On Thu, 9 Mar 2000, Kiran Jonnalagadda wrote:

} > The idea is that most parts of shared data may (implicitly) be read only,
} > so there really is no reason why both processes shouldn't use the same
} > copy.  Saves memory as well as time lost while making copies.
} 
} I was aware of the code base of identical processes not being duplicated
} unnecessarily, but memory too? This is good.

non duplication of the code base is standard in all multiprocessing
systems and has always been AFAIK.  Only the very early OSes did not allow
it (I think) because back then it was not uncommon to write programs that
modified their own code (refer the story of mel).

Philip

===========
Prototype designs always work.
                -- Don Vonada

To subscribe / unsubscribe goto the site www.ilug-bom.org ., click on the mailing list 
button and fill the appropriate information 
and submit. For any other queries contact the ML maintener

Reply via email to