Re: [PATCH v9 12/12] cpuidle/poll_state: Wait for need-resched via tif_need_resched_relaxed_wait()

[Ankur Arora]

Hi Peter,

Could you look tif_need_resched_relaxed_wait() and see if the interface
(and the implementation via smp_cond_load_relaxed_timeout() -- patch-11
in this series, with smp_cond_load_relaxed_timeout() itself being the
patch-1 in the series) seems okay to you.

And just a little bit of history on the interface. You had suggested using
smp_cond_load_relaxed() way back
(https://lore.kernel.org/lkml/20230809134837.GM212435@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx/)
when we were trying to use poll_idle() on arm64. That eventually became
smp_cond_load_relaxed_timeout() and then Rafael suggested this interface
abstracting out all the non idle related details.

Rafael J. Wysocki <rafael@xxxxxxxxxx> writes:

> On Mon, Feb 9, 2026 at 3:43 AM Ankur Arora <ankur.a.arora@xxxxxxxxxx> wrote:
>>
>> The inner loop in poll_idle() polls over the thread_info flags,
>> waiting to see if the thread has TIF_NEED_RESCHED set. The loop
>> exits once the condition is met, or if the poll time limit has
>> been exceeded.
>>
>> To minimize the number of instructions executed in each iteration,
>> the time check is rate-limited. In addition, each loop iteration
>> executes cpu_relax() which on certain platforms provides a hint to
>> the pipeline that the loop busy-waits, allowing the processor to
>> reduce power consumption.
>>
>> Switch over to tif_need_resched_relaxed_wait() instead, since that
>> provides exactly that.
>>
>> However, since we want to minimize power consumption in idle, building
>> of cpuidle/poll_state.c continues to depend on CONFIG_ARCH_HAS_CPU_RELAX
>> as that serves as an indicator that the platform supports an optimized
>> version of tif_need_resched_relaxed_wait() (via
>> smp_cond_load_acquire_timeout()).
>>
>> Cc: "Rafael J. Wysocki" <rafael@xxxxxxxxxx>
>> Cc: Daniel Lezcano <daniel.lezcano@xxxxxxxxxx>
>> Cc: linux-pm@xxxxxxxxxxxxxxx
>> Suggested-by: "Rafael J. Wysocki" <rafael@xxxxxxxxxx>
>> Signed-off-by: Ankur Arora <ankur.a.arora@xxxxxxxxxx>
>
> This is generally fine with me, of course depending on how
> tif_need_resched_relaxed_wait() will work, so conditional on reaching
> an agreement with the arch and scheduler folks feel free to add
>
> Acked-by: Rafael J. Wysocki (Intel) <rafael@xxxxxxxxxx>

Excellent. Thanks Rafael.


Ankur

> to this one.
>
>> ---
>> Notes:
>>   - get rid of unnecessary variable assignments, casts etc
>>
>>  drivers/cpuidle/poll_state.c | 21 +--------------------
>>  1 file changed, 1 insertion(+), 20 deletions(-)
>>
>> diff --git a/drivers/cpuidle/poll_state.c b/drivers/cpuidle/poll_state.c
>> index c7524e4c522a..7443b3e971ba 100644
>> --- a/drivers/cpuidle/poll_state.c
>> +++ b/drivers/cpuidle/poll_state.c
>> @@ -6,41 +6,22 @@
>>  #include <linux/cpuidle.h>
>>  #include <linux/export.h>
>>  #include <linux/irqflags.h>
>> -#include <linux/sched.h>
>> -#include <linux/sched/clock.h>
>>  #include <linux/sched/idle.h>
>>  #include <linux/sprintf.h>
>>  #include <linux/types.h>
>>
>> -#define POLL_IDLE_RELAX_COUNT  200
>> -
>>  static int __cpuidle poll_idle(struct cpuidle_device *dev,
>>                                struct cpuidle_driver *drv, int index)
>>  {
>> -       u64 time_start;
>> -
>> -       time_start = local_clock_noinstr();
>> -
>>         dev->poll_time_limit = false;
>>
>>         raw_local_irq_enable();
>>         if (!current_set_polling_and_test()) {
>> -               unsigned int loop_count = 0;
>>                 u64 limit;
>>
>>                 limit = cpuidle_poll_time(drv, dev);
>>
>> -               while (!need_resched()) {
>> -                       cpu_relax();
>> -                       if (loop_count++ < POLL_IDLE_RELAX_COUNT)
>> -                               continue;
>> -
>> -                       loop_count = 0;
>> -                       if (local_clock_noinstr() - time_start > limit) {
>> -                               dev->poll_time_limit = true;
>> -                               break;
>> -                       }
>> -               }
>> +               dev->poll_time_limit = !tif_need_resched_relaxed_wait(limit);
>>         }
>>         raw_local_irq_disable();
>>
>> --


--
ankur