Skip to contentSkip to conversation

mfd: Make use of software nodes

PATCH In review Andy Shevchenko 5 participants 24 messages Jun 8, 2020 – Jun 24, 2020

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

cover mfd: Make use of software nodes

Some devices would need to have a hierarchy of properties and child nodes passed to the child or children of MFD. For such case we may utilize software nodes, which is superior on device properties.

Add support of software nodes to MFD core and convert one driver to show how it looks like. This allows to get rid of legacy platform data.

The change has been tested on Intel Galileo Gen 2.

Andy Shevchenko (5):
gpio: dwapb: Replace irq_shared flag with fwnode type check
gpio: dwapb: Read GPIO base from snps,gpio-base property
mfd: intel_quark_i2c_gpio: Convert to use software nodes
gpio: dwapb: Get rid of legacy platform data
gpio: dwapb: Define magic number for IRQ and GPIO lines

Heikki Krogerus (1):
mfd: core: Propagate software node group to the sub devices

 drivers/gpio/gpio-dwapb.c                | 44 +++++++++-------
 drivers/mfd/intel_quark_i2c_gpio.c       | 64 +++++++++++-------------
 drivers/mfd/mfd-core.c                   | 31 ++++++++++--
 include/linux/mfd/core.h                 |  3 ++
 include/linux/platform_data/gpio-dwapb.h | 23 ---------
 5 files changed, 85 insertions(+), 80 deletions(-)
 delete mode 100644 include/linux/platform_data/gpio-dwapb.h
Signature · 1 line
2.27.0.rc2
#1 lore raw

PATCH 1/6 gpio: dwapb: Replace irq_shared flag with fwnode type check

Shared IRQ is only enabled for ACPI platforms or pure platform drivers, there is no need to have a special flag for that, since we simple can test port fwnode to be type of OF.

While at it, drop duplicate declaration of loop variable.

---
 drivers/gpio/gpio-dwapb.c                | 5 +----
 drivers/mfd/intel_quark_i2c_gpio.c       | 1 -
 include/linux/platform_data/gpio-dwapb.h | 1 -
 3 files changed, 1 insertion(+), 6 deletions(-)
drivers/gpio/gpio-dwapb.c +1 -4
index 1d8d55bd63aa..a7ca72086511 100644
@@ -418,9 +418,7 @@ static void dwapb_configure_irqs(struct dwapb_gpio *gpio,
 	irq_gc->chip_types[1].type = IRQ_TYPE_EDGE_BOTH;
 	irq_gc->chip_types[1].handler = handle_edge_irq;
 
-	if (!pp->irq_shared) {
-		int i;
-
+	if (to_of_node(pp->fwnode)) {
 		for (i = 0; i < pp->ngpio; i++) {
 			if (pp->irq[i])
 				irq_set_chained_handler_and_data(pp->irq[i],
@@ -601,7 +599,6 @@ static struct dwapb_platform_data *dwapb_gpio_get_pdata(struct device *dev)
 			pp->ngpio = 32;
 		}
 
-		pp->irq_shared	= false;
 		pp->gpio_base	= -1;
 
 		/*
drivers/mfd/intel_quark_i2c_gpio.c +0 -1
index 84ca7902e1df..06b538dd124c 100644
@@ -216,7 +216,6 @@ static int intel_quark_gpio_setup(struct pci_dev *pdev, struct mfd_cell *cell)
 	pdata->properties->ngpio	= INTEL_QUARK_MFD_NGPIO;
 	pdata->properties->gpio_base	= INTEL_QUARK_MFD_GPIO_BASE;
 	pdata->properties->irq[0]	= pdev->irq;
-	pdata->properties->irq_shared	= true;
 
 	cell->platform_data = pdata;
 	cell->pdata_size = sizeof(*pdata);
include/linux/platform_data/gpio-dwapb.h +0 -1
index ff1be737bad6..a63010b6e898 100644
@@ -12,7 +12,6 @@ struct dwapb_port_property {
 	unsigned int	ngpio;
 	unsigned int	gpio_base;
 	int		irq[32];
-	bool		irq_shared;
 };
 
 struct dwapb_platform_data {
Signature · 1 line
2.27.0.rc2
#2 lore raw

PATCH 2/6 gpio: dwapb: Read GPIO base from snps,gpio-base property

In reply to Andy Shevchenko

For backward compatibility with some legacy devices, introduce a new property snps,gpio-base to read GPIO base. Don't advertise to discourage users from utilizing it.

---
 drivers/gpio/gpio-dwapb.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/gpio/gpio-dwapb.c +2 -1
index a7ca72086511..e3d7589434eb 100644
@@ -599,7 +599,8 @@ static struct dwapb_platform_data *dwapb_gpio_get_pdata(struct device *dev)
 			pp->ngpio = 32;
 		}
 
-		pp->gpio_base	= -1;
+		if (fwnode_property_read_u32(fwnode, "snps,gpio-base", &pp->gpio_base))
+			pp->gpio_base = -1;
 
 		/*
 		 * Only port A can provide interrupts in all configurations of
Signature · 1 line
2.27.0.rc2
#3 lore raw

Re: [PATCH v1 2/6] gpio: dwapb: Read GPIO base from snps,gpio-base property

Quoted 5 lines · Andy Shevchenko

For backward compatibility with some legacy devices, introduce a new property snps,gpio-base to read GPIO base. Don't advertise to discourage users from utilizing it.

I usually have a very negative gut reaction to any attempts to hardcode the heavily deprecated use of the gpio_base for the global GPIO numberspace. The reason is clear from drivers/gpio/TODO I think.

I need a really good explanation why this is needed, the only reason people have been pushing this in the past is "oh we are using the sysfs and we don't wanna change the GPIO numbers in our scripts" which I really want to push back on now that we have the chardev and the libgpiod utils.

If this is needed for something driver internal, it should stay in a driver-local
variable.

Yours,
Linus Walleij

#4 lore raw

Re: [PATCH v1 2/6] gpio: dwapb: Read GPIO base from snps,gpio-base property

Quoted 13 lines · Linus Walleij
Quoted 5 lines · Andy Shevchenko

For backward compatibility with some legacy devices, introduce a new property snps,gpio-base to read GPIO base. Don't advertise to discourage users from utilizing it.

I usually have a very negative gut reaction to any attempts to hardcode the heavily deprecated use of the gpio_base for the global GPIO numberspace. The reason is clear from drivers/gpio/TODO I think.

Quoted 5 lines

I need a really good explanation why this is needed, the only reason people have been pushing this in the past is "oh we are using the sysfs and we don't wanna change the GPIO numbers in our scripts" which I really want to push back on now that we have the chardev and the libgpiod utils.

I understand the point, that's why this property is not being advertised.
The problem here is that on existing (legacy) platform we have to deal with an absolute numbers of GPIOs and those numbers can work if we keep the base

Quoted 3 lines

If this is needed for something driver internal, it should stay in a driver-local
variable.

It's a cross drivers variable. The purpose of this series is to unify interface in the GPIO driver how to deal with device properties (so, be agnostic from resource provider at the end).

Everything so far so good, but then we need a mean how not to break the platform. Because dropping the base in this case will immediately make a regression.

We can re-use 'gpio-base' as gpio-mockup is using it silently. So, in such case we won't add any deviation to the property namespace.

Or we can rename above to linux,gpio-base and re-use it here with this name to explicitly show that this is not firmware property.

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#5 lore raw

PATCH 3/6 mfd: core: Propagate software node group to the sub devices

In reply to Andy Shevchenko

From: Heikki Krogerus <[email protected]>

When ever device properties are supplied for a sub device, a software node (fwnode) is actually created and then associated with that device. By allowing the drivers to supply the complete software node group instead of just the properties in it, the drivers can take advantage of the other features the software nodes have on top of supplying the device properties.

---
 drivers/mfd/mfd-core.c   | 31 +++++++++++++++++++++++++++----
 include/linux/mfd/core.h |  3 +++
 2 files changed, 30 insertions(+), 4 deletions(-)
drivers/mfd/mfd-core.c +27 -4
index f5a73af60dd4..1a256f64dc9d 100644
@@ -173,7 +173,24 @@ static int mfd_add_device(struct device *parent, int id,
 			goto fail_alias;
 	}
 
-	if (cell->properties) {
+	/* If software node group exists, use it, otherwise try properties */
+	if (cell->node_group) {
+		struct fwnode_handle *fwnode;
+
+		ret = software_node_register_node_group(cell->node_group);
+		if (ret)
+			goto fail_alias;
+
+		/*
+		 * The very first software node in the group is related to
+		 * the device itself. The rest can be device-less children to
+		 * fulfill the case of sub nodes (LEDs, GPIO keys, etc).
+		 */
+		fwnode = software_node_fwnode(cell->node_group[0]);
+
+		/* Assign this firmware node as a secondary one of the device */
+		set_secondary_fwnode(&pdev->dev, fwnode);
+	} else if (cell->properties) {
 		ret = platform_device_add_properties(pdev, cell->properties);
 		if (ret)
 			goto fail_alias;
@@ -213,18 +230,18 @@ static int mfd_add_device(struct device *parent, int id,
 			if (has_acpi_companion(&pdev->dev)) {
 				ret = acpi_check_resource_conflict(&res[r]);
 				if (ret)
-					goto fail_alias;
+					goto fail_resources;
 			}
 		}
 	}
 
 	ret = platform_device_add_resources(pdev, res, cell->num_resources);
 	if (ret)
-		goto fail_alias;
+		goto fail_resources;
 
 	ret = platform_device_add(pdev);
 	if (ret)
-		goto fail_alias;
+		goto fail_resources;
 
 	if (cell->pm_runtime_no_callbacks)
 		pm_runtime_no_callbacks(&pdev->dev);
@@ -233,6 +250,9 @@ static int mfd_add_device(struct device *parent, int id,
 
 	return 0;
 
