On 08/05/24 15:59, Frederic Weisbecker wrote:
> Le Tue, Apr 30, 2024 at 11:17:31AM +0200, Valentin Schneider a écrit :
>> @@ -228,7 +228,7 @@ void noinstr ct_nmi_exit(void)
>> }
>>
>> /* This NMI interrupted an RCU-idle CPU, restore RCU-idleness. */
>> - trace_rcu_dyntick(TPS("Startirq"), ct_nmi_nesting(), 0, ct_rcu_watching());
>> + trace_rcu_watching(TPS("Endirq"), ct_nmi_nesting(), 0, ct_rcu_watching());
>
> Ah the initial string was wrong and you're fixing it, right?
>
> Should be a seperate patch?
>
No, I'm just creating confusion for everyone involved (including myself) :(
Dynticks start when RCU stops watching, so the naming logic gets flipped on
its head.
If I take the original comment from
bd2b879a1ca5 ("rcu: Add tracing to irq/NMI dyntick-idle transitions")
"""
as argument: "Start" for entering dyntick-idle mode, "Startirq" for
entering it from irq/NMI
"""
So here "Startirq" means "start dyntick mode from IRQ". With the name
change, it should be "Endirq", since RCU stops watching, from IRQ...
I know, a lot of confusion for 0 functional change :(
