[racket] datum-intern-literal and lists?

A few minutes ago, Robby Findler wrote:
> On Tue, Jun 26, 2012 at 10:05 PM, Eli Barzilay <eli at barzilay.org> wrote:
> > 6 hours ago, Danny Yoo wrote:
> >>
> >> [Note: I find a place in Scribble that might help improve speed by
> >> about 15%.  See below.]
> >>
> >> I'll tried adding the fast path computation, with no effective
> >> difference. The statistical profiler is not working well in this
> >> instance.
> >
> > Any idea why?  Did you trace everything under a custodian to catch
> > all threads etc, and ran without places?
> 
> Does the profiler have any way to analyze how accurate it is when it
> says "15%"? How un/lucky does it get?

Well, if you run the code for enough time (and ~25 seconds should be
plenty) then it's statistically impossible that you got unlucky
enough.  And there's the usual thing with profilers: when in doubt,
run it 10 times...

(That "impossible" could probably be translated to a probability, but
I'm on the wrong side of the family for such questions...)


> >> I did not expect steps 2 and 3 to take that much time, but
> >> according to (time ...), they take a significant amount of
> >> time.  Out of the 26 seconds on my machine, it takes about four
> >> seconds to do step 2, and another three seconds to do step 3.
> >>
> >> So about 15% of the time in html-render is spent coersing xexprs
> >> to xml structure! [...]
> >
> > And how much is that time compared to the overall time of starting
> > with the .scrbl sources and ending with the final html pages?
> 
> Not a lot, but we're at the point where we're not going to find big,
> simple changes!

(I'm just wondering whether it would make sense to render via the
scribble/html bindings which might be faster.)

-- 
          ((lambda (x) (x x)) (lambda (x) (x x)))          Eli Barzilay:
                    http://barzilay.org/                   Maze is Life!