[PATCH] mouse button #1 emulation for mac_hid

[PATCH] mouse button #1 emulation for mac_hid

[Soeren Sonnenburg]

[-- Attachment #1: Type: text/plain, Size: 605 bytes --]

Dear all,

I am not sure what the future of mac_hid is (the source says will be
removed, but since 9 years). In case it stays in kernel the attached
patch adds support for mouse button 1 emulation. If this driver stays in
the kernel I would also suggest to rename it - it is not mac specific (I
am using this on a netbook).

Rationale? Broken touchpads, within linux unsupported (multi-)touch pads
where click and hold + moving the mouse does not work.

Soeren
-- 
For the one fact about the future of which we can be certain is that it
will be utterly fantastic. -- Arthur C. Clarke, 1962

[-- Attachment #2: mac_hid_lmb.diff --]
[-- Type: text/x-patch, Size: 3294 bytes --]

diff --git a/drivers/macintosh/mac_hid.c b/drivers/macintosh/mac_hid.c
index cc9f275..3aac90f 100644
--- a/drivers/macintosh/mac_hid.c
+++ b/drivers/macintosh/mac_hid.c
@@ -19,6 +19,7 @@
 static struct input_dev *emumousebtn;
 static int emumousebtn_input_register(void);
 static int mouse_emulate_buttons;
+static int mouse_button1_keycode = KEY_RIGHTSHIFT;	/* right shift key */
 static int mouse_button2_keycode = KEY_RIGHTCTRL;	/* right control key */
 static int mouse_button3_keycode = KEY_RIGHTALT;	/* right option key */
 static int mouse_last_keycode;
@@ -35,6 +36,14 @@ static ctl_table mac_hid_files[] = {
 		.proc_handler	= &proc_dointvec,
 	},
 	{
+		.ctl_name	= DEV_MAC_HID_MOUSE_BUTTON1_KEYCODE,
+		.procname	= "mouse_button1_keycode",
+		.data		= &mouse_button1_keycode,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec,
+	},
+	{
 		.ctl_name	= DEV_MAC_HID_MOUSE_BUTTON2_KEYCODE,
 		.procname	= "mouse_button2_keycode",
 		.data		= &mouse_button2_keycode,
@@ -83,16 +92,24 @@ static struct ctl_table_header *mac_hid_sysctl_header;
 
 int mac_hid_mouse_emulate_buttons(int caller, unsigned int keycode, int down)
 {
+	unsigned int reportcode;
+
 	switch (caller) {
 	case 1:
 		/* Called from keyboard.c */
 		if (mouse_emulate_buttons
-		    && (keycode == mouse_button2_keycode
-			|| keycode == mouse_button3_keycode)) {
+		    && (keycode == mouse_button1_keycode ||
+				keycode == mouse_button2_keycode ||
+				keycode == mouse_button3_keycode)) {
 			if (mouse_emulate_buttons == 1) {
-				input_report_key(emumousebtn,
-						 keycode == mouse_button2_keycode ? BTN_MIDDLE : BTN_RIGHT,
-						 down);
+				if (keycode==mouse_button1_keycode)
+					reportcode=BTN_LEFT;
+				else if (keycode==mouse_button2_keycode)
+					reportcode=BTN_MIDDLE;
+				else
+					reportcode=BTN_RIGHT;
+
+				input_report_key(emumousebtn, reportcode, down);
 				input_sync(emumousebtn);
 				return 1;
 			}
diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h
index 1e4743e..d38f5a0 100644
--- a/include/linux/sysctl.h
+++ b/include/linux/sysctl.h
@@ -913,9 +913,10 @@ enum {
 	DEV_MAC_HID_KEYBOARD_SENDS_LINUX_KEYCODES=1,
 	DEV_MAC_HID_KEYBOARD_LOCK_KEYCODES=2,
 	DEV_MAC_HID_MOUSE_BUTTON_EMULATION=3,
-	DEV_MAC_HID_MOUSE_BUTTON2_KEYCODE=4,
-	DEV_MAC_HID_MOUSE_BUTTON3_KEYCODE=5,
-	DEV_MAC_HID_ADB_MOUSE_SENDS_KEYCODES=6
+	DEV_MAC_HID_MOUSE_BUTTON1_KEYCODE=4,
+	DEV_MAC_HID_MOUSE_BUTTON2_KEYCODE=5,
+	DEV_MAC_HID_MOUSE_BUTTON3_KEYCODE=6,
+	DEV_MAC_HID_ADB_MOUSE_SENDS_KEYCODES=7
 };
 
