From d7037381d00286aa4beb631c401da761ee564c94 Mon Sep 17 00:00:00 2001 From: Bitao Hu Date: Thu, 11 Apr 2024 15:41:33 +0800 Subject: watchdog/softlockup: Low-overhead detection of interrupt storm The following softlockup is caused by interrupt storm, but it cannot be identified from the call tree. Because the call tree is just a snapshot and doesn't fully capture the behavior of the CPU during the soft lockup. watchdog: BUG: soft lockup - CPU#28 stuck for 23s! [fio:83921] ... Call trace: __do_softirq+0xa0/0x37c __irq_exit_rcu+0x108/0x140 irq_exit+0x14/0x20 __handle_domain_irq+0x84/0xe0 gic_handle_irq+0x80/0x108 el0_irq_naked+0x50/0x58 Therefore, it is necessary to report CPU utilization during the softlockup_threshold period (report once every sample_period, for a total of 5 reportings), like this: watchdog: BUG: soft lockup - CPU#28 stuck for 23s! [fio:83921] CPU#28 Utilization every 4s during lockup: #1: 0% system, 0% softirq, 100% hardirq, 0% idle #2: 0% system, 0% softirq, 100% hardirq, 0% idle #3: 0% system, 0% softirq, 100% hardirq, 0% idle #4: 0% system, 0% softirq, 100% hardirq, 0% idle #5: 0% system, 0% softirq, 100% hardirq, 0% idle ... This is helpful in determining whether an interrupt storm has occurred or in identifying the cause of the softlockup. The criteria for determination are as follows: a. If the hardirq utilization is high, then interrupt storm should be considered and the root cause cannot be determined from the call tree. b. If the softirq utilization is high, then the call might not necessarily point at the root cause. c. If the system utilization is high, then analyzing the root cause from the call tree is possible in most cases. The mechanism requires a considerable amount of global storage space when configured for the maximum number of CPUs. Therefore, adding a SOFTLOCKUP_DETECTOR_INTR_STORM Kconfig knob that defaults to "yes" if the max number of CPUs is <= 128. Signed-off-by: Bitao Hu Signed-off-by: Thomas Gleixner Reviewed-by: Douglas Anderson Reviewed-by: Liu Song Link: https://lore.kernel.org/r/20240411074134.30922-5-yaoma@linux.alibaba.com --- kernel/watchdog.c | 99 ++++++++++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 98 insertions(+), 1 deletion(-) (limited to 'kernel/watchdog.c') diff --git a/kernel/watchdog.c b/kernel/watchdog.c index d7b2125503af..ef8ebd31fdab 100644 --- a/kernel/watchdog.c +++ b/kernel/watchdog.c @@ -16,6 +16,8 @@ #include #include #include +#include +#include #include #include #include @@ -35,6 +37,8 @@ static DEFINE_MUTEX(watchdog_mutex); # define WATCHDOG_HARDLOCKUP_DEFAULT 0 #endif +#define NUM_SAMPLE_PERIODS 5 + unsigned long __read_mostly watchdog_enabled; int __read_mostly watchdog_user_enabled = 1; static int __read_mostly watchdog_hardlockup_user_enabled = WATCHDOG_HARDLOCKUP_DEFAULT; @@ -333,6 +337,96 @@ __setup("watchdog_thresh=", watchdog_thresh_setup); static void __lockup_detector_cleanup(void); +#ifdef CONFIG_SOFTLOCKUP_DETECTOR_INTR_STORM +enum stats_per_group { + STATS_SYSTEM, + STATS_SOFTIRQ, + STATS_HARDIRQ, + STATS_IDLE, + NUM_STATS_PER_GROUP, +}; + +static const enum cpu_usage_stat tracked_stats[NUM_STATS_PER_GROUP] = { + CPUTIME_SYSTEM, + CPUTIME_SOFTIRQ, + CPUTIME_IRQ, + CPUTIME_IDLE, +}; + +static