Black screen-dead unit on S3 resume - Intel Atom

Black screen-dead unit on S3 resume - Intel Atom

[Robert Wörle]

Dear List

I am approaching this list now as one of my last hopes i ve to fix S3 on 
Intel Atom based Tablet PC.

The overall description of the situation is as follows.

- entereing into S3 via "echo mem > /sys/power/state will result in a 
suspend into the bios which is comparable to the one when running windows.
The acitivity led starts blinking and it sits there. Power consumption 
measurments show the equally amounts of mA take.

The system then reacts on the power button only with a changing power 
led to static. ( after around 1 sec)
But that is all i can get as reaction.
No hdd activity at all.

So it stucks on the resume situation as early as acpi calls the waking 
vektor. That is my assumption.

Tests taken :
- All /sys/power/pm_test  `scenarios runs fine without any obvious issue.
- All test are done with mininal system scenario with only hdd, usb and 
framebuffer drivers compiled in.
- s3_beep kernel option also doesnt beep on resume , beepdriver is 
compiled in.
- testet various options, like acpi old_ordering, rsdt, sci_force_enable 
and some more
- most tests where done on 2.6.34-rc1.  but others where with 2.6.27 
(openSuse 11.1) and 2.6.20

I am about to recieve a ami bios post code - debug card. That one should 
help to determine the bios side.

So i am stuck without any further debug information beside the dmesg 
from the bootet system and the pm_test runs.


I can attach or mail any log you guys want me to. Also i ve all acpidump 
`ed tables

Any hints, help or suggstions are welcome at this point.

Regards Rob

Black screen-dead unit on S3 resume - Intel Atom

[Robert Wörle]

Dear List

I am approaching this list now as one of my last hopes i ve to fix S3 on
Intel Atom based Tablet PC.

The overall description of the situation is as follows.

- entereing into S3 via "echo mem > /sys/power/state will result in a
suspend into the bios which is comparable to the one when running windows.
The acitivity led starts blinking and it sits there. Power consumption
measurments show the equally amounts of mA take.

The system then reacts on the power button only with a changing power
led to static. ( after around 1 sec)
But that is all i can get as reaction.
No hdd activity at all.

So it stucks on the resume situation as early as acpi calls the waking
vektor. That is my assumption.

Tests taken :
- All /sys/power/pm_test  `scenarios runs fine without any obvious issue.
- All test are done with mininal system scenario with only hdd, usb and
framebuffer drivers compiled in.
- s3_beep kernel option also doesnt beep on resume , beepdriver is
compiled in.
- testet various options, like acpi old_ordering, rsdt, sci_force_enable
and some more
- most tests where done on 2.6.34-rc1.  but others where with 2.6.27
(openSuse 11.1) and 2.6.20

I am about to recieve a ami bios post code - debug card. That one should
help to determine the bios side.

So i am stuck without any further debug information beside the dmesg
from the bootet system and the pm_test runs.


I can attach or mail any log you guys want me to. Also i ve all acpidump
`ed tables

Any hints, help or suggstions are welcome at this point.

Regards Rob

Re: Black screen-dead unit on S3 resume - Intel Atom

[Zhang Rui]

On Mon, 2010-04-19 at 18:00 +0800, Robert Wörle wrote:
> Dear List
> 
> I am approaching this list now as one of my last hopes i ve to fix S3 on
> Intel Atom based Tablet PC.
> 
> The overall description of the situation is as follows.
> 
> - entereing into S3 via "echo mem > /sys/power/state will result in a
> suspend into the bios which is comparable to the one when running windows.
> The acitivity led starts blinking and it sits there. Power consumption
> measurments show the equally amounts of mA take.
> 
Sorry my English is not good.
Do you mean the power consumption is consistent in this state?
Or the power consumption in S3 is the same in both windows and Linux?
Or the power consumption is the same when Linux is in/out of S3?

> The system then reacts on the power button only with a changing power
> led to static. ( after around 1 sec)
> But that is all i can get as reaction.
> No hdd activity at all.
> 
> So it stucks on the resume situation as early as acpi calls the waking
> vektor. That is my assumption.
> 
> Tests taken :
> - All /sys/power/pm_test  `scenarios runs fine without any obvious issue.
> - All test are done with mininal system scenario with only hdd, usb and
> framebuffer drivers compiled in.
> - s3_beep kernel option also doesnt beep on resume , beepdriver is
> compiled in.
> - testet various options, like acpi old_ordering, rsdt, sci_force_enable
> and some more
> - most tests where done on 2.6.34-rc1.  but others where with 2.6.27
> (openSuse 11.1) and 2.6.20
> 
> I am about to recieve a ami bios post code - debug card. That one should
> help to determine the bios side.
> 
> So i am stuck without any further debug information beside the dmesg
> from the bootet system and the pm_test runs.
> 
> 
> I can attach or mail any log you guys want me to. Also i ve all acpidump
> `ed tables
> 
please attach the acpidump of the laptop.
please also attach the dmesg outputa after boot, in 2.6.34-rc1.

thanks,
rui


--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html