From b421b22b00b0011f6a2ce3561176c4e79e640c49 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 21 Apr 2017 12:14:13 +0200 Subject: x86/tsc, sched/clock, clocksource: Use clocksource watchdog to provide stable sync points Currently we keep sched_clock_tick() active for stable TSC in order to keep the per-CPU state semi up-to-date. The (obvious) problem is that by the time we detect TSC is borked, our per-CPU state is also borked. So hook into the clocksource watchdog and call a method after we've found it to still be stable. There's the obvious race where the TSC goes wonky between finding it stable and us running the callback, but closing that is too much work and not really worth it, since we're already detecting TSC wobbles after the fact, so we cannot, per definition, fully avoid funny clock values. And since the watchdog runs less often than the tick, this is also an optimization. Signed-off-by: Peter Zijlstra (Intel) Cc: Linus Torvalds Cc: Mike Galbraith Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar --- kernel/time/clocksource.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'kernel/time') diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c index 93621ae718d3..03918a19cf2d 100644 --- a/kernel/time/clocksource.c +++ b/kernel/time/clocksource.c @@ -233,6 +233,9 @@ static void clocksource_watchdog(unsigned long data) continue; } + if (cs == curr_clocksource && cs->tick_stable) + cs->tick_stable(cs); + if (!(cs->flags & CLOCK_SOURCE_VALID_FOR_HRES) && (cs->flags & CLOCK_SOURCE_IS_CONTINUOUS) && (watchdog->flags & CLOCK_SOURCE_IS_CONTINUOUS)) { -- cgit v1.2.3 From ac1e843f0900bea92fcb47f6205e1f9ffb0d469c Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 21 Apr 2017 12:26:23 +0200 Subject: sched/clock: Remove unused argument to sched_clock_idle_wakeup_event() The argument to sched_clock_idle_wakeup_event() has not been used in a long time. Remove it. Signed-off-by: Peter Zijlstra (Intel) Cc: Linus Torvalds Cc: Mike Galbraith Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar --- kernel/time/tick-sched.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/time') diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c index 64c97fc130c4..9c2dc64e31d8 100644 --- a/kernel/time/tick-sched.c +++ b/kernel/time/tick-sched.c @@ -554,7 +554,7 @@ static void tick_nohz_stop_idle(struct tick_sched *ts, ktime_t now) update_ts_time_stats(smp_processor_id(), ts, now, NULL); ts->idle_active = 0; - sched_clock_idle_wakeup_event(0); + sched_clock_idle_wakeup_event(); } static ktime_t tick_nohz_start_idle(struct tick_sched *ts) -- cgit v1.2.3 From ce6cf9a15d62fd7ee92f4f9bb754883bacf85a3e Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Thu, 11 May 2017 16:36:19 +0200 Subject: nohz: Add hrtimer sanity check Signed-off-by: Frederic Weisbecker Signed-off-by: Ingo Molnar --- kernel/time/tick-sched.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c index 64c97fc130c4..d212bb62bc08 100644 --- a/kernel/time/tick-sched.c +++ b/kernel/time/tick-sched.c @@ -771,8 +771,13 @@ static ktime_t tick_nohz_stop_sched_tick(struct tick_sched *ts, tick = expires; /* Skip reprogram of event if its not changed */ - if (ts->tick_stopped && (expires == dev->next_event)) - goto out; + if (ts->tick_stopped) { + if (hrtimer_active(&ts->sched_timer)) + WARN_ON_ONCE(hrtimer_get_expires(&ts->sched_timer) < dev->next_event); + + if (expires == dev->next_event) + goto out; + } /* * nohz_stop_sched_tick can be called several times before -- cgit v1.2.3 From 411fe24e6b7c283c3a1911450cdba6dd3aaea56e Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Fri, 21 Apr 2017 16:00:54 +0200 Subject: nohz: Fix collision between tick and other hrtimers, again This restores commit: 24b91e360ef5: ("nohz: Fix collision between tick and other hrtimers") ... which got reverted by commit: 558e8e27e73f: ('Revert "nohz: Fix collision between tick and other hrtimers"') ... due to a regression where CPUs spuriously stopped ticking. The bug happened when a tick fired too early past its expected expiration: on IRQ exit the tick was scheduled again to the same deadline but skipped reprogramming because ts->next_tick still kept in cache the deadline. This has been fixed now with resetting ts->next_tick from the tick itself. Extra care has also been taken to prevent from obsolete values throughout CPU hotplug operations. When the tick is stopped and an interrupt occurs afterward, we check on that interrupt exit if the next tick needs to be rescheduled. If it doesn't need any update, we don't want to do anything. In order to check if the tick needs an update, we compare it against the clockevent device deadline. Now that's a problem because the clockevent device is at a lower level than the tick itself if it is implemented on top of hrtimer. Every hrtimer share this clockevent device. So comparing the next tick deadline against the clockevent device deadline is wrong because the device may be programmed for another hrtimer whose deadline collides with the tick. As a result we may end up not reprogramming the tick accidentally. In a worst case scenario under full dynticks mode, the tick stops firing as it is supposed to every 1hz, leaving /proc/stat stalled: Task