[PATCH net v2] net: phy: fix phy_start to consider PHY_IGNORE_INTERRUPT

This condition wasn't adjusted when PHY_IGNORE_INTERRUPT (-2) was added
long ago. In case of PHY_IGNORE_INTERRUPT the MAC interrupt indicates
also PHY state changes and we should do what the symbol says.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
---
v2:
- use phy_interrupt_is_valid() instead of checking for irq > 0
---
 drivers/net/phy/phy.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c
index f3313a129..50ed35a45 100644
--- a/drivers/net/phy/phy.c
+++ b/drivers/net/phy/phy.c

@@ -822,7 +822,7 @@ void phy_start(struct phy_device *phydev)
 		phy_resume(phydev);
 
 		/* make sure interrupts are re-enabled for the PHY */
-		if (phydev->irq != PHY_POLL) {
+		if (phy_interrupt_is_valid(phydev)) {
 			err = phy_enable_interrupts(phydev);
 			if (err < 0)
 				break;

-- 
2.16.1

On 02/07/2018 11:44 AM, Heiner Kallweit wrote:

This condition wasn't adjusted when PHY_IGNORE_INTERRUPT (-2) was added
long ago. In case of PHY_IGNORE_INTERRUPT the MAC interrupt indicates
also PHY state changes and we should do what the symbol says.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
---
v2:
- use phy_interrupt_is_valid() instead of checking for irq > 0

Thanks, could you identify which Fixes: tag we should be using for that?
It would be great to see this backported to -stable

-- 
Florian

Am 08.02.2018 um 00:00 schrieb Florian Fainelli:

On 02/07/2018 11:44 AM, Heiner Kallweit wrote:

quoted

This condition wasn't adjusted when PHY_IGNORE_INTERRUPT (-2) was added
long ago. In case of PHY_IGNORE_INTERRUPT the MAC interrupt indicates
also PHY state changes and we should do what the symbol says.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
---
v2:
- use phy_interrupt_is_valid() instead of checking for irq > 0

Thanks, could you identify which Fixes: tag we should be using for that?
It would be great to see this backported to -stable

Before submitting the change I had a look at the phylib history for
when the wrong behavior was introduced. However it has been there since
introduction of phylib in 2005. Different patches fixed this in few,
but not all places.
Latest one was 84a527a41f38 "net: phylib: fix interrupts re-enablement
in phy_start" which just partially fixed the issue. So we could
declare the change now to fix this fix.

quoted

---
 drivers/net/phy/phy.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c
index f3313a129..50ed35a45 100644
--- a/drivers/net/phy/phy.c
+++ b/drivers/net/phy/phy.c

@@ -822,7 +822,7 @@ void phy_start(struct phy_device *phydev)
 		phy_resume(phydev);
 
 		/* make sure interrupts are re-enabled for the PHY */
-		if (phydev->irq != PHY_POLL) {
+		if (phy_interrupt_is_valid(phydev)) {
 			err = phy_enable_interrupts(phydev);
 			if (err < 0)
 				break;