* [PATCH v1 0/5] cpuidle: governors: teo: Wakeup events classification change and some refinements
@ 2026-01-14 19:42 Rafael J. Wysocki
2026-01-14 19:44 ` [PATCH v1 1/5] cpuidle: governors: teo: Avoid selecting states with zero-size bins Rafael J. Wysocki
` (6 more replies)
0 siblings, 7 replies; 21+ messages in thread
From: Rafael J. Wysocki @ 2026-01-14 19:42 UTC (permalink / raw)
To: Linux PM; +Cc: LKML, Christian Loehle, Doug Smythies
Hi All,
This material has been in my local queue for almost a full development cycle,
so time to post it.
The motivation for the changes in this series is mostly theoretical, but I do
see some idle power improvements from patch [4/5], for example, but nothing
specifically worth reporting.
The first patch simply prevents idle states with zero-size bins from being
selected sometimes when teo_select() runs with stopped tick.
Patch [2/5] avoids counting tick wakeups as intercepts unless there are
sufficiently many intercepts within the tick period range to assume that
the tick wakeup may have clobbered a genuine intercept.
Patch [3/5] simply updates a coefficient in one of the inequalities to be
somewhat easier to interpret (this should be a cosmetic change).
Patch [4/5] changes the criteria used for classifying wakeup events as hits
or intercepts to (hopefully) make the classification work better for large
state bins.
Patch [5/5] refines the idle state lookup based on intercepts to first
consider the state with the maximum intercepts metric, so that state is
always taken into consideration.
Please see the individual patch changelogs for details.
Thanks!
^ permalink raw reply [flat|nested] 21+ messages in thread
* [PATCH v1 1/5] cpuidle: governors: teo: Avoid selecting states with zero-size bins
2026-01-14 19:42 [PATCH v1 0/5] cpuidle: governors: teo: Wakeup events classification change and some refinements Rafael J. Wysocki
@ 2026-01-14 19:44 ` Rafael J. Wysocki
2026-01-21 13:09 ` Christian Loehle
2026-01-14 19:44 ` [PATCH v1 2/5] cpuidle: governors: teo: Avoid fake intercepts produced by tick Rafael J. Wysocki
` (5 subsequent siblings)
6 siblings, 1 reply; 21+ messages in thread
From: Rafael J. Wysocki @ 2026-01-14 19:44 UTC (permalink / raw)
To: Linux PM; +Cc: LKML, Christian Loehle, Doug Smythies
From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
If the last two enabled idle states have the same target residency which
is at least equal to TICK_NSET, teo may select the next-to-last one even
though the size of that state's bin is 0, which is confusing.
Prevent that from happening by adding a target residency check to the
relevant code path.
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
---
drivers/cpuidle/governors/teo.c | 10 ++++++++++
1 file changed, 10 insertions(+)
--- a/drivers/cpuidle/governors/teo.c
+++ b/drivers/cpuidle/governors/teo.c
@@ -388,6 +388,15 @@ static int teo_select(struct cpuidle_dri
while (min_idx < idx &&
drv->states[min_idx].target_residency_ns < TICK_NSEC)
min_idx++;
+
+ /*
+ * Avoid selecting a state with a lower index, but with
+ * the same target residency as the current candidate
+ * one.
+ */
+ if (drv->states[min_idx].target_residency_ns ==
+ drv->states[idx].target_residency_ns)
+ goto constraint;
}
/*
@@ -410,6 +419,7 @@ static int teo_select(struct cpuidle_dri
}
}
+constraint:
/*
* If there is a latency constraint, it may be necessary to select an
* idle state shallower than the current candidate one.
^ permalink raw reply [flat|nested] 21+ messages in thread
* [PATCH v1 2/5] cpuidle: governors: teo: Avoid fake intercepts produced by tick
2026-01-14 19:42 [PATCH v1 0/5] cpuidle: governors: teo: Wakeup events classification change and some refinements Rafael J. Wysocki
2026-01-14 19:44 ` [PATCH v1 1/5] cpuidle: governors: teo: Avoid selecting states with zero-size bins Rafael J. Wysocki
@ 2026-01-14 19:44 ` Rafael J. Wysocki
2026-01-21 13:34 ` Christian Loehle
2026-01-14 19:45 ` [PATCH v1 3/5] cpuidle: governors: teo: Refine tick_intercepts vs total events check Rafael J. Wysocki
` (4 subsequent siblings)
6 siblings, 1 reply; 21+ messages in thread
From: Rafael J. Wysocki @ 2026-01-14 19:44 UTC (permalink / raw)
To: Linux PM; +Cc: LKML, Christian Loehle, Doug Smythies
From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
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>
---
drivers/cpuidle/governors/teo.c | 11 +++++++++++
1 file changed, 11 insertions(+)
--- a/drivers/cpuidle/governors/teo.c
+++ b/drivers/cpuidle/governors/teo.c
@@ -239,6 +239,17 @@ static void teo_update(struct cpuidle_dr
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;
+ }
}
/*
^ permalink raw reply [flat|nested] 21+ messages in thread
* [PATCH v1 3/5] cpuidle: governors: teo: Refine tick_intercepts vs total events check
2026-01-14 19:42 [PATCH v1 0/5] cpuidle: governors: teo: Wakeup events classification change and some refinements Rafael J. Wysocki
2026-01-14 19:44 ` [PATCH v1 1/5] cpuidle: governors: teo: Avoid selecting states with zero-size bins Rafael J. Wysocki
2026-01-14 19:44 ` [PATCH v1 2/5] cpuidle: governors: teo: Avoid fake intercepts produced by tick Rafael J. Wysocki
@ 2026-01-14 19:45 ` Rafael J. Wysocki
2026-01-21 13:36 ` Christian Loehle
2026-01-14 19:46 ` [PATCH v1 4/5] cpuidle: governors: teo: Adjust the classification of wakeup events Rafael J. Wysocki
` (3 subsequent siblings)
6 siblings, 1 reply; 21+ messages in thread
From: Rafael J. Wysocki @ 2026-01-14 19:45 UTC (permalink / raw)
To: Linux PM; +Cc: LKML, Christian Loehle, Doug Smythies
From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Use 2/3 as the proportion coefficient in the check comparing
cpu_data->tick_intercepts with cpu_data->total because it is close
enough to the current one (5/8) and it allows of more straightforward
interpretation (on average, intercepts within the tick period length
are twice as frequent as other events).
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
---
drivers/cpuidle/governors/teo.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
--- a/drivers/cpuidle/governors/teo.c
+++ b/drivers/cpuidle/governors/teo.c
@@ -485,7 +485,7 @@ constraint:
* total wakeup events, do not stop the tick.
*/
if (drv->states[idx].target_residency_ns < TICK_NSEC &&
- cpu_data->tick_intercepts > cpu_data->total / 2 + cpu_data->total / 8)
+ 3 * cpu_data->tick_intercepts >= 2 * cpu_data->total)
duration_ns = TICK_NSEC / 2;
end:
^ permalink raw reply [flat|nested] 21+ messages in thread
* [PATCH v1 4/5] cpuidle: governors: teo: Adjust the classification of wakeup events
2026-01-14 19:42 [PATCH v1 0/5] cpuidle: governors: teo: Wakeup events classification change and some refinements Rafael J. Wysocki
` (2 preceding siblings ...)
2026-01-14 19:45 ` [PATCH v1 3/5] cpuidle: governors: teo: Refine tick_intercepts vs total events check Rafael J. Wysocki
@ 2026-01-14 19:46 ` Rafael J. Wysocki
2026-01-14 19:47 ` [PATCH v1 5/5] cpuidle: governors: teo: Refine intercepts-based idle state lookup Rafael J. Wysocki
` (2 subsequent siblings)
6 siblings, 0 replies; 21+ messages in thread
From: Rafael J. Wysocki @ 2026-01-14 19:46 UTC (permalink / raw)
To: Linux PM; +Cc: LKML, Christian Loehle, Doug Smythies
From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
If differences between target residency values of adjacent idle states
of a given CPU are relatively large, the corresponding idle state bins
used by the teo governors are large either and the rule by which hits
are distinguished from intercepts is inaccurate.
Namely, by that rule, a wakeup event is classified as a hit if the
sleep length (the time till the closest timer other than the tick)
and the measured idle duration, adjusted for the entered idle state
exit latency, fall into the same idle state bin. However, if that bin
is large enough, the actual difference between the sleep length and
the measured idle duration may be significant. It may in fact be
significantly greater than the analogous difference for an event where
the sleep length and the measured idle duration fall into different
bins.
For this reason, amend the rule in question with a check that will
only allow a wakeup event to be counted as a hit if the difference
between the sleep length and the measured idle duration is less than
LATENCY_THRESHOLD_NS (which means that the difference between the
sleep length and the raw measured idle duration is below the sum of
LATENCY_THRESHOLD_NS and 1/2 of the entered idle state exit latency).
Otherwise, the event will be counted as an intercept.
Moreover, since the above change is likely to cause more intercepts
and fewer hits to be counted than before, at least in some cases, also
adjust the check in teo_select() deciding whether or not to take
intercepts into account at all. Specifically, remove idx_hit_sum from
the right-hand side of the inequality in that check on the premise that
intercepts previously counted as hits will now be included in
idx_intercept_sum and there is no reason to subtract the remaining
events (that are still counted as hits) from cpu_data->total. Also
adjust the comment preceding the check in question to reflect that
modification.
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
---
drivers/cpuidle/governors/teo.c | 32 ++++++++++++++++----------------
1 file changed, 16 insertions(+), 16 deletions(-)
--- a/drivers/cpuidle/governors/teo.c
+++ b/drivers/cpuidle/governors/teo.c
@@ -48,13 +48,11 @@
* in accordance with what happened last time.
*
* The "hits" metric reflects the relative frequency of situations in which the
- * sleep length and the idle duration measured after CPU wakeup fall into the
- * same bin (that is, the CPU appears to wake up "on time" relative to the sleep
- * length). In turn, the "intercepts" metric reflects the relative frequency of
- * non-timer wakeup events for which the measured idle duration falls into a bin
- * that corresponds to an idle state shallower than the one whose bin is fallen
- * into by the sleep length (these events are also referred to as "intercepts"
- * below).
+ * sleep length and the idle duration measured after CPU wakeup are close enough
+ * (that is, the CPU appears to wake up "on time" relative to the sleep length).
+ * In turn, the "intercepts" metric reflects the relative frequency of non-timer
+ * wakeup events for which the measured idle duration is measurably less than
+ * the sleep length (these events are also referred to as "intercepts" below).
*
* The governor also counts "intercepts" with the measured idle duration below
* the tick period length and uses this information when deciding whether or not
@@ -253,12 +251,16 @@ static void teo_update(struct cpuidle_dr
}
/*
- * If the measured idle duration falls into the same bin as the sleep
- * length, this is a "hit", so update the "hits" metric for that bin.
+ * If the measured idle duration falls into the same bin as the
+ * sleep length and the difference between them is less than
+ * LATENCY_THRESHOLD_NS, this is a "hit", so update the "hits"
+ * metric for that bin.
+ *
* Otherwise, update the "intercepts" metric for the bin fallen into by
* the measured idle duration.
*/
- if (idx_timer == idx_duration) {
+ if (idx_timer == idx_duration &&
+ cpu_data->sleep_length_ns - measured_ns < LATENCY_THRESHOLD_NS) {
cpu_data->state_bins[idx_timer].hits += PULSE;
} else {
cpu_data->state_bins[idx_duration].intercepts += PULSE;
@@ -381,13 +383,11 @@ static int teo_select(struct cpuidle_dri
}
/*
- * If the sum of the intercepts metric for all of the idle states
- * shallower than the current candidate one (idx) is greater than the
- * sum of the intercepts and hits metrics for the candidate state and
- * all of the deeper states, a shallower idle state is likely to be a
- * better choice.
+ * If intercepts in the range below the target residency of the current
+ * candidate state are the majority of all wakeups, a shallower state is
+ * likely to be a better choice.
*/
- if (2 * idx_intercept_sum > cpu_data->total - idx_hit_sum) {
+ if (2 * idx_intercept_sum > cpu_data->total) {
int min_idx = idx0;
if (tick_nohz_tick_stopped()) {
^ permalink raw reply [flat|nested] 21+ messages in thread
* [PATCH v1 5/5] cpuidle: governors: teo: Refine intercepts-based idle state lookup
2026-01-14 19:42 [PATCH v1 0/5] cpuidle: governors: teo: Wakeup events classification change and some refinements Rafael J. Wysocki
` (3 preceding siblings ...)
2026-01-14 19:46 ` [PATCH v1 4/5] cpuidle: governors: teo: Adjust the classification of wakeup events Rafael J. Wysocki
@ 2026-01-14 19:47 ` Rafael J. Wysocki
2026-01-16 11:52 ` [PATCH v1 0/5] cpuidle: governors: teo: Wakeup events classification change and some refinements Christian Loehle
2026-01-20 15:29 ` [Update][PATCH v1.1 4/5] cpuidle: governors: teo: Adjust the classification of wakeup events Rafael J. Wysocki
6 siblings, 0 replies; 21+ messages in thread
From: Rafael J. Wysocki @ 2026-01-14 19:47 UTC (permalink / raw)
To: Linux PM; +Cc: LKML, Christian Loehle, Doug Smythies
From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
There are cases in which decisions made by the teo governor are
arguably overly conservative.
For instance, suppose that there are 4 idle states and the values of
the intercepts metric for the first 3 of them are 400, 250, and 251,
respectively. If the total sum computed in teo_update() is 1000, the
governor will select idle state 1 (provided that all idle states are
enabled and the scheduler tick has not been stopped) although arguably
idle state 0 would be a better choice because the likelihood of getting
an idle duration below the target residency of idle state 1 is greater
than the likelihood of getting an idle duration between the target
residency of idle state 1 and the target residency of idle state 2.
To address this, refine the candidate idle state lookup based on
intercepts to start at the state with the maximum intercepts metric,
below the deepest enabled one, to avoid the cases in which the search
may stop before reaching that state.
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
---
drivers/cpuidle/governors/teo.c | 46 +++++++++++++++++++++++++++++++++++-----
1 file changed, 41 insertions(+), 5 deletions(-)
--- a/drivers/cpuidle/governors/teo.c
+++ b/drivers/cpuidle/governors/teo.c
@@ -73,12 +73,17 @@
* than the candidate one (it represents the cases in which the CPU was
* likely woken up by a non-timer wakeup source).
*
+ * Also find the idle state with the maximum intercepts metric (if there are
+ * multiple states with the maximum intercetps metric, choose the one with
+ * the highest index).
+ *
* 2. If the second sum computed in step 1 is greater than a half of the sum of
* both metrics for the candidate state bin and all subsequent bins (if any),
* a shallower idle state is likely to be more suitable, so look for it.
*
* - Traverse the enabled idle states shallower than the candidate one in the
- * descending order.
+ * descending order, starting at the state with the maximum intercepts
+ * metric found in step 1.
*
* - For each of them compute the sum of the "intercepts" metrics over all
* of the idle states between it and the candidate one (including the
@@ -306,9 +311,12 @@ static int teo_select(struct cpuidle_dri
s64 latency_req = cpuidle_governor_latency_req(dev->cpu);
ktime_t delta_tick = TICK_NSEC / 2;
unsigned int idx_intercept_sum = 0;
+ unsigned int intercept_max_sum = 0;
unsigned int intercept_sum = 0;
+ unsigned int intercept_max = 0;
unsigned int idx_hit_sum = 0;
unsigned int hit_sum = 0;
+ int intercept_max_idx = -1;
int constraint_idx = 0;
int idx0 = 0, idx = -1;
s64 duration_ns;
@@ -339,17 +347,33 @@ static int teo_select(struct cpuidle_dri
if (!dev->states_usage[0].disable)
idx = 0;
- /* Compute the sums of metrics for early wakeup pattern detection. */
+ /*
+ * Compute the sums of metrics for early wakeup pattern detection and
+ * look for the state bin with the maximum intercepts metric below the
+ * deepest enabled one (if there are multiple states with the maximum
+ * intercepts metric, choose the one with the highest index).
+ */
for (i = 1; i < drv->state_count; i++) {
struct teo_bin *prev_bin = &cpu_data->state_bins[i-1];
+ unsigned int prev_intercepts = prev_bin->intercepts;
struct cpuidle_state *s = &drv->states[i];
/*
* Update the sums of idle state metrics for all of the states
* shallower than the current one.
*/
- intercept_sum += prev_bin->intercepts;
hit_sum += prev_bin->hits;
+ intercept_sum += prev_intercepts;
+ /*
+ * Check if this is the bin with the maximum number of
+ * intercepts so far and in that case update the index of
+ * the state with the maximum intercetps metric.
+ */
+ if (prev_intercepts >= intercept_max) {
+ intercept_max = prev_intercepts;
+ intercept_max_sum = intercept_sum;
+ intercept_max_idx = i - 1;
+ }
if (dev->states_usage[i].disable)
continue;
@@ -411,14 +435,26 @@ static int teo_select(struct cpuidle_dri
}
/*
- * Look for the deepest idle state whose target residency had
+ * If the minimum state index is greater than or equal to the
+ * index of the state with the maximum intercepts metric, there
+ * is no need to look at the shallower states.
+ */
+ if (min_idx >= intercept_max_idx) {
+ idx = min_idx;
+ goto constraint;
+ }
+
+ /*
+ * Look for the deepest idle state at least as deep as the one
+ * with the maximum intercetps metric whose target residency had
* not exceeded the idle duration in over a half of the relevant
* cases in the past.
*
* Take the possible duration limitation present if the tick
* has been stopped already into account.
*/
- for (i = idx - 1, intercept_sum = 0; i >= min_idx; i--) {
+ intercept_sum = idx_intercept_sum - intercept_max_sum;
+ for (i = intercept_max_idx; i >= min_idx; i--) {
intercept_sum += cpu_data->state_bins[i].intercepts;
if (dev->states_usage[i].disable)
^ permalink raw reply [flat|nested] 21+ messages in thread
* Re: [PATCH v1 0/5] cpuidle: governors: teo: Wakeup events classification change and some refinements
2026-01-14 19:42 [PATCH v1 0/5] cpuidle: governors: teo: Wakeup events classification change and some refinements Rafael J. Wysocki
` (4 preceding siblings ...)
2026-01-14 19:47 ` [PATCH v1 5/5] cpuidle: governors: teo: Refine intercepts-based idle state lookup Rafael J. Wysocki
@ 2026-01-16 11:52 ` Christian Loehle
2026-01-16 12:29 ` Rafael J. Wysocki
2026-01-20 15:29 ` [Update][PATCH v1.1 4/5] cpuidle: governors: teo: Adjust the classification of wakeup events Rafael J. Wysocki
6 siblings, 1 reply; 21+ messages in thread
From: Christian Loehle @ 2026-01-16 11:52 UTC (permalink / raw)
To: Rafael J. Wysocki, Linux PM; +Cc: LKML, Doug Smythies
[-- Attachment #1: Type: text/plain, Size: 2148 bytes --]
On 1/14/26 19:42, Rafael J. Wysocki wrote:
> Hi All,
>
> This material has been in my local queue for almost a full development cycle,
> so time to post it.
>
> The motivation for the changes in this series is mostly theoretical, but I do
> see some idle power improvements from patch [4/5], for example, but nothing
> specifically worth reporting.
>
> The first patch simply prevents idle states with zero-size bins from being
> selected sometimes when teo_select() runs with stopped tick.
>
> Patch [2/5] avoids counting tick wakeups as intercepts unless there are
> sufficiently many intercepts within the tick period range to assume that
> the tick wakeup may have clobbered a genuine intercept.
>
> Patch [3/5] simply updates a coefficient in one of the inequalities to be
> somewhat easier to interpret (this should be a cosmetic change).
>
> Patch [4/5] changes the criteria used for classifying wakeup events as hits
> or intercepts to (hopefully) make the classification work better for large
> state bins.
>
> Patch [5/5] refines the idle state lookup based on intercepts to first
> consider the state with the maximum intercepts metric, so that state is
> always taken into consideration.
>
> Please see the individual patch changelogs for details.
>
> Thanks!
>
>
>
Hi Rafael,
I'll do the in-depth review, but have run some tests already.
They are attached, platform is the usual rk3399.
"teo" is mainline, "teo-$i" is with patches 1..$i applied.
There's a regression on teo-4 visible on the intercept heavy IO workloads,
for idle misses that isn't strong enough to reflect in score changes except
for the very slow mtdblock device.
interestingly though there also seems to be a regression in
mapper/dm-slow (dm device with 51ms delay on each IO), which is not
intercept heavy.
Looking at the state residencies it overuses the deepest state2 in
where state1 was preferred for the other teo variants.
I've attached that too for reference.
I'm assuming that is because of the new intercept-logic-exclusion clause.
teo-5 seems to be slightly better than teo-4 here, but still a regression
from the others.
Regards,
Christian
[-- Attachment #2: teo-6.19-patches-rafael-wakeup-classification-dm-slow-residencies.txt --]
[-- Type: text/plain, Size: 9781 bytes --]
====================================
menu
====================================
time
idle_state
-1.0 0.522207
2.0 29.809488
0.0 0.439359
1.0 0.023592
time
idle_state
-1.0 0.386142
0.0 0.415897
2.0 29.693899
1.0 0.009253
time
idle_state
-1.0 0.429166
0.0 0.435448
2.0 29.676407
1.0 0.006326
time
idle_state
-1.0 0.106030
0.0 0.218954
2.0 30.112272
time
idle_state
-1.0 0.170958
2.0 29.984560
0.0 0.293831
1.0 0.001573
time
idle_state
-1.0 0.101308
2.0 30.005663
0.0 0.331625
time
idle_state
-1.0 0.364297
2.0 29.629962
0.0 0.248982
1.0 0.135782
time
idle_state
-1.0 0.385761
0.0 0.637699
2.0 29.422025
time
idle_state
-1.0 0.114566
0.0 0.145572
2.0 30.185000
time
idle_state
-1.0 0.474654
0.0 0.289992
2.0 29.629828
1.0 0.004852
====================================
teo
====================================
time
idle_state
-1.0 0.581934
1.0 0.178947
0.0 8.080984
2.0 21.546238
time
idle_state
-1.0 0.551908
2.0 23.129510
0.0 5.920605
1.0 0.847930
time
idle_state
-1.0 0.341767
0.0 5.217195
2.0 24.714937
1.0 0.142218
time
idle_state
-1.0 0.440228
0.0 4.791386
2.0 23.538069
1.0 1.630066
time
idle_state
-1.0 0.579999
0.0 8.612213
2.0 21.189232
1.0 0.019412
time
idle_state
-1.0 0.497495
0.0 8.350844
1.0 0.516045
2.0 21.049258
time
idle_state
-1.0 0.224099
0.0 2.699585
2.0 12.324405
1.0 15.146087
time
idle_state
-1.0 0.546996
0.0 6.022450
2.0 23.194753
1.0 0.647525
time
idle_state
-1.0 0.621619
2.0 19.158942
1.0 1.085783
0.0 9.539629
time
idle_state
-1.0 0.613194
2.0 17.900722
1.0 2.790429
0.0 9.087842
====================================
teo-1
====================================
time
idle_state
-1.0 0.424316
2.0 22.427511
1.0 1.059403
0.0 6.502851
time
idle_state
-1.0 0.186115
1.0 0.174262
0.0 1.944237
2.0 28.112224
time
idle_state
-1.0 0.469465
0.0 5.272924
2.0 24.462418
1.0 0.220847
time
idle_state
-1.0 0.097441
0.0 0.419606
2.0 29.418198
1.0 0.467028
time
idle_state
-1.0 0.236107
1.0 1.042653
0.0 2.864246
2.0 26.254346
time
idle_state
-1.0 0.536481
2.0 18.735667
0.0 11.073653
1.0 0.072279
time
idle_state
-1.0 0.307125
0.0 5.327118
2.0 22.648541
1.0 2.145863
time
idle_state
-1.0 0.162878
0.0 0.397467
1.0 2.296615
2.0 27.544667
time
idle_state
-1.0 0.393321
0.0 6.907063
2.0 23.059045
1.0 0.057276
time
idle_state
-1.0 0.154064
1.0 0.361785
0.0 0.332219
2.0 29.543256
====================================
teo-2
====================================
time
idle_state
-1.0 0.368612
0.0 6.131738
1.0 0.389804
2.0 23.522502
time
idle_state
-1.0 0.213287
0.0 2.048208
2.0 26.171435
1.0 1.970807
time
idle_state
-1.0 0.255844
1.0 0.421552
0.0 3.914190
2.0 25.800547
time
idle_state
-1.0 0.333126
0.0 4.614121
1.0 0.312318
2.0 25.136069
time
idle_state
-1.0 0.158395
0.0 0.426980
1.0 2.124545
2.0 27.712900
time
idle_state
-1.0 0.378377
2.0 26.404118
1.0 0.422916
0.0 3.210590
time
idle_state
-1.0 0.303689
0.0 4.941710
1.0 1.066197
2.0 24.101117
time
idle_state
-1.0 0.473386
0.0 6.704000
2.0 20.205167
1.0 3.004110
time
idle_state
-1.0 0.431586
0.0 3.934274
2.0 25.804022
1.0 0.216861
time
idle_state
-1.0 0.270506
0.0 2.238385
2.0 27.290971
1.0 0.608233
====================================
teo-3
====================================
time
idle_state
-1.0 0.097063
0.0 0.267526
2.0 29.688347
1.0 0.376698
time
idle_state
-1.0 0.200512
1.0 0.444248
0.0 1.640503
2.0 28.146898
time
idle_state
-1.0 0.311647
2.0 19.779806
0.0 7.135393
1.0 3.184264
time
idle_state
-1.0 0.173353
0.0 0.813484
1.0 0.912186
2.0 28.507607
time
idle_state
-1.0 0.544571
2.0 21.892216
1.0 0.258679
0.0 7.725663
time
idle_state
-1.0 0.122469
0.0 0.790456
2.0 29.350012
1.0 0.156731
time
idle_state
-1.0 0.481248
2.0 24.213318
1.0 0.166705
0.0 5.535685
time
idle_state
-1.0 0.503180
0.0 5.647267
2.0 22.769293
1.0 1.465941
time
idle_state
-1.0 0.148520
2.0 29.775357
1.0 0.110179
0.0 0.345198
time
idle_state
-1.0 0.115748
1.0 0.678066
0.0 0.251607
2.0 29.329033
====================================
teo-4
====================================
time
idle_state
-1.0 0.415130
1.0 0.011785
0.0 0.519576
2.0 29.431846
time
idle_state
-1.0 0.437376
0.0 0.605326
1.0 0.030607
2.0 29.352431
time
idle_state
-1.0 0.139508
0.0 0.108725
1.0 0.005131
2.0 30.165426
time
idle_state
-1.0 0.513020
0.0 0.598344
1.0 0.008301
2.0 29.287957
time
idle_state
-1.0 0.466327
0.0 0.723815
2.0 29.185117
1.0 0.024337
time
idle_state
-1.0 0.497801
0.0 0.687437
2.0 29.158447
1.0 0.066941
time
idle_state
-1.0 0.435365
2.0 29.280457
1.0 0.131756
0.0 0.534759
time
idle_state
-1.0 0.090548
0.0 0.109003
2.0 30.184958
1.0 0.009593
time
idle_state
-1.0 0.374734
2.0 29.505441
0.0 0.494162
1.0 0.023750
time
idle_state
-1.0 0.436251
2.0 29.432858
1.0 0.010065
0.0 0.509389
====================================
teo-5
====================================
====================================
time
idle_state
-1.0 0.263445
0.0 7.687333
2.0 22.316729
1.0 0.128150
time
idle_state
-1.0 0.284426
0.0 0.439623
2.0 28.827033
1.0 0.846147
time
idle_state
-1.0 0.138399
2.0 28.362396
0.0 1.385181
1.0 0.502742
time
idle_state
-1.0 0.103490
0.0 0.133783
2.0 30.145286
1.0 0.017852
time
idle_state
-1.0 0.418341
2.0 29.286523
1.0 0.100358
0.0 0.581828
time
idle_state
-1.0 0.279997
2.0 29.760241
1.0 0.024312
0.0 0.325708
time
idle_state
-1.0 0.530353
2.0 29.086169
1.0 0.096106
0.0 0.703320
time
idle_state
-1.0 0.502074
0.0 0.643404
2.0 29.464513
1.0 0.011626
time
idle_state
-1.0 0.533315
0.0 0.678228
2.0 29.121867
1.0 0.065334
time
idle_state
-1.0 0.091624
0.0 0.193224
2.0 30.098391
1.0 0.020493
Click to add a cell.
