Re: [PATCH 2/2] watchdog: imx7ulp_wdt: needn't wait 2.5 clocks after RCS is done for iMX93

[Guenter Roeck]

On 7/11/24 15:41, Frank Li wrote:
> From: Alice Guo <alice.guo@xxxxxxx>
>
> i.MX93 watchdog needn't wait 2.5 clocks after RCS is done. So set
> post_rcs_wait to false for "fsl,imx93-wdt".
>
> Signed-off-by: Alice Guo <alice.guo@xxxxxxx>
> Reviewed-by: Ye Li <ye.li@xxxxxxx>
> Signed-off-by: Frank Li <Frank.Li@xxxxxxx>
> ---
>   drivers/watchdog/imx7ulp_wdt.c | 1 -
>   1 file changed, 1 deletion(-)
>
> diff --git a/drivers/watchdog/imx7ulp_wdt.c b/drivers/watchdog/imx7ulp_wdt.c
> index 904b9f1873856..3a75a6f98f8f0 100644
> --- a/drivers/watchdog/imx7ulp_wdt.c
> +++ b/drivers/watchdog/imx7ulp_wdt.c
> @@ -405,7 +405,6 @@ static const struct imx_wdt_hw_feature imx8ulp_wdt_hw = {
>   static const struct imx_wdt_hw_feature imx93_wdt_hw = {
>   	.prescaler_enable = true,
>   	.wdog_clock_rate = 125,
> -	.post_rcs_wait = true,
>   };
>   
>   static const struct of_device_id imx7ulp_wdt_dt_ids[] = {
Introducing that flag in the previous patch just to remove it here doesn't
make sense to me, sorry.

What the two changes do together is to disable post_rcs_wait for iMX93.
That is a single logical change, and it can and should be done in a
single patch. If you do that by moving the flag into imx_wdt_hw_feature
or by adding another of_device_is_compatible() is your call.

Guenter