acpi S3 never wakes up

acpi S3 never wakes up

[Hamie]

Hi.

I have an IBM r50p that I'd really really really like to have ACPI usable on. The battery works fine, the temp shows fine, the processor speeds up & slows down. All great, until I try to suspend (to RAM or to disk).

Firstly to RAM.

I hit Fn=F4 (Or cat 3 > /proc/acpi/sleep). System says I'm suspending. & goes to sleep. Fine so far. Unlike APM, the Fn key doesn't wake it up any more, so I press the power button. It LOOKS like it's starting up again, fan starts, CD spins, but no display... Doesn't matter whether I'm running X, or from the console. After wakup, nothing seems to work. Fn-F4 will send me to sleep again though... CTRL-ALT-Fx, does nothing...

Now should I get the lock icon again (I lock my machine with a poweron pass BIOS + Disk, NOT supervisor) like on APM? Or not? SHould I come up directly back where I was before? Or something else? Anyone know? I've seen reference to people having success, but I've had nothing but bad luck with ACPi & suspending so far.

Anyone got any ideas? I really really really hate having to boot my laptop 3x a day...

Oh. I've tried all kinds of kernels. Current is 2.6.6 and 2.6.7, both do exactly the same thing.

TIA

Hamish., 

Re: acpi S3 never wakes up

[Hamie]

Hamie wrote:

>Hi.
>
>I have an IBM r50p that I'd really really really like to have ACPI usable on. The battery works fine, the temp shows fine, the processor speeds up & slows down. All great, until I try to suspend (to RAM or to disk).
>
>Firstly to RAM.
>
>I hit Fn=F4 (Or cat 3 > /proc/acpi/sleep). System says I'm suspending. & goes to sleep. Fine so far. Unlike APM, the Fn key doesn't wake it up any more, so I press the power button. It LOOKS like it's starting up again, fan starts, CD spins, but no display... Doesn't matter whether I'm running X, or from the console. After wakup, nothing seems to work. Fn-F4 will send me to sleep again though... CTRL-ALT-Fx, does nothing...
>
>Now should I get the lock icon again (I lock my machine with a poweron pass BIOS + Disk, NOT supervisor) like on APM? Or not? SHould I come up directly back where I was before? Or something else? Anyone know? I've seen reference to people having success, but I've had nothing but bad luck with ACPi & suspending so far.
>
>Anyone got any ideas? I really really really hate having to boot my laptop 3x a day...
>
>Oh. I've tried all kinds of kernels. Current is 2.6.6 and 2.6.7, both do exactly the same thing.
>
>  
>

Hi all.

Some logs from kern.log... It's definately coming back from suspend... 
Just no video... It even suspends a second time... And wakes... No video 
still..

