This patch (as1402) fixes a bug in the PCI core. When a device is
about to be suspended, the core does a runtime resume because the
device settings may need to be reconfigured. But calling
pm_runtime_resume() is not the right way to do this, because it does
nothing to prevent the device from autosuspending immediately
afterward. The core needs to use pm_runtime_get_sync() and
pm_runtime_put_sync() instead.
Signed-off-by: Alan Stern <stern@xxxxxxxxxxxxxxxxxxx>
CC: Rafael J. Wysocki <rjw@xxxxxxx>
---
Index: usb-2.6/drivers/pci/pci-driver.c
===================================================================
--- usb-2.6.orig/drivers/pci/pci-driver.c
+++ usb-2.6/drivers/pci/pci-driver.c
@@ -623,10 +623,13 @@ static int pci_pm_prepare(struct device
* system from the sleep state, we'll have to prevent it from signaling
* wake-up.
*/
- pm_runtime_resume(dev);
+ pm_runtime_get_sync(dev);
- if (drv && drv->pm && drv->pm->prepare)
+ if (drv && drv->pm && drv->pm->prepare) {
error = drv->pm->prepare(dev);
+ if (error)
+ pm_runtime_put_sync(dev);
+ }
return error;
}
@@ -637,6 +640,7 @@ static void pci_pm_complete(struct devic
if (drv && drv->pm && drv->pm->complete)
drv->pm->complete(dev);
+ pm_runtime_put_sync(dev);
}
#else /* !CONFIG_PM_SLEEP */
--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
