lwn.git/kernel/time/timer.c, branch doc/4.9

timers/core: Convert to hotplug state machine

2016-07-15T08:41:42+00:00

When tearing down, call timers_dead_cpu() before notify_dead(). There is a hidden dependency between: - timers - block multiqueue - rcutree If timers_dead_cpu() comes later than blk_mq_queue_reinit_notify() that latter function causes a RCU stall. Signed-off-by: Richard Cochran Signed-off-by: Anna-Maria Gleixner Reviewed-by: Sebastian Andrzej Siewior Cc: John Stultz Cc: Linus Torvalds Cc: Oleg Nesterov Cc: Peter Zijlstra Cc: Rasmus Villemoes Cc: Thomas Gleixner Cc: rt@linutronix.de Link: http://lkml.kernel.org/r/20160713153337.566790058@linutronix.de Signed-off-by: Ingo Molnar

Merge branch 'timers/fast-wheel' into timers/core

2016-07-07T08:35:28+00:00

timers: Implement optimization for same expiry time in mod_timer()

2016-07-07T08:35:12+00:00

The existing optimization for same expiry time in mod_timer() checks whether the timer expiry time is the same as the new requested expiry time. In the old timer wheel implementation this does not take the slack batching into account, neither does the new implementation evaluate whether the new expiry time will requeue the timer to the same bucket. To optimize that, we can calculate the resulting bucket and check if the new expiry time is different from the current expiry time. This calculation happens outside the base lock held region. If the resulting bucket is the same we can avoid taking the base lock and requeueing the timer. If the timer needs to be requeued then we have to check under the base lock whether the base time has changed between the lockless calculation and taking the lock. If it has changed we need to recalculate under the lock. This optimization takes effect for timers which are enqueued into the less granular wheel levels (1 and above). With a simple test case the functionality has been verified: Before After Match: 5.5% 86.6% Requeue: 94.5% 13.4% Recalc: <0.01% In the non optimized case the timer is requeued in 94.5% of the cases. With the index optimization in place the requeue rate drops to 13.4%. The case where the lockless index calculation has to be redone is less than 0.01%. With a real world test case (networking) we observed the following changes: Before After Match: 97.8% 99.7% Requeue: 2.2% 0.3% Recalc: <0.001% That means two percent fewer lock/requeue/unlock operations done in one of the hot path use cases of timers. Signed-off-by: Anna-Maria Gleixner Signed-off-by: Thomas Gleixner Cc: Arjan van de Ven Cc: Chris Mason Cc: Eric Dumazet Cc: Frederic Weisbecker Cc: George Spelvin Cc: Josh Triplett Cc: Len Brown Cc: Linus Torvalds Cc: Paul E. McKenney Cc: Peter Zijlstra Cc: Rik van Riel Cc: rt@linutronix.de Link: http://lkml.kernel.org/r/20160704094342.778527749@linutronix.de Signed-off-by: Ingo Molnar

timers: Split out index calculation

2016-07-07T08:35:12+00:00

For further optimizations we need to seperate index calculation from queueing. No functional change. Signed-off-by: Anna-Maria Gleixner Signed-off-by: Thomas Gleixner Cc: Arjan van de Ven Cc: Chris Mason Cc: Eric Dumazet Cc: Frederic Weisbecker Cc: George Spelvin Cc: Josh Triplett Cc: Len Brown Cc: Linus Torvalds Cc: Paul E. McKenney Cc: Peter Zijlstra Cc: Rik van Riel Cc: rt@linutronix.de Link: http://lkml.kernel.org/r/20160704094342.691159619@linutronix.de Signed-off-by: Ingo Molnar

timers: Only wake softirq if necessary

2016-07-07T08:35:11+00:00

With the wheel forwading in place and with the HZ=1000 4ms folding we can avoid running the softirq at all. Signed-off-by: Thomas Gleixner Cc: Arjan van de Ven Cc: Chris Mason Cc: Frederic Weisbecker Cc: George Spelvin Cc: Josh Triplett Cc: Len Brown Cc: Linus Torvalds Cc: Paul McKenney Cc: Peter Zijlstra Cc: Rik van Riel Cc: rt@linutronix.de Link: http://lkml.kernel.org/r/20160704094342.607650550@linutronix.de Signed-off-by: Ingo Molnar

