snooze_loop is done often in any system which has fair bit of idle time. So it qualifies for even micro-optimizations.
When breaking the snooze due to timeout, TIF_POLLING_NRFLAG is cleared twice. Clearing the bit invokes atomics. Avoid double clear and thereby avoid one atomic write. dev->poll_time_limit indicates whether the loop was broken due to timeout. Use that instead of defining a new variable. Signed-off-by: Shrikanth Hegde <[email protected]> --- drivers/cpuidle/cpuidle-powernv.c | 5 ++++- drivers/cpuidle/cpuidle-pseries.c | 5 ++++- 2 files changed, 8 insertions(+), 2 deletions(-) diff --git a/drivers/cpuidle/cpuidle-powernv.c b/drivers/cpuidle/cpuidle-powernv.c index 9ebedd972df0..b89e7111e7b8 100644 --- a/drivers/cpuidle/cpuidle-powernv.c +++ b/drivers/cpuidle/cpuidle-powernv.c @@ -95,7 +95,10 @@ static int snooze_loop(struct cpuidle_device *dev, HMT_medium(); ppc64_runlatch_on(); - clear_thread_flag(TIF_POLLING_NRFLAG); + + /* Avoid double clear when breaking */ + if (!dev->poll_time_limit) + clear_thread_flag(TIF_POLLING_NRFLAG); local_irq_disable(); diff --git a/drivers/cpuidle/cpuidle-pseries.c b/drivers/cpuidle/cpuidle-pseries.c index f68c65f1d023..864dd5d6e627 100644 --- a/drivers/cpuidle/cpuidle-pseries.c +++ b/drivers/cpuidle/cpuidle-pseries.c @@ -64,7 +64,10 @@ int snooze_loop(struct cpuidle_device *dev, struct cpuidle_driver *drv, } HMT_medium(); - clear_thread_flag(TIF_POLLING_NRFLAG); + + /* Avoid double clear when breaking */ + if (!dev->poll_time_limit) + clear_thread_flag(TIF_POLLING_NRFLAG); raw_local_irq_disable(); -- 2.43.0
