From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1759184Ab0EZB2R (ORCPT ); Tue, 25 May 2010 21:28:17 -0400 Received: from fgwmail5.fujitsu.co.jp ([192.51.44.35]:45607 "EHLO fgwmail5.fujitsu.co.jp" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1759142Ab0EZB2P (ORCPT ); Tue, 25 May 2010 21:28:15 -0400 X-SecurityPolicyCheck-FJ: OK by FujitsuOutboundMailChecker v1.3.1 Date: Wed, 26 May 2010 10:23:59 +0900 From: KAMEZAWA Hiroyuki To: Peter Zijlstra Cc: Michel Lespinasse , Linus Torvalds , David Howells , Ingo Molnar , Thomas Gleixner , LKML , Andrew Morton , Mike Waychison , Suleiman Souhlal , Ying Han Subject: Re: [PATCH 00/11] V4: rwsem changes + down_read_critical() proposal Message-Id: <20100526102359.da18b89e.kamezawa.hiroyu@jp.fujitsu.com> In-Reply-To: <1274779675.5882.684.camel@twins> References: <1274733081-4623-1-git-send-email-walken@google.com> <1274777239.5882.581.camel@twins> <1274779675.5882.684.camel@twins> Organization: FUJITSU Co. LTD. X-Mailer: Sylpheed 3.0.2 (GTK+ 2.10.14; i686-pc-mingw32) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, 25 May 2010 11:27:55 +0200 Peter Zijlstra wrote: > On Tue, 2010-05-25 at 02:12 -0700, Michel Lespinasse wrote: > > On Tue, May 25, 2010 at 1:47 AM, Peter Zijlstra wrote: > > > So what happened to those patches that dropped mmap_sem during I/O? > > > > Yes, we do have patches trying to release the mmap_sem when a page > > fault for a file backed VMA blocks on accessing the corresponding > > file. We have not given up on these, and we intend to try submitting > > them again. However, these patches do *not* address the case of a page > > fault blocking while trying to get a free page (i.e. when you get > > under high memory pressure). > > But I guess they could, right? Simply make the allocation under mmap_sem > be __GFP_HARDWALL|__GFP_HIGHMEM|__GFP_MOVABLE__GFP_NOWARN or > (GFP_HUGHUSER_MOVABLE & ~(__GFP_WAIT|__GFP_IO|__GFP_FS))|__GFP_NOWARN > > and drop the mmap_sem when that fails. > > > > I really don't like people tinkering with the lock implementations like > > > this. Nor do I like the naming, stats are in no way _critical_. > > > > Critical here refers to the fact that you're not allowed to block > > while holding the unfairly acquired rwsem. > > We usually call that atomic, your 0/n patch didn't explain any of that. > > Also, do you really think doing something like: > > /* > * Check the vma index is within the range and do > * sequential scan until m_index. > */ > vma = NULL; > if ((unsigned long)l < mm->map_count) { > vma = mm->mmap; > while (l-- && vma) > vma = vma->vm_next; > goto out; > } > > with preemption disabled is a _good_ thing? > > People were talking about raising our vma limit of 64k... Yes, at least, my customers want over 64k vmas ;) Hmm..can't we use something speculative lookup for reading maps ? (as we played in several months ago...) Thanks, -Kame