sched/fair: Remove stale FREQUENCY_UTIL comment

On 05/03/2024 15:05, Vincent Guittot wrote:

I'm fine with either and that was my first thought here, too, but it did seem like
the comment was mostly placed there to justify the 'unexpected' high utilization
when explicitly passing FREQUENCY_UTIL and the need to clamp it then.
So removing did feel slightly more natural to me anyway.

So alternatively:

From: Christian Loehle <christian.loehle@arm.com>
Date: Tue, 5 Mar 2024 09:34:41 +0000
Subject: [PATCH] sched/fair: Remove stale FREQUENCY_UTIL mention

effective_cpu_util() flags were removed, so remove mentioning of the
flag.

commit 9c0b4bb7f6303 ("sched/cpufreq: Rework schedutil governor performance estimation")
reworked effective_cpu_util() removing enum cpu_util_type. Modify the
comment accordingly.

Signed-off-by: Christian Loehle <christian.loehle@arm.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/0e2833ee-0939-44e0-82a2-520a585a0153@arm.com

sched/fair: Fix initial util_avg calculation

Change se->load.weight to se_weight(se) in the calculation for the
initial util_avg to avoid unnecessarily inflating the util_avg by 1024
times.

The reason is that se->load.weight has the unit/scale as the scaled-up
load, while cfs_rg->avg.load_avg has the unit/scale as the true task
weight (as mapped directly from the task's nice/priority value). With
CONFIG_32BIT, the scaled-up load is equal to the true task weight. With
CONFIG_64BIT, the scaled-up load is 1024 times the true task weight.
Thus, the current code may inflate the util_avg by 1024 times. The
follow-up capping will not allow the util_avg value to go wild. But the
calculation should have the correct logic.

Signed-off-by: Dawei Li <daweilics@gmail.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Vishal Chourasia <vishalc@linux.ibm.com>
Link: https://lore.kernel.org/r/20240315015916.21545-1-daweilics@gmail.com

scheduler: Remove the now superfluous sentinel elements from ctl_table array

This commit comes at the tail end of a greater effort to remove the
empty elements at the end of the ctl_table arrays (sentinels) which
will reduce the overall build time size of the kernel and run time
memory bloat by ~64 bytes per sentinel (further information Link :
https://lore.kernel.org/all/ZO5Yx5JFogGi%2FcBo@bombadil.infradead.org/)

rm sentinel element from ctl_table arrays

Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Valentin Schneider <vschneid@redhat.com>
Reviewed-by: Valentin Schneider <vschneid@redhat.com>
Signed-off-by: Joel Granados <j.granados@samsung.com>

sched/pelt: Remove shift of thermal clock

The optional shift of the clock used by thermal/hw load avg has been
introduced to handle case where the signal was not always a high frequency
hw signal. Now that cpufreq provides a signal for firmware and
SW pressure, we can remove this exception and always keep this PELT signal
aligned with other signals.
Mark sysctl_sched_migration_cost boot parameter as deprecated

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Tested-by: Lukasz Luba <lukasz.luba@arm.com>
Reviewed-by: Qais Yousef <qyousef@layalina.io>
Reviewed-by: Lukasz Luba <lukasz.luba@arm.com>
Link: https://lore.kernel.org/r/20240326091616.3696851-6-vincent.guittot@linaro.org

sched/cpufreq: Rename arch_update_thermal_pressure() => arch_update_hw_pressure()

Now that cpufreq provides a pressure value to the scheduler, rename
arch_update_thermal_pressure into HW pressure to reflect that it returns
a pressure applied by HW (i.e. with a high frequency change) and not
always related to thermal mitigation but also generated by max current
limitation as an example. Such high frequency signal needs filtering to be
smoothed and provide an value that reflects the average available capacity
into the scheduler time scale.

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Tested-by: Lukasz Luba <lukasz.luba@arm.com>
Reviewed-by: Qais Yousef <qyousef@layalina.io>
Reviewed-by: Lukasz Luba <lukasz.luba@arm.com>
Link: https://lore.kernel.org/r/20240326091616.3696851-5-vincent.guittot@linaro.org

sched/cpufreq: Take cpufreq feedback into account

Aggregate the different pressures applied on the capacity of CPUs and
create a new function that returns the actual capacity of the CPU:
get_actual_cpu_capacity().

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Tested-by: Lukasz Luba <lukasz.luba@arm.com>
Reviewed-by: Lukasz Luba <lukasz.luba@arm.com>
Reviewed-by: Qais Yousef <qyousef@layalina.io>
Link: https://lore.kernel.org/r/20240326091616.3696851-3-vincent.guittot@linaro.org

sched/fair: Fix update of rd->sg_overutilized

sg_overloaded is used instead of sg_overutilized to update
rd->sg_overutilized.

Fixes: 4475cd8bfd9b ("sched/balancing: Simplify the sg_status bitmask and use separate ->overloaded and ->overutilized flags")
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20240404155738.2866102-1-vincent.guittot@linaro.org

sched/balancing: Simplify the sg_status bitmask and use separate ->overloaded and ->overutilized flags

SG_OVERLOADED and SG_OVERUTILIZED flags plus the sg_status bitmask are an
unnecessary complication that only make the code harder to read and slower.

We only ever set them separately:

thule:~/tip> git grep SG_OVER kernel/sched/
kernel/sched/fair.c: set_rd_overutilized_status(rq->rd, SG_OVERUTILIZED);
kernel/sched/fair.c: *sg_status |= SG_OVERLOADED;
kernel/sched/fair.c: *sg_status |= SG_OVERUTILIZED;
kernel/sched/fair.c: *sg_status |= SG_OVERLOADED;
kernel/sched/fair.c: set_rd_overloaded(env->dst_rq->rd, sg_status & SG_OVERLOADED);
kernel/sched/fair.c: sg_status & SG_OVERUTILIZED);
kernel/sched/fair.c: } else if (sg_status & SG_OVERUTILIZED) {
kernel/sched/fair.c: set_rd_overutilized_status(env->dst_rq->rd, SG_OVERUTILIZED);
kernel/sched/sched.h:#define SG_OVERLOADED 0x1 /* More than one runnable task on a CPU. */
kernel/sched/sched.h:#define SG_OVERUTILIZED 0x2 /* One or more CPUs are over-utilized. */
kernel/sched/sched.h: set_rd_overloaded(rq->rd, SG_OVERLOADED);

And use them separately, which results in suboptimal code:

/* update overload indicator if we are at root domain */
set_rd_overloaded(env->dst_rq->rd, sg_status & SG_OVERLOADED);

/* Update over-utilization (tipping point, U >= 0) indicator */
set_rd_overutilized_status(env->dst_rq->rd,

Introduce separate sg_overloaded and sg_overutilized flags in update_sd_lb_stats()
and its lower level functions, and change all of them to 'bool'.

Remove the now unused SG_OVERLOADED and SG_OVERUTILIZED flags.

Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Tested-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Cc: Qais Yousef <qyousef@layalina.io>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/r/ZgVPhODZ8/nbsqbP@gmail.com

sched/fair: Rename set_rd_overutilized_status() to set_rd_overutilized()

The _status() postfix has no real meaning, simplify the naming
and harmonize it with set_rd_overloaded().

Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Qais Yousef <qyousef@layalina.io>
Cc: Shrikanth Hegde <sshegde@linux.ibm.com>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/r/ZgVHq65XKsOZpfgK@gmail.com

sched/fair: Rename SG_OVERLOAD to SG_OVERLOADED

Follow the rename of the root_domain::overloaded flag.

Note that this also matches the SG_OVERUTILIZED flag better.

Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Qais Yousef <qyousef@layalina.io>
Cc: Shrikanth Hegde <sshegde@linux.ibm.com>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/r/ZgVHq65XKsOZpfgK@gmail.com

sched/fair: Rename {set|get}_rd_overload() to {set|get}_rd_overloaded()

Follow the rename of the root_domain::overloaded flag.

Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Qais Yousef <qyousef@layalina.io>
Cc: Shrikanth Hegde <sshegde@linux.ibm.com>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/r/ZgVHq65XKsOZpfgK@gmail.com

sched/fair: Use helper functions to access root_domain::overload

Introduce two helper functions to access & set the root_domain::overload flag:

get_rd_overload()
set_rd_overload()

To make sure code is always following READ_ONCE()/WRITE_ONCE() access methods.

No change in functionality intended.

[ mingo: Renamed the accessors to get_/set_rd_overload(), tidied up the changelog. ]

Suggested-by: Qais Yousef <qyousef@layalina.io>
Signed-off-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Qais Yousef <qyousef@layalina.io>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20240325054505.201995-3-sshegde@linux.ibm.com

sched/fair: Check root_domain::overload value before update

The root_domain::overload flag is 1 when there's any rq
in the root domain that has 2 or more running tasks. (Ie. it's overloaded.)

The root_domain structure itself is a global structure per cpuset island.

The ::overload flag is maintained the following way:

- Set when adding a second task to the runqueue.

- It is cleared in update_sd_lb_stats() during load balance,
if none of the rqs have 2 or more running tasks.

This flag is used during newidle balance to see if its worth doing a full
load balance pass, which can be an expensive operation. If it is set,
then newidle balance will try to aggressively pull a task.

Since commit:

630246a06ae2 ("sched/fair: Clean-up update_sg_lb_stats parameters")

::overload is being written unconditionally, even if it has the same
value. The change in value of this depends on the workload, but on
typical workloads, it doesn't change all that often: a system is
either dominantly overloaded for substantial amounts of time, or not.

Extra writes to this semi-global structure cause unnecessary overhead, extra
bus traffic, etc. - so avoid it as much as possible.

Perf probe stats show that it's worth making this change (numbers are
with patch applied):

1M probe:sched_balance_newidle_L38
139 probe:update_sd_lb_stats_L53 <====== 1->0 writes
129K probe:add_nr_running_L12
74 probe:add_nr_running_L13 <====== 0->1 writes
54K probe:update_sd_lb_stats_L50 <====== reads

These numbers prove that actual change in the ::overload value is (much) less
frequent: L50 is much larger at ~54,000 accesses vs L53+L13 of 139+74.

[ mingo: Rewrote the changelog. ]

Signed-off-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Qais Yousef <qyousef@layalina.io>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Mel Gorman <mgorman@suse.de>
Link: https://lore.kernel.org/r/20240325054505.201995-2-sshegde@linux.ibm.com

sched/fair: Combine EAS check with root_domain::overutilized access

Access to root_domainoverutilized is always used with sched_energy_enabled in
the pattern:

if (sched_energy_enabled && !overutilized)
do something

So modify the helper function to utilize this pattern. This is more
readable code as it would say, do something when root domain is not
overutilized. This function always return true when EAS is disabled.

No change in functionality intended.

Suggested-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20240326152616.380999-1-sshegde@linux.ibm.com

sched/fair: Simplify the continue_balancing logic in sched_balance_newidle()

newidle(CPU_NEWLY_IDLE) balancing doesn't stop the load-balancing if the
continue_balancing flag is reset, but the other two balancing (IDLE, BUSY)
cases do that.

newidle balance stops the load balancing if rq has a task or there
is wakeup pending. The same checks are present in should_we_balance for
newidle. Hence use the return value and simplify continue_balancing
mechanism for newidle. Update the comment surrounding it as well.

No change in functionality intended.

Signed-off-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Link: https://lore.kernel.org/r/20240325153926.274284-1-sshegde@linux.ibm.com

sched/fair: Introduce is_rd_overutilized() helper function to access root_domain::overutilized

The root_domain::overutilized field is READ_ONCE() accessed in
multiple places, which could be simplified with a helper function.

This might also make it more apparent that it needs to be used
only in case of EAS.

No change in functionality intended.

Signed-off-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Qais Yousef <qyousef@layalina.io>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20240307085725.444486-3-sshegde@linux.ibm.com

sched/fair: Add EAS checks before updating root_domain::overutilized

root_domain::overutilized is only used for EAS(energy aware scheduler)
to decide whether to do load balance or not. It is not used if EAS
not possible.

Currently enqueue_task_fair and task_tick_fair accesses, sometime updates
this field. In update_sd_lb_stats it is updated often. This causes cache
contention due to true sharing and burns a lot of cycles. ::overload and
::overutilized are part of the same cacheline. Updating it often invalidates
the cacheline. That causes access to ::overload to slow down due to
false sharing. Hence add EAS check before accessing/updating this field.
EAS check is optimized at compile time or it is a static branch.
Hence it shouldn't cost much.

With the patch, both enqueue_task_fair and newidle_balance don't show
up as hot routines in perf profile.

6.8-rc4:
7.18% swapper [kernel.vmlinux] [k] enqueue_task_fair
6.78% s [kernel.vmlinux] [k] newidle_balance

+patch:
0.14% swapper [kernel.vmlinux] [k] enqueue_task_fair
0.00% swapper [kernel.vmlinux] [k] newidle_balance

While at it: trace_sched_overutilized_tp expect that second argument to
be bool. So do a int to bool conversion for that.

Fixes: 2802bf3cd936 ("sched/fair: Add over-utilization/tipping point indicator")
Signed-off-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Qais Yousef <qyousef@layalina.io>
Reviewed-by: Srikar Dronamraju <srikar@linux.ibm.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20240307085725.444486-2-sshegde@linux.ibm.com

sched/fair: Don't double balance_interval for migrate_misfit

It is not necessarily an indication of the system being busy and
requires a backoff of the load balancer activities. But pushing it high
could mean generally delaying other misfit activities or other type of
imbalances.

Also don't pollute nr_balance_failed because of misfit failures. The
value is used for enabling cache hot migration and in migrate_util/load
types. None of which should be impacted (skewed) by misfit failures.

Signed-off-by: Qais Yousef <qyousef@layalina.io>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20240324004552.999936-5-qyousef@layalina.io

sched/fair: Check if a task has a fitting CPU when updating misfit

If a misfit task is affined to a subset of the possible CPUs, we need to
verify that one of these CPUs can fit it. Otherwise the load balancer
code will continuously trigger needlessly leading the balance_interval
to increase in return and eventually end up with a situation where real
imbalances take a long time to address because of this impossible
imbalance situation.

This can happen in Android world where it's common for background tasks
to be restricted to little cores.

Similarly if we can't fit the biggest core, triggering misfit is
pointless as it is the best we can ever get on this system.

To be able to detect that; we use asym_cap_list to iterate through
capacities in the system to see if the task is able to run at a higher
capacity level based on its p->cpus_ptr. We do that when the affinity
change, a fair task is forked, or when a task switched to fair policy.
We store the max_allowed_capacity in task_struct to allow for cheap
comparison in the fast path.

Improve check_misfit_status() function by removing redundant checks.
misfit_task_load will be 0 if the task can't move to a bigger CPU. And
nohz_balancer_kick() already checks for cpu_check_capacity() before
calling check_misfit_status().

Test:
=====

Add

trace_printk("balance_interval = %lu\n", interval)

in get_sd_balance_interval().

run
if [ "$MASK" != "0" ]; then
adb shell "taskset -a $MASK cat /dev/zero > /dev/null"
fi
sleep 10
// parse ftrace buffer counting the occurrence of each valaue

Where MASK is either:

* 0: no busy task running
* 1: busy task is pinned to 1 cpu; handled today to not cause
misfit
* f: busy task pinned to little cores, simulates busy background
task, demonstrates the problem to be fixed

Results:
========

Note how occurrence of balance_interval = 128 overshoots for MASK = f.

BEFORE
------

MASK=0

1 balance_interval = 175
120 balance_interval = 128
846 balance_interval = 64
55 balance_interval = 63
215 balance_interval = 32
2 balance_interval = 31
2 balance_interval = 16
4 balance_interval = 8
1870 balance_interval = 4
65 balance_interval = 2

MASK=1

27 balance_interval = 175
37 balance_interval = 127
840 balance_interval = 64
167 balance_interval = 63
449 balance_interval = 32
84 balance_interval = 31
304 balance_interval = 16
1156 balance_interval = 8
2781 balance_interval = 4
428 balance_interval = 2

MASK=f

1 balance_interval = 175
1328 balance_interval = 128
44 balance_interval = 64
101 balance_interval = 63
25 balance_interval = 32
5 balance_interval = 31
23 balance_interval = 16
23 balance_interval = 8
4306 balance_interval = 4
177 balance_interval = 2

AFTER
-----

Note how the high values almost disappear for all MASK values. The
system has background tasks that could trigger the problem without
simulate it even with MASK=0.

MASK=0

103 balance_interval = 63
19 balance_interval = 31
194 balance_interval = 8
4827 balance_interval = 4
179 balance_interval = 2

MASK=1

131 balance_interval = 63
1 balance_interval = 31
87 balance_interval = 8
3600 balance_interval = 4
7 balance_interval = 2

MASK=f

8 balance_interval = 127
182 balance_interval = 63
3 balance_interval = 31
9 balance_interval = 16
415 balance_interval = 8
3415 balance_interval = 4
21 balance_interval = 2

Signed-off-by: Qais Yousef <qyousef@layalina.io>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20240324004552.999936-3-qyousef@layalina.io

sched/fair: Fix typos in comments

So I made all speling mistakes / typos red in my editor. Big mistake...

Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: linux-kernel@vger.kernel.org

sched/balancing: Fix a couple of outdated function names in comments

The 'idle_balance()' function hasn't existed for years, and there's no
load_balance_newidle() either - both are sched_balance_newidle() today.

Reported-by: Honglei Wang <jameshongleiwang@126.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/ZfAwNufbiyt/5biu@gmail.com

sched/balancing: Rename find_idlest_cpu() => sched_balance_find_dst_cpu()

Standardize scheduler load-balancing function names on the
sched_balance_() prefix.

Also use 'dst' instead of 'idlest', because it's not really
true that we return the 'idlest' group or CPU, we sort by
idle-exit latency and only return the idlest CPUs from the
lowest-latency set of CPUs.

The true 'idlest' CPUs often remain idle for a long time
and are never returned as long as the system is under-loaded.

Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Link: https://lore.kernel.org/r/20240308111819.1101550-14-mingo@kernel.org

sched/balancing: Rename find_idlest_group() => sched_balance_find_dst_group()

Standardize scheduler load-balancing function names on the
sched_balance_() prefix.

Also use 'dst' instead of 'idlest', because it's not really
true that we return the 'idlest' group or CPU, we sort by
idle-exit latency and only return the idlest CPUs from the
lowest-latency set of CPUs.

The true 'idlest' CPUs often remain idle for a long time
and are never returned as long as the system is under-loaded.

Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Link: https://lore.kernel.org/r/20240308111819.1101550-13-mingo@kernel.org

sched/balancing: Rename find_idlest_group_cpu() => sched_balance_find_dst_group_cpu()

Standardize scheduler load-balancing function names on the
sched_balance_() prefix.

Also use 'dst' instead of 'idlest': while historically correct,
today it's not really true anymore that we return the 'idlest'
group or CPU, we sort by idle-exit latency and only return the
idlest CPUs from the lowest-latency set of CPUs.

The true 'idlest' CPUs often remain idle for a long time
and are never returned as long as the system is under-loaded.

Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Link: https://lore.kernel.org/r/20240308111819.1101550-12-mingo@kernel.org

sched/balancing: Rename newidle_balance() => sched_balance_newidle()

Standardize scheduler load-balancing function names on the
sched_balance_() prefix.

Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Link: https://lore.kernel.org/r/20240308111819.1101550-11-mingo@kernel.org

sched/balancing: Rename update_blocked_averages() => sched_balance_update_blocked_averages()

Standardize scheduler load-balancing function names on the
sched_balance_() prefix.

Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Link: https://lore.kernel.org/r/20240308111819.1101550-10-mingo@kernel.org

sched/balancing: Rename find_busiest_group() => sched_balance_find_src_group()

Make two naming changes:

1)
Standardize scheduler load-balancing function names on the
sched_balance_() prefix.

2)

Similar to find_busiest_queue(), the find_busiest_group() naming
has become a bit of a misnomer: the 'busiest' qualifier to this
function was historically correct but in the current code
in quite a few cases we will not pick the 'busiest' group - but the best
(possible) group we can balance from based on a complex set of
constraints.

So name it a bit more neutrally, similar to the 'src/dst' nomenclature
we are already using when moving tasks between runqueues, and also
use the sched_balance_ prefix: sched_balance_find_src_group().

Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Link: https://lore.kernel.org/r/20240308111819.1101550-9-mingo@kernel.org

sched/balancing: Rename find_busiest_queue() => sched_balance_find_src_rq()

The find_busiest_queue() naming has two small quirks:

- Scheduler functions that deal with runqueues usually have a rq_ prefix
or _rq postfix, but this function has neither.

- Plus the 'busiest' qualifier to this function was historically
correct, but has become somewhat of a misnomer: in quite a few
cases we will not pick the busiest runqueue - but the best
(possible) runqueue we can balance tasks from. So name it a
bit more neutrally, similar to the 'src/dst' nomenclature
we are already using when moving tasks between runqueues.

To fix both quirks, and to standardize scheduler load-balancing
function names on the sched_balance_() prefix, rename the
function to sched_balance_find_src_rq().

Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Link: https://lore.kernel.org/r/20240308111819.1101550-7-mingo@kernel.org

sched/balancing: Rename load_balance() => sched_balance_rq()

Standardize scheduler load-balancing function names on the
sched_balance_() prefix.

Also load_balance() has become somewhat of a misnomer: historically
it was the first and primary load-balancing function that was called,
but with the introduction of sched domains, it's become a lower
layer function that balances runqueues.

Rename it to sched_balance_rq() accordingly.

Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Link: https://lore.kernel.org/r/20240308111819.1101550-6-mingo@kernel.org

sched/balancing: Rename rebalance_domains() => sched_balance_domains()

Standardize scheduler load-balancing function names on the
sched_balance_() prefix.

Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Link: https://lore.kernel.org/r/20240308111819.1101550-5-mingo@kernel.org

sched/balancing: Rename trigger_load_balance() => sched_balance_trigger()

Standardize scheduler load-balancing function names on the
sched_balance_() prefix.

Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Link: https://lore.kernel.org/r/20240308111819.1101550-4-mingo@kernel.org

sched/balancing: Rename run_rebalance_domains() => sched_balance_softirq()

run_rebalance_domains() is a misnomer, as it doesn't only
run rebalance_domains(), but since the introduction of the
NOHZ code it also runs nohz_idle_balance().

Rename it to sched_balance_softirq(), reflecting its more
generic purpose and that it's a softirq handler.

Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Valentin Schneider <vschneid@redhat.com>
Reviewed-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Link: https://lore.kernel.org/r/20240308111819.1101550-2-mingo@kernel.org

sched/balancing: Update comments in 'struct sg_lb_stats' and 'struct sd_lb_stats'

- Align for readability
- Capitalize consistently

Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Valentin Schneider <vschneid@redhat.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20240308105901.1096078-11-mingo@kernel.org

sched/balancing: Vertically align the comments of 'struct sg_lb_stats' and 'struct sd_lb_stats'

Make them easier to read.

Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Valentin Schneider <vschneid@redhat.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20240308105901.1096078-10-mingo@kernel.org

sched/balancing: Update run_rebalance_domains() comments

The first sentence of the comment explaining run_rebalance_domains()
is historic and not true anymore:

* run_rebalance_domains is triggered when needed from the scheduler tick.

... contradicted/modified by the second sentence:

* Also triggered for NOHZ idle balancing (with NOHZ_BALANCE_KICK set).

Avoid that kind of confusion straight away and explain from what
places sched_balance_softirq() is triggered.

Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Acked-by: Valentin Schneider <vschneid@redhat.com>
Link: https://lore.kernel.org/r/20240308105901.1096078-9-mingo@kernel.org

sched/balancing: Fix comments (trying to) refer to NOHZ_BALANCE_KICK

Fix two typos:

- There's no such thing as 'nohz_balancing_kick', the
flag is named 'BALANCE' and is capitalized: NOHZ_BALANCE_KICK.

