From: Ulf Hansson <ulf.hansson@linaro.org>
Date: Mon, 17 Oct 2016 18:17:01 +0000 (+0200)
Subject: PM / Runtime: Don't allow to suspend a device with an active child
X-Git-Tag: v4.10-rc1~152^2~5^2~3
X-Git-Url: https://www.infradead.org/git/?a=commitdiff_plain;h=a8636c89648a;p=linux.git

PM / Runtime: Don't allow to suspend a device with an active child

When resuming a device in __pm_runtime_set_status(), the prerequisite is
that its parent must already be active, else an error code is returned and
the device's status remains suspended.

When suspending a device there is no similar constraints being validated.
Let's change this to make the behaviour consistent, by not allowing to
suspend a device with an active child, unless it has been explicitly set to
ignore its children.

Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
---

diff --git a/drivers/base/power/runtime.c b/drivers/base/power/runtime.c
index 117db71e84cb..60ebb04d8140 100644
--- a/drivers/base/power/runtime.c
+++ b/drivers/base/power/runtime.c
@@ -1028,7 +1028,17 @@ int __pm_runtime_set_status(struct device *dev, unsigned int status)
 		goto out_set;
 
 	if (status == RPM_SUSPENDED) {
-		/* It always is possible to set the status to 'suspended'. */
+		/*
+		 * It is invalid to suspend a device with an active child,
+		 * unless it has been set to ignore its children.
+		 */
+		if (!dev->power.ignore_children &&
+			atomic_read(&dev->power.child_count)) {
+			dev_err(dev, "runtime PM trying to suspend device but active child\n");
+			error = -EBUSY;
+			goto out;
+		}
+
 		if (parent) {
 			atomic_add_unless(&parent->power.child_count, -1, 0);
 			notify_parent = !parent->power.ignore_children;