From mboxrd@z Thu Jan 1 00:00:00 1970 From: Patrick McHardy Subject: Re: Fw: 2.6.15-rc2-mm1 problems (get_cpu / NUMA optimizations) Date: Fri, 25 Nov 2005 03:18:48 +0100 Message-ID: <43867488.1030700@trash.net> References: <20051123.155603.64967447.davem@davemloft.net> <20051124205828.GD31478@sunbeam.de.gnumonks.org> <43863A93.2010406@cosmosbay.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: Harald Welte , Netfilter Development Mailinglist , "David S. Miller" Return-path: To: Eric Dumazet In-Reply-To: <43863A93.2010406@cosmosbay.com> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: netfilter-devel-bounces@lists.netfilter.org Errors-To: netfilter-devel-bounces@lists.netfilter.org List-Id: netfilter-devel.vger.kernel.org Eric Dumazet wrote: > Hum... I was not aware that disabling preemption (in get_cpu()) had this > effect (ie not allowing to sleep). I was using get_cpu() to bind the > thread to the current cpu, maybe there is something better ? set_cpus_allowed() might work. > If not, we can zap get_cpu()/pu_cpu() and use raw_smp_processor_id() as > a hint only, (and dont care of preemption/cpu migration), as the value > is only used as a hint to speedup operations in do_replace(), > do_add_counters(), ipt_unregister_table(), translate_table()... That sounds like a good way. iptables ruleset replacement is and will always be slow anyway.