X-Git-Url: http://git.onelab.eu/?a=blobdiff_plain;f=kernel%2Fposix-timers.c;h=42c24868837ca4bec27b8ffddfbc318263e23c21;hb=9bf4aaab3e101692164d49b7ca357651eb691cb6;hp=f846b77a205e868f4f3d4ea8ebc23b0247230117;hpb=db216c3d5e4c040e557a50f8f5d35d5c415e8c1c;p=linux-2.6.git diff --git a/kernel/posix-timers.c b/kernel/posix-timers.c index f846b77a2..42c248688 100644 --- a/kernel/posix-timers.c +++ b/kernel/posix-timers.c @@ -7,6 +7,9 @@ * * Copyright (C) 2002 2003 by MontaVista Software. * + * 2004-06-01 Fix CLOCK_REALTIME clock/timer TIMER_ABSTIME bug. + * Copyright (C) 2004 Boris Hu + * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by * the Free Software Foundation; either version 2 of the License, or (at @@ -41,6 +44,7 @@ #include #include #include +#include #ifndef div_long_long_rem #include @@ -169,6 +173,12 @@ static spinlock_t idr_lock = SPIN_LOCK_UNLOCKED; */ static struct k_clock posix_clocks[MAX_CLOCKS]; +/* + * We only have one real clock that can be set so we need only one abs list, + * even if we should want to have several clocks with differing resolutions. + */ +static struct k_clock_abs abs_list = {.list = LIST_HEAD_INIT(abs_list.list), + .lock = SPIN_LOCK_UNLOCKED}; #define if_clock_do(clock_fun,alt_fun,parms) \ (!clock_fun) ? alt_fun parms : clock_fun parms @@ -200,8 +210,11 @@ static inline void unlock_timer(struct k_itimer *timr, unsigned long flags) */ static __init int init_posix_timers(void) { - struct k_clock clock_realtime = {.res = CLOCK_REALTIME_RES }; + struct k_clock clock_realtime = {.res = CLOCK_REALTIME_RES, + .abs_struct = &abs_list + }; struct k_clock clock_monotonic = {.res = CLOCK_REALTIME_RES, + .abs_struct = NULL, .clock_get = do_posix_clock_monotonic_gettime, .clock_set = do_posix_clock_monotonic_settime }; @@ -210,9 +223,8 @@ static __init int init_posix_timers(void) register_posix_clock(CLOCK_MONOTONIC, &clock_monotonic); posix_timers_cache = kmem_cache_create("posix_timers_cache", - sizeof (struct k_itimer), 0, 0, 0, 0); + sizeof (struct k_itimer), 0, 0, NULL, NULL); idr_init(&posix_timers_id); - return 0; } @@ -239,19 +251,92 @@ static void tstojiffie(struct timespec *tp, int res, u64 *jiff) (NSEC_JIFFIE_SC - SEC_JIFFIE_SC))) >> SEC_JIFFIE_SC; } +/* + * This function adjusts the timer as needed as a result of the clock + * being set. It should only be called for absolute timers, and then + * under the abs_list lock. It computes the time difference and sets + * the new jiffies value in the timer. It also updates the timers + * reference wall_to_monotonic value. It is complicated by the fact + * that tstojiffies() only handles positive times and it needs to work + * with both positive and negative times. Also, for negative offsets, + * we need to defeat the res round up. + * + * Return is true if there is a new time, else false. + */ +static long add_clockset_delta(struct k_itimer *timr, + struct timespec *new_wall_to) +{ + struct timespec delta; + int sign = 0; + u64 exp; + + set_normalized_timespec(&delta, + new_wall_to->tv_sec - + timr->wall_to_prev.tv_sec, + new_wall_to->tv_nsec - + timr->wall_to_prev.tv_nsec); + if (likely(!(delta.tv_sec | delta.tv_nsec))) + return 0; + if (delta.tv_sec < 0) { + set_normalized_timespec(&delta, + -delta.tv_sec, + 1 - delta.tv_nsec - + posix_clocks[timr->it_clock].res); + sign++; + } + tstojiffie(&delta, posix_clocks[timr->it_clock].res, &exp); + timr->wall_to_prev = *new_wall_to; + timr->it_timer.expires += (sign ? -exp : exp); + return 1; +} + +static void remove_from_abslist(struct k_itimer *timr) +{ + if (!list_empty(&timr->abs_timer_entry)) { + spin_lock(&abs_list.lock); + list_del_init(&timr->abs_timer_entry); + spin_unlock(&abs_list.lock); + } +} + static void schedule_next_timer(struct k_itimer *timr) { + struct timespec new_wall_to; struct now_struct now; + unsigned long seq; - /* Set up the timer for the next interval (if there is one) */ + /* + * Set up the timer for the next interval (if