Machines self-power-up with 2.6.9-rc3 (evo N620c, ASUS, ...)

Machines self-power-up with 2.6.9-rc3 (evo N620c, ASUS, ...)

[Pavel Machek]

Hi!

I'm seeing bad problem with N620c notebook (and have reports of more
machines behaving like this, for example ASUS L8400C.) If I shutdown
machine with lid closed, opening lid will power the machine up. Ouch.
2.6.7 behaves okay.

Ouch, acpi=off makes it even worse [2.6.9-rc3, N620c]. I get some very
strange show on the leds (battery charge led blinks fast?!), then
machine powers up itself. This happens even with lid initially
open. 2.6.7 works as expected.

Any ideas?
								Pavel
-- 
People were complaining that M$ turns users into beta-testers...
...jr ghea gurz vagb qrirybcref, naq gurl frrz gb yvxr vg gung jnl!

Re: [ACPI] Machines self-power-up with 2.6.9-rc3 (evo N620c, ASUS, ...)

[Hiroshi 2 Itoh]

Pavel,

Thanks for your report. I have some suggestions to identify your
problem and help ACPI developers to work more effectively.

The developers are working both fixing ACPI related bugs and applying
work around for bad BIOS behaviors. At this time I think fixing
bugs is more important than applying work around.

So if your problem on some machine is so serious, could you please
give us the BIOS version and whether the machine suspend/resume is
OK in ACPI mode of Windows 2000/XP too? Of course DSDT table check
is highly helpful. http://bugzilla.kernel.org/ is prepared for such
purpose.

I believe such details help the developers to avoid spending much
time to rescue desperate BIOS and to keep tight release schedule.

- Hiro

acpi-devel-admin@lists.sourceforge.net wrote on 2004/10/21 04:15:31:

> Hi!
>
> I'm seeing bad problem with N620c notebook (and have reports of more
> machines behaving like this, for example ASUS L8400C.) If I shutdown
> machine with lid closed, opening lid will power the machine up. Ouch.
> 2.6.7 behaves okay.
>
> Ouch, acpi=off makes it even worse [2.6.9-rc3, N620c]. I get some very
> strange show on the leds (battery charge led blinks fast?!), then
> machine powers up itself. This happens even with lid initially
> open. 2.6.7 works as expected.
>
> Any ideas?
>                         Pavel

Re: [ACPI] Machines self-power-up with 2.6.9-rc3 (evo N620c, ASUS, ...)

[Nigel Cunningham]

Hi.

On Thu, 2004-10-21 at 05:15, Pavel Machek wrote:
> Hi!
> 
> I'm seeing bad problem with N620c notebook (and have reports of more
> machines behaving like this, for example ASUS L8400C.) If I shutdown
> machine with lid closed, opening lid will power the machine up. Ouch.
> 2.6.7 behaves okay.

:> Some people would love to have the machine power up when they open
the lid! Wish my XE3 would do that!

> Ouch, acpi=off makes it even worse [2.6.9-rc3, N620c]. I get some very
> strange show on the leds (battery charge led blinks fast?!), then
> machine powers up itself. This happens even with lid initially
> open. 2.6.7 works as expected.

Nigel
-- 
Nigel Cunningham
Pastoral Worker
Christian Reformed Church of Tuggeranong
PO Box 1004, Tuggeranong, ACT 2901

Many today claim to be tolerant. True tolerance, however, can cope with others
being intolerant.

Re: [ACPI] Machines self-power-up with 2.6.9-rc3 (evo N620c, ASUS, ...)

[Pavel Machek]

Hi!

> > I'm seeing bad problem with N620c notebook (and have reports of more
> > machines behaving like this, for example ASUS L8400C.) If I shutdown
> > machine with lid closed, opening lid will power the machine up. Ouch.
> > 2.6.7 behaves okay.
> 
> :> Some people would love to have the machine power up when they open
> the lid! Wish my XE3 would do that!

:-). Well for some other people it powers up when they unplug AC
power, and *that* is nasty. I'd like my machine to stay powered down
when I tell it so.
								Pavel
-- 
People were complaining that M$ turns users into beta-testers...
...jr ghea gurz vagb qrirybcref, naq gurl frrz gb yvxr vg gung jnl!

Re: [ACPI] Machines self-power-up with 2.6.9-rc3 (evo N620c, ASUS, ...)

[Pavel Machek]

Hi!

