linux-kernel - Re: [PATCH] random: make try_to_generate

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAHk-=wip91atX54zCEo5_OwEjyCkOb7_OUvn8P7o_39tMJSe7A@mail.gmail.com>
Date:   Fri, 18 Oct 2019 18:58:50 -0400
From:   Linus Torvalds <torvalds@...ux-foundation.org>
To:     Jörn Engel <joern@...estorage.com>
Cc:     Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] random: make try_to_generate_entropy() more robust

On Fri, Oct 18, 2019 at 4:42 PM Jörn Engel <joern@...estorage.com> wrote:
>
> Sorry for coming late to the discussion.  I generally like the approach
> in try_to_generate_entropy(), but I think we can do a little better
> still.  Would something like this work?

Hmm. I'm not convinced that the register set is all that random in
general if you have attackers (or - in the absence of an attack - if
it hits in the idle loop a lot), but I do like it for this particular
use where we have that timeout while doing entropy work.

So I think this is potentially a good way to at least improve on the
situation when there is no TSC. Which would remove one worry about
getrandom() on other platforms than the usual development ones.

I'm on a plane at 38,933 ft right now according to the flight tracker,
and about to lose internet access again, but I like it and will take
another look when I'm on the ground and at a hotel.

              Linus