On Wed, Sep 01, 2010 at 07:49:52PM +0800, Ike Panhc wrote:
quoted text
> Your log tells a lot. It tells me if the bluetooth init failed. Then we
> have to turning off and power on again by hw switch to make it alive again.
> 
> So, I believe we have two issue here.
> * when rfkill unblock, sometimes bluetooth init failed.

Yes.

quoted text
> * power up system after rfkill block bluetooth, sometimes the cfgbit = 0xc0000
> Am I correct?

Not completely.

* reboot system after rfkill block bluetooth, sometimes the cfgbit = 0xc0000

I never had cfgbit = 0xc0000 when I power-cycled the system, only when I
rebooted it.
This is why I belive the second point has virtually the same reason as
the first point: bluetooth device initialization fails and thus its
detection by BIOS fails and thus ACPI returns 0xc0000 cfg.
Hence, I belive chances are high that solving the first point does also
solve the second.

While thinking about it... the init problems persisting a reboot could
also mean that not the unblock operation is the real issue but the block
operation - which probably leaves the bluetooth device in kind of a
semi-disabled state or something like that.


regards
   Mario
-- 
Do I contradict myself?
Very well, then I contradict myself, I am large, I contain multitudes.
                                                       -- Walt Whitman