The scheduler IPI has grown weird and wonderful over the years, time
for spring cleaning.
Move all the non-trivial stuff out of it and into a regular smp function
call IPI. This then reduces the schedule_ipi() to most of it's former NOP
glory and ensures to keep the interrupt vector lean and mean.
Aside of that avoiding the full irq_enter() in the x86 IPI implementation
is incorrect as scheduler_ipi() can be instrumented. To work around that
scheduler_ipi() had an irq_enter/exit() hack when heavy work was
pending. This is gone now.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Alexandre Chartre <alexandre.chartre@oracle.com>
Link: https://lkml.kernel.org/r/20200505134058.361859938@linutronix.de
The code is executed with preemption(and interrupts) disabled,
so it's safe to use __this_cpu_write().
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20200421144123.33580-1-songmuchun@bytedance.com
Function sched_init_granularity() is only called from __init
functions, so mark it __init as well.
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
Link: https://lkml.kernel.org/r/20200406074750.56533-1-songmuchun@bytedance.com
In order to prevent possible hardlockup of sched_cfs_period_timer()
loop, loop count is introduced to denote whether to scale quota and
period or not. However, scale is done between forwarding period timer
and refilling cfs bandwidth runtime, which means that period timer is
forwarded with old "period" while runtime is refilled with scaled
"quota".
Move do_sched_cfs_period_timer() before scaling to solve this.
Fixes: 2e8e192263 ("sched/fair: Limit sched_cfs_period_timer() loop to avoid hard lockup")
Signed-off-by: Huaixin Chang <changhuaixin@linux.alibaba.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Ben Segall <bsegall@google.com>
Reviewed-by: Phil Auld <pauld@redhat.com>
Link: https://lkml.kernel.org/r/20200420024421.22442-3-changhuaixin@linux.alibaba.com
After Commit 6e2df0581f ("sched: Fix pick_next_task() vs 'change'
pattern race"), there is no need to expose newidle_balance() as it
is only used within fair.c file. Change this function back to static again.
No functional change.
Reported-by: kbuild test robot <lkp@intel.com>
Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/83cd3030b031ca5d646cd5e225be10e7a0fdd8f5.1587464698.git.yu.c.chen@intel.com
The SD_LOAD_BALANCE flag is set unconditionally for all domains in
sd_init(). By making the sched_domain->flags syctl interface read-only, we
have removed the last piece of code that could clear that flag - as such,
it will now be always present. Rather than to keep carrying it along, we
can work towards getting rid of it entirely.
cpusets don't need it because they can make CPUs be attached to the NULL
domain (e.g. cpuset with sched_load_balance=0), or to a partitioned
root_domain, i.e. a sched_domain hierarchy that doesn't span the entire
system (e.g. root cpuset with sched_load_balance=0 and sibling cpusets with
sched_load_balance=1).
isolcpus apply the same "trick": isolated CPUs are explicitly taken out of
the sched_domain rebuild (using housekeeping_cpumask()), so they get the
NULL domain treatment as well.
Remove the checks against SD_LOAD_BALANCE.
Signed-off-by: Valentin Schneider <valentin.schneider@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20200415210512.805-4-valentin.schneider@arm.com
The last use of that parameter was removed by commit
57abff067a ("sched/fair: Rework find_idlest_group()")
Get rid of the parameter.
Signed-off-by: Valentin Schneider <valentin.schneider@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://lkml.kernel.org/r/20200415210512.805-2-valentin.schneider@arm.com
We only consider group_balance_cpu() after there is no idle
cpu. So, just do comparison before return at these two cases.
Signed-off-by: Peng Wang <rocking@linux.alibaba.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Valentin Schneider <valentin.schneider@arm.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lkml.kernel.org/r/245c792f0e580b3ca342ad61257f4c066ee0f84f.1586594833.git.rocking@linux.alibaba.com
This is mostly a revert of commit:
baa9be4ffb ("sched/fair: Fix throttle_list starvation with low CFS quota")
The primary use of distribute_running was to determine whether to add
throttled entities to the head or the tail of the throttled list. Now
that we always add to the tail, we can remove this field.
The other use of distribute_running is in the slack_timer, so that we
don't start a distribution while one is already running. However, even
in the event that this race occurs, it is fine to have two distributions
running (especially now that distribute grabs the cfs_b->lock to
determine remaining quota before assigning).
Signed-off-by: Josh Don <joshdon@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Phil Auld <pauld@redhat.com>
Tested-by: Phil Auld <pauld@redhat.com>
Link: https://lkml.kernel.org/r/20200410225208.109717-3-joshdon@google.com
There is a race window in which an entity begins throttling before quota
is added to the pool, but does not finish throttling until after we have
finished with distribute_cfs_runtime(). This entity is not observed by
distribute_cfs_runtime() because it was not on the throttled list at the
time that distribution was running. This race manifests as rare
period-length statlls for such entities.
Rather than heavy-weight the synchronization with the progress of
distribution, we can fix this by aborting throttling if bandwidth has
become available. Otherwise, we immediately add the entity to the
throttled list so that it can be observed by a subsequent distribution.
Additionally, we can remove the case of adding the throttled entity to
the head of the throttled list, and simply always add to the tail.
Thanks to 26a8b12747, distribute_cfs_runtime() no longer holds onto
its own pool of runtime. This means that if we do hit the !assign and
distribute_running case, we know that distribution is about to end.
Signed-off-by: Paul Turner <pjt@google.com>
Signed-off-by: Ben Segall <bsegall@google.com>
Signed-off-by: Josh Don <joshdon@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Phil Auld <pauld@redhat.com>
Link: https://lkml.kernel.org/r/20200410225208.109717-2-joshdon@google.com
A negative imbalance value was observed after imbalance calculation,
this happens when the local sched group type is group_fully_busy,
and the average load of local group is greater than the selected
busiest group. Fix this problem by comparing the average load of the
local and busiest group before imbalance calculation formula.
