netdev - Re: [4.9.13] use after free in ipv4

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: <1488905813.9415.403.camel@edumazet-glaptop3.roam.corp.google.com>
Date:   Tue, 07 Mar 2017 08:56:53 -0800
From:   Eric Dumazet <eric.dumazet@...il.com>
To:     Stephen Hemminger <stephen@...workplumber.org>
Cc:     Daniel J Blueman <daniel@...ra.org>,
        Netdev <netdev@...r.kernel.org>,
        David Miller <davem@...emloft.net>,
        Stephen Hemminger <shemminger@...ux-foundation.org>
Subject: Re: [4.9.13] use after free in ipv4_mtu

On Tue, 2017-03-07 at 08:29 -0800, Stephen Hemminger wrote:

> +       WARN_ONCE(strcmp(default_qdisc_ops->id, "fq"),
> +                 "TCP BBR should only be used with FQ qdisc\n");
> +
>  

Why would that be needed, especially for people that properly setup
their qdisc ? Maybe they do not want to force fq on all devices like tun
or taps ;)

Also we intend to provide a fallback in case FQ is not in the qdisc
chain : Some routers admins really want fq_codel (or whatever qdisc)

TCP will handle the pacing itself, at a small cost for the ones that
hate fq.