Jukka Marin writes:
> 
> Michael Hitch writes:
> 
> > The only place I saw that made any check was in
> > kern/kern_fork.c and that was against "maxproc", which is set to NPROC
> > in param.c.  There is probably a limitation in amiga/pmap.c when it is
> > allocating memory for the user page tables.  I think there is an upper
> > limit on the number of pages available for the user page tables.
> 
> Hmm.  I _think_ the MAXUSERS is set properly, after all - I checked my
> kernel using binpatch (symbol _maxproc).  But _why_ did I get the message
> of process table being full when the table had 788 entries and I had only
> 100...150 processes? :-I

Type "limit" at your csh shell prompt, and see what the "maxproc" value is.
I think the default is something like "40".  

Try adding "limit maxproc unlimited" (or "limit maxproc 2048" or
whatever) to your .cshrc file, and see if you have the same problems...

(I'm not sure if the above is the way to solve that problem or not, but
it seems to me that's how I solved the problem when I encountered it
quite a while ago...) 

Later...

Greg Oster

oster@cs.usask.ca
Department of Computational Science
University of Saskatchewan, Saskatoon, Saskatchewan, CANADA