[MODERATED] Re: [PATCH 0/2] more sampling fun 0

On Thu, Feb 20, 2020 at 04:05:49PM +0100, speck for Greg KH wrote:
> On Thu, Feb 20, 2020 at 06:55:10AM -0800, speck for Andi Kleen wrote:
> > > Then we need to stop using RDRAND internally for our "give me a random
> > > number api" which has spread to more and more parts of the kernel.
> > 
> > Only if that API is called frequently enough. AFAIK it is not. 
> 
> It's called by all sorts of places in the kernel today.
> 
> > Normally it's used for rare rekeying of hash tables etc., which
> > doesn't happen very often.
> 
> "normally" :)

I looked through the users. The ones that weren't clearly slow path
were SCTP connection setup, exec, and some strange abuse in bcache
that should probably be fixed anyways.

I suppose maybe for exec, but still not sure it's worth it.

But the exec use is for seeding user space RNG, so that's the one
that might actually be worth it.

-Andi