Your revised version has the disadvantage of not working.
Consider a lock word that's initially 1:
Task 1: Task 2:
cmpb $0,%0 cmpb $0,%0 Both see it as equal to 1
jle 2f jle 2f Both no NOT jump
movb $0, %0 movb $0, %0 Both overwrite it with zero.
When releasing a lock, there is no race condition because only one
task *has* the lock. But when acquiring it, you need the (expensive0
atomic test-and-set to synchronize between different processors.
-
To unsubscribe from this list: send the line "unsubscribe linux-net" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
