Skip to contentSkip to conversation

can: tcan4x5x: Introduce tcan4552/4553

PATCH Superseded Markus Schneider-Pargmann 3 participants 18 messages Jun 21, 2023 – Jul 18, 2023

Subsystems (5)

v2 of 4 Superseded Latest is v4 →

lore
Share & tools
lore

Markus Schneider-Pargmann

author maintainer
#0 lore raw

v2 cover can: tcan4x5x: Introduce tcan4552/4553

Hi everyone,

This series introduces two new chips tcan-4552 and tcan-4553. The generic driver works in general but needs a few small changes. These are caused by the removal of wake and state pins.

In v2 I updated the bindings to use tcan4x5x always as a fallback. The driver now uses the first more specific binding if available. If the given binding does not match the chip that is present, a warning is printed and the correct driver data is loaded instead.

Based on v6.4-rc1.

Best,
Markus

Changes in v2:
- Update the binding documentation to specify tcan4552 and tcan4553 with the tcan4x5x as fallback
- Update the driver to use auto detection as well. If compatible differs from the ID2 register, use the ID2 register and print a warning.
- Small style changes

Previous versions:
v1 - https://lore.kernel.org/lkml/[email protected]

Markus Schneider-Pargmann (6):
dt-bindings: can: tcan4x5x: Add tcan4552 and tcan4553 variants
can: tcan4x5x: Remove reserved register 0x814 from writable table
can: tcan4x5x: Check size of mram configuration
can: tcan4x5x: Rename ID registers to match datasheet
can: tcan4x5x: Add support for tcan4552/4553
can: tcan4x5x: Add error messages in probe

 .../devicetree/bindings/net/can/tcan4x5x.txt  |  11 +-
 drivers/net/can/m_can/m_can.c                 |  16 ++
 drivers/net/can/m_can/m_can.h                 |   1 +
 drivers/net/can/m_can/tcan4x5x-core.c         | 161 ++++++++++++++----
 drivers/net/can/m_can/tcan4x5x-regmap.c       |   1 -
 5 files changed, 155 insertions(+), 35 deletions(-)

base-commit: ac9a78681b921877518763ba0e89202254349d1b

Signature · 1 line
2.40.1
author maintainer
#1 lore raw

PATCH v2 1/6 dt-bindings: can: tcan4x5x: Add tcan4552 and tcan4553 variants

These two new chips do not have state or wake pins.

---
 .../devicetree/bindings/net/can/tcan4x5x.txt          | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)
Documentation/devicetree/bindings/net/can/tcan4x5x.txt +8 -3
index e3501bfa22e9..170e23f0610d 100644
@@ -4,7 +4,10 @@ Texas Instruments TCAN4x5x CAN Controller
 This file provides device node information for the TCAN4x5x interface contains.
 
 Required properties:
-	- compatible: "ti,tcan4x5x"
+	- compatible:
+		"ti,tcan4552", "ti,tcan4x5x"
+		"ti,tcan4553", "ti,tcan4x5x" or
+		"ti,tcan4x5x"
 	- reg: 0
 	- #address-cells: 1
 	- #size-cells: 0
@@ -21,8 +24,10 @@ Optional properties:
 	- reset-gpios: Hardwired output GPIO. If not defined then software
 		       reset.
 	- device-state-gpios: Input GPIO that indicates if the device is in
-			      a sleep state or if the device is active.
-	- device-wake-gpios: Wake up GPIO to wake up the TCAN device.
+			      a sleep state or if the device is active. Not
+			      available with tcan4552/4553.
+	- device-wake-gpios: Wake up GPIO to wake up the TCAN device. Not
+			     available with tcan4552/4553.
 
 Example:
 tcan4x5x: tcan4x5x@0 {
Signature · 1 line
2.40.1
Acked-by on v2 1/6
#2 lore raw

Re: [PATCH v2 1/6] dt-bindings: can: tcan4x5x: Add tcan4552 and tcan4553 variants

Quoted 12 lines · Markus Schneider-Pargmann

These two new chips do not have state or wake pins.

---
 .../devicetree/bindings/net/can/tcan4x5x.txt          | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)
Documentation/devicetree/bindings/net/can/tcan4x5x.txt +0 -0
index e3501bfa22e9..170e23f0610d 100644

Documentation/devicetree/bindings/net/can/tcan4x5x.txt Texas Instruments TCAN4x5x CAN Controller · line 10

Quoted 8 lines
Documentation/devicetree/bindings/net/can/tcan4x5x.txt +4 -1
@@ -4,7 +4,10 @@ Texas Instruments TCAN4x5x CAN Controller
 This file provides device node information for the TCAN4x5x interface contains.
 
 Required properties:
-	- compatible: "ti,tcan4x5x"
+	- compatible:
+		"ti,tcan4552", "ti,tcan4x5x"
+		"ti,tcan4553", "ti,tcan4x5x" or
+		"ti,tcan4x5x"

One more example why wildcards are not allowed. Ehh,

Best regards,
Krzysztof

#3 lore raw

Re: [PATCH v2 1/6] dt-bindings: can: tcan4x5x: Add tcan4552 and tcan4553 variants

In reply to Markus Schneider-Pargmann
Quoted 3 lines · Markus Schneider-Pargmann

These two new chips do not have state or wake pins.

BTW, why did you ignore the tag?

This is a friendly reminder during the review process.

It looks like you received a tag and forgot to add it.

If you do not know the process, here is a short explanation:
Please add Acked-by/Reviewed-by/Tested-by tags when posting new versions. However, there's no need to repost patches *only* to add the tags. The upstream maintainer will do that for acks received on the version they apply.

https://elixir.bootlin.com/linux/v5.17/source/Documentation/process/submitting-patches.rst#L540

If a tag was not added on purpose, please state why and what changed.

