enable controller" which is
> multipurpose and handle activating various SoC peripherals.
> 
> It is unclear whether RTD SoCs have something similar to that so set
> the resets to false for these devices.
> 
> RTL9607C requires the "resets" to be specified so add the corresponding
> if check for the "realtek,rtl9607-usb2phy" compatible.

<form letter>
This is a friendly reminder during the review process.

It looks like you received a tag and forgot to add it.

If you do not know the process, here is a short explanation:
Please add Acked-by/Reviewed-by/Tested-by tags when posting new
versions of patchset, under or above your Signed-off-by tag, unless
patch changed significantly (e.g. new properties added to the DT
bindings). Tag is "received", when provided in a message replied to you
on the mailing list. Tools like b4 can help here. However, there's no
need to repost patches *only* to add the tags. The upstream maintainer
will do that for tags received on the version they apply.

Please read:
https://elixir.bootlin.com/linux/v6.12-rc3/source/Documentation/process/submitting-patches.rst#L577

If a tag was not added on purpose, please state in the patch changelog
or cover letter why and what changed.
</form letter>

Pay attention to the last part. Intentional dropping tag is the same as
"not adding".

Best regards,
Krzysztof

[PATCH v5 4/6] phy: realtek: usb2: introduce reset controller struct

[Rustam Adilov]

In RTL9607C, there is so called "IP Enable Controller" which resemble
reset controller with reset lines and is used for various things like
USB, PCIE, GMAC and such.

Introduce the reset_control struct to this driver to handle deasserting
usb2 phy reset line.

Make use of the function devm_reset_control_array_get_optional_exclusive()
function to get the reset controller and since existing RTD SoCs don't
specify the resets we can have a cleaner code.

Since the vendor usb driver developed by Realtek doesn't assert the reset
line (or in their case clear the register bit), we can reasonably assume
reset_control_assert is not needed here.

Co-developed-by: Michael Zavertkin <misha.zavertkin@mail.ru>
Signed-off-by: Michael Zavertkin <misha.zavertkin@mail.ru>
Signed-off-by: Rustam Adilov <adilov@disroot.org>
---
 drivers/phy/realtek/phy-rtk-usb2.c | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/drivers/phy/realtek/phy-rtk-usb2.c b/drivers/phy/realtek/phy-rtk-usb2.c
index 19297a42a841..aaa07a69224a 100644
--- a/drivers/phy/realtek/phy-rtk-usb2.c
+++ b/drivers/phy/realtek/phy-rtk-usb2.c
@@ -17,6 +17,7 @@
 #include <linux/sys_soc.h>
 #include <linux/mfd/syscon.h>
 #include <linux/phy/phy.h>
+#include <linux/reset.h>
 #include <linux/usb.h>
 
 /* GUSB2PHYACCn register */
@@ -130,6 +131,7 @@ struct rtk_phy {
 	struct phy_cfg *phy_cfg;
 	int num_phy;
 	struct phy_parameter *phy_parameter;
+	struct reset_control *phy_rst;
 
 	struct dentry *debug_dir;
 };
@@ -676,6 +678,15 @@ static int rtk_phy_init(struct phy *phy)
 	if (!rtk_phy)
 		return -EINVAL;
 
+	if (rtk_phy->phy_rst) {
+		ret = reset_control_deassert(rtk_phy->phy_rst);
+
+		if (ret)
+			return ret;
+
+		usleep_range(5000, 6000);
+	}
+
 	for (i = 0; i < rtk_phy->num_phy; i++)
 		ret = do_rtk_phy_init(rtk_phy, i);
 
@@ -1059,6 +1070,11 @@ static int rtk_usb2phy_probe(struct platform_device *pdev)
 
 	rtk_phy->num_phy = phy_cfg->num_phy;
 
+	rtk_phy->phy_rst = devm_reset_control_array_get_optional_exclusive(dev);
+	if (IS_ERR(rtk_phy->phy_rst))
+		return dev_err_probe(dev, PTR_ERR(rtk_phy->phy_rst),
+				     "usb2 phy resets are not working\n");
+
 	ret = parse_phy_data(rtk_phy);
 	if (ret)
 		goto err;
-- 
2.53.0

[PATCH v5 5/6] phy: realtek: usb2: add support for RTL9607C USB2 PHY

[Rustam Adilov]

Add support for the usb2 phy of RTL9607C series based SoCs.
Add the macros and phy config struct for rtl9607.

