user/sven/linux.git/kernel/time, branch v4.10.16

Merge branch 'timers-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

2017-02-19T01:30:36Z

Pull timer fixes from Thomas Gleixner: "Two small fixes:: - Prevent deadlock on the tick broadcast lock. Found and fixed by Mike. - Stop using printk() in the timekeeping debug code to prevent a deadlock against the scheduler" * 'timers-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: timekeeping: Use deferred printk() in debug code tick/broadcast: Prevent deadlock on tick_broadcast_lock

Revert "nohz: Fix collision between tick and other hrtimers"

2017-02-16T20:19:18Z

This reverts commit 24b91e360ef521a2808771633d76ebc68bd5604b and commit 7bdb59f1ad47 ("tick/nohz: Fix possible missing clock reprog after tick soft restart") that depends on it, Pavel reports that it causes occasional boot hangs for him that seem to depend on just how the machine was booted. In particular, his machine hangs at around the PCI fixups of the EHCI USB host controller, but only hangs from cold boot, not from a warm boot. Thomas Gleixner suspecs it's a CPU hotplug interaction, particularly since Pavel also saw suspend/resume issues that seem to be related. We're reverting for now while trying to figure out the root cause. Reported-bisected-and-tested-by: Pavel Machek Acked-by: Frederic Weisbecker Cc: Wanpeng Li Cc: Peter Zijlstra Cc: Rik van Riel Cc: Thomas Gleixner Cc: stable@kernel.org # reverted commits were marked for stable Signed-off-by: Linus Torvalds

