From mboxrd@z Thu Jan 1 00:00:00 1970 From: William Lee Irwin III Date: Sat, 11 Dec 2004 06:03:12 +0000 Subject: Re: pfault V12 : correction to tasklist rss Message-Id: <20041211060312.GV2714@holomorphy.com> List-Id: References: <20041210133859.2443a856.akpm@osdl.org> In-Reply-To: <20041210133859.2443a856.akpm@osdl.org> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Andrew Morton Cc: Hugh Dickins , clameter@sgi.com, torvalds@osdl.org, benh@kernel.crashing.org, nickpiggin@yahoo.com.au, linux-mm@kvack.org, linux-ia64@vger.kernel.org, linux-kernel@vger.kernel.org Hugh Dickins wrote: >>> We have no real way of establishing the ownership of shared pages >>> anyways. Its counted when allocated. But the page may live on afterwards >>> in another process and then not be accounted for although its only user is >>> the new process. On Fri, Dec 10, 2004 at 01:38:59PM -0800, Andrew Morton wrote: > We did lose some accounting accuracy when the pagetable walk and the big > tasklist walks were removed. Bill would probably have more details. Given > that the code as it stood was a complete showstopper, the tradeoff seemed > reasonable. There are several issues, not listed in order of importance here: (1) Workload monitoring with high multiprogramming levels was infeasible. (2) The long address space walks interfered with mmap() and page faults in the monitored processes, disturbing cluster membership and exceeding maximum response times in monitored workloads. (3) There's a general long-running ongoing effort to take on various places tasklist_lock is abused one-by-one to incrementally resolve or otherwise mitigate the rwlock starvation issues. -- wli From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261893AbULKGDd (ORCPT ); Sat, 11 Dec 2004 01:03:33 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261892AbULKGDd (ORCPT ); Sat, 11 Dec 2004 01:03:33 -0500 Received: from holomorphy.com ([207.189.100.168]:3223 "EHLO holomorphy.com") by vger.kernel.org with ESMTP id S261832AbULKGD2 (ORCPT ); Sat, 11 Dec 2004 01:03:28 -0500 Date: Fri, 10 Dec 2004 22:03:12 -0800 From: William Lee Irwin III To: Andrew Morton Cc: Hugh Dickins , clameter@sgi.com, torvalds@osdl.org, benh@kernel.crashing.org, nickpiggin@yahoo.com.au, linux-mm@kvack.org, linux-ia64@vger.kernel.org, linux-kernel@vger.kernel.org Subject: Re: pfault V12 : correction to tasklist rss Message-ID: <20041211060312.GV2714@holomorphy.com> References: <20041210133859.2443a856.akpm@osdl.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20041210133859.2443a856.akpm@osdl.org> Organization: The Domain of Holomorphy User-Agent: Mutt/1.5.6+20040722i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Hugh Dickins wrote: >>> We have no real way of establishing the ownership of shared pages >>> anyways. Its counted when allocated. But the page may live on afterwards >>> in another process and then not be accounted for although its only user is >>> the new process. On Fri, Dec 10, 2004 at 01:38:59PM -0800, Andrew Morton wrote: > We did lose some accounting accuracy when the pagetable walk and the big > tasklist walks were removed. Bill would probably have more details. Given > that the code as it stood was a complete showstopper, the tradeoff seemed > reasonable. There are several issues, not listed in order of importance here: (1) Workload monitoring with high multiprogramming levels was infeasible. (2) The long address space walks interfered with mmap() and page faults in the monitored processes, disturbing cluster membership and exceeding maximum response times in monitored workloads. (3) There's a general long-running ongoing effort to take on various places tasklist_lock is abused one-by-one to incrementally resolve or otherwise mitigate the rwlock starvation issues. -- wli From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Fri, 10 Dec 2004 22:03:12 -0800 From: William Lee Irwin III Subject: Re: pfault V12 : correction to tasklist rss Message-ID: <20041211060312.GV2714@holomorphy.com> References: <20041210133859.2443a856.akpm@osdl.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20041210133859.2443a856.akpm@osdl.org> Sender: owner-linux-mm@kvack.org Return-Path: To: Andrew Morton Cc: Hugh Dickins , clameter@sgi.com, torvalds@osdl.org, benh@kernel.crashing.org, nickpiggin@yahoo.com.au, linux-mm@kvack.org, linux-ia64@vger.kernel.org, linux-kernel@vger.kernel.org List-ID: Hugh Dickins wrote: >>> We have no real way of establishing the ownership of shared pages >>> anyways. Its counted when allocated. But the page may live on afterwards >>> in another process and then not be accounted for although its only user is >>> the new process. On Fri, Dec 10, 2004 at 01:38:59PM -0800, Andrew Morton wrote: > We did lose some accounting accuracy when the pagetable walk and the big > tasklist walks were removed. Bill would probably have more details. Given > that the code as it stood was a complete showstopper, the tradeoff seemed > reasonable. There are several issues, not listed in order of importance here: (1) Workload monitoring with high multiprogramming levels was infeasible. (2) The long address space walks interfered with mmap() and page faults in the monitored processes, disturbing cluster membership and exceeding maximum response times in monitored workloads. (3) There's a general long-running ongoing effort to take on various places tasklist_lock is abused one-by-one to incrementally resolve or otherwise mitigate the rwlock starvation issues. -- wli -- 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: aart@kvack.org