linux-stable.git/kernel/sched/sched.h, branch linux-5.2.y

sched/core: Annotate perf_domain pointer with __rcu

2019-04-03T10:34:31+00:00

This fixes the following sparse errors in sched/fair.c:

  fair.c:6506:14: error: incompatible types in comparison expression
  fair.c:8642:21: error: incompatible types in comparison expression

Using __rcu will also help sparse catch any future bugs.

Signed-off-by: Joel Fernandes (Google) 
Signed-off-by: Peter Zijlstra (Intel) 
[ From an RCU perspective. ]
Reviewed-by: Paul E. McKenney 
Cc: Josh Triplett 
Cc: Lai Jiangshan 
Cc: Linus Torvalds 
Cc: Luc Van Oostenryck 
Cc: Mathieu Desnoyers 
Cc: Mike Galbraith 
Cc: Morten Rasmussen 
Cc: Peter Zijlstra 
Cc: Steven Rostedt 
Cc: Thomas Gleixner 
Cc: keescook@chromium.org
Cc: kernel-hardening@lists.openwall.com
Cc: kernel-team@android.com
Link: https://lkml.kernel.org/r/20190321003426.160260-5-joel@joelfernandes.org
Signed-off-by: Ingo Molnar

sched_domain: Annotate RCU pointers properly

2019-04-03T10:34:31+00:00

The scheduler uses RCU API in various places to access sched_domain
pointers. These cause sparse errors as below.

Many new errors show up because of an annotation check I added to
rcu_assign_pointer(). Let us annotate the pointers correctly which also
will help sparse catch any potential future bugs.

This fixes the following sparse errors:

  rt.c:1681:9: error: incompatible types in comparison expression
  deadline.c:1904:9: error: incompatible types in comparison expression
  core.c:519:9: error: incompatible types in comparison expression
  core.c:1634:17: error: incompatible types in comparison expression
  fair.c:6193:14: error: incompatible types in comparison expression
  fair.c:9883:22: error: incompatible types in comparison expression
  fair.c:9897:9: error: incompatible types in comparison expression
  sched.h:1287:9: error: incompatible types in comparison expression
  topology.c:612:9: error: incompatible types in comparison expression
  topology.c:615:9: error: incompatible types in comparison expression
  sched.h:1300:9: error: incompatible types in comparison expression
  topology.c:618:9: error: incompatible types in comparison expression
  sched.h:1287:9: error: incompatible types in comparison expression
  topology.c:621:9: error: incompatible types in comparison expression
  sched.h:1300:9: error: incompatible types in comparison expression
  topology.c:624:9: error: incompatible types in comparison expression
  topology.c:671:9: error: incompatible types in comparison expression
  stats.c:45:17: error: incompatible types in comparison expression
  fair.c:5998:15: error: incompatible types in comparison expression
  fair.c:5989:15: error: incompatible types in comparison expression
  fair.c:5998:15: error: incompatible types in comparison expression
  fair.c:5989:15: error: incompatible types in comparison expression
  fair.c:6120:19: error: incompatible types in comparison expression
  fair.c:6506:14: error: incompatible types in comparison expression
  fair.c:6515:14: error: incompatible types in comparison expression
  fair.c:6623:9: error: incompatible types in comparison expression
  fair.c:5970:17: error: incompatible types in comparison expression
  fair.c:8642:21: error: incompatible types in comparison expression
  fair.c:9253:9: error: incompatible types in comparison expression
  fair.c:9331:9: error: incompatible types in comparison expression
  fair.c:9519:15: error: incompatible types in comparison expression
  fair.c:9533:14: error: incompatible types in comparison expression
  fair.c:9542:14: error: incompatible types in comparison expression
  fair.c:9567:14: error: incompatible types in comparison expression
  fair.c:9597:14: error: incompatible types in comparison expression
  fair.c:9421:16: error: incompatible types in comparison expression
  fair.c:9421:16: error: incompatible types in comparison expression

Signed-off-by: Joel Fernandes (Google) 
Signed-off-by: Peter Zijlstra (Intel) 
[ From an RCU perspective. ]
Reviewed-by: Paul E. McKenney 
Cc: Josh Triplett 
Cc: Lai Jiangshan 
Cc: Linus Torvalds 
Cc: Luc Van Oostenryck 
Cc: Mathieu Desnoyers 
Cc: Mike Galbraith 
Cc: Morten Rasmussen 
Cc: Peter Zijlstra 
Cc: Steven Rostedt 
Cc: Thomas Gleixner 
Cc: keescook@chromium.org
Cc: kernel-hardening@lists.openwall.com
Cc: kernel-team@android.com
Link: https://lkml.kernel.org/r/20190321003426.160260-3-joel@joelfernandes.org
Signed-off-by: Ingo Molnar

sched/cpufreq: Annotate cpufreq_update_util_data pointer with __rcu

2019-04-03T10:34:31+00:00

Recently I added an RCU annotation check to rcu_assign_pointer(). All
pointers assigned to RCU protected data are to be annotated with __rcu
inorder to be able to use rcu_assign_pointer() similar to checks in
other RCU APIs.

