[PATCH net-next v4 0/3] net: ethernet: oa_tc6: add reset-gpios support

[PATCH net-next v4 0/3] net: ethernet: oa_tc6: add reset-gpios support

[Alessandro Zini]

This series adds optional reset GPIO support for OPEN Alliance 10BASE-T1x
MAC-PHY devices using the OA TC6 library, and documents the property for
the Microchip LAN8650/1 MAC-PHY.

Patch 1 documents the optional reset-gpios property for the Microchip
LAN8650/1.

Patch 2 converts oa_tc6_init() to return an ERR_PTR instead of NULL, so
that -EPROBE_DEFER from the GPIO lookup added in patch 3 reaches the
driver core instead of being turned into -ENODEV.

Patch 3 implements optional reset GPIO handling in the common oa_tc6
library. The reset line is requested asserted, kept asserted for 10 us and
1 ms of settle time is allowed after deassertion for the crystal
oscillator startup, before the OPEN Alliance TC6 communication is started.

Note: In v3, a DT binding patch was added for adi,ad3306.yaml based on
review bot feedback, as adin1140 also calls oa_tc6_init(). However, I have
later realized that ADIN1140/AD3306 does not have an external reset pin...
Therefore, that binding patch has been dropped in v4. Sorry for the
spam.

Changes in v4:
- Dropped patch 2 from v3 ("dt-bindings: net: adi,ad3306: add reset-gpios
  property") because the ADIN1140/AD3306 does not feature a physical reset
  pin.
- Renumbered patches accordingly (now 3 patches).

Changes in v3:
- Picked up Acked-by tag from Krzysztof on patch 1.
- Added patch documenting the property for adi,ad3306 (dropped in v4).
- Added patch converting oa_tc6_init() to ERR_PTR, so that -EPROBE_DEFER
  is not lost.
- In the driver patch, requested the GPIO with GPIOD_OUT_HIGH and dropped
  the now redundant gpiod_set_value_cansleep() call, as suggested by
  Qingfang Deng, propagated the dev_err_probe() return value, reworded
  the comment and sorted the new includes.

Changes in v2:
- Moved reset-gpios handling from the lan865x driver into the common oa_tc6
  library (drivers/net/ethernet/oa_tc6.c), so all MAC-PHY devices can make
  use of it, as suggested by Parthiban Veerasooran.

Alessandro Zini (3):
  dt-bindings: net: microchip,lan8650: add reset-gpios property
  net: ethernet: oa_tc6: return ERR_PTR from oa_tc6_init()
  net: ethernet: oa_tc6: add reset-gpios support

 .../bindings/net/microchip,lan8650.yaml       |  5 ++
 drivers/net/ethernet/adi/adin1140.c           |  4 +-
 .../net/ethernet/microchip/lan865x/lan865x.c  |  4 +-
 drivers/net/ethernet/oa_tc6.c                 | 46 +++++++++++++------
 4 files changed, 42 insertions(+), 17 deletions(-)

-- 
2.55.0

[PATCH net-next v4 1/3] dt-bindings: net: microchip,lan8650: add reset-gpios property

[Alessandro Zini]

Add optional reset-gpios property to describe the connection to the
hardware RESET_N pin of the LAN8650/1 MAC-PHY.

Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@oss.qualcomm.com>
Signed-off-by: Alessandro Zini <alessandro.zini@siemens.com>
---
Changes in v4:
- None

Changes in v3:
- Picked up Acked-by tag.

Changes in v2:
- None
 Documentation/devicetree/bindings/net/microchip,lan8650.yaml | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/Documentation/devicetree/bindings/net/microchip,lan8650.yaml b/Documentation/devicetree/bindings/net/microchip,lan8650.yaml
index 766ff58147ae3..1b543ce307632 100644
--- a/Documentation/devicetree/bindings/net/microchip,lan8650.yaml
+++ b/Documentation/devicetree/bindings/net/microchip,lan8650.yaml
@@ -40,6 +40,11 @@ properties:
       Event.
     maxItems: 1
 
