user/sven/linux.git/kernel/kthread.c, branch v4.19.290

kthread: prevent deadlock when kthread_mod_delayed_work() races with kthread_cancel_delayed_work_sync()

2021-07-11T10:49:31Z

commit 5fa54346caf67b4b1b10b1f390316ae466da4d53 upstream. The system might hang with the following backtrace: schedule+0x80/0x100 schedule_timeout+0x48/0x138 wait_for_common+0xa4/0x134 wait_for_completion+0x1c/0x2c kthread_flush_work+0x114/0x1cc kthread_cancel_work_sync.llvm.16514401384283632983+0xe8/0x144 kthread_cancel_delayed_work_sync+0x18/0x2c xxxx_pm_notify+0xb0/0xd8 blocking_notifier_call_chain_robust+0x80/0x194 pm_notifier_call_chain_robust+0x28/0x4c suspend_prepare+0x40/0x260 enter_state+0x80/0x3f4 pm_suspend+0x60/0xdc state_store+0x108/0x144 kobj_attr_store+0x38/0x88 sysfs_kf_write+0x64/0xc0 kernfs_fop_write_iter+0x108/0x1d0 vfs_write+0x2f4/0x368 ksys_write+0x7c/0xec It is caused by the following race between kthread_mod_delayed_work() and kthread_cancel_delayed_work_sync(): CPU0 CPU1 Context: Thread A Context: Thread B kthread_mod_delayed_work() spin_lock() __kthread_cancel_work() spin_unlock() del_timer_sync() kthread_cancel_delayed_work_sync() spin_lock() __kthread_cancel_work() spin_unlock() del_timer_sync() spin_lock() work->canceling++ spin_unlock spin_lock() queue_delayed_work() // dwork is put into the worker->delayed_work_list spin_unlock() kthread_flush_work() // flush_work is put at the tail of the dwork wait_for_completion() Context: IRQ kthread_delayed_work_timer_fn() spin_lock() list_del_init(&work->node); spin_unlock() BANG: flush_work is not longer linked and will never get proceed. The problem is that kthread_mod_delayed_work() checks work->canceling flag before canceling the timer. A simple solution is to (re)check work->canceling after __kthread_cancel_work(). But then it is not clear what should be returned when __kthread_cancel_work() removed the work from the queue (list) and it can't queue it again with the new @delay. The return value might be used for reference counting. The caller has to know whether a new work has been queued or an existing one was replaced. The proper solution is that kthread_mod_delayed_work() will remove the work from the queue (list) _only_ when work->canceling is not set. The flag must be checked after the timer is stopped and the remaining operations can be done under worker->lock. Note that kthread_mod_delayed_work() could remove the timer and then bail out. It is fine. The other canceling caller needs to cancel the timer as well. The important thing is that the queue (list) manipulation is done atomically under worker->lock. Link: https://lkml.kernel.org/r/20210610133051.15337-3-pmladek@suse.com Fixes: 9a6b06c8d9a220860468a ("kthread: allow to modify delayed kthread work") Signed-off-by: Petr Mladek Reported-by: Martin Liu Cc: Cc: Minchan Kim Cc: Nathan Chancellor Cc: Nick Desaulniers Cc: Oleg Nesterov Cc: Tejun Heo Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds Signed-off-by: Greg Kroah-Hartman

kthread_worker: split code for canceling the delayed work timer

2021-07-11T10:49:31Z

commit 34b3d5344719d14fd2185b2d9459b3abcb8cf9d8 upstream. Patch series "kthread_worker: Fix race between kthread_mod_delayed_work() and kthread_cancel_delayed_work_sync()". This patchset fixes the race between kthread_mod_delayed_work() and kthread_cancel_delayed_work_sync() including proper return value handling. This patch (of 2): Simple code refactoring as a preparation step for fixing a race between kthread_mod_delayed_work() and kthread_cancel_delayed_work_sync(). It does not modify the existing behavior. Link: https://lkml.kernel.org/r/20210610133051.15337-2-pmladek@suse.com Signed-off-by: Petr Mladek Cc: Cc: Martin Liu Cc: Minchan Kim Cc: Nathan Chancellor Cc: Nick Desaulniers Cc: Oleg Nesterov Cc: Tejun Heo Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds Signed-off-by: Greg Kroah-Hartman

kthread: Extract KTHREAD_IS_PER_CPU

2021-02-07T13:48:38Z

[ Upstream commit ac687e6e8c26181a33270efd1a2e2241377924b0 ] There is a need to distinguish geniune per-cpu kthreads from kthreads that happen to have a single CPU affinity. Geniune per-cpu kthreads are kthreads that are CPU affine for correctness, these will obviously have PF_KTHREAD set, but must also have PF_NO_SETAFFINITY set, lest userspace modify their affinity and ruins things. However, these two things are not sufficient, PF_NO_SETAFFINITY is also set on other tasks that have their affinities controlled through other means, like for instance workqueues. Therefore another bit is needed; it turns out kthread_create_per_cpu() already has such a bit: KTHREAD_IS_PER_CPU, which is used to make kthread_park()/kthread_unpark() work correctly. Expose this flag and remove the implicit setting of it from kthread_create_on_cpu(); the io_uring usage of it seems dubious at best. Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Valentin Schneider Tested-by: Valentin Schneider Link: https://lkml.kernel.org/r/20210121103506.557620262@infradead.org Signed-off-by: Sasha Levin

kthread_worker: prevent queuing delayed work from timer_fn when it is being canceled

