I really hate to jump in, as you are all entertaining me quite well, but
this point jumped out at me like the snake in "Honey, I Shrunk the
Audience" at Disney's Epcot...

On Wed, 23 Oct 1996, David S. Miller wrote:

> 	a) Consider that your OS it taking say 200 cycles more than it
> 	   needs to with just trap instruction in/out overhead.  Say
> 	   also that once your application takes control again, it
> 	   will take it only 100 instructions to get to the point
> 	   where it asks the kernel to queue up another disk request.
> 	   I'd say you are losing here.

Taking 200 cycles more than _what_?  Is this 200 cycles we're shaving from
300 cycles, or 600, or 3000?  Of course, it varies, but pick an average
based on real-world usage patterns, and compare the 200 cycles to _that_
number, as that's where the significance shows up. 

You seem to be operating from at least some flawed assumptions here, which
is quite possibly leading to an overstatement of the importance.  Sure,
it's cool and all, but in the real world, does it amount to something
actually perceptible?  (Note that comparing against other OSs is utterly
invalid, as there are far more variables than the syscall implementation.) 

--scott