Re: [PATCH 7/7] cpufreq: Exit governor when failed to start old governor

[zhenglifeng (A)]

On 2025/6/23 23:12, Rafael J. Wysocki wrote:

> On Mon, Jun 23, 2025 at 3:34 PM Lifeng Zheng <zhenglifeng1@xxxxxxxxxx> wrote:
>>
>> Detect the result of starting old governor in cpufreq_set_policy(). If it
>> fails, exit the governor and clear policy->governor.
>>
>> Signed-off-by: Lifeng Zheng <zhenglifeng1@xxxxxxxxxx>
>> ---
>>  drivers/cpufreq/cpufreq.c | 8 +++++---
>>  1 file changed, 5 insertions(+), 3 deletions(-)
>>
>> diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
>> index 7b82ffb50283..2b431845a8a3 100644
>> --- a/drivers/cpufreq/cpufreq.c
>> +++ b/drivers/cpufreq/cpufreq.c
>> @@ -2715,10 +2715,12 @@ static int cpufreq_set_policy(struct cpufreq_policy *policy,
>>         pr_debug("starting governor %s failed\n", policy->governor->name);
>>         if (old_gov) {
>>                 policy->governor = old_gov;
>> -               if (cpufreq_init_governor(policy))
>> +               if (cpufreq_init_governor(policy)) {
>>                         policy->governor = NULL;
>> -               else
>> -                       cpufreq_start_governor(policy);
>> +               } else if (cpufreq_start_governor(policy)) {
>> +                       cpufreq_exit_governor(policy);
> 
> This may introduce a governor module reference imbalance AFAICS.

Sorry, I don't really understand this. Could you explain more? Thanks.

> 
>> +                       policy->governor = NULL;
>> +               }
>>         }
>>
>>         return ret;
>> --
>