[-- Attachment #3: teo-6.19-patches-rafael-wakeup-classification-results.txt --]
[-- Type: text/plain, Size: 44962 bytes --]
device gov iter iops idles idle_miss ratio belows aboves
mmcblk1 menu 0 1557 397820 94403 0.237 94351 52
mmcblk1 menu 1 1561 401416 93736 0.234 93684 52
mmcblk1 menu 2 1570 404934 94086 0.232 94082 4
mmcblk1 menu 3 1560 398710 94644 0.237 94599 45
mmcblk1 menu 4 1570 403904 94315 0.234 94245 70
mmcblk1 teo 0 2326 589868 1491 0.003 1376 115
mmcblk1 teo 1 2329 616170 1527 0.002 1508 19
mmcblk1 teo 2 2328 613920 1494 0.002 1467 27
mmcblk1 teo 3 2329 614426 1560 0.003 1530 30
mmcblk1 teo 4 2329 615786 1568 0.003 1544 24
mmcblk1 teo-1 0 2329 615372 1563 0.003 1536 27
mmcblk1 teo-1 1 2329 612746 1502 0.002 1477 25
mmcblk1 teo-1 2 2329 611428 1644 0.003 1584 60
mmcblk1 teo-1 3 2328 615404 1551 0.003 1515 36
mmcblk1 teo-1 4 2328 613994 1802 0.003 1740 62
mmcblk1 teo-2 0 2328 610470 1498 0.002 1466 32
mmcblk1 teo-2 1 2329 609530 1656 0.003 1611 45
mmcblk1 teo-2 2 2324 585088 1702 0.003 1608 94
mmcblk1 teo-2 3 2329 609768 1716 0.003 1666 50
mmcblk1 teo-2 4 2328 611256 1523 0.002 1501 22
mmcblk1 teo-3 0 2328 609742 1681 0.003 1625 56
mmcblk1 teo-3 1 2328 609009 1679 0.003 1623 56
mmcblk1 teo-3 2 2329 610234 1696 0.003 1641 55
mmcblk1 teo-3 3 2328 611482 1561 0.003 1536 25
mmcblk1 teo-3 4 2329 611874 1567 0.003 1534 33
mmcblk1 teo-4 0 2348 615552 2820 0.005 2807 13
mmcblk1 teo-4 1 2354 619166 2855 0.005 2847 8
mmcblk1 teo-4 2 2355 617392 2806 0.005 2793 13
mmcblk1 teo-4 3 2328 602228 2834 0.005 2816 18
mmcblk1 teo-4 4 2351 614586 2715 0.004 2702 13
mmcblk1 teo-5 0 2330 611972 2821 0.005 2806 15
mmcblk1 teo-5 1 2328 610244 2886 0.005 2869 17
mmcblk1 teo-5 2 2329 600070 1910 0.003 1895 15
mmcblk1 teo-5 3 2352 614102 2652 0.004 2645 7
mmcblk1 teo-5 4 2329 600028 1976 0.003 1962 14
mmcblk2 menu 0 3060 434056 111934 0.258 111929 5
mmcblk2 menu 1 3604 378218 67645 0.179 67639 6
mmcblk2 menu 2 3644 514887 120580 0.234 120559 21
mmcblk2 menu 3 3684 411128 70928 0.173 70914 14
mmcblk2 menu 4 3438 409996 88536 0.216 88524 12
mmcblk2 teo 0 5760 812856 4487 0.006 4452 35
mmcblk2 teo 1 5675 691494 3089 0.004 2993 96
mmcblk2 teo 2 5671 852962 4769 0.006 4717 52
mmcblk2 teo 3 5686 753282 2600 0.003 2541 59
mmcblk2 teo 4 5777 855566 632 0.001 627 5
mmcblk2 teo-1 0 5713 881538 2970 0.003 2963 7
mmcblk2 teo-1 1 5707 872648 2859 0.003 2811 48
mmcblk2 teo-1 2 5715 860354 1612 0.002 1582 30
mmcblk2 teo-1 3 5689 746826 3402 0.005 3328 74
mmcblk2 teo-1 4 5689 655828 1494 0.002 1474 20
mmcblk2 teo-2 0 5638 605932 6022 0.010 6012 10
mmcblk2 teo-2 1 5898 862506 2229 0.003 2218 11
mmcblk2 teo-2 2 5694 740193 2442 0.003 2379 63
mmcblk2 teo-2 3 5699 861458 2257 0.003 2208 49
mmcblk2 teo-2 4 5713 873826 2053 0.002 2005 48
mmcblk2 teo-3 0 5685 693868 2649 0.004 2590 59
mmcblk2 teo-3 1 5716 872350 2295 0.003 2273 22
mmcblk2 teo-3 2 5697 842930 3028 0.004 2998 30
mmcblk2 teo-3 3 5714 864032 1772 0.002 1695 77
mmcblk2 teo-3 4 5683 679870 2549 0.004 2481 68
mmcblk2 teo-4 0 5674 862964 6700 0.008 6684 16
mmcblk2 teo-4 1 5678 811848 5486 0.007 5470 16
mmcblk2 teo-4 2 5713 870238 2841 0.003 2833 8
mmcblk2 teo-4 3 5852 878620 4099 0.005 4091 8
mmcblk2 teo-4 4 5888 894664 5885 0.007 5874 11
mmcblk2 teo-5 0 5672 628654 3866 0.006 3843 23
mmcblk2 teo-5 1 5579 766352 14131 0.018 14118 13
mmcblk2 teo-5 2 5714 869916 2839 0.003 2818 21
mmcblk2 teo-5 3 5672 578388 3475 0.006 3468 7
mmcblk2 teo-5 4 5611 740432 18248 0.025 18239 9
nvme0n1 menu 0 5866 459102 55657 0.121 55585 72
nvme0n1 menu 1 6188 459834 50885 0.111 50778 107
nvme0n1 menu 2 6962 529744 62101 0.117 62012 89
nvme0n1 menu 3 5960 463992 51725 0.111 51561 164
nvme0n1 menu 4 8013 598132 69902 0.117 69892 10
nvme0n1 teo 0 10597 753096 1531 0.002 1442 89
nvme0n1 teo 1 10501 750804 1668 0.002 1561 107
nvme0n1 teo 2 10564 753516 1881 0.002 1786 95
nvme0n1 teo 3 11240 786878 1388 0.002 1366 22
nvme0n1 teo 4 10463 747932 1587 0.002 1550 37
nvme0n1 teo-1 0 10548 752680 1639 0.002 1564 75
nvme0n1 teo-1 1 10398 724296 889 0.001 787 102
nvme0n1 teo-1 2 10564 753492 1871 0.002 1765 106
nvme0n1 teo-1 3 10674 761064 1760 0.002 1667 93
nvme0n1 teo-1 4 10929 777152 1966 0.003 1878 88
nvme0n1 teo-2 0 10906 764638 1472 0.002 1456 16
nvme0n1 teo-2 1 10709 760364 1736 0.002 1661 75
nvme0n1 teo-2 2 10670 757406 1761 0.002 1689 72
nvme0n1 teo-2 3 10471 743518 1625 0.002 1528 97
nvme0n1 teo-2 4 10729 760790 1729 0.002 1677 52
nvme0n1 teo-3 0 11110 785084 1811 0.002 1753 58
nvme0n1 teo-3 1 11001 779932 1860 0.002 1782 78
nvme0n1 teo-3 2 10525 747584 1706 0.002 1550 156
nvme0n1 teo-3 3 11624 807506 1234 0.002 1150 84
nvme0n1 teo-3 4 10435 744040 1723 0.002 1569 154
nvme0n1 teo-4 0 10984 778096 2945 0.004 2937 8
nvme0n1 teo-4 1 10824 767550 2963 0.004 2934 29
nvme0n1 teo-4 2 11536 813280 2763 0.003 2756 7
nvme0n1 teo-4 3 11464 811950 2871 0.004 2863 8
nvme0n1 teo-4 4 10500 746828 2911 0.004 2883 28
nvme0n1 teo-5 0 11156 784900 2542 0.003 2529 13
nvme0n1 teo-5 1 11573 818806 2842 0.003 2835 7
nvme0n1 teo-5 2 11007 781880 2880 0.004 2870 10
nvme0n1 teo-5 3 10522 742995 2397 0.003 2386 11
nvme0n1 teo-5 4 10529 749368 2867 0.004 2858 9
nullb0 menu 0 97218 74494 1860 0.025 1859 1
nullb0 menu 1 97750 83802 3030 0.036 2992 38
nullb0 menu 2 96859 86830 2939 0.034 2923 16
nullb0 menu 3 97792 85786 2811 0.033 2809 2
nullb0 menu 4 97739 86370 3026 0.035 3006 20
nullb0 teo 0 96986 83320 3162 0.038 2966 196
nullb0 teo 1 97758 68370 1863 0.027 1637 226
nullb0 teo 2 97491 84736 4417 0.052 4196 221
nullb0 teo 3 97624 84226 3115 0.037 3051 64
nullb0 teo 4 97486 83702 3798 0.045 3617 181
nullb0 teo-1 0 97614 83620 3864 0.046 3705 159
nullb0 teo-1 1 97426 83024 4841 0.058 4570 271
nullb0 teo-1 2 97895 84302 3365 0.040 3343 22
nullb0 teo-1 3 97802 83556 2866 0.034 2808 58
nullb0 teo-1 4 97452 82140 4648 0.057 4489 159
nullb0 teo-2 0 97115 78638 2560 0.033 2521 39
nullb0 teo-2 1 98063 79152 3413 0.043 3291 122
nullb0 teo-2 2 97819 79352 2699 0.034 2612 87
nullb0 teo-2 3 97711 80408 3444 0.043 3257 187
nullb0 teo-2 4 96996 76586 3674 0.048 3450 224
nullb0 teo-3 0 97219 77306 4216 0.055 3913 303
nullb0 teo-3 1 97742 78320 3250 0.041 3139 111
nullb0 teo-3 2 97875 79322 3144 0.040 3036 108
nullb0 teo-3 3 97799 78166 2386 0.031 2258 128
nullb0 teo-3 4 97543 77132 4206 0.055 3933 273
nullb0 teo-4 0 98087 86676 3001 0.035 2990 11
nullb0 teo-4 1 97711 86866 2879 0.033 2868 11
nullb0 teo-4 2 98402 87054 2843 0.033 2835 8
nullb0 teo-4 3 98304 84442 2830 0.034 2822 8
nullb0 teo-4 4 97182 80068 2755 0.034 2741 14
nullb0 teo-5 0 97408 85380 3233 0.038 3208 25
nullb0 teo-5 1 97735 85558 3018 0.035 3012 6
nullb0 teo-5 2 96778 78558 2468 0.031 2460 8
nullb0 teo-5 3 98012 85802 2993 0.035 2987 6
nullb0 teo-5 4 97567 85564 2868 0.034 2860 8
mtdblock3 menu 0 168 369546 94826 0.257 94821 5
mtdblock3 menu 1 176 346222 83578 0.241 83565 13
mtdblock3 menu 2 165 387704 101194 0.261 101172 22
mtdblock3 menu 3 168 399594 100182 0.251 100138 44
mtdblock3 menu 4 160 485974 117974 0.243 117969 5
mtdblock3 teo 0 260 291392 1621 0.006 1605 16
mtdblock3 teo 1 260 343224 2225 0.006 2207 18
mtdblock3 teo 2 261 911894 1056 0.001 1054 2
mtdblock3 teo 3 261 946920 2504 0.003 2490 14
mtdblock3 teo 4 260 359508 2202 0.006 2185 17
mtdblock3 teo-1 0 258 549900 3889 0.007 3871 18
mtdblock3 teo-1 1 260 903866 3472 0.004 3452 20
mtdblock3 teo-1 2 257 858518 549 0.001 509 40
mtdblock3 teo-1 3 257 772806 2237 0.003 2221 16
mtdblock3 teo-1 4 255 894394 5834 0.007 5818 16
mtdblock3 teo-2 0 260 278426 2139 0.008 2118 21
mtdblock3 teo-2 1 256 926870 3226 0.003 3205 21
mtdblock3 teo-2 2 255 883874 5728 0.006 5719 9
mtdblock3 teo-2 3 255 904428 4382 0.005 4371 11
mtdblock3 teo-2 4 256 752692 4494 0.006 4464 30
mtdblock3 teo-3 0 256 775924 3147 0.004 3095 52
mtdblock3 teo-3 1 257 745188 3040 0.004 3024 16
mtdblock3 teo-3 2 257 665386 3209 0.005 3177 32
mtdblock3 teo-3 3 257 858298 1291 0.002 1246 45
mtdblock3 teo-3 4 262 856940 2092 0.002 2043 49
mtdblock3 teo-4 0 256 915940 2853 0.003 2847 6
mtdblock3 teo-4 1 258 633828 2042 0.003 2028 14
mtdblock3 teo-4 2 256 828430 2988 0.004 2982 6
mtdblock3 teo-4 3 255 900750 4900 0.005 4887 13
mtdblock3 teo-4 4 253 555864 12439 0.022 12431 8
mtdblock3 teo-5 0 255 904272 5026 0.006 5013 13
mtdblock3 teo-5 1 255 908694 5566 0.006 5558 8
mtdblock3 teo-5 2 254 904998 5897 0.007 5887 10
mtdblock3 teo-5 3 255 906142 6171 0.007 6157 14
mtdblock3 teo-5 4 260 301928 2849 0.009 2836 13
test gov i score %change idles idle_miss miss_rt belows aboves
schbench menu 0 281.70 +0.00% 31444 1851 0.059 1840 11
schbench menu 1 271.60 -3.59% 30218 1816 0.060 1808 8
schbench menu 2 290.83 +3.24% 32637 1931 0.059 1920 11
schbench menu 3 292.20 +3.73% 32770 1966 0.060 1952 14
schbench menu 4 295.33 +4.84% 33330 2120 0.064 2106 14
schbench teo 0 301.50 +7.03% 30792 103 0.003 94 9
schbench teo 1 299.13 +6.19% 31736 104 0.003 94 10
schbench teo 2 297.43 +5.58% 30406 83 0.003 78 5
schbench teo 3 297.67 +5.67% 31004 91 0.003 81 10
schbench teo 4 295.87 +5.03% 30206 95 0.003 82 13
schbench teo-1 0 295.87 +5.03% 30432 69 0.002 65 4
schbench teo-1 1 298.97 +6.13% 32200 102 0.003 92 10
schbench teo-1 2 298.57 +5.99% 31950 82 0.003 72 10
schbench teo-1 3 263.63 -6.41% 27220 87 0.003 77 10
schbench teo-1 4 287.53 +2.07% 29042 104 0.004 96 8
schbench teo-2 0 284.00 +0.82% 29824 122 0.004 117 5
schbench teo-2 1 290.90 +3.27% 31508 170 0.005 157 13
schbench teo-2 2 285.07 +1.20% 30058 111 0.004 97 14
schbench teo-2 3 290.07 +2.97% 30914 142 0.005 129 13
schbench teo-2 4 289.03 +2.60% 30294 117 0.004 108 9
schbench teo-3 0 289.30 +2.70% 30506 134 0.004 124 10
schbench teo-3 1 286.07 +1.55% 30868 118 0.004 103 15
schbench teo-3 2 288.87 +2.55% 31574 146 0.005 136 10
schbench teo-3 3 291.43 +3.45% 31414 139 0.004 124 15
schbench teo-3 4 292.97 +4.00% 31468 133 0.004 120 13
schbench teo-4 0 292.03 +3.67% 31541 730 0.023 715 15
schbench teo-4 1 299.30 +6.25% 32908 748 0.023 734 14
schbench teo-4 2 272.20 -3.37% 27826 1028 0.037 1015 13
schbench teo-4 3 264.40 -6.14% 27611 749 0.027 738 11
schbench teo-4 4 290.77 +3.22% 30914 881 0.028 867 14
schbench teo-5 0 275.17 -2.32% 27730 878 0.032 868 10
schbench teo-5 1 288.50 +2.41% 30938 860 0.028 848 12
schbench teo-5 2 287.93 +2.21% 30630 929 0.030 911 18
schbench teo-5 3 288.63 +2.46% 30446 849 0.028 837 12
schbench teo-5 4 288.40 +2.38% 30946 988 0.032 957 31
ebizzy menu 0 10834.00 +0.00% 1066 39 0.037 39 0
ebizzy menu 1 10758.00 -0.70% 1454 55 0.038 55 0
ebizzy menu 2 10545.00 -2.67% 1076 49 0.046 49 0
ebizzy menu 3 10720.00 -1.05% 1390 28 0.020 28 0
ebizzy menu 4 10728.00 -0.98% 1126 51 0.045 51 0
ebizzy teo 0 10754.00 -0.74% 1082 46 0.043 41 5
ebizzy teo 1 10813.00 -0.19% 1098 44 0.040 38 6
ebizzy teo 2 10724.00 -1.02% 1198 40 0.033 38 2
ebizzy teo 3 10721.00 -1.04% 1096 31 0.028 28 3
ebizzy teo 4 10777.00 -0.53% 1042 41 0.039 32 9
ebizzy teo-1 0 10845.00 +0.10% 1036 31 0.030 26 5
ebizzy teo-1 1 10836.00 +0.02% 1263 46 0.036 39 7
ebizzy teo-1 2 10868.00 +0.31% 1028 39 0.038 34 5
ebizzy teo-1 3 10710.00 -1.14% 1404 42 0.030 38 4
ebizzy teo-1 4 10787.00 -0.43% 974 27 0.028 24 3
ebizzy teo-2 0 10742.00 -0.85% 1200 45 0.037 39 6
ebizzy teo-2 1 10754.00 -0.74% 1122 44 0.039 37 7
ebizzy teo-2 2 10733.00 -0.93% 1148 52 0.045 44 8
ebizzy teo-2 3 10795.00 -0.36% 1084 45 0.042 40 5
ebizzy teo-2 4 10766.00 -0.63% 1214 65 0.054 60 5
ebizzy teo-3 0 10732.00 -0.94% 1010 42 0.042 35 7
ebizzy teo-3 1 10717.00 -1.08% 1146 23 0.020 19 4
ebizzy teo-3 2 10764.00 -0.65% 1100 34 0.031 31 3
ebizzy teo-3 3 10754.00 -0.74% 1007 29 0.029 25 4
ebizzy teo-3 4 10678.00 -1.44% 1348 36 0.027 34 2
ebizzy teo-4 0 10762.00 -0.66% 1152 43 0.037 37 6
ebizzy teo-4 1 10767.00 -0.62% 998 37 0.037 33 4
ebizzy teo-4 2 10729.00 -0.97% 1170 49 0.042 42 7
ebizzy teo-4 3 10843.00 +0.08% 1076 34 0.032 29 5
ebizzy teo-4 4 10723.00 -1.02% 1032 50 0.048 45 5
ebizzy teo-5 0 10763.00 -0.66% 1074 37 0.034 33 4
ebizzy teo-5 1 10791.00 -0.40% 1028 27 0.026 24 3
ebizzy teo-5 2 10803.00 -0.29% 1098 41 0.037 37 4
ebizzy teo-5 3 10809.00 -0.23% 1152 29 0.025 26 3
ebizzy teo-5 4 10808.00 -0.24% 1172 46 0.039 41 5
adrestia menu 0 12.00 +0.00% 103064 310 0.003 310 0
adrestia menu 1 12.00 +0.00% 103196 201 0.002 201 0
adrestia menu 2 12.00 +0.00% 103162 237 0.002 237 0
adrestia menu 3 12.00 +0.00% 102962 277 0.003 276 1
adrestia menu 4 12.00 +0.00% 103128 293 0.003 293 0
adrestia teo 0 12.00 +0.00% 102998 54 0.001 34 20
adrestia teo 1 12.00 +0.00% 103242 39 0.000 35 4
adrestia teo 2 12.00 +0.00% 103078 40 0.000 32 8
adrestia teo 3 12.00 +0.00% 103164 24 0.000 21 3
adrestia teo 4 12.00 +0.00% 102938 39 0.000 33 6
adrestia teo-1 0 12.00 +0.00% 103076 43 0.000 31 12
adrestia teo-1 1 8.00 -33.33% 103358 44 0.000 32 12
adrestia teo-1 2 12.00 +0.00% 103050 32 0.000 27 5
adrestia teo-1 3 12.00 +0.00% 103196 30 0.000 27 3
adrestia teo-1 4 12.00 +0.00% 103405 46 0.000 43 3
adrestia teo-2 0 12.00 +0.00% 102992 32 0.000 27 5
adrestia teo-2 1 12.00 +0.00% 102920 53 0.001 45 8
adrestia teo-2 2 11.00 -8.33% 103018 42 0.000 36 6
adrestia teo-2 3 12.00 +0.00% 102942 42 0.000 39 3
adrestia teo-2 4 11.00 -8.33% 103236 46 0.000 44 2
adrestia teo-3 0 12.00 +0.00% 103182 55 0.001 43 12
adrestia teo-3 1 12.00 +0.00% 102854 33 0.000 29 4
adrestia teo-3 2 12.00 +0.00% 102828 35 0.000 29 6
adrestia teo-3 3 12.00 +0.00% 103126 37 0.000 29 8
adrestia teo-3 4 12.00 +0.00% 103104 44 0.000 38 6
adrestia teo-4 0 12.00 +0.00% 102864 51 0.000 46 5
adrestia teo-4 1 12.00 +0.00% 103236 70 0.001 67 3
adrestia teo-4 2 12.00 +0.00% 102976 24 0.000 19 5
adrestia teo-4 3 12.00 +0.00% 103158 45 0.000 39 6
adrestia teo-4 4 12.00 +0.00% 102988 64 0.001 58 6
adrestia teo-5 0 12.00 +0.00% 102900 48 0.000 44 4
adrestia teo-5 1 12.00 +0.00% 102910 23 0.000 22 1
adrestia teo-5 2 12.00 +0.00% 102762 31 0.000 26 5
adrestia teo-5 3 12.00 +0.00% 102898 55 0.001 48 7
adrestia teo-5 4 12.00 +0.00% 103096 46 0.000 44 2
hackbench menu 0 22.11 +0.00% 11298 711 0.063 711 0
hackbench menu 1 21.96 -0.67% 5768 545 0.094 545 0
hackbench menu 2 21.85 -1.17% 6854 405 0.059 405 0
hackbench menu 3 21.83 -1.26% 3840 555 0.145 555 0
hackbench menu 4 21.70 -1.85% 5518 468 0.085 468 0
hackbench teo 0 21.70 -1.85% 3440 46 0.013 42 4
hackbench teo 1 21.85 -1.18% 4258 31 0.007 26 5
hackbench teo 2 21.64 -2.11% 4036 33 0.008 29 4
hackbench teo 3 21.78 -1.48% 4596 37 0.008 33 4
hackbench teo 4 21.75 -1.61% 5088 42 0.008 36 6
hackbench teo-1 0 21.81 -1.32% 5838 40 0.007 34 6
hackbench teo-1 1 21.74 -1.67% 4380 41 0.009 38 3
hackbench teo-1 2 22.09 -0.07% 4946 42 0.008 33 9
hackbench teo-1 3 21.59 -2.32% 4660 35 0.008 25 10
hackbench teo-1 4 21.79 -1.42% 3672 37 0.010 31 6
hackbench teo-2 0 21.83 -1.27% 5092 29 0.006 23 6
hackbench teo-2 1 21.96 -0.66% 5728 42 0.007 35 7
hackbench teo-2 2 21.85 -1.17% 6384 38 0.006 34 4
hackbench teo-2 3 21.90 -0.92% 4180 47 0.011 40 7
hackbench teo-2 4 21.78 -1.47% 4482 36 0.008 32 4
hackbench teo-3 0 21.77 -1.52% 8780 33 0.004 25 8
hackbench teo-3 1 21.66 -2.03% 6282 43 0.007 39 4
hackbench teo-3 2 21.75 -1.61% 3590 38 0.011 34 4
hackbench teo-3 3 21.99 -0.51% 5600 36 0.006 33 3
hackbench teo-3 4 21.72 -1.73% 5332 37 0.007 30 7
hackbench teo-4 0 21.82 -1.32% 8232 44 0.005 38 6
hackbench teo-4 1 21.81 -1.34% 4658 38 0.008 32 6
hackbench teo-4 2 21.93 -0.81% 3308 49 0.015 43 6
hackbench teo-4 3 21.69 -1.87% 3438 66 0.019 58 8
hackbench teo-4 4 21.59 -2.31% 3996 42 0.011 41 1
hackbench teo-5 0 21.87 -1.08% 6076 44 0.007 38 6
hackbench teo-5 1 21.73 -1.71% 3694 35 0.009 32 3
hackbench teo-5 2 21.81 -1.35% 7434 45 0.006 40 5
hackbench teo-5 3 21.67 -1.96% 3270 37 0.011 34 3
hackbench teo-5 4 21.80 -1.41% 8316 43 0.005 39 4
test gov i score %change idles idle_miss miss_rt belows aboves
schbench menu 0 277.37 +0.00% 33528 2311 0.069 2303 8
schbench menu 1 272.40 -1.79% 30756 1924 0.063 1912 12
schbench menu 2 293.30 +5.74% 33068 2100 0.064 2090 10
schbench menu 3 290.83 +4.85% 33377 2129 0.064 2108 21
schbench menu 4 294.00 +6.00% 32986 2018 0.061 2011 7
schbench menu 5 291.57 +5.12% 32426 2007 0.062 1997 10
schbench menu 6 298.83 +7.74% 33396 2134 0.064 2118 16
schbench menu 7 298.47 +7.61% 34437 2066 0.060 2055 11
schbench menu 8 270.30 -2.55% 30340 1980 0.065 1968 12
schbench menu 9 293.43 +5.79% 33378 2062 0.062 2047 15
schbench menu 10 293.90 +5.96% 33884 1972 0.058 1956 16
schbench menu 11 289.33 +4.31% 32596 1927 0.059 1912 15
schbench menu 12 290.50 +4.73% 32252 1880 0.058 1870 10
schbench menu 13 292.73 +5.54% 33240 2001 0.060 1987 14
schbench menu 14 293.03 +5.65% 33122 1941 0.059 1927 14
schbench menu 15 295.90 +6.68% 33996 1846 0.054 1835 11
schbench menu 16 296.80 +7.01% 34566 2077 0.060 2068 9
schbench menu 17 274.37 -1.08% 30586 1807 0.059 1792 15
schbench menu 18 282.30 +1.78% 30962 1909 0.062 1901 8
schbench menu 19 289.53 +4.38% 33072 2059 0.062 2043 16
schbench teo 0 292.40 +5.42% 31699 116 0.004 106 10
schbench teo 1 292.13 +5.32% 31110 95 0.003 84 11
schbench teo 2 291.97 +5.26% 30646 68 0.002 62 6
schbench teo 3 291.07 +4.94% 31634 96 0.003 86 10
schbench teo 4 295.17 +6.42% 32520 90 0.003 83 7
schbench teo 5 285.47 +2.92% 30607 74 0.002 69 5
schbench teo 6 274.70 -0.96% 29076 95 0.003 77 18
schbench teo 7 287.90 +3.80% 32074 103 0.003 93 10
schbench teo 8 290.13 +4.60% 29912 72 0.002 66 6
schbench teo 9 289.13 +4.24% 31542 92 0.003 78 14
schbench teo 10 289.27 +4.29% 31482 111 0.004 104 7
schbench teo 11 288.53 +4.02% 30994 92 0.003 85 7
schbench teo 12 288.17 +3.89% 31202 124 0.004 115 9
schbench teo 13 286.57 +3.32% 30096 94 0.003 85 9
schbench teo 14 288.63 +4.06% 31063 86 0.003 74 12
schbench teo 15 286.60 +3.33% 30734 75 0.002 69 6
schbench teo 16 291.93 +5.25% 30276 98 0.003 94 4
schbench teo 17 289.50 +4.37% 30496 104 0.003 96 8
schbench teo 18 292.23 +5.36% 30792 111 0.004 95 16
schbench teo 19 289.37 +4.33% 31606 129 0.004 116 13
schbench teo-1 0 288.73 +4.10% 30062 81 0.003 71 10
schbench teo-1 1 289.30 +4.30% 29604 75 0.003 71 4
schbench teo-1 2 285.97 +3.10% 29324 96 0.003 90 6
schbench teo-1 3 292.00 +5.27% 30118 87 0.003 82 5
schbench teo-1 4 291.23 +5.00% 31044 103 0.003 90 13
schbench teo-1 5 285.27 +2.85% 30554 88 0.003 84 4
schbench teo-1 6 290.77 +4.83% 31770 90 0.003 83 7
schbench teo-1 7 287.53 +3.66% 30520 74 0.002 68 6
schbench teo-1 8 289.53 +4.38% 30954 99 0.003 90 9
schbench teo-1 9 291.03 +4.92% 31310 84 0.003 76 8
schbench teo-1 10 272.43 -1.78% 29166 89 0.003 79 10
schbench teo-1 11 284.13 +2.44% 31165 96 0.003 89 7
schbench teo-1 12 289.97 +4.54% 29576 102 0.003 92 10
schbench teo-1 13 290.63 +4.78% 30486 94 0.003 85 9
schbench teo-1 14 290.67 +4.80% 30214 91 0.003 85 6
schbench teo-1 15 292.10 +5.31% 30026 91 0.003 85 6
schbench teo-1 16 294.53 +6.19% 31184 121 0.004 113 8
schbench teo-1 17 274.87 -0.90% 29328 88 0.003 80 8
schbench teo-1 18 283.23 +2.11% 29386 98 0.003 93 5
schbench teo-1 19 288.30 +3.94% 31274 105 0.003 96 9
schbench teo-2 0 288.60 +4.05% 30364 109 0.004 99 10
schbench teo-2 1 290.17 +4.61% 31496 109 0.003 103 6
schbench teo-2 2 286.87 +3.43% 31062 111 0.004 106 5
schbench teo-2 3 285.27 +2.85% 30744 125 0.004 110 15
schbench teo-2 4 290.13 +4.60% 30512 103 0.003 92 11
schbench teo-2 5 287.17 +3.53% 29996 106 0.004 95 11
schbench teo-2 6 287.97 +3.82% 29600 113 0.004 101 12
schbench teo-2 7 288.87 +4.15% 30454 80 0.003 70 10
schbench teo-2 8 290.40 +4.70% 31316 138 0.004 124 14
schbench teo-2 9 287.97 +3.82% 31166 135 0.004 125 10
schbench teo-2 10 287.37 +3.61% 31736 140 0.004 130 10
schbench teo-2 11 287.57 +3.68% 31182 153 0.005 137 16
schbench teo-2 12 290.70 +4.81% 30898 159 0.005 145 14
schbench teo-2 13 290.43 +4.71% 30684 117 0.004 106 11
schbench teo-2 14 287.60 +3.69% 30302 119 0.004 108 11
schbench teo-2 15 287.30 +3.58% 30984 132 0.004 122 10
schbench teo-2 16 269.17 -2.96% 28464 121 0.004 108 13
schbench teo-2 17 290.40 +4.70% 31798 139 0.004 122 17
schbench teo-2 18 286.57 +3.32% 30548 126 0.004 117 9
schbench teo-2 19 287.10 +3.51% 30496 108 0.004 97 11
schbench teo-3 0 288.87 +4.15% 30742 138 0.004 121 17
schbench teo-3 1 289.60 +4.41% 30560 100 0.003 90 10
schbench teo-3 2 289.30 +4.30% 30614 134 0.004 125 9
schbench teo-3 3 292.43 +5.43% 30408 115 0.004 105 10
schbench teo-3 4 292.33 +5.39% 30708 118 0.004 104 14
schbench teo-3 5 289.53 +4.38% 30916 132 0.004 126 6
schbench teo-3 6 287.40 +3.62% 30930 133 0.004 123 10
schbench teo-3 7 285.03 +2.76% 29328 111 0.004 103 8
schbench teo-3 8 292.17 +5.34% 29776 115 0.004 103 12
schbench teo-3 9 279.57 +0.79% 28564 139 0.005 129 10
schbench teo-3 10 287.87 +3.79% 31060 117 0.004 108 9
schbench teo-3 11 289.93 +4.53% 30844 122 0.004 111 11
schbench teo-3 12 290.30 +4.66% 30544 125 0.004 116 9
schbench teo-3 13 288.57 +4.04% 30566 125 0.004 110 15
schbench teo-3 14 287.17 +3.53% 30057 132 0.004 119 13
schbench teo-3 15 291.03 +4.92% 31116 127 0.004 114 13
schbench teo-3 16 287.97 +3.82% 31431 125 0.004 111 14
schbench teo-3 17 289.70 +4.45% 30148 112 0.004 104 8
schbench teo-3 18 286.73 +3.37% 30550 113 0.004 106 7
schbench teo-3 19 286.87 +3.43% 31357 140 0.004 130 10
schbench teo-4 0 287.23 +3.55% 30916 796 0.026 776 20
schbench teo-4 1 287.30 +3.58% 31040 762 0.025 748 14
schbench teo-4 2 288.70 +4.08% 31566 863 0.027 837 26
schbench teo-4 3 290.13 +4.60% 30910 859 0.028 840 19
schbench teo-4 4 288.90 +4.16% 31452 788 0.025 775 13
schbench teo-4 5 292.73 +5.54% 31402 831 0.026 810 21
schbench teo-4 6 286.87 +3.43% 30786 827 0.027 807 20
schbench teo-4 7 291.67 +5.16% 31396 973 0.031 949 24
schbench teo-4 8 279.83 +0.89% 30938 827 0.027 803 24
schbench teo-4 9 278.40 +0.37% 27162 1240 0.046 1227 13
schbench teo-4 10 267.73 -3.48% 28802 852 0.030 840 12
schbench teo-4 11 286.93 +3.45% 30738 776 0.025 761 15
schbench teo-4 12 289.70 +4.45% 31094 891 0.029 866 25
schbench teo-4 13 287.77 +3.75% 31064 833 0.027 809 24
schbench teo-4 14 288.80 +4.12% 31202 898 0.029 877 21
schbench teo-4 15 288.97 +4.18% 30724 995 0.032 972 23
schbench teo-4 16 289.10 +4.23% 30926 810 0.026 791 19
schbench teo-4 17 288.83 +4.13% 31826 754 0.024 732 22
schbench teo-4 18 288.73 +4.10% 31268 946 0.030 925 21
schbench teo-4 19 290.93 +4.89% 30764 798 0.026 778 20
schbench teo-5 0 288.40 +3.98% 30134 1107 0.037 1084 23
schbench teo-5 1 290.60 +4.77% 31304 884 0.028 865 19
schbench teo-5 2 288.87 +4.15% 30476 774 0.025 759 15
schbench teo-5 3 289.10 +4.23% 30486 1169 0.038 1150 19
schbench teo-5 4 285.97 +3.10% 31178 981 0.031 958 23
schbench teo-5 5 291.40 +5.06% 31016 968 0.031 943 25
schbench teo-5 6 289.17 +4.25% 30848 876 0.028 858 18
schbench teo-5 7 291.00 +4.91% 31077 883 0.028 864 19
schbench teo-5 8 289.13 +4.24% 31186 907 0.029 884 23
schbench teo-5 9 287.70 +3.72% 31468 817 0.026 794 23
schbench teo-5 10 287.20 +3.54% 31160 804 0.026 785 19
schbench teo-5 11 287.53 +3.66% 30165 747 0.025 731 16
schbench teo-5 12 291.70 +5.17% 31356 857 0.027 840 17
schbench teo-5 13 289.87 +4.51% 31762 834 0.026 816 18
schbench teo-5 14 288.17 +3.89% 30582 901 0.029 888 13
schbench teo-5 15 293.67 +5.88% 31538 919 0.029 899 20
schbench teo-5 16 296.60 +6.93% 32620 894 0.027 875 19
schbench teo-5 17 268.13 -3.33% 28104 906 0.032 885 21
schbench teo-5 18 289.23 +4.28% 30218 1165 0.039 1139 26
schbench teo-5 19 293.93 +5.97% 30578 1143 0.037 1121 22
device gov iter iops idles idle_misses idle_miss_ratio belows aboves
mapper/dm-slow menu 0 115 50130 1937 0.039 1935 2
mapper/dm-slow menu 1 115 47448 1591 0.034 1589 2
mapper/dm-slow menu 2 115 47570 1624 0.034 1622 2
mapper/dm-slow menu 3 115 47478 1606 0.034 1606 0
mapper/dm-slow menu 4 115 46498 1523 0.033 1522 1
mapper/dm-slow menu 5 115 45834 1462 0.032 1453 9
mapper/dm-slow menu 6 115 46984 1629 0.035 1628 1
mapper/dm-slow menu 7 115 47710 1599 0.034 1593 6
mapper/dm-slow menu 8 115 44688 1388 0.031 1387 1
mapper/dm-slow menu 9 115 47720 1608 0.034 1605 3
mapper/dm-slow teo 0 115 44442 876 0.020 864 12
mapper/dm-slow teo 1 115 46884 1095 0.023 1070 25
mapper/dm-slow teo 2 115 45762 1010 0.022 957 53
mapper/dm-slow teo 3 115 48768 1102 0.023 1078 24
mapper/dm-slow teo 4 115 45948 1028 0.022 1017 11
mapper/dm-slow teo 5 115 47964 1054 0.022 1029 25
mapper/dm-slow teo 6 115 47280 936 0.020 919 17
mapper/dm-slow teo 7 115 46464 968 0.021 950 18
mapper/dm-slow teo 8 115 46342 1035 0.022 1025 10
mapper/dm-slow teo 9 115 41460 813 0.020 794 19
mapper/dm-slow teo-1 0 115 49218 1119 0.023 1101 18
mapper/dm-slow teo-1 1 115 44854 895 0.020 882 13
mapper/dm-slow teo-1 2 115 47572 1019 0.021 1004 15
mapper/dm-slow teo-1 3 115 48212 1031 0.021 1016 15
mapper/dm-slow teo-1 4 115 46284 955 0.021 944 11
mapper/dm-slow teo-1 5 115 50868 1099 0.022 1091 8
mapper/dm-slow teo-1 6 115 47966 1019 0.021 1003 16
mapper/dm-slow teo-1 7 115 47624 991 0.021 934 57
mapper/dm-slow teo-1 8 115 46708 1002 0.021 995 7
mapper/dm-slow teo-1 9 115 45328 881 0.019 863 18
mapper/dm-slow teo-2 0 115 46562 1071 0.023 1049 22
mapper/dm-slow teo-2 1 115 47114 1102 0.023 1048 54
mapper/dm-slow teo-2 2 115 46710 1075 0.023 1057 18
mapper/dm-slow teo-2 3 115 47156 1134 0.024 1120 14
mapper/dm-slow teo-2 4 115 46650 1007 0.022 988 19
mapper/dm-slow teo-2 5 115 46302 1055 0.023 1030 25
mapper/dm-slow teo-2 6 115 47572 1105 0.023 1087 18
mapper/dm-slow teo-2 7 115 44404 1017 0.023 1000 17
mapper/dm-slow teo-2 8 115 46516 1113 0.024 1097 16
mapper/dm-slow teo-2 9 115 46624 1062 0.023 1042 20
mapper/dm-slow teo-3 0 115 46596 1070 0.023 1052 18
mapper/dm-slow teo-3 1 115 47490 1076 0.023 1057 19
mapper/dm-slow teo-3 2 115 46808 1101 0.024 1035 66
mapper/dm-slow teo-3 3 115 47366 1082 0.023 1069 13
mapper/dm-slow teo-3 4 115 46384 1148 0.025 1133 15
mapper/dm-slow teo-3 5 115 45712 986 0.022 978 8
mapper/dm-slow teo-3 6 115 43916 983 0.022 972 11
mapper/dm-slow teo-3 7 115 45798 1078 0.024 1062 16
mapper/dm-slow teo-3 8 115 46104 1016 0.022 1007 9
mapper/dm-slow teo-3 9 115 44144 940 0.021 923 17
mapper/dm-slow teo-4 0 115 46186 1571 0.034 1564 7
mapper/dm-slow teo-4 1 115 46560 1640 0.035 1633 7
mapper/dm-slow teo-4 2 115 46776 1611 0.034 1606 5
mapper/dm-slow teo-4 3 115 46736 1638 0.035 1633 5
mapper/dm-slow teo-4 4 115 47370 1659 0.035 1653 6
mapper/dm-slow teo-4 5 115 47034 1654 0.035 1645 9
mapper/dm-slow teo-4 6 115 46268 1610 0.035 1601 9
mapper/dm-slow teo-4 7 115 46778 1612 0.034 1606 6
mapper/dm-slow teo-4 8 115 46914 1647 0.035 1639 8
mapper/dm-slow teo-4 9 115 44730 1429 0.032 1422 7
mapper/dm-slow teo-5 0 115 46884 1581 0.034 1573 8
mapper/dm-slow teo-5 1 115 46968 1677 0.036 1670 7
mapper/dm-slow teo-5 2 115 43722 1395 0.032 1384 11
mapper/dm-slow teo-5 3 115 43374 1326 0.031 1319 7
mapper/dm-slow teo-5 4 115 47308 1674 0.035 1665 9
mapper/dm-slow teo-5 5 115 45372 1512 0.033 1502 10
mapper/dm-slow teo-5 6 115 46778 1639 0.035 1629 10
mapper/dm-slow teo-5 7 115 46864 1668 0.036 1662 6
mapper/dm-slow teo-5 8 115 46926 1689 0.036 1680 9
mapper/dm-slow teo-5 9 115 46810 1601 0.034 1593 8
^ permalink raw reply [flat|nested] 21+ messages in thread
* Re: [PATCH v1 0/5] cpuidle: governors: teo: Wakeup events classification change and some refinements
2026-01-16 11:52 ` [PATCH v1 0/5] cpuidle: governors: teo: Wakeup events classification change and some refinements Christian Loehle
@ 2026-01-16 12:29 ` Rafael J. Wysocki
2026-01-19 9:53 ` Christian Loehle
0 siblings, 1 reply; 21+ messages in thread
From: Rafael J. Wysocki @ 2026-01-16 12:29 UTC (permalink / raw)
To: Christian Loehle; +Cc: Rafael J. Wysocki, Linux PM, LKML, Doug Smythies
On Fri, Jan 16, 2026 at 12:52 PM Christian Loehle
<christian.loehle@arm.com> wrote:
>
> On 1/14/26 19:42, Rafael J. Wysocki wrote:
> > Hi All,
> >
> > This material has been in my local queue for almost a full development cycle,
> > so time to post it.
