svn commit: r239569 - head/etc/rc.d

[Ben Laurie]

On Fri, Sep 14, 2012 at 3:46 PM, RW <rwmaillists at googlemail.com> wrote:
> On Fri, 14 Sep 2012 14:43:53 +0100
> Ben Laurie wrote:
>
>> On Fri, Sep 14, 2012 at 2:38 PM, Bjoern A. Zeeb <bz at freebsd.org>
>> wrote:
>> > 7) send all data to the kernel and hash (arch dependent?) it +
>> > counter value into the buffer on overflow, as in b[n] = H(b[n] + c
>> > + i[n]) in the kernel
>> >    (can control when buffer full and only then take action when
>> >    needed, indepedent on how seed data is chosen, uses standard
>> >    technology)
>>
>> IMO, this is the only good option.
>
> No it isn't. I means that the hashing is unconditional, so anyone that
> needs a faster boot needs to patch the kernel.

Has anyone measured the cost of doing this? Also, if you really want
to turn it off, we could provide a flag.

>  It has no advantage
> whatsoever over a minor change to initrandom.

It absolutely has. It applies to all inputs to /dev/random, not just
those that come from initrandom. Also, should something get to write
to it before initrandom, initrandom's input would still be used.