- Likewise there's no such thing as a 'pending nohz_balance_kick'
either, the NOHZ_BALANCE_KICK flag is all upper-case.

Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Valentin Schneider <vschneid@redhat.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20240308105901.1096078-8-mingo@kernel.org

sched/balancing: Change comment formatting to not overlap Git conflict marker lines

So the scheduler has two such comment blocks, with '=' used as a double underline:

/*
* VRUNTIME
* ========
*

'========' also happens to be a Git conflict marker, throwing off a simple
search in an editor for this pattern.

Change them to '-------' type of underline instead - it looks just as good.

Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Valentin Schneider <vschneid@redhat.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20240308105901.1096078-7-mingo@kernel.org

sched/balancing: Change 'enum cpu_idle_type' to have more natural definitions

The cpu_idle_type enum has the confusingly inverted property
that 'not idle' is 1, and 'idle' is '0'.

This resulted in a number of unnecessary complications in the code.

Reverse the order, remove the CPU_NOT_IDLE type, and convert
all code to a natural boolean form.

It's much more readable:

- enum cpu_idle_type idle = this_rq->idle_balance ?
- CPU_IDLE : CPU_NOT_IDLE;
-
+ enum cpu_idle_type idle = this_rq->idle_balance;

--------------------------------

- if (env->idle == CPU_NOT_IDLE || !busiest->sum_nr_running)
+ if (!env->idle || !busiest->sum_nr_running)

--------------------------------

And gets rid of the double negation in these usages:

- if (env->idle != CPU_NOT_IDLE && env->src_rq->nr_running <= 1)
+ if (env->idle && env->src_rq->nr_running <= 1)

Furthermore, this makes code much more obvious where there's
differentiation between CPU_IDLE and CPU_NEWLY_IDLE.

Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Valentin Schneider <vschneid@redhat.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Cc: "Gautham R. Shenoy" <gautham.shenoy@amd.com>
Link: https://lore.kernel.org/r/20240308105901.1096078-4-mingo@kernel.org

sched/balancing: Switch the 'DEFINE_SPINLOCK(balancing)' spinlock into an 'atomic_t sched_balance_running' flag

The 'balancing' spinlock added in:

08c183f31bdb ("[PATCH] sched: add option to serialize load balancing")

... is taken when the SD_SERIALIZE flag is set in a domain, but in reality it
is a glorified global atomic flag serializing the load-balancing of
those domains.

It doesn't have any explicit locking semantics per se: we just
spin_trylock() it.

Turn it into a ... global atomic flag. This makes it more
clear what is going on here, and reduces overhead and code
size a bit:

# kernel/sched/fair.o: [x86-64 defconfig]

text data bss dec hex filename
60730 2721 104 63555 f843 fair.o.before
60718 2721 104 63543 f837 fair.o.after

Also document the flag a bit.

No change in functionality intended.

Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Valentin Schneider <vschneid@redhat.com>
Cc: Shrikanth Hegde <sshegde@linux.ibm.com>
Link: https://lore.kernel.org/r/20240308105901.1096078-2-mingo@kernel.org

sched/eevdf: Prevent vlag from going out of bounds in reweight_eevdf()

It was possible to have pick_eevdf() return NULL, which then causes a
NULL-deref. This turned out to be due to entity_eligible() returning
falsely negative because of a s64 multiplcation overflow.

Specifically, reweight_eevdf() computes the vlag without considering
the limit placed upon vlag as update_entity_lag() does, and then the
scaling multiplication (remember that weight is 20bit fixed point) can
overflow. This then leads to the new vruntime being weird which then
causes the above entity_eligible() to go side-ways and claim nothing
is eligible.

Thus limit the range of vlag accordingly.

All this was quite rare, but fatal when it does happen.

Closes: https://lore.kernel.org/all/ZhuYyrh3mweP_Kd8@nz.home/
Closes: https://lore.kernel.org/all/CA+9S74ih+45M_2TPUY_mPPVDhNvyYfy1J1ftSix+KjiTVxg8nw@mail.gmail.com/
Closes: https://lore.kernel.org/lkml/202401301012.2ed95df0-oliver.sang@intel.com/
Fixes: eab03c23c2a1 ("sched/eevdf: Fix vruntime adjustment on reweight")
Reported-by: Sergei Trofimovich <slyich@gmail.com>
Reported-by: Igor Raits <igor@gooddata.com>
Reported-by: Breno Leitao <leitao@debian.org>
Reported-by: kernel test robot <oliver.sang@intel.com>
Reported-by: Yujie Liu <yujie.liu@intel.com>
Signed-off-by: Xuewen Yan <xuewen.yan@unisoc.com>
Reviewed-and-tested-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20240422082238.5784-1-xuewen.yan@unisoc.com

sched/eevdf: Fix miscalculation in reweight_entity() when se is not curr

reweight_eevdf() only keeps V unchanged inside itself. When se !=
cfs_rq->curr, it would be dequeued from rb tree first. So that V is
changed and the result is wrong. Pass the original V to reweight_eevdf()
to fix this issue.

Fixes: eab03c23c2a1 ("sched/eevdf: Fix vruntime adjustment on reweight")
Signed-off-by: Tianchen Ding <dtcccc@linux.alibaba.com>
[peterz: flip if() condition for clarity]
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Abel Wu <wuyun.abel@bytedance.com>
Link: https://lkml.kernel.org/r/20240306022133.81008-3-dtcccc@linux.alibaba.com

sched/eevdf: Always update V if se->on_rq when reweighting

reweight_eevdf() needs the latest V to do accurate calculation for new
ve and vd. So update V unconditionally when se is runnable.

Fixes: eab03c23c2a1 ("sched/eevdf: Fix vruntime adjustment on reweight")
Suggested-by: Abel Wu <wuyun.abel@bytedance.com>
Signed-off-by: Tianchen Ding <dtcccc@linux.alibaba.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Abel Wu <wuyun.abel@bytedance.com>
Tested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Tested-by: Chen Yu <yu.c.chen@intel.com>
Link: https://lore.kernel.org/r/20240306022133.81008-2-dtcccc@linux.alibaba.com

sched/numa, mm: do not try to migrate memory to memoryless nodes

Memoryless nodes do not have any memory to migrate to, so, as an
optimization, stop trying it.

Link: https://lkml.kernel.org/r/20240219041920.1183-1-byungchul@sk.com
Link: https://lkml.kernel.org/r/20240216111502.79759-1-byungchul@sk.com
Fixes: c574bbe91703 ("NUMA balancing: optimize page placement for memory tiering system")
Signed-off-by: Byungchul Park <byungchul@sk.com>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Reviewed-by: "Huang, Ying" <ying.huang@intel.com>
Reviewed-by: Phil Auld <pauld@redhat.com>
Reviewed-by: Davidlohr Bueso <dave@stgolabs.net>
Acked-by: David Hildenbrand <david@redhat.com>
Cc: Benjamin Segall <bsegall@google.com>
Cc: Daniel Bristot de Oliveira <bristot@redhat.com>
Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Valentin Schneider <vschneid@redhat.com>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

sched/topology: Rename SD_SHARE_PKG_RESOURCES to SD_SHARE_LLC

SD_SHARE_PKG_RESOURCES is a bit of a misnomer: its naming suggests that
it's sharing all 'package resources' - while in reality it's specifically
for sharing the LLC only.

Rename it to SD_SHARE_LLC to reduce confusion.

[ mingo: Rewrote the confusing changelog as well. ]

Suggested-by: Valentin Schneider <vschneid@redhat.com>
Signed-off-by: Alex Shi <alexs@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Valentin Schneider <vschneid@redhat.com>
Reviewed-by: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
Reviewed-by: Barry Song <baohua@kernel.org>
Link: https://lore.kernel.org/r/20240210113924.1130448-5-alexs@kernel.org

sched/fair: Check the SD_ASYM_PACKING flag in sched_use_asym_prio()

sched_use_asym_prio() checks whether CPU priorities should be used. It
makes sense to check for the SD_ASYM_PACKING() inside the function.
Since both sched_asym() and sched_group_asym() use sched_use_asym_prio(),
remove the now superfluous checks for the flag in various places.

Signed-off-by: Alex Shi <alexs@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Tested-by: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
Reviewed-by: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20240210113924.1130448-4-alexs@kernel.org

sched/fair: Rework sched_use_asym_prio() and sched_asym_prefer()

sched_use_asym_prio() and sched_asym_prefer() are used together in various
places. Consolidate them into a single function sched_asym().

The existing sched_asym() function is only used when collecting statistics
of a scheduling group. Rename it as sched_group_asym(), and remove the
obsolete function description.

This makes the code easier to read. No functional changes.

Signed-off-by: Alex Shi <alexs@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Tested-by: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
Reviewed-by: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20240210113924.1130448-3-alexs@kernel.org

sched/fair: Remove unused parameter from sched_asym()

The 'sds' argument is not used in the sched_asym() function anymore, remove it.

Fixes: c9ca07886aaa ("sched/fair: Do not even the number of busy CPUs via asym_packing")
Signed-off-by: Alex Shi <alexs@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
Reviewed-by: Valentin Schneider <vschneid@redhat.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20240210113924.1130448-2-alexs@kernel.org

sched/fair: Simplify the update_sd_pick_busiest() logic

When comparing the current struct sched_group with the yet-busiest
domain in update_sd_pick_busiest(), if the two groups have the same
group type, we're currently doing a bit of unnecessary work for any
group >= group_misfit_task. We're comparing the two groups, and then
returning only if false (the group in question is not the busiest).

Otherwise, we break out, do an extra unnecessary conditional check that's
vacuously false for any group type > group_fully_busy, and then always
return true.

Let's just return directly in the switch statement instead. This doesn't
change the size of vmlinux with llvm 17 (not surprising given that all
of this is inlined in load_balance()), but it does shrink load_balance()
by 88 bytes on x86. Given that it also improves readability, this seems
worth doing.

Signed-off-by: David Vernet <void@manifault.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Valentin Schneider <vschneid@redhat.com>
Link: https://lore.kernel.org/r/20240206043921.850302-4-void@manifault.com

sched/fair: Do strict inequality check for busiest misfit task group

In update_sd_pick_busiest(), when comparing two sched groups that are
both of type group_misfit_task, we currently consider the new group as
busier than the current busiest group even if the new group has the
same misfit task load as the current busiest group. We can avoid some
unnecessary writes if we instead only consider the newest group to be
the busiest if it has a higher load than the current busiest. This
matches the behavior of other group types where we compare load, such as
two groups that are both overloaded.

Let's update the group_misfit_task type comparison to also only update
the busiest group in the event of strict inequality.

Signed-off-by: David Vernet <void@manifault.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Valentin Schneider <vschneid@redhat.com>
Link: https://lore.kernel.org/r/20240206043921.850302-3-void@manifault.com

sched/fair: Remove unnecessary goto in update_sd_lb_stats()

In update_sd_lb_stats(), when we're iterating over the sched groups that
comprise a sched domain, we're skipping the call to
update_sd_pick_busiest() for the sched group that contains the local /
destination CPU. We use a goto to skip the call, but we could just as
easily check !local_group, as there's no other logic that we need to
skip with the goto. Let's remove the goto, and check for !local_group in
the if statement instead.

Signed-off-by: David Vernet <void@manifault.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Valentin Schneider <vschneid@redhat.com>
Link: https://lore.kernel.org/r/20240206043921.850302-2-void@manifault.com

sched/fair: Take the scheduling domain into account in select_idle_core()

When picking a CPU on task wakeup, select_idle_core() has to take
into account the scheduling domain where the function looks for the CPU.

This is because the "isolcpus" kernel command line option can remove CPUs
from the domain to isolate them from other SMT siblings.

This change replaces the set of CPUs allowed to run the task from
p->cpus_ptr by the intersection of p->cpus_ptr and sched_domain_span(sd)
which is stored in the 'cpus' argument provided by select_idle_cpu().

Fixes: 9fe1f127b913 ("sched/fair: Merge select_idle_core/cpu()")
Signed-off-by: Keisuke Nishimura <keisuke.nishimura@inria.fr>
Signed-off-by: Julia Lawall <julia.lawall@inria.fr>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20240110131707.437301-2-keisuke.nishimura@inria.fr

sched/fair: Take the scheduling domain into account in select_idle_smt()

When picking a CPU on task wakeup, select_idle_smt() has to take
into account the scheduling domain of @target. This is because the
"isolcpus" kernel command line option can remove CPUs from the domain to
isolate them from other SMT siblings.

This fix checks if the candidate CPU is in the target scheduling domain.

Commit:

df3cb4ea1fb6 ("sched/fair: Fix wrong cpu selecting from isolated domain")

... originally introduced this fix by adding the check of the scheduling
domain in the loop.

However, commit:

3e6efe87cd5cc ("sched/fair: Remove redundant check in select_idle_smt()")

... accidentally removed the check. Bring it back.

Fixes: 3e6efe87cd5c ("sched/fair: Remove redundant check in select_idle_smt()")
Signed-off-by: Keisuke Nishimura <keisuke.nishimura@inria.fr>
Signed-off-by: Julia Lawall <julia.lawall@inria.fr>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20240110131707.437301-1-keisuke.nishimura@inria.fr

sched/fair: Add READ_ONCE() and use existing helper function to access ->avg_irq

Use existing helper function cpu_util_irq() instead of open-coding
access to ->avg_irq.

During review it was noted that ->avg_irq could be updated by a
different CPU than the one which is trying to access it.

->avg_irq is updated with WRITE_ONCE(), use READ_ONCE to access it
in order to avoid any compiler optimizations.

Signed-off-by: Shrikanth Hegde <sshegde@linux.vnet.ibm.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20240101154624.100981-3-sshegde@linux.vnet.ibm.com

sched/fair: Use existing helper functions to access ->avg_rt and ->avg_dl

There are helper functions called cpu_util_dl() and cpu_util_rt() which give
the average utilization of DL and RT respectively. But there are a few
places in code where access to these variables is open-coded.

Instead use the helper function so that code becomes simpler and easier to
maintain later on.

No functional changes intended.

Signed-off-by: Shrikanth Hegde <sshegde@linux.vnet.ibm.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20240101154624.100981-2-sshegde@linux.vnet.ibm.com

sched/core: Simplify code by removing duplicate #ifdefs

There's a few cases of nested #ifdefs in the scheduler code
that can be simplified:

#ifdef DEFINE_A
...code block...
#ifdef DEFINE_A <-- This is a duplicate.
...code block...
#endif
#else
#ifndef DEFINE_A <-- This is also duplicate.
...code block...
#endif
#endif

More details about the script and methods used to find these code
patterns can be found at:

https://lore.kernel.org/all/20240118080326.13137-1-sshegde@linux.ibm.com/

No change in functionality intended.

[ mingo: Clarified the changelog. ]

Signed-off-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20240216061433.535522-1-sshegde@linux.ibm.com

sched: fair: move unused stub functions to header

These four functions have a normal definition for CONFIG_FAIR_GROUP_SCHED,
and empty one that is only referenced when FAIR_GROUP_SCHED is disabled
but CGROUP_SCHED is still enabled. If both are turned off, the functions
are still defined but the misisng prototype causes a W=1 warning:

kernel/sched/fair.c:12544:6: error: no previous prototype for 'free_fair_sched_group'
kernel/sched/fair.c:12546:5: error: no previous prototype for 'alloc_fair_sched_group'
kernel/sched/fair.c:12553:6: error: no previous prototype for 'online_fair_sched_group'
kernel/sched/fair.c:12555:6: error: no previous prototype for 'unregister_fair_sched_group'

Move the alternatives into the header as static inline functions with the
correct combination of #ifdef checks to avoid the warning without adding
even more complexity.

[A different patch with the same description got applied by accident
and was later reverted, but the original patch is still missing]

Link: https://lkml.kernel.org/r/20231123110506.707903-4-arnd@kernel.org
Fixes: 7aa55f2a5902 ("sched/fair: Move unused stub functions to header")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: David Woodhouse <dwmw2@infradead.org>
Cc: Dinh Nguyen <dinguyen@kernel.org>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Cc: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Cc: Kees Cook <keescook@chromium.org>
Cc: Masahiro Yamada <masahiroy@kernel.org>
Cc: Matt Turner <mattst88@gmail.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Nathan Chancellor <nathan@kernel.org>
Cc: Nicolas Schier <nicolas@fjasle.eu>
Cc: Palmer Dabbelt <palmer@rivosinc.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Richard Henderson <richard.henderson@linaro.org>
Cc: Richard Weinberger <richard@nod.at>
Cc: Rich Felker <dalias@libc.org>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Cc: Tudor Ambarus <tudor.ambarus@linaro.org>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Cc: Zhihao Cheng <chengzhihao1@huawei.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

sched/fair: Fix tg->load when offlining a CPU

When a CPU is taken offline, the contribution of its cfs_rqs to task_groups'
load may remain and will negatively impact the calculation of the share of
the online CPUs.

To fix this bug, clear the contribution of an offlining CPU to task groups'
load and skip its contribution while it is inactive.

Here's the reproducer of the anomaly, by Imran Khan:

"So far I have encountered only one rather lengthy way of reproducing this issue,
which is as follows:

1. Take a KVM guest (booted with 4 CPUs and can be scaled up to 124 CPUs) and
create 2 custom cgroups: /sys/fs/cgroup/cpu/test_group_1 and /sys/fs/cgroup/
cpu/test_group_2

2. Assign a CPU intensive workload to each of these cgroups and start the
workload.

For my tests I am using following app:

int main(int argc, char *argv[])
{
unsigned long count, i, val;
if (argc != 2) {
printf("usage: ./a.out <number of random nums to generate> \n");
return 0;
}

count = strtoul(argv[1], NULL, 10);

printf("Generating %lu random numbers \n", count);
for (i = 0; i < count; i++) {
val = rand();
val = val % 2;
//usleep(1);
}
printf("Generated %lu random numbers \n", count);
return 0;
}

Also since the system is booted with 4 CPUs, in order to completely load the
system I am also launching 4 instances of same test app under:

/sys/fs/cgroup/cpu/

3. We can see that both of the cgroups get similar CPU time:

# systemd-cgtop --depth 1
Path Tasks %CPU Memory Input/s Output/s
/ 659 - 5.5G - -
/system.slice - - 5.7G - -
/test_group_1 4 - - - -
/test_group_2 3 - - - -
/user.slice 31 - 56.5M - -

Path Tasks %CPU Memory Input/s Output/s
/ 659 394.6 5.5G - -
/test_group_2 3 65.7 - - -
/user.slice 29 55.1 48.0M - -
/test_group_1 4 47.3 - - -
/system.slice - 2.2 5.7G - -

Path Tasks %CPU Memory Input/s Output/s
/ 659 394.8 5.5G - -
/test_group_1 4 62.9 - - -
/user.slice 28 44.9 54.2M - -
/test_group_2 3 44.7 - - -
/system.slice - 0.9 5.7G - -

Path Tasks %CPU Memory Input/s Output/s
/ 659 394.4 5.5G - -
/test_group_2 3 58.8 - - -
/test_group_1 4 51.9 - - -
/user.slice 30 39.3 59.6M - -
/system.slice - 1.9 5.7G - -

Path Tasks %CPU Memory Input/s Output/s
/ 659 394.7 5.5G - -
/test_group_1 4 60.9 - - -
/test_group_2 3 57.9 - - -
/user.slice 28 43.5 36.9M - -
/system.slice - 3.0 5.7G - -

Path Tasks %CPU Memory Input/s Output/s
/ 659 395.0 5.5G - -
/test_group_1 4 66.8 - - -
/test_group_2 3 56.3 - - -
/user.slice 29 43.1 51.8M - -
/system.slice - 0.7 5.7G - -

4. Now move systemd-udevd to one of these test groups, say test_group_1, and
perform scale up to 124 CPUs followed by scale down back to 4 CPUs from the
host side.

5. Run the same workload i.e 4 instances of CPU hogger under /sys/fs/cgroup/cpu
and one instance of CPU hogger each in /sys/fs/cgroup/cpu/test_group_1 and
/sys/fs/cgroup/test_group_2.

It can be seen that test_group_1 (the one where systemd-udevd was moved) is getting
much less CPU time than the test_group_2, even though at this point of time both of
these groups have only CPU hogger running:

# systemd-cgtop --depth 1
Path Tasks %CPU Memory Input/s Output/s
/ 1219 - 5.4G - -
/system.slice - - 5.6G - -
/test_group_1 4 - - - -
/test_group_2 3 - - - -
/user.slice 26 - 91.3M - -

Path Tasks %CPU Memory Input/s Output/s
/ 1221 394.3 5.4G - -
/test_group_2 3 82.7 - - -
/test_group_1 4 14.3 - - -
/system.slice - 0.8 5.6G - -
/user.slice 26 0.4 91.2M - -

Path Tasks %CPU Memory Input/s Output/s
/ 1221 394.6 5.4G - -
/test_group_2 3 67.4 - - -
/system.slice - 24.6 5.6G - -
/test_group_1 4 12.5 - - -
/user.slice 26 0.4 91.2M - -

Path Tasks %CPU Memory Input/s Output/s
/ 1221 395.2 5.4G - -
/test_group_2 3 60.9 - - -
/system.slice - 27.9 5.6G - -
/test_group_1 4 12.2 - - -
/user.slice 26 0.4 91.2M - -

Path Tasks %CPU Memory Input/s Output/s
/ 1221 395.2 5.4G - -
/test_group_2 3 69.4 - - -
/test_group_1 4 13.9 - - -
/user.slice 28 1.6 92.0M - -
/system.slice - 1.0 5.6G - -

Path Tasks %CPU Memory Input/s Output/s
/ 1221 395.6 5.4G - -
/test_group_2 3 59.3 - - -
/test_group_1 4 14.1 - - -
/user.slice 28 1.3 92.2M - -
/system.slice - 0.7 5.6G - -

Path Tasks %CPU Memory Input/s Output/s
/ 1221 395.5 5.4G - -
/test_group_2 3 67.2 - - -
/test_group_1 4 11.5 - - -
/user.slice 28 1.3 92.5M - -
/system.slice - 0.6 5.6G - -

Path Tasks %CPU Memory Input/s Output/s
/ 1221 395.1 5.4G - -
/test_group_2 3 76.8 - - -
/test_group_1 4 12.9 - - -
/user.slice 28 1.3 92.8M - -
/system.slice - 1.2 5.6G - -

From sched_debug data it can be seen that in bad case the load.weight of per-CPU
sched entities corresponding to test_group_1 has reduced significantly and
also load_avg of test_group_1 remains much higher than that of test_group_2,
even though systemd-udevd stopped running long time back and at this point of
time both cgroups just have the CPU hogger app as running entity."

[ mingo: Added details from the original discussion, plus minor edits to the patch. ]

Reported-by: Imran Khan <imran.f.khan@oracle.com>
Tested-by: Imran Khan <imran.f.khan@oracle.com>
Tested-by: Aaron Lu <aaron.lu@intel.com>
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Imran Khan <imran.f.khan@oracle.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Borislav Petkov <bp@alien8.de>
Link: https://lore.kernel.org/r/20231223111545.62135-1-vincent.guittot@linaro.org

sched/fair: Remove unused 'next_buddy_marked' local variable in check_preempt_wakeup_fair()

This variable became unused in:

5e963f2bd465 ("sched/fair: Commit to EEVDF")

Signed-off-by: Wang Jinchao <wangjinchao@xfusion.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/202312141319+0800-wangjinchao@xfusion.com

sched/fair: Use all little CPUs for CPU-bound workloads

Running N CPU-bound tasks on an N CPUs platform:

- with asymmetric CPU capacity

- not being a DynamIq system (i.e. having a PKG level sched domain
without the SD_SHARE_PKG_RESOURCES flag set)

.. might result in a task placement where two tasks run on a big CPU
and none on a little CPU. This placement could be more optimal by
using all CPUs.

