Skip to contentSkip to conversation

mfd: da9063: Support SMBus and I2C mode

PATCH Superseded Mark Jonas 3 participants 6 messages Mar 15, 2021 – Mar 16, 2021

Subsystems (3)

v5 of 6 Superseded Latest is v6 →

lore
Share & tools
lore

Mark Jonas

author
#0 lore raw

PATCH v5 1/1 mfd: da9063: Support SMBus and I2C mode

From: Hubert Streidl <[email protected]>

By default the PMIC DA9063 2-wire interface is SMBus compliant. This means the PMIC will automatically reset the interface when the clock signal ceases for more than the SMBus timeout of 35 ms.

If the I2C driver / device is not capable of creating atomic I2C transactions, a context change can cause a ceasing of the clock signal. This can happen if for example a real-time thread is scheduled. Then the DA9063 in SMBus mode will reset the 2-wire interface. Subsequently a write message could end up in the wrong register. This could cause unpredictable system behavior.

The DA9063 PMIC also supports an I2C compliant mode for the 2-wire interface. This mode does not reset the interface when the clock signal ceases. Thus the problem depicted above does not occur.

This patch tests for the bus functionality "I2C_FUNC_I2C". It can reasonably be assumed that the bus cannot obey SMBus timings if this functionality is set. SMBus commands most probably are emulated in this case which is prone to the latency issue described above.

This patch enables the I2C bus mode if I2C_FUNC_I2C is set or otherwise keeps the default SMBus mode.

---
 drivers/mfd/da9063-i2c.c             | 10 ++++++++++
 include/linux/mfd/da9063/registers.h |  3 +++
 2 files changed, 13 insertions(+)
drivers/mfd/da9063-i2c.c +10 -0
index 3781d0bb7786..e8a022e697c5 100644
@@ -442,6 +442,16 @@ static int da9063_i2c_probe(struct i2c_client *i2c,
 		return ret;
 	}
 
+	/* If SMBus is not available and only I2C is possible, enter I2C mode */
+	if (i2c_check_functionality(i2c->adapter, I2C_FUNC_I2C)) {
+		ret = regmap_clear_bits(da9063->regmap, DA9063_REG_CONFIG_J,
+			  DA9063_TWOWIRE_TO);
+		if (ret < 0) {
+			dev_err(da9063->dev, "Failed to set Two-Wire Bus Mode.\n");
+			return -EIO;
+		}
+	}
+
 	return da9063_device_init(da9063, i2c->irq);
 }
 
include/linux/mfd/da9063/registers.h +3 -0
index 1dbabf1b3cb8..6e0f66a2e727 100644
@@ -1037,6 +1037,9 @@
 #define		DA9063_NONKEY_PIN_AUTODOWN	0x02
 #define		DA9063_NONKEY_PIN_AUTOFLPRT	0x03
 
+/* DA9063_REG_CONFIG_J (addr=0x10F) */
+#define DA9063_TWOWIRE_TO			0x40
+
 /* DA9063_REG_MON_REG_5 (addr=0x116) */
 #define DA9063_MON_A8_IDX_MASK			0x07
 #define		DA9063_MON_A8_IDX_NONE		0x00
Signature · 1 line
2.25.1
Reviewed-by on v5 1/1
#1 lore raw
Quoted 22 lines · Mark Jonas

From: Hubert Streidl <[email protected]>

By default the PMIC DA9063 2-wire interface is SMBus compliant. This means the PMIC will automatically reset the interface when the clock signal ceases for more than the SMBus timeout of 35 ms.

If the I2C driver / device is not capable of creating atomic I2C transactions, a context change can cause a ceasing of the clock signal. This can happen if for example a real-time thread is scheduled. Then the DA9063 in SMBus mode will reset the 2-wire interface. Subsequently a write message could end up in the wrong register. This could cause unpredictable system behavior.

The DA9063 PMIC also supports an I2C compliant mode for the 2-wire interface. This mode does not reset the interface when the clock signal ceases. Thus the problem depicted above does not occur.

