diff mbox

intel-hid: new hid event driver for hotkeys

Message ID 1450337402-4178-1-git-send-email-alex.hung@canonical.com (mailing list archive)
State Changes Requested, archived
Headers show

Commit Message

Alex Hung Dec. 17, 2015, 7:30 a.m. UTC
This driver supports various hid events including hotkeys.
Dell XPS 13 9350 requires it for wireless hotkey.

Andy Lutomirski contributed greatly to this driver.

Signed-off-by: Alex Hung <alex.hung@canonical.com>
---
 MAINTAINERS                      |   6 +
 drivers/platform/x86/Kconfig     |  11 ++
 drivers/platform/x86/Makefile    |   1 +
 drivers/platform/x86/intel-hid.c | 289 +++++++++++++++++++++++++++++++++++++++
 4 files changed, 307 insertions(+)
 create mode 100644 drivers/platform/x86/intel-hid.c

Comments

Darren Hart Dec. 17, 2015, 10:15 p.m. UTC | #1
On Thu, Dec 17, 2015 at 03:30:02PM +0800, Alex Hung wrote:
> This driver supports various hid events including hotkeys.
> Dell XPS 13 9350 requires it for wireless hotkey.
> 
> Andy Lutomirski contributed greatly to this driver.
> 
> Signed-off-by: Alex Hung <alex.hung@canonical.com>

Queued for testing.

Andy, please provide your Reviewed-by if you're happy with it.

Alex, just one question maybe you can answer quickly and save me some time
below:


> +static int intel_hid_pl_suspend_handler(struct device *device)
> +{
> +	intel_hid_set_enable(device, 0);
> +	return 0;
> +}
> +
> +static int intel_hid_pl_resume_handler(struct device *device)
> +{
> +	intel_hid_set_enable(device, 1);
> +	return 0;
> +}

Why not propagate the intel_hid_set_enable() return code? Is it because it just
doesn't really impact suspend/resume, even if we do get an error?
Andy Lutomirski Dec. 17, 2015, 10:50 p.m. UTC | #2
On Thu, Dec 17, 2015 at 2:15 PM, Darren Hart <dvhart@infradead.org> wrote:
> On Thu, Dec 17, 2015 at 03:30:02PM +0800, Alex Hung wrote:
>> This driver supports various hid events including hotkeys.
>> Dell XPS 13 9350 requires it for wireless hotkey.
>>
>> Andy Lutomirski contributed greatly to this driver.
>>
>> Signed-off-by: Alex Hung <alex.hung@canonical.com>
>
> Queued for testing.
>
> Andy, please provide your Reviewed-by if you're happy with it.

Reviewed-and-tested-by: Andy Lutomirski <luto@kernel.org>

Minor nits, though:

+        This driver provides supports for Intel HID event. Some laptops
+        require this driver for hotkey supports.

Should that be "This driver provides support for the Intel HID Event
hotkey interface.  Some laptops require this driver for hotkey
support."?

Also, can we remove the word "Filter" a couple lines up in the Kconfig
file?  I'd guess the the Windows equivalent is a "filter' driver in
the Windows model, but it's not logically filtering anything, and
there's nothing filter-like about the Linux driver.

(I think it's slightly sad that HID is in the name, too, but that's
indeed what it seems to be called.)

>
> Alex, just one question maybe you can answer quickly and save me some time
> below:
>
>
>> +static int intel_hid_pl_suspend_handler(struct device *device)
>> +{
>> +     intel_hid_set_enable(device, 0);
>> +     return 0;
>> +}
>> +
>> +static int intel_hid_pl_resume_handler(struct device *device)
>> +{
>> +     intel_hid_set_enable(device, 1);
>> +     return 0;
>> +}
>
> Why not propagate the intel_hid_set_enable() return code? Is it because it just
> doesn't really impact suspend/resume, even if we do get an error?

That was me, actually.  I definitely don't want to cause suspect to
fail if the ACPI call fails.  At worst we'll report a spurious key
event when we resume.