Testing platform:

Juno-r2:
- 2 big CPUs (1-2), maximum capacity of 1024
- 4 little CPUs (0,3-5), maximum capacity of 383

Testing workload ([1]):

Spawn 6 CPU-bound tasks. During the first 100ms (step 1), each tasks
is affine to a CPU, except for:

- one little CPU which is left idle.
- one big CPU which has 2 tasks affine.

After the 100ms (step 2), remove the cpumask affinity.

Behavior before the patch:

During step 2, the load balancer running from the idle CPU tags sched
domains as:

- little CPUs: 'group_has_spare'. Cf. group_has_capacity() and
group_is_overloaded(), 3 CPU-bound tasks run on a 4 CPUs
sched-domain, and the idle CPU provides enough spare capacity
regarding the imbalance_pct

- big CPUs: 'group_overloaded'. Indeed, 3 tasks run on a 2 CPUs
sched-domain, so the following path is used:

group_is_overloaded()
\-if (sgs->sum_nr_running <= sgs->group_weight) return true;

The following path which would change the migration type to
'migrate_task' is not taken:

calculate_imbalance()
\-if (env->idle != CPU_NOT_IDLE && env->imbalance == 0)

as the local group has some spare capacity, so the imbalance
is not 0.

The migration type requested is 'migrate_util' and the busiest
runqueue is the big CPU's runqueue having 2 tasks (each having a
utilization of 512). The idle little CPU cannot pull one of these
task as its capacity is too small for the task. The following path
is used:

detach_tasks()
\-case migrate_util:
\-if (util > env->imbalance) goto next;

After the patch:

As the number of failed balancing attempts grows (with
'nr_balance_failed'), progressively make it easier to migrate
a big task to the idling little CPU. A similar mechanism is
used for the 'migrate_load' migration type.

Improvement:

Running the testing workload [1] with the step 2 representing
a ~10s load for a big CPU:

Before patch: ~19.3s
After patch: ~18s (-6.7%)

Similar issue reported at:

https://lore.kernel.org/lkml/20230716014125.139577-1-qyousef@layalina.io/

Suggested-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Pierre Gondois <pierre.gondois@arm.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Acked-by: Qais Yousef <qyousef@layalina.io>
Link: https://lore.kernel.org/r/20231206090043.634697-1-pierre.gondois@arm.com

sched/fair: Simplify util_est

With UTIL_EST_FASTUP now being permanent, we can take advantage of the
fact that the ewma jumps directly to a higher utilization at dequeue to
simplify util_est and remove the enqueued field.

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Tested-by: Lukasz Luba <lukasz.luba@arm.com>
Reviewed-by: Lukasz Luba <lukasz.luba@arm.com>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Reviewed-by: Hongyan Xia <hongyan.xia2@arm.com>
Reviewed-by: Alex Shi <alexs@kernel.org>
Link: https://lore.kernel.org/r/20231201161652.1241695-3-vincent.guittot@linaro.org

sched/fair: Remove SCHED_FEAT(UTIL_EST_FASTUP, true)

sched_feat(UTIL_EST_FASTUP) has been added to easily disable the feature
in order to check for possibly related regressions. After 3 years, it has
never been used and no regression has been reported. Let's remove it
and make fast increase a permanent behavior.

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Tested-by: Lukasz Luba <lukasz.luba@arm.com>
Reviewed-by: Lukasz Luba <lukasz.luba@arm.com>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Reviewed-by: Hongyan Xia <hongyan.xia2@arm.com>
Reviewed-by: Tang Yizhou <yizhou.tang@shopee.com>
Reviewed-by: Yanteng Si <siyanteng@loongson.cn> [for the Chinese translation]
Reviewed-by: Alex Shi <alexs@kernel.org>
Link: https://lore.kernel.org/r/20231201161652.1241695-2-vincent.guittot@linaro.org

sched/fair: Update min_vruntime for reweight_entity() correctly

Since reweight_entity() may have chance to change the weight of
cfs_rq->curr entity, we should also update_min_vruntime() if
this is the case

Fixes: eab03c23c2a1 ("sched/eevdf: Fix vruntime adjustment on reweight")
Signed-off-by: Yiwei Lin <s921975628@gmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Abel Wu <wuyun.abel@bytedance.com>
Link: https://lore.kernel.org/r/20231117080106.12890-1-s921975628@gmail.com

sched/cpufreq: Rework schedutil governor performance estimation

The current method to take into account uclamp hints when estimating the
target frequency can end in a situation where the selected target
frequency is finally higher than uclamp hints, whereas there are no real
needs. Such cases mainly happen because we are currently mixing the
traditional scheduler utilization signal with the uclamp performance
hints. By adding these 2 metrics, we loose an important information when
it comes to select the target frequency, and we have to make some
assumptions which can't fit all cases.

Rework the interface between the scheduler and schedutil governor in order
to propagate all information down to the cpufreq governor.

effective_cpu_util() interface changes and now returns the actual
utilization of the CPU with 2 optional inputs:

- The minimum performance for this CPU; typically the capacity to handle
the deadline task and the interrupt pressure. But also uclamp_min
request when available.

- The maximum targeting performance for this CPU which reflects the
maximum level that we would like to not exceed. By default it will be
the CPU capacity but can be reduced because of some performance hints
set with uclamp. The value can be lower than actual utilization and/or
min performance level.

A new sugov_effective_cpu_perf() interface is also available to compute
the final performance level that is targeted for the CPU, after applying
some cpufreq headroom and taking into account all inputs.

With these 2 functions, schedutil is now able to decide when it must go
above uclamp hints. It now also has a generic way to get the min
performance level.

The dependency between energy model and cpufreq governor and its headroom
policy doesn't exist anymore.

eenv_pd_max_util() asks schedutil for the targeted performance after
applying the impact of the waking task.

[ mingo: Refined the changelog & C comments. ]

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Rafael J. Wysocki <rafael@kernel.org>
Link: https://lore.kernel.org/r/20231122133904.446032-2-vincent.guittot@linaro.org

sched/pelt: Avoid underestimation of task utilization

Lukasz Luba reported that a thread's util_est can significantly decrease as
a result of sharing the CPU with other threads.

The use case can be easily reproduced with a periodic task TA that runs 1ms
and sleeps 100us. When the task is alone on the CPU, its max utilization and
its util_est is around 888. If another similar task starts to run on the
same CPU, TA will have to share the CPU runtime and its maximum utilization
will decrease around half the CPU capacity (512) then TA's util_est will
follow this new maximum trend which is only the result of sharing the CPU
with others tasks.

Such situation can be detected with runnable_avg wich is close or
equal to util_avg when TA is alone, but increases above util_avg when TA
shares the CPU with other threads and wait on the runqueue.

[ We prefer an util_est that overestimate rather than under estimate
because in 1st case we will not provide enough performance to the
task which will remain under-provisioned, whereas in the other case we
will create some idle time which will enable to reduce contention and
as a result reduces the util_est so the overestimate will be transient
whereas the underestimate will remain. ]

[ mingo: Refined the changelog, added comments from the LKML discussion. ]

Reported-by: Lukasz Luba <lukasz.luba@arm.com>
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/lkml/CAKfTPtDd-HhF-YiNTtL9i5k0PfJbF819Yxu4YquzfXgwi7voyw@mail.gmail.com/#t
Link: https://lore.kernel.org/r/20231122140119.472110-1-vincent.guittot@linaro.org
Cc: Hongyan Xia <hongyan.xia2@arm.com>

sched/deadline: Introduce deadline servers

Low priority tasks (e.g., SCHED_OTHER) can suffer starvation if tasks
with higher priority (e.g., SCHED_FIFO) monopolize CPU(s).

RT Throttling has been introduced a while ago as a (mostly debug)
countermeasure one can utilize to reserve some CPU time for low priority
tasks (usually background type of work, e.g. workqueues, timers, etc.).
It however has its own problems (see documentation) and the undesired
effect of unconditionally throttling FIFO tasks even when no lower
priority activity needs to run (there are mechanisms to fix this issue
as well, but, again, with their own problems).

Introduce deadline servers to service low priority tasks needs under
starvation conditions. Deadline servers are built extending SCHED_DEADLINE
implementation to allow 2-level scheduling (a sched_deadline entity
becomes a container for lower priority scheduling entities).

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Daniel Bristot de Oliveira <bristot@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/4968601859d920335cf85822eb573a5f179f04b8.1699095159.git.bristot@kernel.org

sched: Unify more update_curr*()

Now that trace_sched_stat_runtime() no longer takes a vruntime
argument, the task specific bits are identical between
update_curr_common() and update_curr().

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>

sched: Remove vruntime from trace_sched_stat_runtime()

Tracing the runtime delta makes sense, observer can sum over time.
Tracing the absolute vruntime makes less sense, inconsistent:
absolute-vs-delta, but also vruntime delta can be computed from
runtime delta.

Removing the vruntime thing also makes the two tracepoint sites
identical, allowing to unify the code in a later patch.

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>

sched: Unify runtime accounting across classes

All classes use sched_entity::exec_start to track runtime and have
copies of the exact same code around to compute runtime.

Collapse all that.

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Daniel Bristot de Oliveira <bristot@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Phil Auld <pauld@redhat.com>
Reviewed-by: Valentin Schneider <vschneid@redhat.com>
Reviewed-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Link: https://lkml.kernel.org/r/54d148a144f26d9559698c4dd82d8859038a7380.1699095159.git.bristot@kernel.org

sched/eevdf: O(1) fastpath for task selection

Since the RB-tree is now sorted by deadline, let's first try the
leftmost entity which has the earliest virtual deadline. I've done
some benchmarks to see its effectiveness.

All the benchmarks are done inside a normal cpu cgroup in a clean
environment with cpu turbo disabled, on a dual-CPU Intel Xeon(R)
Platinum 8260 with 2 NUMA nodes each of which has 24C/48T.

hackbench: process/thread + pipe/socket + 1/2/4/8 groups
netperf: TCP/UDP + STREAM/RR + 24/48/72/96/192 threads
tbench: loopback 24/48/72/96/192 threads
schbench: 1/2/4/8 mthreads

direct: cfs_rq has only one entity
parity: RUN_TO_PARITY
fast: O(1) fastpath
slow: heap search

(%) direct parity fast slow
hackbench 92.95 2.02 4.91 0.12
netperf 68.08 6.60 24.18 1.14
tbench 67.55 11.22 20.61 0.62
schbench 69.91 2.65 25.73 1.71

The above results indicate that this fastpath really makes task
selection more efficient.

Signed-off-by: Abel Wu <wuyun.abel@bytedance.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20231115033647.80785-4-wuyun.abel@bytedance.com

sched/eevdf: Sort the rbtree by virtual deadline

Sort the task timeline by virtual deadline and keep the min_vruntime
in the augmented tree, so we can avoid doubling the worst case cost
and make full use of the cached leftmost node to enable O(1) fastpath
picking in next patch.

Signed-off-by: Abel Wu <wuyun.abel@bytedance.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20231115033647.80785-3-wuyun.abel@bytedance.com

sched/numa: Fix mm numa_scan_seq based unconditional scan

Since commit fc137c0ddab2 ("sched/numa: enhance vma scanning logic")

NUMA Balancing allows updating PTEs to trap NUMA hinting faults if the
task had previously accessed VMA. However unconditional scan of VMAs are
allowed during initial phase of VMA creation until process's
mm numa_scan_seq reaches 2 even though current task had not accessed VMA.

Rationale:
- Without initial scan subsequent PTE update may never happen.
- Give fair opportunity to all the VMAs to be scanned and subsequently
understand the access pattern of all the VMAs.

But it has a corner case where, if a VMA is created after some time,
process's mm numa_scan_seq could be already greater than 2.

For e.g., values of mm numa_scan_seq when VMAs are created by running
mmtest autonuma benchmark briefly looks like:
start_seq=0 : 459
start_seq=2 : 138
start_seq=3 : 144
start_seq=4 : 8
start_seq=8 : 1
start_seq=9 : 1
This results in no unconditional PTE updates for those VMAs created after
some time.

Fix:
- Note down the initial value of mm numa_scan_seq in per VMA start_seq.
- Allow unconditional scan till start_seq + 2.

Result:
SUT: AMD EPYC Milan with 2 NUMA nodes 256 cpus.
base kernel: upstream 6.6-rc6 with Mels patches [1] applied.

kernbench
========== base patched %gain
Amean elsp-128 165.09 ( 0.00%) 164.78 * 0.19%*

Duration User 41404.28 41375.08
Duration System 9862.22 9768.48
Duration Elapsed 519.87 518.72

Ops NUMA PTE updates 1041416.00 831536.00
Ops NUMA hint faults 263296.00 220966.00
Ops NUMA pages migrated 258021.00 212769.00
Ops AutoNUMA cost 1328.67 1114.69

autonumabench

NUMA01_THREADLOCAL
==================
Amean elsp-NUMA01_THREADLOCAL 81.79 (0.00%) 67.74 * 17.18%*

Duration User 54832.73 47379.67
Duration System 75.00 185.75
Duration Elapsed 576.72 476.09

Ops NUMA PTE updates 394429.00 11121044.00
Ops NUMA hint faults 1001.00 8906404.00
Ops NUMA pages migrated 288.00 2998694.00
Ops AutoNUMA cost 7.77 44666.84

Signed-off-by: Raghavendra K T <raghavendra.kt@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Mel Gorman <mgorman@suse.de>
Link: https://lore.kernel.org/r/2ea7cbce80ac7c62e90cbfb9653a7972f902439f.1697816692.git.raghavendra.kt@amd.com

sched/fair: Fix the decision for load balance

should_we_balance is called for the decision to do load-balancing.
When sched ticks invoke this function, only one CPU should return
true. However, in the current code, two CPUs can return true. The
following situation, where b means busy and i means idle, is an
example, because CPU 0 and CPU 2 return true.

[0, 1] [2, 3]
b b i b

This fix checks if there exists an idle CPU with busy sibling(s)
after looking for a CPU on an idle core. If some idle CPUs with busy
siblings are found, just the first one should do load-balancing.

Fixes: b1bfeab9b002 ("sched/fair: Consider the idle state of the whole core for load balance")
Signed-off-by: Keisuke Nishimura <keisuke.nishimura@inria.fr>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Chen Yu <yu.c.chen@intel.com>
Reviewed-by: Shrikanth Hegde <sshegde@linux.vnet.ibm.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lkml.kernel.org/r/20231031133821.1570861-1-keisuke.nishimura@inria.fr

sched/eevdf: Fix vruntime adjustment on reweight

vruntime of the (on_rq && !0-lag) entity needs to be adjusted when
it gets re-weighted, and the calculations can be simplified based
on the fact that re-weight won't change the w-average of all the
entities. Please check the proofs in comments.

But adjusting vruntime can also cause position change in RB-tree
hence require re-queue to fix up which might be costly. This might
be avoided by deferring adjustment to the time the entity actually
leaves tree (dequeue/pick), but that will negatively affect task
selection and probably not good enough either.

Fixes: 147f3efaa241 ("sched/fair: Implement an EEVDF-like scheduling policy")
Signed-off-by: Abel Wu <wuyun.abel@bytedance.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20231107090510.71322-2-wuyun.abel@bytedance.com

sched/fair: use folio_xchg_last_cpupid() in should_numa_migrate_memory()

Convert to use folio_xchg_last_cpupid() in should_numa_migrate_memory().

Link: https://lkml.kernel.org/r/20231018140806.2783514-14-wangkefeng.wang@huawei.com
Signed-off-by: Kefeng Wang <wangkefeng.wang@huawei.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Huang Ying <ying.huang@intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Zi Yan <ziy@nvidia.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

sched/fair: use folio_xchg_access_time() in numa_hint_fault_latency()

Convert to use folio_xchg_access_time() in numa_hint_fault_latency().

Link: https://lkml.kernel.org/r/20231018140806.2783514-9-wangkefeng.wang@huawei.com
Signed-off-by: Kefeng Wang <wangkefeng.wang@huawei.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Huang Ying <ying.huang@intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Zi Yan <ziy@nvidia.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

sched/numa, mm: make numa migrate functions to take a folio

The cpupid (or access time) is stored in the head page for THP, so it is
safely to make should_numa_migrate_memory() and numa_hint_fault_latency()
to take a folio. This is in preparation for large folio numa balancing.

Link: https://lkml.kernel.org/r/20230921074417.24004-7-wangkefeng.wang@huawei.com
Signed-off-by: Kefeng Wang <wangkefeng.wang@huawei.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: "Huang, Ying" <ying.huang@intel.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Zi Yan <ziy@nvidia.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

sched/fair: Remove SIS_PROP

SIS_UTIL seems to work well, lets remove the old thing.

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lkml.kernel.org/r/20231020134337.GD33965@noisy.programming.kicks-ass.net

sched/fair: Use candidate prev/recent_used CPU if scanning failed for cluster wakeup

Chen Yu reports a hackbench regression of cluster wakeup when
hackbench threads equal to the CPU number [1]. Analysis shows
it's because we wake up more on the target CPU even if the
prev_cpu is a good wakeup candidate and leads to the decrease
of the CPU utilization.

Generally if the task's prev_cpu is idle we'll wake up the task
on it without scanning. On cluster machines we'll try to wake up
the task in the same cluster of the target for better cache
affinity, so if the prev_cpu is idle but not sharing the same
cluster with the target we'll still try to find an idle CPU within
the cluster. This will improve the performance at low loads on
cluster machines. But in the issue above, if the prev_cpu is idle
but not in the cluster with the target CPU, we'll try to scan an
idle one in the cluster. But since the system is busy, we're
likely to fail the scanning and use target instead, even if
the prev_cpu is idle. Then leads to the regression.

This patch solves this in 2 steps:
o record the prev_cpu/recent_used_cpu if they're good wakeup
candidates but not sharing the cluster with the target.
o on scanning failure use the prev_cpu/recent_used_cpu if
they're recorded as idle

[1] https://lore.kernel.org/all/ZGzDLuVaHR1PAYDt@chenyu5-mobl1/

Closes: https://lore.kernel.org/all/ZGsLy83wPIpamy6x@chenyu5-mobl1/
Reported-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Yicong Yang <yangyicong@hisilicon.com>
Tested-and-reviewed-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lkml.kernel.org/r/20231019033323.54147-4-yangyicong@huawei.com

sched/fair: Scan cluster before scanning LLC in wake-up path

For platforms having clusters like Kunpeng920, CPUs within the same cluster
have lower latency when synchronizing and accessing shared resources like
cache. Thus, this patch tries to find an idle cpu within the cluster of the
target CPU before scanning the whole LLC to gain lower latency. This
will be implemented in 2 steps in select_idle_sibling():
1. When the prev_cpu/recent_used_cpu are good wakeup candidates, use them
if they're sharing cluster with the target CPU. Otherwise trying to
scan for an idle CPU in the target's cluster.
2. Scanning the cluster prior to the LLC of the target CPU for an
idle CPU to wakeup.

Testing has been done on Kunpeng920 by pinning tasks to one numa and two
numa. On Kunpeng920, Each numa has 8 clusters and each cluster has 4 CPUs.

With this patch, We noticed enhancement on tbench and netperf within one
numa or cross two numa on top of tip-sched-core commit
9b46f1abc6d4 ("sched/debug: Print 'tgid' in sched_show_task()")

tbench results (node 0):
baseline patched
1: 327.2833 372.4623 ( 13.80%)
4: 1320.5933 1479.8833 ( 12.06%)
8: 2638.4867 2921.5267 ( 10.73%)
16: 5282.7133 5891.5633 ( 11.53%)
32: 9810.6733 9877.3400 ( 0.68%)
64: 7408.9367 7447.9900 ( 0.53%)
128: 6203.2600 6191.6500 ( -0.19%)
tbench results (node 0-1):
baseline patched
1: 332.0433 372.7223 ( 12.25%)
4: 1325.4667 1477.6733 ( 11.48%)
8: 2622.9433 2897.9967 ( 10.49%)
16: 5218.6100 5878.2967 ( 12.64%)
32: 10211.7000 11494.4000 ( 12.56%)
64: 13313.7333 16740.0333 ( 25.74%)
128: 13959.1000 14533.9000 ( 4.12%)

netperf results TCP_RR (node 0):
baseline patched
1: 76546.5033 90649.9867 ( 18.42%)
4: 77292.4450 90932.7175 ( 17.65%)
8: 77367.7254 90882.3467 ( 17.47%)
16: 78519.9048 90938.8344 ( 15.82%)
32: 72169.5035 72851.6730 ( 0.95%)
64: 25911.2457 25882.2315 ( -0.11%)
128: 10752.6572 10768.6038 ( 0.15%)

netperf results TCP_RR (node 0-1):
baseline patched
1: 76857.6667 90892.2767 ( 18.26%)
4: 78236.6475 90767.3017 ( 16.02%)
8: 77929.6096 90684.1633 ( 16.37%)
16: 77438.5873 90502.5787 ( 16.87%)
32: 74205.6635 88301.5612 ( 19.00%)
64: 69827.8535 71787.6706 ( 2.81%)
128: 25281.4366 25771.3023 ( 1.94%)

netperf results UDP_RR (node 0):
baseline patched
1: 96869.8400 110800.8467 ( 14.38%)
4: 97744.9750 109680.5425 ( 12.21%)
8: 98783.9863 110409.9637 ( 11.77%)
16: 99575.0235 110636.2435 ( 11.11%)
32: 95044.7250 97622.8887 ( 2.71%)
64: 32925.2146 32644.4991 ( -0.85%)
128: 12859.2343 12824.0051 ( -0.27%)

netperf results UDP_RR (node 0-1):
baseline patched
1: 97202.4733 110190.1200 ( 13.36%)
4: 95954.0558 106245.7258 ( 10.73%)
8: 96277.1958 105206.5304 ( 9.27%)
16: 97692.7810 107927.2125 ( 10.48%)
32: 79999.6702 103550.2999 ( 29.44%)
64: 80592.7413 87284.0856 ( 8.30%)
128: 27701.5770 29914.5820 ( 7.99%)

Note neither Kunpeng920 nor x86 Jacobsville supports SMT, so the SMT branch
in the code has not been tested but it supposed to work.

Chen Yu also noticed this will improve the performance of tbench and
netperf on a 24 CPUs Jacobsville machine, there are 4 CPUs in one
cluster sharing L2 Cache.

[https://lore.kernel.org/lkml/Ytfjs+m1kUs0ScSn@worktop.programming.kicks-ass.net]
Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Barry Song <song.bao.hua@hisilicon.com>
Signed-off-by: Yicong Yang <yangyicong@hisilicon.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Tim Chen <tim.c.chen@linux.intel.com>
Reviewed-by: Chen Yu <yu.c.chen@intel.com>
Reviewed-by: Gautham R. Shenoy <gautham.shenoy@amd.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Tested-and-reviewed-by: Chen Yu <yu.c.chen@intel.com>
Tested-by: Yicong Yang <yangyicong@hisilicon.com>
Link: https://lkml.kernel.org/r/20231019033323.54147-3-yangyicong@huawei.com

sched/fair: Remove unused 'curr' argument from pick_next_entity()

The 'curr' argument of pick_next_entity() has become unused after
the EEVDF changes.

[ mingo: Updated the changelog. ]

Signed-off-by: Yiwei Lin <s921975628@gmail.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20231020055617.42064-1-s921975628@gmail.com

sched/nohz: Update comments about NEWILB_KICK

How ILB is triggered without IPIs is cryptic. Out of mercy for future
code readers, document it in code comments.

The comments are derived from a discussion with Vincent in a past
review.

Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20231020014031.919742-2-joel@joelfernandes.org

sched/fair: Remove duplicate #include

./kernel/sched/fair.c: linux/sched/cond_resched.h is included more than once.

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20231018062759.44375-1-jiapeng.chong@linux.alibaba.com

Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=6907

sched: Fix stop_one_cpu_nowait() vs hotplug

Kuyo reported sporadic failures on a sched_setaffinity() vs CPU
hotplug stress-test -- notably affine_move_task() remains stuck in
wait_for_completion(), leading to a hung-task detector warning.

Specifically, it was reported that stop_one_cpu_nowait(.fn =
migration_cpu_stop) returns false -- this stopper is responsible for
the matching complete().

The race scenario is:

CPU0 CPU1

// doing _cpu_down()

__set_cpus_allowed_ptr()
task_rq_lock();
takedown_cpu()
stop_machine_cpuslocked(take_cpu_down..)

<PREEMPT: cpu_stopper_thread()
MULTI_STOP_PREPARE
...
__set_cpus_allowed_ptr_locked()
affine_move_task()
task_rq_unlock();

<PREEMPT: cpu_stopper_thread()\>
ack_state()
MULTI_STOP_RUN
take_cpu_down()
__cpu_disable();
stop_machine_park();
stopper->enabled = false;
/>
/>
stop_one_cpu_nowait(.fn = migration_cpu_stop);
if (stopper->enabled) // false!!!

That is, by doing stop_one_cpu_nowait() after dropping rq-lock, the
stopper thread gets a chance to preempt and allows the cpu-down for
the target CPU to complete.

OTOH, since stop_one_cpu_nowait() / cpu_stop_queue_work() needs to
issue a wakeup, it must not be ran under the scheduler locks.

Solve this apparent contradiction by keeping preemption disabled over
the unlock + queue_stopper combination:

preempt_disable();
task_rq_unlock(...);
if (!stop_pending)
stop_one_cpu_nowait(...)
preempt_enable();

This respects the lock ordering contraints while still avoiding the
above race. That is, if we find the CPU is online under rq-lock, the
targeted stop_one_cpu_nowait() must succeed.

Apply this pattern to all similar stop_one_cpu_nowait() invocations.

Fixes: 6d337eab041d ("sched: Fix migrate_disable() vs set_cpus_allowed_ptr()")
Reported-by: "Kuyo Chang (張建文)" <Kuyo.Chang@mediatek.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: "Kuyo Chang (張建文)" <Kuyo.Chang@mediatek.com>
Link: https://lkml.kernel.org/r/20231010200442.GA16515@noisy.programming.kicks-ass.net

sched/topology: Rename 'DIE' domain to 'PKG'

While reworking the x86 topology code Thomas tripped over creating a 'DIE' domain
for the package mask. :-)

Since these names are CONFIG_SCHED_DEBUG=y only, rename them to make the
name less ambiguous.

[ Shrikanth Hegde: rename on s390 as well. ]
[ Valentin Schneider: also rename it in the comments. ]
[ mingo: port to recent kernels & find all remaining occurances. ]

Reported-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Valentin Schneider <vschneid@redhat.com>
Acked-by: Mel Gorman <mgorman@suse.de>
Acked-by: Heiko Carstens <hca@linux.ibm.com>
Acked-by: Gautham R. Shenoy <gautham.shenoy@amd.com>
Acked-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20230712141056.GI3100107@hirez.programming.kicks-ass.net

sched/numa: Complete scanning of inactive VMAs when there is no alternative

VMAs are skipped if there is no recent fault activity but this represents
a chicken-and-egg problem as there may be no fault activity if the PTEs
are never updated to trap NUMA hints. There is an indirect reliance on
scanning to be forced early in the lifetime of a task but this may fail
to detect changes in phase behaviour. Force inactive VMAs to be scanned
when all other eligible VMAs have been updated within the same scan
sequence.

Test results in general look good with some changes in performance, both
negative and positive, depending on whether the additional scanning and
faulting was beneficial or not to the workload. The autonuma benchmark
workload NUMA01_THREADLOCAL was picked for closer examination. The workload
creates two processes with numerous threads and thread-local storage that
is zero-filled in a loop. It exercises the corner case where unrelated
threads may skip VMAs that are thread-local to another thread and still
has some VMAs that inactive while the workload executes.

The VMA skipping activity frequency with and without the patch:

6.6.0-rc2-sched-numabtrace-v1
=============================
649 reason=scan_delay
9,094 reason=unsuitable
48,915 reason=shared_ro
143,919 reason=inaccessible
193,050 reason=pid_inactive

6.6.0-rc2-sched-numabselective-v1
=============================
146 reason=seq_completed
622 reason=ignore_pid_inactive

624 reason=scan_delay
6,570 reason=unsuitable
16,101 reason=shared_ro
27,608 reason=inaccessible
41,939 reason=pid_inactive

Note that with the patch applied, the PID activity is ignored
(ignore_pid_inactive) to ensure a VMA with some activity is completely
scanned. In addition, a small number of VMAs are scanned when no other
eligible VMA is available during a single scan window (seq_completed).
The number of times a VMA is skipped due to no PID activity from the
scanning task (pid_inactive) drops dramatically. It is expected that
this will increase the number of PTEs updated for NUMA hinting faults
as well as hinting faults but these represent PTEs that would otherwise
have been missed. The tradeoff is scan+fault overhead versus improving
locality due to migration.

On a 2-socket Cascade Lake test machine, the time to complete the
workload is as follows;

6.6.0-rc2 6.6.0-rc2
sched-numabtrace-v1 sched-numabselective-v1
Min elsp-NUMA01_THREADLOCAL 174.22 ( 0.00%) 117.64 ( 32.48%)
Amean elsp-NUMA01_THREADLOCAL 175.68 ( 0.00%) 123.34 * 29.79%*
Stddev elsp-NUMA01_THREADLOCAL 1.20 ( 0.00%) 4.06 (-238.20%)
CoeffVar elsp-NUMA01_THREADLOCAL 0.68 ( 0.00%) 3.29 (-381.70%)
Max elsp-NUMA01_THREADLOCAL 177.18 ( 0.00%) 128.03 ( 27.74%)

The time to complete the workload is reduced by almost 30%:

6.6.0-rc2 6.6.0-rc2
sched-numabtrace-v1 sched-numabselective-v1 /
Duration User 91201.80 63506.64
Duration System 2015.53 1819.78
Duration Elapsed 1234.77 868.37

In this specific case, system CPU time was not increased but it's not
universally true.

From vmstat, the NUMA scanning and fault activity is as follows;

6.6.0-rc2 6.6.0-rc2
sched-numabtrace-v1 sched-numabselective-v1
Ops NUMA base-page range updates 64272.00 26374386.00
Ops NUMA PTE updates 36624.00 55538.00
Ops NUMA PMD updates 54.00 51404.00
Ops NUMA hint faults 15504.00 75786.00
Ops NUMA hint local faults % 14860.00 56763.00
Ops NUMA hint local percent 95.85 74.90
Ops NUMA pages migrated 1629.00 6469222.00

Both the number of PTE updates and hint faults is dramatically
increased. While this is superficially unfortunate, it represents
ranges that were simply skipped without the patch. As a result
of the scanning and hinting faults, many more pages were also
migrated but as the time to completion is reduced, the overhead
is offset by the gain.

Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Tested-by: Raghavendra K T <raghavendra.kt@amd.com>
Link: https://lore.kernel.org/r/20231010083143.19593-7-mgorman@techsingularity.net

sched/numa: Complete scanning of partial VMAs regardless of PID activity

NUMA Balancing skips VMAs when the current task has not trapped a NUMA
fault within the VMA. If the VMA is skipped then mm->numa_scan_offset
advances and a task that is trapping faults within the VMA may never
fully update PTEs within the VMA.

Force tasks to update PTEs for partially scanned PTEs. The VMA will
be tagged for NUMA hints by some task but this removes some of the
benefit of tracking PID activity within a VMA. A follow-on patch
will mitigate this problem.

The test cases and machines evaluated did not trigger the corner case so
the performance results are neutral with only small changes within the
noise from normal test-to-test variance. However, the next patch makes
the corner case easier to trigger.

Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Tested-by: Raghavendra K T <raghavendra.kt@amd.com>
Link: https://lore.kernel.org/r/20231010083143.19593-6-mgorman@techsingularity.net

sched/numa: Move up the access pid reset logic

Recent NUMA hinting faulting activity is reset approximately every
VMA_PID_RESET_PERIOD milliseconds. However, if the current task has not
accessed a VMA then the reset check is missed and the reset is potentially
deferred forever. Check if the PID activity information should be reset
before checking if the current task recently trapped a NUMA hinting fault.

[ mgorman@techsingularity.net: Rewrite changelog ]

Suggested-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Raghavendra K T <raghavendra.kt@amd.com>
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20231010083143.19593-5-mgorman@techsingularity.net

sched/numa: Trace decisions related to skipping VMAs

NUMA balancing skips or scans VMAs for a variety of reasons. In preparation
for completing scans of VMAs regardless of PID access, trace the reasons
why a VMA was skipped. In a later patch, the tracing will be used to track
if a VMA was forcibly scanned.

Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20231010083143.19593-4-mgorman@techsingularity.net

sched/numa: Rename vma_numab_state::access_pids[] => ::pids_active[], ::next_pid_reset => ::pids_active_reset

The access_pids[] field name is somewhat ambiguous as no PIDs are accessed.
Similarly, it's not clear that next_pid_reset is related to access_pids[].
Rename the fields to more accurately reflect their purpose.

[ mingo: Rename in the comments too. ]

Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20231010083143.19593-3-mgorman@techsingularity.net

sched/topology: Consolidate and clean up access to a CPU's max compute capacity

Remove the rq::cpu_capacity_orig field and use arch_scale_cpu_capacity()
instead.

The scheduler uses 3 methods to get access to a CPU's max compute capacity:

- arch_scale_cpu_capacity(cpu) which is the default way to get a CPU's capacity.

- cpu_capacity_orig field which is periodically updated with
arch_scale_cpu_capacity().

- capacity_orig_of(cpu) which encapsulates rq->cpu_capacity_orig.

There is no real need to save the value returned by arch_scale_cpu_capacity()
in struct rq. arch_scale_cpu_capacity() returns:

- either a per_cpu variable.

- or a const value for systems which have only one capacity.

Remove rq::cpu_capacity_orig and use arch_scale_cpu_capacity() everywhere.

No functional changes.

Some performance tests on Arm64:

- small SMP device (hikey): no noticeable changes
- HMP device (RB5): hackbench shows minor improvement (1-2%)
- large smp (thx2): hackbench and tbench shows minor improvement (1%)

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Link: https://lore.kernel.org/r/20231009103621.374412-2-vincent.guittot@linaro.org

sched/nohz: Remove unnecessarily complex error handling pattern from find_new_ilb()

find_new_ilb() returns nr_cpu_ids on failure - which is the usual
cpumask bitops return pattern, but is weird & unnecessary in this
context: not only is it a global variable, it it is a +1 out of
bounds CPU index and also has different signedness ...

Its only user, kick_ilb(), then checks the return against nr_cpu_ids
to decide to return. There's no other use.

So instead of this, use a standard -1 return on failure to find an
idle CPU, as the argument is signed already.

Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Link: https://lore.kernel.org/r/20231006102518.2452758-4-mingo@kernel.org

sched/nohz: Use consistent variable names in find_new_ilb() and kick_ilb()

Use 'ilb_cpu' consistently in both functions.

Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Link: https://lore.kernel.org/r/20231006102518.2452758-3-mingo@kernel.org

sched/nohz: Update idle load-balancing (ILB) comments

- Fix incorrect/misleading comments,

- clarify some others,

- fix typos & grammar,

- and use more consistent style throughout.

Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Link: https://lore.kernel.org/r/20231006102518.2452758-2-mingo@kernel.org

sched/debug: Add new tracepoint to track compute energy computation

It was useful to track feec() placement decision and debug the spare
capacity and optimization issues vs uclamp_max.

Signed-off-by: Qais Yousef (Google) <qyousef@layalina.io>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20230916232955.2099394-4-qyousef@layalina.io

sched/uclamp: Ignore (util == 0) optimization in feec() when p_util_max = 0

find_energy_efficient_cpu() bails out early if effective util of the
task is 0 as the delta at this point will be zero and there's nothing
for EAS to do. When uclamp is being used, this could lead to wrong
decisions when uclamp_max is set to 0. In this case the task is capped
to performance point 0, but it is actually running and consuming energy
and we can benefit from EAS energy calculations.

Rework the condition so that it bails out when both util and uclamp_min
are 0.

We can do that without needing to use uclamp_task_util(); remove it.

Fixes: d81304bc6193 ("sched/uclamp: Cater for uclamp in find_energy_efficient_cpu()'s early exit condition")
Signed-off-by: Qais Yousef (Google) <qyousef@layalina.io>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20230916232955.2099394-3-qyousef@layalina.io

sched/uclamp: Set max_spare_cap_cpu even if max_spare_cap is 0

When uclamp_max is being used, the util of the task could be higher than
the spare capacity of the CPU, but due to uclamp_max value we force-fit
it there.

The way the condition for checking for max_spare_cap in
find_energy_efficient_cpu() was constructed; it ignored any CPU that has
its spare_cap less than or _equal_ to max_spare_cap. Since we initialize
max_spare_cap to 0; this lead to never setting max_spare_cap_cpu and
hence ending up never performing compute_energy() for this cluster and
missing an opportunity for a better energy efficient placement to honour
uclamp_max setting.

max_spare_cap = 0;
cpu_cap = capacity_of(cpu) - cpu_util(p); // 0 if cpu_util(p) is high

...

util_fits_cpu(...); // will return true if uclamp_max forces it to fit

...

// this logic will fail to update max_spare_cap_cpu if cpu_cap is 0
if (cpu_cap > max_spare_cap) {
max_spare_cap = cpu_cap;
max_spare_cap_cpu = cpu;
}

prev_spare_cap suffers from a similar problem.

Fix the logic by converting the variables into long and treating -1
value as 'not populated' instead of 0 which is a viable and correct
spare capacity value. We need to be careful signed comparison is used
when comparing with cpu_cap in one of the conditions.

Fixes: 1d42509e475c ("sched/fair: Make EAS wakeup placement consider uclamp restrictions")
Signed-off-by: Qais Yousef (Google) <qyousef@layalina.io>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20230916232955.2099394-2-qyousef@layalina.io

sched/fair: Fix warning in bandwidth distribution

We've observed the following warning being hit in
distribute_cfs_runtime():

SCHED_WARN_ON(cfs_rq->runtime_remaining > 0)

We have the following race:

- CPU 0: running bandwidth distribution (distribute_cfs_runtime).
Inspects the local cfs_rq and makes its runtime_remaining positive.
However, we defer unthrottling the local cfs_rq until after
considering all remote cfs_rq's.

- CPU 1: starts running bandwidth distribution from the slack timer. When
it finds the cfs_rq for CPU 0 on the throttled list, it observers the
that the cfs_rq is throttled, yet is not on the CSD list, and has a
positive runtime_remaining, thus triggering the warning in
distribute_cfs_runtime.

To fix this, we can rework the local unthrottling logic to put the local
cfs_rq on a local list, so that any future bandwidth distributions will
realize that the cfs_rq is about to be unthrottled.

Signed-off-by: Josh Don <joshdon@google.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20230922230535.296350-2-joshdon@google.com

sched/fair: Make cfs_rq->throttled_csd_list available on !SMP

This makes the following patch cleaner by avoiding extra CONFIG_SMP
conditionals on the availability of rq->throttled_csd_list.

Signed-off-by: Josh Don <joshdon@google.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20230922230535.296350-1-joshdon@google.com

sched/debug: Remove the /proc/sys/kernel/sched_child_runs_first sysctl

The /proc/sys/kernel/sched_child_runs_first knob is no longer connected since:

5e963f2bd4654 ("sched/fair: Commit to EEVDF")

Remove it.

Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20230920130025.412071-2-bigeasy@linutronix.de

sched/fair: Rename check_preempt_curr() to wakeup_preempt()

The name is a bit opaque - make it clear that this is about wakeup
preemption.

Also rename the ->check_preempt_curr() methods similarly.

Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>

sched/fair: Rename check_preempt_wakeup() to check_preempt_wakeup_fair()

Other scheduling classes already postfix their similar methods
with the class name.

Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>

sched/fair: Ratelimit update to tg->load_avg

When using sysbench to benchmark Postgres in a single docker instance
with sysbench's nr_threads set to nr_cpu, it is observed there are times
update_cfs_group() and update_load_avg() shows noticeable overhead on
a 2sockets/112core/224cpu Intel Sapphire Rapids(SPR):

13.75% 13.74% [kernel.vmlinux] [k] update_cfs_group
10.63% 10.04% [kernel.vmlinux] [k] update_load_avg

Annotate shows the cycles are mostly spent on accessing tg->load_avg
with update_load_avg() being the write side and update_cfs_group() being
the read side. tg->load_avg is per task group and when different tasks
of the same taskgroup running on different CPUs frequently access
tg->load_avg, it can be heavily contended.

E.g. when running postgres_sysbench on a 2sockets/112cores/224cpus Intel
Sappire Rapids, during a 5s window, the wakeup number is 14millions and
migration number is 11millions and with each migration, the task's load
will transfer from src cfs_rq to target cfs_rq and each change involves
an update to tg->load_avg. Since the workload can trigger as many wakeups
and migrations, the access(both read and write) to tg->load_avg can be
unbound. As a result, the two mentioned functions showed noticeable
overhead. With netperf/nr_client=nr_cpu/UDP_RR, the problem is worse:
during a 5s window, wakeup number is 21millions and migration number is
14millions; update_cfs_group() costs ~25% and update_load_avg() costs ~16%.

Reduce the overhead by limiting updates to tg->load_avg to at most once
per ms. The update frequency is a tradeoff between tracking accuracy and
overhead. 1ms is chosen because PELT window is roughly 1ms and it
delivered good results for the tests that I've done. After this change,
the cost of accessing tg->load_avg is greatly reduced and performance
improved. Detailed test results below.

==============================
postgres_sysbench on SPR:
25%
base: 42382±19.8%
patch: 50174±9.5% (noise)

50%
base: 67626±1.3%
patch: 67365±3.1% (noise)

75%
base: 100216±1.2%
patch: 112470±0.1% +12.2%

100%
base: 93671±0.4%
patch: 113563±0.2% +21.2%

==============================
hackbench on ICL:
group=1
base: 114912±5.2%
patch: 117857±2.5% (noise)

group=4
base: 359902±1.6%
patch: 361685±2.7% (noise)

group=8
base: 461070±0.8%
patch: 491713±0.3% +6.6%

group=16
base: 309032±5.0%
patch: 378337±1.3% +22.4%

=============================
hackbench on SPR:
group=1
base: 100768±2.9%
patch: 103134±2.9% (noise)

group=4
base: 413830±12.5%
patch: 378660±16.6% (noise)

group=8
base: 436124±0.6%
patch: 490787±3.2% +12.5%

group=16
base: 457730±3.2%
patch: 680452±1.3% +48.8%

============================
netperf/udp_rr on ICL
25%
base: 114413±0.1%
patch: 115111±0.0% +0.6%

50%
base: 86803±0.5%
patch: 86611±0.0% (noise)

75%
base: 35959±5.3%
patch: 49801±0.6% +38.5%

100%
base: 61951±6.4%
patch: 70224±0.8% +13.4%

===========================
netperf/udp_rr on SPR
25%
base: 104954±1.3%
patch: 107312±2.8% (noise)

50%
base: 55394±4.6%
patch: 54940±7.4% (noise)

75%
base: 13779±3.1%
patch: 36105±1.1% +162%

100%
base: 9703±3.7%
patch: 28011±0.2% +189%

==============================================
netperf/tcp_stream on ICL (all in noise range)
25%
base: 43092±0.1%
patch: 42891±0.5%

50%
base: 19278±14.9%
patch: 22369±7.2%

75%
base: 16822±3.0%
patch: 17086±2.3%

100%
base: 18216±0.6%
patch: 18078±2.9%

===============================================
netperf/tcp_stream on SPR (all in noise range)
25%
base: 34491±0.3%
patch: 34886±0.5%

50%
base: 19278±14.9%
patch: 22369±7.2%

75%
base: 16822±3.0%
patch: 17086±2.3%

100%
base: 18216±0.6%
patch: 18078±2.9%

Reported-by: Nitin Tekchandani <nitin.tekchandani@intel.com>
Suggested-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Aaron Lu <aaron.lu@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Reviewed-by: David Vernet <void@manifault.com>
Tested-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Tested-by: Swapnil Sapkal <Swapnil.Sapkal@amd.com>
Link: https://lkml.kernel.org/r/20230912065808.2530-2-aaron.lu@intel.com

sched/fair: Fix cfs_rq_is_decayed() on !SMP

We don't need to maintain per-queue leaf_cfs_rq_list on !SMP, since
it's used for cfs_rq load tracking & balancing on SMP.

But sched debug interface uses it to print per-cfs_rq stats.

This patch fixes the !SMP version of cfs_rq_is_decayed(), so the
per-queue leaf_cfs_rq_list is also maintained correctly on !SMP,
to fix the warning in assert_list_leaf_cfs_rq().

Fixes: 0a00a354644e ("sched/fair: Delete useless condition in tg_unthrottle_up()")
Reported-by: Leo Yu-Chi Liang <ycliang@andestech.com>
Signed-off-by: Chengming Zhou <zhouchengming@bytedance.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Tested-by: Leo Yu-Chi Liang <ycliang@andestech.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Closes: https://lore.kernel.org/all/ZN87UsqkWcFLDxea@swlinux02/
Link: https://lore.kernel.org/r/20230913132031.2242151-1-chengming.zhou@linux.dev

sched/fair: Fix open-coded numa_nearest_node()

task_numa_placement() searches for a nearest node to migrate by calling
for_each_node_state(). Now that we have numa_nearest_node(), switch to
using it.

Signed-off-by: Yury Norov <yury.norov@gmail.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Mel Gorman <mgorman@suse.de>
Link: https://lore.kernel.org/r/20230819141239.287290-3-yury.norov@gmail.com

sched/eevdf: Fix heap corruption more

Because someone is a flaming idiot... and forgot we have current as
se->on_rq but not actually in the tree itself, and walking rb_parent()
on an entry not in the tree is 'funky' and KASAN complains.

Fixes: 8dafa9d0eb1a ("sched/eevdf: Fix min_deadline heap integrity")
Reported-by: 0599jiangyc@gmail.com
Reported-by: Dmitry Safonov <0x7f454c46@gmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Dmitry Safonov <0x7f454c46@gmail.com>
Link: https://bugzilla.kernel.org/show_bug.cgi?id=218020
Link: https://lkml.kernel.org/r/CAJwJo6ZGXO07%3DQvW4fgQfbsDzQPs9xj5sAQ1zp%3DmAyPMNbHYww%40mail.gmail.com

sched/eevdf: Fix pick_eevdf()

The old pick_eevdf() could fail to find the actual earliest eligible
deadline when it descended to the right looking for min_deadline, but
it turned out that that min_deadline wasn't actually eligible. In that
case we need to go back and search through any left branches we
skipped looking for the actual best _eligible_ min_deadline.

This is more expensive, but still O(log n), and at worst should only
involve descending two branches of the rbtree.

I've run this through a userspace stress test (thank you
tools/lib/rbtree.c), so hopefully this implementation doesn't miss any
corner cases.

