In older versions of the elantech hardware/firmware those bits always
were unset, so it didn't actually matter, but newer versions seem to
use those high bits for something else, screwing up the coordinates
we report to the input layer for those devices.
Signed-off-by: NFlorian Ragwitz <rafl@debian.org>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Apparently hardware vendors now ship elantech touchpads with different version
magic. This options allows for them to be tested easier with the current driver
in order to add their magic to the whitelist later.
Signed-off-by: NFlorian Ragwitz <rafl@debian.org>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

The check determining whether device should use 4- or 6-byte packets
was trying to compare firmware with 2.48, but was failing on majors
greater than 2. The new check ensures that versions like 4.1 are
checked properly.
Signed-off-by: NFlorian Ragwitz <rafl@debian.org>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

The new type of touchpads can be detected via a new query command
0x0c. The clickpad flags are in cap[0]:4 and cap[1]:0 bits.

When the device is detected, the driver now reports only the left
button as the supported buttons so that X11 driver can detect that
the device is Clickpad. A Clickpad device gives the button events
only as the middle button. The kernel driver morphs to the left
button. The real handling of Clickpad is done rather in X driver
side.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

This reverts commit 5e28d8eb since
the magic knock does not work for this model of the touchpad and the
device stays in PS/2 compatibility mode.

Observing behavior of the other OS it appears that parity errors reported
by the keyboard controller are being ignored and the data is processed
as usual. Let's do the same for standard PS/2 protocols (bare, Intellimouse
and Intellimouse Explorer) to provide better compatibility. Thsi should fix
teh following bug:

	https://bugzilla.kernel.org/show_bug.cgi?id=6105

Thanks for Damjan Jovanovic for locating the source of issue and ideas
for the patch.
Tested-by: NDamjan Jovanovic <damjan.jov@gmail.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Tested by a user running Ubuntu 9.10 in the following bug report.

BugLink: http://bugs.launchpad.net/bugs/545307Signed-off-by: NChase Douglas <chase.douglas@canonical.com>
Cc: stable@kernel.org
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h

percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files.  percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.

percpu.h -> slab.h dependency is about to be removed.  Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability.  As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.

  http://userweb.kernel.org/~tj/misc/slabh-sweep.py

The script does the followings.

* Scan files for gfp and slab usages and update includes such that
  only the necessary includes are there.  ie. if only gfp is used,
  gfp.h, if slab is used, slab.h.

* When the script inserts a new include, it looks at the include
  blocks and try to put the new include such that its order conforms
  to its surrounding.  It's put in the include block which contains
  core kernel includes, in the same order that the rest are ordered -
  alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
  doesn't seem to be any matching order.

* If the script can't find a place to put a new include (mostly
  because the file doesn't have fitting include block), it prints out
  an error message indicating which .h file needs to be added to the
  file.

The conversion was done in the following steps.

1. The initial automatic conversion of all .c files updated slightly
   over 4000 files, deleting around 700 includes and adding ~480 gfp.h
   and ~3000 slab.h inclusions.  The script emitted errors for ~400
   files.

2. Each error was manually checked.  Some didn't need the inclusion,
   some needed manual addition while adding it to implementation .h or
   embedding .c file was more appropriate for others.  This step added
   inclusions to around 150 files.

3. The script was run again and the output was compared to the edits
   from #2 to make sure no file was left behind.

4. Several build tests were done and a couple of problems were fixed.
   e.g. lib/decompress_*.c used malloc/free() wrappers around slab
   APIs requiring slab.h to be added manually.

5. The script was run on all .h files but without automatically
   editing them as sprinkling gfp.h and slab.h inclusions around .h
   files could easily lead to inclusion dependency hell.  Most gfp.h
   inclusion directives were ignored as stuff from gfp.h was usually
   wildly available and often used in preprocessor macros.  Each
   slab.h inclusion directive was examined and added manually as
   necessary.

6. percpu.h was updated not to include slab.h.

7. Build test were done on the following configurations and failures
   were fixed.  CONFIG_GCOV_KERNEL was turned off for all tests (as my
   distributed build env didn't work with gcov compiles) and a few
   more options had to be turned off depending on archs to make things
   build (like ipr on powerpc/64 which failed due to missing writeq).

