The original intent here was to track unweighted runqueue load
with less resolution so we could use the least-recently-disturbed
runqueue to choose between 'closely related' load levels.
However, after experimenting with the resolution it turns out
that the following algorithm is highly beneficial for mobile
workloads.
In hmp_domain_min_load:
* If any CPU is zero, the overall load is zero
* If no CPUs are idle, the domain is 'fully loaded'
Additionally, the time since last migration count is used to
discriminate between idle CPUs.
Signed-off-by: Chris Redpath <chris.redpath@arm.com>
Signed-off-by: Liviu Dudau <liviu.dudau@arm.com>
Signed-off-by: Jon Medhurst <tixy@linaro.org>
Track when migrations were performed to runqueues.
Use this to decide between runqueues as migration targets when run
queues in an hmp domain have equal load.
Intention is to spread migration load amongst CPUs more fairly.
When all CPUs in an hmp domain are fully loaded, the existing code
always selects the last CPU as a migration target - this is unfair
and little better than doing no selection.
Signed-off-by: Chris Redpath <chris.redpath@arm.com>
Signed-off-by: Liviu Dudau <liviu.dudau@arm.com>
Signed-off-by: Jon Medhurst <tixy@linaro.org>
The hmp_get_{lightest,heaviest}_task() need to use
__pick_first_entity() to get a pointer to a sched_entity on the rq.
The current is not kept on the rq while running, so its rb-tree node
pointers are no longer valid.
Signed-off-by: Chris Redpath <chris.redpath@arm.com>
Signed-off-by: Liviu Dudau <liviu.dudau@arm.com>
Signed-off-by: Jon Medhurst <tixy@linaro.org>
When we are looking for a task to migrate up, select the heaviest
one in the first 5 runnable on the runqueue.
Likewise, when looking for a task to offload, select the lightest
one in the first 5 runnable on the runqueue.
Ensure task selected is runnable in the target domain.
This change is necessary in order to implement idle pull in a
sensible manner, but here is used in up-migration and offload to
select the correct target task.
Signed-off-by: Chris Redpath <chris.redpath@arm.com>
Signed-off-by: Liviu Dudau <liviu.dudau@arm.com>
Signed-off-by: Jon Medhurst <tixy@linaro.org>
It is sometimes desirable to run a kernel with HMP scheduling enabled
on a system which is not big.LITTLE, e.g. when building a multi-platform
kernel, or when testing a big.LITTLE system with one cluster disabled.
We should therefore allow for the situation where is no little domain.
Signed-off-by: Jon Medhurst <tixy@linaro.org>
Signed-off-by: Mark Brown <broonie@linaro.org>
This patch restricts the allowed cpu mask for rt tasks initially started
with a full cpu mask to the little domain.
An rt task is specified as real time in __setscheduler() which is finally
called for all rt tasks (kernel and user land). In this function we
restrict the allowed cpu mask to the little domain.
This also prevents that a rt tasks can later be pushed to the big domain
because the function find_lowest_rq() will only recognize the allowed cpu
mask of a task to find the new cpu the task runs on.
Current kludges of the patch:
* Since we do not have an API to get the cpu mask of the A7 cluster,
hmp_slow_cpu_mask is made global in arm/kernel/topology.c for now.
* The watchdog_enable() function calls sched_setscheduler() before
kthread_bind() for the cpu specific watchdog kernel threads. The order of
these two calls has to be changed to make this patch work.
Signed-off-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
There is little point in doing a nohz balance kick on a CPU from a
different HMP domain, since the unset SD_LOAD_BALANCE flag on the CPU
domain level prevents tasks from being balanced across clusters
except through the per-task load driven hmp_migrate/hmp_offload paths.
Further, the nohz balance kick is actively harmful to power usage if
all the tasks fit into the little domain since it causes the big
domain to wake up and do a lot of calculation to determine that
there is nothing to do.