> >
> > The motivation for the changes in this series is mostly theoretical, but I do
> > see some idle power improvements from patch [4/5], for example, but nothing
> > specifically worth reporting.
> >
> > The first patch simply prevents idle states with zero-size bins from being
> > selected sometimes when teo_select() runs with stopped tick.
> >
> > Patch [2/5] avoids counting tick wakeups as intercepts unless there are
> > sufficiently many intercepts within the tick period range to assume that
> > the tick wakeup may have clobbered a genuine intercept.
> >
> > Patch [3/5] simply updates a coefficient in one of the inequalities to be
> > somewhat easier to interpret (this should be a cosmetic change).
> >
> > Patch [4/5] changes the criteria used for classifying wakeup events as hits
> > or intercepts to (hopefully) make the classification work better for large
> > state bins.
> >
> > Patch [5/5] refines the idle state lookup based on intercepts to first
> > consider the state with the maximum intercepts metric, so that state is
> > always taken into consideration.
> >
> > Please see the individual patch changelogs for details.
> >
> > Thanks!
> >
> >
Hi Christian,
> Hi Rafael,
> I'll do the in-depth review, but have run some tests already.
> They are attached, platform is the usual rk3399.
> "teo" is mainline, "teo-$i" is with patches 1..$i applied.
Thanks for testing!
> There's a regression on teo-4 visible on the intercept heavy IO workloads,
> for idle misses that isn't strong enough to reflect in score changes except
> for the very slow mtdblock device.
> interestingly though there also seems to be a regression in
> mapper/dm-slow (dm device with 51ms delay on each IO), which is not
> intercept heavy.
> Looking at the state residencies it overuses the deepest state2 in
> where state1 was preferred for the other teo variants.
> I've attached that too for reference.
> I'm assuming that is because of the new intercept-logic-exclusion clause.
So can you please restore that clause to its previous form, while
keeping the other changes in patch 4, and see if the regression is
still there?
> teo-5 seems to be slightly better than teo-4 here, but still a regression
> from the others.
I see.
^ permalink raw reply [flat|nested] 21+ messages in thread
* Re: [PATCH v1 0/5] cpuidle: governors: teo: Wakeup events classification change and some refinements
2026-01-16 12:29 ` Rafael J. Wysocki
@ 2026-01-19 9:53 ` Christian Loehle
2026-01-19 12:09 ` Rafael J. Wysocki
0 siblings, 1 reply; 21+ messages in thread
From: Christian Loehle @ 2026-01-19 9:53 UTC (permalink / raw)
To: Rafael J. Wysocki; +Cc: Linux PM, LKML, Doug Smythies
[-- Attachment #1: Type: text/plain, Size: 10135 bytes --]
On 1/16/26 12:29, Rafael J. Wysocki wrote:
> On Fri, Jan 16, 2026 at 12:52 PM Christian Loehle
> <christian.loehle@arm.com> wrote:
>>
>> On 1/14/26 19:42, Rafael J. Wysocki wrote:
>>> Hi All,
>>>
>>> This material has been in my local queue for almost a full development cycle,
>>> so time to post it.
>>>
>>> The motivation for the changes in this series is mostly theoretical, but I do
>>> see some idle power improvements from patch [4/5], for example, but nothing
>>> specifically worth reporting.
>>>
>>> The first patch simply prevents idle states with zero-size bins from being
>>> selected sometimes when teo_select() runs with stopped tick.
>>>
>>> Patch [2/5] avoids counting tick wakeups as intercepts unless there are
>>> sufficiently many intercepts within the tick period range to assume that
>>> the tick wakeup may have clobbered a genuine intercept.
>>>
>>> Patch [3/5] simply updates a coefficient in one of the inequalities to be
>>> somewhat easier to interpret (this should be a cosmetic change).
>>>
>>> Patch [4/5] changes the criteria used for classifying wakeup events as hits
>>> or intercepts to (hopefully) make the classification work better for large
>>> state bins.
>>>
>>> Patch [5/5] refines the idle state lookup based on intercepts to first
>>> consider the state with the maximum intercepts metric, so that state is
>>> always taken into consideration.
>>>
>>> Please see the individual patch changelogs for details.
>>>
>>> Thanks!
>>>
>>>
>
> Hi Christian,
>
>> Hi Rafael,
>> I'll do the in-depth review, but have run some tests already.
>> They are attached, platform is the usual rk3399.
>> "teo" is mainline, "teo-$i" is with patches 1..$i applied.
>
> Thanks for testing!
>
>> There's a regression on teo-4 visible on the intercept heavy IO workloads,
>> for idle misses that isn't strong enough to reflect in score changes except
>> for the very slow mtdblock device.
>> interestingly though there also seems to be a regression in
>> mapper/dm-slow (dm device with 51ms delay on each IO), which is not
>> intercept heavy.
>> Looking at the state residencies it overuses the deepest state2 in
>> where state1 was preferred for the other teo variants.
>> I've attached that too for reference.
>> I'm assuming that is because of the new intercept-logic-exclusion clause.
>
> So can you please restore that clause to its previous form, while
> keeping the other changes in patch 4, and see if the regression is
> still there?
>
Yep that restores it, I've attached the full results again with teo-4-p and teo-5-p
being teo-4 and teo-5 but with
- if (2 * idx_intercept_sum > cpu_data->total) {
+ if (2 * idx_intercept_sum > cpu_data->total - idx_hit_sum) {
respectively.
To highlight let me just quote the schbench results
test gov i score %change idles idle_miss miss_rt belows aboves
schbench teo-4 0 295.87 +5.51% 29014 1309 0.045 1168 141
schbench teo-4 1 302.20 +7.76% 31642 1408 0.044 1105 303
schbench teo-4 2 301.07 +7.36% 29966 1602 0.053 1393 209
schbench teo-4 3 297.00 +5.91% 29982 1529 0.051 1266 263
schbench teo-4 4 297.20 +5.98% 29432 1419 0.048 1194 225
schbench teo-4 5 299.47 +6.79% 30262 1381 0.046 1189 192
schbench teo-4 6 301.23 +7.42% 29810 1469 0.049 1234 235
schbench teo-4 7 301.00 +7.34% 29916 1511 0.051 1355 156
schbench teo-4 8 299.17 +6.68% 29246 1444 0.049 1296 148
schbench teo-4 9 297.77 +6.18% 30306 1399 0.046 1195 204
schbench teo-4 10 304.03 +8.42% 30904 1346 0.044 1163 183
schbench teo-4 11 300.73 +7.24% 30716 1276 0.042 1089 187
schbench teo-4 12 298.13 +6.31% 29872 1314 0.044 1081 233
schbench teo-4 13 298.73 +6.53% 30386 1288 0.042 1092 196
schbench teo-4 14 300.77 +7.25% 30150 1323 0.044 1074 249
schbench teo-4 15 296.23 +5.63% 32178 1328 0.041 1061 267
schbench teo-4 16 283.50 +1.09% 27472 1282 0.047 1126 156
schbench teo-4 17 300.53 +7.17% 29644 1552 0.052 1375 177
schbench teo-4 18 302.23 +7.77% 31456 1418 0.045 1194 224
schbench teo-4 19 307.43 +9.63% 32518 1259 0.039 968 291
schbench teo-4-p 0 277.53 -1.03% 28054 503 0.018 151 352
schbench teo-4-p 1 304.07 +8.43% 30650 553 0.018 178 375
schbench teo-4-p 2 273.73 -2.39% 27120 509 0.019 163 346
schbench teo-4-p 3 300.20 +7.05% 29932 528 0.018 171 357
schbench teo-4-p 4 300.00 +6.98% 29580 584 0.020 180 404
schbench teo-4-p 5 296.03 +5.56% 30176 514 0.017 172 342
schbench teo-4-p 6 300.50 +7.16% 29806 550 0.018 174 376
schbench teo-4-p 7 271.07 -3.34% 26934 472 0.018 165 307
schbench teo-4-p 8 299.00 +6.62% 30568 564 0.018 175 389
schbench teo-4-p 9 300.80 +7.26% 29660 482 0.016 188 294
schbench teo-4-p 10 300.57 +7.18% 30384 634 0.021 195 439
schbench teo-4-p 11 296.53 +5.74% 29590 599 0.020 221 378
schbench teo-4-p 12 298.77 +6.54% 30664 599 0.020 177 422
schbench teo-4-p 13 302.47 +7.86% 30200 559 0.019 223 336
schbench teo-4-p 14 300.43 +7.13% 30326 562 0.019 166 396
schbench teo-4-p 15 292.87 +4.44% 28798 524 0.018 142 382
schbench teo-4-p 16 299.63 +6.85% 29718 535 0.018 189 346
schbench teo-4-p 17 302.10 +7.73% 31122 584 0.019 221 363
schbench teo-4-p 18 294.13 +4.89% 28312 560 0.020 167 393
schbench teo-4-p 19 302.20 +7.76% 29872 556 0.019 168 388
schbench teo-5 0 299.53 +6.81% 31198 1312 0.042 1062 250
schbench teo-5 1 296.67 +5.79% 30090 1383 0.046 1169 214
schbench teo-5 2 299.60 +6.84% 30010 1566 0.052 1331 235
schbench teo-5 3 300.00 +6.98% 30764 1466 0.048 1239 227
schbench teo-5 4 298.77 +6.54% 30002 1383 0.046 1212 171
schbench teo-5 5 299.30 +6.73% 29848 1481 0.050 1288 193
schbench teo-5 6 298.90 +6.59% 29586 1441 0.049 1260 181
schbench teo-5 7 300.37 +7.11% 30672 1460 0.048 1306 154
schbench teo-5 8 301.00 +7.34% 31454 1383 0.044 1190 193
schbench teo-5 9 298.80 +6.55% 30042 1415 0.047 1194 221
schbench teo-5 10 302.23 +7.77% 30986 1305 0.042 1066 239
schbench teo-5 11 297.30 +6.02% 30112 1405 0.047 1190 215
schbench teo-5 12 299.33 +6.74% 29536 1482 0.050 1214 268
schbench teo-5 13 288.87 +3.01% 28754 1244 0.043 1038 206
schbench teo-5 14 300.17 +7.04% 30232 1517 0.050 1300 217
schbench teo-5 15 301.43 +7.49% 30578 1576 0.052 1287 289
schbench teo-5 16 282.10 +0.60% 29106 996 0.034 834 162
schbench teo-5 17 278.03 -0.86% 27432 1123 0.041 926 197
schbench teo-5 18 304.40 +8.55% 29934 1526 0.051 1296 230
schbench teo-5 19 276.50 -1.40% 28072 1158 0.041 970 188
schbench teo-5-p 0 299.47 +6.79% 29994 587 0.020 231 356
schbench teo-5-p 1 298.27 +6.36% 31600 555 0.018 190 365
schbench teo-5-p 2 297.20 +5.98% 30350 543 0.018 198 345
schbench teo-5-p 3 294.30 +4.95% 29962 523 0.017 192 331
schbench teo-5-p 4 303.30 +8.16% 30366 480 0.016 160 320
schbench teo-5-p 5 298.87 +6.58% 29838 541 0.018 180 361
schbench teo-5-p 6 301.63 +7.56% 30022 480 0.016 168 312
schbench teo-5-p 7 298.47 +6.43% 30210 581 0.019 202 379
schbench teo-5-p 8 298.77 +6.54% 30636 529 0.017 210 319
schbench teo-5-p 9 298.63 +6.49% 30474 564 0.019 185 379
schbench teo-5-p 10 300.10 +7.01% 30808 577 0.019 251 326
schbench teo-5-p 11 295.87 +5.51% 30190 565 0.019 186 379
schbench teo-5-p 12 301.27 +7.43% 29958 466 0.016 160 306
schbench teo-5-p 13 280.83 +0.14% 27916 494 0.018 164 330
schbench teo-5-p 14 294.23 +4.92% 30330 611 0.020 219 392
schbench teo-5-p 15 297.13 +5.96% 30728 656 0.021 195 461
schbench teo-5-p 16 297.97 +6.25% 30242 524 0.017 178 346
schbench teo-5-p 17 301.27 +7.43% 31004 524 0.017 208 316
schbench teo-5-p 18 297.97 +6.25% 29864 565 0.019 211 354
schbench teo-5-p 19 298.43 +6.42% 30462 519 0.017 150 369
The -p variants have less than half the idle_misses and fall into the same order of magnitude
for this test than all the other teo variants.
[-- Attachment #2: teo-6.19-patches-rafael-wakeup-classification-modified-dm-slow-residencies.txt --]
[-- Type: text/plain, Size: 12695 bytes --]
====================================
menu
====================================
time
idle_state
-1.0 0.519347
0.0 9.805539
1.0 2.856356
2.0 17.605928
time
idle_state
-1.0 0.144220
0.0 0.244613
1.0 0.045671
2.0 30.063597
time
idle_state
-1.0 0.171862
2.0 30.087738
0.0 0.173437
1.0 0.034034
time
idle_state
-1.0 0.108919
0.0 0.227121
1.0 0.039202
2.0 30.073220
time
idle_state
-1.0 0.335498
2.0 29.885951
0.0 0.225368
1.0 0.030971
time
idle_state
-1.0 0.349519
2.0 29.838754
0.0 0.224864
1.0 0.024177
time
idle_state
-1.0 0.281127
2.0 29.892060
0.0 0.174007
1.0 0.024905
time
idle_state
-1.0 0.291298
2.0 29.876345
1.0 0.040021
0.0 0.256267
time
idle_state
-1.0 0.109349
0.0 0.141349
1.0 0.024782
2.0 30.182048
time
idle_state
-1.0 0.489200
0.0 0.186743
2.0 29.671824
1.0 0.031454
====================================
teo
====================================
time
idle_state
-1.0 0.099843
0.0 0.083271
2.0 30.096207
1.0 0.108118
time
idle_state
-1.0 0.497552
0.0 0.561951
1.0 3.821100
2.0 25.589928
time
idle_state
-1.0 0.456734
2.0 26.170341
1.0 3.318869
0.0 0.460043
time
idle_state
-1.0 0.365303
1.0 6.034275
0.0 0.789791
2.0 23.212346
time
idle_state
-1.0 0.247448
1.0 0.465332
0.0 0.140499
2.0 29.542321
time
idle_state
-1.0 0.494740
2.0 29.380491
0.0 0.133455
1.0 0.374272
time
idle_state
-1.0 0.233728
1.0 0.295976
2.0 29.726954
0.0 0.137358
time
idle_state
-1.0 0.138549
0.0 0.102419
1.0 0.167945
2.0 29.969321
time
idle_state
-1.0 0.420523
0.0 0.123331
1.0 0.488867
2.0 29.388603
time
idle_state
-1.0 0.159103
2.0 30.029013
0.0 0.096278
1.0 0.103443
====================================
teo-1
====================================
time
idle_state
-1.0 0.138578
0.0 0.097399
1.0 0.126002
2.0 30.054509
time
idle_state
-1.0 0.110219
2.0 30.102737
0.0 0.091943
1.0 0.094109
time
idle_state
-1.0 0.524526
2.0 18.669748
0.0 1.259111
1.0 9.966566
time
idle_state
-1.0 0.457737
2.0 29.225979
1.0 0.588731
0.0 0.124309
time
idle_state
-1.0 0.574399
1.0 6.287790
0.0 0.848161
2.0 22.672145
time
idle_state
-1.0 0.211731
1.0 0.711729
0.0 0.141975
2.0 29.311058
time
idle_state
-1.0 0.107048
2.0 30.058434
0.0 0.102691
1.0 0.126769
time
idle_state
-1.0 0.112842
2.0 30.136715
0.0 0.090312
1.0 0.062225
time
idle_state
-1.0 0.174177
0.0 0.119666
1.0 0.146481
2.0 29.945314
time
idle_state
-1.0 0.423898
2.0 27.731474
0.0 0.295387
1.0 1.922763
====================================
teo-2
====================================
time
idle_state
-1.0 0.564948
0.0 1.247513
1.0 10.059035
2.0 18.531677
time
idle_state
-1.0 0.148360
0.0 0.081269
1.0 0.086353
2.0 30.094240
time
idle_state
-1.0 0.290126
0.0 0.119720
1.0 0.303814
2.0 29.685550
time
idle_state
-1.0 0.108246
0.0 0.094786
1.0 0.138555
2.0 30.078625
time
idle_state
-1.0 0.270287
2.0 24.843992
0.0 0.608318
1.0 4.663533
time
idle_state
-1.0 0.384186
2.0 20.927768
1.0 8.074054
0.0 1.008619
time
idle_state
-1.0 0.106037
0.0 0.058642
1.0 0.042971
2.0 30.176490
time
idle_state
-1.0 0.143492
0.0 0.088795
1.0 0.064343
2.0 30.095477
time
idle_state
-1.0 0.481974
2.0 29.077065
0.0 0.128594
1.0 0.724787
time
idle_state
-1.0 0.351496
0.0 0.855749
1.0 6.761263
2.0 22.416728
====================================
teo-3
====================================
time
idle_state
-1.0 0.380160
2.0 29.588078
0.0 0.111541
1.0 0.327226
time
idle_state
-1.0 0.143215
0.0 0.103571
1.0 0.232043
2.0 29.939322
time
idle_state
-1.0 0.355085
2.0 24.555389
1.0 4.848766
0.0 0.656725
time
idle_state
-1.0 0.130883
2.0 30.073831
0.0 0.094981
1.0 0.097431
time
idle_state
-1.0 0.148319
0.0 0.078862
1.0 0.075049
2.0 30.103286
time
idle_state
-1.0 0.227304
2.0 26.591961
0.0 0.443516
1.0 3.138051
time
idle_state
-1.0 0.170320
2.0 30.038726
0.0 0.083822
1.0 0.095754
time
idle_state
-1.0 0.149993
0.0 0.084013
1.0 0.100105
2.0 30.094225
time
idle_state
-1.0 0.217064
0.0 0.378949
1.0 2.707453
2.0 27.088495
time
idle_state
-1.0 0.385561
2.0 20.580747
0.0 1.034470
1.0 8.399299
====================================
teo-4
====================================
time
idle_state
-1.0 0.664030
2.0 21.272839
0.0 0.986977
1.0 7.473258
time
idle_state
-1.0 0.168210
0.0 0.098375
1.0 0.144143
2.0 29.995653
time
idle_state
-1.0 0.455400
2.0 17.872049
0.0 1.284199
1.0 10.793430
time
idle_state
-1.0 0.161471
2.0 30.042443
0.0 0.089536
1.0 0.108051
time
idle_state
-1.0 0.168592
2.0 30.032260
0.0 0.089051
1.0 0.102059
time
idle_state
-1.0 0.500834
0.0 1.285904
1.0 10.480839
2.0 18.158708
time
idle_state
-1.0 0.119650
2.0 30.083933
1.0 0.101195
0.0 0.085805
time
idle_state
-1.0 0.095695
0.0 0.057199
1.0 0.069387
2.0 30.189172
time
idle_state
-1.0 0.538868
0.0 0.906345
1.0 7.144794
2.0 21.820099
time
idle_state
-1.0 0.663265
2.0 22.466025
0.0 0.856007
1.0 6.426644
====================================
teo-4-p
====================================
====================================
time
idle_state
-1.0 0.434586
0.0 0.637993
1.0 4.554776
2.0 24.785758
time
idle_state
-1.0 0.262746
0.0 0.096834
1.0 0.195826
2.0 29.886917
time
idle_state
-1.0 0.454211
0.0 0.098639
1.0 0.351944
2.0 29.501359
time
idle_state
-1.0 0.672277
2.0 20.776389
0.0 1.001396
1.0 7.937021
time
idle_state
-1.0 0.208609
2.0 27.680218
0.0 0.333571
1.0 2.199513
time
idle_state
-1.0 0.130159
2.0 30.109982
0.0 0.090787
1.0 0.069754
time
idle_state
-1.0 0.160358
0.0 0.078899
1.0 0.099888
2.0 30.085066
time
idle_state
-1.0 0.099877
0.0 0.067514
1.0 0.059376
2.0 30.186329
time
idle_state
-1.0 0.430748
2.0 28.961636
0.0 0.159454
1.0 0.848048
time
idle_state
-1.0 0.304236
2.0 29.821079
0.0 0.091710
1.0 0.186979
====================================
teo-5
====================================
time
idle_state
-1.0 0.147005
0.0 0.084477
1.0 0.091733
2.0 30.093757
time
idle_state
-1.0 0.328327
2.0 29.608717
0.0 0.097709
1.0 0.403149
time
idle_state
-1.0 0.112502
2.0 30.165993
0.0 0.056966
1.0 0.080292
time
idle_state
-1.0 0.212418
0.0 0.078549
1.0 0.138208
2.0 29.988083
time
idle_state
-1.0 0.248723
2.0 25.158518
0.0 0.562121
1.0 4.445537
time
idle_state
-1.0 0.496222
2.0 27.094540
0.0 0.330054
1.0 2.474676
time
idle_state
-1.0 0.360713
0.0 0.959195
1.0 7.631812
2.0 21.455507
time
idle_state
-1.0 0.549453
0.0 1.092732
1.0 8.560580
2.0 20.206515
time
idle_state
-1.0 0.315927
0.0 0.076242
1.0 0.245703
2.0 29.767137
time
idle_state
-1.0 0.390531
2.0 20.664143
0.0 1.021027
1.0 8.352085
====================================
teo-5-p
====================================
====================================
time
idle_state
-1.0 0.519554
2.0 25.601354
0.0 0.503926
1.0 3.771415
time
idle_state
-1.0 0.173239
2.0 30.034349
0.0 0.096779
1.0 0.103212
time
idle_state
-1.0 0.446027
2.0 29.405944
0.0 0.132321
1.0 0.406744
time
idle_state
-1.0 0.482720
0.0 0.130407
1.0 0.469641
2.0 29.329485
time
idle_state
-1.0 0.168540
2.0 30.037565
0.0 0.090790
1.0 0.108998
time
idle_state
-1.0 0.584965
2.0 24.630357
0.0 0.601880
1.0 4.585128
time
idle_state
-1.0 0.284330
2.0 25.979438
0.0 0.481155
1.0 3.665018
time
idle_state
-1.0 0.514197
2.0 23.254127
0.0 0.755923
1.0 5.882899
time
idle_state
-1.0 0.582706
0.0 1.501510
1.0 12.469665
2.0 15.860456
time
idle_state
-1.0 0.145340
0.0 0.182204
1.0 0.914773
2.0 29.146813
[-- Attachment #3: teo-6.19-patches-rafael-wakeup-classification-modified-results.txt --]
[-- Type: text/plain, Size: 117912 bytes --]
device gov iter iops idles idle_miss ratio belows aboves
mmcblk1 menu 0 1443 376496 116621 0.310 95635 20986
mmcblk1 menu 1 1438 382528 118835 0.311 98062 20773
mmcblk1 menu 2 1431 373586 115483 0.309 96031 19452
mmcblk1 menu 3 1436 373868 114351 0.306 94831 19520
mmcblk1 menu 4 1430 373626 115272 0.309 95649 19623
mmcblk1 menu 5 1427 370486 114015 0.308 95165 18850
mmcblk1 menu 6 1424 365030 112342 0.308 94495 17847
mmcblk1 menu 7 1499 388818 121750 0.313 101268 20482
mmcblk1 menu 8 1433 373844 115440 0.309 95672 19768
mmcblk1 menu 9 1438 371774 114616 0.308 95068 19548
mmcblk1 teo 0 2329 611902 2938 0.005 2773 165
mmcblk1 teo 1 2353 583918 752 0.001 557 195
mmcblk1 teo 2 2332 610432 2961 0.005 2690 271
mmcblk1 teo 3 2328 610808 2767 0.005 2660 107
mmcblk1 teo 4 2329 609654 2954 0.005 2712 242
mmcblk1 teo 5 2352 617344 2934 0.005 2786 148
mmcblk1 teo 6 2323 577002 2713 0.005 2585 128
mmcblk1 teo 7 2336 601854 2199 0.004 2040 159
mmcblk1 teo 8 2353 586666 849 0.001 711 138
mmcblk1 teo 9 2354 617866 2920 0.005 2752 168
mmcblk1 teo-1 0 2331 610646 2783 0.005 2590 193
mmcblk1 teo-1 1 2325 590450 2698 0.005 2607 91
mmcblk1 teo-1 2 2353 617896 2833 0.005 2744 89
mmcblk1 teo-1 3 2327 601668 2807 0.005 2666 141
mmcblk1 teo-1 4 2354 613074 2387 0.004 2218 169
mmcblk1 teo-1 5 2329 610208 2770 0.005 2663 107
mmcblk1 teo-1 6 2329 607000 570 0.001 460 110
mmcblk1 teo-1 7 2329 612059 2850 0.005 2742 108
mmcblk1 teo-1 8 2354 617648 2899 0.005 2760 139
mmcblk1 teo-1 9 2348 615826 3040 0.005 2795 245
mmcblk1 teo-2 0 2330 609580 2879 0.005 2657 222
mmcblk1 teo-2 1 2329 612158 2884 0.005 2772 112
mmcblk1 teo-2 2 2328 612214 2835 0.005 2747 88
mmcblk1 teo-2 3 2329 608448 2863 0.005 2693 170
mmcblk1 teo-2 4 2352 617198 2905 0.005 2784 121
mmcblk1 teo-2 5 2331 610861 2724 0.004 2643 81
mmcblk1 teo-2 6 2345 615072 2830 0.005 2765 65
mmcblk1 teo-2 7 2330 609060 2562 0.004 2429 133
mmcblk1 teo-2 8 2329 608388 1072 0.002 918 154
mmcblk1 teo-2 9 2344 614366 2317 0.004 2214 103
mmcblk1 teo-3 0 2355 605962 2160 0.004 2047 113
mmcblk1 teo-3 1 2330 605992 2435 0.004 2357 78
mmcblk1 teo-3 2 2328 599208 2024 0.003 1954 70
mmcblk1 teo-3 3 2330 610964 2866 0.005 2713 153
mmcblk1 teo-3 4 2353 615902 1681 0.003 1581 100
mmcblk1 teo-3 5 2347 612289 2393 0.004 2289 104
mmcblk1 teo-3 6 2354 618096 2845 0.005 2772 73
mmcblk1 teo-3 7 2354 617042 2788 0.005 2702 86
mmcblk1 teo-3 8 2354 615592 2824 0.005 2695 129
mmcblk1 teo-3 9 2353 617606 2919 0.005 2803 116
mmcblk1 teo-4 0 2328 601128 2994 0.005 2837 157
mmcblk1 teo-4 1 2327 607894 2701 0.004 2572 129
mmcblk1 teo-4 2 2324 593894 2845 0.005 2729 116
mmcblk1 teo-4 3 2327 610722 3032 0.005 2939 93
mmcblk1 teo-4 4 2327 610796 2877 0.005 2764 113
mmcblk1 teo-4 5 2328 610184 2849 0.005 2748 101
mmcblk1 teo-4 6 2353 616630 2695 0.004 2652 43
