Pull 'full dynticks' support from Ingo Molnar:
"This tree from Frederic Weisbecker adds a new, (exciting! :-) core
kernel feature to the timer and scheduler subsystems: 'full dynticks',
or CONFIG_NO_HZ_FULL=y.
This feature extends the nohz variable-size timer tick feature from
idle to busy CPUs (running at most one task) as well, potentially
reducing the number of timer interrupts significantly.
This feature got motivated by real-time folks and the -rt tree, but
the general utility and motivation of full-dynticks runs wider than
that:
- HPC workloads get faster: CPUs running a single task should be able
to utilize a maximum amount of CPU power. A periodic timer tick at
HZ=1000 can cause a constant overhead of up to 1.0%. This feature
removes that overhead - and speeds up the system by 0.5%-1.0% on
typical distro configs even on modern systems.
- Real-time workload latency reduction: CPUs running critical tasks
should experience as little jitter as possible. The last remaining
source of kernel-related jitter was the periodic timer tick.
- A single task executing on a CPU is a pretty common situation,
especially with an increasing number of cores/CPUs, so this feature
helps desktop and mobile workloads as well.
The cost of the feature is mainly related to increased timer
reprogramming overhead when a CPU switches its tick period, and thus
slightly longer to-idle and from-idle latency.
Configuration-wise a third mode of operation is added to the existing
two NOHZ kconfig modes:
- CONFIG_HZ_PERIODIC: [formerly !CONFIG_NO_HZ], now explicitly named
as a config option. This is the traditional Linux periodic tick
design: there's a HZ tick going on all the time, regardless of
whether a CPU is idle or not.
- CONFIG_NO_HZ_IDLE: [formerly CONFIG_NO_HZ=y], this turns off the
periodic tick when a CPU enters idle mode.
- CONFIG_NO_HZ_FULL: this new mode, in addition to turning off the
tick when a CPU is idle, also slows the tick down to 1 Hz (one
timer interrupt per second) when only a single task is running on a
CPU.
The .config behavior is compatible: existing !CONFIG_NO_HZ and
CONFIG_NO_HZ=y settings get translated to the new values, without the
user having to configure anything. CONFIG_NO_HZ_FULL is turned off by
default.
This feature is based on a lot of infrastructure work that has been
steadily going upstream in the last 2-3 cycles: related RCU support
and non-periodic cputime support in particular is upstream already.
This tree adds the final pieces and activates the feature. The pull
request is marked RFC because:
- it's marked 64-bit only at the moment - the 32-bit support patch is
small but did not get ready in time.
- it has a number of fresh commits that came in after the merge
window. The overwhelming majority of commits are from before the
merge window, but still some aspects of the tree are fresh and so I
marked it RFC.
- it's a pretty wide-reaching feature with lots of effects - and
while the components have been in testing for some time, the full
combination is still not very widely used. That it's default-off
should reduce its regression abilities and obviously there are no
known regressions with CONFIG_NO_HZ_FULL=y enabled either.
- the feature is not completely idempotent: there is no 100%
equivalent replacement for a periodic scheduler/timer tick. In
particular there's ongoing work to map out and reduce its effects
on scheduler load-balancing and statistics. This should not impact
correctness though, there are no known regressions related to this
feature at this point.
- it's a pretty ambitious feature that with time will likely be
enabled by most Linux distros, and we'd like you to make input on
its design/implementation, if you dislike some aspect we missed.
Without flaming us to crisp! :-)
Future plans:
- there's ongoing work to reduce 1Hz to 0Hz, to essentially shut off
the periodic tick altogether when there's a single busy task on a
CPU. We'd first like 1 Hz to be exposed more widely before we go
for the 0 Hz target though.
- once we reach 0 Hz we can remove the periodic tick assumption from
nr_running>=2 as well, by essentially interrupting busy tasks only
as frequently as the sched_latency constraints require us to do -
once every 4-40 msecs, depending on nr_running.
I am personally leaning towards biting the bullet and doing this in
v3.10, like the -rt tree this effort has been going on for too long -
but the final word is up to you as usual.
More technical details can be found in Documentation/timers/NO_HZ.txt"
* 'timers-nohz-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (39 commits)
sched: Keep at least 1 tick per second for active dynticks tasks
rcu: Fix full dynticks' dependency on wide RCU nocb mode
nohz: Protect smp_processor_id() in tick_nohz_task_switch()
nohz_full: Add documentation.
cputime_nsecs: use math64.h for nsec resolution conversion helpers
nohz: Select VIRT_CPU_ACCOUNTING_GEN from full dynticks config
nohz: Reduce overhead under high-freq idling patterns
nohz: Remove full dynticks' superfluous dependency on RCU tree
nohz: Fix unavailable tick_stop tracepoint in dynticks idle
nohz: Add basic tracing
nohz: Select wide RCU nocb for full dynticks
nohz: Disable the tick when irq resume in full dynticks CPU
nohz: Re-evaluate the tick for the new task after a context switch
nohz: Prepare to stop the tick on irq exit
nohz: Implement full dynticks kick
nohz: Re-evaluate the tick from the scheduler IPI
sched: New helper to prevent from stopping the tick in full dynticks
sched: Kick full dynticks CPU that have more than one task enqueued.
perf: New helper to prevent full dynticks CPUs from stopping tick
perf: Kick full dynticks CPU if events rotation is needed
...