Best regards,
Krzysztof

author maintainer
#4 lore raw

Re: [PATCH v2 1/6] dt-bindings: can: tcan4x5x: Add tcan4552 and tcan4553 variants

Hi Krzysztof,

Quoted 6 lines · Krzysztof Kozlowski
Quoted 3 lines · Markus Schneider-Pargmann

These two new chips do not have state or wake pins.

BTW, why did you ignore the tag?

Thanks for your Acked-by.

I did not add it from v1 because:
- We had a long discussion after you sent your tag
- I changed the binding documentation according to the discussion on v1 as stated in the cover letter:
"- Update the binding documentation to specify tcan4552 and tcan4553 with
the tcan4x5x as fallback"

Thanks,
Markus

author maintainer
#5 lore raw

PATCH v2 2/6 can: tcan4x5x: Remove reserved register 0x814 from writable table

In reply to Markus Schneider-Pargmann

The mentioned register is not writable. It is reserved and should not be written.

Fixes 39dbb21b6a29 ("can: tcan4x5x: Specify separate read/write ranges")
---
 drivers/net/can/m_can/tcan4x5x-regmap.c | 1 -
 1 file changed, 1 deletion(-)
drivers/net/can/m_can/tcan4x5x-regmap.c +0 -1
index 2b218ce04e9f..fafa6daa67e6 100644
@@ -95,7 +95,6 @@ static const struct regmap_range tcan4x5x_reg_table_wr_range[] = {
 	regmap_reg_range(0x000c, 0x0010),
 	/* Device configuration registers and Interrupt Flags*/
 	regmap_reg_range(0x0800, 0x080c),
-	regmap_reg_range(0x0814, 0x0814),
 	regmap_reg_range(0x0820, 0x0820),
 	regmap_reg_range(0x0830, 0x0830),
 	/* M_CAN */
Signature · 1 line
2.40.1
author maintainer
#6 lore raw

PATCH v2 3/6 can: tcan4x5x: Check size of mram configuration

In reply to Markus Schneider-Pargmann

To reduce debugging effort in case the mram is misconfigured, add this size check of the DT configuration. Currently if the mram configuration doesn't fit into the available MRAM it just overwrites other areas of the MRAM.

---
 drivers/net/can/m_can/m_can.c         | 16 ++++++++++++++++
 drivers/net/can/m_can/m_can.h         |  1 +
 drivers/net/can/m_can/tcan4x5x-core.c |  5 +++++
 3 files changed, 22 insertions(+)
drivers/net/can/m_can/m_can.c +16 -0
index a5003435802b..1cdd2675f022 100644
@@ -1887,6 +1887,22 @@ static int register_m_can_dev(struct net_device *dev)
 	return register_candev(dev);
 }
 
+int m_can_check_mram_cfg(struct m_can_classdev *cdev, u32 mram_max_size)
+{
+	u32 total_size;
+
+	total_size = cdev->mcfg[MRAM_TXB].off - cdev->mcfg[MRAM_SIDF].off +
+			cdev->mcfg[MRAM_TXB].num * TXB_ELEMENT_SIZE;
+	if (total_size > mram_max_size) {
+		dev_err(cdev->dev, "Total size of mram config(%u) exceeds mram(%u)\n",
+			total_size, mram_max_size);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(m_can_check_mram_cfg);
+
 static void m_can_of_parse_mram(struct m_can_classdev *cdev,
 				const u32 *mram_config_vals)
 {
drivers/net/can/m_can/m_can.h +1 -0
index a839dc71dc9b..d8150d8128e7 100644
@@ -101,6 +101,7 @@ int m_can_class_register(struct m_can_classdev *cdev);
 void m_can_class_unregister(struct m_can_classdev *cdev);
 int m_can_class_get_clocks(struct m_can_classdev *cdev);
 int m_can_init_ram(struct m_can_classdev *priv);
+int m_can_check_mram_cfg(struct m_can_classdev *cdev, u32 mram_max_size);
 
 int m_can_class_suspend(struct device *dev);
 int m_can_class_resume(struct device *dev);
drivers/net/can/m_can/tcan4x5x-core.c +5 -0
index 2342aa011647..e706518176e4 100644
@@ -80,6 +80,7 @@
 	 TCAN4X5X_MCAN_IR_RF1F)
 
 #define TCAN4X5X_MRAM_START 0x8000
+#define TCAN4X5X_MRAM_SIZE 0x800
 #define TCAN4X5X_MCAN_OFFSET 0x1000
 
 #define TCAN4X5X_CLEAR_ALL_INT 0xffffffff
@@ -307,6 +308,10 @@ static int tcan4x5x_can_probe(struct spi_device *spi)
 	if (!mcan_class)
 		return -ENOMEM;
 
+	ret = m_can_check_mram_cfg(mcan_class, TCAN4X5X_MRAM_SIZE);
+	if (ret)
+		goto out_m_can_class_free_dev;
+
 	priv = cdev_to_priv(mcan_class);
 
 	priv->power = devm_regulator_get_optional(&spi->dev, "vsup");
Signature · 1 line
2.40.1
author maintainer
#7 lore raw

PATCH v2 4/6 can: tcan4x5x: Rename ID registers to match datasheet

In reply to Markus Schneider-Pargmann

The datasheet calls these registers ID1 and ID2. Rename these to avoid confusion.

