Re: Issue with max_threads (and other resources) and highmem

[Dan Kegel]

Rik van Riel wrote:
> 
> On Tue, 23 Oct 2001, Dan Kegel wrote:
> > Rik wrote:
> > > ... A sane upper limit for
> > > max_threads would be 10000, this also keeps in mind the
> > > fact that we only have 32000 possible PIDs, some of which
> > > could be taken by task groups, etc...
> >
> > ?  I thought the 2.4 kernel had switched to 32 bit pid's long ago.
> > Where does the limit of 32000 possible PIDs come from?
> 
> Please take a look at kernel/fork.c::get_pid() ...

Yes, I see the limit is enforced there, but why do we need that limit?

There are probably a bunch of user-space programs that assume
a pid fits in five digits, is that the main reason?
- Dan