standby problem

standby problem

[giskard]

[-- Attachment #1: Type: text/plain, Size: 581 bytes --]

hello,

i wanted to know what's mean this error :)

mrspurr:/home/giskard# echo standby > /sys/power/state
acpi_pm_prepare does not support 1

means that my laptop doesn't supporto standby mode?

another question: i can suspend to ram my laptop without problems (echo
mem > /sys/power/state) but when i try to resume it i get a black screen
and then i need to reboot it. i need to do something else for resume
it? 

i have a:
satellite a80-111
linux 2.6.13 with suspend2.net patch
 
-- 
ciao giskard

non ho peli sulla lingua. e se ce li ho... non sono miei.

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: standby problem

[Pavel Machek]

On St 14-09-05 19:45:45, giskard wrote:
> hello,
> 
> i wanted to know what's mean this error :)
> 
> mrspurr:/home/giskard# echo standby > /sys/power/state
> acpi_pm_prepare does not support 1
> 
> means that my laptop doesn't supporto standby mode?

Probably yes.

> another question: i can suspend to ram my laptop without problems (echo
> mem > /sys/power/state) but when i try to resume it i get a black screen
> and then i need to reboot it. i need to do something else for resume
> it? 

See Doc*/power/video.txt.
								Pavel
-- 
if you have sharp zaurus hardware you don't need... you know my address


-------------------------------------------------------
SF.Net email is sponsored by:
Tame your development challenges with Apache's Geronimo App Server. 
Download it for free - -and be entered to win a 42" plasma tv or your very
own Sony(tm)PSP.  Click here to play: http://sourceforge.net/geronimo.php

RE: standby problem

[Li, Shaohua]

Hi Pavel,
>>
>> i wanted to know what's mean this error :)
>>
>> mrspurr:/home/giskard# echo standby > /sys/power/state
>> acpi_pm_prepare does not support 1
>>
>> means that my laptop doesn't supporto standby mode?
>
>Probably yes.
Is it possible we can just show available pm states in /sys/power/state?
I think we should report the error earlier before we do actual
suspend/resume. (maybe pm_ops should have a .supported_states callback
for all archs) It seems many users are complaining about this after Len
deprecated the proc interface.

Thanks,
Shaohua


-------------------------------------------------------
SF.Net email is sponsored by:
Tame your development challenges with Apache's Geronimo App Server.
Download it for free - -and be entered to win a 42" plasma tv or your very
own Sony(tm)PSP.  Click here to play: http://sourceforge.net/geronimo.php

Re: standby problem

[Pavel Machek]

Hi!

> >> i wanted to know what's mean this error :)
> >>
> >> mrspurr:/home/giskard# echo standby > /sys/power/state
> >> acpi_pm_prepare does not support 1
> >>
> >> means that my laptop doesn't supporto standby mode?
> >
> >Probably yes.
> Is it possible we can just show available pm states in /sys/power/state?
> I think we should report the error earlier before we do actual
> suspend/resume. (maybe pm_ops should have a .supported_states callback
> for all archs) It seems many users are complaining about this after Len
> deprecated the proc interface.

Pavel Troller even had a patch, but it was too ugly. Yes, we should
only list available states in /sys/power/state.
								Pavel

-- 
if you have sharp zaurus hardware you don't need... you know my address


-------------------------------------------------------
SF.Net email is sponsored by:
Tame your development challenges with Apache's Geronimo App Server. 
Download it for free - -and be entered to win a 42" plasma tv or your very
own Sony(tm)PSP.  Click here to play: http://sourceforge.net/geronimo.php

Re: standby problem

[Pavel Troller]

> Pavel Troller even had a patch, but it was too ugly. Yes, we should
> only list available states in /sys/power/state.
Hi!
  I'm still using it but I don't know how to get the list of supported states
otherwise than using my "ugly" ACPI-dependent method and without code
duplication. Could You, Pavel, drop me some tips, I would then make the real
patch and test it.
                      With regards, Pavel Troller


-------------------------------------------------------
SF.Net email is sponsored by:
Tame your development challenges with Apache's Geronimo App Server. 
Download it for free - -and be entered to win a 42" plasma tv or your very
own Sony(tm)PSP.  Click here to play: http://sourceforge.net/geronimo.php

RE: standby problem

[Li, Shaohua]

[-- Attachment #1: Type: text/plain, Size: 803 bytes --]

Hi,
>
>> >> i wanted to know what's mean this error :)
>> >>
>> >> mrspurr:/home/giskard# echo standby > /sys/power/state
>> >> acpi_pm_prepare does not support 1
>> >>
>> >> means that my laptop doesn't supporto standby mode?
>> >
>> >Probably yes.
>> Is it possible we can just show available pm states in
/sys/power/state?
>> I think we should report the error earlier before we do actual
>> suspend/resume. (maybe pm_ops should have a .supported_states
callback
>> for all archs) It seems many users are complaining about this after
Len
>> deprecated the proc interface.
>
>Pavel Troller even had a patch, but it was too ugly. Yes, we should
>only list available states in /sys/power/state.
Ok, how does this one look? Each arch provides a .valid callback.