---
 drivers/net/can/m_can/tcan4x5x-core.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/can/m_can/tcan4x5x-core.c +2 -2
index e706518176e4..fb9375fa20ec 100644
@@ -6,8 +6,8 @@
 
 #define TCAN4X5X_EXT_CLK_DEF 40000000
 
-#define TCAN4X5X_DEV_ID0 0x00
-#define TCAN4X5X_DEV_ID1 0x04
+#define TCAN4X5X_DEV_ID1 0x00
+#define TCAN4X5X_DEV_ID2 0x04
 #define TCAN4X5X_REV 0x08
 #define TCAN4X5X_STATUS 0x0C
 #define TCAN4X5X_ERROR_STATUS_MASK 0x10
Signature · 1 line
2.40.1
author maintainer
#8 lore raw

PATCH v2 5/6 can: tcan4x5x: Add support for tcan4552/4553

In reply to Markus Schneider-Pargmann

tcan4552 and tcan4553 do not have wake or state pins, so they are currently not compatible with the generic driver. The generic driver uses tcan4x5x_disable_state() and tcan4x5x_disable_wake() if the gpios are not defined. These functions use register bits that are not available in tcan4552/4553.

This patch adds support by introducing version information to reflect if the chip has wake and state pins. Also the version is now checked.

---
 drivers/net/can/m_can/tcan4x5x-core.c | 128 +++++++++++++++++++++-----
 1 file changed, 104 insertions(+), 24 deletions(-)
drivers/net/can/m_can/tcan4x5x-core.c +104 -24
index fb9375fa20ec..756acd122075 100644
@@ -7,6 +7,7 @@
 #define TCAN4X5X_EXT_CLK_DEF 40000000
 
 #define TCAN4X5X_DEV_ID1 0x00
+#define TCAN4X5X_DEV_ID1_TCAN 0x4e414354 /* ASCII TCAN */
 #define TCAN4X5X_DEV_ID2 0x04
 #define TCAN4X5X_REV 0x08
 #define TCAN4X5X_STATUS 0x0C
@@ -103,6 +104,13 @@
 #define TCAN4X5X_WD_3_S_TIMER BIT(29)
 #define TCAN4X5X_WD_6_S_TIMER (BIT(28) | BIT(29))
 
+struct tcan4x5x_version_info {
+	u32 id2_register;
+
+	bool has_wake_pin;
+	bool has_state_pin;
+};
+
 static inline struct tcan4x5x_priv *cdev_to_priv(struct m_can_classdev *cdev)
 {
 	return container_of(cdev, struct tcan4x5x_priv, cdev);
@@ -254,18 +262,68 @@ static int tcan4x5x_disable_state(struct m_can_classdev *cdev)
 				  TCAN4X5X_DISABLE_INH_MSK, 0x01);
 }
 
-static int tcan4x5x_get_gpios(struct m_can_classdev *cdev)
+static const struct tcan4x5x_version_info tcan4x5x_generic;
+static const struct of_device_id tcan4x5x_of_match[];
+
+static const struct tcan4x5x_version_info
+*tcan4x5x_find_version_info(struct tcan4x5x_priv *priv, u32 id2_value)
+{
+	for (int i = 0; tcan4x5x_of_match[i].data; ++i) {
+		const struct tcan4x5x_version_info *vinfo =
+			tcan4x5x_of_match[i].data;
+		if (!vinfo->id2_register || id2_value == vinfo->id2_register) {
+			dev_warn(&priv->spi->dev, "TCAN device is %s, please use it in DT\n",
+				 tcan4x5x_of_match[i].compatible);
+			return vinfo;
+		}
+	}
+
+	return &tcan4x5x_generic;
+}
+
+static int tcan4x5x_verify_version(struct tcan4x5x_priv *priv,
+				   const struct tcan4x5x_version_info **info)
+{
+	u32 val;
+	int ret;
+
+	ret = regmap_read(priv->regmap, TCAN4X5X_DEV_ID1, &val);
+	if (ret)
+		return ret;
+
+	if (val != TCAN4X5X_DEV_ID1_TCAN) {
+		dev_err(&priv->spi->dev, "Not a tcan device %x\n", val);
+		return -ENODEV;
+	}
+
+	if (!(*info)->id2_register)
+		return 0;
+
+	ret = regmap_read(priv->regmap, TCAN4X5X_DEV_ID2, &val);
+	if (ret)
+		return ret;
+
+	if ((*info)->id2_register != val)
+		*info = tcan4x5x_find_version_info(priv, val);
+
+	return 0;
+}
+
+static int tcan4x5x_get_gpios(struct m_can_classdev *cdev,
+			      const struct tcan4x5x_version_info *version_info)
 {
 	struct tcan4x5x_priv *tcan4x5x = cdev_to_priv(cdev);
 	int ret;
 
-	tcan4x5x->device_wake_gpio = devm_gpiod_get(cdev->dev, "device-wake",
-						    GPIOD_OUT_HIGH);
-	if (IS_ERR(tcan4x5x->device_wake_gpio)) {
-		if (PTR_ERR(tcan4x5x->device_wake_gpio) == -EPROBE_DEFER)
-			return -EPROBE_DEFER;
+	if (version_info->has_wake_pin) {
+		tcan4x5x->device_wake_gpio = devm_gpiod_get(cdev->dev, "device-wake",
+							    GPIOD_OUT_HIGH);
+		if (IS_ERR(tcan4x5x->device_wake_gpio)) {
+			if (PTR_ERR(tcan4x5x->device_wake_gpio) == -EPROBE_DEFER)
+				return -EPROBE_DEFER;
 
-		tcan4x5x_disable_wake(cdev);
+			tcan4x5x_disable_wake(cdev);
+		}
 	}
 
 	tcan4x5x->reset_gpio = devm_gpiod_get_optional(cdev->dev, "reset",
@@ -277,12 +335,14 @@ static int tcan4x5x_get_gpios(struct m_can_classdev *cdev)
 	if (ret)
 		return ret;
 
