blob: 90aa4531cb67343b03b396610aa0db77c1d1e0ba [file] [log] [blame]
Linus Torvalds1da177e2005-04-16 15:20:36 -07001
2[NMI watchdog is available for x86 and x86-64 architectures]
3
4Is your system locking up unpredictably? No keyboard activity, just
5a frustrating complete hard lockup? Do you want to help us debugging
6such lockups? If all yes then this document is definitely for you.
7
8On many x86/x86-64 type hardware there is a feature that enables
9us to generate 'watchdog NMI interrupts'. (NMI: Non Maskable Interrupt
10which get executed even if the system is otherwise locked up hard).
11This can be used to debug hard kernel lockups. By executing periodic
12NMI interrupts, the kernel can monitor whether any CPU has locked up,
Cyrill Gorcunovafda3352008-06-27 19:43:40 +040013and print out debugging messages if so.
Linus Torvalds1da177e2005-04-16 15:20:36 -070014
15In order to use the NMI watchdog, you need to have APIC support in your
16kernel. For SMP kernels, APIC support gets compiled in automatically. For
17UP, enable either CONFIG_X86_UP_APIC (Processor type and features -> Local
18APIC support on uniprocessors) or CONFIG_X86_UP_IOAPIC (Processor type and
19features -> IO-APIC support on uniprocessors) in your kernel config.
20CONFIG_X86_UP_APIC is for uniprocessor machines without an IO-APIC.
21CONFIG_X86_UP_IOAPIC is for uniprocessor with an IO-APIC. [Note: certain
22kernel debugging options, such as Kernel Stack Meter or Kernel Tracer,
23may implicitly disable the NMI watchdog.]
24
Cyrill Gorcunovafda3352008-06-27 19:43:40 +040025For x86-64, the needed APIC is always compiled in.
Linus Torvalds1da177e2005-04-16 15:20:36 -070026
27Using local APIC (nmi_watchdog=2) needs the first performance register, so
28you can't use it for other purposes (such as high precision performance
29profiling.) However, at least oprofile and the perfctr driver disable the
30local APIC NMI watchdog automatically.
31
32To actually enable the NMI watchdog, use the 'nmi_watchdog=N' boot
33parameter. Eg. the relevant lilo.conf entry:
34
35 append="nmi_watchdog=1"
36
37For SMP machines and UP machines with an IO-APIC use nmi_watchdog=1.
38For UP machines without an IO-APIC use nmi_watchdog=2, this only works
39for some processor types. If in doubt, boot with nmi_watchdog=1 and
40check the NMI count in /proc/interrupts; if the count is zero then
41reboot with nmi_watchdog=2 and check the NMI count. If it is still
42zero then log a problem, you probably have a processor that needs to be
43added to the nmi code.
44
45A 'lockup' is the following scenario: if any CPU in the system does not
46execute the period local timer interrupt for more than 5 seconds, then
47the NMI handler generates an oops and kills the process. This
48'controlled crash' (and the resulting kernel messages) can be used to
49debug the lockup. Thus whenever the lockup happens, wait 5 seconds and
50the oops will show up automatically. If the kernel produces no messages
51then the system has crashed so hard (eg. hardware-wise) that either it
52cannot even accept NMI interrupts, or the crash has made the kernel
53unable to print messages.
54
55Be aware that when using local APIC, the frequency of NMI interrupts
56it generates, depends on the system load. The local APIC NMI watchdog,
57lacking a better source, uses the "cycles unhalted" event. As you may
58guess it doesn't tick when the CPU is in the halted state (which happens
59when the system is idle), but if your system locks up on anything but the
60"hlt" processor instruction, the watchdog will trigger very soon as the
61"cycles unhalted" event will happen every clock tick. If it locks up on
62"hlt", then you are out of luck -- the event will not happen at all and the
63watchdog won't trigger. This is a shortcoming of the local APIC watchdog
64-- unfortunately there is no "clock ticks" event that would work all the
Cyrill Gorcunovafda3352008-06-27 19:43:40 +040065time. The I/O APIC watchdog is driven externally and has no such shortcoming.
Linus Torvalds1da177e2005-04-16 15:20:36 -070066But its NMI frequency is much higher, resulting in a more significant hit
67to the overall system performance.
68
Cyrill Gorcunovafda3352008-06-27 19:43:40 +040069On x86 nmi_watchdog is disabled by default so you have to enable it with
70a boot time parameter.
Linus Torvalds1da177e2005-04-16 15:20:36 -070071
Ingo Molnar1bb3a022008-06-30 08:47:42 +020072NOTE: In kernels prior to 2.4.2-ac18 the NMI-oopser is enabled unconditionally
Cyrill Gorcunovafda3352008-06-27 19:43:40 +040073on x86 SMP boxes.
Linus Torvalds1da177e2005-04-16 15:20:36 -070074
75[ feel free to send bug reports, suggestions and patches to
76 Ingo Molnar <mingo@redhat.com> or the Linux SMP mailing
77 list at <linux-smp@vger.kernel.org> ]
78