From: David Miller <davem@davemloft.net>
To: ilpo.jarvinen@helsinki.fi
Cc: ycheng@google.com, ncardwell@google.com, netdev@vger.kernel.org,
nanditad@google.com, therbert@google.com
Subject: Re: [PATCH 5/5] tcp: skip cwnd moderation in TCP_CA_Open in tcp_try_to_open
Date: Sun, 27 Nov 2011 19:01:20 -0500 (EST) [thread overview]
Message-ID: <20111127.190120.2116987138205753451.davem@davemloft.net> (raw)
In-Reply-To: <alpine.DEB.2.00.1111221146060.31705@wel-95.cs.helsinki.fi>
From: "Ilpo Järvinen" <ilpo.jarvinen@helsinki.fi>
Date: Tue, 22 Nov 2011 13:47:13 +0200 (EET)
> On Mon, 21 Nov 2011, Yuchung Cheng wrote:
>
>> AFAIK cwnd moderation is to lower bursty drops not to discourage
>> (dsack) cheating. I believe the reason is the same in
>> tcp_try_to_open(). We are in common cases, e.g., loss-recovery, this
>> logic hurts performance.
>
> Quote from the patch description: "Senders were overriding cwnd values
> picked during an undo by calling tcp_moderate_cwnd()" ...so after all it
> has to do with undo being limited. IMHO only up to orig_cwnd/2+IW is safe
> due to cheating opportunities. Also FRTO uses orig_cwnd/2 due to same
> reason (it could do the +IW too but IIRC it is only /2 currently). What
> would be the safeguard there after this one is removed? I kind of see your
> point about this particular line being related to burst mitigation but on
> the same time the end result of removal is that undo becomes potentially
> much more aggressive.
I'm apply this patch to net-next now, but Neil and Yucheng are on the hook
to fully look into the issues Ilpo has raised.
Thanks.
next prev parent reply other threads:[~2011-11-28 0:01 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-11-16 18:58 [PATCH 1/5] tcp: make is_dupack a parameter to tcp_fastretrans_alert() Neal Cardwell
2011-11-16 18:58 ` [PATCH 2/5] tcp: use DSACKs that arrive when packets_out is 0 Neal Cardwell
2011-11-17 1:34 ` Ilpo Järvinen
2011-11-27 23:55 ` David Miller
2011-11-16 18:58 ` [PATCH 3/5] tcp: use SACKs and DSACKs that arrive on ACKs below snd_una Neal Cardwell
2011-11-17 1:52 ` Ilpo Järvinen
2011-11-27 23:57 ` David Miller
2011-11-16 18:58 ` [PATCH 4/5] tcp: allow undo from reordered DSACKs Neal Cardwell
2011-11-17 5:18 ` Ilpo Järvinen
2011-11-17 20:49 ` Neal Cardwell
2011-11-22 12:44 ` Ilpo Järvinen
2011-11-27 23:58 ` David Miller
2011-11-28 17:31 ` Neal Cardwell
2011-12-12 12:05 ` [PATCH 1/1] tcp: fix spurious undos in CA_Open Ilpo Järvinen
2011-12-12 19:40 ` Neal Cardwell
2011-12-14 8:57 ` Ilpo Järvinen
2011-12-18 18:23 ` Neal Cardwell
2011-11-16 18:58 ` [PATCH 5/5] tcp: skip cwnd moderation in TCP_CA_Open in tcp_try_to_open Neal Cardwell
2011-11-17 5:14 ` Ilpo Järvinen
2011-11-19 21:08 ` Neal Cardwell
2011-11-20 8:38 ` Ilpo Järvinen
2011-11-21 22:57 ` Yuchung Cheng
2011-11-22 11:47 ` Ilpo Järvinen
2011-11-28 0:01 ` David Miller [this message]
2011-11-28 17:35 ` Neal Cardwell
2011-11-28 20:33 ` Yuchung Cheng
2011-11-29 10:33 ` Ilpo Järvinen
2011-11-20 10:19 ` Alexander Zimmermann
2011-11-17 1:23 ` [PATCH 1/5] tcp: make is_dupack a parameter to tcp_fastretrans_alert() Ilpo Järvinen
2011-11-27 23:55 ` David Miller
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=20111127.190120.2116987138205753451.davem@davemloft.net \
--to=davem@davemloft.net \
--cc=ilpo.jarvinen@helsinki.fi \
--cc=nanditad@google.com \
--cc=ncardwell@google.com \
--cc=netdev@vger.kernel.org \
--cc=therbert@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).