On Sun, 24 Feb 2008, Rafael J. Wysocki wrote:

quoted text
> > Very subtly wrong ;-).
> > 
> > imagine suspending_task == 0xabcdef01. Now task "R" with current ==
> > 0xabcd0000 reads suspending_task while the other cpu is writing to it,
> > and sees 0xabcd0000 (0xef01 was not yet written) -- and mistakenly
> > believes that  "R" == suspending_task.
> > 
> > I agree it is very unlikely, and it will not happen on i386. But what
> > about just using atomic_t suspending_task, and store current->pid into
> > it?
> 
> I'd rather use a lock, frankly.  For example, we can require the readers to
> take pm_sleep_rwsem for reading in order to access that.

That certainly won't work.  Imagine what would happen when the reader 
_was_ the suspending task.

But if you really twist my arm, I'll go along with Pavel's suggestion.

Alan Stern

--
unsubscribe notice
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/