+fail_resources:
+	set_secondary_fwnode(&pdev->dev, NULL);
+	software_node_unregister_node_group(cell->node_group);
 fail_alias:
 	regulator_bulk_unregister_supply_alias(&pdev->dev,
 					       cell->parent_supplies,
@@ -294,6 +314,9 @@ static int mfd_remove_devices_fn(struct device *dev, void *data)
 	pdev = to_platform_device(dev);
 	cell = mfd_get_cell(pdev);
 
+	set_secondary_fwnode(&pdev->dev, NULL);
+	software_node_unregister_node_group(cell->node_group);
+
 	regulator_bulk_unregister_supply_alias(dev, cell->parent_supplies,
 					       cell->num_parent_supplies);
 
include/linux/mfd/core.h +3 -0
index d01d1299e49d..9a998568759f 100644
@@ -72,6 +72,9 @@ struct mfd_cell {
 	/* device properties passed to the sub devices drivers */
 	struct property_entry *properties;
 
+	/* Software node group for the sub device */
+	const struct software_node **node_group;
+
 	/*
 	 * Device Tree compatible string
 	 * See: Documentation/devicetree/usage-model.txt Chapter 2.2 for details
Signature · 1 line
2.27.0.rc2
#6 lore raw

Re: [PATCH v1 3/6] mfd: core: Propagate software node group to the sub devices

Quoted 8 lines · Andy Shevchenko

From: Heikki Krogerus <[email protected]>

When ever device properties are supplied for a sub device, a software node (fwnode) is actually created and then associated with that device. By allowing the drivers to supply the complete software node group instead of just the properties in it, the drivers can take advantage of the other features the software nodes have on top of supplying the device properties.

Quoted 6 lines
---
 drivers/mfd/mfd-core.c   | 31 +++++++++++++++++++++++++++----
 include/linux/mfd/core.h |  3 +++
 2 files changed, 30 insertions(+), 4 deletions(-)

I'm not sure a change to the API is justified presently (same does go for 'properties' really, but as it was only a couple of lines, it didn't seem too intrusive).

My recommendation is to handle this in-house (i.e. locally in-driver) for now. When (if) more users adopt the practice, then we should consider to draw down on line numbers and repetition and make it part of the API.

Signature · 4 lines
Lee Jones [李琼斯]
Senior Technical Lead - Developer Services
Linaro.org │ Open source software for Arm SoCs
Follow Linaro: Facebook | Twitter | Blog
#7 lore raw

Re: [PATCH v1 3/6] mfd: core: Propagate software node group to the sub devices

Quoted 17 lines · Lee Jones
Quoted 8 lines · Andy Shevchenko

From: Heikki Krogerus <[email protected]>

When ever device properties are supplied for a sub device, a software node (fwnode) is actually created and then associated with that device. By allowing the drivers to supply the complete software node group instead of just the properties in it, the drivers can take advantage of the other features the software nodes have on top of supplying the device properties.

Quoted 6 lines
---
 drivers/mfd/mfd-core.c   | 31 +++++++++++++++++++++++++++----
 include/linux/mfd/core.h |  3 +++
 2 files changed, 30 insertions(+), 4 deletions(-)
Quoted 3 lines

I'm not sure a change to the API is justified presently (same does go for 'properties' really, but as it was only a couple of lines, it didn't seem too intrusive).

This is better and comprehensive API, but I heard you.

Quoted 2 lines

My recommendation is to handle this in-house (i.e. locally in-driver) for now.

I think you understand that this is not gonna work (we need to attach fwnode to the child device before it's registration.

Quoted 3 lines

When (if) more users adopt the practice, then we should consider to draw down on line numbers and repetition and make it part of the API.

I briefly looked at the current state of affairs and found that properties are used only for MFD LPSS driver. Would the conversion of that driver to swnodes work for you?

Note, the long prospective is to get rid of platform_add_properties() API completely.

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#8 lore raw

Re: [PATCH v1 3/6] mfd: core: Propagate software node group to the sub devices

Quoted 34 lines · Andy Shevchenko
Quoted 17 lines · Lee Jones
Quoted 8 lines · Andy Shevchenko

From: Heikki Krogerus <[email protected]>

When ever device properties are supplied for a sub device, a software node (fwnode) is actually created and then associated with that device. By allowing the drivers to supply the complete software node group instead of just the properties in it, the drivers can take advantage of the other features the software nodes have on top of supplying the device properties.

Quoted 6 lines
---
 drivers/mfd/mfd-core.c   | 31 +++++++++++++++++++++++++++----
 include/linux/mfd/core.h |  3 +++
 2 files changed, 30 insertions(+), 4 deletions(-)
Quoted 3 lines

I'm not sure a change to the API is justified presently (same does go for 'properties' really, but as it was only a couple of lines, it didn't seem too intrusive).

This is better and comprehensive API, but I heard you.

Quoted 2 lines

My recommendation is to handle this in-house (i.e. locally in-driver) for now.

I think you understand that this is not gonna work (we need to attach fwnode to the child device before it's registration.

Quoted 3 lines

When (if) more users adopt the practice, then we should consider to draw down on line numbers and repetition and make it part of the API.

Quoted 6 lines

I briefly looked at the current state of affairs and found that properties are used only for MFD LPSS driver. Would the conversion of that driver to swnodes work for you?

Note, the long prospective is to get rid of platform_add_properties() API completely.

That's a shame. Do you plan on replacing it with something else?

MFD tends to only interact with the platform_device API, and even platform_device_add_properties() doesn't get involved in the nitty gritty of fwnodes. Instead it acts as a pass-through straight to device_add_properties() which is where the magic happens.

For this to be acceptable you would need to add support into platform_device i.e. platform_device_add_property_group() or some such. I really do not want the MFD API to have knowledge about regarding the particulars i.e. software node registration, secondary fwnodes and the like.

Signature · 4 lines
Lee Jones [李琼斯]
Senior Technical Lead - Developer Services
Linaro.org │ Open source software for Arm SoCs
Follow Linaro: Facebook | Twitter | Blog
#9 lore raw

PATCH 4/6 mfd: intel_quark_i2c_gpio: Convert to use software nodes

In reply to Andy Shevchenko

The driver can provide a software node group instead of passing legacy platform data. This will allow to drop the legacy platform data structures along with unifying a child device driver to use same interface for all property providers, i.e. Device Tree, ACPI, and board files.

---
 drivers/mfd/intel_quark_i2c_gpio.c | 63 ++++++++++++++----------------
 1 file changed, 30 insertions(+), 33 deletions(-)
drivers/mfd/intel_quark_i2c_gpio.c +30 -33
index 06b538dd124c..fb8c0b042ecc 100644
@@ -16,8 +16,8 @@
 #include <linux/clkdev.h>
 #include <linux/clk-provider.h>
 #include <linux/dmi.h>
-#include <linux/platform_data/gpio-dwapb.h>
 #include <linux/platform_data/i2c-designware.h>
+#include <linux/property.h>
 
 /* PCI BAR for register base address */
 #define MFD_I2C_BAR		0
@@ -27,15 +27,6 @@
 #define MFD_ACPI_MATCH_GPIO	0ULL
 #define MFD_ACPI_MATCH_I2C	1ULL
 
-/* The base GPIO number under GPIOLIB framework */
-#define INTEL_QUARK_MFD_GPIO_BASE	8
-
-/* The default number of South-Cluster GPIO on Quark. */
-#define INTEL_QUARK_MFD_NGPIO		8
-
-/* The DesignWare GPIO ports on Quark. */
-#define INTEL_QUARK_GPIO_NPORTS	1
-
 #define INTEL_QUARK_IORES_MEM	0
 #define INTEL_QUARK_IORES_IRQ	1
 
@@ -89,17 +80,44 @@ static struct resource intel_quark_gpio_res[] = {
 	[INTEL_QUARK_IORES_MEM] = {
 		.flags = IORESOURCE_MEM,
 	},
+	[INTEL_QUARK_IORES_IRQ] = {
+		.flags = IORESOURCE_IRQ,
+	},
 };
 
 static struct mfd_cell_acpi_match intel_quark_acpi_match_gpio = {
 	.adr = MFD_ACPI_MATCH_GPIO,
 };
 