-	tcan4x5x->device_state_gpio = devm_gpiod_get_optional(cdev->dev,
-							      "device-state",
-							      GPIOD_IN);
-	if (IS_ERR(tcan4x5x->device_state_gpio)) {
-		tcan4x5x->device_state_gpio = NULL;
-		tcan4x5x_disable_state(cdev);
+	if (version_info->has_state_pin) {
+		tcan4x5x->device_state_gpio = devm_gpiod_get_optional(cdev->dev,
+								      "device-state",
+								      GPIOD_IN);
+		if (IS_ERR(tcan4x5x->device_state_gpio)) {
+			tcan4x5x->device_state_gpio = NULL;
+			tcan4x5x_disable_state(cdev);
+		}
 	}
 
 	return 0;
@@ -299,10 +359,15 @@ static struct m_can_ops tcan4x5x_ops = {
 
 static int tcan4x5x_can_probe(struct spi_device *spi)
 {
+	const struct tcan4x5x_version_info *version_info;
 	struct tcan4x5x_priv *priv;
 	struct m_can_classdev *mcan_class;
 	int freq, ret;
 
+	version_info = of_device_get_match_data(&spi->dev);
+	if (!version_info)
+		version_info = (void *)spi_get_device_id(spi)->driver_data;
+
 	mcan_class = m_can_class_allocate_dev(&spi->dev,
 					      sizeof(struct tcan4x5x_priv));
 	if (!mcan_class)
@@ -361,7 +426,11 @@ static int tcan4x5x_can_probe(struct spi_device *spi)
 	if (ret)
 		goto out_m_can_class_free_dev;
 
-	ret = tcan4x5x_get_gpios(mcan_class);
+	ret = tcan4x5x_verify_version(priv, &version_info);
+	if (ret)
+		goto out_power;
+
+	ret = tcan4x5x_get_gpios(mcan_class, version_info);
 	if (ret)
 		goto out_power;
 
@@ -394,21 +463,32 @@ static void tcan4x5x_can_remove(struct spi_device *spi)
 	m_can_class_free_dev(priv->cdev.net);
 }
 
+static const struct tcan4x5x_version_info tcan4x5x_generic = {
+	.has_state_pin = true,
+	.has_wake_pin = true,
+};
+
+static const struct tcan4x5x_version_info tcan4x5x_tcan4552 = {
+	.id2_register = 0x32353534, /* ASCII = 4552 */
+};
+
+static const struct tcan4x5x_version_info tcan4x5x_tcan4553 = {
+	.id2_register = 0x33353534, /* ASCII = 4553 */
+};
+
 static const struct of_device_id tcan4x5x_of_match[] = {
-	{
-		.compatible = "ti,tcan4x5x",
-	}, {
-		/* sentinel */
-	},
+	{ .compatible = "ti,tcan4552", .data = &tcan4x5x_tcan4552 },
+	{ .compatible = "ti,tcan4553", .data = &tcan4x5x_tcan4553 },
+	{ .compatible = "ti,tcan4x5x", .data = &tcan4x5x_generic },
+	{ /* sentinel */ }
 };
 MODULE_DEVICE_TABLE(of, tcan4x5x_of_match);
 
 static const struct spi_device_id tcan4x5x_id_table[] = {
-	{
-		.name = "tcan4x5x",
-	}, {
-		/* sentinel */
-	},
+	{ .name = "tcan4x5x", .driver_data = (unsigned long)&tcan4x5x_generic, },
+	{ .name = "tcan4552", .driver_data = (unsigned long)&tcan4x5x_tcan4552, },
+	{ .name = "tcan4553", .driver_data = (unsigned long)&tcan4x5x_tcan4553, },
+	{ /* sentinel */ }
 };
 MODULE_DEVICE_TABLE(spi, tcan4x5x_id_table);
 
Signature · 1 line
2.40.1
#9 lore raw

Re: [PATCH v2 5/6] can: tcan4x5x: Add support for tcan4552/4553

Quoted 52 lines · Markus Schneider-Pargmann

tcan4552 and tcan4553 do not have wake or state pins, so they are currently not compatible with the generic driver. The generic driver uses tcan4x5x_disable_state() and tcan4x5x_disable_wake() if the gpios are not defined. These functions use register bits that are not available in tcan4552/4553.

This patch adds support by introducing version information to reflect if the chip has wake and state pins. Also the version is now checked.

---
 drivers/net/can/m_can/tcan4x5x-core.c | 128 +++++++++++++++++++++-----
 1 file changed, 104 insertions(+), 24 deletions(-)
drivers/net/can/m_can/tcan4x5x-core.c +15 -1
index fb9375fa20ec..756acd122075 100644
@@ -7,6 +7,7 @@
 #define TCAN4X5X_EXT_CLK_DEF 40000000
 
 #define TCAN4X5X_DEV_ID1 0x00
+#define TCAN4X5X_DEV_ID1_TCAN 0x4e414354 /* ASCII TCAN */
 #define TCAN4X5X_DEV_ID2 0x04
 #define TCAN4X5X_REV 0x08
 #define TCAN4X5X_STATUS 0x0C
@@ -103,6 +104,13 @@
 #define TCAN4X5X_WD_3_S_TIMER BIT(29)
 #define TCAN4X5X_WD_6_S_TIMER (BIT(28) | BIT(29))
 
+struct tcan4x5x_version_info {
+	u32 id2_register;
+
+	bool has_wake_pin;
+	bool has_state_pin;
+};
+
 static inline struct tcan4x5x_priv *cdev_to_priv(struct m_can_classdev *cdev)
 {
 	return container_of(cdev, struct tcan4x5x_priv, cdev);
@@ -254,18 +262,68 @@ static int tcan4x5x_disable_state(struct m_can_classdev *cdev)
 				  TCAN4X5X_DISABLE_INH_MSK, 0x01);
 }
 
