2017-09-26 07:03:13

by Arvind Yadav

Subject: [PATCH] Input: synaptics-rmi4 - pr_err() strings should end with newlines

pr_err() messages should end with a new-line to avoid other messages
being concatenated.

Signed-off-by: Arvind Yadav <[email protected]>
---
drivers/input/rmi4/rmi_f11.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/input/rmi4/rmi_f11.c b/drivers/input/rmi4/rmi_f11.c
index bc5e37f..28296a9 100644
--- a/drivers/input/rmi4/rmi_f11.c
+++ b/drivers/input/rmi4/rmi_f11.c
@@ -596,7 +596,7 @@ static void rmi_f11_finger_handler(struct f11_data *f11,
/* Possible of having 4 fingers per f_state register */
finger_state = rmi_f11_parse_finger_state(f_state, i);
if (finger_state == F11_RESERVED) {
- pr_err("Invalid finger state[%d]: 0x%02x", i,
+ pr_err("Invalid finger state[%d]: 0x%02x\n", i,
finger_state);
continue;
}
--
1.9.1

2017-09-26 07:15:47

by Benjamin Tissoires

Subject: Re: [PATCH] Input: synaptics-rmi4 - pr_err() strings should end with newlines

Hi Arvind,

On Sep 26 2017 or thereabouts, Arvind Yadav wrote:
> pr_err() messages should end with a new-line to avoid other messages
> being concatenated.
>
> Signed-off-by: Arvind Yadav <[email protected]>
> ---
> drivers/input/rmi4/rmi_f11.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/input/rmi4/rmi_f11.c b/drivers/input/rmi4/rmi_f11.c
> index bc5e37f..28296a9 100644
> --- a/drivers/input/rmi4/rmi_f11.c
> +++ b/drivers/input/rmi4/rmi_f11.c
> @@ -596,7 +596,7 @@ static void rmi_f11_finger_handler(struct f11_data *f11,
> /* Possible of having 4 fingers per f_state register */
> finger_state = rmi_f11_parse_finger_state(f_state, i);
> if (finger_state == F11_RESERVED) {
> - pr_err("Invalid finger state[%d]: 0x%02x", i,
> + pr_err("Invalid finger state[%d]: 0x%02x\n", i,

I am pretty sure not having the '\n' is the correct behavior.

Linus recently complained[1] about people trying to "fix" the printk
behavior, and it resulted that any printk will be shown on its own line
(unless there is an other printk with a KERN_CONT marker, which is
unlikely to happen). So basically such patch doesn't add much.

I am also pretty sure there has already been a thread[2 and others]
about these '\n' and from what I can remember, the end result was that
not having the line termination was the preferred syntax.

Cheers,
Benjamin

[1] https://lkml.org/lkml/2017/8/29/504
[2] https://lkml.org/lkml/2010/11/23/394

> finger_state);
> continue;
> }
> --
> 1.9.1
>