Quoting Alexander Stein (2023-04-25 08:33:46)
> Add the error code of a failed i2c transfer to the error message.
>
> Signed-off-by: Alexander Stein <alexander.stein@xxxxxxxxxxxxxxx>
> ---
> drivers/media/i2c/tc358743.c | 8 ++++----
> 1 file changed, 4 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/media/i2c/tc358743.c b/drivers/media/i2c/tc358743.c
> index 9197fa0b1bc2..ad6a72b2bcf5 100644
> --- a/drivers/media/i2c/tc358743.c
> +++ b/drivers/media/i2c/tc358743.c
> @@ -133,8 +133,8 @@ static void i2c_rd(struct v4l2_subdev *sd, u16 reg, u8 *values, u32 n)
>
> err = i2c_transfer(client->adapter, msgs, ARRAY_SIZE(msgs));
> if (err != ARRAY_SIZE(msgs)) {
> - v4l2_err(sd, "%s: reading register 0x%x from 0x%x failed\n",
> - __func__, reg, client->addr);
> + v4l2_err(sd, "%s: reading register 0x%x from 0x%x failed: %d\n",
Hrm. I was going to ask if %pe [0] would be relevant to print the error, but
these aren't quite PTR_ERR's ... just errors... Though it looks from
lib/vscnprintf.c :: err_ptr() it would still work!
[0] https://www.kernel.org/doc/html/latest/core-api/printk-formats.html#error-pointers
So either way,
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@xxxxxxxxxxxxxxxx>
> + __func__, reg, client->addr, err);
> }
> }
>
> @@ -165,8 +165,8 @@ static void i2c_wr(struct v4l2_subdev *sd, u16 reg, u8 *values, u32 n)
>
> err = i2c_transfer(client->adapter, &msg, 1);
> if (err != 1) {
> - v4l2_err(sd, "%s: writing register 0x%x from 0x%x failed\n",
> - __func__, reg, client->addr);
> + v4l2_err(sd, "%s: writing register 0x%x from 0x%x failed: %d\n",
> + __func__, reg, client->addr, err);
> return;
> }
>
> --
> 2.34.1
>
