netdev - Re: [PATCH v4] tcp: harmonize tcp_vx

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: <4B4F31EB.1000004@gmail.com>
Date:	Thu, 14 Jan 2010 10:02:03 -0500
From:	William Allen Simpson <william.allen.simpson@...il.com>
To:	Patrick McHardy <kaber@...sh.net>
CC:	Joe Perches <joe@...ches.com>, Andi Kleen <andi@...stfloor.org>,
	Linux Kernel Developers <linux-kernel@...r.kernel.org>,
	Linux Kernel Network Developers <netdev@...r.kernel.org>,
	Ilpo Järvinen <ilpo.jarvinen@...sinki.fi>,
	Eric Dumazet <eric.dumazet@...il.com>
Subject: Re: [PATCH v4] tcp: harmonize tcp_vx_rcv header length assumptions

Patrick McHardy wrote:
> In IPv6 it is released in ip6_input.c before the packet is handed to
> the protocol handler.
> 
Thanks, I made a note of it.

Thanks to Andi, I've been using his suggestion to compare the IPv4 and
IPv6 generated code.  Another minor difference between them was IPv4
stored *iph.

So, I added *ip6h, and discovered that it made the code longer.  Later
inline function invocations had also calculated it, so adding it merely
calculated it earlier and stored it.  The code thrashes registers
making room for the additional variable.  I'm guessing that the IPv6
folks had already looked at the issue....  Good work!

So, I tried removing *iph, and that shrunk the IPv4 code some, too.

Thanks to everybody for their comments and review.
--
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