timers: Forward the wheel clock whenever possible

2016-07-07T08:35:11+00:00

The wheel clock is stale when a CPU goes into a long idle sleep. This has the side effect that timers which are queued end up in the outer wheel levels. That results in coarser granularity. To solve this, we keep track of the idle state and forward the wheel clock whenever possible. Signed-off-by: Thomas Gleixner Cc: Arjan van de Ven Cc: Chris Mason Cc: Eric Dumazet Cc: Frederic Weisbecker Cc: George Spelvin Cc: Josh Triplett Cc: Len Brown Cc: Linus Torvalds Cc: Paul E. McKenney Cc: Peter Zijlstra Cc: Rik van Riel Cc: rt@linutronix.de Link: http://lkml.kernel.org/r/20160704094342.512039360@linutronix.de Signed-off-by: Ingo Molnar

timers: Optimize collect_expired_timers() for NOHZ

2016-07-07T08:35:10+00:00

After a NOHZ idle sleep the timer wheel must be forwarded to current jiffies. There might be expired timers so the current code loops and checks the expired buckets for timers. This can take quite some time for long NOHZ idle periods. The pending bitmask in the timer base allows us to do a quick search for the next expiring timer and therefore a fast forward of the base time which prevents pointless long lasting loops. For a 3 seconds idle sleep this reduces the catchup time from ~1ms to 5us. Signed-off-by: Anna-Maria Gleixner Signed-off-by: Thomas Gleixner Cc: Arjan van de Ven Cc: Chris Mason Cc: Eric Dumazet Cc: Frederic Weisbecker Cc: George Spelvin Cc: Josh Triplett Cc: Len Brown Cc: Linus Torvalds Cc: Paul E. McKenney Cc: Peter Zijlstra Cc: Rik van Riel Cc: rt@linutronix.de Link: http://lkml.kernel.org/r/20160704094342.351296290@linutronix.de Signed-off-by: Ingo Molnar

timers: Move __run_timers() function

2016-07-07T08:35:09+00:00

Move __run_timers() below __next_timer_interrupt() and next_pending_bucket() in preparation for __run_timers() NOHZ optimization. No functional change. Signed-off-by: Anna-Maria Gleixner Signed-off-by: Thomas Gleixner Cc: Arjan van de Ven Cc: Chris Mason Cc: Eric Dumazet Cc: Frederic Weisbecker Cc: George Spelvin Cc: Josh Triplett Cc: Len Brown Cc: Linus Torvalds Cc: Paul E. McKenney Cc: Peter Zijlstra Cc: Rik van Riel Cc: rt@linutronix.de Link: http://lkml.kernel.org/r/20160704094342.271872665@linutronix.de Signed-off-by: Ingo Molnar

timers: Remove set_timer_slack() leftovers

2016-07-07T08:35:09+00:00

We now have implicit batching in the timer wheel. The slack API is no longer used, so remove it. Signed-off-by: Thomas Gleixner Cc: Alan Stern Cc: Andrew F. Davis Cc: Arjan van de Ven Cc: Chris Mason Cc: David S. Miller Cc: David Woodhouse Cc: Dmitry Eremin-Solenikov Cc: Eric Dumazet Cc: Frederic Weisbecker Cc: George Spelvin Cc: Greg Kroah-Hartman Cc: Jaehoon Chung Cc: Jens Axboe Cc: John Stultz Cc: Josh Triplett Cc: Len Brown Cc: Linus Torvalds Cc: Mathias Nyman Cc: Pali Rohár Cc: Paul E. McKenney Cc: Peter Zijlstra Cc: Rik van Riel Cc: Sebastian Reichel Cc: Ulf Hansson Cc: linux-block@vger.kernel.org Cc: linux-kernel@vger.kernel.org Cc: linux-mmc@vger.kernel.org Cc: linux-pm@vger.kernel.org Cc: linux-usb@vger.kernel.org Cc: netdev@vger.kernel.org Cc: rt@linutronix.de Link: http://lkml.kernel.org/r/20160704094342.189813118@linutronix.de Signed-off-by: Ingo Molnar