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>
One can trigger an overflow when using ktime_add_ns() on a 32bit
architecture not supporting CONFIG_KTIME_SCALAR.
When passing a very high value for u64 nsec, e.g. 7881299347898368000
the do_div() function converts this value to seconds (7881299347) which
is still to high to pass to the ktime_set() function as long. The result
in is a negative value.
The problem on my system occurs in the tick-sched.c,
tick_nohz_stop_sched_tick() when time_delta is set to
timekeeping_max_deferment(). The check for time_delta < KTIME_MAX is
valid, thus ktime_add_ns() is called with a too large value resulting in
a negative expire value. This leads to an endless loop in the ticker code:
time_delta: 7881299347898368000
expires = ktime_add_ns(last_update, time_delta)
expires: negative value
This fix caps the value to KTIME_MAX.
This error doesn't occurs on 64bit or architectures supporting
CONFIG_KTIME_SCALAR (e.g. ARM, x86-32).
Cc: stable@vger.kernel.org
Signed-off-by: David Engraf <david.engraf@sysgo.com>
[jstultz: Minor tweaks to commit message & header]
Signed-off-by: John Stultz <john.stultz@linaro.org>
The settimeofday01 test in the LTP testsuite effectively does
gettimeofday(current time);
settimeofday(Jan 1, 1970 + 100 seconds);
settimeofday(current time);
This test causes a stack trace to be displayed on the console during the
setting of timeofday to Jan 1, 1970 + 100 seconds:
[ 131.066751] ------------[ cut here ]------------
[ 131.096448] WARNING: at kernel/time/clockevents.c:209 clockevents_program_event+0x135/0x140()
[ 131.104935] Hardware name: Dinar
[ 131.108150] Modules linked in: sg nfsv3 nfs_acl nfsv4 auth_rpcgss nfs dns_resolver fscache lockd sunrpc nf_conntrack_netbios_ns nf_conntrack_broadcast ipt_MASQUERADE ip6table_mangle ip6t_REJECT nf_conntrack_ipv6 nf_defrag_ipv6 iptable_nat nf_nat_ipv4 nf_nat iptable_mangle ipt_REJECT nf_conntrack_ipv4 nf_defrag_ipv4 xt_conntrack nf_conntrack ebtable_filter ebtables ip6table_filter ip6_tables iptable_filter ip_tables kvm_amd kvm sp5100_tco bnx2 i2c_piix4 crc32c_intel k10temp fam15h_power ghash_clmulni_intel amd64_edac_mod pcspkr serio_raw edac_mce_amd edac_core microcode xfs libcrc32c sr_mod sd_mod cdrom ata_generic crc_t10dif pata_acpi radeon i2c_algo_bit drm_kms_helper ttm drm ahci pata_atiixp libahci libata usb_storage i2c_core dm_mirror dm_region_hash dm_log dm_mod
[ 131.176784] Pid: 0, comm: swapper/28 Not tainted 3.8.0+ #6
[ 131.182248] Call Trace:
[ 131.184684] <IRQ> [<ffffffff810612af>] warn_slowpath_common+0x7f/0xc0
[ 131.191312] [<ffffffff8106130a>] warn_slowpath_null+0x1a/0x20
[ 131.197131] [<ffffffff810b9fd5>] clockevents_program_event+0x135/0x140
[ 131.203721] [<ffffffff810bb584>] tick_program_event+0x24/0x30
[ 131.209534] [<ffffffff81089ab1>] hrtimer_interrupt+0x131/0x230
[ 131.215437] [<ffffffff814b9600>] ? cpufreq_p4_target+0x130/0x130
[ 131.221509] [<ffffffff81619119>] smp_apic_timer_interrupt+0x69/0x99
[ 131.227839] [<ffffffff8161805d>] apic_timer_interrupt+0x6d/0x80
[ 131.233816] <EOI> [<ffffffff81099745>] ? sched_clock_cpu+0xc5/0x120
[ 131.240267] [<ffffffff814b9ff0>] ? cpuidle_wrap_enter+0x50/0xa0
[ 131.246252] [<ffffffff814b9fe9>] ? cpuidle_wrap_enter+0x49/0xa0
[ 131.252238] [<ffffffff814ba050>] cpuidle_enter_tk+0x10/0x20
[ 131.257877] [<ffffffff814b9c89>] cpuidle_idle_call+0xa9/0x260
[ 131.263692] [<ffffffff8101c42f>] cpu_idle+0xaf/0x120
[ 131.268727] [<ffffffff815f8971>] start_secondary+0x255/0x257
[ 131.274449] ---[ end trace 1151a50552231615 ]---
When we change the system time to a low value like this, the value of
timekeeper->offs_real will be a negative value.
