Skip to content


Subversion checkout URL

You can clone with
Download ZIP
Commits on Feb 12, 2013
  1. Add several more versions of Clojure 1.5 including 1.5-beta12 and 1.5…

    …-beta13, and 1.5-RC14 and 1.5-RC15.
Commits on Feb 8, 2013
  1. Update README to mention Leiningen 1.7.1 needed as "lein1", and simpl…

    …ify so it doesn't need all of those lein-files/* files any more. Add most recent Clojure releases to and
Commits on Jan 29, 2013
  1. Make RC versions of Clojure to sort correctly relative to alpha, beta…

    …, and final versions.
    Also some minor corrections to the results2graphs usage help message.
Commits on Jan 28, 2013
  1. Correct

Commits on Dec 8, 2012
Commits on Nov 28, 2012
Commits on Nov 27, 2012
Commits on Nov 25, 2012
Commits on Nov 24, 2012
Commits on Nov 7, 2012
  1. Name plot files by number rather than benchmark name

    This is in preparation for when benchmark names will be Clojure
    expressions with funky characters like [ ] * / that tend not to do
    well in file names.
  2. Global rename problem -> benchmark in results2graphs

    It is a better name for what it represents.
Commits on Nov 6, 2012
Commits on Nov 5, 2012
  1. Move towards command line args to pick a set of benchmarks to run, or…

    Andy Fingerhut committed
    … to analyze a saved result file
Commits on Nov 1, 2012
  1. Add a few more cases to my hack for generating a value for…

    …ndor on the fly for my existing XML data files.
  2. Add java.vm.vendor (or at least the first word in the string) to the …

    …platform name used to label curves in the plots
  3. Add java.jv.vendor to the list of properties collected about JVMs use…

    …d in JVMInfo.class and measureproc
Commits on Oct 30, 2012
  1. Modify so that the LANGUAGES variable has the same format wh…

    …ether you specify a Clojure version on the command line or not.
Commits on Oct 25, 2012
  1. Calculate and show in table the amount of parallelism achieved in eac…

    …h run
    Where by "parallelism" I mean the (user+system cpu time) / elapsed
    time, which seems like a reasonable measure of how many CPU cores were
    kept busy during the elapsed time, on average.
Something went wrong with that request. Please try again.