[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <1351691516.4004.80.camel@gandalf.local.home>
Date:	Wed, 31 Oct 2012 09:51:56 -0400
From:	Steven Rostedt <rostedt@...dmis.org>
To:	anish kumar <anish198519851985@...il.com>
Cc:	Frederic Weisbecker <fweisbec@...il.com>,
	Paul McKenney <paulmck@...ux.vnet.ibm.com>,
	Peter Zijlstra <peterz@...radead.org>,
	LKML <linux-kernel@...r.kernel.org>,
	Ingo Molnar <mingo@...nel.org>,
	Thomas Gleixner <tglx@...utronix.de>,
	Andrew Morton <akpm@...ux-foundation.org>,
	Paul Gortmaker <paul.gortmaker@...driver.com>
Subject: Re: [PATCH 2/2] irq_work: Fix racy IRQ_WORK_BUSY flag setting
On Wed, 2012-10-31 at 20:04 +0900, anish kumar wrote:
> > This now does:
> > 
> > 	CPU 1					CPU 2
> > 	-----					-----
> > 	(flags = 0)
> > 	cmpxchg(flags, 0, IRQ_WORK_FLAGS)
> > 	(flags = 3)
> > 	[...]
We can still add here:
						(fetch flags)
> > 	xchg(&flags, IRQ_WORK_BUSY)
> > 	(flags = 2)
> > 	func()
> > 						oflags = cmpxchg(&flags, flags, nflags);
Then the cmpxchg() would fail, and oflags would be 2
> > 						(sees flags = 2)
> > 						if (flags & IRQ_WORK_PENDING)
This should be:
						  if (oflags & IRQ_WORK_PENDING)
> > 							(not true)
> > 						(loop)						
> > 	cmpxchg(flags, 2, 0);
> > 	(flags = 2)
This should be:
	  (flags = 0)
as we described the previous cmpxchg() as failing, flags would still be
2 before this cmpxchg(), and this one would succeed.
-- Steve
> > 						flags = 3
> > 
> > 
> > 
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/
Powered by blists - more mailing lists