From: Jakub Kicinski <kuba@kernel.org>
To: Eric Dumazet <edumazet@google.com>
Cc: "David S . Miller" <davem@davemloft.net>,
Paolo Abeni <pabeni@redhat.com>,
netdev@vger.kernel.org, eric.dumazet@gmail.com,
Soheil Hassas Yeganeh <soheil@google.com>,
Neal Cardwell <ncardwell@google.com>,
Yuchung Cheng <ycheng@google.com>
Subject: Re: [RFC net-next 4/4] tcp: defer regular ACK while processing socket backlog
Date: Thu, 7 Sep 2023 11:00:15 -0700 [thread overview]
Message-ID: <20230907110015.75fdcc5c@kernel.org> (raw)
In-Reply-To: <CANn89iJY8UypOGqSOJo531ny4isPSiTg2xW-rO_xNmnYVVovQw@mail.gmail.com>
On Thu, 7 Sep 2023 19:16:01 +0200 Eric Dumazet wrote:
> > Is it okay if I asked why quickack?
> > Is it related to delay-based CC?
>
> Note the patch is also helping the 'regular' mode, without "quickack 1" .
>
> This is CC related in any way, but some TCP tx zerocopy workload, sending
> one chunk at a time, waiting for the TCP tx zerocopy completion in
> order to proceed for the next chunk,
> because the 'next chunk' is re-using the memory.
>
> The receiver application is not sending back a message (otherwise the
> 'delayed ack' would be piggybacked in the reply),
> and it also does not know what size of the message was expected (so no
> SO_RCVLOWAT or anything could be attempted)
>
> For this kind of workload, it is crucial the last ACK is not delayed, at all.
Interesting. Some folks at Meta were recently looking into parsing RPCs
in the kernel to avoid unnecessary wakeups. Poor man's KCM using BPF
sockmaps. Passing message size hints from the sender would solve so
many problems..
In any case, I don't mean to question the patch :)
next prev parent reply other threads:[~2023-09-07 18:00 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-09-06 20:10 [RFC net-next 0/4] tcp: backlog processing optims Eric Dumazet
2023-09-06 20:10 ` [RFC net-next 1/4] tcp: no longer release socket ownership in tcp_release_cb() Eric Dumazet
2023-09-06 20:10 ` [RFC net-next 2/4] net: sock_release_ownership() cleanup Eric Dumazet
2023-09-06 20:10 ` [RFC net-next 3/4] net: call prot->release_cb() when processing backlog Eric Dumazet
2023-09-06 20:10 ` [RFC net-next 4/4] tcp: defer regular ACK while processing socket backlog Eric Dumazet
2023-09-06 20:31 ` Yuchung Cheng
2023-09-06 23:07 ` Stephen Hemminger
2023-09-07 14:07 ` Neal Cardwell
2023-09-07 14:35 ` Soheil Hassas Yeganeh
[not found] ` <CAA93jw7Fuov-vmxiZdW7My-AVWCOFQo4XVm9bNwAg4Td2CUNCA@mail.gmail.com>
2023-09-07 16:54 ` Eric Dumazet
2023-09-07 17:09 ` Jakub Kicinski
2023-09-07 17:16 ` Eric Dumazet
2023-09-07 18:00 ` Jakub Kicinski [this message]
2023-09-07 18:05 ` Eric Dumazet
2023-09-07 18:10 ` 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=20230907110015.75fdcc5c@kernel.org \
--to=kuba@kernel.org \
--cc=davem@davemloft.net \
--cc=edumazet@google.com \
--cc=eric.dumazet@gmail.com \
--cc=ncardwell@google.com \
--cc=netdev@vger.kernel.org \
--cc=pabeni@redhat.com \
--cc=soheil@google.com \
--cc=ycheng@google.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).