On Mon, Jun 22, 2009 at 01:42:40PM +0200, Josip Rodin wrote:
> > > Then keep an eye on the logs, you want to check both the ones since
> > > boot for the session where you notice WWAN is disabled, as well as the
> > > ones since the boot _previous_ to that one, because something might
> > > have caused thinkpad-acpi to tell the firmware to store in NVRAM that
> > > WWAN should be disabled on the next power up...
> >
> > Gotcha.
>
> Right now I'm thinking that this is either completely screwed up by the
> BIOS or there's a bug in the driver. Even though the BIOS says the WWAN is
> Enabled and the antenna is On, I *always* have it turned off after any
> kind of boot, both under Linux and Windows.
>
> Before hibernation I got it in this state (everything working):
>
> Jun 22 13:34:08 myhostname kernel: thinkpad_acpi: bluetooth_update_rfk: forced rfkill state to 1
> Jun 22 13:34:08 myhostname kernel: thinkpad_acpi: wan_update_rfk: forced rfkill state to 1
>
> Then I had the laptop hibernate, and after restore I had:
>
> Jun 22 13:36:12 myhostname kernel: thinkpad_acpi: bluetooth_update_rfk: forced rfkill state to 0
> Jun 22 13:36:12 myhostname kernel: thinkpad_acpi: wan_update_rfk: forced rfkill state to 0
OK, I think it's definitely something broken in the driver, because under
2.6.31 and 2.6.32 it consistently screws things up through hibernation.
If I just reboot the machine normally, I get the devices back in state 1,
both in Linux and in Windows. Only when I hibernate in Linux does the driver
seem to write the "let's-completely-disable-it" state into the hardware --
even if it was explicitly *not* disabled in the Linux session that initiated
the hibernation.
The /sys/devices/platform/thinkpad_acpi/rfkill/rfkill?/persistent states
are on (1) all the time, but it doesn't actually do what appears to be
logical - keep the switch state that it had in the last session.
--
2. That which causes joy or happiness.
------------------------------------------------------------------------------
Return on Information:
Google Enterprise Search pays you back
Get the facts.
http://p.sf.net/sfu/google-dev2dev
_______________________________________________
ibm-acpi-devel mailing list
ibm-acpi-devel@xxxxxxxxxxxxxxxxxxxxx
https://lists.sourceforge.net/lists/listinfo/ibm-acpi-devel
