From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1759285AbXGXVgp (ORCPT ); Tue, 24 Jul 2007 17:36:45 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753734AbXGXVgi (ORCPT ); Tue, 24 Jul 2007 17:36:38 -0400 Received: from gate.crashing.org ([63.228.1.57]:53731 "EHLO gate.crashing.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753531AbXGXVgh (ORCPT ); Tue, 24 Jul 2007 17:36:37 -0400 Subject: Re: [PATCH 6/8] i386: bitops: Don't mark memory as clobbered unnecessarily From: Benjamin Herrenschmidt To: Linus Torvalds Cc: Satyam Sharma , Linux Kernel Mailing List , David Howells , Nick Piggin , Andi Kleen , Andrew Morton In-Reply-To: References: <20070723160528.22137.84144.sendpatchset@cselinux1.cse.iitk.ac.in> <20070723160558.22137.71943.sendpatchset@cselinux1.cse.iitk.ac.in> <1185270756.5439.256.camel@localhost.localdomain> Content-Type: text/plain Date: Wed, 25 Jul 2007 07:36:17 +1000 Message-Id: <1185312977.5439.281.camel@localhost.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.10.1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Tue, 2007-07-24 at 10:24 -0700, Linus Torvalds wrote: > > On Tue, 24 Jul 2007, Benjamin Herrenschmidt wrote: > > > > In fact, it's more than that... the bitops that return a value are often > > used to have hand-made spinlock semantics. I'm sure we would get funky > > bugs if loads or stores leaked out of the locked region. I think a full > > "memory" clobber should be kept around for those cases. > > Not helpful. > > The CPU ordering constraints for "test_and_set_bit()" and friends are weak > enough that even if you have a full memory clobber, it still wouldn't work > as a lock. > > That's exactly why we have smp_mb__after_set_bit() and friends. On some > architectures (arm, mips, parsic, powerpc), *that* is where the CPU memory > barrier is, because the "test_and_set_bit()" itself is just a > cache-coherent operation, not an actual barrier. Well, as I said, our test_and_set_bit() asm (and in general, the asm for all the atomic ops that -return- a value) have at least some level of barriers in them because of that. We do that because people are abusing them as locks. The smp_mb__after_set_bit() I never quite grokked. We do an mb in there but I suspect we don't need if it's only ever used after test_and_set_bit() because of the above. The smb_mb__before_clear_bit() makes more sense as it's supposed to give clear_bit() a spin_unlock semantic. But we do need the "memory" clobber as well. That's one reason why I like Nick's bitop locks patches, providing -explicit- test_and_set_bit_lock() and clear_bit_unlock(), we can fix a whole lot of things and make sure they have the right barriers and not more. (We save a few useless barriers on POWER that way in the page lock path and it's measureable in his benchmark). Cheers, Ben.