Skip to contentSkip to conversation

platform/x86: intel-vbtn: Convert to pure ACPI driver

RFC PATCH In review Andy Shevchenko 5 participants 13 messages Jan 31, 2018 – Feb 7, 2018 1 message not in the archive x86 platform drivers

lore
Share & tools
lore
#0 lore raw

RFC PATCH 1/1 platform/x86: intel-vbtn: Convert to pure ACPI driver

Remove code duplication by converting driver to pure ACPI one.

---
- not tested

 drivers/platform/x86/intel-vbtn.c | 93 +++++++++------------------------------
 1 file changed, 22 insertions(+), 71 deletions(-)
drivers/platform/x86/intel-vbtn.c +22 -71
index b703d6f5b099..7201effdf37a 100644
@@ -24,6 +24,7 @@ static const struct acpi_device_id intel_vbtn_ids[] = {
 	{"INT33D6", 0},
 	{"", 0},
 };
+MODULE_DEVICE_TABLE(acpi, intel_vbtn_ids);
 
 /* In theory, these are HID usages. */
 static const struct key_entry intel_vbtn_keymap[] = {
@@ -47,9 +48,9 @@ struct intel_vbtn_priv {
 	bool wakeup_mode;
 };
 
-static int intel_vbtn_input_setup(struct platform_device *device)
+static int intel_vbtn_input_setup(struct acpi_device *device)
 {
-	struct intel_vbtn_priv *priv = dev_get_drvdata(&device->dev);
+	struct intel_vbtn_priv *priv = acpi_driver_data(device);
 	int ret;
 
 	priv->input_dev = devm_input_allocate_device(&device->dev);
@@ -60,17 +61,15 @@ static int intel_vbtn_input_setup(struct platform_device *device)
 	if (ret)
 		return ret;
 
-	priv->input_dev->dev.parent = &device->dev;
 	priv->input_dev->name = "Intel Virtual Button driver";
 	priv->input_dev->id.bustype = BUS_HOST;
 
 	return input_register_device(priv->input_dev);
 }
 
-static void notify_handler(acpi_handle handle, u32 event, void *context)
+static void notify_handler(struct acpi_device *device, u32 event)
 {
-	struct platform_device *device = context;
-	struct intel_vbtn_priv *priv = dev_get_drvdata(&device->dev);
+	struct intel_vbtn_priv *priv = acpi_driver_data(device);
 	unsigned int val = !(event & 1); /* Even=press, Odd=release */
 	const struct key_entry *ke_rel;
 	bool autorelease;
@@ -97,10 +96,10 @@ static void notify_handler(acpi_handle handle, u32 event, void *context)
 	dev_dbg(&device->dev, "unknown event index 0x%x\n", event);
 }
 
-static int intel_vbtn_probe(struct platform_device *device)
+static int intel_vbtn_add(struct acpi_device *device)
 {
 	struct acpi_buffer vgbs_output = { ACPI_ALLOCATE_BUFFER, NULL };
-	acpi_handle handle = ACPI_HANDLE(&device->dev);
+	acpi_handle handle = acpi_device_handle(device);
 	struct intel_vbtn_priv *priv;
 	acpi_status status;
 	int err;
@@ -114,11 +113,11 @@ static int intel_vbtn_probe(struct platform_device *device)
 	priv = devm_kzalloc(&device->dev, sizeof(*priv), GFP_KERNEL);
 	if (!priv)
 		return -ENOMEM;
-	dev_set_drvdata(&device->dev, priv);
+	device->driver_data = priv;
 
 	err = intel_vbtn_input_setup(device);
 	if (err) {
-		pr_err("Failed to setup Intel Virtual Button\n");
+		dev_warn(&device->dev, "Failed to setup Intel Virtual Button\n");
 		return err;
 	}
 
@@ -139,33 +138,14 @@ static int intel_vbtn_probe(struct platform_device *device)
 
 	kfree(vgbs_output.pointer);
 
-	status = acpi_install_notify_handler(handle,
-					     ACPI_DEVICE_NOTIFY,
-					     notify_handler,
-					     device);
-	if (ACPI_FAILURE(status))
-		return -EBUSY;
-
 	device_init_wakeup(&device->dev, true);
 	return 0;
 }
 