Fixes: 147f3efaa241 ("sched/fair: Implement an EEVDF-like scheduling policy")
Signed-off-by: Ben Segall <bsegall@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/xm261qego72d.fsf_-_@google.com

sched/eevdf: Fix min_deadline heap integrity

Marek and Biju reported instances of:

"EEVDF scheduling fail, picking leftmost"

which Mike correlated with cgroup scheduling and the min_deadline heap
getting corrupted; some trace output confirms:

> And yeah, min_deadline is hosed somehow:
>
> validate_cfs_rq: --- /
> __print_se: ffff88845cf48080 w: 1024 ve: -58857638 lag: 870381 vd: -55861854 vmd: -66302085 E (11372/tr)
> __print_se: ffff88810d165800 w: 25 ve: -80323686 lag: 22336429 vd: -41496434 vmd: -66302085 E (-1//autogroup-31)
> __print_se: ffff888108379000 w: 25 ve: 0 lag: -57987257 vd: 114632828 vmd: 114632828 N (-1//autogroup-33)
> validate_cfs_rq: min_deadline: -55861854 avg_vruntime: -62278313462 / 1074 = -57987256

Turns out that reweight_entity(), which tries really hard to be fast,
does not do the normal dequeue+update+enqueue pattern but *does* scale
the deadline.

However, it then fails to propagate the updated deadline value up the
heap.

Fixes: 147f3efaa241 ("sched/fair: Implement an EEVDF-like scheduling policy")
Reported-by: Marek Szyprowski <m.szyprowski@samsung.com>
Reported-by: Biju Das <biju.das.jz@bp.renesas.com>
Reported-by: Mike Galbraith <efault@gmx.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Marek Szyprowski <m.szyprowski@samsung.com>
Tested-by: Biju Das <biju.das.jz@bp.renesas.com>
Tested-by: Mike Galbraith <efault@gmx.de>
Link: https://lkml.kernel.org/r/20231006192445.GE743@noisy.programming.kicks-ass.net

sched/eevdf: Fix avg_vruntime()

The expectation is that placing a task at avg_vruntime() makes it
eligible. Turns out there is a corner case where this is not the case.

Specifically, avg_vruntime() relies on the fact that integer division
is a flooring function (eg. it discards the remainder). By this
property the value returned is slightly left of the true average.

However! when the average is a negative (relative to min_vruntime) the
effect is flipped and it becomes a ceil, with the result that the
returned value is just right of the average and thus not eligible.

Fixes: af4cf40470c2 ("sched/fair: Add cfs_rq::avg_vruntime")
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>

sched/eevdf: Also update slice on placement

Tasks that never consume their full slice would not update their slice value.
This means that tasks that are spawned before the sysctl scaling keep their
original (UP) slice length.

Fixes: 147f3efaa241 ("sched/fair: Implement an EEVDF-like scheduling policy")
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20230915124822.847197830@noisy.programming.kicks-ass.net

sched/fair: Fix SMT4 group_smt_balance handling

For SMT4, any group with more than 2 tasks will be marked as
group_smt_balance. Retain the behaviour of group_has_spare by marking
the busiest group as the group which has the least number of idle_cpus.

Also, handle rounding effect of adding (ncores_local + ncores_busy) when
the local is fully idle and busy group imbalance is less than 2 tasks.
Local group should try to pull at least 1 task in this case so imbalance
should be set to 2 instead.

Fixes: fee1759e4f04 ("sched/fair: Determine active load balance for SMT sched groups")
Acked-by: Shrikanth Hegde <sshegde@linux.vnet.ibm.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: http://lkml.kernel.org/r/6cd1633036bb6b651af575c32c2a9608a106702c.camel@linux.intel.com

sched/fair: Optimize should_we_balance() for large SMT systems

should_we_balance() is called in load_balance() to find out if the CPU that
is trying to do the load balance is the right one or not.

With commit:

b1bfeab9b002("sched/fair: Consider the idle state of the whole core for load balance")

the code tries to find an idle core to do the load balancing
and falls back on an idle sibling CPU if there is no idle core.

However, on larger SMT systems, it could be needlessly iterating to find a
idle by scanning all the CPUs in an non-idle core. If the core is not idle,
and first SMT sibling which is idle has been found, then its not needed to
check other SMT siblings for idleness

Lets say in SMT4, Core0 has 0,2,4,6 and CPU0 is BUSY and rest are IDLE.
balancing domain is MC/DIE. CPU2 will be set as the first idle_smt and
same process would be repeated for CPU4 and CPU6 but this is unnecessary.
Since calling is_core_idle loops through all CPU's in the SMT mask, effect
is multiplied by weight of smt_mask. For example,when say 1 CPU is busy,
we would skip loop for 2 CPU's and skip iterating over 8CPU's. That
effect would be more in DIE/NUMA domain where there are more cores.

Testing and performance evaluation
==================================

The test has been done on this system which has 12 cores, i.e 24 small
cores with SMT=4:

lscpu
Architecture: ppc64le
Byte Order: Little Endian
CPU(s): 96
On-line CPU(s) list: 0-95
Model name: POWER10 (architected), altivec supported
Thread(s) per core: 8

Used funclatency bcc tool to evaluate the time taken by should_we_balance(). For
base tip/sched/core the time taken is collected by making the
should_we_balance() noinline. time is in nanoseconds. The values are
collected by running the funclatency tracer for 60 seconds. values are
average of 3 such runs. This represents the expected reduced time with
patch.

tip/sched/core was at commit:

2f88c8e802c8 ("sched/eevdf/doc: Modify the documented knob to base_slice_ns as well")

Results:

------------------------------------------------------------------------------
workload tip/sched/core with_patch(%gain)
------------------------------------------------------------------------------
idle system 809.3 695.0(16.45)
stress ng – 12 threads -l 100 1013.5 893.1(13.49)
stress ng – 24 threads -l 100 1073.5 980.0(9.54)
stress ng – 48 threads -l 100 683.0 641.0(6.55)
stress ng – 96 threads -l 100 2421.0 2300(5.26)
stress ng – 96 threads -l 15 375.5 377.5(-0.53)
stress ng – 96 threads -l 25 635.5 637.5(-0.31)
stress ng – 96 threads -l 35 934.0 891.0(4.83)

Ran schbench(old), hackbench and stress_ng to evaluate the workload
performance between tip/sched/core and with patch.
No modification to tip/sched/core

TL;DR:

Good improvement is seen with schbench. when hackbench and stress_ng
runs for longer good improvement is seen.

------------------------------------------------------------------------------
schbench(old) tip +patch(%gain)
10 iterations sched/core
------------------------------------------------------------------------------
1 Threads
50.0th: 8.00 9.00(-12.50)
75.0th: 9.60 9.00(6.25)
90.0th: 11.80 10.20(13.56)
95.0th: 12.60 10.40(17.46)
99.0th: 13.60 11.90(12.50)
99.5th: 14.10 12.60(10.64)
99.9th: 15.90 14.60(8.18)
2 Threads
50.0th: 9.90 9.20(7.07)
75.0th: 12.60 10.10(19.84)
90.0th: 15.50 12.00(22.58)
95.0th: 17.70 14.00(20.90)
99.0th: 21.20 16.90(20.28)
99.5th: 22.60 17.50(22.57)
99.9th: 30.40 19.40(36.18)
4 Threads
50.0th: 12.50 10.60(15.20)
75.0th: 15.30 12.00(21.57)
90.0th: 18.60 14.10(24.19)
95.0th: 21.30 16.20(23.94)
99.0th: 26.00 20.70(20.38)
99.5th: 27.60 22.50(18.48)
99.9th: 33.90 31.40(7.37)
8 Threads
50.0th: 16.30 14.30(12.27)
75.0th: 20.20 17.40(13.86)
90.0th: 24.50 21.90(10.61)
95.0th: 27.30 24.70(9.52)
99.0th: 35.00 31.20(10.86)
99.5th: 46.40 33.30(28.23)
99.9th: 89.30 57.50(35.61)
16 Threads
50.0th: 22.70 20.70(8.81)
75.0th: 30.10 27.40(8.97)
90.0th: 36.00 32.80(8.89)
95.0th: 39.60 36.40(8.08)
99.0th: 49.20 44.10(10.37)
99.5th: 64.90 50.50(22.19)
99.9th: 143.50 100.60(29.90)
32 Threads
50.0th: 34.60 35.50(-2.60)
75.0th: 48.20 50.50(-4.77)
90.0th: 59.20 62.40(-5.41)
95.0th: 65.20 69.00(-5.83)
99.0th: 80.40 83.80(-4.23)
99.5th: 102.10 98.90(3.13)
99.9th: 727.10 506.80(30.30)

schbench does improve in general. There is some run to run variation with
schbench. Did a validation run to confirm that trend is similar.

------------------------------------------------------------------------------
hackbench tip +patch(%gain)
20 iterations, 50000 loops sched/core
------------------------------------------------------------------------------
Process 10 groups : 11.74 11.70(0.34)
Process 20 groups : 22.73 22.69(0.18)
Process 30 groups : 33.39 33.40(-0.03)
Process 40 groups : 43.73 43.61(0.27)
Process 50 groups : 53.82 54.35(-0.98)
Process 60 groups : 64.16 65.29(-1.76)
thread 10 Time : 12.81 12.79(0.16)
thread 20 Time : 24.63 24.47(0.65)
Process(Pipe) 10 Time : 6.40 6.34(0.94)
Process(Pipe) 20 Time : 10.62 10.63(-0.09)
Process(Pipe) 30 Time : 15.09 14.84(1.66)
Process(Pipe) 40 Time : 19.42 19.01(2.11)
Process(Pipe) 50 Time : 24.04 23.34(2.91)
Process(Pipe) 60 Time : 28.94 27.51(4.94)
thread(Pipe) 10 Time : 6.96 6.87(1.29)
thread(Pipe) 20 Time : 11.74 11.73(0.09)

hackbench shows slight improvement with pipe. Slight degradation in process.

------------------------------------------------------------------------------
stress_ng tip +patch(%gain)
10 iterations 100000 cpu_ops sched/core
------------------------------------------------------------------------------

--cpu=96 -util=100 Time taken : 5.30, 5.01(5.47)
--cpu=48 -util=100 Time taken : 7.94, 6.73(15.24)
--cpu=24 -util=100 Time taken : 11.67, 8.75(25.02)
--cpu=12 -util=100 Time taken : 15.71, 15.02(4.39)
--cpu=96 -util=10 Time taken : 22.71, 22.19(2.29)
--cpu=96 -util=20 Time taken : 12.14, 12.37(-1.89)
--cpu=96 -util=30 Time taken : 8.76, 8.86(-1.14)
--cpu=96 -util=40 Time taken : 7.13, 7.14(-0.14)
--cpu=96 -util=50 Time taken : 6.10, 6.13(-0.49)
--cpu=96 -util=60 Time taken : 5.42, 5.41(0.18)
--cpu=96 -util=70 Time taken : 4.94, 4.94(0.00)
--cpu=96 -util=80 Time taken : 4.56, 4.53(0.66)
--cpu=96 -util=90 Time taken : 4.27, 4.26(0.23)

Good improvement seen with 24 CPUs. In this case only one CPU is busy,
and no core is idle. Decent improvement with 100% utilization case. no
difference in other utilization.

Fixes: b1bfeab9b002 ("sched/fair: Consider the idle state of the whole core for load balance")
Signed-off-by: Shrikanth Hegde <sshegde@linux.vnet.ibm.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20230902081204.232218-1-sshegde@linux.vnet.ibm.com

sched/fair: Make update_entity_lag() static

The function update_entity_lag() is only used inside the kernel/sched/fair.c file.
Make it static.

Signed-off-by: Hao Jia <jiahao.os@bytedance.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20230829030325.69128-1-jiahao.os@bytedance.com

sched/eevdf: Curb wakeup-preemption

Mike and others noticed that EEVDF does like to over-schedule quite a
bit -- which does hurt performance of a number of benchmarks /
workloads.

In particular, what seems to cause over-scheduling is that when lag is
of the same order (or larger) than the request / slice then placement
will not only cause the task to be placed left of current, but also
with a smaller deadline than current, which causes immediate
preemption.

[ notably, lag bounds are relative to HZ ]

Mike suggested we stick to picking 'current' for as long as it's
eligible to run, giving it uninterrupted runtime until it reaches
parity with the pack.

Augment Mike's suggestion by only allowing it to exhaust it's initial
request.

One random data point:

echo NO_RUN_TO_PARITY > /debug/sched/features
perf stat -a -e context-switches --repeat 10 -- perf bench sched messaging -g 20 -t -l 5000

3,723,554 context-switches ( +- 0.56% )
9.5136 +- 0.0394 seconds time elapsed ( +- 0.41% )

echo RUN_TO_PARITY > /debug/sched/features
perf stat -a -e context-switches --repeat 10 -- perf bench sched messaging -g 20 -t -l 5000

2,556,535 context-switches ( +- 0.51% )
9.2427 +- 0.0302 seconds time elapsed ( +- 0.33% )

Suggested-by: Mike Galbraith <umgwanakikbuti@gmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20230816134059.GC982867@hirez.programming.kicks-ass.net

sched/fair: Block nohz tick_stop when cfs bandwidth in use

CFS bandwidth limits and NOHZ full don't play well together. Tasks
can easily run well past their quotas before a remote tick does
accounting. This leads to long, multi-period stalls before such
tasks can run again. Currently, when presented with these conflicting
requirements the scheduler is favoring nohz_full and letting the tick
be stopped. However, nohz tick stopping is already best-effort, there
are a number of conditions that can prevent it, whereas cfs runtime
bandwidth is expected to be enforced.

Make the scheduler favor bandwidth over stopping the tick by setting
TICK_DEP_BIT_SCHED when the only running task is a cfs task with
runtime limit enabled. We use cfs_b->hierarchical_quota to
determine if the task requires the tick.

Add check in pick_next_task_fair() as well since that is where
we have a handle on the task that is actually going to be running.

Add check in sched_can_stop_tick() to cover some edge cases such
as nr_running going from 2->1 and the 1 remains the running task.

Reviewed-By: Ben Segall <bsegall@google.com>
Signed-off-by: Phil Auld <pauld@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20230712133357.381137-3-pauld@redhat.com

sched, cgroup: Restore meaning to hierarchical_quota

In cgroupv2 cfs_b->hierarchical_quota is set to -1 for all task
groups due to the previous fix simply taking the min. It should
reflect a limit imposed at that level or by an ancestor. Even
though cgroupv2 does not require child quota to be less than or
equal to that of its ancestors the task group will still be
constrained by such a quota so this should be shown here. Cgroupv1
continues to set this correctly.

In both cases, add initialization when a new task group is created
based on the current parent's value (or RUNTIME_INF in the case of
root_task_group). Otherwise, the field is wrong until a quota is
changed after creation and __cfs_schedulable() is called.

Fixes: c53593e5cb69 ("sched, cgroup: Don't reject lower cpu.max on ancestors")
Signed-off-by: Phil Auld <pauld@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Ben Segall <bsegall@google.com>
Acked-by: Tejun Heo <tj@kernel.org>
Link: https://lore.kernel.org/r/20230714125746.812891-1-pauld@redhat.com

Revert "sched/fair: Move unused stub functions to header"

Revert commit 7aa55f2a5902 ("sched/fair: Move unused stub functions to
header"), for while it has the right Changelog, the actual patch
content a revert of the previous 4 patches:

f7df852ad6db ("sched: Make task_vruntime_update() prototype visible")
c0bdfd72fbfb ("sched/fair: Hide unused init_cfs_bandwidth() stub")
378be384e01f ("sched: Add schedule_user() declaration")
d55ebae3f312 ("sched: Hide unused sched_update_scaling()")

So in effect this is a revert of a revert and re-applies those
patches.

Fixes: 7aa55f2a5902 ("sched/fair: Move unused stub functions to header")
Reported-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>

sched/fair: remove util_est boosting

There is no need to use runnable_avg when estimating util_est and that
even generates wrong behavior because one includes blocked tasks whereas
the other one doesn't. This can lead to accounting twice the waking task p,
once with the blocked runnable_avg and another one when adding its
util_est.

cpu's runnable_avg is already used when computing util_avg which is then
compared with util_est.

In some situation, feec will not select prev_cpu but another one on the
same performance domain because of higher max_util

Fixes: 7d0583cf9ec7 ("sched/fair, cpufreq: Introduce 'runnable boosting'")
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Tested-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Link: https://lore.kernel.org/r/20230706135144.324311-1-vincent.guittot@linaro.org

sched/fair: Propagate enqueue flags into place_entity()

This allows place_entity() to consider ENQUEUE_WAKEUP and
ENQUEUE_MIGRATED.

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20230531124604.274010996@infradead.org

sched/debug: Rename sysctl_sched_min_granularity to sysctl_sched_base_slice

EEVDF uses this tunable as the base request/slice -- make sure the
name reflects this.

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20230531124604.205287511@infradead.org

sched/fair: Commit to EEVDF

EEVDF is a better defined scheduling policy, as a result it has less
heuristics/tunables. There is no compelling reason to keep CFS around.

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20230531124604.137187212@infradead.org

sched/smp: Use lag to simplify cross-runqueue placement

Using lag is both more correct and simpler when moving between
runqueues.

Notable, min_vruntime() was invented as a cheap approximation of
avg_vruntime() for this very purpose (SMP migration). Since we now
have the real thing; use it.

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20230531124604.068911180@infradead.org

sched/fair: Commit to lag based placement

Removes the FAIR_SLEEPERS code in favour of the new LAG based
placement.

Specifically, the whole FAIR_SLEEPER thing was a very crude
approximation to make up for the lack of lag based placement,
specifically the 'service owed' part. This is important for things
like 'starve' and 'hackbench'.

One side effect of FAIR_SLEEPER is that it caused 'small' unfairness,
specifically, by always ignoring up-to 'thresh' sleeptime it would
have a 50%/50% time distribution for a 50% sleeper vs a 100% runner,
while strictly speaking this should (of course) result in a 33%/67%
split (as CFS will also do if the sleep period exceeds 'thresh').

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20230531124604.000198861@infradead.org

sched/fair: Implement an EEVDF-like scheduling policy

Where CFS is currently a WFQ based scheduler with only a single knob,
the weight. The addition of a second, latency oriented parameter,
makes something like WF2Q or EEVDF based a much better fit.

Specifically, EEVDF does EDF like scheduling in the left half of the
tree -- those entities that are owed service. Except because this is a
virtual time scheduler, the deadlines are in virtual time as well,
which is what allows over-subscription.

EEVDF has two parameters:

- weight, or time-slope: which is mapped to nice just as before

- request size, or slice length: which is used to compute
the virtual deadline as: vd_i = ve_i + r_i/w_i

Basically, by setting a smaller slice, the deadline will be earlier
and the task will be more eligible and ran earlier.

Tick driven preemption is driven by request/slice completion; while
wakeup preemption is driven by the deadline.

Because the tree is now effectively an interval tree, and the
selection is no longer 'leftmost', over-scheduling is less of a
problem.

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20230531124603.931005524@infradead.org

sched/fair: Add lag based placement

With the introduction of avg_vruntime, it is possible to approximate
lag (the entire purpose of introducing it in fact). Use this to do lag
based placement over sleep+wake.

Specifically, the FAIR_SLEEPERS thing places things too far to the
left and messes up the deadline aspect of EEVDF.

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20230531124603.794929315@infradead.org

sched/fair: Remove sched_feat(START_DEBIT)

With the introduction of avg_vruntime() there is no need to use worse
approximations. Take the 0-lag point as starting point for inserting
new tasks.

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20230531124603.722361178@infradead.org

sched/fair: Add cfs_rq::avg_vruntime

In order to move to an eligibility based scheduling policy, we need
to have a better approximation of the ideal scheduler.

Specifically, for a virtual time weighted fair queueing based
scheduler the ideal scheduler will be the weighted average of the
individual virtual runtimes (math in the comment).

As such, compute the weighted average to approximate the ideal
scheduler -- note that the approximation is in the individual task
behaviour, which isn't strictly conformant.

Specifically consider adding a task with a vruntime left of center, in
this case the average will move backwards in time -- something the
ideal scheduler would of course never do.

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20230531124603.654144274@infradead.org

sched: add WF_CURRENT_CPU and externise ttwu

Add WF_CURRENT_CPU wake flag that advices the scheduler to
move the wakee to the current CPU. This is useful for fast on-CPU
context switching use cases.

In addition, make ttwu external rather than static so that
the flag could be passed to it from outside of sched/core.c.

Signed-off-by: Peter Oskolkov <posk@google.com>
Signed-off-by: Andrei Vagin <avagin@google.com>
Acked-by: "Peter Zijlstra (Intel)" <peterz@infradead.org>
Link: https://lore.kernel.org/r/20230308073201.3102738-3-avagin@google.com
Signed-off-by: Kees Cook <keescook@chromium.org>

sched/fair: Stabilize asym cpu capacity system idle cpu selection

select_idle_capacity() not only looks for an idle cpu that fits for the
waking task but also for cpu with highest bandwidth when no cpu fits.
Start the loop with target cpu so it will be selected 1st when no cpu fits
but several cpus shared the same bandwidth. Starting with target cpu
prevents the task to migrate between cpus with same bandwidth at every
wakeup when no cpu fits.

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20230711081359.868862-1-vincent.guittot@linaro.org

sched/fair: Consider the idle state of the whole core for load balance

should_we_balance() traverses the group_balance_mask (AND'ed with lb_env::
cpus) starting from lower numbered CPUs looking for the first idle CPU.

In hybrid x86 systems, the siblings of SMT cores get CPU numbers, before
non-SMT cores:

[0, 1] [2, 3] [4, 5] 6 7 8 9
b i b i b i b i i i

In the figure above, CPUs in brackets are siblings of an SMT core. The
rest are non-SMT cores. 'b' indicates a busy CPU, 'i' indicates an
idle CPU.

We should let a CPU on a fully idle core get the first chance to idle
load balance as it has more CPU capacity than a CPU on an idle SMT
CPU with busy sibling. So for the figure above, if we are running
should_we_balance() to CPU 1, we should return false to let CPU 7 on
idle core to have a chance first to idle load balance.

A partially busy (i.e., of type group_has_spare) local group with SMT 
cores will often have only one SMT sibling busy. If the destination CPU
is a non-SMT core, partially busy, lower-numbered, SMT cores should not
be considered when finding the first idle CPU. 

However, in should_we_balance(), when we encounter idle SMT first in partially
busy core, we prematurely break the search for the first idle CPU.

Higher-numbered, non-SMT cores is not given the chance to have
idle balance done on their behalf. Those CPUs will only be considered
for idle balancing by chance via CPU_NEWLY_IDLE.

Instead, consider the idle state of the whole SMT core.

Signed-off-by: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
Co-developed-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/807bdd05331378ea3bf5956bda87ded1036ba769.1688770494.git.tim.c.chen@linux.intel.com

sched/fair: Implement prefer sibling imbalance calculation between asymmetric groups

In the current prefer sibling load balancing code, there is an implicit
assumption that the busiest sched group and local sched group are
equivalent, hence the tasks to be moved is simply the difference in
number of tasks between the two groups (i.e. imbalance) divided by two.

However, we may have different number of cores between the cluster groups,
say when we take CPU offline or we have hybrid groups. In that case,
we should balance between the two groups such that #tasks/#cores ratio
is the same between the same between both groups. Hence the imbalance
computed will need to reflect this.

Adjust the sibling imbalance computation to take into account of the
above considerations.

Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/4eacbaa236e680687dae2958378a6173654113df.1688770494.git.tim.c.chen@linux.intel.com

sched/fair: Determine active load balance for SMT sched groups

On hybrid CPUs with scheduling cluster enabled, we will need to
consider balancing between SMT CPU cluster, and Atom core cluster.

Below shows such a hybrid x86 CPU with 4 big cores and 8 atom cores.
Each scheduling cluster span a L2 cache.

