diff options
| -rw-r--r-- | Documentation/lockup-watchdogs.txt | 63 | ||||
| -rw-r--r-- | Documentation/nmi_watchdog.txt | 83 | ||||
| -rw-r--r-- | arch/x86/include/asm/inat.h | 5 | ||||
| -rw-r--r-- | arch/x86/include/asm/insn.h | 18 | ||||
| -rw-r--r-- | arch/x86/lib/inat.c | 36 | ||||
| -rw-r--r-- | arch/x86/lib/insn.c | 13 | ||||
| -rw-r--r-- | kernel/watchdog.c | 24 | ||||
| -rw-r--r-- | lib/Kconfig.debug | 18 |
8 files changed, 126 insertions, 134 deletions
diff --git a/Documentation/lockup-watchdogs.txt b/Documentation/lockup-watchdogs.txt new file mode 100644 index 000000000000..d2a36602ca8d --- /dev/null +++ b/Documentation/lockup-watchdogs.txt @@ -0,0 +1,63 @@ +=============================================================== +Softlockup detector and hardlockup detector (aka nmi_watchdog) +=============================================================== + +The Linux kernel can act as a watchdog to detect both soft and hard +lockups. + +A 'softlockup' is defined as a bug that causes the kernel to loop in +kernel mode for more than 20 seconds (see "Implementation" below for +details), without giving other tasks a chance to run. The current +stack trace is displayed upon detection and, by default, the system +will stay locked up. Alternatively, the kernel can be configured to +panic; a sysctl, "kernel.softlockup_panic", a kernel parameter, +"softlockup_panic" (see "Documentation/kernel-parameters.txt" for +details), and a compile option, "BOOTPARAM_HARDLOCKUP_PANIC", are +provided for this. + +A 'hardlockup' is defined as a bug that causes the CPU to loop in +kernel mode for more than 10 seconds (see "Implementation" below for +details), without letting other interrupts have a chance to run. +Similarly to the softlockup case, the current stack trace is displayed +upon detection and the system will stay locked up unless the default +behavior is changed, which can be done through a compile time knob, +"BOOTPARAM_HARDLOCKUP_PANIC", and a kernel parameter, "nmi_watchdog" +(see "Documentation/kernel-parameters.txt" for details). + +The panic option can be used in combination with panic_timeout (this +timeout is set through the confusingly named "kernel.panic" sysctl), +to cause the system to reboot automatically after a specified amount +of time. + +=== Implementation === + +The soft and hard lockup detectors are built on top of the hrtimer and +perf subsystems, respectively. A direct consequence of this is that, +in principle, they should work in any architecture where these +subsystems are present. + +A periodic hrtimer runs to generate interrupts and kick the watchdog +task. An NMI perf event is generated every "watchdog_thresh" +(compile-time initialized to 10 and configurable through sysctl of the +same name) seconds to check for hardlockups. If any CPU in the system +does not receive any hrtimer interrupt during that time the +'hardlockup detector' (the handler for the NMI perf event) will +generate a kernel warning or call panic, depending on the +configuration. + +The watchdog task is a high priority kernel thread that updates a +timestamp every time it is scheduled. If that timestamp is not updated +for 2*watchdog_thresh seconds (the softlockup threshold) the +'softlockup detector' (coded inside the hrtimer callback function) +will dump useful debug information to the system log, after which it +will call panic if it was instructed to do so or resume execution of +other kernel code. + +The period of the hrtimer is 2*watchdog_thresh/5, which means it has +two or three chances to generate an interrupt before the hardlockup +detector kicks in. + +As explained above, a kernel knob is provided that allows +administrators to configure the period of the hrtimer and the perf +event. The right value for a particular environment is a trade-off +between fast response to lockups and detection overhead. diff --git a/Documentation/nmi_watchdog.txt b/Documentation/nmi_watchdog.txt deleted file mode 100644 index bf9f80a98282..000000000000 --- a/Documentation/nmi_watchdog.txt +++ /dev/null @@ -1,83 +0,0 @@ - -[NMI watchdog is available for x86 and x86-64 architectures] - -Is your system locking up unpredictably? No keyboard activity, just -a frustrating complete hard lockup? Do you want to help us debugging -such lockups? If all yes then this document is definitely for you. - -On many x86/x86-64 type hardware there is a feature that enables -us to generate 