From: Jason Davies <jason@jasondavies.com>
Content-Type: text/plain;
	charset=us-ascii
Content-Transfer-Encoding: 7bit
Mime-Version: 1.0 (Mac OS X Mail 11.3 \(3445.6.18\))
Date: Tue, 10 Apr 2018 09:51:51 +0100
References: <84976adb75bef1dfdb12b98c19811278@national.shitposting.agency>
	<CA+vKqYc3X6ZjVNXs0xgsLGekxPCTcLZj7t2vkyBOV_o=2C2qPA@mail.gmail.com>
	<921edfdb-e0e5-8ce4-55d8-ba4e84ef633f@musalbas.com>
	<010e34a3-f9cf-fba1-5482-de06bc350d64@musalbas.com>
	<69fb5cc4-7b3d-e23d-2b7e-cddcd7b2877b@musalbas.com>
To: bitcoin-dev@lists.linuxfoundation.org
In-Reply-To: <69fb5cc4-7b3d-e23d-2b7e-cddcd7b2877b@musalbas.com>
Message-Id: <333F9973-6092-45B7-A87F-32730D752501@jasondavies.com>
Subject: Re: [bitcoin-dev] KETAMINE: Multiple vulnerabilities in
 SecureRandom(), numerous cryptocurrency products affected.
Precedence: list

On 10 Apr 2018, at 00:39, mus@musalbas.com wrote:

> The original disclosure didn't contain any information about the library
> in question, so I did some digging.
>
> I think that the vulnerability disclosure is referring to a pre-2013
> version of jsbn, a JavaScript crypto library. Before it used the CSRNG
> in the Web Crypto API, it tried to use nsIDOMCrypto, but incorrectly did
> a string comparison when checking the browser version.
>
> In practice though, this doesn't really matter, because
> navigator.appVersion < "5" returns true anyway for old browsers. The
> real issue is that modern browsers don't have window.crypto.random
> defined, so Bitcoin wallets using a pre-2013 version of jsbn may not be
> using a CSPRNG, when run on a modern browser.

Yes, it looks like high-quality entropy via crypto.getRandomValues was only
added in Tom Wu's latest version (v1.4) in July 2013.

Note that even with v1.4, it still does not use high-quality entropy for
Internet Explorer, because getRandomValues is provided under window.msCrypto
for that browser.

  http://www-cs-students.stanford.edu/~tjw/jsbn/rng.js

> As is noted though, even if a CSPRNG is used, the library passes the
> output of the CSPRNG through RC4, which generates some biased bits,
> leading to possible private key recovery.

I think this is the real issue: even if high-quality entropy is utilised, the
RNG is RC4-based, which is known to generate biased output.

Finally, note that even Chrome used RC4 for crypto.getRandomValues at one
point (as recently as 2015)!

  https://bugs.chromium.org/p/chromium/issues/detail?id=552749

--
Jason Davies, https://www.jasondavies.com/