/* -*- mode: c; c-basic-offset: 8; indent-tabs-mode: nil; -*- * vim:expandtab:shiftwidth=8:tabstop=8: * * GPL HEADER START * * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 only, * as published by the Free Software Foundation. * * This program is distributed in the hope that it will be useful, but * WITHOUT ANY WARRANTY; without even the implied warranty of * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU * General Public License version 2 for more details (a copy is included * in the LICENSE file that accompanied this code). * * You should have received a copy of the GNU General Public License * version 2 along with this program; If not, see * http://www.sun.com/software/products/lustre/docs/GPLv2.pdf * * Please contact Sun Microsystems, Inc., 4150 Network Circle, Santa Clara, * CA 95054 USA or visit www.sun.com if you need additional information or * have any questions. * * GPL HEADER END */ /* * Copyright 2008 Sun Microsystems, Inc. All rights reserved * Use is subject to license terms. */ /* * This file is part of Lustre, http://www.lustre.org/ * Lustre is a trademark of Sun Microsystems, Inc. * * libcfs/libcfs/watchdog.c * * Author: Jacob Berkman */ #define DEBUG_SUBSYSTEM S_LNET #include #include "tracefile.h" struct lc_watchdog { cfs_timer_t lcw_timer; /* kernel timer */ struct list_head lcw_list; struct timeval lcw_last_touched; cfs_task_t *lcw_task; void (*lcw_callback)(pid_t, void *); void *lcw_data; pid_t lcw_pid; cfs_duration_t lcw_time; /* time until watchdog fires, jiffies */ enum { LC_WATCHDOG_DISABLED, LC_WATCHDOG_ENABLED, LC_WATCHDOG_EXPIRED } lcw_state; }; #ifdef WITH_WATCHDOG /* * The dispatcher will complete lcw_start_completion when it starts, * and lcw_stop_completion when it exits. * Wake lcw_event_waitq to signal timer callback dispatches. */ static struct completion lcw_start_completion; static struct completion lcw_stop_completion; static cfs_waitq_t lcw_event_waitq; /* * Set this and wake lcw_event_waitq to stop the dispatcher. */ enum { LCW_FLAG_STOP = 0 }; static unsigned long lcw_flags = 0; /* * Number of outstanding watchdogs. * When it hits 1, we start the dispatcher. * When it hits 0, we stop the distpatcher. */ static __u32 lcw_refcount = 0; static DECLARE_MUTEX(lcw_refcount_sem); /* * List of timers that have fired that need their callbacks run by the * dispatcher. */ static spinlock_t lcw_pending_timers_lock = SPIN_LOCK_UNLOCKED; /* BH lock! */ static struct list_head lcw_pending_timers = \ CFS_LIST_HEAD_INIT(lcw_pending_timers); /* Last time a watchdog expired */ static cfs_time_t lcw_last_watchdog_time; static int lcw_recent_watchdog_count; static spinlock_t lcw_last_watchdog_lock = SPIN_LOCK_UNLOCKED; static void lcw_dump(struct lc_watchdog *lcw) { cfs_task_t *tsk; #if defined(HAVE_TASKLIST_LOCK) read_lock(&tasklist_lock); #elif defined(HAVE_TASK_RCU) rcu_read_lock(); #else CERROR("unable to dump stack because of missing export\n"); return; #endif ENTRY; tsk = find_task_by_pid(lcw->lcw_pid); if (tsk == NULL) { CWARN("Process " LPPID " was not found in the task list; " "watchdog callback may be incomplete\n", lcw->lcw_pid); } else if (tsk != lcw->lcw_task) { CWARN("The current process " LPPID " did not set the watchdog; " "watchdog callback may be incomplete\n", lcw->lcw_pid); } else { libcfs_debug_dumpstack(tsk); } #if defined(HAVE_TASKLIST_LOCK) read_unlock(&tasklist_lock); #elif defined(HAVE_TASK_RCU) rcu_read_unlock(); #endif EXIT; } static void lcw_cb(ulong_ptr_t data) { struct lc_watchdog *lcw = (struct lc_watchdog *)data; cfs_time_t current_time; cfs_duration_t delta_time; ENTRY; if (lcw->lcw_state != LC_WATCHDOG_ENABLED) { EXIT; return; } lcw->lcw_state = LC_WATCHDOG_EXPIRED; current_time = cfs_time_current(); /* Check to see if we should throttle the watchdog timer to avoid * too many dumps going to the console thus triggering an NMI. * Normally we would not hold the spin lock over the CWARN but in * this case we hold it to ensure non ratelimited lcw_dumps are not * interleaved on the console making them hard to read. */ spin_lock_bh(&lcw_last_watchdog_lock); delta_time = cfs_duration_sec(current_time - lcw_last_watchdog_time); if (delta_time < libcfs_watchdog_ratelimit && lcw_recent_watchdog_count > 3) { CWARN("Refusing to fire watchdog for pid %d: it was inactive " "for %ldms. Rate limiting 1 per %d seconds.\n", (int)lcw->lcw_pid,cfs_duration_sec(lcw->lcw_time) * 1000, libcfs_watchdog_ratelimit); } else { if (delta_time < libcfs_watchdog_ratelimit) { lcw_recent_watchdog_count++; } else { memcpy(&lcw_last_watchdog_time, ¤t_time, sizeof(current_time)); lcw_recent_watchdog_count = 0; } /* This warning should appear on the console, but may not get * into the logs since we're running in a softirq handler */ CWARN("Watchdog triggered for pid %d: it was inactive for %lds\n", (int)lcw->lcw_pid, cfs_duration_sec(lcw->lcw_time)); lcw_dump(lcw); } spin_unlock_bh(&lcw_last_watchdog_lock); spin_lock_bh(&lcw_pending_timers_lock); if (list_empty(&lcw->lcw_list)) { list_add(&lcw->lcw_list, &lcw_pending_timers); cfs_waitq_signal(&lcw_event_waitq); } spin_unlock_bh(&lcw_pending_timers_lock); EXIT; } static int is_watchdog_fired(void) { int rc; if (test_bit(LCW_FLAG_STOP, &lcw_flags)) return 1; spin_lock_bh(&lcw_pending_timers_lock); rc = !list_empty(&lcw_pending_timers); spin_unlock_bh(&lcw_pending_timers_lock); return rc; } static int lcw_dispatch_main(void *data) { int rc = 0; unsigned long flags; struct lc_watchdog *lcw; ENTRY; cfs_daemonize("lc_watchdogd"); SIGNAL_MASK_LOCK(current, flags); sigfillset(¤t->blocked); RECALC_SIGPENDING; SIGNAL_MASK_UNLOCK(current, flags); complete(&lcw_start_completion); while (1) { cfs_wait_event_interruptible(lcw_event_waitq, is_watchdog_fired(), rc); CDEBUG(D_INFO, "Watchdog got woken up...\n"); if (test_bit(LCW_FLAG_STOP, &lcw_flags)) { CDEBUG(D_INFO, "LCW_FLAG_STOP was set, shutting down...\n"); spin_lock_bh(&lcw_pending_timers_lock); rc = !list_empty(&lcw_pending_timers); spin_unlock_bh(&lcw_pending_timers_lock); if (rc) { CERROR("pending timers list was not empty at " "time of watchdog dispatch shutdown\n"); } break; } spin_lock_bh(&lcw_pending_timers_lock); while (!list_empty(&lcw_pending_timers)) { lcw = list_entry(lcw_pending_timers.next, struct lc_watchdog, lcw_list); list_del_init(&lcw->lcw_list); spin_unlock_bh(&lcw_pending_timers_lock); CDEBUG(D_INFO, "found lcw for pid " LPPID ": inactive for " "%lds\n", lcw->lcw_pid, cfs_duration_sec(lcw->lcw_time)); if (lcw->lcw_state != LC_WATCHDOG_DISABLED) lcw->lcw_callback(lcw->lcw_pid, lcw->lcw_data); spin_lock_bh(&lcw_pending_timers_lock); } spin_unlock_bh(&lcw_pending_timers_lock); } complete(&lcw_stop_completion); RETURN(rc); } static void lcw_dispatch_start(void) { int rc; ENTRY; LASSERT(lcw_refcount == 1); init_completion(&lcw_stop_completion); init_completion(&lcw_start_completion); cfs_waitq_init(&lcw_event_waitq); CDEBUG(D_INFO, "starting dispatch thread\n"); rc = kernel_thread(lcw_dispatch_main, NULL, 0); if (rc < 0) { CERROR("error spawning watchdog dispatch thread: %d\n", rc); EXIT; return; } wait_for_completion(&lcw_start_completion); CDEBUG(D_INFO, "watchdog dispatcher initialization complete.\n"); EXIT; } static void lcw_dispatch_stop(void) { ENTRY; LASSERT(lcw_refcount == 0); CDEBUG(D_INFO, "trying to stop watchdog dispatcher.