public inbox for quic@lists.linux.dev 
 help / color / mirror / Atom feed
From: Xin Long <lucien.xin@gmail.com>
To: Paolo Abeni <pabeni@redhat.com>
Cc: network dev <netdev@vger.kernel.org>,
	quic@lists.linux.dev, davem@davemloft.net,  kuba@kernel.org,
	Eric Dumazet <edumazet@google.com>,
	Simon Horman <horms@kernel.org>,
	 Stefan Metzmacher <metze@samba.org>,
	Moritz Buhl <mbuhl@openbsd.org>,
	Tyler Fanelli <tfanelli@redhat.com>,
	 Pengtao He <hepengtao@xiaomi.com>,
	Thomas Dreibholz <dreibh@simula.no>,
	linux-cifs@vger.kernel.org,  Steve French <smfrench@gmail.com>,
	Namjae Jeon <linkinjeon@kernel.org>,
	 Paulo Alcantara <pc@manguebit.com>, Tom Talpey <tom@talpey.com>,
	kernel-tls-handshake@lists.linux.dev,
	 Chuck Lever <chuck.lever@oracle.com>,
	Jeff Layton <jlayton@kernel.org>,
	 Steve Dickson <steved@redhat.com>,
	Hannes Reinecke <hare@suse.de>,
	Alexander Aring <aahringo@redhat.com>,
	 David Howells <dhowells@redhat.com>,
	Matthieu Baerts <matttbe@kernel.org>,
	 John Ericson <mail@johnericson.me>,
	Cong Wang <xiyou.wangcong@gmail.com>,
	 "D . Wythe" <alibuda@linux.alibaba.com>,
	Jason Baron <jbaron@akamai.com>,
	 illiliti <illiliti@protonmail.com>,
	Sabrina Dubroca <sd@queasysnail.net>,
	 Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>,
	Daniel Stenberg <daniel@haxx.se>,
	 Andy Gospodarek <andrew.gospodarek@broadcom.com>
Subject: Re: [PATCH net-next v4 09/15] quic: add congestion control
Date: Thu, 6 Nov 2025 15:24:30 -0500	[thread overview]
Message-ID: <CADvbK_d8WoKJkU7ACK6nzbv7hzxxkAYZ5--DPzVQHsSZbEJnuw@mail.gmail.com> (raw)
In-Reply-To: <43ea4062-75a8-4152-bf19-2eca561036bd@redhat.com>

On Tue, Nov 4, 2025 at 7:02 AM Paolo Abeni <pabeni@redhat.com> wrote:
>
> On 10/29/25 3:35 PM, Xin Long wrote:
> > +/* Compute and update the pacing rate based on congestion window and smoothed RTT. */
> > +static void quic_cong_pace_update(struct quic_cong *cong, u32 bytes, u32 max_rate)
> > +{
> > +     u64 rate;
> > +
> > +     /* rate = N * congestion_window / smoothed_rtt */
> > +     rate = (u64)cong->window * USEC_PER_SEC * 2;
> > +     if (likely(cong->smoothed_rtt))
> > +             rate = div64_ul(rate, cong->smoothed_rtt);
> > +
> > +     WRITE_ONCE(cong->pacing_rate, min_t(u64, rate, max_rate));
> > +     pr_debug("%s: update pacing rate: %u, max rate: %u, srtt: %u\n",
> > +              __func__, cong->pacing_rate, max_rate, cong->smoothed_rtt);
>
> I think you should skip entirely the pacing_rate update when
> `smoothed_rtt == 0`
>
will update it.

