Re: [PATCH] thermal: of: fix cooling_spec.args for 64bit

[Punit Agrawal <punit.agrawal@arm.com>]

Wei Ni <wni@nvidia.com> writes:

> The THERMAL_NO_LIMIT is defined as -1UL, when it is set in DT,
> the of-thermal will use of_parse_phandle_with_args() to parse it.
> Since the args is uint32_t, so in 32bit system, the of framework
> will read it as 0xffffffff, it's the expected value "-1".
> But in 64bit system, this value is read as 0x00000000ffffffff,
> it's not a negative value, it will cause problems.
>
> This change can fix this issue.

If I understand the issue correctly, a patch fixing this is already
merged. Does "commit a940cb34f: thermal: Fix cdev registration with
THERMAL_NO_LIMIT on 64bit" not work for you?

>
> Signed-off-by: Wei Ni <wni@nvidia.com>
> ---
>  drivers/thermal/of-thermal.c | 6 ++++--
>  1 file changed, 4 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/thermal/of-thermal.c b/drivers/thermal/of-thermal.c
> index 9e8c614103ef..8d6b4070dea3 100644
> --- a/drivers/thermal/of-thermal.c
> +++ b/drivers/thermal/of-thermal.c
> @@ -613,8 +613,10 @@ static int thermal_of_populate_bind_params(struct device_node *np,
>  	}
>  	__tbp->cooling_device = cooling_spec.np;
>  	if (cooling_spec.args_count >= 2) { /* at least min and max */
> -		__tbp->min = cooling_spec.args[0];
> -		__tbp->max = cooling_spec.args[1];
> +		__tbp->min = cooling_spec.args[0] == -1U ?
> +			THERMAL_NO_LIMIT : cooling_spec.args[0];
> +		__tbp->max = cooling_spec.args[1] == -1U ?
> +			THERMAL_NO_LIMIT : cooling_spec.args[1];
>  	} else {
>  		pr_err("wrong reference to cooling device, missing limits\n");
>  	}