-static int intel_vbtn_remove(struct platform_device *device)
-{
-	acpi_handle handle = ACPI_HANDLE(&device->dev);
-
-	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 int intel_vbtn_pm_prepare(struct device *dev)
 {
-	struct intel_vbtn_priv *priv = dev_get_drvdata(dev);
+	struct acpi_device *device = to_acpi_device(dev);
+	struct intel_vbtn_priv *priv = acpi_driver_data(device);
 
 	priv->wakeup_mode = true;
 	return 0;
@@ -173,7 +153,8 @@ static int intel_vbtn_pm_prepare(struct device *dev)
 
 static int intel_vbtn_pm_resume(struct device *dev)
 {
-	struct intel_vbtn_priv *priv = dev_get_drvdata(dev);
+	struct acpi_device *device = to_acpi_device(dev);
+	struct intel_vbtn_priv *priv = acpi_driver_data(device);
 
 	priv->wakeup_mode = false;
 	return 0;
@@ -186,46 +167,16 @@ static const struct dev_pm_ops intel_vbtn_pm_ops = {
 	.thaw = intel_vbtn_pm_resume,
 };
 
-static struct platform_driver intel_vbtn_pl_driver = {
+static struct acpi_driver intel_vbtn_driver = {
+	.name = "intel-vbtn",
+	.class = ACPI_BUTTON_CLASS,
+	.ids = intel_vbtn_ids,
+	.ops = {
+		.add = intel_vbtn_add,
+		.notify = notify_handler,
+	},
 	.driver = {
-		.name = "intel-vbtn",
-		.acpi_match_table = intel_vbtn_ids,
 		.pm = &intel_vbtn_pm_ops,
 	},
-	.probe = intel_vbtn_probe,
-	.remove = intel_vbtn_remove,
 };
-MODULE_DEVICE_TABLE(acpi, intel_vbtn_ids);
-
-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, NULL))
-			dev_info(&dev->dev,
-				 "intel-vbtn: created platform device\n");
-
-	return AE_OK;
-}
-
-static int __init intel_vbtn_init(void)
-{
-	acpi_walk_namespace(ACPI_TYPE_DEVICE, ACPI_ROOT_OBJECT,
-			    ACPI_UINT32_MAX, check_acpi_dev, NULL,
-			    (void *)intel_vbtn_ids, NULL);
-
-	return platform_driver_register(&intel_vbtn_pl_driver);
-}
-module_init(intel_vbtn_init);
-
-static void __exit intel_vbtn_exit(void)
-{
-	platform_driver_unregister(&intel_vbtn_pl_driver);
-}
-module_exit(intel_vbtn_exit);
+module_acpi_driver(intel_vbtn_driver);
Signature · 1 line
2.15.1
#1 lore raw
Quoted 1 line · Andy Shevchenko

Remove code duplication by converting driver to pure ACPI one.

+Rafael

I don't see any reason *not* to do this, it seems the acpi driver model provides some of the boilerplate stuff we were doing manually before as a platform device. I'm scratching my head wondering why we did this as a platform-driver in the first place now...

Quoted 184 lines

---
- not tested

 drivers/platform/x86/intel-vbtn.c | 93 +++++++++------------------------------
 1 file changed, 22 insertions(+), 71 deletions(-)
drivers/platform/x86/intel-vbtn.c +22 -71
index b703d6f5b099..7201effdf37a 100644
@@ -24,6 +24,7 @@ static const struct acpi_device_id intel_vbtn_ids[] = {
 	{"INT33D6", 0},
 	{"", 0},
 };
+MODULE_DEVICE_TABLE(acpi, intel_vbtn_ids);
 
 /* In theory, these are HID usages. */
 static const struct key_entry intel_vbtn_keymap[] = {
@@ -47,9 +48,9 @@ struct intel_vbtn_priv {
 	bool wakeup_mode;
 };
 
-static int intel_vbtn_input_setup(struct platform_device *device)
+static int intel_vbtn_input_setup(struct acpi_device *device)
 {
-	struct intel_vbtn_priv *priv = dev_get_drvdata(&device->dev);
+	struct intel_vbtn_priv *priv = acpi_driver_data(device);
 	int ret;
 
 	priv->input_dev = devm_input_allocate_device(&device->dev);
@@ -60,17 +61,15 @@ static int intel_vbtn_input_setup(struct platform_device *device)
 	if (ret)
 		return ret;
 
