From: "Rafael J. Wysocki" <rafael.j.wysocki@intel.com>
Date: Wed, 13 Nov 2019 01:03:24 +0100
Subject: cpuidle: teo: Avoid using "early hits" incorrectly
Git-commit: 63f202e5edf161c2ccffa286a9a701e995427b15
Patch-mainline: 5.5-rc1
References: git-fixes

If the current state with the maximum "early hits" metric in
teo_select() is also the one "matching" the expected idle duration,
it will be used as the candidate one for selection even if its
"misses" metric is greater than its "hits" metric, which is not
correct.

In that case, the candidate state should be shallower than the
current one and its "early hits" metric should be the maximum
among the idle states shallower than the current one.

To make that happen, modify teo_select() to save the index of
the state whose "early hits" metric is the maximum for the
range of states below the current one and go back to that state
if it turns out that the current one should be rejected.

Fixes: 159e48560f51 ("cpuidle: teo: Fix "early hits" handling for disabled idle states")
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Jiri Slaby <jslaby@suse.cz>
---
 drivers/cpuidle/governors/teo.c |   21 +++++++++++++++++----
 1 file changed, 17 insertions(+), 4 deletions(-)

--- a/drivers/cpuidle/governors/teo.c
+++ b/drivers/cpuidle/governors/teo.c
@@ -242,7 +242,7 @@
 	struct teo_cpu *cpu_data = per_cpu_ptr(&teo_cpus, dev->cpu);
 	int latency_req = cpuidle_governor_latency_req(dev->cpu);
 	unsigned int duration_us, hits, misses, early_hits;
-	int max_early_idx, constraint_idx, idx, i;
+	int max_early_idx, prev_max_early_idx, constraint_idx, idx, i;
 	ktime_t delta_tick;
 
 	if (dev->last_state_idx >= 0) {
@@ -259,6 +259,7 @@
 	misses = 0;
 	early_hits = 0;
 	max_early_idx = -1;
+	prev_max_early_idx = -1;
 	constraint_idx = drv->state_count;
 	idx = -1;
 
@@ -311,6 +312,7 @@
 			 */
 			if (!(tick_nohz_tick_stopped() &&
 			      drv->states[idx].target_residency < TICK_USEC)) {
+				prev_max_early_idx = max_early_idx;
 				early_hits = cpu_data->states[i].early_hits;
 				max_early_idx = idx;
 			}
@@ -337,6 +339,7 @@
 		if (early_hits < cpu_data->states[i].early_hits &&
 		    !(tick_nohz_tick_stopped() &&
 		      drv->states[i].target_residency < TICK_USEC)) {
+			prev_max_early_idx = max_early_idx;
 			early_hits = cpu_data->states[i].early_hits;
 			max_early_idx = i;
 		}
@@ -350,9 +353,19 @@
 	 * "early hits" metric, but if that cannot be determined, just use the
 	 * state selected so far.
 	 */
-	if (hits <= misses && max_early_idx >= 0) {
-		idx = max_early_idx;
-		duration_us = drv->states[idx].target_residency;
+	if (hits <= misses) {
+		/*
+		 * The current candidate state is not suitable, so take the one
+		 * whose "early hits" metric is the maximum for the range of
+		 * shallower states.
+		 */
+		if (idx == max_early_idx)
+			max_early_idx = prev_max_early_idx;
+
+		if (max_early_idx >= 0) {
+			idx = max_early_idx;
+			duration_us = drv->states[idx].target_residency;
+		}
 	}
 
 	/*