On Wed, Feb 18, 1998 at 02:43:41PM -0800, Jonathan Stone wrote:

> But what about processors that cant do test-and-set?  Should we
> introduce a new MD syscall or fastpath trap for them, or what?

PostgreSQL will use SysV type semaphores if you don't supply it with a
test-and-set mechanism.  This means taking a pretty hefty performance
hit, though -- people have measured 40% of CPU time being spent in the
semaphore handlers.

-tih
-- 
Popularity is the hallmark of mediocrity.  --Niles Crane, "Frasier"