Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1711832

Re: [PATCH net-next V2 3/3] tap: XDP support

From Jason Wang <jasowang@redhat.com>
Newsgroups linux.kernel
Subject Re: [PATCH net-next V2 3/3] tap: XDP support
Date 2017-08-15 07:10 +0200
Message-ID <ueCh3-5tW-3@gated-at.bofh.it> (permalink)
References <udgBY-315-9@gated-at.bofh.it> <udgBZ-315-43@gated-at.bofh.it> <udrnI-1gO-7@gated-at.bofh.it> <uduEV-3fB-3@gated-at.bofh.it> <ueq6f-6jm-31@gated-at.bofh.it>
Organization linux.* mail to news gateway

Show all headers | View raw



On 2017年08月15日 00:01, Michael S. Tsirkin wrote:
> On Sat, Aug 12, 2017 at 10:48:49AM +0800, Jason Wang wrote:
>>
>> On 2017年08月12日 07:12, Jakub Kicinski wrote:
>>> On Fri, 11 Aug 2017 19:41:18 +0800, Jason Wang wrote:
>>>> This patch tries to implement XDP for tun. The implementation was
>>>> split into two parts:
>>>>
>>>> - fast path: small and no gso packet. We try to do XDP at page level
>>>>     before build_skb(). For XDP_TX, since creating/destroying queues
>>>>     were completely under control of userspace, it was implemented
>>>>     through generic XDP helper after skb has been built. This could be
>>>>     optimized in the future.
>>>> - slow path: big or gso packet. We try to do it after skb was created
>>>>     through generic XDP helpers.
>>>>
>>>> Test were done through pktgen with small packets.
>>>>
>>>> xdp1 test shows ~41.1% improvement:
>>>>
>>>> Before: ~1.7Mpps
>>>> After:  ~2.3Mpps
>>>>
>>>> xdp_redirect to ixgbe shows ~60% improvement:
>>>>
>>>> Before: ~0.8Mpps
>>>> After:  ~1.38Mpps
>>>>
>>>> Suggested-by: Michael S. Tsirkin <mst@redhat.com>
>>>> Signed-off-by: Jason Wang <jasowang@redhat.com>
>>> Looks OK to me now :)
>>>
>>> Out of curiosity, you say the build_skb() is for "small packets", and it
>>> seems you are always reserving the 256B regardless of XDP being
>>> installed.  Does this have no performance impact on non-XDP case?
>> Have a test, only less than 1% were noticed which I think could be ignored.
>>
>> Thanks
> What did you test btw?

Pktgen

>   The biggest issue would be with something like
> UDP with short packets.
>

Note that we do this only when sndbuf is INT_MAX. So this is probably 
not an issue. The only thing matter is more stress to page allocator, 
but according to the result of pktgen it was very small that could be 
ignored.

Thanks

Back to linux.kernel | Previous | NextPrevious in thread | Next in thread | Find similar | Unroll thread


Thread

Re: [PATCH net-next V2 3/3] tap: XDP support Jakub Kicinski <kubakici@wp.pl> - 2017-08-12 01:20 +0200
  Re: [PATCH net-next V2 3/3] tap: XDP support Jason Wang <jasowang@redhat.com> - 2017-08-12 04:50 +0200
    Re: [PATCH net-next V2 3/3] tap: XDP support "Michael S. Tsirkin" <mst@redhat.com> - 2017-08-14 18:10 +0200
      Re: [PATCH net-next V2 3/3] tap: XDP support Jason Wang <jasowang@redhat.com> - 2017-08-15 07:10 +0200
        Re: [PATCH net-next V2 3/3] tap: XDP support "Michael S. Tsirkin" <mst@redhat.com> - 2017-08-16 05:50 +0200

csiph-web