there is one). + * Note: this code uses the abs_timer_lock to protect + * wall_to_prev and must hold it until exp is set, not exactly + * obvious... + + * This function is used for CLOCK_REALTIME* and + * CLOCK_MONOTONIC* timers. If we ever want to handle other + * CLOCKs, the calling code (do_schedule_next_timer) would need + * to pull the "clock" info from the timer and dispatch the + * "other" CLOCKs "next timer" code (which, I suppose should + * also be added to the k_clock structure). + */ if (!timr->it_incr) return; - posix_get_now(&now); do { - posix_bump_timer(timr); - }while (posix_time_before(&timr->it_timer, &now)); + seq = read_seqbegin(&xtime_lock); + new_wall_to = wall_to_monotonic; + posix_get_now(&now); + } while (read_seqretry(&xtime_lock, seq)); + + if (!list_empty(&timr->abs_timer_entry)) { + spin_lock(&abs_list.lock); + add_clockset_delta(timr, &new_wall_to); + posix_bump_timer(timr, now); + + spin_unlock(&abs_list.lock); + } else { + posix_bump_timer(timr, now); + } timr->it_overrun_last = timr->it_overrun; timr->it_overrun = -1; ++timr->it_requeue_pending; @@ -312,7 +397,15 @@ static void timer_notify_task(struct k_itimer *timr) memset(&timr->sigq->info, 0, sizeof(siginfo_t)); - /* Send signal to the process that owns this timer. */ + /* + * Send signal to the process that owns this timer. + + * This code assumes that all the possible abs_lists share the + * same lock (there is only one list at this time). If this is + * not the case, the CLOCK info would need to be used to find + * the proper abs list lock. + */ + timr->sigq->info.si_signo = timr->it_sigev_signo; timr->sigq->info.si_errno = 0; timr->sigq->info.si_code = SI_TIMER; @@ -320,6 +413,9 @@ static void timer_notify_task(struct k_itimer *timr) timr->sigq->info.si_value = timr->it_sigev_value; if (timr->it_incr) timr->sigq->info.si_sys_private = ++timr->it_requeue_pending; + else { + remove_from_abslist(timr); + } if (timr->it_sigev_notify & SIGEV_THREAD_ID) { if (unlikely(timr->it_process->flags & PF_EXITING)) { @@ -350,16 +446,51 @@ static void timer_notify_task(struct k_itimer *timr) * This function gets called when a POSIX.1b interval timer expires. It * is used as a callback from the kernel internal timer. The * run_timer_list code ALWAYS calls with interrutps on. + + * This code is for CLOCK_REALTIME* and CLOCK_MONOTONIC* timers. */ static void posix_timer_fn(unsigned long __data) { struct k_itimer *timr = (struct k_itimer *) __data; unsigned long flags; + unsigned long seq; + struct timespec delta, new_wall_to; + u64 exp = 0; + int do_notify = 1; spin_lock_irqsave(&timr->it_lock, flags); set_timer_inactive(timr); - timer_notify_task(timr); - unlock_timer(timr, flags); + if (!list_empty(&timr->abs_timer_entry)) { + spin_lock(&abs_list.lock); + do { + seq = read_seqbegin(&xtime_lock); + new_wall_to = wall_to_monotonic; + } while (read_seqretry(&xtime_lock, seq)); + set_normalized_timespec(&delta, + new_wall_to.tv_sec - + timr->wall_to_prev.tv_sec, + new_wall_to.tv_nsec - + timr->wall_to_prev.tv_nsec); + if (likely((delta.tv_sec | delta.tv_nsec ) == 0)) { + /* do nothing, timer is on time */ + } else if (delta.tv_sec < 0) { + /* do nothing, timer is already late */ + } else { + /* timer is early due to a clock set */ + tstojiffie(&delta, + posix_clocks[timr->it_clock].res, + &exp); + timr->wall_to_prev = new_wall_to; + timr->it_timer.expires += exp; + add_timer(&timr->it_timer); + do_notify = 0; + } + spin_unlock(&abs_list.lock); + + } + if (do_notify) + timer_notify_task(timr); + unlock_timer(timr, flags); /* hold thru abs lock to keep irq off */ } @@ -397,17 +528,19 @@ static struct k_itimer * alloc_posix_timer(void) if (!tmr) return tmr; memset(tmr, 0, sizeof (struct k_itimer)); - tmr->it_id = (timer_t)-1; + INIT_LIST_HEAD(&tmr->abs_timer_entry); if (unlikely(!