[MUD-Dev] Re: TECH: Distributed Muds
J C Lawrence
claw at 2wire.com
Fri Apr 27 19:31:48 New Zealand Standard Time 2001
On Fri, 27 Apr 2001 19:38:32 -0600
Chris Gray <cg at ami-cg.GraySage.COM> wrote:
>> From: J C Lawrence <claw at 2wire.com>
>>> Chris Gray >> <cg at ami-cg.GraySage.COM> wrote:
>> Given a soft code system and other wise single threaded system
>> timeouts mean that I must either enforce nasty time limits on
>> softcode execution, or I must build full stack fault support into
>> the soft language supports such that I can push the soft context,
>> go do an IO loop, and then return to my soft context. No thanks.
>> The model of decoupling IO, softcode execution, etc is too damn
>> simple and clean to bother trying to munge them back together.
> OK, sure, if you have that model to start with, but if you are
> just starting with a select() loop, putting timeouts on softcode
> execution is a reasonable thing to do.
> In any case, do you really want to have no limit on softcode
Nah, I just want to be able to control it more gracefully than would
be enfoced by a select() hearbeat.
> No matter how many threads you give to your execution system, if
> the bad softcode is hit enough, you are still in trouble. Not only
> in the CPU usage, but possibly also in the memory tied up and
> maybe even locks held.
Yup. You may remember the whole discussion back in 1998 (?) on the
scheduler and event priorities under C&C etc etc etc yada yada which
is intended to a) soften this edge, and b) play executioner on rogue
The critical aspect from my vantage is that I have the time and
opportunity to both try really hard to get the damn thing executed,
and if that doesn't work, to try and rescue graefully without having
to pervert the rest of the system to account for the fact that
things sometimes die.
J C Lawrence claw at kanga.nu
--=| A man is as sane as he is dangerous to his environment |=--
MUD-Dev mailing list
MUD-Dev at kanga.nu
More information about the MUD-Dev