   * x86 and x86_64 UP and SMP allmodconfig and a custom test config.
   * powerpc and powerpc64 SMP allmodconfig
   * sparc and sparc64 SMP allmodconfig
   * ia64 SMP allmodconfig
   * s390 SMP allmodconfig
   * alpha SMP allmodconfig
   * um on x86_64 SMP allmodconfig

8. percpu.h modifications were reverted so that it could be applied as
   a separate patch and serve as bisection point.

Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.
Signed-off-by: NTejun Heo <tj@kernel.org>
Guess-its-ok-by: NChristoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>

With the recent system-wide improvements on suspend/resume and EFI
booting the suspend_resume method of the bcm5974 has broken. When
waking up from the S3 state on the MacBookAir, the trackpad is found
in a yet unknown state, unable to switch to the proper multitouch
mode. The result is a frozen touchpad, and a flood of errors of the
kind

	bcm5974: bad trackpad package, length: 8.

This patch retracts the reset_resume method altogether, falling back
on the generic unbind/rebind functionality of the usb layer until
further investigations can be made as how to reset the device when
booting from efi.
Signed-off-by: NHenrik Rydberg <rydberg@euromail.se>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Enable button release event redirection to the device that got the
button press not only for touchpads with interleaved protocols, but
unconditionally for all Alps touchpads. This is required at least
for the touchpads in Dell Inspiron 8200 and Latitude d630.
Signed-off-by: NMartin Buck <mb-tmp-yvahk-vachg@gromit.dyndns.org>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

When reading data from Geyser 2 touchpads used on post Oct 2005 Apple
PowerBooks the driver was casting X and Y coordinates values to
'signed char'. Testing on one of such PowerBooks I have noticed that
touchpad always generates positive values, but some of them are greater
that 127, and thus, when cast to 'signed char' being interpreted as
a negative.

Such bigger values have been observed infrequently, closer to the
edges of a touchpad, so the problem was not very visible.
Nevertheless, the patch would potentially improve touchpad
driver accuracy.
Signed-off-by: NVadim Zaliva <lord@crocodile.org>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Signed-off-by: NThomas Bächler <thomas@archlinux.org>
Cc: stable@kernel.org
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Set state of the device as "initializing" during and after cleanup
to ensure that unsolicited data from the device is not passed on.
We especially want to avoid processing new device announcements
"0xaa 0x00" that can come up before we perform reconnect operation.
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

There are many many ways one can capitalize "Lifebook B Series"...
Signed-off-by: NJon Dodgson <crayzeejon@gmail.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Signed-off-by: NTai-hwa Liang <avatar@sentelic.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Sentelic probes confuse IBM trackpoints so they stop responding to
TP_READ_ID command. See:

	http://bugzilla.kernel.org/show_bug.cgi?id=14970

Let's move FSP detection lower so it is probed after trackpoint and
others, just before we strat probing for Intellimouse Explorer.
Signed-off-by: NTai-hwa Liang <avatar@sentelic.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

For configurations where Synaptics hardware is present but the Synaptics
extensions support is not compiled in, the mouse is reprobed and a new
device is allocated on every suspend/resume.

During probe, psmouse_switch_protocol() calls psmouse_extensions() with
set_properties=1. This calls the dummy synaptics_init() which returns an
error code, instructing us not to use the synaptics extensions.

During resume, psmouse_reconnect() calls psmouse_extensions() with
set_properties=0, in which case call to synaptics_init() is bypassed and
PSMOUSE_SYNAPTICS is returned. Since the result is different from previous
attempt psmouse_reconnect() fails and full re-probe happens.

Fix this by tweaking the set_properties=0 codepath in psmouse_extensions()
to be more careful about offering PSMOUSE_SYNAPTICS extensions.
Signed-off-by: NDaniel Drake <dsd@laptop.org>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Make bcm5974 report raw multi-touch (MT) data in the form of ABS_MT events.

[dtor@mail.ru: get rid of module option, always report all events]
Signed-off-by: NHenrik Rydberg <rydberg@euromail.se>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

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

sysfs_remove_group() waits for sysfs attributes to be removed, therefore
we do not need to worry about driver-specific attributes being accessed
after driver has been detached from the device. In fact, attempts to take
serio->drv_mutex in attribute methods may lead to the following deadlock:

                                          sysfs_read_file()
                                            fill_read_buffer()
                                              sysfs_get_active_two()
                                                psmouse_attr_show_helper()
                                                  serio_pin_driver()
serio_disconnect_driver()
  mutex_lock(&serio->drv_mutex);
                                <-------->        mutex_lock(&serio_drv_mutex);
    psmouse_disconnect()
      sysfs_remove_group(... psmouse_attr_group);
        ....
        sysfs_deactivate();
          wait_for_completion();

Fix this by removing calls to serio_[un]pin_driver() and functions themselves
and using driver-private mutexes to serialize access to attribute's set()
methods that may change device state.
Signed-off-by: NEric W. Biederman <ebiederm@xmission.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Lifebook protocol can only be activated if we find known DMI signature.
It is useles without DMI.
Reported-by: NRakib Mullick <rakib.mullick@gmail.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

The variable 'dev' is unused in function 'hgpk_register'.
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Instead of doing full-blown reset while suspending or shutting down
the box use lighter form of reset that should take less time.
Tested-by: NAlan Stern <stern@rowland.harvard.edu>
Tested-by: NRafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Properly handle version of the protocol where standard PS/2 packets
from trackpoint are stuffed into middle (byte 3-6) of the standard
ALPS packets when both the touchpad and trackpoint are used together.

The patch is based on work done by Matthew Chapman and additional
research done by David Kubicek and Erik Osterholm:

	https://bugs.launchpad.net/ubuntu/+source/linux/+bug/296610

Many thanks to David Kubicek for his efforts in researching fine points
of this new version of the protocol, especially interaction between pad
and stick in these models.
Signed-off-by: NSebastian Kapfer <sebastian_kapfer@gmx.net>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Relative events are only reported via secondary device therefore device
associated with the touchpad should not advertise these capabilities.
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

The driver does not reference identification strings in DMI tables and
since these strings are no longer required by DMI core we can safely
remove them and save some memory.
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

DMI tables use considerable amount of memory. Mark them as __initconst
so they will be discarded once module is loaded.
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

The touchpad on Acer Aspire 5720, 5520 and some other Aspire models
(signature 0x73, 0x02, 0x50) has a button that can be rocked in 4
different directions. Make the driver to generate BTN_0..BTN_3 events
in response. The Synaptics driver by default maps BTN_0 and BTN_1 to
up and down, so there should be no visible changes with the old setup
that generated BTN_FORWARD and BTN_BACK (also mapped to up and down).
Signed-off-by: NMaxim Levitsky <maximlevitsky@gmail.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Do not start protocol detection assuming that middle mouse is present,
instead let individual protocols explicitly set this capability.
This fixes issue with Synaptics touchpads pretending that they have
middle button when hardware clearly reports otherwise.
Reported-and-tested-by: NAndrey Borzenkov <arvidjaar@mail.ru>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

The main input device of Lifebook touchscreens does not generate
left/right/middle button events and therefore should not be advertising
them in its capabilities.
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Elantech touchpads work in absolute mode and do not generate relative
events so they should not be advertising them.
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Touchkit PS/2 touchscreen does not have left/right/middle buttons and
should not be advertising as capable of generating these events.
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Panasonic CF-72 uses 6-byte protocol and does not need to be tied
to a particular port.
Signed-off-by: NAbner Holsinger <9zabner@gmail.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Commit b7802c5c ("Input: psmouse - use boolean type") caused the
synaptics_hardware variable to be completely useless, as it is
constantly set to 'true' throughout the whole psmouse_extensions().
This was caused by the following hunk in the commit in question

	-       int synaptics_hardware = 0;
	+       bool synaptics_hardware = true;

which is wrong and causes driver to issue extra reset when falling
back to bare PS/2 protocol.
Signed-off-by: NJiri Kosina <jkosina@suse.cz>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

The get parameter function should return a string without a life-feed.
Otherwise you'll see additional empty line in sysfs parameters file.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

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

Reported-and-tested-by: NHarald Dunkel <harald.dunkel@t-online.de>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

 - no spaces between function name and opening parenthesis
 - switch statements were indented too much

This makes checkpatch (and me) happy.
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Apparently some of Toshiba Protege M300 identify themselves as
"Portable PC" in DMI so we need to add that to the DMI table as
well. We need DMI data so we can automatically lower Synaptics
reporting rate from 80 to 40 pps to avoid over-taxing their
keyboard controllers.
Tested-by: NRod Davison <roddavison@gmail.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

cancel_delayed_work() may spin and therefore should not be used in
interrupt contexts.
Acked-by: NMike Rapoport <mike@compulab.co.il>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>