 /* /proc/sys/dev/scsi */
diff --git a/kernel/sysctl_check.c b/kernel/sysctl_check.c
index b6e7aae..60fa7fe 100644
--- a/kernel/sysctl_check.c
+++ b/kernel/sysctl_check.c
@@ -816,6 +816,7 @@ static const struct trans_ctl_table trans_mac_hid_files[] = {
 	/* DEV_MAC_HID_KEYBOARD_SENDS_LINUX_KEYCODES unused */
 	/* DEV_MAC_HID_KEYBOARD_LOCK_KEYCODES unused */
 	{ DEV_MAC_HID_MOUSE_BUTTON_EMULATION,	"mouse_button_emulation" },
+	{ DEV_MAC_HID_MOUSE_BUTTON1_KEYCODE,	"mouse_button1_keycode" },
 	{ DEV_MAC_HID_MOUSE_BUTTON2_KEYCODE,	"mouse_button2_keycode" },
 	{ DEV_MAC_HID_MOUSE_BUTTON3_KEYCODE,	"mouse_button3_keycode" },
 	/* DEV_MAC_HID_ADB_MOUSE_SENDS_KEYCODES unused */

[PATCH] mouse button 1 emulation for mac_hid

[Soeren Sonnenburg]

[-- Attachment #1.1: Type: text/plain, Size: 810 bytes --]

(this is a resent with all the people included in the CC that git
         blame'd they are the authors).

Dear all,

I am not sure what the future of mac_hid is (the source says will be
removed, but since 9 years). In case it stays in kernel the attached
patch adds support for mouse button 1 emulation. If this driver stays in
the kernel I would also suggest to rename it - it is not mac specific (I
am using this on a netbook).

Rationale? Broken touchpads, within linux unsupported (multi-)touchpads
where click and hold + moving the mouse does not work (yet?).

I am using this for over a week now with about that uptime - so it
should be safe.

Soeren
-- 
For the one fact about the future of which we can be certain is that it
will be utterly fantastic. -- Arthur C. Clarke, 1962


[-- Attachment #1.2: mac_hid_lmb.diff --]
[-- Type: text/x-patch, Size: 3294 bytes --]

diff --git a/drivers/macintosh/mac_hid.c b/drivers/macintosh/mac_hid.c
index cc9f275..3aac90f 100644
--- a/drivers/macintosh/mac_hid.c
+++ b/drivers/macintosh/mac_hid.c
@@ -19,6 +19,7 @@
 static struct input_dev *emumousebtn;
 static int emumousebtn_input_register(void);
 static int mouse_emulate_buttons;
+static int mouse_button1_keycode = KEY_RIGHTSHIFT;	/* right shift key */
 static int mouse_button2_keycode = KEY_RIGHTCTRL;	/* right control key */
 static int mouse_button3_keycode = KEY_RIGHTALT;	/* right option key */
 static int mouse_last_keycode;
@@ -35,6 +36,14 @@ static ctl_table mac_hid_files[] = {
 		.proc_handler	= &proc_dointvec,
 	},
 	{
+		.ctl_name	= DEV_MAC_HID_MOUSE_BUTTON1_KEYCODE,
+		.procname	= "mouse_button1_keycode",
+		.data		= &mouse_button1_keycode,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec,
+	},
+	{
 		.ctl_name	= DEV_MAC_HID_MOUSE_BUTTON2_KEYCODE,
 		.procname	= "mouse_button2_keycode",
 		.data		= &mouse_button2_keycode,
@@ -83,16 +92,24 @@ static struct ctl_table_header *mac_hid_sysctl_header;
 