+  reset-gpios:
+    description:
+      GPIO connected to the active-low RESET_N pin of the MAC-PHY.
+    maxItems: 1
+
   spi-max-frequency:
     minimum: 15000000
     maximum: 25000000
-- 
2.55.0

[PATCH net-next v4 2/3] net: ethernet: oa_tc6: return ERR_PTR from oa_tc6_init()

[Alessandro Zini]

oa_tc6_init() returns NULL on failure, so the reason for the failure is
lost and both callers turn it into a plain -ENODEV. In preparation for
acquiring an optional reset GPIO, which can return -EPROBE_DEFER when the
GPIO provider is not available yet, the error code has to reach the driver
core, otherwise the SPI device would never be probed again.

Convert oa_tc6_init() to return an ERR_PTR on failure instead of NULL,
and update the lan865x and adin1140 callers to check with IS_ERR() and
propagate the error with PTR_ERR().

Signed-off-by: Alessandro Zini <alessandro.zini@siemens.com>
---
Changes in v4:
- None (renumbered from 3/4 to 2/3).

Changes in v3:
- New patch to allow proper propagation of error codes, such as
  -EPROBE_DEFER.
 drivers/net/ethernet/adi/adin1140.c           |  4 +--
 .../net/ethernet/microchip/lan865x/lan865x.c  |  4 +--
 drivers/net/ethernet/oa_tc6.c                 | 27 ++++++++++---------
 3 files changed, 18 insertions(+), 17 deletions(-)

diff --git a/drivers/net/ethernet/adi/adin1140.c b/drivers/net/ethernet/adi/adin1140.c
index 93710baca1517..adc6e2ef44f8c 100644
--- a/drivers/net/ethernet/adi/adin1140.c
+++ b/drivers/net/ethernet/adi/adin1140.c
@@ -725,8 +725,8 @@ static int adin1140_probe(struct spi_device *spi)
 	tc6_quirks.quirk_flags = OA_TC6_BROKEN_PHY;
 
 	priv->tc6 = oa_tc6_init(spi, netdev, &tc6_quirks);
-	if (!priv->tc6)
-		return -ENODEV;
+	if (IS_ERR(priv->tc6))
+		return PTR_ERR(priv->tc6);
 
 	ret = devm_add_action_or_reset(&spi->dev, adin1140_oa_tc6_remove,
 				       priv->tc6);
diff --git a/drivers/net/ethernet/microchip/lan865x/lan865x.c b/drivers/net/ethernet/microchip/lan865x/lan865x.c
index 26a2761332a5a..127afb9e9f141 100644
--- a/drivers/net/ethernet/microchip/lan865x/lan865x.c
+++ b/drivers/net/ethernet/microchip/lan865x/lan865x.c
@@ -347,8 +347,8 @@ static int lan865x_probe(struct spi_device *spi)
 	INIT_WORK(&priv->multicast_work, lan865x_multicast_work_handler);
 
 	priv->tc6 = oa_tc6_init(spi, netdev, NULL);
-	if (!priv->tc6) {
-		ret = -ENODEV;
+	if (IS_ERR(priv->tc6)) {
+		ret = PTR_ERR(priv->tc6);
 		goto free_netdev;
 	}
 
diff --git a/drivers/net/ethernet/oa_tc6.c b/drivers/net/ethernet/oa_tc6.c
index 6fcc5f561d560..8c82bc8354ede 100644
--- a/drivers/net/ethernet/oa_tc6.c
+++ b/drivers/net/ethernet/oa_tc6.c
@@ -1453,7 +1453,7 @@ static int oa_tc6_check_ctrl_protection(struct oa_tc6 *tc6)
  * @quirks: device specific modifiers for the OA TC6 protocol.
  *
  * Return: pointer reference to the oa_tc6 structure if the MAC-PHY
- * initialization is successful otherwise NULL.
+ * initialization is successful otherwise an ERR_PTR.
  */
 struct oa_tc6 *oa_tc6_init(struct spi_device *spi, struct net_device *netdev,
 			   struct oa_tc6_quirks *quirks)
@@ -1463,7 +1463,7 @@ struct oa_tc6 *oa_tc6_init(struct spi_device *spi, struct net_device *netdev,
 
 	tc6 = devm_kzalloc(&spi->dev, sizeof(*tc6), GFP_KERNEL);
 	if (!tc6)
-		return NULL;
+		return ERR_PTR(-ENOMEM);
 
 	tc6->spi = spi;
 	tc6->netdev = netdev;
@@ -1476,60 +1476,61 @@ struct oa_tc6 *oa_tc6_init(struct spi_device *spi, struct net_device *netdev,
 
