From mboxrd@z Thu Jan 1 00:00:00 1970 From: Chris Snook Subject: Re: [PATCH 0/24] make atomic_read() behave consistently across all architectures Date: Thu, 16 Aug 2007 15:55:49 -0400 Message-ID: <46C4ABC5.0@redhat.com> References: <20070816003948.GY9645@linux.vnet.ibm.com> <18115.44462.622801.683446@cargo.ozlabs.ibm.com> <20070816020042.GA30650@gondor.apana.org.au> <18115.45316.702491.681906@cargo.ozlabs.ibm.com> <18115.52863.638655.658466@cargo.ozlabs.ibm.com> <20070816053945.GB32442@gondor.apana.org.au> <18115.62741.807704.969977@cargo.ozlabs.ibm.com> <20070816070907.GA964@gondor.apana.org.au> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Herbert Xu , Paul Mackerras , Satyam Sharma , Christoph Lameter , "Paul E. McKenney" , Stefan Richter , Linux Kernel Mailing List , linux-arch@vger.kernel.org, Linus Torvalds , Netdev , Andrew Morton , ak@suse.de, heiko.carstens@de.ibm.com, David Miller , schwidefsky@de.ibm.com, wensong@linux-vs.org, horms@verge.net.au, wjiang@resilience.com, cfriesen@nortel.com, zlynx@acm.org, rpjday@mindspring.com, jesper.juhl@gmail.com, segher@kernel.crashing.org To: =?ISO-8859-1?Q?Ilpo_J=E4rvinen?= Return-path: Received: from mx1.redhat.com ([66.187.233.31]:41452 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750925AbXHPUDW (ORCPT ); Thu, 16 Aug 2007 16:03:22 -0400 In-Reply-To: Sender: netdev-owner@vger.kernel.org List-Id: netdev.vger.kernel.org Ilpo J=E4rvinen wrote: > On Thu, 16 Aug 2007, Herbert Xu wrote: >=20 >> We've been through that already. If it's a busy-wait it >> should use cpu_relax.=20 >=20 > I looked around a bit by using some command lines and ended up wonder= ing=20 > if these are equal to busy-wait case (and should be fixed) or not: >=20 > ./drivers/telephony/ixj.c > 6674: while (atomic_read(&j->DSPWrite) > 0) > 6675- atomic_dec(&j->DSPWrite); >=20 > ...besides that, there are couple of more similar cases in the same f= ile=20 > (with braces)... atomic_dec() already has volatile behavior everywhere, so this is=20 semantically okay, but this code (and any like it) should be calling=20 cpu_relax() each iteration through the loop, unless there's a compellin= g=20 reason not to. I'll allow that for some hardware drivers (possibly thi= s=20 one) such a compelling reason may exist, but hardware-independent core=20 subsystems probably have no excuse. If the maintainer of this code doesn't see a compelling reason not to=20 add cpu_relax() in this loop, then it should be patched. -- Chris