[prev in list] [next in list] [prev in thread] [next in thread] 

List:       dpdk-users
Subject:    Re: [dpdk-users] MLX5 Rx HW Timestamp units
From:       Anshu Boonapalli <anshu.boonapalli () gmail ! com>
Date:       2018-11-29 13:24:11
Message-ID: CA+=pU-KqhEBC20D270M=1ELGKrMzvXGhrd_2HCR2=VajBV2+gQ () mail ! gmail ! com
[Download RAW message or body]

Thank you very much Tom!

Do you know if your work is portable to DPDK 17.11 tip? If not, i will look
into the move to 18.11.

-Anshu

On Wed, Nov 28, 2018 at 3:01 AM Tom Barbette <barbette@kth.se> wrote:

> You actually motivated me to finish re-writing the patch... I just sent i=
t
> to the dev mailing list. You'll also find the 3 commits there :
> https://github.com/tbarbette/dpdk/tree/mlxclock
>
>
> Tom
> ------------------------------
> *De :* Anshu Boonapalli <anshu.boonapalli@gmail.com>
> *Envoy=C3=A9 :* mardi 27 novembre 2018 16:36
> *=C3=80 :* Tom Barbette
> *Cc :* users@dpdk.org
> *Objet :* Re: [dpdk-users] MLX5 Rx HW Timestamp units
>
> hi Tom,
>
> Thanks for clarifying.
>
> I am interested in the patch that allows the query of the current device
> clock values. Please point me to this patch.
>
> -Anshu
>
> On Tue, Nov 27, 2018 at 12:57 AM Tom Barbette <barbette@kth.se> wrote:
>
>> Hi Anshu,
>>
>> No it's the value of an internal Clock. The frequency and the base are
>> both unknown.
>>
>> We have a patch that works (but must be reworked before submission to
>> dev) that allows to query the current device clock value. Then from ther=
e
>> you can derive the base and the frequency. I can send it to you if you w=
ish.
>>
>> Tom
>> ________________________________________
>> De : users <users-bounces@dpdk.org> de la part de Anshu Boonapalli <
>> anshu.boonapalli@gmail.com>
>> Envoy=C3=A9 : mardi 27 novembre 2018 02:41
>> =C3=80 : users@dpdk.org
>> Objet : [dpdk-users] MLX5 Rx HW Timestamp units
>>
>> Anyone know the units of Rx HW timestamps MLX5 provides?
>>
>> I assumed nanoseconds but tests i have run don't match the nanosecond
>> units.
>>
>> Thanks,
>> -Anshu
>>
>
[prev in list] [next in list] [prev in thread] [next in thread]