Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <89452cc1-5322-58f1-0f4a-41c2ad201be7@infradead.org>
Date:   Tue, 30 May 2023 16:32:39 -0700
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.11.0
Subject: Re: [PATCH 2/2] genirq: fasteoi resends interrupt on concurrent
 invoke
Content-Language: en-US
To:     James Gowans <jgowans@amazon.com>,
        Thomas Gleixner <tglx@linutronix.de>
Cc:     linux-kernel@vger.kernel.org, Liao Chang <liaochang1@huawei.com>,
        Marc Zyngier <maz@kernel.org>,
        KarimAllah Raslan <karahmed@amazon.com>,
        Yipeng Zou <zouyipeng@huawei.com>,
        Zhang Jianhua <chris.zjh@huawei.com>
References: <20230530213848.3273006-1-jgowans@amazon.com>
 <20230530213848.3273006-2-jgowans@amazon.com>
From:   Randy Dunlap <rdunlap@infradead.org>
In-Reply-To: <20230530213848.3273006-2-jgowans@amazon.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
Precedence: bulk

Hi--

On 5/30/23 14:38, James Gowans wrote:

> ---
>  kernel/irq/chip.c | 15 ++++++++++++++-
>  1 file changed, 14 insertions(+), 1 deletion(-)
> 
> diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c
> index 49e7bc871fec..42f33e77c16b 100644
> --- a/kernel/irq/chip.c
> +++ b/kernel/irq/chip.c
> @@ -692,8 +692,15 @@ void handle_fasteoi_irq(struct irq_desc *desc)
>  
>  	raw_spin_lock(&desc->lock);
>  
> -	if (!irq_may_run(desc))
> +	/*
> +	 * When an affinity change races with IRQ delivery, the next interrupt
> +	 * can arrive on the new CPU before the original CPU has completed
> +	 * handling the previous one. Mark it as pending and return EOI.
> +	 */
> +	if (!irq_may_run(desc)) {
> +		desc->istate |= IRQS_PENDING;
>  		goto out;
> +	}
>  
>  	desc->istate &= ~(IRQS_REPLAY | IRQS_WAITING);
>  
> @@ -715,6 +722,12 @@ void handle_fasteoi_irq(struct irq_desc *desc)
>  
>  	cond_unmask_eoi_irq(desc, chip);
>  
> +	/*
> +	 * When the race descibed above happens, this will resend the interrupt.

	                 described

> +	 */
> +	if (unlikely(desc->istate & IRQS_PENDING))
> +		check_irq_resend(desc, false);
> +
>  	raw_spin_unlock(&desc->lock);
>  	return;
>  out:

-- 
~Randy