[PATCH v2 0/3] Add Support for MCAN in AM654x-idk

This series adds driver patches to support MCAN in TI's AM654x-idk.

v2: Model the sandby line to the transceiver as a regulator.

Faiz Abbas (3):
  dt-bindings: m_can: Add Documentation for transceiver regulator
  can: m_can: m_can_platform: Add support for enabling transceiver
  arm64: defconfig: Add Support for Bosch M_CAN controllers

 Documentation/devicetree/bindings/net/can/m_can.txt | 3 +++
 arch/arm64/configs/defconfig                        | 3 +++
 drivers/net/can/m_can/m_can_platform.c              | 6 ++++++
 3 files changed, 12 insertions(+)

-- 
2.19.2

Enable configs for supporting Bosch M_CAN controllers.

Signed-off-by: Faiz Abbas <redacted>
---
 arch/arm64/configs/defconfig | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/arch/arm64/configs/defconfig b/arch/arm64/configs/defconfig
index 905109f6814f..b25bbcf691b6 100644
--- a/arch/arm64/configs/defconfig
+++ b/arch/arm64/configs/defconfig

@@ -161,6 +161,9 @@ CONFIG_QRTR=m
 CONFIG_QRTR_SMD=m
 CONFIG_QRTR_TUN=m
 CONFIG_BPF_JIT=y
+CONFIG_CAN=m
+CONFIG_CAN_M_CAN=m
+CONFIG_CAN_M_CAN_PLATFORM=m
 CONFIG_BT=m
 CONFIG_BT_HIDP=m
 # CONFIG_BT_HS is not set

-- 
2.19.2

CAN transceivers on some boards have a standby line which can be
toggled to enable/disable the transceiver. Model this as an optional
fixed xceiver regulator.

Signed-off-by: Faiz Abbas <redacted>
Acked-by: Sriram Dash <redacted>
---
 drivers/net/can/m_can/m_can_platform.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/net/can/m_can/m_can_platform.c b/drivers/net/can/m_can/m_can_platform.c
index 38ea5e600fb8..719468fab507 100644
--- a/drivers/net/can/m_can/m_can_platform.c
+++ b/drivers/net/can/m_can/m_can_platform.c

@@ -6,6 +6,7 @@
 // Copyright (C) 2018-19 Texas Instruments Incorporated - http://www.ti.com/
 
 #include <linux/platform_device.h>
+#include <linux/regulator/consumer.h>
 
 #include "m_can.h"
 

@@ -57,6 +58,7 @@ static int m_can_plat_probe(struct platform_device *pdev)
 {
 	struct m_can_classdev *mcan_class;
 	struct m_can_plat_priv *priv;
+	struct regulator *reg_xceiver;
 	struct resource *res;
 	void __iomem *addr;
 	void __iomem *mram_addr;

@@ -111,6 +113,10 @@ static int m_can_plat_probe(struct platform_device *pdev)
 
 	m_can_init_ram(mcan_class);
 
+	reg_xceiver = devm_regulator_get_optional(&pdev->dev, "xceiver");
+	if (PTR_ERR(reg_xceiver) == -EPROBE_DEFER)
+		return -EPROBE_DEFER;
+
 	ret = m_can_class_register(mcan_class);
 
 failed_ret:

-- 
2.19.2

Some CAN transceivers have a standby line that needs to be asserted
before they can be used. Model this GPIO lines as an optional
fixed-regulator node. Document bindings for the same.

Signed-off-by: Faiz Abbas <redacted>
---
 Documentation/devicetree/bindings/net/can/m_can.txt | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/Documentation/devicetree/bindings/net/can/m_can.txt b/Documentation/devicetree/bindings/net/can/m_can.txt
index ed614383af9c..f17e2a5207dc 100644
--- a/Documentation/devicetree/bindings/net/can/m_can.txt
+++ b/Documentation/devicetree/bindings/net/can/m_can.txt

@@ -48,6 +48,9 @@ Optional Subnode:
 			  that can be used for CAN/CAN-FD modes. See
 			  Documentation/devicetree/bindings/net/can/can-transceiver.txt
 			  for details.
+
+- xceiver-supply: Regulator that powers the CAN transceiver.
+
 Example:
 SoC dtsi:
 m_can1: can@20e8000 {

-- 
2.19.2

Faiz

On 2/17/20 8:28 AM, Faiz Abbas wrote:

Where is this regulator enabled?

Shouldn't the regulator be managed by runtime PM as well?

Dan

On Mon, Feb 17, 2020 at 07:58:34PM +0530, Faiz Abbas wrote:

Some CAN transceivers have a standby line that needs to be asserted
before they can be used. Model this GPIO lines as an optional
fixed-regulator node. Document bindings for the same.

