Re: [PATCH v1] random: block in /dev/urandom

From: "Jason A. Donenfeld" <Jason@zx2c4.com>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Guenter Roeck <linux@roeck-us.net>,
	Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
	Linux Crypto Mailing List <linux-crypto@vger.kernel.org>,
	linux-arch <linux-arch@vger.kernel.org>,
	Dinh Nguyen <dinguyen@kernel.org>, Nick Hu <nickhu@andestech.com>,
	Max Filippov <jcmvbkbc@gmail.com>,
	Palmer Dabbelt <palmer@dabbelt.com>,
	"David S . Miller" <davem@davemloft.net>,
	Yoshinori Sato <ysato@users.sourceforge.jp>,
	Michal Simek <monstr@monstr.eu>, Borislav Petkov <bp@alien8.de>,
	Guo Ren <guoren@kernel.org>,
	Geert Uytterhoeven <geert@linux-m68k.org>,
	Joshua Kinard <kumba@gentoo.org>,
	David Laight <David.Laight@aculab.com>,
	Dominik Brodowski <linux@dominikbrodowski.net>,
	Eric Biggers <ebiggers@google.com>,
	Ard Biesheuvel <ardb@kernel.org>, Arnd Bergmann <arnd@arndb.de>,
	Thomas Gleixner <tglx@linutronix.de>,
	Andy Lutomirski <luto@kernel.org>,
	Kees Cook <keescook@chromium.org>,
	Lennart Poettering <mzxreary@0pointer.de>,
	Konstantin Ryabitsev <konstantin@linuxfoundation.org>,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	"Theodore Ts'o" <tytso@mit.edu>
Subject: Re: [PATCH v1] random: block in /dev/urandom
Date: Tue, 22 Mar 2022 12:36:35 -0600	[thread overview]
Message-ID: <CAHmME9oi-XM--3c4fbM6LydM_Q7CTBdEzKC9fDEE7gXcwUwtcw@mail.gmail.com> (raw)
In-Reply-To: <CAHk-=wiq3bKDdt7noWOaMnDL-yYfFHb1CEsNkk8huq4O7ByetA@mail.gmail.com>

Hi Linus,

On Tue, Mar 22, 2022 at 12:29 PM Linus Torvalds
<torvalds@linux-foundation.org> wrote:
> Christ, how I hate the crazy "no entropy means that we can't use it".
>
> It's a disease, I tell you.
>
> And it seems to be the direct cause of this misfeature.

A disease indeed.

> By all means the code can say "I can't credit this as entropy", but
> the fact that it then doesn't even mix it into the fast pool is just
> wrong, wrong, wrong.
>
> I think *that* is what we should fix. The fact is, urandom has
> long-standing semantics as "don't block", and that it shouldn't care
> about the (often completely insane) entropy crediting rules.
>
> But that "don't care about entropy rules" should then also mean "oh,
> we'll mix things in even if we don't credit entropy".
>
> I hope that's the easy fix you are thinking about.

Yes, exactly. And the patch to fix it is literally 2 lines. I'm
playing with it now and I'll think about it a bit and hopefully have
something for you to pull not before long.

In general, your intuition is correct, I think, that the entropy
crediting scheme is sort of insane and leads to problems. As I wrote
in <https://www.zx2c4.com/projects/linux-rng-5.17-5.18/> at the end,
other RNG schemes like Fortuna don't really suffer from this in the
same way because they're not even counting entropy. This might be
something to look at seriously in the future.

Jason