From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1762121AbXEKODy (ORCPT ); Fri, 11 May 2007 10:03:54 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1758491AbXEKODr (ORCPT ); Fri, 11 May 2007 10:03:47 -0400 Received: from pentafluge.infradead.org ([213.146.154.40]:38406 "EHLO pentafluge.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754699AbXEKODq (ORCPT ); Fri, 11 May 2007 10:03:46 -0400 Date: Fri, 11 May 2007 15:03:37 +0100 From: Christoph Hellwig To: Peter Zijlstra Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org, Oleg Nesterov , Andrew Morton , Ingo Molnar , Thomas Gleixner , Nick Piggin Subject: Re: [PATCH 1/2] scalable rw_mutex Message-ID: <20070511140337.GA3515@infradead.org> Mail-Followup-To: Christoph Hellwig , Peter Zijlstra , linux-mm@kvack.org, linux-kernel@vger.kernel.org, Oleg Nesterov , Andrew Morton , Ingo Molnar , Thomas Gleixner , Nick Piggin References: <20070511131541.992688403@chello.nl> <20070511132321.895740140@chello.nl> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20070511132321.895740140@chello.nl> User-Agent: Mutt/1.4.2.2i X-SRS-Rewrite: SMTP reverse-path rewritten from by pentafluge.infradead.org See http://www.infradead.org/rpr.html Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Fri, May 11, 2007 at 03:15:42PM +0200, Peter Zijlstra wrote: > Scalable reader/writer lock. > > Its scalable in that the read count is a percpu counter and the reader fast > path does not write to a shared cache-line. > > Its not FIFO fair, but starvation proof by alternating readers and writers. While this implementation looks pretty nice I really hate growing more and more locking primitives. Do we have any rwsem user that absolutley needs FIFO semantics or could we convert all user over (in which case the objection above is of course completely moot) From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Fri, 11 May 2007 15:03:37 +0100 From: Christoph Hellwig Subject: Re: [PATCH 1/2] scalable rw_mutex Message-ID: <20070511140337.GA3515@infradead.org> References: <20070511131541.992688403@chello.nl> <20070511132321.895740140@chello.nl> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20070511132321.895740140@chello.nl> Sender: owner-linux-mm@kvack.org Return-Path: To: Peter Zijlstra Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org, Oleg Nesterov , Andrew Morton , Ingo Molnar , Thomas Gleixner , Nick Piggin List-ID: On Fri, May 11, 2007 at 03:15:42PM +0200, Peter Zijlstra wrote: > Scalable reader/writer lock. > > Its scalable in that the read count is a percpu counter and the reader fast > path does not write to a shared cache-line. > > Its not FIFO fair, but starvation proof by alternating readers and writers. While this implementation looks pretty nice I really hate growing more and more locking primitives. Do we have any rwsem user that absolutley needs FIFO semantics or could we convert all user over (in which case the objection above is of course completely moot) -- 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