Re: [PATCH] net: phy: add locking to phy_read_mmd_indirect()/phy_write_mmd_indirect()

Re: [PATCH] net: phy: add locking to phy_read_mmd_indirect()/phy_write_mmd_indirect()

[Florian Fainelli]

Le 08/25/15 01:49, Russell King a écrit :
> The phy layer is missing locking for the above two functions - it
> has been observed that two threads (userspace and the phy worker
> thread) can race, entering the bus ->write or ->read functions
> simultaneously.
> 
> This causes the FEC driver to initialise a completion while another
> thread is waiting on it or while the interrupt is calling complete()
> on it, which causes spinlock unlock-without-lock, spinlock lockups,
> and completion timeouts.
> 
> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

Acked-by: Florian Fainelli <f.fainelli@gmail.com>
Fixes: a59a4d192 ("phy: add the EEE support and the way to access to the
MMD registers.")
Fixes: 0c1d77dfb ("net: libphy: Add phy specific function to access mmd
phy registers")

Thanks!

> ---
>  drivers/net/phy/phy.c | 16 ++++++++++++----
>  1 file changed, 12 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c
> index 1e1fbb049ec6..34fe339f4e80 100644
> --- a/drivers/net/phy/phy.c
> +++ b/drivers/net/phy/phy.c
> @@ -1038,10 +1038,14 @@ int phy_read_mmd_indirect(struct phy_device *phydev, int prtad,
>  	int value = -1;
>  
>  	if (phydrv->read_mmd_indirect == NULL) {
> -		mmd_phy_indirect(phydev->bus, prtad, devad, addr);
> +		struct mii_bus *bus = phydev->bus;
> +
> +		mutex_lock(&bus->mdio_lock);
> +		mmd_phy_indirect(bus, prtad, devad, addr);
>  
>  		/* Read the content of the MMD's selected register */
> -		value = phydev->bus->read(phydev->bus, addr, MII_MMD_DATA);
> +		value = bus->read(bus, addr, MII_MMD_DATA);
> +		mutex_unlock(&bus->mdio_lock);
>  	} else {
>  		value = phydrv->read_mmd_indirect(phydev, prtad, devad, addr);
>  	}
> @@ -1071,10 +1075,14 @@ void phy_write_mmd_indirect(struct phy_device *phydev, int prtad,
>  	struct phy_driver *phydrv = phydev->drv;
>  
>  	if (phydrv->write_mmd_indirect == NULL) {
> -		mmd_phy_indirect(phydev->bus, prtad, devad, addr);
> +		struct mii_bus *bus = phydev->bus;
> +
> +		mutex_lock(&bus->mdio_lock);
> +		mmd_phy_indirect(bus, prtad, devad, addr);
>  
>  		/* Write the data into MMD's selected register */
> -		phydev->bus->write(phydev->bus, addr, MII_MMD_DATA, data);
> +		bus->write(bus, addr, MII_MMD_DATA, data);
> +		mutex_unlock(&bus->mdio_lock);
>  	} else {
>  		phydrv->write_mmd_indirect(phydev, prtad, devad, addr, data);
>  	}
> 


-- 
Florian

Re: [PATCH] net: phy: add locking to phy_read_mmd_indirect()/phy_write_mmd_indirect()

[Russell King - ARM Linux]

On Tue, Aug 25, 2015 at 07:08:24AM -0700, Florian Fainelli wrote:
> Le 08/25/15 01:49, Russell King a écrit :
> > The phy layer is missing locking for the above two functions - it
> > has been observed that two threads (userspace and the phy worker
> > thread) can race, entering the bus ->write or ->read functions
> > simultaneously.
> > 
> > This causes the FEC driver to initialise a completion while another
> > thread is waiting on it or while the interrupt is calling complete()
> > on it, which causes spinlock unlock-without-lock, spinlock lockups,
> > and completion timeouts.
> > 
> > Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
> 
> Acked-by: Florian Fainelli <f.fainelli@gmail.com>
> Fixes: a59a4d192 ("phy: add the EEE support and the way to access to the
> MMD registers.")
> Fixes: 0c1d77dfb ("net: libphy: Add phy specific function to access mmd
> phy registers")

Thanks.  I'm assuming that davem will take the patch?

