Provide a generic p->on_rq because the p->se.on_rq semantics are
unfavourable for lockless wakeups but needed for sched_fair.
In particular, p->on_rq is only cleared when we actually dequeue the
task in schedule() and not on any random dequeue as done by things
like __migrate_task() and __sched_setscheduler().
This also allows us to remove p->se usage from !sched_fair code.
Reviewed-by: Frank Rowand <[email protected]>
Cc: Mike Galbraith <[email protected]>
Cc: Nick Piggin <[email protected]>
Cc: Linus Torvalds <[email protected]>
Cc: Andrew Morton <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
Signed-off-by: Peter Zijlstra <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Use the buddy mechanism to implement yield_task_fair. This
allows us to skip onto the next highest priority se at every
level in the CFS tree, unless doing so would introduce gross
unfairness in CPU time distribution.
We order the buddy selection in pick_next_entity to check
yield first, then last, then next. We need next to be able
to override yield, because it is possible for the "next" and
"yield" task to be different processen in the same sub-tree
of the CFS tree. When they are, we need to go into that
sub-tree regardless of the "yield" hint, and pick the correct
entity once we get to the right level.
Signed-off-by: Rik van Riel <[email protected]>
Signed-off-by: Peter Zijlstra <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
Now rq->rq_sched_info.bkl_count is not used for rq, scroll
rq->bkl_count into it. Thus we can save some space for rq.
Signed-off-by: Yong Zhang <[email protected]>
Signed-off-by: Peter Zijlstra <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
Displaying of group names in /proc/sched_debug was dropped in autogroup
patches. Add group names while displaying cfs_rq and tasks information.
Signed-off-by: Bharata B Rao <[email protected]>
Signed-off-by: Peter Zijlstra <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
A recurring complaint from CFS users is that parallel kbuild has
a negative impact on desktop interactivity. This patch
implements an idea from Linus, to automatically create task
groups. Currently, only per session autogroups are implemented,
but the patch leaves the way open for enhancement.
Implementation: each task's signal struct contains an inherited
pointer to a refcounted autogroup struct containing a task group
pointer, the default for all tasks pointing to the
init_task_group. When a task calls setsid(), a new task group
is created, the process is moved into the new task group, and a
reference to the preveious task group is dropped. Child
processes inherit this task group thereafter, and increase it's
refcount. When the last thread of a process exits, the
process's reference is dropped, such that when the last process
referencing an autogroup exits, the autogroup is destroyed.
At runqueue selection time, IFF a task has no cgroup assignment,
its current autogroup is used.
Autogroup bandwidth is controllable via setting it's nice level
through the proc filesystem:
cat /proc/<pid>/autogroup
Displays the task's group and the group's nice level.
echo <nice level> > /proc/<pid>/autogroup
Sets the task group's shares to the weight of nice <level> task.
Setting nice level is rate limited for !admin users due to the
abuse risk of task group locking.
The feature is enabled from boot by default if
CONFIG_SCHED_AUTOGROUP=y is selected, but can be disabled via
the boot option noautogroup, and can also be turned on/off on
the fly via:
echo [01] > /proc/sys/kernel/sched_autogroup_enabled
... which will automatically move tasks to/from the root task group.
Signed-off-by: Mike Galbraith <[email protected]>
Acked-by: Linus Torvalds <[email protected]>
Acked-by: Peter Zijlstra <[email protected]>
Cc: Markus Trippelsdorf <[email protected]>
Cc: Mathieu Desnoyers <[email protected]>
Cc: Paul Turner <[email protected]>
Cc: Oleg Nesterov <[email protected]>
[ Removed the task_group_path() debug code, and fixed !EVENTFD build failure. ]
Signed-off-by: Ingo Molnar <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
Add more clock information to /proc/sched_debug, Thomas wanted to see
the sched_clock_stable state.
Requested-by: Thomas Gleixner <[email protected]>
Signed-off-by: Peter Zijlstra <[email protected]>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <[email protected]>
By tracking a per-cpu load-avg for each cfs_rq and folding it into a
global task_group load on each tick we can rework tg_shares_up to be
strictly per-cpu.
This should improve cpu-cgroup performance for smp systems
significantly.
[ Paul: changed to use queueing cfs_rq + bug fixes ]
Signed-off-by: Paul Turner <[email protected]>
Signed-off-by: Peter Zijlstra <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
Trivial, use get_nr_threads() helper to read signal->count which we are
going to change.
Like other callers, proc_sched_show_task() doesn't need the exactly
precise nr_threads.
David said:
: Note that get_nr_threads() isn't completely equivalent (it can return 0
: where proc_sched_show_task() will display a 1). But I don't think this
: should be a problem.
