> > Kernel level support for async-IO is something that should be at least
> > thought about. From the UNIX's that I've heard actually do it, they
> > limit you to one outstanding async-IO request per task/thread. This
> > seems to suggest it is not a trivial problem to solve at all.
> >
> > From what I've heard NT allows numerous (unlimited?) outstanding
> > async-IO requests to be queued to the system, and programmers love
> > this. Can someone validate this? It's what I've heard, I want to
> > know if it is true.
> >
>
> VMS (omigosh) has this - you can queue as many async IO requests as
> your process quota will allow. Its a VERY nice feature. I think we
> should borrow it.
If process quota matters, then can I assume that you need to fork() (or
clone()) and then do the IO in a BLOCKING manner? _thought so!_ It might
have a handy API/lib thingy for it, though.
Greg Alexander
http://www.cia-g.com/~sietch/