On Thu, Apr 19, 2012 at 09:43:45AM -0400, Alan Stern wrote:
> Below is a patch that will prevent any PCI device from being put into
> D3 during suspend.  (This is meant to be used without any of the
> earlier diagnostic patches.)  Obviously it's not the final solution; 
> the test in the "if" statement needs to be more discriminating.  :-)
> 
> But before going any farther, I'd like to test if USB wakeup works.  
> So after booting with this patch, make sure that the power/wakeup file
> says "enabled" in the sysfs path for each controller as well as the
> usb1, usb2, 1-2, and 2-2 paths under /sys/bus/usb/devices/.
> 
> Then suspend the computer without using the script, and while it is
> asleep try plugging in a new USB device.  That should cause the
> computer to wake up, if everything is working properly.
I've set "enabled" for following files:

/sys/bus/usb/devices/usb1/power/wakeup
/sys/bus/usb/devices/usb2/power/wakeup
/sys/bus/usb/devices/1-1/power/wakeup
/sys/bus/usb/devices/2-1/power/wakeup
/sys/bus/pci/devices/0000:00:1a.0/power/wakeup
/sys/bus/pci/devices/0000:00:1d.0/power/wakeup

yet it didn't wake up after plugging, though devices light up.


> Index: usb-3.4/drivers/pci/pci.c
> ===================================================================
> --- usb-3.4.orig/drivers/pci/pci.c
> +++ usb-3.4/drivers/pci/pci.c
> @@ -1713,6 +1713,8 @@ int pci_prepare_to_sleep(struct pci_dev
>  	pci_power_t target_state = pci_target_state(dev);
>  	int error;
>  
> +	if (1)
> +		target_state = PCI_D0;
>  	if (target_state == PCI_POWER_ERROR)
>  		return -EIO;
>  
> 
> 

-- 
WBR, wRAR