[PATCH net-next v2] net: phy: micrel: Make the GPIO to be non-exclusive

[PATCH net-next v2] net: phy: micrel: Make the GPIO to be non-exclusive

[Horatiu Vultur]

The same GPIO line can be shared by multiple phys for the coma mode pin.
If that is the case then, all the other phys that share the same line
will failed to be probed because the access to the gpio line is not
non-exclusive.
Fix this by making access to the gpio line to be nonexclusive using flag
GPIOD_FLAGS_BIT_NONEXCLUSIVE. This allows all the other PHYs to be
probed.

Fixes: 738871b09250ee ("net: phy: micrel: add coma mode GPIO")
Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
---
v1->v2:
- add comment describing that there should not be anyone putting back
  the phy in coma mode.
---
 drivers/net/phy/micrel.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c
index e78d0bf69bc3..6f52b4fb6888 100644
--- a/drivers/net/phy/micrel.c
+++ b/drivers/net/phy/micrel.c
@@ -2873,12 +2873,18 @@ static int lan8814_config_init(struct phy_device *phydev)
 	return 0;
 }
 
+/* It is expected that there will not be any 'lan8814_take_coma_mode'
+ * function called in suspend. Because the GPIO line can be shared, so if one of
+ * the phys goes back in coma mode, then all the other PHYs will go, which is
+ * wrong.
+ */
 static int lan8814_release_coma_mode(struct phy_device *phydev)
 {
 	struct gpio_desc *gpiod;
 
 	gpiod = devm_gpiod_get_optional(&phydev->mdio.dev, "coma-mode",
-					GPIOD_OUT_HIGH_OPEN_DRAIN);
+					GPIOD_OUT_HIGH_OPEN_DRAIN |
+					GPIOD_FLAGS_BIT_NONEXCLUSIVE);
 	if (IS_ERR(gpiod))
 		return PTR_ERR(gpiod);
 
-- 
2.33.0

Re: [PATCH net-next v2] net: phy: micrel: Make the GPIO to be non-exclusive

[Andrew Lunn]

On Thu, Aug 25, 2022 at 10:14:47PM +0200, Horatiu Vultur wrote:
> The same GPIO line can be shared by multiple phys for the coma mode pin.
> If that is the case then, all the other phys that share the same line
> will failed to be probed because the access to the gpio line is not
> non-exclusive.
> Fix this by making access to the gpio line to be nonexclusive using flag
> GPIOD_FLAGS_BIT_NONEXCLUSIVE. This allows all the other PHYs to be
> probed.
> 
> Fixes: 738871b09250ee ("net: phy: micrel: add coma mode GPIO")
> Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>

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

    Andrew

Re: [PATCH net-next v2] net: phy: micrel: Make the GPIO to be non-exclusive

[Jakub Kicinski]

On Thu, 25 Aug 2022 22:16:31 +0200 Andrew Lunn wrote:
> On Thu, Aug 25, 2022 at 10:14:47PM +0200, Horatiu Vultur wrote:
> > The same GPIO line can be shared by multiple phys for the coma mode pin.
> > If that is the case then, all the other phys that share the same line
> > will failed to be probed because the access to the gpio line is not
> > non-exclusive.
> > Fix this by making access to the gpio line to be nonexclusive using flag
> > GPIOD_FLAGS_BIT_NONEXCLUSIVE. This allows all the other PHYs to be
> > probed.
> > 
> > Fixes: 738871b09250ee ("net: phy: micrel: add coma mode GPIO")
> > Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>  
> 
> Reviewed-by: Andrew Lunn <andrew@lunn.ch>

The tree name switch in the subject compared to v1 is unintentional?

Re: [PATCH net-next v2] net: phy: micrel: Make the GPIO to be non-exclusive

[Horatiu Vultur]

The 08/26/2022 18:37, Jakub Kicinski wrote:

Hi Jakub,

> 
> On Thu, 25 Aug 2022 22:16:31 +0200 Andrew Lunn wrote:
> > On Thu, Aug 25, 2022 at 10:14:47PM +0200, Horatiu Vultur wrote:
> > > The same GPIO line can be shared by multiple phys for the coma mode pin.
> > > If that is the case then, all the other phys that share the same line
> > > will failed to be probed because the access to the gpio line is not
> > > non-exclusive.
> > > Fix this by making access to the gpio line to be nonexclusive using flag
> > > GPIOD_FLAGS_BIT_NONEXCLUSIVE. This allows all the other PHYs to be
> > > probed.
> > >
> > > Fixes: 738871b09250ee ("net: phy: micrel: add coma mode GPIO")
> > > Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
> >
> > Reviewed-by: Andrew Lunn <andrew@lunn.ch>
> 
> The tree name switch in the subject compared to v1 is unintentional?

Yes I have switch it by mistake.
It should be net. I can send a new version if it is needed.

-- 
/Horatiu