--Andy
--
To unsubscribe from this list: send the line "unsubscribe platform-driver-x86" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Darren Hart Dec. 17, 2015, 11:01 p.m. UTC | #3
On Thu, Dec 17, 2015 at 02:50:39PM -0800, Andy Lutomirski wrote:
> On Thu, Dec 17, 2015 at 2:15 PM, Darren Hart <dvhart@infradead.org> wrote:
> > On Thu, Dec 17, 2015 at 03:30:02PM +0800, Alex Hung wrote:
> >> This driver supports various hid events including hotkeys.
> >> Dell XPS 13 9350 requires it for wireless hotkey.
> >>
> >> Andy Lutomirski contributed greatly to this driver.
> >>
> >> Signed-off-by: Alex Hung <alex.hung@canonical.com>
> >
> > Queued for testing.
> >
> > Andy, please provide your Reviewed-by if you're happy with it.
> 
> Reviewed-and-tested-by: Andy Lutomirski <luto@kernel.org>
> 
> Minor nits, though:
> 
> +        This driver provides supports for Intel HID event. Some laptops
> +        require this driver for hotkey supports.
> 
> Should that be "This driver provides support for the Intel HID Event
> hotkey interface.  Some laptops require this driver for hotkey
> support."?
> 
> Also, can we remove the word "Filter" a couple lines up in the Kconfig
> file?  I'd guess the the Windows equivalent is a "filter' driver in
> the Windows model, but it's not logically filtering anything, and
> there's nothing filter-like about the Linux driver.
> 
> (I think it's slightly sad that HID is in the name, too, but that's
> indeed what it seems to be called.)

These are minor, but I had similar thoughts. I'd be happy to see these included
and respun.

> 
> >
> > Alex, just one question maybe you can answer quickly and save me some time
> > below:
> >
> >
> >> +static int intel_hid_pl_suspend_handler(struct device *device)
> >> +{
> >> +     intel_hid_set_enable(device, 0);
> >> +     return 0;
> >> +}
> >> +
> >> +static int intel_hid_pl_resume_handler(struct device *device)
> >> +{
> >> +     intel_hid_set_enable(device, 1);
> >> +     return 0;
> >> +}
> >
> > Why not propagate the intel_hid_set_enable() return code? Is it because it just
> > doesn't really impact suspend/resume, even if we do get an error?
> 
> That was me, actually.  I definitely don't want to cause suspect to
> fail if the ACPI call fails.  At worst we'll report a spurious key
> event when we resume.
> 

OK, thanks for the clarification (reminder :-)
Alex Hung Dec. 18, 2015, 3:28 p.m. UTC | #4
On Fri, Dec 18, 2015 at 7:01 AM, Darren Hart <dvhart@infradead.org> wrote:
> On Thu, Dec 17, 2015 at 02:50:39PM -0800, Andy Lutomirski wrote:
>> On Thu, Dec 17, 2015 at 2:15 PM, Darren Hart <dvhart@infradead.org> wrote:
>> > On Thu, Dec 17, 2015 at 03:30:02PM +0800, Alex Hung wrote:
>> >> This driver supports various hid events including hotkeys.
>> >> Dell XPS 13 9350 requires it for wireless hotkey.
>> >>
>> >> Andy Lutomirski contributed greatly to this driver.
>> >>
>> >> Signed-off-by: Alex Hung <alex.hung@canonical.com>
>> >
>> > Queued for testing.
>> >
>> > Andy, please provide your Reviewed-by if you're happy with it.
>>
>> Reviewed-and-tested-by: Andy Lutomirski <luto@kernel.org>
>>
>> Minor nits, though:
>>
>> +        This driver provides supports for Intel HID event. Some laptops
>> +        require this driver for hotkey supports.
>>
>> Should that be "This driver provides support for the Intel HID Event
>> hotkey interface.  Some laptops require this driver for hotkey
>> support."?
>>
>> Also, can we remove the word "Filter" a couple lines up in the Kconfig
>> file?  I'd guess the the Windows equivalent is a "filter' driver in
>> the Windows model, but it's not logically filtering anything, and
>> there's nothing filter-like about the Linux driver.
>>
>> (I think it's slightly sad that HID is in the name, too, but that's
>> indeed what it seems to be called.)
>
> These are minor, but I had similar thoughts. I'd be happy to see these included
> and respun.

