Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756903Ab0KLLGu (ORCPT ); Fri, 12 Nov 2010 06:06:50 -0500 Received: from mail-qw0-f46.google.com ([209.85.216.46]:37744 "EHLO mail-qw0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751457Ab0KLLGt convert rfc822-to-8bit (ORCPT ); Fri, 12 Nov 2010 06:06:49 -0500 DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:content-transfer-encoding; b=sHM15uuixdWoARvBnA862h8/GyuEk8HdueKSi7BDQIOEVbS5csfYoR/+K30ogyvlcZ 04x7kAL/cG8fYTUnigrdi1L6fa3v80jX+Y87kKRmRk8fA/6jw3Ag+ZEgdUYysTOnV4LC 5HrOo6BBNK/pmI5yXIjeJApHk6c2VDVgkl3I4= MIME-Version: 1.0 In-Reply-To: <20101112091818.GB5949@cr0.nay.redhat.com> References: <1289489007.17691.1310.camel@edumazet-laptop> <20101112071323.GB5660@cr0.nay.redhat.com> <1289546874.17691.1774.camel@edumazet-laptop> <20101112081945.GA5949@cr0.nay.redhat.com> <20101112091818.GB5949@cr0.nay.redhat.com> Date: Fri, 12 Nov 2010 19:06:47 +0800 Message-ID: Subject: Re: Kernel rwlock design, Multicore and IGMP From: Cypher Wu To: =?ISO-8859-1?Q?Am=E9rico_Wang?= Cc: Yong Zhang , Eric Dumazet , linux-kernel@vger.kernel.org, netdev Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8BIT Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 3004 Lines: 73 2010/11/12 Am?rico Wang : > On Fri, Nov 12, 2010 at 05:09:45PM +0800, Yong Zhang wrote: >>On Fri, Nov 12, 2010 at 4:19 PM, Am?rico Wang wrote: >>> On Fri, Nov 12, 2010 at 08:27:54AM +0100, Eric Dumazet wrote: >>>>Le vendredi 12 novembre 2010 ? 15:13 +0800, Am?rico Wang a ?crit : >>>>> On Fri, Nov 12, 2010 at 11:32:59AM +0800, Cypher Wu wrote: >>>>> >On Thu, Nov 11, 2010 at 11:23 PM, Eric Dumazet wrote: >>>>> >> Le jeudi 11 novembre 2010 ? 21:49 +0800, Cypher Wu a ?crit : >>>>> >> >>>>> >> Hi >>>>> >> >>>>> >> CC netdev, since you ask questions about network stuff _and_ rwlock >>>>> >> >>>>> >> >>>>> >>> I'm using TILEPro and its rwlock in kernel is a liitle different than >>>>> >>> other platforms. It have a priority for write lock that when tried it >>>>> >>> will block the following read lock even if read lock is hold by >>>>> >>> others. Its code can be read in Linux Kernel 2.6.36 in >>>>> >>> arch/tile/lib/spinlock_32.c. >>>>> >> >>>>> >> This seems a bug to me. >>>>> >> >>>>> >> read_lock() can be nested. We used such a schem in the past in iptables >>>>> >> (it can re-enter itself), >>>>> >> and we used instead a spinlock(), but with many discussions with lkml >>>>> >> and Linus himself if I remember well. >>>>> >> >>>>> >It seems not a problem that read_lock() can be nested or not since >>>>> >rwlock doesn't have 'owner', it's just that should we give >>>>> >write_lock() a priority than read_lock() since if there have a lot >>>>> >read_lock()s then they'll starve write_lock(). >>>>> >We should work out a well defined behavior so all the >>>>> >platform-dependent raw_rwlock has to design under that principle. >>>>> >>>> >>>>AFAIK, Lockdep allows read_lock() to be nested. >>>> >>>>> It is a known weakness of rwlock, it is designed like that. :) >>>>> >>>> >>>>Agreed. >>>> >>> >>> Just for record, both Tile and X86 implement rwlock with a write-bias, >>> this somewhat reduces the write-starvation problem. >> >>Are you sure(on x86)? >> >>It seems that we never realize writer-bias rwlock. >> > > Try > > % grep RW_LOCK_BIAS -nr arch/x86 > > *And* read the code to see how it works. :) > > Note, on Tile, it uses a little different algorithm. > It seems that rwlock on x86 and tile have different behavior, x86 use RW_LOCK_BIAS, when read_lock() it will test if the lock is 0, and if so then the read_lock() have to 'spinning', otherwise it dec the lock; when write_lock() tried it first check if lock is It seems that rwlock on x86 and tile have different behavior, x86 use RW_LOCK_BIAS and if so, set lock to 0 and continue, otherwise it will 'spinning'. I'm not very familiar with x86 architecture, but the code seems like working that way. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/