DEFINE_PER_CPU(u16, cpustat_old[NUM_STATS_PER_GROUP]); +static DEFINE_PER_CPU(u8, cpustat_util[NUM_SAMPLE_PERIODS][NUM_STATS_PER_GROUP]); +static DEFINE_PER_CPU(u8, cpustat_tail); + +/* + * We don't need nanosecond resolution. A granularity of 16ms is + * sufficient for our precision, allowing us to use u16 to store + * cpustats, which will roll over roughly every ~1000 seconds. + * 2^24 ~= 16 * 10^6 + */ +static u16 get_16bit_precision(u64 data_ns) +{ + return data_ns >> 24LL; /* 2^24ns ~= 16.8ms */ +} + +static void update_cpustat(void) +{ + int i; + u8 util; + u16 old_stat, new_stat; + struct kernel_cpustat kcpustat; + u64 *cpustat = kcpustat.cpustat; + u8 tail = __this_cpu_read(cpustat_tail); + u16 sample_period_16 = get_16bit_precision(sample_period); + + kcpustat_cpu_fetch(&kcpustat, smp_processor_id()); + + for (i = 0; i < NUM_STATS_PER_GROUP; i++) { + old_stat = __this_cpu_read(cpustat_old[i]); + new_stat = get_16bit_precision(cpustat[tracked_stats[i]]); + util = DIV_ROUND_UP(100 * (new_stat - old_stat), sample_period_16); + __this_cpu_write(cpustat_util[tail][i], util); + __this_cpu_write(cpustat_old[i], new_stat); + } + + __this_cpu_write(cpustat_tail, (tail + 1) % NUM_SAMPLE_PERIODS); +} + +static void print_cpustat(void) +{ + int i, group; + u8 tail = __this_cpu_read(cpustat_tail); + u64 sample_period_second = sample_period; + + do_div(sample_period_second, NSEC_PER_SEC); + + /* + * Outputting the "watchdog" prefix on every line is redundant and not + * concise, and the original alarm information is sufficient for + * positioning in logs, hence here printk() is used instead of pr_crit(). + */ + printk(KERN_CRIT "CPU#%d Utilization every %llus during lockup:\n", + smp_processor_id(), sample_period_second); + + for (i = 0; i < NUM_SAMPLE_PERIODS; i++) { + group = (tail + i) % NUM_SAMPLE_PERIODS; + printk(KERN_CRIT "\t#%d: %3u%% system,\t%3u%% softirq,\t" + "%3u%% hardirq,\t%3u%% idle\n", i + 1, + __this_cpu_read(cpustat_util[group][STATS_SYSTEM]), + __this_cpu_read(cpustat_util[group][STATS_SOFTIRQ]), + __this_cpu_read(cpustat_util[group][STATS_HARDIRQ]), + __this_cpu_read(cpustat_util[group][STATS_IDLE])); + } +} + +static void report_cpu_status(void) +{ + print_cpustat(); +} +#else +static inline void update_cpustat(void) { } +static inline void report_cpu_status(void) { } +#endif + /* * Hard-lockup warnings should be triggered after just a few seconds. Soft- * lockups can have false positives under extreme conditions. So we generally @@ -364,7 +458,7 @@ static void set_sample_period(void) * and hard thresholds) to increment before the * hardlockup detector generates a warning */ - sample_period = get_softlockup_thresh() * ((u64)NSEC_PER_SEC / 5); + sample_period = get_softlockup_thresh() * ((u64)NSEC_PER_SEC / NUM_SAMPLE_PERIODS); watchdog_update_hrtimer_threshold(sample_period); } @@ -504,6 +598,8 @@ static enum hrtimer_restart watchdog_timer_fn(struct hrtimer *hrtimer) */ period_ts = READ_ONCE(*this_cpu_ptr(&watchdog_report_ts)); + update_cpustat(); + /* Reset the interval when touched by known problematic code. */ if (period_ts == SOFTLOCKUP_DELAY_REPORT) { if (unlikely(__this_cpu_read(softlockup_touch_sync))) { @@ -539,6 +635,7 @@ static enum hrtimer_restart watchdog_timer_fn(struct hrtimer *hrtimer) pr_emerg("BUG: soft lockup - CPU#%d stuck for %us! [%s:%d]\n", smp_processor_id(), duration, current->comm, task_pid_nr(current)); + report_cpu_status(); print_modules(); print_irqtrace_events(current); if (regs) -- cgit v1.2.3-59-g8ed1b From e9a9292e2368e9be4a48aae6ff8aafa3433133e6 Mon Sep 17 00:00:00 2001 From: Bitao Hu Date: Thu, 11 Apr 2024 15:41:34 +0800 Subject: watchdog/softlockup: Report the most frequent interrupts When the watchdog determines that the current soft lockup is due to an interrupt storm based on CPU utilization, reporting the most frequent interrupts could be good enough for further troubleshooting. Below is an example of interrupt storm. The call tree does not provide useful information, but analyzing which interrupt caused the soft lockup by comparing the counts of interrupts during the lockup period allows to identify the culprit. [ 638.870231] watchdog: BUG: soft lockup - CPU#9 stuck for 26s! [swapper/9:0] [ 638.870825] CPU#9 Utilization every 4s during lockup: [ 638.871194] #1: 0% system, 0% softirq, 100% hardirq, 0% idle [ 638.871652] #2: 0% system, 0% softirq, 100% hardirq, 0% idle [ 638.872107] #3: 0% system, 0% softirq, 100% hardirq, 0% idle [ 638.872563] #4: 0% system, 0% softirq, 100% hardirq, 0% idle [ 638.873018] #5: 0% system, 0% softirq, 100% hardirq, 0% idle [ 638.873494] CPU#9 Detect HardIRQ Time exceeds 50%. Most frequent HardIRQs: [ 638.873994] #1: 330945 irq#7 [ 638.874236] #2: 31 irq#82 [ 638.874493] #3: 10 irq#10 [ 638.874744] #4: 2 irq#89 [ 638.874992] #5: 1 irq#102 ... [ 638.875313] Call trace: [ 638.875315] __do_softirq+0xa8/0x364 Signed-off-by: Bitao Hu Signed-off-by: Thomas Gleixner Reviewed-by: Liu Song Reviewed-by: Douglas Anderson Link: https://lore.kernel.org/r/20240411074134.30922-6-yaoma@linux.alibaba.com --- kernel/watchdog.c | 116 ++++++++++++++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 112 insertions(+), 4 deletions(-) (limited to 'kernel/watchdog.c') diff --git a/kernel/watchdog.c b/kernel/watchdog.c index ef8ebd31fdab..d12ff74889ed 100644 --- a/kernel/watchdog.c +++ b/kernel/watchdog.c @@ -12,22 +12,25 @@ #define pr_fmt(fmt) "watchdog: " fmt -#include #include -#include #include +#include +#include #include +#include #include +#include #include +#include +#include #include #include + #include #include #include -#include #include -#include static DEFINE_MUTEX(watchdog_mutex); @@ -418,13 +421,105 @@ static void print_cpustat(void) } } +#define HARDIRQ_PERCENT_THRESH 50 +#define NUM_HARDIRQ_REPORT 5 +struct irq_counts { + int irq; + u32 counts; +}; + +static DEFINE_PER_CPU(bool, snapshot_taken); + +/* Tabulate the most frequent interrupts. */ +static void tabulate_irq_count(struct irq_counts *irq_counts, int irq, u32 counts, int rank) +{ + int i; + struct irq_counts new_count = {irq, counts}; + + for (i = 0; i < rank; i++) { + if (counts > irq_counts[i].counts) + swap(new_count, irq_counts[i]); + } +} + +/* + * If the hardirq time exceeds HARDIRQ_PERCENT_THRESH% of the sample_period, + * then the cause of softlockup might be interrupt storm. In this case, it + * would be useful to start interrupt counting. + */ +static bool need_counting_irqs(void) +{ + u8 util; + int tail = __this_cpu_read(cpustat_tail); + + tail = (tail + NUM_HARDIRQ_REPORT - 1) % NUM_HARDIRQ_REPORT; + util = __this_cpu_read(cpustat_util[tail][STATS_HARDIRQ]); + return util > HARDIRQ_PERCENT_THRESH; +} + +static void start_counting_irqs(void) +{ + if (!