Skip to main content

Browse the archive

Staging
v0.8.1
https://github.com/torvalds/linux
02 December 2023, 22:23:51 UTC
Revision 04e2f1741d235ba599037734878d72e57cb302b5 authored by Linus Torvalds on 24 February 2008, 02:05:03 UTC, committed by Linus Torvalds on 24 February 2008, 02:05:03 UTC 
Add memory barrier semantics to wake_up() & co
 
Oleg Nesterov and others have pointed out that on some architectures,
the traditional sequence of

	set_current_state(TASK_INTERRUPTIBLE);
	if (CONDITION)
		return;
	schedule();

is racy wrt another CPU doing

	CONDITION = 1;
	wake_up_process(p);

because while set_current_state() has a memory barrier separating
setting of the TASK_INTERRUPTIBLE state from reading of the CONDITION
variable, there is no such memory barrier on the wakeup side.

Now, wake_up_process() does actually take a spinlock before it reads and
sets the task state on the waking side, and on x86 (and many other
architectures) that spinlock is in fact equivalent to a memory barrier,
but that is not generally guaranteed.  The write that sets CONDITION
could move into the critical region protected by the runqueue spinlock.

However, adding a smp_wmb() to before the spinlock should now order the
writing of CONDITION wrt the lock itself, which in turn is ordered wrt
the accesses within the spinlock (which includes the reading of the old
state).

This should thus close the race (which probably has never been seen in
practice, but since smp_wmb() is a no-op on x86, it's not like this will
make anything worse either on the most common architecture where the
spinlock already gave the required protection).

Acked-by: Oleg Nesterov <oleg@tv-sign.ru>
Acked-by: Dmitry Adamushko <dmitry.adamushko@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
1 parent 0a3716e
Raw File
Tip revision: 04e2f1741d235ba599037734878d72e57cb302b5 authored by Linus Torvalds on 24 February 2008, 02:05:03 UTC
Add memory barrier semantics to wake_up() & co
Tip revision: 04e2f17
ac97_bus.c 
/*
 * Linux driver model AC97 bus interface
 *
 * Author:	Nicolas Pitre
 * Created:	Jan 14, 2005
 * Copyright:	(C) MontaVista Software Inc.
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation; either version 2 of the License, or
 * (at your option) any later version.
 */

#include <linux/module.h>
#include <linux/init.h>
#include <linux/device.h>
#include <linux/string.h>

/*
 * Let drivers decide whether they want to support given codec from their
 * probe method.  Drivers have direct access to the struct snd_ac97 structure and may
 * decide based on the id field amongst other things.
 */
static int ac97_bus_match(struct device *dev, struct device_driver *drv)
{
	return 1;
}

#ifdef CONFIG_PM
static int ac97_bus_suspend(struct device *dev, pm_message_t state)
{
	int ret = 0;

	if (dev->driver && dev->driver->suspend)
		ret = dev->driver->suspend(dev, state);

	return ret;
}

static int ac97_bus_resume(struct device *dev)
{
	int ret = 0;

	if (dev->driver && dev->driver->resume)
		ret = dev->driver->resume(dev);

	return ret;
}
#endif /* CONFIG_PM */

struct bus_type ac97_bus_type = {
	.name		= "ac97",
	.match		= ac97_bus_match,
#ifdef CONFIG_PM
	.suspend	= ac97_bus_suspend,
	.resume		= ac97_bus_resume,
#endif /* CONFIG_PM */
};

static int __init ac97_bus_init(void)
{
	return bus_register(&ac97_bus_type);
}

subsys_initcall(ac97_bus_init);

static void __exit ac97_bus_exit(void)
{
	bus_unregister(&ac97_bus_type);
}

module_exit(ac97_bus_exit);

EXPORT_SYMBOL(ac97_bus_type);

MODULE_LICENSE("GPL");
back to top