16.02.2015, 14:52, "Kirill Tkhai" <tk...@yandex.ru>:
> 16.02.2015, 10:21, "Fengguang Wu" <fengguang...@intel.com>:
>> Hi Peter,
>>
>> We got the below dmesg and the first bad commit is
>>
>> git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
>> revert-d347efeb16d3d5150cb7f8d50b05f388b572840e-1019a359d3dc4b64d0e1e5a5efcb725d5e83994d
>>
>> commit 1019a359d3dc4b64d0e1e5a5efcb725d5e83994d
>> Author: Peter Zijlstra <pet...@infradead.org>
>> AuthorDate: Wed Nov 26 08:44:03 2014 +0800
>> Commit: Ingo Molnar <mi...@kernel.org>
>> CommitDate: Wed Feb 4 07:52:26 2015 +0100
>>
>> sched/deadline: Fix stale yield state
>>
>> When we fail to start the deadline timer in update_curr_dl(), we
>> forget to clear ->dl_yielded, resulting in wrecked time keeping.
>>
>> Since the natural place to clear both ->dl_yielded and ->dl_throttled
>> is in replenish_dl_entity(); both are after all waiting for that event;
>> make it so.
>>
>> Luckily since 67dfa1b756f2 ("sched/deadline: Implement
>> cancel_dl_timer() to use in switched_from_dl()") the
>> task_on_rq_queued() condition in dl_task_timer() must be true, and can
>> therefore call enqueue_task_dl() unconditionally.
>>
>> Reported-by: Wanpeng Li <wanpeng...@linux.intel.com>
>> Signed-off-by: Peter Zijlstra (Intel) <pet...@infradead.org>
>> Cc: Kirill Tkhai <ktk...@parallels.com>
>> Cc: Juri Lelli <juri.le...@arm.com>
>> Cc: Linus Torvalds <torva...@linux-foundation.org>
>> Link:
>> http://lkml.kernel.org/r/1416962647-76792-4-git-send-email-wanpeng...@linux.intel.com
>> Signed-off-by: Ingo Molnar <mi...@kernel.org>
>>
>> ===================================================
>> PARENT COMMIT NOT CLEAN. LOOK OUT FOR WRONG BISECT!
>> ===================================================
>>
>>
>> +------------------------------------------------+------------+------------+------------+
>> | | a7bebf4887 | 1019a359d3
>> | 58e0e18cb8 |
>>
>> +------------------------------------------------+------------+------------+------------+
>> | boot_successes | 886 | 54
>> | 10 |
>> | boot_failures | 314 | 246
>> | 4 |
>> | BUG:kernel_boot_crashed | 48 | 20
>> | 1 |
>> | BUG:kernel_test_crashed | 16 | 2
>> | |
>> | BUG:kernel_boot_hang | 213 | 43
>> | |
>> | Kernel_panic-not_syncing:softlockup:hung_tasks | 37 | 3
>> | |
>> | backtrace:ring_buffer_consumer_thread | 17 | 3
>> | |
>> | EIP_is_at_get_node_by_name | 4 |
>> | |
>> | backtrace:gcov_enable_events | 17 |
>> | |
>> | backtrace:gcov_fs_init | 17 |
>> | |
>> | backtrace:kernel_init_freeable | 20 | 178
>> | 3 |
>> | EIP_is_at_strcmp | 7 |
>> | |
>> | EIP_is_at__slab_alloc | 1 |
>> | |
>> | backtrace:__ring_buffer_alloc | 1 |
>> | |
>> | backtrace:ring_buffer_benchmark_init | 1 |
>> | |
>> | EIP_is_at_lock_release | 1 |
>> | |
>> | backtrace:debug_mutex_init | 1 |
>> | |
>> | EIP_is_at_strncmp | 1 |
>> | |
>> | EIP_is_at_do_raw_spin_unlock | 1 |
>> | |
>> | EIP_is_at__do_softirq | 6 | 2
>> | |
>> | EIP_is_at_gcov_info_filename | 1 |
>> | |
>> | EIP_is_at_ring_buffer_consume | 8 | 1
>> | |
>> | EIP_is_at_raw_spin_unlock_irqrestore | 1 |
>> | |
>> | EIP_is_at_ftrace_ops_list_func | 1 |
>> | |
>> | backtrace:event_trace_self_tests | 2 |
>> | |
>> | backtrace:event_trace_self_tests_init | 2 |
>> | |
>> | EIP_is_at_ftrace_likely_update | 1 |
>> | |
>> | EIP_is_at_rb_reserve_next_event | 1 |
>> | |
>> | EIP_is_at_raw_spin_unlock_irq | 1 |
>> | |
>> | kernel_BUG_at_kernel/sched/deadline.c | 0 | 178
>> | 3 |
>> | invalid_opcode | 0 | 178
>> | 3 |
>> | EIP_is_at_enqueue_task_dl | 0 | 178
>> | 3 |
>> | Kernel_panic-not_syncing:Fatal_exception | 0 | 178
>> | 3 |
>> | backtrace:register_tracer | 0 | 178
>> | 3 |
>> | backtrace:init_wakeup_tracer | 0 | 178
>> | 3 |
>>
>> +------------------------------------------------+------------+------------+------------+
>>
>> [ 6.432010] Testing ftrace regs: PASSED
>> [ 6.530393] Testing tracer wakeup:
>> [ 7.648615] ------------[ cut here ]------------
>> [ 7.649000] kernel BUG at kernel/sched/deadline.c:805!
