Re: enhanced spinlock debugging code for intel

[Tachino Nobuhiro]

Hello,

At Thu, 02 Aug 2001 13:44:24 -0400,
Brent Baccala wrote:
> 
> No longer any race (right?) and we don't lose anything since the one
> processor is about to drop the lock it (presumably) held.  I wonder if
> it should check to make sure the same processor that grabbed the lock is
> releasing it.  Not exactly a bug, and somebody might write code like
> that, but it seems suspicious.  Comments?

  last_lock_processor should be initialized as -1, or you have another
race.

	cpu0					cpu1
----------------------------------------------------------------------------------------

						last_lock_processor is 0

						call spin_unlock...

						asm(spin_lock_string)
						
	call spin_lock...

	if (spin_is_locked(lock)
	  && lock->last_lock_processor == my_processor_id)

	-------------------------------------------
	last_lock_processor and my_processor_id are
	both 0 here. So BUG() is called incorrectly.
	-------------------------------------------
						
	asm(spin_lock_string)

					        last_lock_processor = 1;