in a full dynticks CPU ---------------------------- * hrtimer A is queued 2 seconds ahead * the tick is stopped, scheduled 1 second ahead * tick fires 1 second later * on tick exit, nohz schedules the tick 1 second ahead but sees the clockevent device is already programmed to that deadline, fooled by hrtimer A, the tick isn't rescheduled. * hrtimer A is cancelled before its deadline * tick never fires again until an interrupt happens... In order to fix this, store the next tick deadline to the tick_sched local structure and reuse that value later to check whether we need to reprogram the clock after an interrupt. On the other hand, ts->sleep_length still wants to know about the next clock event and not just the tick, so we want to improve the related comment to avoid confusion. Reported-and-tested-by: Tim Wright Reported-and-tested-by: Pavel Machek Reported-by: James Hartsock Signed-off-by: Frederic Weisbecker Signed-off-by: Thomas Gleixner Acked-by: Rik van Riel Cc: Peter Zijlstra Cc: stable@vger.kernel.org Link: http://lkml.kernel.org/r/1492783255-5051-2-git-send-email-fweisbec@gmail.com Signed-off-by: Ingo Molnar --- kernel/time/tick-sched.c | 37 +++++++++++++++++++++++++++++++------ kernel/time/tick-sched.h | 2 ++ 2 files changed, 33 insertions(+), 6 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c index d212bb62bc08..764d2905e6a5 100644 --- a/kernel/time/tick-sched.c +++ b/kernel/time/tick-sched.c @@ -150,6 +150,12 @@ static void tick_sched_handle(struct tick_sched *ts, struct pt_regs *regs) touch_softlockup_watchdog_sched(); if (is_idle_task(current)) ts->idle_jiffies++; + /* + * In case the current tick fired too early past its expected + * expiration, make sure we don't bypass the next clock reprogramming + * to the same deadline. + */ + ts->next_tick = 0; } #endif update_process_times(user_mode(regs)); @@ -660,6 +666,12 @@ static void tick_nohz_restart(struct tick_sched *ts, ktime_t now) hrtimer_start_expires(&ts->sched_timer, HRTIMER_MODE_ABS_PINNED); else tick_program_event(hrtimer_get_expires(&ts->sched_timer), 1); + + /* + * Reset to make sure next tick stop doesn't get fooled by past + * cached clock deadline. + */ + ts->next_tick = 0; } static ktime_t tick_nohz_stop_sched_tick(struct tick_sched *ts, @@ -771,12 +783,15 @@ static ktime_t tick_nohz_stop_sched_tick(struct tick_sched *ts, tick = expires; /* Skip reprogram of event if its not changed */ - if (ts->tick_stopped) { - if (hrtimer_active(&ts->sched_timer)) - WARN_ON_ONCE(hrtimer_get_expires(&ts->sched_timer) < dev->next_event); - - if (expires == dev->next_event) + if (ts->tick_stopped && (expires == ts->next_tick)) { + /* Sanity check: make sure clockevent is actually programmed */ + if (likely(dev->next_event <= ts->next_tick)) goto out; + + WARN_ON_ONCE(1); + printk_once("basemono: %llu ts->next_tick: %llu dev->next_event: %llu timer->active: %d timer->expires: %llu\n", + basemono, ts->next_tick, dev->next_event, + hrtimer_active(&ts->sched_timer), hrtimer_get_expires(&ts->sched_timer)); } /* @@ -796,6 +811,8 @@ static ktime_t tick_nohz_stop_sched_tick(struct tick_sched *ts, trace_tick_stop(1, TICK_DEP_MASK_NONE); } + ts->next_tick = tick; + /* * If the expiration time == KTIME_MAX, then we simply stop * the tick timer. @@ -811,7 +828,10 @@ static ktime_t tick_nohz_stop_sched_tick(struct tick_sched *ts, else tick_program_event(tick, 1); out: - /* Update the estimated sleep length */ + /* + * Update the estimated sleep length until the next timer + * (not only the tick). + */ ts->sleep_length = ktime_sub(dev->next_event, now); return tick; } @@ -869,6 +889,11 @@ static bool can_stop_idle_tick(int cpu, struct tick_sched *ts) if (unlikely(!cpu_online(cpu))) { if (cpu == tick_do_timer_cpu) tick_do_timer_cpu = TICK_DO_TIMER_NONE; + /* + * Make sure the CPU doesn't get fooled by obsolete tick + * deadline if it comes back online later. + */ + ts->next_tick = 0; return false; } diff --git a/kernel/time/tick-sched.h b/kernel/time/tick-sched.h index bf38226e5c17..075444e3d48e 100644 --- a/kernel/time/tick-sched.h +++ b/kernel/time/tick-sched.h @@ -27,6 +27,7 @@ enum tick_nohz_mode { * timer is modified for nohz sleeps. This is necessary * to resume the tick timer operation in the timeline * when the CPU returns from nohz sleep. + * @next_tick: Next tick to be fired when in dynticks mode. * @tick_stopped: Indicator that the idle tick has been stopped * @idle_jiffies: jiffies at the entry to idle for idle time accounting * @idle_calls: Total number of idle calls @@ -44,6 +45,7 @@ struct tick_sched { unsigned long check_clocks; enum tick_nohz_mode nohz_mode; ktime_t last_tick; + ktime_t next_tick; int inidle; int tick_stopped; unsigned long idle_jiffies; -- cgit v1.2.3 From 43fe8b8eb81eee713400340716cf945f59d21496 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 23 May 2017 23:27:38 +0200 Subject: posix-timers: Make signal printks conditional A recent commit added extra printks for CPU/RT limits. This can result in excessive spam in dmesg. Make the printks conditional on print_fatal_signals. Fixes: e7ea7c9806a2 ("rlimits: Print more information when CPU/RT limits are exceeded") Reported-by: Dave Jones Signed-off-by: Thomas Gleixner Cc: Arun