Pull scheduler fixes from Ingo Molnar:
"Docbook fixes that make 99% of the diffstat, plus a oneliner fix"
* 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched: Ensure update_cfs_shares() is called for parents of continuously-running tasks
sched: Fix some kernel-doc warnings
Commit 3105b86a9f ("mm: sched: numa: Control enabling and disabling of
NUMA balancing if !SCHED_DEBUG") defined numabalancing_enabled to
control the enabling and disabling of automatic NUMA balancing, but it
is never used.
I believe the intention was to use this in place of sched_feat_numa(NUMA).
Currently, if SCHED_DEBUG is not defined, sched_feat_numa(NUMA) will
never be changed from the initial "false".
Signed-off-by: Dave Kleikamp <dave.kleikamp@oracle.com>
Acked-by: Mel Gorman <mgorman@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
We typically update a task_group's shares within the dequeue/enqueue
path. However, continuously running tasks sharing a CPU are not
subject to these updates as they are only put/picked. Unfortunately,
when we reverted f269ae046 (in 17bc14b7), we lost the augmenting
periodic update that was supposed to account for this; resulting in a
potential loss of fairness.
To fix this, re-introduce the explicit update in
update_cfs_rq_blocked_load() [called via entity_tick()].
Reported-by: Max Hailperin <max@gustavus.edu>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Reviewed-by: Paul Turner <pjt@google.com>
Link: http://lkml.kernel.org/n/tip-9545m3apw5d93ubyrotrj31y@git.kernel.org
Cc: <stable@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Smart wake-affine is using node-size as the factor currently, but the overhead
of the mask operation is high.
Thus, this patch introduce the 'sd_llc_size' percpu variable, which will record
the highest cache-share domain size, and make it to be the new factor, in order
to reduce the overhead and make it more reasonable.
Tested-by: Davidlohr Bueso <davidlohr.bueso@hp.com>
Tested-by: Michael Wang <wangyun@linux.vnet.ibm.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Acked-by: Michael Wang <wangyun@linux.vnet.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Link: http://lkml.kernel.org/r/51D5008E.6030102@linux.vnet.ibm.com
[ Tidied up the changelog. ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The wake-affine scheduler feature is currently always trying to pull
the wakee close to the waker. In theory this should be beneficial if
the waker's CPU caches hot data for the wakee, and it's also beneficial
in the extreme ping-pong high context switch rate case.
Testing shows it can benefit hackbench up to 15%.
However, the feature is somewhat blind, from which some workloads
such as pgbench suffer. It's also time-consuming algorithmically.
Testing shows it can damage pgbench up to 50% - far more than the
benefit it brings in the best case.
So wake-affine should be smarter and it should realize when to
stop its thankless effort at trying to find a suitable CPU to wake on.
This patch introduces 'wakee_flips', which will be increased each
time the task flips (switches) its wakee target.
So a high 'wakee_flips' value means the task has more than one
wakee, and the bigger the number, the higher the wakeup frequency.
Now when making the decision on whether to pull or not, pay attention to
the wakee with a high 'wakee_flips', pulling such a task may benefit
the wakee. Also imply that the waker will face cruel competition later,
it could be very cruel or very fast depends on the story behind
'wakee_flips', waker therefore suffers.
Furthermore, if waker also has a high 'wakee_flips', that implies that
multiple tasks rely on it, then waker's higher latency will damage all
of them, so pulling wakee seems to be a bad deal.
Thus, when 'waker->wakee_flips / wakee->wakee_flips' becomes
higher and higher, the cost of pulling seems to be worse and worse.
The patch therefore helps the wake-affine feature to stop its pulling
work when:
wakee->wakee_flips > factor &&
waker->wakee_flips > (factor * wakee->wakee_flips)
The 'factor' here is the number of CPUs in the current CPU's NUMA node,
so a bigger node will lead to more pulling since the trial becomes more
severe.
After applying the patch, pgbench shows up to 40% improvements and no regressions.
Tested with 12 cpu x86 server and tip 3.10.0-rc7.
The percentages in the final column highlight the areas with the biggest wins,
all other areas improved as well:
pgbench base smart
| db_size | clients | tps | | tps |
+---------+---------+-------+ +-------+
| 22 MB | 1 | 10598 | | 10796 |
| 22 MB | 2 | 21257 | | 21336 |
| 22 MB | 4 | 41386 | | 41622 |
| 22 MB | 8 | 51253 | | 57932 |
| 22 MB | 12 | 48570 | | 54000 |
| 22 MB | 16 | 46748 | | 55982 | +19.75%
| 22 MB | 24 | 44346 | | 55847 | +25.93%
| 22 MB | 32 | 43460 | | 54614 | +25.66%
| 7484 MB | 1 | 8951 | | 9193 |
| 7484 MB | 2 | 19233 | | 19240 |
| 7484 MB | 4 | 37239 | | 37302 |
| 7484 MB | 8 | 46087 | | 50018 |
| 7484 MB | 12 | 42054 | | 48763 |
| 7484 MB | 16 | 40765 | | 51633 | +26.66%
| 7484 MB | 24 | 37651 | | 52377 | +39.11%
| 7484 MB | 32 | 37056 | | 51108 | +37.92%
| 15 GB | 1 | 8845 | | 9104 |
| 15 GB | 2 | 19094 | | 19162 |
| 15 GB | 4 | 36979 | | 36983 |
| 15 GB | 8 | 46087 | | 49977 |
| 15 GB | 12 | 41901 | | 48591 |
| 15 GB | 16 | 40147 | | 50651 | +26.16%
| 15 GB | 24 | 37250 | | 52365 | +40.58%
| 15 GB | 32 | 36470 | | 50015 | +37.14%
Signed-off-by: Michael Wang <wangyun@linux.vnet.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/51D50057.9000809@linux.vnet.ibm.com
[ Improved the changelog. ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The bad thing about update_h_load(), which computes hierarchical load
factor for task groups, is that it is called for each task group in the
system before every load balancer run, and since rebalance can be
triggered very often, this function can eat really a lot of cpu time if
there are many cpu cgroups in the system.
Although the situation was improved significantly by commit a35b646
('sched, cgroup: Reduce rq->lock hold times for large cgroup
hierarchies'), the problem still can arise under some kinds of loads,
e.g. when cpus are switching from idle to busy and back very frequently.
For instance, when I start 1000 of processes that wake up every
millisecond on my 8 cpus host, 'top' and 'perf top' show:
Cpu(s): 17.8%us, 24.3%sy, 0.0%ni, 57.9%id, 0.0%wa, 0.0%hi, 0.0%si
Events: 243K cycles
7.57% [kernel] [k] __schedule
7.08% [kernel] [k] timerqueue_add
6.13% libc-2.12.so [.] usleep
Then if I create 10000 *idle* cpu cgroups (no processes in them), cpu
usage increases significantly although the 'wakers' are still executing
in the root cpu cgroup:
Cpu(s): 19.1%us, 48.7%sy, 0.0%ni, 31.6%id, 0.0%wa, 0.0%hi, 0.7%si
Events: 230K cycles
24.56% [kernel] [k] tg_load_down
5.76% [kernel] [k] __schedule
This happens because this particular kind of load triggers 'new idle'
rebalance very frequently, which requires calling update_h_load(),
which, in turn, calls tg_load_down() for every *idle* cpu cgroup even
though it is absolutely useless, because idle cpu cgroups have no tasks
to pull.
This patch tries to improve the situation by making h_load calculation
proceed only when h_load is really necessary. To achieve this, it
substitutes update_h_load() with update_cfs_rq_h_load(), which computes
h_load only for a given cfs_rq and all its ascendants, and makes the
load balancer call this function whenever it considers if a task should
be pulled, i.e. it moves h_load calculations directly to task_h_load().
For h_load of the same cfs_rq not to be updated multiple times (in case
several tasks in the same cgroup are considered during the same balance
run), the patch keeps the time of the last h_load update for each cfs_rq
and breaks calculation when it finds h_load to be uptodate.
The benefit of it is that h_load is computed only for those cfs_rq's,
which really need it, in particular all idle task groups are skipped.
Although this, in fact, moves h_load calculation under rq lock, it
should not affect latency much, because the amount of work done under rq
lock while trying to pull tasks is limited by sched_nr_migrate.
After the patch applied with the setup described above (1000 wakers in
the root cgroup and 10000 idle cgroups), I get:
Cpu(s): 16.9%us, 24.8%sy, 0.0%ni, 58.4%id, 0.0%wa, 0.0%hi, 0.0%si
Events: 242K cycles
7.57% [kernel] [k] __schedule
6.70% [kernel] [k] timerqueue_add
5.93% libc-2.12.so [.] usleep
Signed-off-by: Vladimir Davydov <vdavydov@parallels.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1373896159-1278-1-git-send-email-vdavydov@parallels.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
cfs_rq is declared twice, fix it.
Also use 'se' instead of '&p->se'.
Signed-off-by: Kirill Tkhai <tkhai@yandex.ru>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Link: http://lkml.kernel.org/r/169201374366727@web6d.yandex.ru
Signed-off-by: Ingo Molnar <mingo@kernel.org>
When building the htmldocs (in verbose mode), scripts/kernel-doc
reports the follwing type of warnings:
Warning(kernel/sched/core.c:936): No description found for return value of 'task_curr'
...
Fix those by:
- adding the missing descriptions
- using "Return" sections for the descriptions
Signed-off-by: Yacine Belkadi <yacine.belkadi.1@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1373654747-2389-1-git-send-email-yacine.belkadi.1@gmail.com
[ While at it, fix the cpupri_set() explanation. ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The __cpuinit type of throwaway sections might have made sense
some time ago when RAM was more constrained, but now the savings
do not offset the cost and complications. For example, the fix in
commit 5e427ec2d0 ("x86: Fix bit corruption at CPU resume time")
is a good example of the nasty type of bugs that can be created
with improper use of the various __init prefixes.
After a discussion on LKML[1] it was decided that cpuinit should go
the way of devinit and be phased out. Once all the users are gone,
we can then finally remove the macros themselves from linux/init.h.
This removes all the uses of the __cpuinit macros from C files in
the core kernel directories (kernel, init, lib, mm, and include)
that don't really have a specific maintainer.
[1] https://lkml.org/lkml/2013/5/20/589
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Similar to runnable_load_avg, blocked_load_avg variable, long type is
enough for removed_load in 64 bit or 32 bit machine.
Then we avoid the expensive atomic64 operations on 32 bit machine.
Signed-off-by: Alex Shi <alex.shi@intel.com>
Reviewed-by: Paul Turner <pjt@google.com>
Tested-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1371694737-29336-12-git-send-email-alex.shi@intel.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Since tg->load_avg is smaller than tg->load_weight, we don't need a
atomic64_t variable for load_avg in 32 bit machine.
The same reason for cfs_rq->tg_load_contrib.
The atomic_long_t/unsigned long variable type are more efficient and
convenience for them.
Signed-off-by: Alex Shi <alex.shi@intel.com>
Tested-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1371694737-29336-11-git-send-email-alex.shi@intel.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Since the 'u64 runnable_load_avg, blocked_load_avg' in cfs_rq struct are
smaller than 'unsigned long' cfs_rq->load.weight. We don't need u64
vaiables to describe them. unsigned long is more efficient and convenience.
Signed-off-by: Alex Shi <alex.shi@intel.com>
Reviewed-by: Paul Turner <pjt@google.com>
Tested-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1371694737-29336-10-git-send-email-alex.shi@intel.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Aside from using runnable load average in background, move_tasks is
also the key function in load balance. We need consider the runnable
load average in it in order to make it an apple to apple load
comparison.
Morten had caught a div u64 bug on ARM, thanks!
