netdev - Re: [PATCH v7 net-next 2/2] tcp: switch rtt estimations to usec resolution

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <1393467946.26794.15.camel@edumazet-glaptop2.roam.corp.google.com>
Date:	Wed, 26 Feb 2014 18:25:46 -0800
From:	Eric Dumazet <eric.dumazet@...il.com>
To:	Stephen Hemminger <stephen@...workplumber.org>
Cc:	David Miller <davem@...emloft.net>, Julian Anastasov <ja@....bg>,
	Yuchung Cheng <ycheng@...gle.com>,
	netdev <netdev@...r.kernel.org>,
	Neal Cardwell <ncardwell@...gle.com>,
	Larry Brakmo <brakmo@...gle.com>
Subject: Re: [PATCH v7 net-next 2/2] tcp: switch rtt estimations to usec
 resolution

On Wed, 2014-02-26 at 17:09 -0800, Stephen Hemminger wrote:

> 
> Since srtt is scaled by 8 and is 32 bits, This does reduce the maximum
> possible RTT to 8 minutes or so. Which is fine, just wanted it to be
> noted.

Note that I plan to remove the scaling by 8.

We also are experimenting alternatives to srtt.

Van Jacobson admits he was on drugs at the time he did it ;)

Idea would be to keep latest 3 (or 5) rtt samples, and either
get the maximum value for RTO, or the mininum values for pacing
or congestion modules. This would filter delay acks in the right way.

I'd like to instrument hystart, it seems we can do a bit better than
current situation, before the new stuff.

Thanks !


--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html