Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1615119

Re: [RFC][PATCHv2 1/8] printk: move printk_pending out of per-cpu

From Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com>
Newsgroups linux.kernel
Subject Re: [RFC][PATCHv2 1/8] printk: move printk_pending out of per-cpu
Date 2017-04-03 13:30 +0200
Message-ID <ts8lk-6bt-29@gated-at.bofh.it> (permalink)
References <tqi5s-4RU-3@gated-at.bofh.it> <tqif7-4VF-9@gated-at.bofh.it> <tr4D8-56f-27@gated-at.bofh.it> <tr4Wu-5cZ-17@gated-at.bofh.it>
Organization linux.* mail to news gateway

Show all headers | View raw


On (03/31/17 15:33), Peter Zijlstra wrote:
> On Fri, Mar 31, 2017 at 03:09:50PM +0200, Petr Mladek wrote:
> > On Wed 2017-03-29 18:25:04, Sergey Senozhatsky wrote:
> 
> > >  	if (waitqueue_active(&log_wait)) {
> > > -		this_cpu_or(printk_pending, PRINTK_PENDING_WAKEUP);
> > > +		set_bit(PRINTK_PENDING_WAKEUP, &printk_pending);
> > 
> > We should add here a write barrier:
> > 
> > 	/*
> > 	 * irq_work_queue() uses cmpxchg() and implies the memory
> > 	 * barrier only when the work is queued. An explicit barrier
> > 	 * is needed here to make sure that wake_up_klogd_work_func()
> > 	 * sees printk_pending set even when the work was already queued
> > 	 * because of an other pending event.
> > 	 */
> > 	 smp_wmb();
> > 
> > >  		irq_work_queue(this_cpu_ptr(&wake_up_klogd_work));
> > >  	}
> > >  	preempt_enable();
> 
> smp_mb__after_atomic() is probably better, because if you're not
> ordering with the cmpxchg, you're ordering against a load done by
> cmpxchg to see it doesn't need to do anything.

Petr and Peter, thanks for the review.

can you educate me, what exactly is broken there?

when called from console_unlock(), we have something as follows

	console_unlock()
	{
		for (;;) {
			spin_lock_irqsave();
			...
			spin_unlock_irqrestore();
			...
		}

		spin_unlock_irqrestore();

<<IRQs enabled>>

		if (wake_klogd)
			wake_up_klogd()
			{
				set_bit(PRINTK_PENDING_WAKEUP, &printk_pending);
				irq_work_queue(this_cpu_ptr(&wake_up_klogd_work));
			}
	}


we queue a per-CPU irq_work. given that by the time we execute wake_up_klogd()
we have local IRQs enabled on that CPU. is it possible that we will have that
CPU's irq_work still being queued?


when called from printk_deferred().

I'm still trying to understand what scenario can cause the problem. so
basically on that CPU we have a call into the scheduler/timer which ends
up in printk_deferred()... and then we have console_unlock()->wake_up_klogd()
//* local IRQs enabled but the irq_work is still queued *// and atop of it
we have IRQ that executes that CPU's run_list and fails to see updated
PRINTK_PENDING_WAKEUP bit, because wake_up_klogd() was called on already
queued wake_up_klogd_work. is this the case? if so, can this race happen on
the CPU?

I don't object the barrier, I'm just trying to have a better understanding
what's broken. sorry if I'm missing something very obvious.

	-ss

Back to linux.kernel | Previous | NextPrevious in thread | Next in thread | Find similar | Unroll thread


