Re: [patch V4 part 1 36/36] rcu: Make RCU IRQ enter/exit functions rely on in_nmi()

[Paul E. McKenney]

On Tue, May 05, 2020 at 03:16:38PM +0200, Thomas Gleixner wrote:
> From: Paul E. McKenney <paulmck@xxxxxxxxxx>
> 
> The rcu_nmi_enter_common() and rcu_nmi_exit_common() functions take an
> "irq" parameter that indicates whether these functions are invoked from
> an irq handler (irq==true) or an NMI handler (irq==false).  However,
> recent changes have applied notrace to a few critical functions such
> that rcu_nmi_enter_common() and rcu_nmi_exit_common() many now rely
> on in_nmi().  Note that in_nmi() works no differently than before,
> but rather that tracing is now prohibited in code regions where in_nmi()
> would incorrectly report NMI state.
> 
> Therefore remove the "irq" parameter and inlines rcu_nmi_enter_common() and
> rcu_nmi_exit_common() into rcu_nmi_enter() and rcu_nmi_exit(),
> respectively.

Not a bad job of ghostwriting, actually.  I had forgotten about this
entirely, but did find my February 13th email containing this patch.  ;-)

But why not make the commit log official?

------------------------------------------------------------------------

The rcu_nmi_enter_common() and rcu_nmi_exit_common() functions take an
"irq" parameter that indicates whether these functions have been invoked
from an irq handler (irq==true) or an NMI handler (irq==false).  However,
recent changes have applied notrace to a number of critical functions,
thus allowing rcu_nmi_enter_common() and rcu_nmi_exit_common() to rely
on in_nmi().  Note that in_nmi() works no differently than before.
Instead, tracing is now prohibited in code regions where in_nmi() would
previously have incorrectly reported NMI state.

This commit therefore removes the "irq" parameter and inlines
rcu_nmi_enter_common() and rcu_nmi_exit_common() into rcu_nmi_enter()
and rcu_nmi_exit(), respectively.

------------------------------------------------------------------------

With that commit log, looks good to me!