2020-11-10T11:35:58Z

commit 6993d0fdbee0eb38bfac350aa016f65ad11ed3b1 upstream. There is a small race window when a delayed work is being canceled and the work still might be queued from the timer_fn: CPU0 CPU1 kthread_cancel_delayed_work_sync() __kthread_cancel_work_sync() __kthread_cancel_work() work->canceling++; kthread_delayed_work_timer_fn() kthread_insert_work(); BUG: kthread_insert_work() should not get called when work->canceling is set. Signed-off-by: Zqiang Signed-off-by: Andrew Morton Reviewed-by: Petr Mladek Acked-by: Tejun Heo Cc: Link: https://lkml.kernel.org/r/20201014083030.16895-1-qiang.zhang@windriver.com Signed-off-by: Linus Torvalds Signed-off-by: Greg Kroah-Hartman

kthread: Do not preempt current task if it is going to call schedule()

2020-08-26T08:31:00Z

commit 26c7295be0c5e6da3fa45970e9748be983175b1b upstream. when we create a kthread with ktrhead_create_on_cpu(),the child thread entry is ktread.c:ktrhead() which will be preempted by the parent after call complete(done) while schedule() is not called yet,then the parent will call wait_task_inactive(child) but the child is still on the runqueue, so the parent will schedule_hrtimeout() for 1 jiffy,it will waste a lot of time,especially on startup. parent child ktrhead_create_on_cpu() wait_fo_completion(&done) -----> ktread.c:ktrhead() |----- complete(done);--wakeup and preempted by parent kthread_bind() <------------| |-> schedule();--dequeue here wait_task_inactive(child) | schedule_hrtimeout(1 jiffy) -| So we hope the child just wakeup parent but not preempted by parent, and the child is going to call schedule() soon,then the parent will not call schedule_hrtimeout(1 jiffy) as the child is already dequeue. The same issue for ktrhead_park()&&kthread_parkme(). This patch can save 120ms on rk312x startup with CONFIG_HZ=300. Signed-off-by: Liang Chen Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Steven Rostedt (VMware) Link: https://lkml.kernel.org/r/20200306070133.18335-2-cl@rock-chips.com Signed-off-by: Chanho Park Signed-off-by: Greg Kroah-Hartman

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

2018-08-13T18:25:07Z

Pull scheduler updates from Thomas Gleixner: - Cleanup and improvement of NUMA balancing - Refactoring and improvements to the PELT (Per Entity Load Tracking) code - Watchdog simplification and related cleanups - The usual pile of small incremental fixes and improvements * 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (41 commits) watchdog: Reduce message verbosity stop_machine: Reflow cpu_stop_queue_two_works() sched/numa: Move task_numa_placement() closer to numa_migrate_preferred() sched/numa: Use group_weights to identify if migration degrades locality sched/numa: Update the scan period without holding the numa_group lock sched/numa: Remove numa_has_capacity() sched/numa: Modify migrate_swap() to accept additional parameters sched/numa: Remove unused task_capacity from 'struct numa_stats' sched/numa: Skip nodes that are at 'hoplimit' sched/debug: Reverse the order of printing faults sched/numa: Use task faults only if numa_group is not yet set up sched/numa: Set preferred_node based on best_cpu sched/numa: Simplify load_too_imbalanced() sched/numa: Evaluate move once per node sched/numa: Remove redundant field sched/debug: Show the sum wait time of a task group sched/fair: Remove #ifdefs from scale_rt_capacity() sched/core: Remove get_cpu() from sched_fork() sched/cpufreq: Clarify sugov_get_util() sched/sysctl: Remove unused sched_time_avg_ms sysctl ...