--L2-- --L2-- --L2-- --L2-- ----L2---- -----L2------
[0, 1] [2, 3] [4, 5] [5, 6] [7 8 9 10] [11 12 13 14]
Big Big Big Big Atom Atom
core core core core Module Module

If the busiest group is a big core with both SMT CPUs busy, we should
active load balance if destination group has idle CPU cores. Such
condition is considered by asym_active_balance() in load balancing but not
considered when looking for busiest group and computing load imbalance.
Add this consideration in find_busiest_group() and calculate_imbalance().

In addition, update the logic determining the busier group when one group
is SMT and the other group is non SMT but both groups are partially busy
with idle CPU. The busier group should be the group with idle cores rather
than the group with one busy SMT CPU. We do not want to make the SMT group
the busiest one to pull the only task off SMT CPU and causing the whole core to
go empty.

Otherwise suppose in the search for the busiest group, we first encounter
an SMT group with 1 task and set it as the busiest. The destination
group is an atom cluster with 1 task and we next encounter an atom
cluster group with 3 tasks, we will not pick this atom cluster over the
SMT group, even though we should. As a result, we do not load balance
the busier Atom cluster (with 3 tasks) towards the local atom cluster
(with 1 task). And it doesn't make sense to pick the 1 task SMT group
as the busier group as we also should not pull task off the SMT towards
the 1 task atom cluster and make the SMT core completely empty.

Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/e24f35d142308790f69be65930b82794ef6658a2.1688770494.git.tim.c.chen@linux.intel.com

sched: add throttled time stat for throttled children

We currently export the total throttled time for cgroups that are given
a bandwidth limit. This patch extends this accounting to also account
the total time that each children cgroup has been throttled.

This is useful to understand the degree to which children have been
affected by the throttling control. Children which are not runnable
during the entire throttled period, for example, will not show any
self-throttling time during this period.

Expose this in a new interface, 'cpu.stat.local', which is similar to
how non-hierarchical events are accounted in 'memory.events.local'.

Signed-off-by: Josh Don <joshdon@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Tejun Heo <tj@kernel.org>
Link: https://lore.kernel.org/r/20230620183247.737942-2-joshdon@google.com

sched: don't account throttle time for empty groups

It is easy for a cfs_rq to become throttled even when it has no enqueued
entities (for example, if we have just put_prev()'d the last runnable
task of the cfs_rq, and the cfs_rq is out of quota).

Avoid accounting this time towards total throttle time, since it
otherwise falsely inflates the stats.

Note that the dequeue path is special, since we normally disallow
migrations when a task is in a throttled hierarchy (see
throttled_lb_pair()).

Signed-off-by: Josh Don <joshdon@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20230620183247.737942-1-joshdon@google.com

sched/fair: Use recent_used_cpu to test p->cpus_ptr

When checking whether a recently used CPU can be a potential idle
candidate, recent_used_cpu should be used to test p->cpus_ptr as
p->recent_used_cpu is not equal to recent_used_cpu and candidate
decision is made based on recent_used_cpu here.

Fixes: 89aafd67f28c ("sched/fair: Use prev instead of new target as recent_used_cpu")
Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Phil Auld <pauld@redhat.com>
Acked-by: Mel Gorman <mgorman@suse.de>
Link: https://lore.kernel.org/r/20230620080747.359122-1-linmiaohe@huawei.com

sched/core: Avoid multiple calling update_rq_clock() in __cfsb_csd_unthrottle()

After commit 8ad075c2eb1f ("sched: Async unthrottling for cfs
bandwidth"), we may update the rq clock multiple times in the loop of
__cfsb_csd_unthrottle().

A prior (although less common) instance of this problem exists in
unthrottle_offline_cfs_rqs().

Cure both by ensuring update_rq_clock() is called before the loop and
setting RQCF_ACT_SKIP during the loop, to supress further updates.
The alternative would be pulling update_rq_clock() out of
unthrottle_cfs_rq(), but that gives an even bigger mess.

Fixes: 8ad075c2eb1f ("sched: Async unthrottling for cfs bandwidth")
Reviewed-By: Ben Segall <bsegall@google.com>
Suggested-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Hao Jia <jiahao.os@bytedance.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lkml.kernel.org/r/20230613082012.49615-4-jiahao.os@bytedance.com

sched/fair: Rename variable cpu_util eff_util

cppcheck reports
kernel/sched/fair.c:7436:17: style: Local variable 'cpu_util' shadows outer function [shadowFunction]
unsigned long cpu_util;
^

Clean this up by renaming the variable to eff_util

Signed-off-by: Tom Rix <trix@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Valentin Schneider <vschneid@redhat.com>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Link: https://lore.kernel.org/r/20230611122535.183654-1-trix@redhat.com

sched/fair, cpufreq: Introduce 'runnable boosting'

The responsiveness of the Per Entity Load Tracking (PELT) util_avg in
mobile devices is still considered too low for utilization changes
during task ramp-up.

In Android this manifests in the fact that the first frames of a UI
activity are very prone to be jankframes (a frame which doesn't meet
the required frame rendering time, e.g. 16ms@60Hz) since the CPU
frequency is normally low at this point and has to ramp up quickly.

The beginning of an UI activity is also characterized by the occurrence
of CPU contention, especially on little CPUs. Current little CPUs can
have an original CPU capacity of only ~ 150 which means that the actual
CPU capacity at lower frequency can even be much smaller.

Schedutil maps CPU util_avg into CPU frequency request via:

util = effective_cpu_util(..., cpu_util_cfs(cpu), ...) ->
util = map_util_perf(util) -> freq = map_util_freq(util, ...)

CPU contention for CFS tasks can be detected by 'CPU runnable > CPU
utililization' in cpu_util_cfs_boost() -> cpu_util(..., boost = 1).
Schedutil uses 'runnable boosting' by calling cpu_util_cfs_boost().

To be in sync with schedutil's CPU frequency selection, Energy Aware
Scheduling (EAS) also calls cpu_util(..., boost = 1) during max util
detection.

Moreover, 'runnable boosting' is also used in load-balance for busiest
CPU selection when the migration type is 'migrate_util', i.e. only at
sched domains which don't have the SD_SHARE_PKG_RESOURCES flag set.

Suggested-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20230515115735.296329-3-dietmar.eggemann@arm.com

sched/fair: Refactor CPU utilization functions

There is a lot of code duplication in cpu_util_next() & cpu_util_cfs().

Remove this by allowing cpu_util_next() to be called with p = NULL.
Rename cpu_util_next() to cpu_util() since the '_next' suffix is no
longer necessary to distinct cpu utilization related functions.
Implement cpu_util_cfs(cpu) as cpu_util(cpu, p = NULL, -1).

This will allow to code future related cpu util changes only in one
place, namely in cpu_util().

Signed-off-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20230515115735.296329-2-dietmar.eggemann@arm.com

sched/fair: Don't balance task to its current running CPU

We've run into the case that the balancer tries to balance a migration
disabled task and trigger the warning in set_task_cpu() like below:

------------[ cut here ]------------
WARNING: CPU: 7 PID: 0 at kernel/sched/core.c:3115 set_task_cpu+0x188/0x240
Modules linked in: hclgevf xt_CHECKSUM ipt_REJECT nf_reject_ipv4 <...snip>
CPU: 7 PID: 0 Comm: swapper/7 Kdump: loaded Tainted: G O 6.1.0-rc4+ #1
Hardware name: Huawei TaiShan 2280 V2/BC82AMDC, BIOS 2280-V2 CS V5.B221.01 12/09/2021
pstate: 604000c9 (nZCv daIF +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
pc : set_task_cpu+0x188/0x240
lr : load_balance+0x5d0/0xc60
sp : ffff80000803bc70
x29: ffff80000803bc70 x28: ffff004089e190e8 x27: ffff004089e19040
x26: ffff007effcabc38 x25: 0000000000000000 x24: 0000000000000001
x23: ffff80000803be84 x22: 000000000000000c x21: ffffb093e79e2a78
x20: 000000000000000c x19: ffff004089e19040 x18: 0000000000000000
x17: 0000000000001fad x16: 0000000000000030 x15: 0000000000000000
x14: 0000000000000003 x13: 0000000000000000 x12: 0000000000000000
x11: 0000000000000001 x10: 0000000000000400 x9 : ffffb093e4cee530
x8 : 00000000fffffffe x7 : 0000000000ce168a x6 : 000000000000013e
x5 : 00000000ffffffe1 x4 : 0000000000000001 x3 : 0000000000000b2a
x2 : 0000000000000b2a x1 : ffffb093e6d6c510 x0 : 0000000000000001
Call trace:
set_task_cpu+0x188/0x240
load_balance+0x5d0/0xc60
rebalance_domains+0x26c/0x380
_nohz_idle_balance.isra.0+0x1e0/0x370
run_rebalance_domains+0x6c/0x80
__do_softirq+0x128/0x3d8
____do_softirq+0x18/0x24
call_on_irq_stack+0x2c/0x38
do_softirq_own_stack+0x24/0x3c
__irq_exit_rcu+0xcc/0xf4
irq_exit_rcu+0x18/0x24
el1_interrupt+0x4c/0xe4
el1h_64_irq_handler+0x18/0x2c
el1h_64_irq+0x74/0x78
arch_cpu_idle+0x18/0x4c
default_idle_call+0x58/0x194
do_idle+0x244/0x2b0
cpu_startup_entry+0x30/0x3c
secondary_start_kernel+0x14c/0x190
__secondary_switched+0xb0/0xb4
---[ end trace 0000000000000000 ]---

Further investigation shows that the warning is superfluous, the migration
disabled task is just going to be migrated to its current running CPU.
This is because that on load balance if the dst_cpu is not allowed by the
task, we'll re-select a new_dst_cpu as a candidate. If no task can be
balanced to dst_cpu we'll try to balance the task to the new_dst_cpu
instead. In this case when the migration disabled task is not on CPU it
only allows to run on its current CPU, load balance will select its
current CPU as new_dst_cpu and later triggers the warning above.

The new_dst_cpu is chosen from the env->dst_grpmask. Currently it
contains CPUs in sched_group_span() and if we have overlapped groups it's
possible to run into this case. This patch makes env->dst_grpmask of
group_balance_mask() which exclude any CPUs from the busiest group and
solve the issue. For balancing in a domain with no overlapped groups
the behaviour keeps same as before.

Suggested-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Yicong Yang <yangyicong@hisilicon.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20230530082507.10444-1-yangyicong@huawei.com

sched/fair: Move unused stub functions to header

These four functions have a normal definition for CONFIG_FAIR_GROUP_SCHED,
and empty one that is only referenced when FAIR_GROUP_SCHED is disabled
but CGROUP_SCHED is still enabled. If both are turned off, the functions
are still defined but the misisng prototype causes a W=1 warning:

kernel/sched/fair.c:12544:6: error: no previous prototype for 'free_fair_sched_group'
kernel/sched/fair.c:12546:5: error: no previous prototype for 'alloc_fair_sched_group'
kernel/sched/fair.c:12553:6: error: no previous prototype for 'online_fair_sched_group'
kernel/sched/fair.c:12555:6: error: no previous prototype for 'unregister_fair_sched_group'

Move the alternatives into the header as static inline functions with
the correct combination of #ifdef checks to avoid the warning without
adding even more complexity.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20230522195021.3456768-6-arnd@kernel.org

sched/fair: Hide unused init_cfs_bandwidth() stub

init_cfs_bandwidth() is only used when CONFIG_FAIR_GROUP_SCHED is
enabled, and without this causes a W=1 warning for the missing prototype:

kernel/sched/fair.c:6131:6: error: no previous prototype for 'init_cfs_bandwidth'

The normal implementation is only defined for CONFIG_CFS_BANDWIDTH,
so the stub exists when CFS_BANDWIDTH is disabled but FAIR_GROUP_SCHED
is enabled.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20230522195021.3456768-4-arnd@kernel.org

sched: Hide unused sched_update_scaling()

This function is only used when CONFIG_SMP is enabled, without that there
is no caller and no prototype:

kernel/sched/fair.c:688:5: error: no previous prototype for 'sched_update_scaling' [-Werror=missing-prototypes

Hide the definition in the same #ifdef check as the declaration.

Fixes: 8a99b6833c88 ("sched: Move SCHED_DEBUG sysctl to debugfs")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20230522195021.3456768-2-arnd@kernel.org

sched/fair: Do not even the number of busy CPUs via asym_packing

Now that find_busiest_group() triggers load balancing between a fully_
busy SMT2 core and an idle non-SMT core, it is no longer needed to force
balancing via asym_packing. Use asym_packing only as intended: when there
is high-priority CPU that is idle.

After this change, the same logic apply to SMT and non-SMT local groups.
It makes less sense having a separate function to deal specifically with
SMT. Fold the logic in asym_smt_can_pull_tasks() into sched_asym().

Signed-off-by: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Zhang Rui <rui.zhang@intel.com>
Link: https://lore.kernel.org/r/20230406203148.19182-8-ricardo.neri-calderon@linux.intel.com

sched/fair: Use the busiest group to set prefer_sibling

The prefer_sibling setting acts on the busiest group to move excess tasks
to the local group. This should be done as per request of the child of the
busiest group's sched domain, not the local group's.

Using the flags of the child domain of the local group works fortuitously
if both groups have child domains.

There are cases, however, in which the busiest group's sched domain has
child but the local group's does not. Consider, for instance a non-SMT
core (or an SMT core with only one online sibling) doing load balance with
an SMT core at the MC level. SD_PREFER_SIBLING of the busiest group's child
domain will not be honored. We are left with a fully busy SMT core and an
idle non-SMT core.

Suggested-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Signed-off-by: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Zhang Rui <rui.zhang@intel.com>
Link: https://lore.kernel.org/r/20230406203148.19182-7-ricardo.neri-calderon@linux.intel.com

sched/fair: Keep a fully_busy SMT sched group as busiest

When comparing two fully_busy scheduling groups, keep the current busiest
group if it represents an SMT core. Tasks in such scheduling group share
CPU resources and need more help than tasks in a non-SMT fully_busy group.

Signed-off-by: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Zhang Rui <rui.zhang@intel.com>
Link: https://lore.kernel.org/r/20230406203148.19182-6-ricardo.neri-calderon@linux.intel.com

sched/fair: Let low-priority cores help high-priority busy SMT cores

Using asym_packing priorities within an SMT core is straightforward. Just
follow the priorities that hardware indicates.

When balancing load from an SMT core, also consider the idle state of its
siblings. Priorities do not reflect that an SMT core divides its throughput
among all its busy siblings. They only makes sense when exactly one sibling
is busy.

Indicate that active balance is needed if the destination CPU has lower
priority than the source CPU but the latter has busy SMT siblings.

Make find_busiest_queue() not skip higher-priority SMT cores with more than
busy sibling.

Suggested-by: Valentin Schneider <vschneid@redhat.com>
Signed-off-by: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Zhang Rui <rui.zhang@intel.com>
Link: https://lore.kernel.org/r/20230406203148.19182-5-ricardo.neri-calderon@linux.intel.com

sched/fair: Simplify asym_packing logic for SMT cores

Callers of asym_smt_can_pull_tasks() check the idle state of the
destination CPU and its SMT siblings, if any. No extra checks are needed
in such function.

Since SMT cores divide capacity among its siblings, priorities only really
make sense if only one sibling is active. This is true for SMT2, SMT4,
SMT8, etc. Do not use asym_packing load balance for this case. Instead,
let find_busiest_group() handle imbalances.

When balancing non-SMT cores or at higher scheduling domains (e.g.,
between MC scheduling groups), continue using priorities.

Signed-off-by: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Len Brown <len.brown@intel.com>
Tested-by: Zhang Rui <rui.zhang@intel.com>
Link: https://lore.kernel.org/r/20230406203148.19182-4-ricardo.neri-calderon@linux.intel.com

sched/fair: Only do asym_packing load balancing from fully idle SMT cores

When balancing load between cores, all the SMT siblings of the destination
CPU, if any, must be idle. Otherwise, pulling new tasks degrades the
throughput of the busy SMT siblings. The overall throughput of the system
remains the same.

When balancing load within an SMT core this consideration is not relevant.
Follow the priorities that hardware indicates.

Suggested-by: Valentin Schneider <vschneid@redhat.com>
Signed-off-by: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Zhang Rui <rui.zhang@intel.com>
Link: https://lore.kernel.org/r/20230406203148.19182-3-ricardo.neri-calderon@linux.intel.com

sched/fair: Move is_core_idle() out of CONFIG_NUMA

asym_packing needs this function to determine whether an SMT core is a
suitable destination for load balancing.

Signed-off-by: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Zhang Rui <rui.zhang@intel.com>
Link: https://lore.kernel.org/r/20230406203148.19182-2-ricardo.neri-calderon@linux.intel.com

sched/numa: use hash_32 to mix up PIDs accessing VMA

before: last 6 bits of PID is used as index to store information about
tasks accessing VMA's.

after: hash_32 is used to take of cases where tasks are created over a
period of time, and thus improve collision probability.

Result:
The patch series overall improves autonuma cost.

Kernbench around more than 5% improvement and system time in mmtest
autonuma showed more than 80% improvement

Link: https://lkml.kernel.org/r/d5a9f75513300caed74e5c8570bba9317b963c2b.1677672277.git.raghavendra.kt@amd.com
Signed-off-by: Raghavendra K T <raghavendra.kt@amd.com>
Suggested-by: Peter Zijlstra <peterz@infradead.org>
Cc: Bharata B Rao <bharata@amd.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Disha Talreja <dishaa.talreja@amd.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Mike Rapoport <rppt@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

sched/numa: implement access PID reset logic

This helps to ensure that only recently accessed PIDs scan the VMAs.

Current implementation: (idea supported by PeterZ)

1. Accessing PID information is maintained in two windows.
access_pids[1] being newest.

2. Reset old access PID info i.e. access_pid[0] every (4 *
sysctl_numa_balancing_scan_delay) interval after initial scan delay
period expires.

The above interval seemed to be experimentally optimum since it avoids
frequent reset of access info as well as helps clearing the old access
info regularly. The reset logic is implemented in scan path.

Link: https://lkml.kernel.org/r/f7a675f66d1442d048b4216b2baf94515012c405.1677672277.git.raghavendra.kt@amd.com
Signed-off-by: Raghavendra K T <raghavendra.kt@amd.com>
Suggested-by: Mel Gorman <mgorman@techsingularity.net>
Cc: Bharata B Rao <bharata@amd.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Disha Talreja <dishaa.talreja@amd.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

sched/numa: enhance vma scanning logic

During Numa scanning make sure only relevant vmas of the tasks are
scanned.

Before:
All the tasks of a process participate in scanning the vma even if they
do not access vma in it's lifespan.

Now:
Except cases of first few unconditional scans, if a process do
not touch vma (exluding false positive cases of PID collisions)
tasks no longer scan all vma

Logic used:

1) 6 bits of PID used to mark active bit in vma numab status during
fault to remember PIDs accessing vma. (Thanks Mel)

2) Subsequently in scan path, vma scanning is skipped if current PID
had not accessed vma.

3) First two times we do allow unconditional scan to preserve earlier
behaviour of scanning.

Acknowledgement to Bharata B Rao <bharata@amd.com> for initial patch to
store pid information and Peter Zijlstra <peterz@infradead.org> (Usage of
test and set bit)

Link: https://lkml.kernel.org/r/092f03105c7c1d3450f4636b1ea350407f07640e.1677672277.git.raghavendra.kt@amd.com
Signed-off-by: Raghavendra K T <raghavendra.kt@amd.com>
Suggested-by: Mel Gorman <mgorman@techsingularity.net>
Cc: David Hildenbrand <david@redhat.com>
Cc: Disha Talreja <dishaa.talreja@amd.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Mike Rapoport <rppt@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

sched/numa: apply the scan delay to every new vma

Pach series "sched/numa: Enhance vma scanning", v3.

The patchset proposes one of the enhancements to numa vma scanning
suggested by Mel. This is continuation of [3].

Reposting the rebased patchset to akpm mm-unstable tree (March 1)

Existing mechanism of scan period involves, scan period derived from
per-thread stats. Process Adaptive autoNUMA [1] proposed to gather NUMA
fault stats at per-process level to capture aplication behaviour better.

During that course of discussion, Mel proposed several ideas to enhance
current numa balancing. One of the suggestion was below

Track what threads access a VMA. The suggestion was to use an unsigned
long pid_mask and use the lower bits to tag approximately what threads
access a VMA. Skip VMAs that did not trap a fault. This would be
approximate because of PID collisions but would reduce scanning of areas
the thread is not interested in. The above suggestion intends not to
penalize threads that has no interest in the vma, thus reduce scanning
overhead.

V3 changes are mostly based on PeterZ comments (details below in changes)

Summary of patchset:

Current patchset implements:

1. Delay the vma scanning logic for newly created VMA's so that
additional overhead of scanning is not incurred for short lived tasks
(implementation by Mel)

2. Store the information of tasks accessing VMA in 2 windows. It is
regularly cleared in (4*sysctl_numa_balancing_scan_delay) interval.
The above time is derived from experimenting (Suggested by PeterZ) to
balance between frequent clearing vs obsolete access data

3. hash_32 used to encode task index accessing VMA information

4. VMA's acess information is used to skip scanning for the tasks
which had not accessed VMA

Changes since V2:
patch1:
- Renaming of structure, macro to function,
- Add explanation to heuristics
- Adding more details from result (PeterZ)
Patch2:
- Usage of test and set bit (PeterZ)
- Move storing access PID info to numa_migrate_prep()
- Add a note on fainess among tasks allowed to scan
(PeterZ)
Patch3:
- Maintain two windows of access PID information
(PeterZ supported implementation and Gave idea to extend
to N if needed)
Patch4:
- Apply hash_32 function to track VMA accessing PIDs (PeterZ)

Changes since RFC V1:
- Include Mel's vma scan delay patch
- Change the accessing pid store logic (Thanks Mel)
- Fencing structure / code to NUMA_BALANCING (David, Mel)
- Adding clearing access PID logic (Mel)
- Descriptive change log ( Mike Rapoport)

Things to ponder over:
==========================================

- Improvement to clearing accessing PIDs logic (discussed in-detail in
patch3 itself (Done in this patchset by implementing 2 window history)

- Current scan period is not changed in the patchset, so we do see
frequent tries to scan. Relaxing scan period dynamically could improve
results further.

