Re: "movb" for spin-unlock (was Re: namei() query)

Oliver Xymoron (oxymoron@waste.org)
Tue, 25 Apr 2000 08:57:29 -0500


On Tue, 25 Apr 2000, Kai Henningsen wrote:

> > Which is my point. Or did you not read the text you quoted?
>
> If that was your point, you nicely managed to completely obscure it: you
> actually seemed to say the opposite.

Perhaps that's because my argument was of the form "assume the opposite of
what you're trying to prove and show a contradiction". And was sarcastic
to boot. It's all about context (which I made sure to quote).

> > Except simulation _very_ quickly becomes infeasible. But don't take my
> > word for it - try simulating the possible branches of an arbitrary
> > Turing machine with just 4 states. With binary input and a blank tape
> > even.
>
> Umm, "exponential time" *is* equivalent to "quickly becomes infeasible" in
> my book.

My point in the previous post was that it's much worse than exponential
(which you might try to tackle for problems as large as 100 or 1000,
given a small exponent). The complexity grows faster than any computable
function of n.

--
 "Love the dolphins," she advised him. "Write by W.A.S.T.E.." 

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/