<html><body bgcolor="#FFFFFF"><div><br></div><div>Le 24 août 2010 à 17:58, Dominick Samperi <<a href="mailto:djsamperi@gmail.com">djsamperi@gmail.com</a>> a écrit :<br><br></div><div></div><blockquote type="cite"><div><br><br><div class="gmail_quote">On Sat, Aug 21, 2010 at 2:40 PM, Romain Francois <span dir="ltr"><<a href="mailto:romain@r-enthusiasts.com"><a href="mailto:romain@r-enthusiasts.com">romain@r-enthusiasts.com</a></a>></span> wrote:<br><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
Not that I care, but I'm slightly surprised at the implementation of log1p :<br>
specifically that it throws an exception<font color="#888888"><br>
Romain</font></blockquote><div><br>I changed the code in msvcmath.cpp to return non-signalling NaN when there is<br>a problem instead of throwing an exception. This is what happens when you<br>try sqrt(-1.0), for example.<br></div></div></div></blockquote><div><br></div><div>Thanks. Ill have a look in a few days</div><br><blockquote type="cite"><div><div class="gmail_quote"><div>
I'm attaching the diffs. I don't think I can checkin directly, right?<br></div></div></div></blockquote><div><br></div><div>Right, you can't</div><br><blockquote type="cite"><div><div class="gmail_quote"><div>Thanks,<br>Dominick<br></div></div><br>
</div></blockquote><blockquote type="cite"><div><msvc.diff></div></blockquote><blockquote type="cite"><div><msvcmath.cpp></div></blockquote></body></html>