[1] sched/numa: Process Adaptive autoNUMA
Link: https://lore.kernel.org/lkml/20220128052851.17162-1-bharata@amd.com/T/

[2] RFC V1 Link:
https://lore.kernel.org/all/cover.1673610485.git.raghavendra.kt@amd.com/

[3] V2 Link:
https://lore.kernel.org/lkml/cover.1675159422.git.raghavendra.kt@amd.com/


Results:
Summary: Huge autonuma cost reduction seen in mmtest. Kernbench improvement
is more than 5% and huge system time (80%+) improvement from mmtest autonuma.
(dbench had huge std deviation to post)

kernbench
===========
6.2.0-mmunstable-base 6.2.0-mmunstable-patched
Amean user-256 22002.51 ( 0.00%) 22649.95 * -2.94%*
Amean syst-256 10162.78 ( 0.00%) 8214.13 * 19.17%*
Amean elsp-256 160.74 ( 0.00%) 156.92 * 2.38%*

Duration User 66017.43 67959.84
Duration System 30503.15 24657.03
Duration Elapsed 504.61 493.12

6.2.0-mmunstable-base 6.2.0-mmunstable-patched
Ops NUMA alloc hit 1738835089.00 1738780310.00
Ops NUMA alloc local 1738834448.00 1738779711.00
Ops NUMA base-page range updates 477310.00 392566.00
Ops NUMA PTE updates 477310.00 392566.00
Ops NUMA hint faults 96817.00 87555.00
Ops NUMA hint local faults % 10150.00 2192.00
Ops NUMA hint local percent 10.48 2.50
Ops NUMA pages migrated 86660.00 85363.00
Ops AutoNUMA cost 489.07 442.14

autonumabench
===============
6.2.0-mmunstable-base 6.2.0-mmunstable-patched
Amean syst-NUMA01 399.50 ( 0.00%) 52.05 * 86.97%*
Amean syst-NUMA01_THREADLOCAL 0.21 ( 0.00%) 0.22 * -5.41%*
Amean syst-NUMA02 0.80 ( 0.00%) 0.78 * 2.68%*
Amean syst-NUMA02_SMT 0.65 ( 0.00%) 0.68 * -3.95%*
Amean elsp-NUMA01 313.26 ( 0.00%) 313.11 * 0.05%*
Amean elsp-NUMA01_THREADLOCAL 1.06 ( 0.00%) 1.08 * -1.76%*
Amean elsp-NUMA02 3.19 ( 0.00%) 3.24 * -1.52%*
Amean elsp-NUMA02_SMT 3.72 ( 0.00%) 3.61 * 2.92%*

Duration User 396433.47 324835.96
Duration System 2808.70 376.66
Duration Elapsed 2258.61 2258.12

6.2.0-mmunstable-base 6.2.0-mmunstable-patched
Ops NUMA alloc hit 59921806.00 49623489.00
Ops NUMA alloc miss 0.00 0.00
Ops NUMA interleave hit 0.00 0.00
Ops NUMA alloc local 59920880.00 49622594.00
Ops NUMA base-page range updates 152259275.00 50075.00
Ops NUMA PTE updates 152259275.00 50075.00
Ops NUMA PMD updates 0.00 0.00
Ops NUMA hint faults 154660352.00 39014.00
Ops NUMA hint local faults % 138550501.00 23139.00
Ops NUMA hint local percent 89.58 59.31
Ops NUMA pages migrated 8179067.00 14147.00
Ops AutoNUMA cost 774522.98 195.69


This patch (of 4):

Currently whenever a new task is created we wait for
sysctl_numa_balancing_scan_delay to avoid unnessary scanning overhead.
Extend the same logic to new or very short-lived VMAs.

[raghavendra.kt@amd.com: add initialization in vm_area_dup())]
Link: https://lkml.kernel.org/r/cover.1677672277.git.raghavendra.kt@amd.com
Link: https://lkml.kernel.org/r/7a6fbba87c8b51e67efd3e74285bb4cb311a16ca.1677672277.git.raghavendra.kt@amd.com
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Raghavendra K T <raghavendra.kt@amd.com>
Cc: Bharata B Rao <bharata@amd.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Disha Talreja <dishaa.talreja@amd.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

sched/fair: Fix inaccurate tally of ttwu_move_affine

There are scenarios where non-affine wakeups are incorrectly counted as
affine wakeups by schedstats.

When wake_affine_idle() returns prev_cpu which doesn't equal to
nr_cpumask_bits, it will slip through the check: target == nr_cpumask_bits
in wake_affine() and be counted as if target == this_cpu in schedstats.

Replace target == nr_cpumask_bits with target != this_cpu to make sure
affine wakeups are accurately tallied.

Fixes: 806486c377e33 (sched/fair: Do not migrate if the prev_cpu is idle)
Suggested-by: Daniel Jordan <daniel.m.jordan@oracle.com>
Signed-off-by: Libo Chen <libo.chen@oracle.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Gautham R. Shenoy <gautham.shenoy@amd.com>
Link: https://lore.kernel.org/r/20220810223313.386614-1-libo.chen@oracle.com

sched: Interleave cfs bandwidth timers for improved single thread performance at low utilization

CPU cfs bandwidth controller uses hrtimer. Currently there is no initial
value set. Hence all period timers would align at expiry.
This happens when there are multiple CPU cgroup's.

There is a performance gain that can be achieved here if the timers are
interleaved when the utilization of each CPU cgroup is low and total
utilization of all the CPU cgroup's is less than 50%. If the timers are
interleaved, then the unthrottled cgroup can run freely without many
context switches and can also benefit from SMT Folding. This effect will
be further amplified in SPLPAR environment.

This commit adds a random offset after initializing each hrtimer. This
would result in interleaving the timers at expiry, which helps in achieving
the said performance gain.

This was tested on powerpc platform with 8 core SMT=8. Socket power was
measured when the workload. Benchmarked the stress-ng with power
information. Throughput oriented benchmarks show significant gain up to
25% while power consumption increases up to 15%.

Workload: stress-ng --cpu=32 --cpu-ops=50000.
1CG - 1 cgroup is running.
2CG - 2 cgroups are running together.
Time taken to complete stress-ng in seconds and power is in watts.
each cgroup is throttled at 25% with 100ms as the period value.
6.2-rc6 | with patch
8 core 1CG power 2CG power | 1CG power 2 CG power
27.5 80.6 40 90 | 27.3 82 32.3 104
27.5 81 40.2 91 | 27.5 81 38.7 96
27.7 80 40.1 89 | 27.6 80 29.7 106
27.7 80.1 40.3 94 | 27.6 80 31.5 105

Latency might be affected by this change. That could happen if the CPU was
in a deep idle state which is possible if we interleave the timers. Used
schbench for measuring the latency. Each cgroup is throttled at 25% with
period value is set to 100ms. Numbers are when both the cgroups are
running simultaneously. Latency values don't degrade much. Some
improvement is seen in tail latencies.

6.2-rc6 with patch
Groups: 16
50.0th: 39.5 42.5
75.0th: 924.0 922.0
90.0th: 972.0 968.0
95.0th: 1005.5 994.0
99.0th: 4166.0 2287.0
99.5th: 7314.0 7448.0
99.9th: 15024.0 13600.0

Groups: 32
50.0th: 819.0 463.0
75.0th: 1596.0 918.0
90.0th: 5992.0 1281.5
95.0th: 13184.0 2765.0
99.0th: 21792.0 14240.0
99.5th: 25696.0 18920.0
99.9th: 33280.0 35776.0

Groups: 64
50.0th: 4806.0 3440.0
75.0th: 31136.0 33664.0
90.0th: 54144.0 58752.0
95.0th: 66176.0 67200.0
99.0th: 84736.0 91520.0
99.5th: 97408.0 114048.0
99.9th: 136448.0 140032.0

Initial RFC PATCH, discussions and details on the problem:

Link1: https://lore.kernel.org/lkml/5ae3cb09-8c9a-11e8-75a7-cc774d9bc283@linux.vnet.ibm.com/
Link2: https://lore.kernel.org/lkml/9c57c92c-3e0c-b8c5-4be9-8f4df344a347@linux.vnet.ibm.com/
Suggested-by: Peter Zijlstra <peterz@infradead.org>
Suggested-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Shrikanth Hegde<sshegde@linux.vnet.ibm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Ben Segall <bsegall@google.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20230223185153.1499710-1-sshegde@linux.vnet.ibm.com

sched/core: Avoid selecting the task that is throttled to run when core-sched enable

When {rt, cfs}_rq or dl task is throttled, since cookied tasks
are not dequeued from the core tree, So sched_core_find() and
sched_core_next() may return throttled task, which may
cause throttled task to run on the CPU.

So we add checks in sched_core_find() and sched_core_next()
to make sure that the return is a runnable task that is
not throttled.

Co-developed-by: Cruz Zhao <CruzZhao@linux.alibaba.com>
Signed-off-by: Cruz Zhao <CruzZhao@linux.alibaba.com>
Signed-off-by: Hao Jia <jiahao.os@bytedance.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20230316081806.69544-1-jiahao.os@bytedance.com

sched/fair: Fix imbalance overflow

When local group is fully busy but its average load is above system load,
computing the imbalance will overflow and local group is not the best
target for pulling this load.

Fixes: 0b0695f2b34a ("sched/fair: Rework load_balance()")
Reported-by: Tingjia Cao <tjcao980311@gmail.com>
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Tingjia Cao <tjcao980311@gmail.com>
Link: https://lore.kernel.org/lkml/CABcWv9_DAhVBOq2=W=2ypKE9dKM5s2DvoV8-U0+GDwwuKZ89jQ@mail.gmail.com/T/

sched/fair: Sanitize vruntime of entity being migrated

Commit 829c1651e9c4 ("sched/fair: sanitize vruntime of entity being placed")
fixes an overflowing bug, but ignore a case that se->exec_start is reset
after a migration.

For fixing this case, we delay the reset of se->exec_start after
placing the entity which se->exec_start to detect long sleeping task.

In order to take into account a possible divergence between the clock_task
of 2 rqs, we increase the threshold to around 104 days.

Fixes: 829c1651e9c4 ("sched/fair: sanitize vruntime of entity being placed")
Originally-by: Zhang Qiao <zhangqiao22@huawei.com>
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Zhang Qiao <zhangqiao22@huawei.com>
Link: https://lore.kernel.org/r/20230317160810.107988-1-vincent.guittot@linaro.org

sched: convert to vma iterator

Use the vma iterator so that the iterator can be invalidated or updated to
avoid each caller doing so.

Link: https://lkml.kernel.org/r/20230120162650.984577-23-Liam.Howlett@oracle.com
Signed-off-by: Liam R. Howlett <Liam.Howlett@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

sched/fair: sanitize vruntime of entity being placed

When a scheduling entity is placed onto cfs_rq, its vruntime is pulled
to the base level (around cfs_rq->min_vruntime), so that the entity
doesn't gain extra boost when placed backwards.

However, if the entity being placed wasn't executed for a long time, its
vruntime may get too far behind (e.g. while cfs_rq was executing a
low-weight hog), which can inverse the vruntime comparison due to s64
overflow. This results in the entity being placed with its original
vruntime way forwards, so that it will effectively never get to the cpu.

To prevent that, ignore the vruntime of the entity being placed if it
didn't execute for much longer than the characteristic sheduler time
scale.

[rkagan: formatted, adjusted commit log, comments, cutoff value]
Signed-off-by: Zhang Qiao <zhangqiao22@huawei.com>
Co-developed-by: Roman Kagan <rkagan@amazon.de>
Signed-off-by: Roman Kagan <rkagan@amazon.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20230130122216.3555094-1-rkagan@amazon.de

sched/fair: Remove capacity inversion detection

Remove the capacity inversion detection which is now handled by
util_fits_cpu() returning -1 when we need to continue to look for a
potential CPU with better performance.

This ends up almost reverting patches below except for some comments:
commit da07d2f9c153 ("sched/fair: Fixes for capacity inversion detection")
commit aa69c36f31aa ("sched/fair: Consider capacity inversion in util_fits_cpu()")
commit 44c7b80bffc3 ("sched/fair: Detect capacity inversion")

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20230201143628.270912-3-vincent.guittot@linaro.org

sched/fair: unlink misfit task from cpu overutilized

By taking into account uclamp_min, the 1:1 relation between task misfit
and cpu overutilized is no more true as a task with a small util_avg may
not fit a high capacity cpu because of uclamp_min constraint.

Add a new state in util_fits_cpu() to reflect the case that task would fit
a CPU except for the uclamp_min hint which is a performance requirement.

Use -1 to reflect that a CPU doesn't fit only because of uclamp_min so we
can use this new value to take additional action to select the best CPU
that doesn't match uclamp_min hint.

When util_fits_cpu() returns -1, we will continue to look for a possible
CPU with better performance, which replaces Capacity Inversion detection
with capacity_orig_of() - thermal_load_avg to detect a capacity inversion.

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-and-tested-by: Qais Yousef <qyousef@layalina.io>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Tested-by: Kajetan Puchalski <kajetan.puchalski@arm.com>
Link: https://lore.kernel.org/r/20230201143628.270912-2-vincent.guittot@linaro.org

sched/fair: Limit sched slice duration

In presence of a lot of small weight tasks like sched_idle tasks, normal
or high weight tasks can see their ideal runtime (sched_slice) to increase
to hundreds ms whereas it normally stays below sysctl_sched_latency.

2 normal tasks running on a CPU will have a max sched_slice of 12ms
(half of the sched_period). This means that they will make progress
every sysctl_sched_latency period.

If we now add 1000 idle tasks on the CPU, the sched_period becomes
3006 ms and the ideal runtime of the normal tasks becomes 609 ms.
It will even become 1500ms if the idle tasks belongs to an idle cgroup.
This means that the scheduler will look for picking another waiting task
after 609ms running time (1500ms respectively). The idle tasks change
significantly the way the 2 normal tasks interleave their running time
slot whereas they should have a small impact.

Such long sched_slice can delay significantly the release of resources
as the tasks can wait hundreds of ms before the next running slot just
because of idle tasks queued on the rq.

Cap the ideal_runtime to sysctl_sched_latency to make sure that tasks will
regularly make progress and will not be significantly impacted by
idle/background tasks queued on the rq.

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Tested-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Link: https://lore.kernel.org/r/20230113133613.257342-1-vincent.guittot@linaro.org

sched/core: Adjusting the order of scanning CPU

When select_idle_capacity() starts scanning for an idle CPU, it starts
with target CPU that has already been checked in select_idle_sibling().
So we start checking from the next CPU and try the target CPU at the end.
Similarly for task_numa_assign(), we have just checked numa_migrate_on
of dst_cpu, so start from the next CPU. This also works for
steal_cookie_task(), the first scan must fail and start directly
from the next one.

Signed-off-by: Hao Jia <jiahao.os@bytedance.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Acked-by: Mel Gorman <mgorman@techsingularity.net>
Link: https://lore.kernel.org/r/20221216062406.7812-3-jiahao.os@bytedance.com

sched/numa: Stop an exhastive search if an idle core is found

In update_numa_stats() we try to find an idle cpu on the NUMA node,
preferably an idle core. we can stop looking for the next idle core
or idle cpu after finding an idle core. But we can't stop the
whole loop of scanning the CPU, because we need to calculate
approximate NUMA stats at a point in time. For example,
the src and dst nr_running is needed by task_numa_find_cpu().

Signed-off-by: Hao Jia <jiahao.os@bytedance.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Mel Gorman <mgorman@techsingularity.net>
Link: https://lore.kernel.org/r/20221216062406.7812-2-jiahao.os@bytedance.com

sched: Make const-safe

With a modified container_of() that preserves constness, the compiler
finds some pointers which should have been marked as const. task_of()
also needs to become const-preserving for the !FAIR_GROUP_SCHED case so
that cfs_rq_of() can take a const argument. No change to generated code.

Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20221212144946.2657785-1-willy@infradead.org

sched: Async unthrottling for cfs bandwidth

CFS bandwidth currently distributes new runtime and unthrottles cfs_rq's
inline in an hrtimer callback. Runtime distribution is a per-cpu
operation, and unthrottling is a per-cgroup operation, since a tg walk
is required. On machines with a large number of cpus and large cgroup
hierarchies, this cpus*cgroups work can be too much to do in a single
hrtimer callback: since IRQ are disabled, hard lockups may easily occur.
Specifically, we've found this scalability issue on configurations with
256 cpus, O(1000) cgroups in the hierarchy being throttled, and high
memory bandwidth usage.

To fix this, we can instead unthrottle cfs_rq's asynchronously via a
CSD. Each cpu is responsible for unthrottling itself, thus sharding the
total work more fairly across the system, and avoiding hard lockups.

Signed-off-by: Josh Don <joshdon@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20221117005418.3499691-1-joshdon@google.com

sched/fair: Fixes for capacity inversion detection

Traversing the Perf Domains requires rcu_read_lock() to be held and is
conditional on sched_energy_enabled(). Ensure right protections applied.

Also skip capacity inversion detection for our own pd; which was an
error.

Fixes: 44c7b80bffc3 ("sched/fair: Detect capacity inversion")
Reported-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Signed-off-by: Qais Yousef (Google) <qyousef@layalina.io>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20230112122708.330667-3-qyousef@layalina.io

sched/uclamp: Fix a uninitialized variable warnings

Addresses the following warnings:

> config: riscv-randconfig-m031-20221111
> compiler: riscv64-linux-gcc (GCC) 12.1.0
>
> smatch warnings:
> kernel/sched/fair.c:7263 find_energy_efficient_cpu() error: uninitialized symbol 'util_min'.
> kernel/sched/fair.c:7263 find_energy_efficient_cpu() error: uninitialized symbol 'util_max'.

Fixes: 244226035a1f ("sched/uclamp: Fix fits_capacity() check in feec()")
Reported-by: kernel test robot <lkp@intel.com>
Reported-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: Qais Yousef (Google) <qyousef@layalina.io>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20230112122708.330667-2-qyousef@layalina.io

sched: Move numa_balancing sysctls to its own file

The sysctl_numa_balancing_promote_rate_limit and sysctl_numa_balancing
are part of sched, move them to its own file.

Signed-off-by: Kefeng Wang <wangkefeng.wang@huawei.com>
Signed-off-by: Luis Chamberlain <mcgrof@kernel.org>

sched/fair: use try_cmpxchg in task_numa_work

Use try_cmpxchg instead of cmpxchg (*ptr, old, new) == old in
task_numa_work. x86 CMPXCHG instruction returns success in ZF flag, so
this change saves a compare after cmpxchg (and related move instruction in
front of cmpxchg).

No functional change intended.

Link: https://lkml.kernel.org/r/20220822173956.82525-1-ubizjak@gmail.com
Signed-off-by: Uros Bizjak <ubizjak@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

sched/fair: Check if prev_cpu has highest spare cap in feec()

When evaluating the CPU candidates in the perf domain (pd) containing
the previously used CPU (prev_cpu), find_energy_efficient_cpu()
evaluates the energy of the pd:
- without the task (base_energy)
- with the task placed on prev_cpu (if the task fits)
- with the task placed on the CPU with the highest spare capacity,
prev_cpu being excluded from this set

If prev_cpu is already the CPU with the highest spare capacity,
max_spare_cap_cpu will be the CPU with the second highest spare
capacity.

On an Arm64 Juno-r2, with a workload of 10 tasks at a 10% duty cycle,
when prev_cpu and max_spare_cap_cpu are both valid candidates,
prev_spare_cap > max_spare_cap at ~82%.
Thus the energy of the pd when placing the task on max_spare_cap_cpu
is computed with no possible positive outcome 82% most of the time.

Do not consider max_spare_cap_cpu as a valid candidate if
prev_spare_cap > max_spare_cap.

Signed-off-by: Pierre Gondois <pierre.gondois@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20221006081052.3862167-2-pierre.gondois@arm.com

sched/fair: Consider capacity inversion in util_fits_cpu()

We do consider thermal pressure in util_fits_cpu() for uclamp_min only.
With the exception of the biggest cores which by definition are the max
performance point of the system and all tasks by definition should fit.

Even under thermal pressure, the capacity of the biggest CPU is the
highest in the system and should still fit every task. Except when it
reaches capacity inversion point, then this is no longer true.

We can handle this by using the inverted capacity as capacity_orig in
util_fits_cpu(). Which not only addresses the problem above, but also
ensure uclamp_max now considers the inverted capacity. Force fitting
a task when a CPU is in this adverse state will contribute to making the
thermal throttling last longer.

Signed-off-by: Qais Yousef <qais.yousef@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220804143609.515789-10-qais.yousef@arm.com

sched/fair: Detect capacity inversion

Check each performance domain to see if thermal pressure is causing its
capacity to be lower than another performance domain.

We assume that each performance domain has CPUs with the same
capacities, which is similar to an assumption made in energy_model.c

We also assume that thermal pressure impacts all CPUs in a performance
domain equally.

If there're multiple performance domains with the same capacity_orig, we
will trigger a capacity inversion if the domain is under thermal
pressure.

The new cpu_in_capacity_inversion() should help users to know when
information about capacity_orig are not reliable and can opt in to use
the inverted capacity as the 'actual' capacity_orig.

Signed-off-by: Qais Yousef <qais.yousef@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220804143609.515789-9-qais.yousef@arm.com

sched/uclamp: Cater for uclamp in find_energy_efficient_cpu()'s early exit condition

If the utilization of the woken up task is 0, we skip the energy
calculation because it has no impact.

But if the task is boosted (uclamp_min != 0) will have an impact on task
placement and frequency selection. Only skip if the util is truly
0 after applying uclamp values.

Change uclamp_task_cpu() signature to avoid unnecessary additional calls
to uclamp_eff_get(). feec() is the only user now.

Fixes: 732cd75b8c920 ("sched/fair: Select an energy-efficient CPU on task wake-up")
Signed-off-by: Qais Yousef <qais.yousef@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220804143609.515789-8-qais.yousef@arm.com

sched/uclamp: Make cpu_overutilized() use util_fits_cpu()

So that it is now uclamp aware.

This fixes a major problem of busy tasks capped with UCLAMP_MAX keeping
the system in overutilized state which disables EAS and leads to wasting
energy in the long run.

Without this patch running a busy background activity like JIT
compilation on Pixel 6 causes the system to be in overutilized state
74.5% of the time.

With this patch this goes down to 9.79%.

It also fixes another problem when long running tasks that have their
UCLAMP_MIN changed while running such that they need to upmigrate to
honour the new UCLAMP_MIN value. The upmigration doesn't get triggered
because overutilized state never gets set in this state, hence misfit
migration never happens at tick in this case until the task wakes up
again.

Fixes: af24bde8df202 ("sched/uclamp: Add uclamp support to energy_compute()")
Signed-off-by: Qais Yousef <qais.yousef@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220804143609.515789-7-qais.yousef@arm.com

sched/uclamp: Make asym_fits_capacity() use util_fits_cpu()

Use the new util_fits_cpu() to ensure migration margin and capacity
pressure are taken into account correctly when uclamp is being used
otherwise we will fail to consider CPUs as fitting in scenarios where
they should.

s/asym_fits_capacity/asym_fits_cpu/ to better reflect what it does now.

Fixes: b4c9c9f15649 ("sched/fair: Prefer prev cpu in asymmetric wakeup path")
Signed-off-by: Qais Yousef <qais.yousef@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220804143609.515789-6-qais.yousef@arm.com

sched/uclamp: Make select_idle_capacity() use util_fits_cpu()

Use the new util_fits_cpu() to ensure migration margin and capacity
pressure are taken into account correctly when uclamp is being used
otherwise we will fail to consider CPUs as fitting in scenarios where
they should.

Fixes: b4c9c9f15649 ("sched/fair: Prefer prev cpu in asymmetric wakeup path")
Signed-off-by: Qais Yousef <qais.yousef@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220804143609.515789-5-qais.yousef@arm.com

sched/uclamp: Fix fits_capacity() check in feec()

As reported by Yun Hsiang [1], if a task has its uclamp_min >= 0.8 * 1024,
it'll always pick the previous CPU because fits_capacity() will always
return false in this case.

The new util_fits_cpu() logic should handle this correctly for us beside
more corner cases where similar failures could occur, like when using
UCLAMP_MAX.

We open code uclamp_rq_util_with() except for the clamp() part,
util_fits_cpu() needs the 'raw' values to be passed to it.

Also introduce uclamp_rq_{set, get}() shorthand accessors to get uclamp
value for the rq. Makes the code more readable and ensures the right
rules (use READ_ONCE/WRITE_ONCE) are respected transparently.

[1] https://lists.linaro.org/pipermail/eas-dev/2020-July/001488.html

Fixes: 1d42509e475c ("sched/fair: Make EAS wakeup placement consider uclamp restrictions")
Reported-by: Yun Hsiang <hsiang023167@gmail.com>
Signed-off-by: Qais Yousef <qais.yousef@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220804143609.515789-4-qais.yousef@arm.com

sched/uclamp: Make task_fits_capacity() use util_fits_cpu()

So that the new uclamp rules in regard to migration margin and capacity
pressure are taken into account correctly.

Fixes: a7008c07a568 ("sched/fair: Make task_fits_capacity() consider uclamp restrictions")
Co-developed-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Qais Yousef <qais.yousef@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220804143609.515789-3-qais.yousef@arm.com

sched/uclamp: Fix relationship between uclamp and migration margin

fits_capacity() verifies that a util is within 20% margin of the
capacity of a CPU, which is an attempt to speed up upmigration.