-	priv->input_dev->dev.parent = &device->dev;
 	priv->input_dev->name = "Intel Virtual Button driver";
 	priv->input_dev->id.bustype = BUS_HOST;
 
 	return input_register_device(priv->input_dev);
 }
 
-static void notify_handler(acpi_handle handle, u32 event, void *context)
+static void notify_handler(struct acpi_device *device, u32 event)
 {
-	struct platform_device *device = context;
-	struct intel_vbtn_priv *priv = dev_get_drvdata(&device->dev);
+	struct intel_vbtn_priv *priv = acpi_driver_data(device);
 	unsigned int val = !(event & 1); /* Even=press, Odd=release */
 	const struct key_entry *ke_rel;
 	bool autorelease;
@@ -97,10 +96,10 @@ static void notify_handler(acpi_handle handle, u32 event, void *context)
 	dev_dbg(&device->dev, "unknown event index 0x%x\n", event);
 }
 
-static int intel_vbtn_probe(struct platform_device *device)
+static int intel_vbtn_add(struct acpi_device *device)
 {
 	struct acpi_buffer vgbs_output = { ACPI_ALLOCATE_BUFFER, NULL };
-	acpi_handle handle = ACPI_HANDLE(&device->dev);
+	acpi_handle handle = acpi_device_handle(device);
 	struct intel_vbtn_priv *priv;
 	acpi_status status;
 	int err;
@@ -114,11 +113,11 @@ static int intel_vbtn_probe(struct platform_device *device)
 	priv = devm_kzalloc(&device->dev, sizeof(*priv), GFP_KERNEL);
 	if (!priv)
 		return -ENOMEM;
-	dev_set_drvdata(&device->dev, priv);
+	device->driver_data = priv;
 
 	err = intel_vbtn_input_setup(device);
 	if (err) {
-		pr_err("Failed to setup Intel Virtual Button\n");
+		dev_warn(&device->dev, "Failed to setup Intel Virtual Button\n");
 		return err;
 	}
 
@@ -139,33 +138,14 @@ static int intel_vbtn_probe(struct platform_device *device)
 
 	kfree(vgbs_output.pointer);
 
-	status = acpi_install_notify_handler(handle,
-					     ACPI_DEVICE_NOTIFY,
-					     notify_handler,
-					     device);
-	if (ACPI_FAILURE(status))
-		return -EBUSY;
-
 	device_init_wakeup(&device->dev, true);
 	return 0;
 }
 
-static int intel_vbtn_remove(struct platform_device *device)
-{
-	acpi_handle handle = ACPI_HANDLE(&device->dev);
-
-	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 int intel_vbtn_pm_prepare(struct device *dev)
 {
-	struct intel_vbtn_priv *priv = dev_get_drvdata(dev);
+	struct acpi_device *device = to_acpi_device(dev);
+	struct intel_vbtn_priv *priv = acpi_driver_data(device);
 
 	priv->wakeup_mode = true;
 	return 0;
@@ -173,7 +153,8 @@ static int intel_vbtn_pm_prepare(struct device *dev)
 
 static int intel_vbtn_pm_resume(struct device *dev)
 {
-	struct intel_vbtn_priv *priv = dev_get_drvdata(dev);
+	struct acpi_device *device = to_acpi_device(dev);
+	struct intel_vbtn_priv *priv = acpi_driver_data(device);
 
 	priv->wakeup_mode = false;
 	return 0;
@@ -186,46 +167,16 @@ static const struct dev_pm_ops intel_vbtn_pm_ops = {
 	.thaw = intel_vbtn_pm_resume,
 };
 