>> [ 7.649000] invalid opcode: 0000 [#1] SMP
>> [ 7.649000] CPU: 0 PID: 1 Comm: swapper/0 Not tainted
>> 3.19.0-rc6-00041-g1019a35 #17
>> [ 7.649000] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS
>> 1.7.5-20140531_083030-gandalf 04/01/2014
>> [ 7.649000] task: d3348000 ti: d3344000 task.ti: d3344000
>> [ 7.649000] EIP: 0060:[<c11969a1>] EFLAGS: 00010046 CPU: 0
>> [ 7.649000] EIP is at enqueue_task_dl+0x341/0x1860
>> [ 7.649000] EAX: d2940101 EBX: 00000001 ECX: 00000000 EDX: 0085a491
>> [ 7.649000] ESI: d2940000 EDI: d2940101 EBP: d3345e08 ESP: d3345dc0
>> [ 7.649000] DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068
>> [ 7.649000] CR0: 8005003b CR2: ffffffff CR3: 04e8b000 CR4: 000006d0
>> [ 7.649000] Stack:
>> [ 7.649000] c1170353 00000002 d3345de0 c11ab53c d391ec10 00000046
>> d33485c0 d3348000
>> [ 7.649000] 00000002 d29401b8 d391ec00 d29401b8 00000046 00000005
>> 00000000 d2940000
>> [ 7.649000] d391ec00 c36efc80 d3345e20 c115991a 00000005 00000000
>> d391ec00 00000000
>> [ 7.649000] Call Trace:
>> [ 7.649000] [<c1170353>] ? sched_clock_cpu+0x293/0x2e0
>> [ 7.649000] [<c11ab53c>] ? __lock_is_held+0x6c/0xa0
>> [ 7.649000] [<c115991a>] enqueue_task+0xba/0x220
>> [ 7.649000] [<c115aa4c>] activate_task+0x4c/0x80
>> [ 7.649000] [<c115aaa1>] ttwu_activate+0x21/0x90
>> [ 7.649000] [<c115bd84>] ttwu_do_activate+0x64/0x90
>> [ 7.649000] [<c11689b6>] try_to_wake_up+0x366/0x8b0
>> [ 7.649000] [<c1169000>] wake_up_process+0xa0/0xe0
>> [ 7.649000] [<c114095b>] kthread_stop+0x21b/0xd90
>> [ 7.649000] [<c12c0b27>] trace_selftest_startup_wakeup+0x3b7/0x4b0
>> [ 7.649000] [<c36b1690>] ? wait_for_completion+0x1f0/0x290
>> [ 7.649000] [<c12c1d04>] register_tracer+0x2d4/0x700
>> [ 7.649000] [<c4bd11a4>] init_wakeup_tracer+0x1d/0xea
>> [ 7.649000] [<c4b8f12e>] do_one_initcall+0x278/0x44c
>> [ 7.649000] [<c4bd1187>] ? init_function_trace+0x1f8/0x1f8
>> [ 7.649000] [<c4b8f565>] kernel_init_freeable+0x263/0x424
>> [ 7.649000] [<c3663b4c>] kernel_init+0x1c/0x2a0
>> [ 7.649000] [<c36c2181>] ret_from_kernel_thread+0x21/0x30
>> [ 7.649000] [<c3663b30>] ? rest_init+0x330/0x330
>> [ 7.649000] Code: da e8 94 96 13 00 8b 04 9d 84 f6 77 c4 83 c0 01 89 04
>> 9d 84 f6 77 c4 89 f8 84 c0 74 4d 83 05 c8 14 f2 c4 01 83 15 cc 14 f2 c4 00
>> <0f> 0b 83 05 d0 14 f2 c4 01 83 15 d4 14 f2 c4 00 8d b4 26 00 00
>> [ 7.649000] EIP: [<c11969a1>] enqueue_task_dl+0x341/0x1860 SS:ESP
>> 0068:d3345dc0
>> [ 7.649000] ---[ end trace cb4bdebe33135894 ]---
>> [ 7.649000] Kernel panic - not syncing: Fatal exception
>>
>> git bisect start 58e0e18cb8b1c3351cb70b3cf78d4e3785a181fe
>> bfa76d49576599a4b9f9b7a71f23d73d6dcff735 --
>> git bisect bad 30496a920e97b97e16d078c520eef323344ef997 # 23:32 0-
>> 67 Merge 'amirv/for-upstream' into devel-xian-smoke-201502152116
>> git bisect good d170f4905ee6f93eb12f87d018dd4b9acea18f2c # 00:27 300+
>> 120 0day base guard for 'devel-xian-smoke-201502152116'
>> git bisect bad c5ce28df0e7c01a1de23c36ebdefcd803f2b6cbb # 01:11 0-