> [...]> +/* rfc9002#section-5: Estimating the Round-Trip Time */
> > +void quic_cong_rtt_update(struct quic_cong *cong, u32 time, u32 ack_delay)
> > +{
> > +     u32 adjusted_rtt, rttvar_sample;
> > +
> > +     /* Ignore RTT sample if ACK delay is suspiciously large. */
> > +     if (ack_delay > cong->max_ack_delay * 2)
> > +             return;
> > +
> > +     /* rfc9002#section-5.1: latest_rtt = ack_time - send_time_of_largest_acked */
> > +     cong->latest_rtt = cong->time - time;
> > +
> > +     /* rfc9002#section-5.2: Estimating min_rtt */
> > +     if (!cong->min_rtt_valid) {
> > +             cong->min_rtt = cong->latest_rtt;
> > +             cong->min_rtt_valid = 1;
> > +     }
> > +     if (cong->min_rtt > cong->latest_rtt)
> > +             cong->min_rtt = cong->latest_rtt;
> > +
> > +     if (!cong->is_rtt_set) {
> > +             /* rfc9002#section-5.3:
> > +              *   smoothed_rtt = latest_rtt
> > +              *   rttvar = latest_rtt / 2
> > +              */
> > +             cong->smoothed_rtt = cong->latest_rtt;
> > +             cong->rttvar = cong->smoothed_rtt / 2;
> > +             quic_cong_pto_update(cong);
> > +             cong->is_rtt_set = 1;
> > +             return;
> > +     }
> > +
> > +     /* rfc9002#section-5.3:
> > +      *   adjusted_rtt = latest_rtt
> > +      *   if (latest_rtt >= min_rtt + ack_delay):
> > +      *     adjusted_rtt = latest_rtt - ack_delay
> > +      *   smoothed_rtt = 7/8 * smoothed_rtt + 1/8 * adjusted_rtt
> > +      *   rttvar_sample = abs(smoothed_rtt - adjusted_rtt)
> > +      *   rttvar = 3/4 * rttvar + 1/4 * rttvar_sample
> > +      */
> > +     adjusted_rtt = cong->latest_rtt;
> > +     if (cong->latest_rtt >= cong->min_rtt + ack_delay)
> > +             adjusted_rtt = cong->latest_rtt - ack_delay;
> > +
> > +     cong->smoothed_rtt = (cong->smoothed_rtt * 7 + adjusted_rtt) / 8;
>
> Out of sheer curiosity, is the compiler smart enough to use a 'srl 3'
> for the above?
>
Yes.

266 cong->smoothed_rtt = (cong->smoothed_rtt * 7 + adjusted_rtt) / 8;

0x593d <+77>:  mov    (%rbx),%ecx         ; ecx = cong->smoothed_rtt
0x593f <+79>:  lea    (%rax,%rcx,8),%edx   ; edx = adjusted_rtt + (ecx * 8)
0x5942 <+82>:  sub    %ecx,%edx           ; edx = adjusted_rtt +
(8*ecx) - ecx = ecx*7 + adjusted_rtt
0x5946 <+86>:  shr    $0x3,%edx           ; edx >>= 3 → divide by 8
0x594d <+93>:  mov    %edx,(%rbx)         ; store result back to
cong->smoothed_rtt

Thanks.

  reply	other threads:[~2025-11-06 20:24 UTC|newest]