A more generic solution is to walk the sched domain tree and determine
the intersection of potential idle balance cpus with visibility of
tasks on the current CPU, however HMP domains are more easily
accessible.
Signed-off-by: Chris Redpath <chris.redpath@arm.com>
Initialise the load stats for new tasks so that they do not
see the instability in early task life which makes it so hard to
decide which CPU is appropriate.
Also, change the fork balance algorithm so that the least loaded of
the CPUs in the big cluster is chosen regardless of the bigness of
the parent task.
This is intended to help performance for applications which use
many short-lived tasks. Although best practise is usually to use
a thread pool, apps which do not do this should not be subject to
the randomness of the early stats.
We should ignore real-time threads for forking on big CPUs, but
it is not possible to figure out if a new thread is real-time or
not at the fork stage. Instead, we prevent kernel threads from
getting the initial boost - when they later become real-time they
will only be on big if their compute requirements demand it.
Signed-off-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Signed-off-by: Chris Redpath <chris.redpath@arm.com>
When evaluating a migration we make two calls to hmp_domain_min_load.
This is unnecessary if we pass on the target CPU information from the
hmp_up_migration path.
In hmp_down_migration, we don't consider the load of the target CPUS.
Signed-off-by: Chris Redpath <chris.redpath@arm.com>
The reference patch set always selects the first CPU in an HMP
domain as a migration target. In busy situations, this means that
the migrated thread cannot make immediate use of an idle CPU but
must share a busy one until the load balancer runs across the big
domain.
This patch uses the hmp_domain_min_load function introduced in
global balancing to figure out which of the CPUs is the least busy
and selects that as a migration target - in both directions.
This essentially implements a task-spread strategy and is intended
to maximise performance of migrated threads but is likely
to use more power than the packing strategy previously employed.
Signed-off-by: Chris Redpath <chris.redpath@arm.com>
Normal task and runqueue loading is scaled according to priority
to end up with a weighted load, known as the contribution.
We want the CPU time to be allotted according to priority, but
we also want to make big/little decisions based upon raw load.
It is common, for example, for Android apps following the dev
guide to end up with all their long-running or async action
threads as low priority unless they override the AsyncThread
constructor. All these threads are such low priority that they
become invisible to the hmp_offload routine.
Using unweighted load here allows us to maximise CPU usage in busy
situations.
Signed-off-by: Chris Redpath <chris.redpath@arm.com>
rq->nr_running is the actual number of runnable tasks we wish to use
to determine if a task is alone on a CPU.
Change-Id: Icaf3022e02924ecdc94e14d4146c6fadd9580e2b
Signed-off-by: Chris Redpath <chris.redpath@arm.com>
This patch introduces an extra-check at task up-migration to
prevent overloading the cpus in the faster hmp_domain while the
slower hmp_domain is not fully utilized. The patch also introduces
a periodic balance check that can down-migrate tasks if the faster
domain is oversubscribed and the slower is under-utilized.
Signed-off-by: Morten Rasmussen <morten.rasmussen@arm.com>
Evaluation Patch to investigate using load as a representation of the
amount of POTENTIAL cpu compute capacity used rather than a representation
of the CURRENT cpu compute capacity.
If CPUFreq is enabled, scales load in accordance with frequency.
Powersave/performance CPUFreq governors are detected and scaling is
disabled while these governors are in use. This is because when a
single-frequency governor is in use, potential CPU capacity is static.
So long as the governors and CPUFreq subsystem correctly report the
frequencies available, the scaling should self tune.
Adds an additional file to sysfs to allow this feature to be disabled
for experimentation.
/sys/kernel/hmp/frequency_invariant_load_scale
write 0 to disable, 1 to enable.
Signed-off-by: Chris Redpath <chris.redpath@arm.com>
These functions allow to change the load average period used
in the task load average computation through
/sys/kernel/hmp/load_avg_period_ms. This period is the time
in ms to go from 0 to 0.5 load average while running or the
time from 1 to 0.5 while sleeping.
