[Bug 268176] net/intel-ix-kmod : connection hangs when starting SR-IOV interfaces

https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=268176

            Bug ID: 268176
           Summary: net/intel-ix-kmod : connection hangs when starting
                    SR-IOV interfaces
           Product: Ports & Packages
           Version: Latest
          Hardware: Any
                OS: Any
            Status: New
          Severity: Affects Only Me
          Priority: ---
         Component: Individual Port(s)
          Assignee: ports-bugs@FreeBSD.org
          Reporter: bruno@ruomad.net
                CC: freebsd@intel.com
             Flags: maintainer-feedback?(freebsd@intel.com)
                CC: freebsd@intel.com

Hello,

I tried to setup VF interfaces using SR-IOV in intel-ix-kmod, it kind of works
but as soon as I create the SF interfaces, the connection on the primary
interfaces hangs.
Hardware : X520-DA1
OS : FreeBSD-13.1

I have in my iov config file (/etc/iovctl-ix0.conf) :

<pre>
PF {
    device: ix0;
    num_vfs: 3;
}
DEFAULT {
    passthrough: false;
    allow-set-mac: true;
    allow-promisc: true;
}
VF-1 {
    passthrough: true;
}
VF-2 {
    passthrough: true;
}
</pre>

I start creating VFs with :

<pre>
iovctl -C -f /etc/iovctl-ix0.conf
</pre>

and get one ixv0 plus two ppt (which is expected), but the connection i lost on
ix0.

No messages in log file except initialization messages :
<pre>
Dec  5 16:02:00 amber kernel: ix0: <Intel(R) PRO/10GbE PCI-Express Network
Driver, Version - 3.3.32> port 0x5000-0x501f mem
0x6000000000-0x600007ffff,0x6000080000-0x6000083fff at device 0.0 on pci2
Dec  5 16:02:00 amber kernel: ix0: fw version unknown, nvm version unknown,
etid 80000707
Dec  5 16:02:00 amber kernel: ix0: Using MSI-X interrupts with 9 vectors
Dec  5 16:02:00 amber kernel: ix0: Ethernet address: 80:61:5f:0a:47:2e
Dec  5 16:02:00 amber kernel: ix0: PCI Express Bus: Speed 5.0GT/s Width x8
Dec  5 16:02:00 amber kernel: ppt0 at device 0.130 on pci2
Dec  5 16:02:00 amber kernel: ppt1 at device 0.132 on pci2
</pre>

Any ideas (bug in driver, too cheap hardware ?).

NB : I tried with different versions of the intel driver (compiled manually),
the last one is 3.3.32, but the symptoms remain the same.

Thanks

Bruno

-- 
You are receiving this mail because:
You are the assignee for the bug.