random(4) plugin infrastructure for mulitple RNG in a modular fashion

[Warner Losh]

On Aug 19, 2013, at 1:31 AM, Mark R V Murray wrote:

> 
> On 19 Aug 2013, at 08:18, Warner Losh <imp at bsdimp.com> wrote:
>> If we're going to allow passthrough, we should require the kernel config to explicitly do something to get pass through.
>> 
>> nodevice yarrow
>> device random_passthrough
>> 
>> would be my suggestion.
> 
> I don't think it will sell; folks are asking for GENERIC with a run-time switch to flip between the raw HW generator output and a SW mixer/conditioner.

This is the config for no yarrow and pass through only. If you want both, you should have both and a sysctl/tunable controlling the wiring...

>> I'd go so far as to say that if you have random in your kernel, then you need to specify some "filter" or you get a compile-time error. Specifying yarrow via DEFAULTS or std.foo is fine by me, since both of those can be overriden fairly easily....  I'd also think we'd want to FAIL_PANIC or FAIL_BLOCKING, and have that choice hard wired at some level too, to be explicit about things. But maybe that's gilding things a bit too much and a tunable would suffice…
> 
> Won't sell. Folks are saying they want the choice of the raw output. In GENERIC.

Nothing I've said will preclude it. What's in generic is policy, not mechanism.

> "What Will Sell" may be up for debate and mind-changing; I think that is the route to explore.

Maybe I need to be more articulate, since I'm trying to describe a mechanism for having one or more filters, but having a compiler error when there's zero...

Warner