'watchdog NMI interrupts'. (NMI: Non Maskable Interrupt -which get executed even if the system is otherwise locked up hard). -This can be used to debug hard kernel lockups. By executing periodic -NMI interrupts, the kernel can monitor whether any CPU has locked up, -and print out debugging messages if so. - -In order to use the NMI watchdog, you need to have APIC support in your -kernel. For SMP kernels, APIC support gets compiled in automatically. For -UP, enable either CONFIG_X86_UP_APIC (Processor type and features -> Local -APIC support on uniprocessors) or CONFIG_X86_UP_IOAPIC (Processor type and -features -> IO-APIC support on uniprocessors) in your kernel config. -CONFIG_X86_UP_APIC is for uniprocessor machines without an IO-APIC. -CONFIG_X86_UP_IOAPIC is for uniprocessor with an IO-APIC. [Note: certain -kernel debugging options, such as Kernel Stack Meter or Kernel Tracer, -may implicitly disable the NMI watchdog.] - -For x86-64, the needed APIC is always compiled in. - -Using local APIC (nmi_watchdog=2) needs the first performance register, so -you can't use it for other purposes (such as high precision performance -profiling.) However, at least oprofile and the perfctr driver disable the -local APIC NMI watchdog automatically. - -To actually enable the NMI watchdog, use the 'nmi_watchdog=N' boot -parameter. Eg. the relevant lilo.conf entry: - - append="nmi_watchdog=1" - -For SMP machines and UP machines with an IO-APIC use nmi_watchdog=1. -For UP machines without an IO-APIC use nmi_watchdog=2, this only works -for some processor types. If in doubt, boot with nmi_watchdog=1 and -check the NMI count in /proc/interrupts; if the count is zero then -reboot with nmi_watchdog=2 and check the NMI count. If it is still -zero then log a problem, you probably have a processor that needs to be -added to the nmi code. - -A 'lockup' is the following scenario: if any CPU in the system does not -execute the period local timer interrupt for more than 5 seconds, then -the NMI handler generates an oops and kills the process. This -'controlled crash' (and the resulting kernel messages) can be used to -debug the lockup. Thus whenever the lockup happens, wait 5 seconds and -the oops will show up automatically. If the kernel produces no messages -then the system has crashed so hard (eg. hardware-wise) that either it -cannot even accept NMI interrupts, or the crash has made the kernel -unable to print messages. - -Be aware that when using local APIC, the frequency of NMI interrupts -it generates, depends on the system load. The local APIC NMI watchdog, -lacking a better source, uses the "cycles unhalted" event. As you may -guess it doesn't tick when the CPU is in the halted state (which happens -when the system is idle), but if your system locks up on anything but the -"hlt" processor instruction, the watchdog will trigger very soon as the -"cycles unhalted" event will happen every clock tick. If it locks up on -"hlt", then you are out of luck -- the event will not happen at all and the -watchdog won't trigger. This is a shortcoming of the local APIC watchdog --- unfortunately there is no "clock ticks" event that would work all the -time. The I/O APIC watchdog is driven externally and has no such shortcoming. -But its NMI frequency is much higher, resulting in a more significant hit -to the overall system performance. - -On x86 nmi_watchdog is disabled by default so you have to enable it with -a boot time parameter. - -It's possible to disable the NMI watchdog in run-time by writing "0" to -/proc/sys/kernel/nmi_watchdog. Writing "1" to the same file will re-enable -the NMI watchdog. Notice that you still need to use "nmi_watchdog=" parameter -at boot time. - -NOTE: In kernels prior to 2.4.2-ac18 the NMI-oopser is enabled unconditionally -on x86 SMP boxes. - -[ feel free to send bug reports, suggestions and patches to - Ingo Molnar <[email protected]> or the Linux SMP mailing - list at <[email protected]> ] - diff --git a/arch/x86/include/asm/inat.h b/arch/x86/include/asm/inat.h index 205b063e3e32..74a2e312e8a2 100644 --- a/arch/x86/include/asm/inat.h +++ b/arch/x86/include/asm/inat.h @@ -97,11 +97,12 @@ /* Attribute search APIs */ extern insn_attr_t inat_get_opcode_attribute(insn_byte_t opcode); +extern int inat_get_last_prefix_id(insn_byte_t last_pfx); extern insn_attr_t inat_get_escape_attribute(insn_byte_t opcode, - insn_byte_t last_pfx, + int