From d129479f1fff5c88adbf8dff7649664916d28f81 Mon Sep 17 00:00:00 2001 From: Jisheng Zhang Date: Wed, 18 Dec 2019 05:31:25 +0000 Subject: [PATCH 1/3] watchdog: Remove soft_lockup_hrtimer_cnt and related code After commit 9cf57731b63e ("watchdog/softlockup: Replace "watchdog/%u" threads with cpu_stop_work"), the percpu soft_lockup_hrtimer_cnt is not used any more, so remove it and related code. Signed-off-by: Jisheng Zhang Signed-off-by: Thomas Gleixner Link: https://lore.kernel.org/r/20191218131720.4146aea2@xhacker.debian --- kernel/watchdog.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/kernel/watchdog.c b/kernel/watchdog.c index f41334ef0971..0621301ae8cf 100644 --- a/kernel/watchdog.c +++ b/kernel/watchdog.c @@ -173,7 +173,6 @@ static DEFINE_PER_CPU(struct hrtimer, watchdog_hrtimer); static DEFINE_PER_CPU(bool, softlockup_touch_sync); static DEFINE_PER_CPU(bool, soft_watchdog_warn); static DEFINE_PER_CPU(unsigned long, hrtimer_interrupts); -static DEFINE_PER_CPU(unsigned long, soft_lockup_hrtimer_cnt); static DEFINE_PER_CPU(struct task_struct *, softlockup_task_ptr_saved); static DEFINE_PER_CPU(unsigned long, hrtimer_interrupts_saved); static unsigned long soft_lockup_nmi_warn; @@ -350,8 +349,6 @@ static DEFINE_PER_CPU(struct cpu_stop_work, softlockup_stop_work); */ static int softlockup_fn(void *data) { - __this_cpu_write(soft_lockup_hrtimer_cnt, - __this_cpu_read(hrtimer_interrupts)); __touch_watchdog(); complete(this_cpu_ptr(&softlockup_completion)); From 3a51449b7959f68cc45abe67298e40c7eb57167b Mon Sep 17 00:00:00 2001 From: Petr Mladek Date: Thu, 24 Oct 2019 13:49:26 +0200 Subject: [PATCH 2/3] watchdog/softlockup: Remove obsolete check of last reported task commit 9cf57731b63e ("watchdog/softlockup: Replace "watchdog/%u" threads with cpu_stop_work") ensures that the watchdog is reliably touched during a task switch. As a result the check for an unnoticed task switch is not longer needed. Remove the relevant code, which effectively reverts commit b1a8de1f5343 ("softlockup: make detector be aware of task switch of processes hogging cpu") Signed-off-by: Petr Mladek Signed-off-by: Thomas Gleixner Acked-by: Peter Ziljstra Link: https://lore.kernel.org/r/20191024114928.15377-2-pmladek@suse.com --- kernel/watchdog.c | 18 +----------------- 1 file changed, 1 insertion(+), 17 deletions(-) diff --git a/kernel/watchdog.c b/kernel/watchdog.c index 0621301ae8cf..e3774e9625b7 100644 --- a/kernel/watchdog.c +++ b/kernel/watchdog.c @@ -173,7 +173,6 @@ static DEFINE_PER_CPU(struct hrtimer, watchdog_hrtimer); static DEFINE_PER_CPU(bool, softlockup_touch_sync); static DEFINE_PER_CPU(bool, soft_watchdog_warn); static DEFINE_PER_CPU(unsigned long, hrtimer_interrupts); -static DEFINE_PER_CPU(struct task_struct *, softlockup_task_ptr_saved); static DEFINE_PER_CPU(unsigned long, hrtimer_interrupts_saved); static unsigned long soft_lockup_nmi_warn; @@ -413,22 +412,8 @@ static enum hrtimer_restart watchdog_timer_fn(struct hrtimer *hrtimer) return HRTIMER_RESTART; /* only warn once */ - if (__this_cpu_read(soft_watchdog_warn) == true) { - /* - * When multiple processes are causing softlockups the - * softlockup detector only warns on the first one - * because the code relies on a full