That sounds good. I am removing "filter" in various places and will
resend the patch shortly.

>
>>
>> >
>> > Alex, just one question maybe you can answer quickly and save me some time
>> > below:
>> >
>> >
>> >> +static int intel_hid_pl_suspend_handler(struct device *device)
>> >> +{
>> >> +     intel_hid_set_enable(device, 0);
>> >> +     return 0;
>> >> +}
>> >> +
>> >> +static int intel_hid_pl_resume_handler(struct device *device)
>> >> +{
>> >> +     intel_hid_set_enable(device, 1);
>> >> +     return 0;
>> >> +}
>> >
>> > Why not propagate the intel_hid_set_enable() return code? Is it because it just
>> > doesn't really impact suspend/resume, even if we do get an error?
>>
>> That was me, actually.  I definitely don't want to cause suspect to
>> fail if the ACPI call fails.  At worst we'll report a spurious key
>> event when we resume.
>>
>
> OK, thanks for the clarification (reminder :-)
>
> --
> Darren Hart
> Intel Open Source Technology Center
Dmitry Torokhov Sept. 29, 2016, 4:02 p.m. UTC | #5
Quite a bit late, but still:

On Wed, Dec 16, 2015 at 11:30 PM, Alex Hung <alex.hung@canonical.com> wrote:
> +
> +static int __init intel_hid_init(void)
> +{
> +       acpi_walk_namespace(ACPI_TYPE_DEVICE, ACPI_ROOT_OBJECT,
> +                           ACPI_UINT32_MAX, check_acpi_dev, NULL,
> +                           (void *)intel_hid_ids, NULL);
> +
> +       return platform_driver_register(&intel_hid_pl_driver);
> +}

Why do we need to walk instantiate the device ourselves instead of
having ACPI core do it for us? I also see this pattern in intel-vbtn.c
now.

Thanks.
Andy Lutomirski Sept. 29, 2016, 5:29 p.m. UTC | #6
On Thu, Sep 29, 2016 at 9:02 AM, Dmitry Torokhov
<dmitry.torokhov@gmail.com> wrote:
> Quite a bit late, but still:
>
> On Wed, Dec 16, 2015 at 11:30 PM, Alex Hung <alex.hung@canonical.com> wrote:
>> +
>> +static int __init intel_hid_init(void)
>> +{
>> +       acpi_walk_namespace(ACPI_TYPE_DEVICE, ACPI_ROOT_OBJECT,
>> +                           ACPI_UINT32_MAX, check_acpi_dev, NULL,
>> +                           (void *)intel_hid_ids, NULL);
>> +
>> +       return platform_driver_register(&intel_hid_pl_driver);
>> +}
>
> Why do we need to walk instantiate the device ourselves instead of
> having ACPI core do it for us? I also see this pattern in intel-vbtn.c
> now.

See the comment above check_acpi_dev():

/*
 * Unfortunately, some laptops provide a _HID="INT33D5" device with
 * _CID="PNP0C02".  This causes the pnpacpi scan driver to claim the
 * ACPI node, so no platform device will be created.  The pnpacpi
 * driver rejects this device in subsequent processing, so no physical
 * node is created at all.
 *
 * As a workaround until the ACPI core figures out how to handle
 * this corner case, manually ask the ACPI platform device code to
 * claim the ACPI node.
 */