This resulted in a sparse error:

  kernel//sched/cpufreq.c:41:9: sparse: error: incompatible types in comparison expression (different address spaces)

Fix this by annotating cpufreq_update_util_data pointer with __rcu. This
will also help sparse catch any future RCU misuage bugs.

Signed-off-by: Joel Fernandes (Google) 
[ From an RCU perspective. ]
Signed-off-by: Peter Zijlstra (Intel) 
Reviewed-by: Paul E. McKenney 
Cc: Josh Triplett 
Cc: Lai Jiangshan 
Cc: Linus Torvalds 
Cc: Luc Van Oostenryck 
Cc: Mathieu Desnoyers 
Cc: Mike Galbraith 
Cc: Morten Rasmussen 
Cc: Peter Zijlstra 
Cc: Steven Rostedt 
Cc: Thomas Gleixner 
Cc: keescook@chromium.org
Cc: kernel-hardening@lists.openwall.com
Cc: kernel-team@android.com
Link: https://lkml.kernel.org/r/20190321003426.160260-2-joel@joelfernandes.org
Signed-off-by: Ingo Molnar

Merge branch 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

2019-03-06T16:14:05+00:00

Pull scheduler updates from Ingo Molnar:
 "The main changes in this cycle were:

   - refcount conversions

   - Solve the rq->leaf_cfs_rq_list can of worms for real.

   - improve power-aware scheduling

   - add sysctl knob for Energy Aware Scheduling

   - documentation updates

   - misc other changes"

* 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (34 commits)
  kthread: Do not use TIMER_IRQSAFE
  kthread: Convert worker lock to raw spinlock
  sched/fair: Use non-atomic cpumask_{set,clear}_cpu()
  sched/fair: Remove unused 'sd' parameter from select_idle_smt()
  sched/wait: Use freezable_schedule() when possible
  sched/fair: Prune, fix and simplify the nohz_balancer_kick() comment block
  sched/fair: Explain LLC nohz kick condition
  sched/fair: Simplify nohz_balancer_kick()
  sched/topology: Fix percpu data types in struct sd_data & struct s_data
  sched/fair: Simplify post_init_entity_util_avg() by calling it with a task_struct pointer argument
  sched/fair: Fix O(nr_cgroups) in the load balancing path
  sched/fair: Optimize update_blocked_averages()
  sched/fair: Fix insertion in rq->leaf_cfs_rq_list
  sched/fair: Add tmp_alone_branch assertion
  sched/core: Use READ_ONCE()/WRITE_ONCE() in move_queued_task()/task_rq_lock()
  sched/debug: Initialize sd_sysctl_cpus if !CONFIG_CPUMASK_OFFSTACK
  sched/pelt: Skip updating util_est when utilization is higher than CPU's capacity
  sched/fair: Update scale invariance of PELT
  sched/fair: Move the rq_of() helper function
  sched/core: Convert task_struct.stack_refcount to refcount_t
  ...

sched/fair: Simplify post_init_entity_util_avg() by calling it with a task_struct pointer argument

2019-02-11T07:02:14+00:00

Since commit:

  d03266910a53 ("sched/fair: Fix task group initialization")

the utilization of a sched entity representing a task group is no longer
initialized to any other value than 0. So post_init_entity_util_avg() is
only used for tasks, not for sched_entities.

Make this clear by calling it with a task_struct pointer argument which
also eliminates the entity_is_task(se) if condition in the fork path and
get rid of the stale comment in remove_entity_load_avg() accordingly.

Signed-off-by: Dietmar Eggemann 
Signed-off-by: Peter Zijlstra (Intel) 
Cc: Linus Torvalds 
Cc: Morten Rasmussen 
Cc: Patrick Bellasi 
Cc: Peter Zijlstra 
Cc: Quentin Perret 
Cc: Thomas Gleixner 
Cc: Valentin Schneider 
Cc: Vincent Guittot 
Link: https://lkml.kernel.org/r/20190122162501.12000-1-dietmar.eggemann@arm.com
Signed-off-by: Ingo Molnar

sched/core: Use READ_ONCE()/WRITE_ONCE() in move_queued_task()/task_rq_lock()

2019-02-04T08:13:21+00:00

move_queued_task() synchronizes with task_rq_lock() as follows:

	move_queued_task()		task_rq_lock()

	[S] ->on_rq = MIGRATING		[L] rq = task_rq()
	WMB (__set_task_cpu())		ACQUIRE (rq->lock);
	[S] ->cpu = new_cpu		[L] ->on_rq

where "[L] rq = task_rq()" is ordered before "ACQUIRE (rq->lock)" by an
address dependency and, in turn, "ACQUIRE (rq->lock)" is ordered before
"[L] ->on_rq" by the ACQUIRE itself.