-static int tcan4x5x_get_gpios(struct m_can_classdev *cdev)
+static const struct tcan4x5x_version_info tcan4x5x_generic;
+static const struct of_device_id tcan4x5x_of_match[];
+
+static const struct tcan4x5x_version_info
+*tcan4x5x_find_version_info(struct tcan4x5x_priv *priv, u32 id2_value)
+{
+	for (int i = 0; tcan4x5x_of_match[i].data; ++i) {

drivers/net/can/m_can/tcan4x5x-core.c static int tcan4x5x_disable_state(struct m_can_classdev *cdev) · line 281

Quoted 10 lines
drivers/net/can/m_can/tcan4x5x-core.c +10 -0
+		const struct tcan4x5x_version_info *vinfo =
+			tcan4x5x_of_match[i].data;
+		if (!vinfo->id2_register || id2_value == vinfo->id2_register) {
+			dev_warn(&priv->spi->dev, "TCAN device is %s, please use it in DT\n",
+				 tcan4x5x_of_match[i].compatible);
+			return vinfo;
+		}
+	}
+
+	return &tcan4x5x_generic;

I don't understand what do you want to achieve here. Kernel job is not to validate DTB, so if DTB says you have 4552, there is no need to double check. On the other hand, you have Id register so entire idea of custom compatibles can be dropped and instead you should detect the variant based on the ID.

Best regards,
Krzysztof

author maintainer
#10 lore raw

Re: [PATCH v2 5/6] can: tcan4x5x: Add support for tcan4552/4553

Hi Krzysztof,

Quoted 61 lines · Krzysztof Kozlowski
Quoted 60 lines · Markus Schneider-Pargmann

tcan4552 and tcan4553 do not have wake or state pins, so they are currently not compatible with the generic driver. The generic driver uses tcan4x5x_disable_state() and tcan4x5x_disable_wake() if the gpios are not defined. These functions use register bits that are not available in tcan4552/4553.

This patch adds support by introducing version information to reflect if the chip has wake and state pins. Also the version is now checked.

---
 drivers/net/can/m_can/tcan4x5x-core.c | 128 +++++++++++++++++++++-----
 1 file changed, 104 insertions(+), 24 deletions(-)
drivers/net/can/m_can/tcan4x5x-core.c +23 -1
index fb9375fa20ec..756acd122075 100644
@@ -7,6 +7,7 @@
 #define TCAN4X5X_EXT_CLK_DEF 40000000
 
 #define TCAN4X5X_DEV_ID1 0x00
+#define TCAN4X5X_DEV_ID1_TCAN 0x4e414354 /* ASCII TCAN */
 #define TCAN4X5X_DEV_ID2 0x04
 #define TCAN4X5X_REV 0x08
 #define TCAN4X5X_STATUS 0x0C
@@ -103,6 +104,13 @@
 #define TCAN4X5X_WD_3_S_TIMER BIT(29)
 #define TCAN4X5X_WD_6_S_TIMER (BIT(28) | BIT(29))
 
+struct tcan4x5x_version_info {
+	u32 id2_register;
+
+	bool has_wake_pin;
+	bool has_state_pin;
+};
+
 static inline struct tcan4x5x_priv *cdev_to_priv(struct m_can_classdev *cdev)
 {
 	return container_of(cdev, struct tcan4x5x_priv, cdev);
@@ -254,18 +262,68 @@ static int tcan4x5x_disable_state(struct m_can_classdev *cdev)
 				  TCAN4X5X_DISABLE_INH_MSK, 0x01);
 }
 
-static int tcan4x5x_get_gpios(struct m_can_classdev *cdev)
+static const struct tcan4x5x_version_info tcan4x5x_generic;
+static const struct of_device_id tcan4x5x_of_match[];
+
+static const struct tcan4x5x_version_info
+*tcan4x5x_find_version_info(struct tcan4x5x_priv *priv, u32 id2_value)
+{
+	for (int i = 0; tcan4x5x_of_match[i].data; ++i) {
+		const struct tcan4x5x_version_info *vinfo =
+			tcan4x5x_of_match[i].data;
+		if (!vinfo->id2_register || id2_value == vinfo->id2_register) {
+			dev_warn(&priv->spi->dev, "TCAN device is %s, please use it in DT\n",
+				 tcan4x5x_of_match[i].compatible);
+			return vinfo;
+		}
+	}
Quoted 8 lines

drivers/net/can/m_can/tcan4x5x-core.c static int tcan4x5x_disable_state(struct m_can_classdev *cdev) · line 281

Quoted 2 lines
drivers/net/can/m_can/tcan4x5x-core.c +2 -0
+
+	return &tcan4x5x_generic;

I don't understand what do you want to achieve here. Kernel job is not to validate DTB, so if DTB says you have 4552, there is no need to double check. On the other hand, you have Id register so entire idea of custom compatibles can be dropped and instead you should detect the variant based on the ID.

I can read the ID register but tcan4552 and 4553 do not have two devicetree properties that tcan4550 has, namely state and wake gpios. See v1 discussion about that [1].

In v1 Marc pointed out that mcp251xfd is using an autodetection and warn mechanism which I implemented here as well. [2]

Best,
Markus

[1] https://lore.kernel.org/lkml/[email protected]/
[2] https://lore.kernel.org/lkml/[email protected]/

#11 lore raw

Re: [PATCH v2 5/6] can: tcan4x5x: Add support for tcan4552/4553

Quoted 73 lines · Markus Schneider-Pargmann

Hi Krzysztof,

Quoted 61 lines · Krzysztof Kozlowski
Quoted 60 lines · Markus Schneider-Pargmann

tcan4552 and tcan4553 do not have wake or state pins, so they are currently not compatible with the generic driver. The generic driver uses tcan4x5x_disable_state() and tcan4x5x_disable_wake() if the gpios are not defined. These functions use register bits that are not available in tcan4552/4553.

This patch adds support by introducing version information to reflect if the chip has wake and state pins. Also the version is now checked.