+static const struct software_node intel_quark_gpio_controller_node = {
+	.name = "intel-quark-gpio-controller",
+};
+
+static const struct property_entry intel_quark_gpio_portA_properties[] = {
+	PROPERTY_ENTRY_U32("reg", 0),
+	PROPERTY_ENTRY_U32("snps,nr-gpios", 8),
+	PROPERTY_ENTRY_U32("snps,gpio-base", 8),
+	{ }
+};
+
+static const struct software_node intel_quark_gpio_portA_node = {
+	.name = "portA",
+	.parent = &intel_quark_gpio_controller_node,
+	.properties = intel_quark_gpio_portA_properties,
+};
+
+static const struct software_node *intel_quark_gpio_node_group[] = {
+	&intel_quark_gpio_controller_node,
+	&intel_quark_gpio_portA_node,
+	NULL
+};
+
 static struct mfd_cell intel_quark_mfd_cells[] = {
 	{
 		.id = MFD_GPIO_BAR,
 		.name = "gpio-dwapb",
 		.acpi_match = &intel_quark_acpi_match_gpio,
+		.node_group = intel_quark_gpio_node_group,
 		.num_resources = ARRAY_SIZE(intel_quark_gpio_res),
 		.resources = intel_quark_gpio_res,
 		.ignore_resource_conflicts = true,
@@ -189,36 +207,15 @@ static int intel_quark_i2c_setup(struct pci_dev *pdev, struct mfd_cell *cell)
 
 static int intel_quark_gpio_setup(struct pci_dev *pdev, struct mfd_cell *cell)
 {
-	struct dwapb_platform_data *pdata;
 	struct resource *res = (struct resource *)cell->resources;
-	struct device *dev = &pdev->dev;
 
 	res[INTEL_QUARK_IORES_MEM].start =
 		pci_resource_start(pdev, MFD_GPIO_BAR);
 	res[INTEL_QUARK_IORES_MEM].end =
 		pci_resource_end(pdev, MFD_GPIO_BAR);
 
-	pdata = devm_kzalloc(dev, sizeof(*pdata), GFP_KERNEL);
-	if (!pdata)
-		return -ENOMEM;
-
-	/* For intel quark x1000, it has only one port: portA */
-	pdata->nports = INTEL_QUARK_GPIO_NPORTS;
-	pdata->properties = devm_kcalloc(dev, pdata->nports,
-					 sizeof(*pdata->properties),
-					 GFP_KERNEL);
-	if (!pdata->properties)
-		return -ENOMEM;
-
-	/* Set the properties for portA */
-	pdata->properties->fwnode	= NULL;
-	pdata->properties->idx		= 0;
-	pdata->properties->ngpio	= INTEL_QUARK_MFD_NGPIO;
-	pdata->properties->gpio_base	= INTEL_QUARK_MFD_GPIO_BASE;
-	pdata->properties->irq[0]	= pdev->irq;
-
-	cell->platform_data = pdata;
-	cell->pdata_size = sizeof(*pdata);
+	res[INTEL_QUARK_IORES_IRQ].start = pdev->irq;
+	res[INTEL_QUARK_IORES_IRQ].end = pdev->irq;
 
 	return 0;
 }
Signature · 1 line
2.27.0.rc2
#10 lore raw

Re: [PATCH v1 4/6] mfd: intel_quark_i2c_gpio: Convert to use software nodes

Hi Andy,

drivers/mfd/intel_quark_i2c_gpio.c line 31

Quoted 2 lines · Andy Shevchenko
drivers/mfd/intel_quark_i2c_gpio.c +0 -2
-/* The base GPIO number under GPIOLIB framework */
-#define INTEL_QUARK_MFD_GPIO_BASE      8

OK I see this was around before, sigh.
So it's not your fault. It was introduced in commit 60ae5b9f5cdd8 which I was not involved in reviewing, for the record I would have said "no".

It is exploiting commit 3d2613c4289ff where I allowed pdata to set the base so it is anyway my fault for not noticing :(

But me complaining about this doesn't make things better.

Can we simply DELETE this assignment and just set base to -1 in a separate patch before this patch and see what happens? It's really unsafe to hardcode base like this.

+ PROPERTY_ENTRY_U32("snps,nr-gpios", 8),

This is however fine in principle but just use the existing generic property "ngpios" and save this custom property.

Yours,
Linus Walleij

#11 lore raw

Re: [PATCH v1 4/6] mfd: intel_quark_i2c_gpio: Convert to use software nodes

Quoted 15 lines · Linus Walleij

drivers/mfd/intel_quark_i2c_gpio.c line 31

Quoted 2 lines · Andy Shevchenko
drivers/mfd/intel_quark_i2c_gpio.c +0 -2
-/* The base GPIO number under GPIOLIB framework */
-#define INTEL_QUARK_MFD_GPIO_BASE      8

OK I see this was around before, sigh.
So it's not your fault. It was introduced in commit 60ae5b9f5cdd8 which I was not involved in reviewing, for the record I would have said "no".

It is exploiting commit 3d2613c4289ff where I allowed pdata to set the base so it is anyway my fault for not noticing :(

Quoted 5 lines

But me complaining about this doesn't make things better.

Can we simply DELETE this assignment and just set base to -1 in a separate patch before this patch and see what happens? It's really unsafe to hardcode base like this.

I don't see easy way to do so, because as I explained in previous mail it will affect relation (from firmware) to the fact that the numbering of this controller is static for that platform.

Quoted 4 lines

+ PROPERTY_ENTRY_U32("snps,nr-gpios", 8),

This is however fine in principle but just use the existing generic property "ngpios" and save this custom property.

This is established (not by ACPI!) property. Completely not my fault :-)

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#12 lore raw

PATCH 5/6 gpio: dwapb: Get rid of legacy platform data

In reply to Andy Shevchenko

Platform data is a legacy interface to supply device properties to the driver. In this case we don't have anymore in-kernel users for it. Just remove it for good.

---
 drivers/gpio/gpio-dwapb.c                | 27 +++++++++++++++---------
 include/linux/platform_data/gpio-dwapb.h | 22 -------------------
 2 files changed, 17 insertions(+), 32 deletions(-)
 delete mode 100644 include/linux/platform_data/gpio-dwapb.h
drivers/gpio/gpio-dwapb.c +17 -10
index e3d7589434eb..de01b8943bd2 100644
@@ -23,7 +23,6 @@
 #include <linux/property.h>
 #include <linux/reset.h>
 #include <linux/spinlock.h>
-#include <linux/platform_data/gpio-dwapb.h>
 #include <linux/slab.h>
 
 #include "gpiolib.h"
@@ -66,6 +65,19 @@
 
 #define DWAPB_NR_CLOCKS		2
 
+struct dwapb_port_property {
+	struct fwnode_handle *fwnode;
+	unsigned int idx;
+	unsigned int ngpio;
+	unsigned int gpio_base;
+	int irq[32];
+};
+
+struct dwapb_platform_data {
+	struct dwapb_port_property *properties;
+	unsigned int nports;
+};
+
 struct dwapb_gpio;
 
 #ifdef CONFIG_PM_SLEEP
@@ -633,17 +645,12 @@ static int dwapb_gpio_probe(struct platform_device *pdev)
 	unsigned int i;
 	struct dwapb_gpio *gpio;
 	int err;
+	struct dwapb_platform_data *pdata;
 	struct device *dev = &pdev->dev;
-	struct dwapb_platform_data *pdata = dev_get_platdata(dev);
-
-	if (!pdata) {
-		pdata = dwapb_gpio_get_pdata(dev);
-		if (IS_ERR(pdata))
-			return PTR_ERR(pdata);
-	}
 
-	if (!pdata->nports)
-		return -ENODEV;
+	pdata = dwapb_gpio_get_pdata(dev);
+	if (IS_ERR(pdata))
+		return PTR_ERR(pdata);
 
 	gpio = devm_kzalloc(&pdev->dev, sizeof(*gpio), GFP_KERNEL);
 	if (!gpio)
include/linux/platform_data/gpio-dwapb.h deleted +0 -22
deleted file mode 100644
index a63010b6e898..000000000000
@@ -1,22 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0-only */
-/*
- * Copyright(c) 2014 Intel Corporation.
- */
-
-#ifndef GPIO_DW_APB_H
-#define GPIO_DW_APB_H
-
-struct dwapb_port_property {
-	struct fwnode_handle *fwnode;
-	unsigned int	idx;
-	unsigned int	ngpio;
-	unsigned int	gpio_base;
-	int		irq[32];
-};
-
-struct dwapb_platform_data {
-	struct dwapb_port_property *properties;
-	unsigned int nports;
-};
-
-#endif
Signature · 1 line
2.27.0.rc2
#13 lore raw

Re: [PATCH v1 5/6] gpio: dwapb: Get rid of legacy platform data

Quoted 5 lines · Andy Shevchenko

Platform data is a legacy interface to supply device properties to the driver. In this case we don't have anymore in-kernel users for it. Just remove it for good.

I really like where this series takes us so I hope we can still apply all of this in some form!

Yours,
Linus Walleij

#14 lore raw

Re: [PATCH v1 5/6] gpio: dwapb: Get rid of legacy platform data

Quoted 11 lines · Linus Walleij
Quoted 5 lines · Andy Shevchenko

Platform data is a legacy interface to supply device properties to the driver. In this case we don't have anymore in-kernel users for it. Just remove it for good.