Jun 18 15:59:05 ballbreaker kernel: uhci_hcd 0000:00:1d.0: remove, state 1
Jun 18 15:59:05 ballbreaker kernel: usb usb2: USB disconnect, address 1
Jun 18 15:59:05 ballbreaker kernel: uhci_hcd 0000:00:1d.0: USB bus 2 
deregistered
Jun 18 15:59:05 ballbreaker kernel: uhci_hcd 0000:00:1d.1: remove, state 1
Jun 18 15:59:05 ballbreaker kernel: usb usb3: USB disconnect, address 1
Jun 18 15:59:05 ballbreaker kernel: uhci_hcd 0000:00:1d.1: USB bus 3 
deregistered
Jun 18 15:59:05 ballbreaker kernel: uhci_hcd 0000:00:1d.2: remove, state 1
Jun 18 15:59:05 ballbreaker kernel: usb usb4: USB disconnect, address 1
Jun 18 15:59:06 ballbreaker kernel: uhci_hcd 0000:00:1d.2: USB bus 4 
deregistered
Jun 18 15:59:09 ballbreaker kernel: ehci_hcd 0000:00:1d.7: remove, state 1
Jun 18 15:59:09 ballbreaker kernel: usb usb1: USB disconnect, address 1
Jun 18 15:59:09 ballbreaker kernel: ehci_hcd 0000:00:1d.7: USB bus 1 
deregistered
Jun 18 15:59:33 ballbreaker kernel: PM: Preparing system for suspend
Jun 18 15:59:33 ballbreaker kernel: Stopping tasks: 
================================================|
Jun 18 15:59:33 ballbreaker kernel: PM: Entering state.
Jun 18 15:59:33 ballbreaker kernel:  hwsleep-0304 [767] 
acpi_enter_sleep_state: Entering sleep state [S3]
Jun 18 15:59:33 ballbreaker kernel: Back to C!
Jun 18 15:59:33 ballbreaker kernel: PM: Finishing up.
Jun 18 15:59:33 ballbreaker kernel: PCI: Setting latency timer of device 
0000:00:1d.0 to 64
Jun 18 15:59:33 ballbreaker kernel: PCI: Setting latency timer of device 
0000:00:1d.1 to 64
Jun 18 15:59:33 ballbreaker kernel: PCI: Setting latency timer of device 
0000:00:1d.2 to 64
Jun 18 15:59:33 ballbreaker kernel: PCI: Setting latency timer of device 
0000:00:1d.7 to 64
Jun 18 15:59:33 ballbreaker kernel: PCI: Setting latency timer of device 
0000:00:1f.5 to 64
Jun 18 15:59:33 ballbreaker kernel: Restarting tasks... done
Jun 18 16:00:07 ballbreaker kernel: PM: Preparing system for suspend
Jun 18 16:00:29 ballbreaker kernel: Stopping tasks: 
=================================================|
Jun 18 16:00:29 ballbreaker kernel: PM: Entering state.
Jun 18 16:00:29 ballbreaker kernel:  hwsleep-0304 [921] 
acpi_enter_sleep_state: Entering sleep state [S3]
Jun 18 16:00:29 ballbreaker kernel: Back to C!
Jun 18 16:00:29 ballbreaker kernel: PM: Finishing up.
Jun 18 16:00:29 ballbreaker kernel: PCI: Setting latency timer of device 
0000:00:1d.0 to 64
Jun 18 16:00:29 ballbreaker kernel: PCI: Setting latency timer of device 
0000:00:1d.1 to 64
Jun 18 16:00:29 ballbreaker kernel: PCI: Setting latency timer of device 
0000:00:1d.2 to 64
Jun 18 16:00:29 ballbreaker kernel: PCI: Setting latency timer of device 
0000:00:1d.7 to 64
Jun 18 16:00:29 ballbreaker kernel: PCI: Setting latency timer of device 
0000:00:1f.5 to 64
Jun 18 16:00:29 ballbreaker kernel: Restarting tasks... done

(Then it's power switch time).

Can anyone give me a hint as to where the kernel should be 
re-initialising the video?

>TIA
>
>Hamish., 
>-
>To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
>the body of a message to majordomo@vger.kernel.org
>More majordomo info at  http://vger.kernel.org/majordomo-info.html
>Please read the FAQ at  http://www.tux.org/lkml/
>  
>

Re: acpi S3 never wakes up

[Roger Luethi]

On Fri, 18 Jun 2004 21:24:48 +0100, Hamie wrote:
> Can anyone give me a hint as to where the kernel should be 
> re-initialising the video?

Have you checked Documentation/power/video.txt? It's not terribly
encouraging, but it may give you some answers.

Roger

Re: acpi S3 never wakes up

[Hamie]

Roger Luethi wrote:

>On Fri, 18 Jun 2004 21:24:48 +0100, Hamie wrote:
>  
>
>>Can anyone give me a hint as to where the kernel should be 
>>re-initialising the video?
>>    
>>
>
>Have you checked Documentation/power/video.txt? It's not terribly
>encouraging, but it may give you some answers.
>
>  
>

Thanks for that. I missed it. Apologies to the list for not looking more 
before I threw the toys. Also thanks to michal@harddata.com who gave the 
one that works for the Thinkpad.

In case anyone else is looking, the r50p will resume from S3 when

acpi_sleep=s3_bios

is specified as a boot option.

Shame about the ATI drivers not restoring, but XOrg's X server works 
like a charm (Apart from no hardware 3D acceleration for the FireGL-T2). 
Still, I suppose I don't have a tainted kernel any more :)

H