Skip to contentSkip to conversation

pm diagnostics update, check for errors

PATCH In review David Brownell 3 participants 3 messages May 9, 2005

Subsystems (4)
Driver core Suspend to RAM Hibernation Power management core
lore
Share & tools
lore

David Brownell

author
#0 lore raw

PATCH 1/1 pm diagnostics update, check for errors

I've had this for a while, and converting to a GIT repository is incentive enough to send it out. Mostly it just makes existing diagnostics be more useful, but it also adds some error checking for cases that can happen on non-trivial systems ... or when userspace actions through /sys/.../power/state files break some system integrity rules.

If nobody objects, I'll probably ask to merge this at some point.

- Dave

This patch includes various tweaks in the messaging that appears during system pm state transitions:

* Warn about certain illegal calls in the device tree, like resuming child before parent or suspending parent before child. This could happen easily enough through sysfs, or in some cases when drivers use device_pm_set_parent().

* Be more consistent about dev_dbg() tracing ... do it for resume() and shutdown() too, and never if the driver doesn't have that method.

* Say which type of system sleep state is being entered.

Except for the warnings, these only affect debug messaging.

Signed-off-by David Brownell <[email protected]>
drivers/base/power/resume.c +10 -1
@@ -22,8 +22,17 @@
 
 int resume_device(struct device * dev)
 {
-	if (dev->bus && dev->bus->resume)
+	if (dev->power.pm_parent
+			&& dev->power.pm_parent->power.power_state) {
+		dev_err(dev, "PM: resume from %d, parent %s still %d\n",
+			dev->power.power_state,
+			dev->power.pm_parent->bus_id,
+			dev->power.pm_parent->power.power_state);
+	}
+	if (dev->bus && dev->bus->resume) {
+		dev_dbg(dev,"resuming\n");
 		return dev->bus->resume(dev);
+	}
 	return 0;
 }
drivers/base/power/shutdown.c +7 -6
@@ -25,8 +25,10 @@
 		return 0;
 
 	if (dev->detach_state == DEVICE_PM_OFF) {
-		if (dev->driver && dev->driver->shutdown)
+		if (dev->driver && dev->driver->shutdown) {
+			dev_dbg(dev, "shutdown\n");
 			dev->driver->shutdown(dev);
+		}
 		return 0;
 	}
 	return dpm_runtime_suspend(dev, dev->detach_state);

@@ -52,13 +54,12 @@
 	struct device * dev;
 
 	down_write(&devices_subsys.rwsem);
-	list_for_each_entry_reverse(dev, &devices_subsys.kset.list, kobj.entry) {
-		pr_debug("shutting down %s: ", dev->bus_id);
+	list_for_each_entry_reverse(dev, &devices_subsys.kset.list,
+				kobj.entry) {
 		if (dev->driver && dev->driver->shutdown) {
-			pr_debug("Ok\n");
+			dev_dbg(dev, "shutdown\n");
 			dev->driver->shutdown(dev);
-		} else
-			pr_debug("Ignored.\n");
+		}
 	}
 	up_write(&devices_subsys.rwsem);
drivers/base/power/suspend.c +15 -2
@@ -39,12 +39,25 @@
 {
 	int error = 0;
 
-	dev_dbg(dev, "suspending\n");
+	if (dev->power.power_state) {
+		dev_dbg(dev, "PM: suspend %d-->%d\n",
+			dev->power.power_state, state);
+	}
+	if (dev->power.pm_parent
+			&& dev->power.pm_parent->power.power_state) {
+		dev_err(dev,
+			"PM: suspend %d->%d, parent %s already %d\n",
+			dev->power.power_state, state,
+			dev->power.pm_parent->bus_id,
+			dev->power.pm_parent->power.power_state);
+	}
 
 	dev->power.prev_state = dev->power.power_state;
 
-	if (dev->bus && dev->bus->suspend && !dev->power.power_state)
+	if (dev->bus && dev->bus->suspend && !dev->power.power_state) {
+		dev_dbg(dev, "suspending\n");
 		error = dev->bus->suspend(dev, state);
+	}
 
 	return error;
 }
kernel/power/main.c +3 -3
@@ -156,14 +156,14 @@
 		goto Unlock;
 	}
 
-	pr_debug("PM: Preparing system for suspend\n");
+	pr_debug("PM: Preparing system for %s sleep\n", pm_states[state]);
 	if ((error = suspend_prepare(state)))
 		goto Unlock;
 
-	pr_debug("PM: Entering state.\n");
+	pr_debug("PM: Entering %s sleep\n", pm_states[state]);
 	error = suspend_enter(state);
 
-	pr_debug("PM: Finishing up.\n");
+	pr_debug("PM: Finishing wakeup.\n");
 	suspend_finish(state);
  Unlock:
 	up(&pm_sem);
#1 lore raw

Hi!

Quoted 8 lines

I've had this for a while, and converting to a GIT repository is incentive enough to send it out. Mostly it just makes existing diagnostics be more useful, but it also adds some error checking for cases that can happen on non-trivial systems ... or when userspace actions through /sys/.../power/state files break some system integrity rules.

If nobody objects, I'll probably ask to merge this at some point.

Patch is good, but merging it is going to be slightly tricky because I'll need to touch the same area just after
2.6.12... (pm_message_t-becomes-struct will conflict in trivial way).

I could create pm_message_t-becomes-struct patch early, and put it into good-git repository so that you could put your patch on the top of that. Would that work for you?

[Another alternative is make this go in *real fast* so that it gets in before 2.6.12 and I can place my patch on the top of that.]
Pavel

Signature · 1 line
Boycott Kodak -- for their patent abuse against Java.
#2 lore raw
In reply to David Brownell
Quoted 8 lines · David Brownell

I've had this for a while, and converting to a GIT repository is incentive enough to send it out. Mostly it just makes existing diagnostics be more useful, but it also adds some error checking for cases that can happen on non-trivial systems ... or when userspace actions through /sys/.../power/state files break some system integrity rules.

If nobody objects, I'll probably ask to merge this at some point.

No objections from me.

greg k-h