Re: PostgreSQL
Alfred Perlstein <[email protected]>
| Newsgroups | gmane.os.netbsd.devel.general |
|---|---|
| Message-ID | <[email protected]> |
* Ignatios Souvatzis <[email protected]> [060202 03:38] wrote: > On Wed, Feb 01, 2006 at 06:58:04PM -0800, Alfred Perlstein wrote: > > > Also, I'm pretty sure that synchronization between threads doesn't > > require locked bus cycles, only atomic ops, again another saving. > > If this was the case, I'd be very interested in your definition of > "locked bus cycle" vs. "atomic ops". My mistake, I somehow was thinking of a userspace threading system, where there is one kernel context for many threads. > In a multi-CPU (with non-shared cache) system, you have to implement > atomic operations via some sort of locked bus cycles. Yes. > This is independent of the thread vs. processes-with-a-shared-variable > question. > > P.S.: Why, exactly, is this discussion on tech-net, among others? Initial posted decided to cross post as much as possible? :D (trimmed to tech-mist@) -- - Alfred Perlstein