---
 drivers/net/can/m_can/tcan4x5x-core.c | 128 +++++++++++++++++++++-----
 1 file changed, 104 insertions(+), 24 deletions(-)
drivers/net/can/m_can/tcan4x5x-core.c +23 -1
index fb9375fa20ec..756acd122075 100644
@@ -7,6 +7,7 @@
 #define TCAN4X5X_EXT_CLK_DEF 40000000
 
 #define TCAN4X5X_DEV_ID1 0x00
+#define TCAN4X5X_DEV_ID1_TCAN 0x4e414354 /* ASCII TCAN */
 #define TCAN4X5X_DEV_ID2 0x04
 #define TCAN4X5X_REV 0x08
 #define TCAN4X5X_STATUS 0x0C
@@ -103,6 +104,13 @@
 #define TCAN4X5X_WD_3_S_TIMER BIT(29)
 #define TCAN4X5X_WD_6_S_TIMER (BIT(28) | BIT(29))
 
+struct tcan4x5x_version_info {
+	u32 id2_register;
+
+	bool has_wake_pin;
+	bool has_state_pin;
+};
+
 static inline struct tcan4x5x_priv *cdev_to_priv(struct m_can_classdev *cdev)
 {
 	return container_of(cdev, struct tcan4x5x_priv, cdev);
@@ -254,18 +262,68 @@ static int tcan4x5x_disable_state(struct m_can_classdev *cdev)
 				  TCAN4X5X_DISABLE_INH_MSK, 0x01);
 }
 
-static int tcan4x5x_get_gpios(struct m_can_classdev *cdev)
+static const struct tcan4x5x_version_info tcan4x5x_generic;
+static const struct of_device_id tcan4x5x_of_match[];
+
+static const struct tcan4x5x_version_info
+*tcan4x5x_find_version_info(struct tcan4x5x_priv *priv, u32 id2_value)
+{
+	for (int i = 0; tcan4x5x_of_match[i].data; ++i) {
+		const struct tcan4x5x_version_info *vinfo =
+			tcan4x5x_of_match[i].data;
+		if (!vinfo->id2_register || id2_value == vinfo->id2_register) {
+			dev_warn(&priv->spi->dev, "TCAN device is %s, please use it in DT\n",
+				 tcan4x5x_of_match[i].compatible);
+			return vinfo;
+		}
+	}
Quoted 8 lines

drivers/net/can/m_can/tcan4x5x-core.c static int tcan4x5x_disable_state(struct m_can_classdev *cdev) · line 281

Quoted 2 lines
drivers/net/can/m_can/tcan4x5x-core.c +2 -0
+
+	return &tcan4x5x_generic;

I don't understand what do you want to achieve here. Kernel job is not to validate DTB, so if DTB says you have 4552, there is no need to double check. On the other hand, you have Id register so entire idea of custom compatibles can be dropped and instead you should detect the variant based on the ID.

Quoted 2 lines

I can read the ID register but tcan4552 and 4553 do not have two devicetree properties that tcan4550 has, namely state and wake gpios.

Does not matter, you don't use OF matching to then differentiate handling of GPIOs to then read the register. You first read registers, so everything is auto-detectable.

Quoted 1 line

See v1 discussion about that [1].

Yeah, but your code is different, although maybe we just misunderstood each other. You wrote that you cannot use the GPIOs, so I assumed you need to know the variant before using the GPIOs. Then you need compatibles. It's not the case here. You can read the variant and based on this skip entirely GPIOs as they are entirely missing.

Quoted 3 lines

In v1 Marc pointed out that mcp251xfd is using an autodetection and warn mechanism which I implemented here as well. [2]

But why? Just read the ID and detect the variant based on this. Your DT still can have separate compatibles followed by fallback, that's not a problem.

Best regards,
Krzysztof

author maintainer
#12 lore raw

Re: [PATCH v2 5/6] can: tcan4x5x: Add support for tcan4552/4553

Hi Krzysztof,

Quoted 81 lines · Krzysztof Kozlowski
Quoted 73 lines · Markus Schneider-Pargmann

Hi Krzysztof,

Quoted 61 lines · Krzysztof Kozlowski
Quoted 60 lines · Markus Schneider-Pargmann

tcan4552 and tcan4553 do not have wake or state pins, so they are currently not compatible with the generic driver. The generic driver uses tcan4x5x_disable_state() and tcan4x5x_disable_wake() if the gpios are not defined. These functions use register bits that are not available in tcan4552/4553.

This patch adds support by introducing version information to reflect if the chip has wake and state pins. Also the version is now checked.

---
 drivers/net/can/m_can/tcan4x5x-core.c | 128 +++++++++++++++++++++-----
 1 file changed, 104 insertions(+), 24 deletions(-)
drivers/net/can/m_can/tcan4x5x-core.c +23 -1
index fb9375fa20ec..756acd122075 100644
@@ -7,6 +7,7 @@
 #define TCAN4X5X_EXT_CLK_DEF 40000000
 
 #define TCAN4X5X_DEV_ID1 0x00
+#define TCAN4X5X_DEV_ID1_TCAN 0x4e414354 /* ASCII TCAN */
 #define TCAN4X5X_DEV_ID2 0x04
 #define TCAN4X5X_REV 0x08
 #define TCAN4X5X_STATUS 0x0C
@@ -103,6 +104,13 @@
 #define TCAN4X5X_WD_3_S_TIMER BIT(29)
 #define TCAN4X5X_WD_6_S_TIMER (BIT(28) | BIT(29))
 
+struct tcan4x5x_version_info {
+	u32 id2_register;
+
+	bool has_wake_pin;
+	bool has_state_pin;
+};
+
 static inline struct tcan4x5x_priv *cdev_to_priv(struct m_can_classdev *cdev)
 {
 	return container_of(cdev, struct tcan4x5x_priv, cdev);
@@ -254,18 +262,68 @@ static int tcan4x5x_disable_state(struct m_can_classdev *cdev)
 				  TCAN4X5X_DISABLE_INH_MSK, 0x01);
 }
 
