linux-kernel - Re: [GIT PULL] hash addresses printed with %p

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: <20171129210848.GF6217@eros>
Date:   Thu, 30 Nov 2017 08:08:48 +1100
From:   "Tobin C. Harding" <me@...in.cc>
To:     Linus Torvalds <torvalds@...ux-foundation.org>
Cc:     LKML <linux-kernel@...r.kernel.org>
Subject: Re: [GIT PULL] hash addresses printed with %p

On Wed, Nov 29, 2017 at 11:22:29AM -0800, Linus Torvalds wrote:
> On Tue, Nov 28, 2017 at 8:59 PM, Tobin C. Harding <me@...in.cc> wrote:
> >
> >   git://github.com/tcharding/linux.git tags/printk-hash-pointer-4.15-rc2
> 
> Bah.

Sorry for creating extra work for you.

> What I didn't realize until after pulling this and testing, is that it
> completely breaks '%pK'.

If you haven't wasted enough time on this can you tell me what you mean
by 'completely breaks %pK'?

If I am at fault I do not want to repeat the same mistake again.

I have just re-run my tests and it passes so something must be wrong
with my tests or method. I wrote a module to print various pointers
using %pK (same module that tests the hashing stuff), built the kernel
with the patch set applied then booted the kernel in a VM and inserted
the module (kptr_restrict==0). Confirmed that addresses were
displayed. Then I set kptr_restrict to 2 and re-inserted the
module. Confirmed that pointers were zeroed out when printed with %pK.

> We've marked various sensitive pointers with %pK, but that is now
> _less_ secure than %p is, since it doesn't do the hashing because of
> how you refactored the %pK code out of 'pointer()' into its own
> function.

Oh, I think I get it. You mean that it is better to hash the address for
%pK (kpt_restrict==0) than to zero it out?

> So now %pK ends up using the plain "number()" function. Reading
> through the series I hadn't noticed that the refactoring ended up
> messing with that.
> 
> I'll fix it up somehow.

(I saw the fix in the next email)

thanks,
Tobin.