Use READ_ONCE() to load ->cpu in task_rq() (c.f., task_cpu()) to honor
this address dependency.  Also, mark the accesses to ->cpu and ->on_rq
with READ_ONCE()/WRITE_ONCE() to comply with the LKMM.

Signed-off-by: Andrea Parri 
Signed-off-by: Peter Zijlstra (Intel) 
Cc: Alan Stern 
Cc: Linus Torvalds 
Cc: Mike Galbraith 
Cc: Paul E. McKenney 
Cc: Peter Zijlstra 
Cc: Thomas Gleixner 
Cc: Will Deacon 
Link: https://lkml.kernel.org/r/20190121155240.27173-1-andrea.parri@amarulasolutions.com
Signed-off-by: Ingo Molnar

sched/pelt: Skip updating util_est when utilization is higher than CPU's capacity

2019-02-04T08:13:21+00:00

util_est is mainly meant to be a lower-bound for tasks utilization.
That's why task_util_est() returns the actual util_avg when it's higher
than the estimated utilization.

With new invaraince signal and without any special check on samples
collection, if a task is limited because of thermal capping for
example, we could end up overestimating its utilization and thus
perhaps generating an unwanted frequency spike when the capping is
relaxed... and (even worst) it will take some more activations for the
estimated utilization to converge back to the actual utilization.

Since we cannot easily know if there is idle time in a CPU when a task
completes an activation with a utilization higher then the CPU capacity,
we skip the sampling when utilization is higher than CPU's capacity.

Suggested-by: Patrick Bellasi 
Signed-off-by: Vincent Guittot 
Signed-off-by: Peter Zijlstra (Intel) 
Cc: Linus Torvalds 
Cc: Mike Galbraith 
Cc: Morten.Rasmussen@arm.com
Cc: Peter Zijlstra 
Cc: Thomas Gleixner 
Cc: bsegall@google.com
Cc: dietmar.eggemann@arm.com
Cc: pjt@google.com
Cc: pkondeti@codeaurora.org
Cc: quentin.perret@arm.com
Cc: rjw@rjwysocki.net
Cc: srinivas.pandruvada@linux.intel.com
Cc: thara.gopinath@linaro.org
Link: https://lkml.kernel.org/r/1548257214-13745-4-git-send-email-vincent.guittot@linaro.org
Signed-off-by: Ingo Molnar

sched/fair: Update scale invariance of PELT

2019-02-04T08:13:21+00:00

The current implementation of load tracking invariance scales the
contribution with current frequency and uarch performance (only for
utilization) of the CPU. One main result of this formula is that the
figures are capped by current capacity of CPU. Another one is that the
load_avg is not invariant because not scaled with uarch.

The util_avg of a periodic task that runs r time slots every p time slots
varies in the range :

    U * (1-y^r)/(1-y^p) * y^i < Utilization < U * (1-y^r)/(1-y^p)

with U is the max util_avg value = SCHED_CAPACITY_SCALE

At a lower capacity, the range becomes:

    U * C * (1-y^r')/(1-y^p) * y^i' < Utilization <  U * C * (1-y^r')/(1-y^p)

with C reflecting the compute capacity ratio between current capacity and
max capacity.

so C tries to compensate changes in (1-y^r') but it can't be accurate.

Instead of scaling the contribution value of PELT algo, we should scale the
running time. The PELT signal aims to track the amount of computation of
tasks and/or rq so it seems more correct to scale the running time to
reflect the effective amount of computation done since the last update.

In order to be fully invariant, we need to apply the same amount of
running time and idle time whatever the current capacity. Because running
at lower capacity implies that the task will run longer, we have to ensure
that the same amount of idle time will be applied when system becomes idle
and no idle time has been "stolen". But reaching the maximum utilization
value (SCHED_CAPACITY_SCALE) means that the task is seen as an
always-running task whatever the capacity of the CPU (even at max compute
capacity). In this case, we can discard this "stolen" idle times which
becomes meaningless.

