diff mbox series

[v2,1/3] i2c: imx: Fix reset of I2SR_IAL flag

Message ID 20201002152305.4963-2-ceggers@arri.de
State New
Headers show
Series i2c: imx: Fix handling of arbitration loss | expand

Commit Message

Christian Eggers Oct. 2, 2020, 3:23 p.m. UTC
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>
Cc: stable@vger.kernel.org
---
 drivers/i2c/busses/i2c-imx.c | 15 ++++++++++++---
 1 file changed, 12 insertions(+), 3 deletions(-)

Comments

Krzysztof Kozlowski Oct. 5, 2020, 8:05 a.m. UTC | #1
On Fri, Oct 02, 2020 at 05:23:03PM +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>
> Cc: stable@vger.kernel.org
> ---
>  drivers/i2c/busses/i2c-imx.c | 15 ++++++++++++---
>  1 file changed, 12 insertions(+), 3 deletions(-)


Tested (not extensively) on Vybrid VF500 (Toradex VF50):
Tested-by: Krzysztof Kozlowski <krzk@kernel.org>

The I2C on Vybrid VF500 still works fine. I did not test this actual
condition (arbitration) but only a regular I2C driver (BQ27xxx fuel
gauge).

Best regards,
Krzysztof
Uwe Kleine-König Oct. 6, 2020, 6:05 a.m. UTC | #2
On Fri, Oct 02, 2020 at 05:23:03PM +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>
> Cc: stable@vger.kernel.org
> ---
>  drivers/i2c/busses/i2c-imx.c | 15 ++++++++++++---
>  1 file changed, 12 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/i2c/busses/i2c-imx.c b/drivers/i2c/busses/i2c-imx.c
> index 0ab5381aa012..34648df7f1a6 100644
> --- a/drivers/i2c/busses/i2c-imx.c
> +++ b/drivers/i2c/busses/i2c-imx.c
> @@ -424,7 +424,12 @@ 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;
> +			/*
> +			 * i2sr_clr_opcode is the value to clear all interrupts.
> +			 * Here we want to clear only I2SR_IAL, so we write
> +			 * ~i2sr_clr_opcode with just the I2SR_IAL bit toggled.
> +			 */
> +			temp = ~i2c_imx->hwdata->i2sr_clr_opcode ^ I2SR_IAL;
>  			imx_i2c_write_reg(temp, i2c_imx, IMX_I2C_I2SR);
>  			return -EAGAIN;

Could we please move clearing an irq to a dedicated function? Such that
it looks like:

	/* check for arbitration lost */
	if (temp & I2SR_IAL) {
		i2c_imx_clear_irq(i2c_imx, I2SR_IAL);
		return -EAGAIN;
	}

Then you also don't need to duplicate the describing comment but just
add it to the implementation of i2c_imx_clear_irq().

Best regards
Uwe
diff mbox series

Patch

diff --git a/drivers/i2c/busses/i2c-imx.c b/drivers/i2c/busses/i2c-imx.c
index 0ab5381aa012..34648df7f1a6 100644
--- a/drivers/i2c/busses/i2c-imx.c
+++ b/drivers/i2c/busses/i2c-imx.c
@@ -424,7 +424,12 @@  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;
+			/*
+			 * i2sr_clr_opcode is the value to clear all interrupts.
+			 * Here we want to clear only I2SR_IAL, so we write
+			 * ~i2sr_clr_opcode with just the I2SR_IAL bit toggled.
+			 */
+			temp = ~i2c_imx->hwdata->i2sr_clr_opcode ^ I2SR_IAL;
 			imx_i2c_write_reg(temp, i2c_imx, IMX_I2C_I2SR);
 			return -EAGAIN;
 		}
@@ -623,8 +628,12 @@  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);
+		/*
+		 * i2sr_clr_opcode is the value to clear all interrupts.
+		 * Here we want to clear only I2SR_IIF, so we write
+		 * ~i2sr_clr_opcode with just the I2SR_IIF bit toggled.
+		 */
+		temp = ~i2c_imx->hwdata->i2sr_clr_opcode ^ I2SR_IIF;
 		imx_i2c_write_reg(temp, i2c_imx, IMX_I2C_I2SR);
 		wake_up(&i2c_imx->queue);
 		return IRQ_HANDLED;