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>
In build_sched_groups() we don't need to call get_group() for cpus
which are already covered in previous iterations. Calling get_group()
would mark the group used and eventually leak it since we wouldn't
connect it and not find it again to free it.
This will happen only in cases where sg->cpumask contained more than
one cpu (For any topology level). This patch would free sg's memory
for all cpus leaving the group leader as the group isn't marked used
now.
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/7a61e955abdcbb1dfa9fe493f11a5ec53a11ddd3.1370948150.git.viresh.kumar@linaro.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Most of the stuff from kernel/sched.c was moved to kernel/sched/core.c long time
back and the comments/Documentation never got updated.
I figured it out when I was going through sched-domains.txt and so thought of
fixing it globally.
I haven't crossed check if the stuff that is referenced in sched/core.c by all
these files is still present and hasn't changed as that wasn't the motive behind
this patch.
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/cdff76a265326ab8d71922a1db5be599f20aad45.1370329560.git.viresh.kumar@linaro.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
[ Peter, this is based off of some of my work, I ran it though a few
tests and it passed. I also reviewed it, and added my SOB as I am
somewhat a co-author to it. ]
Based on the patch by Steven Rostedt from previous year:
https://lkml.org/lkml/2012/4/18/517
1)Simplify pull_rt_task() logic: search in pushable tasks of dest runqueue.
The only pullable tasks are the tasks which are pushable in their local rq,
and no others.
2)Remove .leaf_rt_rq_list member of struct rt_rq and functions connected
with it: nobody uses it since now.
Signed-off-by: Kirill Tkhai <tkhai@yandex.ru>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/287571370557898@web7d.yandex.ru
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Dave Jones hit the following bug report:
===============================
[ INFO: suspicious RCU usage. ]
3.10.0-rc2+ #1 Not tainted
-------------------------------
include/linux/rcupdate.h:771 rcu_read_lock() used illegally while idle!
other info that might help us debug this:
RCU used illegally from idle CPU! rcu_scheduler_active = 1, debug_locks = 0
RCU used illegally from extended quiescent state!
2 locks held by cc1/63645:
#0: (&rq->lock){-.-.-.}, at: [<ffffffff816b39fd>] __schedule+0xed/0x9b0
#1: (rcu_read_lock){.+.+..}, at: [<ffffffff8109d645>] cpuacct_charge+0x5/0x1f0
CPU: 1 PID: 63645 Comm: cc1 Not tainted 3.10.0-rc2+ #1 [loadavg: 40.57 27.55 13.39 25/277 64369]
Hardware name: Gigabyte Technology Co., Ltd. GA-MA78GM-S2H/GA-MA78GM-S2H, BIOS F12a 04/23/2010
0000000000000000 ffff88010f78fcf8 ffffffff816ae383 ffff88010f78fd28
ffffffff810b698d ffff88011c092548 000000000023d073 ffff88011c092500
0000000000000001 ffff88010f78fd60 ffffffff8109d7c5 ffffffff8109d645
Call Trace:
[<ffffffff816ae383>] dump_stack+0x19/0x1b
[<ffffffff810b698d>] lockdep_rcu_suspicious+0xfd/0x130
[<ffffffff8109d7c5>] cpuacct_charge+0x185/0x1f0
[<ffffffff8109d645>] ? cpuacct_charge+0x5/0x1f0
[<ffffffff8108dffc>] update_curr+0xec/0x240
[<ffffffff8108f528>] put_prev_task_fair+0x228/0x480
[<ffffffff816b3a71>] __schedule+0x161/0x9b0
[<ffffffff816b4721>] preempt_schedule+0x51/0x80
[<ffffffff816b4800>] ? __cond_resched_softirq+0x60/0x60
[<ffffffff816b6824>] ? retint_careful+0x12/0x2e
[<ffffffff810ff3cc>] ftrace_ops_control_func+0x1dc/0x210
[<ffffffff816be280>] ftrace_call+0x5/0x2f
[<ffffffff816b681d>] ? retint_careful+0xb/0x2e
[<ffffffff816b4805>] ? schedule_user+0x5/0x70
[<ffffffff816b4805>] ? schedule_user+0x5/0x70
[<ffffffff816b6824>] ? retint_careful+0x12/0x2e
------------[ cut here ]------------
What happened was that the function tracer traced the schedule_user() code
that tells RCU that the system is coming back from userspace, and to
add the CPU back to the RCU monitoring.
