Re: SR-IOV: IGB/IGBVF initialisation problem

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Thu, Jul 16, 2009 at 06:33:22PM +1000, Simon Horman wrote:
> Sorry, it was me who sent the previous mail, not Jesse Barnes.
> I didn't intend to pretend to be him :-(
> 
> > Hi,
> > 
> > I am having some trouble initialising the virtual functions
> > of an intel 85276 NIC.
> > 
> > The first problem that  I see is that if I insert the igb
> > module with max_vfs > 0 is the following:
> > 
> >     igb 0000:01:00.1: SR-IOV: bus number out of range
> > 
> > I added some debugging messages to the relevant code
> > and it seems that iov wants to add the vf to bus 2,
> > but bus->subordiante is set to 1.

That's interesting.  I also have an 82576 and my VFs show up on the same
bus as my PFs:

09:00.0 Ethernet controller: Intel Corporation 82576 Gigabit Network Connection (rev 01)
09:00.1 Ethernet controller: Intel Corporation 82576 Gigabit Network Connection (rev 01)
09:10.0 Ethernet controller: Intel Corporation 82576 Virtual Function (rev 01)
09:10.1 Ethernet controller: Intel Corporation 82576 Virtual Function (rev 01)
...
09:11.4 Ethernet controller: Intel Corporation 82576 Virtual Function (rev 01)
09:11.5 Ethernet controller: Intel Corporation 82576 Virtual Function (rev 01)

I wonder if maybe the parent you've plugged the card into doesn't
support ARI ... I can try swapping this card into a slot that doesn't
support ARI and see what happens.  In the meantime, you could look at
lspci -vv of the parent PCI bridge and see if its DevCap2: and DevCtl2:
lines are ARIFwd+ or ARIFwd-.

> > If I disable this check (probably unwise, but I wanted to see what would
> > happen), then the virtual function is indeed added on bus 2, and my
> > lspci looks a bit like this:
> > 
> >     01:00.0 Ethernet controller: Intel Corporation Device 10c9 (rev 01)
> >     01:00.1 Ethernet controller: Intel Corporation Device 10c9 (rev 01)
> >     02:10.0 Class ffff: Intel Corporation Device 10ca (rev ff)
> >     02:10.1 Class ffff: Intel Corporation Device 10ca (rev ff)

Possibly you need a newer lspci ...

> > That is, each of the PFs (01:00.0 and 01:00.1) end up with one VF each.
> > I ran this test with max_vfs=1.
> > 
> > However, when I try to bring up one of these VFs using ifconfig,
> > the kernel panics, as per the log below.
> > 
> > Is this a know problem? Is there something peculiar about my hw,
> > its an Intel 85276 plugged into an HP dc7800? Am I doing something wrong?

I'm not terribly surprised to see the panic given the forcing you were
trying to do ;-)

-- 
Matthew Wilcox				Intel Open Source Technology Centre
"Bill, look, we understand that you're interested in selling us this
operating system, but compare it to ours.  We can't possibly take such
a retrograde step."
--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[Index of Archives]     [DMA Engine]     [Linux Coverity]     [Linux USB]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]     [Greybus]

  Powered by Linux