I've taken a profiling run of the pure aspect version, and had about 50%
of the time spent in Thread.sleep() and Object.wait(). Admittedly I ran
the full 100K iterations with the aspect version, but only about 4000
with the TM, as profiling really slows it down.
- P
Prof. Laurie HENDREN wrote:
>Ok, thanks Pavel. I will take a look.
>
>Laurie
>
>
>+-----------------------------------------------------------------
>| Laurie Hendren --- laurie.hendren@mcgill.ca
>| Associate Dean (Academic), Faculty of Science,
>| Dawson Hall, McGill University, 853 Sherbrooke St W,
>| Montreal QC H3A 2T6 Canada, 514-398-7179, fax 514-398-1774
>+----------------------------------------------------------------
>| For contact and home page info as Professor, Computer Science:
>| http://www.sable.mcgill.ca/~hendren --- hendren@cs.mcgill.ca
>| Research: http://www.sable.mcgill.ca http://aspectbench.org
>+----------------------------------------------------------------
>
>On Sun, 12 Mar 2006, Pavel Avgustinov wrote:
>
>
>
>>I had a brief look at AJHotDraw. The automaton has three states (0, 1,
>>2), and throughout the run there is a single disjunct in state 0, a
>>single disjunct associated with state 1, and a single disjunct in state
>>2. Negative bindings sets are empty, so there really isn't anything
>>large to iterate.
>>
>>I tried running hprof, and was very startled to find 82% or so of the
>>time spent in Object.wait() or Thread.sleep(). I don't understand this,
>>but it definitely seems wrong. (See discussion on IRC about that.)
>>
>>Perhaps that'd be a good starting point for those who have the time to
>>investigate the ajhotdraw slowdown.
>>
>>- P
>>
>>
>>
>>
>>
>
>
>
>
>
>
Received on Sun Mar 12 21:57:46 2006
This archive was generated by hypermail 2.1.8 : Tue Mar 06 2007 - 16:13:27 GMT