Suggested-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Phil Auld <pauld@redhat.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Acked-by: Mel Gorman <mgorman@suse.de>
Signed-off-by: Aubrey Li <aubrey.li@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lkml.kernel.org/r/1585201349-70192-1-git-send-email-aubrey.li@intel.com
Currently, there is a potential race between distribute_cfs_runtime()
and assign_cfs_rq_runtime(). Race happens when cfs_b->runtime is read,
distributes without holding lock and finds out there is not enough
runtime to charge against after distribution. Because
assign_cfs_rq_runtime() might be called during distribution, and use
cfs_b->runtime at the same time.
Fibtest is the tool to test this race. Assume all gcfs_rq is throttled
and cfs period timer runs, slow threads might run and sleep, returning
unused cfs_rq runtime and keeping min_cfs_rq_runtime in their local
pool. If all this happens sufficiently quickly, cfs_b->runtime will drop
a lot. If runtime distributed is large too, over-use of runtime happens.
A runtime over-using by about 70 percent of quota is seen when we
test fibtest on a 96-core machine. We run fibtest with 1 fast thread and
95 slow threads in test group, configure 10ms quota for this group and
see the CPU usage of fibtest is 17.0%, which is far more than the
expected 10%.
On a smaller machine with 32 cores, we also run fibtest with 96
threads. CPU usage is more than 12%, which is also more than expected
10%. This shows that on similar workloads, this race do affect CPU
bandwidth control.
Solve this by holding lock inside distribute_cfs_runtime().
Fixes: c06f04c704 ("sched: Fix potential near-infinite distribute_cfs_runtime() loop")
Reviewed-by: Ben Segall <bsegall@google.com>
Signed-off-by: Huaixin Chang <changhuaixin@linux.alibaba.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/lkml/20200325092602.22471-1-changhuaixin@linux.alibaba.com/
sched/core.c uses update_avg() for rq->avg_idle and sched/fair.c uses an
open-coded version (with the exact same decay factor) for
rq->avg_scan_cost. On top of that, select_idle_cpu() expects to be able to
compare these two fields.
The only difference between the two is that rq->avg_scan_cost is computed
using a pure division rather than a shift. Turns out it actually matters,
first of all because the shifted value can be negative, and the standard
has this to say about it:
"""
The result of E1 >> E2 is E1 right-shifted E2 bit positions. [...] If E1
has a signed type and a negative value, the resulting value is
implementation-defined.
"""
Not only this, but (arithmetic) right shifting a negative value (using 2's
complement) is *not* equivalent to dividing it by the corresponding power
of 2. Let's look at a few examples:
-4 -> 0xF..FC
-4 >> 3 -> 0xF..FF == -1 != -4 / 8
-8 -> 0xF..F8
-8 >> 3 -> 0xF..FF == -1 == -8 / 8
-9 -> 0xF..F7
-9 >> 3 -> 0xF..FE == -2 != -9 / 8
Make update_avg() use a division, and export it to the private scheduler
header to reuse it where relevant. Note that this still lets compilers use
a shift here, but should prevent any unwanted surprise. The disassembly of
select_idle_cpu() remains unchanged on arm64, and ttwu_do_wakeup() gains 2
instructions; the diff sort of looks like this:
- sub x1, x1, x0
+ subs x1, x1, x0 // set condition codes
+ add x0, x1, #0x7
+ csel x0, x0, x1, mi // x0 = x1 < 0 ? x0 : x1
add x0, x3, x0, asr #3
which does the right thing (i.e. gives us the expected result while still
using an arithmetic shift)
Signed-off-by: Valentin Schneider <valentin.schneider@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lkml.kernel.org/r/20200330090127.16294-1-valentin.schneider@arm.com
In update_sg_wakeup_stats(), the comment says:
Computing avg_load makes sense only when group is fully
busy or overloaded.
But, the code below this comment does not check like this.
From reading the code about avg_load in other functions, I
confirm that avg_load should be calculated in fully busy or
overloaded case. The comment is correct and the checking
condition is wrong. So, change that condition.
Fixes: 57abff067a ("sched/fair: Rework find_idlest_group()")
Signed-off-by: Tao Zhou <ouwen210@hotmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Acked-by: Mel Gorman <mgorman@suse.de>
Link: https://lkml.kernel.org/r/Message-ID:
During load_balancing, a group with spare capacity will try to pull some
utilizations from an overloaded group. In such case, the load balance
looks for the runqueue with the highest utilization. Nevertheless, it
should also ensure that there are some pending tasks to pull otherwise
the load balance will fail to pull a task and the spread of the load will
be delayed.
This situation is quite transient but it's possible to highlight the
effect with a short run of sysbench test so the time to spread task impacts
the global result significantly.
Below are the average results for 15 iterations on an arm64 octo core:
sysbench --test=cpu --num-threads=8 --max-requests=1000 run
tip/sched/core +patchset
total time: 172ms 158ms
per-request statistics:
avg: 1.337ms 1.244ms
max: 21.191ms 10.753ms
The average max doesn't fully reflect the wide spread of the value which
ranges from 1.350ms to more than 41ms for the tip/sched/core and from
1.350ms to 21ms with the patch.
Other factors like waiting for an idle load balance or cache hotness
can delay the spreading of the tasks which explains why we can still
have up to 21ms with the patch.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20200312165429.990-1-vincent.guittot@linaro.org
When a cfs rq is throttled, the latter and its child are removed from the
leaf list but their nr_running is not changed which includes staying higher
than 1. When a task is enqueued in this throttled branch, the cfs rqs must
be added back in order to ensure correct ordering in the list but this can
only happens if nr_running == 1.
When cfs bandwidth is used, we call unconditionnaly list_add_leaf_cfs_rq()
when enqueuing an entity to make sure that the complete branch will be
added.
Similarly unthrottle_cfs_rq() can stop adding cfs in the list when a parent
is throttled. Iterate the remaining entity to ensure that the complete
branch will be added in the list.
Reported-by: Christian Borntraeger <borntraeger@de.ibm.com>
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: Christian Borntraeger <borntraeger@de.ibm.com>
Tested-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Cc: stable@vger.kernel.org
Cc: stable@vger.kernel.org #v5.1+
Link: https://lkml.kernel.org/r/20200306135257.25044-1-vincent.guittot@linaro.org
Even when a cgroup is throttled, the group se of a child cgroup can still
be enqueued and its gse->on_rq stays true. When a task is enqueued on such
child, we still have to update the load_avg and increase
h_nr_running of the throttled cfs. Nevertheless, the 1st
for_each_sched_entity() loop is skipped because of gse->on_rq == true and the
2nd loop because the cfs is throttled whereas we have to update both
load_avg with the old h_nr_running and increase h_nr_running in such case.
