Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1697875

Re: [PATCH] genirq/cpuhotplug: Revert "Set force affinity flag on hotplug migration"

From Will Deacon <will.deacon@arm.com>
Newsgroups linux.kernel
Subject Re: [PATCH] genirq/cpuhotplug: Revert "Set force affinity flag on hotplug migration"
Date 2017-07-27 12:30 +0200
Message-ID <u7Odk-4Tl-17@gated-at.bofh.it> (permalink)
References <u7Odj-4Tl-7@gated-at.bofh.it>
Organization linux.* mail to news gateway

Show all headers | View raw


Thanks Thomas, this should fix our nightly testing which is currently
failing the CPU hotplug tests.

On Thu, Jul 27, 2017 at 12:21:11PM +0200, Thomas Gleixner wrote:
> --- a/kernel/irq/cpuhotplug.c
> +++ b/kernel/irq/cpuhotplug.c
> @@ -95,8 +95,13 @@ static bool migrate_one_irq(struct irq_d
>  		affinity = cpu_online_mask;
>  		brokeaff = true;
>  	}
> -
> -	err = irq_do_set_affinity(d, affinity, true);
> +	/*
> +	 * Do not set the force argument of irq_do_set_affinity() as this
> +	 * disables the masking of offline CPUs from the supplied affinity
> +	 * mask and therefor might keep/reassign the irq to the outgoing

Typo: therefore

> +	 * CPU.
> +	 */
> +	err = irq_do_set_affinity(d, affinity, false);

With that fixed:

Acked-by: Will Deacon <will.deacon@arm.com>

Cheers,

Will

Back to linux.kernel | Previous | NextPrevious in thread | Next in thread | Find similar | Unroll thread


Thread

[PATCH] genirq/cpuhotplug: Revert "Set force affinity flag on hotplug  migration" Thomas Gleixner <tglx@linutronix.de> - 2017-07-27 12:30 +0200
  Re: [PATCH] genirq/cpuhotplug: Revert "Set force affinity flag on  hotplug migration" Will Deacon <will.deacon@arm.com> - 2017-07-27 12:30 +0200
  [tip:irq/urgent] genirq/cpuhotplug: Revert "Set force affinity flag  on hotplug migration" tip-bot for Thomas Gleixner <tipbot@zytor.com> - 2017-07-27 15:50 +0200

csiph-web