The default one used is 32 and gives the same load_avg_ratio
computation than without this patch. These functions also allow
to change the up and down threshold of HMP using
/sys/kernel/hmp/{up,down}_threshold. Both must be between 0 and
1024. The thresholds are divided by 1024 before being compared
to the load_avg_ratio.
If /sys/kernel/hmp/load_avg_period_ms is 128 and
/sys/kernel/hmp/up_threshold is 512, a task will be migrated
to a bigger cluster after running for 128ms. Because after
load_avg_period_ms the load average is 0.5 and real up_threshold
us 512 / 1024 = 0.5.
Signed-off-by: Olivier Cozette <olivier.cozette@arm.com>
Signed-off-by: Chris Redpath <chris.redpath@arm.com>
Previously, an offline CPU would always appear to have a zero load
and this would distort the offload functionality used for balancing
big and little domains.
Maintain a mask of online CPUs in each domain and use this instead.
Change-Id: I639b564b2f40cb659af8ceb8bd37f84b8a1fe323
Signed-off-by: Chris Redpath <chris.redpath@arm.com>
If the entity is not a task, it is a cfs group rq. Iterate up to
find the task entity.
Change-Id: I7cab7aba0798f6f14e38ad32e566d90e5937ffbc
Signed-off-by: Chris Redpath <chris.redpath@arm.com>
Adds an extra check intersection of the task affinity mask and the slower
hmp_domain cpumask before down migrating low priority tasks.
Signed-off-by: Morten Rasmussen <morten.rasmussen@arm.com>
We need a way to prevent tasks that are migrating up and down the
hmp_domains from migrating straight on through before the load has
adapted to the new compute capacity of the CPU on the new hmp_domain.
This patch adds a next up/down migration delay that prevents the task
from doing another migration in the same direction until the delay
has expired.
Signed-off-by: Morten Rasmussen <Morten.Rasmussen@arm.com>
Adds ftrace events for key variables related to the entity
load-tracking to help debugging scheduler behaviour. Allows tracing
of load contribution and runqueue residency ratio for both entities
and runqueues as well as entity CPU usage ratio.
Signed-off-by: Morten Rasmussen <Morten.Rasmussen@arm.com>
Introduces a priority threshold which prevents low priority task
from migrating to faster hmp_domains (cpus). This is useful for
user-space software which assigns lower task priority to background
task.
Signed-off-by: Morten Rasmussen <Morten.Rasmussen@arm.com>
This patch introduces forced task migration for moving suitable
currently running tasks between hmp_domains. Task behaviour is likely
to change over time. Tasks running in a less capable hmp_domain may
change to become more demanding and should therefore be migrated up.
They are unlikely go through the select_task_rq_fair() path anytime
soon and therefore need special attention.
This patch introduces a period check (SCHED_TICK) of the currently
running task on all runqueues and sets up a forced migration using
stop_machine_no_wait() if the task needs to be migrated.
Ideally, this should not be implemented by polling all runqueues.
Signed-off-by: Morten Rasmussen <Morten.Rasmussen@arm.com>
This patch introduces the basic SCHED_HMP infrastructure. Each class of
cpus is represented by a hmp_domain and tasks will only be moved between
these domains when their load profiles suggest it is beneficial.
SCHED_HMP relies heavily on the task load-tracking introduced in Paul
Turners fair group scheduling patch set:
<https://lkml.org/lkml/2012/8/23/267>
SCHED_HMP requires that the platform implements arch_get_hmp_domains()
which should set up the platform specific list of hmp_domains. It is
also assumed that the platform disables SD_LOAD_BALANCE for the
appropriate sched_domains.
Tasks placement takes place every time a task is to be inserted into
a runqueue based on its load history. The task placement decision is
based on load thresholds.
There are no restrictions on the number of hmp_domains, however,
multiple (>2) has not been tested and the up/down migration policy is
rather simple.