It seems that the WARN occurs because an hrtimer has been started in the time
between the releasing of the timekeeper lock and the IPI call (via a call to
on_each_cpu) in clock_was_set() in the do_settimeofday() code. The end result
is that a REALTIME_CLOCK timer has been added with softexpires = expires =
KTIME_MAX. The hrtimer_interrupt() fires/is called and the loop at
kernel/hrtimer.c:1289 is executed. In this loop the code subtracts the
clock base's offset (which was set to timekeeper->offs_real in
do_settimeofday()) from the current hrtimer_cpu_base->expiry value (which
was KTIME_MAX):
KTIME_MAX - (a negative value) = overflow
A simple check for an overflow can resolve this problem. Using KTIME_MAX
instead of the overflow value will result in the hrtimer function being run,
and the reprogramming of the timer after that.
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: stable@vger.kernel.org
Reviewed-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Prarit Bhargava <prarit@redhat.com>
[jstultz: Tweaked commit subject]
Signed-off-by: John Stultz <john.stultz@linaro.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>
The current code makes the assumption that a cpu_base lock won't be
held if the CPU corresponding to that cpu_base is offline, which isn't
always true.
If a hrtimer is not queued, then it will not be migrated by
migrate_hrtimers() when a CPU is offlined. Therefore, the hrtimer's
cpu_base may still point to a CPU which has subsequently gone offline
if the timer wasn't enqueued at the time the CPU went down.
Normally this wouldn't be a problem, but a cpu_base's lock is blindly
reinitialized each time a CPU is brought up. If a CPU is brought
online during the period that another thread is performing a hrtimer
operation on a stale hrtimer, then the lock will be reinitialized
under its feet, and a SPIN_BUG() like the following will be observed:
<0>[ 28.082085] BUG: spinlock already unlocked on CPU#0, swapper/0/0
<0>[ 28.087078] lock: 0xc4780b40, value 0x0 .magic: dead4ead, .owner: <none>/-1, .owner_cpu: -1
<4>[ 42.451150] [<c0014398>] (unwind_backtrace+0x0/0x120) from [<c0269220>] (do_raw_spin_unlock+0x44/0xdc)
<4>[ 42.460430] [<c0269220>] (do_raw_spin_unlock+0x44/0xdc) from [<c071b5bc>] (_raw_spin_unlock+0x8/0x30)
<4>[ 42.469632] [<c071b5bc>] (_raw_spin_unlock+0x8/0x30) from [<c00a9ce0>] (__hrtimer_start_range_ns+0x1e4/0x4f8)
<4>[ 42.479521] [<c00a9ce0>] (__hrtimer_start_range_ns+0x1e4/0x4f8) from [<c00aa014>] (hrtimer_start+0x20/0x28)
<4>[ 42.489247] [<c00aa014>] (hrtimer_start+0x20/0x28) from [<c00e6190>] (rcu_idle_enter_common+0x1ac/0x320)
<4>[ 42.498709] [<c00e6190>] (rcu_idle_enter_common+0x1ac/0x320) from [<c00e6440>] (rcu_idle_enter+0xa0/0xb8)
<4>[ 42.508259] [<c00e6440>] (rcu_idle_enter+0xa0/0xb8) from [<c000f268>] (cpu_idle+0x24/0xf0)
<4>[ 42.516503] [<c000f268>] (cpu_idle+0x24/0xf0) from [<c06ed3c0>] (rest_init+0x88/0xa0)
<4>[ 42.524319] [<c06ed3c0>] (rest_init+0x88/0xa0) from [<c0c00978>] (start_kernel+0x3d0/0x434)
As an example, this particular crash occurred when hrtimer_start() was
executed on CPU #0. The code locked the hrtimer's current cpu_base
corresponding to CPU #1. CPU #0 then tried to switch the hrtimer's
cpu_base to an optimal CPU which was online. In this case, it selected
the cpu_base corresponding to CPU #3.
Before it could proceed, CPU #1 came online and reinitialized the
spinlock corresponding to its cpu_base. Thus now CPU #0 held a lock
which was reinitialized. When CPU #0 finally ended up unlocking the
old cpu_base corresponding to CPU #1 so that it could switch to CPU
#3, we hit this SPIN_BUG() above while in switch_hrtimer_base().
CPU #0 CPU #1
---- ----
... <offline>
hrtimer_start()
lock_hrtimer_base(base #1)
... init_hrtimers_cpu()
switch_hrtimer_base() ...
... raw_spin_lock_init(&cpu_base->lock)
raw_spin_unlock(&cpu_base->lock) ...
<spin_bug>
Solve this by statically initializing the lock.
Signed-off-by: Michael Bohan <mbohan@codeaurora.org>
Link: http://lkml.kernel.org/r/1363745965-23475-1-git-send-email-mbohan@codeaurora.org
Cc: stable@vger.kernel.org
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
The comments mention HRTIMER_ABS and HRTIMER_REL, these symbols don't
exist, the proper names are HRTIMER_MODE_ABS and HRTIMER_MODE_REL.
Signed-off-by: David Daney <david.daney@cavium.com>
Cc: Jiri Kosina <trivial@kernel.org>
Link: http://lkml.kernel.org/r/1363202438-21234-1-git-send-email-ddaney.cavm@gmail.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Pull timer changes from Ingo Molnar:
"Main changes:
- ntp: Add CONFIG_RTC_SYSTOHC: a generic RTC driver facility
complementing the existing CONFIG_RTC_HCTOSYS, which uses NTP to
keep the hardware clock updated.