\n"); set_bit(LCW_FLAG_STOP, &lcw_flags); cfs_waitq_signal(&lcw_event_waitq); wait_for_completion(&lcw_stop_completion); CDEBUG(D_INFO, "watchdog dispatcher has shut down.\n"); EXIT; } struct lc_watchdog *lc_watchdog_add(int timeout_ms, void (*callback)(pid_t, void *), void *data) { struct lc_watchdog *lcw = NULL; ENTRY; LIBCFS_ALLOC(lcw, sizeof(*lcw)); if (lcw == NULL) { CDEBUG(D_INFO, "Could not allocate new lc_watchdog\n"); RETURN(ERR_PTR(-ENOMEM)); } lcw->lcw_task = cfs_current(); lcw->lcw_pid = cfs_curproc_pid(); lcw->lcw_time = cfs_time_seconds(timeout_ms) / 1000; lcw->lcw_callback = (callback != NULL) ? callback : lc_watchdog_dumplog; lcw->lcw_data = data; lcw->lcw_state = LC_WATCHDOG_DISABLED; CFS_INIT_LIST_HEAD(&lcw->lcw_list); cfs_timer_init(&lcw->lcw_timer, lcw_cb, lcw); down(&lcw_refcount_sem); if (++lcw_refcount == 1) lcw_dispatch_start(); up(&lcw_refcount_sem); /* Keep this working in case we enable them by default */ if (lcw->lcw_state == LC_WATCHDOG_ENABLED) { do_gettimeofday(&lcw->lcw_last_touched); cfs_timer_arm(&lcw->lcw_timer, lcw->lcw_time + cfs_time_current()); } RETURN(lcw); } EXPORT_SYMBOL(lc_watchdog_add); static void lcw_update_time(struct lc_watchdog *lcw, const char *message) { struct timeval newtime; struct timeval timediff; do_gettimeofday(&newtime); if (lcw->lcw_state == LC_WATCHDOG_EXPIRED) { cfs_timeval_sub(&newtime, &lcw->lcw_last_touched, &timediff); CWARN("Expired watchdog for pid " LPPID " %s after %lu.%.4lus\n", lcw->lcw_pid, message, timediff.tv_sec, timediff.tv_usec / 100); } lcw->lcw_last_touched = newtime; } void lc_watchdog_touch_ms(struct lc_watchdog *lcw, int timeout_ms) { ENTRY; LASSERT(lcw != NULL); spin_lock_bh(&lcw_pending_timers_lock); list_del_init(&lcw->lcw_list); spin_unlock_bh(&lcw_pending_timers_lock); lcw_update_time(lcw, "touched"); lcw->lcw_state = LC_WATCHDOG_ENABLED; cfs_timer_arm(&lcw->lcw_timer, cfs_time_current() + cfs_time_seconds(timeout_ms) / 1000); EXIT; } EXPORT_SYMBOL(lc_watchdog_touch_ms); /* deprecated - use above instead */ void lc_watchdog_touch(struct lc_watchdog *lcw) { lc_watchdog_touch_ms(lcw, (int)cfs_duration_sec(lcw->lcw_time) * 1000); } EXPORT_SYMBOL(lc_watchdog_touch); void lc_watchdog_disable(struct lc_watchdog *lcw) { ENTRY; LASSERT(lcw != NULL); spin_lock_bh(&lcw_pending_timers_lock); if (!list_empty(&lcw->lcw_list)) list_del_init(&lcw->lcw_list); spin_unlock_bh(&lcw_pending_timers_lock); lcw_update_time(lcw, "disabled"); lcw->lcw_state = LC_WATCHDOG_DISABLED; EXIT; } EXPORT_SYMBOL(lc_watchdog_disable); void lc_watchdog_delete(struct lc_watchdog *lcw) { ENTRY; LASSERT(lcw != NULL); cfs_timer_disarm(&lcw->lcw_timer); lcw_update_time(lcw, "deleted"); spin_lock_bh(&lcw_pending_timers_lock); if (!list_empty(&lcw->lcw_list)) list_del_init(&lcw->lcw_list); spin_unlock_bh(&lcw_pending_timers_lock); down(&lcw_refcount_sem); if (--lcw_refcount == 0) lcw_dispatch_stop(); up(&lcw_refcount_sem); LIBCFS_FREE(lcw, sizeof(*lcw)); EXIT; } EXPORT_SYMBOL(lc_watchdog_delete); /* * Provided watchdog handlers */ void lc_watchdog_dumplog(pid_t pid, void *data) { libcfs_debug_dumplog_internal((void *)((long_ptr_t)pid)); } EXPORT_SYMBOL(lc_watchdog_dumplog); #else /* !defined(WITH_WATCHDOG) */ struct lc_watchdog *lc_watchdog_add(int timeout_ms, void (*callback)(pid_t pid, void *), void *data) { static struct lc_watchdog watchdog; return &watchdog; } EXPORT_SYMBOL(lc_watchdog_add); void lc_watchdog_touch_ms(struct lc_watchdog *lcw, int timeout_ms) { } EXPORT_SYMBOL(lc_watchdog_touch_ms); void lc_watchdog_touch(struct lc_watchdog *lcw) { } EXPORT_SYMBOL(lc_watchdog_touch); void lc_watchdog_disable(struct lc_watchdog *lcw) { } EXPORT_SYMBOL(lc_watchdog_disable); void lc_watchdog_delete(struct lc_watchdog *lcw) { } EXPORT_SYMBOL(lc_watchdog_delete); #endif