I noticed the large amount of allocation going on when I was using mandelbrot to test my fixes to the visualizer last week -- I'll take a look. I wrote an allocation hotspot profiler tool that may end up as part of the visualizer in some form, so I can try running it on your code. <div>
<br></div><div><br><br><div class="gmail_quote">On Tue, Jul 24, 2012 at 6:12 PM, Sam Tobin-Hochstadt <span dir="ltr"><<a href="mailto:samth@ccs.neu.edu" target="_blank">samth@ccs.neu.edu</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
I've been trying to use the new futures visualizer (which is great) to<br>
improve my parallel version of the mandelbrot benchmark<br>
(collects/tests/racket/shootout/mandelbrot-futures.rkt). My current<br>
code is attached. You can run it like this:<br>
<br>
$ racket mand.rkt 4000 8 # 8 is the number of futures to use.<br>
<br>
The visualizer shows that each future syncs thousands of times for<br>
allocation. However, I can't find the allocation in the loop at all<br>
(even in the decompiler output). Does anyone have any suggestions on<br>
what might be allocating here, or any strategies I can use for finding<br>
the allocation?<br>
<span class="HOEnZb"><font color="#888888">--<br>
sam th<br>
<a href="mailto:samth@ccs.neu.edu">samth@ccs.neu.edu</a><br>
</font></span><br>_________________________<br>
Racket Developers list:<br>
<a href="http://lists.racket-lang.org/dev" target="_blank">http://lists.racket-lang.org/dev</a><br>
<br></blockquote></div><br></div>