<div dir="ltr"><br><div class="gmail_quote">---------- Forwarded message ----------<br>From: <b class="gmail_sendername">Theo de Raadt</b><span dir="ltr"></span><br>Date: Wed, Sep 9, 2015 at 10:42 AM<br>Subject: Re: getentropy, getrandom, arc4random()<br>To: <a href="mailto:guido@python.org">guido@python.org</a><br><br><br>been speaking to a significant go person.<br>
<br>
confirmed.<br>
<br>
it takes data out of that buffer, and does not zero it behind itself.<br>
obviously for performance reasons.<br>
<br>
same type of thing happens with MT-style engines. in practice, they<br>
can be would backwards. a proper stream cipher cannot be turned<br>
backwards.<br>
<br>
however, that's just an academic observation. or maybe it indicates<br>
that well-financed groups can get it wrong too.<br>
<br>
by the way, chacha arc4random can create random values faster than a<br>
memcpy -- the computation of fresh output is faster than doing<br>
gross-cost of "read" from memory (when cache dirtying is accounted for).<br>
<br>
</div><br><br clear="all"><br>-- <br><div class="gmail_signature">--Guido van Rossum (<a href="http://python.org/~guido" target="_blank">python.org/~guido</a>)</div>
</div>