[PATCH net-next] net: phy: icplus: add config_intr callback

[PATCH net-next] net: phy: icplus: add config_intr callback

[Heiner Kallweit]

Move IRQ configuration for IP101A/G from config_init to config_intr
callback. Reasons:

1. This allows phylib to disable interrupts if needed.
2. Icplus was the only driver supporting interrupts w/o defining a
   config_intr callback. Now we can add a phylib plausibility check
   disabling interrupt mode if one of the two irq-related callbacks
   isn't defined.

I don't own hardware with this PHY, and the change is based on the
datasheet for IP101A LF (which is supposed to be register-compatible
with IP101A/G). Change is compile-tested only.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
---
 drivers/net/phy/icplus.c | 23 ++++++++++++++++-------
 1 file changed, 16 insertions(+), 7 deletions(-)

diff --git a/drivers/net/phy/icplus.c b/drivers/net/phy/icplus.c
index 21ce68964..ad87bd328 100644
--- a/drivers/net/phy/icplus.c
+++ b/drivers/net/phy/icplus.c
@@ -42,8 +42,8 @@ MODULE_LICENSE("GPL");
 #define IP1001_APS_ON			11	/* IP1001 APS Mode  bit */
 #define IP101A_G_APS_ON			2	/* IP101A/G APS Mode bit */
 #define IP101A_G_IRQ_CONF_STATUS	0x11	/* Conf Info IRQ & Status Reg */
-#define	IP101A_G_IRQ_PIN_USED		(1<<15) /* INTR pin used */
-#define	IP101A_G_IRQ_DEFAULT		IP101A_G_IRQ_PIN_USED
+#define	IP101A_G_IRQ_PIN_USED		BIT(15) /* INTR pin used */
+#define	IP101A_G_NO_IRQ			BIT(11) /* IRQ's inactive */
 
 static int ip175c_config_init(struct phy_device *phydev)
 {
@@ -170,11 +170,6 @@ static int ip101a_g_config_init(struct phy_device *phydev)
 	if (c < 0)
 		return c;
 
-	/* INTR pin used: speed/link/duplex will cause an interrupt */
-	c = phy_write(phydev, IP101A_G_IRQ_CONF_STATUS, IP101A_G_IRQ_DEFAULT);
-	if (c < 0)
-		return c;
-
 	/* Enable Auto Power Saving mode */
 	c = phy_read(phydev, IP10XX_SPEC_CTRL_STATUS);
 	c |= IP101A_G_APS_ON;
@@ -201,6 +196,19 @@ static int ip175c_config_aneg(struct phy_device *phydev)
 	return 0;
 }
 
+static int ip101a_g_config_intr(struct phy_device *phydev)
+{
+	u16 val;
+
+	if (phydev->interrupts == PHY_INTERRUPT_ENABLED)
+		/* INTR pin used: Speed/link/duplex will cause an interrupt */
+		val = IP101A_G_IRQ_PIN_USED;
+	else
+		val = IP101A_G_NO_IRQ;
+
+	return phy_write(phydev, IP101A_G_IRQ_CONF_STATUS, val);
+}
+
 static int ip101a_g_ack_interrupt(struct phy_device *phydev)
 {
 	int err = phy_read(phydev, IP101A_G_IRQ_CONF_STATUS);
@@ -234,6 +242,7 @@ static struct phy_driver icplus_driver[] = {
 	.name		= "ICPlus IP101A/G",
 	.phy_id_mask	= 0x0ffffff0,
 	.features	= PHY_BASIC_FEATURES,
+	.config_intr	= ip101a_g_config_intr,
 	.ack_interrupt	= ip101a_g_ack_interrupt,
 	.config_init	= &ip101a_g_config_init,
 	.suspend	= genphy_suspend,
-- 
2.19.1

Re: [PATCH net-next] net: phy: icplus: add config_intr callback

[Andrew Lunn]

On Sun, Nov 11, 2018 at 09:49:12PM +0100, Heiner Kallweit wrote:
> Move IRQ configuration for IP101A/G from config_init to config_intr
> callback. Reasons:
> 
> 1. This allows phylib to disable interrupts if needed.
> 2. Icplus was the only driver supporting interrupts w/o defining a
>    config_intr callback. Now we can add a phylib plausibility check
>    disabling interrupt mode if one of the two irq-related callbacks
>    isn't defined.
> 
> I don't own hardware with this PHY, and the change is based on the
> datasheet for IP101A LF (which is supposed to be register-compatible
> with IP101A/G). Change is compile-tested only.

Hi Heiner

This looks sensible. Thanks for fixing up this driver.

Reviewed-by: Andrew Lunn <andrew@lunn.ch>

    Andrew

Re: [PATCH net-next] net: phy: icplus: add config_intr callback

[Florian Fainelli]

On 11/11/18 12:49 PM, Heiner Kallweit wrote:
> Move IRQ configuration for IP101A/G from config_init to config_intr
> callback. Reasons:
> 
> 1. This allows phylib to disable interrupts if needed.
> 2. Icplus was the only driver supporting interrupts w/o defining a
>    config_intr callback. Now we can add a phylib plausibility check
>    disabling interrupt mode if one of the two irq-related callbacks
>    isn't defined.
> 
> I don't own hardware with this PHY, and the change is based on the
> datasheet for IP101A LF (which is supposed to be register-compatible
> with IP101A/G). Change is compile-tested only.
> 
> Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>

Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>

A bit surprising that this is not a read/modify/write sequence.
-- 
Florian