> Thanks for your report. I have some suggestions to identify your
> problem and help ACPI developers to work more effectively.
> 
> The developers are working both fixing ACPI related bugs and applying
> work around for bad BIOS behaviors. At this time I think fixing
> bugs is more important than applying work around.
> 
> So if your problem on some machine is so serious, could you please
> give us the BIOS version and whether the machine suspend/resume is
> OK in ACPI mode of Windows 2000/XP too? Of course DSDT table check
> is highly helpful. http://bugzilla.kernel.org/ is prepared for such
> purpose.

I was not talking about suspend/resume here, it wakes up even after
plain /sbin/shutdown -h now. I do not have Windows here to try, but it
works well in 2.6.7... 

								Pavel

> > I'm seeing bad problem with N620c notebook (and have reports of more
> > machines behaving like this, for example ASUS L8400C.) If I shutdown
> > machine with lid closed, opening lid will power the machine up. Ouch.
> > 2.6.7 behaves okay.
> >
> > Ouch, acpi=off makes it even worse [2.6.9-rc3, N620c]. I get some very
> > strange show on the leds (battery charge led blinks fast?!), then
> > machine powers up itself. This happens even with lid initially
> > open. 2.6.7 works as expected.
> >
> > Any ideas?


-- 
People were complaining that M$ turns users into beta-testers...
...jr ghea gurz vagb qrirybcref, naq gurl frrz gb yvxr vg gung jnl!

Re: [ACPI] Machines self-power-up with 2.6.9-rc3 (evo N620c, ASUS, ...)

[Nate Lawson]

Pavel Machek wrote:
>>>I'm seeing bad problem with N620c notebook (and have reports of more
>>>machines behaving like this, for example ASUS L8400C.) If I shutdown
>>>machine with lid closed, opening lid will power the machine up. Ouch.
>>>2.6.7 behaves okay.
>>
>>:> Some people would love to have the machine power up when they open
>>the lid! Wish my XE3 would do that!

This problem sounds like a wake GPE is enabled for the lid switch and 
that it has a _PRW that indicates it can wake the system from S5.  If 
this is the case, just disabled the GPE.

> :-). Well for some other people it powers up when they unplug AC
> power, and *that* is nasty. I'd like my machine to stay powered down
> when I tell it so.

This is likely a similar GPE problem.  The GPE for the EC fires even in 
S5.  I think the EC GPE should be disabled in the suspend method.

-- 
Nate

Re: [ACPI] Machines self-power-up with 2.6.9-rc3 (evo N620c, ASUS, ...)

[Li Shaohua]

On Thu, 2004-10-21 at 08:03, Nate Lawson wrote:
> Pavel Machek wrote:
> >>>I'm seeing bad problem with N620c notebook (and have reports of
> more
> >>>machines behaving like this, for example ASUS L8400C.) If I
> shutdown
> >>>machine with lid closed, opening lid will power the machine up.
> Ouch.
> >>>2.6.7 behaves okay.
> >>
> >>:> Some people would love to have the machine power up when they
> open
> >>the lid! Wish my XE3 would do that!
> 
> This problem sounds like a wake GPE is enabled for the lid switch and 
> that it has a _PRW that indicates it can wake the system from S5.  If 
> this is the case, just disabled the GPE.
> 
> > :-). Well for some other people it powers up when they unplug AC
> > power, and *that* is nasty. I'd like my machine to stay powered down
> > when I tell it so.
> 
> This is likely a similar GPE problem.  The GPE for the EC fires even
> in 
> S5.  I think the EC GPE should be disabled in the suspend method.
It could be the wakeup GPE issue, but must note Pavel's system suffer
the problem even with acpi=off. Could you please try boot your system
with acpi=off, and then reboot with acpi=off, what's the result? I
expected the wakeup GPE is disabled by the BIOS in this case.
Anyway, the DSDT can tell us the wakeup GPE info.

Thanks,
Shaohua

Re: [ACPI] Machines self-power-up with 2.6.9-rc3 (evo N620c, ASUS, ...)

[Pavel Machek]

Hi!

> >>>I'm seeing bad problem with N620c notebook (and have reports of more
> >>>machines behaving like this, for example ASUS L8400C.) If I shutdown
> >>>machine with lid closed, opening lid will power the machine up. Ouch.
> >>>2.6.7 behaves okay.
> >>
> >>:> Some people would love to have the machine power up when they open
> >>the lid! Wish my XE3 would do that!
> 
> This problem sounds like a wake GPE is enabled for the lid switch and 
> that it has a _PRW that indicates it can wake the system from S5.  If 
> this is the case, just disabled the GPE.

