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

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

Date:   Wed, 4 Apr 2018 13:13:10 -0400
From:   Steven Rostedt <rostedt@...dmis.org>
To:     Sergey Senozhatsky <sergey.senozhatsky.work@...il.com>
Cc:     Kees Cook <keescook@...omium.org>,
        LKML <linux-kernel@...r.kernel.org>,
        "Tobin C. Harding" <me@...in.cc>,
        Linus Torvalds <torvalds@...ux-foundation.org>,
        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, 4 Apr 2018 11:34:55 +0900
Sergey Senozhatsky <sergey.senozhatsky.work@...il.com> wrote:

> On (04/03/18 18:03), Steven Rostedt wrote:
> >   
> > > he'd want you to change all the trace_printk()s to %px with
> > > justifications, though.  
> > 
> > What trace_printk()s do you want to change? They are throw away
> > functions. trace_printk() is not something that stays in the kernel.
> > It's added during debugging and then removed before submitting what you
> > are working on upstream.  
> 
> Seems that your patch also can fix up bpf_trace_printk() - it doesn't
> even support %px, only hashed %p, which it passes to __trace_printk().
> 

Crap, that should not work, but it does. I need to update the code to
prevent that from happening. I don't want any user space code to be
able to enable this.

Something like this will even prevent modules from disabling the printk
hash...

diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index 2a1d01666dd2..c4af083742b0 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -2834,6 +2834,14 @@ static int alloc_percpu_trace_buffer(void)
 
 static int buffers_allocated;
 
+/* Will become read only after boot */
+static const bool disable_printk_hash;
+
+static void set_printk_hash(bool val)
+{
+	*(bool *)&disable_printk_hash = val;
+}
+
 void trace_printk_init_buffers(void)
 {
 	if (buffers_allocated)
@@ -2864,9 +2872,11 @@ void trace_printk_init_buffers(void)
 
 	buffers_allocated = 1;
 
-	/* This is a debug kernel, allow pointers to be shown */
-	static_branch_enable(&trace_debug);
-	kptr_restrict = 0;
+	if (disable_printk_hash) {
+		/* This is a debug kernel, allow pointers to be shown */
+		static_branch_enable(&trace_debug);
+		kptr_restrict = 0;
+	}
 
 	/*
 	 * trace_printk_init_buffers() can be called by modules.
@@ -8456,10 +8466,13 @@ __init static int tracer_alloc_buffers(void)
 	if (!alloc_cpumask_var(&global_trace.tracing_cpumask, GFP_KERNEL))
 		goto out_free_buffer_mask;
 
+	/* Only trace_printk() in the core kernel can disable hashing */
+	set_printk_hash(true);
 	/* Only allocate trace_printk buffers if a trace_printk exists */
 	if (__stop___trace_bprintk_fmt != __start___trace_bprintk_fmt)
 		/* Must be called before global_trace.buffer is allocated */
 		trace_printk_init_buffers();
+	set_printk_hash(false);
 
 	/* To save memory, keep the ring buffer size to its minimum */
 	if (ring_buffer_expanded)



-- Steve