Identify Synaptics touchpads without multifinger support.

ok miod@

Discard relative movement packets outside of [-127, 127] range to
prevent cursor jumps when using the trackpoint on some lenovo laptops.

Known affected models:
- Lenovo Thinkpad X13 Gen1
- Lenovo Thinkpad T14(s)
- Lenovo Thinkpad E15 Gen3
- Lenovo A475

With help from stsp@

OK stsp@ miod@ deraadt@ bru@

constify struct cfattach

1. Work around a firmware bug of some Elantech-v1 touchpads.

2. Remove obsolete code from the v1-/v2-packet handlers (coordinate
values are ignored if z is 0).

Thanks to sxvghd for providing the base of this patch, and for testing.

Get Elantech v1 touchpad with fw version 0x20022 working.

From sxvghd (at) firemail (dot) cc, on tech@

Fix intermittent failing device initialization seen on some Synaptics
devices which has been re-introduced by pms.c revision 1.92.

ok tb@

1. Run Synaptics touchpads in plain W-mode. 2. Update the AGM-check.

Enabling extended W-mode is unnecessary as long as pms does not process
the additional packets properly.

Thanks to David Demelier for his reports and tests.

ok anton@ mpi@

Unbreak !DIAGNOSTIC build.

From Andrius V vezhlys (at) gmail, on bugs@

Extend the identification and validation of elantech-v4 packets to the
"crc-enabled" format.

ok awolk@

Accept more elantech-v4 models, and apply a test for external buttons.

Thanks to Alexander Cronheim.

ok mpi@

Wait a bit for the set mode command to complete for Synaptics devices to
make sure that the device doesn't get enabled before, which would make it
fail.
Fixes regular occurring 'not in sync yet, discard input' issue seen on
ThinkPad X1 Gen6.

ok deraadt@

Add support for Elantech trackpoints to pms. Extend the list of
"IC types" that identify Elantech-V4 touchpads.

Thanks to Ryan Lennox for help and testing.

ok mpi@

'juming' -> 'jumping' in comment.

Pointed out by Ryan Lennox via tech@

pms: minor cleanups in the alps code.

1. Update the ALPS_*_BEZEL values, which don't match the actual limits
accurately. The new values are used by Linux for all models with the
protocols "V1" and "V2".

2. Models with the ID 0x7331 are not supported properly, remove it
from the list.

3. Report to wsmouse that multiple contacts aren't recognized.

add the alps touchpad that can be found in the ideapad 710s

ok mpi@

Detect touchpad reset announcements. On rare occasions, the touchpad is reset
due to power failure and an announcement is transmitted as input by the device
making it detectable. At this point, the device must be re-enabled to continue
working since any configuration done prior to the reset is lost.

Detection is implemented using a watchdog and enabled for all protocols. So far,
the announcement has been observed on Synaptics by me and on Elantech v4 by
bru@. Extra care has been taken to handle potential false positives: when the
announcement byte sequence is part of a valid input packet.

with help and ok bru@, mpi@

Print out the offending data causing pms sync to fail.

ok bru@, mpi@ (as part of a larger diff)

Adapt the Elantech handlers (v1-3) to the new wsmouse infrastructure.

The new versions use wsmouse_configure, and leave compat-mode handling
to wsmouse.

ok stsp

bug fix: intialize variables in synaptics_get_hwinfo

Query the lower coordinate limits.

Show Synaptics Model IDs.

pms/alps: configure compat mode

Set the weak variant of hysteresis as default.

improved coordinate filters

ok mpi@

Use nitems() when passing all known parameters to wsmouse_configure().

ok bru@

pms/elantech-v4: configure wsmouse for handling compat-mode

ok @matthieu @stsp @mpi

pms/synaptics: configure wsmouse for handling compat-mode

ok @matthieu @stsp @mpi

Improved parameter handling in wsmouse and new ioctls for reading and
setting parameter values.

ok matthieu@

Use the new input functions of wsmouse in mouse and touchscreen drivers.

ok kettenis@

Add support for multitouch input to wsmouse.

This change adds new input-processing functions to wsmouse and
adapts the touchpad drivers.

ok mpi@, shadchin@

pakets -> packets

Support Synaptics touchpads without W mode.

Emulate W mode if a Synaptics model doesn't provide it, and check
for the success of the resolution query during setup.

ok shadchin@, mpi@

Improve tap-and-drag detection for ALPS touchpads.

The ALPS mechanism for gesture detection can produce long delays between
a packet that signals a tap and the next packet, which either signals
the end of the gesture or the start of a drag-operation. The time spans
can exceed the timeouts used by the generic detection mechanism in the
synaptics driver. Reporting both the touch and the release event when
the second packet has arrived ensures that the tap will be recognized.

ok mpi@ shadchin@

fairly trivial sizes for free()

Tweak previous; the Synaptics TrackPoint in my ThinkPad T500 works again.

ok krw@

Don't emit values for 'width' that

/usr/xenocara/driver/xf86-input-synaptics/src/wsconscomm.c

doesn't understand.

'Fixes' synaptics trackpad in Dell L400 laptop so the cursor can
be moved in X. Problem noted and experimental Dell L400 donated by
Paolo Aglialoro. Thanks!

ok miod@ mpi@ jcs@

pms(4): Don't match Elantech v4 devices with firmware versions 0xX7XXXX.
Apparently we don't support these touchpads properly so leave them in
PS/2 compat mode. Regression reported by Remi Locherer on bugs@. And stop
matching devices with firmware versions higher than 0xX8XXXX since we cannot
be sure they work.
ok mpi@

Match newer elantech v4 touchpads, logic taken from Linux.

Based on a submission from and ok jcs@.

Print irq informations in pckbc_set_inputhandler().

ok miod@

Add support for CRC-enabled elantech v3 touchpads.
Patch by Fasse <fasse.f2@gmail.com>
ok mpi@

Use the 0x10 extended capability to determine if the trackstick
buttons are wired to the trackpad and need to be re-routed to the
trackstick. Without this change the buttons on 2015 thinkpads
get picked up as extended buttons that show up as scroll up/down.

Remove the X1 Carbon 2015 (LEN0048) and X250 (LEN0046) from the
top button area/soft buttons quirks list. Also avoid using the
quirk list entirely if the capability bit is set.

Based on work from jcs@ with feedback from mpi@
ok mpi@ jcs@

Complete rewrite of the Elantech clickpads v4 packet parser in order to
support "click-and-drag".

This implements a simple filter to keep track of a moving finger on the
clickpad. For the moment wsmouse(4) and wscons(4) are still unaware of
multitouch events and pms(4) has to do the translation, but this is a
small steps in this direction.

All the good work has been done by Ulf Brosziewski.

Define a new wscons mouse type for Synaptics clickpad devices that lack
physical buttons. This will be used if the acpi pnpid for the mouse
matches a list provided by Synaptics found in Linux.

Combined with the xenocara changes this will remove the need for an xorg
config file for the x240/t440/t540 generation of thinkpads.

ok matthieu@ shadchin@ kettenis@

Only MS IntelliMouse have a Z axis, prevent a false positive triggered
by -Wuninitialized reported by jsg@.

ok stsp@

Fix checking sync for old synaptics touchpad (ver 5.9)
From Thierry Deval <thierry at deval dot be>

ok mpi@

Make sure state changes are properly serialized.

When pms(4) is attached to a touchpad it generally presents two different
wsmouse(4) devices: one for the touchpad itself and one for the clitpad
and/or some interleaved packets. But since both devices are writing to
the same pckbc slot, a race can occur if they try to change the state at
the same time.

So prevent two process opening the two /dev/wsmouse* node at the same time
to corrupt the magic sequences needed to enable/disable the touchpad.

ok schadchin@

add a size argument to free. will be used soon, but for now default to 0.
after discussions with beck deraadt kettenis.

Fix tail packet check in elantech v3 touchpad code. Due to a typo this
code was masking out bits which were also tested in the same expression.
Reported by cppcheck via jsg@ (Expression '(X & 0xfc) != 0x2' is always true).
Tested with Elantech Touchpad, version 3, firmware 0x250f00.

Do no reprobe for a supported protocol when enabling a pointing
device of type PMS_STANDARD, probing during autoconf(9) is enough!.

In such case, the device can be a simple PS/2 mouse, a unsupported
touchpad or a downgraded touchpad if something bad happened. But
it is very unlikely that reprobing will improve the situation and
transform a standard mouse into a multitouch-aware touchpad ;)

This fixes the 12 seconds delay seen on various Dell laptops (E4310
and Latitude D630) when starting Xorg, reported by K��rlis Mi��elsons
on bugs@.

ok stsp@, dcoppa@, shadchin@

Fix jagged diagonal lines (kernel part)

Send WSCONS_EVENT_SYNC every time you call wsmouse_input(). Used to synchronize
and separate events into packets of input data changes occurring at the same
moment in time. For example, motion of a mouse may set the DELTA_X and DELTA_Y
values for one motion, then emit a SYNC.

ok matthieu@. tested edd@, Henri Kemppainen and Alf Schlichting.

