[PATCH] This adds support for keyboards on mid-2011 MacBook Airs

 Hello,

 (Please include me explicitly in replies as I'm not subscribed to the 
 list.)

 This patch adds proper keyboard support for both MacBook Air 4,1 and 
 4,2 models (11" and 13" mid-2011 releases). It is based mostly on work 
 by Joshua Dillon.

 This fixes the issues with the previously-submitted patch by 
 Pieter-Augustijn Van Malleghem (subject: [PATCH] Add MacBookAir4,1 
 keyboard support), namely using an incorrect table, and also contains 
 support for both the 4,1 and 4,2 models instead of just the 4,1 model.

 Thanks,
 Jeff

Hi Jeff,

This patch adds proper keyboard support for both MacBook Air 4,1 and
4,2 models (11" and 13" mid-2011 releases). It is based mostly on
work by Joshua Dillon.

This fixes the issues with the previously-submitted patch by
Pieter-Augustijn Van Malleghem (subject: [PATCH] Add MacBookAir4,1
keyboard support), namely using an incorrect table, and also
contains support for both the 4,1 and 4,2 models instead of just the
4,1 model.

This patch makes no changes to the table logic at all - is that how it
is supposed to be?

Jiri, it seems the bcm5974 part for these devices has already landed,
meaning (some version of) the below is needed in order to have a
functional touchpad. Any chance of squeezing it in?

Thanks,
Henrik

[ adding Nobuhiro to CC ]

On Wed, 5 Oct 2011, Henrik Rydberg wrote:

Hi Jeff,

quoted

This patch adds proper keyboard support for both MacBook Air 4,1 and
4,2 models (11" and 13" mid-2011 releases). It is based mostly on
work by Joshua Dillon.

This fixes the issues with the previously-submitted patch by
Pieter-Augustijn Van Malleghem (subject: [PATCH] Add MacBookAir4,1
keyboard support), namely using an incorrect table, and also
contains support for both the 4,1 and 4,2 models instead of just the
4,1 model.

This patch makes no changes to the table logic at all - is that how it
is supposed to be?

Jiri, it seems the bcm5974 part for these devices has already landed,
meaning (some version of) the below is needed in order to have a
functional touchpad. Any chance of squeezing it in?

Guys,

please check what is in my hid tree (included in linux-next as well) and 
let me know whether there is anything that need to be applied on top of 
that. I believe all the HID-related 4,1 and 4,2 bits are already in there.

Thanks,
Henrik

quoted

From e0f31e3d1a89831239676baa9bb244feee153133 Mon Sep 17 00:00:00 2001
From: Jeff Mitchell <redacted>
Date: Wed, 21 Sep 2011 18:20:06 -0400
Subject: [PATCH] This adds support for keyboards on mid-2011 MacBook Airs
 (both the 11" 4,1 and 13" 4,2 models). Confusingly, 4,1
 corresponds to WELLSPRING6A and 4,2 corresponds to
 WELLSPRING6.

The patch is a cleaned up version of that created by Josh Dillon (jvdillon@gmail.com).

Signed-off-by: Jeff Mitchell <redacted>
---
 drivers/hid/hid-apple.c |   12 ++++++++++++
 drivers/hid/hid-core.c  |    6 ++++++
 drivers/hid/hid-ids.h   |    6 ++++++
 3 files changed, 24 insertions(+), 0 deletions(-)

diff --git a/drivers/hid/hid-apple.c b/drivers/hid/hid-apple.c
index 18b3bc6..03e6fb0 100644
--- a/drivers/hid/hid-apple.c
+++ b/drivers/hid/hid-apple.c