I really like where this series takes us so I hope we can still apply all of this in some form!

Thanks! I hope you see the benefit despite some (GPIO base) limitations.

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#15 lore raw

PATCH 6/6 gpio: dwapb: Define magic number for IRQ and GPIO lines

In reply to Andy Shevchenko

Define maximum number of IRQ and GPIO lines per port and replace magic number by it.

---
 drivers/gpio/gpio-dwapb.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)
drivers/gpio/gpio-dwapb.c +6 -5
index de01b8943bd2..77a86a8eaee0 100644
@@ -63,6 +63,7 @@
 #define GPIO_INTSTATUS_V2	0x3c
 #define GPIO_PORTA_EOI_V2	0x40
 
+#define DWAPB_NR_GPIOS		32
 #define DWAPB_NR_CLOCKS		2
 
 struct dwapb_port_property {
@@ -70,7 +71,7 @@ struct dwapb_port_property {
 	unsigned int idx;
 	unsigned int ngpio;
 	unsigned int gpio_base;
-	int irq[32];
+	int irq[DWAPB_NR_GPIOS];
 };
 
 struct dwapb_platform_data {
@@ -174,7 +175,7 @@ static struct dwapb_gpio_port *dwapb_offs_to_port(struct dwapb_gpio *gpio, unsig
 
 	for (i = 0; i < gpio->nr_ports; i++) {
 		port = &gpio->ports[i];
-		if (port->idx == offs / 32)
+		if (port->idx == offs / DWAPB_NR_GPIOS)
 			return port;
 	}
 
@@ -194,7 +195,7 @@ static void dwapb_toggle_trigger(struct dwapb_gpio *gpio, unsigned int offs)
 
 	pol = dwapb_read(gpio, GPIO_INT_POLARITY);
 	/* Just read the current value right out of the data register */
-	val = gc->get(gc, offs % 32);
+	val = gc->get(gc, offs % DWAPB_NR_GPIOS);
 	if (val)
 		pol &= ~BIT(offs);
 	else
@@ -209,7 +210,7 @@ static u32 dwapb_do_irq(struct dwapb_gpio *gpio)
 	irq_hw_number_t hwirq;
 
 	irq_status = dwapb_read(gpio, GPIO_INTSTATUS);
-	for_each_set_bit(hwirq, &irq_status, 32) {
+	for_each_set_bit(hwirq, &irq_status, DWAPB_NR_GPIOS) {
 		int gpio_irq = irq_find_mapping(gpio->domain, hwirq);
 		u32 irq_type = irq_get_trigger_type(gpio_irq);
 
@@ -608,7 +609,7 @@ static struct dwapb_platform_data *dwapb_gpio_get_pdata(struct device *dev)
 			dev_info(dev,
 				 "failed to get number of gpios for port%d\n",
 				 i);
-			pp->ngpio = 32;
+			pp->ngpio = DWAPB_NR_GPIOS;
 		}
 
 		if (fwnode_property_read_u32(fwnode, "snps,gpio-base", &pp->gpio_base))
Signature · 1 line
2.27.0.rc2
#16 lore raw
In reply to Andy Shevchenko
Quoted 9 lines · Andy Shevchenko

Some devices would need to have a hierarchy of properties and child nodes passed to the child or children of MFD. For such case we may utilize software nodes, which is superior on device properties.

Add support of software nodes to MFD core and convert one driver to show how it looks like. This allows to get rid of legacy platform data.

The change has been tested on Intel Galileo Gen 2.

I am wondering whether we could move the {gpio_base, ngpio, irq_shared} part into the gpio-dwapb.c driver and use either the ACPI-based or platform_device_id-based matching to get the device-specific resources info through the driver_data field. By doing so you wouldn't need to introduce a new "snps,gpio-base"-like property and propagate software_node-based properties, but still you could get rid of the dwapb_platform_data structure since all the info would be locally available.

If ACPI-based matching doesn't uniquely address the Quark GPIO node, then you could just replace the intel_quark_mfd_cells[0].name with something like "gpio-dwapb-quark", which then by the MFD core will be copied to the corresponding platform_device->name due to calling platform_device_alloc() with cell-name passed. That name will be used to match a platform_driver with id_table having that new name added.

-Sergey

Quoted 22 lines

Andy Shevchenko (5):
gpio: dwapb: Replace irq_shared flag with fwnode type check
gpio: dwapb: Read GPIO base from snps,gpio-base property
mfd: intel_quark_i2c_gpio: Convert to use software nodes
gpio: dwapb: Get rid of legacy platform data
gpio: dwapb: Define magic number for IRQ and GPIO lines

Heikki Krogerus (1):
mfd: core: Propagate software node group to the sub devices

 drivers/gpio/gpio-dwapb.c                | 44 +++++++++-------
 drivers/mfd/intel_quark_i2c_gpio.c       | 64 +++++++++++-------------
 drivers/mfd/mfd-core.c                   | 31 ++++++++++--
 include/linux/mfd/core.h                 |  3 ++
 include/linux/platform_data/gpio-dwapb.h | 23 ---------
 5 files changed, 85 insertions(+), 80 deletions(-)
 delete mode 100644 include/linux/platform_data/gpio-dwapb.h
Signature · 1 line
2.27.0.rc2
#17 lore raw
Quoted 11 lines · Serge Semin
Quoted 9 lines · Andy Shevchenko

Some devices would need to have a hierarchy of properties and child nodes passed to the child or children of MFD. For such case we may utilize software nodes, which is superior on device properties.

Add support of software nodes to MFD core and convert one driver to show how it looks like. This allows to get rid of legacy platform data.

The change has been tested on Intel Galileo Gen 2.

Quoted 8 lines

I am wondering whether we could move the {gpio_base, ngpio, irq_shared} part into the gpio-dwapb.c driver and use either the ACPI-based or platform_device_id-based matching to get the device-specific resources info through the driver_data field. By doing so you wouldn't need to introduce a new "snps,gpio-base"-like property and propagate software_node-based properties, but still you could get rid of the dwapb_platform_data structure since all the info would be locally available.

The idea is to get rid of the driver being dependent on some quirks when we may do it clearly and nicely.
We, by applying this series, make (keep) layers independent: board code vs. driver code. Mixing them more is the opposite to what I propose.

WRT property.
snps,gpio-base can be easily changed to *already in use* gpio-base or being both converted to linux,gpio-base to explicitly show people that this is internal stuff that must not be present in firmware nodes.

Quoted 6 lines

If ACPI-based matching doesn't uniquely address the Quark GPIO node, then you could just replace the intel_quark_mfd_cells[0].name with something like "gpio-dwapb-quark", which then by the MFD core will be copied to the corresponding platform_device->name due to calling platform_device_alloc() with cell-name passed. That name will be used to match a platform_driver with id_table having that new name added.

Oh, that doesn't sound right. It makes things ugly.

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#18 lore raw
Quoted 21 lines · Andy Shevchenko
Quoted 11 lines · Serge Semin
Quoted 9 lines · Andy Shevchenko

Some devices would need to have a hierarchy of properties and child nodes passed to the child or children of MFD. For such case we may utilize software nodes, which is superior on device properties.

Add support of software nodes to MFD core and convert one driver to show how it looks like. This allows to get rid of legacy platform data.

The change has been tested on Intel Galileo Gen 2.

Quoted 8 lines

I am wondering whether we could move the {gpio_base, ngpio, irq_shared} part into the gpio-dwapb.c driver and use either the ACPI-based or platform_device_id-based matching to get the device-specific resources info through the driver_data field. By doing so you wouldn't need to introduce a new "snps,gpio-base"-like property and propagate software_node-based properties, but still you could get rid of the dwapb_platform_data structure since all the info would be locally available.

Quoted 2 lines

The idea is to get rid of the driver being dependent on some quirks when we may do it clearly and nicely.

Yes, I've got that and in most of the aspects I like what you suggested in this parchset. But it seems to me that the maintainers are mostly prone to having some of the platform-specifics being locally (in-driver) defined. So I proposed an alternative solution, which might do to satisfy their requirement. Note saying that you want to get rid of the quirks and introducing something like "gpio-base" firmware property seems contradicting a bit.

Quoted 9 lines

We, by applying this series, make (keep) layers independent: board code vs. driver code. Mixing them more is the opposite to what I propose.

WRT property.
snps,gpio-base can be easily changed to *already in use* gpio-base or being both converted to linux,gpio-base to explicitly show people that this is internal stuff that must not be present in firmware nodes.

As I see it the part with "gpio-base" and the irq_shared can be moved to the gpio-dwapb.c driver to be defined as the quark-specific quirks. Adding a property like "gpio-base" seems like a quirk anyway, so I'd leave it defined in the driver.

* Note I don't really like replacing the irq_shared flag with to_of_node() because in general to_of_node() doesn't mean the IRQ line is shared, so selecting the shared and non-shared interrupt request paths based on that macro seems hackish.

Quoted 8 lines
Quoted 6 lines

If ACPI-based matching doesn't uniquely address the Quark GPIO node, then you could just replace the intel_quark_mfd_cells[0].name with something like "gpio-dwapb-quark", which then by the MFD core will be copied to the corresponding platform_device->name due to calling platform_device_alloc() with cell-name passed. That name will be used to match a platform_driver with id_table having that new name added.

