With various drivers wanting to inject idle time; we get people
calling idle routines outside of the idle loop proper.
Therefore we need to be extra careful about not missing
TIF_NEED_RESCHED -> PREEMPT_NEED_RESCHED propagations.
While looking at this, I also realized there's a small window in the
existing idle loop where we can miss TIF_NEED_RESCHED; when it hits
right after the tif_need_resched() test at the end of the loop but
right before the need_resched() test at the start of the loop.
So move preempt_fold_need_resched() out of the loop where we're
guaranteed to have TIF_NEED_RESCHED set.
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/n/tip-x9jgh45oeayzajz2mjt0y7d6@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
In order to combine the preemption and need_resched test we need to
fold the need_resched information into the preempt_count value.
Since the NEED_RESCHED flag is set across CPUs this needs to be an
atomic operation, however we very much want to avoid making
preempt_count atomic, therefore we keep the existing TIF_NEED_RESCHED
infrastructure in place but at 3 sites test it and fold its value into
preempt_count; namely:
- resched_task() when setting TIF_NEED_RESCHED on the current task
- scheduler_ipi() when resched_task() sets TIF_NEED_RESCHED on a
remote task it follows it up with a reschedule IPI
and we can modify the cpu local preempt_count from
there.
- cpu_idle_loop() for when resched_task() found tsk_is_polling().
We use an inverted bitmask to indicate need_resched so that a 0 means
both need_resched and !atomic.
Also remove the barrier() in preempt_enable() between
preempt_enable_no_resched() and preempt_check_resched() to avoid
having to reload the preemption value and allow the compiler to use
the flags of the previuos decrement. I couldn't come up with any sane
reason for this barrier() to be there as preempt_enable_no_resched()
already has a barrier() before doing the decrement.
Suggested-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/n/tip-7a7m5qqbn5pmwnd4wko9u6da@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Mike reported that commit 7d1a9417 ("x86: Use generic idle loop")
regressed several workloads and caused excessive reschedule
interrupts.
The patch in question failed to notice that the x86 code had an
inverted sense of the polling state versus the new generic code (x86:
default polling, generic: default !polling).
Fix the two prominent x86 mwait based idle drivers and introduce a few
new generic polling helpers (fixing the wrong smp_mb__after_clear_bit
usage).
Also switch the idle routines to using tif_need_resched() which is an
immediate TIF_NEED_RESCHED test as opposed to need_resched which will
end up being slightly different.
Reported-by: Mike Galbraith <bitbucket@online.de>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: lenb@kernel.org
Cc: tglx@linutronix.de
Link: http://lkml.kernel.org/n/tip-nc03imb0etuefmzybzj7sprf@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
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>
Moving x86 to the generic idle implementation (commit 7d1a9417 "x86:
Use generic idle loop") wreckaged the stack protector.
I stupidly missed that boot_init_stack_canary() must be inlined from a
function which never returns, but I put that call into
arch_cpu_idle_prepare() which of course returns.
I pondered to play tricks with arch_cpu_idle_prepare() first, but then
I noticed, that the other archs which have implemented the
stackprotector (ARM and SH) do not initialize the canary for the
non-boot cpus.
So I decided to move the boot_init_stack_canary() call into
cpu_startup_entry() ifdeffed with an CONFIG_X86 for now. This #ifdef
is just a temporary measure as I don't want to inflict the
boot_init_stack_canary() call on ARM and SH that late in the cycle.
I'll queue a patch for 3.11 which removes the #ifdef if the ARM/SH
maintainers have no objection.
Reported-by: Wouter van Kesteren <woutershep@gmail.com>
Cc: x86@kernel.org
Cc: Russell King <linux@arm.linux.org.uk>
Cc: Paul Mundt <lethal@linux-sh.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Bjørn Mork reported the following warning when running powertop.
[ 49.289034] ------------[ cut here ]------------
[ 49.289055] WARNING: at kernel/rcutree.c:502 rcu_eqs_exit_common.isra.48+0x3d/0x125()
[ 49.289244] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 3.10.0-bisect-rcu-warn+ #107
[ 49.289251] ffffffff8157d8c8 ffffffff81801e28 ffffffff8137e4e3 ffffffff81801e68
[ 49.289260] ffffffff8103094f ffffffff81801e68 0000000000000000 ffff88023afcd9b0
[ 49.289268] 0000000000000000 0140000000000000 ffff88023bee7700 ffffffff81801e78
[ 49.289276] Call Trace:
[ 49.289285] [<ffffffff8137e4e3>] dump_stack+0x19/0x1b
[ 49.289293] [<ffffffff8103094f>] warn_slowpath_common+0x62/0x7b
[ 49.289300] [<ffffffff8103097d>] warn_slowpath_null+0x15/0x17
[ 49.289306] [<ffffffff810a9006>] rcu_eqs_exit_common.isra.48+0x3d/0x125
[ 49.289314] [<ffffffff81079b49>] ? trace_hardirqs_off_caller+0x37/0xa6
[ 49.289320] [<ffffffff810a9692>] rcu_idle_exit+0x85/0xa8
[ 49.289327] [<ffffffff8107076e>] trace_cpu_idle_rcuidle+0xae/0xff
[ 49.289334] [<ffffffff810708b1>] cpu_startup_entry+0x72/0x115
[ 49.289341] [<ffffffff813689e5>] rest_init+0x149/0x150
[ 49.289347] [<ffffffff8136889c>] ? csum_partial_copy_generic+0x16c/0x16c
[ 49.289355] [<ffffffff81a82d34>] start_kernel+0x3f0/0x3fd
[ 49.289362] [<ffffffff81a8274c>] ? repair_env_string+0x5a/0x5a
[ 49.289368] [<ffffffff81a82481>] x86_64_start_reservations+0x2a/0x2c
[ 49.289375] [<ffffffff81a82550>] x86_64_start_kernel+0xcd/0xd1
[ 49.289379] ---[ end trace 07a1cc95e29e9036 ]---
The warning is that 'rdtp->dynticks' has an unexpected value, which roughly
translates to - the calls to rcu_idle_enter() and rcu_idle_exit() were not
made in the correct order, or otherwise messed up.
