Re: Desktop and S3 suspend the Holy Grail of ACPI?

[Nick Lidakis <nlidakis-H+0wwilmMs3R7s880joybQ@public.gmane.org>]

Len Brown wrote:

>>>...D875PBZ...
>>>      
>>>
>
>On this box S1 works for me, but S3 doesn't resume (below)
>
>Note that you need to enable S3 in the BIOS SETUP to try S3.
>
>  
>
I do have S3 enabled in the BIOS. I just grabbed Debian's 
kernel-source-2.6.3 and compiled it without SMP support. cat 
/sys/power/state gives me standby mem disk.
I tried echo -n disk /sys/power/state (as per the ACPI docs on 
sourceforge) and the system did try to write the contents to disk, but 
failed. I then realized I don't have enough swap space, which I will try 
to exapnd later.

I then tried eco -n mem > /sys/power/state and the machine basically 
shuts down (completely off I think) so fast that I don't get a chance to 
see any console messages. Doesn't resume when I hit a key. If I try to 
hit the power button once, the machine beeps hi-lo (BIOS error I guess?) 
without booting. I have to then hold the power swithc for 4 seconds, 
then turn it on again.

Doing:
# /etc/init.d/acpid stop
echo -n standby > /sys/power/state
the machine does go into standy with no error messages.
I am able to resume by hitting the power button once.

Thank you for the  "acpid stop" tip, I never saw it mentioned in the howto.
The only problem is the case fans are still on even though I have the 
fans set for off at low temperature in the BIOS.
I was hoping for a silent sleeping machine; maybe I'll wait an hour and 
see if the case fans do eventually shut off.


>Note that neither S1 nor S3 will do anything on the SMP kernel.
>(and unfortunately, they will not even tell you that when you try)
>
>-Len
>
>BIOS SETUP: Power/ACPI
>ACPI Suspend State [S3 State] (default is S1)
>
>boot FC2-test1 UP kernel
>
># cat /proc/acpi/sleep
>S0 S1 S3 S4 S5
>
># /etc/init.d/acpid stop
>(to make power button resume event not initiate a shut-down)
>
># echo 1 >/proc/acpi/sleep
>Stopping tasks: ================================|
>psmouse.c: bad data from KBC - timeout bad parity
>hdc: start_power_step(step: 0)
>hdc: completing PM request, suspend
>hda: start_power_step(step: 0)
>hda: start_power_step(step: 1)
>hda: complete_power_step(step: 1, stat: 50, err: 0)
>hda: completing PM request, suspend
>psmouse.c: bad data from KBC - timeout bad parity
>psmouse.c: bad data from KBC - timeout bad parity
>PM: Entering state.
>i8042.c: Can't restore CTR.
>
>
><press power button here to resume>
>
>
>Back to C!
>i8042.c: Can't read CTR while initializing i8042.
>i8042: resume failed
>PM: Finishing up.
>PCI: Setting latency timer of device 0000:00:1d.0 to 64
>PCI: Setting latency timer of device 0000:00:1d.1 to 64
>PCI: Setting latency timer of device 0000:00:1d.2 to 64
>PCI: Setting latency timer of device 0000:00:1d.3 to 64
>hda: Wakeup request inited, waiting for !BSY...
>hda: start_power_step(step: 1000)
>e1000: eth0 NIC Link is Up 1000 Mbps Full Duplex
>blk: queue f7dcb5c0, I/O limit 4095Mb (mask 0xffffffff)
>hda: completing PM request, resume
>hdc: Wakeup request inited, waiting for !BSY...
>hdc: start_power_step(step: 1000)
>hdc: completing PM request, resume
>Restarting tasks... done
>#
>
>system running normally here.
>
>now try S3:
>
># echo 3 >/proc/acpi/sleep
>Stopping tasks: =================================|
>hdc: start_power_step(step: 0)
>hdc: completing PM request, suspend
>hda: start_power_step(step: 0)
>hda: start_power_step(step: 1)
>hda: complete_power_step(step: 1, stat: 50, err: 0)
>hda: completing PM request, suspend
>PM: Entering state.
>
><power is off, press power button here to wake up system>
>
>here a bunch of garbled junk comes out my serial console.
>But video is not restored to my matrox mystique FB.
>ping doesn't work.
>
>  
>

This is on the INTEL D875PBZ, is that correct? What kernel version is 
FC2-test1 UP kernel? I am also assuming debian's kernel 2.6.3 would have 
all the latest ACPI patches?

>Curiously CTRL-ALT-DEL results in the following on the serial console:
>
>quotaoff: numerical sysctl 5 16 8 is obsolete.
>md: stopping all md devices.
>md: md0 switched to read-only mode.
>Restarting system.
>
><and the system reboots here>
>
>So it seems that the system did resume and talk to the keyboard at
>least...
>
>
>
>
>  
>
Len, thank you for your detailed responses, much appreciated.


-------------------------------------------------------
SF.Net is sponsored by: Speed Start Your Linux Apps Now.
Build and deploy apps & Web services for Linux with
a free DVD software kit from IBM. Click Now!
http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click