From mboxrd@z Thu Jan 1 00:00:00 1970 From: KAMEZAWA Hiroyuki Date: Mon, 27 Mar 2006 00:16:56 +0000 Subject: Re: git pull on ia64 linux tree Message-Id: <20060327091656.0bb2f2ee.kamezawa.hiroyu@jp.fujitsu.com> List-Id: References: <200504222203.j3MM3fV17003@unix-os.sc.intel.com> In-Reply-To: <200504222203.j3MM3fV17003@unix-os.sc.intel.com> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: linux-ia64@vger.kernel.org On Sun, 26 Mar 2006 11:01:21 -0800 "Chen, Kenneth W" wrote: > Luck, Tony wrote on Friday, March 24, 2006 3:02 PM > Tony, please queue the following two patches in your next update. They still > apply cleanly on Linus' git-tree as of this morning. Thanks. > > [patch] optimize flush_tlb_range on large numa box > http://marc.theaimsgroup.com/?t4168341600005&r=1&w=2 > I love this patch :) Our performance validation team found ia64_global_tlb_purge() performance problem in do_wp_page()/ia64_global_tlb_purge(), copy-on-write. When they does fork() at the rate of 4-500 fork/sec, 40% of cpu time was spent in copy-on-write's ia64_global_tlb_purge. By this, customer's shell-script cannot scale enough. (we asked them to reduce fork()..) (Because they have already removed their test enironment, I can't ask them to try this patch. sorry.) Thanks, --Kame