[PATCH] cpufreq: Fix implicit type conversion

[PATCH] cpufreq: Fix implicit type conversion

[Jiasheng Jiang]

The variable 'cpu' and 'j' are defined as unsigned int.
However in the for_each_cpu, their values are assigned to -1.
That doesn't make sense and in the cpumask_next() they are
implicitly type conversed to int.
It is universally accepted that the implicit type conversion is
terrible.
Also, having the good programming custom will set an example for
others.
Thus, it might be better to change the definition of 'cpu' and 'j'
from unsigned int to int.

Signed-off-by: Jiasheng Jiang <jiasheng@iscas.ac.cn>
---
 kernel/sched/cpufreq_schedutil.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c
index 4f09afd..4aff4b6 100644
--- a/kernel/sched/cpufreq_schedutil.c
+++ b/kernel/sched/cpufreq_schedutil.c
@@ -409,7 +409,7 @@ static unsigned int sugov_next_freq_shared(struct sugov_cpu *sg_cpu, u64 time)
 	struct sugov_policy *sg_policy = sg_cpu->sg_policy;
 	struct cpufreq_policy *policy = sg_policy->policy;
 	unsigned long util = 0, max = 1;
-	unsigned int j;
+	int j;
 
 	for_each_cpu(j, policy->cpus) {
 		struct sugov_cpu *j_sg_cpu = &per_cpu(sugov_cpu, j);
@@ -746,7 +746,7 @@ static int sugov_start(struct cpufreq_policy *policy)
 {
 	struct sugov_policy *sg_policy = policy->governor_data;
 	void (*uu)(struct update_util_data *data, u64 time, unsigned int flags);
-	unsigned int cpu;
+	int cpu;
 
 	sg_policy->freq_update_delay_ns	= sg_policy->tunables->rate_limit_us * NSEC_PER_USEC;
 	sg_policy->last_freq_update_time	= 0;
@@ -783,7 +783,7 @@ static int sugov_start(struct cpufreq_policy *policy)
 static void sugov_stop(struct cpufreq_policy *policy)
 {
 	struct sugov_policy *sg_policy = policy->governor_data;
-	unsigned int cpu;
+	int cpu;
 
 	for_each_cpu(cpu, policy->cpus)
 		cpufreq_remove_update_util_hook(cpu);
-- 
2.7.4

Re: [PATCH] cpufreq: Fix implicit type conversion

[Rafael J. Wysocki]

On Fri, Oct 29, 2021 at 5:51 AM Jiasheng Jiang <jiasheng@iscas.ac.cn> wrote:
>
> The variable 'cpu' and 'j' are defined as unsigned int.
> However in the for_each_cpu, their values are assigned to -1.
> That doesn't make sense

Yes, it does.

The binary representation of -1 is an all-ones value of the size of
int.  It is perfectly valid to store that value in an unsigned int
variable.

> and in the cpumask_next() they are implicitly type conversed to int.

However, the return type of cpumask_next() is unsigned int.

> It is universally accepted that the implicit type conversion is terrible.

I wouldn't say "terrible", but yes, it is risky when dealing with
variables of different sizes and possible sign-extensions.

In this particular case, I don't see a problem.

> Also, having the good programming custom will set an example for
> others.
> Thus, it might be better to change the definition of 'cpu' and 'j'
> from unsigned int to int.
>
> Signed-off-by: Jiasheng Jiang <jiasheng@iscas.ac.cn>
> ---
>  kernel/sched/cpufreq_schedutil.c | 6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c
> index 4f09afd..4aff4b6 100644
> --- a/kernel/sched/cpufreq_schedutil.c
> +++ b/kernel/sched/cpufreq_schedutil.c
> @@ -409,7 +409,7 @@ static unsigned int sugov_next_freq_shared(struct sugov_cpu *sg_cpu, u64 time)
>         struct sugov_policy *sg_policy = sg_cpu->sg_policy;
>         struct cpufreq_policy *policy = sg_policy->policy;
>         unsigned long util = 0, max = 1;
> -       unsigned int j;
> +       int j;
>
>         for_each_cpu(j, policy->cpus) {
>                 struct sugov_cpu *j_sg_cpu = &per_cpu(sugov_cpu, j);
> @@ -746,7 +746,7 @@ static int sugov_start(struct cpufreq_policy *policy)
>  {
>         struct sugov_policy *sg_policy = policy->governor_data;
>         void (*uu)(struct update_util_data *data, u64 time, unsigned int flags);
> -       unsigned int cpu;
> +       int cpu;
>
>         sg_policy->freq_update_delay_ns = sg_policy->tunables->rate_limit_us * NSEC_PER_USEC;
>         sg_policy->last_freq_update_time        = 0;
> @@ -783,7 +783,7 @@ static int sugov_start(struct cpufreq_policy *policy)
>  static void sugov_stop(struct cpufreq_policy *policy)
>  {
>         struct sugov_policy *sg_policy = policy->governor_data;
> -       unsigned int cpu;
> +       int cpu;
>
>         for_each_cpu(cpu, policy->cpus)
>                 cpufreq_remove_update_util_hook(cpu);
> --
> 2.7.4
>