But when uclamp is used, this 20% margin is problematic because for
example if a task is boosted to 1024, then it will not fit on any CPU
according to fits_capacity() logic.

Or if a task is boosted to capacity_orig_of(medium_cpu). The task will
end up on big instead on the desired medium CPU.

Similar corner cases exist for uclamp and usage of capacity_of().
Slightest irq pressure on biggest CPU for example will make a 1024
boosted task look like it can't fit.

What we really want is for uclamp comparisons to ignore the migration
margin and capacity pressure, yet retain them for when checking the
_actual_ util signal.

For example, task p:

p->util_avg = 300
p->uclamp[UCLAMP_MIN] = 1024

Will fit a big CPU. But

p->util_avg = 900
p->uclamp[UCLAMP_MIN] = 1024

will not, this should trigger overutilized state because the big CPU is
now *actually* being saturated.

Similar reasoning applies to capping tasks with UCLAMP_MAX. For example:

p->util_avg = 1024
p->uclamp[UCLAMP_MAX] = capacity_orig_of(medium_cpu)

Should fit the task on medium cpus without triggering overutilized
state.

Inlined comments expand more on desired behavior in more scenarios.

Introduce new util_fits_cpu() function which encapsulates the new logic.
The new function is not used anywhere yet, but will be used to update
various users of fits_capacity() in later patches.

Fixes: af24bde8df202 ("sched/uclamp: Add uclamp support to energy_compute()")
Signed-off-by: Qais Yousef <qais.yousef@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220804143609.515789-2-qais.yousef@arm.com

sched: use maple tree iterator to walk VMAs

The linked list is slower than walking the VMAs using the maple tree. We
can't use the VMA iterator here because it doesn't support moving to an
earlier position.

Link: https://lkml.kernel.org/r/20220906194824.2110408-49-Liam.Howlett@oracle.com
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Signed-off-by: Liam R. Howlett <Liam.Howlett@Oracle.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Tested-by: Yu Zhao <yuzhao@google.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Davidlohr Bueso <dave@stgolabs.net>
Cc: SeongJae Park <sj@kernel.org>
Cc: Sven Schnelle <svens@linux.ibm.com>
Cc: Will Deacon <will@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm/demotion: update node_is_toptier to work with memory tiers

With memory tier support we can have memory only NUMA nodes in the top
tier from which we want to avoid promotion tracking NUMA faults. Update
node_is_toptier to work with memory tiers. All NUMA nodes are by default
top tier nodes. With lower(slower) memory tiers added we consider all
memory tiers above a memory tier having CPU NUMA nodes as a top memory
tier

[sj@kernel.org: include missed header file, memory-tiers.h]
Link: https://lkml.kernel.org/r/20220820190720.248704-1-sj@kernel.org
[akpm@linux-foundation.org: mm/memory.c needs linux/memory-tiers.h]
[aneesh.kumar@linux.ibm.com: make toptier_distance inclusive upper bound of toptiers]
Link: https://lkml.kernel.org/r/20220830081457.118960-1-aneesh.kumar@linux.ibm.com
Link: https://lkml.kernel.org/r/20220818131042.113280-10-aneesh.kumar@linux.ibm.com
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Reviewed-by: "Huang, Ying" <ying.huang@intel.com>
Acked-by: Wei Xu <weixugc@google.com>
Cc: Alistair Popple <apopple@nvidia.com>
Cc: Bharata B Rao <bharata@amd.com>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Davidlohr Bueso <dave@stgolabs.net>
Cc: Hesham Almatary <hesham.almatary@huawei.com>
Cc: Jagdish Gediya <jvgediya.oss@gmail.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Tim Chen <tim.c.chen@intel.com>
Cc: Yang Shi <shy828301@gmail.com>
Cc: SeongJae Park <sj@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

sched/fair: Move call to list_last_entry() in detach_tasks

Move the call to list_last_entry() in detach_tasks() after testing
loop_max and loop_break.

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20220825122726.20819-4-vincent.guittot@linaro.org

sched/fair: Cleanup loop_max and loop_break

sched_nr_migrate_break is set to a fix value and never changes so we can
replace it by a define SCHED_NR_MIGRATE_BREAK.

Also, we adjust SCHED_NR_MIGRATE_BREAK to be aligned with the init value
of sysctl_sched_nr_migrate which can be init to different values.

Then, use SCHED_NR_MIGRATE_BREAK to init sysctl_sched_nr_migrate.

The behavior stays unchanged unless you modify sysctl_sched_nr_migrate
trough debugfs.

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20220825122726.20819-3-vincent.guittot@linaro.org

sched/fair: Make sure to try to detach at least one movable task

During load balance, we try at most env->loop_max time to move a task.
But it can happen that the loop_max LRU tasks (ie tail of
the cfs_tasks list) can't be moved to dst_cpu because of affinity.
In this case, loop in the list until we found at least one.

The maximum of detached tasks remained the same as before.

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20220825122726.20819-2-vincent.guittot@linaro.org

memory tiering: adjust hot threshold automatically

The promotion hot threshold is workload and system configuration
dependent. So in this patch, a method to adjust the hot threshold
automatically is implemented. The basic idea is to control the number of
the candidate promotion pages to match the promotion rate limit. If the
hint page fault latency of a page is less than the hot threshold, we will
try to promote the page, and the page is called the candidate promotion
page.

If the number of the candidate promotion pages in the statistics interval
is much more than the promotion rate limit, the hot threshold will be
decreased to reduce the number of the candidate promotion pages.
Otherwise, the hot threshold will be increased to increase the number of
the candidate promotion pages.

To make the above method works, in each statistics interval, the total
number of the pages to check (on which the hint page faults occur) and the
hot/cold distribution need to be stable. Because the page tables are
scanned linearly in NUMA balancing, but the hot/cold distribution isn't
uniform along the address usually, the statistics interval should be
larger than the NUMA balancing scan period. So in the patch, the max scan
period is used as statistics interval and it works well in our tests.

Link: https://lkml.kernel.org/r/20220713083954.34196-4-ying.huang@intel.com
Signed-off-by: "Huang, Ying" <ying.huang@intel.com>
Reviewed-by: Baolin Wang <baolin.wang@linux.alibaba.com>
Tested-by: Baolin Wang <baolin.wang@linux.alibaba.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Michal Hocko <mhocko@suse.com>
Cc: osalvador <osalvador@suse.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Rik van Riel <riel@surriel.com>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: Wei Xu <weixugc@google.com>
Cc: Yang Shi <shy828301@gmail.com>
Cc: Zhong Jiang <zhongjiang-ali@linux.alibaba.com>
Cc: Zi Yan <ziy@nvidia.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

memory tiering: rate limit NUMA migration throughput

In NUMA balancing memory tiering mode, if there are hot pages in slow
memory node and cold pages in fast memory node, we need to promote/demote
hot/cold pages between the fast and cold memory nodes.

A choice is to promote/demote as fast as possible. But the CPU cycles and
memory bandwidth consumed by the high promoting/demoting throughput will
hurt the latency of some workload because of accessing inflating and slow
memory bandwidth contention.

A way to resolve this issue is to restrict the max promoting/demoting
throughput. It will take longer to finish the promoting/demoting. But
the workload latency will be better. This is implemented in this patch as
the page promotion rate limit mechanism.

The number of the candidate pages to be promoted to the fast memory node
via NUMA balancing is counted, if the count exceeds the limit specified by
the users, the NUMA balancing promotion will be stopped until the next
second.

A new sysctl knob kernel.numa_balancing_promote_rate_limit_MBps is added
for the users to specify the limit.

Link: https://lkml.kernel.org/r/20220713083954.34196-3-ying.huang@intel.com
Signed-off-by: "Huang, Ying" <ying.huang@intel.com>
Reviewed-by: Baolin Wang <baolin.wang@linux.alibaba.com>
Tested-by: Baolin Wang <baolin.wang@linux.alibaba.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Michal Hocko <mhocko@suse.com>
Cc: osalvador <osalvador@suse.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Rik van Riel <riel@surriel.com>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: Wei Xu <weixugc@google.com>
Cc: Yang Shi <shy828301@gmail.com>
Cc: Zhong Jiang <zhongjiang-ali@linux.alibaba.com>
Cc: Zi Yan <ziy@nvidia.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

memory tiering: hot page selection with hint page fault latency

Patch series "memory tiering: hot page selection", v4.

To optimize page placement in a memory tiering system with NUMA balancing,
the hot pages in the slow memory nodes need to be identified.
Essentially, the original NUMA balancing implementation selects the mostly
recently accessed (MRU) pages to promote. But this isn't a perfect
algorithm to identify the hot pages. Because the pages with quite low
access frequency may be accessed eventually given the NUMA balancing page
table scanning period could be quite long (e.g. 60 seconds). So in this
patchset, we implement a new hot page identification algorithm based on
the latency between NUMA balancing page table scanning and hint page
fault. Which is a kind of mostly frequently accessed (MFU) algorithm.

In NUMA balancing memory tiering mode, if there are hot pages in slow
memory node and cold pages in fast memory node, we need to promote/demote
hot/cold pages between the fast and cold memory nodes.

A choice is to promote/demote as fast as possible. But the CPU cycles and
memory bandwidth consumed by the high promoting/demoting throughput will
hurt the latency of some workload because of accessing inflating and slow
memory bandwidth contention.

A way to resolve this issue is to restrict the max promoting/demoting
throughput. It will take longer to finish the promoting/demoting. But
the workload latency will be better. This is implemented in this patchset
as the page promotion rate limit mechanism.

The promotion hot threshold is workload and system configuration
dependent. So in this patchset, a method to adjust the hot threshold
automatically is implemented. The basic idea is to control the number of
the candidate promotion pages to match the promotion rate limit.

We used the pmbench memory accessing benchmark tested the patchset on a
2-socket server system with DRAM and PMEM installed. The test results are
as follows,

pmbench score promote rate
(accesses/s) MB/s
------------- ------------
base 146887704.1 725.6
hot selection 165695601.2 544.0
rate limit 162814569.8 165.2
auto adjustment 170495294.0 136.9

From the results above,

With hot page selection patch [1/3], the pmbench score increases about
12.8%, and promote rate (overhead) decreases about 25.0%, compared with
base kernel.

With rate limit patch [2/3], pmbench score decreases about 1.7%, and
promote rate decreases about 69.6%, compared with hot page selection
patch.

With threshold auto adjustment patch [3/3], pmbench score increases about
4.7%, and promote rate decrease about 17.1%, compared with rate limit
patch.

Baolin helped to test the patchset with MySQL on a machine which contains
1 DRAM node (30G) and 1 PMEM node (126G).

sysbench /usr/share/sysbench/oltp_read_write.lua \
......
--tables=200 \
--table-size=1000000 \
--report-interval=10 \
--threads=16 \
--time=120

The tps can be improved about 5%.


This patch (of 3):

To optimize page placement in a memory tiering system with NUMA balancing,
the hot pages in the slow memory node need to be identified. Essentially,
the original NUMA balancing implementation selects the mostly recently
accessed (MRU) pages to promote. But this isn't a perfect algorithm to
identify the hot pages. Because the pages with quite low access frequency
may be accessed eventually given the NUMA balancing page table scanning
period could be quite long (e.g. 60 seconds). The most frequently
accessed (MFU) algorithm is better.

So, in this patch we implemented a better hot page selection algorithm.
Which is based on NUMA balancing page table scanning and hint page fault
as follows,

- When the page tables of the processes are scanned to change PTE/PMD
to be PROT_NONE, the current time is recorded in struct page as scan
time.

- When the page is accessed, hint page fault will occur. The scan
time is gotten from the struct page. And The hint page fault
latency is defined as

hint page fault time - scan time

The shorter the hint page fault latency of a page is, the higher the
probability of their access frequency to be higher. So the hint page
fault latency is a better estimation of the page hot/cold.

It's hard to find some extra space in struct page to hold the scan time.
Fortunately, we can reuse some bits used by the original NUMA balancing.

NUMA balancing uses some bits in struct page to store the page accessing
CPU and PID (referring to page_cpupid_xchg_last()). Which is used by the
multi-stage node selection algorithm to avoid to migrate pages shared
accessed by the NUMA nodes back and forth. But for pages in the slow
memory node, even if they are shared accessed by multiple NUMA nodes, as
long as the pages are hot, they need to be promoted to the fast memory
node. So the accessing CPU and PID information are unnecessary for the
slow memory pages. We can reuse these bits in struct page to record the
scan time. For the fast memory pages, these bits are used as before.

For the hot threshold, the default value is 1 second, which works well in
our performance test. All pages with hint page fault latency < hot
threshold will be considered hot.

It's hard for users to determine the hot threshold. So we don't provide a
kernel ABI to set it, just provide a debugfs interface for advanced users
to experiment. We will continue to work on a hot threshold automatic
adjustment mechanism.

The downside of the above method is that the response time to the workload
hot spot changing may be much longer. For example,

- A previous cold memory area becomes hot

- The hint page fault will be triggered. But the hint page fault
latency isn't shorter than the hot threshold. So the pages will
not be promoted.

- When the memory area is scanned again, maybe after a scan period,
the hint page fault latency measured will be shorter than the hot
threshold and the pages will be promoted.

To mitigate this, if there are enough free space in the fast memory node,
the hot threshold will not be used, all pages will be promoted upon the
hint page fault for fast response.

Thanks Zhong Jiang reported and tested the fix for a bug when disabling
memory tiering mode dynamically.

Link: https://lkml.kernel.org/r/20220713083954.34196-1-ying.huang@intel.com
Link: https://lkml.kernel.org/r/20220713083954.34196-2-ying.huang@intel.com
Signed-off-by: "Huang, Ying" <ying.huang@intel.com>
Reviewed-by: Baolin Wang <baolin.wang@linux.alibaba.com>
Tested-by: Baolin Wang <baolin.wang@linux.alibaba.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Rik van Riel <riel@surriel.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Yang Shi <shy828301@gmail.com>
Cc: Zi Yan <ziy@nvidia.com>
Cc: Wei Xu <weixugc@google.com>
Cc: osalvador <osalvador@suse.de>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: Zhong Jiang <zhongjiang-ali@linux.alibaba.com>
Cc: Oscar Salvador <osalvador@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

sched: Rename task_running() to task_on_cpu()

There is some ambiguity about task_running() in that it is unrelated
to TASK_RUNNING but instead tests ->on_cpu. As such, rename the thing
task_on_cpu().

Suggested-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/Yxhkhn55uHZx+NGl@hirez.programming.kicks-ass.net

sched/fair: Cleanup for SIS_PROP

The sched-domain of this cpu is only used for some heuristics when
SIS_PROP is enabled, and it should be irrelevant whether the local
sd_llc is valid or not, since all we care about is target sd_llc
if !SIS_PROP.

Access the local domain only when there is a need.

Signed-off-by: Abel Wu <wuyun.abel@bytedance.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Mel Gorman <mgorman@suse.de>
Link: https://lore.kernel.org/r/20220907112000.1854-6-wuyun.abel@bytedance.com

sched/fair: Default to false in test_idle_cores()

It's uncertain whether idle cores exist or not if shared sched-
domains are not ready, so returning "no idle cores" usually
makes sense.

While __update_idle_core() is an exception, it checks status
of this core and set hint to shared sched-domain if necessary.
So the whole logic of this function depends on the existence
of shared sched-domain, and can certainly bail out early if
it is not available.

It's somehow a little tricky, and as Josh suggested that it
should be transient while the domain isn't ready. So remove
the self-defined default value to make things more clearer.

Signed-off-by: Abel Wu <wuyun.abel@bytedance.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Josh Don <joshdon@google.com>
Acked-by: Mel Gorman <mgorman@techsingularity.net>
Link: https://lore.kernel.org/r/20220907112000.1854-5-wuyun.abel@bytedance.com

sched/fair: Remove useless check in select_idle_core()

The function select_idle_core() only gets called when has_idle_cores
is true which can be possible only when sched_smt_present is enabled.

This change also aligns select_idle_core() with select_idle_smt() in
the way that the caller do the check if necessary.

Signed-off-by: Abel Wu <wuyun.abel@bytedance.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Mel Gorman <mgorman@techsingularity.net>
Link: https://lore.kernel.org/r/20220907112000.1854-4-wuyun.abel@bytedance.com

sched/fair: Avoid double search on same cpu

The prev cpu is checked at the beginning of SIS, and it's unlikely
to be idle before the second check in select_idle_smt(). So we'd
better focus on its SMT siblings.

Signed-off-by: Abel Wu <wuyun.abel@bytedance.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Josh Don <joshdon@google.com>
Acked-by: Mel Gorman <mgorman@techsingularity.net>
Link: https://lore.kernel.org/r/20220907112000.1854-3-wuyun.abel@bytedance.com

sched/fair: Remove redundant check in select_idle_smt()

If two cpus share LLC cache, then the two cores they belong to
are also in the same LLC domain.

Signed-off-by: Abel Wu <wuyun.abel@bytedance.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Josh Don <joshdon@google.com>
Acked-by: Mel Gorman <mgorman@techsingularity.net>
Link: https://lore.kernel.org/r/20220907112000.1854-2-wuyun.abel@bytedance.com

sched/fair: Don't init util/runnable_avg for !fair task

post_init_entity_util_avg() init task util_avg according to the cpu util_avg
at the time of fork, which will decay when switched_to_fair() some time later,
we'd better to not set them at all in the case of !fair task.

Suggested-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Chengming Zhou <zhouchengming@bytedance.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20220818124805.601-10-zhouchengming@bytedance.com

sched/fair: Move task sched_avg attach to enqueue_task_fair()

When wake_up_new_task(), we use post_init_entity_util_avg() to init
util_avg/runnable_avg based on cpu's util_avg at that time, and
attach task sched_avg to cfs_rq.

Since enqueue_task_fair() -> enqueue_entity() -> update_load_avg()
loop will do attach, we can move this work to update_load_avg().

wake_up_new_task(p)
post_init_entity_util_avg(p)
attach_entity_cfs_rq() --> (1)
activate_task(rq, p)
enqueue_task() := enqueue_task_fair()
enqueue_entity() loop
update_load_avg(cfs_rq, se, UPDATE_TG | DO_ATTACH)
if (!se->avg.last_update_time && (flags & DO_ATTACH))
attach_entity_load_avg() --> (2)

This patch move attach from (1) to (2), update related comments too.

Signed-off-by: Chengming Zhou <zhouchengming@bytedance.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20220818124805.601-9-zhouchengming@bytedance.com

sched/fair: Allow changing cgroup of new forked task

commit 7dc603c9028e ("sched/fair: Fix PELT integrity for new tasks")
introduce a TASK_NEW state and an unnessary limitation that would fail
when changing cgroup of new forked task.

Because at that time, we can't handle task_change_group_fair() for new
forked fair task which hasn't been woken up by wake_up_new_task(),
which will cause detach on an unattached task sched_avg problem.

This patch delete this unnessary limitation by adding check before do
detach or attach in task_change_group_fair().

So cpu_cgrp_subsys.can_attach() has nothing to do for fair tasks,
only define it in #ifdef CONFIG_RT_GROUP_SCHED.

Signed-off-by: Chengming Zhou <zhouchengming@bytedance.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20220818124805.601-8-zhouchengming@bytedance.com

sched/fair: Fix another detach on unattached task corner case

commit 7dc603c9028e ("sched/fair: Fix PELT integrity for new tasks")
fixed two load tracking problems for new task, including detach on
unattached new task problem.

There still left another detach on unattached task problem for the task
which has been woken up by try_to_wake_up() and waiting for actually
being woken up by sched_ttwu_pending().

try_to_wake_up(p)
cpu = select_task_rq(p)
if (task_cpu(p) != cpu)
set_task_cpu(p, cpu)
migrate_task_rq_fair()
remove_entity_load_avg() --> unattached
se->avg.last_update_time = 0;
__set_task_cpu()
ttwu_queue(p, cpu)
ttwu_queue_wakelist()
__ttwu_queue_wakelist()

task_change_group_fair()
detach_task_cfs_rq()
detach_entity_cfs_rq()
detach_entity_load_avg() --> detach on unattached task
set_task_rq()
attach_task_cfs_rq()
attach_entity_cfs_rq()
attach_entity_load_avg()

The reason of this problem is similar, we should check in detach_entity_cfs_rq()
that se->avg.last_update_time != 0, before do detach_entity_load_avg().

Signed-off-by: Chengming Zhou <zhouchengming@bytedance.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20220818124805.601-7-zhouchengming@bytedance.com

sched/fair: Combine detach into dequeue when migrating task

When we are migrating task out of the CPU, we can combine detach and
propagation into dequeue_entity() to save the detach_entity_cfs_rq()
in migrate_task_rq_fair().

This optimization is like combining DO_ATTACH in the enqueue_entity()
when migrating task to the CPU. So we don't have to traverse the CFS tree
extra time to do the detach_entity_cfs_rq() -> propagate_entity_cfs_rq(),
which wouldn't be called anymore with this patch's change.

detach_task()
deactivate_task()
dequeue_task_fair()
for_each_sched_entity(se)
dequeue_entity()
update_load_avg() /* (1) */
detach_entity_load_avg()

set_task_cpu()
migrate_task_rq_fair()
detach_entity_cfs_rq() /* (2) */
update_load_avg();
detach_entity_load_avg();
propagate_entity_cfs_rq();
for_each_sched_entity()
update_load_avg()

This patch save the detach_entity_cfs_rq() called in (2) by doing
the detach_entity_load_avg() for a CPU migrating task inside (1)
(the task being the first se in the loop)

Signed-off-by: Chengming Zhou <zhouchengming@bytedance.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20220818124805.601-6-zhouchengming@bytedance.com

sched/fair: Update comments in enqueue/dequeue_entity()

When reading the sched_avg related code, I found the comments in
enqueue/dequeue_entity() are not updated with the current code.

We don't add/subtract entity's runnable_avg from cfs_rq->runnable_avg
during enqueue/dequeue_entity(), those are done only for attach/detach.

This patch updates the comments to reflect the current code working.

Signed-off-by: Chengming Zhou <zhouchengming@bytedance.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20220818124805.601-5-zhouchengming@bytedance.com

sched/fair: Reset sched_avg last_update_time before set_task_rq()

set_task_rq() -> set_task_rq_fair() will try to synchronize the blocked
task's sched_avg when migrate, which is not needed for already detached
task.

task_change_group_fair() will detached the task sched_avg from prev cfs_rq
first, so reset sched_avg last_update_time before set_task_rq() to avoid that.

Signed-off-by: Chengming Zhou <zhouchengming@bytedance.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20220818124805.601-4-zhouchengming@bytedance.com

sched/fair: Remove redundant cpu_cgrp_subsys->fork()

We use cpu_cgrp_subsys->fork() to set task group for the new fair task
in cgroup_post_fork().

Since commit b1e8206582f9 ("sched: Fix yet more sched_fork() races")
has already set_task_rq() for the new fair task in sched_cgroup_fork(),
so cpu_cgrp_subsys->fork() can be removed.

cgroup_can_fork() --> pin parent's sched_task_group
sched_cgroup_fork()
__set_task_cpu()
set_task_rq()
cgroup_post_fork()
ss->fork() := cpu_cgroup_fork()
sched_change_group(..., TASK_SET_GROUP)
task_set_group_fair()
set_task_rq() --> can be removed

After this patch's change, task_change_group_fair() only need to
care about task cgroup migration, make the code much simplier.

Signed-off-by: Chengming Zhou <zhouchengming@bytedance.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Link: https://lore.kernel.org/r/20220818124805.601-3-zhouchengming@bytedance.com