From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Sat, 4 Aug 2007 21:16:38 +0200 From: Ingo Molnar Subject: Re: [PATCH 00/23] per device dirty throttling -v8 Message-ID: <20070804191638.GA23603@elte.hu> References: <20070803123712.987126000@chello.nl> <20070804063217.GA25069@elte.hu> <20070804070737.GA940@elte.hu> <20070804103347.GA1956@elte.hu> <20070804163733.GA31001@elte.hu> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: Sender: owner-linux-mm@kvack.org Return-Path: To: Linus Torvalds Cc: Peter Zijlstra , linux-mm@kvack.org, Linux Kernel Mailing List , miklos@szeredi.hu, akpm@linux-foundation.org, neilb@suse.de, dgc@sgi.com, tomoki.sekiyama.qu@hitachi.com, nikita@clusterfs.com, trond.myklebust@fys.uio.no, yingchao.zhou@gmail.com, richard@rsk.demon.co.uk, david@lang.hm List-ID: * Linus Torvalds wrote: > I ended up using O_NOATIME for the individual object "open()" calls > inside git, and it was an absolutely huge time-saver for the case of > not having "noatime" in the mount options. Certainly more than your > estimated 10% under some loads. yeah, it's much more than that for file-intense stuff like git. My 20-30% estimation was for typical stuff that a desktop user does and which is quite often IO bound. I suspect bootup would be faster with noatime as well. atime is so rarely used that it should be an opt-in thing not an opt-out thing. It should be an opt-in thing even if it had 10 times more applications using it. Ingo -- 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