Re: {MUD} threads

["Chris Lawrence" <clawrenc#xsvr1,cup.hp.com>]

On Aug 23,  8:38pm, Wout Mertens wrote:
> On Thu, 22 Aug 1996, Chris Lawrence wrote:
> > I use fast RAM semephores to signal an executing event to die.  I then
> > rely on the clean-up processing of the event, and its messsages to the
> > DB invalidating the uncommitted changes to do the clean ups.
>
> How do the threads know they have to die? They poll the semaphore? Hmm. If
> they're interpreting, that's simple of course...

I think I have the byte-code translator check the semaphore every time
it leaves a block, and immediately prior to exiting an event.

> I meant more like: All memory /belongs/ to the mud, but threads put their
> stamp on what's theirs. Or have a list of running Actions, that have a
> list of all allocated things. Whatever...

Yup.  As I recall I use OSE memory pools and then let the clean up routines in
the various parties (DB, Event, etc) wipe the pools.


-- 
J C Lawrence                           Internet: coder#ibm,net
---------------(*)               Internet: clawrenc#cup,hp.com
...Honorary Member Clan McFUD -- Teamer's Avenging Monolith...