Re: [PATCH] cpufreq: qcom-cpufreq-hw: fix device node refcount leak in phandle parsing

[Viresh Kumar <viresh.kumar@linaro.org>]

On 09-10-26, 01:24, Haotian Zhang wrote:
> of_parse_phandle_with_args() transfers a reference to the provider node
> in out_args->np, which the caller is responsible for releasing with
> of_node_put().  Neither qcom_get_related_cpus() nor
> qcom_cpufreq_hw_cpu_init() ever puts args.np: they only put cpu_np, which
> is a different node obtained from of_cpu_device_node_get().  The
> reference is therefore leaked on every successful call, and
> qcom_get_related_cpus() leaks one per present CPU.
> 
> Release args.np after the parsed index has been read in both call sites.
> 
> Fixes: 2849dd8bc72b ("cpufreq: qcom-hw: Add support for QCOM cpufreq HW driver")
> Assisted-by: DeepSeek-V4.1-Flash
> Signed-off-by: Haotian Zhang <vulab@iscas.ac.cn>
> ---
>  drivers/cpufreq/qcom-cpufreq-hw.c | 3 +++
>  1 file changed, 3 insertions(+)
> 
> diff --git a/drivers/cpufreq/qcom-cpufreq-hw.c b/drivers/cpufreq/qcom-cpufreq-hw.c
> index 874ff3fb9d97..060a460716f7 100644
> --- a/drivers/cpufreq/qcom-cpufreq-hw.c
> +++ b/drivers/cpufreq/qcom-cpufreq-hw.c
> @@ -322,6 +322,8 @@ static void qcom_get_related_cpus(int index, struct cpumask *m)
>  
>  		if (index == args.args[0])
>  			cpumask_set_cpu(cpu, m);
> +
> +		of_node_put(args.np);
>  	}
>  }
>  
> @@ -554,6 +556,7 @@ static int qcom_cpufreq_hw_cpu_init(struct cpufreq_policy *policy)
>  		return ret;
>  
>  	index = args.args[0];
> +	of_node_put(args.np);
>  	data = &qcom_cpufreq.data[index];
>  
>  	/* HW should be in enabled state to proceed */

Applied. Thanks.

-- 
viresh