Signed-off-by: Faiz Abbas <redacted>
---
 Documentation/devicetree/bindings/net/can/m_can.txt | 3 +++
 1 file changed, 3 insertions(+)

This has moved to DT schema in my tree, so please adjust it and resend.

The supply for a transceiver should go in the transceiver node.

+
 Example:
 SoC dtsi:
 m_can1: can@20e8000 {
-- 
2.19.2

Hi Rob,

On 20/02/20 2:05 am, Rob Herring wrote:

On Mon, Feb 17, 2020 at 07:58:34PM +0530, Faiz Abbas wrote:

quoted

Some CAN transceivers have a standby line that needs to be asserted
before they can be used. Model this GPIO lines as an optional
fixed-regulator node. Document bindings for the same.

Signed-off-by: Faiz Abbas <redacted>
---
 Documentation/devicetree/bindings/net/can/m_can.txt | 3 +++
 1 file changed, 3 insertions(+)

This has moved to DT schema in my tree, so please adjust it and resend.

Ok.

quoted

diff --git a/Documentation/devicetree/bindings/net/can/m_can.txt b/Documentation/devicetree/bindings/net/can/m_can.txt
index ed614383af9c..f17e2a5207dc 100644
--- a/Documentation/devicetree/bindings/net/can/m_can.txt
+++ b/Documentation/devicetree/bindings/net/can/m_can.txt

@@ -48,6 +48,9 @@ Optional Subnode:
 			  that can be used for CAN/CAN-FD modes. See
 			  Documentation/devicetree/bindings/net/can/can-transceiver.txt
 			  for details.
+
+- xceiver-supply: Regulator that powers the CAN transceiver.

The supply for a transceiver should go in the transceiver node.

Marc, while I have you here, do you agree with this?

Thanks,
Faiz

On 2/21/20 9:31 AM, Faiz Abbas wrote:

Hi Rob,

On 20/02/20 2:05 am, Rob Herring wrote:

quoted

On Mon, Feb 17, 2020 at 07:58:34PM +0530, Faiz Abbas wrote:

quoted

Some CAN transceivers have a standby line that needs to be asserted
before they can be used. Model this GPIO lines as an optional
fixed-regulator node. Document bindings for the same.

Signed-off-by: Faiz Abbas <redacted>
---
 Documentation/devicetree/bindings/net/can/m_can.txt | 3 +++
 1 file changed, 3 insertions(+)

This has moved to DT schema in my tree, so please adjust it and resend.

Ok.

quoted

quoted

diff --git a/Documentation/devicetree/bindings/net/can/m_can.txt b/Documentation/devicetree/bindings/net/can/m_can.txt
index ed614383af9c..f17e2a5207dc 100644
--- a/Documentation/devicetree/bindings/net/can/m_can.txt
+++ b/Documentation/devicetree/bindings/net/can/m_can.txt

@@ -48,6 +48,9 @@ Optional Subnode:
 			  that can be used for CAN/CAN-FD modes. See
 			  Documentation/devicetree/bindings/net/can/can-transceiver.txt
 			  for details.
+
+- xceiver-supply: Regulator that powers the CAN transceiver.

The supply for a transceiver should go in the transceiver node.

Marc, while I have you here, do you agree with this?

I'll look into the details later today.

Marc

-- 
Pengutronix e.K.                 | Marc Kleine-Budde           |
Embedded Linux                   | https://www.pengutronix.de  |
Vertretung West/Dortmund         | Phone: +49-231-2826-924     |
Amtsgericht Hildesheim, HRA 2686 | Fax:   +49-5121-206917-5555 |

On 2/21/20 8:54 AM, Faiz Abbas wrote:

Hi Dan,

On 17/02/20 8:40 pm, Dan Murphy wrote:

quoted

Faiz

On 2/17/20 8:28 AM, Faiz Abbas wrote:

quoted

CAN transceivers on some boards have a standby line which can be
toggled to enable/disable the transceiver. Model this as an optional
fixed xceiver regulator.

