<div class="gmail_quote">On Fri, Jan 22, 2010 at 6:36 PM, Yue Li <span dir="ltr"><<a href="mailto:xyly781@gmail.com">xyly781@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
<div class="im"> The second and third points seems to me will provide better</div>
performance, so can I say the first one is the current main cause of<br>
the slow down?</blockquote><div><br></div><div>I do not understand fully your question. The system I am describing is the _new_ one. The old one does not implement _any_ of those three points: it had a terrible slowdown compared to the single-threaded version which was due to the use of hash tables.</div>
<div><br></div><div>The major difference from SBCL in this new version is the size of the thread-local storage. I do not want to fix it beforehand because that would mean allocating always large buffers for all threads. I would like threads to start with small or even no thread-local storage for special variables and then grow from there.</div>
<div><br></div><div>This flexibility has a consequence, which is a small slowdown due to an addition memory read and comparison operations but I am sure the inline version will get much better.</div><div><br></div><div>Juanjo</div>
</div><br>-- <br>Instituto de Física Fundamental, CSIC<br>c/ Serrano, 113b, Madrid 28006 (Spain) <br><a href="http://juanjose.garciaripoll.googlepages.com">http://juanjose.garciaripoll.googlepages.com</a><br>