-static struct platform_driver intel_vbtn_pl_driver = {
+static struct acpi_driver intel_vbtn_driver = {
+	.name = "intel-vbtn",
+	.class = ACPI_BUTTON_CLASS,
+	.ids = intel_vbtn_ids,
+	.ops = {
+		.add = intel_vbtn_add,
+		.notify = notify_handler,
+	},
 	.driver = {
-		.name = "intel-vbtn",
-		.acpi_match_table = intel_vbtn_ids,
 		.pm = &intel_vbtn_pm_ops,
 	},
-	.probe = intel_vbtn_probe,
-	.remove = intel_vbtn_remove,
 };
-MODULE_DEVICE_TABLE(acpi, intel_vbtn_ids);
-
-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, NULL))
-			dev_info(&dev->dev,
-				 "intel-vbtn: created platform device\n");
-
-	return AE_OK;
-}
-
-static int __init intel_vbtn_init(void)
-{
-	acpi_walk_namespace(ACPI_TYPE_DEVICE, ACPI_ROOT_OBJECT,
-			    ACPI_UINT32_MAX, check_acpi_dev, NULL,
-			    (void *)intel_vbtn_ids, NULL);
-
-	return platform_driver_register(&intel_vbtn_pl_driver);
-}
-module_init(intel_vbtn_init);
-
-static void __exit intel_vbtn_exit(void)
-{
-	platform_driver_unregister(&intel_vbtn_pl_driver);
-}
-module_exit(intel_vbtn_exit);
+module_acpi_driver(intel_vbtn_driver);
Signature · 1 line
2.15.1
Signature · 2 lines
Darren Hart
VMware Open Source Technology Center
#2 lore raw
Quoted 3 lines · Darren Hart
Quoted 1 line · Andy Shevchenko

Remove code duplication by converting driver to pure ACPI one.

I have fixed couple issues, but...

Quoted 9 lines

+Rafael

I don't see any reason *not* to do this, it seems the acpi driver model provides
some of the boilerplate stuff we were doing manually before as a platform
device. I'm scratching my head wondering why we did this as a platform-driver in
the first place now...

...I still keep in my review queue in a hope that someone can actually test this on real hardware.

Signature · 2 lines
Andy Shevchenko <[email protected]>
Intel Finland Oy
#3 lore raw
Quoted 28 lines

-----Original Message-----
From: [email protected] [mailto:platform-driver-x86- [email protected]] On Behalf Of Andy Shevchenko
Sent: Monday, February 5, 2018 10:18 AM
To: Darren Hart <[email protected]>
Cc: [email protected]; AceLan Kao <[email protected]>; [email protected]; [email protected]; Limonciello, Mario <[email protected]>; Rafael Wysocki <[email protected]>
Subject: Re: [RFC, PATCH v1] platform/x86: intel-vbtn: Convert to pure ACPI driver

Quoted 3 lines · Darren Hart
Quoted 1 line · Andy Shevchenko

Remove code duplication by converting driver to pure ACPI one.

I have fixed couple issues, but...

Quoted 9 lines

+Rafael

I don't see any reason *not* to do this, it seems the acpi driver model provides
some of the boilerplate stuff we were doing manually before as a platform
device. I'm scratching my head wondering why we did this as a platform-driver in
the first place now...

Quoted 6 lines

...I still keep in my review queue in a hope that someone can actually test this on real hardware.

--
Andy Shevchenko <[email protected]> Intel Finland Oy

Andy,

I'm pretty sure XPS 9365 can test this. I was going to try but
https://patchwork.kernel.org/patch/10194503/ was not able to compile on top of platform-x86/for-next.
That was the only one I saw at patchwork.

drivers/platform/x86/intel-vbtn.c:172:11: error: ‘ACPI_BUTTON_CLASS’ undeclared here (not in a function)
.class = ACPI_BUTTON_CLASS,

           ^

drivers/platform/x86/intel-vbtn.c:178:2: error: unknown field ‘driver’ specified in initializer
.driver = {
^
scripts/Makefile.build:316: recipe for target 'drivers/platform/x86/intel-vbtn.o' failed make[5]: *** [drivers/platform/x86/intel-vbtn.o] Error 1

is that different than the one in your review queue?

Thanks,

A message that is not in the archive (sent off-list or not yet received): CAHp75VeFSuQYu67DdnJDjZDoaLZiPuo7ag7Lz0q6eSCDjayobw@mail.gmail.com

#5 lore raw
Quoted 2 lines

Yeah, that's fixed in the branch.
P.S. Apologies for top-posting, wrote from phone in order to not waste time

Andy,

I tested your review branch on XPS 9365 which uses power button through intel-vbtn. I confirmed it worked properly both with usage in OS and during S2I.

