Re: [PATCH] ACPI / scan: Annotate physical_node_lock in acpi_scan_is_offline()

From: "Rafael J. Wysocki" <rjw@rjwysocki.net>
To: Hanjun Guo <hanjun.guo@linaro.org>
Cc: linux-acpi@vger.kernel.org, Xie XiuQi <xiexiuqi@huawei.com>,
	guohanjun@huawei.com, linux-kernel@vger.kernel.org,
	Toshi Kani <toshi.kani@hp.com>
Subject: Re: [PATCH] ACPI / scan: Annotate physical_node_lock in acpi_scan_is_offline()
Date: Mon, 13 Apr 2015 15:48:53 +0200	[thread overview]
Message-ID: <9790151.kO7BNWOEJr@vostro.rjw.lan> (raw)
In-Reply-To: <552B7DE4.3070008@linaro.org>

On Monday, April 13, 2015 04:27:16 PM Hanjun Guo wrote:
> On 2015年04月11日 07:31, Rafael J. Wysocki wrote:
> > From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
> >
> > acpi_scan_is_offline() may be called under the physical_node_lock
> > of the given device object's parent, so prevent lockdep from
> > complaining about that by annotating that instance with
> > SINGLE_DEPTH_NESTING.
> 
> I think this is trigged by setting acpi_force_hot_remove to 1,
> in acpi_scan_hot_remove():
> 
>          if (device->handler && device->handler->hotplug.demand_offline
>              && !acpi_force_hot_remove) {
>                  if (!acpi_scan_is_offline(device, true))
>                          return -EBUSY;
>          } else {
>                  int error = acpi_scan_try_to_offline(device);
>                  if (error)
>                          return error;
>          }
> 
> then the container device will be removed by acpi_scan_try_to_offline(),
> let's wait for Xiuqi's test result.

I'm not sure what you mean.  demand_offline is 'true' for containers, so
acpi_force_hot_remove doesn't matter here.

-- 
I speak only for myself.
Rafael J. Wysocki, Intel Open Source Technology Center.