Re: [PATCH v3 gnumach] smp: Rearrange IPI sending mechanism

[Samuel Thibault]

Damien Zammit, le ven. 23 févr. 2024 08:14:53 +0000, a ecrit:
> Wait for ICR then just assert the signal.
> Linux and NetBSD does it this way, assume they know better.

But also possibly not. IIRC Almudena had good reasons to set the code
this way.

What problem does this actually solve?

Remember that commit logs should always document the "why the change".

> ---
>  i386/i386/smp.c | 8 +-------
>  1 file changed, 1 insertion(+), 7 deletions(-)
> 
> diff --git a/i386/i386/smp.c b/i386/i386/smp.c
> index 05e9de67..a758eea3 100644
> --- a/i386/i386/smp.c
> +++ b/i386/i386/smp.c
> @@ -54,17 +54,11 @@ static void smp_send_ipi(unsigned apic_id, unsigned 
> vector)
>  
>      cpu_intr_save(&flags);
>  
> -    apic_send_ipi(NO_SHORTHAND, FIXED, PHYSICAL, ASSERT, EDGE, vector, 
> apic_id);
> -
>      do {
>          cpu_pause();
>      } while(lapic->icr_low.delivery_status == SEND_PENDING);
>  
> -    apic_send_ipi(NO_SHORTHAND, FIXED, PHYSICAL, DE_ASSERT, EDGE, vector, 
> apic_id);
> -
> -    do {
> -        cpu_pause();
> -    } while(lapic->icr_low.delivery_status == SEND_PENDING);
> +    apic_send_ipi(NO_SHORTHAND, FIXED, PHYSICAL, ASSERT, EDGE, vector, 
> apic_id);
>  
>      cpu_intr_restore(flags);
>  }
> -- 
> 2.43.0
> 
> 
> 

-- 
Samuel
---
Pour une évaluation indépendante, transparente et rigoureuse !
Je soutiens la Commission d'Évaluation de l'Inria.