From: Paul Guyot (pguyot_at_kallisys.net)
Date: Mon Feb 10 2003 - 10:24:39 PST
Ā (At) 20:06 +0200 10/2/03, Eckhart Köppen écrivait (wrote) :
>One thing to consider is the memory usage pattern. An application that
>would benefit greatly is for example a web browser or a mail client.
>They use relatively big amounts of memory for a short time: When
>processing the incoming data. Then, attachments get stripped of, web
>pages get simplified and everything ends up in NewtonScript land.
Especially if we could avoid transitting in the NewtonScript land
with native C++ endpoints.
> > - This heap isn't garbage collected. Any leak there is definitive.
>
>Yes, but that constraint is even more true on other platforms and they
>manage to do well. With the same uptime requirements.
You're right.
I was just mentionning drawbacks of using C++ for memory allocation.
Now, there are plenty of advantages.
One of the nice thing you can do with C++ memory allocation is to
have your own heap. This also eliminates the leak issue.
A long time ago, I thought about a way to design pure C++ program in
the Newton that would be loaded as a separate task, with globals
(this requires doing some runtime/compilation hack) and that would
talk to the NS world in a thread safe way (thru special APIs). I
think this is how NewtonOS should have been extended 3 years ago.
Paul
-- NPDS: http://newton.kallisys.net:8080/ Apache: http://www.kallisys.com/ -- This is the NewtonTalk list - http://www.newtontalk.net/ for all inquiries List FAQ/Etiquette/Terms: http://www.newtontalk.net/faq.html Official Newton FAQ: http://www.chuma.org/newton/faq/
This archive was generated by hypermail 2.1.5 : Mon Feb 10 2003 - 11:00:01 PST