Because the function tracer does a preempt_disable/enable_notrace() calls
the preempt_enable_notrace() checks the NEED_RESCHED flag. If it is set,
then preempt_schedule() is called. But this is called before the user_exit()
function can inform the kernel that the CPU is no longer in user mode and
needs to be accounted for by RCU.
The fix is to create a new preempt_schedule_context() that checks if
the kernel is still in user mode and if so to switch it to kernel mode
before calling schedule. It also switches back to user mode coming back
from schedule in need be.
The only user of this currently is the preempt_enable_notrace(), which is
only used by the tracing subsystem.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1369423420.6828.226.camel@gandalf.local.home
Signed-off-by: Ingo Molnar <mingo@kernel.org>
I have faced a sequence where the Idle Load Balance was sometime not
triggered for a while on my platform, in the following scenario:
CPU 0 and CPU 1 are running tasks and CPU 2 is idle
CPU 1 kicks the Idle Load Balance
CPU 1 selects CPU 2 as the new Idle Load Balancer
CPU 2 sets NOHZ_BALANCE_KICK for CPU 2
CPU 2 sends a reschedule IPI to CPU 2
While CPU 3 wakes up, CPU 0 or CPU 1 migrates a waking up task A on CPU 2
CPU 2 finally wakes up, runs task A and discards the Idle Load Balance
task A quickly goes back to sleep (before a tick occurs on CPU 2)
CPU 2 goes back to idle with NOHZ_BALANCE_KICK set
Whenever CPU 2 will be selected as the ILB, no reschedule IPI will be sent
because NOHZ_BALANCE_KICK is already set and no Idle Load Balance will be
performed.
We must wait for the sched softirq to be raised on CPU 2 thanks to another
part the kernel to come back to clear NOHZ_BALANCE_KICK.
The proposed solution clears NOHZ_BALANCE_KICK in schedule_ipi if
we can't raise the sched_softirq for the Idle Load Balance.
Change since V1:
- move the clear of NOHZ_BALANCE_KICK in got_nohz_idle_kick if the ILB
can't run on this CPU (as suggested by Peter)
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1370419991-13870-1-git-send-email-vincent.guittot@linaro.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull tracing fixes from Steven Rostedt:
"Two more fixes:
The first one was reported by Mauro Carvalho Chehab, where if a poll()
is done against a trace buffer for a CPU that has never been online,
it will crash the kernel, as buffers are only created when a CPU comes
on line, but the trace files are for all possible CPUs.
This fix is to check if the buffer was allocated and if not return
-EINVAL.
That was the simple fix, the real fix is a bit more complex and not
for a -rc release. We could have the files created when the CPUs come
online. That would require some design changes.
The second one was reported by Peter Zijlstra. If the kernel command
line has ftrace=nop, it will lock up the system on boot up. This is
because the new design for 3.10 has the nop tracer bootstrap the
tracing subsystem. When ftrace=<trace> is defined, when a that tracer
is registered, it starts the tracing, but uses the nop tracer to clear
things out. What happened here was that ftrace=nop caused the
registering of nop to start it and use nop before it was initialized.
The only thing nop needs to have done to initialize it is to have the
tracer point its current_tracer structure member to the nop tracer.
Doing that before registering the nop tracer makes everything work."
