[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAOUHufbm6dLxgZ5=cbz1ogHGXrOfbF=wimpJySDpMRGJ=QPV4w@mail.gmail.com>
Date: Tue, 22 Nov 2022 12:08:07 -0700
From: Yu Zhao <yuzhao@...gle.com>
To: Ivan Babrou <ivan@...udflare.com>
Cc: Linux MM <linux-mm@...ck.org>,
Linux Kernel Network Developers <netdev@...r.kernel.org>,
linux-kernel <linux-kernel@...r.kernel.org>,
Johannes Weiner <hannes@...xchg.org>,
Michal Hocko <mhocko@...nel.org>,
Roman Gushchin <roman.gushchin@...ux.dev>,
Shakeel Butt <shakeelb@...gle.com>,
Muchun Song <songmuchun@...edance.com>,
Andrew Morton <akpm@...ux-foundation.org>,
Eric Dumazet <edumazet@...gle.com>,
"David S. Miller" <davem@...emloft.net>,
Hideaki YOSHIFUJI <yoshfuji@...ux-ipv6.org>,
David Ahern <dsahern@...nel.org>,
Jakub Kicinski <kuba@...nel.org>,
Paolo Abeni <pabeni@...hat.com>, cgroups@...r.kernel.org,
kernel-team <kernel-team@...udflare.com>
Subject: Re: Low TCP throughput due to vmpressure with swap enabled
On Tue, Nov 22, 2022 at 12:05 PM Ivan Babrou <ivan@...udflare.com> wrote:
>
> On Tue, Nov 22, 2022 at 10:59 AM Yu Zhao <yuzhao@...gle.com> wrote:
> >
> > On Mon, Nov 21, 2022 at 5:53 PM Ivan Babrou <ivan@...udflare.com> wrote:
> > >
> > > Hello,
> > >
> > > We have observed a negative TCP throughput behavior from the following commit:
> > >
> > > * 8e8ae645249b mm: memcontrol: hook up vmpressure to socket pressure
> > >
> > > It landed back in 2016 in v4.5, so it's not exactly a new issue.
> > >
> > > The crux of the issue is that in some cases with swap present the
> > > workload can be unfairly throttled in terms of TCP throughput.
> > >
> > > I am able to reproduce this issue in a VM locally on v6.1-rc6 with 8
> > > GiB of RAM with zram enabled.
> >
> > Hi Ivan,
> >
> > If it's not too much trouble, could you try again with the following?
> > CONFIG_LRU_GEN=y
> > CONFIG_LRU_GEN_ENABLED=y
> >
> > I haven't tried it myself. But I'll fix whatever doesn't work for you,
> > since your team is on the top of my customer list :)
>
> We don't have it in production, since there we have 5.15 LTS (our
> kernel team is testing v6.1).
>
> I do have it enabled in my VM running v6.1-rc6 where I was able to
> replicate this, so it doesn't seem to help.
Thanks. I'll look into this and get back to you.
Powered by blists - more mailing lists