- posix-timers: Fix clock_adjtime to always return timex data on
success. This is changing the ABI, but no breakage was expected
and found - caution is warranted nevertheless.
- platform persistent clock improvements/cleanups.
- clockevents: refactor timer broadcast handling to be more generic
and less duplicated with matching architecture code (mostly ARM
motivated.)
- various fixes and cleanups"
* 'timers-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
timers/x86/hpet: Use HPET_COUNTER to specify the hpet counter in vread_hpet()
posix-cpu-timers: Fix nanosleep task_struct leak
clockevents: Fix generic broadcast for FEAT_C3STOP
time, Fix setting of hardware clock in NTP code
hrtimer: Prevent hrtimer_enqueue_reprogram race
clockevents: Add generic timer broadcast function
clockevents: Add generic timer broadcast receiver
timekeeping: Switch HAS_PERSISTENT_CLOCK to ALWAYS_USE_PERSISTENT_CLOCK
x86/time/rtc: Don't print extended CMOS year when reading RTC
x86: Select HAS_PERSISTENT_CLOCK on x86
timekeeping: Add CONFIG_HAS_PERSISTENT_CLOCK option
rtc: Skip the suspend/resume handling if persistent clock exist
timekeeping: Add persistent_clock_exist flag
posix-timers: Fix clock_adjtime to always return timex data on success
Round the calculated scale factor in set_cyc2ns_scale()
NTP: Add a CONFIG_RTC_SYSTOHC configuration
MAINTAINERS: Update John Stultz's email
time: create __getnstimeofday for WARNless calls
Move rt scheduler definitions out of include/linux/sched.h into
new file include/linux/sched/rt.h
Signed-off-by: Clark Williams <williams@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Link: http://lkml.kernel.org/r/20130207094707.7b9f825f@riff.lan
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Move the sysctl-related bits from include/linux/sched.h into
a new file: include/linux/sched/sysctl.h. Then update source
files requiring access to those bits by including the new
header file.
Signed-off-by: Clark Williams <williams@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Link: http://lkml.kernel.org/r/20130207094659.06dced96@riff.lan
Signed-off-by: Ingo Molnar <mingo@kernel.org>
hrtimer_enqueue_reprogram contains a race which could result in
timer.base switch during unlock/lock sequence.
hrtimer_enqueue_reprogram is releasing the lock protecting the timer
base for calling raise_softirq_irqsoff() due to a lock ordering issue
versus rq->lock.
If during that time another CPU calls __hrtimer_start_range_ns() on
the same hrtimer, the timer base might switch, before the current CPU
can lock base->lock again and therefor the unlock_timer_base() call
will unlock the wrong lock.
[ tglx: Added comment and massaged changelog ]
Signed-off-by: Leonid Shatz <leonid.shatz@ravellosystems.com>
Signed-off-by: Izik Eidus <izik.eidus@ravellosystems.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: stable@vger.kernel.org
Link: http://lkml.kernel.org/r/1359981217-389-1-git-send-email-izik.eidus@ravellosystems.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
The update of the hrtimer base offsets on all cpus cannot be made
atomically from the timekeeper.lock held and interrupt disabled region
as smp function calls are not allowed there.
clock_was_set(), which enforces the update on all cpus, is called
either from preemptible process context in case of do_settimeofday()
or from the softirq context when the offset modification happened in
the timer interrupt itself due to a leap second.
In both cases there is a race window for an hrtimer interrupt between
dropping timekeeper lock, enabling interrupts and clock_was_set()
issuing the updates. Any interrupt which arrives in that window will
see the new time but operate on stale offsets.
So we need to make sure that an hrtimer interrupt always sees a
consistent state of time and offsets.
ktime_get_update_offsets() allows us to get the current monotonic time
and update the per cpu hrtimer base offsets from hrtimer_interrupt()
to capture a consistent state of monotonic time and the offsets. The
function replaces the existing ktime_get() calls in hrtimer_interrupt().
The overhead of the new function vs. ktime_get() is minimal as it just
adds two store operations.
This ensures that any changes to realtime or boottime offsets are
noticed and stored into the per-cpu hrtimer base structures, prior to
any hrtimer expiration and guarantees that timers are not expired early.
Signed-off-by: John Stultz <johnstul@us.ibm.com>
Reviewed-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Prarit Bhargava <prarit@redhat.com>
Cc: stable@vger.kernel.org
Link: http://lkml.kernel.org/r/1341960205-56738-8-git-send-email-johnstul@us.ibm.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
We need to update the base offsets from this code and we need to do
that under base->lock. Move the lock held region around the
ktime_get() calls. The ktime_get() calls are going to be replaced with
a function which gets the time and the offsets atomically.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Prarit Bhargava <prarit@redhat.com>
Cc: stable@vger.kernel.org
Signed-off-by: John Stultz <johnstul@us.ibm.com>
Link: http://lkml.kernel.org/r/1341960205-56738-6-git-send-email-johnstul@us.ibm.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
clock_was_set() cannot be called from hard interrupt context because
it calls on_each_cpu().