 	/* Set the SPI controller to pump at realtime priority */
 	tc6->spi->rt = true;
-	if (spi_setup(tc6->spi) < 0)
-		return NULL;
+	ret = spi_setup(tc6->spi);
+	if (ret < 0)
+		return ERR_PTR(ret);
 
 	tc6->spi_ctrl_tx_buf = devm_kzalloc(&tc6->spi->dev,
 					    OA_TC6_CTRL_SPI_BUF_SIZE,
 					    GFP_KERNEL);
 	if (!tc6->spi_ctrl_tx_buf)
-		return NULL;
+		return ERR_PTR(-ENOMEM);
 
 	tc6->spi_ctrl_rx_buf = devm_kzalloc(&tc6->spi->dev,
 					    OA_TC6_CTRL_SPI_BUF_SIZE,
 					    GFP_KERNEL);
 	if (!tc6->spi_ctrl_rx_buf)
-		return NULL;
+		return ERR_PTR(-ENOMEM);
 
 	tc6->spi_data_tx_buf = devm_kzalloc(&tc6->spi->dev,
 					    OA_TC6_SPI_DATA_BUF_SIZE,
 					    GFP_KERNEL);
 	if (!tc6->spi_data_tx_buf)
-		return NULL;
+		return ERR_PTR(-ENOMEM);
 
 	tc6->spi_data_rx_buf = devm_kzalloc(&tc6->spi->dev,
 					    OA_TC6_SPI_DATA_BUF_SIZE,
 					    GFP_KERNEL);
 	if (!tc6->spi_data_rx_buf)
-		return NULL;
+		return ERR_PTR(-ENOMEM);
 
 	/* Check the PROTE bit status so that we can reset the device */
 	ret = oa_tc6_check_ctrl_protection(tc6);
 	if (ret) {
 		dev_err(&tc6->spi->dev,
 			"Failed to check the protection mode: %d\n", ret);
-		return NULL;
+		return ERR_PTR(ret);
 	}
 
 	ret = oa_tc6_sw_reset_macphy(tc6);
 	if (ret) {
 		dev_err(&tc6->spi->dev,
 			"MAC-PHY software reset failed: %d\n", ret);
-		return NULL;
+		return ERR_PTR(ret);
 	}
 
 	ret = oa_tc6_unmask_macphy_error_interrupts(tc6);
 	if (ret) {
 		dev_err(&tc6->spi->dev,
 			"MAC-PHY error interrupts unmask failed: %d\n", ret);
-		return NULL;
+		return ERR_PTR(ret);
 	}
 
 	ret = oa_tc6_phy_init(tc6);
 	if (ret) {
 		dev_err(&tc6->spi->dev,
 			"MAC internal PHY initialization failed: %d\n", ret);
-		return NULL;
+		return ERR_PTR(ret);
 	}
 
 	ret = oa_tc6_enable_data_transfer(tc6);
@@ -1570,7 +1571,7 @@ struct oa_tc6 *oa_tc6_init(struct spi_device *spi, struct net_device *netdev,
 
 phy_exit:
 	oa_tc6_phy_exit(tc6);
-	return NULL;
+	return ERR_PTR(ret);
 }
 EXPORT_SYMBOL_GPL(oa_tc6_init);
 
-- 
2.55.0

[PATCH net-next v4 3/3] net: ethernet: oa_tc6: add reset-gpios support

[Alessandro Zini]

OPEN Alliance 10BASE-T1x MAC-PHY Serial Interface specification
(v1.1, section 8.2) defines an external RESET pin as an optional reset
source for the MAC-PHY.