mmcblk1 teo-4 7 2331 612432 2813 0.005 2775 38
mmcblk1 teo-4 8 2327 611018 3006 0.005 2872 134
mmcblk1 teo-4 9 2328 608176 2524 0.004 2421 103
mmcblk1 teo-4-p 0 2325 601110 2882 0.005 2752 130
mmcblk1 teo-4-p 1 2328 603308 2321 0.004 2186 135
mmcblk1 teo-4-p 2 2328 608972 2641 0.004 2544 97
mmcblk1 teo-4-p 3 2328 610876 2807 0.005 2699 108
mmcblk1 teo-4-p 4 2329 610488 2767 0.005 2673 94
mmcblk1 teo-4-p 5 2328 611694 2921 0.005 2747 174
mmcblk1 teo-4-p 6 2328 606892 2825 0.005 2573 252
mmcblk1 teo-4-p 7 2328 609104 2593 0.004 2532 61
mmcblk1 teo-4-p 8 2328 600757 2181 0.004 2030 151
mmcblk1 teo-4-p 9 2327 611572 2890 0.005 2764 126
mmcblk1 teo-5 0 2325 608952 2865 0.005 2773 92
mmcblk1 teo-5 1 2327 611764 2786 0.005 2760 26
mmcblk1 teo-5 2 2327 605612 922 0.002 816 106
mmcblk1 teo-5 3 2352 615110 2641 0.004 2569 72
mmcblk1 teo-5 4 2342 614792 2832 0.005 2808 24
mmcblk1 teo-5 5 2330 595648 1634 0.003 1593 41
mmcblk1 teo-5 6 2336 609844 2753 0.005 2687 66
mmcblk1 teo-5 7 2328 605354 2374 0.004 2343 31
mmcblk1 teo-5 8 2327 609176 2993 0.005 2806 187
mmcblk1 teo-5 9 2328 604798 2475 0.004 2409 66
mmcblk1 teo-5-p 0 2324 582690 2624 0.005 2550 74
mmcblk1 teo-5-p 1 2352 617580 2837 0.005 2753 84
mmcblk1 teo-5-p 2 2329 608834 2813 0.005 2651 162
mmcblk1 teo-5-p 3 2327 610390 2843 0.005 2748 95
mmcblk1 teo-5-p 4 2327 610572 2827 0.005 2754 73
mmcblk1 teo-5-p 5 2337 609930 2647 0.004 2530 117
mmcblk1 teo-5-p 6 2327 611742 2871 0.005 2752 119
mmcblk1 teo-5-p 7 2354 583804 630 0.001 517 113
mmcblk1 teo-5-p 8 2356 618362 2892 0.005 2775 117
mmcblk1 teo-5-p 9 2330 611336 2840 0.005 2677 163
mmcblk2 menu 0 3051 456810 139262 0.305 118947 20315
mmcblk2 menu 1 3057 453379 137155 0.303 118251 18904
mmcblk2 menu 2 3169 485506 180571 0.372 160169 20402
mmcblk2 menu 3 3187 462152 129591 0.280 111125 18466
mmcblk2 menu 4 3228 419740 115469 0.275 97028 18441
mmcblk2 menu 5 2999 478520 150868 0.315 129788 21080
mmcblk2 menu 6 3298 454910 127772 0.281 106547 21225
mmcblk2 menu 7 3319 464330 129251 0.278 110840 18411
mmcblk2 menu 8 3308 439864 119087 0.271 98808 20279
mmcblk2 menu 9 3105 434734 127504 0.293 108498 19006
mmcblk2 teo 0 5701 792730 2659 0.003 2546 113
mmcblk2 teo 1 5702 846552 3753 0.004 3460 293
mmcblk2 teo 2 5693 784536 3814 0.005 3522 292
mmcblk2 teo 3 5716 860000 2210 0.003 2099 111
mmcblk2 teo 4 5661 798828 6918 0.009 6746 172
mmcblk2 teo 5 5729 806836 1144 0.001 969 175
mmcblk2 teo 6 5694 848008 1563 0.002 1444 119
mmcblk2 teo 7 5682 714696 3858 0.005 3648 210
mmcblk2 teo 8 5710 779412 3351 0.004 3218 133
mmcblk2 teo 9 5683 862370 3538 0.004 3396 142
mmcblk2 teo-1 0 5695 766808 3285 0.004 3189 96
mmcblk2 teo-1 1 5692 732834 2782 0.004 2693 89
mmcblk2 teo-1 2 5662 740156 5546 0.007 5419 127
mmcblk2 teo-1 3 5689 684176 2497 0.004 2387 110
mmcblk2 teo-1 4 5897 895554 4937 0.006 4726 211
mmcblk2 teo-1 5 5685 748122 3239 0.004 3045 194
mmcblk2 teo-1 6 5693 785466 3396 0.004 3295 101
mmcblk2 teo-1 7 5694 787166 2646 0.003 2469 177
mmcblk2 teo-1 8 5705 804006 1921 0.002 1811 110
mmcblk2 teo-1 9 5711 859884 1844 0.002 1754 90
mmcblk2 teo-2 0 5795 729662 3757 0.005 3660 97
mmcblk2 teo-2 1 5708 849006 2106 0.002 1929 177
mmcblk2 teo-2 2 5698 752708 3821 0.005 3634 187
mmcblk2 teo-2 3 5708 868914 2870 0.003 2751 119
mmcblk2 teo-2 4 5676 614316 3453 0.006 3371 82
mmcblk2 teo-2 5 5709 853710 2350 0.003 2234 116
mmcblk2 teo-2 6 5881 871322 3345 0.004 3256 89
mmcblk2 teo-2 7 5687 642576 2688 0.004 2614 74
mmcblk2 teo-2 8 5696 802124 5997 0.007 5859 138
mmcblk2 teo-2 9 5656 511244 3612 0.007 3472 140
mmcblk2 teo-3 0 5885 852004 2141 0.003 2039 102
mmcblk2 teo-3 1 5872 863788 3324 0.004 3211 113
mmcblk2 teo-3 2 5663 678382 3703 0.005 3593 110
mmcblk2 teo-3 3 5690 823314 3470 0.004 3382 88
mmcblk2 teo-3 4 5715 859068 2524 0.003 2378 146
mmcblk2 teo-3 5 5656 719636 6441 0.009 6274 167
mmcblk2 teo-3 6 5664 658722 5915 0.009 5812 103
mmcblk2 teo-3 7 5835 828394 1711 0.002 1588 123
mmcblk2 teo-3 8 5907 855572 899 0.001 802 97
mmcblk2 teo-3 9 5707 810014 2700 0.003 2629 71
mmcblk2 teo-4 0 5671 699652 4059 0.006 4024 35
mmcblk2 teo-4 1 5663 796996 5626 0.007 5468 158
mmcblk2 teo-4 2 5644 774476 6845 0.009 6772 73
mmcblk2 teo-4 3 5681 674864 2380 0.004 2304 76
mmcblk2 teo-4 4 5889 871780 2445 0.003 2339 106
mmcblk2 teo-4 5 5909 854778 176 0.000 111 65
mmcblk2 teo-4 6 5714 777034 939 0.001 774 165
mmcblk2 teo-4 7 5616 772408 10097 0.013 10034 63
mmcblk2 teo-4 8 5622 558203 6542 0.012 6505 37
mmcblk2 teo-4 9 5661 850298 7006 0.008 6889 117
mmcblk2 teo-4-p 0 5712 863390 2497 0.003 2354 143
mmcblk2 teo-4-p 1 5698 864548 1715 0.002 1632 83
mmcblk2 teo-4-p 2 5714 863730 2677 0.003 2565 112
mmcblk2 teo-4-p 3 5714 866830 2962 0.003 2798 164
mmcblk2 teo-4-p 4 5713 837344 792 0.001 622 170
mmcblk2 teo-4-p 5 5698 849502 3072 0.004 2944 128
mmcblk2 teo-4-p 6 5690 852502 3450 0.004 3354 96
mmcblk2 teo-4-p 7 5664 633928 3672 0.006 3555 117
mmcblk2 teo-4-p 8 5908 856518 1095 0.001 941 154
mmcblk2 teo-4-p 9 5694 824166 2407 0.003 2227 180
mmcblk2 teo-5 0 5655 637798 4729 0.007 4678 51
mmcblk2 teo-5 1 5889 894962 3988 0.004 3903 85
mmcblk2 teo-5 2 5674 855046 4560 0.005 4500 60
mmcblk2 teo-5 3 5649 737462 4844 0.007 4806 38
mmcblk2 teo-5 4 5466 781282 24597 0.031 24467 130
mmcblk2 teo-5 5 5648 641718 5754 0.009 5568 186
mmcblk2 teo-5 6 5640 748816 7131 0.010 7061 70
mmcblk2 teo-5 7 5428 780041 28193 0.036 28115 78
mmcblk2 teo-5 8 5664 743116 3059 0.004 2998 61
mmcblk2 teo-5 9 5696 831331 1232 0.001 1173 59
mmcblk2 teo-5-p 0 5907 895032 2768 0.003 2618 150
mmcblk2 teo-5-p 1 5666 609912 3687 0.006 3498 189
mmcblk2 teo-5-p 2 5678 652393 2881 0.004 2752 129
mmcblk2 teo-5-p 3 5716 871564 2891 0.003 2791 100
mmcblk2 teo-5-p 4 5707 783458 2871 0.004 2781 90
mmcblk2 teo-5-p 5 5720 867670 3002 0.003 2851 151
mmcblk2 teo-5-p 6 5686 748360 3502 0.005 3390 112
mmcblk2 teo-5-p 7 5688 709086 1980 0.003 1826 154
mmcblk2 teo-5-p 8 5655 708258 6099 0.009 6007 92
mmcblk2 teo-5-p 9 5717 864870 2547 0.003 2403 144
nvme0n1 menu 0 5026 402866 71270 0.177 52222 19048
nvme0n1 menu 1 5552 443098 74527 0.168 53968 20559
nvme0n1 menu 2 5260 425112 75635 0.178 55284 20351
nvme0n1 menu 3 8787 635368 92057 0.145 71556 20501
nvme0n1 menu 4 5464 434588 75555 0.174 56619 18936
nvme0n1 menu 5 5534 437722 73660 0.168 54405 19255
nvme0n1 menu 6 5518 429864 71965 0.167 53669 18296
nvme0n1 menu 7 6802 519490 90524 0.174 70684 19840
nvme0n1 menu 8 6265 485546 80299 0.165 60456 19843
nvme0n1 menu 9 5467 434256 76099 0.175 57203 18896
nvme0n1 teo 0 10585 750894 2334 0.003 2048 286
nvme0n1 teo 1 10495 746168 3389 0.005 2909 480
nvme0n1 teo 2 11521 813042 3312 0.004 2866 446
nvme0n1 teo 3 10478 742582 1595 0.002 1418 177
nvme0n1 teo 4 10735 759906 1963 0.003 1810 153
nvme0n1 teo 5 11168 790226 3220 0.004 2850 370
nvme0n1 teo 6 10440 739194 1211 0.002 1007 204
nvme0n1 teo 7 10719 754992 2855 0.004 2448 407
nvme0n1 teo 8 10457 742860 1999 0.003 1717 282
nvme0n1 teo 9 10723 750762 1216 0.002 1037 179
nvme0n1 teo-1 0 10621 754202 3112 0.004 2756 356
nvme0n1 teo-1 1 10668 754330 1757 0.002 1436 321
nvme0n1 teo-1 2 10459 733296 2182 0.003 2006 176
nvme0n1 teo-1 3 11346 801012 3005 0.004 2699 306
nvme0n1 teo-1 4 10435 738676 612 0.001 437 175
nvme0n1 teo-1 5 10737 757850 1631 0.002 1526 105
nvme0n1 teo-1 6 10436 731580 2136 0.003 1925 211
nvme0n1 teo-1 7 10705 754172 1173 0.002 972 201
nvme0n1 teo-1 8 10436 742248 2231 0.003 2065 166
nvme0n1 teo-1 9 10804 761666 2498 0.003 2324 174
nvme0n1 teo-2 0 11112 787050 3308 0.004 2948 360
nvme0n1 teo-2 1 10608 749840 1118 0.001 985 133
nvme0n1 teo-2 2 10770 764556 3233 0.004 2985 248
nvme0n1 teo-2 3 10594 750202 1745 0.002 1633 112
nvme0n1 teo-2 4 10528 745329 1607 0.002 1441 166
nvme0n1 teo-2 5 10711 759934 2876 0.004 2745 131
nvme0n1 teo-2 6 10694 756418 3083 0.004 2800 283
nvme0n1 teo-2 7 10808 764226 2621 0.003 2403 218
nvme0n1 teo-2 8 10505 741516 1800 0.002 1681 119
nvme0n1 teo-2 9 11429 806750 3258 0.004 2974 284
nvme0n1 teo-3 0 10445 736708 1969 0.003 1903 66
nvme0n1 teo-3 1 10725 752562 812 0.001 699 113
nvme0n1 teo-3 2 10433 743248 2866 0.004 2749 117
nvme0n1 teo-3 3 10733 762146 3015 0.004 2916 99
nvme0n1 teo-3 4 11588 814266 2691 0.003 2497 194
nvme0n1 teo-3 5 11161 789852 2570 0.003 2481 89
nvme0n1 teo-3 6 10464 744400 2697 0.004 2470 227
nvme0n1 teo-3 7 10727 756358 785 0.001 670 115
nvme0n1 teo-3 8 10427 737060 971 0.001 828 143
nvme0n1 teo-3 9 11262 793820 2101 0.003 1975 126
nvme0n1 teo-4 0 10540 745888 864 0.001 741 123
nvme0n1 teo-4 1 10702 754880 716 0.001 585 131
nvme0n1 teo-4 2 10531 746718 2758 0.004 2470 288
nvme0n1 teo-4 3 10780 761054 2805 0.004 2585 220
nvme0n1 teo-4 4 11002 781084 2824 0.004 2686 138
nvme0n1 teo-4 5 10688 753876 2775 0.004 2577 198
nvme0n1 teo-4 6 10652 755318 1838 0.002 1695 143
nvme0n1 teo-4 7 11020 777710 1676 0.002 1551 125
nvme0n1 teo-4 8 10559 750244 2348 0.003 2190 158
nvme0n1 teo-4 9 10777 761542 1784 0.002 1662 122
nvme0n1 teo-4-p 0 10471 742270 1613 0.002 1433 180
nvme0n1 teo-4-p 1 10744 752344 1814 0.002 1656 158
nvme0n1 teo-4-p 2 10478 740942 2069 0.003 1959 110
nvme0n1 teo-4-p 3 10744 760184 2911 0.004 2556 355
nvme0n1 teo-4-p 4 10537 750784 2741 0.004 2549 192
nvme0n1 teo-4-p 5 11894 839150 3098 0.004 2857 241
nvme0n1 teo-4-p 6 10600 752718 1966 0.003 1770 196
nvme0n1 teo-4-p 7 11058 780202 1676 0.002 1486 190
nvme0n1 teo-4-p 8 10518 731898 1650 0.002 1411 239
nvme0n1 teo-4-p 9 10698 757768 2938 0.004 2613 325
nvme0n1 teo-5 0 10501 745952 2107 0.003 1963 144
nvme0n1 teo-5 1 10731 760952 2541 0.003 2435 106
nvme0n1 teo-5 2 10524 724324 1341 0.002 1176 165
nvme0n1 teo-5 3 10680 755682 2082 0.003 1965 117
nvme0n1 teo-5 4 10512 743696 906 0.001 824 82
nvme0n1 teo-5 5 10731 762428 3091 0.004 2934 157
nvme0n1 teo-5 6 10452 744728 2360 0.003 2242 118
nvme0n1 teo-5 7 11478 809550 3072 0.004 2907 165
nvme0n1 teo-5 8 10432 738524 1688 0.002 1544 144
nvme0n1 teo-5 9 10734 756216 505 0.001 396 109
nvme0n1 teo-5-p 0 10826 762866 2733 0.004 2355 378
nvme0n1 teo-5-p 1 11231 794422 3225 0.004 2959 266
nvme0n1 teo-5-p 2 10504 744906 2444 0.003 2242 202
nvme0n1 teo-5-p 3 10639 753520 2521 0.003 2200 321
nvme0n1 teo-5-p 4 10531 748482 3322 0.004 2957 365
nvme0n1 teo-5-p 5 11172 788240 1915 0.002 1605 310
nvme0n1 teo-5-p 6 10908 773606 3235 0.004 2931 304
nvme0n1 teo-5-p 7 11405 806548 3147 0.004 2932 215
nvme0n1 teo-5-p 8 10654 756320 3257 0.004 2927 330
nvme0n1 teo-5-p 9 11028 778448 2912 0.004 2599 313
sda menu 0 914 458290 159821 0.349 140147 19674
sda menu 1 851 518848 182249 0.351 162972 19277
sda menu 2 887 541418 194831 0.360 174749 20082
sda menu 3 851 524610 197903 0.377 177427 20476
sda menu 4 850 525586 191754 0.365 171097 20657
sda menu 5 887 543680 196250 0.361 175415 20835
sda menu 6 857 529326 185632 0.351 165228 20404
sda menu 7 900 539138 191203 0.355 170732 20471
sda menu 8 871 514152 191282 0.372 170863 20419
sda menu 9 893 481884 172076 0.357 153744 18332
sda teo 0 1629 985284 926 0.001 782 144
sda teo 1 1630 1020098 3197 0.003 3029 168
sda teo 2 1637 953748 2957 0.003 2740 217
sda teo 3 1629 1008518 1009 0.001 815 194
sda teo 4 1647 1005468 735 0.001 609 126
sda teo 5 1641 822122 3571 0.004 3409 162
sda teo 6 1630 992350 2926 0.003 2760 166
sda teo 7 1647 988134 480 0.000 256 224
sda teo 8 1632 982006 2411 0.002 2268 143
sda teo 9 1628 805014 803 0.001 594 209
sda teo-1 0 1650 819198 3022 0.004 2759 263
sda teo-1 1 1648 737572 2122 0.003 1932 190
sda teo-1 2 1644 994814 520 0.001 360 160
sda teo-1 3 1633 924068 2379 0.003 2256 123
sda teo-1 4 1630 924010 2710 0.003 2602 108
sda teo-1 5 1643 1037002 2856 0.003 2754 102
sda teo-1 6 1629 995810 2881 0.003 2780 101
sda teo-1 7 1629 1027435 2599 0.003 2463 136
sda teo-1 8 1629 741596 2850 0.004 2693 157
sda teo-1 9 1626 958394 2878 0.003 2718 160
sda teo-2 0 1640 1030394 4898 0.005 4726 172
sda teo-2 1 1645 938574 3304 0.004 3144 160
sda teo-2 2 1634 915432 2094 0.002 1976 118
sda teo-2 3 1635 987666 2877 0.003 2773 104
sda teo-2 4 1626 841040 3451 0.004 3320 131
sda teo-2 5 1645 916892 2908 0.003 2796 112
sda teo-2 6 1627 990686 3048 0.003 2888 160
sda teo-2 7 1628 1025299 2939 0.003 2861 78
sda teo-2 8 1635 1016134 2098 0.002 1958 140
sda teo-2 9 1641 1009210 4085 0.004 3965 120
sda teo-3 0 1650 728638 909 0.001 768 141
sda teo-3 1 1630 977878 2865 0.003 2763 102
sda teo-3 2 1630 981243 901 0.001 770 131
sda teo-3 3 1640 835847 3508 0.004 3407 101
sda teo-3 4 1632 910148 2424 0.003 2313 111
sda teo-3 5 1650 751912 3501 0.005 3336 165
sda teo-3 6 1645 979578 916 0.001 769 147
sda teo-3 7 1653 700100 2411 0.003 2302 109
sda teo-3 8 1648 833250 489 0.001 393 96
sda teo-3 9 1629 764804 731 0.001 572 159
sda teo-4 0 1623 970630 7198 0.007 7127 71
sda teo-4 1 1615 1017032 10396 0.010 10303 93
sda teo-4 2 1573 894480 33012 0.037 32921 91
sda teo-4 3 1620 859323 3932 0.005 3816 116
sda teo-4 4 1628 1016788 2658 0.003 2612 46
sda teo-4 5 1627 1010060 3865 0.004 3775 90
sda teo-4 6 1590 956068 21585 0.023 21473 112
sda teo-4 7 1612 1012148 9659 0.010 9614 45
sda teo-4 8 1630 952508 2095 0.002 1960 135
sda teo-4 9 1604 874512 13939 0.016 13854 85
sda teo-4-p 0 1630 1016320 2843 0.003 2684 159
sda teo-4-p 1 1641 970262 2954 0.003 2786 168
sda teo-4-p 2 1645 1019642 2727 0.003 2606 121
sda teo-4-p 3 1628 762076 2877 0.004 2553 324
sda teo-4-p 4 1626 908362 2960 0.003 2787 173
sda teo-4-p 5 1631 972010 2705 0.003 2582 123
sda teo-4-p 6 1629 930382 2770 0.003 2639 131
sda teo-4-p 7 1641 938488 604 0.001 458 146
sda teo-4-p 8 1641 992768 990 0.001 840 150
sda teo-4-p 9 1628 904150 2973 0.003 2809 164
sda teo-5 0 1633 989694 1795 0.002 1728 67
sda teo-5 1 1623 984366 4332 0.004 4152 180
sda teo-5 2 1626 1004042 3144 0.003 3064 80
sda teo-5 3 1631 886814 2828 0.003 2614 214
sda teo-5 4 1605 775144 13890 0.018 13724 166
sda teo-5 5 1624 743920 3992 0.005 3917 75
sda teo-5 6 1653 657916 478 0.001 331 147
sda teo-5 7 1630 799216 2900 0.004 2786 114
sda teo-5 8 1635 932904 4162 0.004 4050 112
sda teo-5 9 1611 892454 11780 0.013 11670 110
sda teo-5-p 0 1632 923770 1702 0.002 1386 316
sda teo-5-p 1 1631 859976 2802 0.003 2418 384
sda teo-5-p 2 1630 1026432 2767 0.003 2656 111
sda teo-5-p 3 1629 1014834 590 0.001 472 118
sda teo-5-p 4 1638 949080 4872 0.005 4669 203
sda teo-5-p 5 1628 1013538 2860 0.003 2725 135
sda teo-5-p 6 1628 948944 992 0.001 834 158
sda teo-5-p 7 1643 1033465 2745 0.003 2621 124
sda teo-5-p 8 1639 906786 1006 0.001 878 128
sda teo-5-p 9 1626 715660 2696 0.004 2551 145
mtdblock3 menu 0 154 374880 131088 0.350 111396 19692
mtdblock3 menu 1 151 422820 135538 0.321 118235 17303
mtdblock3 menu 2 132 486228 164749 0.339 146417 18332
mtdblock3 menu 3 144 407212 146228 0.359 126654 19574
mtdblock3 menu 4 167 328526 109689 0.334 90367 19322
mtdblock3 menu 5 168 364234 121480 0.334 102098 19382
mtdblock3 menu 6 140 423880 153459 0.362 134032 19427
mtdblock3 menu 7 186 279764 84779 0.303 64987 19792
mtdblock3 menu 8 150 447254 138292 0.309 120292 18000
mtdblock3 menu 9 170 311228 102250 0.329 84318 17932
mtdblock3 teo 0 261 912934 2746 0.003 2638 108
mtdblock3 teo 1 256 764556 3986 0.005 3918 68
mtdblock3 teo 2 260 906018 2397 0.003 2229 168
mtdblock3 teo 3 256 906950 2304 0.003 2198 106
mtdblock3 teo 4 256 533356 5075 0.010 4927 148
mtdblock3 teo 5 255 911468 3336 0.004 3255 81
mtdblock3 teo 6 256 827172 3257 0.004 3156 101
mtdblock3 teo 7 257 719934 2804 0.004 2715 89
mtdblock3 teo 8 258 479528 2464 0.005 2342 122
mtdblock3 teo 9 258 586072 3426 0.006 3338 88
mtdblock3 teo-1 0 258 510920 4207 0.008 3804 403
mtdblock3 teo-1 1 257 801532 1174 0.001 1014 160
mtdblock3 teo-1 2 258 896070 3269 0.004 3109 160
mtdblock3 teo-1 3 256 819626 2837 0.003 2736 101
mtdblock3 teo-1 4 262 903436 806 0.001 685 121
mtdblock3 teo-1 5 257 712534 3311 0.005 3202 109
mtdblock3 teo-1 6 257 750034 2755 0.004 2635 120
mtdblock3 teo-1 7 256 911194 5528 0.006 5360 168
mtdblock3 teo-1 8 256 888464 2756 0.003 2645 111
mtdblock3 teo-1 9 255 909178 4253 0.005 4157 96
mtdblock3 teo-2 0 259 517246 3351 0.006 3282 69
mtdblock3 teo-2 1 256 890048 1353 0.002 1263 90
mtdblock3 teo-2 2 259 509644 2101 0.004 2009 92
mtdblock3 teo-2 3 256 915324 3230 0.004 3159 71
mtdblock3 teo-2 4 256 907364 1886 0.002 1828 58
mtdblock3 teo-2 5 259 424006 3491 0.008 3384 107
mtdblock3 teo-2 6 258 606754 1324 0.002 1248 76
mtdblock3 teo-2 7 262 939862 1922 0.002 1841 81
mtdblock3 teo-2 8 257 513084 5515 0.011 5400 115
mtdblock3 teo-2 9 262 945954 2747 0.003 2645 102
mtdblock3 teo-3 0 253 905688 8478 0.009 8412 66
mtdblock3 teo-3 1 258 601240 3658 0.006 3583 75
mtdblock3 teo-3 2 254 907660 5515 0.006 5404 111
mtdblock3 teo-3 3 252 758772 10805 0.014 10727 78
mtdblock3 teo-3 4 261 906302 1338 0.001 1254 84
mtdblock3 teo-3 5 260 370876 2771 0.007 2721 50
mtdblock3 teo-3 6 256 821390 4319 0.005 4283 36
mtdblock3 teo-3 7 258 607659 2273 0.004 2200 73
mtdblock3 teo-3 8 256 895490 2422 0.003 2294 128
mtdblock3 teo-3 9 256 916842 7220 0.008 7139 81
mtdblock3 teo-4 0 250 890514 14531 0.016 14501 30
mtdblock3 teo-4 1 255 758434 4527 0.006 4485 42
mtdblock3 teo-4 2 258 446396 2987 0.007 2961 26
mtdblock3 teo-4 3 261 453260 1917 0.004 1884 33
mtdblock3 teo-4 4 247 705740 22283 0.032 22224 59
mtdblock3 teo-4 5 248 740716 17636 0.024 17569 67
mtdblock3 teo-4 6 256 911312 1377 0.002 1342 35
mtdblock3 teo-4 7 254 845856 6890 0.008 6802 88
mtdblock3 teo-4 8 251 631104 16048 0.025 15983 65
mtdblock3 teo-4 9 257 779876 3353 0.004 3309 44
mtdblock3 teo-4-p 0 255 872784 2444 0.003 2369 75
mtdblock3 teo-4-p 1 259 395130 3736 0.009 3628 108
mtdblock3 teo-4-p 2 257 797682 3319 0.004 3235 84
mtdblock3 teo-4-p 3 258 393938 3949 0.010 3872 77
mtdblock3 teo-4-p 4 259 899511 5355 0.006 5169 186
mtdblock3 teo-4-p 5 259 549306 2671 0.005 2588 83
mtdblock3 teo-4-p 6 257 826008 2776 0.003 2691 85
mtdblock3 teo-4-p 7 257 466122 7635 0.016 7536 99
mtdblock3 teo-4-p 8 256 920086 3627 0.004 3477 150
mtdblock3 teo-4-p 9 256 910830 2368 0.003 2264 104
mtdblock3 teo-5 0 257 683388 2562 0.004 2516 46
mtdblock3 teo-5 1 253 897532 7529 0.008 7498 31
mtdblock3 teo-5 2 260 390444 3302 0.008 3264 38
mtdblock3 teo-5 3 255 388000 11643 0.030 11559 84
mtdblock3 teo-5 4 259 458896 3526 0.008 3458 68
mtdblock3 teo-5 5 259 488898 2634 0.005 2600 34
mtdblock3 teo-5 6 254 904758 5270 0.006 5246 24
mtdblock3 teo-5 7 258 408352 3802 0.009 3758 44
mtdblock3 teo-5 8 242 847632 28778 0.034 28739 39
mtdblock3 teo-5 9 249 659764 16515 0.025 16470 45
mtdblock3 teo-5-p 0 254 886750 4686 0.005 4572 114
mtdblock3 teo-5-p 1 256 916092 595 0.001 472 123
mtdblock3 teo-5-p 2 257 773550 3558 0.005 3448 110
mtdblock3 teo-5-p 3 256 886680 1959 0.002 1849 110
mtdblock3 teo-5-p 4 260 269350 3409 0.013 3302 107
mtdblock3 teo-5-p 5 256 890416 1977 0.002 1875 102
mtdblock3 teo-5-p 6 258 611094 3353 0.005 3286 67
mtdblock3 teo-5-p 7 258 651734 2592 0.004 2480 112
mtdblock3 teo-5-p 8 256 916952 2848 0.003 2773 75
mtdblock3 teo-5-p 9 256 833438 3247 0.004 3155 92
nullb0 menu 0 99086 83766 22957 0.274 2611 20346
nullb0 menu 1 98836 87023 23729 0.273 2807 20922
nullb0 menu 2 99683 86746 23562 0.272 2801 20761
nullb0 menu 3 99256 87346 23814 0.273 2992 20822
nullb0 menu 4 99282 86318 23926 0.277 2779 21147
nullb0 menu 5 99453 82164 22386 0.272 2419 19967
nullb0 menu 6 98422 86482 23589 0.273 2677 20912
nullb0 menu 7 98691 70436 18111 0.257 1610 16501
nullb0 menu 8 99059 84498 22762 0.269 2749 20013
nullb0 menu 9 98480 87444 23902 0.273 2802 21100
nullb0 teo 0 98997 84748 2799 0.033 2733 66
nullb0 teo 1 98983 86392 3365 0.039 3042 323
nullb0 teo 2 99933 84289 3493 0.041 2988 505
nullb0 teo 3 99170 87020 3306 0.038 3012 294
nullb0 teo 4 98627 81402 3353 0.041 2913 440
nullb0 teo 5 98952 83336 3611 0.043 3081 530
nullb0 teo 6 99391 85478 3076 0.036 2862 214
nullb0 teo 7 98836 83064 3348 0.040 2884 464
nullb0 teo 8 99005 86781 3040 0.035 2896 144
nullb0 teo 9 99062 86574 3194 0.037 2921 273
nullb0 teo-1 0 99535 79018 3288 0.042 2690 598
nullb0 teo-1 1 98710 82341 3299 0.040 2833 466
nullb0 teo-1 2 99282 87326 3940 0.045 3367 573
nullb0 teo-1 3 98647 85458 3366 0.039 3026 340
nullb0 teo-1 4 99175 86524 2901 0.034 2801 100
nullb0 teo-1 5 99378 86576 3036 0.035 2861 175
nullb0 teo-1 6 98825 85538 4180 0.049 3125 1055
nullb0 teo-1 7 99300 87116 2889 0.033 2821 68
nullb0 teo-1 8 99162 86888 2892 0.033 2802 90
nullb0 teo-1 9 99073 87200 2897 0.033 2813 84
nullb0 teo-2 0 98773 85718 2969 0.035 2881 88
nullb0 teo-2 1 98253 79552 3093 0.039 2792 301
nullb0 teo-2 2 99901 82222 2801 0.034 2668 133
nullb0 teo-2 3 98720 86946 2883 0.033 2858 25
