public inbox for netdev@vger.kernel.org 
 help / color / mirror / Atom feed
From: Stephen Hemminger <stephen@networkplumber•org>
To: David Miller <davem@davemloft•net>
Cc: caleb.raitto@gmail•com, mst@redhat•com, jasowang@redhat•com,
	netdev@vger•kernel.org, caraitto@google•com
Subject: Re: [PATCH net-next] virtio_net: force_napi_tx module param.
Date: Mon, 30 Jul 2018 12:51:46 -0700	[thread overview]
Message-ID: <20180730125146.19c37975@xeon-e3> (raw)
In-Reply-To: <20180729.090027.1373538625446665385.davem@davemloft.net>

On Sun, 29 Jul 2018 09:00:27 -0700 (PDT)
David Miller <davem@davemloft•net> wrote:

> From: Caleb Raitto <caleb.raitto@gmail•com>
> Date: Mon, 23 Jul 2018 16:11:19 -0700
> 
> > From: Caleb Raitto <caraitto@google•com>
> > 
> > The driver disables tx napi if it's not certain that completions will
> > be processed affine with tx service.
> > 
> > Its heuristic doesn't account for some scenarios where it is, such as
> > when the queue pair count matches the core but not hyperthread count.
> > 
> > Allow userspace to override the heuristic. This is an alternative
> > solution to that in the linked patch. That added more logic in the
> > kernel for these cases, but the agreement was that this was better left
> > to user control.
> > 
> > Do not expand the existing napi_tx variable to a ternary value,
> > because doing so can break user applications that expect
> > boolean ('Y'/'N') instead of integer output. Add a new param instead.
> > 
> > Link: https://patchwork.ozlabs.org/patch/725249/
> > Acked-by: Willem de Bruijn <willemb@google•com>
> > Acked-by: Jon Olson <jonolson@google•com>
> > Signed-off-by: Caleb Raitto <caraitto@google•com>  
> 
> So I looked into the history surrounding these issues.
> 
> First of all, it's always ends up turning out crummy when drivers start
> to set affinities themselves.  The worst possible case is to do it
> _conditionally_, and that is exactly what virtio_net is doing.
> 
> From the user's perspective, this provides a really bad experience.
> 
> So if I have a 32-queue device and there are 32 cpus, you'll do all
> the affinity settings, stopping Irqbalanced from doing anything
> right?
> 
> So if I add one more cpu, you'll say "oops, no idea what to do in
> this situation" and not touch the affinities at all?
> 
> That makes no sense at all.
> 
> If the driver is going to set affinities at all, OWN that decision
> and set it all the time to something reasonable.
> 
> Or accept that you shouldn't be touching this stuff in the first place
> and leave the affinities alone.
> 
> Right now we're kinda in a situation where the driver has been setting
> affinities in the ncpus==nqueues cases for some time, so we can't stop
> doing it.
> 
> Which means we have to set them in all cases to make the user
> experience sane again.
> 
> I looked at the linked to patch again:
> 
> 	https://patchwork.ozlabs.org/patch/725249/
> 
> And I think the strategy should be made more generic, to get rid of
> the hyperthreading assumptions.  I also agree that the "assign
> to first N cpus" logic doesn't make much sense either.
> 
> Just distribute across the available cpus evenly, and be done with it.
> If you have 64 cpus and 32 queues, this assigns queues to every other
> cpu.
> 
> Then we don't need this weird new module parameter.

I wonder if it would be possible to give irqbalanced hints
with irq_set_affinity_hint instead of doing direct affinity setting?

  parent reply	other threads:[~2018-07-30 21:28 UTC|newest]

Thread overview: 33+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-07-23 23:11 [PATCH net-next] virtio_net: force_napi_tx module param Caleb Raitto
2018-07-24  0:53 ` Stephen Hemminger
2018-07-24  1:23   ` Willem de Bruijn
2018-07-24 10:42 ` Michael S. Tsirkin
2018-07-24 14:01   ` Willem de Bruijn
2018-07-24 18:38     ` Michael S. Tsirkin
2018-07-24 20:52       ` Willem de Bruijn
2018-07-24 22:23         ` Michael S. Tsirkin
2018-07-24 22:31           ` Willem de Bruijn
2018-07-24 22:46             ` Michael S. Tsirkin
2018-07-25  0:02               ` Willem de Bruijn
2018-07-25  0:17               ` Jon Olson
2018-07-30  6:06                 ` Jason Wang
2018-08-01 22:27                   ` Michael S. Tsirkin
2018-08-28 19:57                   ` Willem de Bruijn
2018-08-29  7:56                     ` Jason Wang
2018-08-29 13:01                       ` Willem de Bruijn
2018-09-09 23:07                         ` Willem de Bruijn
2018-09-10  5:59                           ` Jason Wang
2018-07-29 16:00 ` David Miller
2018-07-29 20:33   ` Michael S. Tsirkin
2018-07-29 20:36     ` David Miller
2018-07-29 21:09     ` Willem de Bruijn
2018-07-29 21:32   ` Willem de Bruijn
2018-07-31 12:34     ` Michael S. Tsirkin
2018-08-01 15:46       ` Willem de Bruijn
2018-08-01 15:56         ` Willem de Bruijn
2018-08-01 22:25         ` Michael S. Tsirkin
2018-08-01 22:43           ` Willem de Bruijn
2018-08-01 22:48             ` Michael S. Tsirkin
2018-08-01 23:33               ` Willem de Bruijn
2018-07-30 19:51   ` Stephen Hemminger [this message]
2018-07-31  1:41     ` Jason Wang

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20180730125146.19c37975@xeon-e3 \
    --to=stephen@networkplumber$(echo .)org \
    --cc=caleb.raitto@gmail$(echo .)com \
    --cc=caraitto@google$(echo .)com \
    --cc=davem@davemloft$(echo .)net \
    --cc=jasowang@redhat$(echo .)com \
    --cc=mst@redhat$(echo .)com \
    --cc=netdev@vger$(echo .)kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox