This is the mail archive of the
mailing list for the GCC project.
Re: [v3] tr1/random: implement random_device
Albert Chin wrote:
Sure! I was working under the (possibly mistaken) assumption that
/dev/random is becoming more and more widespread, but we can definitely
improve the thing either on the configury side or in the implementation
proper: I would suggest either filing a detailed enhancement DR or
providing a patch...
On systems, without /dev/random, would allowing an option to specify a
path to a socket that entropy is provided on be viable? Systems
without /dev/random sometimes use PRNGd to gather entropy.