Re: [Linuxwacom-discuss] X60t: Suspend twice will kill the tablet

[Kristoffer <my...@tr...>]

Good evening,

On Thu, 2007-05-10 at 23:42 +0200, Magnus Vigerlöf wrote:
> [..]
> > BEGIN xf86WcmProc dev=0x825d0c8 priv=0x82478b0 type=stylus(stylus)
> > flags=1041 fd=-1 what=ON opening /dev/ttyS0
> > usbDetect
> > initializing ISDV4 tablet
> > initializing as ISDV4 model
> > getting ISDV4 Ranges
> > initializing ISDV4 tablet
> > initializing as ISDV4 model
> > getting ISDV4 Ranges
> > Wacom Query ISDV4 error magic error
> > Failed to open device (fd=39)
> > Closing device
> > xf86WcmProc ON FAILED
> 
> It appear to restart after the first 'getting ISDV4 Ranges' here and then it 
> looks like the driver get out-of-sync with the tablet and cannot proceed any 
> further.
> 
> > Does anybody know where to look further?
> 
> I'll have to pass the flag to someone with more knowledge of the serial 
> tablets...

Is it possible that this is actually not even a wacom-driver problem but
hidden somewhere deep in X and the kernel suspending? There's at least
nothing in `dmesg` or /var/log/messages.
I know for sure though, that the serial line has been reset (there was
at least an attempt) while coming back to life.

It's especially weird, because it works (at least) once. Just to make
sure, it works with 2.6.20+ too? 

Does X eventually not bind and release the serial well after the first
suspend? Because if it were a driver issue, the following should be
printed:

wcmISDV4.c Line 130 : 
ErrorF("Wacom xf86WcmWrite error : %s", strerror(errno));

shouldn't it?

It's getting late and I'll stop talking now. Wish you all a nice
weekend.

Yours,
Kristoffer