* tag 'trace-fixes-v3.10-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
ring-buffer: Do not poll non allocated cpu buffers
tracing: Fix crash when ftrace=nop on the kernel command line
The tracing infrastructure sets up for possible CPUs, but it uses
the ring buffer polling, it is possible to call the ring buffer
polling code with a CPU that hasn't been allocated. This will cause
a kernel oops when it access a ring buffer cpu buffer that is part
of the possible cpus but hasn't been allocated yet as the CPU has never
been online.
Reported-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Tested-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
migration_call() will do all the things that update_runtime() does.
So let's remove it.
Furthermore, there is potential risk that the current code will catch
BUG_ON at line 689 of rt.c when do cpu hotplug while there are realtime
threads running because of enabling runtime twice while the rt_runtime
may already changed.
Signed-off-by: Neil Zhang <zhangwm@marvell.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1365685499-26515-1-git-send-email-zhangwm@marvell.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
In autogroup_create(), a tg is allocated and added to the task_groups
list. If CONFIG_RT_GROUP_SCHED is set, this tg is then modified while on
the list, without locking. This can race with someone walking the list,
like __enable_runtime() during CPU unplug, and result in a use-after-free
bug.
To fix this, move sched_online_group(), which adds the tg to the list,
to the end of the autogroup_create() function after the modification.
Signed-off-by: Gerald Schaefer <gerald.schaefer@de.ibm.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1369411669-46971-2-git-send-email-gerald.schaefer@de.ibm.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Fix kernel-doc warnings in kernel/auditfilter.c:
Warning(kernel/auditfilter.c:1029): Excess function parameter 'loginuid' description in 'audit_receive_filter'
Warning(kernel/auditfilter.c:1029): Excess function parameter 'sessionid' description in 'audit_receive_filter'
Warning(kernel/auditfilter.c:1029): Excess function parameter 'sid' description in 'audit_receive_filter'
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Cc: Eric Paris <eparis@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull tracing fix from Steven Rostedt:
"Masami Hiramatsu fixed another bug. This time returning a proper
result in event_enable_func(). After checking the return status of
try_module_get(), it returned the status of try_module_get().
But try_module_get() returns 0 on failure, which is success for
event_enable_func()"
* tag 'trace-fixes-v3.10-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
tracing: Return -EBUSY when event_enable_func() fails to get module
If ftrace=<tracer> is on the kernel command line, when that tracer is
registered, it will be initiated by tracing_set_tracer() to execute that
tracer.
The nop tracer is just a stub tracer that is used to have no tracer
enabled. It is assigned at early bootup as it is the default tracer.
But if ftrace=nop is on the kernel command line, the registering of the
nop tracer will call tracing_set_tracer() which will try to execute
the nop tracer. But it expects tr->current_trace to be assigned something
as it usually is assigned to the nop tracer. As it hasn't been assigned
to anything yet, it causes the system to crash.
The simple fix is to move the tr->current_trace = nop before registering
the nop tracer. The functionality is still the same as the nop tracer
doesn't do anything anyway.
Reported-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Pull kmemleak patches from Catalin Marinas:
"Kmemleak now scans all the writable and non-executable module sections
to avoid false positives (previously it was only scanning specific
sections and missing .ref.data)."
* tag 'kmemleak-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/cmarinas/linux-aarch64:
kmemleak: No need for scanning specific module sections
kmemleak: Scan all allocated, writeable and not executable module sections
As kmemleak now scans all module sections that are allocated, writable
and non executable, there's no need to scan individual sections that
might reference data.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Acked-by: Rusty Russell <rusty@rustcorp.com.au>
Instead of just picking data sections by name (names that start
with .data, .bss or .ref.data), use the section flags and scan all
sections that are allocated, writable and not executable. Which should
cover all sections of a module that might reference data.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
[catalin.marinas@arm.com: removed unused 'name' variable]
[catalin.marinas@arm.com: collapsed 'if' blocks]
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Acked-by: Rusty Russell <rusty@rustcorp.com.au>
Pull workqueue fixes from Tejun Heo:
"Three more workqueue regression fixes.