Add support for an optional reset GPIO in oa_tc6_init(). The GPIO is
requested asserted, kept asserted for 10 us and, once deasserted, 1 ms of
settle time is allowed for the crystal oscillator startup before the
initial TC6 control protection check and software reset are performed.

As the reset handling lives in the common library, it is available to all
MAC-PHY devices using oa_tc6_init() that provide an external reset pin.

Suggested-by: Parthiban Veerasooran <parthiban.veerasooran@microchip.com>
Signed-off-by: Alessandro Zini <alessandro.zini@siemens.com>
---
Changes in v4:
- Clarified commit message regarding devices with external reset pins
  (renumbered from 4/4 to 3/3).

Changes in v3:
- Request the GPIO with GPIOD_OUT_HIGH and drop the now redundant
  gpiod_set_value_cansleep(..., 1) call, as suggested by Qingfang Deng.
- Return the error from dev_err_probe() so that -EPROBE_DEFER is
  propagated to the driver core.
- Reworded comment to stay within 80 columns.
- Sorted the new includes into the existing include block.

Changes in v2:
- Moved reset GPIO handling from lan865x.c to common oa_tc6.c as suggested
  by Parthiban Veerasooran.
 drivers/net/ethernet/oa_tc6.c | 19 +++++++++++++++++++
 1 file changed, 19 insertions(+)

diff --git a/drivers/net/ethernet/oa_tc6.c b/drivers/net/ethernet/oa_tc6.c
index 8c82bc8354ede..364027c39fa4a 100644
--- a/drivers/net/ethernet/oa_tc6.c
+++ b/drivers/net/ethernet/oa_tc6.c
@@ -6,6 +6,8 @@
  */
 
 #include <linux/bitfield.h>
+#include <linux/delay.h>
+#include <linux/gpio/consumer.h>
 #include <linux/iopoll.h>
 #include <linux/interrupt.h>
 #include <linux/mdio.h>
@@ -88,6 +90,7 @@ struct oa_tc6 {
 	bool disable_traffic;
 	bool prot_ctrl;
 	enum oa_tc6_quirk_flag quirk_flags;
+	struct gpio_desc *reset_gpio;
 };
 
 enum oa_tc6_header_type {
@@ -1504,6 +1507,22 @@ struct oa_tc6 *oa_tc6_init(struct spi_device *spi, struct net_device *netdev,
 	if (!tc6->spi_data_rx_buf)
 		return ERR_PTR(-ENOMEM);
 
+	tc6->reset_gpio = devm_gpiod_get_optional(&spi->dev, "reset",
+						  GPIOD_OUT_HIGH);
+	if (IS_ERR(tc6->reset_gpio))
+		return ERR_PTR(dev_err_probe(&spi->dev,
+					     PTR_ERR(tc6->reset_gpio),
+					     "failed to get reset gpio\n"));
+
+	if (tc6->reset_gpio) {
+		/* Keep the reset asserted for 10 us and then allow 1 ms of
+		 * settle time for the crystal oscillator startup.
+		 */
+		fsleep(10);
+		gpiod_set_value_cansleep(tc6->reset_gpio, 0);
+		fsleep(1000);
+	}
+
 	/* Check the PROTE bit status so that we can reset the device */
 	ret = oa_tc6_check_ctrl_protection(tc6);
 	if (ret) {
-- 
2.55.0

Re: [PATCH net-next v4 3/3] net: ethernet: oa_tc6: add reset-gpios support

[Parthiban Veerasooran]

