netdev - Re: [RFC PATCH v2] bridge: make it possible for packets to traverse the bridge without hitting netfilter

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: <1520612114.29061.15.camel@infradead.org>
Date:   Fri, 09 Mar 2018 16:15:14 +0000
From:   David Woodhouse <dwmw2@...radead.org>
To:     David Miller <davem@...emloft.net>
Cc:     fw@...len.de, pablo@...filter.org, rga@...zon.de,
        bridge@...ts.linux-foundation.org, stephen@...workplumber.org,
        netdev@...r.kernel.org, linux-kernel@...r.kernel.org,
        aliguori@...zon.com, nbd@...nwrt.org
Subject: Re: [RFC PATCH v2] bridge: make it possible for packets to
 traverse the bridge without hitting netfilter



On Fri, 2018-03-09 at 10:57 -0500, David Miller wrote:
> From: David Woodhouse <dwmw2@...radead.org>
> Date: Fri, 09 Mar 2018 15:31:15 +0000
> 
> > Eschewing a 15% speedup on the basis that "well, even though we've had
> > three of these already for a decade, we're worried that adding a fourth
> > might open the floodgates to further patches" does seem a little odd to
> > me, FWIW.
> 
> The cost we are dealing with is a fundamental one which is a result of
> the hook design.
> 
> Indirect calls are killer.
> 
> Indirect calls are even more killer now in the age of Spectre and
> retpolines.

Imre's 15% measurement was, obviously, before that. We should redo it
and confirm the numbers.

> I definitely would rather see the fundamental issue addressed rather
> than poking at it randomly with knobs for this case and that.

Yeah. What do you think of the suggestion I made — that a given hook
should automatically disable itself if it tautologically does nothing?
Coupled with notifiers for when the rules change, to re-enable the
hooks again? I confess I don't even know how realistic that is.
Download attachment "smime.p7s" of type "application/x-pkcs7-signature" (5213 bytes)