kthread, tracing: Don't expose half-written comm when creating kthreads

2018-07-26T13:59:33Z

There is a window for racing when printing directly to task->comm, allowing other threads to see a non-terminated string. The vsnprintf function fills the buffer, counts the truncated chars, then finally writes the \0 at the end. creator other vsnprintf: fill (not terminated) count the rest trace_sched_waking(p): ... memcpy(comm, p->comm, TASK_COMM_LEN) write \0 The consequences depend on how 'other' uses the string. In our case, it was copied into the tracing system's saved cmdlines, a buffer of adjacent TASK_COMM_LEN-byte buffers (note the 'n' where 0 should be): crash-arm64> x/1024s savedcmd->saved_cmdlines | grep 'evenk' 0xffffffd5b3818640: "irq/497-pwr_evenkworker/u16:12" ...and a strcpy out of there would cause stack corruption: [224761.522292] Kernel panic - not syncing: stack-protector: Kernel stack is corrupted in: ffffff9bf9783c78 crash-arm64> kbt | grep 'comm\|trace_print_context' #6 0xffffff9bf9783c78 in trace_print_context+0x18c(+396) comm (char [16]) = "irq/497-pwr_even" crash-arm64> rd 0xffffffd4d0e17d14 8 ffffffd4d0e17d14: 2f71726900000000 5f7277702d373934 ....irq/497-pwr_ ffffffd4d0e17d24: 726f776b6e657665 3a3631752f72656b evenkworker/u16: ffffffd4d0e17d34: f9780248ff003231 cede60e0ffffff9b 12..H.x......`.. ffffffd4d0e17d44: cede60c8ffffffd4 00000fffffffffd4 .....`.......... The workaround in e09e28671 (use strlcpy in __trace_find_cmdline) was likely needed because of this same bug. Solved by vsnprintf:ing to a local buffer, then using set_task_comm(). This way, there won't be a window where comm is not terminated. Link: http://lkml.kernel.org/r/20180726071539.188015-1-snild@sony.com Cc: stable@vger.kernel.org Fixes: bc0c38d139ec7 ("ftrace: latency tracer infrastructure") Reviewed-by: Steven Rostedt (VMware) Signed-off-by: Snild Dolkow Signed-off-by: Steven Rostedt (VMware)

kthread: Simplify kthread_park() completion

2018-07-03T07:20:44Z

Oleg explains the reason we could hit park+park is that smpboot_update_cpumask_percpu_thread()'s for_each_cpu_and(cpu, &tmp, cpu_online_mask) smpboot_park_kthread(); turns into: for ((cpu) = 0; (cpu) < 1; (cpu)++, (void)mask, (void)and) smpboot_park_kthread(); on UP, ignoring the mask. But since we just completely removed that function, this is no longer relevant. So revert commit: b1f5b378e126 ("kthread: Allow kthread_park() on a parked kthread") Suggested-by: Oleg Nesterov Signed-off-by: Peter Zijlstra (Intel) Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar

kthread, sched/core: Fix kthread_parkme() (again...)

2018-07-03T07:17:30Z

Gaurav reports that commit: 85f1abe0019f ("kthread, sched/wait: Fix kthread_parkme() completion issue") isn't working for him. Because of the following race: > controller Thread CPUHP Thread > takedown_cpu > kthread_park > kthread_parkme > Set KTHREAD_SHOULD_PARK > smpboot_thread_fn > set Task interruptible > > > wake_up_process > if (!(p->state & state)) > goto out; > > Kthread_parkme > SET TASK_PARKED > schedule > raw_spin_lock(&rq->lock) > ttwu_remote > waiting for __task_rq_lock > context_switch > > finish_lock_switch > > > > Case TASK_PARKED > kthread_park_complete > > > SET Running Furthermore, Oleg noticed that the whole scheduler TASK_PARKED handling is buggered because the TASK_DEAD thing is done with preemption disabled, the current code can still complete early on preemption :/ So basically revert that earlier fix and go with a variant of the alternative mentioned in the commit. Promote TASK_PARKED to special state to avoid the store-store issue on task->state leading to the WARN in kthread_unpark() -> __kthread_bind(). But in addition, add wait_task_inactive() to kthread_park() to ensure the task really is PARKED when we return from kthread_park(). This avoids the whole kthread still gets migrated nonsense -- although it would be really good to get this done differently. Reported-by: Gaurav Kohli Signed-off-by: Peter Zijlstra (Intel) Cc: Linus Torvalds Cc: Oleg Nesterov Cc: Peter Zijlstra Cc: Thomas Gleixner Fixes: 85f1abe0019f ("kthread, sched/wait: Fix kthread_parkme() completion issue") Signed-off-by: Ingo Molnar

kthread: Allow kthread_park() on a parked kthread

2018-05-25T06:03:51Z

The following commit: 85f1abe0019f ("kthread, sched/wait: Fix kthread_parkme() completion issue") added a WARN() in the case where we call kthread_park() on an already parked thread, because the old code wasn't doing the right thing there and it wasn't at all clear that would happen. It turns out, this does in fact happen, so we have to deal with it. Instead of potentially returning early, also wait for the completion. This does however mean we have to use complete_all() and re-initialize the completion on re-use. Reported-by: LKP Tested-by: Meelis Roos Signed-off-by: Peter Zijlstra (Intel) Cc: Linus Torvalds Cc: Peter Zijlstra Cc: kernel test robot Cc: wfg@linux.intel.com Cc: Thomas Gleixner Fixes: 85f1abe0019f ("kthread, sched/wait: Fix kthread_parkme() completion issue") Link: http://lkml.kernel.org/r/20180504091142.GI12235@hirez.programming.kicks-ass.net Signed-off-by: Ingo Molnar