[patch 00/24] Input patches for 2.6.13
Hi Vojtech, I have a bunch of changes for input subsystem I would like to get in. Half of the stuff is patches I picked from your quilt tree (ALPS dualpoint resume fix, ALPS tapping mode, couple HID changes, etc) plus some of my changes (usb_to_input_id, small acecad cleanup, add modalias for serio bus, couple of input cleanups). The full changelog is at the end of the mail. Could you please "bless" these for Linus to pull: git pull rsync://rsync.kernel.org/pub/scm/linux/kernel/git/dtor/input.git Thanks! -- Dmitry Changelog: == author Sergey Vlasov <[EMAIL PROTECTED]> Sun, 24 Jul 2005 10:53:32 -0500 Input: synaptics - fix setting packet size on passthrough port. Synaptics driver used child->type to select either 3-byte or 4-byte packet size for the pass-through port; this gives wrong results for the newer protocols. Change the check to use child->pktsize instead. Signed-off-by: Sergey Vlasov <[EMAIL PROTECTED]> Signed-off-by: Dmitry Torokhov <[EMAIL PROTECTED]> -- author Vojtech Pavlik <[EMAIL PROTECTED]> Sun, 24 Jul 2005 10:50:03 -0500 Input: check keycodesize when adjusting keymaps When changing key mappings we need to make sure that the new keycode value can be stored in dev->keycodesize bytes. Signed-off-by: Vojtech Pavlik <[EMAIL PROTECTED]> Signed-off-by: Dmitry Torokhov <[EMAIL PROTECTED]> -- author Dmitry Torokhov <[EMAIL PROTECTED]> Fri, 15 Jul 2005 11:51:56 -0500 Input: i8042 - don't use negation to mark AUX data Currently i8042_command() negates data coming from the AUX port of keyboard controller; this is not a very reliable indicator. Change i8042_command() to fail if response to I8042_CMD_AUX_LOOP is not coming from AUX channel and get rid of negation. Based on patch by Vojtech Pavlik. Signed-off-by: Dmitry Torokhov <[EMAIL PROTECTED]> -- author Vojtech Pavlik <[EMAIL PROTECTED]> Fri, 15 Jul 2005 11:50:08 -0500 Input: psmouse - wheel mice (imps, exps) always have 3rd button There are wheel mice that respond to Logitech probes and report that they have only 2 buttons (such as e-Aser mouse) and this stops the wheel from being used as a middle button. Change the driver to always report BTN_MIDDLE capability if a wheel is present. Also, never reset BTN_RIGHT capability in logips2pp code - there are no Logitech mice that have only one button and if some other mice happen to respond to Logitech's query we could do the wrong thing. Signed-off-by: Vojtech Pavlik <[EMAIL PROTECTED]> Signed-off-by: Dmitry Torokhov <[EMAIL PROTECTED]> -- author Adam Kropelin <[EMAIL PROTECTED]> Mon, 11 Jul 2005 11:09:32 -0500 Input: HID - only report events coming from interrupts to hiddev Currently hid-core follows the same code path for input reports regardless of whether they are a result of interrupt transfers or control transfers. That leads to interrupt events erroneously being reported to hiddev for regular control transfers. Prior to 2.6.12 the problem was mitigated by the fact that reporting to hiddev is supressed if the field value has not changed, which is often the case. Said filtering was removed in 2.6.12-rc1 which means any input reports fetched via control transfers result in hiddev interrupt events. This behavior can quickly lead to a feedback loop where a userspace app, in response to interrupt events, issues control transfers which in turn create more interrupt events. This patch prevents input reports that arrive via control transfers from being reported to hiddev as interrupt events. Signed-off-by: Adam Kropelin <[EMAIL PROTECTED]> Signed-off-by: Vojtech Pavlik <[EMAIL PROTECTED]> Signed-off-by: Dmitry Torokhov <[EMAIL PROTECTED]> -- author Michael Prokop <[EMAIL PROTECTED]> Mon, 11 Jul 2005 11:09:10 -0500 Input: elo - fix help in Kconfig (wrong module name) Signed-off-by: Vojtech Pavlik <[EMAIL PROTECTED]> Signed-off-by: Dmitry Torokhov <[EMAIL PROTECTED]> -- author Tobias Klauser <[EMAIL PROTECTED]> Mon, 11 Jul 2005 11:08:56 -0500 Input: joydev - remove custom conversion from jiffies to msecs Replace the MSECS() macro with the jiffies_to_msecs() function provided in jiffies.h Signed-off-by: Tobias Klauser <[EMAIL PROTECTED]> Signed-off-by: Domen Puncer <[EMAIL PROTECTED]> Signed-off-by: Vojtech Pavlik <[EMAIL PROTECTED]> Signed-off-by: Dmitry Torokhov <[EMAIL PROTECTED]> -- author Luca T <[EMAIL PROTECTED]> Mon, 11 Jul 2005 11:08:40 -0500 Input: HID - add a quirk for Aashima Trust (06d6:0025) gamepad Signed-off-by: Vojtech Pavlik <[EMAIL PROTECTED]> Signed-off-by: Dmitry Torokhov <[EMAIL PROTECTED]> -- author Peter Osterlund <[EMAIL PROTECTED]> Mon, 11 Jul 2005 11:08:04 -0500 Input: ALPS - unconditionally enable tapping mode The condition in alps_init() was also inverted and the driver was enabling tapping mode only if it was already enabled.
