libvirt users - Feb 2014 - SR-IOV: no traffic isolation between VFs with Broadcom 10Gbps cards

Hi all,

I'm testing on debian/unstable SR-IOV feature with Broadcom BCM57810 
cards and KVM hypervisor:

Compiled against library: libvirt 1.2.1
Using library: libvirt 1.2.1
Using API: QEMU 1.2.1
Running hypervisor: QEMU 1.7.0

bnx2x
-> firmware 7.8.17
-> driver from kernel 3.12.7

8 VFs are created on the first PF. For each VF, a specific mac address 
is set manually using "ip link set eth0 vf x mac xx:xx:xx:xx:xx" 
command. I run several KVM guests with PCI passthrough (same kernel, 
bnx2x driver and firmware as the host), performance is close to bare metal.

Well, that sounds good, until I start capturing the traffic inside each 
VM: host traffic is visible as well as traffic destined to other VM. 
It's like if internal card switching was inoperable. I made several 
tests with different kernels, different PCIe Passthrough method 
assignments for libvirt. All failed.

Has anyone successfully experiment SR-IOV with Broadcom cards on linux ?

-----

Some details:

01:00.0 Ethernet controller: Broadcom Corporation NetXtreme II BCM57810 
10 Gigabit Ethernet (rev 10)
01:00.1 Ethernet controller: Broadcom Corporation NetXtreme II BCM57810 
10 Gigabit Ethernet (rev 10)

01:09.0 Ethernet controller: Broadcom Corporation NetXtreme II BCM57810 
10 Gigabit Ethernet Virtual Function
01:09.1 Ethernet controller: Broadcom Corporation NetXtreme II BCM57810 
10 Gigabit Ethernet Virtual Function
01:09.2 Ethernet controller: Broadcom Corporation NetXtreme II BCM57810 
10 Gigabit Ethernet Virtual Function
01:09.3 Ethernet controller: Broadcom Corporation NetXtreme II BCM57810 
10 Gigabit Ethernet Virtual Function
01:09.4 Ethernet controller: Broadcom Corporation NetXtreme II BCM57810 
10 Gigabit Ethernet Virtual Function
01:09.5 Ethernet controller: Broadcom Corporation NetXtreme II BCM57810 
10 Gigabit Ethernet Virtual Function
01:09.6 Ethernet controller: Broadcom Corporation NetXtreme II BCM57810 
10 Gigabit Ethernet Virtual Function
01:09.7 Ethernet controller: Broadcom Corporation NetXtreme II BCM57810 
10 Gigabit Ethernet Virtual Function


# virsh nodedev-dumpxml pci_0000_01_09_0
<device>
   <name>pci_0000_01_09_0</name>
   <path>/sys/devices/pci0000:00/0000:00:01.0/0000:01:09.0</path>
   <parent>pci_0000_00_01_0</parent>
   <driver>
     <name>vfio-pci</name>
   </driver>
   <capability type='pci'>
     <domain>0</domain>
     <bus>1</bus>
     <slot>9</slot>
     <function>0</function>
     <product id='0x16af'>NetXtreme II BCM57810 10 Gigabit
Ethernet
Virtual Function</product>
     <vendor id='0x14e4'>Broadcom Corporation</vendor>
     <capability type='phys_function'>
       <address domain='0x0000' bus='0x01'
slot='0x00' function='0x1'/>
     </capability>
     <iommuGroup number='35'>
       <address domain='0x0000' bus='0x01'
slot='0x09' function='0x0'/>
     </iommuGroup>
   </capability>
</device>


# virsh nodedev-dumpxml pci_0000_01_09_1
<device>
   <name>pci_0000_01_09_1</name>
   <path>/sys/devices/pci0000:00/0000:00:01.0/0000:01:09.1</path>
   <parent>pci_0000_00_01_0</parent>
   <driver>
     <name>vfio-pci</name>
   </driver>
   <capability type='pci'>
     <domain>0</domain>
     <bus>1</bus>
     <slot>9</slot>
     <function>1</function>
     <product id='0x16af'>NetXtreme II BCM57810 10 Gigabit
Ethernet
Virtual Function</product>
     <vendor id='0x14e4'>Broadcom Corporation</vendor>
     <capability type='phys_function'>
       <address domain='0x0000' bus='0x01'
slot='0x00' function='0x1'/>
     </capability>
     <iommuGroup number='36'>
       <address domain='0x0000' bus='0x01'
slot='0x09' function='0x1'/>
     </iommuGroup>
   </capability>
</device>


Guest A XML:
	...
     <hostdev mode='subsystem' type='pci'
managed='yes'>
       <source>
         <address domain='0x0000' bus='0x01'
slot='0x09' function='0x0'/>
       </source>
     </hostdev>
	...


