Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1450788 > unrolled thread
| Started by | Oleg Nesterov <oleg@redhat.com> |
|---|---|
| First post | 2016-07-26 21:00 +0200 |
| Last post | 2016-07-27 12:50 +0200 |
| Articles | 4 — 3 participants |
Back to article view | Back to linux.kernel
This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by
below is the oldest one visible, not the original post.
[PATCH 0/1] (Was: introduce for_each_process_thread_{break,continue}() helpers) Oleg Nesterov <oleg@redhat.com> - 2016-07-26 21:00 +0200
[PATCH 1/1] stop_machine: touch_nmi_watchdog() after MULTI_STOP_PREPARE Oleg Nesterov <oleg@redhat.com> - 2016-07-26 21:00 +0200
Re: [PATCH 1/1] stop_machine: touch_nmi_watchdog() after MULTI_STOP_PREPARE Thomas Gleixner <tglx@linutronix.de> - 2016-07-27 10:10 +0200
[tip:core/urgent] stop_machine: Touch_nmi_watchdog() after MULTI_STOP_PREPARE tip-bot for Oleg Nesterov <tipbot@zytor.com> - 2016-07-27 12:50 +0200
| From | Oleg Nesterov <oleg@redhat.com> |
|---|---|
| Date | 2016-07-26 21:00 +0200 |
| Subject | [PATCH 0/1] (Was: introduce for_each_process_thread_{break,continue}() helpers) |
| Message-ID | <rZfK9-5Fs-9@gated-at.bofh.it> |
On 07/25, Oleg Nesterov wrote: > > IMHO this makes sense in any case, but mostly this is preparation for > another change: show_state_filter() should be preemptible. But this needs > more discussion, I'll write another email/patch when I fully understand > the hard-lockup caused by sysrq-t. Yes, we need to do something with show_state_filter() anyway, I think. OTOH, I believe this simple change in multi_cpu_stop() makes sense too regardless. Oleg.
[toc] | [next] | [standalone]
| From | Oleg Nesterov <oleg@redhat.com> |
|---|---|
| Date | 2016-07-26 21:00 +0200 |
| Subject | [PATCH 1/1] stop_machine: touch_nmi_watchdog() after MULTI_STOP_PREPARE |
| Message-ID | <rZfKa-5Fs-13@gated-at.bofh.it> |
| In reply to | #1450788 |
Suppose that stop_machine(fn) hangs because fn() hangs. In this case NMI
hard-lockup can be triggered on another CPU which does nothing wrong and
the trace from nmi_panic() won't help to investigate the problem.
And this change "fixes" the problem we (seem to) hit in practice.
- stop_two_cpus(0, 1) races with show_state_filter() running on CPU_0.
- CPU_1 already spins in MULTI_STOP_PREPARE state, it detects the soft
lockup and tries to report the problem.
- show_state_filter() enables preemption, CPU_0 calls multi_cpu_stop()
which goes to MULTI_STOP_DISABLE_IRQ state and disables interrupts.
- CPU_1 spends more than 10 seconds trying to flush the log buffer to
the slow serial console.
- NMI interrupt on CPU_0 (which now waits for CPU_1) calls nmi_panic().
Reported-by: Wang Shu <shuwang@redhat.com>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
---
kernel/stop_machine.c | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c
index a467e6c..4a1ca5f 100644
--- a/kernel/stop_machine.c
+++ b/kernel/stop_machine.c
@@ -21,6 +21,7 @@
#include <linux/smpboot.h>
#include <linux/atomic.h>
#include <linux/lglock.h>
+#include <linux/nmi.h>
/*
* Structure to determine completion condition and record errors. May
@@ -209,6 +210,13 @@ static int multi_cpu_stop(void *data)
break;
}
ack_state(msdata);
+ } else if (curstate > MULTI_STOP_PREPARE) {
+ /*
+ * At this stage all other CPUs we depend on must spin
+ * in the same loop. Any reason for hard-lockup should
+ * be detected and reported on their side.
+ */
+ touch_nmi_watchdog();
}
} while (curstate != MULTI_STOP_EXIT);
--
2.5.0
[toc] | [prev] | [next] | [standalone]
| From | Thomas Gleixner <tglx@linutronix.de> |
|---|---|
| Date | 2016-07-27 10:10 +0200 |
| Subject | Re: [PATCH 1/1] stop_machine: touch_nmi_watchdog() after MULTI_STOP_PREPARE |
| Message-ID | <rZs4G-5fF-13@gated-at.bofh.it> |
| In reply to | #1450791 |
On Tue, 26 Jul 2016, Oleg Nesterov wrote:
> Suppose that stop_machine(fn) hangs because fn() hangs. In this case NMI
> hard-lockup can be triggered on another CPU which does nothing wrong and
> the trace from nmi_panic() won't help to investigate the problem.
