On 2017/6/24 11:40, Yunsheng Lin wrote:
> Hi, Andrew
> 
> On 2017/6/24 11:12, Andrew Lunn wrote:
>>> +int phy_loopback(struct phy_device *phydev, bool enable)
>>> +{
>>> +   struct phy_driver *phydrv = to_phy_driver(phydev->mdio.dev.driver);
>>> +   int ret = 0;
>>> +
>>> +   if (enable && phydev->loopback_enabled)
>>> +           return -EBUSY;
>>> +
>>> +   if (!enable && !phydev->loopback_enabled)
>>> +           return -EINVAL;
>>> +
>>> +   if (phydev->drv && phydrv->set_loopback)
>>> +           ret = phydrv->set_loopback(phydev, enable);
>>
>>      else
>>              ret = -EOPNOTSUPP;
>>
>>> +
>>> +   if (ret)
>>> +           return ret;
>>> +
>>> +   phydev->loopback_enabled = enable;
>>> +
>>> +   return 0;
>>> +}
>>> +EXPORT_SYMBOL(phy_loopback);
>>
>> One of the comments we made of the PHY code in the hns driver is that
>> its locking is completely broken. You have made the same error
>> here. The core needs to hold the mutex while calling into the PHY
>> driver.
> Do you mean hns_nic_config_phy_loopback need to hold the mutex while
> calling phy_loopback? and other place that calling phy_* function?
I took some time looking into how to take mutex in phy core, here is what
I find:
phy_resume call phydrv->resume without take mutex.
if phy driver implement resume function, for example marvell_resume, then

static int marvell_resume(struct phy_device *phydev)
{
        int err;

        /* Resume the fiber mode first */
        if (!(phydev->supported & SUPPORTED_FIBRE)) {
                err = phy_write(phydev, MII_MARVELL_PHY_PAGE, MII_M1111_FIBER);
                if (err < 0)
                        goto error;

                /* With the page set, use the generic resume */
                err = genphy_resume(phydev);
                if (err < 0)
                        goto error;

                /* Then, the copper link */
                err = phy_write(phydev, MII_MARVELL_PHY_PAGE, MII_M1111_COPPER);
                if (err < 0)
                        goto error;
        }

the code above executes without holding a mutex expect genphy_resume

        /* With the page set, use the generic resume */
        return genphy_resume(phydev);

error:
        phy_write(phydev, MII_MARVELL_PHY_PAGE, MII_M1111_COPPER);
        return err;
}

So I think the correct way to hold a lock is in phy_* function, not in genphy_*,
and current genphy_resume and phy_resume is broken in this way.
Please let me know if I misunderstand the mutex taking in phy core.

Best Regards
Yunsheng Lin

Reply via email to