On Sun, Sep 14, 2003 at 07:10:51PM +0200, Matthias Buelow wrote:
> Jason Thorpe writes:
> 
> >Uh, it's a significant performance boost even on modern hardware for 
> >the types of applications where it is intended to be used.
> 
> What are these applications?  I can't right now think of any which
> produce a steady stream of fork/exec, enough to satisfy the CPU with
> only these operations...  Even huge, make-directed build jobs spend
> almost all of their time in compilers, or on disk i/o, etc.

"If you say so".  I suggest that instead of just continuing to make
unsupported assertions, you actually do some measurements.

Jason and I did.  When Jason added real vfork() to NetBSD again several
years ago, he measured a speedup of several minutes in a system build
just by adjusting make and the compilers.  When I adjusted groff and
some of the other build tools to use vfork, we again won by a couple of
*minutes* in system build time.