[mythtv-users] Remote to turn *on* mythtv?

Leighton Brough brough at baremetalsoft.com
Tue Jul 17 18:27:37 UTC 2007


Craig Huff wrote:
> Problem:  I can't get my remote to wake up my system.  I want to be able to wake from the suspend-to-disk state, which I *think* is S4 (important that I have this right -- see below)
>
> Earlier in this thread Leighton reported that he was using the Microsoft MCE USB remote to wake his system from the suspend-to-RAM state.
>
> I have an Asus M2NPV-VM motherboard based system which I am still configuring and I have a (newly purchased) Microsoft MCE USB remote attached.
>
> I have all the mobo jumpers repositioned to supply standby power to the keyboard and the backpanel USB ports, where I have the MCE receiver attached.
>
> I *can* wake my system up with the keyboard I'm using while I set it up, but eventually it has to go ;-).
>
> I have turned on the BIOS setting for allowing/enabling PCI events to wake the system up.
>
> I have tried doing the following before suspending to disk:
> #  echo USB0 > /proc/acpi/wakeup
> #  echo USB2 > /proc/acpi/wakeup
> #  echo PS2K > /proc/acpi/wakeup
> and then suspending with (IIRC):
> #  echo disk > /sys/power/state
>
> Since I plan to use Suspend2 (now called TuxOnIce! I think) and already have in installed, I tried that, too, after the three echoes to /proc/acpi/wakeup with:
>   hibernate
>
> I've tried various keys on the remote and can see the receiver feedback LED respond, but the system stays off until I use the keyboard or front panel power switch.
>
> After performing the three echoes to /proc/acpi/wakeup, I get the following with this command:
> # cat /proc/acpi/wakeup | grep enabled
> PS2M    4    enabled
> PS2K    4    enabled
> USB0    4    enabled
> USB2    4    enabled
>
> I interpret this to mean the USB ports, keyboard and mouse are all enabled to wake the system from the S4 state.
>
> Have I left any stones unturned or is this mobo doomed to not wake from S4 via this IR receiver?
>
>   
The only other stones I can suggest to turn are:

The only button on the MCE remote which will wake the system is the 
power button. So make sure this is the one you are pressing. For me this 
is the only button which causes the receiver LED to flash when the 
system is suspended.

Some time ago, using an earlier kernel (sometime before 2.6.18 - I 
forget which) I found I had to manually set the power state of the USB 
receiver then unload the USB driver module to prevent this from turning 
it off when suspending. I forget the exact files involved, but I think 
it was something like /sys/.../power. I guess it's unlikely this is your 
problem, but if you're trying to exhaust all possibilities it might be 
worth having a rummage in /sys too. Or for that matter you could try 
another kernel, since it seems the support for ACPI and power management 
continues to improve over time. Of course your kernel needs to be built 
with ACPI support too, but I don't think you'd see /proc/acpi if it wasn't.

There's a kernel command line parameter called "acpi_os_name" which 
allows you to specify the name which the OS reports to the ACPI 
subsystem, which in turn may alter the behaviour of this. Try Googling 
for this, it might give you some ideas. Also, there are some ACPI 
support issues with Linux, due to the Microsoft versus Intel compiler 
being used to build the ACPI tables. Some folks have found they need to 
fix the ACPI tables to get features to work. Again, Googling for Linux 
and ACPI might give you some things to try.

When I was first experimenting with this I temporarily installed Windows 
2000 (or maybe it was XP) just to prove the HW worked OK. You can set 
the "allow this device to wake the system" option for the USB receiver, 
then hibernate and try the remote power button. This was (relatively) 
quick and easy to get going, and gave me the confidence to persevere 
with the required Linux config. That may be something else you can try.

The only other thing I can think of to suggest is to try both enabling 
and disabling the BIOS setting for PCI event wakeup (I don't have this 
setting in my BIOS). But you've probably already tried that.

For what it's worth, when I cat /proc/acpi/wakeup I also see state 4 for 
all devices, even though my system wakes from S3. I'm not sure the "4" 
means "S4" or the ACPI device power state the USB device will be put 
into on suspend (related to but different from the S1 - S5 system power 
states).

Leighton




More information about the mythtv-users mailing list