cpuidle: governors: teo: Avoid fake intercepts produced by tick

Tick wakeups can lead to fake intercepts that may skew idle state
selection towards shallow states, so it is better to avoid counting
them as intercepts.

For this purpose, add a check causing teo_update() to only count
tick wakeups as intercepts if intercepts within the tick period
range are at least twice as frequent as any other events.

Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: Christian Loehle <christian.loehle@arm.com>
Link: https://patch.msgid.link/3404606.44csPzL39Z@rafael.j.wysocki

diff --git a/drivers/cpuidle/governors/teo.c b/drivers/cpuidle/governors/teo.c
index 9820ef36a664969f25f5077bf3008465af67eae4..5434584af040d12a91ab62eb3cb5d7f90de34b3d 100644 (file)
--- a/drivers/cpuidle/governors/teo.c
+++ b/drivers/cpuidle/governors/teo.c
@@ -239,6 +239,17 @@ static void teo_update(struct cpuidle_driver *drv, struct cpuidle_device *dev)
                        cpu_data->state_bins[drv->state_count-1].hits += PULSE;
                        return;
                }
+               /*
+                * If intercepts within the tick period range are not frequent
+                * enough, count this wakeup as a hit, since it is likely that
+                * the tick has woken up the CPU because an expected intercept
+                * was not there.  Otherwise, one of the intercepts may have
+                * been incidentally preceded by the tick wakeup.
+                */
+               if (3 * cpu_data->tick_intercepts < 2 * total) {
+                       cpu_data->state_bins[idx_timer].hits += PULSE;
+                       return;
+               }
        }
 
        /*