This patch tests for the bus functionality "I2C_FUNC_I2C". It can reasonably be assumed that the bus cannot obey SMBus timings if this functionality is set. SMBus commands most probably are emulated in this case which is prone to the latency issue described above.

Quoted 5 lines

This patch enables the I2C bus mode if I2C_FUNC_I2C is set or otherwise keeps the default SMBus mode.

From an I2C point of view, this is the correct approach:

#2 lore raw
In reply to Mark Jonas
Quoted 25 lines · Mark Jonas

From: Hubert Streidl <[email protected]>

By default the PMIC DA9063 2-wire interface is SMBus compliant. This means the PMIC will automatically reset the interface when the clock signal ceases for more than the SMBus timeout of 35 ms.

If the I2C driver / device is not capable of creating atomic I2C transactions, a context change can cause a ceasing of the clock signal. This can happen if for example a real-time thread is scheduled. Then the DA9063 in SMBus mode will reset the 2-wire interface. Subsequently a write message could end up in the wrong register. This could cause unpredictable system behavior.

The DA9063 PMIC also supports an I2C compliant mode for the 2-wire interface. This mode does not reset the interface when the clock signal ceases. Thus the problem depicted above does not occur.

This patch tests for the bus functionality "I2C_FUNC_I2C". It can reasonably be assumed that the bus cannot obey SMBus timings if this functionality is set. SMBus commands most probably are emulated in this case which is prone to the latency issue described above.

This patch enables the I2C bus mode if I2C_FUNC_I2C is set or otherwise keeps the default SMBus mode.

Quoted 6 lines
---
 drivers/mfd/da9063-i2c.c             | 10 ++++++++++
 include/linux/mfd/da9063/registers.h |  3 +++
 2 files changed, 13 insertions(+)

Code looks good to me now, thanks.

However, this doesn't look like it would pass checkpatch.

Have you tried to build with W=1 and checkpatch?

Quoted 35 lines
drivers/mfd/da9063-i2c.c +10 -0
index 3781d0bb7786..e8a022e697c5 100644
@@ -442,6 +442,16 @@ static int da9063_i2c_probe(struct i2c_client *i2c,
 		return ret;
 	}
 
+	/* If SMBus is not available and only I2C is possible, enter I2C mode */
+	if (i2c_check_functionality(i2c->adapter, I2C_FUNC_I2C)) {
+		ret = regmap_clear_bits(da9063->regmap, DA9063_REG_CONFIG_J,
+			  DA9063_TWOWIRE_TO);
+		if (ret < 0) {
+			dev_err(da9063->dev, "Failed to set Two-Wire Bus Mode.\n");
+			return -EIO;
+		}
+	}
+
 	return da9063_device_init(da9063, i2c->irq);
 }
 
include/linux/mfd/da9063/registers.h +3 -0
index 1dbabf1b3cb8..6e0f66a2e727 100644
@@ -1037,6 +1037,9 @@
 #define		DA9063_NONKEY_PIN_AUTODOWN	0x02
 #define		DA9063_NONKEY_PIN_AUTOFLPRT	0x03
 
+/* DA9063_REG_CONFIG_J (addr=0x10F) */
+#define DA9063_TWOWIRE_TO			0x40
+
 /* DA9063_REG_MON_REG_5 (addr=0x116) */
 #define DA9063_MON_A8_IDX_MASK			0x07
 #define		DA9063_MON_A8_IDX_NONE		0x00
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
#3 lore raw

PATCH v5 1/1 mfd: da9063: Support SMBus and I2C mode

Hi Lee,

Quoted 5 lines

Code looks good to me now, thanks.

However, this doesn't look like it would pass checkpatch.

Have you tried to build with W=1 and checkpatch?

Yes, we used checkpatch.pl.

    $ ./scripts/checkpatch.pl 0001-mfd-da9063-Support-SMBus-and-I2C-mode.v5
    total: 0 errors, 0 warnings, 25 lines checked

0001-mfd-da9063-Support-SMBus-and-I2C-mode.v5 has no obvious style problems and is ready for submission.

