Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 25 Aug 2020 11:43:45 +0200
From:      Michael Tuexen <Michael.Tuexen@lurchi.franken.de>
To:        Liang Tian <l.tian.email@gmail.com>
Cc:        freebsd-net <freebsd-net@freebsd.org>
Subject:   Re: Appropriate Byte Counting during Congestion Avoidance
Message-ID:  <2854347B-8C13-485A-AA23-B895B6ACD714@lurchi.franken.de>
In-Reply-To: <51C01869-3FC2-43A8-AC9D-4A6F961F93B7@lurchi.franken.de>
References:  <CAJhigrgajo%2BXAfY7PQA5zPr_HWOqTjfThnGkX9St2NqbgiTprg@mail.gmail.com> <51C01869-3FC2-43A8-AC9D-4A6F961F93B7@lurchi.franken.de>

next in thread | previous in thread | raw e-mail | index | archive | help
> On 19. Aug 2020, at 10:14, Michael Tuexen =
<Michael.Tuexen@lurchi.franken.de> wrote:
>=20
>> On 19. Aug 2020, at 06:51, Liang Tian <l.tian.email@gmail.com> wrote:
>>=20
>> Hi everyone,
>>=20
>> We noticed CWND is growing much slower than expected during =
congestion
>> avoidance with new reno, and we came to this piece of code in
>> cc_ack_received() at tcp_input.c:353
>>=20
>> if (type =3D=3D CC_ACK) {
>> ....
>>       if (tp->snd_cwnd > tp->snd_ssthresh) {
>>           tp->t_bytes_acked +=3D min(tp->ccv->bytes_this_ack,
>>                nsegs * V_tcp_abc_l_var * tcp_maxseg(tp));
>>           if (tp->t_bytes_acked >=3D tp->snd_cwnd) {
>>               tp->t_bytes_acked -=3D tp->snd_cwnd;
>>               tp->ccv->flags |=3D CCF_ABC_SENTAWND;
>>           }
>> The increment of t_bytes_acked is capped at 2*maxseg.
>> The description of the sysctl variable tcp_abc_l_var(default value 2) =
is
>> "Cap the max cwnd increment during slow-start to this number of =
segments"
>> After reading RFC3465, it doesn't look like this cap should be =
applied
>> here since this is clearly not during slow-start.
>> We've seen in some cases the receiver is ACKing every 16 packets, and
>> CWND is growing at 1/8 of the expected rate because of this.
>>=20
>> I would appreciate your opinion on this. Thanks a lot.
> Hi Liang,
>=20
> thanks for bringing this up. I agree. A patch is under review:
> https://reviews.freebsd.org/D26120
Now fixed in https://svnweb.freebsd.org/changeset/base/364754

Thanks again for reporting.

Best regards
Michael
>=20
> Best regards
> Michael
>>=20
>> Regards,
>> Liang
>> _______________________________________________
>> freebsd-net@freebsd.org mailing list
>> https://lists.freebsd.org/mailman/listinfo/freebsd-net
>> To unsubscribe, send any mail to =
"freebsd-net-unsubscribe@freebsd.org"
>=20
> _______________________________________________
> freebsd-net@freebsd.org mailing list
> https://lists.freebsd.org/mailman/listinfo/freebsd-net
> To unsubscribe, send any mail to "freebsd-net-unsubscribe@freebsd.org"




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?2854347B-8C13-485A-AA23-B895B6ACD714>