Use a working magic sequence to configure absolute mode for elantech v2.
The synaptics driver now attaches to elantech v2 touchpads. It seems these
pads were always falling back to compat mode, which went unnoticed until r1.47.

Fix based on a patch submitted by Cedric Tessier:
http://marc.info/?l=openbsd-bugs&m=136904396518528&w=2
which in turn was probably based on parts of this patch:
https://launchpadlibrarian.net/73099683/elantech.patch
which updates the Linux driver code but isn't part of Torvald's Linux.

Behaviour of elantech firmware versions seems to vary quite a bit so print
the firmware version at attach time for reference in future bug reports.

Fix tested by frantisek holop and Peter J. Philipp.
ok mpi@

When /dev/wsmouse is opened right after resume, some synaptics touchpads
are still taking time to wake up. So make the pms(4) driver wait a little
longer before giving up on them. Fixes mouse after resume on my x130e TP.
ok mpi, earlier versions tested by weerd with no regressions seen

Only free the per-protocol descriptor if a touchpad cannot be correctly
identified during attach. This prevents the driver to downgrade itself
to use the standard mouse protocol and reattach a new wsmouse child if
something bad happens when we re-query the hardware (during resume for
example).

Discussed with and ok stsp@

Correctly initialized the width value to 0 instead of passing garbage
to wsmouse_input() when no finger are reported.

Pointed out by Maxime Villard.

send proper z value to elantech_send_input in previous v4 commit

pointed out by Owain G. Ainsworth

add elantech v4 (clickpad) support
tested on an asus ux21a

ok mpi@ stsp@

Revert Active PS/2 support for now, until we find a solution suitable
for chipsets which stopped working.

ok mpi

Enable active PS/2 multiplexing if available.
Supported for i386 and amd64 except SMALL_KERNEL.

Based on Miod's former work on this subject.

ok mpi

Revert r1.37 now that we stop after the first matching protocol, tested
by Peter J. Philipp, thanks.

Move the IntelliMouse protocol definition after Elantech ones, because some
touchpads support both of them and we want to pick the latter.

Reported by Peter J. Philipp, thanks.

Don't probe for all supported protocols this can confuse some touchpads and
makes it harder to pick the right protocol if a device answers to more than
one magic sequence.

Tested by many on tech@, ok shadchin@

Once we've detected an elantech touchpad, stop probing for other hardware
versions. Fixes a problem reported by Peter J. Philipp where the pms driver
spits out "not in sync yet" messages after attaching to elantech v2 hardware.
Fix tested by me on v3 hardware, and by Peter on v2 hardware.
ok mpi@ shadchin@

Enable untested Elantech touchpad v1 and v2 code. I'd like some test reports.
ok mpi@ shadchin@

Fix memory leak in error paths for synaptics and alps (pointed stsp@).
Also small simplify elantech part.

ok stsp@, mpi@

Handle elantech touchpad buttons in a common code path for all
hardware variants. ok shadchin mikeb mpi

Don't OR flags into an uninitialized variable in elantech hardware v1 code.
Fix by oga, via mikeb.

Add support for Elantech touchpads to pms(4). This allows the X synaptics(4)
input driver to configure these touchpads properly, e.g. toggle tap-to-click,
two-finger scrolling, edge-scrolling, etc. So far these pads were working
only in PS/2 mouse compatibility mode.

Support for up to hardware version 3 has been implemented. But only version 3
has been tested so versions 1 and 2 remain disabled via #ifdef notyet until
someone tests (bugs in this code could seriously mess up the mouse and I
don't want to end up shipping such bugs in 5.3).

help and ok mpi@ shadchin@

Fix for ClickPad. On issue pointed gilles@ and matthieu@

ok gilles@, matthieu@, mpi@, miod@, deraadt@

Pass a width value corresponding to one finger for ALPS touchpads when the
reported pressure is non-null. Fix the use of ALPS touchpads with recent
(>1.6) xf86-input-synaptics drivers, issue reported by janis at cieti lv.

Tested by yasuoka@ and janis, ok shadchin@

Added yet another magic for my ALPS touchpad.

ok mpi@

Adds support for a slightly different version of the ALPS protocol,
which send 'interleaved' PS2 packets in between absolute ALPS packets.
Dell laptops (E6xxx and E5500 at least) have this kind of touchpad.

ok and some rework mpi@, tested okan@.

Blacklist a dualpoint model which clitpad sends non compatible PS2 packets.

Issue reported by deraadt@

renaming synaptics_pt_* -> pms_sec_*
(synaptics and alps uses this for trackpoint)

ok mpi@

Correctly handle clitpad packets for some ALPS models.

reported by espie@, not reported by miod@

with and ok shadchin@, ok espie@

Blacklist alps devices that might have a clitpad untill the driver get fixed

Fix the synchronization of some alps pads, issue reported by Bryan (brakeb)
on misc@

Similar diff tested by okan@ and Mathieu (ptr.jetable)

ok shadchin@

Add support for ALPS touchpads.

This also fix the "slow touchpad" problem due to the fact that some ALPS
devices respond to some Synaptics messages and stay in un half-configured
state.

Tested by Rivo Nurges, armani@, shadchin@ and matthieu@

ok shadchin@

revert the last commit. pms_enable_alps() does not guarantee
the accuracy detection ALPS.

ok mpi@

Adding stub for ALPS touchpad, which not supported yet.
This fixed "react strangely" when the pms driver probes
for a synaptic-compatible touchpad.

Tested and ok mpi@

Adding support Synaptics touchpad. Thanks all for test.

ok miod@, matthieu@

remove sc->sc_kbcslot, as is always PCKBC_AUX_SLOT
ok krw, miod

cleanup pms_enable_intelli: magic numbers -> defines
ok krw, miod

fix crash:
after boot and before the first activation (wsmoused or X),
sc->protocol is not initialized, and any call ioctl lead to crash
(example 'wsconsctl -a')

thanks and ok krw@

Add a common interface for various devices

This is to make it easier to add additional mice types in future

ok krw@, miod@, nicm@

Aesthetic improvements to make it appear more normal and knf'y.

No functional change. From Alexandr Scadchin again.

More cleanup. Add DEVNAME() macro and generic functions for various
actions (set/get/reset/enable/disable) to wrap pms_cmd() invocations.

No functional change. From Alexandr Scadchin again.

Minor code cleanups. No functional change. From Alexandr Shadchin.

ok nicm@

Unify some error messages to simplify reset code. No functional
change.

ok nicm@

From shadchin.

Ansify non-ansified function definitions. No functional change.

ok nicm@

From shadchin.

Remove some duplicated code. No need to flush/reset at attach and then
again at PMS_STATE_CHANGE events.

Tested my myself, ian@ and nicm@ without regressions on usual
problematic machines.

From shadchin.

Fix some pms(4) vs wsmoused issues. From Alexandr Shadchin via
tech@. Tested on the usual problematic machines by myself, nicm@
and ian@.

Fix Ian@'s pms keyboard/mouse while keeping my eeePC working.

For posterity: there is one time when pckbc commands must be thrown,
and one time when pckbc commands must be polled.

Diff from shadchin.

ok miod@

The attach function can return 1 now because there is no more priority
scheme relating to pmsi.
from alexandr.shadchin@gmail.com

Make sure the SEND_DEV_ID command is sent synchronously, so the response buffer
is valid when we look at it. Repairs intellimouse (former pmsi) operation.

Merge pmsi(4) into pms(4). From Alexandr Shadchin on tech@.

Toggle the mouse through the disabled and enabled states when suspending
and restoring. Fixes most mice.

use NULL to point to the response buffer, which avoids confusion; ok fgsch

Correct a 10-year old typo inherited from NetBSD by renaming psm.c,
psm_intelli.c and psmreg.h into pms.c, pms_intelli.c and pmsreg.h.

ok deraadt@

Discard relative movement packets outside of [-127, 127] range to
prevent cursor jumps when using the trackpoint on some lenovo laptops.

Known affected models:
- Lenovo Thinkpad X13 Gen1
- Lenovo Thinkpad T14(s)
- Lenovo Thinkpad E15 Gen3
- Lenovo A475

With help from stsp@

OK stsp@ miod@ deraadt@ bru@

constify struct cfattach

1. Work around a firmware bug of some Elantech-v1 touchpads.

2. Remove obsolete code from the v1-/v2-packet handlers (coordinate
values are ignored if z is 0).

Thanks to sxvghd for providing the base of this patch, and for testing.