Guest B XML:

	...
     <hostdev mode='subsystem' type='pci'
managed='yes'>
       <source>
         <address domain='0x0000' bus='0x01'
slot='0x09' function='0x1'/>
       </source>
     </hostdev>
	...


-- 
Université de Nantes - Direction des Systèmes d'Information

On 02/04/2014 05:10 PM, Yoann Juet wrote:
> Hi all,
>
> I'm testing on debian/unstable SR-IOV feature with Broadcom BCM57810
> cards and KVM hypervisor:
>
> Compiled against library: libvirt 1.2.1
> Using library: libvirt 1.2.1
> Using API: QEMU 1.2.1
> Running hypervisor: QEMU 1.7.0
>
> bnx2x
> -> firmware 7.8.17
> -> driver from kernel 3.12.7
>
> 8 VFs are created on the first PF. For each VF, a specific mac address
> is set manually using "ip link set eth0 vf x mac xx:xx:xx:xx:xx"
command.
Instead of using <hostdev>, you should instead try using <interface
type='hostdev'>, which will allow you to specify the mac address for
the
interface directly in the guest's XML config (rather than needing to do
it separately). Here's a link to documentation on this feature:

 
http://wiki.libvirt.org/page/Networking#PCI_Passthrough_of_host_network_devices

(look down to the section titled "Assignment with <interface
type='hostdev'>")

Or even better, use <interface type='network'> in your guest
config
(still put the <mac address='xx:xx:xx:xx:xx:xx'/> element in each
one),
and define a libvirt network which is a pool of SRIOV VFs - this is
described further down the same page.

This will not make a difference to the issue you describe below, but it
should make managing your guest config and lifecycle much simpler.

> I run several KVM guests with PCI passthrough (same kernel, bnx2x
> driver and firmware as the host), performance is close to bare metal.
>
> Well, that sounds good, until I start capturing the traffic inside
> each VM: host traffic is visible as well as traffic destined to other
> VM. It's like if internal card switching was inoperable. I made
> several tests with different kernels, different PCIe Passthrough
> method assignments for libvirt. All failed.
Define "failed". Do you mean that the cards communicated, but the
guests
can see each others' traffic? Or do you mean that they see traffic from
each other, but can't seem to communicate normally?

If the problem is the latter, then make sure the PF (eth0 for you, I
guess) has status UP and RUNNING before you start the guests.

For the former, I'm not clear on the internal rules of switching of an
SRIOV card. I think in most cases, the SRIOV card's internal switch may
need to make everything from each VF visible to all other VFs, because
the physical switch it's connected to may not mirror back traffic that
really does need to go from one guest to the other. 802.1Qbh (which
libvirt supports via the <virtualport type='802.1Qbh'> element)
does
this differently, requiring all traffic to travel out to the switch,
with the switch making the decision about what gets mirrored back, but
you need an 802.1Qbh-capable switch for that.
>
> Has anyone successfully experiment SR-IOV with Broadcom cards on linux ?
>
> -----
>
> Some details:
>
> 01:00.0 Ethernet controller: Broadcom Corporation NetXtreme II
> BCM57810 10 Gigabit Ethernet (rev 10)
> 01:00.1 Ethernet controller: Broadcom Corporation NetXtreme II
> BCM57810 10 Gigabit Ethernet (rev 10)
>
> 01:09.0 Ethernet controller: Broadcom Corporation NetXtreme II
> BCM57810 10 Gigabit Ethernet Virtual Function
> 01:09.1 Ethernet controller: Broadcom Corporation NetXtreme II
> BCM57810 10 Gigabit Ethernet Virtual Function
> 01:09.2 Ethernet controller: Broadcom Corporation NetXtreme II
> BCM57810 10 Gigabit Ethernet Virtual Function
> 01:09.3 Ethernet controller: Broadcom Corporation NetXtreme II
> BCM57810 10 Gigabit Ethernet Virtual Function
> 01:09.4 Ethernet controller: Broadcom Corporation NetXtreme II
> BCM57810 10 Gigabit Ethernet Virtual Function
> 01:09.5 Ethernet controller: Broadcom Corporation NetXtreme II
> BCM57810 10 Gigabit Ethernet Virtual Function
> 01:09.6 Ethernet controller: Broadcom Corporation NetXtreme II
> BCM57810 10 Gigabit Ethernet Virtual Function
> 01:09.7 Ethernet controller: Broadcom Corporation NetXtreme II
> BCM57810 10 Gigabit Ethernet Virtual Function
>
>
> # virsh nodedev-dumpxml pci_0000_01_09_0
> <device>
>   <name>pci_0000_01_09_0</name>
>  
<path>/sys/devices/pci0000:00/0000:00:01.0/0000:01:09.0</path>
>   <parent>pci_0000_00_01_0</parent>
>   <driver>
>     <name>vfio-pci</name>
>   </driver>
>   <capability type='pci'>
>     <domain>0</domain>
>     <bus>1</bus>
>     <slot>9</slot>
>     <function>0</function>
>     <product id='0x16af'>NetXtreme II BCM57810 10 Gigabit
Ethernet
> Virtual Function</product>
>     <vendor id='0x14e4'>Broadcom Corporation</vendor>
>     <capability type='phys_function'>
>       <address domain='0x0000' bus='0x01'
slot='0x00' function='0x1'/>
>     </capability>
>     <iommuGroup number='35'>
>       <address domain='0x0000' bus='0x01'
slot='0x09' function='0x0'/>
>     </iommuGroup>
>   </capability>
> </device>
>
>
> # virsh nodedev-dumpxml pci_0000_01_09_1
> <device>
>   <name>pci_0000_01_09_1</name>
>  
<path>/sys/devices/pci0000:00/0000:00:01.0/0000:01:09.1</path>
>   <parent>pci_0000_00_01_0</parent>
>   <driver>
>     <name>vfio-pci</name>
>   </driver>
>   <capability type='pci'>
>     <domain>0</domain>
>     <bus>1</bus>
>     <slot>9</slot>
>     <function>1</function>
>     <product id='0x16af'>NetXtreme II BCM57810 10 Gigabit
Ethernet
> Virtual Function</product>
>     <vendor id='0x14e4'>Broadcom Corporation</vendor>
>     <capability type='phys_function'>
>       <address domain='0x0000' bus='0x01'
slot='0x00' function='0x1'/>
>     </capability>
>     <iommuGroup number='36'>
>       <address domain='0x0000' bus='0x01'
slot='0x09' function='0x1'/>
>     </iommuGroup>
>   </capability>
> </device>
>
>
> Guest A XML:
>     ...
>     <hostdev mode='subsystem' type='pci'
managed='yes'>
>       <source>
>         <address domain='0x0000' bus='0x01'
slot='0x09' function='0x0'/>
>       </source>
>     </hostdev>
>     ...
>
>
> Guest B XML:
>
>     ...
>     <hostdev mode='subsystem' type='pci'
managed='yes'>
>       <source>
>         <address domain='0x0000' bus='0x01'
slot='0x09' function='0x1'/>
>       </source>
>     </hostdev>
>     ...
>
>
>
>
> _______________________________________________
> libvirt-users mailing list
> libvirt-users@redhat.com
> https://www.redhat.com/mailman/listinfo/libvirt-users

