Message-ID: <aM5bizfTTTAH5Xoa@krava>
Date: Sat, 20 Sep 2025 09:45:15 +0200
From: Jiri Olsa <olsajiri@...il.com>
To: Steven Rostedt <rostedt@...dmis.org>
Cc: "Masami Hiramatsu (Google)" <mhiramat@...nel.org>,
	Peter Zijlstra <peterz@...radead.org>,
	Steven Rostedt <rostedt@...nel.org>,
	Menglong Dong <menglong8.dong@...il.com>, tglx@...utronix.de,
	mingo@...hat.com, bp@...en8.de, dave.hansen@...ux.intel.com,
	x86@...nel.org, hpa@...or.com, kees@...nel.org,
	samitolvanen@...gle.com, rppt@...nel.org, luto@...nel.org,
	ast@...nel.org, andrii@...nel.org, linux-kernel@...r.kernel.org,
	bpf@...r.kernel.org
Subject: Re: [PATCH] tracing: fgraph: Protect return handler from recursion
 loop

On Fri, Sep 19, 2025 at 11:27:46AM -0400, Steven Rostedt wrote:
> On Fri, 19 Sep 2025 20:57:36 +0900
> "Masami Hiramatsu (Google)" <mhiramat@...nel.org> wrote:
> 
> > From: Masami Hiramatsu (Google) <mhiramat@...nel.org>
> > 
> > function_graph_enter_regs() prevents itself from recursion by
> > ftrace_test_recursion_trylock(), but __ftrace_return_to_handler(),
> > which is called at the exit, does not prevent such recursion.
> > Therefore, while it can prevent recursive calls from
> > fgraph_ops::entryfunc(), it is not able to prevent recursive calls
> > to fgraph from fgraph_ops::retfunc(), resulting in a recursive loop.
> > This can lead an unexpected recursion bug reported by Menglong.
> > 
> >  is_endbr() is called in __ftrace_return_to_handler -> fprobe_return
> >   -> kprobe_multi_link_exit_handler -> is_endbr.  
> 
> So basically its if the handler for the return part calls something that it
> is tracing, it can trigger the recursion?
> 
> > 
> > To fix this issue, acquire ftrace_test_recursion_trylock() in the
> > __ftrace_return_to_handler() after unwind the shadow stack to mark
> > this section must prevent recursive call of fgraph inside user-defined
> > fgraph_ops::retfunc().
> > 
> > This is essentially a fix to commit 4346ba160409 ("fprobe: Rewrite
> > fprobe on function-graph tracer"), because before that fgraph was
> > only used from the function graph tracer. Fprobe allowed user to run
> > any callbacks from fgraph after that commit.
> 
> I would actually say it's because before this commit, the return handler
> callers never called anything that the entry handlers didn't already call.
> If there was recursion, the entry handler would catch it (and the entry
> tells fgraph if the exit handler should be called).
> 
> The difference here is with fprobes, you can have the exit handler calling
> functions that the entry handler does not, which exposes more cases where
> recursion could happen.

so IIUC we have return kprobe multi probe on is_endbr and now we do:
	
	is_endbr()
	{ -> function_graph_enter_regs installs return probe
	  ...
	} -> __ftrace_return_to_handler
	       fprobe_return
	         kprobe_multi_link_exit_handler
	           is_endbr
		   { -> function_graph_enter_regs installs return probe
		     ...
		   } -> __ftrace_return_to_handler
		          fprobe_return
		            kprobe_multi_link_exit_handler
			      is_endbr
			      { -> function_graph_enter_regs installs return probe
			        ...
			      } -> __ftrace_return_to_handler
			           ... recursion


with the fix:

	is_endbr()
	{ -> function_graph_enter_regs installs return probe
	  ...
	} -> __ftrace_return_to_handler
	       fprobe_return
	         kprobe_multi_link_exit_handler
	           ...
	           is_endbr
		   { ->  function_graph_enter_regs
		           ftrace_test_recursion_trylock fails and we do NOT install return probe
                     ...
		   }


there's is_endbr call also in kprobe_multi_link_handler, but it won't
trigger recursion, because function_graph_enter_regs already uses
ftrace_test_recursion_trylock 


if above is correct then the fix looks good to me

Acked-by: Jiri Olsa <jolsa@...nel.org>

thanks,
jirka


> 
> > 
> > Reported-by: Menglong Dong <menglong8.dong@...il.com>
> > Closes: https://lore.kernel.org/all/20250918120939.1706585-1-dongml2@chinatelecom.cn/
> > Fixes: 4346ba160409 ("fprobe: Rewrite fprobe on function-graph tracer")
> > Cc: stable@...r.kernel.org
> > Signed-off-by: Masami Hiramatsu (Google) <mhiramat@...nel.org>
> > ---
> >  kernel/trace/fgraph.c |   12 ++++++++++++
> >  1 file changed, 12 insertions(+)
> > 
> > diff --git a/kernel/trace/fgraph.c b/kernel/trace/fgraph.c
> > index 1e3b32b1e82c..08dde420635b 100644
> > --- a/kernel/trace/fgraph.c
> > +++ b/kernel/trace/fgraph.c
> > @@ -815,6 +815,7 @@ __ftrace_return_to_handler(struct ftrace_regs *fregs, unsigned long frame_pointe
> >  	unsigned long bitmap;
> >  	unsigned long ret;
> >  	int offset;
> > +	int bit;
> >  	int i;
> >  
> >  	ret_stack = ftrace_pop_return_trace(&trace, &ret, frame_pointer, &offset);
> > @@ -829,6 +830,15 @@ __ftrace_return_to_handler(struct ftrace_regs *fregs, unsigned long frame_pointe
> >  	if (fregs)
> >  		ftrace_regs_set_instruction_pointer(fregs, ret);
> >  
> > +	bit = ftrace_test_recursion_trylock(trace.func, ret);
> > +	/*
> > +	 * This must be succeeded because the entry handler returns before
> > +	 * modifying the return address if it is nested. Anyway, we need to
> > +	 * avoid calling user callbacks if it is nested.
> > +	 */
> > +	if (WARN_ON_ONCE(bit < 0))
> 
> I'm not so sure we need the warn on here. We should probably hook it to the
> recursion detection infrastructure that the function tracer has.
> 
> The reason I would say not to have the warn on, is because we don't have a
> warn on for recursion happening at the entry handler. Because this now is
> exposed by fprobe allowing different routines to be called at exit than
> what is used in entry, it can easily be triggered.
> 
> -- Steve
> 
> 
> 
> > +		goto out;
> > +
> >  #ifdef CONFIG_FUNCTION_GRAPH_RETVAL
> >  	trace.retval = ftrace_regs_get_return_value(fregs);
> >  #endif
> > @@ -852,6 +862,8 @@ __ftrace_return_to_handler(struct ftrace_regs *fregs, unsigned long frame_pointe
> >  		}
> >  	}
> >  
> > +	ftrace_test_recursion_unlock(bit);
> > +out:
> >  	/*
> >  	 * The ftrace_graph_return() may still access the current
> >  	 * ret_stack structure, we need to make sure the update of
> 

Powered by blists - more mailing lists