In article <20030220174151.A22733@noc.untraceable.net>,
Andrew Brown <atatat@atatdot.net> wrote:
>i've just committed the top down changes to uvm that i've been working
>on for...well...a while.  at this point, it's undergone rigorous
>testing and survived everything i threw at it.

[...]

>and as you can plainly, see, there is only one area now, [3], into
>which either of them can grow.  this means (or will mean, ultimately),
>that your heap can grow larger, or you can mmap() larger objects.

Great job andrew! To explain the implications of this a bit further:

Before we had a ~3G area on the i386 which was carved 1G for sbrk() memory
and 2G for mmap() memory. Now a process can request > 1G of sbrk() *OR*
> 2G of mmap(). 

Also on small machines [vax, acorn26 etc.] we had to make to provide
a small MAXDSIZE (the limit for sbrk()) so that small memory footprint
machines would be able to boot. Now we don't have to specify MAXDSIZE
at all.

christos