Thanks,

#6 lore raw
In reply to a message not in the archive
Quoted 36 lines · CET Andy Shevchenko

Yeah, that's fixed in the branch.

P.S. Apologies for top-posting, wrote from phone in order to not waste time

Quoted 30 lines · On Tuesday, February 6, 2018, <[email protected]> wrote:
Quoted 2 lines

-----Original Message-----
From: [email protected] [mailto:

platform-driver-x86-

Quoted 6 lines

[email protected]] On Behalf Of Andy Shevchenko
Sent: Monday, February 5, 2018 10:18 AM
To: Darren Hart <[email protected]>
Cc: [email protected]; AceLan Kao <[email protected]>; [email protected]; [email protected]; Limonciello, Mario <[email protected]>;

Rafael

Quoted 2 lines

Wysocki <[email protected]>
Subject: Re: [RFC, PATCH v1] platform/x86: intel-vbtn: Convert to pure

ACPI driver

Quoted 17 lines · Darren Hart
Quoted 3 lines · Andy Shevchenko
Quoted 1 line

Remove code duplication by converting driver to pure ACPI one.

I have fixed couple issues, but...

Quoted 9 lines

+Rafael

I don't see any reason *not* to do this, it seems the acpi driver model provides
some of the boilerplate stuff we were doing manually before as a platform
device. I'm scratching my head wondering why we did this as a platform-driver in
the first place now...

It looks like I have overlooked this.

Generally speaking, binding drivers to ACPI device objects is a bad idea, because it is sort of what binding drivers to struct device_node in OF would be. It was an unfortunate design choice for struct acpi_device to be based on struct device in the first place, but undoing it is not worth the effort and pain.

In some instances (PCI, PNP, I2C, SPI etc) there is a device object to bind a driver to already and the corresponding struct acpi_device is the "companion" of it. In those cases binding a driver to the struct acpi_device itself clearly doesn't make sense.

In the cases like the $subject one it is less clear, but for symmetry and general confusion avoidance it is better to have a struct device representing the "physical" thing *in* *addition* to the struct acpi_device and use the latter as the "companion" of it. That's why we have the platform device in this particular case.

Please drop this patch, it doesn't go in the right direction IMO.

Thanks,
Rafael

#7 lore raw
Quoted 51 lines · CET Rafael J. Wysocki
Quoted 36 lines · CET Andy Shevchenko

Yeah, that's fixed in the branch.

P.S. Apologies for top-posting, wrote from phone in order to not waste time

Quoted 30 lines · On Tuesday, February 6, 2018, <[email protected]> wrote:
Quoted 2 lines

-----Original Message-----
From: [email protected] [mailto:

platform-driver-x86-

Quoted 6 lines

[email protected]] On Behalf Of Andy Shevchenko
Sent: Monday, February 5, 2018 10:18 AM
To: Darren Hart <[email protected]>
Cc: [email protected]; AceLan Kao <[email protected]>; [email protected]; [email protected]; Limonciello, Mario <[email protected]>;

Rafael

Quoted 2 lines

Wysocki <[email protected]>
Subject: Re: [RFC, PATCH v1] platform/x86: intel-vbtn: Convert to pure

ACPI driver

Quoted 17 lines · Darren Hart
Quoted 3 lines · Andy Shevchenko
Quoted 1 line

Remove code duplication by converting driver to pure ACPI one.

I have fixed couple issues, but...

Quoted 9 lines

+Rafael

I don't see any reason *not* to do this, it seems the acpi driver model provides
some of the boilerplate stuff we were doing manually before as a platform
device. I'm scratching my head wondering why we did this as a platform-driver in
the first place now...

It looks like I have overlooked this.

Generally speaking, binding drivers to ACPI device objects is a bad idea, because it is sort of what binding drivers to struct device_node in OF would be. It was an unfortunate design choice for struct acpi_device to be based on struct device in the first place, but undoing it is not worth the effort and pain.

In some instances (PCI, PNP, I2C, SPI etc) there is a device object to bind a driver to already and the corresponding struct acpi_device is the "companion" of it. In those cases binding a driver to the struct acpi_device itself clearly doesn't make sense.

