[coyotos-dev] Need for revised invocation
Jonathan S. Shapiro
shap at eros-os.org
Tue Feb 7 04:25:59 EST 2006
On Mon, 2006-02-06 at 14:24 -0800, Charles Landau wrote:
> At 3:38 PM -0500 2/6/06, Jonathan S. Shapiro wrote:
> >Why should I have to spend an entire kernel thread (4k register state,
> >4k stack) to set a single word to 1?
> I was referring to a new kernel object that *acts* like a kernel
> thread obeying a known simple program (set word to 1), but is
> implemented optimally. I think you are calling this a "spin-agent".
> It's an agent, but I don't see the "spin" part.
Ah. So now you have proposed *two* kernel agents:
1. The "set word to 1" agent
2. The "message forwarding agent"
I find the ad-hoc growth of this list of kernel agents disconcerting. It
seems to suggest that the primitive abstraction set isn't working well.
More information about the coyotos-dev