Thread overview: 48+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-10-29 14:35 [PATCH net-next v4 00/15] net: introduce QUIC infrastructure and core subcomponents Xin Long
2025-10-29 14:35 ` [PATCH net-next v4 01/15] net: define IPPROTO_QUIC and SOL_QUIC constants Xin Long
2025-11-04  9:20   ` Paolo Abeni
2025-10-29 14:35 ` [PATCH net-next v4 02/15] net: build socket infrastructure for QUIC protocol Xin Long
2025-10-29 16:22   ` Stefan Metzmacher
2025-10-29 19:57     ` Xin Long
2025-10-30 11:29       ` Stefan Metzmacher
2025-10-30 14:13         ` Xin Long
2025-10-30 14:17           ` Stefan Metzmacher
2025-10-30 14:28             ` Xin Long
2025-11-04  9:38   ` Paolo Abeni
2025-11-05 22:20     ` Xin Long
2025-10-29 14:35 ` [PATCH net-next v4 03/15] quic: provide common utilities and data structures Xin Long
2025-11-04  9:55   ` Paolo Abeni
2025-10-29 14:35 ` [PATCH net-next v4 04/15] quic: provide family ops for address and protocol Xin Long
2025-11-04 10:27   ` Paolo Abeni
2025-11-06  1:01     ` Xin Long
2025-10-29 14:35 ` [PATCH net-next v4 05/15] quic: provide quic.h header files for kernel and userspace Xin Long
2025-10-29 14:35 ` [PATCH net-next v4 06/15] quic: add stream management Xin Long
2025-11-04 11:05   ` Paolo Abeni
2025-11-06  1:27     ` Xin Long
2025-11-06  8:51       ` Paolo Abeni
2025-11-06 16:22         ` Xin Long
2025-10-29 14:35 ` [PATCH net-next v4 07/15] quic: add connection id management Xin Long
2025-10-29 14:35 ` [PATCH net-next v4 08/15] quic: add path management Xin Long
2025-11-04 11:50   ` Paolo Abeni
2025-11-06  1:28     ` Xin Long
2025-10-29 14:35 ` [PATCH net-next v4 09/15] quic: add congestion control Xin Long
2025-11-04 12:02   ` Paolo Abeni
2025-11-06 20:24     ` Xin Long [this message]
2025-10-29 14:35 ` [PATCH net-next v4 10/15] quic: add packet number space Xin Long
2025-11-04 12:17   ` Paolo Abeni
2025-11-06 16:40     ` Xin Long
2025-10-29 14:35 ` [PATCH net-next v4 11/15] quic: add crypto key derivation and installation Xin Long
2025-10-29 14:35 ` [PATCH net-next v4 12/15] quic: add crypto packet encryption and decryption Xin Long
2025-10-29 14:35 ` [PATCH net-next v4 13/15] quic: add timer management Xin Long
2025-11-04 12:33   ` Paolo Abeni
2025-11-06 16:49     ` Xin Long
2025-11-13 21:23       ` Xin Long
2025-10-29 14:35 ` [PATCH net-next v4 14/15] quic: add frame encoder and decoder base Xin Long
2025-11-04 12:47   ` Paolo Abeni
2025-11-06 17:22     ` Xin Long
2025-11-13 21:26       ` Xin Long
2025-10-29 14:35 ` [PATCH net-next v4 15/15] quic: add packet builder and parser base Xin Long
2025-11-04 14:44   ` Paolo Abeni
2025-11-06 19:24     ` Xin Long
2025-11-04  2:41 ` [PATCH net-next v4 00/15] net: introduce QUIC infrastructure and core subcomponents Jakub Kicinski
2025-11-05 22:19   ` Xin Long

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=CADvbK_d8WoKJkU7ACK6nzbv7hzxxkAYZ5--DPzVQHsSZbEJnuw@mail.gmail.com \
    --to=lucien.xin@gmail.com \
    --cc=aahringo@redhat.com \
    --cc=alibuda@linux.alibaba.com \
    --cc=andrew.gospodarek@broadcom.com \
    --cc=chuck.lever@oracle.com \
    --cc=daniel@haxx.se \
    --cc=davem@davemloft.net \
    --cc=dhowells@redhat.com \
    --cc=dreibh@simula.no \
    --cc=edumazet@google.com \
    --cc=hare@suse.de \
    --cc=hepengtao@xiaomi.com \
    --cc=horms@kernel.org \
    --cc=illiliti@protonmail.com \
    --cc=jbaron@akamai.com \
    --cc=jlayton@kernel.org \
    --cc=kernel-tls-handshake@lists.linux.dev \
    --cc=kuba@kernel.org \
    --cc=linkinjeon@kernel.org \
    --cc=linux-cifs@vger.kernel.org \
    --cc=mail@johnericson.me \
    --cc=marcelo.leitner@gmail.com \
    --cc=matttbe@kernel.org \
    --cc=mbuhl@openbsd.org \
    --cc=metze@samba.org \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=pc@manguebit.com \
    --cc=quic@lists.linux.dev \
    --cc=sd@queasysnail.net \
    --cc=smfrench@gmail.com \
    --cc=steved@redhat.com \
    --cc=tfanelli@redhat.com \
    --cc=tom@talpey.com \
    --cc=xiyou.wangcong@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