>> 72 Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next
>> git bisect good 57ee062e4baa9532e5a3835a2d185590a3fc8bd3 # 01:51 300+
>> 94 Merge branch 'r8152'
>> git bisect bad c08f8467939e7d2eebcba7cf2330242c4f53f2f7 # 02:10 0-
>> 103 Merge tag 'pci-v3.20-changes' of
>> git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci
>> git bisect bad 5b9b28a63f2e47dac5ff3a2503bfe3ade8796aa0 # 02:24 0-
>> 5 Merge branch 'sched-core-for-linus' of
>> git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
>> git bisect good 30d46827c2744f56bb31460007f2d16455f10720 # 02:41 300+
>> 107 Merge tag 'regulator-v3.20' of
>> git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator
>> git bisect good 8308756f45a12e2ff4f7749c2694fc83cdef0be9 # 03:11 300+
>> 101 Merge branch 'locking-core-for-linus' of
>> git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
>> git bisect good 4ac30cf74b308fb01338e660d3471cd490a7958a # 03:36 300+
>> 25 perf tools: Do not use __perf_session__process_events() directly
>> git bisect bad 9659e1eeee28f7025b6545934d644d19e9c6e603 # 03:48 0-
>> 2 sched/deadline: Remove cpu_active_mask from cpudl_find()
>> git bisect good 3847b272248a3a4ed70d20392cc0454917f7713b # 04:23 300+
>> 89 Merge branch 'sched/urgent' into sched/core
>> git bisect good 4c195c8a1967ff8bee13a811518a99db04618ab7 # 04:59 300+
>> 118 Merge branch 'sched/urgent' into sched/core, to merge fixes before
>> applying new patches
>> git bisect bad 1019a359d3dc4b64d0e1e5a5efcb725d5e83994d # 05:10 39-
>> 193 sched/deadline: Fix stale yield state
>> git bisect good a7bebf488791aa1036f3e6629daf01d01f705dcb # 05:34 300+
>> 103 sched/deadline: Fix hrtick for a non-leftmost task
>> # first bad commit: [1019a359d3dc4b64d0e1e5a5efcb725d5e83994d]
>> sched/deadline: Fix stale yield state
>> git bisect good a7bebf488791aa1036f3e6629daf01d01f705dcb # 05:55 900+
>> 314 sched/deadline: Fix hrtick for a non-leftmost task
>> # extra tests with DEBUG_INFO
>> git bisect good 1019a359d3dc4b64d0e1e5a5efcb725d5e83994d # 06:11 900+
>> 0 sched/deadline: Fix stale yield state
>> # extra tests on HEAD of linux-devel/devel-xian-smoke-201502152116
>> git bisect bad 58e0e18cb8b1c3351cb70b3cf78d4e3785a181fe # 06:11 0-
>> 4 0day head guard for 'devel-xian-smoke-201502152116'
>> # extra tests on tree/branch linus/master
>> git bisect bad a9724125ad014decf008d782e60447c811391326 # 06:11 0-
>> 1 Merge tag 'tty-3.20-rc1' of
>> git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty
>> # extra tests with first bad commit reverted
>> # extra tests on tree/branch linus/master
>> git bisect bad a9724125ad014decf008d782e60447c811391326 # 06:16 0-
>> 1 Merge tag 'tty-3.20-rc1' of
>> git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty
>> # extra tests on tree/branch next/master
>> git bisect bad b8acf73194186a5cba86812eb4ba17b897f0e13e # 06:26 0-
>> 35 Add linux-next specific files for 20150213
>>
>> This script may reproduce the error.