Thanks-to: Morten Rasmussen <morten.rasmussen@arm.com>
Signed-off-by: Alex Shi <alex.shi@intel.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1371694737-29336-8-git-send-email-alex.shi@intel.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
They are the base values in load balance, update them with rq runnable
load average, then the load balance will consider runnable load avg
naturally.
We also try to include the blocked_load_avg as cpu load in balancing,
but that cause kbuild performance drop 6% on every Intel machine, and
aim7/oltp drop on some of 4 CPU sockets machines.
Or only add blocked_load_avg into get_rq_runable_load, hackbench still
drop a little on NHM EX.
Signed-off-by: Alex Shi <alex.shi@intel.com>
Reviewed-by: Gu Zheng <guz.fnst@cn.fujitsu.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1371694737-29336-7-git-send-email-alex.shi@intel.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The woken migrated task will __synchronize_entity_decay(se); in
migrate_task_rq_fair, then it needs to set
`se->avg.last_runnable_update -= (-se->avg.decay_count) << 20' before
update_entity_load_avg, in order to avoid sleep time is updated twice
for se.avg.load_avg_contrib in both __syncchronize and
update_entity_load_avg.
However if the sleeping task is woken up from the same cpu, it miss
the last_runnable_update before update_entity_load_avg(se, 0, 1), then
the sleep time was used twice in both functions. So we need to remove
the double sleep time accounting.
Paul also contributed some code comments in this commit.
Signed-off-by: Alex Shi <alex.shi@intel.com>
Reviewed-by: Paul Turner <pjt@google.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1371694737-29336-5-git-send-email-alex.shi@intel.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
We need to initialize the se.avg.{decay_count, load_avg_contrib} for a
new forked task. Otherwise random values of above variables cause a
mess when a new task is enqueued:
enqueue_task_fair
enqueue_entity
enqueue_entity_load_avg
and make fork balancing imbalance due to incorrect load_avg_contrib.
Further more, Morten Rasmussen notice some tasks were not launched at
once after created. So Paul and Peter suggest giving a start value for
new task runnable avg time same as sched_slice().
PeterZ said:
> So the 'problem' is that our running avg is a 'floating' average; ie. it
> decays with time. Now we have to guess about the future of our newly
> spawned task -- something that is nigh impossible seeing these CPU
> vendors keep refusing to implement the crystal ball instruction.
>
> So there's two asymptotic cases we want to deal well with; 1) the case
> where the newly spawned program will be 'nearly' idle for its lifetime;
> and 2) the case where its cpu-bound.
>
> Since we have to guess, we'll go for worst case and assume its
> cpu-bound; now we don't want to make the avg so heavy adjusting to the
> near-idle case takes forever. We want to be able to quickly adjust and
> lower our running avg.
>
> Now we also don't want to make our avg too light, such that it gets
> decremented just for the new task not having had a chance to run yet --
> even if when it would run, it would be more cpu-bound than not.
>
> So what we do is we make the initial avg of the same duration as that we
> guess it takes to run each task on the system at least once -- aka
> sched_slice().
>
> Of course we can defeat this with wakeup/fork bombs, but in the 'normal'
> case it should be good enough.
Paul also contributed most of the code comments in this commit.
Signed-off-by: Alex Shi <alex.shi@intel.com>
Reviewed-by: Gu Zheng <guz.fnst@cn.fujitsu.com>
Reviewed-by: Paul Turner <pjt@google.com>
[peterz; added explanation of sched_slice() usage]
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1371694737-29336-4-git-send-email-alex.shi@intel.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Remove CONFIG_FAIR_GROUP_SCHED that covers the runnable info, then
we can use runnable load variables.
Also remove 2 CONFIG_FAIR_GROUP_SCHED setting which is not in reverted
patch(introduced in 9ee474f), but also need to revert.
Signed-off-by: Alex Shi <alex.shi@intel.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/51CA76A3.3050207@intel.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
default_cfs_period(), do_sched_cfs_period_timer(), do_sched_cfs_slack_timer()
already defined previously, no need to declare again.
Signed-off-by: Michael Wang <wangyun@linux.vnet.ibm.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/51AD8808.7020608@linux.vnet.ibm.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Directly use rq to save some code.
Signed-off-by: Michael Wang <wangyun@linux.vnet.ibm.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/51AD87EB.1070605@linux.vnet.ibm.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Commit 78becc2709 ("sched: Use an accessor to read the rq clock")
introduces rq_clock(), which obsoletes the use of the "rq" variable
in expire_cfs_rq_runtime() and triggers this build warning:
kernel/sched/fair.c: In function 'expire_cfs_rq_runtime':
kernel/sched/fair.c:2159:13: warning: unused variable 'rq' [-Wunused-variable]
Signed-off-by: Kamalesh Babulal <kamalesh@linux.vnet.ibm.com>
Acked-by: Frederic Weisbecker <fweisbec@gmail.com>
Acked-by: Paul Turner <pjt@google.com>
Cc: peterz@infradead.org
Link: http://lkml.kernel.org/r/1369904660-14169-1-git-send-email-kamalesh@linux.vnet.ibm.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Read the runqueue clock through an accessor. This
prepares for adding a debugging infrastructure to
detect missing or redundant calls to update_rq_clock()
between a scheduler's entry and exit point.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Li Zhong <zhong@linux.vnet.ibm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Paul Turner <pjt@google.com>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1365724262-20142-6-git-send-email-fweisbec@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
In this function we are making use of rq->clock right before the
update of the rq clock, let's just call update_rq_clock() just
before that to avoid using a stale rq clock value.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Li Zhong <zhong@linux.vnet.ibm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Paul Turner <pjt@google.com>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1365724262-20142-5-git-send-email-fweisbec@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Because we may update the execution time in
sched_group_set_shares()->update_cfs_shares()->reweight_entity()->update_curr()
before reweighting the entity while setting the group shares and this requires
an uptodate version of the runqueue clock.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Li Zhong <zhong@linux.vnet.ibm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Paul Turner <pjt@google.com>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1365724262-20142-3-git-send-email-fweisbec@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
It is a few instructions more efficent to and slightly more
readable to use this_rq()-> instead of cpu_rq(smp_processor_id())-> .
Size comparison of kernel/sched/fair.o:
text data bss dec hex filename
27972 122 26 28120 6dd8 fair.o.before
27956 122 26 28104 6dc8 fair.o.after
Signed-off-by: Nathan Zimmer <nzimmer@sgi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1368116643-87971-1-git-send-email-nzimmer@sgi.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
These inlines are only used by kernel/sched/fair.c so they do
not need to be present in the main kernel/sched/sched.h file.
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Link: http://lkml.kernel.org/r/1366398650-31599-3-git-send-email-paul.gortmaker@windriver.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The full dynticks tree needs the latest RCU and sched
upstream updates in order to fix some dependencies.
Merge a common upstream merge point that has these
updates.
Conflicts:
include/linux/perf_event.h
kernel/rcutree.h
kernel/rcutree_plugin.h
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
On my SMP platform which is made of 5 cores in 2 clusters, I
have the nr_busy_cpu field of sched_group_power struct that is
not null when the platform is fully idle - which makes the
scheduler unhappy.
The root cause is:
During the boot sequence, some CPUs reach the idle loop and set
their NOHZ_IDLE flag while waiting for others CPUs to boot. But
the nr_busy_cpus field is initialized later with the assumption
that all CPUs are in the busy state whereas some CPUs have
already set their NOHZ_IDLE flag.
More generally, the NOHZ_IDLE flag must be initialized when new
sched_domains are created in order to ensure that NOHZ_IDLE and
nr_busy_cpus are aligned.
This condition can be ensured by adding a synchronize_rcu()
between the destruction of old sched_domains and the creation of
new ones so the NOHZ_IDLE flag will not be updated with old
sched_domain once it has been initialized. But this solution
introduces a additionnal latency in the rebuild sequence that is
called during cpu hotplug.
As suggested by Frederic Weisbecker, another solution is to have
the same rcu lifecycle for both NOHZ_IDLE and sched_domain
struct. A new nohz_idle field is added to sched_domain so both
status and sched_domain will share the same RCU lifecycle and
will be always synchronized. In addition, there is no more need
to protect nohz_idle against concurrent access as it is only
modified by 2 exclusive functions called by local cpu.
This solution has been prefered to the creation of a new struct
with an extra pointer indirection for sched_domain.
The synchronization is done at the cost of :
- An additional indirection and a rcu_dereference for accessing nohz_idle.
- We use only the nohz_idle field of the top sched_domain.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: linaro-kernel@lists.linaro.org
Cc: peterz@infradead.org
Cc: fweisbec@gmail.com
Cc: pjt@google.com
Cc: rostedt@goodmis.org
Cc: efault@gmx.de
Link: http://lkml.kernel.org/r/1366729142-14662-1-git-send-email-vincent.guittot@linaro.org
[ Fixed !NO_HZ build bug. ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Commit 88b8dac0 makes load_balance() consider other cpus in its
group. But, in that, there is no code for preventing to
re-select dst-cpu. So, same dst-cpu can be selected over and
over.
This patch add functionality to load_balance() in order to
exclude cpu which is selected once. We prevent to re-select
dst_cpu via env's cpus, so now, env's cpus is a candidate not
only for src_cpus, but also dst_cpus.
With this patch, we can remove lb_iterations and
max_lb_iterations, because we decide whether we can go ahead or
not via env's cpus.
Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Tested-by: Jason Low <jason.low2@hp.com>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Davidlohr Bueso <davidlohr.bueso@hp.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1366705662-3587-7-git-send-email-iamjoonsoo.kim@lge.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
This name doesn't represent specific meaning.
So rename it to imply it's purpose.
Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Tested-by: Jason Low <jason.low2@hp.com>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Davidlohr Bueso <davidlohr.bueso@hp.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1366705662-3587-6-git-send-email-iamjoonsoo.kim@lge.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Currently, LBF_ALL_PINNED is cleared after affinity check is
passed. So, if task migration is skipped by small load value or
small imbalance value in move_tasks(), we don't clear
LBF_ALL_PINNED. At last, we trigger 'redo' in load_balance().
Imbalance value is often so small that any tasks cannot be moved
to other cpus and, of course, this situation may be continued
after we change the target cpu. So this patch move up affinity
check code and clear LBF_ALL_PINNED before evaluating load value
in order to mitigate useless redoing overhead.
In addition, re-order some comments correctly.
Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Tested-by: Jason Low <jason.low2@hp.com>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Davidlohr Bueso <davidlohr.bueso@hp.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1366705662-3587-5-git-send-email-iamjoonsoo.kim@lge.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Commit 88b8dac0 makes load_balance() consider other cpus in its
group, regardless of idle type. When we do NEWLY_IDLE balancing,
we should not consider it, because a motivation of NEWLY_IDLE
balancing is to turn this cpu to non idle state if needed. This
is not the case of other cpus. So, change code not to consider
other cpus for NEWLY_IDLE balancing.
With this patch, assign 'if (pulled_task) this_rq->idle_stamp =
0' in idle_balance() is corrected, because NEWLY_IDLE balancing
doesn't consider other cpus. Assigning to 'this_rq->idle_stamp'
is now valid.
Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Tested-by: Jason Low <jason.low2@hp.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Davidlohr Bueso <davidlohr.bueso@hp.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1366705662-3587-4-git-send-email-iamjoonsoo.kim@lge.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
After commit 88b8dac0, dst-cpu can be changed in load_balance(),
then we can't know cpu_idle_type of dst-cpu when load_balance()
return positive. So, add explicit cpu_idle_type checking.
Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Tested-by: Jason Low <jason.low2@hp.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Davidlohr Bueso <davidlohr.bueso@hp.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1366705662-3587-3-git-send-email-iamjoonsoo.kim@lge.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
cur_ld_moved is reset if env.flags hit LBF_NEED_BREAK.
