[U-Boot] [PATCH 3/5] i2c: imx: Optimize the i2c device recovery solution

[Peng Fan]

From: Fugang Duan <b38611 at freescale.com>

>From i2c spec, if device pull down the SDA line that causes
i2c bus dead, host can send out 9 clock to let device release
SDA.

But for some special device like pfuze100, it pull down SDA line
and the solution cannot take effort.

The patch just add NACK and STOP signal after 8 dummy clock, and pmic
can release SDA line after the recovery. Test case catch 375 times of
i2c hang, and all are recovered.

Signed-off-by: Peng Fan <Peng.Fan at freescale.com>
Signed-off-by: Fugang Duan <B38611 at freescale.com>
---
 arch/arm/imx-common/i2c-mxv7.c | 27 +++++++++++++++++++++++++--
 1 file changed, 25 insertions(+), 2 deletions(-)

diff --git a/arch/arm/imx-common/i2c-mxv7.c b/arch/arm/imx-common/i2c-mxv7.c
index 1a632e7..b1d4ec4 100644
--- a/arch/arm/imx-common/i2c-mxv7.c
+++ b/arch/arm/imx-common/i2c-mxv7.c
@@ -33,13 +33,36 @@ static int force_idle_bus(void *priv)
 
 	printf("%s: sda=%d scl=%d sda.gp=0x%x scl.gp=0x%x\n", __func__,
 		sda, scl, p->sda.gp, p->scl.gp);
+	gpio_direction_output(p->scl.gp, 1);
+	udelay(1000);
 	/* Send high and low on the SCL line */
 	for (i = 0; i < 9; i++) {
-		gpio_direction_output(p->scl.gp, 0);
+		gpio_direction_output(p->scl.gp, 1);
 		udelay(50);
-		gpio_direction_input(p->scl.gp);
+		gpio_direction_output(p->scl.gp, 0);
 		udelay(50);
 	}
+
+	/* Simulate the NACK */
+	gpio_direction_output(p->sda.gp, 1);
+	udelay(50);
+	gpio_direction_output(p->scl.gp, 1);
+	udelay(50);
+	gpio_direction_output(p->scl.gp, 0);
+	udelay(50);
+
+	/* Simulate the STOP signal */
+	gpio_direction_output(p->sda.gp, 0);
+	udelay(50);
+	gpio_direction_output(p->scl.gp, 1);
+	udelay(50);
+	gpio_direction_output(p->sda.gp, 1);
+	udelay(50);
+
+	/* Get the bus status */
+	gpio_direction_input(p->sda.gp);
+	gpio_direction_input(p->scl.gp);
+
 	start_time = get_timer(0);
 	for (;;) {
 		sda = gpio_get_value(p->sda.gp);
-- 
1.8.4