Quoted 7 lines

In the cases like the $subject one it is less clear, but for symmetry and general confusion avoidance it is better to have a struct device representing the "physical" thing *in* *addition* to the struct acpi_device and use the latter as the "companion" of it. That's why we have the platform device in this particular case.

Please drop this patch, it doesn't go in the right direction IMO.

BTW, please CC patches with ACPI material to linux-acpi. They are less likely to be overlooked by me then and the others on that list may have opinions on them too.

#8 lore raw
Quoted 3 lines · On Wednesday 07 February 2018 10:47:57 Rafael J. Wysocki wrote:

BTW, please CC patches with ACPI material to linux-acpi. They are less likely to be overlooked by me then and the others on that list may have opinions on them too.

Maybe we can update MAINTAINERS file for acpi based platform-x86 drivers to include linux-acpi list? Darren, Andy, any idea?

Signature · 2 lines
Pali Rohár
[email protected]
#9 lore raw
Quoted 10 lines

-----Original Message-----
From: Pali Rohár [mailto:[email protected]]
Sent: Wednesday, February 7, 2018 3:55 AM
To: Rafael J. Wysocki <[email protected]>
Cc: Andy Shevchenko <[email protected]>; Limonciello, Mario <[email protected]>; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Subject: Re: [RFC, PATCH v1] platform/x86: intel-vbtn: Convert to pure ACPI driver

Quoted 8 lines
Quoted 3 lines · On Wednesday 07 February 2018 10:47:57 Rafael J. Wysocki wrote:

BTW, please CC patches with ACPI material to linux-acpi. They are less likely to be overlooked by me then and the others on that list may have opinions on them too.

Maybe we can update MAINTAINERS file for acpi based platform-x86 drivers to include linux-acpi list? Darren, Andy, any idea?

I would also wonder if the intel-vbtn driver should still live in platform-x86 as it's now pure ACPI driver with this change Andy proposed.

Btw Andy for your v2 with your fixups that were in your review branch feel free to add:

When you resubmit it.

#10 lore raw
Quoted 27 lines · On Wed, 2018-02-07 at 13:08 +0000, [email protected] wrote:
Quoted 18 lines

-----Original Message-----
From: Pali Rohár [mailto:[email protected]]
Sent: Wednesday, February 7, 2018 3:55 AM
To: Rafael J. Wysocki <[email protected]>
Cc: Andy Shevchenko <[email protected]>; Limonciello, Mario <[email protected]>; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Subject: Re: [RFC, PATCH v1] platform/x86: intel-vbtn: Convert to pure ACPI driver

Quoted 5 lines · On Wednesday 07 February 2018 10:47:57 Rafael J. Wysocki wrote:

BTW, please CC patches with ACPI material to linux-acpi. They are less likely
to be overlooked by me then and the others on that list may have opinions on
them too.

Quoted 4 lines

Maybe we can update MAINTAINERS file for acpi based platform-x86 drivers
to include linux-acpi list? Darren, Andy, any idea?

I would also wonder if the intel-vbtn driver should still live in platform-x86 as it's
now pure ACPI driver with this change Andy proposed.

Quoted 6 lines

Btw Andy for your v2 with your fixups that were in your review branch feel free
to add:

When you resubmit it.

Mario, Rafael clarifies the matter and NACKed this one (which kinda was expected by me, that's why RFC was in the first place).

I'm going to drop this patch from my queue.

Signature · 2 lines
Andy Shevchenko <[email protected]>
Intel Finland Oy
#11 lore raw
Quoted 33 lines · CET Andy Shevchenko
Quoted 32 lines · On Wed, 2018-02-07 at 13:08 +0000, [email protected] wrote:
Quoted 18 lines

-----Original Message-----
From: Pali Rohár [mailto:[email protected]]
Sent: Wednesday, February 7, 2018 3:55 AM
To: Rafael J. Wysocki <[email protected]>
Cc: Andy Shevchenko <[email protected]>; Limonciello, Mario <[email protected]>; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Subject: Re: [RFC, PATCH v1] platform/x86: intel-vbtn: Convert to pure ACPI driver

Quoted 5 lines · On Wednesday 07 February 2018 10:47:57 Rafael J. Wysocki wrote:

BTW, please CC patches with ACPI material to linux-acpi. They are less likely
to be overlooked by me then and the others on that list may have opinions on
them too.