Oh, that doesn't sound right. It makes things ugly.

I may have said that a bit unclearly. The only thing you'd need to do is to add an unique name to the Quark GPIO cell, like:
drivers/mfd/intel_quark_i2c_gpio.c:

static struct mfd_cell intel_quark_mfd_cells[] = {
        {
                .name = "gpio-dwapb-quark",
        }

Then make the gpio-dwapb.c driver being compatible with that device by declaring the id_table with that device name and passing the table to the DW APB GPIO "struct platform_driver" descriptor. The MFD/platform cores already provide the functionality of matching those two device and driver. If ACPI node uniquely defines the Quark GPIO with all that quirks applicable then you wouldn't even need to do the platform_device-driver-based matching. Just use the acpi_device_id to get the quirks flags/descriptors.

The rest of the patchset concerning passing software_node's down to the individual MFD sub-device and declaring "reg" and "snps,nr-gpios" in there seems pretty much acceptable to me.

* Though indeed it would be better to mark the "snps,nr-gpios" as deprecated in the DT schema and have the "ngpios" supported as well.

-Sergey

Quoted 4 lines
Signature · 3 lines
-- 
With Best Regards,
Andy Shevchenko
#19 lore raw
Quoted 22 lines · Serge Semin
Quoted 21 lines · Andy Shevchenko
Quoted 19 lines · Serge Semin
Quoted 9 lines · Andy Shevchenko

Some devices would need to have a hierarchy of properties and child nodes passed to the child or children of MFD. For such case we may utilize software nodes, which is superior on device properties.

Add support of software nodes to MFD core and convert one driver to show how it looks like. This allows to get rid of legacy platform data.

The change has been tested on Intel Galileo Gen 2.

I am wondering whether we could move the {gpio_base, ngpio, irq_shared} part into the gpio-dwapb.c driver and use either the ACPI-based or platform_device_id-based matching to get the device-specific resources info through the driver_data field. By doing so you wouldn't need to introduce a new "snps,gpio-base"-like property and propagate software_node-based properties, but still you could get rid of the dwapb_platform_data structure since all the info would be locally available.

Quoted 6 lines
Quoted 2 lines

The idea is to get rid of the driver being dependent on some quirks when we may do it clearly and nicely.

Yes, I've got that and in most of the aspects I like what you suggested in this parchset. But it seems to me that the maintainers are mostly prone to having some of the platform-specifics being locally (in-driver) defined.

You are a maintainer of the dwapb-gpio. Is it what you insist?

Quoted 2 lines

So I proposed an alternative solution, which might do to satisfy their requirement.

I'm puzzled whom about you are talking.

Quoted 3 lines

Note saying that you want to get rid of the quirks and introducing something like "gpio-base" firmware property seems contradicting a bit.

Maybe I need to elaborate that under quirks I meant quirk-clean GPIO driver, so, it wouldn't care about what platform(s) require base and what do not.

Quoted 9 lines
Quoted 8 lines

We, by applying this series, make (keep) layers independent: board code vs. driver code. Mixing them more is the opposite to what I propose.

WRT property.
snps,gpio-base can be easily changed to *already in use* gpio-base or being both converted to linux,gpio-base to explicitly show people that this is internal stuff that must not be present in firmware nodes.

Quoted 4 lines

As I see it the part with "gpio-base" and the irq_shared can be moved to the gpio-dwapb.c driver to be defined as the quark-specific quirks. Adding a property like "gpio-base" seems like a quirk anyway, so I'd leave it defined in the driver.

Huh?! The whole idea is make GPIO driver agnostic from platforms and their quirks.

Quoted 4 lines

* Note I don't really like replacing the irq_shared flag with to_of_node() because in general to_of_node() doesn't mean the IRQ line is shared, so selecting the shared and non-shared interrupt request paths based on that macro seems hackish.

This I can understand, but can you propose better alternative?

Quoted 17 lines
Quoted 8 lines
Quoted 6 lines

If ACPI-based matching doesn't uniquely address the Quark GPIO node, then you could just replace the intel_quark_mfd_cells[0].name with something like "gpio-dwapb-quark", which then by the MFD core will be copied to the corresponding platform_device->name due to calling platform_device_alloc() with cell-name passed. That name will be used to match a platform_driver with id_table having that new name added.

Oh, that doesn't sound right. It makes things ugly.

I may have said that a bit unclearly. The only thing you'd need to do is to add an unique name to the Quark GPIO cell, like:
drivers/mfd/intel_quark_i2c_gpio.c:

static struct mfd_cell intel_quark_mfd_cells[] = {
        {
                .name = "gpio-dwapb-quark",
        }
Quoted 7 lines

Then make the gpio-dwapb.c driver being compatible with that device by declaring the id_table with that device name and passing the table to the DW APB GPIO "struct platform_driver" descriptor. The MFD/platform cores already provide the functionality of matching those two device and driver. If ACPI node uniquely defines the Quark GPIO with all that quirks applicable then you wouldn't even need to do the platform_device-driver-based matching. Just use the acpi_device_id to get the quirks flags/descriptors.

What you are talking about? Can you provide a code we can discuss?

Quoted 2 lines

* Though indeed it would be better to mark the "snps,nr-gpios" as deprecated in the DT schema and have the "ngpios" supported as well.

Actually I used to have that piece in my first patches, but decided to remove due to this property being (semi-)internal. I would like to hear GPIO maintainers about this.

Bart, Linus?

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#20 lore raw
Quoted 27 lines · Andy Shevchenko
Quoted 25 lines · Serge Semin
Quoted 23 lines · Andy Shevchenko
Quoted 11 lines · Serge Semin
Quoted 9 lines · Andy Shevchenko

Some devices would need to have a hierarchy of properties and child nodes passed to the child or children of MFD. For such case we may utilize software nodes, which is superior on device properties.

Add support of software nodes to MFD core and convert one driver to show how it looks like. This allows to get rid of legacy platform data.

The change has been tested on Intel Galileo Gen 2.

Quoted 8 lines

I am wondering whether we could move the {gpio_base, ngpio, irq_shared} part into the gpio-dwapb.c driver and use either the ACPI-based or platform_device_id-based matching to get the device-specific resources info through the driver_data field. By doing so you wouldn't need to introduce a new "snps,gpio-base"-like property and propagate software_node-based properties, but still you could get rid of the dwapb_platform_data structure since all the info would be locally available.

The idea is to get rid of the driver being dependent on some quirks when we may do it clearly and nicely.

Quoted 5 lines
Quoted 3 lines

Yes, I've got that and in most of the aspects I like what you suggested in this parchset. But it seems to me that the maintainers are mostly prone to having some of the platform-specifics being locally (in-driver) defined.

You are a maintainer of the dwapb-gpio. Is it what you insist?

That was a discussable suggestion of how the patches could be improved to still have the software-node functionality and to get rid of the legacy platform data passing around.

Quoted 5 lines
Quoted 3 lines

So I proposed an alternative solution, which might do to satisfy their requirement.

I'm puzzled whom about you are talking.

I don't have a habit to say of myself like that) So obviously I was talking about Linus and Lee in the previous sentence mostly referring to the Linus' words regarding setting the GPIO base and having the "gpio-base" property at all.

Quoted 7 lines
Quoted 4 lines

Note saying that you want to get rid of the quirks and introducing something like "gpio-base" firmware property seems contradicting a bit.

Maybe I need to elaborate that under quirks I meant quirk-clean GPIO driver, so, it wouldn't care about what platform(s) require base and what do not.

My point is that no mater how you pass the gpio-base value it will always be a quirk. The same thing is with the irq_shared flag. Both of them are specific to the Quark-platform. They are used to tune the DW APB GPIO driver up so one would create a GPIO device working well with the Quark-specific DW APB GPIO block.

Quoted 10 lines
Quoted 10 lines
Quoted 9 lines

We, by applying this series, make (keep) layers independent: board code vs. driver code. Mixing them more is the opposite to what I propose.

WRT property.
snps,gpio-base can be easily changed to *already in use* gpio-base or being both converted to linux,gpio-base to explicitly show people that this is internal stuff that must not be present in firmware nodes.

Quoted 7 lines
Quoted 4 lines

As I see it the part with "gpio-base" and the irq_shared can be moved to the gpio-dwapb.c driver to be defined as the quark-specific quirks. Adding a property like "gpio-base" seems like a quirk anyway, so I'd leave it defined in the driver.

Huh?! The whole idea is make GPIO driver agnostic from platforms and their quirks.

As I said above having the gpio-base value set is the platform-specific thing in any case. So no mater whether you pass it via the software_node/properties or the private platform-data structure, the DW APB GPIO driver will have to handle those parameters in some special way, which is quirk-prone since normal platforms don't have such peculiar requirements.

Quoted 7 lines
Quoted 5 lines

* Note I don't really like replacing the irq_shared flag with to_of_node() because in general to_of_node() doesn't mean the IRQ line is shared, so selecting the shared and non-shared interrupt request paths based on that macro seems hackish.

This I can understand, but can you propose better alternative?

