Re: [PATCH v5] softirq: Preserve interrupt context during IRQ exit

[Frederic Weisbecker]

Hi Karl,

Le Wed, Sep 30, 2026 at 09:14:32PM +0200, Karl Mehltretter a écrit :
> +static inline bool softirq_handle_begin(void)
>  {
> -	__local_bh_disable_ip(_RET_IP_, SOFTIRQ_OFFSET);
> +	bool from_irq_exit = in_hardirq();
> +
> +	if (!from_irq_exit) {

Is this variable necessary?

> +		__local_bh_disable_ip(_RET_IP_, SOFTIRQ_OFFSET);
> +		return false;
> +	}
> +
> +	/*
> +	 * Only reached from irq_exit(), with HARDIRQ_OFFSET still set.
> +	 * Replace it with SOFTIRQ_OFFSET before handle_softirqs() enables
> +	 * interrupts. Use the raw operation to preserve the preemption
> +	 * disable location recorded by irq_enter_rcu(), and update lockdep
> +	 * directly.
> +	 */
> +	__preempt_count_sub(HARDIRQ_OFFSET - SOFTIRQ_OFFSET);
> +	lockdep_softirqs_off(_RET_IP_);
> +	WARN_ON_ONCE(irq_count() != SOFTIRQ_OFFSET);
> +
> +	return true;
>  }
>  
> -static inline void softirq_handle_end(void)
> +static inline void softirq_handle_end(bool from_irq_exit)
>  {
> -	__local_bh_enable(SOFTIRQ_OFFSET);
> -	WARN_ON_ONCE(in_interrupt());
> +	if (!from_irq_exit) {
> +		__local_bh_enable(SOFTIRQ_OFFSET);
> +		WARN_ON_ONCE(in_interrupt());
> +		return;
> +	}
> +
> +	lockdep_softirqs_on(_RET_IP_);
> +	__preempt_count_add(HARDIRQ_OFFSET - SOFTIRQ_OFFSET);
> +	WARN_ON_ONCE(irq_count() != HARDIRQ_OFFSET);
>  }
>  
>  static inline void ksoftirqd_run_begin(void)
> @@ -758,10 +788,21 @@ static inline void __irq_exit_rcu(void)
>  		invoke_softirq();
>  	}
>  
> +	/*
> +	 * Wake the timer thread even if the interrupt hit a softirq or a
> +	 * section with BHs disabled. Only nested interrupts and NMIs are
> +	 * excluded.
> +	 */
>  	if (IS_ENABLED(CONFIG_IRQ_FORCED_THREADING) && force_irqthreads() &&
> -	    local_timers_pending_force_th() && !(in_nmi() | in_hardirq()))
> +	    local_timers_pending_force_th() &&
> +	    (in_nmi() | hardirq_count()) == HARDIRQ_OFFSET)

NMIs shouldn't ever take this path, right?

>  		wake_timersd();
>  
> +	/*
> +	 * tick_irq_exit() relies on in_hardirq() being false for the
> +	 * outermost interrupt.
> +	 */
> +	preempt_count_sub(HARDIRQ_OFFSET);
>  	tick_irq_exit();

Can we also extend the HARDIRQ_OFFSET context coverage to tick_irq_exit() ?
>From a quick look I haven't found anything that would prevent from that. You just need
to turn the "if (!in_hardirq())" test to "if hardirq_count()) == HARDIRQ_OFFSET"

Thanks.

-- 
Frederic Weisbecker
SUSE Labs