Timings for random

[Eric S. Raymond]

Hal Murray <hmurray at megapathdsl.net>:
> > What about the OpenSSL RAND_bytes()? 
> 
> It's slightly faster than RAND_pseudo_bytes()  :)  ??

And what we're now using.  I finished the cleanup last night; everything
goes through OpenSSL now, the local MD5 and SHA-1 code is gone, and the
depenendency on libsodium has been abolished.

Hal, thanks for nudging me about this.  Everybody, thanks for the
supporting analysis.  We dropped about 400 LOC.
-- 
		<a href="http://www.catb.org/~esr/">Eric S. Raymond</a>