Alas I can't at the moment. But see my comment below.

Quoted 26 lines
Quoted 18 lines
Quoted 9 lines
Quoted 7 lines

If ACPI-based matching doesn't uniquely address the Quark GPIO node, then you could just replace the intel_quark_mfd_cells[0].name with something like "gpio-dwapb-quark", which then by the MFD core will be copied to the corresponding platform_device->name due to calling platform_device_alloc() with cell-name passed. That name will be used to match a platform_driver with id_table having that new name added.

Oh, that doesn't sound right. It makes things ugly.

I may have said that a bit unclearly. The only thing you'd need to do is to add an unique name to the Quark GPIO cell, like:
drivers/mfd/intel_quark_i2c_gpio.c:

static struct mfd_cell intel_quark_mfd_cells[] = {
        {
                .name = "gpio-dwapb-quark",
        }
Quoted 7 lines

Then make the gpio-dwapb.c driver being compatible with that device by declaring the id_table with that device name and passing the table to the DW APB GPIO "struct platform_driver" descriptor. The MFD/platform cores already provide the functionality of matching those two device and driver. If ACPI node uniquely defines the Quark GPIO with all that quirks applicable then you wouldn't even need to do the platform_device-driver-based matching. Just use the acpi_device_id to get the quirks flags/descriptors.

Quoted 1 line

What you are talking about? Can you provide a code we can discuss?

Here is what I suggest in step-by-step:
Patch 1) Move the Quark-specific GPIO-base and irq_shared parameters definition
to the gpio-dwapb.c driver.
Patch 2) Retrieve the GPIO-base and IRQ-shared parameters only for the DW APB
GPIO block living in the Quark platform (by means of either ACPI or the platform-device ID data).
Patch 3) Introduce the software_node propagation functionality (AFAIU Lee
wants it to be a part of the platform-device core instead of being created on the MFD level).
Patch 4) Convert the intel_quark_i2c_gpio driver to using the software nodes
instead of creating the dwapb_platform_data instance (nearly matches your patch 4 except it won't introduce any "gpio-base"-like property).
Patch 5) Get rid of the global dwapb_platform_data utilization (matches your

         patch 5).

Patch 6) Introduce the IRQ and GPIO lines macro (matches your patch 6).

By doing as I suggested you don't have to pass around the GPIO-base value and the IRQ-shared flag. They will be initialized and utilized locally in the DW APB GPIO driver if Quark-specific DW APB GPIO block is detected (patches 1 and 2). In that case a software-node created in Patch 3 and 4 will be a normal firmware node described by the DW bindings. You'll still be able to get rid of the legacy global platform-data in the Patch 5.

The spirit of your original patchset will still be preserved: introduce the software-nodes propagation interface and use it to pass the generic DW APB GPIO parameters. The only difference is that we'd move the GPIO-base and IRQ-shared functionality fully into the DW APB GPIO driver. By doing so we'd have:
1) DW APB GPIO Quark-specifics localized in a single gpio-dwapb driver. In the current implementation we have them distributed between two drivers:
intel_quark_i2c_gpio initializes the GPIO-base and IRQ-shared parameters, gpio-dwapb uses them for setup. Note there is no any other platform with DW APB GPIO which needs the GPIO-base and IRQ-shared flag being setup. Why do we need to have such complicated interface then if we can identify whether the particular DW APB GPIO is a Quark-specific block or not? Moreover having those specifics in both drivers in fact means having quirks in both of them.
2) Quirk-clean intel_quark_i2c_gpio driver which creates a normal DW APB GPIO device with software-nodes fully compatible with DT binding. DW APB GPIO driver will detect whether the block is Quark-specific and will activate the corresponding quirks if it is.
3) Still legacy global platform-data removed.

-Sergey

Quoted 15 lines
Quoted 3 lines

* Though indeed it would be better to mark the "snps,nr-gpios" as deprecated in the DT schema and have the "ngpios" supported as well.

Actually I used to have that piece in my first patches, but decided to remove due to this property being (semi-)internal. I would like to hear GPIO maintainers about this.

Bart, Linus?

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#21 lore raw
Quoted 5 lines · Serge Semin
Quoted 4 lines · Andy Shevchenko
Quoted 3 lines · Serge Semin
Quoted 2 lines · Andy Shevchenko
Quoted 1 line · Serge Semin

On Mon, Jun 08, 2020 at 04:42:54PM +0300, Andy Shevchenko wrote:

...

Quoted 26 lines
Quoted 13 lines
Quoted 12 lines
Quoted 11 lines
Quoted 8 lines

I am wondering whether we could move the {gpio_base, ngpio, irq_shared} part into the gpio-dwapb.c driver and use either the ACPI-based or platform_device_id-based matching to get the device-specific resources info through the driver_data field. By doing so you wouldn't need to introduce a new "snps,gpio-base"-like property and propagate software_node-based properties, but still you could get rid of the dwapb_platform_data structure since all the info would be locally available.

The idea is to get rid of the driver being dependent on some quirks when we may do it clearly and nicely.

Quoted 5 lines
Quoted 3 lines

Yes, I've got that and in most of the aspects I like what you suggested in this parchset. But it seems to me that the maintainers are mostly prone to having some of the platform-specifics being locally (in-driver) defined.

You are a maintainer of the dwapb-gpio. Is it what you insist?

That was a discussable suggestion of how the patches could be improved to still have the software-node functionality and to get rid of the legacy platform data passing around.

Quoted 3 lines
Quoted 2 lines

So I proposed an alternative solution, which might do to satisfy their requirement.

Quoted 6 lines
Quoted 1 line

I'm puzzled whom about you are talking.

I don't have a habit to say of myself like that) So obviously I was talking about Linus and Lee in the previous sentence mostly referring to the Linus' words regarding setting the GPIO base and having the "gpio-base" property at all.

Linus replied to three messages out of 7, seems you read only the first one. That's why I have been puzzled.

Quoted 11 lines
Quoted 6 lines
Quoted 3 lines

Note saying that you want to get rid of the quirks and introducing something like "gpio-base" firmware property seems contradicting a bit.

Maybe I need to elaborate that under quirks I meant quirk-clean GPIO driver, so, it wouldn't care about what platform(s) require base and what do not.

My point is that no mater how you pass the gpio-base value it will always be a quirk. The same thing is with the irq_shared flag. Both of them are specific to the Quark-platform. They are used to tune the DW APB GPIO driver up so one would create a GPIO device working well with the Quark-specific DW APB GPIO block.

Precisely!

Quoted 15 lines
Quoted 14 lines
Quoted 13 lines
Quoted 8 lines

We, by applying this series, make (keep) layers independent: board code vs. driver code. Mixing them more is the opposite to what I propose.

WRT property.
snps,gpio-base can be easily changed to *already in use* gpio-base or being both converted to linux,gpio-base to explicitly show people that this is internal stuff that must not be present in firmware nodes.

As I see it the part with "gpio-base" and the irq_shared can be moved to the gpio-dwapb.c driver to be defined as the quark-specific quirks. Adding a property like "gpio-base" seems like a quirk anyway, so I'd leave it defined in the driver.

Quoted 7 lines
Quoted 1 line

Huh?! The whole idea is make GPIO driver agnostic from platforms and their quirks.

As I said above having the gpio-base value set is the platform-specific thing in any case. So no mater whether you pass it via the software_node/properties or the private platform-data structure, the DW APB GPIO driver will have to handle those parameters in some special way, which is quirk-prone since normal platforms don't have such peculiar requirements.

Seems you are proposing layering violation without seeing it. Let me explain the design of the drivers in Linux kernel.

There are basically few possible concepts how drivers are handling quirks (because we know that most of the hardware support is a set of quirks here and there):
1) platform provides a board files and via platform data structures supplies quirks to the certain driver
2) platform provides a firmware node (ACPI, DT, ...) and unified driver handles it thru the fwnode interface
3) driver is split to be a library (or core part) + glue drivers full of quirks
4) driver has embedded quirks and supplies them based on IDs (PCI ID, ACPI ID, compatible string, ID table, etc)
5) ...missed something? ...

What I'm proposing is turn 1 to 2 for Quark case, what you are proposing is absent on the picture, i.e.:
x) platform provides a board file (intel_quark_i2c_gpio.c) and the driver has embedded quirks based on ID.

This is not what we want, definitely.

...

Quoted 5 lines
Quoted 1 line

What you are talking about? Can you provide a code we can discuss?

Here is what I suggest in step-by-step:
Patch 1) Move the Quark-specific GPIO-base and irq_shared parameters definition
to the gpio-dwapb.c driver.

Layering violation: spreading board code over the kernel.

Quoted 3 lines

Patch 2) Retrieve the GPIO-base and IRQ-shared parameters only for the DW APB
GPIO block living in the Quark platform (by means of either ACPI or the platform-device ID data).

Same. (The rest doesn't matter. I dropped it because it's basically the same what I proposed)

Quoted 16 lines

By doing as I suggested you don't have to pass around the GPIO-base value and the IRQ-shared flag. They will be initialized and utilized locally in the DW APB GPIO driver if Quark-specific DW APB GPIO block is detected (patches 1 and 2). In that case a software-node created in Patch 3 and 4 will be a normal firmware node described by the DW bindings. You'll still be able to get rid of the legacy global platform-data in the Patch 5.

