netdev - Re: [PATCH net] net, sched: fix soft lockup in tc

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: <585AD503.2070300@iogearbox.net>
Date:   Wed, 21 Dec 2016 20:16:19 +0100
From:   Daniel Borkmann <daniel@...earbox.net>
To:     Cong Wang <xiyou.wangcong@...il.com>
CC:     David Miller <davem@...emloft.net>,
        Shahar Klein <shahark@...lanox.com>,
        Or Gerlitz <gerlitz.or@...il.com>,
        Roi Dayan <roid@...lanox.com>, Jiri Pirko <jiri@...lanox.com>,
        John Fastabend <john.fastabend@...il.com>,
        Linux Kernel Network Developers <netdev@...r.kernel.org>
Subject: Re: [PATCH net] net, sched: fix soft lockup in tc_classify

On 12/21/2016 07:51 PM, Cong Wang wrote:
> On Wed, Dec 21, 2016 at 9:04 AM, Daniel Borkmann <daniel@...earbox.net> wrote:
>> What happens is that in tc_ctl_tfilter(), thread A allocates a new
>> tp, initializes it, sets tp_created to 1, and calls into tp->ops->change()
>> with it. In that classifier callback we had to unlock/lock the rtnl
>> mutex and returned with -EAGAIN. One reason why we need to drop there
>> is, for example, that we need to request an action module to be loaded.
>
> Excellent catch!
>
> But why do we have to replay the request here? Shouldn't we just return
> EAGAIN to user-space and let user-space decide to retry or not?
> Replaying is the root of the evil here.

Right, this behavior is already pretty old (2005), see history
tree 8d7c694553dc ("[PKT_SCHED]: act_api.c: drop rtnl for loading
modules") and 437293de63d8 ("[PKT_SCHED]: cls_api.c: drop rtnl
for loading modules"), some binaries could rely on that behavior
in one way or another I'd presume.