Message ID | E1qgoNZ-007a4I-W1@rmk-PC.armlinux.org.uk (mailing list archive) |
---|---|
State | Accepted |
Commit | ea5968cd7d6e6f237cfc6d00aa29ff1dc9230b2f |
Delegated to: | Netdev Maintainers |
Headers | show |
Series | net: phy: avoid race when erroring stopping PHY | expand |
On 9/14/23 08:35, Russell King (Oracle) wrote: > Move the call to start auto-negotiation inside the lock in the PHYLIB > state machine, calling the locked variant _phy_start_aneg(). This > avoids unnecessarily releasing and re-acquiring the lock. > > Tested-by: Jijie Shao <shaojijie@huawei.com> > Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c index 990d387b31bd..5bb33af2a4cb 100644 --- a/drivers/net/phy/phy.c +++ b/drivers/net/phy/phy.c @@ -1489,14 +1489,15 @@ void phy_state_machine(struct work_struct *work) break; } + if (needs_aneg) { + err = _phy_start_aneg(phydev); + func = &_phy_start_aneg; + } + mutex_unlock(&phydev->lock); - if (needs_aneg) { - err = phy_start_aneg(phydev); - func = &phy_start_aneg; - } else if (do_suspend) { + if (do_suspend) phy_suspend(phydev); - } if (err == -ENODEV) return;