Bug 1977921

1. Feature Overview
        a. Name of feature:
           SR-IOV support over VirtIO devices

        b. Feature description:
            We are able to load drivers on VirtIO Physical functions, but after binding to vSwitch we don't see any VF's.
            There is no option to enable SRIOV for VirtIO devices.
            Talking to one of the developers (Yan VugenFirer), they needed to discuss internally the timeframe for the implementation.

2. Feature Details
        a. Architectures: Intel / AMD (x86_64), POWER LE (ppc64le), IBM Z (s390x), 64-bit ARM (aarch64):
           x86_64
        b. Bugzilla dependencies:
        c. Drivers or hardware dependencies, including a specific platform or CPU:
             Mellanox ConnectX network cards.
        d. Library or other software dependencies:
        e. Upstream acceptance information, including Linus' kernel version in which the         
           feature appears and the date on which this feature was accepted or is targeted
           for acceptance into Linus' kernel:
        f. External links:
        g. Severity (H,M,L): High (required for Hardware Enablement), Medium, Low:
            High
        h. Feature required by date (for example, the date on which hardware requiring this feature is planned for launch):
3. Business Justification
        a. Why is this feature needed?
            Required to support a project for our customers.
        b. What hardware or software does this enable?
            Mellanox ConnectX network cards.
        c. If hardware, is it on-board in a system (eg, LOM) or an add-on card?
            Add-on card.
        d. Business impact?
            Needed by a major customer.
        e. What market problems / audience does it address?
4. QE Test Plan
5. Primary contact at Red Hat, email, phone (chat)
        a. Andre Beausoleil 
        b. abeausol
6. Primary contact at Partner, email, phone (chat)
        a. Sunil Rani
        b. sunrani

        a. Alaa Hleihel
        b. ahleihel

Hi, Yan

If possible, could you please provide the detailed test steps for this bug ?

(In reply to Yanghang Liu from comment #2)
> Hi, Yan
> 
> If possible, could you please provide the detailed test steps for this bug ?

When the feature will be available, I will provide the test steps.

For now, it is impossible to created VFs for virtio-net devices on a Windows host.

Hi Yan, 


I have a few questions about this bug that I want to confirm with you in advance.

Thanks a lot for your help.

(1) 
>  Drivers or hardware dependencies, including a specific platform or CPU: Mellanox ConnectX network cards.

Do we have to use this "Mellanox ConnectX network cards" to test and reproduce this bug?

Can it be replaced by other SRIOV nic（like 82599ES、BCM57810） or non-SRIOV nic ?


(2)
> For now, it is impossible to created VFs for virtio-net devices on a Windows *host*

Is it acceptable that QE reproduce and test this bug in a Windows *vm* ?

If the answer is yes, please help check the question 3 as well.


(3) 
> VirtIO Physical Functions

What does "VirtIO Physical functions" refer to ?

I am not sure if this "VirtIO Physical functions" is similar to the devices I use in the following two test scenarios.

Could you please help confirm that ?

Test scenarios 1:

(3.1.1) create a bridge and add a PF interface into this bridge on the host 

(3.1.2) start a Windows vm with a virtio net pci device

The qemu cmd line of the "virtio net pci device" is as following:
-netdev tap,fd=40,id=hostnet0,vhost=on,vhostfd=41 \
-device virtio-net-pci,netdev=hostnet0,id=net0,mac=f4:8e:38:c3:85:85,bus=pci.1,addr=0x0 \

Test scenarios 2:

(3.2.1) choose a PF and then bind the PF driver to the vfio-pci

(3.2.2) start a Windows vm with this PF

The qemu cmd line of the PF is as following:
-device vfio-pci,host=$domain:$bus:$device.$function,id=hostdev0 \

(4) 
>  binding to vSwitch

Is this step similar to the one in the following link? 

https://docs.microsoft.com/en-us/windows-server/virtualization/hyper-v/get-started/create-a-virtual-switch-for-hyper-v-virtual-machines

Do I need to do any additional operations for binding Physical functions to vSwitch  ?

After evaluating this issue, there are no plans to address it further or fix it in an upcoming release.  Therefore, it is being closed.  If plans change such that this issue will be fixed in an upcoming release, then the bug can be reopened.

Issue migration from Bugzilla to Jira is in process at this time. This will be the last message in Jira copied from the Bugzilla bug.

This BZ has been automatically migrated to the issues.redhat.com Red Hat Issue Tracker. All future work related to this report will be managed there.

To find the migrated issue, look in the "Links" section for a direct link to the new issue location. The issue key will have an icon of 2 footprints next to it, and begin with "RHEL-" followed by an integer.  You can also find this issue by visiting https://issues.redhat.com/issues/?jql= and searching the "Bugzilla Bug" field for this BZ's number, e.g. a search like:

"Bugzilla Bug" = 1234567

In the event you have trouble locating or viewing this issue, you can file an issue by sending mail to rh-issues.