[PATCH v6 0/3] i2c: imx: Fix handling of arbitration loss

[PATCH v6 0/3] i2c: imx: Fix handling of arbitration loss

[Christian Eggers]

Changes in v5:
---------------
- Added missing "Tested-By" tags.

Changes in v4:
---------------
- Extend comment (W1C vs. W0C)

Changes in v3:
---------------
- dedicated function for clearing an irq

Changes in v2:
---------------
- Don't accidently clear additional status flags on Vybrid
(reported by Uwe Kleine-Koenig)

[PATCH v6 1/3] i2c: imx: Fix reset of I2SR_IAL flag

[Christian Eggers]

According to the "VFxxx Controller Reference Manual" (and the comment
block starting at line 97), Vybrid requires writing a one for clearing
an interrupt flag. Syncing the method for clearing I2SR_IIF in
i2c_imx_isr().

Signed-off-by: Christian Eggers <ceggers@arri.de>
Fixes: 4b775022f6fd ("i2c: imx: add struct to hold more configurable quirks")
Reviewed-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Acked-by: Oleksij Rempel <o.rempel@pengutronix.de>
Cc: stable@vger.kernel.org
---
 drivers/i2c/busses/i2c-imx.c | 22 ++++++++++++++++------
 1 file changed, 16 insertions(+), 6 deletions(-)

diff --git a/drivers/i2c/busses/i2c-imx.c b/drivers/i2c/busses/i2c-imx.c
index 0ab5381aa012..028f8a626410 100644
--- a/drivers/i2c/busses/i2c-imx.c
+++ b/drivers/i2c/busses/i2c-imx.c
@@ -412,6 +412,19 @@ static void i2c_imx_dma_free(struct imx_i2c_struct *i2c_imx)
 	dma->chan_using = NULL;
 }
 
+static void i2c_imx_clear_irq(struct imx_i2c_struct *i2c_imx, unsigned int bits)
+{
+	unsigned int temp;
+
+	/*
+	 * i2sr_clr_opcode is the value to clear all interrupts. Here we want to
+	 * clear only <bits>, so we write ~i2sr_clr_opcode with just <bits>
+	 * toggled. This is required because i.MX needs W0C and Vybrid uses W1C.
+	 */
+	temp = ~i2c_imx->hwdata->i2sr_clr_opcode ^ bits;
+	imx_i2c_write_reg(temp, i2c_imx, IMX_I2C_I2SR);
+}
+
 static int i2c_imx_bus_busy(struct imx_i2c_struct *i2c_imx, int for_busy, bool atomic)
 {
 	unsigned long orig_jiffies = jiffies;
@@ -424,8 +437,7 @@ static int i2c_imx_bus_busy(struct imx_i2c_struct *i2c_imx, int for_busy, bool a
 
 		/* check for arbitration lost */
 		if (temp & I2SR_IAL) {
-			temp &= ~I2SR_IAL;
-			imx_i2c_write_reg(temp, i2c_imx, IMX_I2C_I2SR);
+			i2c_imx_clear_irq(i2c_imx, I2SR_IAL);
 			return -EAGAIN;
 		}
 
@@ -469,7 +481,7 @@ static int i2c_imx_trx_complete(struct imx_i2c_struct *i2c_imx, bool atomic)
 		 */
 		readb_poll_timeout_atomic(addr, regval, regval & I2SR_IIF, 5, 1000 + 100);
 		i2c_imx->i2csr = regval;
-		imx_i2c_write_reg(0, i2c_imx, IMX_I2C_I2SR);
+		i2c_imx_clear_irq(i2c_imx, I2SR_IIF | I2SR_IAL);
 	} else {
 		wait_event_timeout(i2c_imx->queue, i2c_imx->i2csr & I2SR_IIF, HZ / 10);
 	}
@@ -623,9 +635,7 @@ static irqreturn_t i2c_imx_isr(int irq, void *dev_id)
 	if (temp & I2SR_IIF) {
 		/* save status register */
 		i2c_imx->i2csr = temp;
-		temp &= ~I2SR_IIF;
-		temp |= (i2c_imx->hwdata->i2sr_clr_opcode & I2SR_IIF);
-		imx_i2c_write_reg(temp, i2c_imx, IMX_I2C_I2SR);
+		i2c_imx_clear_irq(i2c_imx, I2SR_IIF);
 		wake_up(&i2c_imx->queue);
 		return IRQ_HANDLED;
 	}
-- 
Christian Eggers
Embedded software developer