>
> Thanks.
>
> --
> Dmitry
Dmitry Torokhov Sept. 30, 2016, 10:44 p.m. UTC | #7
On Thu, Sep 29, 2016 at 10:29:49AM -0700, Andy Lutomirski wrote:
> On Thu, Sep 29, 2016 at 9:02 AM, Dmitry Torokhov
> <dmitry.torokhov@gmail.com> wrote:
> > Quite a bit late, but still:
> >
> > On Wed, Dec 16, 2015 at 11:30 PM, Alex Hung <alex.hung@canonical.com> wrote:
> >> +
> >> +static int __init intel_hid_init(void)
> >> +{
> >> +       acpi_walk_namespace(ACPI_TYPE_DEVICE, ACPI_ROOT_OBJECT,
> >> +                           ACPI_UINT32_MAX, check_acpi_dev, NULL,
> >> +                           (void *)intel_hid_ids, NULL);
> >> +
> >> +       return platform_driver_register(&intel_hid_pl_driver);
> >> +}
> >
> > Why do we need to walk instantiate the device ourselves instead of
> > having ACPI core do it for us? I also see this pattern in intel-vbtn.c
> > now.
> 
> See the comment above check_acpi_dev():
> 
> /*
>  * Unfortunately, some laptops provide a _HID="INT33D5" device with
>  * _CID="PNP0C02".  This causes the pnpacpi scan driver to claim the
>  * ACPI node, so no platform device will be created.  The pnpacpi
>  * driver rejects this device in subsequent processing, so no physical
>  * node is created at all.
>  *
>  * As a workaround until the ACPI core figures out how to handle
>  * this corner case, manually ask the ACPI platform device code to
>  * claim the ACPI node.
>  */

Ah, I missed that (because I originally looked at intel-vbtn). Does
INT33D6 also have the same issue?
Andy Lutomirski Sept. 30, 2016, 11:06 p.m. UTC | #8
On Fri, Sep 30, 2016 at 3:44 PM, Dmitry Torokhov
<dmitry.torokhov@gmail.com> wrote:
> On Thu, Sep 29, 2016 at 10:29:49AM -0700, Andy Lutomirski wrote:
>> On Thu, Sep 29, 2016 at 9:02 AM, Dmitry Torokhov
>> <dmitry.torokhov@gmail.com> wrote:
>> > Quite a bit late, but still:
>> >
>> > On Wed, Dec 16, 2015 at 11:30 PM, Alex Hung <alex.hung@canonical.com> wrote:
>> >> +
>> >> +static int __init intel_hid_init(void)
>> >> +{
>> >> +       acpi_walk_namespace(ACPI_TYPE_DEVICE, ACPI_ROOT_OBJECT,
>> >> +                           ACPI_UINT32_MAX, check_acpi_dev, NULL,
>> >> +                           (void *)intel_hid_ids, NULL);
>> >> +
>> >> +       return platform_driver_register(&intel_hid_pl_driver);
>> >> +}
>> >
>> > Why do we need to walk instantiate the device ourselves instead of
>> > having ACPI core do it for us? I also see this pattern in intel-vbtn.c
>> > now.
>>
>> See the comment above check_acpi_dev():
>>
>> /*
>>  * Unfortunately, some laptops provide a _HID="INT33D5" device with
>>  * _CID="PNP0C02".  This causes the pnpacpi scan driver to claim the
>>  * ACPI node, so no platform device will be created.  The pnpacpi
>>  * driver rejects this device in subsequent processing, so no physical
>>  * node is created at all.
>>  *
>>  * As a workaround until the ACPI core figures out how to handle
>>  * this corner case, manually ask the ACPI platform device code to
>>  * claim the ACPI node.
>>  */
>
> Ah, I missed that (because I originally looked at intel-vbtn). Does
> INT33D6 also have the same issue?

I don't know, because I don't have that device at all.  Alex?

--Andy
--
To unsubscribe from this list: send the line "unsubscribe platform-driver-x86" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
diff mbox

Patch

diff --git a/MAINTAINERS b/MAINTAINERS
index 9bff63c..083b704 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -5502,6 +5502,12 @@  T:	git git://git.code.sf.net/p/intel-sas/isci
 S:	Supported
 F:	drivers/scsi/isci/
 
+INTEL HID EVENT FILTER DRIVER
+M:	Alex Hung <alex.hung@canonical.com>
+L:	platform-driver-x86@vger.kernel.org
+S:	Maintained
+F:	drivers/platform/x86/intel-hid.c
+
 INTEL IDLE DRIVER
 M:	Len Brown <lenb@kernel.org>
 L:	linux-pm@vger.kernel.org
diff --git a/drivers/platform/x86/Kconfig b/drivers/platform/x86/Kconfig
index 1089eaa..9d31221 100644
--- a/drivers/platform/x86/Kconfig
+++ b/drivers/platform/x86/Kconfig
@@ -731,6 +731,17 @@  config ACPI_CMPC
 	  keys as input device, backlight device, tablet and accelerometer
 	  devices.
 
