633aae2 "Input: i8042 - switch to using dev_pm_ops" removed handling
for PMSG_THAW, since we do not need to do anything during freeze and
thus it was thougt that thaw is not needed as well. However, there is
a period when interrupts are kept off, and if key happens to be pressed
during that time KBC becomes jammed. To avoid the jam we simply need
to poll KBC once during thaw.

Signed-off-by: Alan Jenkins <alan-jenkins@tuffmail.co.uk>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>

Signed-off-by: Elliott Sales de Andrade <quantum.analyst@gmail.com>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>

We used to make kseriod freezable to prevent unnecessary attempts at
resuming keyboard and mouse before taking hibernation image when suspend
and hibernation were sharing PM operations. Now that they are separated
and we don't risk resuming during 'thaw' we don't need to freeze kseriod
anymore. This will allow us to start resetting mouse and keyboard a bit
earlier, before rest of the userspace comes back up.

Acked-by: Rafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>

ps2if->irq is unsigned so the test does not work.

Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>

We can't use msleep() while holding a spinlock, moreower serio's write()
method is supposed to be useable from inettrupt context. Let's do what
i8042 does and poll the status register every 50 us (with udelay).

Reported-by: Marjan Fojkar <marjan@pajkc.eu>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>

Gigabyte netbook model M1022M requires i8042.noloop, otherwise AUX port
will not detected and the touchpad will not work. Unfortunately chassis
type in DMI set to "Other" and thus generic laptop entry does not fire
on it.

Reported-by: Darryl Bond <dbond@nrggos.com.au>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>

Some hardware (such as Dell laptops) signal a variety of events through
the i8042 controller, even if these don't map to keyboard events. Add
support for drivers to filter the i8042 event stream in order to respond
to these events and (if appropriate) block them from entering the input
stream.

Signed-off-by: Matthew Garrett <mjg@redhat.com>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>

We need to protect not only i8042 status and data register from concurrent
access from IRQ 1 and 12 but the rest of the shared state as well, so let's
move release of i8042_lock in i8042_interrupt() a little bit further down.

Signed-off-by: Dmitry Torokhov <dtor@mail.ru>

Also fix annotation of ps2_test() - it can'be __init since it is called
from __devinit code.

Signed-off-by: Dmitry Torokhov <dtor@mail.ru>