<div dir="ltr"><br><div class="gmail_extra"><div class="gmail_quote">On Mon, Jun 17, 2013 at 6:18 PM, Carl Worth <span dir="ltr"><<a href="mailto:cworth@cworth.org" target="_blank">cworth@cworth.org</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">José Fonseca <<a href="mailto:jose.r.fonseca@gmail.com">jose.r.fonseca@gmail.com</a>> writes:<br>
>> However it fails to build on windows due to random(). Could you modify<br>
>> random_level() to use rand() , or add a os::random() abstraction that does<br>
>> something sensible for windows?<br>
><br>
> I went ahead and did this.<br>
<br>
</div>Thanks, José. My apologies for not getting to this little task you asked<br>
for. I totally dropped the ball on this one, (to the point where someone<br>
asked me about these performance improvements recently and I couldn't<br>
recall why there weren't on master).<br>
<br>
It will be nice to have this in place on the master branch now.<br></blockquote><div><br></div><div style>No prob. I was trying to catch up with my starred/flagged apitrace emails when I came across this. It was easier than I thought -- initially I somehow thought that rand() and random() had different interfaces, and would require fiddling with the function that used them, but rand/random have precisely the same interface (0..RAND_MAX), so it was actually trivial.</div>
<div><br></div><div style>Jose</div><div> </div></div></div></div>