+config INTEL_HID_EVENT
+	tristate "INTEL HID Event Filter"
+	depends on ACPI
+	depends on INPUT
+	help
+	 This driver provides supports for Intel HID event. Some laptops
+	 require this driver for hotkey supports.
+
+	 To compile this driver as a module, choose M here: the module will
+	 be called intel_hid.
+
 config INTEL_SCU_IPC
 	bool "Intel SCU IPC Support"
 	depends on X86_INTEL_MID
diff --git a/drivers/platform/x86/Makefile b/drivers/platform/x86/Makefile
index 3ca78a3..c7766a4 100644
--- a/drivers/platform/x86/Makefile
+++ b/drivers/platform/x86/Makefile
@@ -41,6 +41,7 @@  obj-$(CONFIG_ACPI_TOSHIBA)	+= toshiba_acpi.o
 obj-$(CONFIG_TOSHIBA_BT_RFKILL)	+= toshiba_bluetooth.o
 obj-$(CONFIG_TOSHIBA_HAPS)	+= toshiba_haps.o
 obj-$(CONFIG_TOSHIBA_WMI)	+= toshiba-wmi.o
+obj-$(CONFIG_INTEL_HID_EVENT)	+= intel-hid.o
 obj-$(CONFIG_INTEL_SCU_IPC)	+= intel_scu_ipc.o
 obj-$(CONFIG_INTEL_SCU_IPC_UTIL) += intel_scu_ipcutil.o
 obj-$(CONFIG_INTEL_MFLD_THERMAL) += intel_mid_thermal.o
diff --git a/drivers/platform/x86/intel-hid.c b/drivers/platform/x86/intel-hid.c
new file mode 100644
index 0000000..02292ac
--- /dev/null
+++ b/drivers/platform/x86/intel-hid.c
@@ -0,0 +1,289 @@ 
+/*
+ *  Intel HID event filter driver for Windows 8
+ *
+ *  Copyright (C) 2015 Alex Hung <alex.hung@canonical.com>
+ *  Copyright (C) 2015 Andrew Lutomirski <luto@kernel.org>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/input.h>
+#include <linux/platform_device.h>
+#include <linux/input/sparse-keymap.h>
+#include <linux/acpi.h>
+#include <acpi/acpi_bus.h>
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Alex Hung");
+
+static const struct acpi_device_id intel_hid_ids[] = {
+	{"INT33D5", 0},
+	{"", 0},
+};
+
+/* In theory, these are HID usages. */
+static const struct key_entry intel_hid_keymap[] = {
+	/* 1: LSuper (Page 0x07, usage 0xE3) -- unclear what to do */
+	/* 2: Toggle SW_ROTATE_LOCK -- easy to implement if seen in wild */
+	{ KE_KEY, 3, { KEY_NUMLOCK } },
+	{ KE_KEY, 4, { KEY_HOME } },
+	{ KE_KEY, 5, { KEY_END } },
+	{ KE_KEY, 6, { KEY_PAGEUP } },
+	{ KE_KEY, 4, { KEY_PAGEDOWN } },
+	{ KE_KEY, 4, { KEY_HOME } },
+	{ KE_KEY, 8, { KEY_RFKILL } },
+	{ KE_KEY, 9, { KEY_POWER } },
+	{ KE_KEY, 11, { KEY_SLEEP } },
+	/* 13 has two different meanings in the spec -- ignore it. */
+	{ KE_KEY, 14, { KEY_STOPCD } },
+	{ KE_KEY, 15, { KEY_PLAYPAUSE } },
+	{ KE_KEY, 16, { KEY_MUTE } },
+	{ KE_KEY, 17, { KEY_VOLUMEUP } },
+	{ KE_KEY, 18, { KEY_VOLUMEDOWN } },
+	{ KE_KEY, 19, { KEY_BRIGHTNESSUP } },
+	{ KE_KEY, 20, { KEY_BRIGHTNESSDOWN } },
+	/* 27: wake -- needs special handling */
+	{ KE_END },
+};
+
+struct intel_hid_priv {
+	struct input_dev *input_dev;
+};
+
+static int intel_hid_set_enable(struct device *device, int enable)
+{
+	union acpi_object arg0 = { ACPI_TYPE_INTEGER };
+	struct acpi_object_list args = { 1, &arg0 };
+	acpi_status status;