The same sequence can happen during dequeue when se moves to parent before
breaking in the 1st loop.
Note that the update of load_avg will effectively happen only once in order
to sync up to the throttled time. Next call for updating load_avg will stop
early because the clock stays unchanged.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Fixes: 6d4d22468d ("sched/fair: Reorder enqueue/dequeue_task_fair path")
Link: https://lkml.kernel.org/r/20200306084208.12583-1-vincent.guittot@linaro.org
When a cfs_rq is throttled, its group entity is dequeued and its running
tasks are removed. We must update runnable_avg with the old h_nr_running
and update group_se->runnable_weight with the new h_nr_running at each
level of the hierarchy.
Reviewed-by: Ben Segall <bsegall@google.com>
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Fixes: 9f68395333 ("sched/pelt: Add a new runnable average signal")
Link: https://lkml.kernel.org/r/20200227154115.8332-1-vincent.guittot@linaro.org
Qian Cai reported the following bug:
The linux-next commit ff7db0bf24 ("sched/numa: Prefer using an idle CPU as a
migration target instead of comparing tasks") introduced a boot warning,
[ 86.520534][ T1] WARNING: suspicious RCU usage
[ 86.520540][ T1] 5.6.0-rc3-next-20200227 #7 Not tainted
[ 86.520545][ T1] -----------------------------
[ 86.520551][ T1] kernel/sched/fair.c:5914 suspicious rcu_dereference_check() usage!
[ 86.520555][ T1]
[ 86.520555][ T1] other info that might help us debug this:
[ 86.520555][ T1]
[ 86.520561][ T1]
[ 86.520561][ T1] rcu_scheduler_active = 2, debug_locks = 1
[ 86.520567][ T1] 1 lock held by systemd/1:
[ 86.520571][ T1] #0: ffff8887f4b14848 (&mm->mmap_sem#2){++++}, at: do_page_fault+0x1d2/0x998
[ 86.520594][ T1]
[ 86.520594][ T1] stack backtrace:
[ 86.520602][ T1] CPU: 1 PID: 1 Comm: systemd Not tainted 5.6.0-rc3-next-20200227 #7
task_numa_migrate() checks for idle cores when updating NUMA-related statistics.
This relies on reading a RCU-protected structure in test_idle_cores() via this
call chain
task_numa_migrate
-> update_numa_stats
-> numa_idle_core
-> test_idle_cores
While the locking could be fine-grained, it is more appropriate to acquire
the RCU lock for the entire scan of the domain. This patch removes the
warning triggered at boot time.
Reported-by: Qian Cai <cai@lca.pw>
Reviewed-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Fixes: ff7db0bf24 ("sched/numa: Prefer using an idle CPU as a migration target instead of comparing tasks")
Link: https://lkml.kernel.org/r/20200227191804.GJ3818@techsingularity.net
Building against the tip/sched/core as ff7db0bf24 ("sched/numa: Prefer
using an idle CPU as a migration target instead of comparing tasks") with
the arm64 defconfig (which doesn't have CONFIG_SCHED_SMT set) leads to:
kernel/sched/fair.c:1525:20: warning: 'test_idle_cores' declared 'static' but never defined [-Wunused-function]
static inline bool test_idle_cores(int cpu, bool def);
^~~~~~~~~~~~~~~
Rather than define it in its own CONFIG_SCHED_SMT #define island, bunch it
up with test_idle_cores().
Reported-by: Anshuman Khandual <anshuman.khandual@arm.com>
Reported-by: Naresh Kamboju <naresh.kamboju@linaro.org>
Reviewed-by: Lukasz Luba <lukasz.luba@arm.com>
[mgorman@techsingularity.net: Edit changelog, minor style change]
Signed-off-by: Valentin Schneider <valentin.schneider@arm.com>
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Fixes: ff7db0bf24 ("sched/numa: Prefer using an idle CPU as a migration target instead of comparing tasks")
Link: https://lkml.kernel.org/r/20200303110258.1092-3-mgorman@techsingularity.net
Thermal pressure follows pelt signals which means the decay period for
thermal pressure is the default pelt decay period. Depending on SoC
characteristics and thermal activity, it might be beneficial to decay
thermal pressure slower, but still in-tune with the pelt signals. One way
to achieve this is to provide a command line parameter to set a decay
shift parameter to an integer between 0 and 10.
Signed-off-by: Thara Gopinath <thara.gopinath@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lkml.kernel.org/r/20200222005213.3873-10-thara.gopinath@linaro.org
cpu_capacity initially reflects the maximum possible capacity of a CPU.
Thermal pressure on a CPU means this maximum possible capacity is
unavailable due to thermal events. This patch subtracts the average
thermal pressure for a CPU from its maximum possible capacity so that
cpu_capacity reflects the remaining maximum capacity.
Signed-off-by: Thara Gopinath <thara.gopinath@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lkml.kernel.org/r/20200222005213.3873-8-thara.gopinath@linaro.org
Introduce support in scheduler periodic tick and other CFS bookkeeping
APIs to trigger the process of computing average thermal pressure for a
CPU. Also consider avg_thermal.load_avg in others_have_blocked which
allows for decay of pelt signals.
Signed-off-by: Thara Gopinath <thara.gopinath@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lkml.kernel.org/r/20200222005213.3873-7-thara.gopinath@linaro.org
sgs->group_weight is not set while gathering statistics in
update_sg_wakeup_stats(). This means that a group can be classified as
fully busy with 0 running tasks if utilization is high enough.
This path is mainly used for fork and exec.
Fixes: 57abff067a ("sched/fair: Rework find_idlest_group()")
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Mel Gorman <mgorman@techsingularity.net>
Link: https://lore.kernel.org/r/20200218144534.4564-1-vincent.guittot@linaro.org
When domains are imbalanced or overloaded a search of all CPUs on the
target domain is searched and compared with task_numa_compare. In some
circumstances, a candidate is found that is an obvious win.
o A task can move to an idle CPU and an idle CPU is found
o A swap candidate is found that would move to its preferred domain
This patch terminates the search when either condition is met.