Signed-off-by: Morten Rasmussen <Morten.Rasmussen@arm.com>
This patch adds load_avg_ratio to each task. The load_avg_ratio is a
variant of load_avg_contrib which is not scaled by the task priority. It
is calculated like this:
runnable_avg_sum * NICE_0_LOAD / (runnable_avg_period + 1).
Signed-off-by: Morten Rasmussen <Morten.Rasmussen@arm.com>
With the frame-work for runnable tracking now fully in place. Per-entity usage
tracking is a simple and low-overhead addition.
Signed-off-by: Paul Turner <pjt@google.com>
Reviewed-by: Ben Segall <bsegall@google.com>
Pull timer fix from Thomas Gleixner:
"Correct an ordering issue in the tick broadcast code. I really wish
we'd get compensation for pain and suffering for each line of code we
write to work around dysfunctional timer hardware."
* 'timers-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
tick: Fix tick_broadcast_pending_mask not cleared
Pull perf fixes from Ingo Molnar:
"Three small fixlets"
* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
hw_breakpoint: Use cpu_possible_mask in {reserve,release}_bp_slot()
hw_breakpoint: Fix cpu check in task_bp_pinned(cpu)
kprobes: Fix arch_prepare_kprobe to handle copy insn failures
Pull x86 fixes from Peter Anvin:
"This series fixes a couple of build failures, and fixes MTRR cleanup
and memory setup on very specific memory maps.
Finally, it fixes triggering backtraces on all CPUs, which was
inadvertently disabled on x86."
* 'x86/urgent' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/efi: Fix dummy variable buffer allocation
x86: Fix trigger_all_cpu_backtrace() implementation
x86: Fix section mismatch on load_ucode_ap
x86: fix build error and kconfig for ia32_emulation and binfmt
range: Do not add new blank slot with add_range_with_merge
x86, mtrr: Fix original mtrr range get for mtrr_cleanup
The recent modification in the cpuidle framework consolidated the
timer broadcast code across the different drivers by setting a new
flag in the idle state. It tells the cpuidle core code to enter/exit
the broadcast mode for the cpu when entering a deep idle state. The
broadcast timer enter/exit is no longer handled by the back-end
driver.
This change made the local interrupt to be enabled *before* calling
CLOCK_EVENT_NOTIFY_EXIT.
On a tegra114, a four cores system, when the flag has been introduced
in the driver, the following warning appeared:
WARNING: at kernel/time/tick-broadcast.c:578 tick_broadcast_oneshot_control
CPU: 2 PID: 0 Comm: swapper/2 Not tainted 3.10.0-rc3-next-20130529+ #15
[<c00667f8>] (tick_broadcast_oneshot_control+0x1a4/0x1d0) from [<c0065cd0>] (tick_notify+0x240/0x40c)
[<c0065cd0>] (tick_notify+0x240/0x40c) from [<c0044724>] (notifier_call_chain+0x44/0x84)
[<c0044724>] (notifier_call_chain+0x44/0x84) from [<c0044828>] (raw_notifier_call_chain+0x18/0x20)
[<c0044828>] (raw_notifier_call_chain+0x18/0x20) from [<c00650cc>] (clockevents_notify+0x28/0x170)
[<c00650cc>] (clockevents_notify+0x28/0x170) from [<c033f1f0>] (cpuidle_idle_call+0x11c/0x168)
[<c033f1f0>] (cpuidle_idle_call+0x11c/0x168) from [<c000ea94>] (arch_cpu_idle+0x8/0x38)
[<c000ea94>] (arch_cpu_idle+0x8/0x38) from [<c005ea80>] (cpu_startup_entry+0x60/0x134)
[<c005ea80>] (cpu_startup_entry+0x60/0x134) from [<804fe9a4>] (0x804fe9a4)
I don't have the hardware, so I wasn't able to reproduce the warning
but after looking a while at the code, I deduced the following:
1. the CPU2 enters a deep idle state and sets the broadcast timer
2. the timer expires, the tick_handle_oneshot_broadcast function is
called, setting the tick_broadcast_pending_mask and waking up the
idle cpu CPU2
3. the CPU2 exits idle handles the interrupt and then invokes
tick_broadcast_oneshot_control with CLOCK_EVENT_NOTIFY_EXIT which
runs the following code:
[...]
if (dev->next_event.tv64 == KTIME_MAX)
goto out;
if (cpumask_test_and_clear_cpu(cpu,
tick_broadcast_pending_mask))
goto out;
[...]