Using the option --strict we get a check hint that the broken line of the regmap_clear_bits() is not aligned. We tried but were not able to make the tool happy. This matches our experience with this check hint and previous patches.

Also compiling Linux 5.10.14 with our patch and W=1 does not yield a warning.

    $ make W=1
      CALL    scripts/checksyscalls.sh
      CALL    scripts/atomic/check-atomics.sh
      CHK     include/generated/compile.h
      CC [M]  drivers/mfd/da9063-i2c.o
      LD [M]  drivers/mfd/da9063.o
      Kernel: arch/arm/boot/Image is ready
      Kernel: arch/arm/boot/zImage is ready
      MODPOST Module.symvers
      LD [M]  drivers/mfd/da9063.ko
Quoted 31 lines

drivers/mfd/da9063-i2c.c static int da9063_i2c_probe(struct i2c_client *i2c, · line 447

Quoted 11 lines
drivers/mfd/da9063-i2c.c index renamed from drivers/mfd/da9063-i2c.c +0 -0

3781d0bb7786..e8a022e697c5 100644

drivers/mfd/da9063-i2c.c +3 -0
@@ -442,6 +442,16 @@ static int da9063_i2c_probe(struct i2c_client *i2c,
 		return ret;
 	}
 
+	/* If SMBus is not available and only I2C is possible, enter I2C mode */
+	if (i2c_check_functionality(i2c->adapter, I2C_FUNC_I2C)) {
+		ret = regmap_clear_bits(da9063->regmap,

DA9063_REG_CONFIG_J,

drivers/mfd/da9063-i2c.c static int da9063_i2c_probe(struct i2c_client *i2c,

Quoted 3 lines
drivers/mfd/da9063-i2c.c +3 -0
+			  DA9063_TWOWIRE_TO);
+		if (ret < 0) {
+			dev_err(da9063->dev, "Failed to set Two-Wire Bus

Mode.\n");

Quoted 15 lines
drivers/mfd/da9063-i2c.c +4 -0
+			return -EIO;
+		}
+	}
+
 	return da9063_device_init(da9063, i2c->irq);  }
 
include/linux/mfd/da9063/registers.h +0 -0
index 1dbabf1b3cb8..6e0f66a2e727 100644
@@ -1037,6 +1037,9 @@
 #define		DA9063_NONKEY_PIN_AUTODOWN	0x02
 #define		DA9063_NONKEY_PIN_AUTOFLPRT	0x03
 
Quoted 6 lines

include/linux/mfd/da9063/registers.h line 1045

Quoted 6 lines
include/linux/mfd/da9063/registers.h +3 -0
+/* DA9063_REG_CONFIG_J (addr=0x10F) */
+#define DA9063_TWOWIRE_TO			0x40
+
 /* DA9063_REG_MON_REG_5 (addr=0x116) */
 #define DA9063_MON_A8_IDX_MASK			0x07
 #define		DA9063_MON_A8_IDX_NONE		0x00

Cheers,
Mark

#4 lore raw
Quoted 13 lines · On Tue, 16 Mar 2021, Jonas Mark (BT-FIR/ENG1-Grb) wrote:

Hi Lee,

Quoted 5 lines

Code looks good to me now, thanks.

However, this doesn't look like it would pass checkpatch.

Have you tried to build with W=1 and checkpatch?

Yes, we used checkpatch.pl.

    $ ./scripts/checkpatch.pl 0001-mfd-da9063-Support-SMBus-and-I2C-mode.v5
    total: 0 errors, 0 warnings, 25 lines checked
Quoted 6 lines

0001-mfd-da9063-Support-SMBus-and-I2C-mode.v5 has no obvious style problems and is ready for submission.

Using the option --strict we get a check hint that the broken line of the regmap_clear_bits() is not aligned. We tried but were not able to make the tool happy. This matches our experience with this check hint and previous patches.

Lining up the first char of the broken line with the open parenthesis will make checkpatch happy. It's easier if you configure your editor to always do this, rather than with strict tabs.

Quoted 51 lines

Also compiling Linux 5.10.14 with our patch and W=1 does not yield a warning.