-static int tcan4x5x_get_gpios(struct m_can_classdev *cdev)
+static const struct tcan4x5x_version_info tcan4x5x_generic;
+static const struct of_device_id tcan4x5x_of_match[];
+
+static const struct tcan4x5x_version_info
+*tcan4x5x_find_version_info(struct tcan4x5x_priv *priv, u32 id2_value)
+{
+	for (int i = 0; tcan4x5x_of_match[i].data; ++i) {
+		const struct tcan4x5x_version_info *vinfo =
+			tcan4x5x_of_match[i].data;
+		if (!vinfo->id2_register || id2_value == vinfo->id2_register) {
+			dev_warn(&priv->spi->dev, "TCAN device is %s, please use it in DT\n",
+				 tcan4x5x_of_match[i].compatible);
+			return vinfo;
+		}
+	}
Quoted 8 lines

drivers/net/can/m_can/tcan4x5x-core.c static int tcan4x5x_disable_state(struct m_can_classdev *cdev) · line 281

Quoted 2 lines
drivers/net/can/m_can/tcan4x5x-core.c +2 -0
+
+	return &tcan4x5x_generic;

I don't understand what do you want to achieve here. Kernel job is not to validate DTB, so if DTB says you have 4552, there is no need to double check. On the other hand, you have Id register so entire idea of custom compatibles can be dropped and instead you should detect the variant based on the ID.

Quoted 2 lines

I can read the ID register but tcan4552 and 4553 do not have two devicetree properties that tcan4550 has, namely state and wake gpios.

Does not matter, you don't use OF matching to then differentiate handling of GPIOs to then read the register. You first read registers, so everything is auto-detectable.

Quoted 7 lines
Quoted 1 line

See v1 discussion about that [1].

Yeah, but your code is different, although maybe we just misunderstood each other. You wrote that you cannot use the GPIOs, so I assumed you need to know the variant before using the GPIOs. Then you need compatibles. It's not the case here. You can read the variant and based on this skip entirely GPIOs as they are entirely missing.

The version information is always readable for that chip, regardless of state and wake GPIOs as far as I know. So yes it is possible to setup the GPIOs based on the content of the ID register.

I personally would prefer separate compatibles. The binding documentation needs to address that wake and state GPIOs are not available for tcan4552/4553. I think having compatibles that are for these chips would make sense then. However this is my opinion, you are the maintainer.

Best,
Markus

Quoted 13 lines
Quoted 4 lines

In v1 Marc pointed out that mcp251xfd is using an autodetection and warn mechanism which I implemented here as well. [2]

But why? Just read the ID and detect the variant based on this. Your DT still can have separate compatibles followed by fallback, that's not a problem.

Best regards,
Krzysztof

#13 lore raw

Re: [PATCH v2 5/6] can: tcan4x5x: Add support for tcan4552/4553

Quoted 11 lines · Markus Schneider-Pargmann
Quoted 6 lines

Yeah, but your code is different, although maybe we just misunderstood each other. You wrote that you cannot use the GPIOs, so I assumed you need to know the variant before using the GPIOs. Then you need compatibles. It's not the case here. You can read the variant and based on this skip entirely GPIOs as they are entirely missing.

The version information is always readable for that chip, regardless of state and wake GPIOs as far as I know. So yes it is possible to setup the GPIOs based on the content of the ID register.

Quoted 5 lines

I personally would prefer separate compatibles. The binding documentation needs to address that wake and state GPIOs are not available for tcan4552/4553. I think having compatibles that are for these chips would make sense then. However this is my opinion, you are the maintainer.

We do not talk about compatibles in the bindings here. This is discussion about your driver. The entire logic of validating DTB is flawed and not needed. Detect the variant and act based on this.

Best regards,
Krzysztof

author maintainer
#14 lore raw

Re: [PATCH v2 5/6] can: tcan4x5x: Add support for tcan4552/4553

Hi Krzysztof,

Quoted 18 lines · Krzysztof Kozlowski
Quoted 11 lines · Markus Schneider-Pargmann
Quoted 6 lines

Yeah, but your code is different, although maybe we just misunderstood each other. You wrote that you cannot use the GPIOs, so I assumed you need to know the variant before using the GPIOs. Then you need compatibles. It's not the case here. You can read the variant and based on this skip entirely GPIOs as they are entirely missing.

The version information is always readable for that chip, regardless of state and wake GPIOs as far as I know. So yes it is possible to setup the GPIOs based on the content of the ID register.

Quoted 5 lines

I personally would prefer separate compatibles. The binding documentation needs to address that wake and state GPIOs are not available for tcan4552/4553. I think having compatibles that are for these chips would make sense then. However this is my opinion, you are the maintainer.

Quoted 3 lines

We do not talk about compatibles in the bindings here. This is discussion about your driver. The entire logic of validating DTB is flawed and not needed. Detect the variant and act based on this.

I thought it was about the bindings, sorry.

So to summarize the compatibles ti,tcan4552 and ti,tcan4553 are fine. But the driver should use the ID register for detection and not compare the detected variant with the given compatible?

In my opinion it is useful to have an error messages that says there is something wrong with the devicetree as this can be very helpful for the developers who bringup new devices. This helps to quickly find issues with the devicetree.

@Marc: What do you think? Maybe I misinterpreted your mail from last version?

Best,
Markus

#15 lore raw

Re: [PATCH v2 5/6] can: tcan4x5x: Add support for tcan4552/4553

Quoted 16 lines · Markus Schneider-Pargmann
Quoted 10 lines
Quoted 9 lines

The version information is always readable for that chip, regardless of state and wake GPIOs as far as I know. So yes it is possible to setup the GPIOs based on the content of the ID register.