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: HRA 57918
Persoenlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: HRB 54477
Geschaeftsfuehrer: Dr. Michael Neuhaeuser; Stephan Schenk; Walter Trauninger; Markus Zeiler

[PATCH v6 2/3] i2c: imx: Check for I2SR_IAL after every byte

[Christian Eggers]

Arbitration Lost (IAL) can happen after every single byte transfer. If
arbitration is lost, the I2C hardware will autonomously switch from
master mode to slave. If a transfer is not aborted in this state,
consecutive transfers will not be executed by the hardware and will
timeout.

Signed-off-by: Christian Eggers <ceggers@arri.de>
Tested (not extensively) on Vybrid VF500 (Toradex VF50):
Tested-by: Krzysztof Kozlowski <krzk@kernel.org>
Acked-by: Oleksij Rempel <o.rempel@pengutronix.de>
Cc: stable@vger.kernel.org
---
 drivers/i2c/busses/i2c-imx.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/drivers/i2c/busses/i2c-imx.c b/drivers/i2c/busses/i2c-imx.c
index 028f8a626410..69ce5eea9b5a 100644
--- a/drivers/i2c/busses/i2c-imx.c
+++ b/drivers/i2c/busses/i2c-imx.c
@@ -490,6 +490,16 @@ static int i2c_imx_trx_complete(struct imx_i2c_struct *i2c_imx, bool atomic)
 		dev_dbg(&i2c_imx->adapter.dev, "<%s> Timeout\n", __func__);
 		return -ETIMEDOUT;
 	}
+
+	/* check for arbitration lost */
+	if (i2c_imx->i2csr & I2SR_IAL) {
+		dev_dbg(&i2c_imx->adapter.dev, "<%s> Arbitration lost\n", __func__);
+		i2c_imx_clear_irq(i2c_imx, I2SR_IAL);
+
+		i2c_imx->i2csr = 0;
+		return -EAGAIN;
+	}
+
 	dev_dbg(&i2c_imx->adapter.dev, "<%s> TRX complete\n", __func__);
 	i2c_imx->i2csr = 0;
 	return 0;
-- 
Christian Eggers
Embedded software developer

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: HRA 57918
Persoenlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: HRB 54477
Geschaeftsfuehrer: Dr. Michael Neuhaeuser; Stephan Schenk; Walter Trauninger; Markus Zeiler

[PATCH v6 3/3] i2c: imx: Don't generate STOP condition if arbitration has been lost

[Christian Eggers]

If arbitration is lost, the master automatically changes to slave mode.
I2SR_IBB may or may not be reset by hardware. Raising a STOP condition
by resetting I2CR_MSTA has no effect and will not clear I2SR_IBB.

So calling i2c_imx_bus_busy() is not required and would busy-wait until
timeout.

Signed-off-by: Christian Eggers <ceggers@arri.de>
Tested (not extensively) on Vybrid VF500 (Toradex VF50):
Tested-by: Krzysztof Kozlowski <krzk@kernel.org>
Acked-by: Oleksij Rempel <o.rempel@pengutronix.de>
Cc: stable@vger.kernel.org # Requires trivial backporting, simple remove
                           # the 3rd argument from the calls to
                           # i2c_imx_bus_busy().
---
 drivers/i2c/busses/i2c-imx.c | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