RTL9607C requires to clear a "force host disconnect" bit in the
specific register (which is at an offset from reg_wrap_vstatus)
before proceeding with phy parameter writes. Since it belongs into
the vstatus register region, it doesn't need bytes swapping.

Add the bool variable to the driver data struct and hide this whole
procedure under the if statement that checks this new variable.

Add the appropriate little endian read and write functions for rtl9607
and assign them to its phy config struct.

As mentioned earlier, the readl/writel are native endian on MIPS arch
if SWAP_IO_SPACE is not enabled. Since enabling SWAP_IO_SPACE results
in boot hang on RTL9607C machine, wrapping le32 around readl/writel
should be a good compromise, but swab32 could be also work.

Co-developed-by: Michael Zavertkin <misha.zavertkin@mail.ru>
Signed-off-by: Michael Zavertkin <misha.zavertkin@mail.ru>
Signed-off-by: Rustam Adilov <adilov@disroot.org>
---
 drivers/phy/realtek/phy-rtk-usb2.c | 57 ++++++++++++++++++++++++++++++
 1 file changed, 57 insertions(+)

diff --git a/drivers/phy/realtek/phy-rtk-usb2.c b/drivers/phy/realtek/phy-rtk-usb2.c
index aaa07a69224a..7671f8a9ff61 100644
--- a/drivers/phy/realtek/phy-rtk-usb2.c
+++ b/drivers/phy/realtek/phy-rtk-usb2.c
@@ -26,6 +26,12 @@
 #define PHY_VCTRL_SHIFT 8
 #define PHY_REG_DATA_MASK 0xff
 
+#define PHY_9607_VSTS_BUSY BIT(17)
+#define PHY_9607_NEW_REG_REQ BIT(13)
+
+#define PHY_9607_FORCE_DISCONNECT_REG 0x10
+#define PHY_9607_FORCE_DISCONNECT_BIT BIT(5)
+
 #define GET_LOW_NIBBLE(addr) ((addr) & 0x0f)
 #define GET_HIGH_NIBBLE(addr) (((addr) & 0xf0) >> 4)
 
@@ -109,6 +115,7 @@ struct phy_cfg {
 
 	u32 (*read)(void __iomem *reg);
 	void (*write)(u32 val, void __iomem *reg);
+	bool force_host_disconnect;
 };
 
 struct phy_parameter {
@@ -146,6 +153,18 @@ static void rtk_usb2phy_write(u32 val, void __iomem *reg)
 	writel(val, reg);
 }
 
+static u32 rtk_usb2phy_read_le(void __iomem *reg)
+{
+	return le32_to_cpu(readl(reg));
+}
+
+static void rtk_usb2phy_write_le(u32 val, void __iomem *reg)
+{
+	u32 tmp = cpu_to_le32(val);
+
+	writel(tmp, reg);
+}
+
 /* mapping 0xE0 to 0 ... 0xE7 to 7, 0xF0 to 8 ,,, 0xF7 to 15 */
 static inline int page_addr_to_array_index(u8 addr)
 {
@@ -600,6 +619,16 @@ static int do_rtk_phy_init(struct rtk_phy *rtk_phy, int index)
 		goto do_toggle;
 	}
 
+	if (phy_cfg->force_host_disconnect) {
+		/* disable force-host-disconnect */
+		u32 temp = readl(phy_reg->reg_wrap_vstatus + PHY_9607_FORCE_DISCONNECT_REG);
+
+		temp &= ~PHY_9607_FORCE_DISCONNECT_BIT;
+		writel(temp, phy_reg->reg_wrap_vstatus + PHY_9607_FORCE_DISCONNECT_REG);
+
+		msleep(10);
+	}
+
 	/* Set page 0 */
 	phy_data_page = phy_cfg->page0;
 	rtk_phy_set_page(phy_reg, 0);
@@ -1374,6 +1403,33 @@ static const struct phy_cfg rtd1315e_phy_cfg = {
 	.write = rtk_usb2phy_write,
 };
 