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Valentin Schneider <valentin.schneider@arm.com>
Cc: Phil Auld <pauld@redhat.com>
Cc: Hillf Danton <hdanton@sina.com>
Link: https://lore.kernel.org/r/20200224095223.13361-14-mgorman@techsingularity.net
When swapping tasks for NUMA balancing, it is preferred that tasks move
to or remain on their preferred node. When considering an imbalance,
encourage tasks to move to their preferred node and discourage tasks from
moving away from their preferred node.
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Valentin Schneider <valentin.schneider@arm.com>
Cc: Phil Auld <pauld@redhat.com>
Cc: Hillf Danton <hdanton@sina.com>
Link: https://lore.kernel.org/r/20200224095223.13361-13-mgorman@techsingularity.net
Multiple tasks can attempt to select and idle CPU but fail because
numa_migrate_on is already set and the migration fails. Instead of failing,
scan for an alternative idle CPU. select_idle_sibling is not used because
it requires IRQs to be disabled and it ignores numa_migrate_on allowing
multiple tasks to stack. This scan may still fail if there are idle
candidate CPUs due to races but if this occurs, it's best that a task
stay on an available CPU that move to a contended one.
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Valentin Schneider <valentin.schneider@arm.com>
Cc: Phil Auld <pauld@redhat.com>
Cc: Hillf Danton <hdanton@sina.com>
Link: https://lore.kernel.org/r/20200224095223.13361-12-mgorman@techsingularity.net
task_numa_find_cpu() can scan a node multiple times. Minimally it scans to
gather statistics and later to find a suitable target. In some cases, the
second scan will simply pick an idle CPU if the load is not imbalanced.
This patch caches information on an idle core while gathering statistics
and uses it immediately if load is not imbalanced to avoid a second scan
of the node runqueues. Preference is given to an idle core rather than an
idle SMT sibling to avoid packing HT siblings due to linearly scanning the
node cpumask.
As a side-effect, even when the second scan is necessary, the importance
of using select_idle_sibling is much reduced because information on idle
CPUs is cached and can be reused.
Note that this patch actually makes is harder to move to an idle CPU
as multiple tasks can race for the same idle CPU due to a race checking
numa_migrate_on. This is addressed in the next patch.
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Valentin Schneider <valentin.schneider@arm.com>
Cc: Phil Auld <pauld@redhat.com>
Cc: Hillf Danton <hdanton@sina.com>
Link: https://lore.kernel.org/r/20200224095223.13361-11-mgorman@techsingularity.net
Take into account the new runnable_avg signal to classify a group and to
mitigate the volatility of util_avg in face of intensive migration or
new task with random utilization.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: "Dietmar Eggemann <dietmar.eggemann@arm.com>"
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Valentin Schneider <valentin.schneider@arm.com>
Cc: Phil Auld <pauld@redhat.com>
Cc: Hillf Danton <hdanton@sina.com>
Link: https://lore.kernel.org/r/20200224095223.13361-10-mgorman@techsingularity.net
Now that runnable_load_avg has been removed, we can replace it by a new
signal that will highlight the runnable pressure on a cfs_rq. This signal
track the waiting time of tasks on rq and can help to better define the
state of rqs.
At now, only util_avg is used to define the state of a rq:
A rq with more that around 80% of utilization and more than 1 tasks is
considered as overloaded.
But the util_avg signal of a rq can become temporaly low after that a task
migrated onto another rq which can bias the classification of the rq.
When tasks compete for the same rq, their runnable average signal will be
higher than util_avg as it will include the waiting time and we can use
this signal to better classify cfs_rqs.
The new runnable_avg will track the runnable time of a task which simply
adds the waiting time to the running time. The runnable _avg of cfs_rq
will be the /Sum of se's runnable_avg and the runnable_avg of group entity
will follow the one of the rq similarly to util_avg.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: "Dietmar Eggemann <dietmar.eggemann@arm.com>"
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Valentin Schneider <valentin.schneider@arm.com>
Cc: Phil Auld <pauld@redhat.com>
Cc: Hillf Danton <hdanton@sina.com>
Link: https://lore.kernel.org/r/20200224095223.13361-9-mgorman@techsingularity.net
Now that runnable_load_avg is no more used, we can remove it to make
space for a new signal.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: "Dietmar Eggemann <dietmar.eggemann@arm.com>"
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Valentin Schneider <valentin.schneider@arm.com>
Cc: Phil Auld <pauld@redhat.com>
Cc: Hillf Danton <hdanton@sina.com>
Link: https://lore.kernel.org/r/20200224095223.13361-8-mgorman@techsingularity.net
The standard load balancer generally tries to keep the number of running
tasks or idle CPUs balanced between NUMA domains. The NUMA balancer allows
tasks to move if there is spare capacity but this causes a conflict and
utilisation between NUMA nodes gets badly skewed. This patch uses similar
logic between the NUMA balancer and load balancer when deciding if a task
migrating to its preferred node can use an idle CPU.
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>
Cc: Valentin Schneider <valentin.schneider@arm.com>
Cc: Phil Auld <pauld@redhat.com>
Cc: Hillf Danton <hdanton@sina.com>
Link: https://lore.kernel.org/r/20200224095223.13361-7-mgorman@techsingularity.net
Similarly to what has been done for the normal load balancer, we can
replace runnable_load_avg by load_avg in numa load balancing and track the
other statistics like the utilization and the number of running tasks to
get to better view of the current state of a node.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: "Dietmar Eggemann <dietmar.eggemann@arm.com>"
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Valentin Schneider <valentin.schneider@arm.com>
Cc: Phil Auld <pauld@redhat.com>
Cc: Hillf Danton <hdanton@sina.com>
Link: https://lore.kernel.org/r/20200224095223.13361-6-mgorman@techsingularity.net
The walk through the cgroup hierarchy during the enqueue/dequeue of a task
is split in 2 distinct parts for throttled cfs_rq without any added value
but making code less readable.
Change the code ordering such that everything related to a cfs_rq
(throttled or not) will be done in the same loop.
