enable_irq_wake() problems

[Simon Kagstrom]

Hi!

As I said in an earlier mail, I'm working on accelerometer wakeup from
suspend. I have some problems with enable_irq_wake(), however. Even with the
device powered down and no interrupt generated as far as I can see, it wakes
up directly after suspend. Part of the log can be found below:

   [   63.640000] lis302dl spi0.1: Found lis302-2 (bottom)
   [   63.640000] Interrupt 1
   [   77.065000] maboo: 0x00000000
   [   77.065000] wake enabled for irq 16
   [   83.405000] Syncing filesystems ... done.
   [   83.410000] PM: Preparing system for mem sleep
   ...
   [   83.720000] neo1973-pm-gsm neo1973-pm-gsm.0: LATE suspend: 0x00000000
   [   83.720000] neo1973-button neo1973-button.0: LATE suspend: 0x00000000
   [   83.720000] spi_s3c24xx_gpio spi_s3c24xx_gpio.1: LATE suspend: 0x00000000
   [   83.720000] glamo3362 glamo3362.0: LATE suspend: 0x00000000
   [   83.720000] suspending dma channel 0
   [   83.720000] suspending dma channel 1
   [   83.720000] suspending dma channel 2
   [   83.720000] suspending dma channel 3
   [   83.720000] GSTATUS3 0x303e13f0
   [   83.720000] GSTATUS4 0x00000000
   [   83.720000] glamo3362 glamo3362.0: EARLY resume
   [   83.720000] spi_s3c24xx_gpio spi_s3c24xx_gpio.1: EARLY resume
   [   83.720000] neo1973-button neo1973-button.0: EARLY resume
   [   83.720000] neo1973-pm-gsm neo1973-pm-gsm.0: EARLY resume

The "0x00000000" is printouts of GSTATUS4 which I added - and as you can see
this is 0 all the way. So I think the problem is something else. Any clues?
Apparently it works well for the modem.

I'm also not sure how the resume process really works, does the board wake up
in U-boot first?

// Simon