Signed-off-by: Oleg Nesterov <[email protected]>
Acked-by: David Howells <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Acked-by: Roland McGrath <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
* 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (49 commits)
stop_machine: Move local variable closer to the usage site in cpu_stop_cpu_callback()
sched, wait: Use wrapper functions
sched: Remove a stale comment
ondemand: Make the iowait-is-busy time a sysfs tunable
ondemand: Solve a big performance issue by counting IOWAIT time as busy
sched: Intoduce get_cpu_iowait_time_us()
sched: Eliminate the ts->idle_lastupdate field
sched: Fold updating of the last_update_time_info into update_ts_time_stats()
sched: Update the idle statistics in get_cpu_idle_time_us()
sched: Introduce a function to update the idle statistics
sched: Add a comment to get_cpu_idle_time_us()
cpu_stop: add dummy implementation for UP
sched: Remove rq argument to the tracepoints
rcu: need barrier() in UP synchronize_sched_expedited()
sched: correctly place paranioa memory barriers in synchronize_sched_expedited()
sched: kill paranoia check in synchronize_sched_expedited()
sched: replace migration_thread with cpu_stop
stop_machine: reimplement using cpu_stop
cpu_stop: implement stop_cpu[s]()
sched: Fix select_idle_sibling() logic in select_task_rq_fair()
...
With CONFIG_PROVE_RCU=y, a warning can be triggered:
$ cat /proc/sched_debug
...
kernel/cgroup.c:1649 invoked rcu_dereference_check() without protection!
...
Both cgroup_path() and task_group() should be called with either
rcu_read_lock or cgroup_mutex held.
The rcu_dereference_check() does include cgroup_lock_is_held(), so we
know that this lock is not held. Therefore, in a CONFIG_PREEMPT kernel,
to say nothing of a CONFIG_PREEMPT_RT kernel, the original code could
have ended up copying a string out of the freelist.
This patch inserts RCU read-side primitives needed to prevent this
scenario.
Signed-off-by: Li Zefan <[email protected]>
Signed-off-by: Paul E. McKenney <[email protected]>
Latencytop clearing sum_exec_runtime via proc_sched_set_task() breaks
task_times(). Other places in kernel use nvcsw and nivcsw, which are
being cleared as well, Clear task statistics only.
Reported-by: Török Edwin <[email protected]>
Signed-off-by: Mike Galbraith <[email protected]>
Cc: Hidetoshi Seto <[email protected]>
Cc: Arjan van de Ven <[email protected]>
Signed-off-by: Peter Zijlstra <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
Both avg_overlap and avg_wakeup had an inherent problem in that their accuracy
was detrimentally affected by cross-cpu wakeups, this because we are missing
the necessary call to update_curr(). This can't be fixed without increasing
overhead in our already too fat fastpath.
Additionally, with recent load balancing changes making us prefer to place tasks
in an idle cache domain (which is good for compute bound loads), communicating
tasks suffer when a sync wakeup, which would enable affine placement, is turned
into a non-sync wakeup by SYNC_LESS. With one task on the runqueue, wake_affine()
rejects the affine wakeup request, leaving the unfortunate where placed, taking
frequent cache misses.
Remove it, and recover some fastpath cycles.
Signed-off-by: Mike Galbraith <[email protected]>
Signed-off-by: Peter Zijlstra <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
Testing the load which led to this heuristic (nfs4 kbuild) shows that it has
outlived it's usefullness. With intervening load balancing changes, I cannot
see any difference with/without, so recover there fastpath cycles.
Signed-off-by: Mike Galbraith <[email protected]>
Signed-off-by: Peter Zijlstra <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
Put all statistic fields of sched_entity in one struct, sched_statistics,
and embed it into sched_entity.
This change allows to memset the sched_statistics to 0 when needed (for
instance when forking), avoiding bugs of non initialized fields.
Signed-off-by: Lucas De Marchi <[email protected]>
Signed-off-by: Peter Zijlstra <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
Convert locks which cannot be sleeping locks in preempt-rt to
raw_spinlocks.
Signed-off-by: Thomas Gleixner <[email protected]>
Acked-by: Peter Zijlstra <[email protected]>
Acked-by: Ingo Molnar <[email protected]>
This build warning:
kernel/sched.c: In function 'set_task_cpu':
kernel/sched.c:2070: warning: unused variable 'old_rq'
Made me realize that the forced2_migrations stat looks pretty
pointless (and a misnomer) - remove it.
Cc: Peter Zijlstra <[email protected]>
Cc: Mike Galbraith <[email protected]>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <[email protected]>
As scaling now takes place on all kind of cpu add/remove events a user
that configures values via proc should be able to configure if his set
values are still rescaled or kept whatever happens.
As the comments state that log2 was just a second guess that worked the
interface is not just designed for on/off, but to choose a scaling type.