The spirit of your original patchset will still be preserved: introduce the software-nodes propagation interface and use it to pass the generic DW APB GPIO parameters. The only difference is that we'd move the GPIO-base and IRQ-shared functionality fully into the DW APB GPIO driver. By doing so we'd have:
1) DW APB GPIO Quark-specifics localized in a single gpio-dwapb driver. In the current implementation we have them distributed between two drivers:
intel_quark_i2c_gpio initializes the GPIO-base and IRQ-shared parameters, gpio-dwapb uses them for setup. Note there is no any other platform with DW APB GPIO which needs the GPIO-base and IRQ-shared flag being setup. Why do

Quoted 8 lines

we need to have such complicated interface then if we can identify whether the particular DW APB GPIO is a Quark-specific block or not? Moreover having those specifics in both drivers in fact means having quirks in both of them.
2) Quirk-clean intel_quark_i2c_gpio driver which creates a normal DW APB GPIO device with software-nodes fully compatible with DT binding. DW APB GPIO driver will detect whether the block is Quark-specific and will activate the corresponding quirks if it is.
3) Still legacy global platform-data removed.

I see. But you mistakenly made a conclusion 1). It will be spread to two drivers.
And in 2) you considered (wrongly) that MFD driver is not a board code.

Again MFD driver _is_ a board code or i.o.w. quirk driver for specific Quark (MFD) device.

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#22 lore raw
Quoted 37 lines · Andy Shevchenko
Quoted 5 lines · Serge Semin
Quoted 4 lines · Andy Shevchenko
Quoted 3 lines · Serge Semin
Quoted 2 lines · Andy Shevchenko
Quoted 1 line · Serge Semin

On Mon, Jun 08, 2020 at 04:42:54PM +0300, Andy Shevchenko wrote:

...

Quoted 28 lines
Quoted 13 lines
Quoted 12 lines
Quoted 11 lines
Quoted 8 lines

I am wondering whether we could move the {gpio_base, ngpio, irq_shared} part into the gpio-dwapb.c driver and use either the ACPI-based or platform_device_id-based matching to get the device-specific resources info through the driver_data field. By doing so you wouldn't need to introduce a new "snps,gpio-base"-like property and propagate software_node-based properties, but still you could get rid of the dwapb_platform_data structure since all the info would be locally available.

The idea is to get rid of the driver being dependent on some quirks when we may do it clearly and nicely.

Quoted 5 lines
Quoted 3 lines

Yes, I've got that and in most of the aspects I like what you suggested in this parchset. But it seems to me that the maintainers are mostly prone to having some of the platform-specifics being locally (in-driver) defined.

You are a maintainer of the dwapb-gpio. Is it what you insist?

That was a discussable suggestion of how the patches could be improved to still have the software-node functionality and to get rid of the legacy platform data passing around.

Quoted 4 lines
Quoted 2 lines

So I proposed an alternative solution, which might do to satisfy their requirement.

I'm puzzled whom about you are talking.

Quoted 7 lines
Quoted 4 lines

I don't have a habit to say of myself like that) So obviously I was talking about Linus and Lee in the previous sentence mostly referring to the Linus' words regarding setting the GPIO base and having the "gpio-base" property at all.

Linus replied to three messages out of 7, seems you read only the first one. That's why I have been puzzled.

Actually I've read all of them and from what I could see, he didn't like a code setting gpio-base regardless whether it has been created before your patchset or your explanation why it was necessary.

Quoted 54 lines
Quoted 5 lines
Quoted 5 lines
Quoted 4 lines

Note saying that you want to get rid of the quirks and introducing something like "gpio-base" firmware property seems contradicting a bit.

Quoted 7 lines
Quoted 2 lines

Maybe I need to elaborate that under quirks I meant quirk-clean GPIO driver, so, it wouldn't care about what platform(s) require base and what do not.

My point is that no mater how you pass the gpio-base value it will always be a quirk. The same thing is with the irq_shared flag. Both of them are specific to the Quark-platform. They are used to tune the DW APB GPIO driver up so one would create a GPIO device working well with the Quark-specific DW APB GPIO block.

Precisely!

Quoted 15 lines
Quoted 14 lines
Quoted 13 lines
Quoted 8 lines

We, by applying this series, make (keep) layers independent: board code vs. driver code. Mixing them more is the opposite to what I propose.

WRT property.
snps,gpio-base can be easily changed to *already in use* gpio-base or being both converted to linux,gpio-base to explicitly show people that this is internal stuff that must not be present in firmware nodes.

As I see it the part with "gpio-base" and the irq_shared can be moved to the gpio-dwapb.c driver to be defined as the quark-specific quirks. Adding a property like "gpio-base" seems like a quirk anyway, so I'd leave it defined in the driver.

Quoted 7 lines
Quoted 1 line

Huh?! The whole idea is make GPIO driver agnostic from platforms and their quirks.

As I said above having the gpio-base value set is the platform-specific thing in any case. So no mater whether you pass it via the software_node/properties or the private platform-data structure, the DW APB GPIO driver will have to handle those parameters in some special way, which is quirk-prone since normal platforms don't have such peculiar requirements.

Seems you are proposing layering violation without seeing it. Let me explain the design of the drivers in Linux kernel.

There are basically few possible concepts how drivers are handling quirks (because we know that most of the hardware support is a set of quirks here and there):
1) platform provides a board files and via platform data structures supplies quirks to the certain driver
2) platform provides a firmware node (ACPI, DT, ...) and unified driver handles it thru the fwnode interface
3) driver is split to be a library (or core part) + glue drivers full of quirks
4) driver has embedded quirks and supplies them based on IDs (PCI ID, ACPI ID, compatible string, ID table, etc)
5) ...missed something? ...

Quoted 6 lines

What I'm proposing is turn 1 to 2 for Quark case, what you are proposing is absent on the picture, i.e.:
x) platform provides a board file (intel_quark_i2c_gpio.c) and the driver has embedded quirks based on ID.

This is not what we want, definitely.

From the list you provided it's still not obvious why what I suggested wasn't good, because it's perfectly fine to have both ACPI/DT-based firmware node (entry 2) and quirks based on IDs (entry 4), which plenty of the kernel drivers do. The difference between the most of those drivers and what I suggested is that by bonding a device with a driver they provide !device!-quirks, but not the !platform!-quirks. While the platform quirks and platform properties are normally provided by means of the platform data, firmware nodes, glue layers, etc. Moving some of the platform-quirks to a driver you called "layering violation". Well, I've never met that definition in the kernel before (have you just come up with it?), but at least I see what you meant.

Anyway there are GPIO-drivers which still use the device IDs to get the platform quirks or get the GPIO-base from an of node alias (gpio-zynq.c, gpio-vf610.c, gpio-zx.c, gpio-mxc.c, gpio-mxs.c). There are even some, which either use a static variable to redistribute the GPIO-base between all available GPIO-chips (gpio-brcmstb.c, gpio-sta2x11.c, gpio-omap.c) or set a fixed GPIO-based value (like gpio-xlp.c, gpio-iop.c, gpio-ep93xx.c, gpio-vx855.c, gpio-cs5535.c, gpio-sch311x.c, gpio-loongson.c, gpio-loongson1.c, gpio-ath79.c, gpio-octeon.c), or even get the GPIO-base value from some hardware register (gpio-merrifield.c, gpio-intel-mid.c). I am pretty sure the examples of having the locally-defined platform quirks and the concepts 1, 2, 3 and 4 at some extent utilized in a single driver can be found in another subsystems too.

I am not saying, that the approaches utilized in those drivers are ideal. Those are just examples, that the platform specifics can be reflected in the corresponding drivers and the so called "layering violation" is allowed at some circumstances. Linus, correct me if I am wrong.

Getting back to this patchset. As I see it, the main problem here is connected with two parameters:
- GPIO-base. You suggest to update the gpio-dwapb.c driver so one would support a firmware property like "gpio-base". It's not good, since the property will be implicitly supported by OF API as well and nothing will prevent a user from using it. Even though you said that we won't advertise that property, some user may try to define it in dts anyway, which can be easily missed on review.
- IRQ-shared. As I said before it's not good to replace the irq_shared flag with the to_of_node() macro. Because having to_of_node() returned an of-node doesn't mean the IRQs can't be shared.

As I see it the convenience provided by your patchset in relation to the GPIO-base and IRQ-shared properties doesn't overcome the problems denoted above. IMO it would be better either to move the GPIO-base and the IRQ-shared parameters definition to the gpio-dwapb.c driver despite of the so called "layering violation" or just leave them in the MFD driver. Linus, please join the discussion. Do you have any better idea of what to do with these properties?

-Sergey

Quoted 53 lines

...

Quoted 5 lines
Quoted 1 line

What you are talking about? Can you provide a code we can discuss?

Here is what I suggest in step-by-step:
Patch 1) Move the Quark-specific GPIO-base and irq_shared parameters definition
to the gpio-dwapb.c driver.

Layering violation: spreading board code over the kernel.