nullb0 teo-2 4 98881 86948 3219 0.037 3049 170
nullb0 teo-2 5 98442 84058 2907 0.035 2832 75
nullb0 teo-2 6 99210 86662 3558 0.041 3237 321
nullb0 teo-2 7 99440 87028 2917 0.034 2860 57
nullb0 teo-2 8 99332 80036 2966 0.037 2803 163
nullb0 teo-2 9 98432 83230 2970 0.036 2867 103
nullb0 teo-3 0 98842 83504 3152 0.038 3016 136
nullb0 teo-3 1 98355 86828 3685 0.042 3358 327
nullb0 teo-3 2 98929 87006 2851 0.033 2809 42
nullb0 teo-3 3 98795 84358 3461 0.041 3201 260
nullb0 teo-3 4 99462 81254 2944 0.036 2820 124
nullb0 teo-3 5 98875 87172 3728 0.043 3386 342
nullb0 teo-3 6 98494 85348 3681 0.043 3297 384
nullb0 teo-3 7 98180 86864 2872 0.033 2833 39
nullb0 teo-3 8 99380 80650 2768 0.034 2659 109
nullb0 teo-3 9 99173 86480 3504 0.041 3221 283
nullb0 teo-4 0 98421 82990 3307 0.040 3124 183
nullb0 teo-4 1 99794 69938 2183 0.031 2111 72
nullb0 teo-4 2 98564 83608 2607 0.031 2597 10
nullb0 teo-4 3 98625 85356 3552 0.042 3290 262
nullb0 teo-4 4 98534 84396 3634 0.043 3325 309
nullb0 teo-4 5 98794 86918 3523 0.041 3320 203
nullb0 teo-4 6 99060 86324 3658 0.042 3376 282
nullb0 teo-4 7 99823 86814 2818 0.032 2812 6
nullb0 teo-4 8 98927 82376 3294 0.040 3086 208
nullb0 teo-4 9 98943 86806 2840 0.033 2839 1
nullb0 teo-4-p 0 98339 86996 3656 0.042 3304 352
nullb0 teo-4-p 1 99212 87055 2875 0.033 2822 53
nullb0 teo-4-p 2 98987 83119 3242 0.039 3013 229
nullb0 teo-4-p 3 98057 86690 3702 0.043 3368 334
nullb0 teo-4-p 4 99265 86268 3066 0.036 2930 136
nullb0 teo-4-p 5 98608 86442 3093 0.036 2995 98
nullb0 teo-4-p 6 99434 86988 3836 0.044 3414 422
nullb0 teo-4-p 7 99042 85148 3146 0.037 3022 124
nullb0 teo-4-p 8 99203 81808 2673 0.033 2594 79
nullb0 teo-4-p 9 99128 84132 3054 0.036 2935 119
nullb0 teo-5 0 98861 86644 3237 0.037 3140 97
nullb0 teo-5 1 99510 86606 2895 0.033 2871 24
nullb0 teo-5 2 99231 85920 3099 0.036 3042 57
nullb0 teo-5 3 98416 82463 3155 0.038 2997 158
nullb0 teo-5 4 99179 82584 3116 0.038 2941 175
nullb0 teo-5 5 98634 86905 2829 0.033 2824 5
nullb0 teo-5 6 98739 84148 3130 0.037 2942 188
nullb0 teo-5 7 99245 86920 2843 0.033 2835 8
nullb0 teo-5 8 99582 85904 3341 0.039 3205 136
nullb0 teo-5 9 99695 86640 2836 0.033 2832 4
nullb0 teo-5-p 0 98635 84786 3556 0.042 3208 348
nullb0 teo-5-p 1 98244 85872 3280 0.038 3116 164
nullb0 teo-5-p 2 98446 87198 3767 0.043 3369 398
nullb0 teo-5-p 3 99078 85092 3477 0.041 3139 338
nullb0 teo-5-p 4 99177 78206 2433 0.031 2363 70
nullb0 teo-5-p 5 98696 85216 3262 0.038 3102 160
nullb0 teo-5-p 6 99376 86124 3338 0.039 3108 230
nullb0 teo-5-p 7 99885 86000 3598 0.042 3327 271
nullb0 teo-5-p 8 98992 83978 3480 0.041 3128 352
nullb0 teo-5-p 9 99465 75044 2463 0.033 2333 130
device gov iter iops idles idle_misses idle_miss_ratio belows aboves
mapper/dm-slow menu 0 115 49976 12729 0.255 1767 10962
mapper/dm-slow menu 1 115 48099 12394 0.258 1622 10772
mapper/dm-slow menu 2 115 45546 11405 0.250 1341 10064
mapper/dm-slow menu 3 115 44204 10840 0.245 1185 9655
mapper/dm-slow menu 4 115 47964 12382 0.258 1665 10717
mapper/dm-slow menu 5 115 47892 12364 0.258 1688 10676
mapper/dm-slow menu 6 115 45908 11679 0.254 1473 10206
mapper/dm-slow menu 7 115 45944 11578 0.252 1415 10163
mapper/dm-slow menu 8 115 47440 12048 0.254 1518 10530
mapper/dm-slow menu 9 115 48518 12627 0.260 1786 10841
mapper/dm-slow teo 0 115 46316 1704 0.037 1518 186
mapper/dm-slow teo 1 115 47910 1859 0.039 1691 168
mapper/dm-slow teo 2 115 47054 1770 0.038 1603 167
mapper/dm-slow teo 3 115 46792 1783 0.038 1598 185
mapper/dm-slow teo 4 115 46916 1850 0.039 1648 202
mapper/dm-slow teo 5 115 47116 1765 0.037 1622 143
mapper/dm-slow teo 6 115 42134 1446 0.034 1294 152
mapper/dm-slow teo 7 115 44814 1617 0.036 1437 180
mapper/dm-slow teo 8 115 47382 1888 0.040 1704 184
mapper/dm-slow teo 9 115 45976 1676 0.036 1501 175
mapper/dm-slow teo-1 0 115 43578 1549 0.036 1360 189
mapper/dm-slow teo-1 1 115 45796 1640 0.036 1458 182
mapper/dm-slow teo-1 2 115 47698 1804 0.038 1638 166
mapper/dm-slow teo-1 3 115 46686 1728 0.037 1587 141
mapper/dm-slow teo-1 4 115 47308 1793 0.038 1617 176
mapper/dm-slow teo-1 5 115 45230 1688 0.037 1497 191
mapper/dm-slow teo-1 6 115 44948 1614 0.036 1397 217
mapper/dm-slow teo-1 7 115 46346 1704 0.037 1500 204
mapper/dm-slow teo-1 8 115 44014 1531 0.035 1352 179
mapper/dm-slow teo-1 9 115 46356 1787 0.039 1597 190
mapper/dm-slow teo-2 0 115 47121 1791 0.038 1654 137
mapper/dm-slow teo-2 1 115 46562 1681 0.036 1553 128
mapper/dm-slow teo-2 2 115 46832 1760 0.038 1614 146
mapper/dm-slow teo-2 3 115 43006 1426 0.033 1325 101
mapper/dm-slow teo-2 4 115 45562 1645 0.036 1494 151
mapper/dm-slow teo-2 5 115 47136 1687 0.036 1570 117
mapper/dm-slow teo-2 6 115 46062 1657 0.036 1539 118
mapper/dm-slow teo-2 7 115 42834 1462 0.034 1329 133
mapper/dm-slow teo-2 8 115 46700 1810 0.039 1670 140
mapper/dm-slow teo-2 9 115 47418 1783 0.038 1635 148
mapper/dm-slow teo-3 0 115 45634 1647 0.036 1545 102
mapper/dm-slow teo-3 1 115 45828 1618 0.035 1483 135
mapper/dm-slow teo-3 2 115 46108 1697 0.037 1577 120
mapper/dm-slow teo-3 3 115 46768 1627 0.035 1520 107
mapper/dm-slow teo-3 4 115 44964 1568 0.035 1460 108
mapper/dm-slow teo-3 5 115 46112 1648 0.036 1520 128
mapper/dm-slow teo-3 6 115 47354 1701 0.036 1608 93
mapper/dm-slow teo-3 7 115 46496 1707 0.037 1589 118
mapper/dm-slow teo-3 8 115 44094 1487 0.034 1386 101
mapper/dm-slow teo-3 9 115 46454 1612 0.035 1515 97
mapper/dm-slow teo-4 0 115 41464 1305 0.031 1243 62
mapper/dm-slow teo-4 1 115 45886 1582 0.034 1499 83
mapper/dm-slow teo-4 2 115 47338 1710 0.036 1631 79
mapper/dm-slow teo-4 3 115 45460 1548 0.034 1478 70
mapper/dm-slow teo-4 4 115 45156 1564 0.035 1483 81
mapper/dm-slow teo-4 5 115 47344 1644 0.035 1568 76
mapper/dm-slow teo-4 6 115 46120 1634 0.035 1543 91
mapper/dm-slow teo-4 7 115 47252 1704 0.036 1630 74
mapper/dm-slow teo-4 8 115 46350 1694 0.037 1602 92
mapper/dm-slow teo-4 9 115 43206 1458 0.034 1370 88
mapper/dm-slow teo-4-p 0 115 39874 1258 0.032 1116 142
mapper/dm-slow teo-4-p 1 115 45428 1648 0.036 1509 139
mapper/dm-slow teo-4-p 2 115 47050 1789 0.038 1660 129
mapper/dm-slow teo-4-p 3 115 38592 1138 0.029 1025 113
mapper/dm-slow teo-4-p 4 115 47388 1720 0.036 1604 116
mapper/dm-slow teo-4-p 5 115 45250 1569 0.035 1454 115
mapper/dm-slow teo-4-p 6 115 45600 1582 0.035 1464 118
mapper/dm-slow teo-4-p 7 115 46096 1667 0.036 1528 139
mapper/dm-slow teo-4-p 8 115 45866 1681 0.037 1554 127
mapper/dm-slow teo-4-p 9 115 44680 1577 0.035 1460 117
mapper/dm-slow teo-5 0 115 46252 1593 0.034 1529 64
mapper/dm-slow teo-5 1 115 46458 1705 0.037 1602 103
mapper/dm-slow teo-5 2 115 45392 1592 0.035 1516 76
mapper/dm-slow teo-5 3 115 45074 1571 0.035 1500 71
mapper/dm-slow teo-5 4 115 45642 1586 0.035 1512 74
mapper/dm-slow teo-5 5 115 47036 1807 0.038 1703 104
mapper/dm-slow teo-5 6 115 47310 1692 0.036 1612 80
mapper/dm-slow teo-5 7 115 44742 1518 0.034 1446 72
mapper/dm-slow teo-5 8 115 46880 1696 0.036 1613 83
mapper/dm-slow teo-5 9 115 47012 1648 0.035 1572 76
mapper/dm-slow teo-5-p 0 115 47288 1829 0.039 1684 145
mapper/dm-slow teo-5-p 1 115 45814 1665 0.036 1524 141
mapper/dm-slow teo-5-p 2 115 47440 1782 0.038 1641 141
mapper/dm-slow teo-5-p 3 115 46976 1843 0.039 1681 162
mapper/dm-slow teo-5-p 4 115 45910 1653 0.036 1508 145
mapper/dm-slow teo-5-p 5 115 43436 1550 0.036 1425 125
mapper/dm-slow teo-5-p 6 115 45440 1599 0.035 1475 124
mapper/dm-slow teo-5-p 7 115 46520 1737 0.037 1585 152
mapper/dm-slow teo-5-p 8 115 47300 1788 0.038 1643 145
mapper/dm-slow teo-5-p 9 115 44770 1537 0.034 1402 135
test gov i score %change idles idle_miss miss_rt belows aboves
schbench menu 0 280.43 +0.00% 31034 2481 0.080 2161 320
schbench menu 1 300.57 +7.18% 34442 2789 0.081 2464 325
schbench menu 2 269.43 -3.92% 29314 2053 0.070 1787 266
schbench menu 3 291.70 +4.02% 32508 2514 0.077 2249 265
schbench menu 4 292.33 +4.24% 32892 2675 0.081 2408 267
schbench menu 5 288.97 +3.05% 33440 2839 0.085 2521 318
schbench menu 6 297.47 +6.08% 33320 2507 0.075 2171 336
schbench menu 7 293.33 +4.60% 33244 2674 0.080 2397 277
schbench menu 8 293.70 +4.73% 32726 2455 0.075 2191 264
schbench menu 9 290.13 +3.46% 33208 2802 0.084 2492 310
schbench menu 10 292.03 +4.14% 32795 2514 0.077 2241 273
schbench menu 11 289.23 +3.14% 32012 2525 0.079 2252 273
schbench menu 12 295.30 +5.30% 33842 2806 0.083 2467 339
schbench menu 13 292.07 +4.15% 32830 2624 0.080 2303 321
schbench menu 14 292.43 +4.28% 32940 2520 0.077 2248 272
schbench menu 15 290.97 +3.76% 31776 2410 0.076 2094 316
schbench menu 16 293.03 +4.49% 32894 2542 0.077 2259 283
schbench menu 17 291.53 +3.96% 32125 2290 0.071 2002 288
schbench menu 18 292.70 +4.38% 33098 2514 0.076 2196 318
schbench menu 19 291.23 +3.85% 32936 2604 0.079 2326 278
schbench teo 0 301.07 +7.36% 31214 627 0.020 163 464
schbench teo 1 297.27 +6.01% 30458 606 0.020 141 465
schbench teo 2 296.63 +5.78% 29718 473 0.016 125 348
schbench teo 3 297.30 +6.02% 30459 526 0.017 125 401
schbench teo 4 298.97 +6.61% 29794 607 0.020 134 473
schbench teo 5 300.63 +7.20% 30708 568 0.018 152 416
schbench teo 6 298.77 +6.54% 30310 533 0.018 142 391
schbench teo 7 300.20 +7.05% 31044 598 0.019 164 434
schbench teo 8 303.13 +8.09% 30922 546 0.018 143 403
schbench teo 9 306.03 +9.13% 32396 514 0.016 149 365
schbench teo 10 303.80 +8.33% 31912 617 0.019 149 468
schbench teo 11 291.17 +3.83% 30524 521 0.017 140 381
schbench teo 12 280.93 +0.18% 28110 479 0.017 134 345
schbench teo 13 298.50 +6.44% 28940 519 0.018 140 379
schbench teo 14 303.17 +8.11% 30184 519 0.017 129 390
schbench teo 15 288.07 +2.72% 28772 493 0.017 121 372
schbench teo 16 289.43 +3.21% 28340 504 0.018 144 360
schbench teo 17 295.13 +5.24% 29493 610 0.021 143 467
schbench teo 18 301.73 +7.60% 31242 541 0.017 150 391
schbench teo 19 297.10 +5.94% 30222 552 0.018 167 385
schbench teo-1 0 302.27 +7.79% 30060 509 0.017 125 384
schbench teo-1 1 299.87 +6.93% 30492 632 0.021 168 464
schbench teo-1 2 302.60 +7.91% 31618 583 0.018 156 427
schbench teo-1 3 295.97 +5.54% 29132 578 0.020 149 429
schbench teo-1 4 301.90 +7.66% 30555 600 0.020 173 427
schbench teo-1 5 296.57 +5.76% 30030 546 0.018 154 392
schbench teo-1 6 299.37 +6.75% 29766 511 0.017 141 370
schbench teo-1 7 307.53 +9.66% 30794 499 0.016 133 366
schbench teo-1 8 274.67 -2.05% 28088 481 0.017 150 331
schbench teo-1 9 298.10 +6.30% 30440 533 0.018 138 395
schbench teo-1 10 299.83 +6.92% 31182 593 0.019 150 443
schbench teo-1 11 296.47 +5.72% 30644 524 0.017 130 394
schbench teo-1 12 298.17 +6.33% 29682 575 0.019 155 420
schbench teo-1 13 299.97 +6.97% 30150 499 0.017 130 369
schbench teo-1 14 297.77 +6.18% 30966 607 0.020 168 439
schbench teo-1 15 297.33 +6.03% 30602 597 0.020 164 433
schbench teo-1 16 297.67 +6.15% 30310 664 0.022 137 527
schbench teo-1 17 287.13 +2.39% 28712 573 0.020 165 408
schbench teo-1 18 297.77 +6.18% 30838 607 0.020 170 437
schbench teo-1 19 299.03 +6.63% 30380 547 0.018 143 404
schbench teo-2 0 301.10 +7.37% 30260 529 0.017 303 226
schbench teo-2 1 301.67 +7.57% 29945 412 0.014 232 180
schbench teo-2 2 300.70 +7.23% 29752 557 0.019 313 244
schbench teo-2 3 293.83 +4.78% 29852 531 0.018 300 231
schbench teo-2 4 301.37 +7.47% 30502 605 0.020 319 286
schbench teo-2 5 299.93 +6.95% 30486 533 0.017 279 254
schbench teo-2 6 300.13 +7.02% 29736 532 0.018 282 250
schbench teo-2 7 288.73 +2.96% 28630 551 0.019 305 246
schbench teo-2 8 299.50 +6.80% 30264 513 0.017 279 234
schbench teo-2 9 300.00 +6.98% 30278 507 0.017 280 227
schbench teo-2 10 302.43 +7.85% 30218 490 0.016 265 225
schbench teo-2 11 300.17 +7.04% 29705 513 0.017 299 214
schbench teo-2 12 301.53 +7.52% 29106 454 0.016 264 190
schbench teo-2 13 296.17 +5.61% 30030 544 0.018 294 250
schbench teo-2 14 298.30 +6.37% 30593 529 0.017 266 263
schbench teo-2 15 299.90 +6.94% 29704 438 0.015 250 188
schbench teo-2 16 303.60 +8.26% 29732 453 0.015 266 187
schbench teo-2 17 298.57 +6.47% 29850 477 0.016 274 203
schbench teo-2 18 298.33 +6.38% 29580 465 0.016 266 199
schbench teo-2 19 300.33 +7.10% 31002 536 0.017 289 247
schbench teo-3 0 295.57 +5.40% 29430 519 0.018 305 214
schbench teo-3 1 298.27 +6.36% 29584 542 0.018 299 243
schbench teo-3 2 302.17 +7.75% 30752 477 0.016 276 201
schbench teo-3 3 297.50 +6.09% 29410 466 0.016 249 217
schbench teo-3 4 298.27 +6.36% 29834 518 0.017 301 217
schbench teo-3 5 298.93 +6.60% 30028 458 0.015 236 222
schbench teo-3 6 298.47 +6.43% 29316 518 0.018 272 246
schbench teo-3 7 303.03 +8.06% 29536 478 0.016 263 215
schbench teo-3 8 303.93 +8.38% 30676 488 0.016 249 239
schbench teo-3 9 298.10 +6.30% 31134 564 0.018 277 287
schbench teo-3 10 299.27 +6.72% 30920 519 0.017 288 231
schbench teo-3 11 297.93 +6.24% 30336 566 0.019 291 275
schbench teo-3 12 300.27 +7.07% 30700 540 0.018 308 232
schbench teo-3 13 304.77 +8.68% 31072 463 0.015 241 222
schbench teo-3 14 303.67 +8.29% 30986 537 0.017 270 267
schbench teo-3 15 286.70 +2.24% 29182 472 0.016 275 197
schbench teo-3 16 294.63 +5.06% 29098 433 0.015 240 193
schbench teo-3 17 298.77 +6.54% 29050 534 0.018 266 268
schbench teo-3 18 300.03 +6.99% 30176 570 0.019 311 259
schbench teo-3 19 295.63 +5.42% 29474 516 0.018 277 239
schbench teo-4 0 295.87 +5.51% 29014 1309 0.045 1168 141
schbench teo-4 1 302.20 +7.76% 31642 1408 0.044 1105 303
schbench teo-4 2 301.07 +7.36% 29966 1602 0.053 1393 209
schbench teo-4 3 297.00 +5.91% 29982 1529 0.051 1266 263
schbench teo-4 4 297.20 +5.98% 29432 1419 0.048 1194 225
schbench teo-4 5 299.47 +6.79% 30262 1381 0.046 1189 192
schbench teo-4 6 301.23 +7.42% 29810 1469 0.049 1234 235
schbench teo-4 7 301.00 +7.34% 29916 1511 0.051 1355 156
schbench teo-4 8 299.17 +6.68% 29246 1444 0.049 1296 148
schbench teo-4 9 297.77 +6.18% 30306 1399 0.046 1195 204
schbench teo-4 10 304.03 +8.42% 30904 1346 0.044 1163 183
schbench teo-4 11 300.73 +7.24% 30716 1276 0.042 1089 187
schbench teo-4 12 298.13 +6.31% 29872 1314 0.044 1081 233
schbench teo-4 13 298.73 +6.53% 30386 1288 0.042 1092 196
schbench teo-4 14 300.77 +7.25% 30150 1323 0.044 1074 249
schbench teo-4 15 296.23 +5.63% 32178 1328 0.041 1061 267
schbench teo-4 16 283.50 +1.09% 27472 1282 0.047 1126 156
schbench teo-4 17 300.53 +7.17% 29644 1552 0.052 1375 177
schbench teo-4 18 302.23 +7.77% 31456 1418 0.045 1194 224
schbench teo-4 19 307.43 +9.63% 32518 1259 0.039 968 291
schbench teo-4-p 0 277.53 -1.03% 28054 503 0.018 151 352
schbench teo-4-p 1 304.07 +8.43% 30650 553 0.018 178 375
schbench teo-4-p 2 273.73 -2.39% 27120 509 0.019 163 346
schbench teo-4-p 3 300.20 +7.05% 29932 528 0.018 171 357
schbench teo-4-p 4 300.00 +6.98% 29580 584 0.020 180 404
schbench teo-4-p 5 296.03 +5.56% 30176 514 0.017 172 342
schbench teo-4-p 6 300.50 +7.16% 29806 550 0.018 174 376
schbench teo-4-p 7 271.07 -3.34% 26934 472 0.018 165 307
schbench teo-4-p 8 299.00 +6.62% 30568 564 0.018 175 389
schbench teo-4-p 9 300.80 +7.26% 29660 482 0.016 188 294
schbench teo-4-p 10 300.57 +7.18% 30384 634 0.021 195 439
schbench teo-4-p 11 296.53 +5.74% 29590 599 0.020 221 378
schbench teo-4-p 12 298.77 +6.54% 30664 599 0.020 177 422
schbench teo-4-p 13 302.47 +7.86% 30200 559 0.019 223 336
schbench teo-4-p 14 300.43 +7.13% 30326 562 0.019 166 396
schbench teo-4-p 15 292.87 +4.44% 28798 524 0.018 142 382
schbench teo-4-p 16 299.63 +6.85% 29718 535 0.018 189 346
schbench teo-4-p 17 302.10 +7.73% 31122 584 0.019 221 363
schbench teo-4-p 18 294.13 +4.89% 28312 560 0.020 167 393
schbench teo-4-p 19 302.20 +7.76% 29872 556 0.019 168 388
schbench teo-5 0 299.53 +6.81% 31198 1312 0.042 1062 250
schbench teo-5 1 296.67 +5.79% 30090 1383 0.046 1169 214
schbench teo-5 2 299.60 +6.84% 30010 1566 0.052 1331 235
schbench teo-5 3 300.00 +6.98% 30764 1466 0.048 1239 227
schbench teo-5 4 298.77 +6.54% 30002 1383 0.046 1212 171
schbench teo-5 5 299.30 +6.73% 29848 1481 0.050 1288 193
schbench teo-5 6 298.90 +6.59% 29586 1441 0.049 1260 181
schbench teo-5 7 300.37 +7.11% 30672 1460 0.048 1306 154
schbench teo-5 8 301.00 +7.34% 31454 1383 0.044 1190 193
schbench teo-5 9 298.80 +6.55% 30042 1415 0.047 1194 221
schbench teo-5 10 302.23 +7.77% 30986 1305 0.042 1066 239
schbench teo-5 11 297.30 +6.02% 30112 1405 0.047 1190 215
schbench teo-5 12 299.33 +6.74% 29536 1482 0.050 1214 268
schbench teo-5 13 288.87 +3.01% 28754 1244 0.043 1038 206
schbench teo-5 14 300.17 +7.04% 30232 1517 0.050 1300 217
schbench teo-5 15 301.43 +7.49% 30578 1576 0.052 1287 289
schbench teo-5 16 282.10 +0.60% 29106 996 0.034 834 162
schbench teo-5 17 278.03 -0.86% 27432 1123 0.041 926 197
schbench teo-5 18 304.40 +8.55% 29934 1526 0.051 1296 230
schbench teo-5 19 276.50 -1.40% 28072 1158 0.041 970 188
schbench teo-5-p 0 299.47 +6.79% 29994 587 0.020 231 356
schbench teo-5-p 1 298.27 +6.36% 31600 555 0.018 190 365
schbench teo-5-p 2 297.20 +5.98% 30350 543 0.018 198 345
schbench teo-5-p 3 294.30 +4.95% 29962 523 0.017 192 331
schbench teo-5-p 4 303.30 +8.16% 30366 480 0.016 160 320
schbench teo-5-p 5 298.87 +6.58% 29838 541 0.018 180 361
schbench teo-5-p 6 301.63 +7.56% 30022 480 0.016 168 312
schbench teo-5-p 7 298.47 +6.43% 30210 581 0.019 202 379
schbench teo-5-p 8 298.77 +6.54% 30636 529 0.017 210 319
schbench teo-5-p 9 298.63 +6.49% 30474 564 0.019 185 379
schbench teo-5-p 10 300.10 +7.01% 30808 577 0.019 251 326
schbench teo-5-p 11 295.87 +5.51% 30190 565 0.019 186 379
schbench teo-5-p 12 301.27 +7.43% 29958 466 0.016 160 306
schbench teo-5-p 13 280.83 +0.14% 27916 494 0.018 164 330
schbench teo-5-p 14 294.23 +4.92% 30330 611 0.020 219 392
schbench teo-5-p 15 297.13 +5.96% 30728 656 0.021 195 461
schbench teo-5-p 16 297.97 +6.25% 30242 524 0.017 178 346
schbench teo-5-p 17 301.27 +7.43% 31004 524 0.017 208 316
schbench teo-5-p 18 297.97 +6.25% 29864 565 0.019 211 354
schbench teo-5-p 19 298.43 +6.42% 30462 519 0.017 150 369
ebizzy menu 0 10720.00 +0.00% 1198 311 0.260 31 280
ebizzy menu 1 10672.00 -0.45% 1214 336 0.277 42 294
ebizzy menu 2 10666.00 -0.50% 1052 280 0.266 43 237
ebizzy menu 3 10672.00 -0.45% 1004 250 0.249 23 227
ebizzy menu 4 10688.00 -0.30% 1104 288 0.261 47 241
ebizzy menu 5 10730.00 +0.09% 1138 271 0.238 22 249
ebizzy menu 6 10726.00 +0.06% 1147 291 0.254 44 247
ebizzy menu 7 10732.00 +0.11% 1144 313 0.274 42 271
ebizzy menu 8 10763.00 +0.40% 1156 293 0.253 18 275
ebizzy menu 9 10687.00 -0.31% 1106 272 0.246 18 254
ebizzy menu 10 10710.00 -0.09% 1035 253 0.244 26 227
ebizzy menu 11 10756.00 +0.34% 1152 289 0.251 35 254
ebizzy menu 12 10751.00 +0.29% 1148 291 0.253 32 259
ebizzy menu 13 10713.00 -0.07% 1112 282 0.254 32 250
ebizzy menu 14 10725.00 +0.05% 1082 279 0.258 38 241
ebizzy menu 15 10702.00 -0.17% 976 240 0.246 33 207
ebizzy menu 16 10748.00 +0.26% 962 223 0.232 27 196
ebizzy menu 17 10684.00 -0.34% 1058 271 0.256 39 232
ebizzy menu 18 10775.00 +0.51% 1118 292 0.261 32 260
ebizzy menu 19 10774.00 +0.50% 1174 282 0.240 34 248
ebizzy teo 0 10684.00 -0.34% 1088 131 0.120 67 64
ebizzy teo 1 10732.00 +0.11% 1106 92 0.083 42 50
ebizzy teo 2 10720.00 +0.00% 1108 125 0.113 59 66
ebizzy teo 3 10718.00 -0.02% 1096 114 0.104 55 59
ebizzy teo 4 10667.00 -0.49% 1086 159 0.146 59 100
ebizzy teo 5 10674.00 -0.43% 1224 81 0.066 37 44
ebizzy teo 6 10717.00 -0.03% 1054 98 0.093 48 50
ebizzy teo 7 10723.00 +0.03% 1254 102 0.081 42 60
ebizzy teo 8 10667.00 -0.49% 1124 157 0.140 78 79
ebizzy teo 9 10702.00 -0.17% 1090 147 0.135 60 87
ebizzy teo 10 10728.00 +0.07% 1018 142 0.139 57 85
ebizzy teo 11 10647.00 -0.68% 1084 134 0.124 57 77
ebizzy teo 12 10748.00 +0.26% 1082 147 0.136 53 94
ebizzy teo 13 10705.00 -0.14% 1076 121 0.112 54 67
ebizzy teo 14 10750.00 +0.28% 1120 110 0.098 60 50
ebizzy teo 15 10700.00 -0.19% 1094 117 0.107 55 62
ebizzy teo 16 10676.00 -0.41% 944 135 0.143 50 85
ebizzy teo 17 10720.00 +0.00% 902 117 0.130 42 75
ebizzy teo 18 10711.00 -0.08% 1016 130 0.128 54 76
ebizzy teo 19 10753.00 +0.31% 1058 125 0.118 55 70
ebizzy teo-1 0 10752.00 +0.30% 1168 102 0.087 52 50
ebizzy teo-1 1 10742.00 +0.21% 1046 110 0.105 41 69
ebizzy teo-1 2 10711.00 -0.08% 1323 120 0.091 63 57
ebizzy teo-1 3 10706.00 -0.13% 1100 95 0.086 36 59
ebizzy teo-1 4 10693.00 -0.25% 1116 129 0.116 53 76
ebizzy teo-1 5 10747.00 +0.25% 1056 138 0.131 62 76
ebizzy teo-1 6 10683.00 -0.35% 1144 115 0.101 57 58
ebizzy teo-1 7 10743.00 +0.21% 1080 129 0.119 50 79
ebizzy teo-1 8 10674.00 -0.43% 1266 135 0.107 66 69
ebizzy teo-1 9 10646.00 -0.69% 1050 132 0.126 57 75
ebizzy teo-1 10 10701.00 -0.18% 1036 100 0.097 47 53
ebizzy teo-1 11 10666.00 -0.50% 1204 131 0.109 56 75
ebizzy teo-1 12 10736.00 +0.15% 1186 104 0.088 56 48
ebizzy teo-1 13 10711.00 -0.08% 1104 113 0.102 61 52