In addition, the same steps ordering is used when updating a cfs_rq:
- update_load_avg
- update_cfs_group
- update *h_nr_running
This reordering enables the use of h_nr_running in PELT algorithm.
No functional and performance changes are expected and have been noticed
during tests.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: "Dietmar Eggemann <dietmar.eggemann@arm.com>"
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Valentin Schneider <valentin.schneider@arm.com>
Cc: Phil Auld <pauld@redhat.com>
Cc: Hillf Danton <hdanton@sina.com>
Link: https://lore.kernel.org/r/20200224095223.13361-5-mgorman@techsingularity.net
sched:sched_stick_numa is meant to fire when a task is unable to migrate
to the preferred node but from the trace, it's possibile to tell the
difference between "no CPU found", "migration to idle CPU failed" and
"tasks could not be swapped". Extend the tracepoint accordingly.
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
[ Minor edits. ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>
Cc: Valentin Schneider <valentin.schneider@arm.com>
Cc: Phil Auld <pauld@redhat.com>
Cc: Hillf Danton <hdanton@sina.com>
Link: https://lore.kernel.org/r/20200224095223.13361-4-mgorman@techsingularity.net
sched:sched_stick_numa is meant to fire when a task is unable to migrate
to the preferred node. The case where no candidate CPU could be found is
not traced which is an important gap. The tracepoint is not fired when
the task is not allowed to run on any CPU on the preferred node or the
task is already running on the target CPU but neither are interesting
corner cases.
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>
Cc: Valentin Schneider <valentin.schneider@arm.com>
Cc: Phil Auld <pauld@redhat.com>
Cc: Hillf Danton <hdanton@sina.com>
Link: https://lore.kernel.org/r/20200224095223.13361-3-mgorman@techsingularity.net
Capacity-awareness in the wake-up path previously involved disabling
wake_affine in certain scenarios. We have just made select_idle_sibling()
capacity-aware, so this isn't needed anymore.
Remove wake_cap() entirely.
Signed-off-by: Morten Rasmussen <morten.rasmussen@arm.com>
[Changelog tweaks]
Signed-off-by: Valentin Schneider <valentin.schneider@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
[Changelog tweaks]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lkml.kernel.org/r/20200206191957.12325-5-valentin.schneider@arm.com
Issue
=====
On asymmetric CPU capacity topologies, we currently rely on wake_cap() to
drive select_task_rq_fair() towards either:
- its slow-path (find_idlest_cpu()) if either the previous or
current (waking) CPU has too little capacity for the waking task
- its fast-path (select_idle_sibling()) otherwise
Commit:
3273163c67 ("sched/fair: Let asymmetric CPU configurations balance at wake-up")
points out that this relies on the assumption that "[...]the CPU capacities
within an SD_SHARE_PKG_RESOURCES domain (sd_llc) are homogeneous".
This assumption no longer holds on newer generations of big.LITTLE
systems (DynamIQ), which can accommodate CPUs of different compute capacity
within a single LLC domain. To hopefully paint a better picture, a regular
big.LITTLE topology would look like this:
+---------+ +---------+
| L2 | | L2 |
+----+----+ +----+----+
|CPU0|CPU1| |CPU2|CPU3|
+----+----+ +----+----+
^^^ ^^^
LITTLEs bigs
which would result in the following scheduler topology:
DIE [ ] <- sd_asym_cpucapacity
MC [ ] [ ] <- sd_llc
0 1 2 3
Conversely, a DynamIQ topology could look like:
+-------------------+
| L3 |
+----+----+----+----+
| L2 | L2 | L2 | L2 |
+----+----+----+----+
|CPU0|CPU1|CPU2|CPU3|
+----+----+----+----+
^^^^^ ^^^^^
LITTLEs bigs
which would result in the following scheduler topology:
MC [ ] <- sd_llc, sd_asym_cpucapacity
0 1 2 3
What this means is that, on DynamIQ systems, we could pass the wake_cap()
test (IOW presume the waking task fits on the CPU capacities of some LLC
domain), thus go through select_idle_sibling().
This function operates on an LLC domain, which here spans both bigs and
LITTLEs, so it could very well pick a CPU of too small capacity for the
task, despite there being fitting idle CPUs - it very much depends on the
CPU iteration order, on which we have absolutely no guarantees
capacity-wise.
Implementation
==============
Introduce yet another select_idle_sibling() helper function that takes CPU
capacity into account. The policy is to pick the first idle CPU which is
big enough for the task (task_util * margin < cpu_capacity). If no
idle CPU is big enough, we pick the idle one with the highest capacity.
Unlike other select_idle_sibling() helpers, this one operates on the
sd_asym_cpucapacity sched_domain pointer, which is guaranteed to span all
known CPU capacities in the system. As such, this will work for both
"legacy" big.LITTLE (LITTLEs & bigs split at MC, joined at DIE) and for
newer DynamIQ systems (e.g. LITTLEs and bigs in the same MC domain).
Note that this limits the scope of select_idle_sibling() to
select_idle_capacity() for asymmetric CPU capacity systems - the LLC domain
will not be scanned, and no further heuristic will be applied.
Signed-off-by: Morten Rasmussen <morten.rasmussen@arm.com>
Signed-off-by: Valentin Schneider <valentin.schneider@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Quentin Perret <qperret@google.com>
Link: https://lkml.kernel.org/r/20200206191957.12325-2-valentin.schneider@arm.com
Fix kernel-doc warning in kernel/sched/fair.c, caused by a recent
function parameter removal:
../kernel/sched/fair.c:3526: warning: Excess function parameter 'flags' description in 'attach_entity_load_avg'
Fixes: a4f9a0e51b ("sched/fair: Remove redundant call to cpufreq_update_util()")
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
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>
Link: https://lkml.kernel.org/r/cbe964e4-6879-fd08-41c9-ef1917414af4@infradead.org
The following XFS commit:
8ab39f11d9 ("xfs: prevent CIL push holdoff in log recovery")
changed the logic from using bound workqueues to using unbound
workqueues. Functionally this makes sense but it was observed at the
time that the dbench performance dropped quite a lot and CPU migrations
were increased.
