Re: [PATCH V4] exit: trigger panic when global init has exited

Hi, Eric, Oleg Any comment? >From the previous discussions, i think this change is necessary, but we need to confirm that move the decrement of signal->live is a safe.Here are some of my considerations There are three places that are going to be called besides do_exit(). 1.

[PATCH V4] exit: trigger panic when global init has exited

From: Qianli Zhao When init sub-threads running on different CPUs exit at the same time, zap_pid_ns_processe()->BUG() may be happened(timing is as below),move panic() before set PF_EXITING to fix this problem. In addition,if panic() after other sub-threads finish do_exit(), some key variab

[PATCH V4] exit: trigger panic when global init has exited

From: Qianli Zhao When init sub-threads running on different CPUs exit at the same time, zap_pid_ns_processe()->BUG() may be happened(timing is as below),move panic() before set PF_EXITING to fix this problem. In addition,if panic() after other sub-threads finish do_exit(), some key variab

Re: [PATCH V3] exit: trigger panic when global init has exited

t; documented. That is why I said it should be documented at least in the > changelog. Ok. I will update the changelog as you suggest. Oleg Nesterov 于2021年3月25日周四 上午2:12写道： > > Hi, > > On 03/23, qianli zhao wrote: > > > > Hi,Oleg > > > > > You certainl

Re: [PATCH V3] exit: trigger panic when global init has exited

hreads finish do_exit(),these variables of sub-task will be lost,and we cannot parse the coredump of the init process through the tool normally such as "gcore". Oleg Nesterov 于2021年3月23日周二 下午5:00写道： > > On 03/23, qianli zhao wrote: > > > > Hi,Oleg > > > > &g

Re: [PATCH V3] exit: trigger panic when global init has exited

erstand each other. > > If we move atomic_dec_and_test(signal->live) and do > > if (group_dead && is_global_init) > panic(...); > > > before setting PF_EXITING like your patch does, then zap_pid_ns_processes() > simply won'

Re: [PATCH V3] exit: trigger panic when global init has exited

safe or not,from my analysis, no side effects have been found. Would you like tell me how to prove that or give me some suggestion? Thanks Eric W. Biederman 于2021年3月19日周五 上午3:09写道： > > Oleg Nesterov writes: > > > On 03/18, qianli zhao wrote: > >> > >> Hi,Ol

Re: [PATCH V3] exit: trigger panic when global init has exited

tcode=0x%08x\n") exit_notify() find_alive_thread() //no alive sub-threads zap_pid_ns_processes() //CONFIG_PID_NS is not set BUG() Oleg Nesterov 于2021年3月20日周六 上午12:32写

Re: [PATCH V3] exit: trigger panic when global init has exited

get a useable coredump. -*/ - if (unlikely(is_global_init(tsk))) - panic("Attempted to kill init! exitcode=0x%08x\n", - tsk->signal->group_exit_code ?: (int)code); - Eric W. Biederman 于2021年3月19日周五 上午3:09写道： > &g

Re: [PATCH V3] exit: trigger panic when global init has exited

ier before other init's sub-threads exit Oleg Nesterov 于2021年3月19日周五 上午2:05写道： > > On 03/18, qianli zhao wrote: > > > > Hi,Oleg > > > > Thank you for your reply. > > > > >> When init sub-threads running on different CPUs exit at the same time, > >

Re: [PATCH V3] exit: trigger panic when global init has exited

es() being called when init do_exit(). In addition, the patch also protects the init process state to successfully get usable init coredump. In my test,this patch works. Oleg Nesterov 于2021年3月17日周三 下午10:38写道： > > On 03/17, Qianli Zhao wrote: > > > > From: Qianli Zhao > >

[PATCH V3] exit: trigger panic when global init has exited

From: Qianli Zhao When init sub-threads running on different CPUs exit at the same time, zap_pid_ns_processe()->BUG() may be happened. And every thread status is abnormal after exit(PF_EXITING set,task->mm=NULL etc), which makes it difficult to parse coredump from fulldump normally. In

Re: [PATCH V2] exit: trigger panic when global init has exited

Hi Eric,Oleg > As Oleg pointer out we need to do something like the code below. > > diff --git a/kernel/exit.c b/kernel/exit.c > index 04029e35e69a..bc676c06ef9a 100644 > --- a/kernel/exit.c > +++ b/kernel/exit.c > @@ -785,15 +785,16 @@ void __noreturn do_exit(long code) >

