netdev - Re: [PATCH net] dctcp: avoid bogus doubling of cwnd after loss

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] [day] [month] [year] [list]

Date:   Mon, 31 Oct 2016 15:18:58 -0400 (EDT)
From:   David Miller <davem@...emloft.net>
To:     fw@...len.de
Cc:     netdev@...r.kernel.org, daniel@...earbox.net, brakmo@...com,
        agshew@...il.com, glenn.judd@...ganstanley.com
Subject: Re: [PATCH net] dctcp: avoid bogus doubling of cwnd after loss

From: Florian Westphal <fw@...len.de>
Date: Fri, 28 Oct 2016 18:43:11 +0200

> If a congestion control module doesn't provide .undo_cwnd function,
> tcp_undo_cwnd_reduction() will set cwnd to
> 
>    tp->snd_cwnd = max(tp->snd_cwnd, tp->snd_ssthresh << 1);
> 
> ... which makes sense for reno (it sets ssthresh to half the current cwnd),
> but it makes no sense for dctcp, which sets ssthresh based on the current
> congestion estimate.
> 
> This can cause severe growth of cwnd (eventually overflowing u32).
> 
> Fix this by saving last cwnd on loss and restore cwnd based on that,
> similar to cubic and other algorithms.
> 
> Fixes: e3118e8359bb7c ("net: tcp: add DCTCP congestion control algorithm")
> Cc: Lawrence Brakmo <brakmo@...com>
> Cc: Andrew Shewmaker <agshew@...il.com>
> Cc: Glenn Judd <glenn.judd@...ganstanley.com>
> Acked-by: Daniel Borkmann <daniel@...earbox.net>
> Signed-off-by: Florian Westphal <fw@...len.de>

Applied and queued up for -stable, thanks Florian.