[ltp] IrDa and double boot WinXP/Linux problem

Harry Popov linux-thinkpad@linux-thinkpad.org
Fri, 30 Sep 2005 11:38:05 +0300


On Thursday 29 September 2005 03:44, Richard Neill wrote:
> Harry Popov wrote:
> > Hi all,
> >
> > I have T600X with double boot Win XP and SuSe 9.1 Personal.
> > Every time after booting XP my infrared is not working under
> > Linux. I have to enter bios and initialize hardware to be able
> > to enable the infrared port. Every thing works fine until I
> > boot again Win XP. After that no IrDa under Linux again, the
> > device is disabled:
>
> A guess: are you powering off the machine, or just
> resetting/rebooting it? If the latter, the hardware will remain
> initialised between OSes, and this may be the cause of your
> problem. Try powering the machine off for 5 seconds after you
> exit windows - this will cause the registers in the IR chipset to
> reset, and you may have better luck.
>
> If that fails, doesn't the tpctl package have a command-line
> method to configure hardware? You might be able to hack something
> into rc.local[*] tpctl was written for the 600 series, so the
> author may have some more info on this problem.
>
> However, thes are just guesses - hopefully, you will get a
> definitive reply from someone more knowledgeable about the 600s.
>
> Richard
>
>
>
> [*]Eg, in pseudocode:
>    Stop IrDa service.  (which probably failed to start anyway)
>    Remove IR modules
>    tpctl --sort-out-the-mess --use-magic-incantation
>    modprobe the IR modules
>    Start the IR service.

Thanks for help. 

Here is what Chris Schumann <cschumann@twp-llc.com> suggested. I tryed and it 
works:
>
IIRC, the 600X has a Quick Boot option in the BIOS. Turn that OFF, so the
BIOS will assign hardware properties to devices before the OS does. XP
will reassign the IO addresses and IRQs as it sees fit if that option is
on, and they could well be different than what Linux saw after a freshboot.

You may have to deal with XP finding "new" hardware because the devices
are in different memory locations, but that should only happen once.
>

Regards, Harry Popov