^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1) // SPDX-License-Identifier: GPL-2.0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3) * Watchdog support on powerpc systems.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 5) * Copyright 2017, IBM Corporation.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 6) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 7) * This uses code from arch/sparc/kernel/nmi.c and kernel/watchdog.c
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 8) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 9)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 10) #define pr_fmt(fmt) "watchdog: " fmt
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 11)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 12) #include <linux/kernel.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 13) #include <linux/param.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 14) #include <linux/init.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 15) #include <linux/percpu.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 16) #include <linux/cpu.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 17) #include <linux/nmi.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 18) #include <linux/module.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 19) #include <linux/export.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 20) #include <linux/kprobes.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 21) #include <linux/hardirq.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 22) #include <linux/reboot.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 23) #include <linux/slab.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 24) #include <linux/kdebug.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 25) #include <linux/sched/debug.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 26) #include <linux/delay.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 27) #include <linux/smp.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 28)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 29) #include <asm/paca.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 30)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 31) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 32) * The powerpc watchdog ensures that each CPU is able to service timers.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 33) * The watchdog sets up a simple timer on each CPU to run once per timer
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 34) * period, and updates a per-cpu timestamp and a "pending" cpumask. This is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 35) * the heartbeat.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 36) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 37) * Then there are two systems to check that the heartbeat is still running.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 38) * The local soft-NMI, and the SMP checker.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 39) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 40) * The soft-NMI checker can detect lockups on the local CPU. When interrupts
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 41) * are disabled with local_irq_disable(), platforms that use soft-masking
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 42) * can leave hardware interrupts enabled and handle them with a masked
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 43) * interrupt handler. The masked handler can send the timer interrupt to the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 44) * watchdog's soft_nmi_interrupt(), which appears to Linux as an NMI
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 45) * interrupt, and can be used to detect CPUs stuck with IRQs disabled.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 46) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 47) * The soft-NMI checker will compare the heartbeat timestamp for this CPU
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 48) * with the current time, and take action if the difference exceeds the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 49) * watchdog threshold.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 50) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 51) * The limitation of the soft-NMI watchdog is that it does not work when
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 52) * interrupts are hard disabled or otherwise not being serviced. This is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 53) * solved by also having a SMP watchdog where all CPUs check all other
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 54) * CPUs heartbeat.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 55) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 56) * The SMP checker can detect lockups on other CPUs. A gobal "pending"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 57) * cpumask is kept, containing all CPUs which enable the watchdog. Each
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 58) * CPU clears their pending bit in their heartbeat timer. When the bitmask
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 59) * becomes empty, the last CPU to clear its pending bit updates a global
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 60) * timestamp and refills the pending bitmask.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 61) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 62) * In the heartbeat timer, if any CPU notices that the global timestamp has
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 63) * not been updated for a period exceeding the watchdog threshold, then it
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 64) * means the CPU(s) with their bit still set in the pending mask have had
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 65) * their heartbeat stop, and action is taken.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 66) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 67) * Some platforms implement true NMI IPIs, which can be used by the SMP
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 68) * watchdog to detect an unresponsive CPU and pull it out of its stuck
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 69) * state with the NMI IPI, to get crash/debug data from it. This way the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 70) * SMP watchdog can detect hardware interrupts off lockups.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 71) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 72)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 73) static cpumask_t wd_cpus_enabled __read_mostly;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 74)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 75) static u64 wd_panic_timeout_tb __read_mostly; /* timebase ticks until panic */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 76) static u64 wd_smp_panic_timeout_tb __read_mostly; /* panic other CPUs */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 77)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 78) static u64 wd_timer_period_ms __read_mostly; /* interval between heartbeat */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 79)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 80) static DEFINE_PER_CPU(struct hrtimer, wd_hrtimer);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 81) static DEFINE_PER_CPU(u64, wd_timer_tb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 82)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 83) /* SMP checker bits */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 84) static unsigned long __wd_smp_lock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 85) static cpumask_t wd_smp_cpus_pending;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 86) static cpumask_t wd_smp_cpus_stuck;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 87) static u64 wd_smp_last_reset_tb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 88)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 89) static inline void wd_smp_lock(unsigned long *flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 90) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 91) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 92) * Avoid locking layers if possible.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 93) * This may be called from low level interrupt handlers at some