- Fix unbalanced unlock in trylock failure path of manage_workers().
This shouldn't happen often in the wild but is possible.
- While making schedule_work() and friends inline, they become
unavailable to !GPL modules. Allow !GPL modules to access basic
stuff - system_wq and queue_*work_on() - so that schedule_work()
and friends can be used.
- During boot, the unbound NUMA support code allocates a cpumask for
each possible node using alloc_cpumask_var_node(), which ends up
trying to allocate node-specific memory even for offline nodes
triggering BUG in the memory alloc code. Use NUMA_NO_NODE for
offline nodes."
* 'for-3.10-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq:
workqueue: don't perform NUMA-aware allocations on offline nodes in wq_numa_init()
workqueue: Make schedule_work() available again to non GPL modules
workqueue: correct handling of the pool spin_lock
Pull RCU fixes from Paul McKenney:
"A couple of fixes for RCU regressions:
- A boneheaded boolean-logic bug that resulted in excessive delays on
boot, hibernation and suspend that was reported by Borislav Petkov,
Bjørn Mork, and Joerg Roedel. The fix inserts a single "!".
- A fix for a boot-time splat due to allocating from bootmem too late
in boot, fix courtesy of Sasha Levin with additional help from
Yinghai Lu."
* 'rcu/urgent' of git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu:
rcu: Don't allocate bootmem from rcu_init()
rcu: Fix comparison sense in rcu_needs_cpu()
argv_split(empty_or_all_spaces) happily succeeds, it simply returns
argc == 0 and argv[0] == NULL. Change call_usermodehelper_exec() to
check sub_info->path != NULL to avoid the crash.
This is the minimal fix, todo:
- perhaps we should change argv_split() to return NULL or change the
callers.
- kill or justify ->path[0] check
- narrow the scope of helper_lock()
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-By: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: stable@vger.kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull tracing fixes from Steven Rostedt:
"This includes a fix to a memory leak when adding filters to traces.
Also, Masami Hiramatsu fixed up some minor bugs that were discovered
by sparse."
* tag 'trace-fixes-v3.10-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
tracing/kprobes: Make print_*probe_event static
tracing/kprobes: Fix a sparse warning for incorrect type in assignment
tracing/kprobes: Use rcu_dereference_raw for tp->files
tracing: Fix leaks of filter preds
Pull perf fixes from Thomas Gleixner:
- Fix for a task exit cleanup race caused by a missing a preempt
disable
- Cleanup of the event notification functions with a massive reduction
of duplicated code
* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf: Factor out auxiliary events notification
perf: Fix EXIT event notification
Pull timer fixes from Thomas Gleixner:
- Cure for not using zalloc in the first place, which leads to random
crashes with CPUMASK_OFF_STACK.
- Revert a user space visible change which broke udev
- Add a missing cpu_online early return introduced by the new full
dyntick conversions
- Plug a long standing race in the timer wheel cpu hotplug code.
Sigh...
- Cleanup NOHZ per cpu data on cpu down to prevent stale data on cpu
up.
* 'timers-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
time: Revert ALWAYS_USE_PERSISTENT_CLOCK compile time optimizaitons
timer: Don't reinitialize the cpu base lock during CPU_UP_PREPARE
tick: Don't invoke tick_nohz_stop_sched_tick() if the cpu is offline
tick: Cleanup NOHZ per cpu data on cpu down
tick: Use zalloc_cpumask_var for allocating offstack cpumasks
Pull core fixes from Thomas Gleixner:
- Two fixlets for the fallout of the generic idle task conversion
- Documentation update
* 'core-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
rcu/idle: Wrap cpu-idle poll mode within rcu_idle_enter/exit
idle: Fix hlt/nohlt command-line handling in new generic idle
kthread: Document ways of reducing OS jitter due to per-CPU kthreads