For fixing the widely reported leap seconds issue it is necessary to
call it from hard interrupt context, i.e. the timer tick code, which
does the timekeeping updates.
Provide a new function which denotes it in the hrtimer cpu base
structure of the cpu on which it is called and raise the hrtimer
softirq. We then execute the clock_was_set() notificiation from
softirq context in run_hrtimer_softirq(). The hrtimer softirq is
rarely used, so polling the flag there is not a performance issue.
[ tglx: Made it depend on CONFIG_HIGH_RES_TIMERS. We really should get
rid of all this ifdeffery ASAP ]
Signed-off-by: John Stultz <johnstul@us.ibm.com>
Reported-by: Jan Engelhardt <jengelh@inai.de>
Reviewed-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Prarit Bhargava <prarit@redhat.com>
Cc: stable@vger.kernel.org
Link: http://lkml.kernel.org/r/1341960205-56738-2-git-send-email-johnstul@us.ibm.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
* 'timers-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
hrtimer: Fix extra wakeups from __remove_hrtimer()
timekeeping: add arch_offset hook to ktime_get functions
clocksource: Avoid selecting mult values that might overflow when adjusted
time: Improve documentation of timekeeeping_adjust()
__remove_hrtimer() attempts to reprogram the clockevent device when
the timer being removed is the next to expire. However,
__remove_hrtimer() reprograms the clockevent *before* removing the
timer from the timerqueue and thus when hrtimer_force_reprogram()
finds the next timer to expire it finds the timer we're trying to
remove.
This is especially noticeable when the system switches to NOHz mode
and the system tick is removed. The timer tick is removed from the
system but the clockevent is programmed to wakeup in another HZ
anyway.
Silence the extra wakeup by removing the timer from the timerqueue
before calling hrtimer_force_reprogram() so that we actually program
the clockevent for the next timer to expire.
This was broken by 998adc3 "hrtimers: Convert hrtimers to use
timerlist infrastructure".
Signed-off-by: Jeff Ohlstein <johlstei@codeaurora.org>
Cc: stable@vger.kernel.org
Link: http://lkml.kernel.org/r/1321660030-8520-1-git-send-email-johlstei@codeaurora.org
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
The changed files were only including linux/module.h for the
EXPORT_SYMBOL infrastructure, and nothing else. Revector them
onto the isolated export header for faster compile times.
Nothing to see here but a whole lot of instances of:
-#include <linux/module.h>
+#include <linux/export.h>
This commit is only changing the kernel dir; next targets
will probably be mm, fs, the arch dirs, etc.
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
commit 9ec2690758 ("timerfd: Manage cancelable timers in timerfd")
introduced a CONFIG_HIGHRES_TIMERS (should be CONFIG_HIGH_RES_TIMERS)
typo, which caused applications depending on CLOCK_REALTIME timers to
become sluggy due to the fact that the time base of the realtime
timers was not updated when the wall clock time was set.
This causes anything from 100% CPU use for some applications to odd
delays and hickups.
Reported-bisected-and-tested-by: Anca Emanuel <anca.emanuel@gmail.com>
Tested-by: Linus Torvalds <torvalds@linux-foundation.org>
Fatfingered-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The ordering of the clock bases is historical due to the
CLOCK_REALTIME and CLOCK_MONOTONIC constants. Now the hrtimer bases
have their own enumeration due to the gap between CLOCK_MONOTONIC and
CLOCK_BOOTTIME. So we can be more clever as most timers end up on the
CLOCK_MONOTONIC base due to the virtue of POSIX declaring that
relative CLOCK_REALTIME timers are not affected by time changes. In
desktop environments this is slowly changing as applications switch to
absolute timers, but I've observed empty CLOCK_REALTIME bases often
enough. There is no performance penalty or overhead when
CLOCK_REALTIME timers are active, but in case they are not we don't
skip over a full cache line.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Peter Zijlstra <peterz@infradead.org>
Instead of iterating over all possible timer bases avoid it by marking
the active bases in the cpu base.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Peter Zijlstra <peterz@infradead.org>
Peter is concerned about the extra scan of CLOCK_REALTIME_COS in the
timer interrupt. Yes, I did not think about it, because the solution
was so elegant. I didn't like the extra list in timerfd when it was
proposed some time ago, but with a rcu based list the list walk it's
less horrible than the original global lock, which was held over the
list iteration.
Requested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Peter Zijlstra <peterz@infradead.org>
Some applications must be aware of clock realtime being set
backward. A simple example is a clock applet which arms a timer for
the next minute display. If clock realtime is set backward then the
applet displays a stale time for the amount of time which the clock
was set backwards. Due to that applications poll the time because we
don't have an interface.
Extend the timerfd interface by adding a flag which puts the timer
onto a different internal realtime clock. All timers on this clock are
expired whenever the clock was set.