In order to achieve this time scaling, a new clock_pelt is created per rq.
The increase of this clock scales with current capacity when something
is running on rq and synchronizes with clock_task when rq is idle. With
this mechanism, we ensure the same running and idle time whatever the
current capacity. This also enables to simplify the pelt algorithm by
removing all references of uarch and frequency and applying the same
contribution to utilization and loads. Furthermore, the scaling is done
only once per update of clock (update_rq_clock_task()) instead of during
each update of sched_entities and cfs/rt/dl_rq of the rq like the current
implementation. This is interesting when cgroup are involved as shown in
the results below:

On a hikey (octo Arm64 platform).
Performance cpufreq governor and only shallowest c-state to remove variance
generated by those power features so we only track the impact of pelt algo.

each test runs 16 times:

	./perf bench sched pipe
	(higher is better)
	kernel	tip/sched/core     + patch
	        ops/seconds        ops/seconds         diff
	cgroup
	root    59652(+/- 0.18%)   59876(+/- 0.24%)    +0.38%
	level1  55608(+/- 0.27%)   55923(+/- 0.24%)    +0.57%
	level2  52115(+/- 0.29%)   52564(+/- 0.22%)    +0.86%

	hackbench -l 1000
	(lower is better)
	kernel	tip/sched/core     + patch
	        duration(sec)      duration(sec)        diff
	cgroup
	root    4.453(+/- 2.37%)   4.383(+/- 2.88%)     -1.57%
	level1  4.859(+/- 8.50%)   4.830(+/- 7.07%)     -0.60%
	level2  5.063(+/- 9.83%)   4.928(+/- 9.66%)     -2.66%

Then, the responsiveness of PELT is improved when CPU is not running at max
capacity with this new algorithm. I have put below some examples of
duration to reach some typical load values according to the capacity of the
CPU with current implementation and with this patch. These values has been
computed based on the geometric series and the half period value:

  Util (%)     max capacity  half capacity(mainline)  half capacity(w/ patch)
  972 (95%)    138ms         not reachable            276ms
  486 (47.5%)  30ms          138ms                     60ms
  256 (25%)    13ms           32ms                     26ms

On my hikey (octo Arm64 platform) with schedutil governor, the time to
reach max OPP when starting from a null utilization, decreases from 223ms
with current scale invariance down to 121ms with the new algorithm.

Signed-off-by: Vincent Guittot 
Signed-off-by: Peter Zijlstra (Intel) 
Cc: Linus Torvalds 
Cc: Mike Galbraith 
Cc: Morten.Rasmussen@arm.com
Cc: Peter Zijlstra 
Cc: Thomas Gleixner 
Cc: bsegall@google.com
Cc: dietmar.eggemann@arm.com
Cc: patrick.bellasi@arm.com
Cc: pjt@google.com
Cc: pkondeti@codeaurora.org
Cc: quentin.perret@arm.com
Cc: rjw@rjwysocki.net
Cc: srinivas.pandruvada@linux.intel.com
Cc: thara.gopinath@linaro.org
Link: https://lkml.kernel.org/r/1548257214-13745-3-git-send-email-vincent.guittot@linaro.org
Signed-off-by: Ingo Molnar

sched/fair: Move the rq_of() helper function

2019-02-04T08:13:21+00:00

Move rq_of() helper function so it can be used in pelt.c

[ mingo: Improve readability while at it. ]

Signed-off-by: Vincent Guittot 
Signed-off-by: Peter Zijlstra (Intel) 
Cc: Linus Torvalds 
Cc: Mike Galbraith 
Cc: Morten.Rasmussen@arm.com
Cc: Peter Zijlstra 
Cc: Thomas Gleixner 
Cc: bsegall@google.com
Cc: dietmar.eggemann@arm.com
Cc: patrick.bellasi@arm.com
Cc: pjt@google.com
Cc: pkondeti@codeaurora.org
Cc: quentin.perret@arm.com
Cc: rjw@rjwysocki.net
Cc: srinivas.pandruvada@linux.intel.com
Cc: thara.gopinath@linaro.org
Link: https://lkml.kernel.org/r/1548257214-13745-2-git-send-email-vincent.guittot@linaro.org
Signed-off-by: Ingo Molnar

sched/core: Give DCE a fighting chance

2019-01-27T11:29:37+00:00

All that fancy new Energy-Aware scheduling foo is hidden behind a
static_key, which is awesome if you have the stuff enabled in your
config.

However, when you lack all the prerequisites it doesn't make any sense
to pretend we'll ever actually run this, so provide a little more clue
to the compiler so it can more agressively delete the code.

   text    data     bss     dec     hex filename
  50297     976      96   51369    c8a9 defconfig-build/kernel/sched/fair.o
  49227     944      96   50267    c45b defconfig-build/kernel/sched/fair.o

Signed-off-by: Peter Zijlstra (Intel) 
Cc: Linus Torvalds 
Cc: Peter Zijlstra 
Cc: Thomas Gleixner 
Signed-off-by: Ingo Molnar