> Thanks!
> 
> > ---
> >  drivers/net/phy/phy.c | 16 ++++++++++++----
> >  1 file changed, 12 insertions(+), 4 deletions(-)
> > 
> > diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c
> > index 1e1fbb049ec6..34fe339f4e80 100644
> > --- a/drivers/net/phy/phy.c
> > +++ b/drivers/net/phy/phy.c
> > @@ -1038,10 +1038,14 @@ int phy_read_mmd_indirect(struct phy_device *phydev, int prtad,
> >  	int value = -1;
> >  
> >  	if (phydrv->read_mmd_indirect == NULL) {
> > -		mmd_phy_indirect(phydev->bus, prtad, devad, addr);
> > +		struct mii_bus *bus = phydev->bus;
> > +
> > +		mutex_lock(&bus->mdio_lock);
> > +		mmd_phy_indirect(bus, prtad, devad, addr);
> >  
> >  		/* Read the content of the MMD's selected register */
> > -		value = phydev->bus->read(phydev->bus, addr, MII_MMD_DATA);
> > +		value = bus->read(bus, addr, MII_MMD_DATA);
> > +		mutex_unlock(&bus->mdio_lock);
> >  	} else {
> >  		value = phydrv->read_mmd_indirect(phydev, prtad, devad, addr);
> >  	}
> > @@ -1071,10 +1075,14 @@ void phy_write_mmd_indirect(struct phy_device *phydev, int prtad,
> >  	struct phy_driver *phydrv = phydev->drv;
> >  
> >  	if (phydrv->write_mmd_indirect == NULL) {
> > -		mmd_phy_indirect(phydev->bus, prtad, devad, addr);
> > +		struct mii_bus *bus = phydev->bus;
> > +
> > +		mutex_lock(&bus->mdio_lock);
> > +		mmd_phy_indirect(bus, prtad, devad, addr);
> >  
> >  		/* Write the data into MMD's selected register */
> > -		phydev->bus->write(phydev->bus, addr, MII_MMD_DATA, data);
> > +		bus->write(bus, addr, MII_MMD_DATA, data);
> > +		mutex_unlock(&bus->mdio_lock);
> >  	} else {
> >  		phydrv->write_mmd_indirect(phydev, prtad, devad, addr, data);
> >  	}
> > 
> 
> 
> -- 
> Florian

-- 
FTTC broadband for 0.8mile line: currently at 10.5Mbps down 400kbps up
according to speedtest.net.

Re: [PATCH] net: phy: add locking to phy_read_mmd_indirect()/phy_write_mmd_indirect()

[David Miller]

From: Florian Fainelli <f.fainelli@gmail.com>
Date: Tue, 25 Aug 2015 07:08:24 -0700

> Le 08/25/15 01:49, Russell King a écrit :
>> The phy layer is missing locking for the above two functions - it
>> has been observed that two threads (userspace and the phy worker
>> thread) can race, entering the bus ->write or ->read functions
>> simultaneously.
>> 
>> This causes the FEC driver to initialise a completion while another
>> thread is waiting on it or while the interrupt is calling complete()
>> on it, which causes spinlock unlock-without-lock, spinlock lockups,
>> and completion timeouts.
>> 
>> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
> 
> Acked-by: Florian Fainelli <f.fainelli@gmail.com>
> Fixes: a59a4d192 ("phy: add the EEE support and the way to access to the
> MMD registers.")
> Fixes: 0c1d77dfb ("net: libphy: Add phy specific function to access mmd
> phy registers")

Applied, thanks.