So if there is no next event scheduled for CPU2, we fulfil the
first condition and jump out without clearing the
tick_broadcast_pending_mask.
4. CPU2 goes to deep idle again and calls
tick_broadcast_oneshot_control with CLOCK_NOTIFY_EVENT_ENTER but
with the tick_broadcast_pending_mask set for CPU2, triggering the
warning.
The issue only surfaced due to the modifications of the cpuidle
framework, which resulted in interrupts being enabled before the call
to the clockevents code. If the call happens before interrupts have
been enabled, the warning cannot trigger, because there is still the
event pending which caused the broadcast timer expiry.
Move the check for the next event below the check for the pending bit,
so the pending bit gets cleared whether an event is scheduled on the
cpu or not.
[ tglx: Massaged changelog ]
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Reported-and-tested-by: Joseph Lo <josephl@nvidia.com>
Cc: Stephen Warren <swarren@nvidia.com>
Cc: linux-arm-kernel@lists.infradead.org
Cc: linaro-kernel@lists.linaro.org
Link: http://lkml.kernel.org/r/1371485735-31249-1-git-send-email-daniel.lezcano@linaro.org
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Pull scheduler fixes from Ingo Molnar:
"Two smaller fixes - plus a context tracking tracing fix that is a bit
bigger"
* 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
tracing/context-tracking: Add preempt_schedule_context() for tracing
sched: Fix clear NOHZ_BALANCE_KICK
sched/x86: Construct all sibling maps if smt
Pull perf fixes from Ingo Molnar:
"Four fixes. The mmap ones are unfortunately larger than desired -
fuzzing uncovered bugs that needed perf context life time management
changes to fix properly"
* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf/x86: Fix broken PEBS-LL support on SNB-EP/IVB-EP
perf: Fix mmap() accounting hole
perf: Fix perf mmap bugs
kprobes: Fix to free gone and unused optprobes
Pull cpu idle fixes from Thomas Gleixner:
- Add a missing irq enable. Fallout of the idle conversion
- Fix stackprotector wreckage caused by the idle conversion
* 'core-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
idle: Enable interrupts in the weak arch_cpu_idle() implementation
idle: Add the stack canary init to cpu_startup_entry()
Pull timer fixes from Thomas Gleixner:
- Fix inconstinant clock usage in virtual time accounting
- Fix a build error in KVM caused by the NOHZ work
- Remove a pointless timekeeping duty assignment which breaks NOHZ
- Use a proper notifier return value to avoid random behaviour
* 'timers-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
tick: Remove useless timekeeping duty attribution to broadcast source
nohz: Fix notifier return val that enforce timekeeping
kvm: Move guest entry/exit APIs to context_tracking
vtime: Use consistent clocks among nohz accounting
trinity fuzzer triggered WARN_ONCE("Can't find any breakpoint
slot") in arch_install_hw_breakpoint() but the problem is not
arch-specific.
The problem is, task_bp_pinned(cpu) checks "cpu == iter->cpu"
but this doesn't account the "all cpus" events with iter->cpu <
0.
This means that, say, register_user_hw_breakpoint(tsk) can
happily create the arbitrary number > HBP_NUM of breakpoints
which can not be activated. toggle_bp_task_slot() is equally
wrong by the same reason and nr_task_bp_pinned[] can have
negative entries.
Simple test:
# perl -e 'sleep 1 while 1' &
# perf record -e mem:0x10,mem:0x10,mem:0x10,mem:0x10,mem:0x10 -p `pidof perl`
Before this patch this triggers the same problem/WARN_ON(),
after the patch it correctly fails with -ENOSPC.