Currently this allows none, log and linear, but more important it allwos
us to keep the interface even if someone has an even better idea how to
scale the values.
Signed-off-by: Christian Ehrhardt <[email protected]>
Signed-off-by: Peter Zijlstra <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
WAKEUP_RUNNING was an experiment, not sure why that ever ended up being
merged...
Signed-off-by: Peter Zijlstra <[email protected]>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <[email protected]>
Rate limit newidle to migration_cost. It's a win for all
stages of sysbench oltp tests.
Signed-off-by: Mike Galbraith <[email protected]>
Cc: Peter Zijlstra <[email protected]>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <[email protected]>
Create a new wakeup preemption mode, preempt towards tasks that run
shorter on avg. It sets next buddy to be sure we actually run the task
we preempted for.
Test results:
root@twins:~# while :; do :; done &
[1] 6537
root@twins:~# while :; do :; done &
[2] 6538
root@twins:~# while :; do :; done &
[3] 6539
root@twins:~# while :; do :; done &
[4] 6540
root@twins:/home/peter# ./latt -c4 sleep 4
Entries: 48 (clients=4)
Averages:
------------------------------
Max 4750 usec
Avg 497 usec
Stdev 737 usec
root@twins:/home/peter# echo WAKEUP_RUNNING > /debug/sched_features
root@twins:/home/peter# ./latt -c4 sleep 4
Entries: 48 (clients=4)
Averages:
------------------------------
Max 14 usec
Avg 5 usec
Stdev 3 usec
Disabled by default - needs more testing.
Signed-off-by: Peter Zijlstra <[email protected]>
Acked-by: Mike Galbraith <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
LKML-Reference: <new-submission>
For counting how long an application has been waiting for
(disk) IO, there currently is only the HZ sample driven
information available, while for all other counters in this
class, a high resolution version is available via
CONFIG_SCHEDSTATS.
In order to make an improved bootchart tool possible, we also
need a higher resolution version of the iowait time.
This patch below adds this scheduler statistic to the kernel.
Signed-off-by: Arjan van de Ven <[email protected]>
Signed-off-by: Peter Zijlstra <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
There are some points which refer the per-cpu value "runqueues" directly.
sched.c provides nice abstraction, such as cpu_rq() and this_rq(),
so we should use these macros when looking runqueues.
Signed-off-by: Hitoshi Mitake <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
Impact: cleanup, new schedstat ABI
Since they are used on in statistics and are always set to zero, the
following fields from struct rq have been removed: yld_exp_empty,
yld_act_empty and yld_both_empty.
Both Sched Debug and SCHEDSTAT_VERSION versions has also been
incremented since ABIs have been changed.
The schedtop tool has been updated to properly handle new version of
schedstat:
http://rt.wiki.kernel.org/index.php/Schedtop_utility
Signed-off-by: Luis Henriques <[email protected]>
Acked-by: Gregory Haskins <[email protected]>
Acked-by: Peter Zijlstra <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
The jiffies value was being printed for each CPU, which does not seem to make
sense. Moved jiffies to system section.
Signed-off-by: Luis Henriques <[email protected]>
Acked-by: Peter Zijlstra <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
Introduce a new avg_wakeup statistic.
avg_wakeup is a measure of how frequently a task wakes up other tasks, it
represents the average time between wakeups, with a limit of avg_runtime
for when it doesn't wake up anybody.
Signed-off-by: Peter Zijlstra <[email protected]>
Signed-off-by: Mike Galbraith <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
Impact: avoid accessing NULL tg.css->cgroup
In commit 0a0db8f5c9, I removed checking
NULL tg.css->cgroup, but I realized I was wrong when I found reading
/proc/sched_debug can race with cgroup_create().
Signed-off-by: Li Zefan <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
Impact: extend information in /proc/sched_debug
This patch adds uid information in sched_debug for CONFIG_USER_SCHED
Signed-off-by: Arun R Bharadwaj <[email protected]>
Acked-by: Peter Zijlstra <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
Impact: extend /proc/sched_debug info
Since the statistics of a group entity isn't exported directly from the
kernel, it becomes difficult to obtain some of the group statistics.
For example, the current method to obtain exec time of a group entity
is not always accurate. One has to read the exec times of all
the tasks(/proc/<pid>/sched) in the group and add them. This method
fails (or becomes difficult) if we want to collect stats of a group over
a duration where tasks get created and terminated.
This patch makes it easier to obtain group stats by directly including
them in /proc/sched_debug. Stats like group exec time would help user
programs (like LTP) to accurately measure the group fairness.