Signed-off-by: Faiz Abbas <redacted>
Acked-by: Sriram Dash <redacted>
---
  drivers/net/can/m_can/m_can_platform.c | 6 ++++++
  1 file changed, 6 insertions(+)

diff --git a/drivers/net/can/m_can/m_can_platform.c

b/drivers/net/can/m_can/m_can_platform.c
index 38ea5e600fb8..719468fab507 100644

--- a/drivers/net/can/m_can/m_can_platform.c
+++ b/drivers/net/can/m_can/m_can_platform.c

@@ -6,6 +6,7 @@

  // Copyright (C) 2018-19 Texas Instruments Incorporated -
http://www.ti.com/
    #include <linux/platform_device.h>
+#include <linux/regulator/consumer.h>
    #include "m_can.h"
  @@ -57,6 +58,7 @@ static int m_can_plat_probe(struct platform_device
*pdev)
  {
      struct m_can_classdev *mcan_class;
      struct m_can_plat_priv *priv;
+    struct regulator *reg_xceiver;
      struct resource *res;
      void __iomem *addr;
      void __iomem *mram_addr;

@@ -111,6 +113,10 @@ static int m_can_plat_probe(struct

platform_device *pdev)
        m_can_init_ram(mcan_class);
  +    reg_xceiver = devm_regulator_get_optional(&pdev->dev, "xceiver");
+    if (PTR_ERR(reg_xceiver) == -EPROBE_DEFER)
+        return -EPROBE_DEFER;
+

Where is this regulator enabled?

I have set regulator-boot-on flag in the dt so this didn't require an
enable.

Please don't do this, please handle it properly.

quoted

Shouldn't the regulator be managed by runtime PM as well?

If so, then make the runtime pm kconfig option mandatory.

Marc

-- 
Pengutronix e.K.                 | Marc Kleine-Budde           |
Embedded Linux                   | https://www.pengutronix.de  |
Vertretung West/Dortmund         | Phone: +49-231-2826-924     |
Amtsgericht Hildesheim, HRA 2686 | Fax:   +49-5121-206917-5555 |

Hi Dan,

On 17/02/20 8:40 pm, Dan Murphy wrote:

Faiz

On 2/17/20 8:28 AM, Faiz Abbas wrote:

quoted

CAN transceivers on some boards have a standby line which can be
toggled to enable/disable the transceiver. Model this as an optional
fixed xceiver regulator.

Signed-off-by: Faiz Abbas <redacted>
Acked-by: Sriram Dash <redacted>
---
  drivers/net/can/m_can/m_can_platform.c | 6 ++++++
  1 file changed, 6 insertions(+)

diff --git a/drivers/net/can/m_can/m_can_platform.c

b/drivers/net/can/m_can/m_can_platform.c
index 38ea5e600fb8..719468fab507 100644

--- a/drivers/net/can/m_can/m_can_platform.c
+++ b/drivers/net/can/m_can/m_can_platform.c

@@ -6,6 +6,7 @@

  // Copyright (C) 2018-19 Texas Instruments Incorporated -
http://www.ti.com/
    #include <linux/platform_device.h>
+#include <linux/regulator/consumer.h>
    #include "m_can.h"
  @@ -57,6 +58,7 @@ static int m_can_plat_probe(struct platform_device
*pdev)
  {
      struct m_can_classdev *mcan_class;
      struct m_can_plat_priv *priv;
+    struct regulator *reg_xceiver;
      struct resource *res;
      void __iomem *addr;
      void __iomem *mram_addr;

@@ -111,6 +113,10 @@ static int m_can_plat_probe(struct

platform_device *pdev)
        m_can_init_ram(mcan_class);
  +    reg_xceiver = devm_regulator_get_optional(&pdev->dev, "xceiver");
+    if (PTR_ERR(reg_xceiver) == -EPROBE_DEFER)
+        return -EPROBE_DEFER;
+

Where is this regulator enabled?

I have set regulator-boot-on flag in the dt so this didn't require an
enable.

Shouldn't the regulator be managed by runtime PM as well?

Let me try this out.

Thanks,
Faiz

Hi Marc,

On 21/02/20 2:01 pm, Marc Kleine-Budde wrote:

On 2/21/20 9:31 AM, Faiz Abbas wrote:

quoted

Hi Rob,

On 20/02/20 2:05 am, Rob Herring wrote:

quoted

On Mon, Feb 17, 2020 at 07:58:34PM +0530, Faiz Abbas wrote:

quoted

Some CAN transceivers have a standby line that needs to be asserted
before they can be used. Model this GPIO lines as an optional
fixed-regulator node. Document bindings for the same.

Signed-off-by: Faiz Abbas <redacted>
---
 Documentation/devicetree/bindings/net/can/m_can.txt | 3 +++
 1 file changed, 3 insertions(+)

This has moved to DT schema in my tree, so please adjust it and resend.