Reported-by: Vince Weaver <vincent.weaver@maine.edu>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: <stable@vger.kernel.org>
Link: http://lkml.kernel.org/r/20130620155006.GA6324@redhat.com
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>
Vince's fuzzer once again found holes. This time it spotted a leak in
the locked page accounting.
When an event had redirected output and its close() was the last
reference to the buffer we didn't have a vm context to undo accounting.
Change the code to destroy the buffer on the last munmap() and detach
all redirected events at that time. This provides us the right context
to undo the vm accounting.
Reported-and-tested-by: Vince Weaver <vincent.weaver@maine.edu>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20130604084421.GI8923@twins.programming.kicks-ass.net
Cc: <stable@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Joshua reported: Commit cd7b304dfa (x86, range: fix missing merge
during add range) broke mtrr cleanup on his setup in 3.9.5.
corresponding commit in upstream is fbe06b7bae.
The reason is add_range_with_merge could generate blank spot.
We could avoid that by searching new expanded start/end, that
new range should include all connected ranges in range array.
At last add the new expanded start/end to the range array.
Also move up left array so do not add new blank slot in the
range array.
-v2: move left array to avoid enhance add_range()
-v3: include fix from Joshua about memmove declaring when
DYN_DEBUG is used.
Reported-by: Joshua Covington <joshuacov@googlemail.com>
Tested-by: Joshua Covington <joshuacov@googlemail.com>
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Link: http://lkml.kernel.org/r/1371154622-8929-3-git-send-email-yinghai@kernel.org
Cc: <stable@vger.kernel.org> v3.9
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Pull VFS fixes from Al Viro:
"Several fixes + obvious cleanup (you've missed a couple of open-coded
can_lookup() back then)"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
snd_pcm_link(): fix a leak...
use can_lookup() instead of direct checks of ->i_op->lookup
move exit_task_namespaces() outside of exit_notify()
fput: task_work_add() can fail if the caller has passed exit_task_work()
ncpfs: fix rmdir returns Device or resource busy
exit_notify() does exit_task_namespaces() after
forget_original_parent(). This was needed to ensure that ->nsproxy
can't be cleared prematurely, an exiting child we are going to
reparent can do do_notify_parent() and use the parent's (ours) pid_ns.
However, after 32084504 "pidns: use task_active_pid_ns in
do_notify_parent" ->nsproxy != NULL is no longer needed, we rely
on task_active_pid_ns().
Move exit_task_namespaces() from exit_notify() to do_exit(), after
exit_fs() and before exit_task_work().
This solves the problem reported by Andrey, free_ipc_ns()->shm_destroy()
does fput() which needs task_work_add().
Note: this particular problem can be fixed if we change fput(), and
that change makes sense anyway. But there is another reason to move
the callsite. The original reason for exit_task_namespaces() from
the middle of exit_notify() was subtle and it has already gone away,
now this looks confusing. And this allows us do simplify exit_notify(),
we can avoid unlock/lock(tasklist) and we can use ->exit_state instead
of PF_EXITING in forget_original_parent().
Reported-by: Andrey Vagin <avagin@openvz.org>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: "Eric W. Biederman" <ebiederm@xmission.com>
Acked-by: Andrey Vagin <avagin@openvz.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
PARISC bootup triggers the warning at kernel/cpu/idle.c:96. That's
caused by the weak arch_cpu_idle() implementation, which is provided
to avoid that architectures implement idle_poll over and over.
The switchover to polling mode happens in the first call of the weak
arch_cpu_idle() implementation, but that code fails to reenable
interrupts and therefor triggers the warning.
Fix this by enabling interrupts in the weak arch_cpu_idle() code.
[ tglx: Made the changelog match the patch ]
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Reviewed-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com>
Link: http://lkml.kernel.org/r/1371236142.2726.43.camel@dabdike
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Pull RCU fixes from Paul McKenney:
"I must confess that this past merge window was not RCU's best showing.