Thread

[RFC][PATCHv2 0/8] printk: introduce printing kernel thread Sergey Senozhatsky <sergey.senozhatsky@gmail.com> - 2017-03-29 11:30 +0200
  [RFC][PATCHv2 2/8] printk: introduce printing kernel thread Sergey Senozhatsky <sergey.senozhatsky@gmail.com> - 2017-03-29 11:30 +0200
    Re: [RFC][PATCHv2 2/8] printk: introduce printing kernel thread Petr Mladek <pmladek@suse.com> - 2017-04-04 11:10 +0200
      Re: [RFC][PATCHv2 2/8] printk: introduce printing kernel thread Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com> - 2017-04-04 11:40 +0200
    Re: [RFC][PATCHv2 2/8] printk: introduce printing kernel thread Pavel Machek <pavel@ucw.cz> - 2017-04-06 19:20 +0200
      Re: [RFC][PATCHv2 2/8] printk: introduce printing kernel thread Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com> - 2017-04-07 07:20 +0200
        Re: [RFC][PATCHv2 2/8] printk: introduce printing kernel thread Pavel Machek <pavel@ucw.cz> - 2017-04-07 09:30 +0200
          Re: [RFC][PATCHv2 2/8] printk: introduce printing kernel thread Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com> - 2017-04-07 10:20 +0200
            Re: [RFC][PATCHv2 2/8] printk: introduce printing kernel thread Pavel Machek <pavel@ucw.cz> - 2017-04-07 14:10 +0200
  [RFC][PATCHv2 5/8] sysrq: switch to printk.emergency mode in unsafe places Sergey Senozhatsky <sergey.senozhatsky@gmail.com> - 2017-03-29 11:30 +0200
    Re: [RFC][PATCHv2 5/8] sysrq: switch to printk.emergency mode in  unsafe places Petr Mladek <pmladek@suse.com> - 2017-03-31 17:40 +0200
      Re: [RFC][PATCHv2 5/8] sysrq: switch to printk.emergency mode in  unsafe places Sergey Senozhatsky <sergey.senozhatsky@gmail.com> - 2017-04-01 02:10 +0200
  [RFC][PATCHv2 1/8] printk: move printk_pending out of per-cpu Sergey Senozhatsky <sergey.senozhatsky@gmail.com> - 2017-03-29 11:40 +0200
    Re: [RFC][PATCHv2 1/8] printk: move printk_pending out of per-cpu Petr Mladek <pmladek@suse.com> - 2017-03-31 15:20 +0200
      Re: [RFC][PATCHv2 1/8] printk: move printk_pending out of per-cpu Peter Zijlstra <peterz@infradead.org> - 2017-03-31 15:40 +0200
        Re: [RFC][PATCHv2 1/8] printk: move printk_pending out of per-cpu Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com> - 2017-04-03 13:30 +0200
          Re: [RFC][PATCHv2 1/8] printk: move printk_pending out of per-cpu Petr Mladek <pmladek@suse.com> - 2017-04-03 14:50 +0200
  [RFC][PATCHv2 6/8] kexec: switch to printk.emergency mode in unsafe places Sergey Senozhatsky <sergey.senozhatsky@gmail.com> - 2017-03-29 11:40 +0200
    Re: [RFC][PATCHv2 6/8] kexec: switch to printk.emergency mode in  unsafe places Petr Mladek <pmladek@suse.com> - 2017-03-31 17:40 +0200
  [RFC][PATCHv2 8/8] printk: enable printk offloading Sergey Senozhatsky <sergey.senozhatsky@gmail.com> - 2017-03-29 11:40 +0200
    Re: [printk]  fbc14616f4:  BUG:kernel_reboot-without-warning_in_test_stage Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com> - 2017-03-31 04:40 +0200
      Re: [printk]  fbc14616f4:  BUG:kernel_reboot-without-warning_in_test_stage Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com> - 2017-03-31 06:10 +0200
        Re: [printk]  fbc14616f4:  BUG:kernel_reboot-without-warning_in_test_stage Ye Xiaolong <xiaolong.ye@intel.com> - 2017-03-31 08:50 +0200
          Re: [printk]  fbc14616f4:  BUG:kernel_reboot-without-warning_in_test_stage Sergey Senozhatsky <sergey.senozhatsky@gmail.com> - 2017-03-31 16:50 +0200
            Re: [printk]  fbc14616f4: BUG:kernel_reboot-without-warning_in_test_stage ebiederm@xmission.com (Eric W. Biederman) - 2017-03-31 17:40 +0200
              Re: [printk]  fbc14616f4:  BUG:kernel_reboot-without-warning_in_test_stage Jan Kara <jack@suse.cz> - 2017-04-03 11:40 +0200
                Re: [printk]  fbc14616f4:  BUG:kernel_reboot-without-warning_in_test_stage Petr Mladek <pmladek@suse.com> - 2017-04-03 12:10 +0200
                Re: [printk]  fbc14616f4:  BUG:kernel_reboot-without-warning_in_test_stage Pavel Machek <pavel@ucw.cz> - 2017-04-06 19:40 +0200
                Re: [printk]  fbc14616f4:  BUG:kernel_reboot-without-warning_in_test_stage Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com> - 2017-04-07 06:50 +0200
                Re: [printk]  fbc14616f4:  BUG:kernel_reboot-without-warning_in_test_stage Pavel Machek <pavel@ucw.cz> - 2017-04-07 09:20 +0200