+static const struct phy_cfg rtl9607_phy_cfg = {
+	.page0_size = MAX_USB_PHY_PAGE0_DATA_SIZE,
+	.page0 = { [0] = {0xe0, 0x95},
+		   [4] = {0xe4, 0x6a},
+		  [12] = {0xf3, 0x31}, },
+	.page1_size = MAX_USB_PHY_PAGE1_DATA_SIZE,
+	.page1 = { [0] = {0xe0, 0x26}, },
+	.page2_size = MAX_USB_PHY_PAGE2_DATA_SIZE,
+	.page2 = { [7] = {0xe7, 0x33}, },
+	.num_phy = 1,
+	.check_efuse_version = CHECK_EFUSE_V2,
+	.efuse_dc_driving_rate = EFUS_USB_DC_CAL_RATE,
+	.dc_driving_mask = 0x1f,
+	.efuse_dc_disconnect_rate = EFUS_USB_DC_DIS_RATE,
+	.dc_disconnect_mask = 0xf,
+	.usb_dc_disconnect_at_page0 = true,
+	.do_toggle = true,
+	.driving_updated_for_dev_dis = 0x8,
+	.is_double_sensitivity_mode = true,
+	.vstatus_offset = 0xc,
+	.vstatus_busy = PHY_9607_VSTS_BUSY,
+	.new_reg_req = PHY_9607_NEW_REG_REQ,
+	.read = rtk_usb2phy_read_le,
+	.write = rtk_usb2phy_write_le,
+	.force_host_disconnect = true,
+};
+
 static const struct of_device_id usbphy_rtk_dt_match[] = {
 	{ .compatible = "realtek,rtd1295-usb2phy", .data = &rtd1295_phy_cfg },
 	{ .compatible = "realtek,rtd1312c-usb2phy", .data = &rtd1312c_phy_cfg },
@@ -1384,6 +1440,7 @@ static const struct of_device_id usbphy_rtk_dt_match[] = {
 	{ .compatible = "realtek,rtd1395-usb2phy-2port", .data = &rtd1395_phy_cfg_2port },
 	{ .compatible = "realtek,rtd1619-usb2phy", .data = &rtd1619_phy_cfg },
 	{ .compatible = "realtek,rtd1619b-usb2phy", .data = &rtd1619b_phy_cfg },
+	{ .compatible = "realtek,rtl9607-usb2phy", .data = &rtl9607_phy_cfg },
 	{},
 };
 MODULE_DEVICE_TABLE(of, usbphy_rtk_dt_match);
-- 
2.53.0

[PATCH v5 6/6] phy: realtek: Make configs available for MACH_REALTEK_RTL

[Rustam Adilov]

Add the MACH_REALTEK_RTL to the if statement to make the config
options available for Realtek RTL SoCs as well.

Signed-off-by: Rustam Adilov <adilov@disroot.org>
---
 drivers/phy/realtek/Kconfig | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/phy/realtek/Kconfig b/drivers/phy/realtek/Kconfig
index 75ac7e7c31ae..76f9215d8b94 100644
--- a/drivers/phy/realtek/Kconfig
+++ b/drivers/phy/realtek/Kconfig
@@ -3,7 +3,7 @@
 # Phy drivers for Realtek platforms
 #
 
-if ARCH_REALTEK || COMPILE_TEST
+if ARCH_REALTEK || MACH_REALTEK_RTL || COMPILE_TEST
 
 config PHY_RTK_RTD_USB2PHY
 	tristate "Realtek RTD USB2 PHY Transceiver Driver"
@@ -29,4 +29,4 @@ config PHY_RTK_RTD_USB3PHY
 	  DWC3 USB IP. This driver will do the PHY initialization
 	  of the parameters.
 
-endif # ARCH_REALTEK || COMPILE_TEST
+endif # ARCH_REALTEK || MACH_REALTEK_RTL || COMPILE_TEST
-- 
2.53.0

Re: [PATCH v5 0/6] phy: realtek: usb2: support for RTL9607C USB2 PHY

[Krzysztof Kozlowski]