+
+	arg0.integer.value = enable;
+	status = acpi_evaluate_object(ACPI_HANDLE(device), "HDSM", &args, NULL);
+	if (!ACPI_SUCCESS(status)) {
+		dev_warn(device, "failed to %sable hotkeys\n",
+			 enable ? "en" : "dis");
+		return -EIO;
+	}
+
+	return 0;
+}
+
+static int intel_hid_pl_suspend_handler(struct device *device)
+{
+	intel_hid_set_enable(device, 0);
+	return 0;
+}
+
+static int intel_hid_pl_resume_handler(struct device *device)
+{
+	intel_hid_set_enable(device, 1);
+	return 0;
+}
+
+static const struct dev_pm_ops intel_hid_pl_pm_ops = {
+	.suspend  = intel_hid_pl_suspend_handler,
+	.resume  = intel_hid_pl_resume_handler,
+};
+
+static int intel_hid_input_setup(struct platform_device *device)
+{
+	struct intel_hid_priv *priv = dev_get_drvdata(&device->dev);
+	int ret;
+
+	priv->input_dev = input_allocate_device();
+	if (!priv->input_dev)
+		return -ENOMEM;
+
+	ret = sparse_keymap_setup(priv->input_dev, intel_hid_keymap, NULL);
+	if (ret)
+		goto err_free_device;
+
+	priv->input_dev->dev.parent = &device->dev;
+	priv->input_dev->name = "Intel HID events";
+	priv->input_dev->id.bustype = BUS_HOST;
+	set_bit(KEY_RFKILL, priv->input_dev->keybit);
+
+	ret = input_register_device(priv->input_dev);
+	if (ret)
+		goto err_free_device;
+
+	return 0;
+
+err_free_device:
+		input_free_device(priv->input_dev);
+		return ret;
+}
+
+static void intel_hid_input_destroy(struct platform_device *device)
+{
+	struct intel_hid_priv *priv = dev_get_drvdata(&device->dev);
+
+	input_unregister_device(priv->input_dev);
+}
+
+static void notify_handler(acpi_handle handle, u32 event, void *context)
+{
+	struct platform_device *device = context;
+	struct intel_hid_priv *priv = dev_get_drvdata(&device->dev);
+	unsigned long long ev_index;
+	acpi_status status;
+
+	/* The platform spec only defines one event code: 0xC0. */
+	if (event != 0xc0) {
+		dev_warn(&device->dev, "received unknown event (0x%x)\n",
+			 event);
+		return;
+	}
+
+	status = acpi_evaluate_integer(handle, "HDEM", NULL, &ev_index);
+	if (!ACPI_SUCCESS(status)) {
+		dev_warn(&device->dev, "failed to get event index\n");
+		return;
+	}
+
+	if (!sparse_keymap_report_event(priv->input_dev, ev_index, 1, true))
+		dev_info(&device->dev, "unknown event index 0x%llx\n",
+			 ev_index);
+}
+
+static int intel_hid_probe(struct platform_device *device)
+{
+	acpi_handle handle = ACPI_HANDLE(&device->dev);
+	struct intel_hid_priv *priv;
+	unsigned long long mode;
+	acpi_status status;
+	int err;
+
+	status = acpi_evaluate_integer(handle, "HDMM", NULL, &mode);
+	if (!ACPI_SUCCESS(status)) {
+		dev_warn(&device->dev, "failed to read mode\n");
+		return -ENODEV;
+	}
+
+	if (mode != 0) {
+		/*
+		 * This driver only implements "simple" mode.  There appear
+		 * to be no other modes, but we should be paranoid and check
+		 * for compatibility.
+		 */
+		dev_info(&device->dev, "platform is not in simple mode\n");