Get Elantech v1 touchpad with fw version 0x20022 working.

From sxvghd (at) firemail (dot) cc, on tech@

Fix intermittent failing device initialization seen on some Synaptics
devices which has been re-introduced by pms.c revision 1.92.

ok tb@

1. Run Synaptics touchpads in plain W-mode. 2. Update the AGM-check.

Enabling extended W-mode is unnecessary as long as pms does not process
the additional packets properly.

Thanks to David Demelier for his reports and tests.

ok anton@ mpi@

Unbreak !DIAGNOSTIC build.

From Andrius V vezhlys (at) gmail, on bugs@

Extend the identification and validation of elantech-v4 packets to the
"crc-enabled" format.

ok awolk@

Accept more elantech-v4 models, and apply a test for external buttons.

Thanks to Alexander Cronheim.

ok mpi@

Wait a bit for the set mode command to complete for Synaptics devices to
make sure that the device doesn't get enabled before, which would make it
fail.
Fixes regular occurring 'not in sync yet, discard input' issue seen on
ThinkPad X1 Gen6.

ok deraadt@

Add support for Elantech trackpoints to pms. Extend the list of
"IC types" that identify Elantech-V4 touchpads.

Thanks to Ryan Lennox for help and testing.

ok mpi@

'juming' -> 'jumping' in comment.

Pointed out by Ryan Lennox via tech@

pms: minor cleanups in the alps code.

1. Update the ALPS_*_BEZEL values, which don't match the actual limits
accurately. The new values are used by Linux for all models with the
protocols "V1" and "V2".

2. Models with the ID 0x7331 are not supported properly, remove it
from the list.

3. Report to wsmouse that multiple contacts aren't recognized.

add the alps touchpad that can be found in the ideapad 710s

ok mpi@

Detect touchpad reset announcements. On rare occasions, the touchpad is reset
due to power failure and an announcement is transmitted as input by the device
making it detectable. At this point, the device must be re-enabled to continue
working since any configuration done prior to the reset is lost.

Detection is implemented using a watchdog and enabled for all protocols. So far,
the announcement has been observed on Synaptics by me and on Elantech v4 by
bru@. Extra care has been taken to handle potential false positives: when the
announcement byte sequence is part of a valid input packet.

with help and ok bru@, mpi@

Print out the offending data causing pms sync to fail.

ok bru@, mpi@ (as part of a larger diff)

Adapt the Elantech handlers (v1-3) to the new wsmouse infrastructure.

The new versions use wsmouse_configure, and leave compat-mode handling
to wsmouse.

ok stsp

bug fix: intialize variables in synaptics_get_hwinfo

Query the lower coordinate limits.

Show Synaptics Model IDs.

pms/alps: configure compat mode

Set the weak variant of hysteresis as default.

improved coordinate filters

ok mpi@

Use nitems() when passing all known parameters to wsmouse_configure().

ok bru@

pms/elantech-v4: configure wsmouse for handling compat-mode

ok @matthieu @stsp @mpi

pms/synaptics: configure wsmouse for handling compat-mode

ok @matthieu @stsp @mpi

Improved parameter handling in wsmouse and new ioctls for reading and
setting parameter values.

ok matthieu@

Use the new input functions of wsmouse in mouse and touchscreen drivers.

ok kettenis@

Add support for multitouch input to wsmouse.

This change adds new input-processing functions to wsmouse and
adapts the touchpad drivers.

ok mpi@, shadchin@

pakets -> packets

Support Synaptics touchpads without W mode.

Emulate W mode if a Synaptics model doesn't provide it, and check
for the success of the resolution query during setup.

ok shadchin@, mpi@

Improve tap-and-drag detection for ALPS touchpads.

The ALPS mechanism for gesture detection can produce long delays between
a packet that signals a tap and the next packet, which either signals
the end of the gesture or the start of a drag-operation. The time spans
can exceed the timeouts used by the generic detection mechanism in the
synaptics driver. Reporting both the touch and the release event when
the second packet has arrived ensures that the tap will be recognized.

ok mpi@ shadchin@

fairly trivial sizes for free()

Tweak previous; the Synaptics TrackPoint in my ThinkPad T500 works again.

ok krw@

Don't emit values for 'width' that

/usr/xenocara/driver/xf86-input-synaptics/src/wsconscomm.c

doesn't understand.

'Fixes' synaptics trackpad in Dell L400 laptop so the cursor can
be moved in X. Problem noted and experimental Dell L400 donated by
Paolo Aglialoro. Thanks!

ok miod@ mpi@ jcs@

pms(4): Don't match Elantech v4 devices with firmware versions 0xX7XXXX.
Apparently we don't support these touchpads properly so leave them in
PS/2 compat mode. Regression reported by Remi Locherer on bugs@. And stop
matching devices with firmware versions higher than 0xX8XXXX since we cannot
be sure they work.
ok mpi@

Match newer elantech v4 touchpads, logic taken from Linux.

Based on a submission from and ok jcs@.

Print irq informations in pckbc_set_inputhandler().

ok miod@

Add support for CRC-enabled elantech v3 touchpads.
Patch by Fasse <fasse.f2@gmail.com>
ok mpi@

Use the 0x10 extended capability to determine if the trackstick
buttons are wired to the trackpad and need to be re-routed to the
trackstick. Without this change the buttons on 2015 thinkpads
get picked up as extended buttons that show up as scroll up/down.

Remove the X1 Carbon 2015 (LEN0048) and X250 (LEN0046) from the
top button area/soft buttons quirks list. Also avoid using the
quirk list entirely if the capability bit is set.

Based on work from jcs@ with feedback from mpi@
ok mpi@ jcs@

Complete rewrite of the Elantech clickpads v4 packet parser in order to
support "click-and-drag".

This implements a simple filter to keep track of a moving finger on the
clickpad. For the moment wsmouse(4) and wscons(4) are still unaware of
multitouch events and pms(4) has to do the translation, but this is a
small steps in this direction.

All the good work has been done by Ulf Brosziewski.

Define a new wscons mouse type for Synaptics clickpad devices that lack
physical buttons. This will be used if the acpi pnpid for the mouse
matches a list provided by Synaptics found in Linux.

Combined with the xenocara changes this will remove the need for an xorg
config file for the x240/t440/t540 generation of thinkpads.

ok matthieu@ shadchin@ kettenis@

Only MS IntelliMouse have a Z axis, prevent a false positive triggered
by -Wuninitialized reported by jsg@.

ok stsp@

Fix checking sync for old synaptics touchpad (ver 5.9)
From Thierry Deval <thierry at deval dot be>

ok mpi@

Make sure state changes are properly serialized.

When pms(4) is attached to a touchpad it generally presents two different
wsmouse(4) devices: one for the touchpad itself and one for the clitpad
and/or some interleaved packets. But since both devices are writing to
the same pckbc slot, a race can occur if they try to change the state at
the same time.

So prevent two process opening the two /dev/wsmouse* node at the same time
to corrupt the magic sequences needed to enable/disable the touchpad.

ok schadchin@

add a size argument to free. will be used soon, but for now default to 0.
after discussions with beck deraadt kettenis.

Fix tail packet check in elantech v3 touchpad code. Due to a typo this
code was masking out bits which were also tested in the same expression.
Reported by cppcheck via jsg@ (Expression '(X & 0xfc) != 0x2' is always true).
Tested with Elantech Touchpad, version 3, firmware 0x250f00.

Do no reprobe for a supported protocol when enabling a pointing
device of type PMS_STANDARD, probing during autoconf(9) is enough!.

In such case, the device can be a simple PS/2 mouse, a unsupported
touchpad or a downgraded touchpad if something bad happened. But
it is very unlikely that reprobing will improve the situation and
transform a standard mouse into a multitouch-aware touchpad ;)

This fixes the 12 seconds delay seen on various Dell laptops (E4310
and Latitude D630) when starting Xorg, reported by K��rlis Mi��elsons
on bugs@.

ok stsp@, dcoppa@, shadchin@

Fix jagged diagonal lines (kernel part)

Send WSCONS_EVENT_SYNC every time you call wsmouse_input(). Used to synchronize
and separate events into packets of input data changes occurring at the same
moment in time. For example, motion of a mouse may set the DELTA_X and DELTA_Y
values for one motion, then emit a SYNC.

ok matthieu@. tested edd@, Henri Kemppainen and Alf Schlichting.

