Subject: Re: rough SA patch (works on SMP)
To: Nathan J. Williams <>
From: Greywolf <>
List: tech-kern
Date: 06/23/2003 23:02:58
Thus spake Nathan J. Williams ("NJW> ") sometime Today...

NJW> Frank van der Linden <> writes:
NJW> > On Mon, Jun 23, 2003 at 04:50:32AM -0400, Stephan Uphoff wrote:
NJW> > > FYI: The patch only allows a single SA LWP to be active in
NJW> > >      userspace.
NJW> >
NJW> > I assume you mean for an SMP configuration? I believe that restriction
NJW> > has always been there (so far).
NJW> For UP, it's not a restriction, it's a deliberate property of SAs. In
NJW> general, the number of SA LWPs in userspace should always be less than
NJW> or equal to the number of CPUs in the system.


Do you mean "active SA LWPs"?

Otherwise, wouldn't the restriction mean that sa_create (or whatever
it's called) would automatically fail, and SA LWP would be completely

I'm sorry if I'm pointing out the bloody obvious, or if I'm
misunderstanding completely...

NJW>         - Nathan

Oh, can I also surmise -- on a slightly different tangent -- that the
primary use thus far for SA/LWP is more or less for function-while-
display (under X, for example (i.e. XMMS/mozilla), so that stuff can just
run in the background without the overhead of fork()/IPC?

[I think I can see the benefit of that, actually.  Trying to jump through
the hoops of fork()/IPC for something like a browser would be REALLY

Opinion survey:  Would dump(8) benefit from SA/LWP?

SA:  I have *no* idea.