PTHREADED -> P_SA

[Marcel Moolenaar]

On Sat, Jun 14, 2003 at 06:27:49PM +0800, David Xu wrote:
> 
> > On Sat, Jun 14, 2003 at 02:56:36PM +0800, David Xu wrote:
> > > 
> > > > I will rename flag P_THREADED to P_SA in kernel, any objections ?
> > > > here is the patch:
> > > > http://people.freebsd.org/~davidxu/P_SA.diff
> > > > 
> > > > It's only semantics..
> > > > but it's been changes before.. is this the final name?
> > > 
> > > Yes, I think it should be final.
> > 
> > P_THREADED says more than P_SA for someone who's not intimate
> > with the code. I for one don't know what SA stands for, but
> > apparently it's some kind of threading model/implementation
> > if I can believe the comments. In that case I don't know what's
> > so wrong about P_THREADED.
> > 
> > So, why is P_SA better than P_THREADED?
> > 
> If you find there is 1:1 threading mode Jeff added in kernel,
> I bet you will understand why I need to change it to P_SA.
> why shouldn't 1:1 code be called as P_THREADED ? I think this
> flags is confusing. P_SA tells you that the threading mode
> a process is using is Scheduler Activation based --- Anderson's
> theory.

Yes, that is a logical explanation. Maybe it helps to have
the term Scheduler Activation mentioned somewhere in the context
of P_SA. Or, maybe use P_SCHEDACT?

I don't object to P_SA, but I know that if I look at the code
that references P_SA in a couple of weeks/months, I'm going to
wonder what it means all over again. But then again, that may
be just me :-)

-- 
 Marcel Moolenaar	  USPA: A-39004		 marcel at xcllnt.net