linux-kernel - Re: [BUG] msr-trace.h:42 suspicious rcu_dereference

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: <20161121091543.45f49945@gandalf.local.home>
Date:   Mon, 21 Nov 2016 09:15:43 -0500
From:   Steven Rostedt <rostedt@...dmis.org>
To:     Peter Zijlstra <peterz@...radead.org>
Cc:     Jiri Olsa <jolsa@...hat.com>,
        "Paul E. McKenney" <paulmck@...ux.vnet.ibm.com>,
        linux-kernel@...r.kernel.org, Ingo Molnar <mingo@...nel.org>,
        Josh Triplett <josh@...htriplett.org>,
        Andi Kleen <andi@...stfloor.org>,
        Jan Stancek <jstancek@...hat.com>,
        Borislav Petkov <bp@...en8.de>
Subject: Re: [BUG] msr-trace.h:42 suspicious rcu_dereference_check() usage!

On Mon, 21 Nov 2016 13:58:30 +0100
Peter Zijlstra <peterz@...radead.org> wrote:

> On Mon, Nov 21, 2016 at 10:34:25AM +0100, Jiri Olsa wrote:
> 
> > > diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c
> > > index 0888a879120f..d6c6aa80675f 100644
> > > --- a/arch/x86/kernel/process.c
> > > +++ b/arch/x86/kernel/process.c
> > > @@ -357,7 +357,7 @@ static void amd_e400_idle(void)
> > >  	if (!amd_e400_c1e_detected) {
> > >  		u32 lo, hi;
> > >  
> > > -		rdmsr(MSR_K8_INT_PENDING_MSG, lo, hi);
> > > +		RCU_NONIDLE(rdmsr(MSR_K8_INT_PENDING_MSG, lo, hi));
> > >  
> > >  		if (lo & K8_INTP_C1E_ACTIVE_MASK) {
> > >  			amd_e400_c1e_detected = true;  
> 
> OK, so while looking at this again, I don't like this ether :/
> 
> Problem with this one is that it always adds the RCU fiddling overhead,
> even when we're not tracing.
> 
> I could do an rdmsr_notrace() for this one, dunno if its important.

But that would neglect the point of tracing rdmsr. What about:

	/* tracepoints require RCU enabled */
	if (trace_read_msr_enabled())
		RCU_NONIDLE(rdmsr(MSR_K8_INT_PENDING_MSG, lo, hi));
	else
		rdmsr(MSR_K8_INT_PENDING_MSG, lo, hi);

-- Steve