The current pattern of the task migration is straight-forward. With XFS,
an IO issuer delegates work to xlog_cil_push_work ()on an unbound kworker.
This runs on a nearby CPU and on completion, dbench wakes up on its old CPU
as it is still idle and no migration occurs. dbench then queues the real
IO on the blk_mq_requeue_work() work item which runs on a bound kworker
which is forced to run on the same CPU as dbench. When IO completes,
the bound kworker wakes dbench but as the kworker is a bound but,
real task, the CPU is not considered idle and dbench gets migrated by
select_idle_sibling() to a new CPU. dbench may ping-pong between two CPUs
for a while but ultimately it starts a round-robin of all CPUs sharing
the same LLC. High-frequency migration on each IO completion has poor
performance overall. It has negative implications both in commication
costs and power management. mpstat confirmed that at low thread counts
that all CPUs sharing an LLC has low level of activity.
Note that even if the CIL patch was reverted, there still would
be migrations but the impact is less noticeable. It turns out that
individually the scheduler, XFS, blk-mq and workqueues all made sensible
decisions but in combination, the overall effect was sub-optimal.
This patch special cases the IO issue/completion pattern and allows
a bound kworker waker and a task wakee to stack on the same CPU if
there is a strong chance they are directly related. The expectation
is that the kworker is likely going back to sleep shortly. This is not
guaranteed as the IO could be queued asynchronously but there is a very
strong relationship between the task and kworker in this case that would
justify stacking on the same CPU instead of migrating. There should be
few concerns about kworker starvation given that the special casing is
only when the kworker is the waker.
DBench on XFS
MMTests config: io-dbench4-async modified to run on a fresh XFS filesystem
UMA machine with 8 cores sharing LLC
5.5.0-rc7 5.5.0-rc7
tipsched-20200124 kworkerstack
Amean 1 22.63 ( 0.00%) 20.54 * 9.23%*
Amean 2 25.56 ( 0.00%) 23.40 * 8.44%*
Amean 4 28.63 ( 0.00%) 27.85 * 2.70%*
Amean 8 37.66 ( 0.00%) 37.68 ( -0.05%)
Amean 64 469.47 ( 0.00%) 468.26 ( 0.26%)
Stddev 1 1.00 ( 0.00%) 0.72 ( 28.12%)
Stddev 2 1.62 ( 0.00%) 1.97 ( -21.54%)
Stddev 4 2.53 ( 0.00%) 3.58 ( -41.19%)
Stddev 8 5.30 ( 0.00%) 5.20 ( 1.92%)
Stddev 64 86.36 ( 0.00%) 94.53 ( -9.46%)
NUMA machine, 48 CPUs total, 24 CPUs share cache
5.5.0-rc7 5.5.0-rc7
tipsched-20200124 kworkerstack-v1r2
Amean 1 58.69 ( 0.00%) 30.21 * 48.53%*
Amean 2 60.90 ( 0.00%) 35.29 * 42.05%*
Amean 4 66.77 ( 0.00%) 46.55 * 30.28%*
Amean 8 81.41 ( 0.00%) 68.46 * 15.91%*
Amean 16 113.29 ( 0.00%) 107.79 * 4.85%*
Amean 32 199.10 ( 0.00%) 198.22 * 0.44%*
Amean 64 478.99 ( 0.00%) 477.06 * 0.40%*
Amean 128 1345.26 ( 0.00%) 1372.64 * -2.04%*
Stddev 1 2.64 ( 0.00%) 4.17 ( -58.08%)
Stddev 2 4.35 ( 0.00%) 5.38 ( -23.73%)
Stddev 4 6.77 ( 0.00%) 6.56 ( 3.00%)
Stddev 8 11.61 ( 0.00%) 10.91 ( 6.04%)
Stddev 16 18.63 ( 0.00%) 19.19 ( -3.01%)
Stddev 32 38.71 ( 0.00%) 38.30 ( 1.06%)
Stddev 64 100.28 ( 0.00%) 91.24 ( 9.02%)
Stddev 128 186.87 ( 0.00%) 160.34 ( 14.20%)
Dbench has been modified to report the time to complete a single "load
file". This is a more meaningful metric for dbench that a throughput
metric as the benchmark makes many different system calls that are not
throughput-related
Patch shows a 9.23% and 48.53% reduction in the time to process a load
file with the difference partially explained by the number of CPUs sharing
a LLC. In a separate run, task migrations were almost eliminated by the
patch for low client counts. In case people have issue with the metric
used for the benchmark, this is a comparison of the throughputs as
reported by dbench on the NUMA machine.
dbench4 Throughput (misleading but traditional)
5.5.0-rc7 5.5.0-rc7
tipsched-20200124 kworkerstack-v1r2
Hmean 1 321.41 ( 0.00%) 617.82 * 92.22%*
Hmean 2 622.87 ( 0.00%) 1066.80 * 71.27%*
Hmean 4 1134.56 ( 0.00%) 1623.74 * 43.12%*
Hmean 8 1869.96 ( 0.00%) 2212.67 * 18.33%*
Hmean 16 2673.11 ( 0.00%) 2806.13 * 4.98%*
Hmean 32 3032.74 ( 0.00%) 3039.54 ( 0.22%)
Hmean 64 2514.25 ( 0.00%) 2498.96 * -0.61%*
Hmean 128 1778.49 ( 0.00%) 1746.05 * -1.82%*
Note that this is somewhat specific to XFS and ext4 shows no performance
difference as it does not rely on kworkers in the same way. No major
problem was observed running other workloads on different machines although
not all tests have completed yet.
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20200128154006.GD3466@techsingularity.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Currently we loop through all threads of a core to evaluate if the core is
idle or not. This is unnecessary. If a thread of a core is not idle, skip
evaluating other threads of a core. Also while clearing the cpumask, bits
of all CPUs of a core can be cleared in one-shot.
