From mboxrd@z Thu Jan 1 00:00:00 1970 From: Waiman Long Subject: Re: [PATCH v2] introduce atomic_pointer to fix a race condition in cancelable mcs spinlocks Date: Mon, 02 Jun 2014 13:42:34 -0400 Message-ID: <538CB78A.8090500@hp.com> References: <20140602162525.GH16155@laptop.programming.kicks-ass.net> <1401729145.5185.5.camel@buesod1.americas.hpqcorp.net> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Cc: Linus Torvalds , Peter Zijlstra , Mikulas Patocka , "James E.J. Bottomley" , Helge Deller , John David Anglin , Parisc List , Linux Kernel Mailing List , Paul McKenney , "Vinod, Chegu" , Thomas Gleixner , Rik van Riel , Andrew Morton , Peter Anvin , Andi Kleen , "Chandramouleeswaran, Aswin" , "Norton, Scott J" , Jason Low To: Davidlohr Bueso Return-path: In-Reply-To: <1401729145.5185.5.camel@buesod1.americas.hpqcorp.net> List-ID: List-Id: linux-parisc.vger.kernel.org On 06/02/2014 01:12 PM, Davidlohr Bueso wrote: > On Mon, 2014-06-02 at 10:09 -0700, Linus Torvalds wrote: >> On Mon, Jun 2, 2014 at 9:25 AM, Peter Zijlstra wrote: >>> And I can't say I'm a particular fan of these ops either, as alternative >>> I'm almost inclined to just exclude parisc from using opt spinning. >> Please do. > I agree, this is the best way out of this mess. Furthermore, it would > also be nice to consolidate opt spinning in a common CONFIG option -- > right now mutexes and rwsems create their own dependencies. > I would suggest adding a RWSEM_SPIN_ON_OWNER to control opt spinning in rwsem. Currently MUTEX_SPIN_ON_OWNER is doing that for mutex, and it is disabled when mutex debugging is turned on. So I think it is better to allow them to be disabled separately. -Longman