@@ -493,6 +493,18 @@ static const struct hid_device_id apple_devices[] = {
 		.driver_data = APPLE_HAS_FN | APPLE_ISO_KEYBOARD },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING5_JIS),
 		.driver_data = APPLE_HAS_FN | APPLE_RDESC_JIS },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING6A_ANSI),
+		.driver_data = APPLE_HAS_FN },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING6A_ISO),
+		.driver_data = APPLE_HAS_FN | APPLE_ISO_KEYBOARD },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING6A_JIS),
+		.driver_data = APPLE_HAS_FN | APPLE_RDESC_JIS },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING6_ANSI),
+		.driver_data = APPLE_HAS_FN },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING6_ISO),
+		.driver_data = APPLE_HAS_FN | APPLE_ISO_KEYBOARD },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING6_JIS),
+		.driver_data = APPLE_HAS_FN | APPLE_RDESC_JIS },
 	{ HID_BLUETOOTH_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_ALU_WIRELESS_2009_ANSI),
 		.driver_data = APPLE_NUMLOCK_EMULATION | APPLE_HAS_FN },
 	{ HID_BLUETOOTH_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_ALU_WIRELESS_2009_ISO),

diff --git a/drivers/hid/hid-core.c b/drivers/hid/hid-core.c
index 242353d..96344ff0 100644
--- a/drivers/hid/hid-core.c
+++ b/drivers/hid/hid-core.c

@@ -1340,6 +1340,12 @@ static const struct hid_device_id hid_have_special_driver[] = {
 	{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING5_ANSI) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING5_ISO) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING5_JIS) },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING6A_ANSI) },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING6A_ISO) },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING6A_JIS) },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING6_ANSI) },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING6_ISO) },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING6_JIS) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_ALU_REVB_ANSI) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_ALU_REVB_ISO) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_ALU_REVB_JIS) },

diff --git a/drivers/hid/hid-ids.h b/drivers/hid/hid-ids.h
index 7484e1b..8b37036 100644
--- a/drivers/hid/hid-ids.h
+++ b/drivers/hid/hid-ids.h

@@ -109,6 +109,12 @@
 #define USB_DEVICE_ID_APPLE_WELLSPRING5_ANSI	0x0245
 #define USB_DEVICE_ID_APPLE_WELLSPRING5_ISO	0x0246
 #define USB_DEVICE_ID_APPLE_WELLSPRING5_JIS	0x0247
+#define USB_DEVICE_ID_APPLE_WELLSPRING6A_ANSI	0x0249
+#define USB_DEVICE_ID_APPLE_WELLSPRING6A_ISO	0x024a
+#define USB_DEVICE_ID_APPLE_WELLSPRING6A_JIS	0x024b
+#define USB_DEVICE_ID_APPLE_WELLSPRING6_ANSI	0x024c
+#define USB_DEVICE_ID_APPLE_WELLSPRING6_ISO	0x024d
+#define USB_DEVICE_ID_APPLE_WELLSPRING6_JIS	0x024e
 #define USB_DEVICE_ID_APPLE_ALU_REVB_ANSI	0x024f
 #define USB_DEVICE_ID_APPLE_ALU_REVB_ISO	0x0250
 #define USB_DEVICE_ID_APPLE_ALU_REVB_JIS	0x0251

-- 
1.7.6.2

-- 
Jiri Kosina
SUSE Labs

please check what is in my hid tree (included in linux-next as well) and 
let me know whether there is anything that need to be applied on top of 
that. I believe all the HID-related 4,1 and 4,2 bits are already in there.

Finding the right tree is the big challenge these days. ;-)

Looking at the twin.jikos.cz tree, the upstream and for-next branches
have the same content with regard to these devices. Both 4,1 and 4,2
are included in the special driver and mouse ignore lists. Whether the
keyboard handling is correct, I cannot say, but it looks consistent.

All good for now, it other words. Sorry about the fuzz.

Thanks,
Henrik

On Wed, 5 Oct 2011, Henrik Rydberg wrote:

quoted

please check what is in my hid tree (included in linux-next as well) and 
let me know whether there is anything that need to be applied on top of 
that. I believe all the HID-related 4,1 and 4,2 bits are already in there.

Finding the right tree is the big challenge these days. ;-)

Indeed :)

I will be moving my trees back to kernel.org as soon as it is possible.

Looking at the twin.jikos.cz tree, the upstream and for-next branches 
have the same content with regard to these devices. Both 4,1 and 4,2 are 
included in the special driver and mouse ignore lists. Whether the 
keyboard handling is correct, I cannot say, but it looks consistent.

