From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
To: Paolo Abeni <pabeni@redhat.com>
Cc: netdev@vger.kernel.org, "David S. Miller" <davem@davemloft.net>,
Eric Dumazet <edumazet@google.com>,
Jakub Kicinski <kuba@kernel.org>,
Jesper Dangaard Brouer <hawk@kernel.org>,
Thomas Gleixner <tglx@linutronix.de>,
Wander Lairson Costa <wander@redhat.com>,
Yan Zhai <yan@cloudflare.com>
Subject: Re: [PATCH v3 net-next 2/4] net: Allow to use SMP threads for backlog NAPI.
Date: Tue, 5 Mar 2024 11:52:18 +0100 [thread overview]
Message-ID: <20240305105218.EXitUftO@linutronix.de> (raw)
In-Reply-To: <ae886b7975751a2c148fa4addce26c456678c735.camel@redhat.com>
On 2024-03-05 11:44:06 [+0100], Paolo Abeni wrote:
> > I am not aware of a perf regression.
>
> I probably inferred too much from the udp lookback case.
Yes. This was real and is gone now :)
> > Jakub was worried about a possible regression with this and while asking
> > nobody came up with an actual use case where this is used. So it is as
> > he suggested, optional for everyone but forced-enabled for RT where it
> > is required.
> > I had RH benchmarking this and based on their 25Gbe and 50Gbe NICs and
> > the results look good. If anything it looked a bit better with this on
> > the 50Gbe NICs but since those NICs have RSS…
> >
> > I have this default off so that nobody complains and yet has to
> > possibility to test and see if it leads to a problem. If not, we could
> > enable it by default and after a few cycles and then remove the IPI code
> > a few cycles later with absent complains.
>
> I think this late in the cycle is better to keep backlog threads off by
> default.
Sure.
> Thanks,
>
> Paolo
Seastian
next prev parent reply other threads:[~2024-03-05 10:52 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-02-28 12:05 [PATCH v3 net-next 0/3] net: Provide SMP threads for backlog NAPI Sebastian Andrzej Siewior
2024-02-28 12:05 ` [PATCH v3 net-next 1/4] net: Remove conditional threaded-NAPI wakeup based on task state Sebastian Andrzej Siewior
2024-02-28 12:05 ` [PATCH v3 net-next 2/4] net: Allow to use SMP threads for backlog NAPI Sebastian Andrzej Siewior
2024-03-05 10:08 ` Paolo Abeni
2024-03-05 10:35 ` Sebastian Andrzej Siewior
2024-03-05 10:44 ` Paolo Abeni
2024-03-05 10:52 ` Sebastian Andrzej Siewior [this message]
2024-03-05 15:23 ` Jakub Kicinski
2024-03-05 15:33 ` Eric Dumazet
2024-03-06 8:18 ` Sebastian Andrzej Siewior
2024-02-28 12:05 ` [PATCH v3 net-next 3/4] net: Use backlog-NAPI to clean up the defer_list Sebastian Andrzej Siewior
2024-03-05 9:55 ` Paolo Abeni
2024-03-05 10:25 ` Sebastian Andrzej Siewior
2024-03-05 10:36 ` Paolo Abeni
2024-03-05 10:40 ` Sebastian Andrzej Siewior
2024-02-28 12:05 ` [PATCH v3 net-next 4/4] net: Rename rps_lock to backlog_lock Sebastian Andrzej Siewior
2024-03-05 3:30 ` [PATCH v3 net-next 0/3] net: Provide SMP threads for backlog NAPI Jakub Kicinski
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=20240305105218.EXitUftO@linutronix.de \
--to=bigeasy@linutronix.de \
--cc=davem@davemloft.net \
--cc=edumazet@google.com \
--cc=hawk@kernel.org \
--cc=kuba@kernel.org \
--cc=netdev@vger.kernel.org \
--cc=pabeni@redhat.com \
--cc=tglx@linutronix.de \
--cc=wander@redhat.com \
--cc=yan@cloudflare.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).