[patch 00/24] Input patches for 2.6.13
Hi Vojtech, I have a bunch of changes for input subsystem I would like to get in. Half of the stuff is patches I picked from your quilt tree (ALPS dualpoint resume fix, ALPS tapping mode, couple HID changes, etc) plus some of my changes (usb_to_input_id, small acecad cleanup, add modalias for serio bus, couple of input cleanups). The full changelog is at the end of the mail. Could you please bless these for Linus to pull: git pull rsync://rsync.kernel.org/pub/scm/linux/kernel/git/dtor/input.git Thanks! -- Dmitry Changelog: == author Sergey Vlasov [EMAIL PROTECTED] Sun, 24 Jul 2005 10:53:32 -0500 Input: synaptics - fix setting packet size on passthrough port. Synaptics driver used child-type to select either 3-byte or 4-byte packet size for the pass-through port; this gives wrong results for the newer protocols. Change the check to use child-pktsize instead. Signed-off-by: Sergey Vlasov [EMAIL PROTECTED] Signed-off-by: Dmitry Torokhov [EMAIL PROTECTED] -- author Vojtech Pavlik [EMAIL PROTECTED] Sun, 24 Jul 2005 10:50:03 -0500 Input: check keycodesize when adjusting keymaps When changing key mappings we need to make sure that the new keycode value can be stored in dev-keycodesize bytes. Signed-off-by: Vojtech Pavlik [EMAIL PROTECTED] Signed-off-by: Dmitry Torokhov [EMAIL PROTECTED] -- author Dmitry Torokhov [EMAIL PROTECTED] Fri, 15 Jul 2005 11:51:56 -0500 Input: i8042 - don't use negation to mark AUX data Currently i8042_command() negates data coming from the AUX port of keyboard controller; this is not a very reliable indicator. Change i8042_command() to fail if response to I8042_CMD_AUX_LOOP is not coming from AUX channel and get rid of negation. Based on patch by Vojtech Pavlik. Signed-off-by: Dmitry Torokhov [EMAIL PROTECTED] -- author Vojtech Pavlik [EMAIL PROTECTED] Fri, 15 Jul 2005 11:50:08 -0500 Input: psmouse - wheel mice (imps, exps) always have 3rd button There are wheel mice that respond to Logitech probes and report that they have only 2 buttons (such as e-Aser mouse) and this stops the wheel from being used as a middle button. Change the driver to always report BTN_MIDDLE capability if a wheel is present. Also, never reset BTN_RIGHT capability in logips2pp code - there are no Logitech mice that have only one button and if some other mice happen to respond to Logitech's query we could do the wrong thing. Signed-off-by: Vojtech Pavlik [EMAIL PROTECTED] Signed-off-by: Dmitry Torokhov [EMAIL PROTECTED] -- author Adam Kropelin [EMAIL PROTECTED] Mon, 11 Jul 2005 11:09:32 -0500 Input: HID - only report events coming from interrupts to hiddev Currently hid-core follows the same code path for input reports regardless of whether they are a result of interrupt transfers or control transfers. That leads to interrupt events erroneously being reported to hiddev for regular control transfers. Prior to 2.6.12 the problem was mitigated by the fact that reporting to hiddev is supressed if the field value has not changed, which is often the case. Said filtering was removed in 2.6.12-rc1 which means any input reports fetched via control transfers result in hiddev interrupt events. This behavior can quickly lead to a feedback loop where a userspace app, in response to interrupt events, issues control transfers which in turn create more interrupt events. This patch prevents input reports that arrive via control transfers from being reported to hiddev as interrupt events. Signed-off-by: Adam Kropelin [EMAIL PROTECTED] Signed-off-by: Vojtech Pavlik [EMAIL PROTECTED] Signed-off-by: Dmitry Torokhov [EMAIL PROTECTED] -- author Michael Prokop [EMAIL PROTECTED] Mon, 11 Jul 2005 11:09:10 -0500 Input: elo - fix help in Kconfig (wrong module name) Signed-off-by: Vojtech Pavlik [EMAIL PROTECTED] Signed-off-by: Dmitry Torokhov [EMAIL PROTECTED] -- author Tobias Klauser [EMAIL PROTECTED] Mon, 11 Jul 2005 11:08:56 -0500 Input: joydev - remove custom conversion from jiffies to msecs Replace the MSECS() macro with the jiffies_to_msecs() function provided in jiffies.h Signed-off-by: Tobias Klauser [EMAIL PROTECTED] Signed-off-by: Domen Puncer [EMAIL PROTECTED] Signed-off-by: Vojtech Pavlik [EMAIL PROTECTED] Signed-off-by: Dmitry Torokhov [EMAIL PROTECTED] -- author Luca T [EMAIL PROTECTED] Mon, 11 Jul 2005 11:08:40 -0500 Input: HID - add a quirk for Aashima Trust (06d6:0025) gamepad Signed-off-by: Vojtech Pavlik [EMAIL PROTECTED] Signed-off-by: Dmitry Torokhov [EMAIL PROTECTED] -- author Peter Osterlund [EMAIL PROTECTED] Mon, 11 Jul 2005 11:08:04 -0500 Input: ALPS - unconditionally enable tapping mode The condition in alps_init() was also inverted and the driver was enabling tapping mode only if it was already enabled. Signed-off-by: Peter Osterlund [EMAIL PROTECTED] Signed-off-by:
