Re: [PATCH 5.10 00/11] 5.10.123-rc1 review

[Jason A. Donenfeld]

Hi Jon,

On Thu, Jun 16, 2022 at 11:11:25AM +0100, Jon Hunter wrote:
> On 16/06/2022 10:46, Ron Economos wrote:
> > On 6/16/22 1:48 AM, Jon Hunter wrote:
> > I'm also seeing this on RISC-V. 5.15 and 5.17, but not 5.18.
> That's good to know. I don't see this on 5.18 either, just 5.10, 5.15 
> and 5.17.

Right. 5.18 has
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=48bff1053c172e6c7f340e506027d118147c8b7f
which I thought was a bit too risky to backport. So on 5.18 you're not
seeing this behavior because /dev/urandom is always seeded magically and
hence there is never an opportunity to warn about unseeded randomness.
Maybe we can think about backporting that if no issues come up after
several months, but I'd rather tread lightly with that one.

Anyway, as mentioned in the other message, I've got a patch I'll send
out in a minute for the unwanted pr_warn().

Jason