Ok.

quoted

quoted

diff --git a/Documentation/devicetree/bindings/net/can/m_can.txt b/Documentation/devicetree/bindings/net/can/m_can.txt
index ed614383af9c..f17e2a5207dc 100644
--- a/Documentation/devicetree/bindings/net/can/m_can.txt
+++ b/Documentation/devicetree/bindings/net/can/m_can.txt

@@ -48,6 +48,9 @@ Optional Subnode:
 			  that can be used for CAN/CAN-FD modes. See
 			  Documentation/devicetree/bindings/net/can/can-transceiver.txt
 			  for details.
+
+- xceiver-supply: Regulator that powers the CAN transceiver.

The supply for a transceiver should go in the transceiver node.

Marc, while I have you here, do you agree with this?

I'll look into the details later today.

Sure. Be sure to take another look at my attempt to use the transceiver
with a phy driver some time ago.

https://lore.kernel.org/patchwork/patch/1006238/

Thanks,
Faiz

Marc,

On 26/02/20 2:40 pm, Faiz Abbas wrote:

Hi Marc,

On 21/02/20 2:01 pm, Marc Kleine-Budde wrote:

quoted

On 2/21/20 9:31 AM, Faiz Abbas wrote:

quoted

Hi Rob,

On 20/02/20 2:05 am, Rob Herring wrote:

quoted

On Mon, Feb 17, 2020 at 07:58:34PM +0530, Faiz Abbas wrote:

quoted

Some CAN transceivers have a standby line that needs to be asserted
before they can be used. Model this GPIO lines as an optional
fixed-regulator node. Document bindings for the same.

Signed-off-by: Faiz Abbas <redacted>
---
 Documentation/devicetree/bindings/net/can/m_can.txt | 3 +++
 1 file changed, 3 insertions(+)

This has moved to DT schema in my tree, so please adjust it and resend.

Ok.

quoted

quoted

diff --git a/Documentation/devicetree/bindings/net/can/m_can.txt b/Documentation/devicetree/bindings/net/can/m_can.txt
index ed614383af9c..f17e2a5207dc 100644
--- a/Documentation/devicetree/bindings/net/can/m_can.txt
+++ b/Documentation/devicetree/bindings/net/can/m_can.txt

@@ -48,6 +48,9 @@ Optional Subnode:
 			  that can be used for CAN/CAN-FD modes. See
 			  Documentation/devicetree/bindings/net/can/can-transceiver.txt
 			  for details.
+
+- xceiver-supply: Regulator that powers the CAN transceiver.

The supply for a transceiver should go in the transceiver node.

Marc, while I have you here, do you agree with this?

I'll look into the details later today.

Sure. Be sure to take another look at my attempt to use the transceiver
with a phy driver some time ago.

https://lore.kernel.org/patchwork/patch/1006238/

Do you have any comments?

Thanks,
Faiz

Hi Marc,

On 02/03/20 1:46 pm, Faiz Abbas wrote:

Marc,

On 26/02/20 2:40 pm, Faiz Abbas wrote:

quoted

Hi Marc,

On 21/02/20 2:01 pm, Marc Kleine-Budde wrote:

quoted

On 2/21/20 9:31 AM, Faiz Abbas wrote:

quoted

Hi Rob,

On 20/02/20 2:05 am, Rob Herring wrote:

quoted

On Mon, Feb 17, 2020 at 07:58:34PM +0530, Faiz Abbas wrote:

quoted

Some CAN transceivers have a standby line that needs to be asserted
before they can be used. Model this GPIO lines as an optional
fixed-regulator node. Document bindings for the same.

Signed-off-by: Faiz Abbas <redacted>
---
 Documentation/devicetree/bindings/net/can/m_can.txt | 3 +++
 1 file changed, 3 insertions(+)

This has moved to DT schema in my tree, so please adjust it and resend.

Ok.

quoted

quoted

diff --git a/Documentation/devicetree/bindings/net/can/m_can.txt b/Documentation/devicetree/bindings/net/can/m_can.txt
index ed614383af9c..f17e2a5207dc 100644
--- a/Documentation/devicetree/bindings/net/can/m_can.txt
+++ b/Documentation/devicetree/bindings/net/can/m_can.txt

@@ -48,6 +48,9 @@ Optional Subnode:
 			  that can be used for CAN/CAN-FD modes. See
 			  Documentation/devicetree/bindings/net/can/can-transceiver.txt
 			  for details.
+
+- xceiver-supply: Regulator that powers the CAN transceiver.

The supply for a transceiver should go in the transceiver node.

Marc, while I have you here, do you agree with this?

I'll look into the details later today.

Sure. Be sure to take another look at my attempt to use the transceiver
with a phy driver some time ago.

https://lore.kernel.org/patchwork/patch/1006238/

Do you have any comments?

Gentle ping.

Thanks,
Faiz