[PATCH V2] exit: trigger panic when global init has exited

From: Qianli Zhao When init sub-threads running on different CPUs exit at the same time, zap_pid_ns_processe()->BUG() may be happened. And every thread status is abnormal after exit(PF_EXITING set,task->mm=NULL etc), which makes it difficult to parse coredump from fulldump normally. In

Re: [PATCH] exit: trigger panic when init process is set to SIGNAL_GROUP_EXIT

Hi, Eric Thank you for your suggestion > At the start of your changelog and your patch subject you describe what > you are doing but not why. For the next revision of the patch please > lead with the why it makes what you are trying to do much easier to > understand. got it. > > It does not

Re: [PATCH] exit: trigger panic when init process is set to SIGNAL_GROUP_EXIT

ark this stat,prevent other init threads from continuing to exit In addition i use siglock to protect tsk->signal->flags. > And iiuc with this patch the kernel will crash if init's sub-thread execs, > signal_group_exit() returns T in this case. Oleg Nesterov 于2021年3月10日周三 上午2:27写道

[PATCH] exit: trigger panic when init process is set to SIGNAL_GROUP_EXIT

From: Qianli Zhao Once any init thread finds SIGNAL_GROUP_EXIT, trigger panic immediately instead of last thread of global init has exited, and do not allow other init threads to exit, protect task/memory state of all sub-threads for get reliable init coredump [ 24.705376] Kernel panic

Re: [PATCH V6] kthread: Add debugobject support

Hi,tglx Would you like to continue to review the new patch set?I made some changes according to your suggestion. Unless this change will not be considered or unnecessary. Thanks Qianli Zhao 于2020年10月12日周一 上午11:00写道： > > From: Qianli Zhao > > kthread_work is not covered by d

[PATCH V6] kthread: Add debugobject support

From: Qianli Zhao kthread_work is not covered by debug objects, but the same problems as with regular work objects apply. Some of the issues like reinitialization of an active kthread_work are hard to debug because the problem manifests itself later in a completely different context. Add

Re: [PATCH v5] kthread: Add debugobject support

Hi,Thomas Thanks for your reply On Thu, 1 Oct 2020 at 22:34, Thomas Gleixner wrote: > > On Mon, Aug 17 2020 at 14:37, Qianli Zhao wrote: > > From: Qianli Zhao > > > > Add debugobject support to track the life time of kthread_work > > which is used to detect reini

[tip: timers/core] timers: Mask invalid flags in do_init_timer()

The following commit has been merged into the timers/core branch of tip: Commit-ID: b952caf2d5ca898cc10d63be7722ae7a5daca696 Gitweb: https://git.kernel.org/tip/b952caf2d5ca898cc10d63be7722ae7a5daca696 Author:Qianli Zhao AuthorDate:Thu, 13 Aug 2020 23:03:14 +08:00

Re: [PATCH v5] kthread: Add debugobject support

Dear maintainer Is this change ignored or rejected? I'm not sure who is the maintainer of kthread, please help give me a definite reply Thanks On Mon, 17 Aug 2020 at 14:37, Qianli Zhao wrote: > > From: Qianli Zhao > > Add debugobject support to track the life time of kthread_

[PATCH v4] workqueue: Warn when work flush own workqueue

From: Qianli Zhao If a workqueue flushes itself then that will lead to a deadlock. Print a warning and a stack trace when this happens. crash> ps 10856 PIDPPID CPU TASK ST COMM 108562 2 ffc873428080 UN [kworker/u16:15] crash> bt 10856 PID: 10856

Re: [kthread] 2e7d8748eb: last_state.is_incomplete_run

owing commit (built with gcc-9): > > commit: 2e7d8748eba7e32150cbd4f57129ea77d1255892 ("[RFC V2] kthread: add > object debug support") > url: > https://github.com/0day-ci/linux/commits/Qianli-Zhao/kthread-add-object-debug-support/20200812-131719 > base: https://git.k

[PATCH v3] workqueue: Warn when work flush own workqueue

From: Qianli Zhao If a workqueue flushes itself then that will lead to a deadlock. Print a warning and a stack trace when this happens. crash> ps 10856 PIDPPID CPU TASK ST COMM 108562 2 ffc873428080 UN [kworker/u16:15] crash> bt 10856 PID: 10856

Re: [PATCH v2] workqueue: Warn when work flush own workqueue

Markus Thanks for your suggestion,and sorry for my poor wording. On Tue, Aug 25, 2020 at 4:00 PM Markus Elfring wrote: > > > Flushing own workqueue or work self in work context will lead to > > a deadlock. > > I imagine that the wording “or work self” can become clearer another bit. > > > >

