"Brian C. Grayson" writes:
> Before I go into gory details, it appears to me that the
> pthreads system allows race conditions when used on an SMP
> system.

If you mean "MIT pthreads", well...

1) We don't have SMP yet.
2) We aren't adopting MIT pthreads -- it is just a package
   available right now for the users who want it. Some people get some 
   advantage out of it, but you really should bitch to the authors of
   the MIT pthreads package about problems you have with it.

Repeating, we aren't going to be using the MIT pthreads package as our 
userland threads system.

We are currently working on making sure our normal libc is thread safe 
(you may have seen Klaus Klein's commits doing that work in recent
weeks) and we are also working on getting kernel support for userland
threads done.

Perry