>>
>> ----------------------------------------------------------------------------
>> #!/bin/bash
>>
>> kernel=$1
>>
>> kvm=(
>> qemu-system-x86_64
>> -cpu kvm64
>> -enable-kvm
>> -kernel $kernel
>> -m 320
>> -smp 1
>> -net nic,vlan=1,model=e1000
>> -net user,vlan=1
>> -boot order=nc
>> -no-reboot
>> -watchdog i6300esb
>> -rtc base=localtime
>> -serial stdio
>> -display none
>> -monitor null
>> )
>>
>> append=(
>> hung_task_panic=1
>> earlyprintk=ttyS0,115200
>> rd.udev.log-priority=err
>> systemd.log_target=journal
>> systemd.log_level=warning
>> debug
>> apic=debug
>> sysrq_always_enabled
>> rcupdate.rcu_cpu_stall_timeout=100
>> panic=-1
>> softlockup_panic=1
>> nmi_watchdog=panic
>> oops=panic
>> load_ramdisk=2
>> prompt_ramdisk=0
>> console=ttyS0,115200
>> console=tty0
>> vga=normal
>> root=/dev/ram0
>> rw
>> drbd.minor_count=8
>> )
>>
>> "${kvm[@]}" --append "${append[*]}"
>> ----------------------------------------------------------------------------
>
> Don't we have the following problem?
>
> current->state = TASK_INTERRUPTIBLE;
> schedule()
> deactivate_task()
> dequeue_task_dl()
> update_curr_dl()
> start_dl_timer()
> __dequeue_task_dl()
>
> ...timer fires...
>
> dl_task_timer()
> enqueue_task_dl() /* queues on dl_rq */
>
> ...try_to_wake_up()
>
> enqueue_dl_entity()
> BUG_ON(on_dl_rq())
>
> Compile-tested fix:
>
> kernel/sched/deadline.c | 4 ++++
> 1 file changed, 4 insertions(+)
> diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c
> index a027799..1b78598 100644
> --- a/kernel/sched/deadline.c
> +++ b/kernel/sched/deadline.c
> @@ -541,6 +541,10 @@ static enum hrtimer_restart dl_task_timer(struct hrtimer
> *timer)
>
> sched_clock_tick();
> update_rq_clock(rq);
> + if (!p->on_rq) {
> + replenish_dl_entity(dl_se, dl_se);
> + goto unlock;
> + }
> enqueue_task_dl(rq, p, ENQUEUE_REPLENISH);
> if (dl_task(rq->curr))
> check_preempt_curr_dl(rq, p, 0);
We shouldn't enqueue migrating tasks. Please, try this one instead ;)
[PATCH] sched/dl: Prevent enqueue of a sleeping task in dl_task_timer()
A deadline task may be throttled and dequeued at the same time.
This happens, when it becomes throttled in schedule(), which
is called to go to sleep:
current->state = TASK_INTERRUPTIBLE;
schedule()
deactivate_task()
dequeue_task_dl()
update_curr_dl()
start_dl_timer()
__dequeue_task_dl()
prev->on_rq = 0;
Later the timer fires, but the task is still dequeued:
dl_task_timer()
enqueue_task_dl() /* queues on dl_rq; on_rq remains 0 */
Someone wakes it up:
try_to_wake_up()
enqueue_dl_entity()
BUG_ON(on_dl_rq())
Patch fixes this problem, it prevents queueing !on_rq tasks
on dl_rq.
Signed-off-by: Kirill Tkhai <ktk...@parallels.com>
diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c
index a027799..784ba68 100644
--- a/kernel/sched/deadline.c
+++ b/kernel/sched/deadline.c
@@ -541,6 +541,10 @@ static enum hrtimer_restart dl_task_timer(struct hrtimer
*timer)
sched_clock_tick();
update_rq_clock(rq);
+ if (!task_on_rq_queued(p)) {
+ replenish_dl_entity(dl_se, dl_se);
+ goto unlock;
+ }
enqueue_task_dl(rq, p, ENQUEUE_REPLENISH);
if (dl_task(rq->curr))
check_preempt_curr_dl(rq, p, 0);
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