The timerfd core records the monotonic offset when the timer is
created. When the timer is armed, then the current offset is compared
to the previous recorded offset. When it has changed, then
timerfd_settime returns -ECANCELED. When a timer is read the offset is
compared and if it changed -ECANCELED returned to user space. Periodic
timers are not rearmed in the cancelation case.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: John Stultz <johnstul@us.ibm.com>
Cc: Chris Friesen <chris.friesen@genband.com>
Tested-by: Kay Sievers <kay.sievers@vrfy.org>
Cc: "Kirill A. Shutemov" <kirill@shutemov.name>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Davide Libenzi <davidel@xmailserver.org>
Reviewed-by: Alexander Shishkin <virtuoso@slind.org>
Link: http://lkml.kernel.org/r/%3Calpine.LFD.2.02.1104271359580.3323%40ionos%3E
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Make clock_was_set() unconditional and rename hres_timers_resume to
hrtimers_resume. This is a preparatory patch for hrtimers which are
cancelled when clock realtime was set.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Sedat and Bruno reported RCU stalls which turned out to be caused by
the following;
sched_init() calls init_rt_bandwidth() which calls hrtimer_init()
_BEFORE_ hrtimers_init() is called. While not entirely correct this
worked because hrtimer_init() only accessed statically initialized
data (hrtimer_bases.clock_base[CLOCK_MONOTONIC])
Commit e06383db9 (hrtimers: extend hrtimer base code to handle more
then 2 clockids) added an indirection to the hrtimer_bases.clock_base
lookup to avoid gap handling in the hot path. The table which is used
for the translataion from CLOCK_ID to HRTIMER_BASE index is
initialized at runtime in hrtimers_init(). So the early call of the
scheduler code translates CLOCK_MONOTONIC to HRTIMER_BASE_REALTIME.
Thus the rt_bandwith timer ends up on CLOCK_REALTIME. If the timer is
armed and the wall clock time is set (e.g. ntpdate in the early boot
process - which also gives the problem deterministic behaviour
i.e. magic recovery after N hours), then the timer ends up with an
expiry time far into the future. That breaks the RT throttler
mechanism as rt runtime is accumulated and never cleared, so the rt
throttler detects a false cpu hog condition and blocks all RT tasks
until the timer finally expires. That in turn stalls the RCU thread of
TINYRCU which leads to an huge amount of RCU callbacks piling up.
Make the translation table statically initialized, so we are back to
the status of <= 2.6.39.
Reported-and-tested-by: Sedat Dilek <sedat.dilek@gmail.com>
Reported-by: Bruno Prémont <bonbons@linux-vserver.org>
Cc: John stultz <johnstul@us.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/%3Calpine.LFD.2.02.1104282353140.3005%40ionos%3E
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
* 'timers-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (62 commits)
posix-clocks: Check write permissions in posix syscalls
hrtimer: Remove empty hrtimer_init_hres_timer()
hrtimer: Update hrtimer->state documentation
hrtimer: Update base[CLOCK_BOOTTIME].offset correctly
timers: Export CLOCK_BOOTTIME via the posix timers interface
timers: Add CLOCK_BOOTTIME hrtimer base
time: Extend get_xtime_and_monotonic_offset() to also return sleep
time: Introduce get_monotonic_boottime and ktime_get_boottime
hrtimers: extend hrtimer base code to handle more then 2 clockids
ntp: Remove redundant and incorrect parameter check
mn10300: Switch do_timer() to xtimer_update()
posix clocks: Introduce dynamic clocks
posix-timers: Cleanup namespace
posix-timers: Add support for fd based clocks
x86: Add clock_adjtime for x86
posix-timers: Introduce a syscall for clock tuning.
time: Splitout compat timex accessors
ntp: Add ADJ_SETOFFSET mode bit
time: Introduce timekeeping_inject_offset
posix-timer: Update comment
...
Fix up new system-call-related conflicts in
arch/x86/ia32/ia32entry.S
arch/x86/include/asm/unistd_32.h
arch/x86/include/asm/unistd_64.h
arch/x86/kernel/syscall_table_32.S
(name_to_handle_at()/open_by_handle_at() vs clock_adjtime()), and some
due to movement of get_jiffies_64() in:
kernel/time.c
In complex subsystems like mac80211 structures can contain several
timers and work structs, so identifying a specific instance from the
call trace and object type output of debugobjects can be hard.
Allow the subsystems which support debugobjects to provide a hint
function. This function returns a pointer to a kernel address
(preferrably the objects callback function) which is printed along
with the debugobjects type.
Add hint methods for timer_list, work_struct and hrtimer.
[ tglx: Massaged changelog, made it compile ]
Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
LKML-Reference: <20110307085809.GA9334@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
We calculate the current time of each clock base by adding an offset
to clock_monotonic. The offset for the clock bases is set in
retrigger_next_event() which is called when we switch a cpu to highres
mode or when the clock was set.
Add the missing update for clock boottime.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: John Stultz <johnstul@us.ibm.com>
CLOCK_MONOTONIC stops while the system is in suspend. This is because
to applications system suspend is invisible. However, there is a
growing set of applications that are wanting to be suspend-aware,
but do not want to deal with the complications of CLOCK_REALTIME
(which might jump around if settimeofday is called).
