[PATCH] hid: wellspring device quirks

From: Henrik Rydberg <redacted>

This patch adds the HID_QUIRK_IGNORE_MOUSE quirk for the Apple USB
multitouch devices based on BCM5974, aka Wellspring. These devices
will instead be picked up by the new bcm5974 driver.

Signed-off-by: Henrik Rydberg <redacted>
---

The bcm5974 driver is currently in Andrew Morton's mm tree:
input-bcm5974-055-smoother-motion-irq-simplification.patch

This patch is against the linux-next tree: v2.6.26-rc9-7639-g60392c2

diff --git a/drivers/hid/usbhid/hid-quirks.c b/drivers/hid/usbhid/hid-quirks.c
index 61e78a4..b15f882 100644
--- a/drivers/hid/usbhid/hid-quirks.c
+++ b/drivers/hid/usbhid/hid-quirks.c

@@ -654,12 +654,12 @@ static const struct hid_blacklist {
 	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_ALU_WIRELESS_ANSI, HID_QUIRK_APPLE_NUMLOCK_EMULATION | HID_QUIRK_APPLE_HAS_FN },
 	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_ALU_WIRELESS_ISO, HID_QUIRK_APPLE_NUMLOCK_EMULATION | HID_QUIRK_APPLE_HAS_FN | HID_QUIRK_APPLE_ISO_KEYBOARD },
 	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_ALU_WIRELESS_JIS, HID_QUIRK_APPLE_NUMLOCK_EMULATION | HID_QUIRK_APPLE_HAS_FN },
-	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING_ANSI, HID_QUIRK_APPLE_HAS_FN },
-	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING_ISO, HID_QUIRK_APPLE_HAS_FN | HID_QUIRK_APPLE_ISO_KEYBOARD },
-	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING_JIS, HID_QUIRK_APPLE_HAS_FN },
-	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING2_ANSI, HID_QUIRK_APPLE_HAS_FN },
-	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING2_ISO, HID_QUIRK_APPLE_HAS_FN | HID_QUIRK_APPLE_ISO_KEYBOARD },
-	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING2_JIS, HID_QUIRK_APPLE_HAS_FN },
+	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING_ANSI, HID_QUIRK_APPLE_HAS_FN | HID_QUIRK_IGNORE_MOUSE },
+	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING_ISO, HID_QUIRK_APPLE_HAS_FN | HID_QUIRK_APPLE_ISO_KEYBOARD | HID_QUIRK_IGNORE_MOUSE},
+	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING_JIS, HID_QUIRK_APPLE_HAS_FN | HID_QUIRK_IGNORE_MOUSE},
+	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING2_ANSI, HID_QUIRK_APPLE_HAS_FN | HID_QUIRK_IGNORE_MOUSE},
+	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING2_ISO, HID_QUIRK_APPLE_HAS_FN | HID_QUIRK_APPLE_ISO_KEYBOARD | HID_QUIRK_IGNORE_MOUSE },
+	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING2_JIS, HID_QUIRK_APPLE_HAS_FN  | HID_QUIRK_IGNORE_MOUSE },
 	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_FOUNTAIN_TP_ONLY, HID_QUIRK_APPLE_NUMLOCK_EMULATION | HID_QUIRK_APPLE_HAS_FN | HID_QUIRK_IGNORE_MOUSE },
 	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_GEYSER1_TP_ONLY, HID_QUIRK_APPLE_NUMLOCK_EMULATION | HID_QUIRK_APPLE_HAS_FN | HID_QUIRK_IGNORE_MOUSE },
 

On Sun, 13 Jul 2008, Henrik Rydberg wrote:

This patch adds the HID_QUIRK_IGNORE_MOUSE quirk for the Apple USB
multitouch devices based on BCM5974, aka Wellspring. These devices
will instead be picked up by the new bcm5974 driver.

Hi Henrik,

is the mouse not working at all with standard HID driver?

I'd prefer to postpone merging this patch until bcm5974 hits mainline if the device works in a limited way with standard HID driver, so that we don't introduce regression.

If the device doesn't work with HID driver at all though, it is probably perfectly fine to merge it right away. Please let me know, I have no knowledge about this particular device.

Thanks a lot,

-- 
Jiri Kosina
SUSE Labs

Jiri Kosina wrote:

On Sun, 13 Jul 2008, Henrik Rydberg wrote:

quoted

This patch adds the HID_QUIRK_IGNORE_MOUSE quirk for the Apple USB
multitouch devices based on BCM5974, aka Wellspring. These devices
will instead be picked up by the new bcm5974 driver.

Hi Henrik,

is the mouse not working at all with standard HID driver?

Yes, it is working out-of-the-box, only poorly.

I'd prefer to postpone merging this patch until bcm5974 hits mainline if
the device works in a limited way with standard HID driver, so that we
don't introduce regression.

Absolutely, this patch should be applied together with the bcm5974 driver,
or not be applied. The driver patch is here:

http://userweb.kernel.org/~akpm/mmotm/broken-out/input-bcm5974-055-smoother-motion-irq-simplification.patch

If the device doesn't work with HID driver at all though, it is probably
perfectly fine to merge it right away. Please let me know, I have no
knowledge about this particular device.

It is fine to leave it, but I would of course prefer to have the driver included together with this patch.

Thanks!
Henrik Rydberg

On Tue, Jul 22, 2008 at 03:31:42PM +0200, Henrik Rydberg wrote:

Jiri Kosina wrote:

quoted

On Sun, 13 Jul 2008, Henrik Rydberg wrote:

quoted

This patch adds the HID_QUIRK_IGNORE_MOUSE quirk for the Apple USB
multitouch devices based on BCM5974, aka Wellspring. These devices
will instead be picked up by the new bcm5974 driver.

Hi Henrik,

is the mouse not working at all with standard HID driver?

Yes, it is working out-of-the-box, only poorly.

quoted

I'd prefer to postpone merging this patch until bcm5974 hits mainline if
the device works in a limited way with standard HID driver, so that we
don't introduce regression.

Absolutely, this patch should be applied together with the bcm5974 driver,
or not be applied. The driver patch is here:

http://userweb.kernel.org/~akpm/mmotm/broken-out/input-bcm5974-055-smoother-motion-irq-simplification.patch

quoted

If the device doesn't work with HID driver at all though, it is probably
perfectly fine to merge it right away. Please let me know, I have no
knowledge about this particular device.

It is fine to leave it, but I would of course prefer to have the driver included together with this patch.

Jiri,

Should I take the quirks patch through my tree when I'm ready to apply
bcm5974?

-- 
Dmitry

On Tue, 22 Jul 2008, Dmitry Torokhov wrote:

Should I take the quirks patch through my tree when I'm ready to apply bcm5974?

Hi Dmitry,

that would be the best option, thanks a lot. Please feel free to add

	Signed-off-by: Jiri Kosina [off-list ref]

to Henrik's quirk addition patch when you are going to merge it.

Thanks,

-- 
Jiri Kosina
SUSE Labs