I personally would prefer separate compatibles. The binding documentation needs to address that wake and state GPIOs are not available for tcan4552/4553. I think having compatibles that are for these chips would make sense then. However this is my opinion, you are the maintainer.

Quoted 3 lines

We do not talk about compatibles in the bindings here. This is discussion about your driver. The entire logic of validating DTB is flawed and not needed. Detect the variant and act based on this.

I thought it was about the bindings, sorry.

Quoted 8 lines

So to summarize the compatibles ti,tcan4552 and ti,tcan4553 are fine. But the driver should use the ID register for detection and not compare the detected variant with the given compatible?

In my opinion it is useful to have an error messages that says there is something wrong with the devicetree as this can be very helpful for the developers who bringup new devices. This helps to quickly find issues with the devicetree.

That's not a current policy for other drivers, so this shouldn't be really special. Kernel is poor in validating DTS. It's not its job. It's the job of the DT schema.

Best regards,
Krzysztof

#16 lore raw

Re: [PATCH v2 5/6] can: tcan4x5x: Add support for tcan4552/4553

Quoted 22 lines · Krzysztof Kozlowski
Quoted 20 lines · Markus Schneider-Pargmann
Quoted 10 lines
Quoted 9 lines

The version information is always readable for that chip, regardless of state and wake GPIOs as far as I know. So yes it is possible to setup the GPIOs based on the content of the ID register.

I personally would prefer separate compatibles. The binding documentation needs to address that wake and state GPIOs are not available for tcan4552/4553. I think having compatibles that are for these chips would make sense then. However this is my opinion, you are the maintainer.

Quoted 3 lines

We do not talk about compatibles in the bindings here. This is discussion about your driver. The entire logic of validating DTB is flawed and not needed. Detect the variant and act based on this.

I thought it was about the bindings, sorry.

So to summarize the compatibles ti,tcan4552 and ti,tcan4553 are fine. But the driver should use the ID register for detection and not compare the detected variant with the given compatible?

Quoted 8 lines
Quoted 4 lines

In my opinion it is useful to have an error messages that says there is something wrong with the devicetree as this can be very helpful for the developers who bringup new devices. This helps to quickly find issues with the devicetree.

That's not a current policy for other drivers, so this shouldn't be really special. Kernel is poor in validating DTS. It's not its job. It's the job of the DT schema.

Fine with me.

I decided to have a check of the auto-detected chip variant against the specified one in the mcp251xfd driver, as it widely used with raspi boards, where commonly DT overlays are used. It also helps remote diagnostics of people, who don't focus on kernel development.

regards,
Marc

Signature · 4 lines
Pengutronix e.K.                 | Marc Kleine-Budde          |
Embedded Linux                   | https://www.pengutronix.de |
Vertretung Nürnberg              | Phone: +49-5121-206917-129 |
Amtsgericht Hildesheim, HRA 2686 | Fax:   +49-5121-206917-9   |
author maintainer
#17 lore raw

PATCH v2 6/6 can: tcan4x5x: Add error messages in probe

In reply to Markus Schneider-Pargmann

To be able to understand issues during probe easier, add error messages if something fails.

---
 drivers/net/can/m_can/tcan4x5x-core.c | 26 ++++++++++++++++++++------
 1 file changed, 20 insertions(+), 6 deletions(-)
drivers/net/can/m_can/tcan4x5x-core.c +20 -6
index 756acd122075..e30faa1cf893 100644
@@ -397,6 +397,8 @@ static int tcan4x5x_can_probe(struct spi_device *spi)
 
 	/* Sanity check */
 	if (freq < 20000000 || freq > TCAN4X5X_EXT_CLK_DEF) {
+		dev_err(&spi->dev, "Clock frequency is out of supported range %d\n",
+			freq);
 		ret = -ERANGE;
 		goto out_m_can_class_free_dev;
 	}
@@ -415,32 +417,44 @@ static int tcan4x5x_can_probe(struct spi_device *spi)
 	/* Configure the SPI bus */
 	spi->bits_per_word = 8;
 	ret = spi_setup(spi);
-	if (ret)
+	if (ret) {
+		dev_err(&spi->dev, "SPI setup failed %d\n", ret);
 		goto out_m_can_class_free_dev;
+	}
 
 	ret = tcan4x5x_regmap_init(priv);
-	if (ret)
+	if (ret) {
+		dev_err(&spi->dev, "regmap init failed %d\n", ret);
 		goto out_m_can_class_free_dev;
+	}
 
 	ret = tcan4x5x_power_enable(priv->power, 1);
-	if (ret)
+	if (ret) {
+		dev_err(&spi->dev, "Enabling regulator failed %d\n", ret);
 		goto out_m_can_class_free_dev;
+	}
 
 	ret = tcan4x5x_verify_version(priv, &version_info);
 	if (ret)
 		goto out_power;
 
 	ret = tcan4x5x_get_gpios(mcan_class, version_info);
-	if (ret)
+	if (ret) {
+		dev_err(&spi->dev, "Getting gpios failed %d\n", ret);
 		goto out_power;
+	}
 
 	ret = tcan4x5x_init(mcan_class);
-	if (ret)
+	if (ret) {
+		dev_err(&spi->dev, "tcan initialization failed %d\n", ret);
 		goto out_power;
+	}
 
 	ret = m_can_class_register(mcan_class);
-	if (ret)
+	if (ret) {
+		dev_err(&spi->dev, "Failed registering m_can device %d\n", ret);
 		goto out_power;
+	}
 
 	netdev_info(mcan_class->net, "TCAN4X5X successfully initialized.\n");
 	return 0;
Signature · 1 line
2.40.1