From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752314Ab2LDUXS (ORCPT ); Tue, 4 Dec 2012 15:23:18 -0500 Received: from mx1.redhat.com ([209.132.183.28]:24176 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751144Ab2LDUXR (ORCPT ); Tue, 4 Dec 2012 15:23:17 -0500 From: Jeff Moyer To: Jens Axboe Cc: "linux-kernel\@vger.kernel.org" , "linux-mm\@kvack.org" , Zach Brown Subject: Re: [patch,v2] bdi: add a user-tunable cpu_list for the bdi flusher threads References: <50BE5988.3050501@fusionio.com> X-PGP-KeyID: 1F78E1B4 X-PGP-CertKey: F6FE 280D 8293 F72C 65FD 5A58 1FF8 A7CA 1F78 E1B4 X-PCLoadLetter: What the f**k does that mean? Date: Tue, 04 Dec 2012 15:23:13 -0500 In-Reply-To: <50BE5988.3050501@fusionio.com> (Jens Axboe's message of "Tue, 4 Dec 2012 21:14:00 +0100") Message-ID: User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Jens Axboe writes: > On 2012-12-03 19:53, Jeff Moyer wrote: >> Hi, >> >> In realtime environments, it may be desirable to keep the per-bdi >> flusher threads from running on certain cpus. This patch adds a >> cpu_list file to /sys/class/bdi/* to enable this. The default is to tie >> the flusher threads to the same numa node as the backing device (though >> I could be convinced to make it a mask of all cpus to avoid a change in >> behaviour). > > Looks sane, and I think defaulting to the home node is a sane default. > One comment: > >> + ret = cpulist_parse(buf, newmask); >> + if (!ret) { >> + spin_lock(&bdi->wb_lock); >> + task = wb->task; >> + if (task) >> + get_task_struct(task); >> + spin_unlock(&bdi->wb_lock); > > bdi->wb_lock needs to be bh safe. The above should have caused lockdep > warnings for you. No lockdep complaints. I'll double check that's enabled (but I usually have it enabled...). >> @@ -437,6 +488,14 @@ static int bdi_forker_thread(void *ptr) >> spin_lock_bh(&bdi->wb_lock); >> bdi->wb.task = task; >> spin_unlock_bh(&bdi->wb_lock); >> + mutex_lock(&bdi->flusher_cpumask_mutex); >> + ret = set_cpus_allowed_ptr(task, >> + bdi->flusher_cpumask); >> + mutex_unlock(&bdi->flusher_cpumask_mutex); > > It'd be very useful if we had a kthread_create_cpu_on_cpumask() instead > of a _node() variant, since the latter could easily be implemented on > top of the former. But not really a show stopper for the patch... Hmm, if it isn't too scary, I might give this a try. Thanks! Jeff