Quoted 3 lines

Patch 2) Retrieve the GPIO-base and IRQ-shared parameters only for the DW APB
GPIO block living in the Quark platform (by means of either ACPI or the platform-device ID data).

Same. (The rest doesn't matter. I dropped it because it's basically the same what I proposed)

Quoted 16 lines

By doing as I suggested you don't have to pass around the GPIO-base value and the IRQ-shared flag. They will be initialized and utilized locally in the DW APB GPIO driver if Quark-specific DW APB GPIO block is detected (patches 1 and 2). In that case a software-node created in Patch 3 and 4 will be a normal firmware node described by the DW bindings. You'll still be able to get rid of the legacy global platform-data in the Patch 5.

The spirit of your original patchset will still be preserved: introduce the software-nodes propagation interface and use it to pass the generic DW APB GPIO parameters. The only difference is that we'd move the GPIO-base and IRQ-shared functionality fully into the DW APB GPIO driver. By doing so we'd have:
1) DW APB GPIO Quark-specifics localized in a single gpio-dwapb driver. In the current implementation we have them distributed between two drivers:
intel_quark_i2c_gpio initializes the GPIO-base and IRQ-shared parameters, gpio-dwapb uses them for setup. Note there is no any other platform with DW APB GPIO which needs the GPIO-base and IRQ-shared flag being setup. Why do

Quoted 8 lines

we need to have such complicated interface then if we can identify whether the particular DW APB GPIO is a Quark-specific block or not? Moreover having those specifics in both drivers in fact means having quirks in both of them.
2) Quirk-clean intel_quark_i2c_gpio driver which creates a normal DW APB GPIO device with software-nodes fully compatible with DT binding. DW APB GPIO driver will detect whether the block is Quark-specific and will activate the corresponding quirks if it is.
3) Still legacy global platform-data removed.

I see. But you mistakenly made a conclusion 1). It will be spread to two drivers.
And in 2) you considered (wrongly) that MFD driver is not a board code.

Again MFD driver _is_ a board code or i.o.w. quirk driver for specific Quark (MFD) device.

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#23 lore raw
Quoted 7 lines · Serge Semin
Quoted 6 lines · Andy Shevchenko
Quoted 5 lines · Serge Semin
Quoted 4 lines · Andy Shevchenko
Quoted 3 lines · Serge Semin
Quoted 2 lines · Andy Shevchenko
Quoted 1 line · Serge Semin

On Mon, Jun 08, 2020 at 04:42:54PM +0300, Andy Shevchenko wrote:

...

Quoted 5 lines
Quoted 1 line

Linus replied to three messages out of 7, seems you read only the first one.

Actually I've read all of them and from what I could see, he didn't like a code setting gpio-base regardless whether it has been created before your patchset or your explanation why it was necessary.

Yes, and then he realized what happened in the past. TBH I also do not like this kind of thing, but we may not get rid of it because we will get a regression.
The compromise (which is used in several cases of my knowledge, but I believe there are more) is to use internal property. This keeps layers separate and individual drivers nicer and neater.

...

Quoted 51 lines
Quoted 51 lines
Quoted 4 lines

My point is that no mater how you pass the gpio-base value it will always be a quirk. The same thing is with the irq_shared flag. Both of them are specific to the Quark-platform. They are used to tune the DW APB GPIO driver up so one would create a GPIO device working well with the Quark-specific DW APB GPIO block.

Precisely!

Quoted 15 lines
Quoted 14 lines
Quoted 13 lines
Quoted 8 lines

We, by applying this series, make (keep) layers independent: board code vs. driver code. Mixing them more is the opposite to what I propose.

WRT property.
snps,gpio-base can be easily changed to *already in use* gpio-base or being both converted to linux,gpio-base to explicitly show people that this is internal stuff that must not be present in firmware nodes.

As I see it the part with "gpio-base" and the irq_shared can be moved to the gpio-dwapb.c driver to be defined as the quark-specific quirks. Adding a property like "gpio-base" seems like a quirk anyway, so I'd leave it defined in the driver.

Quoted 7 lines
Quoted 1 line

Huh?! The whole idea is make GPIO driver agnostic from platforms and their quirks.

As I said above having the gpio-base value set is the platform-specific thing in any case. So no mater whether you pass it via the software_node/properties or the private platform-data structure, the DW APB GPIO driver will have to handle those parameters in some special way, which is quirk-prone since normal platforms don't have such peculiar requirements.

Seems you are proposing layering violation without seeing it. Let me explain the design of the drivers in Linux kernel.

There are basically few possible concepts how drivers are handling quirks (because we know that most of the hardware support is a set of quirks here and there):
1) platform provides a board files and via platform data structures supplies quirks to the certain driver
2) platform provides a firmware node (ACPI, DT, ...) and unified driver handles it thru the fwnode interface
3) driver is split to be a library (or core part) + glue drivers full of quirks
4) driver has embedded quirks and supplies them based on IDs (PCI ID, ACPI ID, compatible string, ID table, etc)
5) ...missed something? ...

What I'm proposing is turn 1 to 2 for Quark case, what you are proposing is absent on the picture, i.e.:
x) platform provides a board file (intel_quark_i2c_gpio.c) and the driver has embedded quirks based on ID.

Quoted 6 lines
Quoted 1 line

This is not what we want, definitely.

From the list you provided it's still not obvious why what I suggested wasn't good, because it's perfectly fine to have both ACPI/DT-based firmware node (entry 2) and quirks based on IDs (entry 4), which plenty of the kernel drivers do.

And I'm not talking about combinations of 2+4, what I'm talking about is 1+4 which *is* layering violation and a bad idea to start with.

Quoted 9 lines

The difference between the most of those drivers and what I suggested is that by bonding a device with a driver they provide !device!-quirks, but not the !platform!-quirks. While the platform quirks and platform properties are normally provided by means of the platform data, firmware nodes, glue layers, etc. Moving some of the platform-quirks to a driver you called "layering violation". Well, I've never met that definition in the kernel before

(have you just come up with it?),

Nope, I heard it many times during reviews.

Quoted 5 lines

but at least I see
what you meant.

Anyway there are GPIO-drivers which still use the device IDs to get the platform quirks

Maybe you missed that we have more players here than simple dwapb-gpio? How many of them are part of MFD (being used as MFD cells)?

Quoted 7 lines

or get the GPIO-base from an of node alias (gpio-zynq.c, gpio-vf610.c, gpio-zx.c, gpio-mxc.c, gpio-mxs.c). There are even some, which either use a static variable to redistribute the GPIO-base between all available GPIO-chips (gpio-brcmstb.c, gpio-sta2x11.c, gpio-omap.c) or set a fixed GPIO-based value (like gpio-xlp.c, gpio-iop.c, gpio-ep93xx.c, gpio-vx855.c, gpio-cs5535.c, gpio-sch311x.c, gpio-loongson.c, gpio-loongson1.c, gpio-ath79.c, gpio-octeon.c),

So, the question is above.

Quoted 2 lines

or even get the GPIO-base value from some hardware register (gpio-merrifield.c, gpio-intel-mid.c).

These examples are not good, they are not part of MFD and the latter one actually should be a glue driver to gpio-pxa.c.

Quoted 4 lines

I am pretty sure
the examples of having the locally-defined platform quirks and the concepts 1, 2, 3 and 4 at some extent utilized in a single driver can be found in another subsystems too.

I'm pretty sure layering violation can be found in many places in the kernel. It doesn't mean we have to take bad or different examples as suitable.

Quoted 4 lines

I am not saying, that the approaches utilized in those drivers are ideal. Those are just examples, that the platform specifics can be reflected in the corresponding drivers and the so called "layering violation" is allowed at some circumstances. Linus, correct me if I am wrong.

It depends how a certain driver is being used. In our case we don't need to spread board code over the kernel, we may be smarter than that!

Quoted 8 lines

Getting back to this patchset. As I see it, the main problem here is connected with two parameters:
- GPIO-base. You suggest to update the gpio-dwapb.c driver so one would support a firmware property like "gpio-base". It's not good, since the property will be implicitly supported by OF API as well and nothing will prevent a user from using it. Even though you said that we won't advertise that property, some user may try to define it in dts anyway, which can be easily missed on review.

No, if we don't advertise that and if we add "linux," prefix (see DWC3 for example) to be explicit what this is used for and why.

Quoted 3 lines

- IRQ-shared. As I said before it's not good to replace the irq_shared flag with the to_of_node() macro. Because having to_of_node() returned an of-node doesn't mean the IRQs can't be shared.

This is a valid point, but I would ask you, as a more familiar guy with the OF/DT system, what suggestion would be better?

Quoted 7 lines

As I see it the convenience provided by your patchset in relation to the GPIO-base and IRQ-shared properties doesn't overcome the problems denoted above. IMO it would be better either to move the GPIO-base and the IRQ-shared parameters definition to the gpio-dwapb.c driver despite of the so called "layering violation" or just leave them in the MFD driver. Linus, please join the discussion. Do you have any better idea of what to do with these properties?

Moving them to gpio-dwapb is a silly move. Better to do nothing if you insist, but I consider that is non-constructive.

--
With Best Regards,
Andy Shevchenko