diff --git a/drivers/i2c/busses/i2c-imx.c b/drivers/i2c/busses/i2c-imx.c
index 69ce5eea9b5a..e98356f3db84 100644
--- a/drivers/i2c/busses/i2c-imx.c
+++ b/drivers/i2c/busses/i2c-imx.c
@@ -615,6 +615,8 @@ static void i2c_imx_stop(struct imx_i2c_struct *i2c_imx, bool atomic)
 		/* Stop I2C transaction */
 		dev_dbg(&i2c_imx->adapter.dev, "<%s>\n", __func__);
 		temp = imx_i2c_read_reg(i2c_imx, IMX_I2C_I2CR);
+		if (!(temp & I2CR_MSTA))
+			i2c_imx->stopped = 1;
 		temp &= ~(I2CR_MSTA | I2CR_MTX);
 		if (i2c_imx->dma)
 			temp &= ~I2CR_DMAEN;
@@ -778,9 +780,12 @@ static int i2c_imx_dma_read(struct imx_i2c_struct *i2c_imx,
 		 */
 		dev_dbg(dev, "<%s> clear MSTA\n", __func__);
 		temp = imx_i2c_read_reg(i2c_imx, IMX_I2C_I2CR);
+		if (!(temp & I2CR_MSTA))
+			i2c_imx->stopped = 1;
 		temp &= ~(I2CR_MSTA | I2CR_MTX);
 		imx_i2c_write_reg(temp, i2c_imx, IMX_I2C_I2CR);
-		i2c_imx_bus_busy(i2c_imx, 0, false);
+		if (!i2c_imx->stopped)
+			i2c_imx_bus_busy(i2c_imx, 0, false);
 	} else {
 		/*
 		 * For i2c master receiver repeat restart operation like:
@@ -905,9 +910,12 @@ static int i2c_imx_read(struct imx_i2c_struct *i2c_imx, struct i2c_msg *msgs,
 				dev_dbg(&i2c_imx->adapter.dev,
 					"<%s> clear MSTA\n", __func__);
 				temp = imx_i2c_read_reg(i2c_imx, IMX_I2C_I2CR);
+				if (!(temp & I2CR_MSTA))
+					i2c_imx->stopped =  1;
 				temp &= ~(I2CR_MSTA | I2CR_MTX);
 				imx_i2c_write_reg(temp, i2c_imx, IMX_I2C_I2CR);
-				i2c_imx_bus_busy(i2c_imx, 0, atomic);
+				if (!i2c_imx->stopped)
+					i2c_imx_bus_busy(i2c_imx, 0, atomic);
 			} else {
 				/*
 				 * For i2c master receiver repeat restart operation like:
-- 
Christian Eggers
Embedded software developer

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: HRA 57918
Persoenlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: HRB 54477
Geschaeftsfuehrer: Dr. Michael Neuhaeuser; Stephan Schenk; Walter Trauninger; Markus Zeiler

Re: [PATCH v6 0/3] i2c: imx: Fix handling of arbitration loss

[Wolfram Sang]

[-- Attachment #1: Type: text/plain, Size: 203 bytes --]


> Changes in v5:

Changes in v6 were missing... Because patch 1 was updated, I reverted it
from for-current and will apply this series to for-next instead to give
it some more time for testing.


[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: [PATCH v6 1/3] i2c: imx: Fix reset of I2SR_IAL flag

[Wolfram Sang]

[-- Attachment #1: Type: text/plain, Size: 621 bytes --]

On Fri, Oct 09, 2020 at 01:03:18PM +0200, Christian Eggers wrote:
> According to the "VFxxx Controller Reference Manual" (and the comment
> block starting at line 97), Vybrid requires writing a one for clearing
> an interrupt flag. Syncing the method for clearing I2SR_IIF in
> i2c_imx_isr().
> 
> Signed-off-by: Christian Eggers <ceggers@arri.de>
> Fixes: 4b775022f6fd ("i2c: imx: add struct to hold more configurable quirks")
> Reviewed-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
> Acked-by: Oleksij Rempel <o.rempel@pengutronix.de>
> Cc: stable@vger.kernel.org

Applied to for-next, thanks!


[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: [PATCH v6 2/3] i2c: imx: Check for I2SR_IAL after every byte

[Wolfram Sang]

[-- Attachment #1: Type: text/plain, Size: 651 bytes --]

On Fri, Oct 09, 2020 at 01:03:19PM +0200, Christian Eggers wrote:
> Arbitration Lost (IAL) can happen after every single byte transfer. If
> arbitration is lost, the I2C hardware will autonomously switch from
> master mode to slave. If a transfer is not aborted in this state,
> consecutive transfers will not be executed by the hardware and will
> timeout.
> 
> Signed-off-by: Christian Eggers <ceggers@arri.de>
> Tested (not extensively) on Vybrid VF500 (Toradex VF50):
> Tested-by: Krzysztof Kozlowski <krzk@kernel.org>
> Acked-by: Oleksij Rempel <o.rempel@pengutronix.de>
> Cc: stable@vger.kernel.org

Applied to for-next, thanks!