quiet cycle to - * re-arm. The second process prevents the quiet cycle - * and never gets reported. Use task pointers to detect - * this. - */ - if (__this_cpu_read(softlockup_task_ptr_saved) != - current) { - __this_cpu_write(soft_watchdog_warn, false); - __touch_watchdog(); - } + if (__this_cpu_read(soft_watchdog_warn) == true) return HRTIMER_RESTART; - } if (softlockup_all_cpu_backtrace) { /* Prevent multiple soft-lockup reports if one cpu is already @@ -444,7 +429,6 @@ 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)); - __this_cpu_write(softlockup_task_ptr_saved, current); print_modules(); print_irqtrace_events(current); if (regs) From 11e31f608b499f044f24b20be73f1dcab3e43f8a Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Thu, 16 Jan 2020 19:17:02 +0100 Subject: [PATCH 3/3] watchdog/softlockup: Enforce that timestamp is valid on boot Robert reported that during boot the watchdog timestamp is set to 0 for one second which is the indicator for a watchdog reset. The reason for this is that the timestamp is in seconds and the time is taken from sched clock and divided by ~1e9. sched clock starts at 0 which means that for the first second during boot the watchdog timestamp is 0, i.e. reset. Use ULONG_MAX as the reset indicator value so the watchdog works correctly right from the start. ULONG_MAX would only conflict with a real timestamp if the system reaches an uptime of 136 years on 32bit and almost eternity on 64bit. Reported-by: Robert Richter Signed-off-by: Thomas Gleixner Link: https://lore.kernel.org/r/87o8v3uuzl.fsf@nanos.tec.linutronix.de --- kernel/watchdog.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/kernel/watchdog.c b/kernel/watchdog.c index e3774e9625b7..b6b1f54a7837 100644 --- a/kernel/watchdog.c +++ b/kernel/watchdog.c @@ -161,6 +161,8 @@ static void lockup_detector_update_enable(void) #ifdef CONFIG_SOFTLOCKUP_DETECTOR +#define SOFTLOCKUP_RESET ULONG_MAX + /* Global variables, exported for sysctl */ unsigned int __read_mostly softlockup_panic = CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC_VALUE; @@ -272,7 +274,7 @@ notrace void touch_softlockup_watchdog_sched(void) * Preemption can be enabled. It doesn't matter which CPU's timestamp * gets zeroed here, so use the raw_ operation. */ - raw_cpu_write(watchdog_touch_ts, 0); + raw_cpu_write(watchdog_touch_ts, SOFTLOCKUP_RESET); } notrace void touch_softlockup_watchdog(void) @@ -296,14 +298,14 @@ void touch_all_softlockup_watchdogs(void) * the softlockup check. */ for_each_cpu(cpu, &watchdog_allowed_mask) - per_cpu(watchdog_touch_ts, cpu) = 0; + per_cpu(watchdog_touch_ts, cpu) = SOFTLOCKUP_RESET; wq_watchdog_touch(-1); } void touch_softlockup_watchdog_sync(void) { __this_cpu_write(softlockup_touch_sync, true); - __this_cpu_write(watchdog_touch_ts, 0); + __this_cpu_write(watchdog_touch_ts, SOFTLOCKUP_RESET); } static int is_softlockup(unsigned long touch_ts) @@ -379,7 +381,7 @@ static enum hrtimer_restart watchdog_timer_fn(struct hrtimer *hrtimer) /* .. and repeat */ hrtimer_forward_now(hrtimer, ns_to_ktime(sample_period)); - if (touch_ts == 0) { + if (touch_ts == SOFTLOCKUP_RESET) { if (unlikely(__this_cpu_read(softlockup_touch_sync))) { /* * If the time stamp was touched atomically