So, there is possibility that we miss doing resched_cpu().
Correct it as changing position of resched_cpu()
before checking LBF_NEED_BREAK.
Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Tested-by: Jason Low <jason.low2@hp.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Davidlohr Bueso <davidlohr.bueso@hp.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1366705662-3587-2-git-send-email-iamjoonsoo.kim@lge.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The current update of the rq's load can be erroneous when RT
tasks are involved.
The update of the load of a rq that becomes idle, is done only
if the avg_idle is less than sysctl_sched_migration_cost. If RT
tasks and short idle duration alternate, the runnable_avg will
not be updated correctly and the time will be accounted as idle
time when a CFS task wakes up.
A new idle_enter function is called when the next task is the
idle function so the elapsed time will be accounted as run time
in the load of the rq, whatever the average idle time is. The
function update_rq_runnable_avg is removed from idle_balance.
When a RT task is scheduled on an idle CPU, the update of the
rq's load is not done when the rq exit idle state because CFS's
functions are not called. Then, the idle_balance, which is
called just before entering the idle function, updates the rq's
load and makes the assumption that the elapsed time since the
last update, was only running time.
As a consequence, the rq's load of a CPU that only runs a
periodic RT task, is close to LOAD_AVG_MAX whatever the running
duration of the RT task is.
A new idle_exit function is called when the prev task is the
idle function so the elapsed time will be accounted as idle time
in the rq's load.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Cc: linaro-kernel@lists.linaro.org
Cc: peterz@infradead.org
Cc: pjt@google.com
Cc: fweisbec@gmail.com
Cc: efault@gmx.de
Link: http://lkml.kernel.org/r/1366302867-5055-1-git-send-email-vincent.guittot@linaro.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
A comment in function rebalance_domains() mentions
arch_init_sched_domains(), but that function does not exist
anymore. The proper function is init_sched_domains().
Signed-off-by: Libin <huawei.libin@huawei.com>
Cc: <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1364814841-49156-1-git-send-email-huawei.libin@huawei.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
At this point tsk_cache_hot is always true, so no need to check it.
Signed-off-by: Zhang Hang <bob.zhanghang@huawei.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/51650107.9040606@huawei.com
[ Also remove unnecessary schedstat #ifdefs. ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
We are planning to convert the dynticks Kconfig options layout
into a choice menu. The user must be able to easily pick
any of the following implementations: constant periodic tick,
idle dynticks, full dynticks.
As this implies a mutual exclusion, the two dynticks implementions
need to converge on the selection of a common Kconfig option in order
to ease the sharing of a common infrastructure.
It would thus seem pretty natural to reuse CONFIG_NO_HZ to
that end. It already implements all the idle dynticks code
and the full dynticks depends on all that code for now.
So ideally the choice menu would propose CONFIG_NO_HZ_IDLE and
CONFIG_NO_HZ_EXTENDED then both would select CONFIG_NO_HZ.
On the other hand we want to stay backward compatible: if
CONFIG_NO_HZ is set in an older config file, we want to
enable CONFIG_NO_HZ_IDLE by default.
But we can't afford both at the same time or we run into
a circular dependency:
1) CONFIG_NO_HZ_IDLE and CONFIG_NO_HZ_EXTENDED both select
CONFIG_NO_HZ
2) If CONFIG_NO_HZ is set, we default to CONFIG_NO_HZ_IDLE
We might be able to support that from Kconfig/Kbuild but it
may not be wise to introduce such a confusing behaviour.
So to solve this, create a new CONFIG_NO_HZ_COMMON option
which gathers the common code between idle and full dynticks
(that common code for now is simply the idle dynticks code)
and select it from their referring Kconfig.
Then we'll later create CONFIG_NO_HZ_IDLE and map CONFIG_NO_HZ
to it for backward compatibility.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Chris Metcalf <cmetcalf@tilera.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Geoff Levand <geoff@infradead.org>
Cc: Gilad Ben Yossef <gilad@benyossef.com>
Cc: Hakan Akkan <hakanakkan@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Kevin Hilman <khilman@linaro.org>
Cc: Li Zhong <zhong@linux.vnet.ibm.com>
Cc: Namhyung Kim <namhyung.kim@lge.com>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Paul Gortmaker <paul.gortmaker@windriver.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
As default_scale_{freq,smt}_power() and update_rt_power() are
used in kernel/sched/fair.c only, annotate them as static
functions.
Signed-off-by: Li Zefan <lizefan@huawei.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/5135A7AF.8010900@huawei.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull scheduler changes from Ingo Molnar:
"Main changes:
- scheduler side full-dynticks (user-space execution is undisturbed
and receives no timer IRQs) preparation changes that convert the
cputime accounting code to be full-dynticks ready, from Frederic
Weisbecker.
- Initial sched.h split-up changes, by Clark Williams
- select_idle_sibling() performance improvement by Mike Galbraith:
" 1 tbench pair (worst case) in a 10 core + SMT package:
pre 15.22 MB/sec 1 procs
post 252.01 MB/sec 1 procs "
- sched_rr_get_interval() ABI fix/change. We think this detail is not
used by apps (so it's not an ABI in practice), but lets keep it
under observation.
- misc RT scheduling cleanups, optimizations"
* 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (24 commits)
sched/rt: Add <linux/sched/rt.h> header to <linux/init_task.h>
cputime: Remove irqsave from seqlock readers
sched, powerpc: Fix sched.h split-up build failure
cputime: Restore CPU_ACCOUNTING config defaults for PPC64
sched/rt: Move rt specific bits into new header file
sched/rt: Add a tuning knob to allow changing SCHED_RR timeslice
sched: Move sched.h sysctl bits into separate header
sched: Fix signedness bug in yield_to()
sched: Fix select_idle_sibling() bouncing cow syndrome
sched/rt: Further simplify pick_rt_task()
sched/rt: Do not account zero delta_exec in update_curr_rt()
cputime: Safely read cputime of full dynticks CPUs
kvm: Prepare to add generic guest entry/exit callbacks
cputime: Use accessors to read task cputime stats
cputime: Allow dynamic switch between tick/virtual based cputime accounting
cputime: Generic on-demand virtual cputime accounting
cputime: Move default nsecs_to_cputime() to jiffies based cputime file
cputime: Librarize per nsecs resolution cputime definitions
cputime: Avoid multiplication overflow on utime scaling
context_tracking: Export context state for generic vtime
...
Fix up conflict in kernel/context_tracking.c due to comment additions.
If the previous CPU is cache affine and idle, select it.
The current implementation simply traverses the sd_llc domain,
taking the first idle CPU encountered, which walks buddy pairs
hand in hand over the package, inflicting excruciating pain.
1 tbench pair (worst case) in a 10 core + SMT package:
pre 15.22 MB/sec 1 procs
post 252.01 MB/sec 1 procs
Signed-off-by: Mike Galbraith <bitbucket@online.de>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1359371965.5783.127.camel@marge.simpson.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
a4c96ae319 "sched: Unthrottle rt runqueues in
__disable_runtime()" turned the unthrottle_offline_cfs_rqs
function into a static symbol, which now triggers a warning
about it being potentially unused:
kernel/sched/fair.c:2055:13: warning: 'unthrottle_offline_cfs_rqs' defined but not used [-Wunused-function]
Marking it __maybe_unused shuts up the gcc warning and lets the
compiler safely drop the function body when it's not being used.
To reproduce, build the ARM bcm2835_defconfig.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Cc: Peter Boonstoppel <pboonstoppel@nvidia.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Turner <pjt@google.com>
Cc: linux-arm-kernel@list.infradead.org
Link: http://lkml.kernel.org/r/1359123276-15833-6-git-send-email-arnd@arndb.de
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The caller of sched_sliced() should pass se.cfs_rq and se as the
arguments, however in sched_rr_get_interval() we gave it
rq.cfs_rq and se, which made the following computation obviously
wrong.
The change was introduced by commit:
77034937dc sched: fix crash in sys_sched_rr_get_interval()
... 5 years ago, while it had been the correct 'cfs_rq_of' before
the commit. The change seems to be irrelevant to the commit
msg, which was to return a 0 timeslice for tasks that are on an
idle runqueue. So I believe that was just a plain typo.
Signed-off-by: Zhu Yanhai <gaoyang.zyh@taobao.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Paul Turner <pjt@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: http://lkml.kernel.org/r/1357621012-15039-1-git-send-email-gaoyang.zyh@taobao.com
[ Since this is an ABI and an old bug, we'll test this via a
slow upstream route, to hopefully discover any app breakage. ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
task_numa_placement() oopsed on NULL p->mm when task_numa_fault() got
called in the handling of break_ksm() for ksmd. That might be a
peculiar case, which perhaps KSM could takes steps to avoid? but it's
more robust if task_numa_placement() allows for such a possibility.
Signed-off-by: Hugh Dickins <hughd@google.com>
Acked-by: Mel Gorman <mgorman@suse.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Michal Hocko reported that the following build error occurs if
CONFIG_NUMA_BALANCING is set without THP support
kernel/sched/fair.c: In function ‘task_numa_work’:
kernel/sched/fair.c:932:55: error: call to ‘__build_bug_failed’ declared with attribute error: BUILD_BUG failed
The problem is that HPAGE_PMD_SHIFT triggers a BUILD_BUG() on
!CONFIG_TRANSPARENT_HUGEPAGE. This patch addresses the problem.
Reported-by: Michal Hocko <mhocko@suse.cz>
Signed-off-by: Mel Gorman <mgorman@suse.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.18 (GNU/Linux)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=ivQa
-----END PGP SIGNATURE-----
Merge tag 'balancenuma-v11' of git://git.kernel.org/pub/scm/linux/kernel/git/mel/linux-balancenuma
Pull Automatic NUMA Balancing bare-bones from Mel Gorman:
"There are three implementations for NUMA balancing, this tree
(balancenuma), numacore which has been developed in tip/master and
autonuma which is in aa.git.
In almost all respects balancenuma is the dumbest of the three because
its main impact is on the VM side with no attempt to be smart about
scheduling. In the interest of getting the ball rolling, it would be
desirable to see this much merged for 3.8 with the view to building
scheduler smarts on top and adapting the VM where required for 3.9.
The most recent set of comparisons available from different people are
mel: https://lkml.org/lkml/2012/12/9/108
mingo: https://lkml.org/lkml/2012/12/7/331
tglx: https://lkml.org/lkml/2012/12/10/437
srikar: https://lkml.org/lkml/2012/12/10/397
The results are a mixed bag. In my own tests, balancenuma does
reasonably well. It's dumb as rocks and does not regress against
mainline. On the other hand, Ingo's tests shows that balancenuma is
incapable of converging for this workloads driven by perf which is bad
but is potentially explained by the lack of scheduler smarts. Thomas'
results show balancenuma improves on mainline but falls far short of
numacore or autonuma. Srikar's results indicate we all suffer on a
large machine with imbalanced node sizes.
My own testing showed that recent numacore results have improved
dramatically, particularly in the last week but not universally.
We've butted heads heavily on system CPU usage and high levels of
migration even when it shows that overall performance is better.
There are also cases where it regresses. Of interest is that for
specjbb in some configurations it will regress for lower numbers of
warehouses and show gains for higher numbers which is not reported by
the tool by default and sometimes missed in treports. Recently I
reported for numacore that the JVM was crashing with
NullPointerExceptions but currently it's unclear what the source of
this problem is. Initially I thought it was in how numacore batch
handles PTEs but I'm no longer think this is the case. It's possible
numacore is just able to trigger it due to higher rates of migration.
These reports were quite late in the cycle so I/we would like to start
with this tree as it contains much of the code we can agree on and has
not changed significantly over the last 2-3 weeks."