For these applications, I propose a new clockid: CLOCK_BOOTTIME.
CLOCK_BOOTTIME is idential to CLOCK_MONOTONIC, except it also
includes any time spent in suspend.
This patch add hrtimer base for CLOCK_BOOTTIME, using
get_monotonic_boottime/ktime_get_boottime, to allow
in kernel users to set timers against.
CC: Jamie Lokier <jamie@shareable.org>
CC: Thomas Gleixner <tglx@linutronix.de>
CC: Alexander Shishkin <virtuoso@slind.org>
CC: Arve Hjønnevåg <arve@android.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
The hrtimer code is written mainly with CLOCK_REALTIME and CLOCK_MONOTONIC
in mind. These are clockids 0 and 1 resepctively. However, if we are
to introduce any new hrtimer bases, using new clockids, we have to skip
the cputimers (clockids 2,3) as well as other clockids that may not impelement
timers.
This patch adds a little bit of indirection between the clockid and
the base, so that we can extend the base by one when we add
a new clockid at number 7 or so.
CC: Jamie Lokier <jamie@shareable.org>
CC: Thomas Gleixner <tglx@linutronix.de>
CC: Alexander Shishkin <virtuoso@slind.org>
CC: Arve Hjønnevåg <arve@android.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
The hrtimer code accesses timekeeping variables under
xtime_lock. Provide a sensible accessor function and use it.
[ tglx: Removed the conditionals, unused variable, fixed codingstyle
and massaged changelog ]
Signed-off-by: Torben Hohn <torbenh@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: johnstul@us.ibm.com
Cc: yong.zhang0@gmail.com
Cc: hch@infradead.org
LKML-Reference: <20110127145905.23248.30458.stgit@localhost>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
* 'for-2.6.38' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/percpu: (30 commits)
gameport: use this_cpu_read instead of lookup
x86: udelay: Use this_cpu_read to avoid address calculation
x86: Use this_cpu_inc_return for nmi counter
x86: Replace uses of current_cpu_data with this_cpu ops
x86: Use this_cpu_ops to optimize code
vmstat: User per cpu atomics to avoid interrupt disable / enable
irq_work: Use per cpu atomics instead of regular atomics
cpuops: Use cmpxchg for xchg to avoid lock semantics
x86: this_cpu_cmpxchg and this_cpu_xchg operations
percpu: Generic this_cpu_cmpxchg() and this_cpu_xchg support
percpu,x86: relocate this_cpu_add_return() and friends
connector: Use this_cpu operations
xen: Use this_cpu_inc_return
taskstats: Use this_cpu_ops
random: Use this_cpu_inc_return
fs: Use this_cpu_inc_return in buffer.c
highmem: Use this_cpu_xx_return() operations
vmstat: Use this_cpu_inc_return for vm statistics
x86: Support for this_cpu_add, sub, dec, inc_return
percpu: Generic support for this_cpu_add, sub, dec, inc_return
...
Fixed up conflicts: in arch/x86/kernel/{apic/nmi.c, apic/x2apic_uv_x.c, process.c}
as per Tejun.
* 'timers-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
MAINTAINERS: Update timer related entries
timers: Use this_cpu_read
timerqueue: Make timerqueue_getnext() static inline
hrtimer: fix timerqueue conversion flub
hrtimers: Convert hrtimers to use timerlist infrastructure
timers: Fixup allmodconfig build issue
timers: Rename timerlist infrastructure to timerqueue
timers: Introduce timerlist infrastructure.
hrtimer: Remove stale comment on curr_timer
timer: Warn when del_timer_sync() is called in hardirq context
timer: Del_timer_sync() can be used in softirq context
timer: Make try_to_del_timer_sync() the same on SMP and UP
posix-timers: Annotate lock_timer()
timer: Permit statically-declared work with deferrable timers
time: Use ARRAY_SIZE macro in timecompare.c
timer: Initialize the field slack of timer_list
timer_list: Remove alignment padding on 64 bit when CONFIG_TIMER_STATS
time: Compensate for rounding on odd-frequency clocksources
Fix up trivial conflict in MAINTAINERS
__get_cpu_var() can be replaced with this_cpu_read and will then use a
single read instruction with implied address calculation to access the
correct per cpu instance.
However, the address of a per cpu variable passed to __this_cpu_read()
cannot be determined (since it's an implied address conversion through
segment prefixes). Therefore apply this only to uses of __get_cpu_var
where the address of the variable is not used.
Cc: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: Hugh Dickins <hughd@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Acked-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Christoph Lameter <cl@linux.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
In converting the hrtimers to timerqueue, I missed
a spot in hrtimer_run_queues where we loop running
timers. We end up not pulling the new next value out
and instead just use the last next value, causing
boot time hangs in some cases.
The proper fix is to pull timerqueue_getnext each iteration
instead of using a local next value.