timers: Switch to a non-cascading wheel

2016-07-07T08:35:09+00:00

The current timer wheel has some drawbacks: 1) Cascading: Cascading can be an unbound operation and is completely pointless in most cases because the vast majority of the timer wheel timers are canceled or rearmed before expiration. (They are used as timeout safeguards, not as real timers to measure time.) 2) No fast lookup of the next expiring timer: In NOHZ scenarios the first timer soft interrupt after a long NOHZ period must fast forward the base time to the current value of jiffies. As we have no way to find the next expiring timer fast, the code loops linearly and increments the base time one by one and checks for expired timers in each step. This causes unbound overhead spikes exactly in the moment when we should wake up as fast as possible. After a thorough analysis of real world data gathered on laptops, workstations, webservers and other machines (thanks Chris!) I came to the conclusion that the current 'classic' timer wheel implementation can be modified to address the above issues. The vast majority of timer wheel timers is canceled or rearmed before expiry. Most of them are timeouts for networking and other I/O tasks. The nature of timeouts is to catch the exception from normal operation (TCP ack timed out, disk does not respond, etc.). For these kinds of timeouts the accuracy of the timeout is not really a concern. Timeouts are very often approximate worst-case values and in case the timeout fires, we already waited for a long time and performance is down the drain already. The few timers which actually expire can be split into two categories: 1) Short expiry times which expect halfways accurate expiry 2) Long term expiry times are inaccurate today already due to the batching which is done for NOHZ automatically and also via the set_timer_slack() API. So for long term expiry timers we can avoid the cascading property and just leave them in the less granular outer wheels until expiry or cancelation. Timers which are armed with a timeout larger than the wheel capacity are no longer cascaded. We expire them with the longest possible timeout (6+ days). We have not observed such timeouts in our data collection, but at least we handle them, applying the rule of the least surprise. To avoid extending the wheel levels for HZ=1000 so we can accomodate the longest observed timeouts (5 days in the network conntrack code) we reduce the first level granularity on HZ=1000 to 4ms, which effectively is the same as the HZ=250 behaviour. From our data analysis there is nothing which relies on that 1ms granularity and as a side effect we get better batching and timer locality for the networking code as well. Contrary to the classic wheel the granularity of the next wheel is not the capacity of the first wheel. The granularities of the wheels are in the currently chosen setting 8 times the granularity of the previous wheel. So for HZ=250 we end up with the following granularity levels: Level Offset Granularity Range 0 0 4 ms 0 ms - 252 ms 1 64 32 ms 256 ms - 2044 ms (256ms - ~2s) 2 128 256 ms 2048 ms - 16380 ms (~2s - ~16s) 3 192 2048 ms (~2s) 16384 ms - 131068 ms (~16s - ~2m) 4 256 16384 ms (~16s) 131072 ms - 1048572 ms (~2m - ~17m) 5 320 131072 ms (~2m) 1048576 ms - 8388604 ms (~17m - ~2h) 6 384 1048576 ms (~17m) 8388608 ms - 67108863 ms (~2h - ~18h) 7 448 8388608 ms (~2h) 67108864 ms - 536870911 ms (~18h - ~6d) That's a worst case inaccuracy of 12.5% for the timers which are queued at the beginning of a level. So the new wheel concept addresses the old issues: 1) Cascading is avoided completely 2) By keeping the timers in the bucket until expiry/cancelation we can track the buckets which have timers enqueued in a bucket bitmap and therefore can look up the next expiring timer very fast and O(1). A further benefit of the concept is that the slack calculation which is done on every timer start is no longer necessary because the granularity levels provide natural batching already. Our extensive testing with various loads did not show any performance degradation vs. the current wheel implementation. This patch does not address the 'fast lookup' issue as we wanted to make sure that there is no regression introduced by the wheel redesign. The optimizations are in follow up patches. This patch contains fixes from Anna-Maria Gleixner and Richard Cochran. Signed-off-by: Thomas Gleixner Cc: Arjan van de Ven Cc: Chris Mason Cc: Eric Dumazet Cc: Frederic Weisbecker Cc: George Spelvin Cc: Josh Triplett Cc: Len Brown Cc: Linus Torvalds Cc: Paul E. McKenney Cc: Peter Zijlstra Cc: Rik van Riel Cc: rt@linutronix.de Link: http://lkml.kernel.org/r/20160704094342.108621834@linutronix.de Signed-off-by: Ingo Molnar