On 16.12.2018 18:02, Florian Fainelli wrote: > Le 12/16/18 à 7:52 AM, Heiner Kallweit a écrit : >> So far phy_error() silently stops the PHY state machine. If the network >> driver doesn't inform about a MDIO error then the user may wonder why >> his network is down. So let's inform the user. >> >> Signed-off-by: Heiner Kallweit <hkallwe...@gmail.com> > > Reviewed-by: Florian Fainelli <f.faine...@gmail.com> > > Kind of similar to the netdev watchdog on a TX timeout, I wonder if we > should not just put a WARN() here to get a complete stack trace to help > debug those events? > But it's good that you mention the WARN(). This is the better option in the other patch I submitted today. I'll prepare an update to this patch.
>> --- >> drivers/net/phy/phy.c | 2 ++ >> 1 file changed, 2 insertions(+) >> >> diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c >> index 890ae1d73..a898fa411 100644 >> --- a/drivers/net/phy/phy.c >> +++ b/drivers/net/phy/phy.c >> @@ -739,6 +739,8 @@ static void phy_error(struct phy_device *phydev) >> phydev->state = PHY_HALTED; >> mutex_unlock(&phydev->lock); >> >> + phydev_err(phydev, "stopping PHY state machine due to error\n"); >> + >> phy_trigger_machine(phydev); >> } >> >> > >
