Subj : Re: Lock Free -- where to start
To   : comp.programming.threads
From : Oliver S.
Date : Tue Oct 04 2005 07:50 pm

> Moreover, some "lock free papers" define the "lock free" term in
> such a way that old good spin _locks_ become lock free synchroni-
> zation primitives!

Yes, that's wrong. Lock-free algorithms are by definition algorithms
that never leave a data-structure in a locked state; and that's not
true for spinlocks.

.