[PATCH v2] workqueue: Warn when work flush own workqueue

From: Qianli Zhao Flushing own workqueue or work self in work context will lead to a deadlock. Catch this incorrect usage and issue a warning when issue happened crash> ps 10856 PIDPPID CPU TASK ST COMM 108562 2 ffc873428080 UN [kworker/u16:15] crash> bt

[PATCH] workqueue: Warning when work try to flush own workqueue

From: Qianli Zhao In a work process context,flush own workqueue or work self will cause process blocked(enter state D),leading to a deadlock,catch this wrong use,warn when the issue happened crash> ps 10856 PIDPPID CPU TASK ST COMM 108562 2 ffc873428080

[PATCH v5] kthread: Add debugobject support

From: Qianli Zhao Add debugobject support to track the life time of kthread_work which is used to detect reinitialization/free active object problems Add kthread_init_work_onstack()/kthread_init_delayed_work_onstack() for kthread onstack support If we reinitialize a kthread_work that has been

[PATCH v4] kthread: Add debugobject support

From: Qianli Zhao Add debugobject support to track the life time of kthread_work which is used to detect reinitialization/free active object problems Add kthread_init_work_onstack()/kthread_init_delayed_work_onstack() for kthread onstack support If we reinitialize a kthread_work that has been

Re: [PATCH v3] kthread: add objectdebug support

Hi,Stephen Thanks for your suggestion, i will improve my patch. Thanks. On Sat, Aug 15, 2020 at 4:18 PM Stephen Boyd wrote: > > Quoting Qianli Zhao (2020-08-13 02:55:16) > > From: Qianli Zhao > > > > Add debugobject support to track the life time of kthread_work >

Re: [PATCH] timer: mask unnecessary set of flags in do_init_timer

Thomas Gleixner 于2020年8月13日周四 下午6:46写道： > > Qianli Zhao writes: > > Please start the first word after the colon with an upper case letter. > > > do_init_timer can specify flags of timer_list, > > Please write do_init_timer() so it's entirely clear that

[PATCH v2] timer: Mask illegal set of flags in do_init_timer()

From: Qianli Zhao do_init_timer() can specify flags of timer_list, only TIMER_DEFFERABLE, TIMER_PINNED, TIMER_IRQSAFE are legal do a sanity check, mask and warning illegal set of flags Signed-off-by: Qianli Zhao --- V2: - update changelog - mask and warning illegal set --- include/linux

[PATCH v3] kthread: add objectdebug support

From: Qianli Zhao Add debugobject support to track the life time of kthread_work which is used to detect reinitialization/free active object problems Add kthread_init_work_onstack/kthread_init_delayed_work_onstack for kthread onstack support If we reinitialize a kthread_work that has been

[RFC V2] kthread: add object debug support

From: Qianli Zhao Add debugobject support to track the life time of kthread_work which is used to detect reinitialization/free active object problems Add kthread_init_work_onstack/kthread_init_delayed_work_onstack for kthread onstack support Signed-off-by: Qianli Zhao --- I got an crash issue

[RFC] kthread: add object debug support

From: Qianli Zhao Add debugobject support to track the life time of kthread_work which is used to detect reinitialization/free active object problems Add kthread_init_work_onstack/kthread_init_delayed_work_onstack for kthread onstack support Signed-off-by: Qianli Zhao --- include/linux

[PATCH] timer: mask unnecessary set of flags in do_init_timer

From: Qianli Zhao do_init_timer can specify flags of timer_list, but this function does not expect to specify the CPU or idx. If user invoking do_init_timer and specify CPU, The result may not what we expected. E.g: do_init_timer invoked in core2,and specify flags 0x1 final result flags is 0x3

[PATCH] firewire: firewire-cdev.h: Avoid the use of one-element array

From: Qianli Zhao There is a regular need in the kernel to provide a way to declare having a dynamically sized set of trailing elements in a structure. Kernel code should always use “flexible array members”[1] for these cases. The older style of one-element or zero-length arrays should no longer

[PATCH] mm: slab: Avoid the use of one-element array and use struct_size() helper

From: Qianli Zhao There is a regular need in the kernel to provide a way to declare having a dynamically sized set of trailing elements in a structure. Kernel code should always use “flexible array members”[1] for these cases. The older style of one-element or zero-length arrays should no longer