From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756116Ab1LGNqN (ORCPT ); Wed, 7 Dec 2011 08:46:13 -0500 Received: from mailout-de.gmx.net ([213.165.64.23]:53822 "HELO mailout-de.gmx.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with SMTP id S1755801Ab1LGNqL (ORCPT ); Wed, 7 Dec 2011 08:46:11 -0500 X-Authenticated: #14349625 X-Provags-ID: V01U2FsdGVkX1/6iWJzNEH6xTS8fjy4BwwrDUwYtubPbC4oRCqoGk EEWlnXhKZyaQYS Subject: Re: [patch] intel_idle: convert i7300_idle_lock to raw spinlock From: Mike Galbraith To: Steven Rostedt Cc: linux-rt-users , Thomas Gleixner , LKML In-Reply-To: <1323263113.2832.7.camel@pippen.local.home> References: <1323258522.5057.73.camel@marge.simson.net> <1323263113.2832.7.camel@pippen.local.home> Content-Type: text/plain; charset="UTF-8" Date: Wed, 07 Dec 2011 14:46:00 +0100 Message-ID: <1323265560.5057.99.camel@marge.simson.net> Mime-Version: 1.0 X-Mailer: Evolution 2.32.1 Content-Transfer-Encoding: 7bit X-Y-GMX-Trusted: 0 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, 2011-12-07 at 08:05 -0500, Steven Rostedt wrote: > Looks as this should also go mainline (Cc'd LKML) > > Note, as I want to release 3.0.12-rt30 today, this wont make it. It will > have to wait till -rt31. No hurry. (20 synchronized core jitter test was horrid). > -- Steve > > On Wed, 2011-12-07 at 12:48 +0100, Mike Galbraith wrote: > > 24 core Intel box's first exposure to 3.0.12-rt30-rc3 didn't go well. > > > > [ 27.104159] i7300_idle: loaded v1.55 > > [ 27.104192] BUG: scheduling while atomic: swapper/2/0/0x00000002 > > [ 27.104309] Pid: 0, comm: swapper/2 Tainted: G N 3.0.12-rt30-rc3-rt #1 > > [ 27.104317] Call Trace: > > [ 27.104338] [] dump_trace+0x85/0x2e0 > > [ 27.104372] [] thread_return+0x12b/0x30b > > [ 27.104381] [] schedule+0x29/0xb0 > > [ 27.104389] [] rt_spin_lock_slowlock+0xc5/0x240 > > [ 27.104401] [] i7300_idle_notifier+0x3f/0x360 [i7300_idle] > > [ 27.104415] [] notifier_call_chain+0x37/0x70 > > [ 27.104426] [] __atomic_notifier_call_chain+0x48/0x70 > > [ 27.104439] [] cpu_idle+0x89/0xb0 > > [ 27.104449] bad: scheduling from the idle thread! > > > > Signed-off-by: Mike Galbraith > > > > --- > > drivers/idle/i7300_idle.c | 8 ++++---- > > 1 file changed, 4 insertions(+), 4 deletions(-) > > > > Index: linux-3.2-rt/drivers/idle/i7300_idle.c > > =================================================================== > > --- linux-3.2-rt.orig/drivers/idle/i7300_idle.c > > +++ linux-3.2-rt/drivers/idle/i7300_idle.c > > @@ -75,7 +75,7 @@ static unsigned long past_skip; > > > > static struct pci_dev *fbd_dev; > > > > -static spinlock_t i7300_idle_lock; > > +static raw_spinlock_t i7300_idle_lock; > > static int i7300_idle_active; > > > > static u8 i7300_idle_thrtctl_saved; > > @@ -457,7 +457,7 @@ static int i7300_idle_notifier(struct no > > idle_begin_time = ktime_get(); > > } > > > > - spin_lock_irqsave(&i7300_idle_lock, flags); > > + raw_spin_lock_irqsave(&i7300_idle_lock, flags); > > if (val == IDLE_START) { > > > > cpumask_set_cpu(smp_processor_id(), idle_cpumask); > > @@ -506,7 +506,7 @@ static int i7300_idle_notifier(struct no > > } > > } > > end: > > - spin_unlock_irqrestore(&i7300_idle_lock, flags); > > + raw_spin_unlock_irqrestore(&i7300_idle_lock, flags); > > return 0; > > } > > > > @@ -554,7 +554,7 @@ struct debugfs_file_info { > > > > static int __init i7300_idle_init(void) > > { > > - spin_lock_init(&i7300_idle_lock); > > + raw_spin_lock_init(&i7300_idle_lock); > > total_us = 0; > > > > if (i7300_idle_platform_probe(&fbd_dev, &ioat_dev, forceload)) > > > >