Reported-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Converts the hrtimer code to use the new timerlist infrastructure
Signed-off-by: John Stultz <john.stultz@linaro.org>
LKML Reference: <1290136329-18291-3-git-send-email-john.stultz@linaro.org>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
CC: Alessandro Zummo <a.zummo@towertech.it>
CC: Thomas Gleixner <tglx@linutronix.de>
CC: Richard Cochran <richardcochran@gmail.com>
The race is described as follows:
CPU X CPU Y
remove_hrtimer
// state & QUEUED == 0
timer->state = CALLBACK
unlock timer base
timer->f(n) //very long
hrtimer_start
lock timer base
remove_hrtimer // no effect
hrtimer_enqueue
timer->state = CALLBACK |
QUEUED
unlock timer base
hrtimer_start
lock timer base
remove_hrtimer
mode = INACTIVE
// CALLBACK bit lost!
switch_hrtimer_base
CALLBACK bit not set:
timer->base
changes to a
different CPU.
lock this CPU's timer base
The bug was introduced with commit ca109491f (hrtimer: removing all ur
callback modes) in 2.6.29
[ tglx: Feed new state via local variable and add a comment. ]
Signed-off-by: Salman Qazi <sqazi@google.com>
Cc: akpm@linux-foundation.org
Cc: Peter Zijlstra <peterz@infradead.org>
LKML-Reference: <20101012142351.8485.21823.stgit@dungbeetle.mtv.corp.google.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: stable@kernel.org
No real bugs I believe, just some dead code.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: andi@firstfloor.org
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
* 'timers-timekeeping-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
um: Fix read_persistent_clock fallout
kgdb: Do not access xtime directly
powerpc: Clean up obsolete code relating to decrementer and timebase
powerpc: Rework VDSO gettimeofday to prevent time going backwards
clocksource: Add __clocksource_updatefreq_hz/khz methods
x86: Convert common clocksources to use clocksource_register_hz/khz
timekeeping: Make xtime and wall_to_monotonic static
hrtimer: Cleanup direct access to wall_to_monotonic
um: Convert to use read_persistent_clock
timkeeping: Fix update_vsyscall to provide wall_to_monotonic offset
powerpc: Cleanup xtime usage
powerpc: Simplify update_vsyscall
time: Kill off CONFIG_GENERIC_TIME
time: Implement timespec_add
x86: Fix vtime/file timestamp inconsistencies
Trivial conflicts in Documentation/feature-removal-schedule.txt
Much less trivial conflicts in arch/powerpc/kernel/time.c resolved as
per Thomas' earlier merge commit 47916be4e2 ("Merge branch
'powerpc.cherry-picks' into timers/clocksource")
Provides an accessor function to replace hrtimer.c's
direct access of wall_to_monotonic.
This will allow wall_to_monotonic to be made static as
planned in Documentation/feature-removal-schedule.txt
Signed-off-by: John Stultz <johnstul@us.ibm.com>
LKML-Reference: <1279068988-21864-9-git-send-email-johnstul@us.ibm.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
In the new push model, all idle CPUs indeed go into nohz mode. There is
still the concept of idle load balancer (performing the load balancing
on behalf of all the idle cpu's in the system). Busy CPU kicks the nohz
balancer when any of the nohz CPUs need idle load balancing.
The kickee CPU does the idle load balancing on behalf of all idle CPUs
instead of the normal idle balance.
This addresses the below two problems with the current nohz ilb logic:
* the idle load balancer continued to have periodic ticks during idle and
wokeup frequently, even though it did not have any rebalancing to do on
behalf of any of the idle CPUs.
* On x86 and CPUs that have APIC timer stoppage on idle CPUs, this
periodic wakeup can result in a periodic additional interrupt on a CPU
doing the timer broadcast.
Also currently we are migrating the unpinned timers from an idle to the cpu
doing idle load balancing (when all the cpus in the system are idle,
there is no idle load balancing cpu and timers get added to the same idle cpu
where the request was made. So the existing optimization works only on semi idle
system).
And In semi idle system, we no longer have periodic ticks on the idle load
balancer CPU. Using that cpu will add more delays to the timers than intended
(as that cpu's timer base may not be uptodate wrt jiffies etc). This was
causing mysterious slowdowns during boot etc.
For now, in the semi idle case, use the nearest busy cpu for migrating timers
from an idle cpu. This is good for power-savings anyway.
Signed-off-by: Venkatesh Pallipadi <venki@google.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>
LKML-Reference: <1274486981.2840.46.camel@sbs-t61.sc.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
hrtimer_get_softirq_time() has it's own xtime lock protection, so it's
safe to use plain __current_kernel_time() and avoid the double seqlock
loop.
Signed-off-by: Stanislaw Gruszka <stf_xl@wp.pl>
LKML-Reference: <20100525214912.GA1934@r2bh72.net.upc.cz>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
The current version of schedule_hrtimeout() always uses the
monotonic clock. Some system calls such as mq_timedsend()
and mq_timedreceive(), however, require the use of the wall
clock due to the definition of the system call.
This patch provides the infrastructure to use schedule_hrtimeout()
with a CLOCK_REALTIME timer.
Signed-off-by: Carsten Emde <C.Emde@osadl.org>
Tested-by: Pradyumna Sampath <pradysam@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Arjan van de Veen <arjan@infradead.org>
LKML-Reference: <20100402204331.167439615@osadl.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Linus Torvalds