Re: [PATCH RESEND] i2c: at91: Use dev_err_probe() instead of dev_err()

[Nicolas Ferre]

On 25/08/2023 at 16:32, Yann Sionneau wrote:
> Change
> if (IS_ERR(x)) { dev_err(...); return PTR_ERR(x); }
> into
> return dev_err_probe()
>
> Also, return the correct error instead of hardcoding -ENODEV
> This change has also the advantage of handling the -EPROBE_DEFER situation.

Is it found using a tool like Coccinelle or you just ran into it and 
figured out it could be good to change?

Regards,
  Nicolas

> Signed-off-by: Yann Sionneau <yann@xxxxxxxxxxxx>
> ---
>   drivers/i2c/busses/i2c-at91-core.c | 7 +++----
>   1 file changed, 3 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/i2c/busses/i2c-at91-core.c b/drivers/i2c/busses/i2c-at91-core.c
> index 05ad3bc3578a..b7bc17b0e5f0 100644
> --- a/drivers/i2c/busses/i2c-at91-core.c
> +++ b/drivers/i2c/busses/i2c-at91-core.c
> @@ -227,10 +227,9 @@ static int at91_twi_probe(struct platform_device *pdev)
>          platform_set_drvdata(pdev, dev);
>
>          dev->clk = devm_clk_get(dev->dev, NULL);
> -       if (IS_ERR(dev->clk)) {
> -               dev_err(dev->dev, "no clock defined\n");
> -               return -ENODEV;
> -       }
> +       if (IS_ERR(dev->clk))
> +               return dev_err_probe(dev->dev, PTR_ERR(dev->clk), "no clock defined\n");
> +
>          clk_prepare_enable(dev->clk);
>
>          snprintf(dev->adapter.name, sizeof(dev->adapter.name), "AT91");
> --
> 2.34.1