From: Jason Xing <kerneljasonxing@gmail.com>
To: Paolo Abeni <pabeni@redhat.com>
Cc: willemdebruijn.kernel@gmail.com, davem@davemloft.net,
dsahern@kernel.org, edumazet@google.com, kuba@kernel.org,
netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
bpf@vger.kernel.org, Jason Xing <kernelxing@tencent.com>
Subject: Re: [PATCH net] udp: fix memory schedule error
Date: Tue, 21 Feb 2023 21:39:00 +0800 [thread overview]
Message-ID: <CAL+tcoD8PzL4khHq44z27qSHHGkcC4YUa91E3h+ki7O0u3SshQ@mail.gmail.com> (raw)
In-Reply-To: <48429c16fdaee59867df5ef487e73d4b1bf099af.camel@redhat.com>
On Tue, Feb 21, 2023 at 8:27 PM Paolo Abeni <pabeni@redhat.com> wrote:
>
> On Tue, 2023-02-21 at 19:03 +0800, Jason Xing wrote:
> > From: Jason Xing <kernelxing@tencent.com>
> >
> > Quoting from the commit 7c80b038d23e ("net: fix sk_wmem_schedule()
> > and sk_rmem_schedule() errors"):
> >
> > "If sk->sk_forward_alloc is 150000, and we need to schedule 150001 bytes,
> > we want to allocate 1 byte more (rounded up to one page),
> > instead of 150001"
>
> I'm wondering if this would cause measurable (even small) performance
> regression? Specifically under high packet rate, with BH and user-space
> processing happening on different CPUs.
>
> Could you please provide the relevant performance figures?
Sure, I've done some basic tests on my machine as below.
Environment: 16 cpus, 60G memory
Server: run "iperf3 -s -p [port]" command and start 500 processes.
Client: run "iperf3 -u -c 127.0.0.1 -p [port]" command and start 500 processes.
Running such tests makes sure that the util output of every cpu is
higher than 15% which is observed through top command.
Here're some before/after numbers by using the "sar -n DEV 10 2" command.
Before: rxpck/s 2000, txpck/s 2000, rxkB/s 64054.69, txkB/s 64054.69
After: rxpck/s 2000, txpck/s 2000, rxkB/s 64054.58, txkB/s 64054.58
So I don't see much impact on the results.
In theory, I have no clue about why it could cause some regression?
Maybe the memory allocation is not that enough compared to the
original code?
Thanks,
Jason
>
> Thanks!
>
> Paolo
>
next prev parent reply other threads:[~2023-02-21 13:39 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-02-21 11:03 [PATCH net] udp: fix memory schedule error Jason Xing
2023-02-21 12:27 ` Paolo Abeni
2023-02-21 12:35 ` Eric Dumazet
2023-02-21 13:44 ` Jason Xing
2023-02-21 13:39 ` Jason Xing [this message]
2023-02-21 14:46 ` Paolo Abeni
2023-02-21 15:46 ` Jason Xing
2023-02-22 3:47 ` Jason Xing
2023-02-23 8:39 ` Paolo Abeni
2023-02-23 9:07 ` Jason Xing
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=CAL+tcoD8PzL4khHq44z27qSHHGkcC4YUa91E3h+ki7O0u3SshQ@mail.gmail.com \
--to=kerneljasonxing@gmail.com \
--cc=bpf@vger.kernel.org \
--cc=davem@davemloft.net \
--cc=dsahern@kernel.org \
--cc=edumazet@google.com \
--cc=kernelxing@tencent.com \
--cc=kuba@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=netdev@vger.kernel.org \
--cc=pabeni@redhat.com \
--cc=willemdebruijn.kernel@gmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
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;
as well as URLs for NNTP newsgroup(s).