[racket-dev] [plt] Push #24367: master branch updated

Yesterday, Matthew Flatt wrote:
> At Tue, 28 Feb 2012 07:21:24 -0600, Robby Findler wrote:
> > Last I heard, Eli was saying that there was something seriously
> > wrong with 'raco setup' on two cores. Did that ever get resolved?
> 
> Commits 012ef60cd545ba and 534886dbe4b6ad (yesterday) were in
> response and improved things on my machine, so it's probably worth
> checking again.

I did a bunch of builds with these two commits in but before Ryan's
change, and it looks like the main problem (2 cores being about twice
slower than a single one) is resolved:

    setup-plt -D -j 1  965.63s user 63.00s system  96% cpu 17:49.50 total
    setup-plt -D -j 2  860.61s user 77.72s system 139% cpu 11:12.83 total
    setup-plt -D -j 3  900.29s user 85.15s system 157% cpu 10:24.68 total
    setup-plt -D -j 4  939.24s user 96.29s system 137% cpu 12:34.89 total

For reference, these are the times for a run when there's nothing to
do (so it's probably an indicator of the overheads).

    setup-plt -D -j 1  1.85s user 0.47s system  99% cpu 2.343 total
    setup-plt -D -j 2  3.59s user 0.90s system 144% cpu 3.113 total
    setup-plt -D -j 3  4.17s user 1.16s system 166% cpu 3.199 total
    setup-plt -D -j 4  4.70s user 1.39s system 181% cpu 3.356 total

And here's the same two after Ryan's change:

    setup-plt -D -j 1  966.43s user 61.98s system  95% cpu 17:54.79 total
    setup-plt -D -j 2  807.60s user 75.75s system 168% cpu  8:44.09 total
    setup-plt -D -j 4  855.21s user 87.56s system 242% cpu  6:28.22 total

  No-op scan:
    setup-plt -D -j 1  1.86s user 0.45s system  98% cpu 2.340 total
    setup-plt -D -j 2  3.57s user 0.90s system 144% cpu 3.099 total
    setup-plt -D -j 3  4.32s user 1.20s system 182% cpu 3.030 total
    setup-plt -D -j 4  4.85s user 1.39s system 200% cpu 3.117 total

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