memory limits....

[Jerry Feldman]

On Thursday 19 May 2005 2:55 pm, markw at mohawksoft.com wrote:
> Classically, text segments are not "swapped," but if they are modified at
> load time, they may be paged out.
Not true. They are mapped in as read-only. Text segments are generally never 
paged out today.

> I also don't like to use the term "swapped." I think "paged" is sort of
> better, because the virtual memory mapping of the 386 works in 4K or
> (optional 4M pentium) pages. Swapping I associate with the 80286 segment
> based swapping.
I agree that swapping is not a good term. 
Historically, swapping predates virtual memory. It goes back to time sharing 
where entire processes would get swapped out. As virtual memory was brought 
into the Unix world, the term swap along with swap space was retained, and 
used in other operating systems. Virtually every non-embedded system today 
uses a sophisticated demand-paged system based on the least-recently-used 
algorithm. Note that the VAX systems used working sets and the hardware did 
not have the bits needed for LRU. 

Today's memory management tries to be smart, such as not creating .bss pages 
until they are specifically used.
-- 
Jerry Feldman <gaf at blu.org>
Boston Linux and Unix user group
http://www.blu.org PGP key id:C5061EA9
PGP Key fingerprint:053C 73EC 3AC1 5C44 3E14 9245 FB00 3ED5 C506 1EA9