[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <eaddd974-abdf-0a96-40a9-936cfcd64d3e@intel.com>
Date: Fri, 23 Mar 2018 16:34:18 -0700
From: Jesus Sanchez-Palencia <jesus.sanchez-palencia@...el.com>
To: Thomas Gleixner <tglx@...utronix.de>
Cc: netdev@...r.kernel.org, jhs@...atatu.com, xiyou.wangcong@...il.com,
jiri@...nulli.us, vinicius.gomes@...el.com,
richardcochran@...il.com, anna-maria@...utronix.de,
henrik@...tad.us, john.stultz@...aro.org, levi.pearson@...man.com,
edumazet@...gle.com, willemb@...gle.com, mlichvar@...hat.com
Subject: Re: [RFC v3 net-next 13/18] net/sched: Introduce the TBS Qdisc
Hi Thomas,
On 03/23/2018 01:49 AM, Thomas Gleixner wrote:
> On Thu, 22 Mar 2018, Jesus Sanchez-Palencia wrote:
>> On 03/22/2018 03:11 PM, Thomas Gleixner wrote:
>> So, are you just opposing to the case where sorting off + offload off is used?
>> (i.e. the scheduled FIFO case)
>
> FIFO does not make any sense if your packets have a fixed transmission
> time. I yet have to see a reasonable explanation why FIFO in the context of
> time ordered would be a good thing.
On context of tbs, the scheduled FIFO was developed just so consistency was kept
between all 4 variants, basically (sw best-effort or hw offload vs sorting
enabled or sorting disabled).
I don't have any strong argument in favor of this mode at the moment, so I will
just remove it on a next version - unless someone else brings up a valid use
case for it, of course.
Thanks for the feedback,
Jesus
Powered by blists - more mailing lists