[PATCH]: Autosleep patch.

[PATCH]: Autosleep patch.

[Christopher Sacchi]

This patch (listed below), changes an old function in autosleep.c in
kernel version 3.5. It's code is listed from the first -- line to the
2nd to last -- line.
This patch was checked with checkpatch.pl with no errors:
--
--- power/autosleep.c	2012-07-21 20:58:29.000000000 +0000
+++ power/autosleepnew.c	2012-08-06 10:52:26.000000000 +0000
@@ -30,7 +30,7 @@ static void try_to_suspend(struct work_s
 	if (!pm_get_wakeup_count(&initial_count, true))
 		goto out;

-	mutex_lock(&autosleep_lock);
+	mutex_lock_interruptable();

 	if (!pm_save_wakeup_count(initial_count)) {
 		mutex_unlock(&autosleep_lock);
Signed-off-by: Christopher P. Sacchi <chris.sacchi@gmail.com>
--

-- 
Christopher

Re: [PATCH]: Autosleep patch.

[Randy Dunlap]

On 08/06/2012 08:13 AM, Christopher Sacchi wrote:

> This patch (listed below), changes an old function in autosleep.c in
> kernel version 3.5. It's code is listed from the first -- line to the
> 2nd to last -- line.
> This patch was checked with checkpatch.pl with no errors:
> --
> --- power/autosleep.c	2012-07-21 20:58:29.000000000 +0000
> +++ power/autosleepnew.c	2012-08-06 10:52:26.000000000 +0000


The file names should include the full path from the top level of the
kernel source tree.  E.g.:

--- linux.orig/kernel/power/autosleep.c
+++ linux/kernel/power/autosleep.c

This patch uses the name autosleepnew.c, which is not what is
required/expected.

Your previous patches were for wireless (mac80211) components, so they
needed to go to the linux-wireless mailing list.  This patch is not
about wireless, so don't send it to that mailing list, although you
could send it to the linux-pm@vger.kernel.org mailing list.
Please see the MAINTAINERS file for where to send patches.


> @@ -30,7 +30,7 @@ static void try_to_suspend(struct work_s
>  	if (!pm_get_wakeup_count(&initial_count, true))
>  		goto out;
> 
> -	mutex_lock(&autosleep_lock);
> +	mutex_lock_interruptable();
> 
>  	if (!pm_save_wakeup_count(initial_count)) {
>  		mutex_unlock(&autosleep_lock);
> Signed-off-by: Christopher P. Sacchi <chris.sacchi@gmail.com>
> --

Signed-off-by: line is not part of the patch.  It should be near the
end of the patch description, before the actual patch.

Again, please read and follow Documentation/SubmittingPatches.

thanks.

-- 
~Randy

Re: [PATCH]: Autosleep patch.

[Guenter Roeck]

On Mon, Aug 06, 2012 at 08:38:17AM -0700, Randy Dunlap wrote:
> On 08/06/2012 08:13 AM, Christopher Sacchi wrote:
> 
> > This patch (listed below), changes an old function in autosleep.c in
> > kernel version 3.5. It's code is listed from the first -- line to the
> > 2nd to last -- line.
> > This patch was checked with checkpatch.pl with no errors:
> > --
> > --- power/autosleep.c	2012-07-21 20:58:29.000000000 +0000
> > +++ power/autosleepnew.c	2012-08-06 10:52:26.000000000 +0000
> 
> 
> The file names should include the full path from the top level of the
> kernel source tree.  E.g.:
> 
> --- linux.orig/kernel/power/autosleep.c
> +++ linux/kernel/power/autosleep.c
> 
> This patch uses the name autosleepnew.c, which is not what is
> required/expected.
> 
> Your previous patches were for wireless (mac80211) components, so they
> needed to go to the linux-wireless mailing list.  This patch is not
> about wireless, so don't send it to that mailing list, although you
> could send it to the linux-pm@vger.kernel.org mailing list.
> Please see the MAINTAINERS file for where to send patches.
> 
> 
> > @@ -30,7 +30,7 @@ static void try_to_suspend(struct work_s
> >  	if (!pm_get_wakeup_count(&initial_count, true))
> >  		goto out;
> > 
> > -	mutex_lock(&autosleep_lock);
> > +	mutex_lock_interruptable();
> > 

Plus, this code obviously never compiled. Nor does the submitter ever explain
what he actually tries to accomplish.

Guenter

> >  	if (!pm_save_wakeup_count(initial_count)) {
> >  		mutex_unlock(&autosleep_lock);
> > Signed-off-by: Christopher P. Sacchi <chris.sacchi@gmail.com>
> > --
> 
> Signed-off-by: line is not part of the patch.  It should be near the
> end of the patch description, before the actual patch.
> 
> Again, please read and follow Documentation/SubmittingPatches.
> 
> thanks.
> 
> -- 
> ~Randy
> 

Re: [PATCH]: Autosleep patch.

[Rafael J. Wysocki]

On Tuesday, August 07, 2012, Guenter Roeck wrote:
> On Mon, Aug 06, 2012 at 08:38:17AM -0700, Randy Dunlap wrote:
> > On 08/06/2012 08:13 AM, Christopher Sacchi wrote:
> > 
> > > This patch (listed below), changes an old function in autosleep.c in
> > > kernel version 3.5. It's code is listed from the first -- line to the
> > > 2nd to last -- line.
> > > This patch was checked with checkpatch.pl with no errors:
> > > --
> > > --- power/autosleep.c	2012-07-21 20:58:29.000000000 +0000
> > > +++ power/autosleepnew.c	2012-08-06 10:52:26.000000000 +0000
> > 
> > 
> > The file names should include the full path from the top level of the
> > kernel source tree.  E.g.:
> > 
> > --- linux.orig/kernel/power/autosleep.c
> > +++ linux/kernel/power/autosleep.c
> > 
> > This patch uses the name autosleepnew.c, which is not what is
> > required/expected.
> > 
> > Your previous patches were for wireless (mac80211) components, so they
> > needed to go to the linux-wireless mailing list.  This patch is not
> > about wireless, so don't send it to that mailing list, although you
> > could send it to the linux-pm@vger.kernel.org mailing list.
> > Please see the MAINTAINERS file for where to send patches.
> > 
> > 
> > > @@ -30,7 +30,7 @@ static void try_to_suspend(struct work_s
> > >  	if (!pm_get_wakeup_count(&initial_count, true))
> > >  		goto out;
> > > 
> > > -	mutex_lock(&autosleep_lock);
> > > +	mutex_lock_interruptable();
> > > 
> 
> Plus, this code obviously never compiled. Nor does the submitter ever explain
> what he actually tries to accomplish.

We've discussed this offlist and as far as I can see it there is no bug and
nothing needs to be fixed.

Thanks,
Rafael