On Sun, 2017-04-02 at 20:42 +0200, Andrew Lunn wrote: > Have you run lockdep tests on this? I think it is normal to take the > rtnl lock first, then the phydev lock. Try some ethtool operations > and see if you get a splat.
You are right. I actually realized that was wrong last night ;-) It will splat with the order in open/close. Just a stupid thinko (I added the phy lock in a second pass). I'll fix that. > Also, do you need to be worried about mdio operations? Do you need to > take that lock as well, since there might be other PHYs, or an > Ethernet switch on the bus? There could be I suppose, none of the setups I've seen with those SoC has done it but nothing prevents it, I'll have a look. Thanks. Ben.