ebizzy teo-1 14 10696.00 -0.22% 1052 126 0.120 55 71
ebizzy teo-1 15 10678.00 -0.39% 1042 109 0.105 46 63
ebizzy teo-1 16 9879.00 -7.85% 1229 130 0.106 68 62
ebizzy teo-1 17 10649.00 -0.66% 1174 117 0.100 55 62
ebizzy teo-1 18 10738.00 +0.17% 1110 131 0.118 62 69
ebizzy teo-1 19 10705.00 -0.14% 974 115 0.118 43 72
ebizzy teo-2 0 10703.00 -0.16% 1140 143 0.125 83 60
ebizzy teo-2 1 9799.00 -8.59% 1260 104 0.083 66 38
ebizzy teo-2 2 10626.00 -0.88% 1006 93 0.092 57 36
ebizzy teo-2 3 10661.00 -0.55% 1072 90 0.084 51 39
ebizzy teo-2 4 10690.00 -0.28% 1130 108 0.096 55 53
ebizzy teo-2 5 10694.00 -0.24% 1118 114 0.102 71 43
ebizzy teo-2 6 10715.00 -0.05% 1004 111 0.111 63 48
ebizzy teo-2 7 10742.00 +0.21% 1034 119 0.115 67 52
ebizzy teo-2 8 10766.00 +0.43% 986 117 0.119 66 51
ebizzy teo-2 9 10682.00 -0.35% 1178 103 0.087 69 34
ebizzy teo-2 10 10711.00 -0.08% 1128 106 0.094 73 33
ebizzy teo-2 11 9874.00 -7.89% 1039 86 0.083 50 36
ebizzy teo-2 12 10704.00 -0.15% 915 84 0.092 48 36
ebizzy teo-2 13 10699.00 -0.20% 1032 81 0.078 49 32
ebizzy teo-2 14 10737.00 +0.16% 1026 94 0.092 60 34
ebizzy teo-2 15 10640.00 -0.75% 1016 107 0.105 57 50
ebizzy teo-2 16 10786.00 +0.62% 1034 99 0.096 54 45
ebizzy teo-2 17 10655.00 -0.61% 1162 138 0.119 72 66
ebizzy teo-2 18 10808.00 +0.82% 1016 117 0.115 74 43
ebizzy teo-2 19 10732.00 +0.11% 940 113 0.120 62 51
ebizzy teo-3 0 10706.00 -0.13% 1294 104 0.080 54 50
ebizzy teo-3 1 10761.00 +0.38% 960 112 0.117 55 57
ebizzy teo-3 2 10733.00 +0.12% 1178 90 0.076 62 28
ebizzy teo-3 3 10625.00 -0.89% 1318 117 0.089 67 50
ebizzy teo-3 4 10711.00 -0.08% 1126 103 0.091 65 38
ebizzy teo-3 5 10688.00 -0.30% 1216 91 0.075 58 33
ebizzy teo-3 6 10686.00 -0.32% 1092 90 0.082 58 32
ebizzy teo-3 7 10750.00 +0.28% 1058 100 0.095 63 37
ebizzy teo-3 8 10741.00 +0.20% 1098 100 0.091 62 38
ebizzy teo-3 9 10677.00 -0.40% 970 101 0.104 49 52
ebizzy teo-3 10 10698.00 -0.21% 1008 85 0.084 46 39
ebizzy teo-3 11 10721.00 +0.01% 1166 128 0.110 70 58
ebizzy teo-3 12 10753.00 +0.31% 1122 112 0.100 62 50
ebizzy teo-3 13 10730.00 +0.09% 1132 85 0.075 54 31
ebizzy teo-3 14 10731.00 +0.10% 1156 143 0.124 84 59
ebizzy teo-3 15 10707.00 -0.12% 949 106 0.112 64 42
ebizzy teo-3 16 10701.00 -0.18% 1086 108 0.099 68 40
ebizzy teo-3 17 10729.00 +0.08% 1070 93 0.087 48 45
ebizzy teo-3 18 10743.00 +0.21% 1232 120 0.097 71 49
ebizzy teo-3 19 9898.00 -7.67% 1080 112 0.104 65 47
ebizzy teo-4 0 10703.00 -0.16% 1248 72 0.058 59 13
ebizzy teo-4 1 10727.00 +0.07% 1134 77 0.068 64 13
ebizzy teo-4 2 10676.00 -0.41% 1093 71 0.065 65 6
ebizzy teo-4 3 9734.00 -9.20% 1034 75 0.073 67 8
ebizzy teo-4 4 10713.00 -0.07% 1228 92 0.075 72 20
ebizzy teo-4 5 10713.00 -0.07% 1096 82 0.075 76 6
ebizzy teo-4 6 10776.00 +0.52% 988 93 0.094 72 21
ebizzy teo-4 7 9674.00 -9.76% 1176 76 0.065 62 14
ebizzy teo-4 8 9696.00 -9.55% 1054 73 0.069 67 6
ebizzy teo-4 9 10672.00 -0.45% 1206 103 0.085 80 23
ebizzy teo-4 10 10682.00 -0.35% 1052 88 0.084 72 16
ebizzy teo-4 11 10717.00 -0.03% 1160 91 0.078 77 14
ebizzy teo-4 12 10725.00 +0.05% 1054 72 0.068 62 10
ebizzy teo-4 13 10736.00 +0.15% 964 83 0.086 74 9
ebizzy teo-4 14 9881.00 -7.83% 1114 69 0.062 61 8
ebizzy teo-4 15 10707.00 -0.12% 1238 87 0.070 71 16
ebizzy teo-4 16 10717.00 -0.03% 1212 86 0.071 72 14
ebizzy teo-4 17 10741.00 +0.20% 1042 81 0.078 71 10
ebizzy teo-4 18 10698.00 -0.21% 1030 89 0.086 77 12
ebizzy teo-4 19 10735.00 +0.14% 1174 70 0.060 61 9
ebizzy teo-4-p 0 10768.00 +0.45% 1256 113 0.090 58 55
ebizzy teo-4-p 1 10737.00 +0.16% 1240 120 0.097 52 68
ebizzy teo-4-p 2 10632.00 -0.82% 1402 126 0.090 65 61
ebizzy teo-4-p 3 10713.00 -0.07% 1114 112 0.101 58 54
ebizzy teo-4-p 4 10704.00 -0.15% 1062 119 0.112 66 53
ebizzy teo-4-p 5 10671.00 -0.46% 1226 134 0.109 55 79
ebizzy teo-4-p 6 10672.00 -0.45% 1086 111 0.102 45 66
ebizzy teo-4-p 7 10712.00 -0.07% 1044 129 0.124 60 69
ebizzy teo-4-p 8 9863.00 -7.99% 1108 138 0.125 59 79
ebizzy teo-4-p 9 10767.00 +0.44% 1066 150 0.141 58 92
ebizzy teo-4-p 10 10679.00 -0.38% 1036 123 0.119 46 77
ebizzy teo-4-p 11 10693.00 -0.25% 1261 124 0.098 60 64
ebizzy teo-4-p 12 10690.00 -0.28% 1106 124 0.112 57 67
ebizzy teo-4-p 13 10687.00 -0.31% 1044 131 0.125 59 72
ebizzy teo-4-p 14 10748.00 +0.26% 1188 114 0.096 51 63
ebizzy teo-4-p 15 10747.00 +0.25% 1184 125 0.106 51 74
ebizzy teo-4-p 16 10713.00 -0.07% 1244 139 0.112 74 65
ebizzy teo-4-p 17 10674.00 -0.43% 1036 125 0.121 48 77
ebizzy teo-4-p 18 10681.00 -0.36% 1386 132 0.095 59 73
ebizzy teo-4-p 19 10735.00 +0.14% 1224 119 0.097 65 54
ebizzy teo-5 0 10730.00 +0.09% 1054 73 0.069 67 6
ebizzy teo-5 1 10697.00 -0.21% 1152 87 0.076 80 7
ebizzy teo-5 2 10728.00 +0.07% 1311 70 0.053 64 6
ebizzy teo-5 3 10762.00 +0.39% 1248 78 0.062 68 10
ebizzy teo-5 4 10667.00 -0.49% 1108 84 0.076 75 9
ebizzy teo-5 5 10665.00 -0.51% 954 97 0.102 71 26
ebizzy teo-5 6 10657.00 -0.59% 908 85 0.094 75 10
ebizzy teo-5 7 10643.00 -0.72% 1104 76 0.069 65 11
ebizzy teo-5 8 10714.00 -0.06% 1298 83 0.064 77 6
ebizzy teo-5 9 9937.00 -7.30% 1328 76 0.057 71 5
ebizzy teo-5 10 10722.00 +0.02% 1197 76 0.063 67 9
ebizzy teo-5 11 10683.00 -0.35% 1082 85 0.079 75 10
ebizzy teo-5 12 10707.00 -0.12% 1010 78 0.077 69 9
ebizzy teo-5 13 10697.00 -0.21% 1234 72 0.058 66 6
ebizzy teo-5 14 10650.00 -0.65% 1252 82 0.065 70 12
ebizzy teo-5 15 10718.00 -0.02% 1147 69 0.060 61 8
ebizzy teo-5 16 10685.00 -0.33% 1222 90 0.074 77 13
ebizzy teo-5 17 10756.00 +0.34% 898 59 0.066 51 8
ebizzy teo-5 18 10540.00 -1.68% 1100 86 0.078 78 8
ebizzy teo-5 19 10731.00 +0.10% 996 70 0.070 61 9
ebizzy teo-5-p 0 10715.00 -0.05% 1048 142 0.135 66 76
ebizzy teo-5-p 1 10725.00 +0.05% 1022 122 0.119 48 74
ebizzy teo-5-p 2 10710.00 -0.09% 1116 116 0.104 57 59
ebizzy teo-5-p 3 10681.00 -0.36% 1298 130 0.100 67 63
ebizzy teo-5-p 4 9797.00 -8.61% 1170 120 0.103 67 53
ebizzy teo-5-p 5 10733.00 +0.12% 1202 142 0.118 65 77
ebizzy teo-5-p 6 10735.00 +0.14% 1128 118 0.105 65 53
ebizzy teo-5-p 7 10693.00 -0.25% 1110 129 0.116 59 70
ebizzy teo-5-p 8 10689.00 -0.29% 1174 119 0.101 52 67
ebizzy teo-5-p 9 10651.00 -0.64% 1022 135 0.132 57 78
ebizzy teo-5-p 10 10803.00 +0.77% 1106 114 0.103 48 66
ebizzy teo-5-p 11 10712.00 -0.07% 952 111 0.117 46 65
ebizzy teo-5-p 12 10659.00 -0.57% 1142 122 0.107 54 68
ebizzy teo-5-p 13 10671.00 -0.46% 1086 121 0.111 52 69
ebizzy teo-5-p 14 10676.00 -0.41% 1120 138 0.123 66 72
ebizzy teo-5-p 15 10793.00 +0.68% 1128 120 0.106 57 63
ebizzy teo-5-p 16 10655.00 -0.61% 1276 152 0.119 77 75
ebizzy teo-5-p 17 10644.00 -0.71% 990 110 0.111 51 59
ebizzy teo-5-p 18 10682.00 -0.35% 1096 122 0.111 49 73
ebizzy teo-5-p 19 10739.00 +0.18% 988 111 0.112 38 73
adrestia menu 0 12.00 +0.00% 103068 684 0.007 423 261
adrestia menu 1 12.00 +0.00% 102916 866 0.008 613 253
adrestia menu 2 12.00 +0.00% 102962 682 0.007 437 245
adrestia menu 3 12.00 +0.00% 103220 997 0.010 723 274
adrestia menu 4 12.00 +0.00% 102894 786 0.008 548 238
adrestia menu 5 12.00 +0.00% 102774 730 0.007 491 239
adrestia menu 6 12.00 +0.00% 102906 739 0.007 522 217
adrestia menu 7 12.00 +0.00% 103096 748 0.007 486 262
adrestia menu 8 12.00 +0.00% 102952 812 0.008 515 297
adrestia menu 9 12.00 +0.00% 102904 756 0.007 535 221
adrestia menu 10 12.00 +0.00% 103130 811 0.008 532 279
adrestia menu 11 12.00 +0.00% 102976 834 0.008 570 264
adrestia menu 12 12.00 +0.00% 102926 709 0.007 492 217
adrestia menu 13 12.00 +0.00% 103128 831 0.008 610 221
adrestia menu 14 12.00 +0.00% 102994 897 0.009 633 264
adrestia menu 15 12.00 +0.00% 103116 711 0.007 463 248
adrestia menu 16 12.00 +0.00% 102954 658 0.006 414 244
adrestia menu 17 12.00 +0.00% 103124 787 0.008 450 337
adrestia menu 18 12.00 +0.00% 103102 492 0.005 185 307
adrestia menu 19 12.00 +0.00% 103228 937 0.009 661 276
adrestia teo 0 12.00 +0.00% 103092 155 0.002 51 104
adrestia teo 1 12.00 +0.00% 102944 158 0.002 59 99
adrestia teo 2 12.00 +0.00% 103038 144 0.001 45 99
adrestia teo 3 12.00 +0.00% 102906 166 0.002 53 113
adrestia teo 4 12.00 +0.00% 103136 141 0.001 53 88
adrestia teo 5 12.00 +0.00% 103246 191 0.002 66 125
adrestia teo 6 8.00 -33.33% 103168 134 0.001 41 93
adrestia teo 7 12.00 +0.00% 103068 196 0.002 62 134
adrestia teo 8 11.00 -8.33% 103412 188 0.002 62 126
adrestia teo 9 12.00 +0.00% 103208 130 0.001 55 75
adrestia teo 10 12.00 +0.00% 103186 221 0.002 72 149
adrestia teo 11 12.00 +0.00% 103022 178 0.002 57 121
adrestia teo 12 12.00 +0.00% 103240 158 0.002 63 95
adrestia teo 13 12.00 +0.00% 102862 119 0.001 44 75
adrestia teo 14 12.00 +0.00% 102934 144 0.001 56 88
adrestia teo 15 12.00 +0.00% 102802 117 0.001 44 73
adrestia teo 16 12.00 +0.00% 102844 122 0.001 45 77
adrestia teo 17 12.00 +0.00% 103084 127 0.001 53 74
adrestia teo 18 12.00 +0.00% 103034 144 0.001 54 90
adrestia teo 19 12.00 +0.00% 103016 170 0.002 59 111
adrestia teo-1 0 12.00 +0.00% 102932 151 0.001 52 99
adrestia teo-1 1 12.00 +0.00% 102870 117 0.001 52 65
adrestia teo-1 2 11.00 -8.33% 103108 145 0.001 56 89
adrestia teo-1 3 12.00 +0.00% 103004 141 0.001 47 94
adrestia teo-1 4 12.00 +0.00% 103306 155 0.002 63 92
adrestia teo-1 5 12.00 +0.00% 102936 124 0.001 43 81
adrestia teo-1 6 11.00 -8.33% 103338 193 0.002 62 131
adrestia teo-1 7 12.00 +0.00% 103105 172 0.002 59 113
adrestia teo-1 8 11.00 -8.33% 103560 200 0.002 76 124
adrestia teo-1 9 12.00 +0.00% 102972 106 0.001 44 62
adrestia teo-1 10 12.00 +0.00% 102912 136 0.001 51 85
adrestia teo-1 11 12.00 +0.00% 102922 161 0.002 52 109
adrestia teo-1 12 12.00 +0.00% 102998 165 0.002 58 107
adrestia teo-1 13 12.00 +0.00% 103104 222 0.002 68 154
adrestia teo-1 14 11.00 -8.33% 103314 175 0.002 50 125
adrestia teo-1 15 12.00 +0.00% 102920 128 0.001 58 70
adrestia teo-1 16 12.00 +0.00% 103324 155 0.002 63 92
adrestia teo-1 17 11.00 -8.33% 103504 163 0.002 56 107
adrestia teo-1 18 12.00 +0.00% 103136 180 0.002 61 119
adrestia teo-1 19 12.00 +0.00% 102893 128 0.001 46 82
adrestia teo-2 0 11.00 -8.33% 103342 136 0.001 50 86
adrestia teo-2 1 12.00 +0.00% 102947 128 0.001 51 77
adrestia teo-2 2 11.00 -8.33% 103462 160 0.002 71 89
adrestia teo-2 3 12.00 +0.00% 103248 184 0.002 78 106
adrestia teo-2 4 12.00 +0.00% 103174 105 0.001 59 46
adrestia teo-2 5 12.00 +0.00% 103134 133 0.001 58 75
adrestia teo-2 6 12.00 +0.00% 103374 147 0.001 70 77
adrestia teo-2 7 12.00 +0.00% 103024 135 0.001 60 75
adrestia teo-2 8 12.00 +0.00% 103418 139 0.001 63 76
adrestia teo-2 9 12.00 +0.00% 103084 149 0.001 68 81
adrestia teo-2 10 12.00 +0.00% 103286 178 0.002 73 105
adrestia teo-2 11 12.00 +0.00% 102846 145 0.001 60 85
adrestia teo-2 12 12.00 +0.00% 102960 176 0.002 69 107
adrestia teo-2 13 8.00 -33.33% 103098 132 0.001 67 65
adrestia teo-2 14 12.00 +0.00% 102990 146 0.001 69 77
adrestia teo-2 15 12.00 +0.00% 103018 119 0.001 52 67
adrestia teo-2 16 12.00 +0.00% 103194 189 0.002 81 108
adrestia teo-2 17 12.00 +0.00% 103060 129 0.001 67 62
adrestia teo-2 18 12.00 +0.00% 103186 186 0.002 71 115
adrestia teo-2 19 12.00 +0.00% 103182 190 0.002 75 115
adrestia teo-3 0 12.00 +0.00% 102876 126 0.001 55 71
adrestia teo-3 1 12.00 +0.00% 102980 133 0.001 60 73
adrestia teo-3 2 12.00 +0.00% 103184 112 0.001 55 57
adrestia teo-3 3 12.00 +0.00% 103030 161 0.002 80 81
adrestia teo-3 4 11.00 -8.33% 103386 139 0.001 64 75
adrestia teo-3 5 12.00 +0.00% 103176 141 0.001 59 82
adrestia teo-3 6 12.00 +0.00% 103212 159 0.002 59 100
adrestia teo-3 7 12.00 +0.00% 103038 123 0.001 53 70
adrestia teo-3 8 12.00 +0.00% 103052 123 0.001 70 53
adrestia teo-3 9 11.00 -8.33% 103364 130 0.001 59 71
adrestia teo-3 10 8.00 -33.33% 103506 109 0.001 56 53
adrestia teo-3 11 12.00 +0.00% 103140 123 0.001 63 60
adrestia teo-3 12 12.00 +0.00% 102926 122 0.001 55 67
adrestia teo-3 13 12.00 +0.00% 103012 149 0.001 74 75
adrestia teo-3 14 12.00 +0.00% 103202 145 0.001 67 78
adrestia teo-3 15 12.00 +0.00% 102918 128 0.001 59 69
adrestia teo-3 16 12.00 +0.00% 103070 143 0.001 67 76
adrestia teo-3 17 11.00 -8.33% 103308 148 0.001 67 81
adrestia teo-3 18 12.00 +0.00% 102996 118 0.001 55 63
adrestia teo-3 19 11.00 -8.33% 103438 155 0.001 64 91
adrestia teo-4 0 12.00 +0.00% 103080 139 0.001 97 42
adrestia teo-4 1 12.00 +0.00% 103184 93 0.001 54 39
adrestia teo-4 2 12.00 +0.00% 103506 173 0.002 107 66
adrestia teo-4 3 12.00 +0.00% 103244 129 0.001 73 56
adrestia teo-4 4 12.00 +0.00% 103042 150 0.001 97 53
adrestia teo-4 5 12.00 +0.00% 102928 90 0.001 63 27
adrestia teo-4 6 12.00 +0.00% 102952 82 0.001 59 23
adrestia teo-4 7 12.00 +0.00% 103020 135 0.001 70 65
adrestia teo-4 8 12.00 +0.00% 103022 126 0.001 76 50
adrestia teo-4 9 12.00 +0.00% 102866 117 0.001 67 50
adrestia teo-4 10 12.00 +0.00% 103114 116 0.001 80 36
adrestia teo-4 11 12.00 +0.00% 102906 75 0.001 50 25
adrestia teo-4 12 12.00 +0.00% 103068 138 0.001 78 60
adrestia teo-4 13 12.00 +0.00% 103248 98 0.001 68 30
adrestia teo-4 14 8.00 -33.33% 103488 132 0.001 81 51
adrestia teo-4 15 12.00 +0.00% 103010 108 0.001 76 32
adrestia teo-4 16 12.00 +0.00% 103062 114 0.001 72 42
adrestia teo-4 17 12.00 +0.00% 103234 140 0.001 83 57
adrestia teo-4 18 12.00 +0.00% 102980 80 0.001 55 25
adrestia teo-4 19 12.00 +0.00% 103060 92 0.001 60 32
adrestia teo-4-p 0 12.00 +0.00% 102948 165 0.002 60 105
adrestia teo-4-p 1 12.00 +0.00% 103114 150 0.001 55 95
adrestia teo-4-p 2 12.00 +0.00% 103168 139 0.001 58 81
adrestia teo-4-p 3 11.00 -8.33% 103085 154 0.001 62 92
adrestia teo-4-p 4 12.00 +0.00% 103060 153 0.001 60 93
adrestia teo-4-p 5 12.00 +0.00% 103148 189 0.002 70 119
adrestia teo-4-p 6 12.00 +0.00% 102918 118 0.001 44 74
adrestia teo-4-p 7 12.00 +0.00% 103056 120 0.001 55 65
adrestia teo-4-p 8 12.00 +0.00% 102996 159 0.002 59 100
adrestia teo-4-p 9 12.00 +0.00% 103184 118 0.001 49 69
adrestia teo-4-p 10 12.00 +0.00% 102948 139 0.001 56 83
adrestia teo-4-p 11 11.00 -8.33% 103558 129 0.001 48 81
adrestia teo-4-p 12 12.00 +0.00% 103080 165 0.002 60 105
adrestia teo-4-p 13 11.00 -8.33% 103268 140 0.001 51 89
adrestia teo-4-p 14 8.00 -33.33% 103150 129 0.001 52 77
adrestia teo-4-p 15 12.00 +0.00% 103186 208 0.002 79 129
adrestia teo-4-p 16 12.00 +0.00% 103054 118 0.001 41 77
adrestia teo-4-p 17 12.00 +0.00% 103141 160 0.002 55 105
adrestia teo-4-p 18 12.00 +0.00% 103376 168 0.002 58 110
adrestia teo-4-p 19 12.00 +0.00% 103562 215 0.002 74 141
adrestia teo-5 0 12.00 +0.00% 103032 109 0.001 59 50
adrestia teo-5 1 12.00 +0.00% 103002 122 0.001 63 59
adrestia teo-5 2 12.00 +0.00% 103168 182 0.002 105 77
adrestia teo-5 3 8.00 -33.33% 103202 114 0.001 67 47
adrestia teo-5 4 12.00 +0.00% 103050 173 0.002 101 72
adrestia teo-5 5 12.00 +0.00% 103076 92 0.001 58 34
adrestia teo-5 6 12.00 +0.00% 103064 105 0.001 72 33
adrestia teo-5 7 12.00 +0.00% 103032 134 0.001 78 56
adrestia teo-5 8 12.00 +0.00% 103118 150 0.001 84 66
adrestia teo-5 9 12.00 +0.00% 103126 111 0.001 72 39
adrestia teo-5 10 8.00 -33.33% 103358 114 0.001 71 43
adrestia teo-5 11 12.00 +0.00% 102866 86 0.001 56 30
adrestia teo-5 12 12.00 +0.00% 103078 135 0.001 87 48
adrestia teo-5 13 12.00 +0.00% 103092 172 0.002 95 77
adrestia teo-5 14 12.00 +0.00% 103010 111 0.001 78 33
adrestia teo-5 15 12.00 +0.00% 103090 121 0.001 80 41
adrestia teo-5 16 12.00 +0.00% 102968 185 0.002 101 84
adrestia teo-5 17 11.00 -8.33% 103294 174 0.002 97 77
adrestia teo-5 18 12.00 +0.00% 103060 85 0.001 62 23
adrestia teo-5 19 11.00 -8.33% 103262 166 0.002 97 69
adrestia teo-5-p 0 12.00 +0.00% 102998 154 0.001 64 90
adrestia teo-5-p 1 12.00 +0.00% 103112 150 0.001 57 93
adrestia teo-5-p 2 12.00 +0.00% 102936 126 0.001 57 69
adrestia teo-5-p 3 12.00 +0.00% 103044 129 0.001 54 75
adrestia teo-5-p 4 12.00 +0.00% 102978 135 0.001 62 73
adrestia teo-5-p 5 12.00 +0.00% 103000 124 0.001 41 83
adrestia teo-5-p 6 12.00 +0.00% 103310 155 0.002 64 91
adrestia teo-5-p 7 11.00 -8.33% 103392 210 0.002 70 140
adrestia teo-5-p 8 12.00 +0.00% 103068 167 0.002 54 113
adrestia teo-5-p 9 12.00 +0.00% 102910 114 0.001 39 75
adrestia teo-5-p 10 12.00 +0.00% 102854 98 0.001 46 52
adrestia teo-5-p 11 12.00 +0.00% 103050 151 0.001 55 96
adrestia teo-5-p 12 12.00 +0.00% 102982 144 0.001 46 98
adrestia teo-5-p 13 12.00 +0.00% 103028 123 0.001 49 74
adrestia teo-5-p 14 12.00 +0.00% 103154 175 0.002 54 121
adrestia teo-5-p 15 12.00 +0.00% 103286 158 0.002 64 94
adrestia teo-5-p 16 8.00 -33.33% 103516 170 0.002 63 107
adrestia teo-5-p 17 12.00 +0.00% 102980 137 0.001 50 87
adrestia teo-5-p 18 11.00 -8.33% 103224 151 0.001 55 96
adrestia teo-5-p 19 11.00 -8.33% 103156 170 0.002 67 103
hackbench menu 0 21.78 +0.00% 11352 1135 0.100 892 243
hackbench menu 1 22.28 +2.31% 9280 703 0.076 458 245
hackbench menu 2 22.08 +1.40% 3744 722 0.193 460 262
hackbench menu 3 21.85 +0.32% 5510 672 0.122 467 205
hackbench menu 4 21.91 +0.60% 7222 766 0.106 526 240
hackbench menu 5 21.96 +0.84% 6620 583 0.088 376 207
hackbench menu 6 21.73 -0.24% 3252 638 0.196 395 243
hackbench menu 7 22.28 +2.31% 5446 842 0.155 577 265
hackbench menu 8 21.94 +0.72% 4576 787 0.172 562 225
hackbench menu 9 21.76 -0.09% 3292 648 0.197 458 190
hackbench menu 10 22.07 +1.32% 6908 837 0.121 575 262
hackbench menu 11 21.96 +0.82% 3584 773 0.216 508 265
hackbench menu 12 22.22 +2.04% 4322 803 0.186 570 233
hackbench menu 13 22.11 +1.53% 4750 764 0.161 530 234
hackbench menu 14 21.87 +0.43% 5262 673 0.128 468 205
hackbench menu 15 22.04 +1.18% 4676 756 0.162 577 179
hackbench menu 16 21.85 +0.33% 5482 753 0.137 435 318
hackbench menu 17 21.88 +0.48% 3020 662 0.219 453 209
hackbench menu 18 22.15 +1.72% 3694 846 0.229 611 235
hackbench menu 19 21.98 +0.92% 8648 773 0.089 579 194
hackbench teo 0 22.15 +1.69% 8956 180 0.020 59 121
hackbench teo 1 21.80 +0.08% 5542 180 0.032 49 131
hackbench teo 2 21.89 +0.51% 3324 185 0.056 58 127
hackbench teo 3 22.03 +1.14% 3352 188 0.056 60 128
hackbench teo 4 21.97 +0.89% 3676 170 0.046 58 112
hackbench teo 5 22.16 +1.75% 7590 168 0.022 54 114
hackbench teo 6 22.07 +1.32% 6798 182 0.027 57 125
hackbench teo 7 21.96 +0.81% 3398 176 0.052 58 118
hackbench teo 8 21.86 +0.38% 3544 190 0.054 65 125
hackbench teo 9 22.01 +1.05% 5126 187 0.036 62 125
hackbench teo 10 21.90 +0.57% 4076 184 0.045 53 131
hackbench teo 11 21.92 +0.67% 6324 192 0.030 63 129
hackbench teo 12 22.03 +1.15% 5900 181 0.031 59 122
hackbench teo 13 22.02 +1.11% 5850 154 0.026 38 116
hackbench teo 14 22.01 +1.06% 8394 179 0.021 50 129
hackbench teo 15 22.10 +1.48% 3736 190 0.051 48 142
hackbench teo 16 22.17 +1.81% 7300 172 0.024 45 127
hackbench teo 17 22.10 +1.48% 4900 161 0.033 46 115
hackbench teo 18 21.91 +0.60% 4240 186 0.044 64 122
hackbench teo 19 22.09 +1.44% 4212 164 0.039 46 118
hackbench teo-1 0 21.96 +0.81% 6384 193 0.030 61 132
hackbench teo-1 1 22.02 +1.10% 9136 183 0.020 55 128
hackbench teo-1 2 21.94 +0.73% 8724 198 0.023 71 127
hackbench teo-1 3 21.97 +0.88% 7806 163 0.021 51 112
hackbench teo-1 4 21.93 +0.69% 9348 183 0.020 57 126
hackbench teo-1 5 22.02 +1.09% 5600 185 0.033 55 130
hackbench teo-1 6 21.93 +0.67% 4978 204 0.041 67 137
hackbench teo-1 7 22.15 +1.69% 4718 175 0.037 48 127
hackbench teo-1 8 22.00 +1.03% 5786 187 0.032 61 126
hackbench teo-1 9 21.92 +0.66% 3990 161 0.040 55 106
hackbench teo-1 10 21.95 +0.78% 8884 202 0.023 54 148
hackbench teo-1 11 22.02 +1.12% 5890 184 0.031 57 127
hackbench teo-1 12 21.90 +0.56% 3722 159 0.043 53 106
hackbench teo-1 13 21.93 +0.69% 3434 201 0.059 60 141
hackbench teo-1 14 21.86 +0.39% 3156 170 0.054 46 124
hackbench teo-1 15 21.98 +0.90% 3540 157 0.044 60 97
hackbench teo-1 16 22.05 +1.26% 3912 159 0.041 40 119
hackbench teo-1 17 22.07 +1.36% 7250 179 0.025 55 124
hackbench teo-1 18 22.00 +1.01% 4840 193 0.040 58 135
hackbench teo-1 19 21.98 +0.94% 11040 170 0.015 56 114
hackbench teo-2 0 22.06 +1.28% 4466 163 0.036 55 108
hackbench teo-2 1 22.03 +1.16% 7262 181 0.025 56 125
hackbench teo-2 2 22.01 +1.06% 3272 155 0.047 60 95
hackbench teo-2 3 21.97 +0.88% 6170 177 0.029 52 125
hackbench teo-2 4 21.98 +0.92% 5216 177 0.034 59 118
hackbench teo-2 5 21.95 +0.77% 12813 183 0.014 66 117
hackbench teo-2 6 22.01 +1.07% 5840 210 0.036 78 132
hackbench teo-2 7 21.97 +0.89% 4870 170 0.035 58 112
hackbench teo-2 8 22.01 +1.06% 3824 199 0.052 65 134
hackbench teo-2 9 22.00 +1.00% 5240 163 0.031 61 102
hackbench teo-2 10 21.94 +0.76% 4692 171 0.036 62 109
hackbench teo-2 11 21.96 +0.85% 4734 180 0.038 68 112
hackbench teo-2 12 21.77 -0.05% 4114 178 0.043 66 112
hackbench teo-2 13 21.83 +0.24% 3474 156 0.045 59 97
hackbench teo-2 14 21.94 +0.74% 4514 170 0.038 57 113
hackbench teo-2 15 22.03 +1.13% 4404 175 0.040 71 104
hackbench teo-2 16 22.03 +1.14% 6342 164 0.026 61 103
hackbench teo-2 17 21.87 +0.41% 5186 177 0.034 65 112