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 94) * point in future.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 95) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 96) raw_local_irq_save(*flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 97) hard_irq_disable(); /* Make it soft-NMI safe */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 98) while (unlikely(test_and_set_bit_lock(0, &__wd_smp_lock))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 99) raw_local_irq_restore(*flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 100) spin_until_cond(!test_bit(0, &__wd_smp_lock));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 101) raw_local_irq_save(*flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 102) hard_irq_disable();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 103) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 104) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 105)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 106) static inline void wd_smp_unlock(unsigned long *flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 107) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 108) clear_bit_unlock(0, &__wd_smp_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 109) raw_local_irq_restore(*flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 110) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 111)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 112) static void wd_lockup_ipi(struct pt_regs *regs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 113) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 114) int cpu = raw_smp_processor_id();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 115) u64 tb = get_tb();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 116)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 117) pr_emerg("CPU %d Hard LOCKUP\n", cpu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 118) pr_emerg("CPU %d TB:%lld, last heartbeat TB:%lld (%lldms ago)\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 119) cpu, tb, per_cpu(wd_timer_tb, cpu),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 120) tb_to_ns(tb - per_cpu(wd_timer_tb, cpu)) / 1000000);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 121) print_modules();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 122) print_irqtrace_events(current);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 123) if (regs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 124) show_regs(regs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 125) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 126) dump_stack();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 127)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 128) /* Do not panic from here because that can recurse into NMI IPI layer */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 129) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 130)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 131) static void set_cpumask_stuck(const struct cpumask *cpumask, u64 tb)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 132) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 133) cpumask_or(&wd_smp_cpus_stuck, &wd_smp_cpus_stuck, cpumask);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 134) cpumask_andnot(&wd_smp_cpus_pending, &wd_smp_cpus_pending, cpumask);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 135) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 136) * See wd_smp_clear_cpu_pending()
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 137) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 138) smp_mb();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 139) if (cpumask_empty(&wd_smp_cpus_pending)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 140) wd_smp_last_reset_tb = tb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 141) cpumask_andnot(&wd_smp_cpus_pending,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 142) &wd_cpus_enabled,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 143) &wd_smp_cpus_stuck);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 144) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 145) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 146) static void set_cpu_stuck(int cpu, u64 tb)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 147) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 148) set_cpumask_stuck(cpumask_of(cpu), tb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 149) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 150)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 151) static void watchdog_smp_panic(int cpu, u64 tb)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 152) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 153) unsigned long flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 154) int c;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 155)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 156) wd_smp_lock(&flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 157) /* Double check some things under lock */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 158) if ((s64)(tb - wd_smp_last_reset_tb) < (s64)wd_smp_panic_timeout_tb)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 159) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 160) if (cpumask_test_cpu(cpu, &wd_smp_cpus_pending))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 161) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 162) if (cpumask_weight(&wd_smp_cpus_pending) == 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 163) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 164)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 165) pr_emerg("CPU %d detected hard LOCKUP on other CPUs %*pbl\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 166) cpu, cpumask_pr_args(&wd_smp_cpus_pending));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 167) pr_emerg("CPU %d TB:%lld, last SMP heartbeat TB:%lld (%lldms ago)\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 168) cpu, tb, wd_smp_last_reset_tb,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 169) tb_to_ns(tb - wd_smp_last_reset_tb) / 1000000);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 170)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 171) if (!sysctl_hardlockup_all_cpu_backtrace) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 172) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 173) * Try to trigger the stuck CPUs, unless we are going to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 174) * get a backtrace on all of them anyway.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 175) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 176) for_each_cpu(c, &wd_smp_cpus_pending) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 177) if (c == cpu)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 178) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 179) smp_send_nmi_ipi(c, wd_lockup_ipi, 1000000);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 180) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 181) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 182)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 183) /* Take the stuck CPUs out of the watch group */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 184) set_cpumask_stuck(&wd_smp_cpus_pending, tb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 185)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 186) wd_smp_unlock(&flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 187)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 188) printk_safe_flush();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 189) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 190) * printk_safe_flush() seems to require another print
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 191) * before anything actually goes out to console.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 192) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 193) if (sysctl_hardlockup_all_cpu_backtrace)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 194) trigger_allbutself_cpu_backtrace();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 195)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 196) if (hardlockup_panic)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 197) nmi_panic(NULL, "Hard LOCKUP");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 198)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 199) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 200)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 201) out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 202) wd_smp_unlock(&flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 203) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 204)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 205) static void wd_smp_clear_cpu_pending(int cpu, u64 tb)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 206) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 207) if (!cpumask_test_cpu(cpu, &wd_smp_cpus_pending)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 208) if (unlikely(cpumask_test_cpu(cpu, &wd_smp_cpus_stuck))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 209) struct pt_regs *regs = get_irq_regs();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 210) unsigned long flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 211)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 212) wd_smp_lock(&flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 213)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 214) pr_emerg("CPU %d became unstuck TB:%lld\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 215) cpu, tb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 216) print_irqtrace_events(current);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 217) if (regs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 218) show_regs(regs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 219) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 220) dump_stack();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 221)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 222) cpumask_clear_cpu(cpu, &wd_smp_cpus_stuck);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 223) wd_smp_unlock(&flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 224) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 225) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 226) * The last CPU to clear pending should have reset the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 227) * watchdog so we generally should not find it empty