I tried disabling all events in /proc/acpi/wakeup, and that did not
change anything.

> >:-). Well for some other people it powers up when they unplug AC
> >power, and *that* is nasty. I'd like my machine to stay powered down
> >when I tell it so.
> 
> This is likely a similar GPE problem.  The GPE for the EC fires even in 
> S5.  I think the EC GPE should be disabled in the suspend method.

How did this hangling changed between 2.6.7 and 2.6.9?
								Pavel
-- 
People were complaining that M$ turns users into beta-testers...
...jr ghea gurz vagb qrirybcref, naq gurl frrz gb yvxr vg gung jnl!

Re: [ACPI] Machines self-power-up with 2.6.9-rc3 (evo N620c, ASUS, ...)

[Pavel Machek]

[-- Attachment #1: Type: text/plain, Size: 935 bytes --]

Hi!

> > > :-). Well for some other people it powers up when they unplug AC
> > > power, and *that* is nasty. I'd like my machine to stay powered down
> > > when I tell it so.
> > 
> > This is likely a similar GPE problem.  The GPE for the EC fires even
> > in 
> > S5.  I think the EC GPE should be disabled in the suspend method.
> It could be the wakeup GPE issue, but must note Pavel's system suffer
> the problem even with acpi=off. Could you please try boot your system
> with acpi=off, and then reboot with acpi=off, what's the result? I
> expected the wakeup GPE is disabled by the BIOS in this case.
> Anyway, the DSDT can tell us the wakeup GPE info.

You want me to boot with acpi=off twice and see if machine powers down
okay in second case?

Compressed DSDT is attached.
								Pavel
-- 
People were complaining that M$ turns users into beta-testers...
...jr ghea gurz vagb qrirybcref, naq gurl frrz gb yvxr vg gung jnl!

[-- Attachment #2: delme.dsdt.gz --]
[-- Type: application/octet-stream, Size: 12237 bytes --]

Re: [ACPI] Machines self-power-up with 2.6.9-rc3 (evo N620c, ASUS, ...)

[Li Shaohua]

On Thu, 2004-10-21 at 18:37, Pavel Machek wrote:
> Hi!
> 
> > > > :-). Well for some other people it powers up when they unplug AC
> > > > power, and *that* is nasty. I'd like my machine to stay powered down
> > > > when I tell it so.
> > > 
> > > This is likely a similar GPE problem.  The GPE for the EC fires even
> > > in 
> > > S5.  I think the EC GPE should be disabled in the suspend method.
> > It could be the wakeup GPE issue, but must note Pavel's system suffer
> > the problem even with acpi=off. Could you please try boot your system
> > with acpi=off, and then reboot with acpi=off, what's the result? I
> > expected the wakeup GPE is disabled by the BIOS in this case.
> > Anyway, the DSDT can tell us the wakeup GPE info.
> 
> You want me to boot with acpi=off twice and see if machine powers down
> okay in second case?
Yes, indeed.
> 
> Compressed DSDT is attached.
The dmesg shows your system can be waked up by Lid from S4 but not S5.
But there is another device (looks like the LPC bridge) can wakeup your
system from S5.

Thanks,
Shaohua

Re: [ACPI] Machines self-power-up with 2.6.9-rc3 (evo N620c, ASUS, ...)

[Pavel Machek]

Hi!

> > > > > :-). Well for some other people it powers up when they unplug AC
> > > > > power, and *that* is nasty. I'd like my machine to stay powered down
> > > > > when I tell it so.
> > > > 
> > > > This is likely a similar GPE problem.  The GPE for the EC fires even
> > > > in 
> > > > S5.  I think the EC GPE should be disabled in the suspend method.
> > > It could be the wakeup GPE issue, but must note Pavel's system suffer
> > > the problem even with acpi=off. Could you please try boot your system
> > > with acpi=off, and then reboot with acpi=off, what's the result? I
> > > expected the wakeup GPE is disabled by the BIOS in this case.
> > > Anyway, the DSDT can tell us the wakeup GPE info.
> > 
> > You want me to boot with acpi=off twice and see if machine powers down
> > okay in second case?
> Yes, indeed.

Okay, I did poweron, boot with acpi=off, reboot (again with acpi=off),
then shutdown -h now. I got the same strange lightshow at the leds and
reboot instead of powerdown.
								Pavel
-- 
People were complaining that M$ turns users into beta-testers...
...jr ghea gurz vagb qrirybcref, naq gurl frrz gb yvxr vg gung jnl!