Raghavan --- kernel/time/posix-cpu-timers.c | 24 ++++++++++++++++-------- 1 file changed, 16 insertions(+), 8 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/posix-cpu-timers.c b/kernel/time/posix-cpu-timers.c index 1370f067fb51..d2a1e6dd0291 100644 --- a/kernel/time/posix-cpu-timers.c +++ b/kernel/time/posix-cpu-timers.c @@ -825,8 +825,10 @@ static void check_thread_timers(struct task_struct *tsk, * At the hard limit, we just die. * No need to calculate anything else now. */ - pr_info("CPU Watchdog Timeout (hard): %s[%d]\n", - tsk->comm, task_pid_nr(tsk)); + if (print_fatal_signals) { + pr_info("CPU Watchdog Timeout (hard): %s[%d]\n", + tsk->comm, task_pid_nr(tsk)); + } __group_send_sig_info(SIGKILL, SEND_SIG_PRIV, tsk); return; } @@ -838,8 +840,10 @@ static void check_thread_timers(struct task_struct *tsk, soft += USEC_PER_SEC; sig->rlim[RLIMIT_RTTIME].rlim_cur = soft; } - pr_info("RT Watchdog Timeout (soft): %s[%d]\n", - tsk->comm, task_pid_nr(tsk)); + if (print_fatal_signals) { + pr_info("RT Watchdog Timeout (soft): %s[%d]\n", + tsk->comm, task_pid_nr(tsk)); + } __group_send_sig_info(SIGXCPU, SEND_SIG_PRIV, tsk); } } @@ -936,8 +940,10 @@ static void check_process_timers(struct task_struct *tsk, * At the hard limit, we just die. * No need to calculate anything else now. */ - pr_info("RT Watchdog Timeout (hard): %s[%d]\n", - tsk->comm, task_pid_nr(tsk)); + if (print_fatal_signals) { + pr_info("RT Watchdog Timeout (hard): %s[%d]\n", + tsk->comm, task_pid_nr(tsk)); + } __group_send_sig_info(SIGKILL, SEND_SIG_PRIV, tsk); return; } @@ -945,8 +951,10 @@ static void check_process_timers(struct task_struct *tsk, /* * At the soft limit, send a SIGXCPU every second. */ - pr_info("CPU Watchdog Timeout (soft): %s[%d]\n", - tsk->comm, task_pid_nr(tsk)); + if (print_fatal_signals) { + pr_info("CPU Watchdog Timeout (soft): %s[%d]\n", + tsk->comm, task_pid_nr(tsk)); + } __group_send_sig_info(SIGXCPU, SEND_SIG_PRIV, tsk); if (soft < hard) { soft++; -- cgit v1.2.3 From d3ba5a9a345b1243276f8a982e1bce557c2504fd Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Fri, 26 May 2017 12:03:11 +0300 Subject: posix-timers: Make posix_clocks immutable There are no more modular users providing a posix clock. The register function is now pointless so the posix clock array can be initialized statically at compile time and the array including the various k_clock structs can be marked 'const'. Inspired by changes in the Grsecurity patch set, but done proper. [ tglx: Massaged changelog and fixed the POSIX_TIMER=n case ] Signed-off-by: Christoph Hellwig Signed-off-by: Thomas Gleixner Cc: Mike Travis Cc: Dimitri Sivanich Link: http://lkml.kernel.org/r/20170526090311.3377-3-hch@lst.de --- kernel/time/alarmtimer.c | 89 ++++++++++--------- kernel/time/posix-clock.c | 2 +- kernel/time/posix-cpu-timers.c | 34 +++----- kernel/time/posix-timers.c | 191 +++++++++++++++++++---------------------- 4 files changed, 146 insertions(+), 170 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index 5cb5b0008d97..4f4cc3509b30 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -307,38 +307,6 @@ static int alarmtimer_resume(struct device *dev) } #endif -static void alarmtimer_freezerset(ktime_t absexp, enum alarmtimer_type type) -{ - struct alarm_base *base; - unsigned long flags; - ktime_t delta; - - switch(type) { - case ALARM_REALTIME: - base = &alarm_bases[ALARM_REALTIME]; - type = ALARM_REALTIME_FREEZER; - break; - case ALARM_BOOTTIME: - base = &alarm_bases[ALARM_BOOTTIME]; - type = ALARM_BOOTTIME_FREEZER; - break; - default: - WARN_ONCE(1, "Invalid alarm type: %d\n", type); - return; - } - - delta = ktime_sub(absexp, base->gettime()); - - spin_lock_irqsave(&freezer_delta_lock, flags); - if (!freezer_delta || (delta < freezer_delta)) { - freezer_delta = delta; - freezer_expires = absexp; - freezer_alarmtype = type; - } - spin_unlock_irqrestore(&freezer_delta_lock, flags); -} - - /** * alarm_init - Initialize an alarm structure * @alarm: ptr to alarm to be initialized @@ -488,6 +456,38 @@ u64 alarm_forward_now(struct alarm *alarm, ktime_t interval) } EXPORT_SYMBOL_GPL(alarm_forward_now); +#ifdef CONFIG_POSIX_TIMERS + +static void alarmtimer_freezerset(ktime_t absexp, enum alarmtimer_type type) +{ + struct alarm_base *base; + unsigned long flags; + ktime_t delta; + + switch(type) { + case ALARM_REALTIME: + base = &alarm_bases[ALARM_REALTIME]; + type = ALARM_REALTIME_FREEZER; + break; + case ALARM_BOOTTIME: + base = &alarm_bases[ALARM_BOOTTIME]; + type = ALARM_BOOTTIME_FREEZER; + break; + default: + WARN_ONCE(1, "Invalid alarm type: %d\n", type); + return; + } + + delta = ktime_sub(absexp, base->gettime()); + + spin_lock_irqsave(&freezer_delta_lock, flags); + if (!freezer_delta || (delta < freezer_delta)) { + freezer_delta = delta; + freezer_expires = absexp; + freezer_alarmtype = type; + } + spin_unlock_irqrestore(&freezer_delta_lock, flags); +} /** * clock2alarm - helper that converts from clockid to alarmtypes @@ -846,6 +846,17 @@ out: return ret; } +const struct k_clock alarm_clock = { + .clock_getres = alarm_clock_getres, + .clock_get = alarm_clock_get, + .timer_create = alarm_timer_create, + .timer_set = alarm_timer_set, + .timer_del = alarm_timer_del, + .timer_get = alarm_timer_get, + .nsleep = alarm_timer_nsleep, +}; +#endif /* CONFIG_POSIX_TIMERS */ + /* Suspend hook structures */ static const struct dev_pm_ops alarmtimer_pm_ops = { @@ -871,23 +882,9 @@ static int __init alarmtimer_init(void) struct platform_device *pdev; int error = 0; int i; - struct k_clock alarm_clock = { - .clock_getres = alarm_clock_getres, - .clock_get = alarm_clock_get, - .timer_create = alarm_timer_create, - .timer_set = alarm_timer_set, - .timer_del = alarm_timer_del, - .timer_get = alarm_timer_get, - .nsleep = alarm_timer_nsleep, - }; alarmtimer_rtc_timer_init(); - if (IS_ENABLED(CONFIG_POSIX_TIMERS)) { - posix_timers_register_clock(CLOCK_REALTIME_ALARM, &alarm_clock); - posix_timers_register_clock(CLOCK_BOOTTIME_ALARM, &alarm_clock); - } - /* Initialize alarm bases */ alarm_bases[ALARM_REALTIME].base_clockid = CLOCK_REALTIME; alarm_bases[ALARM_REALTIME].gettime = &ktime_get_real; diff --git a/kernel/time/posix-clock.c b/kernel/time/posix-clock.c index 31d588d37a17..7e453005e078 100644 --- a/kernel/time/posix-clock.c +++ b/kernel/time/posix-clock.c @@ -434,7 +434,7 @@ static int pc_timer_settime(struct k_itimer *kit, int flags, return err; } -struct k_clock clock_posix_dynamic = { +const struct k_clock clock_posix_dynamic = { .clock_getres = pc_clock_getres, .clock_set = pc_clock_settime, .clock_get = pc_clock_gettime, diff --git a/kernel/time/posix-cpu-timers.c b/kernel/time/posix-cpu-timers.c index 1370f067fb51..1a522b39f19d 100644 --- a/kernel/time/posix-cpu-timers.c +++ b/kernel/time/posix-cpu-timers.c @@ -1413,7 +1413,7 @@ static int thread_cpu_timer_create(struct k_itimer *timer) return posix_cpu_timer_create(timer); } -struct k_clock clock_posix_cpu = { +const struct k_clock clock_posix_cpu = { .clock_getres = posix_cpu_clock_getres, .clock_set = posix_cpu_clock_set, .clock_get = posix_cpu_clock_get, @@ -1425,24 +1425,16 @@ struct k_clock clock_posix_cpu = { .timer_get = posix_cpu_timer_get, }; -static __init int init_posix_cpu_timers(void) -{ - struct k_clock process = { - .clock_getres = process_cpu_clock_getres, - .clock_get = process_cpu_clock_get, - .timer_create = process_cpu_timer_create, - .nsleep = process_cpu_nsleep, - .nsleep_restart = process_cpu_nsleep_restart, - }; - struct k_clock thread = { - .clock_getres = thread_cpu_clock_getres, - .clock_get = thread_cpu_clock_get, - .timer_create = thread_cpu_timer_create, - }; - - posix_timers_register_clock(CLOCK_PROCESS_CPUTIME_ID, &process); - posix_timers_register_clock(CLOCK_THREAD_CPUTIME_ID, &thread); +const struct k_clock clock_process = { + .clock_getres = process_cpu_clock_getres, + .clock_get = process_cpu_clock_get, + .timer_create = process_cpu_timer_create, + .nsleep = process_cpu_nsleep, + .nsleep_restart = process_cpu_nsleep_restart, +}; - return 0; -} -__initcall(init_posix_cpu_timers); +const struct k_clock clock_thread = { + .clock_getres = thread_cpu_clock_getres, + .clock_get = thread_cpu_clock_get, + .timer_create = thread_cpu_timer_create, +}; diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c index 4d7b2ce09c27..0c0cccfa3586 100644 --- a/kernel/time/posix-timers.c +++ b/kernel/time/posix-timers.c @@ -125,8 +125,6 @@ static DEFINE_SPINLOCK(hash_lock); * which we beg off on and pass to do_sys_settimeofday(). */ -static struct k_clock posix_clocks[MAX_CLOCKS]; - /* * These ones are defined below. */ @@ -280,74 +278,87 @@ static int posix_get_hrtimer_res(clockid_t which_clock, struct timespec64 *tp) return 0; } + +static const struct k_clock clock_realtime = { + .clock_getres = posix_get_hrtimer_res, + .clock_get = posix_clock_realtime_get, + .clock_set = posix_clock_realtime_set, + .clock_adj = posix_clock_realtime_adj, + .nsleep = common_nsleep, + .nsleep_restart = hrtimer_nanosleep_restart, + .timer_create = common_timer_create, + .timer_set = common_timer_set, + .timer_get = common_timer_get, + .timer_del = common_timer_del, +}; + +static const struct k_clock clock_monotonic = { + .clock_getres = posix_get_hrtimer_res, + .clock_get = posix_ktime_get_ts, + .nsleep = common_nsleep, + .nsleep_restart = hrtimer_nanosleep_restart, + .timer_create = common_timer_create, + .timer_set = common_timer_set, + .timer_get = common_timer_get, + .timer_del = common_timer_del, +}; + +static const struct k_clock clock_monotonic_raw = { + .clock_getres = posix_get_hrtimer_res, + .clock_get = posix_get_monotonic_raw, +}; + +static const struct k_clock clock_realtime_coarse = { + .clock_getres = posix_get_coarse_res, + .clock_get = posix_get_realtime_coarse, +}; + +static const struct k_clock clock_monotonic_coarse = { + .clock_getres = posix_get_coarse_res, + .clock_get = posix_get_monotonic_coarse, +}; + +static const struct k_clock clock_tai = { + .clock_getres = posix_get_hrtimer_res, + .clock_get = posix_get_tai, + .nsleep = common_nsleep, + .nsleep_restart = hrtimer_nanosleep_restart, + .timer_create = common_timer_create, + .timer_set = common_timer_set, + .timer_get = common_timer_get, + .timer_del = common_timer_del, +}; + +static const struct k_clock clock_boottime = { + .clock_getres = posix_get_hrtimer_res, + .clock_get = posix_get_boottime, + .nsleep = common_nsleep, + .nsleep_restart = hrtimer_nanosleep_restart, + .timer_create = common_timer_create, + .timer_set = common_timer_set, + .timer_get = common_timer_get, + .timer_del = common_timer_del, +}; + +static const struct k_clock * const posix_clocks[] = { + [CLOCK_REALTIME] = &clock_realtime, + [CLOCK_MONOTONIC] = &clock_monotonic, + [CLOCK_PROCESS_CPUTIME_ID] = &clock_process, + [CLOCK_THREAD_CPUTIME_ID] = &clock_thread, + [CLOCK_MONOTONIC_RAW] = &clock_monotonic_raw, + [CLOCK_REALTIME_COARSE] = &clock_realtime_coarse, + [CLOCK_MONOTONIC_COARSE] = &clock_monotonic_coarse, + [CLOCK_BOOTTIME] = &clock_boottime, + [CLOCK_REALTIME_ALARM] = &alarm_clock, + [CLOCK_BOOTTIME_ALARM] = &alarm_clock, + [CLOCK_TAI] = &clock_tai, +}; + /* * Initialize everything, well, just everything in Posix clocks/timers ;) */ static __init int init_posix_timers(void) { - struct k_clock clock_realtime = { - .clock_getres = posix_get_hrtimer_res, - .clock_get = posix_clock_realtime_get, - .clock_set = posix_clock_realtime_set, - .clock_adj = posix_clock_realtime_adj, - .nsleep = common_nsleep, - .nsleep_restart = hrtimer_nanosleep_restart, - .timer_create = common_timer_create, - .timer_set = common_timer_set, - .timer_get = common_timer_get, - .timer_del = common_timer_del, - }; - struct k_clock clock_monotonic = { - .clock_getres = posix_get_hrtimer_res, - .clock_get = posix_ktime_get_ts, - .nsleep = common_nsleep, - .nsleep_restart = hrtimer_nanosleep_restart, - .timer_create = common_timer_create, - .timer_set = common_timer_set, - .timer_get = common_timer_get, - .timer_del = common_timer_del, - }; - struct k_clock clock_monotonic_raw = { - .clock_getres = posix_get_hrtimer_res, - .clock_get = posix_get_monotonic_raw, - }; - struct k_clock clock_realtime_coarse = { - .clock_getres = posix_get_coarse_res, - .clock_get = posix_get_realtime_coarse, - }; - struct k_clock clock_monotonic_coarse = { - .clock_getres = posix_get_coarse_res, - .clock_get = posix_get_monotonic_coarse, - }; - struct k_clock clock_tai = { - .clock_getres = posix_get_hrtimer_res, - .clock_get = posix_get_tai, - .nsleep = common_nsleep, - .nsleep_restart = hrtimer_nanosleep_restart, - .timer_create = common_timer_create, - .timer_set = common_timer_set, - .timer_get = common_timer_get, - .timer_del = common_timer_del, - }; - struct k_clock clock_boottime = { - .clock_getres = posix_get_hrtimer_res, - .clock_get = posix_get_boottime, - .nsleep = common_nsleep, - .nsleep_restart = hrtimer_nanosleep_restart, - .timer_create = common_timer_create, - .timer_set = common_timer_set, - .timer_get = common_timer_get, - .timer_del = common_timer_del, - }; - - posix_timers_register_clock(CLOCK_REALTIME, &clock_realtime); - posix_timers_register_clock(CLOCK_MONOTONIC, &clock_monotonic); - posix_timers_register_clock(CLOCK_MONOTONIC_RAW, &clock_monotonic_raw); - posix_timers_register_clock(CLOCK_REALTIME_COARSE, &clock_realtime_coarse); - posix_timers_register_clock(CLOCK_MONOTONIC_COARSE, &clock_monotonic_coarse); - posix_timers_register_clock(CLOCK_BOOTTIME, &clock_boottime); - posix_timers_register_clock(CLOCK_TAI, &clock_tai); - posix_timers_cache = kmem_cache_create("posix_timers_cache", sizeof (struct k_itimer), 0, SLAB_PANIC, NULL); @@ -521,30 +532,6 @@ static struct pid *good_sigevent(sigevent_t * event) return task_pid(rtn); } -void posix_timers_register_clock(const clockid_t clock_id, - struct k_clock *new_clock) -{ - if ((unsigned) clock_id >= MAX_CLOCKS) { - printk(KERN_WARNING "POSIX clock register failed for clock_id %d\n", - clock_id); - return; - } - - if (!new_clock->clock_get) { - printk(KERN_WARNING "POSIX clock id %d lacks clock_get()\n", - clock_id); - return; - } - if (!new_clock->clock_getres) { - printk(KERN_WARNING "POSIX clock id %d lacks clock_getres()\n", - clock_id); - return; - } - - posix_clocks[clock_id] = *new_clock; -} -EXPORT_SYMBOL_GPL(posix_timers_register_clock); - static struct k_itimer * alloc_posix_timer(void) { struct k_itimer *tmr; @@ -581,15 +568,15 @@ static void release_posix_timer(struct k_itimer *tmr, int it_id_set) call_rcu(&tmr->it.rcu, k_itimer_rcu_free); } -static struct k_clock *clockid_to_kclock(const clockid_t id) +static const struct k_clock *clockid_to_kclock(const clockid_t id) { if (id < 0) return (id & CLOCKFD_MASK) == CLOCKFD ? &clock_posix_dynamic : &clock_posix_cpu; - if (id >= MAX_CLOCKS || !posix_clocks[id].clock_getres) + if (id >= ARRAY_SIZE(posix_clocks) || !posix_clocks[id]) return NULL; - return &posix_clocks[id]; + return posix_clocks[id]; } static int common_timer_create(struct k_itimer *new_timer) @@ -604,7 +591,7 @@ SYSCALL_DEFINE3(timer_create, const clockid_t, which_clock, struct sigevent __user *, timer_event_spec, timer_t __user *, created_timer_id) { - struct k_clock *kc = clockid_to_kclock(which_clock); + const struct k_clock *kc = clockid_to_kclock(which_clock); struct k_itimer *new_timer; int error, new_timer_id; sigevent_t event; @@ -781,7 +768,7 @@ SYSCALL_DEFINE2(timer_gettime, timer_t, timer_id, struct itimerspec64 cur_setting64; struct itimerspec cur_setting; struct k_itimer *timr; - struct k_clock *kc; + const struct k_clock *kc; unsigned long flags; int ret = 0; @@ -890,7 +877,7 @@ SYSCALL_DEFINE4(timer_settime, timer_t, timer_id, int, flags, struct