Thanks,
Shaohua

[-- Attachment #2: pm_states_cleanup.patch --]
[-- Type: application/octet-stream, Size: 2158 bytes --]



---

 linux-2.6.14-rc1-root/drivers/acpi/sleep/main.c |    8 ++++++++
 linux-2.6.14-rc1-root/include/linux/pm.h        |    1 +
 linux-2.6.14-rc1-root/kernel/power/main.c       |    5 ++++-
 3 files changed, 13 insertions(+), 1 deletion(-)

diff -puN kernel/power/main.c~pm_states_cleanup kernel/power/main.c
--- linux-2.6.14-rc1/kernel/power/main.c~pm_states_cleanup	2005-09-19 16:32:53.000000000 +0800
+++ linux-2.6.14-rc1-root/kernel/power/main.c	2005-09-19 16:51:12.000000000 +0800
@@ -167,6 +167,8 @@ static int enter_state(suspend_state_t s
 {
 	int error;
 
+	if (pm_ops->valid && !pm_ops->valid(state))
+		return -ENODEV;
 	if (down_trylock(&pm_sem))
 		return -EBUSY;
 
@@ -236,7 +238,8 @@ static ssize_t state_show(struct subsyst
 	char * s = buf;
 
 	for (i = 0; i < PM_SUSPEND_MAX; i++) {
-		if (pm_states[i])
+		if (pm_states[i] && pm_ops && (!pm_ops->valid
+			||(pm_ops->valid && pm_ops->valid(i))))
 			s += sprintf(s,"%s ",pm_states[i]);
 	}
 	s += sprintf(s,"\n");
diff -puN include/linux/pm.h~pm_states_cleanup include/linux/pm.h
--- linux-2.6.14-rc1/include/linux/pm.h~pm_states_cleanup	2005-09-19 16:33:35.000000000 +0800
+++ linux-2.6.14-rc1-root/include/linux/pm.h	2005-09-19 16:50:38.000000000 +0800
@@ -170,6 +170,7 @@ typedef int __bitwise suspend_disk_metho
 
 struct pm_ops {
 	suspend_disk_method_t pm_disk_mode;
+	int (*valid)(suspend_state_t state);
 	int (*prepare)(suspend_state_t state);
 	int (*enter)(suspend_state_t state);
 	int (*finish)(suspend_state_t state);
diff -puN drivers/acpi/sleep/main.c~pm_states_cleanup drivers/acpi/sleep/main.c
--- linux-2.6.14-rc1/drivers/acpi/sleep/main.c~pm_states_cleanup	2005-09-19 16:42:26.000000000 +0800
+++ linux-2.6.14-rc1-root/drivers/acpi/sleep/main.c	2005-09-19 16:51:33.000000000 +0800
@@ -158,7 +158,15 @@ int acpi_suspend(u32 acpi_state)
 	return -EINVAL;
 }
 
+static int acpi_pm_state_valid(suspend_state_t pm_state)
+{
+	u32 acpi_state = acpi_suspend_states[pm_state];
+
+	return sleep_states[acpi_state];
+}
+
 static struct pm_ops acpi_pm_ops = {
+	.valid = acpi_pm_state_valid,
 	.prepare = acpi_pm_prepare,
 	.enter = acpi_pm_enter,
 	.finish = acpi_pm_finish,
_

Re: standby problem

[Pavel Machek]

On Po 19-09-05 17:00:25, Li, Shaohua wrote:
> Hi,
> >
> >> >> i wanted to know what's mean this error :)
> >> >>
> >> >> mrspurr:/home/giskard# echo standby > /sys/power/state
> >> >> acpi_pm_prepare does not support 1
> >> >>
> >> >> means that my laptop doesn't supporto standby mode?
> >> >
> >> >Probably yes.
> >> Is it possible we can just show available pm states in
> /sys/power/state?
> >> I think we should report the error earlier before we do actual
> >> suspend/resume. (maybe pm_ops should have a .supported_states
> callback
> >> for all archs) It seems many users are complaining about this after
> Len
> >> deprecated the proc interface.
> >
> >Pavel Troller even had a patch, but it was too ugly. Yes, we should
> >only list available states in /sys/power/state.
> Ok, how does this one look? Each arch provides a .valid callback.

