On Fri, 19 Nov 1999, matthew green wrote:

>    options         MAXBSIZE=32768
> 
> perhaps we should use this in std.sparc ?  this seems to be a major
> problem with NetBSD -- and it's worse for the sparc64 which doesn't
> have the luxury of easily moving to 0xf000000 as something else
> already wants to be there (i believe).  eeh, do we want it for 
> std.sparc64 too?

It's not all that great a solution. I actually had started in on
a fix for sparc that would make sure that the allocation of buffers
never got too large for the kernel VM space, so that at least the
machine would always boot, but I had the impression that it wasn't
worth completing because UVM would be here any day now and fix
that.

If we're not going to have UVM for 1.5, I guess I should try to
cons this up again. For testing, however, I'd need a bunch of
volunteers with machines with lots of memory. (I have just the
SS20. This obviously affects the 5 and 10 as well, and lord knows
what else.)

cjs
--
Curt Sampson  <cjs@cynic.net>   917 532 4208   De gustibus, aut bene aut nihil.
The most widely ported operating system in the world: http://www.netbsd.org