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 228) * here if our CPU was clear. However it could happen
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 229) * due to a rare race with another CPU taking the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 230) * last CPU out of the mask concurrently.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 231) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 232) * We can't add a warning for it. But just in case
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 233) * there is a problem with the watchdog that is causing
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 234) * the mask to not be reset, try to kick it along here.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 235) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 236) if (unlikely(cpumask_empty(&wd_smp_cpus_pending)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 237) goto none_pending;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 238) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 239) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 240) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 241)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 242) cpumask_clear_cpu(cpu, &wd_smp_cpus_pending);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 243)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 244) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 245) * Order the store to clear pending with the load(s) to check all
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 246) * words in the pending mask to check they are all empty. This orders
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 247) * with the same barrier on another CPU. This prevents two CPUs
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 248) * clearing the last 2 pending bits, but neither seeing the other's
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 249) * store when checking if the mask is empty, and missing an empty
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 250) * mask, which ends with a false positive.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 251) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 252) smp_mb();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 253) if (cpumask_empty(&wd_smp_cpus_pending)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 254) unsigned long flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 255)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 256) none_pending:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 257) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 258) * Double check under lock because more than one CPU could see
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 259) * a clear mask with the lockless check after clearing their
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 260) * pending bits.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 261) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 262) wd_smp_lock(&flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 263) if (cpumask_empty(&wd_smp_cpus_pending)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 264) wd_smp_last_reset_tb = tb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 265) cpumask_andnot(&wd_smp_cpus_pending,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 266) &wd_cpus_enabled,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 267) &wd_smp_cpus_stuck);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 268) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 269) wd_smp_unlock(&flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 270) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 271) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 272)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 273) static void watchdog_timer_interrupt(int cpu)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 274) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 275) u64 tb = get_tb();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 276)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 277) per_cpu(wd_timer_tb, cpu) = tb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 278)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 279) wd_smp_clear_cpu_pending(cpu, tb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 280)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 281) if ((s64)(tb - wd_smp_last_reset_tb) >= (s64)wd_smp_panic_timeout_tb)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 282) watchdog_smp_panic(cpu, tb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 283) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 284)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 285) void soft_nmi_interrupt(struct pt_regs *regs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 286) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 287) unsigned long flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 288) int cpu = raw_smp_processor_id();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 289) u64 tb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 290)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 291) if (!cpumask_test_cpu(cpu, &wd_cpus_enabled))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 292) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 293)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 294) nmi_enter();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 295)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 296) __this_cpu_inc(irq_stat.soft_nmi_irqs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 297)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 298) tb = get_tb();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 299) if (tb - per_cpu(wd_timer_tb, cpu) >= wd_panic_timeout_tb) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 300) wd_smp_lock(&flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 301) if (cpumask_test_cpu(cpu, &wd_smp_cpus_stuck)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 302) wd_smp_unlock(&flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 303) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 304) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 305) set_cpu_stuck(cpu, tb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 306)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 307) pr_emerg("CPU %d self-detected hard LOCKUP @ %pS\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 308) cpu, (void *)regs->nip);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 309) pr_emerg("CPU %d TB:%lld, last heartbeat TB:%lld (%lldms ago)\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 310) cpu, tb, per_cpu(wd_timer_tb, cpu),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 311) tb_to_ns(tb - per_cpu(wd_timer_tb, cpu)) / 1000000);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 312) print_modules();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 313) print_irqtrace_events(current);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 314) show_regs(regs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 315)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 316) wd_smp_unlock(&flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 317)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 318) if (sysctl_hardlockup_all_cpu_backtrace)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 319) trigger_allbutself_cpu_backtrace();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 320)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 321) if (hardlockup_panic)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 322) nmi_panic(regs, "Hard LOCKUP");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 323) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 324) if (wd_panic_timeout_tb < 0x7fffffff)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 325) mtspr(SPRN_DEC, wd_panic_timeout_tb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 326)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 327) out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 328) nmi_exit();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 329) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 330)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 331) static enum hrtimer_restart watchdog_timer_fn(struct hrtimer *hrtimer)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 332) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 333) int cpu = smp_processor_id();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 334)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 335) if (!