                Re: [printk]  fbc14616f4:  BUG:kernel_reboot-without-warning_in_test_stage Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com> - 2017-04-07 09:50 +0200
                Re: [printk]  fbc14616f4:  BUG:kernel_reboot-without-warning_in_test_stage Pavel Machek <pavel@ucw.cz> - 2017-04-07 10:20 +0200
                Re: [printk]  fbc14616f4:  BUG:kernel_reboot-without-warning_in_test_stage Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com> - 2017-04-07 14:20 +0200
                Re: [printk]  fbc14616f4:  BUG:kernel_reboot-without-warning_in_test_stage Pavel Machek <pavel@ucw.cz> - 2017-04-07 14:50 +0200
                Re: [printk]  fbc14616f4:  BUG:kernel_reboot-without-warning_in_test_stage Steven Rostedt <rostedt@goodmis.org> - 2017-04-07 16:50 +0200
                Re: [printk]  fbc14616f4:  BUG:kernel_reboot-without-warning_in_test_stage Sergey Senozhatsky <sergey.senozhatsky@gmail.com> - 2017-04-07 17:20 +0200
                Re: [printk]  fbc14616f4:  BUG:kernel_reboot-without-warning_in_test_stage Peter Zijlstra <peterz@infradead.org> - 2017-04-07 17:30 +0200
                Re: [printk]  fbc14616f4:  BUG:kernel_reboot-without-warning_in_test_stage Sergey Senozhatsky <sergey.senozhatsky@gmail.com> - 2017-04-07 17:50 +0200
                Re: [printk]  fbc14616f4: BUG:kernel_reboot-without-warning_in_test_stage ebiederm@xmission.com (Eric W. Biederman) - 2017-04-09 20:30 +0200
                Re: [printk]  fbc14616f4:  BUG:kernel_reboot-without-warning_in_test_stage Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com> - 2017-04-10 06:50 +0200
                Re: [printk]  fbc14616f4:  BUG:kernel_reboot-without-warning_in_test_stage Pavel Machek <pavel@ucw.cz> - 2017-04-09 12:20 +0200
                Re: [printk]  fbc14616f4:  BUG:kernel_reboot-without-warning_in_test_stage Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com> - 2017-04-10 07:00 +0200
                Re: [printk]  fbc14616f4:  BUG:kernel_reboot-without-warning_in_test_stage Petr Mladek <pmladek@suse.com> - 2017-04-10 14:00 +0200
                Re: [printk]  fbc14616f4:  BUG:kernel_reboot-without-warning_in_test_stage Steven Rostedt <rostedt@goodmis.org> - 2017-04-07 16:40 +0200
                Re: [printk]  fbc14616f4:  BUG:kernel_reboot-without-warning_in_test_stage Pavel Machek <pavel@ucw.cz> - 2017-04-09 12:00 +0200
              Re: [printk]  fbc14616f4:  BUG:kernel_reboot-without-warning_in_test_stage Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com> - 2017-04-03 13:00 +0200
          Re: [printk]  fbc14616f4:  BUG:kernel_reboot-without-warning_in_test_stage Ye Xiaolong <xiaolong.ye@intel.com> - 2017-04-05 09:40 +0200
            Re: [printk]  fbc14616f4:  BUG:kernel_reboot-without-warning_in_test_stage Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com> - 2017-04-05 10:50 +0200
    Re: [RFC][PATCHv2 8/8] printk: enable printk offloading Petr Mladek <pmladek@suse.com> - 2017-04-03 17:50 +0200
      Re: [RFC][PATCHv2 8/8] printk: enable printk offloading Sergey Senozhatsky <sergey.senozhatsky@gmail.com> - 2017-04-04 14:30 +0200
  [RFC][PATCHv2 4/8] pm: switch to printk.emergency mode in unsafe places Sergey Senozhatsky <sergey.senozhatsky@gmail.com> - 2017-03-29 11:40 +0200
    Re: [RFC][PATCHv2 4/8] pm: switch to printk.emergency mode in unsafe  places Petr Mladek <pmladek@suse.com> - 2017-03-31 17:10 +0200
    Re: [RFC][PATCHv2 4/8] pm: switch to printk.emergency mode in unsafe  places Pavel Machek <pavel@ucw.cz> - 2017-04-06 19:30 +0200
      Re: [RFC][PATCHv2 4/8] pm: switch to printk.emergency mode in unsafe  places Andreas Mohr <andi@lisas.de> - 2017-04-09 13:00 +0200
        Re: [RFC][PATCHv2 4/8] pm: switch to printk.emergency mode in unsafe  places Petr Mladek <pmladek@suse.com> - 2017-04-10 14:30 +0200
          Re: [RFC][PATCHv2 4/8] pm: switch to printk.emergency mode in unsafe  places Sergey Senozhatsky <sergey.senozhatsky@gmail.com> - 2017-04-10 16:40 +0200
  [RFC][PATCHv2 7/8] printk: add printk emergency_mode parameter Sergey Senozhatsky <sergey.senozhatsky@gmail.com> - 2017-03-29 11:40 +0200
    Re: [RFC][PATCHv2 7/8] printk: add printk emergency_mode parameter Petr Mladek <pmladek@suse.com> - 2017-04-03 17:30 +0200
      Re: [RFC][PATCHv2 7/8] printk: add printk emergency_mode parameter Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com> - 2017-04-04 10:30 +0200

csiph-web