Re: kprobes broken since 0d00449c7a28 ("x86: Replace ist_enter() with nmi_enter()")

[Steven Rostedt <rostedt@goodmis.org>]

On Fri, 29 Jan 2021 14:01:03 -0500
Steven Rostedt <rostedt@goodmis.org> wrote:

> On Fri, 29 Jan 2021 18:59:43 +0100
> Peter Zijlstra <peterz@infradead.org> wrote:
> 
> > On Fri, Jan 29, 2021 at 09:45:48AM -0800, Alexei Starovoitov wrote:  
> > > Same things apply to bpf side. We can statically prove safety for
> > > ftrace and kprobe attaching whereas to deal with NMI situation we
> > > have to use run-time checks for recursion prevention, etc.    
> > 
> > I have no idea what you're saying. You can attach to functions that are
> > called with random locks held, you can create kprobes in some very
> > sensitive places.
> > 
> > What can you staticlly prove about that?  
> 
> I think the main difference is, if you attach a kprobe or ftrace function,
> you can theoretically analyze the location before you do the attachment.
> 
> Does, the NMI context mean "in_nmi()" returns true? Because there's cases
> in ftrace callbacks where that is checked (like the stack tracer). And
> having ftrace return true for "in_nmi()" will break a lot of existing
> utilities.

Specifically, kprobe and ftrace callbacks may have this:

	if (in_nmi())
		return;

	raw_spin_lock_irqsave(&lock, flags);
	[..]
	raw_spin_unlock_irqrestore(&lock, flags);

Which is totally fine to have, but the above only works if "in_nmi()"
returns true only if you are in a real NMI.

The stack tracer code does exactly the above.

-- Steve