Use a working magic sequence to configure absolute mode for elantech v2.
The synaptics driver now attaches to elantech v2 touchpads. It seems these
pads were always falling back to compat mode, which went unnoticed until r1.47.

Fix based on a patch submitted by Cedric Tessier:
http://marc.info/?l=openbsd-bugs&m=136904396518528&w=2
which in turn was probably based on parts of this patch:
https://launchpadlibrarian.net/73099683/elantech.patch
which updates the Linux driver code but isn't part of Torvald's Linux.

Behaviour of elantech firmware versions seems to vary quite a bit so print
the firmware version at attach time for reference in future bug reports.

Fix tested by frantisek holop and Peter J. Philipp.
ok mpi@

When /dev/wsmouse is opened right after resume, some synaptics touchpads
are still taking time to wake up. So make the pms(4) driver wait a little
longer before giving up on them. Fixes mouse after resume on my x130e TP.
ok mpi, earlier versions tested by weerd with no regressions seen

Only free the per-protocol descriptor if a touchpad cannot be correctly
identified during attach. This prevents the driver to downgrade itself
to use the standard mouse protocol and reattach a new wsmouse child if
something bad happens when we re-query the hardware (during resume for
example).

Discussed with and ok stsp@

Correctly initialized the width value to 0 instead of passing garbage
to wsmouse_input() when no finger are reported.

Pointed out by Maxime Villard.

send proper z value to elantech_send_input in previous v4 commit

pointed out by Owain G. Ainsworth

add elantech v4 (clickpad) support
tested on an asus ux21a

ok mpi@ stsp@

Revert Active PS/2 support for now, until we find a solution suitable
for chipsets which stopped working.

ok mpi

Enable active PS/2 multiplexing if available.
Supported for i386 and amd64 except SMALL_KERNEL.

Based on Miod's former work on this subject.

ok mpi

Revert r1.37 now that we stop after the first matching protocol, tested
by Peter J. Philipp, thanks.

Move the IntelliMouse protocol definition after Elantech ones, because some
touchpads support both of them and we want to pick the latter.

Reported by Peter J. Philipp, thanks.

Don't probe for all supported protocols this can confuse some touchpads and
makes it harder to pick the right protocol if a device answers to more than
one magic sequence.

Tested by many on tech@, ok shadchin@

Once we've detected an elantech touchpad, stop probing for other hardware
versions. Fixes a problem reported by Peter J. Philipp where the pms driver
spits out "not in sync yet" messages after attaching to elantech v2 hardware.
Fix tested by me on v3 hardware, and by Peter on v2 hardware.
ok mpi@ shadchin@

Enable untested Elantech touchpad v1 and v2 code. I'd like some test reports.
ok mpi@ shadchin@

Fix memory leak in error paths for synaptics and alps (pointed stsp@).
Also small simplify elantech part.

ok stsp@, mpi@

Handle elantech touchpad buttons in a common code path for all
hardware variants. ok shadchin mikeb mpi

Don't OR flags into an uninitialized variable in elantech hardware v1 code.
Fix by oga, via mikeb.

Add support for Elantech touchpads to pms(4). This allows the X synaptics(4)
input driver to configure these touchpads properly, e.g. toggle tap-to-click,
two-finger scrolling, edge-scrolling, etc. So far these pads were working
only in PS/2 mouse compatibility mode.

Support for up to hardware version 3 has been implemented. But only version 3
has been tested so versions 1 and 2 remain disabled via #ifdef notyet until
someone tests (bugs in this code could seriously mess up the mouse and I
don't want to end up shipping such bugs in 5.3).

help and ok mpi@ shadchin@

Fix for ClickPad. On issue pointed gilles@ and matthieu@

ok gilles@, matthieu@, mpi@, miod@, deraadt@

Pass a width value corresponding to one finger for ALPS touchpads when the
reported pressure is non-null. Fix the use of ALPS touchpads with recent
(>1.6) xf86-input-synaptics drivers, issue reported by janis at cieti lv.

Tested by yasuoka@ and janis, ok shadchin@

Added yet another magic for my ALPS touchpad.

ok mpi@

Adds support for a slightly different version of the ALPS protocol,
which send 'interleaved' PS2 packets in between absolute ALPS packets.
Dell laptops (E6xxx and E5500 at least) have this kind of touchpad.

ok and some rework mpi@, tested okan@.

Blacklist a dualpoint model which clitpad sends non compatible PS2 packets.

Issue reported by deraadt@

renaming synaptics_pt_* -> pms_sec_*
(synaptics and alps uses this for trackpoint)

ok mpi@

Correctly handle clitpad packets for some ALPS models.

reported by espie@, not reported by miod@

with and ok shadchin@, ok espie@

Blacklist alps devices that might have a clitpad untill the driver get fixed

Fix the synchronization of some alps pads, issue reported by Bryan (brakeb)
on misc@

Similar diff tested by okan@ and Mathieu (ptr.jetable)

ok shadchin@

Add support for ALPS touchpads.

This also fix the "slow touchpad" problem due to the fact that some ALPS
devices respond to some Synaptics messages and stay in un half-configured
state.

Tested by Rivo Nurges, armani@, shadchin@ and matthieu@

ok shadchin@

revert the last commit. pms_enable_alps() does not guarantee
the accuracy detection ALPS.

ok mpi@

Adding stub for ALPS touchpad, which not supported yet.
This fixed "react strangely" when the pms driver probes
for a synaptic-compatible touchpad.

Tested and ok mpi@

Adding support Synaptics touchpad. Thanks all for test.

ok miod@, matthieu@

remove sc->sc_kbcslot, as is always PCKBC_AUX_SLOT
ok krw, miod

cleanup pms_enable_intelli: magic numbers -> defines
ok krw, miod

fix crash:
after boot and before the first activation (wsmoused or X),
sc->protocol is not initialized, and any call ioctl lead to crash
(example 'wsconsctl -a')

thanks and ok krw@

Add a common interface for various devices

This is to make it easier to add additional mice types in future

ok krw@, miod@, nicm@

Aesthetic improvements to make it appear more normal and knf'y.

No functional change. From Alexandr Scadchin again.

More cleanup. Add DEVNAME() macro and generic functions for various
actions (set/get/reset/enable/disable) to wrap pms_cmd() invocations.

No functional change. From Alexandr Scadchin again.

Minor code cleanups. No functional change. From Alexandr Shadchin.

ok nicm@

Unify some error messages to simplify reset code. No functional
change.

ok nicm@

From shadchin.

Ansify non-ansified function definitions. No functional change.

ok nicm@

From shadchin.

Remove some duplicated code. No need to flush/reset at attach and then
again at PMS_STATE_CHANGE events.

Tested my myself, ian@ and nicm@ without regressions on usual
problematic machines.

From shadchin.

Fix some pms(4) vs wsmoused issues. From Alexandr Shadchin via
tech@. Tested on the usual problematic machines by myself, nicm@
and ian@.

Fix Ian@'s pms keyboard/mouse while keeping my eeePC working.

For posterity: there is one time when pckbc commands must be thrown,
and one time when pckbc commands must be polled.

Diff from shadchin.

ok miod@

The attach function can return 1 now because there is no more priority
scheme relating to pmsi.
from alexandr.shadchin@gmail.com

Make sure the SEND_DEV_ID command is sent synchronously, so the response buffer
is valid when we look at it. Repairs intellimouse (former pmsi) operation.

Merge pmsi(4) into pms(4). From Alexandr Shadchin on tech@.

Toggle the mouse through the disabled and enabled states when suspending
and restoring. Fixes most mice.

use NULL to point to the response buffer, which avoids confusion; ok fgsch

Correct a 10-year old typo inherited from NetBSD by renaming psm.c,
psm_intelli.c and psmreg.h into pms.c, pms_intelli.c and pmsreg.h.

ok deraadt@

constify struct cfattach

1. Work around a firmware bug of some Elantech-v1 touchpads.

2. Remove obsolete code from the v1-/v2-packet handlers (coordinate
values are ignored if z is 0).

Thanks to sxvghd for providing the base of this patch, and for testing.

Get Elantech v1 touchpad with fw version 0x20022 working.

From sxvghd (at) firemail (dot) cc, on tech@

Fix intermittent failing device initialization seen on some Synaptics
devices which has been re-introduced by pms.c revision 1.92.

ok tb@

1. Run Synaptics touchpads in plain W-mode. 2. Update the AGM-check.

Enabling extended W-mode is unnecessary as long as pms does not process
the additional packets properly.

Thanks to David Demelier for his reports and tests.

ok anton@ mpi@

Unbreak !DIAGNOSTIC build.

