[PATCH] pm: clear -EAGAIN in dpm_prepare

[PATCH] pm: clear -EAGAIN in dpm_prepare

[Sebastian Ott]

pm: clear -EAGAIN in dpm_prepare

When the last device in the dpm list is unregistered directly after its
prepare() callback returned with -EAGAIN, the return code is passed to
the calling function, resulting in a resume failure. Prevent this by
clearing the return code after -EAGAIN.

Signed-off-by: Sebastian Ott <sebott@linux.vnet.ibm.com>
---
 drivers/base/power/main.c |    1 +
 1 file changed, 1 insertion(+)

Index: git.linux-2.6/drivers/base/power/main.c
===================================================================
--- git.linux-2.6.orig/drivers/base/power/main.c	2009-06-30 12:27:33.000000000 +0200
+++ git.linux-2.6/drivers/base/power/main.c	2009-06-30 12:38:26.000000000 +0200
@@ -762,6 +762,7 @@ static int dpm_prepare(pm_message_t stat
 			dev->power.status = DPM_ON;
 			if (error == -EAGAIN) {
 				put_device(dev);
+				error = 0;
 				continue;
 			}
 			printk(KERN_ERR "PM: Failed to prepare device %s "

Re: [PATCH] pm: clear -EAGAIN in dpm_prepare

[Rafael J. Wysocki]

On Tuesday 30 June 2009, Sebastian Ott wrote:
> pm: clear -EAGAIN in dpm_prepare
> 
> When the last device in the dpm list is unregistered directly after its
> prepare() callback returned with -EAGAIN, the return code is passed to
> the calling function, resulting in a resume failure. Prevent this by
> clearing the return code after -EAGAIN.
> 
> Signed-off-by: Sebastian Ott <sebott@linux.vnet.ibm.com>
> ---

Thanks, the patch is now in the linux-next branch of the suspend-2.6 tree.

Best,
Rafael