+		return -ENODEV;
+	}
+
+	priv = devm_kzalloc(&device->dev,
+			    sizeof(struct intel_hid_priv *), GFP_KERNEL);
+	if (!priv)
+		return -ENOMEM;
+	dev_set_drvdata(&device->dev, priv);
+
+	err = intel_hid_input_setup(device);
+	if (err) {
+		pr_err("Failed to setup Intel HID hotkeys\n");
+		return err;
+	}
+
+	status = acpi_install_notify_handler(handle,
+					     ACPI_DEVICE_NOTIFY,
+					     notify_handler,
+					     device);
+	if (ACPI_FAILURE(status)) {
+		err = -EBUSY;
+		goto err_remove_input;
+	}
+
+	err = intel_hid_set_enable(&device->dev, 1);
+	if (err)
+		goto err_remove_notify;
+
+	return 0;
+
+err_remove_notify:
+	acpi_remove_notify_handler(handle, ACPI_DEVICE_NOTIFY, notify_handler);
+
+err_remove_input:
+	intel_hid_input_destroy(device);
+
+	return err;
+}
+
+static int intel_hid_remove(struct platform_device *device)
+{
+	acpi_handle handle = ACPI_HANDLE(&device->dev);
+
+	acpi_remove_notify_handler(handle, ACPI_DEVICE_NOTIFY, notify_handler);
+	intel_hid_input_destroy(device);
+	intel_hid_set_enable(&device->dev, 0);
+	acpi_remove_notify_handler(handle, ACPI_DEVICE_NOTIFY, notify_handler);
+
+	/*
+	 * Even if we failed to shut off the event stream, we can still
+	 * safely detach from the device.
+	 */
+	return 0;
+}
+
+static struct platform_driver intel_hid_pl_driver = {
+	.driver = {
+		.name = "intel-hid",
+		.acpi_match_table = intel_hid_ids,
+		.pm = &intel_hid_pl_pm_ops,
+	},
+	.probe = intel_hid_probe,
+	.remove = intel_hid_remove,
+};
+MODULE_DEVICE_TABLE(acpi, intel_hid_ids);
+
+/*
+ * Unfortunately, some laptops provide a _HID="INT33D5" device with
+ * _CID="PNP0C02".  This causes the pnpacpi scan driver to claim the
+ * ACPI node, so no platform device will be created.  The pnpacpi
+ * driver rejects this device in subsequent processing, so no physical
+ * node is created at all.
+ *
+ * As a workaround until the ACPI core figures out how to handle
+ * this corner case, manually ask the ACPI platform device code to
+ * claim the ACPI node.
+ */
+static acpi_status __init
+check_acpi_dev(acpi_handle handle, u32 lvl, void *context, void **rv)
+{
+	const struct acpi_device_id *ids = context;
+	struct acpi_device *dev;
+
+	if (acpi_bus_get_device(handle, &dev) != 0)
+		return AE_OK;
+
+	if (acpi_match_device_ids(dev, ids) == 0)
+		if (acpi_create_platform_device(dev))
+			dev_info(&dev->dev,
+				 "intel-hid: created platform device\n");
+
+	return AE_OK;
+}
+
+static int __init intel_hid_init(void)
+{
+	acpi_walk_namespace(ACPI_TYPE_DEVICE, ACPI_ROOT_OBJECT,
+			    ACPI_UINT32_MAX, check_acpi_dev, NULL,
+			    (void *)intel_hid_ids, NULL);
+
+	return platform_driver_register(&intel_hid_pl_driver);
+}
+module_init(intel_hid_init);
+
+static void __exit intel_hid_exit(void)
+{
+	platform_driver_unregister(&intel_hid_pl_driver);
+}
+module_exit(intel_hid_exit);