From Andrius V vezhlys (at) gmail, on bugs@

Extend the identification and validation of elantech-v4 packets to the
"crc-enabled" format.

ok awolk@

Accept more elantech-v4 models, and apply a test for external buttons.

Thanks to Alexander Cronheim.

ok mpi@

Wait a bit for the set mode command to complete for Synaptics devices to
make sure that the device doesn't get enabled before, which would make it
fail.
Fixes regular occurring 'not in sync yet, discard input' issue seen on
ThinkPad X1 Gen6.

ok deraadt@

Add support for Elantech trackpoints to pms. Extend the list of
"IC types" that identify Elantech-V4 touchpads.

Thanks to Ryan Lennox for help and testing.

ok mpi@

'juming' -> 'jumping' in comment.

Pointed out by Ryan Lennox via tech@

pms: minor cleanups in the alps code.

1. Update the ALPS_*_BEZEL values, which don't match the actual limits
accurately. The new values are used by Linux for all models with the
protocols "V1" and "V2".

2. Models with the ID 0x7331 are not supported properly, remove it
from the list.

3. Report to wsmouse that multiple contacts aren't recognized.

add the alps touchpad that can be found in the ideapad 710s

ok mpi@

Detect touchpad reset announcements. On rare occasions, the touchpad is reset
due to power failure and an announcement is transmitted as input by the device
making it detectable. At this point, the device must be re-enabled to continue
working since any configuration done prior to the reset is lost.

Detection is implemented using a watchdog and enabled for all protocols. So far,
the announcement has been observed on Synaptics by me and on Elantech v4 by
bru@. Extra care has been taken to handle potential false positives: when the
announcement byte sequence is part of a valid input packet.

with help and ok bru@, mpi@

Print out the offending data causing pms sync to fail.

ok bru@, mpi@ (as part of a larger diff)

Adapt the Elantech handlers (v1-3) to the new wsmouse infrastructure.

The new versions use wsmouse_configure, and leave compat-mode handling
to wsmouse.

ok stsp

bug fix: intialize variables in synaptics_get_hwinfo

Query the lower coordinate limits.

Show Synaptics Model IDs.

pms/alps: configure compat mode

Set the weak variant of hysteresis as default.

improved coordinate filters

ok mpi@

Use nitems() when passing all known parameters to wsmouse_configure().

ok bru@

pms/elantech-v4: configure wsmouse for handling compat-mode

ok @matthieu @stsp @mpi

pms/synaptics: configure wsmouse for handling compat-mode

ok @matthieu @stsp @mpi

Improved parameter handling in wsmouse and new ioctls for reading and
setting parameter values.

ok matthieu@

Use the new input functions of wsmouse in mouse and touchscreen drivers.

ok kettenis@

Add support for multitouch input to wsmouse.

This change adds new input-processing functions to wsmouse and
adapts the touchpad drivers.

ok mpi@, shadchin@

pakets -> packets

Support Synaptics touchpads without W mode.

Emulate W mode if a Synaptics model doesn't provide it, and check
for the success of the resolution query during setup.

ok shadchin@, mpi@

Improve tap-and-drag detection for ALPS touchpads.

The ALPS mechanism for gesture detection can produce long delays between
a packet that signals a tap and the next packet, which either signals
the end of the gesture or the start of a drag-operation. The time spans
can exceed the timeouts used by the generic detection mechanism in the
synaptics driver. Reporting both the touch and the release event when
the second packet has arrived ensures that the tap will be recognized.

ok mpi@ shadchin@

fairly trivial sizes for free()

Tweak previous; the Synaptics TrackPoint in my ThinkPad T500 works again.

ok krw@

Don't emit values for 'width' that

/usr/xenocara/driver/xf86-input-synaptics/src/wsconscomm.c

doesn't understand.

'Fixes' synaptics trackpad in Dell L400 laptop so the cursor can
be moved in X. Problem noted and experimental Dell L400 donated by
Paolo Aglialoro. Thanks!

ok miod@ mpi@ jcs@

pms(4): Don't match Elantech v4 devices with firmware versions 0xX7XXXX.
Apparently we don't support these touchpads properly so leave them in
PS/2 compat mode. Regression reported by Remi Locherer on bugs@. And stop
matching devices with firmware versions higher than 0xX8XXXX since we cannot
be sure they work.
ok mpi@

Match newer elantech v4 touchpads, logic taken from Linux.

Based on a submission from and ok jcs@.

Print irq informations in pckbc_set_inputhandler().

ok miod@

Add support for CRC-enabled elantech v3 touchpads.
Patch by Fasse <fasse.f2@gmail.com>
ok mpi@

Use the 0x10 extended capability to determine if the trackstick
buttons are wired to the trackpad and need to be re-routed to the
trackstick. Without this change the buttons on 2015 thinkpads
get picked up as extended buttons that show up as scroll up/down.

Remove the X1 Carbon 2015 (LEN0048) and X250 (LEN0046) from the
top button area/soft buttons quirks list. Also avoid using the
quirk list entirely if the capability bit is set.

Based on work from jcs@ with feedback from mpi@
ok mpi@ jcs@

Complete rewrite of the Elantech clickpads v4 packet parser in order to
support "click-and-drag".

This implements a simple filter to keep track of a moving finger on the
clickpad. For the moment wsmouse(4) and wscons(4) are still unaware of
multitouch events and pms(4) has to do the translation, but this is a
small steps in this direction.

All the good work has been done by Ulf Brosziewski.

Define a new wscons mouse type for Synaptics clickpad devices that lack
physical buttons. This will be used if the acpi pnpid for the mouse
matches a list provided by Synaptics found in Linux.

Combined with the xenocara changes this will remove the need for an xorg
config file for the x240/t440/t540 generation of thinkpads.

ok matthieu@ shadchin@ kettenis@

Only MS IntelliMouse have a Z axis, prevent a false positive triggered
by -Wuninitialized reported by jsg@.

ok stsp@

Fix checking sync for old synaptics touchpad (ver 5.9)
From Thierry Deval <thierry at deval dot be>

ok mpi@

Make sure state changes are properly serialized.

When pms(4) is attached to a touchpad it generally presents two different
wsmouse(4) devices: one for the touchpad itself and one for the clitpad
and/or some interleaved packets. But since both devices are writing to
the same pckbc slot, a race can occur if they try to change the state at
the same time.

So prevent two process opening the two /dev/wsmouse* node at the same time
to corrupt the magic sequences needed to enable/disable the touchpad.

ok schadchin@

add a size argument to free. will be used soon, but for now default to 0.
after discussions with beck deraadt kettenis.

Fix tail packet check in elantech v3 touchpad code. Due to a typo this
code was masking out bits which were also tested in the same expression.
Reported by cppcheck via jsg@ (Expression '(X & 0xfc) != 0x2' is always true).
Tested with Elantech Touchpad, version 3, firmware 0x250f00.

Do no reprobe for a supported protocol when enabling a pointing
device of type PMS_STANDARD, probing during autoconf(9) is enough!.

In such case, the device can be a simple PS/2 mouse, a unsupported
touchpad or a downgraded touchpad if something bad happened. But
it is very unlikely that reprobing will improve the situation and
transform a standard mouse into a multitouch-aware touchpad ;)

This fixes the 12 seconds delay seen on various Dell laptops (E4310
and Latitude D630) when starting Xorg, reported by K��rlis Mi��elsons
on bugs@.

ok stsp@, dcoppa@, shadchin@

Fix jagged diagonal lines (kernel part)

Send WSCONS_EVENT_SYNC every time you call wsmouse_input(). Used to synchronize
and separate events into packets of input data changes occurring at the same
moment in time. For example, motion of a mouse may set the DELTA_X and DELTA_Y
values for one motion, then emit a SYNC.

ok matthieu@. tested edd@, Henri Kemppainen and Alf Schlichting.

Use a working magic sequence to configure absolute mode for elantech v2.
The synaptics driver now attaches to elantech v2 touchpads. It seems these
pads were always falling back to compat mode, which went unnoticed until r1.47.

Fix based on a patch submitted by Cedric Tessier:
http://marc.info/?l=openbsd-bugs&m=136904396518528&w=2
which in turn was probably based on parts of this patch:
https://launchpadlibrarian.net/73099683/elantech.patch
which updates the Linux driver code but isn't part of Torvald's Linux.

Behaviour of elantech firmware versions seems to vary quite a bit so print
the firmware version at attach time for reference in future bug reports.

Fix tested by frantisek holop and Peter J. Philipp.
ok mpi@

When /dev/wsmouse is opened right after resume, some synaptics touchpads
are still taking time to wake up. So make the pms(4) driver wait a little
longer before giving up on them. Fixes mouse after resume on my x130e TP.
ok mpi, earlier versions tested by weerd with no regressions seen

Only free the per-protocol descriptor if a touchpad cannot be correctly
identified during attach. This prevents the driver to downgrade itself
to use the standard mouse protocol and reattach a new wsmouse child if
something bad happens when we re-query the hardware (during resume for
example).