An example output of group stats from /proc/sched_debug:
cfs_rq[3]:/3/a/1
.exec_clock : 89.598007
.MIN_vruntime : 0.000001
.min_vruntime : 256300.970506
.max_vruntime : 0.000001
.spread : 0.000000
.spread0 : -25373.372248
.nr_running : 0
.load : 0
.yld_exp_empty : 0
.yld_act_empty : 0
.yld_both_empty : 0
.yld_count : 4474
.sched_switch : 0
.sched_count : 40507
.sched_goidle : 12686
.ttwu_count : 15114
.ttwu_local : 11950
.bkl_count : 67
.nr_spread_over : 0
.shares : 0
.se->exec_start : 113676.727170
.se->vruntime : 1592.612714
.se->sum_exec_runtime : 89.598007
.se->wait_start : 0.000000
.se->sleep_start : 0.000000
.se->block_start : 0.000000
.se->sleep_max : 0.000000
.se->block_max : 0.000000
.se->exec_max : 1.000282
.se->slice_max : 1.999750
.se->wait_max : 54.981093
.se->wait_sum : 217.610521
.se->wait_count : 50
.se->load.weight : 2
Signed-off-by: Bharata B Rao <[email protected]>
Acked-by: Srivatsa Vaddagiri <[email protected]>
Acked-by: Dhaval Giani <[email protected]>
Acked-by: Peter Zijlstra <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
Impact: clean up and fix debug info printout
While looking over the sched_debug code I noticed that we printed the rq
schedstats for every cfs_rq, ammend this.
Also change nr_spead_over into an int, and fix a little buglet in
min_vruntime printing.
Signed-off-by: Peter Zijlstra <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
Impact: cleanup
cfs->tg is initialized in init_tg_cfs_entry() with tg != NULL, and
will never be invalidated to NULL. And the underlying cgroup of a
valid task_group is always valid.
Same for rt->tg.
Signed-off-by: Li Zefan <[email protected]>
Acked-by: Peter Zijlstra <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
Impact: change /proc/sched/debug from rw-r--r-- to r--r--r--
/proc/sched_debug is read-only.
Signed-off-by: Li Zefan <[email protected]>
Acked-by: Peter Zijlstra <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
lock_task_sighand() make sure task->sighand is being protected,
so we do not need rcu_read_lock().
[ exec() will get task->sighand->siglock before change task->sighand! ]
But code using rcu_read_lock() _just_ to protect lock_task_sighand()
only appear in procfs. (and some code in procfs use lock_task_sighand()
without such redundant protection.)
Other subsystem may put lock_task_sighand() into rcu_read_lock()
critical region, but these rcu_read_lock() are used for protecting
"for_each_process()", "find_task_by_vpid()" etc. , not for protecting
lock_task_sighand().
Signed-off-by: Lai Jiangshan <[email protected]>
[ok from Oleg]
Signed-off-by: Alexey Dobriyan <[email protected]>
Yanmin Zhang reported:
Comparing with 2.6.25, volanoMark has big regression with kernel 2.6.26-rc1.
It's about 50% on my 8-core stoakley, 16-core tigerton, and Itanium Montecito.
With bisect, I located the following patch:
| 18d95a2832 is first bad commit
| commit 18d95a2832
| Author: Peter Zijlstra <[email protected]>
| Date: Sat Apr 19 19:45:00 2008 +0200
|
| sched: fair-group: SMP-nice for group scheduling
Revert it so that we get v2.6.25 behavior.
Bisected-by: Yanmin Zhang <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
this replaces the rq->clock stuff (and possibly cpu_clock()).
- architectures that have an 'imperfect' hardware clock can set
CONFIG_HAVE_UNSTABLE_SCHED_CLOCK
- the 'jiffie' window might be superfulous when we update tick_gtod
before the __update_sched_clock() call in sched_clock_tick()
- cpu_clock() might be implemented as:
sched_clock_cpu(smp_processor_id())
if the accuracy proves good enough - how far can TSC drift in a
single jiffie when considering the filtering and idle hooks?
[ [email protected]: various fixes and cleanups ]
Signed-off-by: Peter Zijlstra <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
Add some extra debug output so we can get a better overview of the
full hierarchy.
We print the cgroup path after each cfs_rq, so we can see what group
we're looking at.
Signed-off-by: Peter Zijlstra <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
improve affine wakeups. Maintain the 'overlap' metric based on CFS's
sum_exec_runtime - which means the amount of time a task executes
after it wakes up some other task.
Use the 'overlap' for the wakeup decisions: if the 'overlap' is short,
it means there's strong workload coupling between this task and the
woken up task. If the 'overlap' is large then the workload is decoupled
and the scheduler will move them to separate CPUs more easily.
( Also slightly move the preempt_check within try_to_wake_up() - this has
no effect on functionality but allows 'early wakeups' (for still-on-rq
tasks) to be correctly accounted as well.)
Signed-off-by: Ingo Molnar <[email protected]>
Jonathan Corbet