 int mac_hid_mouse_emulate_buttons(int caller, unsigned int keycode, int down)
 {
+	unsigned int reportcode;
+
 	switch (caller) {
 	case 1:
 		/* Called from keyboard.c */
 		if (mouse_emulate_buttons
-		    && (keycode == mouse_button2_keycode
-			|| keycode == mouse_button3_keycode)) {
+		    && (keycode == mouse_button1_keycode ||
+				keycode == mouse_button2_keycode ||
+				keycode == mouse_button3_keycode)) {
 			if (mouse_emulate_buttons == 1) {
-				input_report_key(emumousebtn,
-						 keycode == mouse_button2_keycode ? BTN_MIDDLE : BTN_RIGHT,
-						 down);
+				if (keycode==mouse_button1_keycode)
+					reportcode=BTN_LEFT;
+				else if (keycode==mouse_button2_keycode)
+					reportcode=BTN_MIDDLE;
+				else
+					reportcode=BTN_RIGHT;
+
+				input_report_key(emumousebtn, reportcode, down);
 				input_sync(emumousebtn);
 				return 1;
 			}
diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h
index 1e4743e..d38f5a0 100644
--- a/include/linux/sysctl.h
+++ b/include/linux/sysctl.h
@@ -913,9 +913,10 @@ enum {
 	DEV_MAC_HID_KEYBOARD_SENDS_LINUX_KEYCODES=1,
 	DEV_MAC_HID_KEYBOARD_LOCK_KEYCODES=2,
 	DEV_MAC_HID_MOUSE_BUTTON_EMULATION=3,
-	DEV_MAC_HID_MOUSE_BUTTON2_KEYCODE=4,
-	DEV_MAC_HID_MOUSE_BUTTON3_KEYCODE=5,
-	DEV_MAC_HID_ADB_MOUSE_SENDS_KEYCODES=6
+	DEV_MAC_HID_MOUSE_BUTTON1_KEYCODE=4,
+	DEV_MAC_HID_MOUSE_BUTTON2_KEYCODE=5,
+	DEV_MAC_HID_MOUSE_BUTTON3_KEYCODE=6,
+	DEV_MAC_HID_ADB_MOUSE_SENDS_KEYCODES=7
 };
 
 /* /proc/sys/dev/scsi */
diff --git a/kernel/sysctl_check.c b/kernel/sysctl_check.c
index b6e7aae..60fa7fe 100644
--- a/kernel/sysctl_check.c
+++ b/kernel/sysctl_check.c
@@ -816,6 +816,7 @@ static const struct trans_ctl_table trans_mac_hid_files[] = {
 	/* DEV_MAC_HID_KEYBOARD_SENDS_LINUX_KEYCODES unused */
 	/* DEV_MAC_HID_KEYBOARD_LOCK_KEYCODES unused */
 	{ DEV_MAC_HID_MOUSE_BUTTON_EMULATION,	"mouse_button_emulation" },
+	{ DEV_MAC_HID_MOUSE_BUTTON1_KEYCODE,	"mouse_button1_keycode" },
 	{ DEV_MAC_HID_MOUSE_BUTTON2_KEYCODE,	"mouse_button2_keycode" },
 	{ DEV_MAC_HID_MOUSE_BUTTON3_KEYCODE,	"mouse_button3_keycode" },
 	/* DEV_MAC_HID_ADB_MOUSE_SENDS_KEYCODES unused */

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

Re: [PATCH] mouse button 1 emulation for mac_hid

[Dmitry Torokhov]

Hi Soeren,

On Mon, Nov 09, 2009 at 11:28:51AM +0100, Soeren Sonnenburg wrote:
> (this is a resent with all the people included in the CC that git
>          blame'd they are the authors).
> 
> Dear all,
> 
> I am not sure what the future of mac_hid is (the source says will be
> removed, but since 9 years). In case it stays in kernel the attached
> patch adds support for mouse button 1 emulation. If this driver stays in
> the kernel I would also suggest to rename it - it is not mac specific (I
> am using this on a netbook).
> 
> Rationale? Broken touchpads, within linux unsupported (multi-)touchpads
> where click and hold + moving the mouse does not work (yet?).
> 

I would rather fix the touchpads so they work properly than pile more
quirks in kernel.

Thanks.

-- 
Dmitry

Re: [PATCH] mouse button 1 emulation for mac_hid

[Soeren Sonnenburg]

[-- Attachment #1: Type: text/plain, Size: 1380 bytes --]

On Fri, 2009-11-13 at 21:42 -0800, Dmitry Torokhov wrote:
> Hi Soeren,

Hi Dmitry,

> On Mon, Nov 09, 2009 at 11:28:51AM +0100, Soeren Sonnenburg wrote:
> > (this is a resent with all the people included in the CC that git
> >          blame'd they are the authors).
> > 
> > Dear all,
> > 
> > I am not sure what the future of mac_hid is (the source says will be
> > removed, but since 9 years). In case it stays in kernel the attached
> > patch adds support for mouse button 1 emulation. If this driver stays in
> > the kernel I would also suggest to rename it - it is not mac specific (I
> > am using this on a netbook).
> > 
> > Rationale? Broken touchpads, within linux unsupported (multi-)touchpads
> > where click and hold + moving the mouse does not work (yet?).
> > 
> 
> I would rather fix the touchpads so they work properly than pile more
> quirks in kernel.

Well I guess this whole mouse button emulation should not exist (not
only not in the kernel). But given that it is hard to fix a touchpad
that has no or only one mouse button ... I don't know.

