From mboxrd@z Thu Jan 1 00:00:00 1970 From: Peter Zijlstra Subject: Re: What if a TLB flush needed to sleep? Date: Thu, 27 Mar 2008 14:20:52 +0100 Message-ID: <1206624052.8514.570.camel@twins> References: <1FE6DD409037234FAB833C420AA843ECE9DF60@orsmsx424.amr.corp.intel.com> Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: Sender: linux-arch-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-ID: To: Christoph Lameter Cc: Thomas Gleixner , "Luck, Tony" , linux-arch-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org, linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org On Wed, 2008-03-26 at 18:19 -0700, Christoph Lameter wrote: > On Wed, 26 Mar 2008, Thomas Gleixner wrote: > > > Please use a mutex, not a semaphore. semaphores should only be used > > when you need a counting sempahore. > > Seems that mutexes are mainly useful for 2 processor systems since they > do not allow concurrent read sections. We want multiple processors able > to reclaim pages within the same vma or file concurrently. This means > processors need to be able to concurrently walk potentially long lists of > vmas. confusion between semaphores and rwsems From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from pentafluge.infradead.org ([213.146.154.40]:40731 "EHLO pentafluge.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1758985AbYC0NV1 (ORCPT ); Thu, 27 Mar 2008 09:21:27 -0400 Subject: Re: What if a TLB flush needed to sleep? From: Peter Zijlstra In-Reply-To: References: <1FE6DD409037234FAB833C420AA843ECE9DF60@orsmsx424.amr.corp.intel.com> Content-Type: text/plain Date: Thu, 27 Mar 2008 14:20:52 +0100 Message-ID: <1206624052.8514.570.camel@twins> Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Sender: linux-arch-owner@vger.kernel.org List-ID: To: Christoph Lameter Cc: Thomas Gleixner , "Luck, Tony" , linux-arch@vger.kernel.org, linux-mm@kvack.org, linux-kernel@vger.kernel.org Message-ID: <20080327132052.RaoLdB3GENAOa4uKqjuqZ4HDMvbZ0A3HWiFEqiWrxSc@z> On Wed, 2008-03-26 at 18:19 -0700, Christoph Lameter wrote: > On Wed, 26 Mar 2008, Thomas Gleixner wrote: > > > Please use a mutex, not a semaphore. semaphores should only be used > > when you need a counting sempahore. > > Seems that mutexes are mainly useful for 2 processor systems since they > do not allow concurrent read sections. We want multiple processors able > to reclaim pages within the same vma or file concurrently. This means > processors need to be able to concurrently walk potentially long lists of > vmas. confusion between semaphores and rwsems