This series contains three more fixes for RCU regressions:
1. A fix to __DECLARE_TRACE_RCU() that causes it to act as an
interrupt from idle rather than as a task switch from idle.
This change is needed due to the recent use of _rcuidle()
tracepoints that can be invoked from interrupt handlers as well
as from idle. Without this fix, invoking _rcuidle() tracepoints
from interrupt handlers results in splats and (more seriously)
confusion on RCU's part as to whether a given CPU is idle or not.
This confusion can in turn result in too-short grace periods and
therefore random memory corruption.
2. A fix to a subtle deadlock that could result due to RCU doing
a wakeup while holding one of its rcu_node structure's locks.
Although the probability of occurrence is low, it really
does happen. The fix, courtesy of Steven Rostedt, uses
irq_work_queue() to avoid the deadlock.
3. A fix to a silent deadlock (invisible to lockdep) due to the
interaction of timeouts posted by RCU debug code enabled by
CONFIG_PROVE_RCU_DELAY=y, grace-period initialization, and CPU
hotplug operations. This will not occur in production kernels,
but really does occur in randconfig testing. Diagnosis courtesy
of Steven Rostedt"
* 'rcu/urgent' of git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu:
rcu: Fix deadlock with CPU hotplug, RCU GP init, and timer migration
rcu: Don't call wakeup() with rcu_node structure ->lock held
trace: Allow idle-safe tracepoints to be called from irq
Merge misc fixes from Andrew Morton:
"Bunch of fixes and one little addition to math64.h"
* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (27 commits)
include/linux/math64.h: add div64_ul()
mm: memcontrol: fix lockless reclaim hierarchy iterator
frontswap: fix incorrect zeroing and allocation size for frontswap_map
kernel/audit_tree.c:audit_add_tree_rule(): protect `rule' from kill_rules()
mm: migration: add migrate_entry_wait_huge()
ocfs2: add missing lockres put in dlm_mig_lockres_handler
mm/page_alloc.c: fix watermark check in __zone_watermark_ok()
drivers/misc/sgi-gru/grufile.c: fix info leak in gru_get_config_info()
aio: fix io_destroy() regression by using call_rcu()
rtc-at91rm9200: use shadow IMR on at91sam9x5
rtc-at91rm9200: add shadow interrupt mask
rtc-at91rm9200: refactor interrupt-register handling
rtc-at91rm9200: add configuration support
rtc-at91rm9200: add match-table compile guard
fs/ocfs2/namei.c: remove unecessary ERROR when removing non-empty directory
swap: avoid read_swap_cache_async() race to deadlock while waiting on discard I/O completion
drivers/rtc/rtc-twl.c: fix missing device_init_wakeup() when booted with device tree
cciss: fix broken mutex usage in ioctl
audit: wait_for_auditd() should use TASK_UNINTERRUPTIBLE
drivers/rtc/rtc-cmos.c: fix accidentally enabling rtc channel
...
audit_add_tree_rule() must set 'rule->tree = NULL;' firstly, to protect
the rule itself freed in kill_rules().
The reason is when it is killed, the 'rule' itself may have already
released, we should not access it. one example: we add a rule to an
inode, just at the same time the other task is deleting this inode.
The work flow for adding a rule:
audit_receive() -> (need audit_cmd_mutex lock)
audit_receive_skb() ->
audit_receive_msg() ->
audit_receive_filter() ->
audit_add_rule() ->
audit_add_tree_rule() -> (need audit_filter_mutex lock)
...
unlock audit_filter_mutex
get_tree()
...
iterate_mounts() -> (iterate all related inodes)
tag_mount() ->
tag_trunk() ->
create_trunk() -> (assume it is 1st rule)
fsnotify_add_mark() ->
fsnotify_add_inode_mark() -> (add mark to inode->i_fsnotify_marks)
...
get_tree(); (each inode will get one)
...