Looks good to me.
								Pavel

-- 
if you have sharp zaurus hardware you don't need... you know my address


-------------------------------------------------------
SF.Net email is sponsored by:
Tame your development challenges with Apache's Geronimo App Server. 
Download it for free - -and be entered to win a 42" plasma tv or your very
own Sony(tm)PSP.  Click here to play: http://sourceforge.net/geronimo.php

Re: standby problem

[Randy.Dunlap]

On Mon, 19 Sep 2005, Pavel Troller wrote:

> > Pavel Troller even had a patch, but it was too ugly. Yes, we should
> > only list available states in /sys/power/state.
> Hi!
>   I'm still using it but I don't know how to get the list of supported states
> otherwise than using my "ugly" ACPI-dependent method and without code
> duplication. Could You, Pavel, drop me some tips, I would then make the real
> patch and test it.
>                       With regards, Pavel Troller

The suggestion of a .supported_states callback seemed very
reasonable to me.  I've been working on something similar
to that in my free time, so between it and your patch (which
I'll find or you could repost it for review), we should be
able to make some progress.

-- 
~Randy


-------------------------------------------------------
SF.Net email is sponsored by:
Tame your development challenges with Apache's Geronimo App Server. 
Download it for free - -and be entered to win a 42" plasma tv or your very
own Sony(tm)PSP.  Click here to play: http://sourceforge.net/geronimo.php

Re: standby problem

[Randy.Dunlap]

On Mon, 19 Sep 2005 17:00:25 +0800 Li, Shaohua wrote:

> Hi,
> >
> >> >> i wanted to know what's mean this error :)
> >> >>
> >> >> mrspurr:/home/giskard# echo standby > /sys/power/state
> >> >> acpi_pm_prepare does not support 1
> >> >>
> >> >> means that my laptop doesn't supporto standby mode?
> >> >
> >> >Probably yes.
> >> Is it possible we can just show available pm states in
> /sys/power/state?
> >> I think we should report the error earlier before we do actual
> >> suspend/resume. (maybe pm_ops should have a .supported_states
> callback
> >> for all archs) It seems many users are complaining about this after
> Len
> >> deprecated the proc interface.
> >
> >Pavel Troller even had a patch, but it was too ugly. Yes, we should
> >only list available states in /sys/power/state.
> Ok, how does this one look? Each arch provides a .valid callback.

With this patch, I see in /sys/power/state either:
standby disk		# if CONFIG_SOFTWARE_SUSPEND=y

or just:
standby			# if CONFIG_SOFTWARE_SUSPEND=n

Does this mean that my poor old BIOS cannot do STR or did I miss
a config option?  Yes, I have CONFIG_ACPI_SLEEP=y.
Is there another config option for STR?

I think it would be great if we could make the config options
something more user-friendly, like Suspend to Disk and
Standby/Suspend to RAM.  Or at least mention Suspend to RAM
in the Help text....  It would also make sense (from a user's
perspective) to have them together (next to each other).

---
~Randy
You can't do anything without having to do something else first.
-- Belefant's Law


-------------------------------------------------------
SF.Net email is sponsored by:
Tame your development challenges with Apache's Geronimo App Server. 
Download it for free - -and be entered to win a 42" plasma tv or your very
own Sony(tm)PSP.  Click here to play: http://sourceforge.net/geronimo.php

RE: standby problem

[Li, Shaohua]

>> >
>> >> >> i wanted to know what's mean this error :)
>> >> >>
>> >> >> mrspurr:/home/giskard# echo standby > /sys/power/state
>> >> >> acpi_pm_prepare does not support 1
>> >> >>
>> >> >> means that my laptop doesn't supporto standby mode?
>> >> >
>> >> >Probably yes.
>> >> Is it possible we can just show available pm states in
>> /sys/power/state?
>> >> I think we should report the error earlier before we do actual
>> >> suspend/resume. (maybe pm_ops should have a .supported_states
>> callback
>> >> for all archs) It seems many users are complaining about this
after
>> Len
>> >> deprecated the proc interface.
>> >
>> >Pavel Troller even had a patch, but it was too ugly. Yes, we should
>> >only list available states in /sys/power/state.
>> Ok, how does this one look? Each arch provides a .valid callback.
>
>With this patch, I see in /sys/power/state either:
>standby disk		# if CONFIG_SOFTWARE_SUSPEND=y
>
>or just:
>standby			# if CONFIG_SOFTWARE_SUSPEND=n
>
>Does this mean that my poor old BIOS cannot do STR or did I miss
>a config option?  Yes, I have CONFIG_ACPI_SLEEP=y.
>Is there another config option for STR?
It's interesting one system support standby but not support STR. Your
dmesg will tell us which states the system supports.