(tmr->sigq = sigqueue_alloc()))) { kmem_cache_free(posix_timers_cache, tmr); - tmr = 0; + tmr = NULL; } return tmr; } -static void release_posix_timer(struct k_itimer *tmr) +#define IT_ID_SET 1 +#define IT_ID_NOT_SET 0 +static void release_posix_timer(struct k_itimer *tmr, int it_id_set) { - if (tmr->it_id != -1) { + if (it_id_set) { unsigned long flags; spin_lock_irqsave(&idr_lock, flags); idr_remove(&posix_timers_id, tmr->it_id); @@ -429,10 +562,11 @@ sys_timer_create(clockid_t which_clock, { int error = 0; struct k_itimer *new_timer = NULL; - timer_t new_timer_id; - struct task_struct *process = 0; + int new_timer_id; + struct task_struct *process = NULL; unsigned long flags; sigevent_t event; + int it_id_set = IT_ID_NOT_SET; if ((unsigned) which_clock >= MAX_CLOCKS || !posix_clocks[which_clock].res) @@ -443,19 +577,38 @@ sys_timer_create(clockid_t which_clock, return -EAGAIN; spin_lock_init(&new_timer->it_lock); - do { - if (unlikely(!idr_pre_get(&posix_timers_id, GFP_KERNEL))) { - error = -EAGAIN; - new_timer->it_id = (timer_t)-1; - goto out; - } - spin_lock_irq(&idr_lock); - new_timer_id = (timer_t) idr_get_new(&posix_timers_id, - (void *) new_timer); - spin_unlock_irq(&idr_lock); - } while (unlikely(new_timer_id == -1)); + retry: + if (unlikely(!idr_pre_get(&posix_timers_id, GFP_KERNEL))) { + error = -EAGAIN; + goto out; + } + spin_lock_irq(&idr_lock); + error = idr_get_new(&posix_timers_id, + (void *) new_timer, + &new_timer_id); + spin_unlock_irq(&idr_lock); + if (error == -EAGAIN) + goto retry; + else if (error) { + /* + * Wierd looking, but we return EAGAIN if the IDR is + * full (proper POSIX return value for this) + */ + error = -EAGAIN; + goto out; + } + + it_id_set = IT_ID_SET; + new_timer->it_id = (timer_t) new_timer_id; + new_timer->it_clock = which_clock; + new_timer->it_incr = 0; + new_timer->it_overrun = -1; + init_timer(&new_timer->it_timer); + new_timer->it_timer.expires = 0; + new_timer->it_timer.data = (unsigned long) new_timer; + new_timer->it_timer.function = posix_timer_fn; + set_timer_inactive(new_timer); - new_timer->it_id = new_timer_id; /* * return the timer_id now. The next step is hard to * back out if there is an error. @@ -470,6 +623,10 @@ sys_timer_create(clockid_t which_clock, error = -EFAULT; goto out; } + new_timer->it_sigev_notify = event.sigev_notify; + new_timer->it_sigev_signo = event.sigev_signo; + new_timer->it_sigev_value = event.sigev_value; + read_lock(&tasklist_lock); if ((process = good_sigevent(&event))) { /* @@ -489,13 +646,14 @@ sys_timer_create(clockid_t which_clock, */ spin_lock_irqsave(&process->sighand->siglock, flags); if (!(process->flags & PF_EXITING)) { + new_timer->it_process = process; list_add(&new_timer->list, &process->signal->posix_timers); spin_unlock_irqrestore(&process->sighand->siglock, flags); get_task_struct(process); } else { spin_unlock_irqrestore(&process->sighand->siglock, flags); - process = 0; + process = NULL; } } read_unlock(&tasklist_lock); @@ -503,35 +661,27 @@ sys_timer_create(clockid_t which_clock, error = -EINVAL; goto out; } - new_timer->it_sigev_notify = event.sigev_notify; - new_timer->it_sigev_signo = event.sigev_signo; - new_timer->it_sigev_value = event.sigev_value; } else { new_timer->it_sigev_notify = SIGEV_SIGNAL; new_timer->it_sigev_signo = SIGALRM; new_timer->it_sigev_value.sival_int = new_timer->it_id; process = current->group_leader; spin_lock_irqsave(&process->sighand->siglock, flags); + new_timer->it_process = process; list_add(&new_timer->list, &process->signal->posix_timers); spin_unlock_irqrestore(&process->sighand->siglock, flags); } - new_timer->it_clock = which_clock; - new_timer->it_incr = 0; - new_timer->it_overrun = -1; - init_timer(&new_timer->it_timer); - new_timer->it_timer.expires = 0; - new_timer->it_timer.data = (unsigned long) new_timer; - new_timer->it_timer.function = posix_timer_fn; - set_timer_inactive(new_timer); - - /* - * Once we set the process, it can be found so do it last... + /* + * In the case of the timer belonging to another task, after + * the task is unlocked, the timer is owned