Discussed with and ok stsp@

Correctly initialized the width value to 0 instead of passing garbage
to wsmouse_input() when no finger are reported.

Pointed out by Maxime Villard.

send proper z value to elantech_send_input in previous v4 commit

pointed out by Owain G. Ainsworth

add elantech v4 (clickpad) support
tested on an asus ux21a

ok mpi@ stsp@

Revert Active PS/2 support for now, until we find a solution suitable
for chipsets which stopped working.

ok mpi

Enable active PS/2 multiplexing if available.
Supported for i386 and amd64 except SMALL_KERNEL.

Based on Miod's former work on this subject.

ok mpi

Revert r1.37 now that we stop after the first matching protocol, tested
by Peter J. Philipp, thanks.

Move the IntelliMouse protocol definition after Elantech ones, because some
touchpads support both of them and we want to pick the latter.

Reported by Peter J. Philipp, thanks.

Don't probe for all supported protocols this can confuse some touchpads and
makes it harder to pick the right protocol if a device answers to more than
one magic sequence.

Tested by many on tech@, ok shadchin@

Once we've detected an elantech touchpad, stop probing for other hardware
versions. Fixes a problem reported by Peter J. Philipp where the pms driver
spits out "not in sync yet" messages after attaching to elantech v2 hardware.
Fix tested by me on v3 hardware, and by Peter on v2 hardware.
ok mpi@ shadchin@

Enable untested Elantech touchpad v1 and v2 code. I'd like some test reports.
ok mpi@ shadchin@

Fix memory leak in error paths for synaptics and alps (pointed stsp@).
Also small simplify elantech part.

ok stsp@, mpi@

Handle elantech touchpad buttons in a common code path for all
hardware variants. ok shadchin mikeb mpi

Don't OR flags into an uninitialized variable in elantech hardware v1 code.
Fix by oga, via mikeb.

Add support for Elantech touchpads to pms(4). This allows the X synaptics(4)
input driver to configure these touchpads properly, e.g. toggle tap-to-click,
two-finger scrolling, edge-scrolling, etc. So far these pads were working
only in PS/2 mouse compatibility mode.

Support for up to hardware version 3 has been implemented. But only version 3
has been tested so versions 1 and 2 remain disabled via #ifdef notyet until
someone tests (bugs in this code could seriously mess up the mouse and I
don't want to end up shipping such bugs in 5.3).

help and ok mpi@ shadchin@

Fix for ClickPad. On issue pointed gilles@ and matthieu@

ok gilles@, matthieu@, mpi@, miod@, deraadt@

Pass a width value corresponding to one finger for ALPS touchpads when the
reported pressure is non-null. Fix the use of ALPS touchpads with recent
(>1.6) xf86-input-synaptics drivers, issue reported by janis at cieti lv.

Tested by yasuoka@ and janis, ok shadchin@

Added yet another magic for my ALPS touchpad.

ok mpi@

Adds support for a slightly different version of the ALPS protocol,
which send 'interleaved' PS2 packets in between absolute ALPS packets.
Dell laptops (E6xxx and E5500 at least) have this kind of touchpad.

ok and some rework mpi@, tested okan@.

Blacklist a dualpoint model which clitpad sends non compatible PS2 packets.

Issue reported by deraadt@

renaming synaptics_pt_* -> pms_sec_*
(synaptics and alps uses this for trackpoint)

ok mpi@

Correctly handle clitpad packets for some ALPS models.

reported by espie@, not reported by miod@

with and ok shadchin@, ok espie@

Blacklist alps devices that might have a clitpad untill the driver get fixed

Fix the synchronization of some alps pads, issue reported by Bryan (brakeb)
on misc@

Similar diff tested by okan@ and Mathieu (ptr.jetable)

ok shadchin@

Add support for ALPS touchpads.

This also fix the "slow touchpad" problem due to the fact that some ALPS
devices respond to some Synaptics messages and stay in un half-configured
state.

Tested by Rivo Nurges, armani@, shadchin@ and matthieu@

ok shadchin@

revert the last commit. pms_enable_alps() does not guarantee
the accuracy detection ALPS.

ok mpi@

Adding stub for ALPS touchpad, which not supported yet.
This fixed "react strangely" when the pms driver probes
for a synaptic-compatible touchpad.

Tested and ok mpi@

Adding support Synaptics touchpad. Thanks all for test.

ok miod@, matthieu@

remove sc->sc_kbcslot, as is always PCKBC_AUX_SLOT
ok krw, miod

cleanup pms_enable_intelli: magic numbers -> defines
ok krw, miod

fix crash:
after boot and before the first activation (wsmoused or X),
sc->protocol is not initialized, and any call ioctl lead to crash
(example 'wsconsctl -a')

thanks and ok krw@

Add a common interface for various devices

This is to make it easier to add additional mice types in future

ok krw@, miod@, nicm@

Aesthetic improvements to make it appear more normal and knf'y.

No functional change. From Alexandr Scadchin again.

More cleanup. Add DEVNAME() macro and generic functions for various
actions (set/get/reset/enable/disable) to wrap pms_cmd() invocations.

No functional change. From Alexandr Scadchin again.

Minor code cleanups. No functional change. From Alexandr Shadchin.

ok nicm@

Unify some error messages to simplify reset code. No functional
change.

ok nicm@

From shadchin.

Ansify non-ansified function definitions. No functional change.

ok nicm@

From shadchin.

Remove some duplicated code. No need to flush/reset at attach and then
again at PMS_STATE_CHANGE events.

Tested my myself, ian@ and nicm@ without regressions on usual
problematic machines.

From shadchin.

Fix some pms(4) vs wsmoused issues. From Alexandr Shadchin via
tech@. Tested on the usual problematic machines by myself, nicm@
and ian@.

Fix Ian@'s pms keyboard/mouse while keeping my eeePC working.

For posterity: there is one time when pckbc commands must be thrown,
and one time when pckbc commands must be polled.

Diff from shadchin.

ok miod@

The attach function can return 1 now because there is no more priority
scheme relating to pmsi.
from alexandr.shadchin@gmail.com

Make sure the SEND_DEV_ID command is sent synchronously, so the response buffer
is valid when we look at it. Repairs intellimouse (former pmsi) operation.

Merge pmsi(4) into pms(4). From Alexandr Shadchin on tech@.

Toggle the mouse through the disabled and enabled states when suspending
and restoring. Fixes most mice.

use NULL to point to the response buffer, which avoids confusion; ok fgsch

Correct a 10-year old typo inherited from NetBSD by renaming psm.c,
psm_intelli.c and psmreg.h into pms.c, pms_intelli.c and pmsreg.h.

ok deraadt@

1. Work around a firmware bug of some Elantech-v1 touchpads.

2. Remove obsolete code from the v1-/v2-packet handlers (coordinate
values are ignored if z is 0).

Get Elantech v1 touchpad with fw version 0x20022 working.

From sxvghd (at) firemail (dot) cc, on tech@

Fix intermittent failing device initialization seen on some Synaptics
devices which has been re-introduced by pms.c revision 1.92.

ok tb@

1. Run Synaptics touchpads in plain W-mode. 2. Update the AGM-check.

Enabling extended W-mode is unnecessary as long as pms does not process
the additional packets properly.

Thanks to David Demelier for his reports and tests.

ok anton@ mpi@

Unbreak !DIAGNOSTIC build.

From Andrius V vezhlys (at) gmail, on bugs@

Extend the identification and validation of elantech-v4 packets to the
"crc-enabled" format.

ok awolk@

Accept more elantech-v4 models, and apply a test for external buttons.

Thanks to Alexander Cronheim.

ok mpi@

Wait a bit for the set mode command to complete for Synaptics devices to
make sure that the device doesn't get enabled before, which would make it
fail.
Fixes regular occurring 'not in sync yet, discard input' issue seen on
ThinkPad X1 Gen6.

ok deraadt@

Add support for Elantech trackpoints to pms. Extend the list of
"IC types" that identify Elantech-V4 touchpads.

Thanks to Ryan Lennox for help and testing.

ok mpi@

'juming' -> 'jumping' in comment.

Pointed out by Ryan Lennox via tech@

pms: minor cleanups in the alps code.

1. Update the ALPS_*_BEZEL values, which don't match the actual limits
accurately. The new values are used by Linux for all models with the
protocols "V1" and "V2".

