In article <1992Sep7.080727.22080@klaava.Helsinki.FI> torvalds@klaava.Helsinki.FI (Linus Benedict Torvalds) writes:
quoted text
>
>And as people have been very concerned about the 16MB limit - don't
>panic.  It's not inherent to anything like the DOS 640kB limit, and is
>there just because it was a bit easier to code that way - correcting it
>later won't break anything.  So while this version and the next may have
>this limitation, it will go away (cleanly) - I just have to code some
>more. 
>
>In case anybody is wondering why there is a 16MB limit in the pl3 code,
>it's very simple: I use a character array to store the swap-page-counts,
>and that array just happens to be 4096 bytes long, as that's the biggest
>easily allocable unit of memory in the current kernel.  So I just
>allocate 4096 bytes: 4096 swap-page-counters = 16MB swapspace.  So the
>only thing that I have to do to allow bigger swapfiles is to allocate
>more memory - SMOP, but irritating and boring. 
>
>               Linus

Is the new swapping code faster?  If so, I don't have a problem with
the 16 MB limit.  If a system has the maximum amount of RAM, which
is presently 16 MB, and needs more than 16 MB swap space on a regular
basis, isn't it slow?  Using an additional swap partition or file for 
more than 16 MB doesn't seem like a bad idea if they aren't needed
on a regular basis.  This gives the benefit of quicker swapping for
regular use (if it is indeed faster), and only reduces performance
when the extra space is needed (which might be way over on the
opposite end of the disk!).  If more than 32 MB of memory (16 MB
RAM, 16 MB swap) is needed on a regular basis, wouldn't it be better
to focus on adding better support for more RAM?  Do any Linux users
use more than 32 MB on a regular basis?

Comments?  Suggestions?  Please post to the newsgroup for further
discussion.  Also, please correct me if I am missing something.

-- 
Jim H.
*
* James L. Henrickson
* ujlh@sunyit.edu        "Some day I might have a real .signature!"