itimerspec new_spec, old_spec; struct k_itimer *timr; unsigned long flag; - struct k_clock *kc; + const struct k_clock *kc; int error = 0; if (!new_setting) @@ -939,7 +926,7 @@ static int common_timer_del(struct k_itimer *timer) static inline int timer_delete_hook(struct k_itimer *timer) { - struct k_clock *kc = clockid_to_kclock(timer->it_clock); + const struct k_clock *kc = clockid_to_kclock(timer->it_clock); if (WARN_ON_ONCE(!kc || !kc->timer_del)) return -EINVAL; @@ -1018,7 +1005,7 @@ void exit_itimers(struct signal_struct *sig) SYSCALL_DEFINE2(clock_settime, const clockid_t, which_clock, const struct timespec __user *, tp) { - struct k_clock *kc = clockid_to_kclock(which_clock); + const struct k_clock *kc = clockid_to_kclock(which_clock); struct timespec64 new_tp64; struct timespec new_tp; @@ -1035,7 +1022,7 @@ SYSCALL_DEFINE2(clock_settime, const clockid_t, which_clock, SYSCALL_DEFINE2(clock_gettime, const clockid_t, which_clock, struct timespec __user *,tp) { - struct k_clock *kc = clockid_to_kclock(which_clock); + const struct k_clock *kc = clockid_to_kclock(which_clock); struct timespec64 kernel_tp64; struct timespec kernel_tp; int error; @@ -1055,7 +1042,7 @@ SYSCALL_DEFINE2(clock_gettime, const clockid_t, which_clock, SYSCALL_DEFINE2(clock_adjtime, const clockid_t, which_clock, struct timex __user *, utx) { - struct k_clock *kc = clockid_to_kclock(which_clock); + const struct k_clock *kc = clockid_to_kclock(which_clock); struct timex ktx; int err; @@ -1078,7 +1065,7 @@ SYSCALL_DEFINE2(clock_adjtime, const clockid_t, which_clock, SYSCALL_DEFINE2(clock_getres, const clockid_t, which_clock, struct timespec __user *, tp) { - struct k_clock *kc = clockid_to_kclock(which_clock); + const struct k_clock *kc = clockid_to_kclock(which_clock); struct timespec64 rtn_tp64; struct timespec rtn_tp; int error; @@ -1110,7 +1097,7 @@ SYSCALL_DEFINE4(clock_nanosleep, const clockid_t, which_clock, int, flags, const struct timespec __user *, rqtp, struct timespec __user *, rmtp) { - struct k_clock *kc = clockid_to_kclock(which_clock); + const struct k_clock *kc = clockid_to_kclock(which_clock); struct timespec64 t64; struct timespec t; @@ -1136,7 +1123,7 @@ SYSCALL_DEFINE4(clock_nanosleep, const clockid_t, which_clock, int, flags, long clock_nanosleep_restart(struct restart_block *restart_block) { clockid_t which_clock = restart_block->nanosleep.clockid; - struct k_clock *kc = clockid_to_kclock(which_clock); + const struct k_clock *kc = clockid_to_kclock(which_clock); if (WARN_ON_ONCE(!kc || !kc->nsleep_restart)) return -EINVAL; -- cgit v1.2.3 From b6b3b80fceb175c825ad6c72659e0a72e201fc5f Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Sat, 27 May 2017 12:23:47 +0200 Subject: alarmtimer: Fix posix-timer constification fallout Some freezer related variables are only used when either CONFIG_POSIX_TIMER or CONFIG_RTC_CLASS are enabled. Hide them when both are off. Fixes: d3ba5a9a345b ("posix-timers: Make posix_clocks immutable") Reported-by: Ingo Molnar Signed-off-by: Thomas Gleixner Cc: Christoph Helwig --- kernel/time/alarmtimer.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'kernel/time') diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index 4f4cc3509b30..4cfebfff848d 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -45,11 +45,13 @@ static struct alarm_base { clockid_t base_clockid; } alarm_bases[ALARM_NUMTYPE]; +#if defined(CONFIG_POSIX_TIMERS) || defined(CONFIG_RTC_CLASS) /* freezer information to handle clock_nanosleep triggered wakeups */ static enum alarmtimer_type freezer_alarmtype; static ktime_t freezer_expires; static ktime_t freezer_delta; static DEFINE_SPINLOCK(freezer_delta_lock); +#endif static struct wakeup_source *ws; -- cgit v1.2.3 From 7c25904508afef134d7a9d2ad1690ee6554a1faa Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Mon, 15 May 2017 14:56:50 +0200 Subject: nohz: Reset next_tick cache even when the timer has no regs Handle tick interrupts whose regs are NULL, out of general paranoia. It happens when hrtimer_interrupt() is called from non-interrupt contexts, such as hotplug CPU down events. Signed-off-by: Frederic Weisbecker Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar --- kernel/time/tick-sched.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'kernel/time') diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c index 764d2905e6a5..e3043873fcdc 100644 --- a/kernel/time/tick-sched.c +++ b/kernel/time/tick-sched.c @@ -1202,6 +1202,8 @@ static enum hrtimer_restart tick_sched_timer(struct hrtimer *timer) */ if (regs) tick_sched_handle(ts, regs); + else + ts->next_tick = 0; /* No need to reprogram if we are in idle or full dynticks mode */ if (unlikely(ts->tick_stopped)) -- cgit v1.2.3 From f4781e76f90df7aec400635d73ea4c35ee1d4765 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 30 May 2017 23:15:34 +0200 Subject: alarmtimer: Prevent overflow of relative timers Andrey reported a alartimer related RCU stall while fuzzing the kernel with syzkaller. The reason for this is an overflow in ktime_add() which brings the resulting time into negative space and causes immediate expiry of the timer. The following rearm with a small interval does not bring the timer back into positive space due to the same issue. This results in a permanent firing alarmtimer which hogs the CPU. Use ktime_add_safe() instead which detects the overflow and clamps the result to KTIME_SEC_MAX. Reported-by: Andrey Konovalov Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: Kostya Serebryany Cc: syzkaller Cc: John Stultz Cc: Dmitry Vyukov Cc: stable@vger.kernel.org Link: http://lkml.kernel.org/r/20170530211655.802921648@linutronix.de --- kernel/time/alarmtimer.