2. Models with the ID 0x7331 are not supported properly, remove it
from the list.

3. Report to wsmouse that multiple contacts aren't recognized.

add the alps touchpad that can be found in the ideapad 710s

ok mpi@

Detect touchpad reset announcements. On rare occasions, the touchpad is reset
due to power failure and an announcement is transmitted as input by the device
making it detectable. At this point, the device must be re-enabled to continue
working since any configuration done prior to the reset is lost.

Detection is implemented using a watchdog and enabled for all protocols. So far,
the announcement has been observed on Synaptics by me and on Elantech v4 by
bru@. Extra care has been taken to handle potential false positives: when the
announcement byte sequence is part of a valid input packet.

with help and ok bru@, mpi@

Print out the offending data causing pms sync to fail.

ok bru@, mpi@ (as part of a larger diff)

Adapt the Elantech handlers (v1-3) to the new wsmouse infrastructure.

The new versions use wsmouse_configure, and leave compat-mode handling
to wsmouse.

ok stsp

bug fix: intialize variables in synaptics_get_hwinfo

Query the lower coordinate limits.

Show Synaptics Model IDs.

pms/alps: configure compat mode

Set the weak variant of hysteresis as default.

improved coordinate filters

ok mpi@

Use nitems() when passing all known parameters to wsmouse_configure().

ok bru@

pms/elantech-v4: configure wsmouse for handling compat-mode

ok @matthieu @stsp @mpi

pms/synaptics: configure wsmouse for handling compat-mode

ok @matthieu @stsp @mpi

Improved parameter handling in wsmouse and new ioctls for reading and
setting parameter values.

ok matthieu@

Use the new input functions of wsmouse in mouse and touchscreen drivers.

ok kettenis@

Add support for multitouch input to wsmouse.

This change adds new input-processing functions to wsmouse and
adapts the touchpad drivers.

ok mpi@, shadchin@

pakets -> packets

Support Synaptics touchpads without W mode.

Emulate W mode if a Synaptics model doesn't provide it, and check
for the success of the resolution query during setup.

ok shadchin@, mpi@

Improve tap-and-drag detection for ALPS touchpads.

The ALPS mechanism for gesture detection can produce long delays between
a packet that signals a tap and the next packet, which either signals
the end of the gesture or the start of a drag-operation. The time spans
can exceed the timeouts used by the generic detection mechanism in the
synaptics driver. Reporting both the touch and the release event when
the second packet has arrived ensures that the tap will be recognized.

ok mpi@ shadchin@

fairly trivial sizes for free()

Tweak previous; the Synaptics TrackPoint in my ThinkPad T500 works again.

ok krw@

Don't emit values for 'width' that

/usr/xenocara/driver/xf86-input-synaptics/src/wsconscomm.c

doesn't understand.

'Fixes' synaptics trackpad in Dell L400 laptop so the cursor can
be moved in X. Problem noted and experimental Dell L400 donated by
Paolo Aglialoro. Thanks!

ok miod@ mpi@ jcs@

pms(4): Don't match Elantech v4 devices with firmware versions 0xX7XXXX.
Apparently we don't support these touchpads properly so leave them in
PS/2 compat mode. Regression reported by Remi Locherer on bugs@. And stop
matching devices with firmware versions higher than 0xX8XXXX since we cannot
be sure they work.
ok mpi@

Match newer elantech v4 touchpads, logic taken from Linux.

Based on a submission from and ok jcs@.

Print irq informations in pckbc_set_inputhandler().

ok miod@

Add support for CRC-enabled elantech v3 touchpads.
Patch by Fasse <fasse.f2@gmail.com>
ok mpi@

Use the 0x10 extended capability to determine if the trackstick
buttons are wired to the trackpad and need to be re-routed to the
trackstick. Without this change the buttons on 2015 thinkpads
get picked up as extended buttons that show up as scroll up/down.

Remove the X1 Carbon 2015 (LEN0048) and X250 (LEN0046) from the
top button area/soft buttons quirks list. Also avoid using the
quirk list entirely if the capability bit is set.

Based on work from jcs@ with feedback from mpi@
ok mpi@ jcs@

Complete rewrite of the Elantech clickpads v4 packet parser in order to
support "click-and-drag".

This implements a simple filter to keep track of a moving finger on the
clickpad. For the moment wsmouse(4) and wscons(4) are still unaware of
multitouch events and pms(4) has to do the translation, but this is a
small steps in this direction.

All the good work has been done by Ulf Brosziewski.

Define a new wscons mouse type for Synaptics clickpad devices that lack
physical buttons. This will be used if the acpi pnpid for the mouse
matches a list provided by Synaptics found in Linux.

Combined with the xenocara changes this will remove the need for an xorg
config file for the x240/t440/t540 generation of thinkpads.

ok matthieu@ shadchin@ kettenis@

Only MS IntelliMouse have a Z axis, prevent a false positive triggered
by -Wuninitialized reported by jsg@.

ok stsp@

Fix checking sync for old synaptics touchpad (ver 5.9)
From Thierry Deval <thierry at deval dot be>

ok mpi@

Make sure state changes are properly serialized.

When pms(4) is attached to a touchpad it generally presents two different
wsmouse(4) devices: one for the touchpad itself and one for the clitpad
and/or some interleaved packets. But since both devices are writing to
the same pckbc slot, a race can occur if they try to change the state at
the same time.

So prevent two process opening the two /dev/wsmouse* node at the same time
to corrupt the magic sequences needed to enable/disable the touchpad.

ok schadchin@

add a size argument to free. will be used soon, but for now default to 0.
after discussions with beck deraadt kettenis.

Fix tail packet check in elantech v3 touchpad code. Due to a typo this
code was masking out bits which were also tested in the same expression.
Reported by cppcheck via jsg@ (Expression '(X & 0xfc) != 0x2' is always true).
Tested with Elantech Touchpad, version 3, firmware 0x250f00.

Do no reprobe for a supported protocol when enabling a pointing
device of type PMS_STANDARD, probing during autoconf(9) is enough!.

In such case, the device can be a simple PS/2 mouse, a unsupported
touchpad or a downgraded touchpad if something bad happened. But
it is very unlikely that reprobing will improve the situation and
transform a standard mouse into a multitouch-aware touchpad ;)

This fixes the 12 seconds delay seen on various Dell laptops (E4310
and Latitude D630) when starting Xorg, reported by K��rlis Mi��elsons
on bugs@.

ok stsp@, dcoppa@, shadchin@

Fix jagged diagonal lines (kernel part)

Send WSCONS_EVENT_SYNC every time you call wsmouse_input(). Used to synchronize
and separate events into packets of input data changes occurring at the same
moment in time. For example, motion of a mouse may set the DELTA_X and DELTA_Y
values for one motion, then emit a SYNC.

ok matthieu@. tested edd@, Henri Kemppainen and Alf Schlichting.

Use a working magic sequence to configure absolute mode for elantech v2.
The synaptics driver now attaches to elantech v2 touchpads. It seems these
pads were always falling back to compat mode, which went unnoticed until r1.47.

Fix based on a patch submitted by Cedric Tessier:
http://marc.info/?l=openbsd-bugs&m=136904396518528&w=2
which in turn was probably based on parts of this patch:
https://launchpadlibrarian.net/73099683/elantech.patch
which updates the Linux driver code but isn't part of Torvald's Linux.

Behaviour of elantech firmware versions seems to vary quite a bit so print
the firmware version at attach time for reference in future bug reports.

Fix tested by frantisek holop and Peter J. Philipp.
ok mpi@

When /dev/wsmouse is opened right after resume, some synaptics touchpads
are still taking time to wake up. So make the pms(4) driver wait a little
longer before giving up on them. Fixes mouse after resume on my x130e TP.
ok mpi, earlier versions tested by weerd with no regressions seen

Only free the per-protocol descriptor if a touchpad cannot be correctly
identified during attach. This prevents the driver to downgrade itself
to use the standard mouse protocol and reattach a new wsmouse child if
something bad happens when we re-query the hardware (during resume for
example).

Discussed with and ok stsp@

Correctly initialized the width value to 0 instead of passing garbage
to wsmouse_input() when no finger are reported.

Pointed out by Maxime Villard.

send proper z value to elantech_send_input in previous v4 commit

pointed out by Owain G. Ainsworth

add elantech v4 (clickpad) support
tested on an asus ux21a

ok mpi@ stsp@

Revert Active PS/2 support for now, until we find a solution suitable
for chipsets which stopped working.

ok mpi

Enable active PS/2 multiplexing if available.
Supported for i386 and amd64 except SMALL_KERNEL.

Based on Miod's former work on this subject.

ok mpi

Revert r1.37 now that we stop after the first matching protocol, tested
by Peter J. Philipp, thanks.

