<div dir="ltr"><div class="gmail_extra"><div class="gmail_quote">On Sun, May 22, 2016 at 6:18 PM, Kent Borg <span dir="ltr"><<a href="mailto:kentborg@borg.org" target="_blank">kentborg@borg.org</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><br></blockquote><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Let me try my own experiment:<br>
<br>
# strace ssh-keygen -t rsa</blockquote><div>.... </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
read(3, "\255J\373\231\323\256\251^\314\207MqkC\332\222^\352\275\307\373\351bM\267\273\260$G\232\301\r", 32) = 32<br>
close(3) = 0<br>
[...]<br>
<br>
(Was I supposed to say "dsa"? Okay...tried that too, same result.)<br></blockquote><div>..... </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Looks to me like it read 256-bits. I would have expected it would have read more, just to waste if nothing else.<br>
No where near using up 4096-bits (if "using up" even is real). Maybe do both DSA and RSA? It still would only "use" 1/8 of a 4096-bit pool.</blockquote><div><br></div><div><br></div><div>Since the read() returns the count a solution is for the process to sleep some reasonable rand() seconds or nanosleep() do some math to know</div><div>how many more to request then request more bits.<br><br></div><div><div class="HOEnZb"></div></div></div>As others have noted this is a known issue yet it is not a common computation. Where it might be common other tricks can leverage modest entropy bit counts returned by the read. And if both common and important additional hardware and local services makes sense. <br><br clear="all"><div><br></div>-- <br><div class="gmail_signature"><div dir="ltr"> T o m M i t c h e l l</div></div>
</div></div>