Anyway, the patch is really small/non-intrusive but if it is not
appropriate  I can - thanks to git - easily have it in some local
branch :)

Soeren
-- 
For the one fact about the future of which we can be certain is that it
will be utterly fantastic. -- Arthur C. Clarke, 1962

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

Re: [PATCH] mouse button 1 emulation for mac_hid

[Jiri Kosina]

On Fri, 13 Nov 2009, Dmitry Torokhov wrote:

> > I am not sure what the future of mac_hid is (the source says will be
> > removed, but since 9 years). In case it stays in kernel the attached
> > patch adds support for mouse button 1 emulation. If this driver stays in
> > the kernel I would also suggest to rename it - it is not mac specific (I
> > am using this on a netbook).
> > 
> > Rationale? Broken touchpads, within linux unsupported (multi-)touchpads
> > where click and hold + moving the mouse does not work (yet?).
> 
> I would rather fix the touchpads so they work properly than pile more
> quirks in kernel.

Hi Dmitry,

in exactly what way are you planning to "fix the touchpads"?

-- 
Jiri Kosina
SUSE Labs, Novell Inc.

Re: [PATCH] mouse button 1 emulation for mac_hid

[Dmitry Torokhov]

On Mon, Nov 16, 2009 at 01:59:37PM +0100, Jiri Kosina wrote:
> On Fri, 13 Nov 2009, Dmitry Torokhov wrote:
> 
> > > I am not sure what the future of mac_hid is (the source says will be
> > > removed, but since 9 years). In case it stays in kernel the attached
> > > patch adds support for mouse button 1 emulation. If this driver stays in
> > > the kernel I would also suggest to rename it - it is not mac specific (I
> > > am using this on a netbook).
> > > 
> > > Rationale? Broken touchpads, within linux unsupported (multi-)touchpads
> > > where click and hold + moving the mouse does not work (yet?).
> > 
> > I would rather fix the touchpads so they work properly than pile more
> > quirks in kernel.
> 
> Hi Dmitry,
> 
> in exactly what way are you planning to "fix the touchpads"?
> 

Hi Jiri,

It would be helpful to know which ones are currently. Obviously we can't
help with physically broken ones but I don't believe it is kernel's task
to deal with physically broken hardware. The ones that are not
physically broken - PS/2 should have fairly decent hardware emulation
mode (at least all I tried supported tap and trag while working as
regular PS/2 mouse) - so that can be used while we figuring out native
protocol. USB - appletouch, bcm - are there any issues at the moment?

-- 
Dmitry

Re: [PATCH] mouse button 1 emulation for mac_hid

[Soeren Sonnenburg]

[-- Attachment #1: Type: text/plain, Size: 1860 bytes --]

On Mon, 2009-11-16 at 09:05 -0800, Dmitry Torokhov wrote:
> On Mon, Nov 16, 2009 at 01:59:37PM +0100, Jiri Kosina wrote:
> > On Fri, 13 Nov 2009, Dmitry Torokhov wrote:
> > 
> > > > I am not sure what the future of mac_hid is (the source says will be
> > > > removed, but since 9 years). In case it stays in kernel the attached
> > > > patch adds support for mouse button 1 emulation. If this driver stays in
> > > > the kernel I would also suggest to rename it - it is not mac specific (I
> > > > am using this on a netbook).
> > > > 
> > > > Rationale? Broken touchpads, within linux unsupported (multi-)touchpads
> > > > where click and hold + moving the mouse does not work (yet?).
> > > 
> > > I would rather fix the touchpads so they work properly than pile more
> > > quirks in kernel.
> > 
> > Hi Dmitry,
> > 
> > in exactly what way are you planning to "fix the touchpads"?
> > 
> 
> Hi Jiri,
> 
> It would be helpful to know which ones are currently. Obviously we can't
> help with physically broken ones but I don't believe it is kernel's task
> to deal with physically broken hardware. The ones that are not
> physically broken - PS/2 should have fairly decent hardware emulation
> mode (at least all I tried supported tap and trag while working as
> regular PS/2 mouse) - so that can be used while we figuring out native
> protocol. USB - appletouch, bcm - are there any issues at the moment?

Ignoring broken hardware, there is the problem or feature that some
touchpads have the mouse *under the touchpad*. So when you click there
and want to move the mouse - well good luck. Of course this is something
that needs to be ultimately fixed in the synaptics xorg driver.

Soeren
-- 
For the one fact about the future of which we can be certain is that it
will be utterly fantastic. -- Arthur C. Clarke, 1962

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 836 bytes --]