Quoted 4 lines

Maybe we can update MAINTAINERS file for acpi based platform-x86 drivers
to include linux-acpi list? Darren, Andy, any idea?

I would also wonder if the intel-vbtn driver should still live in platform-x86 as it's
now pure ACPI driver with this change Andy proposed.

Btw Andy for your v2 with your fixups that were in your review branch feel free
to add:

Quoted 6 lines
Quoted 1 line

When you resubmit it.

Mario, Rafael clarifies the matter and NACKed this one (which kinda was expected by me, that's why RFC was in the first place).

I'm going to drop this patch from my queue.

Thanks!

#12 lore raw
In reply to Pali Rohár
Quoted 7 lines · CET Pali Rohár
Quoted 3 lines · On Wednesday 07 February 2018 10:47:57 Rafael J. Wysocki wrote:

BTW, please CC patches with ACPI material to linux-acpi. They are less likely to be overlooked by me then and the others on that list may have opinions on them too.

Maybe we can update MAINTAINERS file for acpi based platform-x86 drivers to include linux-acpi list? Darren, Andy, any idea?

You could, but generally, if you have ACPI material in your patch, CCing linux-acpi won't hurt.

#13 lore raw
In reply to Darren Hart
Quoted 9 lines · CET Darren Hart
Quoted 1 line · Andy Shevchenko

Remove code duplication by converting driver to pure ACPI one.

+Rafael

I don't see any reason *not* to do this, it seems the acpi driver model provides some of the boilerplate stuff we were doing manually before as a platform device. I'm scratching my head wondering why we did this as a platform-driver in the first place now...

We talked about it quite a bit before, I guess it needs to be documented somewhere, finally. :-)

If you have a PCI device, you have a struct pci_dev representing it and then there may be a "companion" struct acpi_device associated with it. In that case you clearly bind a driver to the struct pci_dev.

This case is really analogous except that the platform bus type is sort of "fake" and it doesn't appear to be necessary to use it.

Thanks,
Rafael

Quoted 183 lines
Quoted 181 lines

---
- not tested

 drivers/platform/x86/intel-vbtn.c | 93 +++++++++------------------------------
 1 file changed, 22 insertions(+), 71 deletions(-)
drivers/platform/x86/intel-vbtn.c +22 -71
index b703d6f5b099..7201effdf37a 100644
@@ -24,6 +24,7 @@ static const struct acpi_device_id intel_vbtn_ids[] = {
 	{"INT33D6", 0},
 	{"", 0},
 };
+MODULE_DEVICE_TABLE(acpi, intel_vbtn_ids);
 
 /* In theory, these are HID usages. */
 static const struct key_entry intel_vbtn_keymap[] = {
@@ -47,9 +48,9 @@ struct intel_vbtn_priv {
 	bool wakeup_mode;
 };
 
-static int intel_vbtn_input_setup(struct platform_device *device)
+static int intel_vbtn_input_setup(struct acpi_device *device)
 {
-	struct intel_vbtn_priv *priv = dev_get_drvdata(&device->dev);
+	struct intel_vbtn_priv *priv = acpi_driver_data(device);
 	int ret;
 
 	priv->input_dev = devm_input_allocate_device(&device->dev);
@@ -60,17 +61,15 @@ static int intel_vbtn_input_setup(struct platform_device *device)
 	if (ret)
 		return ret;
 
-	priv->input_dev->dev.parent = &device->dev;
 	priv->input_dev->name = "Intel Virtual Button driver";
 	priv->input_dev->id.bustype = BUS_HOST;
 
 	return input_register_device(priv->input_dev);
 }
 
-static void notify_handler(acpi_handle handle, u32 event, void *context)
+static void notify_handler(struct acpi_device *device, u32 event)
 {
-	struct platform_device *device = context;
-	struct intel_vbtn_priv *priv = dev_get_drvdata(&device->dev);
+	struct intel_vbtn_priv *priv = acpi_driver_data(device);
 	unsigned int val = !(event & 1); /* Even=press, Odd=release */
 	const struct key_entry *ke_rel;
 	bool autorelease;
@@ -97,10 +96,10 @@ static void notify_handler(acpi_handle handle, u32 event, void *context)
 	dev_dbg(&device->dev, "unknown event index 0x%x\n", event);
 }
 
