On 6/7/05, Dmitry Torokhov <dtor_core@ameritech.net> wrote:
> On Monday 06 June 2005 02:28, Adam Morley wrote:
> > Hi Dimitry,
> >
> > On 6/5/05, Dmitry Torokhov <dtor_core@ameritech.net> wrote:
> > > On Sunday 05 June 2005 21:24, Adam Morley wrote:
> > >  > When I do a mem suspend (echo mem > /sys/power/state), either through
> > > > a lid switch ACPI action, or manually echo'ing the parameter, the
> > > > mouse doesn't work after un-suspending.  It seems like it is no longer
> > > > detected/initialized.  cat'ing the device file doesn't produce output,
> > > > and gpm and X don't get mouse inputs.
> > >
> > > Could you please try booting 2.6.12-rc5 with "i8042.debug" on the kernel
> > > command line; suspend, resume and post your dmesg?
> >
> > Sure.  Here it is.  Suspend was done using acpid using a lid action.
> > psmouse was modprobe -r'ed before suspend and modprobe'ed back in
> > after resume.
> >
> 
> We are trying to resume but KBC signals timeout condition every time we
> ping AUX port:
> 
> > drivers/input/serio/i8042.c: 60 -> i8042 (command) [220701]
> > drivers/input/serio/i8042.c: 47 -> i8042 (parameter) [220701]
> > drivers/input/serio/i8042.c: d4 -> i8042 (command) [220703]
> > drivers/input/serio/i8042.c: f2 -> i8042 (parameter) [220703]
> > drivers/input/serio/i8042.c: fe <- i8042 (interrupt, AUX, 12, timeout) [220725]
> > drivers/input/serio/i8042.c: d4 -> i8042 (command) [220726]
> > drivers/input/serio/i8042.c: ed -> i8042 (parameter) [220726]
> > drivers/input/serio/i8042.c: fe <- i8042 (interrupt, AUX, 12, timeout) [220747]
> > drivers/input/serio/i8042.c: 60 -> i8042 (command) [220748]
> > drivers/input/serio/i8042.c: 45 -> i8042 (parameter) [220748]
> > drivers/input/serio/i8042.c: 60 -> i8042 (command) [220943]
> > drivers/input/serio/i8042.c: 47 -> i8042 (parameter) [220943]
> > drivers/input/serio/i8042.c: d4 -> i8042 (command) [220943]
> > drivers/input/serio/i8042.c: f2 -> i8042 (parameter) [220943]
> > drivers/input/serio/i8042.c: fe <- i8042 (interrupt, AUX, 12, timeout) [220965]
> 
> Could you please try the patch below?

Ok, patch applied (against 2.6.12-rc5, clean, offset 2 lines for both
hunk 3 and 4).  Mouse still doesn't work on resume.  dmesg
w/i8042.debug set on kernel command line attached covering one
suspend/resume.

Thanks a bunch!  let me know if more data will help.

-- 
adam