__this_cpu_read(snapshot_taken)) { + kstat_snapshot_irqs(); + __this_cpu_write(snapshot_taken, true); + } +} + +static void stop_counting_irqs(void) +{ + __this_cpu_write(snapshot_taken, false); +} + +static void print_irq_counts(void) +{ + unsigned int i, count; + struct irq_counts irq_counts_sorted[NUM_HARDIRQ_REPORT] = { + {-1, 0}, {-1, 0}, {-1, 0}, {-1, 0}, {-1, 0} + }; + + if (__this_cpu_read(snapshot_taken)) { + for_each_active_irq(i) { + count = kstat_get_irq_since_snapshot(i); + tabulate_irq_count(irq_counts_sorted, i, count, NUM_HARDIRQ_REPORT); + } + + /* + * Outputting the "watchdog" prefix on every line is redundant and not + * concise, and the original alarm information is sufficient for + * positioning in logs, hence here printk() is used instead of pr_crit(). + */ + printk(KERN_CRIT "CPU#%d Detect HardIRQ Time exceeds %d%%. Most frequent HardIRQs:\n", + smp_processor_id(), HARDIRQ_PERCENT_THRESH); + + for (i = 0; i < NUM_HARDIRQ_REPORT; i++) { + if (irq_counts_sorted[i].irq == -1) + break; + + printk(KERN_CRIT "\t#%u: %-10u\tirq#%d\n", + i + 1, irq_counts_sorted[i].counts, + irq_counts_sorted[i].irq); + } + + /* + * If the hardirq time is less than HARDIRQ_PERCENT_THRESH% in the last + * sample_period, then we suspect the interrupt storm might be subsiding. + */ + if (!need_counting_irqs()) + stop_counting_irqs(); + } +} + static void report_cpu_status(void) { print_cpustat(); + print_irq_counts(); } #else static inline void update_cpustat(void) { } static inline void report_cpu_status(void) { } +static inline bool need_counting_irqs(void) { return false; } +static inline void start_counting_irqs(void) { } +static inline void stop_counting_irqs(void) { } #endif /* @@ -528,6 +623,18 @@ static int is_softlockup(unsigned long touch_ts, unsigned long now) { if ((watchdog_enabled & WATCHDOG_SOFTOCKUP_ENABLED) && watchdog_thresh) { + /* + * If period_ts has not been updated during a sample_period, then + * in the subsequent few sample_periods, period_ts might also not + * be updated, which could indicate a potential softlockup. In + * this case, if we suspect the cause of the potential softlockup + * might be interrupt storm, then we need to count the interrupts + * to find which interrupt is storming. + */ + if (time_after_eq(now, period_ts + get_softlockup_thresh() / NUM_SAMPLE_PERIODS) && + need_counting_irqs()) + start_counting_irqs(); + /* Warn about unreasonable delays. */ if (time_after(now, period_ts + get_softlockup_thresh())) return now - touch_ts; @@ -550,6 +657,7 @@ static DEFINE_PER_CPU(struct cpu_stop_work, softlockup_stop_work); static int softlockup_fn(void *data) { update_touch_ts(); + stop_counting_irqs(); complete(this_cpu_ptr(&softlockup_completion)); return 0; -- cgit v1.2.3-59-g8ed1b From 