lock audit_filter_mutex
The work flow for deleting an inode:
__destroy_inode() ->
fsnotify_inode_delete() ->
__fsnotify_inode_delete() ->
fsnotify_clear_marks_by_inode() -> (get mark from inode->i_fsnotify_marks)
fsnotify_destroy_mark() ->
fsnotify_destroy_mark_locked() ->
audit_tree_freeing_mark() ->
evict_chunk() ->
...
tree->goner = 1
...
kill_rules() -> (assume current->audit_context == NULL)
call_rcu() -> (rule->tree != NULL)
audit_free_rule_rcu() ->
audit_free_rule()
...
audit_schedule_prune() -> (assume current->audit_context == NULL)
kthread_run() -> (need audit_cmd_mutex and audit_filter_mutex lock)
prune_one() -> (delete it from prue_list)
put_tree(); (match the original get_tree above)
Signed-off-by: Chen Gang <gang.chen@asianux.com>
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>
audit_log_start() does wait_for_auditd() in a loop until
audit_backlog_wait_time passes or audit_skb_queue has a room.
If signal_pending() is true this becomes a busy-wait loop, schedule() in
TASK_INTERRUPTIBLE won't block.
Thanks to Guy for fully investigating and explaining the problem.
(akpm: that'll cause the system to lock up on a non-preemptible
uniprocessor kernel)
(Guy: "Our customer was in fact running a uniprocessor machine, and they
reported a system hang.")
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Reported-by: Guy Streeter <streeter@redhat.com>
Cc: Eric Paris <eparis@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The dmesg_restrict sysctl currently covers the syslog method for access
dmesg, however /dev/kmsg isn't covered by the same protections. Most
people haven't noticed because util-linux dmesg(1) defaults to using the
syslog method for access in older versions. With util-linux dmesg(1)
defaults to reading directly from /dev/kmsg.
To fix /dev/kmsg, let's compare the existing interfaces and what they
allow:
- /proc/kmsg allows:
- open (SYSLOG_ACTION_OPEN) if CAP_SYSLOG since it uses a destructive
single-reader interface (SYSLOG_ACTION_READ).
- everything, after an open.
- syslog syscall allows:
- anything, if CAP_SYSLOG.
- SYSLOG_ACTION_READ_ALL and SYSLOG_ACTION_SIZE_BUFFER, if
dmesg_restrict==0.
- nothing else (EPERM).
The use-cases were:
- dmesg(1) needs to do non-destructive SYSLOG_ACTION_READ_ALLs.
- sysklog(1) needs to open /proc/kmsg, drop privs, and still issue the
destructive SYSLOG_ACTION_READs.
AIUI, dmesg(1) is moving to /dev/kmsg, and systemd-journald doesn't
clear the ring buffer.
Based on the comments in devkmsg_llseek, it sounds like actions besides
reading aren't going to be supported by /dev/kmsg (i.e.
SYSLOG_ACTION_CLEAR), so we have a strict subset of the non-destructive
syslog syscall actions.
To this end, move the check as Josh had done, but also rename the
constants to reflect their new uses (SYSLOG_FROM_CALL becomes
SYSLOG_FROM_READER, and SYSLOG_FROM_FILE becomes SYSLOG_FROM_PROC).
SYSLOG_FROM_READER allows non-destructive actions, and SYSLOG_FROM_PROC
allows destructive actions after a capabilities-constrained
SYSLOG_ACTION_OPEN check.
- /dev/kmsg allows:
- open if CAP_SYSLOG or dmesg_restrict==0
- reading/polling, after open
Addresses https://bugzilla.redhat.com/show_bug.cgi?id=903192
[akpm@linux-foundation.org: use pr_warn_once()]
Signed-off-by: Kees Cook <keescook@chromium.org>
Reported-by: Christian Kujau <lists@nerdbynature.de>
Tested-by: Josh Boyer <jwboyer@redhat.com>
Cc: Kay Sievers <kay@vrfy.org>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
