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]

Date:   Wed, 29 Nov 2017 13:05:39 -0800
From:   Linus Torvalds <torvalds@...ux-foundation.org>
To:     Joe Perches <joe@...ches.com>
Cc:     "Tobin C. Harding" <me@...in.cc>,
        Kees Cook <keescook@...omium.org>,
        LKML <linux-kernel@...r.kernel.org>
Subject: Re: [GIT PULL] hash addresses printed with %p

On Wed, Nov 29, 2017 at 12:54 PM, Joe Perches <joe@...ches.com> wrote:
>
> I'd prefer a global sed of '%pK' to '%pxK' and remove '%pK' altogether

No, we really don't want %pK to become %pxK.

Most of the %pK users absolutely do not want the real hex address.
They are things like the socket pointers in /proc etc. The exact thing
that the hashing does well, and that %pK was so useless for.

Anyway, with the current tree, the "leaking_addresses script gives
almost no results for me (and I have kptr_restrict set to 0).

I haven't seen any breakage, but I may change my mind if bug reports
end up being illegible. And maybe I just haven't triggered anything
that might use a pointer.

Annotated perf disassembly is broken for me right now, but that
happens to me with alarming regularity because perf uses the wrong
vmlinux file or something, so it is probably not related (the kernel
symbol show up properly in the profile, I just don't get the
disassembly).

               Linus