On Wed, 30 Jul 1997, Colin Wood wrote:

> > Paul Goyette also complained of problems with the newest current giving
> > fifo overruns. I wonder what we changed.

Beats me. :-(

> I believe I remember seeing something about Scott having changed the way
> splraise semantics worked, or am I just mis-remembering something from all
> the NetBSD lists I'm on??? ;-)

Until recently, splraise() wasn't even used.  I wouldn't use it until I
was reasonably sure I had squeezed every last cycle out of it that I
could.  Even so, this only affected spltty(), not splzs(), splsched(), or
splhigh().

--scott