11a921909fea230cf7afcd6842a9452f3720b61b Mon Sep 17 00:00:00 2001 From: Joel Granados Date: Tue, 27 Jun 2023 15:30:19 +0200 Subject: kernel misc: Remove the now superfluous sentinel elements from ctl_table array This commit comes at the tail end of a greater effort to remove the empty elements at the end of the ctl_table arrays (sentinels) which will reduce the overall build time size of the kernel and run time memory bloat by ~64 bytes per sentinel (further information Link : https://lore.kernel.org/all/ZO5Yx5JFogGi%2FcBo@bombadil.infradead.org/) Remove the sentinel from ctl_table arrays. Reduce by one the values used to compare the size of the adjusted arrays. Signed-off-by: Joel Granados --- kernel/acct.c | 1 - kernel/exit.c | 1 - kernel/hung_task.c | 1 - kernel/kexec_core.c | 1 - kernel/latencytop.c | 1 - kernel/panic.c | 1 - kernel/pid_namespace.c | 1 - kernel/pid_sysctl.h | 1 - kernel/reboot.c | 1 - kernel/signal.c | 1 - kernel/stackleak.c | 1 - kernel/sysctl.c | 2 -- kernel/ucount.c | 3 +-- kernel/utsname_sysctl.c | 1 - kernel/watchdog.c | 2 -- 15 files changed, 1 insertion(+), 18 deletions(-) (limited to 'kernel/watchdog.c') diff --git a/kernel/acct.c b/kernel/acct.c index 986c8214dabf..179848ad33e9 100644 --- a/kernel/acct.c +++ b/kernel/acct.c @@ -84,7 +84,6 @@ static struct ctl_table kern_acct_table[] = { .mode = 0644, .proc_handler = proc_dointvec, }, - { } }; static __init int kernel_acct_sysctls_init(void) diff --git a/kernel/exit.c b/kernel/exit.c index 41a12630cbbc..cd3aa9042f1a 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -94,7 +94,6 @@ static struct ctl_table kern_exit_table[] = { .mode = 0644, .proc_handler = proc_douintvec, }, - { } }; static __init int kernel_exit_sysctls_init(void) diff --git a/kernel/hung_task.c b/kernel/hung_task.c index b2fc2727d654..1d92016b0b3c 100644 --- a/kernel/hung_task.c +++ b/kernel/hung_task.c @@ -314,7 +314,6 @@ static struct ctl_table hung_task_sysctls[] = { .proc_handler = proc_dointvec_minmax, .extra1 = SYSCTL_NEG_ONE, }, - {} }; static void __init hung_task_sysctl_init(void) diff --git a/kernel/kexec_core.c b/kernel/kexec_core.c index 0e96f6b24344..9112d69d68b0 100644 --- a/kernel/kexec_core.c +++ b/kernel/kexec_core.c @@ -948,7 +948,6 @@ static struct ctl_table kexec_core_sysctls[] = { .mode = 0644, .proc_handler = kexec_limit_handler, }, - { } }; static int __init kexec_core_sysctl_init(void) diff --git a/kernel/latencytop.c b/kernel/latencytop.c index 781249098cb6..84c53285f499 100644 --- a/kernel/latencytop.c +++ b/kernel/latencytop.c @@ -85,7 +85,6 @@ static struct ctl_table latencytop_sysctl[] = { .mode = 0644, .proc_handler = sysctl_latencytop, }, - {} }; #endif diff --git a/kernel/panic.c b/kernel/panic.c index 747c3f3d289a..8bff183d6180 100644 --- a/kernel/panic.c +++ b/kernel/panic.c @@ -100,7 +100,6 @@ static struct ctl_table kern_panic_table[] = { .mode = 0644, .proc_handler = proc_douintvec, }, - { } }; static __init int kernel_panic_sysctls_init(void) diff --git a/kernel/pid_namespace.c b/kernel/pid_namespace.c index 7ade20e95232..dc48fecfa1dc 100644 --- a/kernel/pid_namespace.c +++ b/kernel/pid_namespace.c @@ -307,7 +307,6 @@ static struct ctl_table pid_ns_ctl_table[] = { .extra1 = SYSCTL_ZERO, .extra2 = &pid_max, }, - { } }; #endif /* CONFIG_CHECKPOINT_RESTORE */ diff --git a/kernel/pid_sysctl.h b/kernel/pid_sysctl.h index 2ee41a3a1dfd..fe9fb991dc42 100644 --- a/kernel/pid_sysctl.h +++ b/kernel/pid_sysctl.h @@ -41,7 +41,6 @@ static struct ctl_table pid_ns_ctl_table_vm[] = { .extra1 = SYSCTL_ZERO, .extra2 = SYSCTL_TWO, }, - { } }; static inline void register_pid_ns_sysctl_table_vm(void) { diff --git a/kernel/reboot.c b/kernel/reboot.c index 22c16e2564cc..f05dbde2c93f 100644 --- a/kernel/reboot.c +++ b/kernel/reboot.c @@ -1295,7 +1295,6 @@ static struct ctl_table kern_reboot_table[] = { .mode = 0644, .proc_handler = proc_dointvec, }, - { } }; static void __init kernel_reboot_sysctls_init(void) diff --git a/kernel/signal.c b/kernel/signal.c index 7bdbcf1b78d0..01c4c46a51a8 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -4840,7 +4840,6 @@ static struct ctl_table signal_debug_table[] = { .proc_handler = proc_dointvec }, #endif - { } }; static int __init init_signal_sysctls(void) diff --git a/kernel/stackleak.c b/kernel/stackleak.c index 34c9d81eea94..d099f3affcf1 100644 --- a/kernel/stackleak.c +++ b/kernel/stackleak.c @@ -54,7 +54,6 @@ static struct ctl_table stackleak_sysctls[] = { .extra1 = SYSCTL_ZERO, .extra2 = SYSCTL_ONE, }, - {} }; static int __init stackleak_sysctls_init(void) diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 81cc974913bb..e0b917328cf9 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -2034,7 +2034,6 @@ static struct ctl_table kern_table[] = { .extra2 = SYSCTL_INT_MAX, }, #endif - { } }; static struct ctl_table vm_table[] = { @@ -2240,7 +2239,6 @@ static struct ctl_table vm_table[] = { .extra2 = (void *)&mmap_rnd_compat_bits_max, }, #endif - { } }; int __init sysctl_init_bases(void) diff --git a/kernel/ucount.c b/kernel/ucount.c index 4aa6166cb856..e196da0204dc 100644 --- a/kernel/ucount.c +++ b/kernel/ucount.c @@ -87,7 +87,6 @@ static struct ctl_table user_table[] = { UCOUNT_ENTRY("max_fanotify_groups"), UCOUNT_ENTRY("max_fanotify_marks"), #endif - { } }; #endif /* CONFIG_SYSCTL */ @@ -96,7 +95,7 @@ bool setup_userns_sysctls(struct user_namespace *ns) #ifdef CONFIG_SYSCTL struct ctl_table *tbl; - BUILD_BUG_ON(ARRAY_SIZE(user_table) != UCOUNT_COUNTS + 1); + BUILD_BUG_ON(ARRAY_SIZE(user_table) != UCOUNT_COUNTS); setup_sysctl_set(&ns->set, &set_root, set_is_seen); tbl = kmemdup(user_table, sizeof(user_table), GFP_KERNEL); if (tbl) { diff --git a/kernel/utsname_sysctl.c b/kernel/utsname_sysctl.c index 019e3a1566cf..76a772072557 100644 --- a/kernel/utsname_sysctl.c +++ b/kernel/utsname_sysctl.c @@ -120,7 +120,6 @@ static struct ctl_table uts_kern_table[] = { .proc_handler = proc_do_uts_string, .poll = &domainname_poll, }, - {} }; #ifdef CONFIG_PROC_SYSCTL diff --git a/kernel/watchdog.c b/kernel/watchdog.c index d7b2125503af..4e472d416525 100644 --- a/kernel/watchdog.c +++ b/kernel/watchdog.c @@ -950,7 +950,6 @@ static struct ctl_table watchdog_sysctls[] = { }, #endif /* CONFIG_SMP */ #endif - {} }; static struct ctl_table watchdog_hardlockup_sysctl[] = { @@ -963,7 +962,6 @@ static struct ctl_table watchdog_hardlockup_sysctl[] = { .extra1 = SYSCTL_ZERO, .extra2 = SYSCTL_ONE, }, - {} }; static void __init watchdog_sysctl_init(void) -- cgit v1.2.3-59-g8ed1b From 602ba77361160d99c77e3dadf7d891364f8c71b3 Mon Sep 17 00:00:00 2001 From: Song Liu Date: Mon, 29 Apr 2024 23:02:35 -0700 Subject: watchdog: handle comma separated nmi_watchdog command line Per the document, the kernel can accept comma separated command line like nmi_watchdog=nopanic,0. However, the code doesn't really handle it. Fix the kernel to handle it properly. Link: https://lkml.kernel.org/r/20240430060236.1878002-1-song@kernel.org Signed-off-by: Song Liu Cc: Peter Zijlstra Signed-off-by: Andrew Morton --- kernel/watchdog.c | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'kernel/watchdog.c') diff --git a/kernel/watchdog.c b/kernel/watchdog.c index d7b2125503af..7f54484de16f 100644 --- a/kernel/watchdog.c +++ b/kernel/watchdog.c @@ -71,6 +71,7 @@ void __init hardlockup_detector_disable(void) static int __init hardlockup_panic_setup(char *str) { +next: if (!strncmp(str, "panic", 5)) hardlockup_panic = 1; else if (!strncmp(str, "nopanic", 7)) @@ -79,6 +80,12 @@ static int __init hardlockup_panic_setup(char *str) watchdog_hardlockup_user_enabled = 0; else if (!strncmp(str, "1", 1)) watchdog_hardlockup_user_enabled = 1; + while (*(str++)) { + if (*str == ',') { + str++; + goto next; + } + } return 1; } __setup("nmi_watchdog=", hardlockup_panic_setup); -- cgit v1.2.3-59-g8ed1b From 393fb313a2e150b768e4850658679e2afff431e9 Mon Sep 17 00:00:00 2001 From: Song Liu Date: Mon, 29 Apr 2024 23:02:36 -0700 Subject: watchdog: allow nmi watchdog to use raw perf event NMI watchdog permanently consumes one hardware counters per CPU on the system. For systems that use many hardware counters, this causes more aggressive time multiplexing of perf events. OTOH, some CPUs (mostly Intel) support "ref-cycles" event, which is rarely used. Add kernel cmdline arg nmi_watchdog=rNNN to configure the watchdog to use raw event. For example, on Intel CPUs, we can use "r300" to configure the watchdog to use ref-cycles event. If the raw event does not work, fall back to use "cycles". [akpm@linux-foundation.org: fix kerneldoc] Link: https://lkml.kernel.org/r/20240430060236.1878002-2-song@kernel.org Signed-off-by: Song Liu Cc: Peter Zijlstra Cc: "Matthew Wilcox (Oracle)" Signed-off-by: Andrew Morton --- Documentation/admin-guide/kernel-parameters.txt | 5 +-- include/linux/nmi.h | 2 ++ kernel/watchdog.c | 2 ++ kernel/watchdog_perf.c | 46 +++++++++++++++++++++++++ 4 files changed, 53 insertions(+), 2 deletions(-) (limited to 'kernel/watchdog.c') diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt index 902ecd92a29f..1fa79a3d0d1a 100644 --- a/Documentation/admin-guide/kernel-parameters.txt +++ b/Documentation/admin-guide/kernel-parameters.txt @@ -3773,10 +3773,12 @@ Format: [state][,regs][,debounce][,die] nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels - Format: [panic,][nopanic,][num] + Format: [panic,][nopanic,][rNNN,][num] Valid num: 0 or 1 0 - turn hardlockup detector in nmi_watchdog off 1 - turn hardlockup detector in nmi_watchdog on + rNNN - configure the watchdog with raw perf event 0xNNN + When panic is specified, panic when an NMI watchdog timeout occurs (or 'nopanic' to not panic on an NMI watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set) @@ -7464,4 +7466,3 @@ memory, and other data can't be written using xmon commands. off xmon is disabled. - diff --git a/include/linux/nmi.h b/include/linux/nmi.h index f53438eae815..a8dfb38c9bb6 100644 --- a/include/linux/nmi.h +++ b/include/linux/nmi.h @@ -105,10 +105,12 @@ void watchdog_hardlockup_check(unsigned int cpu, struct pt_regs *regs); extern void hardlockup_detector_perf_stop(void); extern void hardlockup_detector_perf_restart(void); extern void hardlockup_detector_perf_cleanup(void); +extern void hardlockup_config_perf_event(const char *str); #else static inline void hardlockup_detector_perf_stop(void) { } static inline void hardlockup_detector_perf_restart(void) { } static inline void hardlockup_detector_perf_cleanup(void) { } +static inline void hardlockup_config_perf_event(const char *str) { } #endif void watchdog_hardlockup_stop(void); diff --git a/kernel/watchdog.c b/kernel/watchdog.c index 7f54484de16f..ab0129b15f25 100644 --- a/kernel/watchdog.c +++ b/kernel/watchdog.c @@ -80,6 +80,8 @@ next: watchdog_hardlockup_user_enabled = 0; else if (!strncmp(str, "1", 1)) watchdog_hardlockup_user_enabled = 1; + else if (!strncmp(str, "r", 1)) + hardlockup_config_perf_event(str + 1); while (*(str++)) { if (*str == ',') { str++; diff --git a/kernel/watchdog_perf.c b/kernel/watchdog_perf.c index 8ea00c4a24b2..5f7d1f0d4268 100644 --- a/kernel/watchdog_perf.c +++ b/kernel/watchdog_perf.c @@ -90,6 +90,14 @@ static struct perf_event_attr wd_hw_attr = { .disabled = 1, }; +static struct perf_event_attr fallback_wd_hw_attr = { + .type = PERF_TYPE_HARDWARE, + .config = PERF_COUNT_HW_CPU_CYCLES, + .size = sizeof(struct perf_event_attr), + .pinned = 1, + .disabled = 1, +}; + /* Callback function for perf event subsystem */ static void watchdog_overflow_callback(struct perf_event *event, struct perf_sample_data *data, @@ -122,6 +130,13 @@ static int hardlockup_detector_event_create(void) /* Try to register using hardware perf events */ evt = perf_event_create_kernel_counter(wd_attr, cpu, NULL, watchdog_overflow_callback, NULL); + if (IS_ERR(evt)) { + wd_attr = &fallback_wd_hw_attr; + wd_attr->sample_period = hw_nmi_get_sample_period(watchdog_thresh); + evt = perf_event_create_kernel_counter(wd_attr, cpu, NULL, + watchdog_overflow_callback, NULL); + } + if (IS_ERR(evt)) { pr_debug("Perf event create on CPU %d failed with %ld\n", cpu, PTR_ERR(evt)); @@ -259,3 +274,34 @@ int __init watchdog_hardlockup_probe(void) } return ret; } + +/** + * hardlockup_config_perf_event - Overwrite config of wd_hw_attr. + * + * @str: number which identifies the raw perf event to use + */ +void __init hardlockup_config_perf_event(const char *str) +{ + u64 config; + char buf[24]; + char *comma = strchr(str, ','); + + if (!comma) { + if (kstrtoull(str, 16, &config)) + return; + } else { + unsigned int len = comma - str; + + if (len >= sizeof(buf)) + return; + + if (strscpy(buf, str, sizeof(buf)) < 0) + return; + buf[len] = 0; + if (kstrtoull(buf, 16, &config)) + return; + } + + wd_hw_attr.type = PERF_TYPE_RAW; + wd_hw_attr.config = config; +} -- cgit v1.2.3-59-g8ed1b