>
> And this change "fixes" the problem we (seem to) hit in practice.
>
> - stop_two_cpus(0, 1) races with show_state_filter() running on CPU_0.
>
> - CPU_1 already spins in MULTI_STOP_PREPARE state, it detects the soft
> lockup and tries to report the problem.
>
> - show_state_filter() enables preemption, CPU_0 calls multi_cpu_stop()
> which goes to MULTI_STOP_DISABLE_IRQ state and disables interrupts.
>
> - CPU_1 spends more than 10 seconds trying to flush the log buffer to
> the slow serial console.
>
> - NMI interrupt on CPU_0 (which now waits for CPU_1) calls nmi_panic().
>
> Reported-by: Wang Shu <shuwang@redhat.com>
> Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
> ---
> kernel/stop_machine.c | 8 ++++++++
> 1 file changed, 8 insertions(+)
>
> diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c
> index a467e6c..4a1ca5f 100644
> --- a/kernel/stop_machine.c
> +++ b/kernel/stop_machine.c
> @@ -21,6 +21,7 @@
> #include <linux/smpboot.h>
> #include <linux/atomic.h>
> #include <linux/lglock.h>
> +#include <linux/nmi.h>
>
> /*
> * Structure to determine completion condition and record errors. May
> @@ -209,6 +210,13 @@ static int multi_cpu_stop(void *data)
> break;
> }
> ack_state(msdata);
> + } else if (curstate > MULTI_STOP_PREPARE) {
> + /*
> + * At this stage all other CPUs we depend on must spin
> + * in the same loop. Any reason for hard-lockup should
> + * be detected and reported on their side.
> + */
> + touch_nmi_watchdog();
> }
> } while (curstate != MULTI_STOP_EXIT);
>
> --
> 2.5.0
>
>
>
[toc] | [prev] | [next] | [standalone]
| From | tip-bot for Oleg Nesterov <tipbot@zytor.com> |
|---|---|
| Date | 2016-07-27 12:50 +0200 |
| Subject | [tip:core/urgent] stop_machine: Touch_nmi_watchdog() after MULTI_STOP_PREPARE |
| Message-ID | <rZuzw-6GJ-17@gated-at.bofh.it> |
| In reply to | #1450791 |
Commit-ID: ce4f06dcbb5d6d04d202f1b81ac72d5679dcdfc0
Gitweb: http://git.kernel.org/tip/ce4f06dcbb5d6d04d202f1b81ac72d5679dcdfc0
Author: Oleg Nesterov <oleg@redhat.com>
AuthorDate: Tue, 26 Jul 2016 20:57:36 +0200
Committer: Ingo Molnar <mingo@kernel.org>
CommitDate: Wed, 27 Jul 2016 11:12:11 +0200
stop_machine: Touch_nmi_watchdog() after MULTI_STOP_PREPARE
Suppose that stop_machine(fn) hangs because fn() hangs. In this case NMI
hard-lockup can be triggered on another CPU which does nothing wrong and
the trace from nmi_panic() won't help to investigate the problem.
And this change "fixes" the problem we (seem to) hit in practice.
- stop_two_cpus(0, 1) races with show_state_filter() running on CPU_0.
- CPU_1 already spins in MULTI_STOP_PREPARE state, it detects the soft
lockup and tries to report the problem.
- show_state_filter() enables preemption, CPU_0 calls multi_cpu_stop()
which goes to MULTI_STOP_DISABLE_IRQ state and disables interrupts.
- CPU_1 spends more than 10 seconds trying to flush the log buffer to
the slow serial console.
- NMI interrupt on CPU_0 (which now waits for CPU_1) calls nmi_panic().
Reported-by: Wang Shu <shuwang@redhat.com>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Dave Anderson <anderson@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Tejun Heo <tj@kernel.org>
Link: http://lkml.kernel.org/r/20160726185736.GB4088@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
---
kernel/stop_machine.c | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c
index a467e6c..4a1ca5f 100644
--- a/kernel/stop_machine.c
+++ b/kernel/stop_machine.c
@@ -21,6 +21,7 @@
#include <linux/smpboot.h>
#include <linux/atomic.h>
#include <linux/lglock.h>
+#include <linux/nmi.h>
/*
* Structure to determine completion condition and record errors. May
@@ -209,6 +210,13 @@ static int multi_cpu_stop(void *data)
break;
}
ack_state(msdata);
+ } else if (curstate > MULTI_STOP_PREPARE) {
+ /*
+ * At this stage all other CPUs we depend on must spin
+ * in the same loop. Any reason for hard-lockup should
+ * be detected and reported on their side.
+ */
+ touch_nmi_watchdog();
}
} while (curstate != MULTI_STOP_EXIT);
[toc] | [prev] | [standalone]
Back to top | Article view | linux.kernel
csiph-web