Perfect, thanks for verification.

All good for now, it other words. Sorry about the fuzz.

No problem at all, the situation with the trees is now simply too 
confusing. Thanks a lot for checking.

-- 
Jiri Kosina
SUSE Labs

On 10/05/2011 06:13 AM, Jiri Kosina wrote:

On Wed, 5 Oct 2011, Henrik Rydberg wrote:

quoted

quoted

please check what is in my hid tree (included in linux-next as well) and 
let me know whether there is anything that need to be applied on top of 
that. I believe all the HID-related 4,1 and 4,2 bits are already in there.

Finding the right tree is the big challenge these days. ;-)

Indeed :)

I will be moving my trees back to kernel.org as soon as it is possible.

Yep, sorry -- I honestly had no idea where to look for the "right" tree.

quoted

Looking at the twin.jikos.cz tree, the upstream and for-next branches 
have the same content with regard to these devices. Both 4,1 and 4,2 are 
included in the special driver and mouse ignore lists. Whether the 
keyboard handling is correct, I cannot say, but it looks consistent.

Perfect, thanks for verification.

Great.

To clear up an earlier point: AFAIK the new MBAs have the same keyboard
layout as normal MacBooks and hence uses the same table. An earlier
patch that was floating around added another table for the new MBAs but
if you looked at the mappings it was the exact same table as the normal
MacBook one.

--Jeff

Great.

To clear up an earlier point: AFAIK the new MBAs have the same keyboard
layout as normal MacBooks and hence uses the same table. An earlier
patch that was floating around added another table for the new MBAs but
if you looked at the mappings it was the exact same table as the normal
MacBook one.

I see. The current tree actually contains two identical tables. Yikes.
Using your patch would have prevented this, but it might be too late
now. If so, the (untested) patch below might fix things up. Does it
still work for you?

Thanks,
Henrik

From deed345085e90505c4534116cc58e0bee98a431f Mon Sep 17 00:00:00 2001

From: Henrik Rydberg <redacted>
Date: Wed, 5 Oct 2011 19:27:27 +0200
Subject: [PATCH] hid-apple: Remove duplicate key table

Commit d762cc290 introduced keyboard support for the WELLSPRING6A
devices, but the added key table is actually identical to the default
one. This patch removes the duplicate.

Signed-off-by: Henrik Rydberg <redacted>
---
 drivers/hid/hid-apple.c |   25 -------------------------
 1 files changed, 0 insertions(+), 25 deletions(-)

diff --git a/drivers/hid/hid-apple.c b/drivers/hid/hid-apple.c
index 2bab9ab..9bc7b03 100644
--- a/drivers/hid/hid-apple.c
+++ b/drivers/hid/hid-apple.c

@@ -82,28 +82,6 @@ static const struct apple_key_translation macbookair_fn_keys[] = {
 	{ }
 };
 
-static const struct apple_key_translation macbookair4_fn_keys[] = {
-	{ KEY_BACKSPACE, KEY_DELETE },
-	{ KEY_ENTER,	KEY_INSERT },
-	{ KEY_F1,	KEY_BRIGHTNESSDOWN, APPLE_FLAG_FKEY },
-	{ KEY_F2,	KEY_BRIGHTNESSUP,   APPLE_FLAG_FKEY },
-	{ KEY_F3,	KEY_SCALE,          APPLE_FLAG_FKEY },
-	{ KEY_F4,	KEY_DASHBOARD,      APPLE_FLAG_FKEY },
-	{ KEY_F5,	KEY_KBDILLUMDOWN,   APPLE_FLAG_FKEY },
-	{ KEY_F6,	KEY_KBDILLUMUP,     APPLE_FLAG_FKEY },
-	{ KEY_F7,	KEY_PREVIOUSSONG,   APPLE_FLAG_FKEY },
-	{ KEY_F8,	KEY_PLAYPAUSE,      APPLE_FLAG_FKEY },
-	{ KEY_F9,	KEY_NEXTSONG,       APPLE_FLAG_FKEY },
-	{ KEY_F10,	KEY_MUTE,           APPLE_FLAG_FKEY },
-	{ KEY_F11,	KEY_VOLUMEDOWN,     APPLE_FLAG_FKEY },
-	{ KEY_F12,	KEY_VOLUMEUP,       APPLE_FLAG_FKEY },
-	{ KEY_UP,	KEY_PAGEUP },
-	{ KEY_DOWN,	KEY_PAGEDOWN },
-	{ KEY_LEFT,	KEY_HOME },
-	{ KEY_RIGHT,	KEY_END },
-	{ }
-};
-
 static const struct apple_key_translation apple_fn_keys[] = {
 	{ KEY_BACKSPACE, KEY_DELETE },
 	{ KEY_ENTER,	KEY_INSERT },