hackbench teo-2 18 21.98 +0.94% 7058 188 0.027 53 135
hackbench teo-2 19 21.79 +0.04% 3798 170 0.045 65 105
hackbench teo-3 0 21.93 +0.72% 5495 182 0.033 63 119
hackbench teo-3 1 21.84 +0.27% 3108 162 0.052 54 108
hackbench teo-3 2 22.17 +1.79% 6460 176 0.027 58 118
hackbench teo-3 3 21.95 +0.77% 3988 162 0.041 60 102
hackbench teo-3 4 21.99 +0.96% 5474 192 0.035 67 125
hackbench teo-3 5 22.19 +1.88% 3656 161 0.044 64 97
hackbench teo-3 6 21.77 -0.06% 5270 181 0.034 62 119
hackbench teo-3 7 21.85 +0.32% 3986 198 0.050 73 125
hackbench teo-3 8 21.92 +0.67% 6596 170 0.026 62 108
hackbench teo-3 9 22.08 +1.39% 3696 190 0.051 63 127
hackbench teo-3 10 21.88 +0.45% 4857 168 0.035 68 100
hackbench teo-3 11 21.76 -0.09% 4829 193 0.040 61 132
hackbench teo-3 12 22.21 +1.97% 4142 178 0.043 63 115
hackbench teo-3 13 21.94 +0.73% 3326 175 0.053 67 108
hackbench teo-3 14 21.92 +0.63% 4380 156 0.036 55 101
hackbench teo-3 15 22.00 +1.01% 3535 146 0.041 56 90
hackbench teo-3 16 22.08 +1.37% 4076 181 0.044 73 108
hackbench teo-3 17 22.03 +1.17% 6474 158 0.024 64 94
hackbench teo-3 18 21.83 +0.22% 6770 191 0.028 59 132
hackbench teo-3 19 21.97 +0.86% 5226 133 0.025 48 85
hackbench teo-4 0 22.05 +1.26% 4066 166 0.041 78 88
hackbench teo-4 1 21.85 +0.31% 3594 130 0.036 66 64
hackbench teo-4 2 22.09 +1.41% 3249 148 0.046 80 68
hackbench teo-4 3 21.88 +0.44% 5356 155 0.029 71 84
hackbench teo-4 4 22.01 +1.06% 4972 138 0.028 70 68
hackbench teo-4 5 21.91 +0.62% 6214 152 0.024 74 78
hackbench teo-4 6 21.91 +0.62% 6490 155 0.024 69 86
hackbench teo-4 7 21.94 +0.75% 3988 140 0.035 70 70
hackbench teo-4 8 21.88 +0.48% 4420 170 0.038 70 100
hackbench teo-4 9 22.05 +1.23% 4284 151 0.035 75 76
hackbench teo-4 10 22.22 +2.04% 8606 169 0.020 71 98
hackbench teo-4 11 21.86 +0.39% 3188 150 0.047 82 68
hackbench teo-4 12 21.97 +0.87% 3828 159 0.042 77 82
hackbench teo-4 13 21.97 +0.87% 3918 145 0.037 60 85
hackbench teo-4 14 21.88 +0.45% 3208 134 0.042 72 62
hackbench teo-4 15 21.91 +0.60% 3770 154 0.041 69 85
hackbench teo-4 16 21.96 +0.85% 4094 164 0.040 77 87
hackbench teo-4 17 21.95 +0.77% 5058 171 0.034 80 91
hackbench teo-4 18 22.04 +1.21% 5792 167 0.029 80 87
hackbench teo-4 19 21.82 +0.17% 4026 145 0.036 71 74
hackbench teo-4-p 0 21.82 +0.18% 3174 175 0.055 43 132
hackbench teo-4-p 1 21.66 -0.53% 4894 177 0.036 44 133
hackbench teo-4-p 2 21.82 +0.21% 6162 196 0.032 67 129
hackbench teo-4-p 3 21.86 +0.35% 8884 172 0.019 55 117
hackbench teo-4-p 4 21.96 +0.82% 3350 191 0.057 59 132
hackbench teo-4-p 5 21.99 +0.98% 5858 173 0.030 49 124
hackbench teo-4-p 6 22.03 +1.15% 3890 183 0.047 62 121
hackbench teo-4-p 7 22.10 +1.46% 7378 194 0.026 51 143
hackbench teo-4-p 8 22.09 +1.41% 6200 166 0.027 56 110
hackbench teo-4-p 9 21.99 +0.98% 4516 204 0.045 84 120
hackbench teo-4-p 10 22.01 +1.07% 4744 206 0.043 60 146
hackbench teo-4-p 11 22.01 +1.07% 3170 197 0.062 43 154
hackbench teo-4-p 12 21.94 +0.76% 4346 181 0.042 51 130
hackbench teo-4-p 13 22.02 +1.09% 6724 208 0.031 68 140
hackbench teo-4-p 14 22.00 +1.03% 3406 177 0.052 56 121
hackbench teo-4-p 15 22.04 +1.19% 6078 204 0.034 63 141
hackbench teo-4-p 16 22.02 +1.08% 5722 161 0.028 45 116
hackbench teo-4-p 17 21.97 +0.88% 6806 166 0.024 49 117
hackbench teo-4-p 18 22.08 +1.39% 3750 145 0.039 40 105
hackbench teo-4-p 19 22.07 +1.36% 6414 196 0.031 57 139
hackbench teo-5 0 21.91 +0.61% 3946 145 0.037 69 76
hackbench teo-5 1 21.91 +0.62% 3800 149 0.039 75 74
hackbench teo-5 2 21.99 +0.95% 6420 168 0.026 73 95
hackbench teo-5 3 21.77 -0.06% 4950 150 0.030 72 78
hackbench teo-5 4 22.05 +1.23% 4992 156 0.031 76 80
hackbench teo-5 5 21.98 +0.90% 5786 152 0.026 79 73
hackbench teo-5 6 22.04 +1.20% 3976 142 0.036 70 72
hackbench teo-5 7 22.03 +1.17% 4436 157 0.035 70 87
hackbench teo-5 8 21.86 +0.39% 5742 151 0.026 67 84
hackbench teo-5 9 22.15 +1.71% 11334 137 0.012 65 72
hackbench teo-5 10 22.00 +1.00% 3242 170 0.052 76 94
hackbench teo-5 11 22.00 +1.02% 5508 147 0.027 76 71
hackbench teo-5 12 22.01 +1.07% 7898 154 0.019 73 81
hackbench teo-5 13 21.97 +0.86% 9344 172 0.018 81 91
hackbench teo-5 14 22.00 +1.00% 5216 149 0.029 77 72
hackbench teo-5 15 21.90 +0.54% 4326 163 0.038 73 90
hackbench teo-5 16 22.00 +1.01% 6686 153 0.023 60 93
hackbench teo-5 17 22.06 +1.30% 4696 144 0.031 65 79
hackbench teo-5 18 22.06 +1.30% 7710 159 0.021 78 81
hackbench teo-5 19 22.02 +1.13% 6468 134 0.021 65 69
hackbench teo-5-p 0 22.00 +1.04% 6954 162 0.023 46 116
hackbench teo-5-p 1 22.25 +2.18% 9166 192 0.021 39 153
hackbench teo-5-p 2 21.95 +0.79% 4250 177 0.042 54 123
hackbench teo-5-p 3 22.07 +1.31% 4686 191 0.041 57 134
hackbench teo-5-p 4 21.89 +0.49% 5596 193 0.034 59 134
hackbench teo-5-p 5 21.89 +0.53% 3522 174 0.049 44 130
hackbench teo-5-p 6 22.11 +1.51% 4254 170 0.040 44 126
hackbench teo-5-p 7 21.99 +0.96% 3872 167 0.043 54 113
hackbench teo-5-p 8 21.93 +0.71% 5740 188 0.033 53 135
hackbench teo-5-p 9 21.98 +0.90% 3254 197 0.061 48 149
hackbench teo-5-p 10 22.07 +1.33% 4518 195 0.043 49 146
hackbench teo-5-p 11 21.98 +0.93% 4034 194 0.048 48 146
hackbench teo-5-p 12 22.02 +1.08% 3624 173 0.048 52 121
hackbench teo-5-p 13 22.04 +1.19% 8528 164 0.019 51 113
hackbench teo-5-p 14 22.08 +1.38% 4810 190 0.040 60 130
hackbench teo-5-p 15 21.96 +0.84% 6498 172 0.026 40 132
hackbench teo-5-p 16 21.84 +0.30% 6934 184 0.027 44 140
hackbench teo-5-p 17 22.08 +1.37% 5470 154 0.028 50 104
hackbench teo-5-p 18 21.95 +0.76% 3830 163 0.043 52 111
hackbench teo-5-p 19 21.96 +0.85% 6318 169 0.027 46 123
^ permalink raw reply [flat|nested] 21+ messages in thread
* Re: [PATCH v1 0/5] cpuidle: governors: teo: Wakeup events classification change and some refinements
2026-01-19 9:53 ` Christian Loehle
@ 2026-01-19 12:09 ` Rafael J. Wysocki
2026-01-19 16:20 ` Doug Smythies
0 siblings, 1 reply; 21+ messages in thread
From: Rafael J. Wysocki @ 2026-01-19 12:09 UTC (permalink / raw)
To: Christian Loehle; +Cc: Rafael J. Wysocki, Linux PM, LKML, Doug Smythies
On Mon, Jan 19, 2026 at 10:53 AM Christian Loehle
<christian.loehle@arm.com> wrote:
>
> On 1/16/26 12:29, Rafael J. Wysocki wrote:
> > On Fri, Jan 16, 2026 at 12:52 PM Christian Loehle
> > <christian.loehle@arm.com> wrote:
> >>
> >> On 1/14/26 19:42, Rafael J. Wysocki wrote:
> >>> Hi All,
> >>>
> >>> This material has been in my local queue for almost a full development cycle,
> >>> so time to post it.
> >>>
> >>> The motivation for the changes in this series is mostly theoretical, but I do
> >>> see some idle power improvements from patch [4/5], for example, but nothing
> >>> specifically worth reporting.
> >>>
> >>> The first patch simply prevents idle states with zero-size bins from being
> >>> selected sometimes when teo_select() runs with stopped tick.
> >>>
> >>> Patch [2/5] avoids counting tick wakeups as intercepts unless there are
> >>> sufficiently many intercepts within the tick period range to assume that
> >>> the tick wakeup may have clobbered a genuine intercept.
> >>>
> >>> Patch [3/5] simply updates a coefficient in one of the inequalities to be
> >>> somewhat easier to interpret (this should be a cosmetic change).
> >>>
> >>> Patch [4/5] changes the criteria used for classifying wakeup events as hits
> >>> or intercepts to (hopefully) make the classification work better for large
> >>> state bins.
> >>>
> >>> Patch [5/5] refines the idle state lookup based on intercepts to first
> >>> consider the state with the maximum intercepts metric, so that state is
> >>> always taken into consideration.
> >>>
> >>> Please see the individual patch changelogs for details.
> >>>
> >>> Thanks!
> >>>
> >>>
> >
> > Hi Christian,
> >
> >> Hi Rafael,
> >> I'll do the in-depth review, but have run some tests already.
> >> They are attached, platform is the usual rk3399.
> >> "teo" is mainline, "teo-$i" is with patches 1..$i applied.
> >
> > Thanks for testing!
> >
> >> There's a regression on teo-4 visible on the intercept heavy IO workloads,
> >> for idle misses that isn't strong enough to reflect in score changes except
> >> for the very slow mtdblock device.
> >> interestingly though there also seems to be a regression in
> >> mapper/dm-slow (dm device with 51ms delay on each IO), which is not
> >> intercept heavy.
> >> Looking at the state residencies it overuses the deepest state2 in
> >> where state1 was preferred for the other teo variants.
> >> I've attached that too for reference.
> >> I'm assuming that is because of the new intercept-logic-exclusion clause.
> >
> > So can you please restore that clause to its previous form, while
> > keeping the other changes in patch 4, and see if the regression is
> > still there?
> >
>
> Yep that restores it, I've attached the full results again with teo-4-p and teo-5-p
> being teo-4 and teo-5 but with
> - if (2 * idx_intercept_sum > cpu_data->total) {
> + if (2 * idx_intercept_sum > cpu_data->total - idx_hit_sum) {
>
> respectively.
>
> To highlight let me just quote the schbench results
> test gov i score %change idles idle_miss miss_rt belows aboves
> schbench teo-4 0 295.87 +5.51% 29014 1309 0.045 1168 141
> schbench teo-4 1 302.20 +7.76% 31642 1408 0.044 1105 303
> schbench teo-4 2 301.07 +7.36% 29966 1602 0.053 1393 209
> schbench teo-4 3 297.00 +5.91% 29982 1529 0.051 1266 263
> schbench teo-4 4 297.20 +5.98% 29432 1419 0.048 1194 225
> schbench teo-4 5 299.47 +6.79% 30262 1381 0.046 1189 192
> schbench teo-4 6 301.23 +7.42% 29810 1469 0.049 1234 235
> schbench teo-4 7 301.00 +7.34% 29916 1511 0.051 1355 156
> schbench teo-4 8 299.17 +6.68% 29246 1444 0.049 1296 148
> schbench teo-4 9 297.77 +6.18% 30306 1399 0.046 1195 204
> schbench teo-4 10 304.03 +8.42% 30904 1346 0.044 1163 183
> schbench teo-4 11 300.73 +7.24% 30716 1276 0.042 1089 187
> schbench teo-4 12 298.13 +6.31% 29872 1314 0.044 1081 233
> schbench teo-4 13 298.73 +6.53% 30386 1288 0.042 1092 196
> schbench teo-4 14 300.77 +7.25% 30150 1323 0.044 1074 249
> schbench teo-4 15 296.23 +5.63% 32178 1328 0.041 1061 267
> schbench teo-4 16 283.50 +1.09% 27472 1282 0.047 1126 156
> schbench teo-4 17 300.53 +7.17% 29644 1552 0.052 1375 177
> schbench teo-4 18 302.23 +7.77% 31456 1418 0.045 1194 224
> schbench teo-4 19 307.43 +9.63% 32518 1259 0.039 968 291
> schbench teo-4-p 0 277.53 -1.03% 28054 503 0.018 151 352
> schbench teo-4-p 1 304.07 +8.43% 30650 553 0.018 178 375
> schbench teo-4-p 2 273.73 -2.39% 27120 509 0.019 163 346
> schbench teo-4-p 3 300.20 +7.05% 29932 528 0.018 171 357
> schbench teo-4-p 4 300.00 +6.98% 29580 584 0.020 180 404
> schbench teo-4-p 5 296.03 +5.56% 30176 514 0.017 172 342
> schbench teo-4-p 6 300.50 +7.16% 29806 550 0.018 174 376
> schbench teo-4-p 7 271.07 -3.34% 26934 472 0.018 165 307
> schbench teo-4-p 8 299.00 +6.62% 30568 564 0.018 175 389
> schbench teo-4-p 9 300.80 +7.26% 29660 482 0.016 188 294
> schbench teo-4-p 10 300.57 +7.18% 30384 634 0.021 195 439
> schbench teo-4-p 11 296.53 +5.74% 29590 599 0.020 221 378
> schbench teo-4-p 12 298.77 +6.54% 30664 599 0.020 177 422
> schbench teo-4-p 13 302.47 +7.86% 30200 559 0.019 223 336
> schbench teo-4-p 14 300.43 +7.13% 30326 562 0.019 166 396
> schbench teo-4-p 15 292.87 +4.44% 28798 524 0.018 142 382
> schbench teo-4-p 16 299.63 +6.85% 29718 535 0.018 189 346
> schbench teo-4-p 17 302.10 +7.73% 31122 584 0.019 221 363
> schbench teo-4-p 18 294.13 +4.89% 28312 560 0.020 167 393
> schbench teo-4-p 19 302.20 +7.76% 29872 556 0.019 168 388
> schbench teo-5 0 299.53 +6.81% 31198 1312 0.042 1062 250
> schbench teo-5 1 296.67 +5.79% 30090 1383 0.046 1169 214
> schbench teo-5 2 299.60 +6.84% 30010 1566 0.052 1331 235
> schbench teo-5 3 300.00 +6.98% 30764 1466 0.048 1239 227
> schbench teo-5 4 298.77 +6.54% 30002 1383 0.046 1212 171
> schbench teo-5 5 299.30 +6.73% 29848 1481 0.050 1288 193
> schbench teo-5 6 298.90 +6.59% 29586 1441 0.049 1260 181
> schbench teo-5 7 300.37 +7.11% 30672 1460 0.048 1306 154
> schbench teo-5 8 301.00 +7.34% 31454 1383 0.044 1190 193
> schbench teo-5 9 298.80 +6.55% 30042 1415 0.047 1194 221
> schbench teo-5 10 302.23 +7.77% 30986 1305 0.042 1066 239
> schbench teo-5 11 297.30 +6.02% 30112 1405 0.047 1190 215
> schbench teo-5 12 299.33 +6.74% 29536 1482 0.050 1214 268
> schbench teo-5 13 288.87 +3.01% 28754 1244 0.043 1038 206
> schbench teo-5 14 300.17 +7.04% 30232 1517 0.050 1300 217
> schbench teo-5 15 301.43 +7.49% 30578 1576 0.052 1287 289
> schbench teo-5 16 282.10 +0.60% 29106 996 0.034 834 162
> schbench teo-5 17 278.03 -0.86% 27432 1123 0.041 926 197
> schbench teo-5 18 304.40 +8.55% 29934 1526 0.051 1296 230
> schbench teo-5 19 276.50 -1.40% 28072 1158 0.041 970 188
> schbench teo-5-p 0 299.47 +6.79% 29994 587 0.020 231 356
> schbench teo-5-p 1 298.27 +6.36% 31600 555 0.018 190 365
> schbench teo-5-p 2 297.20 +5.98% 30350 543 0.018 198 345
> schbench teo-5-p 3 294.30 +4.95% 29962 523 0.017 192 331
> schbench teo-5-p 4 303.30 +8.16% 30366 480 0.016 160 320
> schbench teo-5-p 5 298.87 +6.58% 29838 541 0.018 180 361
> schbench teo-5-p 6 301.63 +7.56% 30022 480 0.016 168 312
> schbench teo-5-p 7 298.47 +6.43% 30210 581 0.019 202 379
> schbench teo-5-p 8 298.77 +6.54% 30636 529 0.017 210 319
> schbench teo-5-p 9 298.63 +6.49% 30474 564 0.019 185 379
> schbench teo-5-p 10 300.10 +7.01% 30808 577 0.019 251 326
> schbench teo-5-p 11 295.87 +5.51% 30190 565 0.019 186 379
> schbench teo-5-p 12 301.27 +7.43% 29958 466 0.016 160 306
> schbench teo-5-p 13 280.83 +0.14% 27916 494 0.018 164 330
> schbench teo-5-p 14 294.23 +4.92% 30330 611 0.020 219 392
> schbench teo-5-p 15 297.13 +5.96% 30728 656 0.021 195 461
> schbench teo-5-p 16 297.97 +6.25% 30242 524 0.017 178 346
> schbench teo-5-p 17 301.27 +7.43% 31004 524 0.017 208 316
> schbench teo-5-p 18 297.97 +6.25% 29864 565 0.019 211 354
> schbench teo-5-p 19 298.43 +6.42% 30462 519 0.017 150 369
>
> The -p variants have less than half the idle_misses and fall into the same order of magnitude
> for this test than all the other teo variants.
Thanks for the data!
I'll send an update of patch 4 without the intercept-logic-exclusion clause.
I thought that it would be necessary, but that does not turn out to be the case.
^ permalink raw reply [flat|nested] 21+ messages in thread
* RE: [PATCH v1 0/5] cpuidle: governors: teo: Wakeup events classification change and some refinements
2026-01-19 12:09 ` Rafael J. Wysocki
@ 2026-01-19 16:20 ` Doug Smythies
0 siblings, 0 replies; 21+ messages in thread
From: Doug Smythies @ 2026-01-19 16:20 UTC (permalink / raw)
To: 'Rafael J. Wysocki', 'Christian Loehle'
Cc: 'Linux PM', 'LKML', Doug Smythies
[-- Attachment #1: Type: text/plain, Size: 1459 bytes --]
Hi All,
I have been testing this series, but have nothing significant to report, so far.
On 2026.01.19 Rafael J. Wysocki wrote:
>On Mon, Jan 19, 2026 at 10:53 AM Christian Loehle wrote:
>> On 1/16/26 12:29, Rafael J. Wysocki wrote:
>>> On Fri, Jan 16, 2026 at 12:52 PM Christian Loehle wrote:
>>>> On 1/14/26 19:42, Rafael J. Wysocki wrote:
...snip...
>>>> There's a regression on teo-4 visible on the intercept heavy IO workloads,
>>>> for idle misses that isn't strong enough to reflect in score changes except
>>>> for the very slow mtdblock device.
>>>> interestingly though there also seems to be a regression in
>>>> mapper/dm-slow (dm device with 51ms delay on each IO), which is not
>>>> intercept heavy.
>>>> Looking at the state residencies it overuses the deepest state2 in
>>>> where state1 was preferred for the other teo variants.
I did observe similar in one test, an increase in the use of idle
state 2. The relevant graphs are attached.
Legend:
rc5 = kernel 6.19-rc5
rjw = kernel 6.19-rc5 + this 5 patch set.
Workflow:
My version of "critical-jobs", an attempt to do similar to the non-free SPECjbb (that Artem sometimes reports on).
Sweep from 4000 to 4600 jobs per second, and 10 disk lookups per job,
and "500" (arbitrary) units of work per lookup. 500 Gigabyte data file:
While "rc5" completed considerably faster than "rjw", I don't yet know if it is repeatable.
... snip ...
... Doug
[-- Attachment #2: 0_residency.png --]
[-- Type: image/png, Size: 57204 bytes --]
[-- Attachment #3: 1_residency.png --]
[-- Type: image/png, Size: 57785 bytes --]
[-- Attachment #4: 2_above.png --]
[-- Type: image/png, Size: 56239 bytes --]
[-- Attachment #5: 2_residency.png --]
[-- Type: image/png, Size: 76961 bytes --]
^ permalink raw reply [flat|nested] 21+ messages in thread
* [Update][PATCH v1.1 4/5] cpuidle: governors: teo: Adjust the classification of wakeup events
2026-01-14 19:42 [PATCH v1 0/5] cpuidle: governors: teo: Wakeup events classification change and some refinements Rafael J. Wysocki
` (5 preceding siblings ...)
2026-01-16 11:52 ` [PATCH v1 0/5] cpuidle: governors: teo: Wakeup events classification change and some refinements Christian Loehle
@ 2026-01-20 15:29 ` Rafael J. Wysocki
2026-01-25 17:21 ` Doug Smythies
6 siblings, 1 reply; 21+ messages in thread
From: Rafael J. Wysocki @ 2026-01-20 15:29 UTC (permalink / raw)
To: Linux PM; +Cc: LKML, Christian Loehle, Doug Smythies
From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
If differences between target residency values of adjacent idle states
of a given CPU are relatively large, the corresponding idle state bins
used by the teo governors are large either and the rule by which hits
are distinguished from intercepts is inaccurate.
Namely, by that rule, a wakeup event is classified as a hit if the
sleep length (the time till the closest timer other than the tick)
and the measured idle duration, adjusted for the entered idle state
exit latency, fall into the same idle state bin. However, if that bin
is large enough, the actual difference between the sleep length and
the measured idle duration may be significant. It may in fact be
significantly greater than the analogous difference for an event where
the sleep length and the measured idle duration fall into different
bins.
For this reason, amend the rule in question with a check that will
only allow a wakeup event to be counted as a hit if the difference
between the sleep length and the measured idle duration is less than
LATENCY_THRESHOLD_NS (which means that the difference between the
sleep length and the raw measured idle duration is below the sum of
LATENCY_THRESHOLD_NS and 1/2 of the entered idle state exit latency).
Otherwise, the event will be counted as an intercept.
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
---
v1 -> v1.1
* Drop the change in teo_select() along with the corresponding
part of the changelog (after receiving testing feedback from
Christian)
---
drivers/cpuidle/governors/teo.c | 32 ++++++++++++++++----------------
1 file changed, 16 insertions(+), 16 deletions(-)
--- a/drivers/cpuidle/governors/teo.c
+++ b/drivers/cpuidle/governors/teo.c
@@ -48,13 +48,11 @@
* in accordance with what happened last time.
*
* The "hits" metric reflects the relative frequency of situations in which the
- * sleep length and the idle duration measured after CPU wakeup fall into the
- * same bin (that is, the CPU appears to wake up "on time" relative to the sleep
- * length). In turn, the "intercepts" metric reflects the relative frequency of
- * non-timer wakeup events for which the measured idle duration falls into a bin
- * that corresponds to an idle state shallower than the one whose bin is fallen
- * into by the sleep length (these events are also referred to as "intercepts"
- * below).
+ * sleep length and the idle duration measured after CPU wakeup are close enough
+ * (that is, the CPU appears to wake up "on time" relative to the sleep length).
+ * In turn, the "intercepts" metric reflects the relative frequency of non-timer
+ * wakeup events for which the measured idle duration is measurably less than
+ * the sleep length (these events are also referred to as "intercepts" below).
*
* The governor also counts "intercepts" with the measured idle duration below
* the tick period length and uses this information when deciding whether or not
@@ -253,12 +251,16 @@ static void teo_update(struct cpuidle_dr
}
/*
- * If the measured idle duration falls into the same bin as the sleep
- * length, this is a "hit", so update the "hits" metric for that bin.
+ * If the measured idle duration falls into the same bin as the
+ * sleep length and the difference between them is less than
+ * LATENCY_THRESHOLD_NS, this is a "hit", so update the "hits"
+ * metric for that bin.
+ *
* Otherwise, update the "intercepts" metric for the bin fallen into by
* the measured idle duration.