Move the IntelliMouse protocol definition after Elantech ones, because some
touchpads support both of them and we want to pick the latter.

Reported by Peter J. Philipp, thanks.

Don't probe for all supported protocols this can confuse some touchpads and
makes it harder to pick the right protocol if a device answers to more than
one magic sequence.

Tested by many on tech@, ok shadchin@

Once we've detected an elantech touchpad, stop probing for other hardware
versions. Fixes a problem reported by Peter J. Philipp where the pms driver
spits out "not in sync yet" messages after attaching to elantech v2 hardware.
Fix tested by me on v3 hardware, and by Peter on v2 hardware.
ok mpi@ shadchin@

Enable untested Elantech touchpad v1 and v2 code. I'd like some test reports.
ok mpi@ shadchin@

Fix memory leak in error paths for synaptics and alps (pointed stsp@).
Also small simplify elantech part.

ok stsp@, mpi@

Handle elantech touchpad buttons in a common code path for all
hardware variants. ok shadchin mikeb mpi

Don't OR flags into an uninitialized variable in elantech hardware v1 code.
Fix by oga, via mikeb.

Add support for Elantech touchpads to pms(4). This allows the X synaptics(4)
input driver to configure these touchpads properly, e.g. toggle tap-to-click,
two-finger scrolling, edge-scrolling, etc. So far these pads were working
only in PS/2 mouse compatibility mode.

Support for up to hardware version 3 has been implemented. But only version 3
has been tested so versions 1 and 2 remain disabled via #ifdef notyet until
someone tests (bugs in this code could seriously mess up the mouse and I
don't want to end up shipping such bugs in 5.3).

help and ok mpi@ shadchin@

Fix for ClickPad. On issue pointed gilles@ and matthieu@

ok gilles@, matthieu@, mpi@, miod@, deraadt@

Pass a width value corresponding to one finger for ALPS touchpads when the
reported pressure is non-null. Fix the use of ALPS touchpads with recent
(>1.6) xf86-input-synaptics drivers, issue reported by janis at cieti lv.

Tested by yasuoka@ and janis, ok shadchin@

Added yet another magic for my ALPS touchpad.

ok mpi@

Adds support for a slightly different version of the ALPS protocol,
which send 'interleaved' PS2 packets in between absolute ALPS packets.
Dell laptops (E6xxx and E5500 at least) have this kind of touchpad.

ok and some rework mpi@, tested okan@.

Blacklist a dualpoint model which clitpad sends non compatible PS2 packets.

Issue reported by deraadt@

renaming synaptics_pt_* -> pms_sec_*
(synaptics and alps uses this for trackpoint)

ok mpi@

Correctly handle clitpad packets for some ALPS models.

reported by espie@, not reported by miod@

with and ok shadchin@, ok espie@

Blacklist alps devices that might have a clitpad untill the driver get fixed

Fix the synchronization of some alps pads, issue reported by Bryan (brakeb)
on misc@

Similar diff tested by okan@ and Mathieu (ptr.jetable)

ok shadchin@

Add support for ALPS touchpads.

This also fix the "slow touchpad" problem due to the fact that some ALPS
devices respond to some Synaptics messages and stay in un half-configured
state.

Tested by Rivo Nurges, armani@, shadchin@ and matthieu@

ok shadchin@

revert the last commit. pms_enable_alps() does not guarantee
the accuracy detection ALPS.

ok mpi@

Adding stub for ALPS touchpad, which not supported yet.
This fixed "react strangely" when the pms driver probes
for a synaptic-compatible touchpad.

Tested and ok mpi@

Adding support Synaptics touchpad. Thanks all for test.

ok miod@, matthieu@

remove sc->sc_kbcslot, as is always PCKBC_AUX_SLOT
ok krw, miod

cleanup pms_enable_intelli: magic numbers -> defines
ok krw, miod

fix crash:
after boot and before the first activation (wsmoused or X),
sc->protocol is not initialized, and any call ioctl lead to crash
(example 'wsconsctl -a')

thanks and ok krw@

Add a common interface for various devices

This is to make it easier to add additional mice types in future

ok krw@, miod@, nicm@

Aesthetic improvements to make it appear more normal and knf'y.

No functional change. From Alexandr Scadchin again.

More cleanup. Add DEVNAME() macro and generic functions for various
actions (set/get/reset/enable/disable) to wrap pms_cmd() invocations.

No functional change. From Alexandr Scadchin again.

Minor code cleanups. No functional change. From Alexandr Shadchin.

ok nicm@

Unify some error messages to simplify reset code. No functional
change.

ok nicm@

From shadchin.

Ansify non-ansified function definitions. No functional change.

ok nicm@

From shadchin.

Remove some duplicated code. No need to flush/reset at attach and then
again at PMS_STATE_CHANGE events.

Tested my myself, ian@ and nicm@ without regressions on usual
problematic machines.

From shadchin.

Fix some pms(4) vs wsmoused issues. From Alexandr Shadchin via
tech@. Tested on the usual problematic machines by myself, nicm@
and ian@.

Fix Ian@'s pms keyboard/mouse while keeping my eeePC working.

For posterity: there is one time when pckbc commands must be thrown,
and one time when pckbc commands must be polled.

Diff from shadchin.

ok miod@

The attach function can return 1 now because there is no more priority
scheme relating to pmsi.
from alexandr.shadchin@gmail.com

Make sure the SEND_DEV_ID command is sent synchronously, so the response buffer
is valid when we look at it. Repairs intellimouse (former pmsi) operation.

Merge pmsi(4) into pms(4). From Alexandr Shadchin on tech@.

Toggle the mouse through the disabled and enabled states when suspending
and restoring. Fixes most mice.

use NULL to point to the response buffer, which avoids confusion; ok fgsch

Correct a 10-year old typo inherited from NetBSD by renaming psm.c,
psm_intelli.c and psmreg.h into pms.c, pms_intelli.c and pmsreg.h.

ok deraadt@

Get Elantech v1 touchpad with fw version 0x20022 working.

From sxvghd (at) firemail (dot) cc, on tech@

Fix intermittent failing device initialization seen on some Synaptics
devices which has been re-introduced by pms.c revision 1.92.

ok tb@

1. Run Synaptics touchpads in plain W-mode. 2. Update the AGM-check.

Enabling extended W-mode is unnecessary as long as pms does not process
the additional packets properly.

Thanks to David Demelier for his reports and tests.

ok anton@ mpi@

Unbreak !DIAGNOSTIC build.

From Andrius V vezhlys (at) gmail, on bugs@

Extend the identification and validation of elantech-v4 packets to the
"crc-enabled" format.

ok awolk@

Accept more elantech-v4 models, and apply a test for external buttons.

Thanks to Alexander Cronheim.

ok mpi@

Wait a bit for the set mode command to complete for Synaptics devices to
make sure that the device doesn't get enabled before, which would make it
fail.
Fixes regular occurring 'not in sync yet, discard input' issue seen on
ThinkPad X1 Gen6.

ok deraadt@

Add support for Elantech trackpoints to pms. Extend the list of
"IC types" that identify Elantech-V4 touchpads.

Thanks to Ryan Lennox for help and testing.

ok mpi@

'juming' -> 'jumping' in comment.

Pointed out by Ryan Lennox via tech@

pms: minor cleanups in the alps code.

1. Update the ALPS_*_BEZEL values, which don't match the actual limits
accurately. The new values are used by Linux for all models with the
protocols "V1" and "V2".

2. Models with the ID 0x7331 are not supported properly, remove it
from the list.

3. Report to wsmouse that multiple contacts aren't recognized.

add the alps touchpad that can be found in the ideapad 710s

ok mpi@

Detect touchpad reset announcements. On rare occasions, the touchpad is reset
due to power failure and an announcement is transmitted as input by the device
making it detectable. At this point, the device must be re-enabled to continue
working since any configuration done prior to the reset is lost.

Detection is implemented using a watchdog and enabled for all protocols. So far,
the announcement has been observed on Synaptics by me and on Elantech v4 by
bru@. Extra care has been taken to handle potential false positives: when the
announcement byte sequence is part of a valid input packet.

with help and ok bru@, mpi@

Print out the offending data causing pms sync to fail.

ok bru@, mpi@ (as part of a larger diff)

Adapt the Elantech handlers (v1-3) to the new wsmouse infrastructure.

The new versions use wsmouse_configure, and leave compat-mode handling
to wsmouse.

ok stsp

bug fix: intialize variables in synaptics_get_hwinfo

Query the lower coordinate limits.

Show Synaptics Model IDs.

pms/alps: configure compat mode

Set the weak variant of hysteresis as default.