On Thu, Dec 17, 1998 at 10:29:47PM +0100, Ignatios Souvatzis wrote:
> All this talk about "overcommitting memory" confuses, at least in the text,
> "memory" with "address space".
> 
> The 4.4BSD VM (and some others, if I read the thread right) allows to seperate
> the two.
> 
> To the very least, please don't mention sbrk() again. It's just a special
> case of an unnamed mmap(), for our VM systems..
> 
> This might be the reason why bookkeeping, and limiting, of the real swap
> needed, is more difficult than it used to be. A 2nd problem is, that pages
> might be shared between processes.... even unnamed mmap() regions, if I read
> the mmap(2) man page right...

I started to read Charles Cranors thesis about the UVM virtual memory system
now integrated in NetBSD (after the 1.3 release), and he states that UVM, 
unlike the old Mach VM used earlier, detects running out of swap and sends 
an error signal to a process if it tries to access a page for which anonymous
backing store would be needed, but isn't (currently) available. (SIGSEGV, I
guess).

So part of this discussion can be dropped.

	-is