Re: [PATCH 1/3] cpufreq: Allow passing NULL as the argument for unregistering a driver

[Huang Rui <ray.huang@amd.com>]

On Fri, Mar 25, 2022 at 01:42:26PM +0800, Limonciello, Mario wrote:
> Currently the arguments passed to `cpufreq_unregister_driver` are matched
> against the currently registered driver.
> 
> This means that the only way for a driver to be unregistered is if it's
> module is unloaded.  Loosen that restriction to allow other kernel modules
> to unregister a registered driver.
> 
> Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
> ---
>  drivers/cpufreq/cpufreq.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
> index 80f535cc8a75..4711c17a89bb 100644
> --- a/drivers/cpufreq/cpufreq.c
> +++ b/drivers/cpufreq/cpufreq.c
> @@ -2885,10 +2885,10 @@ int cpufreq_unregister_driver(struct cpufreq_driver *driver)
>  {
>  	unsigned long flags;
>  
> -	if (!cpufreq_driver || (driver != cpufreq_driver))
> +	if (!cpufreq_driver)
>  		return -EINVAL;
>  
> -	pr_debug("unregistering driver %s\n", driver->name);
> +	pr_debug("unregistering driver %s\n", cpufreq_driver->name);
>  

This is amd-pstate only use case, I suggest we only modify the amd-pstate
driver, and don't impact generic cpufreq. Actually, only acpi-cpufreq could
be registered earlier than amd-pstate. So how about exposing the
acpi_cpufreq_exit(), and call this to free the acpi_cpufreq_driver in
amd-pstate directly?

Thanks,
Ray