Re: [PATCH v2] tracing: Make tracepoint lockdep check actually test something

[Peter Zijlstra]

On Fri, Mar 10, 2023 at 05:28:56PM -0500, Steven Rostedt wrote:

> @@ -249,9 +248,7 @@ static inline struct tracepoint *tracepoint_ptr_deref(tracepoint_ptr_t *p)
>  				TP_ARGS(args),				\
>  				TP_CONDITION(cond), 0);			\
>  		if (IS_ENABLED(CONFIG_LOCKDEP) && (cond)) {		\
> -			rcu_read_lock_sched_notrace();			\
> -			rcu_dereference_sched(__tracepoint_##name.funcs);\
> -			rcu_read_unlock_sched_notrace();		\
> +			WARN_ON_ONCE(!rcu_is_watching());		\
>  		}							\
>  	}								\
>  	__DECLARE_TRACE_RCU(name, PARAMS(proto), PARAMS(args),		\

Yep, that makes heaps more sense. If you so care you can save one more
line and make the {} go away too.

In any case,

Acked-by: Peter Zijlstra (Intel) <peterz@xxxxxxxxxxxxx>