timekeeping: Use deferred printk() in debug code

2017-02-15T10:46:08Z

We cannot do printk() from tk_debug_account_sleep_time(), because tk_debug_account_sleep_time() is called under tk_core seq lock. The reason why printk() is unsafe there is that console_sem may invoke scheduler (up()->wake_up_process()->activate_task()), which, in turn, can return back to timekeeping code, for instance, via get_time()->ktime_get(), deadlocking the system on tk_core seq lock. [ 48.950592] ====================================================== [ 48.950622] [ INFO: possible circular locking dependency detected ] [ 48.950622] 4.10.0-rc7-next-20170213+ #101 Not tainted [ 48.950622] ------------------------------------------------------- [ 48.950622] kworker/0:0/3 is trying to acquire lock: [ 48.950653] (tk_core){----..}, at: [] retrigger_next_event+0x4c/0x90 [ 48.950683] but task is already holding lock: [ 48.950683] (hrtimer_bases.lock){-.-...}, at: [] retrigger_next_event+0x38/0x90 [ 48.950714] which lock already depends on the new lock. [ 48.950714] the existing dependency chain (in reverse order) is: [ 48.950714] -> #5 (hrtimer_bases.lock){-.-...}: [ 48.950744] _raw_spin_lock_irqsave+0x50/0x64 [ 48.950775] lock_hrtimer_base+0x28/0x58 [ 48.950775] hrtimer_start_range_ns+0x20/0x5c8 [ 48.950775] __enqueue_rt_entity+0x320/0x360 [ 48.950805] enqueue_rt_entity+0x2c/0x44 [ 48.950805] enqueue_task_rt+0x24/0x94 [ 48.950836] ttwu_do_activate+0x54/0xc0 [ 48.950836] try_to_wake_up+0x248/0x5c8 [ 48.950836] __setup_irq+0x420/0x5f0 [ 48.950836] request_threaded_irq+0xdc/0x184 [ 48.950866] devm_request_threaded_irq+0x58/0xa4 [ 48.950866] omap_i2c_probe+0x530/0x6a0 [ 48.950897] platform_drv_probe+0x50/0xb0 [ 48.950897] driver_probe_device+0x1f8/0x2cc [ 48.950897] __driver_attach+0xc0/0xc4 [ 48.950927] bus_for_each_dev+0x6c/0xa0 [ 48.950927] bus_add_driver+0x100/0x210 [ 48.950927] driver_register+0x78/0xf4 [ 48.950958] do_one_initcall+0x3c/0x16c [ 48.950958] kernel_init_freeable+0x20c/0x2d8 [ 48.950958] kernel_init+0x8/0x110 [ 48.950988] ret_from_fork+0x14/0x24 [ 48.950988] -> #4 (&rt_b->rt_runtime_lock){-.-...}: [ 48.951019] _raw_spin_lock+0x40/0x50 [ 48.951019] rq_offline_rt+0x9c/0x2bc [ 48.951019] set_rq_offline.part.2+0x2c/0x58 [ 48.951049] rq_attach_root+0x134/0x144 [ 48.951049] cpu_attach_domain+0x18c/0x6f4 [ 48.951049] build_sched_domains+0xba4/0xd80 [ 48.951080] sched_init_smp+0x68/0x10c [ 48.951080] kernel_init_freeable+0x160/0x2d8 [ 48.951080] kernel_init+0x8/0x110 [ 48.951080] ret_from_fork+0x14/0x24 [ 48.951110] -> #3 (&rq->lock){-.-.-.}: [ 48.951110] _raw_spin_lock+0x40/0x50 [ 48.951141] task_fork_fair+0x30/0x124 [ 48.951141] sched_fork+0x194/0x2e0 [ 48.951141] copy_process.part.5+0x448/0x1a20 [ 48.951171] _do_fork+0x98/0x7e8 [ 48.951171] kernel_thread+0x2c/0x34 [ 48.951171] rest_init+0x1c/0x18c [ 48.951202] start_kernel+0x35c/0x3d4 [ 48.951202] 0x8000807c [ 48.951202] -> #2 (&p->pi_lock){-.-.-.}: [ 48.951232] _raw_spin_lock_irqsave+0x50/0x64 [ 48.951232] try_to_wake_up+0x30/0x5c8 [ 48.951232] up+0x4c/0x60 [ 48.951263] __up_console_sem+0x2c/0x58 [ 48.951263] console_unlock+0x3b4/0x650 [ 48.951263] vprintk_emit+0x270/0x474 [ 48.951293] vprintk_default+0x20/0x28 [ 48.951293] printk+0x20/0x30 [ 48.951324] kauditd_hold_skb+0x94/0xb8 [ 48.951324] kauditd_thread+0x1a4/0x56c [ 48.951324] kthread+0x104/0x148 [ 48.951354] ret_from_fork+0x14/0x24 [ 48.951354] -> #1 ((console_sem).lock){-.....}: [ 48.951385] _raw_spin_lock_irqsave+0x50/0x64 [ 48.951385] down_trylock+0xc/0x2c [ 48.951385] __down_trylock_console_sem+0x24/0x80 [ 48.951385] console_trylock+0x10/0x8c [ 48.951416] vprintk_emit+0x264/0x474 [ 48.951416] vprintk_default+0x20/0x28 [ 48.951416] printk+0x20/0x30 [ 48.951446] tk_debug_account_sleep_time+0x5c/0x70 [ 48.951446] __timekeeping_inject_sleeptime.constprop.3+0x170/0x1a0 [ 48.951446] timekeeping_resume+0x218/0x23c [ 48.951477] syscore_resume+0x94/0x42c [ 48.951477] suspend_enter+0x554/0x9b4 [ 48.951477] suspend_devices_and_enter+0xd8/0x4b4 [ 48.951507] enter_state+0x934/0xbd4 [ 48.951507] pm_suspend+0x14/0x70 [ 48.951507] state_store+0x68/0xc8 [ 48.951538] kernfs_fop_write+0xf4/0x1f8 [ 48.951538] __vfs_write+0x1c/0x114 [ 48.951538] vfs_write+0xa0/0x168 [ 48.951568] SyS_write+0x3c/0x90 [ 48.951568] __sys_trace_return+0x0/0x10 [ 48.951568] -> #0 (tk_core){----..}: [ 48.951599] lock_acquire+0xe0/0x294 [ 48.951599] ktime_get_update_offsets_now+0x5c/0x1d4 [ 48.951629] retrigger_next_event+0x4c/0x90 [ 48.951629] on_each_cpu+0x40/0x7c [ 48.951629] clock_was_set_work+0x14/0x20 [ 48.951660] process_one_work+0x2b4/0x808 [ 48.951660] worker_thread+0x3c/0x550 [ 48.951660] kthread+0x104/0x148 [ 48.951690] ret_from_fork+0x14/0x24 [ 48.951690] other info that might help us debug this: [ 48.951690] Chain exists of: tk_core --> &rt_b->rt_runtime_lock --> hrtimer_bases.lock [ 48.951721] Possible unsafe locking scenario: [ 48.951721] CPU0 CPU1 [ 48.951721] ---- ---- [ 48.951721] lock(hrtimer_bases.lock); [ 48.951751] lock(&rt_b->rt_runtime_lock); [ 48.951751] lock(hrtimer_bases.lock); [ 48.951751] lock(tk_core); [ 48.951782] *** DEADLOCK *** [ 48.951782] 3 locks held by kworker/0:0/3: [ 48.951782] #0: ("events"){.+.+.+}, at: [] process_one_work+0x1f8/0x808 [ 48.951812] #1: (hrtimer_work){+.+...}, at: [] process_one_work+0x1f8/0x808 [ 48.951843] #2: (hrtimer_bases.lock){-.-...}, at: [] retrigger_next_event+0x38/0x90 [ 48.951843] stack backtrace: [ 48.951873] CPU: 0 PID: 3 Comm: kworker/0:0 Not tainted 4.10.0-rc7-next-20170213+ [ 48.951904] Workqueue: events clock_was_set_work [ 48.951904] [] (unwind_backtrace) from [] (show_stack+0x10/0x14) [ 48.951934] [] (show_stack) from [] (dump_stack+0xac/0xe0) [ 48.951934] [] (dump_stack) from [] (print_circular_bug+0x1d0/0x308) [ 48.951965] [] (print_circular_bug) from [] (validate_chain+0xf50/0x1324) [ 48.951965] [] (validate_chain) from [] (__lock_acquire+0x468/0x7e8) [ 48.951995] [] (__lock_acquire) from [] (lock_acquire+0xe0/0x294) [ 48.951995] [] (lock_acquire) from [] (ktime_get_update_offsets_now+0x5c/0x1d4) [ 48.952026] [] (ktime_get_update_offsets_now) from [] (retrigger_next_event+0x4c/0x90) [ 48.952026] [] (retrigger_next_event) from [] (on_each_cpu+0x40/0x7c) [ 48.952056] [] (on_each_cpu) from [] (clock_was_set_work+0x14/0x20) [ 48.952056] [] (clock_was_set_work) from [] (process_one_work+0x2b4/0x808) [ 48.952087] [] (process_one_work) from [] (worker_thread+0x3c/0x550) [ 48.952087] [] (worker_thread) from [] (kthread+0x104/0x148) [ 48.952087] [] (kthread) from [] (ret_from_fork+0x14/0x24) Replace printk() with printk_deferred(), which does not call into the scheduler. Fixes: 0bf43f15db85 ("timekeeping: Prints the amounts of time spent during suspend") Reported-and-tested-by: Tony Lindgren Signed-off-by: Sergey Senozhatsky Cc: Petr Mladek Cc: Sergey Senozhatsky Cc: Peter Zijlstra Cc: "Rafael J . Wysocki" Cc: Steven Rostedt Cc: John Stultz Cc: "[4.9+]" Link: http://lkml.kernel.org/r/20170215044332.30449-1-sergey.senozhatsky@gmail.com Signed-off-by: Thomas Gleixner