On Tue, Apr 21, 2026 at 12:19:35AM +0500, Rustam Adilov wrote:
> This patch series for Realtek USB2 PHY driver adds support for RTL9607C
> USB2 PHY.
> 
> RTL9607C is a big endian MIPS CPU which is quite far from RTD series SoCs
> supported by realtek usb2 phy driver, but the phy initilization is found
> to be very indentical in most areas.
> 
> Most of the code was based on the Realtek's usb driver from the GPL tarball
> in [1] and adjusted to fit into the realtek usb2 phy driver code format.
> 
> The patch series was split into smaller patches that add/change something
> in the driver that are not exactly related to RTL9607C and that also
> helps for easier review. That also means, patch 5 depends on all the prior
> patches that come before it.
> 
> USB2 PHY on RTL9607C is primarly used for its internal OHCI/EHCI controllers.
> 
> [1] - https://github.com/jameywine/GPL-for-GP3000/blob/main/linux-5.10.x/arch/mips/rtl9607c/usb.c
> 
> ---
> Changelog in v5:
> Mostly addressing LLM review
> - Patch 1
>  - changed int to u32 type for new_reg_req and vstatus_busy data fields.
>  - changed comments in rtk_phy_read/write from PHY_NEW_REG_REQ to phy_reg->new_reg_req.
> - Patch 2
>  - explained readl/writel native endianess issue in more detail.
>  - explained why vstatus register doesn't need byte swapping.
> - Patch 4
>  - moved reset_control_deassert to rtk_phy_init function to keep it outside of for loop.
>  - changed msleep(5) to usleep_range(5000, 6000).
>  - explained why reset_control_assert is not needed.
> - Patch 5
>  - explained readl/writel native endianess issue here as well.
>  - explained why FORCE_DISCONNECT_REG doesn't need byte swapping.
> - Link to v4: https://lore.kernel.org/linux-phy/20260406181228.25892-1-adilov@disroot.org/
> 
> Changelog in v4:
> - Patch 2
>  - moved the le variations of read/write functions to Patch 5 where it is actually used because
>    otherwise, it results in unused errors when only Patch 2 is applied.
>  - updated the commit message to to point the reason for le32 wrappers around readl/writel.
> - Patch 3
>  - added "Reviewed by Krzysztof Kozlowski"

Where?

Best regards,
Krzysztof

Re: [PATCH v5 0/6] phy: realtek: usb2: support for RTL9607C USB2 PHY

[Rustam Adilov]

On 2026-04-21 07:09, Krzysztof Kozlowski wrote:
> On Tue, Apr 21, 2026 at 12:19:35AM +0500, Rustam Adilov wrote:
>> This patch series for Realtek USB2 PHY driver adds support for RTL9607C
>> USB2 PHY.
>> 
>> RTL9607C is a big endian MIPS CPU which is quite far from RTD series SoCs
>> supported by realtek usb2 phy driver, but the phy initilization is found
>> to be very indentical in most areas.
>> 
>> Most of the code was based on the Realtek's usb driver from the GPL tarball
>> in [1] and adjusted to fit into the realtek usb2 phy driver code format.
>> 
>> The patch series was split into smaller patches that add/change something
>> in the driver that are not exactly related to RTL9607C and that also
>> helps for easier review. That also means, patch 5 depends on all the prior
>> patches that come before it.
>> 
>> USB2 PHY on RTL9607C is primarly used for its internal OHCI/EHCI controllers.
>> 
>> [1] - https://github.com/jameywine/GPL-for-GP3000/blob/main/linux-5.10.x/arch/mips/rtl9607c/usb.c
>> 
>> ---
>> Changelog in v5:
>> Mostly addressing LLM review
>> - Patch 1
>>  - changed int to u32 type for new_reg_req and vstatus_busy data fields.
>>  - changed comments in rtk_phy_read/write from PHY_NEW_REG_REQ to phy_reg->new_reg_req.
>> - Patch 2
>>  - explained readl/writel native endianess issue in more detail.
>>  - explained why vstatus register doesn't need byte swapping.
>> - Patch 4
>>  - moved reset_control_deassert to rtk_phy_init function to keep it outside of for loop.
>>  - changed msleep(5) to usleep_range(5000, 6000).
>>  - explained why reset_control_assert is not needed.
>> - Patch 5
>>  - explained readl/writel native endianess issue here as well.
>>  - explained why FORCE_DISCONNECT_REG doesn't need byte swapping.
>> - Link to v4: https://lore.kernel.org/linux-phy/20260406181228.25892-1-adilov@disroot.org/
>> 
>> Changelog in v4:
>> - Patch 2
>>  - moved the le variations of read/write functions to Patch 5 where it is actually used because
>>    otherwise, it results in unused errors when only Patch 2 is applied.
>>  - updated the commit message to to point the reason for le32 wrappers around readl/writel.
>> - Patch 3
>>  - added "Reviewed by Krzysztof Kozlowski"
> 
> Where?

Oi, it must have been lost in transmission because i added the tag manually to .patch after
format-patch command instead of adding it to commit message. So it disappeared in v5.

Sorry about that. Unintentional blunder on my part.

> Best regards,
> Krzysztof

Thanks,
Rustam