linux-kernel - Re: [PATCH v2] tracing, printk: Force no hashing when trace

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 for Android: free password hash cracker in your pocket

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

Date:   Wed, 4 Apr 2018 12:04:42 -0700
From:   Linus Torvalds <torvalds@...ux-foundation.org>
To:     Steven Rostedt <rostedt@...dmis.org>
Cc:     Sergey Senozhatsky <sergey.senozhatsky.work@...il.com>,
        Kees Cook <keescook@...omium.org>,
        LKML <linux-kernel@...r.kernel.org>,
        "Tobin C. Harding" <me@...in.cc>,
        Andrew Morton <akpm@...ux-foundation.org>,
        David Laight <David.Laight@...lab.com>,
        Peter Zijlstra <peterz@...radead.org>,
        Ingo Molnar <mingo@...nel.org>,
        Thomas Gleixner <tglx@...utronix.de>,
        Petr Mladek <pmladek@...e.com>
Subject: Re: [PATCH v2] tracing, printk: Force no hashing when trace_printk()
 is used

On Wed, Apr 4, 2018 at 10:13 AM, Steven Rostedt <rostedt@...dmis.org> wrote:
>
> Something like this will even prevent modules from disabling the printk
> hash...

That still seems broken.

The *natural* thing to do would seem to be to tie the hash to the
printk state, kind of like the percpu buffers that safe_printk() and
friends use.

Modifying the hash global is fundamentally broken, since some problem
that happens *during* tracing - on another CPU entirely - would now
have the hashing disabled.

So at the *very* least this would need to be percpu logic, but even
that is honestly broken since an NMI might come in and want to printk
too.

Why don't you just use %px? That avoids all of these hacks.

So NAK on this stupid "enable and disable hashing that is
fundamentally broken" approach.

                     Linus