Collecting ticks on a Power 9 SMT 8 system around select_idle_core
while running schbench shows us
(units are in ticks, hence lesser is better)
Without patch
N Min Max Median Avg Stddev
x 130 151 1083 284 322.72308 144.41494
With patch
N Min Max Median Avg Stddev Improvement
x 164 88 610 201 225.79268 106.78943 30.03%
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Valentin Schneider <valentin.schneider@arm.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Acked-by: Mel Gorman <mgorman@techsingularity.net>
Link: https://lkml.kernel.org/r/20191206172422.6578-1-srikar@linux.vnet.ibm.com
The CPU load balancer balances between different domains to spread load
and strives to have equal balance everywhere. Communicating tasks can
migrate so they are topologically close to each other but these decisions
are independent. On a lightly loaded NUMA machine, two communicating tasks
pulled together at wakeup time can be pushed apart by the load balancer.
In isolation, the load balancer decision is fine but it ignores the tasks
data locality and the wakeup/LB paths continually conflict. NUMA balancing
is also a factor but it also simply conflicts with the load balancer.
This patch allows a fixed degree of imbalance of two tasks to exist
between NUMA domains regardless of utilisation levels. In many cases,
this prevents communicating tasks being pulled apart. It was evaluated
whether the imbalance should be scaled to the domain size. However, no
additional benefit was measured across a range of workloads and machines
and scaling adds the risk that lower domains have to be rebalanced. While
this could change again in the future, such a change should specify the
use case and benefit.
The most obvious impact is on netperf TCP_STREAM -- two simple
communicating tasks with some softirq offload depending on the
transmission rate.
2-socket Haswell machine 48 core, HT enabled
netperf-tcp -- mmtests config config-network-netperf-unbound
baseline lbnuma-v3
Hmean 64 568.73 ( 0.00%) 577.56 * 1.55%*
Hmean 128 1089.98 ( 0.00%) 1128.06 * 3.49%*
Hmean 256 2061.72 ( 0.00%) 2104.39 * 2.07%*
Hmean 1024 7254.27 ( 0.00%) 7557.52 * 4.18%*
Hmean 2048 11729.20 ( 0.00%) 13350.67 * 13.82%*
Hmean 3312 15309.08 ( 0.00%) 18058.95 * 17.96%*
Hmean 4096 17338.75 ( 0.00%) 20483.66 * 18.14%*
Hmean 8192 25047.12 ( 0.00%) 27806.84 * 11.02%*
Hmean 16384 27359.55 ( 0.00%) 33071.88 * 20.88%*
Stddev 64 2.16 ( 0.00%) 2.02 ( 6.53%)
Stddev 128 2.31 ( 0.00%) 2.19 ( 5.05%)
Stddev 256 11.88 ( 0.00%) 3.22 ( 72.88%)
Stddev 1024 23.68 ( 0.00%) 7.24 ( 69.43%)
Stddev 2048 79.46 ( 0.00%) 71.49 ( 10.03%)
Stddev 3312 26.71 ( 0.00%) 57.80 (-116.41%)
Stddev 4096 185.57 ( 0.00%) 96.15 ( 48.19%)
Stddev 8192 245.80 ( 0.00%) 100.73 ( 59.02%)
Stddev 16384 207.31 ( 0.00%) 141.65 ( 31.67%)
In this case, there was a sizable improvement to performance and
a general reduction in variance. However, this is not univeral.
For most machines, the impact was roughly a 3% performance gain.
Ops NUMA base-page range updates 19796.00 292.00
Ops NUMA PTE updates 19796.00 292.00
Ops NUMA PMD updates 0.00 0.00
Ops NUMA hint faults 16113.00 143.00
Ops NUMA hint local faults % 8407.00 142.00
Ops NUMA hint local percent 52.18 99.30
Ops NUMA pages migrated 4244.00 1.00
Without the patch, only 52.18% of sampled accesses are local. In an
earlier changelog, 100% of sampled accesses are local and indeed on
most machines, this was still the case. In this specific case, the
local sampled rates was 99.3% but note the "base-page range updates"
and "PTE updates". The activity with the patch is negligible as were
the number of faults. The small number of pages migrated were related to
shared libraries. A 2-socket Broadwell showed better results on average
but are not presented for brevity as the performance was similar except
it showed 100% of the sampled NUMA hints were local. The patch holds up
for a 4-socket Haswell, an AMD EPYC and AMD Epyc 2 machine.
For dbench, the impact depends on the filesystem used and the number of
clients. On XFS, there is little difference as the clients typically
communicate with workqueues which have a separate class of scheduler
problem at the moment. For ext4, performance is generally better,
particularly for small numbers of clients as NUMA balancing activity is
negligible with the patch applied.
A more interesting example is the Facebook schbench which uses a
number of messaging threads to communicate with worker threads. In this
configuration, one messaging thread is used per NUMA node and the number of
worker threads is varied. The 50, 75, 90, 95, 99, 99.5 and 99.9 percentiles
for response latency is then reported.
Lat 50.00th-qrtle-1 44.00 ( 0.00%) 37.00 ( 15.91%)
Lat 75.00th-qrtle-1 53.00 ( 0.00%) 41.00 ( 22.64%)
Lat 90.00th-qrtle-1 57.00 ( 0.00%) 42.00 ( 26.32%)
Lat 95.00th-qrtle-1 63.00 ( 0.00%) 43.00 ( 31.75%)
Lat 99.00th-qrtle-1 76.00 ( 0.00%) 51.00 ( 32.89%)
Lat 99.50th-qrtle-1 89.00 ( 0.00%) 52.00 ( 41.57%)
Lat 99.90th-qrtle-1 98.00 ( 0.00%) 55.00 ( 43.88%)
Lat 50.00th-qrtle-2 42.00 ( 0.00%) 42.00 ( 0.00%)
Lat 75.00th-qrtle-2 48.00 ( 0.00%) 47.00 ( 2.08%)
Lat 90.00th-qrtle-2 53.00 ( 0.00%) 52.00 ( 1.89%)
Lat 95.00th-qrtle-2 55.00 ( 0.00%) 53.00 ( 3.64%)
Lat 99.00th-qrtle-2 62.00 ( 0.00%) 60.00 ( 3.23%)
Lat 99.50th-qrtle-2 63.00 ( 0.00%) 63.00 ( 0.00%)
Lat 99.90th-qrtle-2 68.00 ( 0.00%) 66.00 ( 2.94%
For higher worker threads, the differences become negligible but it's
interesting to note the difference in wakeup latency at low utilisation
and mpstat confirms that activity was almost all on one node until
the number of worker threads increase.
