Upper memory limit

[Michael Hudson]

Siegfried Gonzi <siegfried.gonzi at kfunigraz.ac.at> writes:

> Michael Hudson wrote:
> 
> > At any rate, if the OS can't claim back a processes memory when said
> > process terminates, it's definitely Time To Get A Real OS.  Even Macs
> > can do this now...
> 
> Oh man, the classic Mac OS 9.0 was the reason for me to jump onto the
> Windows bandwagon eventually. On my Mac rebooting every 30min due to
> memory problems were normal.

That's what the "now" was about; I meant OS X.

> You can even hurt a Unix machine (I once had the problem with Clean on a
> Sun). 

Oh yeah, running out of VM is never nice.  But what (I think) you're
describing is that what a process does affects a process that runs
after the first one has finished... that sounds very odd.

> I have to emphasize that I also used Common Lisp (Allegro) on my
> machine and some calculations consumed a hour execution time or so;
> but I never noticed any memory problems (and my Lisp and Python
> programming style is quite the same).

ACL probably has a more advanced approach to memory, but Python
doesn't do anything out of the ordinary.  I'm pretty sure all it does
it call the platform malloc() and free() (even with pymalloc, that's
still how the arenas are allocated).

> > Summary: if you're on 95/98(/Me?), get off, right now.
> 
> I am actually using Windows XP (NT 5.1 to be correct) on a Celeron 1000
> MHz, 256 MB RAM laptop.

Ah.

> To make it clear: Python gives memory back (the task manager shows
> it), but, I do not know,...

Can't help you further, sorry...

Cheers,
M.

-- 
  Important data should not be entrusted to Pinstripe, as it may
  eat it and make loud belching noises.
   -- from the announcement of the beta of "Pinstripe" aka. Redhat 7.0