[dpdk-users] KNI Threads/Cores

[Cliff Burdick]

Ferruh, sorry.

On Wed, Jun 8, 2016 at 12:48 PM, Cliff Burdick <shaklee3 at gmail.com> wrote:

> Hi Yigit, when you say "I guess it's possible" is that not common? I would
> think that the amount of traffic people would want to forward to Linux for
> normal DPDK applications would be quite small. If you had to have a core
> dedicated for KNI on each interface that would get very wasteful.
>
> On Wed, Jun 8, 2016 at 12:31 PM, Ferruh Yigit <ferruh.yigit at intel.com>
> wrote:
>
>> On 6/8/2016 5:30 PM, Cliff Burdick wrote:
>> > Hi, I have an application with two sockets where each core I'm planning
>> to
>> > transmit and receive a fairly large amount of traffic per core. Each
>> core
>> > right now handles a single queue of either TX or RX of a given port.
>> Across
>> > all the cores, I may be processing up to 12 ports. I also need to handle
>> > things like ARP and ping, so I'm going to add in the KNI driver to
>> handle
>> > that. Since the amount of traffic I'm expecting that I'll need to
>> forward
>> > to Linux is very small, it seems like I should be able to dedicate one
>> > lcore per socket to handle this functionality and have the dataplane
>> cores
>> > pass the traffic off to this core using rte_kni_tx_burst().
>> >
>> > My question is, first of all, is this possible? It seems like I can
>> > configure the KNI driver to start in "single thread" mode. From that
>> point,
>> > I want to initialize one KNI device for each port, and have each kernel
>> > lcore on each processor handle that traffic. I believe if I call
>> > rte_kni_alloc with core_id set to the kernel lcore for each device,
>> then in
>> > the end I'll have something like 6 KNI devices on socket one being
>> handled
>> > by lcore 0, and 6 KNI devices on socket 2 being handled by lcore 31 as
>> an
>> > example. Then my threads that are handling the dataplane tx/rx can
>> simply
>> > be passed a pointer to their respective rte_kni device. Does this sound
>> > correct?
>>
>> If rte_kni module used "single thread" mode, kernel core_id is not used
>> at all. For single thread mode, a single thread created, this is used to
>> for all kni devices and not able to pin to any specific lcore.
>>
>> For what you have described, first need to insert module with
>> kthread_mode=multiple param. This will create a kernel thread per kni
>> interface. But I guess it is possible to provide same
>> rte_kni_conf->core_id for some of them, and yes rte_kni_conf->force_pin
>> is required, otherwise core_id is not useful. According your sample,
>> first 6 kni devices will have core_id value 0, and other 6 kni devices
>> will have core_id value 31, with all have force_bind set. This will
>> create 12 kernel threads, will bind 6 of them to core 0 and other 6 to
>> core 31.
>>
>> >
>> > Also, the sample says the core affinity needs to be set using taskset.
>> Is
>> > that already taken care of with conf.core_id in rte_kni_alloc or do I
>> still
>> > need to set it?
>> >
>> > Thanks
>> >
>>
>>
>