Date:   Tue, 31 Jul 2018 15:34:08 +0300
From:   "Michael S. Tsirkin" <mst@...hat.com>
To:     Willem de Bruijn <willemdebruijn.kernel@...il.com>
Cc:     David Miller <davem@...emloft.net>, caleb.raitto@...il.com,
        Jason Wang <jasowang@...hat.com>,
        Network Development <netdev@...r.kernel.org>,
        Caleb Raitto <caraitto@...gle.com>
Subject: Re: [PATCH net-next] virtio_net: force_napi_tx module param.

On Sun, Jul 29, 2018 at 05:32:56PM -0400, Willem de Bruijn wrote:
> On Sun, Jul 29, 2018 at 12:01 PM David Miller <davem@...emloft.net> wrote:
> >
> > From: Caleb Raitto <caleb.raitto@...il.com>
> > Date: Mon, 23 Jul 2018 16:11:19 -0700
> >
> > > From: Caleb Raitto <caraitto@...gle.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@...gle.com>
> > > Acked-by: Jon Olson <jonolson@...gle.com>
> > > Signed-off-by: Caleb Raitto <caraitto@...gle.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.
> 
> Sounds good to me.

So e.g. we could set an affinity hint to a group of CPUs that
might transmit to this queue.

> > If you have 64 cpus and 32 queues, this assigns queues to every other
> > cpu.
> 
> Striping half the number of queues as cores on a hyperthreaded system
> with two logical cores per physical core will allocate all queues on
> only half the physical cores.
> 
> But it is probably not safe to make any assumptions on virtual to
> physical core mapping, anyway, which makes the simplest strategy is
> preferable.

Powered by blists - more mailing lists