1Using RCU's CPU Stall Detector
2
3The rcu_cpu_stall_suppress module parameter enables RCU's CPU stall
4detector, which detects conditions that unduly delay RCU grace periods.
5This module parameter enables CPU stall detection by default, but
6may be overridden via boot-time parameter or at runtime via sysfs.
7The stall detector's idea of what constitutes "unduly delayed" is
8controlled by a set of kernel configuration variables and cpp macros:
9
10CONFIG_RCU_CPU_STALL_TIMEOUT
11
12	This kernel configuration parameter defines the period of time
13	that RCU will wait from the beginning of a grace period until it
14	issues an RCU CPU stall warning.  This time period is normally
15	ten seconds.
16
17RCU_SECONDS_TILL_STALL_RECHECK
18
19	This macro defines the period of time that RCU will wait after
20	issuing a stall warning until it issues another stall warning
21	for the same stall.  This time period is normally set to three
22	times the check interval plus thirty seconds.
23
24RCU_STALL_RAT_DELAY
25
26	The CPU stall detector tries to make the offending CPU print its
27	own warnings, as this often gives better-quality stack traces.
28	However, if the offending CPU does not detect its own stall in
29	the number of jiffies specified by RCU_STALL_RAT_DELAY, then
30	some other CPU will complain.  This delay is normally set to
31	two jiffies.
32
33When a CPU detects that it is stalling, it will print a message similar
34to the following:
35
36INFO: rcu_sched_state detected stall on CPU 5 (t=2500 jiffies)
37
38This message indicates that CPU 5 detected that it was causing a stall,
39and that the stall was affecting RCU-sched.  This message will normally be
40followed by a stack dump of the offending CPU.  On TREE_RCU kernel builds,
41RCU and RCU-sched are implemented by the same underlying mechanism,
42while on TREE_PREEMPT_RCU kernel builds, RCU is instead implemented
43by rcu_preempt_state.
44
45On the other hand, if the offending CPU fails to print out a stall-warning
46message quickly enough, some other CPU will print a message similar to
47the following:
48
49INFO: rcu_bh_state detected stalls on CPUs/tasks: { 3 5 } (detected by 2, 2502 jiffies)
50
51This message indicates that CPU 2 detected that CPUs 3 and 5 were both
52causing stalls, and that the stall was affecting RCU-bh.  This message
53will normally be followed by stack dumps for each CPU.  Please note that
54TREE_PREEMPT_RCU builds can be stalled by tasks as well as by CPUs,
55and that the tasks will be indicated by PID, for example, "P3421".
56It is even possible for a rcu_preempt_state stall to be caused by both
57CPUs -and- tasks, in which case the offending CPUs and tasks will all
58be called out in the list.
59
60Finally, if the grace period ends just as the stall warning starts
61printing, there will be a spurious stall-warning message:
62
63INFO: rcu_bh_state detected stalls on CPUs/tasks: { } (detected by 4, 2502 jiffies)
64
65This is rare, but does happen from time to time in real life.
66
67So your kernel printed an RCU CPU stall warning.  The next question is
68"What caused it?"  The following problems can result in RCU CPU stall
69warnings:
70
71o	A CPU looping in an RCU read-side critical section.
72
73o	A CPU looping with interrupts disabled.  This condition can
74	result in RCU-sched and RCU-bh stalls.
75
76o	A CPU looping with preemption disabled.  This condition can
77	result in RCU-sched stalls and, if ksoftirqd is in use, RCU-bh
78	stalls.
79
80o	A CPU looping with bottom halves disabled.  This condition can
81	result in RCU-sched and RCU-bh stalls.
82
83o	For !CONFIG_PREEMPT kernels, a CPU looping anywhere in the kernel
84	without invoking schedule().
85
86o	A CPU-bound real-time task in a CONFIG_PREEMPT kernel, which might
87	happen to preempt a low-priority task in the middle of an RCU
88	read-side critical section.   This is especially damaging if
89	that low-priority task is not permitted to run on any other CPU,
90	in which case the next RCU grace period can never complete, which
91	will eventually cause the system to run out of memory and hang.
92	While the system is in the process of running itself out of
93	memory, you might see stall-warning messages.
94
95o	A CPU-bound real-time task in a CONFIG_PREEMPT_RT kernel that
96	is running at a higher priority than the RCU softirq threads.
97	This will prevent RCU callbacks from ever being invoked,
98	and in a CONFIG_TREE_PREEMPT_RCU kernel will further prevent
99	RCU grace periods from ever completing.  Either way, the
100	system will eventually run out of memory and hang.  In the
101	CONFIG_TREE_PREEMPT_RCU case, you might see stall-warning
102	messages.
103
104o	A hardware or software issue shuts off the scheduler-clock
105	interrupt on a CPU that is not in dyntick-idle mode.  This
106	problem really has happened, and seems to be most likely to
107	result in RCU CPU stall warnings for CONFIG_NO_HZ=n kernels.
108
109o	A bug in the RCU implementation.
110
111o	A hardware failure.  This is quite unlikely, but has occurred
112	at least once in real life.  A CPU failed in a running system,
113	becoming unresponsive, but not causing an immediate crash.
114	This resulted in a series of RCU CPU stall warnings, eventually
115	leading the realization that the CPU had failed.
116
117The RCU, RCU-sched, and RCU-bh implementations have CPU stall warning.
118SRCU does not have its own CPU stall warnings, but its calls to
119synchronize_sched() will result in RCU-sched detecting RCU-sched-related
120CPU stalls.  Please note that RCU only detects CPU stalls when there is
121a grace period in progress.  No grace period, no CPU stall warnings.
122
123To diagnose the cause of the stall, inspect the stack traces.
124The offending function will usually be near the top of the stack.
125If you have a series of stall warnings from a single extended stall,
126comparing the stack traces can often help determine where the stall
127is occurring, which will usually be in the function nearest the top of
128that portion of the stack which remains the same from trace to trace.
129If you can reliably trigger the stall, ftrace can be quite helpful.
130
131RCU bugs can often be debugged with the help of CONFIG_RCU_TRACE.
132