netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* igb VF allocation with quirk_i82576_sriov
@ 2009-09-22  5:19 Chris Wright
  2009-09-22 15:41 ` Alexander Duyck
  0 siblings, 1 reply; 3+ messages in thread
From: Chris Wright @ 2009-09-22  5:19 UTC (permalink / raw)
  To: John Ronciak; +Cc: netdev, e1000-devel

Is this known to work?  During recent virt testing for upcoming Fedora 12,
a box w/out SR-IOV support in BIOS was using quirk to create VF BAR space,
VF allocation worked enough to assign a device to the guest, but igbvf
was not actually functioning properly in the guest.

Is it worth debugging this further, or is it already a known issue?

thanks,
-chris

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: igb VF allocation with quirk_i82576_sriov
  2009-09-22  5:19 igb VF allocation with quirk_i82576_sriov Chris Wright
@ 2009-09-22 15:41 ` Alexander Duyck
  2009-09-23  5:12   ` Chris Wright
  0 siblings, 1 reply; 3+ messages in thread
From: Alexander Duyck @ 2009-09-22 15:41 UTC (permalink / raw)
  To: Chris Wright
  Cc: e1000-devel@lists.sourceforge.net, netdev@vger.kernel.org,
	Ronciak, John

Chris Wright wrote:
> Is this known to work?  During recent virt testing for upcoming Fedora 12,
> a box w/out SR-IOV support in BIOS was using quirk to create VF BAR space,
> VF allocation worked enough to assign a device to the guest, but igbvf
> was not actually functioning properly in the guest.
> 
> Is it worth debugging this further, or is it already a known issue?

You could be experiencing one of a couple different issues.

First when you say you started SR-IOV on a box w/out SR-IOV support I 
assume you are using "pci=assign-busses" in order to reserve the bus 
space for the VFs, is that correct?  Also while your system may not 
support SR-IOV does it at least support VT-d?  Without VT-d support you 
won't be able to assign a device to the guest.

My recommendations for further testing would be to test a VF on the host 
kernel to see if that works.  If it does then you could also try direct 
assigning an entire port to see if that works.  If the entire port 
doesn't work then you probably don't have VT-d enabled.

Thanks,

Alex

------------------------------------------------------------------------------
Come build with us! The BlackBerry® Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay 
ahead of the curve. Join us from November 9-12, 2009. Register now!
http://p.sf.net/sfu/devconf

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: igb VF allocation with quirk_i82576_sriov
  2009-09-22 15:41 ` Alexander Duyck
@ 2009-09-23  5:12   ` Chris Wright
  0 siblings, 0 replies; 3+ messages in thread
From: Chris Wright @ 2009-09-23  5:12 UTC (permalink / raw)
  To: Alexander Duyck
  Cc: Chris Wright, netdev@vger.kernel.org, Ronciak, John,
	e1000-devel@lists.sourceforge.net

* Alexander Duyck (alexander.h.duyck@intel.com) wrote:
> Chris Wright wrote:
>> Is this known to work?  During recent virt testing for upcoming Fedora 12,
>> a box w/out SR-IOV support in BIOS was using quirk to create VF BAR space,
>> VF allocation worked enough to assign a device to the guest, but igbvf
>> was not actually functioning properly in the guest.
>>
>> Is it worth debugging this further, or is it already a known issue?
>
> You could be experiencing one of a couple different issues.
>
> First when you say you started SR-IOV on a box w/out SR-IOV support I  
> assume you are using "pci=assign-busses" in order to reserve the bus  
> space for the VFs, is that correct?  Also while your system may not  
> support SR-IOV does it at least support VT-d?  Without VT-d support you  
> won't be able to assign a device to the guest.

VT-d was definitely there, as for the rest I'll have to ask the tester
for more details.  I just wanted to verify that it's a known working
combo before spending more time on it.

Regarding the bus numbering, I don't think there's a bus issue.
The PF+VFs all stay w/in same bus segment despite large offset and the
stride (IIRC, this was only device on bus 2, a dual port igb on .0 and .1.
the offset is 128 and stride is 2, so even w/ 8 VFs the max device would
be something like 2:11.7 or 2:12.0).

> My recommendations for further testing would be to test a VF on the host  
> kernel to see if that works.  If it does then you could also try direct  
> assigning an entire port to see if that works.  If the entire port  
> doesn't work then you probably don't have VT-d enabled.

Yeah, IIRC, igbvf at least loaded on the host (on the guest too, after
unbinding host driver).  I didn't get a chance to see if VF passed
traffic on the host, and from the report, it wasn't able to get a dhcp
address in the guest.  Will dig into it a bit more after plumbers.

thanks,
-chris

------------------------------------------------------------------------------
Come build with us! The BlackBerry® Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay 
ahead of the curve. Join us from November 9-12, 2009. Register now!
http://p.sf.net/sfu/devconf

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2009-09-23  5:12 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2009-09-22  5:19 igb VF allocation with quirk_i82576_sriov Chris Wright
2009-09-22 15:41 ` Alexander Duyck
2009-09-23  5:12   ` Chris Wright

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).