[PATCH 0/2] net: phy: air_en8811h: fix LED GPIO not surviving MCU restart

[PATCH 0/2] net: phy: air_en8811h: fix LED GPIO not surviving MCU restart

[Vitaliy Sochnev]

The EN8811H PHY's LED GPIO pins are only ever configured as outputs
once, in .probe(). Every .config_init() call after the first restarts
the PHY's MD32 MCU, which resets that GPIO configuration back to
inputs, so the LEDs silently stop reflecting link/activity state after
the first link renegotiation, ifdown/ifup, or resume. Patch 1 fixes
that. Patch 2 is a small follow-on cleanup enabled by patch 1's
removal of code from .probe(), kept as a separate patch so patch 1
stays a minimal, clean Fixes: change.

Verified on hardware (Nokia XG-040G-MD, Airoha AN7581) via OpenWrt:
before the fix, the LED stayed dark after the board's first link
renegotiation post-boot; after the fix, it lit correctly through
repeated link up/down cycles.

Vitaliy Sochnev (2):
  net: phy: air_en8811h: move LED GPIO configuration to config_init
  net: phy: air_en8811h: simplify en8811h_probe() trailing return

 drivers/net/phy/air_en8811h.c | 24 ++++++++++++------------
 1 file changed, 12 insertions(+), 12 deletions(-)

-- 
2.55.0

[PATCH 1/2] net: phy: air_en8811h: move LED GPIO configuration to config_init

[Vitaliy Sochnev]

The LED GPIO pins (GPIO3/4/5, mapped to LED2/LED1/LED0) are only ever
configured as outputs once, in .probe(). But .config_init() restarts
the MD32 MCU via en8811h_restart_mcu() on every call after the first
(priv->mcu_needs_restart), and that restart resets buckpbus-mapped MCU
state, including EN8811H_GPIO_OUTPUT. As a result the LED GPIOs fall
back to inputs after the first event that re-triggers .config_init()
(link renegotiation, ifdown/ifup, resume), and the PHY's LEDs stop
reflecting link/activity state even though they worked right after
probe.

Move the GPIO-as-output configuration from .probe() to the end of
.config_init(), so it is reapplied every time the MCU may have been
restarted.

Fixes: 71e79430117d ("net: phy: air_en8811h: Add the Airoha EN8811H PHY driver")
Suggested-by: Mikhail Zhilkin <csharper2005@gmail.com>
Signed-off-by: Vitaliy Sochnev <sochnev.v.74@gmail.com>
---
 drivers/net/phy/air_en8811h.c | 18 +++++++++++-------
 1 file changed, 11 insertions(+), 7 deletions(-)

diff --git a/drivers/net/phy/air_en8811h.c b/drivers/net/phy/air_en8811h.c
index edd49c193e47..0eeb7b9a4e26 100644
--- a/drivers/net/phy/air_en8811h.c
+++ b/drivers/net/phy/air_en8811h.c
@@ -1173,13 +1173,6 @@ static int en8811h_probe(struct phy_device *phydev)
 	if (ret)
 		return ret;
 
-	/* Configure led gpio pins as output */
-	ret = air_phy_buckpbus_reg_modify(phydev, EN8811H_GPIO_OUTPUT,
-					  EN8811H_GPIO_OUTPUT_345,
-					  EN8811H_GPIO_OUTPUT_345);
-	if (ret < 0)
-		return ret;
-
 	return 0;
 }
 
@@ -1324,6 +1317,17 @@ static int en8811h_config_init(struct phy_device *phydev)
 		return ret;
 	}
 
+	/* Configure led gpio pins as output. Must be redone on every
+	 * .config_init(), not just once in .probe(): en8811h_restart_mcu()
+	 * resets buckpbus-mapped MCU state (incl. this register) on every
+	 * call after the first, e.g. on link renegotiation or ifup/ifdown.
+	 */
+	ret = air_phy_buckpbus_reg_modify(phydev, EN8811H_GPIO_OUTPUT,
+					  EN8811H_GPIO_OUTPUT_345,
+					  EN8811H_GPIO_OUTPUT_345);
+	if (ret < 0)
+		return ret;
+
 	return 0;
 }
 
-- 
2.55.0

[PATCH 2/2] net: phy: air_en8811h: simplify en8811h_probe() trailing return

[Vitaliy Sochnev]

en8811h_clk_provider_setup() is now the last statement in the
function, so the intermediate "if (ret) return ret; return 0;" is
redundant.

Signed-off-by: Vitaliy Sochnev <sochnev.v.74@gmail.com>
---
 drivers/net/phy/air_en8811h.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/drivers/net/phy/air_en8811h.c b/drivers/net/phy/air_en8811h.c
index 0eeb7b9a4e26..8f4498cf0a7c 100644
--- a/drivers/net/phy/air_en8811h.c
+++ b/drivers/net/phy/air_en8811h.c
@@ -1169,11 +1169,7 @@ static int en8811h_probe(struct phy_device *phydev)
 
 	priv->phydev = phydev;
 	/* Co-Clock Output */
-	ret = en8811h_clk_provider_setup(&phydev->mdio.dev, &priv->hw);
-	if (ret)
-		return ret;
-
-	return 0;
+	return en8811h_clk_provider_setup(&phydev->mdio.dev, &priv->hw);
 }
 
 static int an8811hb_config_serdes_polarity(struct phy_device *phydev)
-- 
2.55.0

Re: [PATCH 0/2] net: phy: air_en8811h: fix LED GPIO not surviving MCU restart

[Paolo Abeni]

On 8/23/26 3:06 PM, Vitaliy Sochnev wrote:
> The EN8811H PHY's LED GPIO pins are only ever configured as outputs
> once, in .probe(). Every .config_init() call after the first restarts
> the PHY's MD32 MCU, which resets that GPIO configuration back to
> inputs, so the LEDs silently stop reflecting link/activity state after
> the first link renegotiation, ifdown/ifup, or resume. Patch 1 fixes
> that. Patch 2 is a small follow-on cleanup enabled by patch 1's
> removal of code from .probe(), kept as a separate patch so patch 1
> stays a minimal, clean Fixes: change.

Patch 2/2 is net-next material, please re-submit after the merge window.

/P

Re: [PATCH 0/2] net: phy: air_en8811h: fix LED GPIO not surviving MCU restart

[patchwork-bot+netdevbpf]

Hello:

This series was applied to netdev/net.git (main)
by Paolo Abeni <pabeni@redhat.com>:

On Sun, 23 Aug 2026 14:06:35 +0100 you wrote:
> The EN8811H PHY's LED GPIO pins are only ever configured as outputs
> once, in .probe(). Every .config_init() call after the first restarts
> the PHY's MD32 MCU, which resets that GPIO configuration back to
> inputs, so the LEDs silently stop reflecting link/activity state after
> the first link renegotiation, ifdown/ifup, or resume. Patch 1 fixes
> that. Patch 2 is a small follow-on cleanup enabled by patch 1's
> removal of code from .probe(), kept as a separate patch so patch 1
> stays a minimal, clean Fixes: change.
> 
> [...]

Here is the summary with links:
  - [1/2] net: phy: air_en8811h: move LED GPIO configuration to config_init
    https://git.kernel.org/netdev/net/c/03b4702fc5e3
  - [2/2] net: phy: air_en8811h: simplify en8811h_probe() trailing return
    (no matching commit)

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