[PATCH] iio: accel: mc3230: Leave chip in standby on mount matrix probe failure

[PATCH] iio: accel: mc3230: Leave chip in standby on mount matrix probe failure

[Xiangwu Lai]

mc3230_probe() switches the chip to WAKE mode before reading the
mount matrix from firmware, but when the mount matrix cannot be read
it returns directly and leaves the chip in WAKE mode.  The
iio_device_register() failure path, in contrast, correctly puts the
chip back into STANDBY, so the mount matrix error path is clearly
missing the same cleanup.

The mount matrix is read from firmware only and does not need the
chip at all, so move it before the WAKE transition.  That way the
chip is only woken up once the remaining steps cannot fail anymore.

Fixes: e59c6acfd5fa ("iio: accel: mc3230: add mount matrix support")
Signed-off-by: laixiangwu <laixiangwu@cmss.chinamobile.com>
---
 drivers/iio/accel/mc3230.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/iio/accel/mc3230.c b/drivers/iio/accel/mc3230.c
index b45897210..9ee71da7e 100644
--- a/drivers/iio/accel/mc3230.c
+++ b/drivers/iio/accel/mc3230.c
@@ -183,14 +183,14 @@ static int mc3230_probe(struct i2c_client *client)
 	indio_dev->channels = mc3230_channels;
 	indio_dev->num_channels = ARRAY_SIZE(mc3230_channels);
 
-	ret = mc3230_set_opcon(data, MC3230_MODE_OPCON_WAKE);
-	if (ret < 0)
-		return ret;
-
 	ret = iio_read_mount_matrix(&client->dev, &data->orientation);
 	if (ret)
 		return ret;
 
+	ret = mc3230_set_opcon(data, MC3230_MODE_OPCON_WAKE);
+	if (ret < 0)
+		return ret;
+
 	ret = iio_device_register(indio_dev);
 	if (ret < 0) {
 		dev_err(&client->dev, "device_register failed\n");

Re: [PATCH] iio: accel: mc3230: Leave chip in standby on mount matrix probe failure

[Arnav Kapoor]

On Tue, Nov 10, 2026 at 03:02:06PM +0800, Xiangwu Lai wrote:
> The mount matrix is read from firmware only and does not need the
> chip at all, so move it before the WAKE transition.  That way the
> chip is only woken up once the remaining steps cannot fail anymore.

Reviewed-by: Arnav Kapoor <kapoorarnav43@gmail.com>