* tag 'balancenuma-v11' of git://git.kernel.org/pub/scm/linux/kernel/git/mel/linux-balancenuma: (50 commits)
mm/rmap, migration: Make rmap_walk_anon() and try_to_unmap_anon() more scalable
mm/rmap: Convert the struct anon_vma::mutex to an rwsem
mm: migrate: Account a transhuge page properly when rate limiting
mm: numa: Account for failed allocations and isolations as migration failures
mm: numa: Add THP migration for the NUMA working set scanning fault case build fix
mm: numa: Add THP migration for the NUMA working set scanning fault case.
mm: sched: numa: Delay PTE scanning until a task is scheduled on a new node
mm: sched: numa: Control enabling and disabling of NUMA balancing if !SCHED_DEBUG
mm: sched: numa: Control enabling and disabling of NUMA balancing
mm: sched: Adapt the scanning rate if a NUMA hinting fault does not migrate
mm: numa: Use a two-stage filter to restrict pages being migrated for unlikely task<->node relationships
mm: numa: migrate: Set last_nid on newly allocated page
mm: numa: split_huge_page: Transfer last_nid on tail page
mm: numa: Introduce last_nid to the page frame
sched: numa: Slowly increase the scanning period as NUMA faults are handled
mm: numa: Rate limit setting of pte_numa if node is saturated
mm: numa: Rate limit the amount of memory that is migrated between nodes
mm: numa: Structures for Migrate On Fault per NUMA migration rate limiting
mm: numa: Migrate pages handled during a pmd_numa hinting fault
mm: numa: Migrate on reference policy
...
This reverts commit f269ae0469.
It turns out it causes a very noticeable interactivity regression with
CONFIG_SCHED_AUTOGROUP (test-case: "make -j32" of the kernel in a
terminal window, while scrolling in a browser - the autogrouping means
that the two end up in separate cgroups, and the browser should be
smooth as silk despite the high load).
Says Paul Turner:
"It seems that the update-throttling on the wake-side is reducing the
interactive tasks' ability to preempt. While I suspect the right
longer term answer here is force these updates only in the
cross-cgroup case; this is less trivial. For this release I believe
the right answer is either going to be a revert or restore the updates
on the enqueue-side."
Reported-by: Linus Torvalds <torvalds@linux-foundation.org>
Bisected-by: Mike Galbraith <efault@gmx.de>
Acked-by: Paul Turner <pjt@google.com>
Acked-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Due to the fact that migrations are driven by the CPU a task is running
on there is no point tracking NUMA faults until one task runs on a new
node. This patch tracks the first node used by an address space. Until
it changes, PTE scanning is disabled and no NUMA hinting faults are
trapped. This should help workloads that are short-lived, do not care
about NUMA placement or have bound themselves to a single node.
This takes advantage of the logic in "mm: sched: numa: Implement slow
start for working set sampling" to delay when the checks are made. This
will take advantage of processes that set their CPU and node bindings
early in their lifetime. It will also potentially allow any initial load
balancing to take place.
Signed-off-by: Mel Gorman <mgorman@suse.de>
This patch adds Kconfig options and kernel parameters to allow the
enabling and disabling of automatic NUMA balancing. The existance
of such a switch was and is very important when debugging problems
related to transparent hugepages and we should have the same for
automatic NUMA placement.
Signed-off-by: Mel Gorman <mgorman@suse.de>
The PTE scanning rate and fault rates are two of the biggest sources of
system CPU overhead with automatic NUMA placement. Ideally a proper policy
would detect if a workload was properly placed, schedule and adjust the
PTE scanning rate accordingly. We do not track the necessary information
to do that but we at least know if we migrated or not.
This patch scans slower if a page was not migrated as the result of a
NUMA hinting fault up to sysctl_numa_balancing_scan_period_max which is
now higher than the previous default. Once every minute it will reset
the scanner in case of phase changes.
This is hilariously crude and the numbers are arbitrary. Workloads will
converge quite slowly in comparison to what a proper policy should be able
to do. On the plus side, we will chew up less CPU for workloads that have
no need for automatic balancing.
Signed-off-by: Mel Gorman <mgorman@suse.de>
Currently the rate of scanning for an address space is controlled
by the individual tasks. The next scan is simply determined by
2*p->numa_scan_period.
The 2*p->numa_scan_period is arbitrary and never changes. At this point
there is still no proper policy that decides if a task or process is
properly placed. It just scans and assumes the next NUMA fault will
place it properly. As it is assumed that pages will get properly placed
over time, increase the scan window each time a fault is incurred. This
is a big assumption as noted in the comments.
It should be noted that changing to p->numa_scan_period will increase
system CPU usage because now the scanning rate has effectively doubled.
If that is a problem then the min_rate should be made 200ms instead of
restoring the 2* logic.
Signed-off-by: Mel Gorman <mgorman@suse.de>
If there are a large number of NUMA hinting faults and all of them
are resulting in migrations it may indicate that memory is just
bouncing uselessly around. NUMA balancing cost is likely exceeding
any benefit from locality. Rate limit the PTE updates if the node
is migration rate-limited. As noted in the comments, this distorts
the NUMA faulting statistics.
Signed-off-by: Mel Gorman <mgorman@suse.de>
Add a 1 second delay before starting to scan the working set of
a task and starting to balance it amongst nodes.
[ note that before the constant per task WSS sampling rate patch
the initial scan would happen much later still, in effect that
patch caused this regression. ]
The theory is that short-run tasks benefit very little from NUMA
placement: they come and go, and they better stick to the node
they were started on. As tasks mature and rebalance to other CPUs
and nodes, so does their NUMA placement have to change and so
does it start to matter more and more.
In practice this change fixes an observable kbuild regression:
# [ a perf stat --null --repeat 10 test of ten bzImage builds to /dev/shm ]
!NUMA:
45.291088843 seconds time elapsed ( +- 0.40% )
45.154231752 seconds time elapsed ( +- 0.36% )
+NUMA, no slow start:
46.172308123 seconds time elapsed ( +- 0.30% )
46.343168745 seconds time elapsed ( +- 0.25% )
+NUMA, 1 sec slow start:
45.224189155 seconds time elapsed ( +- 0.25% )
45.160866532 seconds time elapsed ( +- 0.17% )
and it also fixes an observable perf bench (hackbench) regression:
# perf stat --null --repeat 10 perf bench sched messaging
-NUMA:
-NUMA: 0.246225691 seconds time elapsed ( +- 1.31% )
+NUMA no slow start: 0.252620063 seconds time elapsed ( +- 1.13% )
+NUMA 1sec delay: 0.248076230 seconds time elapsed ( +- 1.35% )
The implementation is simple and straightforward, most of the patch
deals with adding the /proc/sys/kernel/numa_balancing_scan_delay_ms tunable
knob.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Rik van Riel <riel@redhat.com>
[ Wrote the changelog, ran measurements, tuned the default. ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Mel Gorman <mgorman@suse.de>
Reviewed-by: Rik van Riel <riel@redhat.com>
Previously, to probe the working set of a task, we'd use
a very simple and crude method: mark all of its address
space PROT_NONE.
That method has various (obvious) disadvantages:
- it samples the working set at dissimilar rates,
giving some tasks a sampling quality advantage
over others.
- creates performance problems for tasks with very
large working sets
- over-samples processes with large address spaces but
which only very rarely execute
Improve that method by keeping a rotating offset into the
address space that marks the current position of the scan,
and advance it by a constant rate (in a CPU cycles execution
proportional manner). If the offset reaches the last mapped
address of the mm then it then it starts over at the first
address.
The per-task nature of the working set sampling functionality in this tree
allows such constant rate, per task, execution-weight proportional sampling
of the working set, with an adaptive sampling interval/frequency that
goes from once per 100ms up to just once per 8 seconds. The current
sampling volume is 256 MB per interval.
As tasks mature and converge their working set, so does the
sampling rate slow down to just a trickle, 256 MB per 8
seconds of CPU time executed.
This, beyond being adaptive, also rate-limits rarely
executing systems and does not over-sample on overloaded
systems.
[ In AutoNUMA speak, this patch deals with the effective sampling
rate of the 'hinting page fault'. AutoNUMA's scanning is
currently rate-limited, but it is also fundamentally
single-threaded, executing in the knuma_scand kernel thread,
so the limit in AutoNUMA is global and does not scale up with
the number of CPUs, nor does it scan tasks in an execution
proportional manner.
So the idea of rate-limiting the scanning was first implemented
in the AutoNUMA tree via a global rate limit. This patch goes
beyond that by implementing an execution rate proportional
working set sampling rate that is not implemented via a single
global scanning daemon. ]
[ Dan Carpenter pointed out a possible NULL pointer dereference in the
first version of this patch. ]
Based-on-idea-by: Andrea Arcangeli <aarcange@redhat.com>
Bug-Found-By: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Rik van Riel <riel@redhat.com>
[ Wrote changelog and fixed bug. ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Mel Gorman <mgorman@suse.de>
Reviewed-by: Rik van Riel <riel@redhat.com>
NOTE: This patch is based on "sched, numa, mm: Add fault driven
placement and migration policy" but as it throws away all the policy
to just leave a basic foundation I had to drop the signed-offs-by.
This patch creates a bare-bones method for setting PTEs pte_numa in the
context of the scheduler that when faulted later will be faulted onto the
node the CPU is running on. In itself this does nothing useful but any
placement policy will fundamentally depend on receiving hints on placement
from fault context and doing something intelligent about it.
Signed-off-by: Mel Gorman <mgorman@suse.de>
Acked-by: Rik van Riel <riel@redhat.com>
While per-entity load-tracking is generally useful, beyond computing shares
distribution, e.g. runnable based load-balance (in progress), governors,
power-management, etc.
These facilities are not yet consumers of this data. This may be trivially
reverted when the information is required; but avoid paying the overhead for
calculations we will not use until then.
Signed-off-by: Paul Turner <pjt@google.com>
Reviewed-by: Ben Segall <bsegall@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20120823141507.422162369@google.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
__update_entity_runnable_avg forms the core of maintaining an entity's runnable
load average. In this function we charge the accumulated run-time since last
update and handle appropriate decay. In some cases, e.g. a waking task, this
time interval may be much larger than our period unit.
Fortunately we can exploit some properties of our series to perform decay for a
blocked update in constant time and account the contribution for a running
update in essentially-constant* time.
[*]: For any running entity they should be performing updates at the tick which
gives us a soft limit of 1 jiffy between updates, and we can compute up to a
32 jiffy update in a single pass.
C program to generate the magic constants in the arrays:
#include <math.h>
#include <stdio.h>
#define N 32
#define WMULT_SHIFT 32
const long WMULT_CONST = ((1UL << N) - 1);
double y;
long runnable_avg_yN_inv[N];
void calc_mult_inv() {
int i;
double yn = 0;
printf("inverses\n");
for (i = 0; i < N; i++) {
yn = (double)WMULT_CONST * pow(y, i);
runnable_avg_yN_inv[i] = yn;
printf("%2d: 0x%8lx\n", i, runnable_avg_yN_inv[i]);
}
printf("\n");
}
long mult_inv(long c, int n) {
return (c * runnable_avg_yN_inv[n]) >> WMULT_SHIFT;
}
void calc_yn_sum(int n)
{
int i;
double sum = 0, sum_fl = 0, diff = 0;
/*
* We take the floored sum to ensure the sum of partial sums is never
* larger than the actual sum.
