Revision history for Bench

0.10    2014-05-14 (SHARYANTO)

        - Use Benchmark as the backend instead of doing our own timing (we
          haven't factored out the overhead of null loop, for example).


0.09    2012-08-28 (SHARYANTO)

        - No functional changes. Some POD tweaks. Add -v (version).


0.08    2012-08-05 (SHARYANTO)

        - Shuffle order of execution


0.07    2012-08-05 (SHARYANTO)

        - bin/bench: Add options -h, -M/m, -n, -I, -b


0.06    2012-08-04 (SHARYANTO)

        - bin/bench: Run at least 1s for each sub

        - Show who's fastest at the end (and by how much)


0.05    2012-07-27 (SHARYANTO)

        - bin/bench: Forgot to support benchmarking multiple codes


0.04    2012-07-27 (SHARYANTO)

        - Handle division by zero error [CT]

        - Add script: bin/bench


0.03    2011-03-31 (SHARYANTO)

        [INCOMPATIBLE CHANGES]

        - By default now only run code at most 100 times or 1s.

        - Changed syntax for specifying multiple codes (suggested by
          educated_foo).

        - Use s all the time, but ms for per-call times.

        [FIXES]

        - A couple of small fixes.


0.02    2011-03-30 (SHARYANTO)

        [ENHANCEMENTS]

        - Display time in ms when appropriate.

        - Reduce benchmarking overhead (especially when n is negative).


0.01    2011-03-30 (SHARYANTO)

        - First release.