    $ make W=1
      CALL    scripts/checksyscalls.sh
      CALL    scripts/atomic/check-atomics.sh
      CHK     include/generated/compile.h
      CC [M]  drivers/mfd/da9063-i2c.o
      LD [M]  drivers/mfd/da9063.o
      Kernel: arch/arm/boot/Image is ready
      Kernel: arch/arm/boot/zImage is ready
      MODPOST Module.symvers
      LD [M]  drivers/mfd/da9063.ko
Quoted 37 lines

drivers/mfd/da9063-i2c.c static int da9063_i2c_probe(struct i2c_client *i2c, · line 447

Quoted 11 lines
drivers/mfd/da9063-i2c.c index renamed from drivers/mfd/da9063-i2c.c +0 -0

3781d0bb7786..e8a022e697c5 100644

drivers/mfd/da9063-i2c.c +3 -0
@@ -442,6 +442,16 @@ static int da9063_i2c_probe(struct i2c_client *i2c,
 		return ret;
 	}
 
+	/* If SMBus is not available and only I2C is possible, enter I2C mode */
+	if (i2c_check_functionality(i2c->adapter, I2C_FUNC_I2C)) {
+		ret = regmap_clear_bits(da9063->regmap,

DA9063_REG_CONFIG_J,

drivers/mfd/da9063-i2c.c static int da9063_i2c_probe(struct i2c_client *i2c,

Quoted 3 lines
drivers/mfd/da9063-i2c.c +3 -0
+			  DA9063_TWOWIRE_TO);
+		if (ret < 0) {
+			dev_err(da9063->dev, "Failed to set Two-Wire Bus

Mode.\n");

Quoted 21 lines
drivers/mfd/da9063-i2c.c +4 -0
+			return -EIO;
+		}
+	}
+
 	return da9063_device_init(da9063, i2c->irq);  }
 
include/linux/mfd/da9063/registers.h +3 -0
index 1dbabf1b3cb8..6e0f66a2e727 100644
@@ -1037,6 +1037,9 @@
 #define		DA9063_NONKEY_PIN_AUTODOWN	0x02
 #define		DA9063_NONKEY_PIN_AUTOFLPRT	0x03
 
+/* DA9063_REG_CONFIG_J (addr=0x10F) */
+#define DA9063_TWOWIRE_TO			0x40
+
 /* DA9063_REG_MON_REG_5 (addr=0x116) */
 #define DA9063_MON_A8_IDX_MASK			0x07
 #define		DA9063_MON_A8_IDX_NONE		0x00
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
#5 lore raw
In reply to Jonas Mark (BT-FIR/ENG1-Grb)
Quoted 10 lines · On Tue, 16 Mar 2021, Jonas Mark (BT-FIR/ENG1-Grb) wrote:

Hi Lee,

Quoted 5 lines

Code looks good to me now, thanks.

However, this doesn't look like it would pass checkpatch.

Have you tried to build with W=1 and checkpatch?

Yes, we used checkpatch.pl.

Quoted 8 lines
    $ ./scripts/checkpatch.pl 0001-mfd-da9063-Support-SMBus-and-I2C-mode.v5
    total: 0 errors, 0 warnings, 25 lines checked

0001-mfd-da9063-Support-SMBus-and-I2C-mode.v5 has no obvious style problems and is ready for submission.

Using the option --strict we get a check hint that the broken line of the regmap_clear_bits() is not aligned. We tried but were not able to make the tool happy. This matches our experience with this check hint and previous patches.

Also compiling Linux 5.10.14 with our patch and W=1 does not yield a warning.

FYI, you should be using -next for upstream development.

Quoted 10 lines
    $ make W=1
      CALL    scripts/checksyscalls.sh
      CALL    scripts/atomic/check-atomics.sh
      CHK     include/generated/compile.h
      CC [M]  drivers/mfd/da9063-i2c.o
      LD [M]  drivers/mfd/da9063.o
      Kernel: arch/arm/boot/Image is ready
      Kernel: arch/arm/boot/zImage is ready
      MODPOST Module.symvers
      LD [M]  drivers/mfd/da9063.ko
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