tick/broadcast: Prevent deadlock on tick_broadcast_lock

2017-02-13T08:49:31Z

tick_broadcast_lock is taken from interrupt context, but the following call chain takes the lock without disabling interrupts: [ 12.703736] _raw_spin_lock+0x3b/0x50 [ 12.703738] tick_broadcast_control+0x5a/0x1a0 [ 12.703742] intel_idle_cpu_online+0x22/0x100 [ 12.703744] cpuhp_invoke_callback+0x245/0x9d0 [ 12.703752] cpuhp_thread_fun+0x52/0x110 [ 12.703754] smpboot_thread_fn+0x276/0x320 So the following deadlock can happen: lock(tick_broadcast_lock); lock(tick_broadcast_lock); intel_idle_cpu_online() is the only place which violates the calling convention of tick_broadcast_control(). This was caused by the removal of the smp function call in course of the cpu hotplug rework. Instead of slapping local_irq_disable/enable() at the call site, we can relax the calling convention and handle it in the core code, which makes the whole machinery more robust. Fixes: 29d7bbada98e ("intel_idle: Remove superfluous SMP fuction call") Reported-by: Gabriel C Signed-off-by: Mike Galbraith Cc: Ruslan Ruslichenko Cc: Jiri Slaby Cc: Greg KH Cc: Borislav Petkov Cc: lwn@lwn.net Cc: Andrew Morton Cc: Linus Torvalds Cc: Anna-Maria Gleixner Cc: Sebastian Siewior Cc: stable Link: http://lkml.kernel.org/r/1486953115.5912.4.camel@gmx.de Signed-off-by: Thomas Gleixner