-static int intel_vbtn_probe(struct platform_device *device)
+static int intel_vbtn_add(struct acpi_device *device)
 {
 	struct acpi_buffer vgbs_output = { ACPI_ALLOCATE_BUFFER, NULL };
-	acpi_handle handle = ACPI_HANDLE(&device->dev);
+	acpi_handle handle = acpi_device_handle(device);
 	struct intel_vbtn_priv *priv;
 	acpi_status status;
 	int err;
@@ -114,11 +113,11 @@ static int intel_vbtn_probe(struct platform_device *device)
 	priv = devm_kzalloc(&device->dev, sizeof(*priv), GFP_KERNEL);
 	if (!priv)
 		return -ENOMEM;
-	dev_set_drvdata(&device->dev, priv);
+	device->driver_data = priv;
 
 	err = intel_vbtn_input_setup(device);
 	if (err) {
-		pr_err("Failed to setup Intel Virtual Button\n");
+		dev_warn(&device->dev, "Failed to setup Intel Virtual Button\n");
 		return err;
 	}
 
@@ -139,33 +138,14 @@ static int intel_vbtn_probe(struct platform_device *device)
 
 	kfree(vgbs_output.pointer);
 
-	status = acpi_install_notify_handler(handle,
-					     ACPI_DEVICE_NOTIFY,
-					     notify_handler,
-					     device);
-	if (ACPI_FAILURE(status))
-		return -EBUSY;
-
 	device_init_wakeup(&device->dev, true);
 	return 0;
 }
 
-static int intel_vbtn_remove(struct platform_device *device)
-{
-	acpi_handle handle = ACPI_HANDLE(&device->dev);
-
-	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 int intel_vbtn_pm_prepare(struct device *dev)
 {
-	struct intel_vbtn_priv *priv = dev_get_drvdata(dev);
+	struct acpi_device *device = to_acpi_device(dev);
+	struct intel_vbtn_priv *priv = acpi_driver_data(device);
 
 	priv->wakeup_mode = true;
 	return 0;
@@ -173,7 +153,8 @@ static int intel_vbtn_pm_prepare(struct device *dev)
 
 static int intel_vbtn_pm_resume(struct device *dev)
 {
-	struct intel_vbtn_priv *priv = dev_get_drvdata(dev);
+	struct acpi_device *device = to_acpi_device(dev);
+	struct intel_vbtn_priv *priv = acpi_driver_data(device);
 
 	priv->wakeup_mode = false;
 	return 0;
@@ -186,46 +167,16 @@ static const struct dev_pm_ops intel_vbtn_pm_ops = {
 	.thaw = intel_vbtn_pm_resume,
 };
 
-static struct platform_driver intel_vbtn_pl_driver = {
+static struct acpi_driver intel_vbtn_driver = {
+	.name = "intel-vbtn",
+	.class = ACPI_BUTTON_CLASS,
+	.ids = intel_vbtn_ids,
+	.ops = {
+		.add = intel_vbtn_add,
+		.notify = notify_handler,
+	},
 	.driver = {
-		.name = "intel-vbtn",
-		.acpi_match_table = intel_vbtn_ids,
 		.pm = &intel_vbtn_pm_ops,
 	},
-	.probe = intel_vbtn_probe,
-	.remove = intel_vbtn_remove,
 };
-MODULE_DEVICE_TABLE(acpi, intel_vbtn_ids);
-
-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, NULL))
-			dev_info(&dev->dev,
-				 "intel-vbtn: created platform device\n");
-
-	return AE_OK;
-}
-
-static int __init intel_vbtn_init(void)
-{
-	acpi_walk_namespace(ACPI_TYPE_DEVICE, ACPI_ROOT_OBJECT,
-			    ACPI_UINT32_MAX, check_acpi_dev, NULL,
-			    (void *)intel_vbtn_ids, NULL);
-
-	return platform_driver_register(&intel_vbtn_pl_driver);
-}
-module_init(intel_vbtn_init);
-
-static void __exit intel_vbtn_exit(void)
-{
-	platform_driver_unregister(&intel_vbtn_pl_driver);
-}
-module_exit(intel_vbtn_exit);
+module_acpi_driver(intel_vbtn_driver);