From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752834AbZLAREJ (ORCPT ); Tue, 1 Dec 2009 12:04:09 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1752187AbZLAREI (ORCPT ); Tue, 1 Dec 2009 12:04:08 -0500 Received: from moutng.kundenserver.de ([212.227.17.8]:61795 "EHLO moutng.kundenserver.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752359AbZLAREH (ORCPT ); Tue, 1 Dec 2009 12:04:07 -0500 From: Arnd Bergmann To: Linus Torvalds Subject: Re: [rfc] "fair" rw spinlocks Date: Tue, 1 Dec 2009 18:03:24 +0100 User-Agent: KMail/1.12.2 (Linux/2.6.31-14-generic; KDE/4.3.2; x86_64; ; ) Cc: Nick Piggin , "Paul E. McKenney" , Linux Kernel Mailing List References: <20091123145409.GA29627@wotan.suse.de> <20091130171333.GG21639@wotan.suse.de> In-Reply-To: MIME-Version: 1.0 Content-Type: Text/Plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Message-Id: <200912011803.24664.arnd@arndb.de> X-Provags-ID: V01U2FsdGVkX19ab+QImrZPJan+7V/z0Hf0krMIi+efW9QI/Xc zeZUBj9a/KBolNn48cacAKL5tZXMFHv5mIxWj7sbNBQVKZBwYx nzR5mVkE5LW/f1UFn6WhQ== Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Monday 30 November 2009, Linus Torvalds wrote: > The best option really would be to try to make it all use RCU, rather than > paper over things. That really should improve performance. Are there any writers at interrupt time? If not, another option might be to first convert all the readers that can happen from interrupts to RCU, which lets us get rid of the irq disable in the write path. Step two could either be the conversion of all nested readers to RCU, or your tasklist_read_lock(), which should then be safe. Doing both of these lets us turn tasklist_lock into a plain spinlock, possibly followed by converting remaining readers that show measurable lock contention. Arnd <><