Re: CVS commit: src/sys/netinet

[Antti Kantee <pooka%cs.hut.fi@localhost>]

On Wed Aug 11 2010 at 17:46:07 +0100, David Laight wrote:
> > Use kpause() instead of DELAY() and sleep a minimum of 1 tick.
> > This is possible now since softints have a thread context.  It's
> > also not a very frequent code path.  Addresses ABI issue with delay
> > (kern/40505).
> 
> 1) How many copies of this function can actually run at the same time ?
> 2) How many kernel threads are available for softints ??
> 3) How many other places might kernel threads sleep for timeout ???
> 
> It might be that they will all get stuck!
> 
> Actually, although kernel worker threads can sleep, in general they
> shouldn't do so unless they are just waiting for locks and similar
> short duration sleeps.
> I remember seeing solaris systems stuck when more than 1 (or a few)
> drivers slept in the kernel threads.

What exactly is your point?  Are you saying sleeping is worse than
a busyloop?