by the other task + * and may cease to exist at any time. Don't use or modify + * new_timer after the unlock call. */ - new_timer->it_process = process; + out: if (error) - release_posix_timer(new_timer); + release_posix_timer(new_timer, it_id_set); return error; } @@ -626,8 +776,7 @@ do_timer_gettime(struct k_itimer *timr, struct itimerspec *cur_setting) if (expires) { if (timr->it_requeue_pending & REQUEUE_PENDING || (timr->it_sigev_notify & ~SIGEV_THREAD_ID) == SIGEV_NONE) { - while (posix_time_before(&timr->it_timer, &now)) - posix_bump_timer(timr); + posix_bump_timer(timr, now); expires = timr->it_timer.expires; } else @@ -703,11 +852,10 @@ sys_timer_getoverrun(timer_t timer_id) * time to it to get the proper time for the timer. */ static int adjust_abs_time(struct k_clock *clock, struct timespec *tp, - int abs, u64 *exp) + int abs, u64 *exp, struct timespec *wall_to) { struct timespec now; struct timespec oc = *tp; - struct timespec wall_to_mono; u64 jiffies_64_f; int rtn =0; @@ -715,15 +863,15 @@ static int adjust_abs_time(struct k_clock *clock, struct timespec *tp, /* * The mask pick up the 4 basic clocks */ - if (!(clock - &posix_clocks[0]) & ~CLOCKS_MASK) { + if (!((clock - &posix_clocks[0]) & ~CLOCKS_MASK)) { jiffies_64_f = do_posix_clock_monotonic_gettime_parts( - &now, &wall_to_mono); + &now, wall_to); /* * If we are doing a MONOTONIC clock */ if((clock - &posix_clocks[0]) & CLOCKS_MONO){ - now.tv_sec += wall_to_mono.tv_sec; - now.tv_nsec += wall_to_mono.tv_nsec; + now.tv_sec += wall_to->tv_sec; + now.tv_nsec += wall_to->tv_nsec; } } else { /* @@ -813,6 +961,8 @@ do_timer_settime(struct k_itimer *timr, int flags, #else del_timer(&timr->it_timer); #endif + remove_from_abslist(timr); + timr->it_requeue_pending = (timr->it_requeue_pending + 2) & ~REQUEUE_PENDING; timr->it_overrun_last = 0; @@ -827,24 +977,25 @@ do_timer_settime(struct k_itimer *timr, int flags, if (adjust_abs_time(clock, &new_setting->it_value, flags & TIMER_ABSTIME, - &expire_64)) { + &expire_64, &(timr->wall_to_prev))) { return -EINVAL; } timr->it_timer.expires = (unsigned long)expire_64; tstojiffie(&new_setting->it_interval, clock->res, &expire_64); timr->it_incr = (unsigned long)expire_64; - /* - * For some reason the timer does not fire immediately if expires is - * equal to jiffies, so the timer notify function is called directly. - * We do not even queue SIGEV_NONE timers! + * We do not even queue SIGEV_NONE timers! But we do put them + * in the abs list so we can do that right. */ - if (((timr->it_sigev_notify & ~SIGEV_THREAD_ID) != SIGEV_NONE)) { - if (timr->it_timer.expires == jiffies) - timer_notify_task(timr); - else - add_timer(&timr->it_timer); + if (((timr->it_sigev_notify & ~SIGEV_THREAD_ID) != SIGEV_NONE)) + add_timer(&timr->it_timer); + + if (flags & TIMER_ABSTIME && clock->abs_struct) { + spin_lock(&clock->abs_struct->lock); + list_add_tail(&(timr->abs_timer_entry), + &(clock->abs_struct->list)); + spin_unlock(&clock->abs_struct->lock); } return 0; } @@ -878,7 +1029,7 @@ retry: if (!posix_clocks[timr->it_clock].timer_set) error = do_timer_settime(timr, flags, &new_spec, rtn); else - error = posix_clocks[timr->it_clock].timer_set(timr, + error = posix_clocks[timr->it_clock].timer_set(timr, flags, &new_spec, rtn); unlock_timer(timr, flag); @@ -911,6 +1062,8 @@ static inline int do_timer_delete(struct k_itimer *timer) #else del_timer(&timer->it_timer); #endif + remove_from_abslist(timer); + return 0; } @@ -952,7 +1105,7 @@ retry_delete: timer->it_process = NULL; } unlock_timer(timer, flags); - release_posix_timer(timer); + release_posix_timer(timer, IT_ID_SET); return 0; } /* @@ -989,7 +1142,7 @@ retry_delete: timer->it_process = NULL; } unlock_timer(timer, flags); - release_posix_timer(timer); + release_posix_timer(timer, IT_ID_SET); } /* @@ -1153,13 +1306,93 @@ static void nanosleep_wake_up(unsigned long __data) * On locking, clock_was_set() is called from update_wall_clock which * holds (or has held for it) a write_lock_irq( xtime_lock) and is * called from the timer bh code. Thus we need the irq save locks. + * + * Also, on the call from update_wall_clock, that is done as part of a + * softirq thing. We don't want to delay the system that much (possibly + * long list of timers to fix), so we defer that work to keventd. */ static DECLARE_WAIT_QUEUE_HEAD(nanosleep_abs_wqueue); +static DECLARE_WORK(clock_was_set_work, (void(*)(void*))clock_was_set, NULL); + +static DECLARE_MUTEX(clock_was_set_lock); void clock_was_set(void) { + struct k_itimer *timr; + struct timespec new_wall_to; + LIST_HEAD(cws_list); + unsigned long seq; + + + if (unlikely(in_interrupt())) { + schedule_work(&clock_was_set_work); + return; + } wake_up_all(&nanosleep_abs_wqueue); + + /* + * Check if there exist TIMER_ABSTIME timers to correct. + * + * Notes on locking: This code is run in task context with irq + * on. We CAN be interrupted! All other usage of the abs list + * lock is under the timer lock which holds the irq lock as + * well. We REALLY don't want to scan the whole list with the + * interrupt system off, AND we would like a sequence lock on + * this code as well. Since we assume that the clock will not + * be set often, it seems ok to take and release the irq lock + * for each timer. In fact add_timer will do this, so this is + * not an issue. So we know when we are done, we will move the + * whole list to a new location. Then as we process each entry, + * we will move it to the actual list again. This way, when our + * copy is empty, we are done. We are not all that concerned + * about preemption so we will use a semaphore lock to protect + * aginst reentry. This way we will not stall another + * processor. It is possible that this may delay some timers + * that should have expired, given the new clock, but even this + * will be minimal as we will always update to the current time, + * even if it was set by a task that is waiting for entry to + * this code. Timers that expire too early will be caught by + * the expire code and restarted. + + * Absolute timers that repeat are left in the abs list while + * waiting for the task to pick up the signal. This means we + * may find timers that are not in the "add_timer" list, but are + * in the abs list. We do the same thing for these, save + * putting them back in the "add_timer" list. (Note, these are + * left in the abs list mainly to indicate that they are + * ABSOLUTE timers, a fact that is used by the re-arm code, and + * for which we have no other flag.) + + */ + + down(&clock_was_set_lock); + spin_lock_irq(&abs_list.lock); + list_splice_init(&abs_list.list, &cws_list); + spin_unlock_irq(&abs_list.lock); + do { + do { + seq = read_seqbegin(&xtime_lock); + new_wall_to = wall_to_monotonic; + } while (read_seqretry(&xtime_lock, seq)); + + spin_lock_irq(&abs_list.lock); + if (list_empty(&cws_list)) { + spin_unlock_irq(&abs_list.lock); + break; + } + timr = list_entry(cws_list.next, struct k_itimer, + abs_timer_entry); + + list_del_init(&timr->abs_timer_entry); + if (add_clockset_delta(timr, &new_wall_to) && + del_timer(&timr->it_timer)) /* timer run yet? */ + add_timer(&timr->it_timer); + list_add(&timr->abs_timer_entry, &abs_list.list); + spin_unlock_irq(&abs_list.lock); + } while (1); + + up(&clock_was_set_lock); } long clock_nanosleep_restart(struct restart_block *restart_block); @@ -1202,7 +1435,7 @@ sys_clock_nanosleep(clockid_t which_clock, int flags, long do_clock_nanosleep(clockid_t which_clock, int flags, struct timespec *tsave) { - struct timespec t; + struct timespec t, dum; struct timer_list new_timer; DECLARE_WAITQUEUE(abs_wqueue, current); u64 rq_time = (u64)0; @@ -1242,7 +1475,7 @@ do_clock_nanosleep(clockid_t which_clock, int flags, struct timespec *tsave) t = *tsave; if (abs || !rq_time) { adjust_abs_time(&posix_clocks[which_clock], &t, abs, - &rq_time); + &rq_time, &dum); rq_time += (t.tv_sec || t.tv_nsec); }