Hi Alessandro,

On 19/09/26 4:44 am, Alessandro Zini wrote:
> 
>   #include <linux/bitfield.h>
> +#include <linux/delay.h>
> +#include <linux/gpio/consumer.h>
>   #include <linux/iopoll.h>
>   #include <linux/interrupt.h>
>   #include <linux/mdio.h>
> @@ -88,6 +90,7 @@ struct oa_tc6 {
>          bool disable_traffic;
>          bool prot_ctrl;
>          enum oa_tc6_quirk_flag quirk_flags;
> +       struct gpio_desc *reset_gpio;
>   };
> 
>   enum oa_tc6_header_type {
> @@ -1504,6 +1507,22 @@ struct oa_tc6 *oa_tc6_init(struct spi_device *spi, struct net_device *netdev,
>          if (!tc6->spi_data_rx_buf)
>                  return ERR_PTR(-ENOMEM);
> 
> +       tc6->reset_gpio = devm_gpiod_get_optional(&spi->dev, "reset",
> +                                                 GPIOD_OUT_HIGH);
> +       if (IS_ERR(tc6->reset_gpio))
> +               return ERR_PTR(dev_err_probe(&spi->dev,
> +                                            PTR_ERR(tc6->reset_gpio),
> +                                            "failed to get reset gpio\n"));
> +
> +       if (tc6->reset_gpio) {
> +               /* Keep the reset asserted for 10 us and then allow 1 ms of
> +                * settle time for the crystal oscillator startup.
> +                */
> +               fsleep(10);
> +               gpiod_set_value_cansleep(tc6->reset_gpio, 0);
> +               fsleep(1000);
> +       }
> +

How do we ensure that the reset has completed before proceeding? Should 
we check for a reset-complete status indication instead of relying only 
on fixed delays?

Also, do we still need oa_tc6_sw_reset_macphy() when reset_gpio is 
implemented? My understanding is that the software reset path is mainly 
needed for MAC-PHYs that do not expose a reset GPIO. In that case, it 
may be cleaner to prefer the GPIO reset when available and fall back to 
the software reset otherwise, rather than always performing both resets.

Best regards,
Parthiban V>          /* Check the PROTE bit status so that we can reset 
the device */
>          ret = oa_tc6_check_ctrl_protection(tc6);
>          if (ret) {
> --
> 2.55.0

RE: [PATCH net-next v4 3/3] net: ethernet: oa_tc6: add reset-gpios support

[Selvamani Rajagopal]

> -----Original Message-----
> From: Parthiban Veerasooran <parthiban.veerasooran@microchip.com>
> Sent: Tuesday, September 22, 2026 2:00 AM


> 
> Also, do we still need oa_tc6_sw_reset_macphy() when reset_gpio is
> implemented? My understanding is that the software reset path is mainly
> needed for MAC-PHYs that do not expose a reset GPIO. In that case, it
> may be cleaner to prefer the GPIO reset when available and fall back to
> the software reset otherwise, rather than always performing both resets.

Register based reset inherently assumes that MACPHY responds to SPI 
read/write transactions. Here is one use case in using GPIO based reset in addition
to register based reset. One could implement a recovery mechanism when 
MACPHY locks up the SPI interface for some reason, 

As we know, the current code just sets disable_traffic flag to prevent the interface from 
working, when SPI transaction fails.

> 
> Best regards,
> Parthiban V> /* Check the PROTE bit status so that we can reset
> the device */
> > ret = oa_tc6_check_ctrl_protection(tc6);
> > if (ret) {
> > --
> > 2.55.0

Re: [PATCH net-next v4 0/3] net: ethernet: oa_tc6: add reset-gpios support

[patchwork-bot+netdevbpf]

Hello:

This series was applied to netdev/net-next.git (main)
by Jakub Kicinski <kuba@kernel.org>:

On Sat, 19 Sep 2026 01:14:47 +0200 you wrote:
> This series adds optional reset GPIO support for OPEN Alliance 10BASE-T1x
> MAC-PHY devices using the OA TC6 library, and documents the property for
> the Microchip LAN8650/1 MAC-PHY.
> 
> Patch 1 documents the optional reset-gpios property for the Microchip
> LAN8650/1.
> 
> [...]

Here is the summary with links:
  - [net-next,v4,1/3] dt-bindings: net: microchip,lan8650: add reset-gpios property
    https://git.kernel.org/netdev/net-next/c/7afc23c9884d
  - [net-next,v4,2/3] net: ethernet: oa_tc6: return ERR_PTR from oa_tc6_init()
    https://git.kernel.org/netdev/net-next/c/15358c2c3dca
  - [net-next,v4,3/3] net: ethernet: oa_tc6: add reset-gpios support
    https://git.kernel.org/netdev/net-next/c/8d5dab574a95

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html