Hi!
Here are some small cleanups, please apply,
Pavel
Index: linux/include/linux/serio.h
===================================================================
--- linux.orig/include/linux/serio.h 2003-06-24 11:54:39.000000000 +0200
+++ linux/include/linux/serio.h 2003-05-06 00:51:25.000000000 +0200
@@ -18,10 +18,7 @@
#include <linux/list.h>
-struct serio;
-
struct serio {
-
void *private;
void *driver;
char *name;
@@ -46,7 +43,6 @@
};
struct serio_dev {
-
void *private;
char *name;
Index: linux/drivers/input/power.c
===================================================================
--- linux.orig/drivers/input/power.c 2003-06-24 11:54:39.000000000 +0200
+++ linux/drivers/input/power.c 2003-04-18 16:19:02.000000000 +0200
@@ -45,9 +45,7 @@
static int suspend_button_pushed = 0;
static void suspend_button_task_handler(void *data)
{
- //extern void pm_do_suspend(void);
udelay(200); /* debounce */
- //pm_do_suspend();
suspend_button_pushed = 0;
}
@@ -67,8 +65,6 @@
case KEY_SUSPEND:
printk("Powering down entire device\n");
- //pm_send_all(PM_SUSPEND, dev);
-
if (!suspend_button_pushed) {
suspend_button_pushed = 1;
schedule_work(&suspend_button_task);
--
When do you have a heart between your knees?
[Johanka's followup: and *two* hearts?]
This needs to be migrated to the new power management code.
> ===================================================================
> --- linux.orig/drivers/input/power.c 2003-06-24 11:54:39.000000000 +0200
> +++ linux/drivers/input/power.c 2003-04-18 16:19:02.000000000 +0200
> @@ -45,9 +45,7 @@
> static int suspend_button_pushed = 0;
> static void suspend_button_task_handler(void *data)
> {
> - //extern void pm_do_suspend(void);
> udelay(200); /* debounce */
> - //pm_do_suspend();
> suspend_button_pushed = 0;
> }
>
> @@ -67,8 +65,6 @@
> case KEY_SUSPEND:
> printk("Powering down entire device\n");
>
> - //pm_send_all(PM_SUSPEND, dev);
> -
> if (!suspend_button_pushed) {
> suspend_button_pushed = 1;
> schedule_work(&suspend_button_task);
>
>
Hi!
> This needs to be migrated to the new power management code.
What exactly should it do? Suspend machine? Then it needs to do
equivalent of "echo 3 > /proc/acpi/sleep", but I do not think we have
generic interface for that...
Perhaps some comment should be put there?
Pavel
> > ===================================================================
> > --- linux.orig/drivers/input/power.c 2003-06-24 11:54:39.000000000 +0200
> > +++ linux/drivers/input/power.c 2003-04-18 16:19:02.000000000 +0200
> > @@ -45,9 +45,7 @@
> > static int suspend_button_pushed = 0;
> > static void suspend_button_task_handler(void *data)
> > {
> > - //extern void pm_do_suspend(void);
> > udelay(200); /* debounce */
> > - //pm_do_suspend();
> > suspend_button_pushed = 0;
> > }
> >
> > @@ -67,8 +65,6 @@
> > case KEY_SUSPEND:
> > printk("Powering down entire device\n");
> >
> > - //pm_send_all(PM_SUSPEND, dev);
> > -
> > if (!suspend_button_pushed) {
> > suspend_button_pushed = 1;
> > schedule_work(&suspend_button_task);
> >
> >
>
--
When do you have a heart between your knees?
[Johanka's followup: and *two* hearts?]
On Wed, Jul 09, 2003 at 01:14:19AM +0200, Pavel Machek wrote:
> > This needs to be migrated to the new power management code.
>
> What exactly should it do? Suspend machine? Then it needs to do
> equivalent of "echo 3 > /proc/acpi/sleep", but I do not think we have
> generic interface for that...
It looks like it was intended to call an old version of the suspend
code on ARM devices - probably the power button on the iPAQ.
The correct function (in the ARM tree) is now called "suspend()" and
deals with suspending the devices and then whatever is needed to cause
the CPU to go into deep sleep - ie, the user visible "power off" state.
