linux-kernel - [PATCH v1 09/10] PM: sleep: Propagate power.set

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <47028492.fMDQidcC6G@rjwysocki.net>
Date: Tue, 11 Feb 2025 22:21:23 +0100
From: "Rafael J. Wysocki" <rjw@...ysocki.net>
To: Linux PM <linux-pm@...r.kernel.org>
Cc: LKML <linux-kernel@...r.kernel.org>,
 Alan Stern <stern@...land.harvard.edu>, Ulf Hansson <ulf.hansson@...aro.org>,
 Johan Hovold <johan@...nel.org>,
 Manivannan Sadhasivam <manivannan.sadhasivam@...aro.org>,
 Jon Hunter <jonathanh@...dia.com>
Subject:
 [PATCH v1 09/10] PM: sleep: Propagate power.set_active in dependency chains

From: Rafael J. Wysocki <rafael.j.wysocki@...el.com>

After preparing pm_runtime_force_resume() for dealing with devices
having power.set_active set, make the core propagate that flag in
dependency chains, so that subordinate device are resumed along with
the ones that they depend on, but take exceptions into account.

Namely, do not set power.set_active for devices that have never had
runtime PM enabled, for parents that have power.ignore_children set and
for suppliers coming from device links with DL_FLAG_PM_RUNTIME unset.

Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@...el.com>
---
 drivers/base/power/main.c |   25 ++++++++++++++++---------
 1 file changed, 16 insertions(+), 9 deletions(-)

--- a/drivers/base/power/main.c
+++ b/drivers/base/power/main.c
@@ -1189,18 +1189,31 @@
 	return PMSG_ON;
 }

+static void dpm_cond_set_active(struct device *dev, bool cond)
+{
+	if (cond && !pm_runtime_no_support(dev))
+		dev->power.set_active = true;
+}
+
 static void dpm_superior_set_must_resume(struct device *dev)
 {
+	bool set_active = dev->power.set_active;
 	struct device_link *link;
 	int idx;

-	if (dev->parent)
+	if (dev->parent) {
 		dev->parent->power.must_resume = true;
+		dpm_cond_set_active(dev->parent, set_active &&
+				    !dev->parent->power.ignore_children);
+	}

 	idx = device_links_read_lock();

-	list_for_each_entry_rcu_locked(link, &dev->links.suppliers, c_node)
+	list_for_each_entry_rcu_locked(link, &dev->links.suppliers, c_node) {
 		link->supplier->power.must_resume = true;
+		dpm_cond_set_active(link->supplier, set_active &&
+				    (link->flags & DL_FLAG_PM_RUNTIME));
+	}

 	device_links_read_unlock(idx);
 }
@@ -1277,13 +1290,7 @@
 		dev->power.must_resume = true;

 	if (dev->power.must_resume) {
-		if (dev_pm_test_driver_flags(dev, DPM_FLAG_SMART_SUSPEND) &&
-		    !pm_runtime_no_support(dev)) {
-			dev->power.set_active = true;
-			if (dev->parent && !dev->parent->power.ignore_children &&
-			    !pm_runtime_no_support(dev->parent))
-				dev->parent->power.set_active = true;
-		}
+		dpm_cond_set_active(dev, dev_pm_test_driver_flags(dev, DPM_FLAG_SMART_SUSPEND));
 		dpm_superior_set_must_resume(dev);
 	}