*/
printf("sum y^n\n");
printf(" %8s %8s %8s\n", "exact", "floor", "error");
for (i = 1; i <= n; i++) {
sum = (y * sum + y * 1024);
sum_fl = floor(y * sum_fl+ y * 1024);
printf("%2d: %8.0f %8.0f %8.0f\n", i, sum, sum_fl,
sum_fl - sum);
}
printf("\n");
}
void calc_conv(long n) {
long old_n;
int i = -1;
printf("convergence (LOAD_AVG_MAX, LOAD_AVG_MAX_N)\n");
do {
old_n = n;
n = mult_inv(n, 1) + 1024;
i++;
} while (n != old_n);
printf("%d> %ld\n", i - 1, n);
printf("\n");
}
void main() {
y = pow(0.5, 1/(double)N);
calc_mult_inv();
calc_conv(1024);
calc_yn_sum(N);
}
[ Compile with -lm ]
Signed-off-by: Paul Turner <pjt@google.com>
Reviewed-by: Ben Segall <bsegall@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20120823141507.277808946@google.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Now that our measurement intervals are small (~1ms) we can amortize the posting
of update_shares() to be about each period overflow. This is a large cost
saving for frequently switching tasks.
Signed-off-by: Paul Turner <pjt@google.com>
Reviewed-by: Ben Segall <bsegall@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20120823141507.200772172@google.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Now that running entities maintain their own load-averages the work we must do
in update_shares() is largely restricted to the periodic decay of blocked
entities. This allows us to be a little less pessimistic regarding our
occupancy on rq->lock and the associated rq->clock updates required.
Signed-off-by: Paul Turner <pjt@google.com>
Reviewed-by: Ben Segall <bsegall@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20120823141507.133999170@google.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Now that the machinery in place is in place to compute contributed load in a
bottom up fashion; replace the shares distribution code within update_shares()
accordingly.
Signed-off-by: Paul Turner <pjt@google.com>
Reviewed-by: Ben Segall <bsegall@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20120823141507.061208672@google.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
With bandwidth control tracked entities may cease execution according to user
specified bandwidth limits. Charging this time as either throttled or blocked
however, is incorrect and would falsely skew in either direction.
What we actually want is for any throttled periods to be "invisible" to
load-tracking as they are removed from the system for that interval and
contribute normally otherwise.
Do this by moderating the progression of time to omit any periods in which the
entity belonged to a throttled hierarchy.
Signed-off-by: Paul Turner <pjt@google.com>
Reviewed-by: Ben Segall <bsegall@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20120823141506.998912151@google.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Entities of equal weight should receive equitable distribution of cpu time.
This is challenging in the case of a task_group's shares as execution may be
occurring on multiple cpus simultaneously.
To handle this we divide up the shares into weights proportionate with the load
on each cfs_rq. This does not however, account for the fact that the sum of
the parts may be less than one cpu and so we need to normalize:
load(tg) = min(runnable_avg(tg), 1) * tg->shares
Where runnable_avg is the aggregate time in which the task_group had runnable
children.
Signed-off-by: Paul Turner <pjt@google.com>
Reviewed-by: Ben Segall <bsegall@google.com>.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20120823141506.930124292@google.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Unlike task entities who have a fixed weight, group entities instead own a
fraction of their parenting task_group's shares as their contributed weight.
Compute this fraction so that we can correctly account hierarchies and shared
entity nodes.
Signed-off-by: Paul Turner <pjt@google.com>
Reviewed-by: Ben Segall <bsegall@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20120823141506.855074415@google.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Maintain a global running sum of the average load seen on each cfs_rq belonging
to each task group so that it may be used in calculating an appropriate
shares:weight distribution.
Signed-off-by: Paul Turner <pjt@google.com>
Reviewed-by: Ben Segall <bsegall@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20120823141506.792901086@google.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
When a running entity blocks we migrate its tracked load to
cfs_rq->blocked_runnable_avg. In the sleep case this occurs while holding
rq->lock and so is a natural transition. Wake-ups however, are potentially
asynchronous in the presence of migration and so special care must be taken.
We use an atomic counter to track such migrated load, taking care to match this
with the previously introduced decay counters so that we don't migrate too much
load.
Signed-off-by: Paul Turner <pjt@google.com>
Reviewed-by: Ben Segall <bsegall@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20120823141506.726077467@google.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Since we are now doing bottom up load accumulation we need explicit
notification when a task has been re-parented so that the old hierarchy can be
updated.
Adds: migrate_task_rq(struct task_struct *p, int next_cpu)
(The alternative is to do this out of __set_task_cpu, but it was suggested that
this would be a cleaner encapsulation.)
Signed-off-by: Paul Turner <pjt@google.com>
Reviewed-by: Ben Segall <bsegall@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20120823141506.660023400@google.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
We are currently maintaining:
runnable_load(cfs_rq) = \Sum task_load(t)
For all running children t of cfs_rq. While this can be naturally updated for
tasks in a runnable state (as they are scheduled); this does not account for
the load contributed by blocked task entities.
This can be solved by introducing a separate accounting for blocked load:
blocked_load(cfs_rq) = \Sum runnable(b) * weight(b)
Obviously we do not want to iterate over all blocked entities to account for
their decay, we instead observe that:
runnable_load(t) = \Sum p_i*y^i
and that to account for an additional idle period we only need to compute:
y*runnable_load(t).
This means that we can compute all blocked entities at once by evaluating:
blocked_load(cfs_rq)` = y * blocked_load(cfs_rq)
Finally we maintain a decay counter so that when a sleeping entity re-awakens
we can determine how much of its load should be removed from the blocked sum.
Signed-off-by: Paul Turner <pjt@google.com>
Reviewed-by: Ben Segall <bsegall@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20120823141506.585389902@google.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
For a given task t, we can compute its contribution to load as:
task_load(t) = runnable_avg(t) * weight(t)
On a parenting cfs_rq we can then aggregate:
runnable_load(cfs_rq) = \Sum task_load(t), for all runnable children t
Maintain this bottom up, with task entities adding their contributed load to
the parenting cfs_rq sum. When a task entity's load changes we add the same
delta to the maintained sum.
Signed-off-by: Paul Turner <pjt@google.com>
Reviewed-by: Ben Segall <bsegall@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20120823141506.514678907@google.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Since runqueues do not have a corresponding sched_entity we instead embed a
sched_avg structure directly.
Signed-off-by: Ben Segall <bsegall@google.com>
Reviewed-by: Paul Turner <pjt@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20120823141506.442637130@google.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Instead of tracking averaging the load parented by a cfs_rq, we can track
entity load directly. With the load for a given cfs_rq then being the sum
of its children.
To do this we represent the historical contribution to runnable average
within each trailing 1024us of execution as the coefficients of a
geometric series.
We can express this for a given task t as:
runnable_sum(t) = \Sum u_i * y^i, runnable_avg_period(t) = \Sum 1024 * y^i
load(t) = weight_t * runnable_sum(t) / runnable_avg_period(t)
Where: u_i is the usage in the last i`th 1024us period (approximately 1ms)
~ms and y is chosen such that y^k = 1/2. We currently choose k to be 32 which
roughly translates to about a sched period.
Signed-off-by: Paul Turner <pjt@google.com>
Reviewed-by: Ben Segall <bsegall@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20120823141506.372695337@google.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
As per the recent discussion with Mike and Linus, make it easier to
test with/without this feature. No change in default behavior.
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: http://lkml.kernel.org/n/tip-izoxq4haeg4mTognnDbwcevt@git.kernel.org
Pull scheduler changes from Ingo Molnar:
"Continued quest to clean up and enhance the cputime code by Frederic
Weisbecker, in preparation for future tickless kernel features.
Other than that, smallish changes."
Fix up trivial conflicts due to additions next to each other in arch/{x86/}Kconfig
* 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (24 commits)
cputime: Make finegrained irqtime accounting generally available
cputime: Gather time/stats accounting config options into a single menu
ia64: Reuse system and user vtime accounting functions on task switch
ia64: Consolidate user vtime accounting
vtime: Consolidate system/idle context detection
cputime: Use a proper subsystem naming for vtime related APIs
sched: cpu_power: enable ARCH_POWER
sched/nohz: Clean up select_nohz_load_balancer()
sched: Fix load avg vs. cpu-hotplug
sched: Remove __ARCH_WANT_INTERRUPTS_ON_CTXSW
sched: Fix nohz_idle_balance()
sched: Remove useless code in yield_to()
sched: Add time unit suffix to sched sysctl knobs
sched/debug: Limit sd->*_idx range on sysctl
sched: Remove AFFINE_WAKEUPS feature flag
s390: Remove leftover account_tick_vtime() header
cputime: Consolidate vtime handling on context switch
sched: Move cputime code to its own file
cputime: Generalize CONFIG_VIRT_CPU_ACCOUNTING
tile: Remove SD_PREFER_LOCAL leftover
...
This reverts commit 970e178985.
Nikolay Ulyanitsky reported thatthe 3.6-rc5 kernel has a 15-20%
performance drop on PostgreSQL 9.2 on his machine (running "pgbench").
Borislav Petkov was able to reproduce this, and bisected it to this
commit 970e178985 ("sched: Improve scalability via 'CPU buddies' ...")
apparently because the new single-idle-buddy model simply doesn't find
idle CPU's to reschedule on aggressively enough.
Mike Galbraith suspects that it is likely due to the user-mode spinlocks
in PostgreSQL not reacting well to preemption, but we don't really know
the details - I'll just revert the commit for now.
There are hopefully other approaches to improve scheduler scalability
without it causing these kinds of downsides.
Reported-by: Nikolay Ulyanitsky <lystor@gmail.com>
Bisected-by: Borislav Petkov <bp@alien8.de>
Acked-by: Mike Galbraith <efault@gmx.de>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
There is no load_balancer to be selected now. It just sets the
state of the nohz tick to stop.
So rename the function, pass the 'cpu' as a parameter and then
remove the useless call from tick_nohz_restart_sched_tick().
[ s/set_nohz_tick_stopped/nohz_balance_enter_idle/g
s/clear_nohz_tick_stopped/nohz_balance_exit_idle/g ]
Signed-off-by: Alex Shi <alex.shi@intel.com>
Acked-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Venkatesh Pallipadi <venki@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1347261059-24747-1-git-send-email-alex.shi@intel.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
On tickless systems, one CPU runs load balance for all idle CPUs.
The cpu_load of this CPU is updated before starting the load balance
of each other idle CPUs. We should instead update the cpu_load of
the balance_cpu.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Venkatesh Pallipadi <venki@google.com>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Link: http://lkml.kernel.org/r/1347509486-8688-1-git-send-email-vincent.guittot@linaro.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Fix two kernel-doc warnings in kernel/sched/fair.c:
Warning(kernel/sched/fair.c:3660): Excess function parameter 'cpus' description in 'update_sg_lb_stats'
Warning(kernel/sched/fair.c:3806): Excess function parameter 'cpus' description in 'update_sd_lb_stats'
Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/50303714.3090204@xenotime.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
migrate_tasks() uses _pick_next_task_rt() to get tasks from the
real-time runqueues to be migrated. When rt_rq is throttled
_pick_next_task_rt() won't return anything, in which case
migrate_tasks() can't move all threads over and gets stuck in an
infinite loop.
Instead unthrottle rt runqueues before migrating tasks.
Additionally: move unthrottle_offline_cfs_rqs() to rq_offline_fair()
Signed-off-by: Peter Boonstoppel <pboonstoppel@nvidia.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Turner <pjt@google.com>
Link: http://lkml.kernel.org/r/5FBF8E85CA34454794F0F7ECBA79798F379D3648B7@HQMAIL04.nvidia.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Since power saving code was removed from sched now, the implement
code is out of service in this function, and even pollute other logical.
like, 'want_sd' never has chance to be set '0', that remove the effect
of SD_WAKE_AFFINE here.
So, clean up the obsolete code, includes SD_PREFER_LOCAL.
Signed-off-by: Alex Shi <alex.shi@intel.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/5028F431.6000306@intel.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
As we already have dst_rq in lb_env, using or changing "this_rq" do not
make sense.