> Instead of using <hostdev>, you should instead try using
<interface
> type='hostdev'>, which will allow you to specify the mac address
for the
> interface directly in the guest's XML config (rather than needing to do
> it separately). Here's a link to documentation on this feature:
>
>
http://wiki.libvirt.org/page/Networking#PCI_Passthrough_of_host_network_devices
>
> (look down to the section titled "Assignment with <interface
> type='hostdev'>")
>
> Or even better, use <interface type='network'> in your guest
config
> (still put the <mac address='xx:xx:xx:xx:xx:xx'/> element in
each one),
> and define a libvirt network which is a pool of SRIOV VFs - this is
> described further down the same page.
>
> This will not make a difference to the issue you describe below, but it
> should make managing your guest config and lifecycle much simpler.
I also conducted experiments with these XML configs and, as you said, it 
didn't make a difference to the issue observed. For most production use, 
my sample xml block is clearly not convenient.
> Define "failed". Do you mean that the cards communicated, but the
guests
> can see each others' traffic? Or do you mean that they see traffic from
> each other, but can't seem to communicate normally?
>
> If the problem is the latter, then make sure the PF (eth0 for you, I
> guess) has status UP and RUNNING before you start the guests.
>
> For the former, I'm not clear on the internal rules of switching of an
> SRIOV card. I think in most cases, the SRIOV card's internal switch may
> need to make everything from each VF visible to all other VFs, because
> the physical switch it's connected to may not mirror back traffic that
> really does need to go from one guest to the other. 802.1Qbh (which
> libvirt supports via the <virtualport type='802.1Qbh'>
element) does
> this differently, requiring all traffic to travel out to the switch,
> with the switch making the decision about what gets mirrored back, but
> you need an 802.1Qbh-capable switch for that.
What I meant by "failed" was a lack of traffic switching. I thought
that
the PF driver was responsible for configuring L2 switching on the NIC. 
Some others manufacturers, such as Emulex, have an internal switch. This 
is an extract from Emulex Whitepaper:

"I/O’s between VFs on the same PF can be processed by the adapter using 
an internal Layer 2 switch, eliminating routing through a physical switch"

I doubt that Broadcom has a specific and unsecure behavior for PF/VF 
communications. Unfortunately, I do not have an Intel or Emulex NIC with 
SR-IOV feature to check it.

Perhaps, I missed a configuration parameter somewhere or the 
driver/firmware used is broken... It looks really weird.

-- 
Université de Nantes - Direction des Systèmes d'Information