The root cause is we have a synaptics touchpad device that fails to be
identified as a genuine synaptics device, so the ps2 subsystem fails back
to the generic psmouse driver.
The psmouse driver does *not* issue a reset to device on reconnect, which
is ps2 slang for resume. Had it been probed correctly, it would have received
a reset on reconnect from the synaptics driver.
So we've succeeded in making the psmouse driver a little more robust but have
yet to resolve the root cause, which is to correctly identify and configure the touchpad
as a genuine synaptic device.
We're working on getting the datasheet for this part.
This fix has finally been distilled and has been proposed for Linus to
pull:
http:// git.kernel. org/?p= linux/kernel/ git/dtor/ input.git; a=commit; h=ef110b24e28f3 6620f63dab94708 a17c7e267358
The root cause is we have a synaptics touchpad device that fails to be
identified as a genuine synaptics device, so the ps2 subsystem fails back
to the generic psmouse driver.
The psmouse driver does *not* issue a reset to device on reconnect, which
is ps2 slang for resume. Had it been probed correctly, it would have received
a reset on reconnect from the synaptics driver.
So we've succeeded in making the psmouse driver a little more robust but have
yet to resolve the root cause, which is to correctly identify and configure the touchpad
as a genuine synaptic device.
We're working on getting the datasheet for this part.