From: David Miller Subject: Re: [PATCH] limit rt cache size Date: Tue, 08 Aug 2006 17:32:56 -0700 (PDT) Message-ID: <20060808.173256.64863348.davem@davemloft.net> References: <200608090123.01123.ak@suse.de> <20060808.171152.58440635.davem@davemloft.net> <200608090224.04158.ak@suse.de> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: akepner@sgi.com, kuznet@ms2.inr.ac.ru, dev@sw.ru, netdev@vger.kernel.org Return-path: Received: from dsl027-180-168.sfo1.dsl.speakeasy.net ([216.27.180.168]:23728 "EHLO sunset.davemloft.net") by vger.kernel.org with ESMTP id S1030261AbWHIAdQ (ORCPT ); Tue, 8 Aug 2006 20:33:16 -0400 To: ak@suse.de In-Reply-To: <200608090224.04158.ak@suse.de> Sender: netdev-owner@vger.kernel.org List-Id: netdev.vger.kernel.org From: Andi Kleen Date: Wed, 9 Aug 2006 02:24:04 +0200 > Yes, but even with dynamic growth you still need some upper boundary > (otherwise a DOS could eat all your memory). And it would need > to be figured out what it is. Absolutely. Otherwise the GC'ing of the routing cache would break horribly. > BTW does dynamic shrink after a load spike make sense too? I'm not sure about that bit yet. > Or do some cheesy default and document the options to change > it clearly and wait for feedback from users on what works for > them? I think this makes sense. An initial upper limit of 1 million hash table slots seems reasonable at first. That's 8MB on a 64-bit machine.