From mboxrd@z Thu Jan  1 00:00:00 1970
From: kgene.kim@samsung.com (Kukjin Kim)
Date: Fri, 30 Sep 2011 15:42:23 +0900
Subject: [patch] ARM: smpboot: Enable interrupts after marking CPU
	online/active
In-Reply-To: <alpine.LFD.2.02.1109292211060.2711@ionos>
References: <20110908215314.829452535@linutronix.de>
	<CADGdYn6V_eQJZ4=irMbSS7XJj4MJVeALxcrWdD7o2bGs1sEJZQ@mail.gmail.com>
	<20110913133258.GA6267@n2100.arm.linux.org.uk>
	<CAKfTPtCWuiOM5+sOwdh1FzXZb1EFT__tVEpRPY5suGOxpYTdQQ@mail.gmail.com>
	<20110913175312.GB6267@n2100.arm.linux.org.uk>
	<20110923084001.GP17169@n2100.arm.linux.org.uk>
	<009101cc7e7b$1ebece00$5c3c6a00$%kim@samsung.com>
	<alpine.LFD.2.02.1109292211060.2711@ionos>
Message-ID: <016601cc7f3c$1ce1b0e0$56a512a0$%kim@samsung.com>
To: linux-arm-kernel@lists.infradead.org
List-Id: linux-arm-kernel.lists.infradead.org

Thomas Gleixner wrote:
> 
> On Thu, 29 Sep 2011, Kukjin Kim wrote:
> > Russell King - ARM Linux wrote:
> > But as Amit said, current EXYNOS4 SoCs which are using SPI for local timers
> > has problem that the irq_set_affinity() method is called in
> > percpu_timer_setup() before CPU1 becomes online with the commit
> > 5dfc54e087c15f823ee9b6541d2f0f314e69cbed ("ARM: GIC: avoid routing
> > interrupts to offline CPUs"). So I will check again.
> 
> Why is a per cpu interrupt having an irq_set_affinity() function at all?
> 
Actually, the interrupt of MCT in EXYNOS4210 is SPI(Shared Peripheral Interrupt), which can be routed to any CPUs.
Since its default affinity is 0 so if it is used on CPU1, needs to irq_set_affinity to CPU1 before using it.

Thanks.

Best regards,
Kgene.
--
Kukjin Kim <kgene.kim@samsung.com>, Senior Engineer,
SW Solution Development Team, Samsung Electronics Co., Ltd.