Re: [PATCH v20 00/19] tracing: fprobe: function_graph: Multi-function graph and fprobe on fgraph

[Masami Hiramatsu (Google) <mhiramat@kernel.org>]

On Sat, 7 Dec 2024 09:47:10 -0500
Steven Rostedt <rostedt@goodmis.org> wrote:

> On Sat, 7 Dec 2024 15:51:36 +0900
> Masami Hiramatsu (Google) <mhiramat@kernel.org> wrote:
> 
> > This checks the handler is called with preempt off.
> > 
> > On x86_64, the ftrace_graph_func calls function_graph_enter_regs() with
> > ftrace_test_recursion_trylock() as below;
> > 
> > void ftrace_graph_func(unsigned long ip, unsigned long parent_ip,
> > 		       struct ftrace_ops *op, struct ftrace_regs *fregs)
> > {
> > 	struct pt_regs *regs = &arch_ftrace_regs(fregs)->regs;
> > 	unsigned long *stack = (unsigned long *)kernel_stack_pointer(regs);
> > 	unsigned long return_hooker = (unsigned long)&return_to_handler;
> > 	unsigned long *parent = (unsigned long *)stack;
> > 	int bit;
> > 
> > 	if (unlikely(skip_ftrace_return()))
> > 		return;
> > 
> > 	bit = ftrace_test_recursion_trylock(ip, *parent);
> > 	if (bit < 0)
> > 		return;
> > 
> > 	if (!function_graph_enter_regs(*parent, ip, 0, parent, fregs))
> > 		*parent = return_hooker;
> > 
> > 	ftrace_test_recursion_unlock(bit);
> > }
> > 
> > However, arm64 version does not;
> 
> Hmm, I think we can move that recursion check out of the arch/x86 code
> and into ftrace_graph_enter_regs().

OK, that's reasonable.

Thanks,

> 
> -- Steve


-- 
Masami Hiramatsu (Google) <mhiramat@kernel.org>