*/
- if (idx_timer == idx_duration) {
+ if (idx_timer == idx_duration &&
+ cpu_data->sleep_length_ns - measured_ns < LATENCY_THRESHOLD_NS) {
cpu_data->state_bins[idx_timer].hits += PULSE;
} else {
cpu_data->state_bins[idx_duration].intercepts += PULSE;
^ permalink raw reply [flat|nested] 21+ messages in thread
* Re: [PATCH v1 1/5] cpuidle: governors: teo: Avoid selecting states with zero-size bins
2026-01-14 19:44 ` [PATCH v1 1/5] cpuidle: governors: teo: Avoid selecting states with zero-size bins Rafael J. Wysocki
@ 2026-01-21 13:09 ` Christian Loehle
2026-01-23 20:46 ` Rafael J. Wysocki
0 siblings, 1 reply; 21+ messages in thread
From: Christian Loehle @ 2026-01-21 13:09 UTC (permalink / raw)
To: Rafael J. Wysocki, Linux PM; +Cc: LKML, Doug Smythies
On 1/14/26 19:44, Rafael J. Wysocki wrote:
> From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
>
> If the last two enabled idle states have the same target residency which
> is at least equal to TICK_NSET, teo may select the next-to-last one even
s/TICK_NSET/TICK_NSEC
> though the size of that state's bin is 0, which is confusing.
>
> Prevent that from happening by adding a target residency check to the
> relevant code path.
>
> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
> ---
> drivers/cpuidle/governors/teo.c | 10 ++++++++++
> 1 file changed, 10 insertions(+)
>
> --- a/drivers/cpuidle/governors/teo.c
> +++ b/drivers/cpuidle/governors/teo.c
> @@ -388,6 +388,15 @@ static int teo_select(struct cpuidle_dri
> while (min_idx < idx &&
> drv->states[min_idx].target_residency_ns < TICK_NSEC)
> min_idx++;
> +
> + /*
> + * Avoid selecting a state with a lower index, but with
> + * the same target residency as the current candidate
> + * one.
> + */
> + if (drv->states[min_idx].target_residency_ns ==
> + drv->states[idx].target_residency_ns)
We need to check that min_idx isn't disabled though, otherwise we now skip a
potential (enabled) idx==1 if min_idx==2 and min_idx is disabled.
Other than that LGTM and with that check and the nit above:
Reviewed-by: Christian Loehle <christian.loehle@arm.com>
> + goto constraint;
> }
>
> /*
> @@ -410,6 +419,7 @@ static int teo_select(struct cpuidle_dri
> }
> }
>
> +constraint:
> /*
> * If there is a latency constraint, it may be necessary to select an
> * idle state shallower than the current candidate one.
>
>
>
^ permalink raw reply [flat|nested] 21+ messages in thread
* Re: [PATCH v1 2/5] cpuidle: governors: teo: Avoid fake intercepts produced by tick
2026-01-14 19:44 ` [PATCH v1 2/5] cpuidle: governors: teo: Avoid fake intercepts produced by tick Rafael J. Wysocki
@ 2026-01-21 13:34 ` Christian Loehle
0 siblings, 0 replies; 21+ messages in thread
From: Christian Loehle @ 2026-01-21 13:34 UTC (permalink / raw)
To: Rafael J. Wysocki, Linux PM; +Cc: LKML, Doug Smythies
On 1/14/26 19:44, Rafael J. Wysocki wrote:
> From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
>
> 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>
> ---
> drivers/cpuidle/governors/teo.c | 11 +++++++++++
> 1 file changed, 11 insertions(+)
>
> --- a/drivers/cpuidle/governors/teo.c
> +++ b/drivers/cpuidle/governors/teo.c
> @@ -239,6 +239,17 @@ static void teo_update(struct cpuidle_dr
> 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;
> + }
> }
Reviewed-by: Christian Loehle <christian.loehle@arm.com>
Makes sense to me, let me try to find something that triggers this (often)
^ permalink raw reply [flat|nested] 21+ messages in thread
* Re: [PATCH v1 3/5] cpuidle: governors: teo: Refine tick_intercepts vs total events check
2026-01-14 19:45 ` [PATCH v1 3/5] cpuidle: governors: teo: Refine tick_intercepts vs total events check Rafael J. Wysocki
@ 2026-01-21 13:36 ` Christian Loehle
0 siblings, 0 replies; 21+ messages in thread
From: Christian Loehle @ 2026-01-21 13:36 UTC (permalink / raw)
To: Rafael J. Wysocki, Linux PM; +Cc: LKML, Doug Smythies
On 1/14/26 19:45, Rafael J. Wysocki wrote:
> From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
>
> Use 2/3 as the proportion coefficient in the check comparing
> cpu_data->tick_intercepts with cpu_data->total because it is close
> enough to the current one (5/8) and it allows of more straightforward
> interpretation (on average, intercepts within the tick period length
> are twice as frequent as other events).
>
> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
> ---
> drivers/cpuidle/governors/teo.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> --- a/drivers/cpuidle/governors/teo.c
> +++ b/drivers/cpuidle/governors/teo.c
> @@ -485,7 +485,7 @@ constraint:
> * total wakeup events, do not stop the tick.
> */
> if (drv->states[idx].target_residency_ns < TICK_NSEC &&
> - cpu_data->tick_intercepts > cpu_data->total / 2 + cpu_data->total / 8)
> + 3 * cpu_data->tick_intercepts >= 2 * cpu_data->total)
> duration_ns = TICK_NSEC / 2;
>
Sure, I guess the 2 and 8 was just as arbitrary to avoid a division.
Reviewed-by: Christian Loehle <christian.loehle@arm.com>
^ permalink raw reply [flat|nested] 21+ messages in thread
* Re: [PATCH v1 1/5] cpuidle: governors: teo: Avoid selecting states with zero-size bins
2026-01-21 13:09 ` Christian Loehle
@ 2026-01-23 20:46 ` Rafael J. Wysocki
2026-01-26 9:18 ` Christian Loehle
0 siblings, 1 reply; 21+ messages in thread
From: Rafael J. Wysocki @ 2026-01-23 20:46 UTC (permalink / raw)
To: Christian Loehle; +Cc: Rafael J. Wysocki, Linux PM, LKML, Doug Smythies
On Wed, Jan 21, 2026 at 2:10 PM Christian Loehle
<christian.loehle@arm.com> wrote:
>
> On 1/14/26 19:44, Rafael J. Wysocki wrote:
> > From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
> >
> > If the last two enabled idle states have the same target residency which
> > is at least equal to TICK_NSET, teo may select the next-to-last one even
>
> s/TICK_NSET/TICK_NSEC
Yup, thanks!
> > though the size of that state's bin is 0, which is confusing.
> >
> > Prevent that from happening by adding a target residency check to the
> > relevant code path.
> >
> > Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
> > ---
> > drivers/cpuidle/governors/teo.c | 10 ++++++++++
> > 1 file changed, 10 insertions(+)
> >
> > --- a/drivers/cpuidle/governors/teo.c
> > +++ b/drivers/cpuidle/governors/teo.c
> > @@ -388,6 +388,15 @@ static int teo_select(struct cpuidle_dri
> > while (min_idx < idx &&
> > drv->states[min_idx].target_residency_ns < TICK_NSEC)
> > min_idx++;
> > +
> > + /*
> > + * Avoid selecting a state with a lower index, but with
> > + * the same target residency as the current candidate
> > + * one.
> > + */
> > + if (drv->states[min_idx].target_residency_ns ==
> > + drv->states[idx].target_residency_ns)
>
> We need to check that min_idx isn't disabled though, otherwise we now skip a
> potential (enabled) idx==1 if min_idx==2 and min_idx is disabled.
Not really because idx is the current candidate state and it is
enabled. We'll use idx if this check is true, not min_idx.
So I think I only need to fix the typo above.
> Other than that LGTM and with that check and the nit above:
>
> Reviewed-by: Christian Loehle <christian.loehle@arm.com>
Thanks!
> > + goto constraint;
> > }
> >
> > /*
> > @@ -410,6 +419,7 @@ static int teo_select(struct cpuidle_dri
> > }
> > }
> >
> > +constraint:
> > /*
> > * If there is a latency constraint, it may be necessary to select an
> > * idle state shallower than the current candidate one.
> >
> >
> >
>
^ permalink raw reply [flat|nested] 21+ messages in thread
* RE: [Update][PATCH v1.1 4/5] cpuidle: governors: teo: Adjust the classification of wakeup events
2026-01-20 15:29 ` [Update][PATCH v1.1 4/5] cpuidle: governors: teo: Adjust the classification of wakeup events Rafael J. Wysocki
@ 2026-01-25 17:21 ` Doug Smythies
2026-01-25 19:38 ` Christian Loehle
0 siblings, 1 reply; 21+ messages in thread
From: Doug Smythies @ 2026-01-25 17:21 UTC (permalink / raw)
To: 'Rafael J. Wysocki', 'Christian Loehle'
Cc: 'LKML', Doug Smythies, 'Linux PM'
[-- Attachment #1: Type: text/plain, Size: 3092 bytes --]
On 2026.01.20 07:30 Rafael J. Wysocki wrote:
> From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
>
> If differences between target residency values of adjacent idle states
> of a given CPU are relatively large, the corresponding idle state bins
> used by the teo governors are large either and the rule by which hits
> are distinguished from intercepts is inaccurate.
>
> Namely, by that rule, a wakeup event is classified as a hit if the
> sleep length (the time till the closest timer other than the tick)
> and the measured idle duration, adjusted for the entered idle state
> exit latency, fall into the same idle state bin. However, if that bin
> is large enough, the actual difference between the sleep length and
> the measured idle duration may be significant. It may in fact be
> significantly greater than the analogous difference for an event where
> the sleep length and the measured idle duration fall into different
> bins.
>
> For this reason, amend the rule in question with a check that will
> only allow a wakeup event to be counted as a hit if the difference
> between the sleep length and the measured idle duration is less than
> LATENCY_THRESHOLD_NS (which means that the difference between the
> sleep length and the raw measured idle duration is below the sum of
> LATENCY_THRESHOLD_NS and 1/2 of the entered idle state exit latency).
> Otherwise, the event will be counted as an intercept.
>
> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
> ---
>
> v1 -> v1.1
> * Drop the change in teo_select() along with the corresponding
> part of the changelog (after receiving testing feedback from
> Christian)
With this updated patch I have not observed any difference in testing
results or power consumption between kernels without or with the
5 patch set:
c66de7fc0157 (HEAD -> rjw-1-1) cpuidle: governors: teo: Adjust the classification of wakeup events
25f70be81668 Revert "cpuidle: governors: teo: Adjust the classification of wakeup events"
f0ae302c4635 cpuidle: governors: teo: Refine intercepts-based idle state lookup
f5ad355214de cpuidle: governors: teo: Adjust the classification of wakeup events
1c5b66c336ea cpuidle: governors: teo: Refine tick_intercepts vs total events check
36148eea2ec2 cpuidle: governors: teo: Avoid fake intercepts produced by tick
8b1ad7bc8a7f cpuidle: governors: teo: Avoid selecting states with zero-size bins
0f61b1860cc3 (tag: v6.19-rc5, origin/master, origin/HEAD, master) Linux 6.19-rc5
My test system:
Processor: Intel(R) Core(TM) i5-10600K CPU @ 4.10GHz, 6 cores 12 CPUs.
HWP: Enabled.
state0/name:POLL
state1/name:C1_ACPI
state2/name:C2_ACPI
state3/name:C3_ACPI
@Christian: I noticed that you like "idle misses" in test results. I have added
percent "idle misses" to my test results. An example graph is attached.
Legend:
rc5 = kernel 6.19-rc5
rjw = kernel 6.19-rc5 + original 5 patch set
rjw-1-1 = kernel 6.19-rc5 + current 5 patch set
See also my previous email [1] about the original 5 patch set:
[1] https://lore.kernel.org/linux-pm/003201dc895f$8cfb2540$a6f16fc0$@telus.net/
... Doug
[-- Attachment #2: misses.png --]
[-- Type: image/png, Size: 59639 bytes --]
^ permalink raw reply [flat|nested] 21+ messages in thread
* Re: [Update][PATCH v1.1 4/5] cpuidle: governors: teo: Adjust the classification of wakeup events
2026-01-25 17:21 ` Doug Smythies
@ 2026-01-25 19:38 ` Christian Loehle
0 siblings, 0 replies; 21+ messages in thread
From: Christian Loehle @ 2026-01-25 19:38 UTC (permalink / raw)
To: Doug Smythies, 'Rafael J. Wysocki'
Cc: 'LKML', 'Linux PM'
On 1/25/26 17:21, Doug Smythies wrote:
> On 2026.01.20 07:30 Rafael J. Wysocki wrote:
>> From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
>>
>> If differences between target residency values of adjacent idle states
>> of a given CPU are relatively large, the corresponding idle state bins
>> used by the teo governors are large either and the rule by which hits
>> are distinguished from intercepts is inaccurate.
>>
>> Namely, by that rule, a wakeup event is classified as a hit if the
>> sleep length (the time till the closest timer other than the tick)
>> and the measured idle duration, adjusted for the entered idle state
>> exit latency, fall into the same idle state bin. However, if that bin
>> is large enough, the actual difference between the sleep length and
>> the measured idle duration may be significant. It may in fact be
>> significantly greater than the analogous difference for an event where
>> the sleep length and the measured idle duration fall into different
>> bins.
>>
>> For this reason, amend the rule in question with a check that will
>> only allow a wakeup event to be counted as a hit if the difference
>> between the sleep length and the measured idle duration is less than
>> LATENCY_THRESHOLD_NS (which means that the difference between the
>> sleep length and the raw measured idle duration is below the sum of
>> LATENCY_THRESHOLD_NS and 1/2 of the entered idle state exit latency).
>> Otherwise, the event will be counted as an intercept.
>>
>> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
>> ---
>>
>> v1 -> v1.1
>> * Drop the change in teo_select() along with the corresponding
>> part of the changelog (after receiving testing feedback from
>> Christian)
>
> With this updated patch I have not observed any difference in testing
> results or power consumption between kernels without or with the
> 5 patch set:
>
> c66de7fc0157 (HEAD -> rjw-1-1) cpuidle: governors: teo: Adjust the classification of wakeup events
> 25f70be81668 Revert "cpuidle: governors: teo: Adjust the classification of wakeup events"
> f0ae302c4635 cpuidle: governors: teo: Refine intercepts-based idle state lookup
> f5ad355214de cpuidle: governors: teo: Adjust the classification of wakeup events
> 1c5b66c336ea cpuidle: governors: teo: Refine tick_intercepts vs total events check
> 36148eea2ec2 cpuidle: governors: teo: Avoid fake intercepts produced by tick
> 8b1ad7bc8a7f cpuidle: governors: teo: Avoid selecting states with zero-size bins
> 0f61b1860cc3 (tag: v6.19-rc5, origin/master, origin/HEAD, master) Linux 6.19-rc5
>
> My test system:
> Processor: Intel(R) Core(TM) i5-10600K CPU @ 4.10GHz, 6 cores 12 CPUs.
> HWP: Enabled.
> state0/name:POLL
> state1/name:C1_ACPI
> state2/name:C2_ACPI
> state3/name:C3_ACPI
>
> @Christian: I noticed that you like "idle misses" in test results. I have added
> percent "idle misses" to my test results. An example graph is attached.
Thanks for the test!
So rjw-1-1 looks pretty much the same on rc5 as well, that's good.
I guess since the v1 of 4/5 change that did have an affect was added by Rafael
to mitigate the more liberal intercept counting, that should be fine then.
Interestingly here (in overall idle_misses) rjw (with 4/5 v1) has significantly
fewer idle_misses, while in state2 aboves it was much worse as per your previous
test results.
>
> Legend:
> rc5 = kernel 6.19-rc5
> rjw = kernel 6.19-rc5 + original 5 patch set
> rjw-1-1 = kernel 6.19-rc5 + current 5 patch set
>
> See also my previous email [1] about the original 5 patch set:
>
> [1] https://lore.kernel.org/linux-pm/003201dc895f$8cfb2540$a6f16fc0$@telus.net/
>
> ... Doug
>
^ permalink raw reply [flat|nested] 21+ messages in thread
* Re: [PATCH v1 1/5] cpuidle: governors: teo: Avoid selecting states with zero-size bins
2026-01-23 20:46 ` Rafael J. Wysocki
@ 2026-01-26 9:18 ` Christian Loehle
2026-01-26 11:40 ` Rafael J. Wysocki
0 siblings, 1 reply; 21+ messages in thread
From: Christian Loehle @ 2026-01-26 9:18 UTC (permalink / raw)
To: Rafael J. Wysocki; +Cc: Linux PM, LKML, Doug Smythies
On 1/23/26 20:46, Rafael J. Wysocki wrote:
> On Wed, Jan 21, 2026 at 2:10 PM Christian Loehle
> <christian.loehle@arm.com> wrote:
>>
>> On 1/14/26 19:44, Rafael J. Wysocki wrote:
>>> From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
>>>
>>> If the last two enabled idle states have the same target residency which
>>> is at least equal to TICK_NSET, teo may select the next-to-last one even
>>
>> s/TICK_NSET/TICK_NSEC
>
> Yup, thanks!
>
>>> though the size of that state's bin is 0, which is confusing.
>>>
>>> Prevent that from happening by adding a target residency check to the
>>> relevant code path.
>>>
>>> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
>>> ---
>>> drivers/cpuidle/governors/teo.c | 10 ++++++++++
>>> 1 file changed, 10 insertions(+)
>>>
>>> --- a/drivers/cpuidle/governors/teo.c
>>> +++ b/drivers/cpuidle/governors/teo.c
>>> @@ -388,6 +388,15 @@ static int teo_select(struct cpuidle_dri
>>> while (min_idx < idx &&
>>> drv->states[min_idx].target_residency_ns < TICK_NSEC)
>>> min_idx++;
>>> +
>>> + /*
>>> + * Avoid selecting a state with a lower index, but with
>>> + * the same target residency as the current candidate
>>> + * one.
>>> + */
>>> + if (drv->states[min_idx].target_residency_ns ==
>>> + drv->states[idx].target_residency_ns)
>>
>> We need to check that min_idx isn't disabled though, otherwise we now skip a
>> potential (enabled) idx==1 if min_idx==2 and min_idx is disabled.
>
> Not really because idx is the current candidate state and it is
> enabled. We'll use idx if this check is true, not min_idx.
>
Are you sure?
I meant initially:
for (i = intercept_max_idx; i >= min_idx; i--) {
intercept_sum += cpu_data->state_bins[i].intercepts;
if (dev->states_usage[i].disable)
continue;
idx = i;
if (2 * intercept_sum > idx_intercept_sum)
break;
}
might skip an idx==3 if it enters with min_idx==2 (sorry, messed up the +-1 in the initial mail)
even though idx==3 might have the same residency as idx==2.
So if idx==2 is disabled we could've selected idx==3, but now won't and will go for idx==1 or
whatever is the next shallower enabled state.
Additionally an issue with this and patch 5/5:
if (min_idx >= intercept_max_idx) {
idx = min_idx;
goto constraint; // CL: this will just select min_idx
}
will use min_idx even though it might be disabled and also the scenario
I think we should just add something like
------8<-------
cpuidle: teo: Fix intercept-logic selecting disabled
Prevent min_idx to be set to a disabled state, which could lead to
both a disabled state being returned by teo, but also an equally good
state being skipped because it has a higher index than a disabled state.
Signed-off-by: Christian Loehle <christian.loehle@arm.com>
---
drivers/cpuidle/governors/teo.c | 8 +++++---
1 file changed, 5 insertions(+), 3 deletions(-)
diff --git a/drivers/cpuidle/governors/teo.c b/drivers/cpuidle/governors/teo.c
index 4cf6302f99ad..94c5ef5df467 100644
--- a/drivers/cpuidle/governors/teo.c
+++ b/drivers/cpuidle/governors/teo.c
@@ -420,9 +420,11 @@ static int teo_select(struct cpuidle_driver *drv, struct cpuidle_device *dev,
* candidate one whose target residency is at least
* equal to the tick period length.
*/
- while (min_idx < idx &&
- drv->states[min_idx].target_residency_ns < TICK_NSEC)
- min_idx++;
+ while (i < idx && drv->states[i].target_residency_ns < TICK_NSEC) {
+ i++;
+ if (!dev->states_usage[i].disable)
+ min_idx = i;
+ }
/*
* Avoid selecting a state with a lower index, but with
--
2.34.1
^ permalink raw reply related [flat|nested] 21+ messages in thread
* Re: [PATCH v1 1/5] cpuidle: governors: teo: Avoid selecting states with zero-size bins
2026-01-26 9:18 ` Christian Loehle
@ 2026-01-26 11:40 ` Rafael J. Wysocki
2026-01-26 12:05 ` Christian Loehle
0 siblings, 1 reply; 21+ messages in thread
From: Rafael J. Wysocki @ 2026-01-26 11:40 UTC (permalink / raw)
To: Christian Loehle; +Cc: Rafael J. Wysocki, Linux PM, LKML, Doug Smythies
On Mon, Jan 26, 2026 at 10:18 AM Christian Loehle
<christian.loehle@arm.com> wrote:
>
> On 1/23/26 20:46, Rafael J. Wysocki wrote:
> > On Wed, Jan 21, 2026 at 2:10 PM Christian Loehle
> > <christian.loehle@arm.com> wrote:
> >>
> >> On 1/14/26 19:44, Rafael J. Wysocki wrote:
> >>> From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
> >>>
> >>> If the last two enabled idle states have the same target residency which
> >>> is at least equal to TICK_NSET, teo may select the next-to-last one even
> >>
> >> s/TICK_NSET/TICK_NSEC
> >
> > Yup, thanks!
> >
> >>> though the size of that state's bin is 0, which is confusing.
> >>>
> >>> Prevent that from happening by adding a target residency check to the
> >>> relevant code path.
> >>>
> >>> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
> >>> ---
> >>> drivers/cpuidle/governors/teo.c | 10 ++++++++++
> >>> 1 file changed, 10 insertions(+)
> >>>
> >>> --- a/drivers/cpuidle/governors/teo.c
> >>> +++ b/drivers/cpuidle/governors/teo.c
> >>> @@ -388,6 +388,15 @@ static int teo_select(struct cpuidle_dri
> >>> while (min_idx < idx &&
> >>> drv->states[min_idx].target_residency_ns < TICK_NSEC)
> >>> min_idx++;
> >>> +
> >>> + /*
> >>> + * Avoid selecting a state with a lower index, but with
> >>> + * the same target residency as the current candidate
> >>> + * one.
> >>> + */
> >>> + if (drv->states[min_idx].target_residency_ns ==
> >>> + drv->states[idx].target_residency_ns)
> >>
> >> We need to check that min_idx isn't disabled though, otherwise we now skip a
> >> potential (enabled) idx==1 if min_idx==2 and min_idx is disabled.
> >
> > Not really because idx is the current candidate state and it is
> > enabled. We'll use idx if this check is true, not min_idx.
> >
>
> Are you sure?
Yeah, pretty much.
> I meant initially:
>
> for (i = intercept_max_idx; i >= min_idx; i--) {
> intercept_sum += cpu_data->state_bins[i].intercepts;
>
> if (dev->states_usage[i].disable)
> continue;
>
> idx = i;
> if (2 * intercept_sum > idx_intercept_sum)
> break;
> }
> might skip an idx==3 if it enters with min_idx==2 (sorry, messed up the +-1 in the initial mail)
> even though idx==3 might have the same residency as idx==2.
> So if idx==2 is disabled we could've selected idx==3, but now won't and will go for idx==1 or
> whatever is the next shallower enabled state.
But that's after patch [5/5] that has problems (I have a new version
of it ready to send, will do that later today).
Look at the original 6.19-rc code and patch [1/1] by itself and you'll
see what I mean.
Patch [1/1] only adds a check and a jump to a new label (at which
point idx is going to be used), nothing more.
> Additionally an issue with this and patch 5/5:
>
> if (min_idx >= intercept_max_idx) {
> idx = min_idx;
> goto constraint; // CL: this will just select min_idx
> }
>
> will use min_idx even though it might be disabled and also the scenario
Again, after patch [5/5] which adds the code quoted above.
> I think we should just add something like
>
> ------8<-------
>
> cpuidle: teo: Fix intercept-logic selecting disabled
>
> Prevent min_idx to be set to a disabled state, which could lead to
> both a disabled state being returned by teo, but also an equally good
> state being skipped because it has a higher index than a disabled state.
>
> Signed-off-by: Christian Loehle <christian.loehle@arm.com>
> ---
> drivers/cpuidle/governors/teo.c | 8 +++++---
> 1 file changed, 5 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/cpuidle/governors/teo.c b/drivers/cpuidle/governors/teo.c
> index 4cf6302f99ad..94c5ef5df467 100644
> --- a/drivers/cpuidle/governors/teo.c
> +++ b/drivers/cpuidle/governors/teo.c
> @@ -420,9 +420,11 @@ static int teo_select(struct cpuidle_driver *drv, struct cpuidle_device *dev,
> * candidate one whose target residency is at least
> * equal to the tick period length.
> */
> - while (min_idx < idx &&
> - drv->states[min_idx].target_residency_ns < TICK_NSEC)
> - min_idx++;
> + while (i < idx && drv->states[i].target_residency_ns < TICK_NSEC) {
> + i++;
> + if (!dev->states_usage[i].disable)
> + min_idx = i;
> + }
>
> /*
> * Avoid selecting a state with a lower index, but with
> --
After fixing patch [5/5] that's not needed any more.
^ permalink raw reply [flat|nested] 21+ messages in thread
* Re: [PATCH v1 1/5] cpuidle: governors: teo: Avoid selecting states with zero-size bins
2026-01-26 11:40 ` Rafael J. Wysocki
@ 2026-01-26 12:05 ` Christian Loehle
0 siblings, 0 replies; 21+ messages in thread
From: Christian Loehle @ 2026-01-26 12:05 UTC (permalink / raw)
To: Rafael J. Wysocki; +Cc: Linux PM, LKML, Doug Smythies
On 1/26/26 11:40, Rafael J. Wysocki wrote:
> On Mon, Jan 26, 2026 at 10:18 AM Christian Loehle
> <christian.loehle@arm.com> wrote:
>>
>> On 1/23/26 20:46, Rafael J. Wysocki wrote:
>>> On Wed, Jan 21, 2026 at 2:10 PM Christian Loehle
>>> <christian.loehle@arm.com> wrote:
>>>>
>>>> On 1/14/26 19:44, Rafael J. Wysocki wrote:
>>>>> From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
>>>>>
>>>>> If the last two enabled idle states have the same target residency which
>>>>> is at least equal to TICK_NSET, teo may select the next-to-last one even
>>>>
>>>> s/TICK_NSET/TICK_NSEC
>>>
>>> Yup, thanks!
>>>
>>>>> though the size of that state's bin is 0, which is confusing.
>>>>>
>>>>> Prevent that from happening by adding a target residency check to the
>>>>> relevant code path.
>>>>>
>>>>> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
>>>>> ---
>>>>> drivers/cpuidle/governors/teo.c | 10 ++++++++++
>>>>> 1 file changed, 10 insertions(+)
>>>>>
>>>>> --- a/drivers/cpuidle/governors/teo.c
>>>>> +++ b/drivers/cpuidle/governors/teo.c
>>>>> @@ -388,6 +388,15 @@ static int teo_select(struct cpuidle_dri
>>>>> while (min_idx < idx &&
>>>>> drv->states[min_idx].target_residency_ns < TICK_NSEC)
>>>>> min_idx++;
>>>>> +
>>>>> + /*
>>>>> + * Avoid selecting a state with a lower index, but with
>>>>> + * the same target residency as the current candidate
>>>>> + * one.
>>>>> + */
>>>>> + if (drv->states[min_idx].target_residency_ns ==
>>>>> + drv->states[idx].target_residency_ns)
>>>>
>>>> We need to check that min_idx isn't disabled though, otherwise we now skip a
>>>> potential (enabled) idx==1 if min_idx==2 and min_idx is disabled.
>>>
>>> Not really because idx is the current candidate state and it is
>>> enabled. We'll use idx if this check is true, not min_idx.
>>>
>>
>> Are you sure?
>
> Yeah, pretty much.
>
>> I meant initially:
>>
>> for (i = intercept_max_idx; i >= min_idx; i--) {
>> intercept_sum += cpu_data->state_bins[i].intercepts;
>>
>> if (dev->states_usage[i].disable)
>> continue;
>>
>> idx = i;
>> if (2 * intercept_sum > idx_intercept_sum)
>> break;
>> }
>> might skip an idx==3 if it enters with min_idx==2 (sorry, messed up the +-1 in the initial mail)
>> even though idx==3 might have the same residency as idx==2.
>> So if idx==2 is disabled we could've selected idx==3, but now won't and will go for idx==1 or
>> whatever is the next shallower enabled state.
>
> But that's after patch [5/5] that has problems (I have a new version
> of it ready to send, will do that later today).
>
> Look at the original 6.19-rc code and patch [1/1] by itself and you'll
> see what I mean.
>
Duh you're right, sorry about the noise :/
^ permalink raw reply [flat|nested] 21+ messages in thread
end of thread, other threads:[~2026-01-26 12:05 UTC | newest]
Thread overview: 21+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-01-14 19:42 [PATCH v1 0/5] cpuidle: governors: teo: Wakeup events classification change and some refinements Rafael J. Wysocki
2026-01-14 19:44 ` [PATCH v1 1/5] cpuidle: governors: teo: Avoid selecting states with zero-size bins Rafael J. Wysocki
2026-01-21 13:09 ` Christian Loehle
2026-01-23 20:46 ` Rafael J. Wysocki
2026-01-26 9:18 ` Christian Loehle
2026-01-26 11:40 ` Rafael J. Wysocki
2026-01-26 12:05 ` Christian Loehle
2026-01-14 19:44 ` [PATCH v1 2/5] cpuidle: governors: teo: Avoid fake intercepts produced by tick Rafael J. Wysocki
2026-01-21 13:34 ` Christian Loehle
2026-01-14 19:45 ` [PATCH v1 3/5] cpuidle: governors: teo: Refine tick_intercepts vs total events check Rafael J. Wysocki
2026-01-21 13:36 ` Christian Loehle
2026-01-14 19:46 ` [PATCH v1 4/5] cpuidle: governors: teo: Adjust the classification of wakeup events Rafael J. Wysocki
2026-01-14 19:47 ` [PATCH v1 5/5] cpuidle: governors: teo: Refine intercepts-based idle state lookup Rafael J. Wysocki
2026-01-16 11:52 ` [PATCH v1 0/5] cpuidle: governors: teo: Wakeup events classification change and some refinements Christian Loehle
2026-01-16 12:29 ` Rafael J. Wysocki
2026-01-19 9:53 ` Christian Loehle
2026-01-19 12:09 ` Rafael J. Wysocki
2026-01-19 16:20 ` Doug Smythies
2026-01-20 15:29 ` [Update][PATCH v1.1 4/5] cpuidle: governors: teo: Adjust the classification of wakeup events Rafael J. Wysocki
2026-01-25 17:21 ` Doug Smythies
2026-01-25 19:38 ` Christian Loehle
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox