2024-04-15 09:50:53

by Dawei Li

Subject: [PATCH] powercap: Avoid explicit cpumask allocation on stack

In general it's preferable to avoid placing cpumasks on the stack, as
for large values of NR_CPUS these can consume significant amounts of
stack space and make stack overflows more likely.

Use cpumask_weight_and() to avoid the need for a temporary cpumask on
the stack.

Signed-off-by: Dawei Li <[email protected]>
---
drivers/powercap/dtpm_cpu.c | 8 ++------
1 file changed, 2 insertions(+), 6 deletions(-)

diff --git a/drivers/powercap/dtpm_cpu.c b/drivers/powercap/dtpm_cpu.c
index bc90126f1b5f..6b6f51b21550 100644
--- a/drivers/powercap/dtpm_cpu.c
+++ b/drivers/powercap/dtpm_cpu.c
@@ -43,13 +43,11 @@ static u64 set_pd_power_limit(struct dtpm *dtpm, u64 power_limit)
struct dtpm_cpu *dtpm_cpu = to_dtpm_cpu(dtpm);
struct em_perf_domain *pd = em_cpu_get(dtpm_cpu->cpu);
struct em_perf_state *table;
- struct cpumask cpus;
unsigned long freq;
u64 power;
int i, nr_cpus;

- cpumask_and(&cpus, cpu_online_mask, to_cpumask(pd->cpus));
- nr_cpus = cpumask_weight(&cpus);
+ nr_cpus = cpumask_weight_and(cpu_online_mask, to_cpumask(pd->cpus));

rcu_read_lock();
table = em_perf_state_from_pd(pd);
@@ -123,11 +121,9 @@ static int update_pd_power_uw(struct dtpm *dtpm)
struct dtpm_cpu *dtpm_cpu = to_dtpm_cpu(dtpm);
struct em_perf_domain *em = em_cpu_get(dtpm_cpu->cpu);
struct em_perf_state *table;
- struct cpumask cpus;
int nr_cpus;

- cpumask_and(&cpus, cpu_online_mask, to_cpumask(em->cpus));
- nr_cpus = cpumask_weight(&cpus);
+ nr_cpus = cpumask_weight_and(cpu_online_mask, to_cpumask(em->cpus));

rcu_read_lock();
table = em_perf_state_from_pd(em);
--
2.27.0

2024-04-16 11:34:51

by Rafael J. Wysocki

Subject: Re: [PATCH] powercap: Avoid explicit cpumask allocation on stack

On Mon, Apr 15, 2024 at 11:48 AM Dawei Li <[email protected]> wrote:
>
> In general it's preferable to avoid placing cpumasks on the stack, as
> for large values of NR_CPUS these can consume significant amounts of
> stack space and make stack overflows more likely.
>
> Use cpumask_weight_and() to avoid the need for a temporary cpumask on
> the stack.
>
> Signed-off-by: Dawei Li <[email protected]>
> ---
> drivers/powercap/dtpm_cpu.c | 8 ++------
> 1 file changed, 2 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/powercap/dtpm_cpu.c b/drivers/powercap/dtpm_cpu.c
> index bc90126f1b5f..6b6f51b21550 100644
> --- a/drivers/powercap/dtpm_cpu.c
> +++ b/drivers/powercap/dtpm_cpu.c
> @@ -43,13 +43,11 @@ static u64 set_pd_power_limit(struct dtpm *dtpm, u64 power_limit)
> struct dtpm_cpu *dtpm_cpu = to_dtpm_cpu(dtpm);
> struct em_perf_domain *pd = em_cpu_get(dtpm_cpu->cpu);
> struct em_perf_state *table;
> - struct cpumask cpus;
> unsigned long freq;
> u64 power;
> int i, nr_cpus;
>
> - cpumask_and(&cpus, cpu_online_mask, to_cpumask(pd->cpus));
> - nr_cpus = cpumask_weight(&cpus);
> + nr_cpus = cpumask_weight_and(cpu_online_mask, to_cpumask(pd->cpus));
>
> rcu_read_lock();
> table = em_perf_state_from_pd(pd);
> @@ -123,11 +121,9 @@ static int update_pd_power_uw(struct dtpm *dtpm)
> struct dtpm_cpu *dtpm_cpu = to_dtpm_cpu(dtpm);
> struct em_perf_domain *em = em_cpu_get(dtpm_cpu->cpu);
> struct em_perf_state *table;
> - struct cpumask cpus;
> int nr_cpus;
>
> - cpumask_and(&cpus, cpu_online_mask, to_cpumask(em->cpus));
> - nr_cpus = cpumask_weight(&cpus);
> + nr_cpus = cpumask_weight_and(cpu_online_mask, to_cpumask(em->cpus));
>
> rcu_read_lock();
> table = em_perf_state_from_pd(em);
> --

Applied as 6.10 material, thanks!