tick/nohz: Fix possible missing clock reprog after tick soft restart

2017-02-10T08:43:48Z

ts->next_tick keeps track of the next tick deadline in order to optimize clock programmation on irq exit and avoid redundant clock device writes. Now if ts->next_tick missed an update, we may spuriously miss a clock reprog later as the nohz code is fooled by an obsolete next_tick value. This is what happens here on a specific path: when we observe an expired timer from the nohz update code on irq exit, we perform a soft tick restart which simply fires the closest possible tick without actually exiting the nohz mode and restoring a periodic state. But we forget to update ts->next_tick accordingly. As a result, after the next tick resulting from such soft tick restart, the nohz code sees a stale value on ts->next_tick which doesn't match the clock deadline that just expired. If that obsolete ts->next_tick value happens to collide with the actual next tick deadline to be scheduled, we may spuriously bypass the clock reprogramming. In the worst case, the tick may never fire again. Fix this with a ts->next_tick reset on soft tick restart. Signed-off-by: Frederic Weisbecker Reviewed: Wanpeng Li Acked-by: Rik van Riel Cc: Peter Zijlstra Cc: stable@vger.kernel.org Link: http://lkml.kernel.org/r/1486485894-29173-1-git-send-email-fweisbec@gmail.com Signed-off-by: Thomas Gleixner

nohz: Fix collision between tick and other hrtimers

2017-01-11T09:41:33Z

When the tick is stopped and an interrupt occurs afterward, we check on that interrupt exit if the next tick needs to be rescheduled. If it doesn't need any update, we don't want to do anything. In order to check if the tick needs an update, we compare it against the clockevent device deadline. Now that's a problem because the clockevent device is at a lower level than the tick itself if it is implemented on top of hrtimer. Every hrtimer share this clockevent device. So comparing the next tick deadline against the clockevent device deadline is wrong because the device may be programmed for another hrtimer whose deadline collides with the tick. As a result we may end up not reprogramming the tick accidentally. In a worst case scenario under full dynticks mode, the tick stops firing as it is supposed to every 1hz, leaving /proc/stat stalled: Task in a full dynticks CPU ---------------------------- * hrtimer A is queued 2 seconds ahead * the tick is stopped, scheduled 1 second ahead * tick fires 1 second later * on tick exit, nohz schedules the tick 1 second ahead but sees the clockevent device is already programmed to that deadline, fooled by hrtimer A, the tick isn't rescheduled. * hrtimer A is cancelled before its deadline * tick never fires again until an interrupt happens... In order to fix this, store the next tick deadline to the tick_sched local structure and reuse that value later to check whether we need to reprogram the clock after an interrupt. On the other hand, ts->sleep_length still wants to know about the next clock event and not just the tick, so we want to improve the related comment to avoid confusion. Reported-by: James Hartsock Signed-off-by: Frederic Weisbecker Reviewed-by: Wanpeng Li Acked-by: Peter Zijlstra Acked-by: Rik van Riel Link: http://lkml.kernel.org/r/1483539124-5693-1-git-send-email-fweisbec@gmail.com Cc: stable@vger.kernel.org Signed-off-by: Thomas Gleixner

ktime: Cleanup ktime_set() usage

2016-12-25T16:21:22Z

ktime_set(S,N) was required for the timespec storage type and is still useful for situations where a Seconds and Nanoseconds part of a time value needs to be converted. For anything where the Seconds argument is 0, this is pointless and can be replaced with a simple assignment. Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra

ktime: Get rid of the union

2016-12-25T16:21:22Z

ktime is a union because the initial implementation stored the time in scalar nanoseconds on 64 bit machine and in a endianess optimized timespec variant for 32bit machines. The Y2038 cleanup removed the timespec variant and switched everything to scalar nanoseconds. The union remained, but become completely pointless. Get rid of the union and just keep ktime_t as simple typedef of type s64. The conversion was done with coccinelle and some manual mopping up. Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra

clocksource: Use a plain u64 instead of cycle_t

2016-12-25T10:04:12Z

There is no point in having an extra type for extra confusion. u64 is unambiguous. Conversion was done with the following coccinelle script: @rem@ @@ -typedef u64 cycle_t; @fix@ typedef cycle_t; @@ -cycle_t +u64 Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: John Stultz

Replace with globally

2016-12-24T19:46:01Z

This was entirely automated, using the script by Al: PATT='^[[:blank:]]*#[[:blank:]]*include[[:blank:]]*' sed -i -e "s!$PATT!#include !" \ $(git grep -l "$PATT"|grep -v ^include/linux/uaccess.h) to do the replacement at the end of the merge window. Requested-by: Al Viro Signed-off-by: Linus Torvalds