@@ -208,9 +186,6 @@ static int hidinput_apple_event(struct hid_device *hid, struct input_dev *input,
 		else if (hid->product >= USB_DEVICE_ID_APPLE_WELLSPRING6_ANSI &&
 				hid->product <= USB_DEVICE_ID_APPLE_WELLSPRING6_JIS)
 			table = macbookair_fn_keys;
-		else if (hid->product >= USB_DEVICE_ID_APPLE_WELLSPRING6A_ANSI &&
-				hid->product <= USB_DEVICE_ID_APPLE_WELLSPRING6A_JIS)
-			table = macbookair4_fn_keys;
 		else if (hid->product < 0x21d || hid->product >= 0x300)
 			table = powerbook_fn_keys;
 		else

-- 
1.7.6.4

On Wed, 5 Oct 2011, Henrik Rydberg wrote:

quoted

To clear up an earlier point: AFAIK the new MBAs have the same keyboard
layout as normal MacBooks and hence uses the same table. An earlier
patch that was floating around added another table for the new MBAs but
if you looked at the mappings it was the exact same table as the normal
MacBook one.

I see. The current tree actually contains two identical tables. Yikes.
Using your patch would have prevented this, but it might be too late
now. If so, the (untested) patch below might fix things up. Does it
still work for you?

I actually already have the very similar patch in my tree ... I really 
hope for soonish resurrection of kernel.org to avoid this duplication of 
work ...

You can check my tree on twin.jikos.cz to see the changes though.

Thanks guys,

-- 
Jiri Kosina
SUSE Labs

On Wed, Oct 05, 2011 at 07:28:31PM +0200, Jiri Kosina wrote:

On Wed, 5 Oct 2011, Henrik Rydberg wrote:

quoted

quoted

To clear up an earlier point: AFAIK the new MBAs have the same keyboard
layout as normal MacBooks and hence uses the same table. An earlier
patch that was floating around added another table for the new MBAs but
if you looked at the mappings it was the exact same table as the normal
MacBook one.

I see. The current tree actually contains two identical tables. Yikes.
Using your patch would have prevented this, but it might be too late
now. If so, the (untested) patch below might fix things up. Does it
still work for you?

I actually already have the very similar patch in my tree ...

Identical, in fact. Doh! :-)

I really hope for soonish resurrection of kernel.org to avoid this
duplication of work ...

Hear, hear.

Cheers,
Henrik

On 10/05/2011 01:43 PM, Henrik Rydberg wrote:

On Wed, Oct 05, 2011 at 07:28:31PM +0200, Jiri Kosina wrote:

quoted

On Wed, 5 Oct 2011, Henrik Rydberg wrote:

quoted

quoted

To clear up an earlier point: AFAIK the new MBAs have the same keyboard
layout as normal MacBooks and hence uses the same table. An earlier
patch that was floating around added another table for the new MBAs but
if you looked at the mappings it was the exact same table as the normal
MacBook one.

I see. The current tree actually contains two identical tables. Yikes.
Using your patch would have prevented this, but it might be too late
now. If so, the (untested) patch below might fix things up. Does it
still work for you?

I believe that that patch should remove the duplication without any
problem, so everything should still work so long as the other bits are
still there (adding the wellspring6/6a entries to the other files).

Thanks,
Jeff