							Thanx, Paul

> Signed-off-by: Paul E. McKenney <paulmck@xxxxxxxxxx>
> Signed-off-by: Peter Zijlstra (Intel) <peterz@xxxxxxxxxxxxx>
> Signed-off-by: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
> ---
>  kernel/rcu/tree.c |   47 +++++++++++++++--------------------------------
>  1 file changed, 15 insertions(+), 32 deletions(-)
> 
> --- a/kernel/rcu/tree.c
> +++ b/kernel/rcu/tree.c
> @@ -627,16 +627,18 @@ noinstr void rcu_user_enter(void)
>  }
>  #endif /* CONFIG_NO_HZ_FULL */
>  
> -/*
> +/**
> + * rcu_nmi_exit - inform RCU of exit from NMI context
> + *
>   * If we are returning from the outermost NMI handler that interrupted an
>   * RCU-idle period, update rdp->dynticks and rdp->dynticks_nmi_nesting
>   * to let the RCU grace-period handling know that the CPU is back to
>   * being RCU-idle.
>   *
> - * If you add or remove a call to rcu_nmi_exit_common(), be sure to test
> + * If you add or remove a call to rcu_nmi_exit(), be sure to test
>   * with CONFIG_RCU_EQS_DEBUG=y.
>   */
> -static __always_inline void rcu_nmi_exit_common(bool irq)
> +noinstr void rcu_nmi_exit(void)
>  {
>  	struct rcu_data *rdp = this_cpu_ptr(&rcu_data);
>  
> @@ -667,7 +669,7 @@ static __always_inline void rcu_nmi_exit
>  	trace_rcu_dyntick(TPS("Startirq"), rdp->dynticks_nmi_nesting, 0, atomic_read(&rdp->dynticks));
>  	WRITE_ONCE(rdp->dynticks_nmi_nesting, 0); /* Avoid store tearing. */
>  
> -	if (irq)
> +	if (!in_nmi())
>  		rcu_prepare_for_idle();
>  	instr_end();
>  
> @@ -675,22 +677,11 @@ static __always_inline void rcu_nmi_exit
>  	rcu_dynticks_eqs_enter();
>  	// ... but is no longer watching here.
>  
> -	if (irq)
> +	if (!in_nmi())
>  		rcu_dynticks_task_enter();
>  }
>  
>  /**
> - * rcu_nmi_exit - inform RCU of exit from NMI context
> - *
> - * If you add or remove a call to rcu_nmi_exit(), be sure to test
> - * with CONFIG_RCU_EQS_DEBUG=y.
> - */
> -void noinstr rcu_nmi_exit(void)
> -{
> -	rcu_nmi_exit_common(false);
> -}
> -
> -/**
>   * rcu_irq_exit - inform RCU that current CPU is exiting irq towards idle
>   *
>   * Exit from an interrupt handler, which might possibly result in entering
> @@ -712,7 +703,7 @@ void noinstr rcu_nmi_exit(void)
>  void noinstr rcu_irq_exit(void)
>  {
>  	lockdep_assert_irqs_disabled();
> -	rcu_nmi_exit_common(true);
> +	rcu_nmi_exit();
>  }
>  
>  /*
> @@ -801,7 +792,7 @@ void noinstr rcu_user_exit(void)
>  #endif /* CONFIG_NO_HZ_FULL */
>  
>  /**
> - * rcu_nmi_enter_common - inform RCU of entry to NMI context
> + * rcu_nmi_enter - inform RCU of entry to NMI context
>   * @irq: Is this call from rcu_irq_enter?
>   *
>   * If the CPU was idle from RCU's viewpoint, update rdp->dynticks and
> @@ -810,10 +801,10 @@ void noinstr rcu_user_exit(void)
>   * long as the nesting level does not overflow an int.  (You will probably
>   * run out of stack space first.)
>   *
> - * If you add or remove a call to rcu_nmi_enter_common(), be sure to test
> + * If you add or remove a call to rcu_nmi_enter(), be sure to test
>   * with CONFIG_RCU_EQS_DEBUG=y.
>   */
> -static __always_inline void rcu_nmi_enter_common(bool irq)
> +noinstr void rcu_nmi_enter(void)
>  {
>  	long incby = 2;
>  	struct rcu_data *rdp = this_cpu_ptr(&rcu_data);
> @@ -831,18 +822,18 @@ static __always_inline void rcu_nmi_ente
>  	 */
>  	if (rcu_dynticks_curr_cpu_in_eqs()) {
>  
> -		if (irq)
> +		if (!in_nmi())
>  			rcu_dynticks_task_exit();
>  
>  		// RCU is not watching here ...
>  		rcu_dynticks_eqs_exit();
>  		// ... but is watching here.
>  
> -		if (irq)
> +		if (!in_nmi())
>  			rcu_cleanup_after_idle();
>  
>  		incby = 1;
> -	} else if (irq) {
> +	} else if (!in_nmi()) {
>  		instr_begin();
>  		if (tick_nohz_full_cpu(rdp->cpu) &&
>  		    rdp->dynticks_nmi_nesting == DYNTICK_IRQ_NONIDLE &&
> @@ -877,14 +868,6 @@ static __always_inline void rcu_nmi_ente
>  }
>  
>  /**
> - * rcu_nmi_enter - inform RCU of entry to NMI context
> - */
> -noinstr void rcu_nmi_enter(void)
> -{
> -	rcu_nmi_enter_common(false);
> -}
> -
> -/**
>   * rcu_irq_enter - inform RCU that current CPU is entering irq away from idle
>   *
>   * Enter an interrupt handler, which might possibly result in exiting
> @@ -909,7 +892,7 @@ noinstr void rcu_nmi_enter(void)
>  noinstr void rcu_irq_enter(void)
>  {
>  	lockdep_assert_irqs_disabled();
> -	rcu_nmi_enter_common(true);
> +	rcu_nmi_enter();
>  }
>  
>  /*
>