[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: [PATCH v6 3/3] i2c: imx: Don't generate STOP condition if arbitration has been lost

[Wolfram Sang]

[-- Attachment #1: Type: text/plain, Size: 828 bytes --]

On Fri, Oct 09, 2020 at 01:03:20PM +0200, Christian Eggers wrote:
> If arbitration is lost, the master automatically changes to slave mode.
> I2SR_IBB may or may not be reset by hardware. Raising a STOP condition
> by resetting I2CR_MSTA has no effect and will not clear I2SR_IBB.
> 
> So calling i2c_imx_bus_busy() is not required and would busy-wait until
> timeout.
> 
> Signed-off-by: Christian Eggers <ceggers@arri.de>
> Tested (not extensively) on Vybrid VF500 (Toradex VF50):
> Tested-by: Krzysztof Kozlowski <krzk@kernel.org>
> Acked-by: Oleksij Rempel <o.rempel@pengutronix.de>
> Cc: stable@vger.kernel.org # Requires trivial backporting, simple remove
>                            # the 3rd argument from the calls to
>                            # i2c_imx_bus_busy().

Applied to for-next, thanks!


[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: [PATCH v6 0/3] i2c: imx: Fix handling of arbitration loss

[Christian Eggers]

Hi Wolfram,

On Saturday, 10 October 2020, 13:06:58 CEST, Wolfram Sang wrote:
> > Changes in v5:
> Changes in v6 were missing... Because patch 1 was updated, I reverted it
> from for-current and will apply this series to for-next instead to give
> it some more time for testing.

can may patches be merged into for-current for -RC2 or RC3?

Best regards
Christian

Re: [PATCH v6 1/3] i2c: imx: Fix reset of I2SR_IAL flag

[Christian Eggers]

On Saturday, 10 October 2020, 13:09:20 CET, Wolfram Sang wrote:
> On Fri, Oct 09, 2020 at 01:03:18PM +0200, Christian Eggers wrote:
> > According to the "VFxxx Controller Reference Manual" (and the comment
> > block starting at line 97), Vybrid requires writing a one for clearing
> > an interrupt flag. Syncing the method for clearing I2SR_IIF in
> > i2c_imx_isr().
> > 
> > Signed-off-by: Christian Eggers <ceggers@arri.de>
> > Fixes: 4b775022f6fd ("i2c: imx: add struct to hold more configurable quirks")
> > Reviewed-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
> > Acked-by: Oleksij Rempel <o.rempel@pengutronix.de>
> > Cc: stable@vger.kernel.org
> 
> Applied to for-next, thanks!
> 
I cannot find my patches in kernel/git/wsa/linux.git, branch "for-next".
Did they get lost?

regards
Christian

Re: [PATCH v6 1/3] i2c: imx: Fix reset of I2SR_IAL flag

[Wolfram Sang]

[-- Attachment #1: Type: text/plain, Size: 414 bytes --]


> > Applied to for-next, thanks!
> >
> I cannot find my patches in kernel/git/wsa/linux.git, branch "for-next".
> Did they get lost?

It seems like it :( I seem to have forgotten to push out and then
continued to to work on another computer with an older state. It is
highly embarassing but I can't find out what exactly happened. I hope I
didn't lose more patches. I am really sorry!

Pushed to for-next again.


[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

[PATCH v6 0/3] i2c: imx: Fix handling of arbitration loss

[Christian Eggers]

Changes in v6:
---------------
- Fixed commit (W1C vs. W0C)
- Extended patch 1/3 for atomic case
- Added acks by Oleksij Rempel

Changes in v5:
---------------
- Added missing "Tested-By" tags.

Changes in v4:
---------------
- Extend comment (W1C vs. W0C)

Changes in v3:
---------------
- dedicated function for clearing an irq

Changes in v2:
---------------
- Don't accidently clear additional status flags on Vybrid
(reported by Uwe Kleine-Koenig)