Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S263129AbTFGLow (ORCPT ); Sat, 7 Jun 2003 07:44:52 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S263150AbTFGLow (ORCPT ); Sat, 7 Jun 2003 07:44:52 -0400 Received: from dp.samba.org ([66.70.73.150]:26857 "EHLO lists.samba.org") by vger.kernel.org with ESMTP id S263129AbTFGLov convert rfc822-to-8bit (ORCPT ); Sat, 7 Jun 2003 07:44:51 -0400 From: Paul Mackerras MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8BIT Message-ID: <16097.53169.687777.155890@argo.ozlabs.ibm.com> Date: Sat, 7 Jun 2003 21:42:41 +1000 To: =?iso-8859-1?Q?J=F6rn?= Engel Cc: Linus Torvalds , Steven Cole , linux-kernel@vger.kernel.org Subject: Re: [Patch] 2.5.70-bk11 zlib merge #4 pure magic In-Reply-To: <20030607100217.GB24694@wohnheim.fh-wedel.de> References: <20030606183126.GA10487@wohnheim.fh-wedel.de> <20030606183247.GB10487@wohnheim.fh-wedel.de> <20030606183920.GC10487@wohnheim.fh-wedel.de> <20030606185210.GE10487@wohnheim.fh-wedel.de> <20030606192325.GG10487@wohnheim.fh-wedel.de> <20030606192814.GH10487@wohnheim.fh-wedel.de> <20030606200051.GI10487@wohnheim.fh-wedel.de> <20030606201306.GJ10487@wohnheim.fh-wedel.de> <16097.45833.384548.319399@argo.ozlabs.ibm.com> <20030607100217.GB24694@wohnheim.fh-wedel.de> X-Mailer: VM 7.16 under Emacs 21.3.2 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1267 Lines: 30 J?rn Engel writes: > On Sat, 7 June 2003 19:40:25 +1000, Paul Mackerras wrote: > > > > Your change won't affect PPP, since pppd already refuses to use > > windowBits == 8 (as a workaround for this bug). > > Seems like I have misread the ppp code then. In that case, please > remove the ppp part from the previous patch or use this one instead, > Linus. I meant that pppd (i.e. the user-level part of PPP) would refuse to negotiate windowBits == 8 with the peer, so from that point of view it doesn't matter if the kernel driver accepts it or not, since it will never be asked to accept it (by pppd). It is better on the whole if the kernel driver doesn't accept it since that is one less exploitable thing in the kernel (although you would have to be root to exploit it). But if the zlib code also refuses to use windowBits == 8, it then doesn't matter whether the ppp_deflate code accepts it, from either point of view. On the whole I would say that the change to ppp-comp.h should go in, for now at least. Paul. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/