Date: Mon, 8 Dec 2008 12:23:17 -0500 (EST)
From: Steven Rostedt <rostedt@goodmis.org>
To: Tim Blechmann <tim@klingt.org>
cc: linux-rt-users <linux-rt-users@vger.kernel.org>,
       LKML <linux-kernel@vger.kernel.org>,
       Peter Zijlstra <peterz@infradead.org>,
       Thomas Gleixner <tglx@linutronix.de>, Ingo Molnar <mingo@elte.hu>
Subject: Re: [PATCH] [RT] avoid preemption in memory controller code
In-Reply-To: <1228756056.7862.8.camel@thinkpad>
Message-ID: <alpine.DEB.1.10.0812081218471.26257@gandalf.stny.rr.com>
References: <1228746854-23875-1-git-send-email-tim@klingt.org>  <alpine.DEB.1.10.0812081138470.26257@gandalf.stny.rr.com> <1228756056.7862.8.camel@thinkpad>
User-Agent: Alpine 1.10 (DEB 962 2008-03-14)
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1288
Lines: 49


On Mon, 8 Dec 2008, Tim Blechmann wrote:

> > > the lru_lock of struct mem_group_per_zone is used to avoid preemption
> > > during the __mem_cgroup_stat_add_safe function.
> > > therefore, a raw_spinlock_t should be used.
> > 
> > What is the reason that this must avoid preemption? 
> 
> it guards a call to smp_processor_id() in __mem_cgroup_stat_add_safe().
> see http://article.gmane.org/gmane.linux.rt.user/3690

We could simply create a 
	preempt_disable_rt();
that will only disable preemption when CONFIG_PREEMPT_RT is set.
and then we could add

	int cpu;

	preempt_disable_rt();
	cpu = smp_processor_id();
	stat->cpustat[cpu].count[idx] += val;
	preempt_enable_rt();

Or something similar.

-- Steve

> 
> > Is there another
> > way to solve this? I rather not just add a raw spinlock if we can
> > help it.
> 
> not sure, maybe one can disable preemption for that specific function?
> 
> tim
> 
> --
> tim@klingt.org
> http://tim.klingt.org
> 
> Art is either a complaint or do something else
>   John Cage quoting Jasper Johns
> 
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/