And Bjørn's painstaking debugging indicated that this happens when the idle
loop enters the poll mode. Looking at the poll function cpu_idle_poll(), and
the implementation of trace_cpu_idle_rcuidle(), the problem becomes very clear:
cpu_idle_poll() lacks calls to rcu_idle_enter/exit(), and trace_cpu_idle_rcuidle()
calls them in the reverse order - first rcu_idle_exit(), and then rcu_idle_enter().
Hence the even/odd alternative sequencing of rdtp->dynticks goes for a toss.
And powertop readily triggers this because powertop uses the idle-tracing
infrastructure extensively.
So, to fix this, wrap the code in cpu_idle_poll() within rcu_idle_enter/exit(),
so that it blends properly with the calls inside trace_cpu_idle_rcuidle() and
thus get the function ordering right.
Reported-and-tested-by: Bjørn Mork <bjorn@mork.no>
Cc: Paul McKenney <paulmck@linux.vnet.ibm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Dipankar Sarma <dipankar@in.ibm.com>
Signed-off-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com>
Link: http://lkml.kernel.org/r/519169BF.4080208@linux.vnet.ibm.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Pull core timer updates from Ingo Molnar:
"The main changes in this cycle's merge are:
- Implement shadow timekeeper to shorten in kernel reader side
blocking, by Thomas Gleixner.
- Posix timers enhancements by Pavel Emelyanov:
- allocate timer ID per process, so that exact timer ID allocations
can be re-created be checkpoint/restore code.
- debuggability and tooling (/proc/PID/timers, etc.) improvements.
- suspend/resume enhancements by Feng Tang: on certain new Intel Atom
processors (Penwell and Cloverview), there is a feature that the
TSC won't stop in S3 state, so the TSC value won't be reset to 0
after resume. This can be taken advantage of by the generic via
the CLOCK_SOURCE_SUSPEND_NONSTOP flag: instead of using the RTC to
recover/approximate sleep time, the main (and precise) clocksource
can be used.
- Fix /proc/timer_list for 4096 CPUs by Nathan Zimmer: on so many
CPUs the file goes beyond 4MB of size and thus the current
simplistic seqfile approach fails. Convert /proc/timer_list to a
proper seq_file with its own iterator.
- Cleanups and refactorings of the core timekeeping code by John
Stultz.
- International Atomic Clock time is managed by the NTP code
internally currently but not exposed externally. Separate the TAI
code out and add CLOCK_TAI support and TAI support to the hrtimer
and posix-timer code, by John Stultz.
- Add deep idle support enhacement to the broadcast clockevents core
timer code, by Daniel Lezcano: add an opt-in CLOCK_EVT_FEAT_DYNIRQ
clockevents feature (which will be utilized by future clockevents
driver updates), which allows the use of IRQ affinities to avoid
spurious wakeups of idle CPUs - the right CPU with an expiring
timer will be woken.
- Add new ARM bcm281xx clocksource driver, by Christian Daudt
- ... various other fixes and cleanups"
* 'timers-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (52 commits)
clockevents: Set dummy handler on CPU_DEAD shutdown
timekeeping: Update tk->cycle_last in resume
posix-timers: Remove unused variable
clockevents: Switch into oneshot mode even if broadcast registered late
timer_list: Convert timer list to be a proper seq_file
timer_list: Split timer_list_show_tickdevices
posix-timers: Show sigevent info in proc file
posix-timers: Introduce /proc/PID/timers file
posix timers: Allocate timer id per process (v2)
timekeeping: Make sure to notify hrtimers when TAI offset changes
hrtimer: Fix ktime_add_ns() overflow on 32bit architectures
hrtimer: Add expiry time overflow check in hrtimer_interrupt
timekeeping: Shorten seq_count region
timekeeping: Implement a shadow timekeeper
timekeeping: Delay update of clock->cycle_last
timekeeping: Store cycle_last value in timekeeper struct as well
ntp: Remove ntp_lock, using the timekeeping locks to protect ntp state
timekeeping: Simplify tai updating from do_adjtimex
timekeeping: Hold timekeepering locks in do_adjtimex and hardpps
timekeeping: Move ADJ_SETOFFSET to top level do_adjtimex()
...
All idle functions in arch/* are more or less the same, plus minus a
few bugs and extra instrumentation, tickless support and other
optional items.
Implement a generic idle function which resembles the functionality
found in arch/. Provide weak arch_cpu_idle_* functions which can be
overridden by the architecture code if needed.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Paul McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Reviewed-by: Cc: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com>
Cc: Magnus Damm <magnus.damm@gmail.com>
Link: http://lkml.kernel.org/r/20130321215233.646635455@linutronix.de
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
For now this calls cpu_idle(), but in the long run we want to move the
cpu bringup code to the core and therefor we add a state argument.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Paul McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Reviewed-by: Cc: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com>
Cc: Magnus Damm <magnus.damm@gmail.com>
Link: http://lkml.kernel.org/r/20130321215233.583190032@linutronix.de
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Peter Zijlstra
James Bottomley
Thomas Gleixner
Srivatsa S. Bhat
Linus Torvalds