1===============================================================
2Softlockup detector and hardlockup detector (aka nmi_watchdog)
3===============================================================
4
5The Linux kernel can act as a watchdog to detect both soft and hard
6lockups.
7
8A 'softlockup' is defined as a bug that causes the kernel to loop in
9kernel mode for more than 20 seconds (see "Implementation" below for
10details), without giving other tasks a chance to run. The current
11stack trace is displayed upon detection and, by default, the system
12will stay locked up. Alternatively, the kernel can be configured to
13panic; a sysctl, "kernel.softlockup_panic", a kernel parameter,
14"softlockup_panic" (see "Documentation/admin-guide/kernel-parameters.rst" for
15details), and a compile option, "BOOTPARAM_SOFTLOCKUP_PANIC", are
16provided for this.
17
18A 'hardlockup' is defined as a bug that causes the CPU to loop in
19kernel mode for more than 10 seconds (see "Implementation" below for
20details), without letting other interrupts have a chance to run.
21Similarly to the softlockup case, the current stack trace is displayed
22upon detection and the system will stay locked up unless the default
23behavior is changed, which can be done through a sysctl,
24'hardlockup_panic', a compile time knob, "BOOTPARAM_HARDLOCKUP_PANIC",
25and a kernel parameter, "nmi_watchdog"
26(see "Documentation/admin-guide/kernel-parameters.rst" for details).
27
28The panic option can be used in combination with panic_timeout (this
29timeout is set through the confusingly named "kernel.panic" sysctl),
30to cause the system to reboot automatically after a specified amount
31of time.
32
33Implementation
34==============
35
36The soft and hard lockup detectors are built on top of the hrtimer and
37perf subsystems, respectively. A direct consequence of this is that,
38in principle, they should work in any architecture where these
39subsystems are present.
40
41A periodic hrtimer runs to generate interrupts and kick the watchdog
42task. An NMI perf event is generated every "watchdog_thresh"
43(compile-time initialized to 10 and configurable through sysctl of the
44same name) seconds to check for hardlockups. If any CPU in the system
45does not receive any hrtimer interrupt during that time the
46'hardlockup detector' (the handler for the NMI perf event) will
47generate a kernel warning or call panic, depending on the
48configuration.
49
50The watchdog task is a high priority kernel thread that updates a
51timestamp every time it is scheduled. If that timestamp is not updated
52for 2*watchdog_thresh seconds (the softlockup threshold) the
53'softlockup detector' (coded inside the hrtimer callback function)
54will dump useful debug information to the system log, after which it
55will call panic if it was instructed to do so or resume execution of
56other kernel code.
57
58The period of the hrtimer is 2*watchdog_thresh/5, which means it has
59two or three chances to generate an interrupt before the hardlockup
60detector kicks in.
61
62As explained above, a kernel knob is provided that allows
63administrators to configure the period of the hrtimer and the perf
64event. The right value for a particular environment is a trade-off
65between fast response to lockups and detection overhead.
66
67By default, the watchdog runs on all online cores.  However, on a
68kernel configured with NO_HZ_FULL, by default the watchdog runs only
69on the housekeeping cores, not the cores specified in the "nohz_full"
70boot argument.  If we allowed the watchdog to run by default on
71the "nohz_full" cores, we would have to run timer ticks to activate
72the scheduler, which would prevent the "nohz_full" functionality
73from protecting the user code on those cores from the kernel.
74Of course, disabling it by default on the nohz_full cores means that
75when those cores do enter the kernel, by default we will not be
76able to detect if they lock up.  However, allowing the watchdog
77to continue to run on the housekeeping (non-tickless) cores means
78that we will continue to detect lockups properly on those cores.
79
80In either case, the set of cores excluded from running the watchdog
81may be adjusted via the kernel.watchdog_cpumask sysctl.  For
82nohz_full cores, this may be useful for debugging a case where the
83kernel seems to be hanging on the nohz_full cores.
84