Absolutely nothing to do with swsusp I'm afraid. 8)
> > > ===================================================================
> > > --- linux.orig/drivers/input/power.c 2003-06-24 11:54:39.000000000 +0200
> > > +++ linux/drivers/input/power.c 2003-04-18 16:19:02.000000000 +0200
> > > @@ -45,9 +45,7 @@
> > > static int suspend_button_pushed = 0;
> > > static void suspend_button_task_handler(void *data)
> > > {
> > > - //extern void pm_do_suspend(void);
> > > udelay(200); /* debounce */
> > > - //pm_do_suspend();
> > > suspend_button_pushed = 0;
> > > }
--
Russell King ([email protected]) The developer of ARM Linux
http://www.arm.linux.org.uk/personal/aboutme.html
Hi!
> > > This needs to be migrated to the new power management code.
> >
> > What exactly should it do? Suspend machine? Then it needs to do
> > equivalent of "echo 3 > /proc/acpi/sleep", but I do not think we have
> > generic interface for that...
>
> It looks like it was intended to call an old version of the suspend
> code on ARM devices - probably the power button on the iPAQ.
>
> The correct function (in the ARM tree) is now called "suspend()" and
> deals with suspending the devices and then whatever is needed to cause
> the CPU to go into deep sleep - ie, the user visible "power off"
> state.
So perhaps we need to add machine_suspend_ram() and
machine_suspend_disk() to reboot.h? We *do* want to have some generic
interface...
> Absolutely nothing to do with swsusp I'm afraid. 8)
That's okay, I'm doing suspend-to-RAM, too.
Pavel
> > > > ===================================================================
> > > > --- linux.orig/drivers/input/power.c 2003-06-24 11:54:39.000000000 +0200
> > > > +++ linux/drivers/input/power.c 2003-04-18 16:19:02.000000000 +0200
> > > > @@ -45,9 +45,7 @@
> > > > static int suspend_button_pushed = 0;
> > > > static void suspend_button_task_handler(void *data)
> > > > {
> > > > - //extern void pm_do_suspend(void);
> > > > udelay(200); /* debounce */
> > > > - //pm_do_suspend();
> > > > suspend_button_pushed = 0;
> > > > }
>
--
When do you have a heart between your knees?
[Johanka's followup: and *two* hearts?]
> > > This needs to be migrated to the new power management code.
> >
> > What exactly should it do? Suspend machine? Then it needs to do
> > equivalent of "echo 3 > /proc/acpi/sleep", but I do not think we have
> > generic interface for that...
>
> It looks like it was intended to call an old version of the suspend
> code on ARM devices - probably the power button on the iPAQ.
Yeap. I wrote that code for use on a iPAQ. I was hoping for a generic
interface for this some time in 2.5.X.
> The correct function (in the ARM tree) is now called "suspend()" and
> deals with suspending the devices and then whatever is needed to cause
> the CPU to go into deep sleep - ie, the user visible "power off" state.
Needs updating :-)
On Wed, Jul 09, 2003 at 01:32:48AM +0200, Pavel Machek wrote:
> So perhaps we need to add machine_suspend_ram() and
> machine_suspend_disk() to reboot.h? We *do* want to have some generic
> interface...
Sounds good.
--
Russell King ([email protected]) The developer of ARM Linux
http://www.arm.linux.org.uk/personal/aboutme.html
> On Wed, Jul 09, 2003 at 01:32:48AM +0200, Pavel Machek wrote:
> > So perhaps we need to add machine_suspend_ram() and
> > machine_suspend_disk() to reboot.h? We *do* want to have some generic
> > interface...
>
> Sounds good.
This would be really nice :-) Very handy for PDAs.
On Wed, 9 Jul 2003, Russell King wrote:
> On Wed, Jul 09, 2003 at 01:32:48AM +0200, Pavel Machek wrote:
> > So perhaps we need to add machine_suspend_ram() and
> > machine_suspend_disk() to reboot.h? We *do* want to have some generic
> > interface...
>
> Sounds good.
There is a similar thing in development at
bk://ldm.bkbits.net/linux-2.5-power/
Things are still a bit rough, and there isn't much documentation, as I'm
busy trying to get it all work correctly before OLS.
FWIW, I will be talking about this at OLS (and not the posted topic). A
paper describing the work is here:
http://kernel.org/pub/linux/kernel/people/mochel/doc/papers/power-ols2003.tar.gz
I should have more to say in the next week or two..
-pat