lpfx_id, insn_attr_t esc_attr); extern insn_attr_t inat_get_group_attribute(insn_byte_t modrm, - insn_byte_t last_pfx, + int lpfx_id, insn_attr_t esc_attr); extern insn_attr_t inat_get_avx_attribute(insn_byte_t opcode, insn_byte_t vex_m, diff --git a/arch/x86/include/asm/insn.h b/arch/x86/include/asm/insn.h index 74df3f1eddfd..48eb30a86062 100644 --- a/arch/x86/include/asm/insn.h +++ b/arch/x86/include/asm/insn.h @@ -96,12 +96,6 @@ struct insn { #define X86_VEX_P(vex) ((vex) & 0x03) /* VEX3 Byte2, VEX2 Byte1 */ #define X86_VEX_M_MAX 0x1f /* VEX3.M Maximum value */ -/* The last prefix is needed for two-byte and three-byte opcodes */ -static inline insn_byte_t insn_last_prefix(struct insn *insn) -{ - return insn->prefixes.bytes[3]; -} - extern void insn_init(struct insn *insn, const void *kaddr, int x86_64); extern void insn_get_prefixes(struct insn *insn); extern void insn_get_opcode(struct insn *insn); @@ -160,6 +154,18 @@ static inline insn_byte_t insn_vex_p_bits(struct insn *insn) return X86_VEX_P(insn->vex_prefix.bytes[2]); } +/* Get the last prefix id from last prefix or VEX prefix */ +static inline int insn_last_prefix_id(struct insn *insn) +{ + if (insn_is_avx(insn)) + return insn_vex_p_bits(insn); /* VEX_p is a SIMD prefix id */ + + if (insn->prefixes.bytes[3]) + return inat_get_last_prefix_id(insn->prefixes.bytes[3]); + + return 0; +} + /* Offset of each field from kaddr */ static inline int insn_offset_rex_prefix(struct insn *insn) { diff --git a/arch/x86/lib/inat.c b/arch/x86/lib/inat.c index 88ad5fbda6e1..c1f01a8e9f65 100644 --- a/arch/x86/lib/inat.c +++ b/arch/x86/lib/inat.c @@ -29,46 +29,46 @@ insn_attr_t inat_get_opcode_attribute(insn_byte_t opcode) return inat_primary_table[opcode]; } -insn_attr_t inat_get_escape_attribute(insn_byte_t opcode, insn_byte_t last_pfx, +int inat_get_last_prefix_id(insn_byte_t last_pfx) +{ + insn_attr_t lpfx_attr; + + lpfx_attr = inat_get_opcode_attribute(last_pfx); + return inat_last_prefix_id(lpfx_attr); +} + +insn_attr_t inat_get_escape_attribute(insn_byte_t opcode, int lpfx_id, insn_attr_t esc_attr) { const insn_attr_t *table; - insn_attr_t lpfx_attr; - int n, m = 0; + int n; n = inat_escape_id(esc_attr); - if (last_pfx) { - lpfx_attr = inat_get_opcode_attribute(last_pfx); - m = inat_last_prefix_id(lpfx_attr); - } + table = inat_escape_tables[n][0]; if (!table) return 0; - if (inat_has_variant(table[opcode]) && m) { - table = inat_escape_tables[n][m]; + if (inat_has_variant(table[opcode]) && lpfx_id) { + table = inat_escape_tables[n][lpfx_id]; if (!table) return 0; } return table[opcode]; } -insn_attr_t inat_get_group_attribute(insn_byte_t modrm, insn_byte_t last_pfx, +insn_attr_t inat_get_group_attribute(insn_byte_t modrm, int lpfx_id, insn_attr_t grp_attr) { const insn_attr_t *table; - insn_attr_t lpfx_attr; - int n, m = 0; + int n; n = inat_group_id(grp_attr); - if (last_pfx) { - lpfx_attr = inat_get_opcode_attribute(last_pfx); - m = inat_last_prefix_id(lpfx_attr); - } + table = inat_group_tables[n][0]; if (!table) return inat_group_common_attribute(grp_attr); - if (inat_has_variant(table[X86_MODRM_REG(modrm)]) && m) { - table = inat_group_tables[n][m]; + if (inat_has_variant(table[X86_MODRM_REG(modrm)]) && lpfx_id) { + table = inat_group_tables[n][lpfx_id]; if (!table) return inat_group_common_attribute(grp_attr); } diff --git a/arch/x86/lib/insn.c b/arch/x86/lib/insn.c index 5a1f9f3e3fbb..25feb1ae71c5 100644 --- a/arch/x86/lib/insn.c +++ b/arch/x86/lib/insn.c @@ -185,7 +185,8 @@ err_out: void insn_get_opcode(struct insn *insn) { struct insn_field *opcode = &insn->opcode; - insn_byte_t op, pfx; + insn_byte_t op; + int pfx_id; if (opcode->got) return; if (!insn->prefixes.got) @@ -212,8 +213,8 @@ void insn_get_opcode(struct insn *insn) /* Get escaped opcode */ op = get_next(insn_byte_t, insn); opcode->bytes[opcode->nbytes++] = op; - pfx = insn_last_prefix(insn); - insn->attr = inat_get_escape_attribute(op, pfx, insn->attr); + pfx_id = insn_last_prefix_id(insn); + insn->attr = inat_get_escape_attribute(op, pfx_id, insn->attr); } if (inat_must_vex(insn->attr)) insn->attr = 0; /* This instruction is bad */ @@ -235,7 +236,7 @@ err_out: void insn_get_modrm(struct insn *insn) { struct insn_field *modrm = &insn->modrm; - insn_byte_t pfx, mod; + insn_byte_t pfx_id, mod; if (modrm->got) return; if (!insn->opcode.got) @@ -246,8 +247,8 @@ void insn_get_modrm(struct insn *insn) modrm->value = mod; modrm->nbytes = 1; if (inat_is_group(insn->attr)) { - pfx = insn_last_prefix(insn); - insn->attr = inat_get_group_attribute(mod, pfx, + pfx_id = insn_last_prefix_id(insn); + insn->attr = inat_get_group_attribute(mod, pfx_id, insn->attr); if (insn_is_avx(insn) && !inat_accept_vex(insn->attr)) insn->attr = 0; /* This is bad */ diff --git a/kernel/watchdog.c b/kernel/watchdog.c index d117262deba3..14bc092fb12c 100644 --- a/kernel/watchdog.c +++ b/kernel/watchdog.c @@ -3,12 +3,9 @@ * * started by Don Zickus, Copyright (C) 2010 Red Hat, Inc. * - * this code detects hard lockups: incidents in where on a CPU - * the kernel does not respond to anything except NMI. - * - * Note: Most of this code is borrowed heavily from softlockup.c, - * so thanks to Ingo for the initial implementation. - * Some chunks also taken from arch/x86/kernel/apic/nmi.c, thanks + * Note: Most of this code is borrowed heavily from the original softlockup + * detector, so thanks to Ingo for the initial implementation. + * Some chunks also taken from the old x86-specific nmi watchdog code, thanks * to those contributors as well. */ @@ -117,9 +114,10 @@ static unsigned long get_sample_period(void) { /* * convert watchdog_thresh from seconds to ns - * the divide by 5 is to give hrtimer 5 chances to - * increment before the hardlockup detector generates - * a warning + * the divide by 5 is to give hrtimer several chances (two + * or three with the current relation between the soft + * and hard thresholds) to increment before the + * hardlockup detector generates a warning */ return get_softlockup_thresh() * (NSEC_PER_SEC / 5); } @@ -336,9 +334,11 @@ static int watchdog(void *unused) set_current_state(TASK_INTERRUPTIBLE); /* - * Run briefly once per second to reset the softlockup timestamp. - * If this gets delayed for more than 60 seconds then the - * debug-printout triggers in watchdog_timer_fn(). + * Run briefly (kicked by the hrtimer callback function) once every + * get_sample_period() seconds (4 seconds by default) to reset the + * softlockup timestamp. If this gets delayed for more than + * 2*watchdog_thresh seconds then the debug-printout triggers in + * watchdog_timer_fn(). */ while (!kthread_should_stop()) { __touch_watchdog(); diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 8745ac7d1f75..9739c0b45e93 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -166,18 +166,21 @@ config LOCKUP_DETECTOR hard and soft lockups. Softlockups are bugs that cause the kernel to loop in kernel - mode for more than 60 seconds, without giving other tasks a + mode for more than 20 seconds, without giving other tasks a chance to run. The current stack trace is displayed upon detection and the system will stay locked up. Hardlockups are bugs that cause the CPU to loop in kernel mode - for more than 60 seconds, without letting other interrupts have a + for more than 10 seconds, without letting other interrupts have a chance to run. The current stack trace is displayed upon detection and the system will stay locked up. The overhead should be minimal. A periodic hrtimer runs to - generate interrupts and kick the watchdog task every 10-12 seconds. - An NMI is generated every 60 seconds or so to check for hardlockups. + generate interrupts and kick the watchdog task every 4 seconds. + An NMI is generated every 10 seconds or so to check for hardlockups. + + The frequency of hrtimer and NMI events and the soft and hard lockup + thresholds can be controlled through the sysctl watchdog_thresh. config HARDLOCKUP_DETECTOR def_bool LOCKUP_DETECTOR && PERF_EVENTS && HAVE_PERF_EVENTS_NMI && \ @@ -189,7 +192,8 @@ config BOOTPARAM_HARDLOCKUP_PANIC help Say Y here to enable the kernel to panic on "hard lockups", which are bugs that cause the kernel to loop in kernel - mode with interrupts disabled for more than 60 seconds. + mode with interrupts disabled for more than 10 seconds (configurable + using the watchdog_thresh sysctl). Say N if unsure. @@ -206,8 +210,8 @@ config BOOTPARAM_SOFTLOCKUP_PANIC help Say Y here to enable the kernel to panic on "soft lockups", which are bugs that cause the kernel to loop in kernel - mode for more than 60 seconds, without giving other tasks a - chance to run. + mode for more than 20 seconds (configurable using the watchdog_thresh + sysctl), without giving other tasks a chance to run. The panic can be used in combination with panic_timeout, to cause the system to reboot automatically after a |