From mboxrd@z Thu Jan 1 00:00:00 1970 From: Nivedita Singhvi Subject: Re: PMTU issues due to TOS field manipulation (for DSCP) Date: Wed, 10 Dec 2003 13:11:41 -0800 Sender: netdev-bounce@oss.sgi.com Message-ID: <3FD78C0D.4030609@us.ibm.com> References: <20031210203433.3747b7fd.ak@suse.de> <3FD7826F.1050308@us.ibm.com> <20031210215527.239fb33f.ak@suse.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Cc: ruddk@us.ibm.com, netdev@oss.sgi.com, chester.f.johnson@intel.com Return-path: To: Andi Kleen In-Reply-To: <20031210215527.239fb33f.ak@suse.de> Errors-to: netdev-bounce@oss.sgi.com List-Id: netdev.vger.kernel.org Andi Kleen wrote: > Only the router would need to change (and rewrite ICMP messages, which is a > bit nasty, but then compatibility is not always fun) True, but I get a sinking feeling when I hear we have to change router/icmp code. The word "only" seems inappropriate ;). > -Andi > > P.S.: I am not opposed to fixing linux for this, just I have my doubts > that fixing all end hosts is a practical solution for the problem. The reason it seems simpler to me, at least, and I don't speak for Kevin or the person at Intel who first reported the problem, is that I see only a few people (well, only Intel :)) running into the problem. So if there were a patch available or the fix were in some release, only the few end hosts who actually needed it would need to upgrade. Otherwise, rollout would be evolutionary. But also open to other approaches... thanks, Nivedita