This patch will replace "this_rq" with dst_rq in load_balance, and we
don't need to change "this_rq" while process LBF_SOME_PINNED any more.
Signed-off-by: Michael Wang <wangyun@linux.vnet.ibm.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/501F8357.3070102@linux.vnet.ibm.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
It should be sched_nr_latency so fix it before it annoys me more.
Signed-off-by: Borislav Petkov <borislav.petkov@amd.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1344435364-18632-1-git-send-email-bp@amd64.org
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Peter Portante reported that for large cgroup hierarchies (and or on
large CPU counts) we get immense lock contention on rq->lock and stuff
stops working properly.
His workload was a ton of processes, each in their own cgroup,
everybody idling except for a sporadic wakeup once every so often.
It was found that:
schedule()
idle_balance()
load_balance()
local_irq_save()
double_rq_lock()
update_h_load()
walk_tg_tree(tg_load_down)
tg_load_down()
Results in an entire cgroup hierarchy walk under rq->lock for every
new-idle balance and since new-idle balance isn't throttled this
results in a lot of work while holding the rq->lock.
This patch does two things, it removes the work from under rq->lock
based on the good principle of race and pray which is widely employed
in the load-balancer as a whole. And secondly it throttles the
update_h_load() calculation to max once per jiffy.
I considered excluding update_h_load() for new-idle balance
all-together, but purely relying on regular balance passes to update
this data might not work out under some rare circumstances where the
new-idle busiest isn't the regular busiest for a while (unlikely, but
a nightmare to debug if someone hits it and suffers).
Cc: pjt@google.com
Cc: Larry Woodman <lwoodman@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Reported-by: Peter Portante <pportant@redhat.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/n/tip-aaarrzfpnaam7pqrekofu8a6@git.kernel.org
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
With this patch struct ld_env will have a pointer of the load balancing
cpumask and we don't need to pass a cpumask around anymore.
Signed-off-by: Michael Wang <wangyun@linux.vnet.ibm.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/4FFE8665.3080705@linux.vnet.ibm.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Current load balance scheme requires only one cpu in a
sched_group (balance_cpu) to look at other peer sched_groups for
imbalance and pull tasks towards itself from a busy cpu. Tasks
thus pulled by balance_cpu could later get picked up by cpus
that are in the same sched_group as that of balance_cpu.
This scheme however fails to pull tasks that are not allowed to
run on balance_cpu (but are allowed to run on other cpus in its
sched_group). That can affect fairness and in some worst case
scenarios cause starvation.
Consider a two core (2 threads/core) system running tasks as
below:
Core0 Core1
/ \ / \
C0 C1 C2 C3
| | | |
v v v v
F0 T1 F1 [idle]
T2
F0 = SCHED_FIFO task (pinned to C0)
F1 = SCHED_FIFO task (pinned to C2)
T1 = SCHED_OTHER task (pinned to C1)
T2 = SCHED_OTHER task (pinned to C1 and C2)
F1 could become a cpu hog, which will starve T2 unless C1 pulls
it. Between C0 and C1 however, C0 is required to look for
imbalance between cores, which will fail to pull T2 towards
Core0. T2 will starve eternally in this case. The same scenario
can arise in presence of non-rt tasks as well (say we replace F1
with high irq load).
We tackle this problem by having balance_cpu move pinned tasks
to one of its sibling cpus (where they can run). We first check
if load balance goal can be met by ignoring pinned tasks,
failing which we retry move_tasks() with a new env->dst_cpu.
This patch modifies load balance semantics on who can move load
towards a given cpu in a given sched_domain.
Before this patch, a given_cpu or a ilb_cpu acting on behalf of
an idle given_cpu is responsible for moving load to given_cpu.
With this patch applied, balance_cpu can in addition decide on
moving some load to a given_cpu.
There is a remote possibility that excess load could get moved
as a result of this (balance_cpu and given_cpu/ilb_cpu deciding
*independently* and at *same* time to move some load to a
given_cpu). However we should see less of such conflicting
decisions in practice and moreover subsequent load balance
cycles should correct the excess load moved to given_cpu.
Signed-off-by: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Signed-off-by: Prashanth Nageshappa <prashanth@linux.vnet.ibm.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/4FE06CDB.2060605@linux.vnet.ibm.com
[ minor edits ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
While load balancing, if all tasks on the source runqueue are pinned,
we retry after excluding the corresponding source cpu. However, loop counters
env.loop and env.loop_break are not reset before retrying, which can lead
to failure in moving the tasks. In this patch we reset env.loop and
env.loop_break to their inital values before we retry.
Signed-off-by: Prashanth Nageshappa <prashanth@linux.vnet.ibm.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/4FE06EEF.2090709@linux.vnet.ibm.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Members of 'struct lb_env' are not in appropriate order to reuse compiler
added padding on 64bit architectures. In this patch we reorder those struct
members and help reduce the size of the structure from 96 bytes to 80
bytes on 64 bit architectures.
Suggested-by: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Signed-off-by: Prashanth Nageshappa <prashanth@linux.vnet.ibm.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/4FE06DDE.7000403@linux.vnet.ibm.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Traversing an entire package is not only expensive, it also leads to tasks
bouncing all over a partially idle and possible quite large package. Fix
that up by assigning a 'buddy' CPU to try to motivate. Each buddy may try
to motivate that one other CPU, if it's busy, tough, it may then try its
SMT sibling, but that's all this optimization is allowed to cost.
Sibling cache buddies are cross-wired to prevent bouncing.
4 socket 40 core + SMT Westmere box, single 30 sec tbench runs, higher is better:
clients 1 2 4 8 16 32 64 128
..........................................................................
pre 30 41 118 645 3769 6214 12233 14312
post 299 603 1211 2418 4697 6847 11606 14557
A nice increase in performance.
Signed-off-by: Mike Galbraith <efault@gmx.de>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1339471112.7352.32.camel@marge.simpson.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Fix lots of new kernel-doc warnings in kernel/sched/fair.c:
Warning(kernel/sched/fair.c:3625): No description found for parameter 'env'
Warning(kernel/sched/fair.c:3625): Excess function parameter 'sd' description in 'update_sg_lb_stats'
Warning(kernel/sched/fair.c:3735): No description found for parameter 'env'
Warning(kernel/sched/fair.c:3735): Excess function parameter 'sd' description in 'update_sd_pick_busiest'
Warning(kernel/sched/fair.c:3735): Excess function parameter 'this_cpu' description in 'update_sd_pick_busiest'
.. more warnings
Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Often when we run into mis-shapen topologies the balance iteration
fails to update the cpu power properly and we'll end up in /0 traps.
Always initialize the cpu-power to a semi-sane value so that we can
at least boot the machine, even if the load-balancer might not
function correctly.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/n/tip-3lbhyj25sr169ha7z3qht5na@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Weird topologies can lead to asymmetric domain setups. This needs
further consideration since these setups are typically non-minimal
too.
For now, make it work by adding an extra mask selecting which CPUs
are allowed to iterate up.
The topology that triggered it is the one from David Rientjes:
10 20 20 30
20 10 20 20
20 20 10 20
30 20 20 10
resulting in boxes that wouldn't even boot.
Reported-by: David Rientjes <rientjes@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/n/tip-3p86l9cuaqnxz7uxsojmz5rm@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Since nr_cpus_allowed is used outside of sched/rt.c and wants to be
used outside of there more, move it to a more natural site.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/n/tip-kr61f02y9brwzkh6x53pdptm@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
We could re-read rq->rt_avg after we validated it was smaller than
total, invalidating the check and resulting in an unintended negative.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: David Rientjes <rientjes@google.com>
Link: http://lkml.kernel.org/r/1337688268.9698.29.camel@twins
Signed-off-by: Ingo Molnar <mingo@kernel.org>
SD_OVERLAP exists to allow overlapping groups, overlapping groups
appear in NUMA topologies that aren't fully connected.
The typical result of not fully connected NUMA is that each cpu (or
rather node) will have different spans for a particular distance.
However due to how sched domains are traversed -- only the first cpu
in the mask goes one level up -- the next level only cares about the
spans of the cpus that went up.
Due to this two things were observed to be broken:
- build_overlap_sched_groups() -- since its possible the cpu we're
building the groups for exists in multiple (or all) groups, the
selection criteria of the first group didn't ensure there was a cpu
for which is was true that cpumask_first(span) == cpu. Thus load-
balancing would terminate.
- update_group_power() -- assumed that the cpu span of the first
group of the domain was covered by all groups of the child domain.
The above explains why this isn't true, so deal with it.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: David Rientjes <rientjes@google.com>
Link: http://lkml.kernel.org/r/1337788843.9783.14.camel@laptop
Signed-off-by: Ingo Molnar <mingo@kernel.org>
It's been broken forever (i.e. it's not scheduling in a power
aware fashion), as reported by Suresh and others sending
patches, and nobody cares enough to fix it properly ...
so remove it to make space free for something better.
There's various problems with the code as it stands today, first
and foremost the user interface which is bound to topology
levels and has multiple values per level. This results in a
state explosion which the administrator or distro needs to
master and almost nobody does.
Furthermore large configuration state spaces aren't good, it
means the thing doesn't just work right because it's either
under so many impossibe to meet constraints, or even if
there's an achievable state workloads have to be aware of
it precisely and can never meet it for dynamic workloads.
So pushing this kind of decision to user-space was a bad idea
even with a single knob - it's exponentially worse with knobs
on every node of the topology.
There is a proposal to replace the user interface with a single
3 state knob:
sched_balance_policy := { performance, power, auto }
where 'auto' would be the preferred default which looks at things
like Battery/AC mode and possible cpufreq state or whatever the hw
exposes to show us power use expectations - but there's been no
progress on it in the past many months.
Aside from that, the actual implementation of the various knobs
is known to be broken. There have been sporadic attempts at
fixing things but these always stop short of reaching a mergable
state.
Therefore this wholesale removal with the hopes of spurring
people who care to come forward once again and work on a
coherent replacement.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Arjan van de Ven <arjan@linux.intel.com>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Vaidyanathan Srinivasan <svaidy@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/1326104915.2442.53.camel@twins
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Group imbalance is meant to deal with situations where affinity masks
and sched domains don't align well, such as 3 cpus from one group and
6 from another. In this case the domain based balancer will want to
put an equal amount of tasks on each side even though they don't have
equal cpus.
Currently group_imb is set whenever two cpus of a group have a weight
difference of at least one avg task and the heaviest cpu has at least
two tasks. A group with imbalance set will always be picked as busiest
and a balance pass will be forced.
The problem is that even if there are no affinity masks this stuff can
trigger and cause weird balancing decisions, eg. the observed
behaviour was that of 6 cpus, 5 had 2 and 1 had 3 tasks, due to the
difference of 1 avg load (they all had the same weight) and nr_running
being >1 the group_imbalance logic triggered and did the weird thing
of pulling more load instead of trying to move the 1 excess task to
the other domain of 6 cpus that had 5 cpu with 2 tasks and 1 cpu with
1 task.
Curb the group_imbalance stuff by making the nr_running condition
weaker by also tracking the min_nr_running and using the difference in
nr_running over the set instead of the absolute max nr_running.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/n/tip-9s7dedozxo8kjsb9kqlrukkf@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
While investigating why the load-balancer did funny I found that the
rq->cpu_load[] tables were completely screwy.. a bit more digging
revealed that the updates that got through were missing ticks followed
by a catchup of 2 ticks.
The catchup assumes the cpu was idle during that time (since only nohz
can cause missed ticks and the machine is idle etc..) this means that
esp. the higher indices were significantly lower than they ought to
be.
The reason for this is that its not correct to compare against jiffies
on every jiffy on any other cpu than the cpu that updates jiffies.
