linux-next: manual merge of the suspend tree with the input tree

linux-next: manual merge of the suspend tree with the input tree

[Stephen Rothwell]

Hi Rafael,

Today's linux-next merge of the suspend tree got a conflict in
drivers/input/serio/serio.c between commits
386d8772980be01b94bd463ea1e745732d7eb502 ("Input: serio - use device core
to create 'id' attribute group") and
ddf1ffbd40c92ff1e58c45fa96d309788f7beb60 ("Input: serio - let device core
tell us if device was registered") from the input tree and commit
f543d6795ccfbd443a042e012a6aea15e4f4c87a ("PM: Allow serio input devices
to suspend/resume asynchronously") from the suspend tree.

I fixed it up (see below) but am not sure it this is the correct fix.  I
can carry this fix as necessary.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

diff --cc drivers/input/serio/serio.c
index c3b626e,42970e4..0000000
--- a/drivers/input/serio/serio.c
+++ b/drivers/input/serio/serio.c
@@@ -547,15 -555,22 +547,17 @@@ static void serio_add_port(struct seri
  	}
  
  	list_add_tail(&serio->node, &serio_list);
 +
  	if (serio->start)
  		serio->start(serio);
 +
  	error = device_add(&serio->dev);
  	if (error)
 -		printk(KERN_ERR
 -			"serio: device_add() failed for %s (%s), error: %d\n",
 +		dev_err(&serio->dev,
 +			"device_add() failed for %s (%s), error: %d\n",
  			serio->phys, serio->name, error);
 -	else {
 -		serio->registered = true;
 -		error = sysfs_create_group(&serio->dev.kobj, &serio_id_attr_group);
 -		if (error)
 -			printk(KERN_ERR
 -				"serio: sysfs_create_group() failed for %s (%s), error: %d\n",
 -				serio->phys, serio->name, error);
++	else
+ 		device_enable_async_suspend(&serio->dev, true);
 -	}
  }
  
  /*

Re: linux-next: manual merge of the suspend tree with the input tree

[Dmitry Torokhov]

On Thu, Jan 07, 2010 at 03:07:23PM +1100, Stephen Rothwell wrote:
> Hi Rafael,
> 
> Today's linux-next merge of the suspend tree got a conflict in
> drivers/input/serio/serio.c between commits
> 386d8772980be01b94bd463ea1e745732d7eb502 ("Input: serio - use device core
> to create 'id' attribute group") and
> ddf1ffbd40c92ff1e58c45fa96d309788f7beb60 ("Input: serio - let device core
> tell us if device was registered") from the input tree and commit
> f543d6795ccfbd443a042e012a6aea15e4f4c87a ("PM: Allow serio input devices
> to suspend/resume asynchronously") from the suspend tree.
> 
> I fixed it up (see below) but am not sure it this is the correct fix.  I
> can carry this fix as necessary.

I think the fix is fine, however, do we still need to do async serio
suspend, now that we sped up resetting psmouse and atkbd? Rafael?

-- 
Dmitry

Re: linux-next: manual merge of the suspend tree with the input tree

[Rafael J. Wysocki]

On Thursday 07 January 2010, Dmitry Torokhov wrote:
> On Thu, Jan 07, 2010 at 03:07:23PM +1100, Stephen Rothwell wrote:
> > Hi Rafael,
> > 
> > Today's linux-next merge of the suspend tree got a conflict in
> > drivers/input/serio/serio.c between commits
> > 386d8772980be01b94bd463ea1e745732d7eb502 ("Input: serio - use device core
> > to create 'id' attribute group") and
> > ddf1ffbd40c92ff1e58c45fa96d309788f7beb60 ("Input: serio - let device core
> > tell us if device was registered") from the input tree and commit
> > f543d6795ccfbd443a042e012a6aea15e4f4c87a ("PM: Allow serio input devices
> > to suspend/resume asynchronously") from the suspend tree.
> > 
> > I fixed it up (see below) but am not sure it this is the correct fix.  I
> > can carry this fix as necessary.
> 
> I think the fix is fine, however, do we still need to do async serio
> suspend, now that we sped up resetting psmouse and atkbd? Rafael?

Well, serio suspend is about 30 ms total on my test boxes that may be
reduced to 15 ms by async.  Not much, but still.

Please let me know what you think.

Rafael