On 12/16/2015 01:17 PM, Laine Stump wrote:

On 12/16/2015 12:35 PM, Vlad Yasevich wrote:

(BTW, Cisco's enic driver, on the other hand, doesn't support setting VF MAC addresses via
a netlink message to the PF *at all* (so libvirt has to make special accommodations), but

Looking at upstream, it looks like it offers support for setting VF mac via VFINFO data in
the netlink message. May be it got fixed?

Interesting. If I had one of those systems of my own to test on, I'd give it a try. The only one I have access to is running a 2.6.32 RHEL6 kernel though :-/

It also occurred to me during my copious "time to think" over the holidays that device assignment using <interface type='hostdev'> wouldn't work for enic cards if that wasn't supported, and I recently saw a bug filed where someone was doing exactly that on a 3.10 kernel (RHEL7.x), so yeah it must have been fixed between 2.6 and 3.10).

To pick up this discussion again: Moshe - am I correct to assume that you've filed a bug somewhere about the mlx driver not accepting 00:00:00:00:00:00?

In the meantime, if setting the MAC address to, e.g., 00:00:00:00:00:01 works for you (as implied when you said you use "ip link set dev p4p2 vf 0 mac 0") then you could temporarily work around the problem by setting all the VFs to something like that at boot time - libvirt would then succeed at setting them *back* to those addresses when it was finished with each device.