This patch cludges around it by only doing the catch-up stuff from
nohz_idle_balance() and doing the regular stuff unconditionally from
the tick.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: pjt@google.com
Cc: Venkatesh Pallipadi <venki@google.com>
Link: http://lkml.kernel.org/n/tip-tp4kj18xdd5aj4vvj0qg55s2@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Patches c22402a2f ("sched/fair: Let minimally loaded cpu balance the
group") and 0ce90475 ("sched/fair: Add some serialization to the
sched_domain load-balance walk") are horribly broken so revert them.
The problem is that while it sounds good to have the minimally loaded
cpu do the pulling of more load, the way we walk the domains there is
absolutely no guarantee this cpu will actually get to the domain. In
fact its very likely it wont. Therefore the higher up the tree we get,
the less likely it is we'll balance at all.
The first of mask always walks up, while sucky in that it accumulates
load on the first cpu and needs extra passes to spread it out at least
guarantees a cpu gets up that far and load-balancing happens at all.
Since its now always the first and idle cpus should always be able to
balance so they get a task as fast as possible we can also do away
with the added serialization.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/n/tip-rpuhs5s56aiv1aw7khv9zkw6@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Since the sched_domain walk is completely unserialized (!SD_SERIALIZE)
it is possible that multiple cpus in the group get elected to do the
next level. Avoid this by adding some serialization.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/n/tip-vqh9ai6s0ewmeakjz80w4qz6@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Currently we let the leftmost (or first idle) cpu ascend the
sched_domain tree and perform load-balancing. The result is that the
busiest cpu in the group might be performing this function and pull
more load to itself. The next load balance pass will then try to
equalize this again.
Change this to pick the least loaded cpu to perform higher domain
balancing.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/n/tip-v8zlrmgmkne3bkcy9dej1fvm@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Since there's a PID space limit of 30bits (see
futex.h:FUTEX_TID_MASK) and allocating that many tasks (assuming a
lower bound of 2 pages per task) would still take 8T of memory it
seems reasonable to say that unsigned int is sufficient for
rq->nr_running.
When we do get anywhere near that amount of tasks I suspect other
things would go funny, load-balancer load computations would really
need to be hoisted to 128bit etc.
So save a few bytes and convert rq->nr_running and friends to
unsigned int.
Suggested-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/n/tip-y3tvyszjdmbibade5bw8zl81@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Commits 367456c756 ("sched: Ditch per cgroup task lists for
load-balancing") and 5d6523ebd ("sched: Fix load-balance wreckage")
left some more wreckage.
By setting loop_max unconditionally to ->nr_running load-balancing
could take a lot of time on very long runqueues (hackbench!). So keep
the sysctl as max limit of the amount of tasks we'll iterate.
Furthermore, the min load filter for migration completely fails with
cgroups since inequality in per-cpu state can easily lead to such
small loads :/
Furthermore the change to add new tasks to the tail of the queue
instead of the head seems to have some effect.. not quite sure I
understand why.
Combined these fixes solve the huge hackbench regression reported by
Tim when hackbench is ran in a cgroup.
Reported-by: Tim Chen <tim.c.chen@linux.intel.com>
Acked-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/1335365763.28150.267.camel@twins
[ got rid of the CONFIG_PREEMPT tuning and made small readability edits ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull scheduler fixes from Ingo Molnar.
* 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
cpusets: Remove an unused variable
sched/rt: Improve pick_next_highest_task_rt()
sched: Fix select_fallback_rq() vs cpu_active/cpu_online
sched/x86/smp: Do not enable IRQs over calibrate_delay()
sched: Fix compiler warning about declared inline after use
MAINTAINERS: Update email address for SCHEDULER and PERF EVENTS
kernel/sched/fair.c:420: warning: 'account_cfs_rq_runtime' declared inline after being called
kernel/sched/fair.c:420: warning: previous declaration of 'account_cfs_rq_runtime' was here
kernel/sched/fair.c:1165: warning: 'return_cfs_rq_runtime' declared inlineafter being called
kernel/sched/fair.c:1165: warning: previous declaration of 'return_cfs_rq_runtime' was here
Reported-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/20120321200717.49BB4A024E@akpm.mtv.corp.google.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull scheduler changes for v3.4 from Ingo Molnar
* 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (27 commits)
printk: Make it compile with !CONFIG_PRINTK
sched/x86: Fix overflow in cyc2ns_offset
sched: Fix nohz load accounting -- again!
sched: Update yield() docs
printk/sched: Introduce special printk_sched() for those awkward moments
sched/nohz: Correctly initialize 'next_balance' in 'nohz' idle balancer
sched: Cleanup cpu_active madness
sched: Fix load-balance wreckage
sched: Clean up parameter passing of proc_sched_autogroup_set_nice()
sched: Ditch per cgroup task lists for load-balancing
sched: Rename load-balancing fields
sched: Move load-balancing arguments into helper struct
sched/rt: Do not submit new work when PI-blocked
sched/rt: Prevent idle task boosting
sched/wait: Add __wake_up_all_locked() API
sched/rt: Document scheduler related skip-resched-check sites
sched/rt: Use schedule_preempt_disabled()
sched/rt: Add schedule_preempt_disabled()
sched/rt: Do not throttle when PI boosting
sched/rt: Keep period timer ticking when rt throttling is active
...
The 'next_balance' field of 'nohz' idle balancer must be initialized
to jiffies. Since jiffies is initialized to negative 300 seconds the
'nohz' idle balancer does not run for the first 300s (5mins) after
bootup. If no new processes are spawed or no idle cycles happen, the
load on the cpus will remain unbalanced for that duration.
Signed-off-by: Diwakar Tundlam <dtundlam@nvidia.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1DD7BFEDD3147247B1355BEFEFE4665237994F30EF@HQMAIL04.nvidia.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Commit 367456c ("sched: Ditch per cgroup task lists for
load-balancing") completely wrecked load-balancing due to
a few silly mistakes.
Correct those and remove more pointless code.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/n/tip-zk04ihygwxn7qqrlpaf73b0r@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Per cgroup load-balance has numerous problems, chief amongst them that
there is no real sane order in them. So stop pretending it makes sense
and enqueue all tasks on a single list.
This also allows us to more easily fix the fwd progress issue
uncovered by the lock-break stuff. Rotate the list on failure to
migreate and limit the total iterations to nr_running (which with
releasing the lock isn't strictly accurate but close enough).
Also add a filter that skips very light tasks on the first attempt
around the list, this attempts to avoid shooting whole cgroups around
without affecting over balance.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: pjt@google.com
Link: http://lkml.kernel.org/n/tip-tx8yqydc7eimgq7i4rkc3a4g@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Passing large sets of similar arguments all around the load-balancer
gets tiresom when you want to modify something. Stick them all in a
helper structure and pass the structure around.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: pjt@google.com
Link: http://lkml.kernel.org/n/tip-5slqz0vhsdzewrfk9eza1aon@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@elte.hu>
So here's a boot tested patch on top of Jason's series that does
all the cleanups I talked about and turns jump labels into a
more intuitive to use facility. It should also address the
various misconceptions and confusions that surround jump labels.
Typical usage scenarios:
#include <linux/static_key.h>
struct static_key key = STATIC_KEY_INIT_TRUE;
if (static_key_false(&key))
do unlikely code
else
do likely code
Or:
if (static_key_true(&key))
do likely code
else
do unlikely code
The static key is modified via:
static_key_slow_inc(&key);
...
static_key_slow_dec(&key);
The 'slow' prefix makes it abundantly clear that this is an
expensive operation.
I've updated all in-kernel code to use this everywhere. Note
that I (intentionally) have not pushed through the rename
blindly through to the lowest levels: the actual jump-label
patching arch facility should be named like that, so we want to
decouple jump labels from the static-key facility a bit.
On non-jump-label enabled architectures static keys default to
likely()/unlikely() branches.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Acked-by: Jason Baron <jbaron@redhat.com>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Cc: a.p.zijlstra@chello.nl
Cc: mathieu.desnoyers@efficios.com
Cc: davem@davemloft.net
Cc: ddaney.cavm@gmail.com
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: http://lkml.kernel.org/r/20120222085809.GA26397@elte.hu
Signed-off-by: Ingo Molnar <mingo@elte.hu>
select_idle_sibling() is called from select_task_rq_fair(), which
already has the RCU read lock held.
Signed-off-by: Nikunj A. Dadhania <nikunj@linux.vnet.ibm.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20120217030409.11748.12491.stgit@abhimanyu
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Commit 1ac9bc69 ("sched/tracing: Add a new tracepoint for sleeptime")
added a new sched:sched_stat_sleeptime tracepoint.
It's broken: the first sample we get on a task might be bad because
of a stale sleep_start value that wasn't reset at the last task switch
because the tracepoint was not active.
It also breaks the existing schedstat samples due to the side
effects of:
- se->statistics.sleep_start = 0;
...
- se->statistics.block_start = 0;
Nor do I see means to fix it without adding overhead to the scheduler
fast path, which I'm not willing to for the sake of redundant
instrumentation.
Most importantly, sleep time information can already be constructed
by tracing context switches and wakeups, and taking the timestamp
difference between the schedule-out, the wakeup and the schedule-in.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Andrew Vagin <avagin@openvz.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Link: http://lkml.kernel.org/n/tip-pc4c9qhl8q6vg3bs4j6k0rbd@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This also fixes the following compilation warning on !SMP:
CC kernel/sched/fair.o
kernel/sched/fair.c:218:36: warning: 'max_load_balance_interval' defined but not used [-Wunused-variable]
Signed-off-by: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/4F2754A0.9090306@ct.jp.nec.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
With a lot of small tasks, the softirq sched is nearly never called
when no_hz is enabled. In this case load_balance() is mainly called
with the newly_idle mode which doesn't update the cpu_power.
Add a next_update field which ensure a maximum update period when
there is short activity.
Having stale cpu_power information can skew the load-balancing
decisions, this is cured by the guaranteed update.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1323717668-2143-1-git-send-email-vincent.guittot@linaro.org
With the recent nohz scheduler changes, rq's nohz flag
'NOHZ_TICK_STOPPED' and its associated state doesn't get cleared
immediately after the cpu exits idle. This gets cleared as part
of the next tick seen on that cpu.
For the cpu offline support, we need to clear this state
manually. Fix it by registering a cpu notifier, which clears the
nohz idle load balance state for this rq explicitly during the
CPU_DYING notification.
There won't be any nohz updates for that cpu, after the
CPU_DYING notification. But lets be extra paranoid and skip
updating the nohz state in the select_nohz_load_balancer() if
the cpu is not in active state anymore.
Reported-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com>
Reviewed-and-tested-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com>
Tested-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1327026538.16150.40.camel@sbsiddha-desk.sc.intel.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Eric and David reported dead machines and traced it to commit
a195f004 ("sched: Fix load-balance lock-breaking"), it turns out
there's still a scenario where we can end up re-trying forever.
Since there is no strict forward progress guarantee in the
load-balance iteration we can get stuck re-retrying the same
task-set over and over.
Creating a forward progress guarantee with the existing
structure is somewhat non-trivial, for now simply terminate the
retry loop after a few tries.
Reported-by: Eric Dumazet <eric.dumazet@gmail.com>
Tested-by: Eric Dumazet <eric.dumazet@gmail.com>
Reported-by: David Ahern <dsahern@gmail.com>
[ logic cleanup as suggested by Eric ]
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Link: http://lkml.kernel.org/r/1326297936.2442.157.camel@twins
Signed-off-by: Ingo Molnar <mingo@elte.hu>
If CONFIG_SCHEDSTATS is defined, the kernel maintains
information about how long the task was sleeping or
in the case of iowait, blocking in the kernel before
getting woken up.
This will be useful for sleep time profiling.
Note: this information is only provided for sched_fair.
