[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <9034CBD80F070943B59700D7F8149ED9024EB81CAF@SC-VEXCH4.marvell.com>
Date: Sun, 19 Oct 2014 23:39:23 -0700
From: Neil Zhang <zhangwm@...vell.com>
To: Greg KH <gregkh@...uxfoundation.org>
CC: "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: RE: [PATCH] drivers: base: update cpu offline info when do hotplug
Greg,
-----Original Message-----
From: Greg KH [mailto:gregkh@...uxfoundation.org]
Sent: 2014年10月20日 12:44
To: Neil Zhang
Cc: linux-kernel@...r.kernel.org
Subject: Re: [PATCH] drivers: base: update cpu offline info when do hotplug
On Mon, Oct 20, 2014 at 11:29:08AM +0800, Neil Zhang wrote:
> The current per-cpu offline info won't be updated if it is hotplugged
> in/out by a kernel governer.
> Let's update it via cpu notifier.
>
> Signed-off-by: Neil Zhang <zhangwm@...vell.com>
> ---
> drivers/base/cpu.c | 25 +++++++++++++++++++++++++
> 1 file changed, 25 insertions(+)
>
> diff --git a/drivers/base/cpu.c b/drivers/base/cpu.c index
> 006b1bc..9d61824 100644
> --- a/drivers/base/cpu.c
> +++ b/drivers/base/cpu.c
> @@ -418,10 +418,35 @@ static void __init
> cpu_dev_register_generic(void) #endif }
>
> +static int device_hotplug_notifier(struct notifier_block *nfb,
> + unsigned long action, void *hcpu) {
> + unsigned int cpu = (unsigned long)hcpu;
> + struct device *dev = get_cpu_device(cpu);
> + int ret;
> +
> + switch (action & ~CPU_TASKS_FROZEN) {
> + case CPU_ONLINE:
> + dev->offline = false;
> + ret = NOTIFY_OK;
> + break;
> + case CPU_DYING:
> + dev->offline = true;
> + ret = NOTIFY_OK;
> + break;
> + default:
> + ret = NOTIFY_DONE;
> + break;
> + }
> +
> + return ret;
> +}
> +
> void __init cpu_dev_init(void)
> {
> if (subsys_system_register(&cpu_subsys, cpu_root_attr_groups))
> panic("Failed to register CPU subsystem");
>
> cpu_dev_register_generic();
> + cpu_notifier(device_hotplug_notifier, 0);
> }
How much noise is this going to cause on a big/little system that constantly hot unplug/plugs processors all of the time?
Can you explain more what kind of noise will be introduced on a big/little system?
As I know IKS on arm will use cpu_suspend way to power down a core.
But I don't know well about other architectures.
Please give your suggestions.
Thanks!
greg k-h
Best Regards,
Neil Zhang
Powered by blists - more mailing lists