Thanks,
Shaohua


-------------------------------------------------------
SF.Net email is sponsored by:
Tame your development challenges with Apache's Geronimo App Server.
Download it for free - -and be entered to win a 42" plasma tv or your very
own Sony(tm)PSP.  Click here to play: http://sourceforge.net/geronimo.php

Re: standby problem

[Pavel Machek]

Hi!

> > >Pavel Troller even had a patch, but it was too ugly. Yes, we should
> > >only list available states in /sys/power/state.
> > Ok, how does this one look? Each arch provides a .valid callback.
> 
> With this patch, I see in /sys/power/state either:
> standby disk		# if CONFIG_SOFTWARE_SUSPEND=y
> 
> or just:
> standby			# if CONFIG_SOFTWARE_SUSPEND=n
> 
> Does this mean that my poor old BIOS cannot do STR or did I miss
> a config option?  Yes, I have CONFIG_ACPI_SLEEP=y.
> Is there another config option for STR?

I think its your poor old BIOS/hardware. 

> I think it would be great if we could make the config options
> something more user-friendly, like Suspend to Disk and
> Standby/Suspend to RAM.  Or at least mention Suspend to RAM
> in the Help text....  It would also make sense (from a user's
> perspective) to have them together (next to each other).

I guess patches are accepted.. but be warned: SWSUSP is generic
option, while ACPI_SLEEP is ... well, ACPI related.
								Pavel
-- 
if you have sharp zaurus hardware you don't need... you know my address


-------------------------------------------------------
SF.Net email is sponsored by:
Tame your development challenges with Apache's Geronimo App Server. 
Download it for free - -and be entered to win a 42" plasma tv or your very
own Sony(tm)PSP.  Click here to play: http://sourceforge.net/geronimo.php

Re: standby problem

[Randy.Dunlap]

On Tue, 20 Sep 2005 13:31:11 +0800 Li, Shaohua wrote:

> >> >
> >> >> >> i wanted to know what's mean this error :)
> >> >> >>
> >> >> >> mrspurr:/home/giskard# echo standby > /sys/power/state
> >> >> >> acpi_pm_prepare does not support 1
> >> >> >>
> >> >> >> means that my laptop doesn't supporto standby mode?
> >> >> >
> >> >> >Probably yes.
> >> >> Is it possible we can just show available pm states in
> >> /sys/power/state?
> >> >> I think we should report the error earlier before we do actual
> >> >> suspend/resume. (maybe pm_ops should have a .supported_states
> >> callback
> >> >> for all archs) It seems many users are complaining about this
> after
> >> Len
> >> >> deprecated the proc interface.
> >> >
> >> >Pavel Troller even had a patch, but it was too ugly. Yes, we should
> >> >only list available states in /sys/power/state.
> >> Ok, how does this one look? Each arch provides a .valid callback.
> >
> >With this patch, I see in /sys/power/state either:
> >standby disk		# if CONFIG_SOFTWARE_SUSPEND=y
> >
> >or just:
> >standby			# if CONFIG_SOFTWARE_SUSPEND=n
> >
> >Does this mean that my poor old BIOS cannot do STR or did I miss
> >a config option?  Yes, I have CONFIG_ACPI_SLEEP=y.
> >Is there another config option for STR?
> It's interesting one system support standby but not support STR. Your
> dmesg will tell us which states the system supports.

Yes, it does tell us :(

[42949386.700000] ACPI: (supports S0 S1 S4 S5)

Thanks,
---
~Randy
You can't do anything without having to do something else first.
-- Belefant's Law


-------------------------------------------------------
SF.Net email is sponsored by:
Tame your development challenges with Apache's Geronimo App Server. 
Download it for free - -and be entered to win a 42" plasma tv or your very
own Sony(tm)PSP.  Click here to play: http://sourceforge.net/geronimo.php