Other scheduling classes may choose to provide this in
the future.
Note: the delay includes the time spent on the runqueue
as well.
Signed-off-by: Arun Sharma <asharma@fb.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Andrew Vagin <avagin@openvz.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Link: http://lkml.kernel.org/r/1324512940-32060-2-git-send-email-asharma@fb.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
There is a small race between try_to_wake_up() and sched_move_task(),
which is trying to move the process being woken up.
try_to_wake_up() on CPU0 sched_move_task() on CPU1
--------------------------------+---------------------------------
raw_spin_lock_irqsave(p->pi_lock)
task_waking_fair()
->p.se.vruntime -= cfs_rq->min_vruntime
ttwu_queue()
->send reschedule IPI to CPU1
raw_spin_unlock_irqsave(p->pi_lock)
task_rq_lock()
-> tring to aquire both p->pi_lock and
rq->lock with IRQ disabled
task_move_group_fair()
-> p.se.vruntime
-= (old)cfs_rq->min_vruntime
+= (new)cfs_rq->min_vruntime
task_rq_unlock()
(via IPI)
sched_ttwu_pending()
raw_spin_lock(rq->lock)
ttwu_do_activate()
...
enqueue_entity()
child.se->vruntime += cfs_rq->min_vruntime
raw_spin_unlock(rq->lock)
As a result, vruntime of the process becomes far bigger than min_vruntime,
if (new)cfs_rq->min_vruntime >> (old)cfs_rq->min_vruntime.
This patch fixes this problem by just ignoring such process in
task_move_group_fair(), because the vruntime has already been normalized in
task_waking_fair().
Signed-off-by: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Tejun Heo <tj@kernel.org>
Link: http://lkml.kernel.org/r/20111215143741.df82dd50.nishimura@mxp.nes.nec.co.jp
Signed-off-by: Ingo Molnar <mingo@elte.hu>
There is a small race between do_fork() and sched_move_task(), which is
trying to move the child.
do_fork() sched_move_task()
--------------------------------+---------------------------------
copy_process()
sched_fork()
task_fork_fair()
-> vruntime of the child is initialized
based on that of the parent.
-> we can see the child in "tasks" file now.
task_rq_lock()
task_move_group_fair()
-> child.se.vruntime
-= (old)cfs_rq->min_vruntime
+= (new)cfs_rq->min_vruntime
task_rq_unlock()
wake_up_new_task()
...
enqueue_entity()
child.se.vruntime += cfs_rq->min_vruntime
As a result, vruntime of the child becomes far bigger than min_vruntime,
if (new)cfs_rq->min_vruntime >> (old)cfs_rq->min_vruntime.
This patch fixes this problem by just ignoring such process in
task_move_group_fair(), because the vruntime has already been normalized in
task_fork_fair().
Signed-off-by: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Tejun Heo <tj@kernel.org>
Link: http://lkml.kernel.org/r/20111215143607.2ee12c5d.nishimura@mxp.nes.nec.co.jp
Signed-off-by: Ingo Molnar <mingo@elte.hu>
There is a small race between task_fork_fair() and sched_move_task(),
which is trying to move the parent.
task_fork_fair() sched_move_task()
--------------------------------+---------------------------------
cfs_rq = task_cfs_rq(current)
-> cfs_rq is the "old" one.
curr = cfs_rq->curr
-> curr is set to the parent.
task_rq_lock()
dequeue_task()
->parent.se.vruntime -= (old)cfs_rq->min_vruntime
enqueue_task()
->parent.se.vruntime += (new)cfs_rq->min_vruntime
task_rq_unlock()
raw_spin_lock_irqsave(rq->lock)
se->vruntime = curr->vruntime
-> vruntime of the child is set to that of the parent
which has already been updated by sched_move_task().
se->vruntime -= (old)cfs_rq->min_vruntime.
raw_spin_unlock_irqrestore(rq->lock)
As a result, vruntime of the child becomes far bigger than expected,
if (new)cfs_rq->min_vruntime >> (old)cfs_rq->min_vruntime.
This patch fixes this problem by setting "cfs_rq" and "curr" after
holding the rq->lock.
Signed-off-by: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Acked-by: Paul Turner <pjt@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Tejun Heo <tj@kernel.org>
Link: http://lkml.kernel.org/r/20111215143655.662676b0.nishimura@mxp.nes.nec.co.jp
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The current lock break relies on contention on the rq locks, something
which might never come because we've got IRQs disabled. Or will be
very likely because on anything with more than 2 cpus a synchronized
load-balance pass will very likely cause contention on the rq locks.
Also the sched_nr_migrate thing fails when it gets trapped the loops
of either the cgroup muck in load_balance_fair() or the move_tasks()
load condition.
Instead, use the new lb_flags field to propagate break/abort
conditions for all these loops and create a new loop outside the irq
disabled on the break being required.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/n/tip-tsceb6w61q0gakmsccix6xxi@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Replace the all_pinned argument with a flags field so that we can add
some extra controls throughout that entire call chain.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/n/tip-33kevm71m924ok1gpxd720v3@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Mike reported a 13% drop in netperf TCP_RR performance due to the
new remote wakeup code. Suresh too noticed some performance issues
with it.
Reducing the IPIs to only cross cache domains solves the observed
performance issues.
Reported-by: Suresh Siddha <suresh.b.siddha@intel.com>
Reported-by: Mike Galbraith <efault@gmx.de>
Acked-by: Suresh Siddha <suresh.b.siddha@intel.com>
Acked-by: Mike Galbraith <efault@gmx.de>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Chris Mason <chris.mason@oracle.com>
Cc: Dave Kleikamp <dave.kleikamp@oracle.com>
Link: http://lkml.kernel.org/r/1323338531.17673.7.camel@twins
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Yong Zhang reported:
> [ INFO: suspicious RCU usage. ]
> kernel/sched/fair.c:5091 suspicious rcu_dereference_check() usage!
This is due to the sched_domain stuff being RCU protected and
commit 0b005cf5 ("sched, nohz: Implement sched group, domain
aware nohz idle load balancing") overlooking this fact.
The sd variable only lives inside the for_each_domain() block,
so we only need to wrap that.
Reported-by: Yong Zhang <yong.zhang0@gmail.com>
Tested-by: Yong Zhang <yong.zhang0@gmail.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Link: http://lkml.kernel.org/r/1323264728.32012.107.camel@twins
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Intention is to set the NOHZ_BALANCE_KICK flag for the 'ilb_cpu'. Not
for the 'cpu' which is the local cpu. Fix the typo.
Reported-by: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1323199594.1984.18.camel@sbsiddha-desk.sc.intel.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
cpu bit in the nohz.idle_cpu_mask are reset in the first busy tick after
exiting idle. So during nohz_idle_balance(), intention is to double
check if the cpu that is part of the idle_cpu_mask is indeed idle before
going ahead in performing idle balance for that cpu.
Fix the cpu typo in the idle_cpu() check during nohz_idle_balance().
Reported-by: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1323199177.1984.12.camel@sbsiddha-desk.sc.intel.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
nr_busy_cpus in the sched_group_power indicates whether the group
is semi idle or not. This helps remove the is_semi_idle_group() and simplify
the find_new_ilb() in the context of finding an optimal cpu that can do
idle load balancing.
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/20111202010832.656983582@sbsiddha-desk.sc.intel.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
When there are many logical cpu's that enter and exit idle often, members of
the global nohz data structure are getting modified very frequently causing
lot of cache-line contention.
Make the nohz idle load balancing more scalabale by using the sched domain
topology and 'nr_busy_cpu's in the struct sched_group_power.
Idle load balance is kicked on one of the idle cpu's when there is atleast
one idle cpu and:
- a busy rq having more than one task or
- a busy rq's scheduler group that share package resources (like HT/MC
siblings) and has more than one member in that group busy or
- for the SD_ASYM_PACKING domain, if the lower numbered cpu's in that
domain are idle compared to the busy ones.
This will help in kicking the idle load balancing request only when
there is a potential imbalance. And once it is mostly balanced, these kicks will
be minimized.
These changes helped improve the workload that is context switch intensive
between number of task pairs by 2x on a 8 socket NHM-EX based system.
Reported-by: Tim Chen <tim.c.chen@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/20111202010832.602203411@sbsiddha-desk.sc.intel.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Introduce nr_busy_cpus in the struct sched_group_power [Not in sched_group
because sched groups are duplicated for the SD_OVERLAP scheduler domain]
and for each cpu that enters and exits idle, this parameter will
be updated in each scheduler group of the scheduler domain that this cpu
belongs to.
To avoid the frequent update of this state as the cpu enters
and exits idle, the update of the stat during idle exit is
delayed to the first timer tick that happens after the cpu becomes busy.
This is done using NOHZ_IDLE flag in the struct rq's nohz_flags.
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/20111202010832.555984323@sbsiddha-desk.sc.intel.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Introduce nohz_flags in the struct rq, which will track these two flags
for now.
NOHZ_TICK_STOPPED keeps track of the tick stopped status that gets set when
the tick is stopped. It will be used to update the nohz idle load balancer data
structures during the first busy tick after the tick is restarted. At this
first busy tick after tickless idle, NOHZ_TICK_STOPPED flag will be reset.
This will minimize the nohz idle load balancer status updates that currently
happen for every tickless exit, making it more scalable when there
are many logical cpu's that enter and exit idle often.
NOHZ_BALANCE_KICK will track the need for nohz idle load balance
on this rq. This will replace the nohz_balance_kick in the rq, which was
not being updated atomically.
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/20111202010832.499438999@sbsiddha-desk.sc.intel.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This is another case where we are on our way to schedule(),
so can save a useless clock update and resulting microscopic
vruntime update.
Signed-off-by: Mike Galbraith <efault@gmx.de>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1321971686.6855.18.camel@marge.simson.net
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Instead of going through the scheduler domain hierarchy multiple times
(for giving priority to an idle core over an idle SMT sibling in a busy
core), start with the highest scheduler domain with the SD_SHARE_PKG_RESOURCES
flag and traverse the domain hierarchy down till we find an idle group.
This cleanup also addresses an issue reported by Mike where the recent
changes returned the busy thread even in the presence of an idle SMT
sibling in single socket platforms.
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Tested-by: Mike Galbraith <efault@gmx.de>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1321556904.15339.25.camel@sbsiddha-desk.sc.intel.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This tracepoint shows how long a task is sleeping in uninterruptible state.
E.g. it may show how long and where a mutex is waited for.
Signed-off-by: Andrew Vagin <avagin@openvz.org>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1322471015-107825-8-git-send-email-avagin@openvz.org
Signed-off-by: Ingo Molnar <mingo@elte.hu>
There's too many sched*.[ch] files in kernel/, give them their own
directory.
(No code changed, other than Makefile glue added.)
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Linus Torvalds
Dave Kleikamp
Peter Zijlstra
Michael Wang
Vladimir Davydov
Kirill Tkhai
Yacine Belkadi
Paul Gortmaker
Kamalesh Babulal
Alex Shi
Frederic Weisbecker
Ingo Molnar
Nathan Zimmer
Vincent Guittot
Joonsoo Kim
Libin
Zhang Hang
Andrei Epure
Li Zefan
Mike Galbraith
Arnd Bergmann
Viresh Kumar
Zhu Yanhai
Hugh Dickins
Mel Gorman
Peter Zijlstra
Paul Turner
Ben Segall
Ingo Molnar
Randy Dunlap
Peter Boonstoppel
Borislav Petkov
Srivatsa Vaddagiri
Prashanth Nageshappa
Mike Galbraith
Diwakar Tundlam
Nikunj A. Dadhania
Hiroshi Shimamoto
Suresh Siddha
Arun Sharma
Daisuke Nishimura
Mike Galbraith
Andrew Vagin