(watchdog_enabled & NMI_WATCHDOG_ENABLED))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 336) return HRTIMER_NORESTART;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 337)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 338) if (!cpumask_test_cpu(cpu, &watchdog_cpumask))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 339) return HRTIMER_NORESTART;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 340)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 341) watchdog_timer_interrupt(cpu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 342)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 343) hrtimer_forward_now(hrtimer, ms_to_ktime(wd_timer_period_ms));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 344)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 345) return HRTIMER_RESTART;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 346) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 347)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 348) void arch_touch_nmi_watchdog(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 349) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 350) unsigned long ticks = tb_ticks_per_usec * wd_timer_period_ms * 1000;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 351) int cpu = smp_processor_id();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 352) u64 tb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 353)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 354) if (!cpumask_test_cpu(cpu, &watchdog_cpumask))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 355) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 356)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 357) tb = get_tb();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 358) if (tb - per_cpu(wd_timer_tb, cpu) >= ticks) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 359) per_cpu(wd_timer_tb, cpu) = tb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 360) wd_smp_clear_cpu_pending(cpu, tb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 361) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 362) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 363) EXPORT_SYMBOL(arch_touch_nmi_watchdog);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 364)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 365) static void start_watchdog(void *arg)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 366) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 367) struct hrtimer *hrtimer = this_cpu_ptr(&wd_hrtimer);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 368) int cpu = smp_processor_id();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 369) unsigned long flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 370)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 371) if (cpumask_test_cpu(cpu, &wd_cpus_enabled)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 372) WARN_ON(1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 373) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 374) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 375)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 376) if (!(watchdog_enabled & NMI_WATCHDOG_ENABLED))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 377) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 378)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 379) if (!cpumask_test_cpu(cpu, &watchdog_cpumask))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 380) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 381)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 382) wd_smp_lock(&flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 383) cpumask_set_cpu(cpu, &wd_cpus_enabled);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 384) if (cpumask_weight(&wd_cpus_enabled) == 1) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 385) cpumask_set_cpu(cpu, &wd_smp_cpus_pending);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 386) wd_smp_last_reset_tb = get_tb();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 387) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 388) wd_smp_unlock(&flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 389)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 390) *this_cpu_ptr(&wd_timer_tb) = get_tb();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 391)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 392) hrtimer_init(hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 393) hrtimer->function = watchdog_timer_fn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 394) hrtimer_start(hrtimer, ms_to_ktime(wd_timer_period_ms),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 395) HRTIMER_MODE_REL_PINNED);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 396) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 397)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 398) static int start_watchdog_on_cpu(unsigned int cpu)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 399) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 400) return smp_call_function_single(cpu, start_watchdog, NULL, true);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 401) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 402)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 403) static void stop_watchdog(void *arg)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 404) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 405) struct hrtimer *hrtimer = this_cpu_ptr(&wd_hrtimer);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 406) int cpu = smp_processor_id();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 407) unsigned long flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 408)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 409) if (!cpumask_test_cpu(cpu, &wd_cpus_enabled))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 410) return; /* Can happen in CPU unplug case */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 411)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 412) hrtimer_cancel(hrtimer);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 413)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 414) wd_smp_lock(&flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 415) cpumask_clear_cpu(cpu, &wd_cpus_enabled);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 416) wd_smp_unlock(&flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 417)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 418) wd_smp_clear_cpu_pending(cpu, get_tb());
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 419) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 420)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 421) static int stop_watchdog_on_cpu(unsigned int cpu)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 422) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 423) return smp_call_function_single(cpu, stop_watchdog, NULL, true);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 424) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 425)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 426) static void watchdog_calc_timeouts(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 427) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 428) wd_panic_timeout_tb = watchdog_thresh * ppc_tb_freq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 429)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 430) /* Have the SMP detector trigger a bit later */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 431) wd_smp_panic_timeout_tb = wd_panic_timeout_tb * 3 / 2;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 432)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 433) /* 2/5 is the factor that the perf based detector uses */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 434) wd_timer_period_ms = watchdog_thresh * 1000 * 2 / 5;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 435) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 436)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 437) void watchdog_nmi_stop(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 438) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 439) int cpu;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 440)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 441) for_each_cpu(cpu, &wd_cpus_enabled)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 442) stop_watchdog_on_cpu(cpu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 443) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 444)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 445) void watchdog_nmi_start(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 446) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 447) int cpu;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 448)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 449) watchdog_calc_timeouts();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 450) for_each_cpu_and(cpu, cpu_online_mask, &watchdog_cpumask)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 451) start_watchdog_on_cpu(cpu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 452) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 453)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 454) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 455) * Invoked from core watchdog init.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 456) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 457) int __init watchdog_nmi_probe(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 458) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 459) int err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 460)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 461) err = cpuhp_setup_state_nocalls(CPUHP_AP_ONLINE_DYN,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 462) "powerpc/watchdog:online",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 463) start_watchdog_on_cpu,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 464) stop_watchdog_on_cpu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 465) if (err < 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 466) pr_warn("could not be initialized");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 467) return err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 468) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 469) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 470) }