Skip to contentSkip to conversation

driver core: Respect all error codes from dev_pm_domain_attach()

PATCH Guenter Roeck 2 participants 2 messages May 15, 2018

lore
Share & tools
lore
#0 lore raw
Quoted 4 lines · Ulf Hansson

The limitation of being able to check only for -EPROBE_DEFER from dev_pm_domain_attach() has been removed. Hence let's respect all error codes and bail out accordingly.

AFAICS this patch causes all drivers/devices to fail instantiating if dev_pm_domain_set() is called in the device initialization path. That seems to be a systemic problem, since dev_pm_domain_set() must only be called for unbound devices.

In practice, I see the problem when trying to boot beagle or overo with qemu (the Linaro version). Of course, that doesn't mean much because that is not real hardware. However, I am not surprised that all devices instantiated through, for example, omap_device_build_from_dt() fail to instantiate. Instrumentation confirms that dev_pm_domain_set() is called prior to platform_drv_probe().

Guenter

Quoted 39 lines
---
 drivers/base/platform.c | 17 ++++++++---------
 1 file changed, 8 insertions(+), 9 deletions(-)
drivers/base/platform.c +8 -9
index 8075ddc70a17..9460139d9b02 100644
@@ -572,17 +572,16 @@ static int platform_drv_probe(struct device *_dev)
 		return ret;
 
 	ret = dev_pm_domain_attach(_dev, true);
-	if (ret != -EPROBE_DEFER) {
-		if (drv->probe) {
-			ret = drv->probe(dev);
-			if (ret)
-				dev_pm_domain_detach(_dev, true);
-		} else {
-			/* don't fail if just dev_pm_domain_attach failed */
-			ret = 0;
-		}
+	if (ret)
+		goto out;
+
+	if (drv->probe) {
+		ret = drv->probe(dev);
+		if (ret)
+			dev_pm_domain_detach(_dev, true);
 	}
 
+out:
 	if (drv->prevent_deferred_probe && ret == -EPROBE_DEFER) {
 		dev_warn(_dev, "probe deferral not supported\n");
 		ret = -ENXIO;
Signature · 1 line
2.7.4
#1 lore raw
Quoted 11 lines · Guenter Roeck
Quoted 4 lines · Ulf Hansson

The limitation of being able to check only for -EPROBE_DEFER from dev_pm_domain_attach() has been removed. Hence let's respect all error codes and bail out accordingly.

AFAICS this patch causes all drivers/devices to fail instantiating if dev_pm_domain_set() is called in the device initialization path. That seems to be a systemic problem, since dev_pm_domain_set() must only be called for unbound devices.

Quoted 6 lines

In practice, I see the problem when trying to boot beagle or overo with qemu (the Linaro version). Of course, that doesn't mean much because that is not real hardware. However, I am not surprised that all devices instantiated through, for example, omap_device_build_from_dt() fail to instantiate. Instrumentation confirms that dev_pm_domain_set() is called prior to platform_drv_probe().

Your observations are correct, my bad! Others have also reported the problems.

This should solve it:
https://patchwork.kernel.org/patch/10398597/

[...]

Kind regards
Uffe