From mboxrd@z Thu Jan 1 00:00:00 1970 Received: by rn-out-0910.google.com with SMTP id j40so45004rnf.4 for ; Tue, 29 Apr 2008 07:52:59 -0700 (PDT) Message-ID: <661de9470804290752w1dc0cfb3k72e81d828a45765e@mail.gmail.com> Date: Tue, 29 Apr 2008 20:22:58 +0530 From: "Balbir Singh" Subject: Re: Page Faults slower in 2.6.25-rc9 than 2.6.23 In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline References: Sender: owner-linux-mm@kvack.org Return-Path: To: Hugh Dickins Cc: Ross Biro , linux-mm@kvack.org, lkml , Kamalesh Babulal List-ID: On Tue, Apr 29, 2008 at 7:27 PM, Hugh Dickins wrote: > On Tue, 29 Apr 2008, Ross Biro wrote: > > I don't know if this has been noticed before. I was benchmarking my > > page table relocation code and I noticed that on 2.6.25-rc9 page > > faults take 10% more time than on 2.6.22. This is using lmbench > > running on an intel x86_64 system. The good news is that the page > > table relocation code now only adds a 1.6% slow down to page faults. > > Do you have CONFIG_CGROUP_MEM_RES_CTLR=y in 2.6.25? > That added about 20% to my lmbench "Page Fault" tests (with > adverse effect on several others e.g. the fork, exec, sh group). > Hmm.. strange.. I don't remember the overhead being so bad (I'll relook at my old numbers). I'll try and git-bisect this one > Try the same kernel with boot option "cgroup_disable=memory", > that should recoup most (but not quite all) of the slowdown; > or rebuild with n to CGROUP_MEM_RES_CTLR. > > But your "Mmap Latency" went up 425% ?? > That's really way of the mark > Hugh > Balbir -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org