From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andi Kleen Subject: Re: [PATCH 00/25] mm: Preemptibility -v7 Date: Tue, 25 Jan 2011 11:45:28 -0800 Message-ID: References: <20110125173111.720927511@chello.nl> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from lo.gmane.org ([80.91.229.12]:56031 "EHLO lo.gmane.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751828Ab1AYUkF (ORCPT ); Tue, 25 Jan 2011 15:40:05 -0500 Received: from list by lo.gmane.org with local (Exim 4.69) (envelope-from ) id 1PhpgJ-0006gi-8W for linux-arch@vger.kernel.org; Tue, 25 Jan 2011 21:40:03 +0100 Received: from jfdmzpr02-ext.jf.intel.com ([134.134.137.71]) by main.gmane.org with esmtp (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Tue, 25 Jan 2011 21:40:03 +0100 Received: from ak by jfdmzpr02-ext.jf.intel.com with local (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Tue, 25 Jan 2011 21:40:03 +0100 Sender: linux-arch-owner@vger.kernel.org List-ID: To: linux-arch@vger.kernel.org Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.orglinux-kernel@vger.kernel.org Peter Zijlstra writes: > This patch-set makes part of the mm a lot more preemptible. It converts > i_mmap_lock and anon_vma->lock to mutexes and makes mmu_gather fully > preemptible. > > The main motivation was making mm_take_all_locks() preemptible, since it > appears people are nesting hundreds of spinlocks there. Just curious: why is mm_take_all_locks() a problem? As far as I can see it's just used when starting KVM or GRU the first time. Is that a common situation? -Andi -- Andi Kleen Intel Open Source Technology Center From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753702Ab1AYUpJ (ORCPT ); Tue, 25 Jan 2011 15:45:09 -0500 Received: from lo.gmane.org ([80.91.229.12]:56205 "EHLO lo.gmane.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753409Ab1AYUpH (ORCPT ); Tue, 25 Jan 2011 15:45:07 -0500 X-Injected-Via-Gmane: http://gmane.org/ To: linux-kernel@vger.kernel.org From: Andi Kleen Subject: Re: [PATCH 00/25] mm: Preemptibility -v7 Date: Tue, 25 Jan 2011 11:45:28 -0800 Message-ID: References: <20110125173111.720927511@chello.nl> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Complaints-To: usenet@dough.gmane.org Cc: linux-mm@kvack.org, linux-arch@vger.kernel.org X-Gmane-NNTP-Posting-Host: jfdmzpr02-ext.jf.intel.com User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/23.2 (gnu/linux) Cancel-Lock: sha1:po9hh0br9vzHOJ/DqLJuEZ77lE8= Cc: linux-kernel@vger.kernel.org, linux-arch@vger.kernel.org Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Peter Zijlstra writes: > This patch-set makes part of the mm a lot more preemptible. It converts > i_mmap_lock and anon_vma->lock to mutexes and makes mmu_gather fully > preemptible. > > The main motivation was making mm_take_all_locks() preemptible, since it > appears people are nesting hundreds of spinlocks there. Just curious: why is mm_take_all_locks() a problem? As far as I can see it's just used when starting KVM or GRU the first time. Is that a common situation? -Andi -- Andi Kleen Intel Open Source Technology Center From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail144.messagelabs.com (mail144.messagelabs.com [216.82.254.51]) by kanga.kvack.org (Postfix) with SMTP id 992A46B0092 for ; Tue, 25 Jan 2011 15:50:14 -0500 (EST) Received: from list by lo.gmane.org with local (Exim 4.69) (envelope-from ) id 1Phpq2-0003yG-0l for linux-mm@kvack.org; Tue, 25 Jan 2011 21:50:06 +0100 Received: from jfdmzpr02-ext.jf.intel.com ([134.134.137.71]) by main.gmane.org with esmtp (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Tue, 25 Jan 2011 21:50:06 +0100 Received: from ak by jfdmzpr02-ext.jf.intel.com with local (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Tue, 25 Jan 2011 21:50:06 +0100 From: Andi Kleen Subject: Re: [PATCH 00/25] mm: Preemptibility -v7 Date: Tue, 25 Jan 2011 11:45:28 -0800 Message-ID: References: <20110125173111.720927511@chello.nl> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: owner-linux-mm@kvack.org To: linux-mm@kvack.org Cc: linux-kernel@vger.kernel.org, linux-arch@vger.kernel.org List-ID: Peter Zijlstra writes: > This patch-set makes part of the mm a lot more preemptible. It converts > i_mmap_lock and anon_vma->lock to mutexes and makes mmu_gather fully > preemptible. > > The main motivation was making mm_take_all_locks() preemptible, since it > appears people are nesting hundreds of spinlocks there. Just curious: why is mm_take_all_locks() a problem? As far as I can see it's just used when starting KVM or GRU the first time. Is that a common situation? -Andi -- Andi Kleen Intel Open Source Technology Center -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Fight unfair telecom policy in Canada: sign http://dissolvethecrtc.ca/ Don't email: email@kvack.org