Hackbench generally showed neutral results across a range of machines.
This is different to earlier versions of the patch which allowed imbalances
for higher degrees of utilisation. perf bench pipe showed negligible
differences in overall performance as the differences are very close to
the noise.
An earlier prototype of the patch showed major regressions for NAS C-class
when running with only half of the available CPUs -- 20-30% performance
hits were measured at the time. With this version of the patch, the impact
is negligible with small gains/losses within the noise measured. This is
because the number of threads far exceeds the small imbalance the aptch
cares about. Similarly, there were report of regressions for the autonuma
benchmark against earlier versions but again, normal load balancing now
applies for that workload.
In general, the patch simply seeks to avoid unnecessary cross-node
migrations in the basic case where imbalances are very small. For low
utilisation communicating workloads, this patch generally behaves better
with less NUMA balancing activity. For high utilisation, there is no
change in behaviour.
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Valentin Schneider <valentin.schneider@arm.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Acked-by: Phil Auld <pauld@redhat.com>
Tested-by: Phil Auld <pauld@redhat.com>
Link: https://lkml.kernel.org/r/20200114101319.GO3466@techsingularity.net
sched_idle_cpu() isn't used for non SMP configuration and with a recent
change, we have started getting following warning:
kernel/sched/fair.c:5221:12: warning: ‘sched_idle_cpu’ defined but not used [-Wunused-function]
Fix that by defining sched_idle_cpu() only for SMP configurations.
Fixes: 323af6deaf ("sched/fair: Load balance aggressively for SCHED_IDLE CPUs")
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>
Link: https://lore.kernel.org/r/f0554f590687478b33914a4aff9f0e6a62886d44.1579499907.git.viresh.kumar@linaro.org
With commit
bef69dd878 ("sched/cpufreq: Move the cfs_rq_util_change() call to cpufreq_update_util()")
update_load_avg() has become the central point for calling cpufreq
(not including the update of blocked load). This change helps to
simplify further the number of calls to cpufreq_update_util() and to
remove last redundant ones. With update_load_avg(), we are now sure
that cpufreq_update_util() will be called after every task attachment
to a cfs_rq and especially after propagating this event down to the
util_avg of the root cfs_rq, which is the level that is used by
cpufreq governors like schedutil to set the frequency of a CPU.
The SCHED_CPUFREQ_MIGRATION flag forces an early call to cpufreq when
the migration happens in a cgroup whereas util_avg of root cfs_rq is
not yet updated and this call is duplicated with the one that happens
immediately after when the migration event reaches the root cfs_rq.
The dedicated flag SCHED_CPUFREQ_MIGRATION is now useless and can be
removed. The interface of attach_entity_load_avg() can also be
simplified accordingly.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Link: https://lkml.kernel.org/r/1579083620-24943-1-git-send-email-vincent.guittot@linaro.org
commit bf475ce0a3 ("sched/fair: Add per-CPU min capacity to
sched_group_capacity") introduced per-cpu min_capacity.
commit e3d6d0cb66 ("sched/fair: Add sched_group per-CPU max capacity")
introduced per-cpu max_capacity.
In the SD_OVERLAP case, the local variable 'capacity' represents the sum
of CPU capacity of all CPUs in the first sched group (sg) of the sched
domain (sd).
It is erroneously used to calculate sg's min and max CPU capacity.
To fix this use capacity_of(cpu) instead of 'capacity'.
The code which achieves this via cpu_rq(cpu)->sd->groups->sgc->capacity
(for rq->sd != NULL) can be removed since it delivers the same value as
capacity_of(cpu) which is currently only used for the (!rq->sd) case
(see update_cpu_capacity()).
An sg of the lowest sd (rq->sd or sd->child == NULL) represents a single
CPU (and hence sg->sgc->capacity == capacity_of(cpu)).
Signed-off-by: Peng Liu <iwtbavbm@gmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Valentin Schneider <valentin.schneider@arm.com>
Link: https://lkml.kernel.org/r/20200104130828.GA7718@iZj6chx1xj0e0buvshuecpZ
Move the code of calculation for delta_sum/delta_avg to where
it is really needed to be done.
Signed-off-by: Peng Wang <rocking@linux.alibaba.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lkml.kernel.org/r/20200103114400.17668-1-rocking@linux.alibaba.com
The fair scheduler performs periodic load balance on every CPU to check
if it can pull some tasks from other busy CPUs. The duration of this
periodic load balance is set to sd->balance_interval for the idle CPUs
and is calculated by multiplying the sd->balance_interval with the
sd->busy_factor (set to 32 by default) for the busy CPUs. The
multiplication is done for busy CPUs to avoid doing load balance too
often and rather spend more time executing actual task. While that is
the right thing to do for the CPUs busy with SCHED_OTHER or SCHED_BATCH
tasks, it may not be the optimal thing for CPUs running only SCHED_IDLE
tasks.
With the recent enhancements in the fair scheduler around SCHED_IDLE
CPUs, we now prefer to enqueue a newly-woken task to a SCHED_IDLE
CPU instead of other busy or idle CPUs. The same reasoning should be
applied to the load balancer as well to make it migrate tasks more
aggressively to a SCHED_IDLE CPU, as that will reduce the scheduling
latency of the migrated (SCHED_OTHER) tasks.
This patch makes minimal changes to the fair scheduler to do the next
load balance soon after the last non SCHED_IDLE task is dequeued from a
runqueue, i.e. making the CPU SCHED_IDLE. Also the sd->busy_factor is
ignored while calculating the balance_interval for such CPUs. This is
done to avoid delaying the periodic load balance by few hundred
milliseconds for SCHED_IDLE CPUs.
This is tested on ARM64 Hikey620 platform (octa-core) with the help of
rt-app and it is verified, using kernel traces, that the newly
SCHED_IDLE CPU does load balancing shortly after it becomes SCHED_IDLE
and pulls tasks from other busy CPUs.
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lkml.kernel.org/r/e485827eb8fe7db0943d6f3f6e0f5a4a70272781.1578471925.git.viresh.kumar@linaro.org
