[ovs-dev] [PATCH 0/7 RFC] OVS-DPDK flow offload with rte_flow

[Darrell Ball]

On 8/29/17, 7:25 PM, "Yuanhan Liu" <yliu at fridaylinux.org> wrote:

    On Wed, Aug 30, 2017 at 01:32:10AM +0000, Darrell Ball wrote:
    > 
    >     >         Though that being said, this patchset still has issues unresolved. The
    >     >         major issue is that maybe most NIC (for instance, Mellanox and Intel)
    >     >         can not support a pure MARK action. It has to be used together with a
    >     >         QUEUE action, which in turn needs a queue index. That comes to the issue:
    >     >         the queue index is not given in the flow context. To make it work, patch
    >     >         5 just set the queue index to 0, which is obviously wrong. One possible
    >     >         solution is to record the rxq and pass it down to the flow creation
    >     >         stage. It would be much better, but it's still far away from being perfect.
    >     >         Because it might have changed the steering rules stealthily, which may
    >     >         break the default RSS setup by OVS-DPDK.
    >     > 
    >     > If this cannot be solved by removing this restriction, I guess another alternative is to actively
    >     > manage flow-queue associations.
    >     
    >     do you mean let user provide the set_queue action?
    > 
    > I mean in the worst case, if the restriction cannot be lifted, we might to need to do flow distribution
    > across queues with additional added logic/tracking,
    
    Like the way mentioned above: record the rxq at recv stage and pass it
    down to flow_put?
    
    > because we would not really want to keep this restriction
    > without a workaround.
    
    For sure.
    
    > This would be a fair bit of work, however.
    
    Indeed. Another reason I don't really like above solution is it changes
    the core logic of OVS (just for OVS-DPDK use case). What's worse, it
    doesn't even solve the issue perfectly: it's still a workaround!
    
    That being said, if we have to provide a workaround, I think maybe a
    simple round-robin queue distribution is better, judging it's our first
    attempt to have hw offload? Later, we could either improve the workaround,
    or remove the restriction from DPDK side (if it's possible). Good to you?
    
    > Alternatively, pushing this to the user in the general may be too much overhead; what do you think ?.
    > As a user specification, it could be optional, however ?
    
    Yes, that's what I was thinking. For the users that know their NIC supports
    MARK with QUEUE action, they could not add such option. For others, it's
    suggested to add it. Or, it's a must (if we do not do workaround like
    above). Makes sense?

Hi Yuanhan

I responded to the last e-mail since there is some overlap in the discussion on some
points.


    
    >     >         
    >     >         The reason I still want to send it out is to get more comments/thoughts
    >     >         from community on this whole patchset. Meanwhile, I will try to resolve
    >     >         the QUEUE action issue.
    >     >         
    >     >         Note that it's disabled by default, which can be enabled by:
    >     >         
    >     >             $ ovs-vsctl set Open_vSwitch . other_config:hw-offload=true
    >     > 
    >     > Maybe per in-port configuration would alleviate the issue to a certain degree.
    >     
    >     Yes, it could be done. I choose it for following reasons:
    >     
    >     - the option is already there, used by tc offloads.
    >     - it also simplifies the (first) patchset a bit, IMO.
    > 
    > Of course, I understand.
    >     
    >     However, I'm okay with making it per port. What's your suggestion for
    >     this? Making "hw-offload" be port, or introducing another one? If so,
    >     what's your suggestion on the naming?
    >     
    > I am not suggesting to drop the global configuration
    > Mainly we ‘consider’ additional per interface configuration because of the restriction with
    > queue action we discuss. This reduces the scope of the queue remapping from what RSS would yield
    > with HWOL.
    > I would expect that when such configuration is done (if it were done), that
    > typically multiple ports would be configured, since traffic flows bi-directionally at least.
    > 
    > If we were to do this, one of the possibilities would be something like:
    > ovs-vsctl set Interface dpdk0 other_config:hw-offload=true
    
    What's the scope between the global one and this one then? We simply
    ignore the globle one for OVS-DPDK?

[Darrell]
If we were to have such an option (and it is a pretty big ‘if’), then more
specific scope (i.e. interface scope here) would eclipse global scope I think.
Maybe initially, we try to keep it as simple as possible, but I could see some other use cases
for interface level config., like HW resource contention for multiport NICs ?

    >     Thanks for the review. BTW, would you please add me in 'to' or 'cc'
    >     list while replying to me?  Otherwise, it's easy to get missed: too
    >     many emails :/
    > 
    > of course
    
    Thank you!
    
    	--yliu