Re: [net-next PATCH V5] qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE

[Jamal Hadi Salim <jhs@mojatatu.com>]

On 10/01/14 10:55, Tom Herbert wrote:
> On Wed, Oct 1, 2014 at 6:17 AM, Jamal Hadi Salim <jhs@mojatatu.com> wrote:
>> On 09/30/14 14:20, David Miller wrote:
>>>
>>> From: Jamal Hadi Salim <jhs@mojatatu.com>
>>> Date: Tue, 30 Sep 2014 07:07:37 -0400
>>>
>>>> Note, there are benefits as you have shown - but i would not
>>>> consider those to be standard use cases (actully one which would
>>>> have shown clear win is the VM thing Rusty was after).
>>>
>>>
>>> I completely disagree, you will see at least decreased cpu utilization
>>> for a very common case, bulk single stream transfers.
>>>
>>
>>
>> So lets say the common use case is:
>> = modern day cpu (pick some random cpu)
>> = 1-10 Gbps ethernet (not 100mbps)
>> = 1-24 tcp or udp bulk (you said one, Jesper had 24 which sounds better)
>>
>> Run with test cases:
>> a) unchanged (no bulking code added at all)
>> vs
>> b) bulking code added and used
>> vs
>> c) bulking code added and *not* used
>>
>> Jesper's results are comparing #b and #c.
>>
>> And if #b + #c are slightly worse or equal then we have a win;->
>>

BTW: meant to say if #b and #c are slightly worse than #a then we have
a win.

>> Again, I do believe things like traffic generators or the VM io
>> or something like tuntap that crosses user space will have a clear
>> benefit (but are those common use cases?).
>>
> You're making this much more complicated that it actually is. The
> algorithm is simple-- queue wakes up, finds out how exactly many bytes
> to dequeue, and performs dequeue of enough packets under one lock.

It is not about bql.
The issue is: if i am going to attempt to do a bulk transfer
every single time (with new code) and for the common use case the
result is "no need to do bulking" then you just added extra code that
is unnecessary for that common case.
Even a single extra if statement at high packet rate is still
costly and would be easy to observe.

>The
> should be a benefit when transmitting high rate as we know that
> reducing locking is generally a win.

You mean amortizing the cost of the lock not removing a lock?
Yes, of course. That is if the added code ends up being hit
meaningfully. Jesper said (and it was my experience as well)
that it was _hard_ to achieve bulking in such a case.
The fear here is in the common case (if we say the
bulk transfer is a common case) infact that code is reduced to be
a per-packet as opposed to a burst of packets, then there is
no win.
The tests should clarify, no?

cheers,
jamal