netdev - Re: [RFC PATCH net-next] net: add an entry for CONFIG_NET_RX_BUSY

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAL+tcoDZ2VDCd00ydv-RzMudq=d+jVukiDLgs7RpsJwvGqBp1Q@mail.gmail.com>
Date: Tue, 23 Jul 2024 23:09:57 +0800
From: Jason Xing <kerneljasonxing@...il.com>
To: Eric Dumazet <edumazet@...gle.com>
Cc: davem@...emloft.net, kuba@...nel.org, pabeni@...hat.com, horms@...nel.org, 
	netdev@...r.kernel.org, Jason Xing <kernelxing@...cent.com>
Subject: Re: [RFC PATCH net-next] net: add an entry for CONFIG_NET_RX_BUSY_POLL

On Tue, Jul 23, 2024 at 10:57 PM Eric Dumazet <edumazet@...gle.com> wrote:
>
> On Tue, Jul 23, 2024 at 3:57 PM Jason Xing <kerneljasonxing@...il.com> wrote:
> >
> > From: Jason Xing <kernelxing@...cent.com>
> >
> > When I was doing performance test on unix_poll(), I found out that
> > accessing sk->sk_ll_usec when calling sock_poll()->sk_can_busy_loop()
> > occupies too much time, which causes around 16% degradation. So I
> > decided to turn off this config, which cannot be done apparently
> > before this patch.
>
> Too many CONFIG_ options, distros will enable it anyway.
>
> In my builds, offset of sk_ll_usec is 0xe8.
>
> Are you using some debug options or an old tree ?
>
> I can not understand how a 16% degradation can occur, reading a field
> in a cache line which contains read mostly fields for af_unix socket.
>
> I think you need to provide more details / analysis, and perhaps come
> to a different conclusion.

Thanks for your comments.

I'm also confused about the result. The design of the cache line is
correct from my perspective because they are all read mostly fields as
you said.

I was doing some tests by using libmicro[1] and found this line '41.30
│      test  %r14d,%r14d' by using perf. So I realised that there is
something strange here. Then I disable that config, the result turns
out to be better than before. One of my colleagues can prove it.

In this patch, I described a story about why I would like to let
people disable/enable it, but investigating this part may be another
different thing, I think. I will keep trying.

[1]: https://github.com/redhat-performance/libMicro.git
running 'https://github.com/redhat-performance/libMicro.git' to see the results

>
> >
> > Signed-off-by: Jason Xing <kernelxing@...cent.com>
> > ---
> > More data not much related if you're interested:
> >   5.82 │      mov   0x18(%r13),%rdx
> >   0.03 │      mov   %rsi,%r12
> >   1.76 │      mov   %rdi,%rbx
> >        │    sk_can_busy_loop():
> >   0.50 │      mov   0x104(%rdx),%r14d
> >  41.30 │      test  %r14d,%r14d
> > Note: I run 'perf record -e  L1-dcache-load-misses' to diagnose
> > ---
> >  net/Kconfig | 4 +++-
> >  1 file changed, 3 insertions(+), 1 deletion(-)
> >
> > diff --git a/net/Kconfig b/net/Kconfig
> > index d27d0deac0bf..1f1b793984fe 100644
> > --- a/net/Kconfig
> > +++ b/net/Kconfig
> > @@ -335,8 +335,10 @@ config CGROUP_NET_CLASSID
> >           being used in cls_cgroup and for netfilter matching.
> >
> >  config NET_RX_BUSY_POLL
> > -       bool
> > +       bool "Low latency busy poll timeout"
> >         default y if !PREEMPT_RT || (PREEMPT_RT && !NETCONSOLE)
> > +       help
> > +         Approximate time in us to spin waiting for packets on the device queue.
>
> Wrong comment. It is a y/n choice, no 'usec' at this stage.

Oh, I see.

Thanks,
Jason

>
> >
> >  config BQL
> >         bool
> > --
> > 2.37.3
> >