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index 5cb5b0008d97..2b2e032b4eb4 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -387,7 +387,7 @@ void alarm_start_relative(struct alarm *alarm, ktime_t start) { struct alarm_base *base = &alarm_bases[alarm->type]; - start = ktime_add(start, base->gettime()); + start = ktime_add_safe(start, base->gettime()); alarm_start(alarm, start); } EXPORT_SYMBOL_GPL(alarm_start_relative); @@ -475,7 +475,7 @@ u64 alarm_forward(struct alarm *alarm, ktime_t now, ktime_t interval) overrun++; } - alarm->node.expires = ktime_add(alarm->node.expires, interval); + alarm->node.expires = ktime_add_safe(alarm->node.expires, interval); return overrun; } EXPORT_SYMBOL_GPL(alarm_forward); @@ -666,7 +666,7 @@ static int alarm_timer_set(struct k_itimer *timr, int flags, ktime_t now; now = alarm_bases[timr->it.alarm.alarmtimer.type].gettime(); - exp = ktime_add(now, exp); + exp = ktime_add_safe(now, exp); } alarm_start(&timr->it.alarm.alarmtimer, exp); -- cgit v1.2.3 From ff86bf0c65f14346bf2440534f9ba5ac232c39a0 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 30 May 2017 23:15:35 +0200 Subject: alarmtimer: Rate limit periodic intervals The alarmtimer code has another source of potentially rearming itself too fast. Interval timers with a very samll interval have a similar CPU hog effect as the previously fixed overflow issue. The reason is that alarmtimers do not implement the normal protection against this kind of problem which the other posix timer use: timer expires -> queue signal -> deliver signal -> rearm timer This scheme brings the rearming under scheduler control and prevents permanently firing timers which hog the CPU. Bringing this scheme to the alarm timer code is a major overhaul because it lacks all the necessary mechanisms completely. So for a quick fix limit the interval to one jiffie. This is not problematic in practice as alarmtimers are usually backed by an RTC for suspend which have 1 second resolution. It could be therefor argued that the resolution of this clock should be set to 1 second in general, but that's outside the scope of this fix. Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: Kostya Serebryany Cc: syzkaller Cc: John Stultz Cc: Dmitry Vyukov Cc: stable@vger.kernel.org Link: http://lkml.kernel.org/r/20170530211655.896767100@linutronix.de --- kernel/time/alarmtimer.c | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'kernel/time') diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index 2b2e032b4eb4..ee2f4202d82a 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -660,6 +660,14 @@ static int alarm_timer_set(struct k_itimer *timr, int flags, /* start the timer */ timr->it.alarm.interval = timespec64_to_ktime(new_setting->it_interval); + + /* + * Rate limit to the tick as a hot fix to prevent DOS. Will be + * mopped up later. + */ + if (timr->it.alarm.interval < TICK_NSEC) + timr->it.alarm.interval = TICK_NSEC; + exp = timespec64_to_ktime(new_setting->it_value); /* Convert (if necessary) to absolute time */ if (flags != TIMER_ABSTIME) { -- cgit v1.2.3 From 18c700c4e3d0a37c43a2df8b8f740121d4dac645 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 30 May 2017 23:15:36 +0200 Subject: alarmtimer: Remove pointless config conditional Having a IF_ENABLED(CONFIG_POSIX_TIMERS) inside of a #ifdef CONFIG_POSIX_TIMERS section is pointless. Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: John Stultz Link: http://lkml.kernel.org/r/20170530211655.975218056@linutronix.de --- kernel/time/alarmtimer.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index e645dcc7d4ee..2a8675f9aac5 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -520,8 +520,7 @@ static enum alarmtimer_restart alarm_handle_timer(struct alarm *alarm, spin_lock_irqsave(&ptr->it_lock, flags); if ((ptr->it_sigev_notify & ~SIGEV_THREAD_ID) != SIGEV_NONE) { - if (IS_ENABLED(CONFIG_POSIX_TIMERS) && - posix_timer_event(ptr, 0) != 0) + if (posix_timer_event(ptr, 0)) ptr->it_overrun++; } -- cgit v1.2.3 From a81129e5a189973abd661704b261f8aad9325407 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 30 May 2017 23:15:37 +0200 Subject: posix-timers: Remove unused export of posix_timer_event() Since the removal of the mmtimer driver the export is not longer needed. Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: John Stultz Link: http://lkml.kernel.org/r/20170530211656.052744418@linutronix.de --- kernel/time/posix-timers.c | 1 - 1 file changed, 1 deletion(-) (limited to 'kernel/time') diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c index 0c0cccfa3586..44d486590e6e 100644 --- a/kernel/time/posix-timers.c +++ b/kernel/time/posix-timers.c @@ -442,7 +442,6 @@ int posix_timer_event(struct k_itimer *timr, int si_private) /* If we failed to send the signal the timer stops. */ return ret > 0; } -EXPORT_SYMBOL_GPL(posix_timer_event); /* * This function gets called when a POSIX.1b interval timer expires. It -- cgit v1.2.3 From 3a06c7ac24f9f24ec059cd77c2dbdf7fbfd0aaaf Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 30 May 2017 23:15:38 +0200 Subject: posix-clocks: Remove interval timer facility and mmap/fasync callbacks The only user of this facility is ptp_clock, which does not implement any of those functions. Remove them to prevent accidental users. Especially the interval timer interfaces are now more or less impossible to implement because the necessary infrastructure has been confined to the core code. Aside of that it's really complex to make these callbacks implemented according to spec as the alarm timer implementation demonstrates. If at all then a nanosleep callback might be a reasonable extension. For now keep just what ptp_clock needs. Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: John Stultz Link: http://lkml.kernel.org/r/20170530211656.145036286@linutronix.de --- kernel/time/posix-clock.c | 113 ---------------------------------------------- 1 file changed, 113 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/posix-clock.c b/kernel/time/posix-clock.c index 7e453005e078..bd4fb785652f 100644 --- a/kernel/time/posix-clock.c +++ b/kernel/time/posix-clock.c @@ -82,38 +82,6 @@ static unsigned int posix_clock_poll(struct file *fp, poll_table *wait) return result; } -static int posix_clock_fasync(int fd, struct file *fp, int on) -{ - struct posix_clock *clk = get_posix_clock(fp); - int err = 0; - - if (!clk) - return -ENODEV; - - if (clk->ops.fasync) - err = clk->ops.fasync(clk, fd, fp, on); - - put_posix_clock(clk); - - return err; -} - -static int posix_clock_mmap(struct file *fp, struct vm_area_struct *vma) -{ - struct posix_clock *clk = get_posix_clock(fp); - int err = -ENODEV; - - if (!clk) - return -ENODEV; - - if (clk->ops.mmap) - err = clk->ops.mmap(clk, vma); - - put_posix_clock(clk); - - return err; -} - static long posix_clock_ioctl(struct file *fp, unsigned int cmd, unsigned long arg) { @@ -199,8 +167,6 @@ static const struct file_operations posix_clock_file_operations = { .unlocked_ioctl = posix_clock_ioctl, .open = posix_clock_open, .release = posix_clock_release, - .fasync = posix_clock_fasync, - .mmap = posix_clock_mmap, #ifdef CONFIG_COMPAT .compat_ioctl = posix_clock_compat_ioctl, #endif @@ -359,88 +325,9 @@ out: return err; } -static int pc_timer_create(struct k_itimer *kit) -{ - clockid_t id = kit->it_clock; - struct posix_clock_desc cd; - int err; - - err = get_clock_desc(id, &cd); - if (err) - return err; - - if (cd.clk->ops.timer_create) - err = cd.clk->ops.timer_create(cd.clk, kit); - else - err = -EOPNOTSUPP; - - put_clock_desc(&cd); - - return err; -} - -static int pc_timer_delete(struct k_itimer *kit) -{ - clockid_t id = kit->it_clock; - struct posix_clock_desc cd; - int err; - - err = get_clock_desc(id, &cd); - if (err) - return err; - - if (cd.clk->ops.timer_delete) - err = cd.clk->ops.timer_delete(cd.clk, kit); - else - err = -EOPNOTSUPP; - - put_clock_desc(&cd); - - return err; -} - -static void pc_timer_gettime(struct k_itimer *kit, struct itimerspec64 *ts) -{ - clockid_t id = kit->it_clock; - struct posix_clock_desc cd; - - if (get_clock_desc(id, &cd)) - return; - - if (cd.clk->ops.timer_gettime) - cd.clk->ops.timer_gettime(cd.clk, kit, ts); - - put_clock_desc(&cd); -} - -static int pc_timer_settime(struct k_itimer *kit, int flags, - struct itimerspec64 *ts, struct itimerspec64 *old) -{ - clockid_t id = kit->it_clock; - struct posix_clock_desc cd; - int err; - - err = get_clock_desc(id, &cd); - if (err) - return err; - - if (cd.clk->ops.timer_settime) - err = cd.clk->ops.timer_settime(cd.clk, kit, flags, ts, old); - else - err = -EOPNOTSUPP; - - put_clock_desc(&cd); - - return err; -} - const struct k_clock clock_posix_dynamic = { .clock_getres = pc_clock_getres, .clock_set = pc_clock_settime, .clock_get = pc_clock_gettime, .clock_adj = pc_clock_adjtime, - .timer_create = pc_timer_create, - .timer_set = pc_timer_settime, - .timer_del = pc_timer_delete, - .timer_get = pc_timer_gettime, }; -- cgit v1.2.3 From 6631fa12c105e326bbe5fb215eb216e86c90d1ba Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 30 May 2017 23:15:39 +0200 Subject: posix-timers: Avoid gazillions of forward declarations Move it below the actual implementations as there are new callbacks coming which would require even more forward declarations. Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: John Stultz Link: http://lkml.kernel.org/r/20170530211656.238209952@linutronix.de --- kernel/time/posix-timers.c | 190 +++++++++++++++++++++------------------------ 1 file changed, 89 insertions(+), 101 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c index 44d486590e6e..b60b655dfbcd 100644 --- a/kernel/time/posix-timers.c +++ b/kernel/time/posix-timers.c @@ -69,6 +69,9 @@ static struct kmem_cache *posix_timers_cache; static DEFINE_HASHTABLE(posix_timers_hashtable, 9); static DEFINE_SPINLOCK(hash_lock); +static const struct k_clock * const posix_clocks[]; +static const struct k_clock *clockid_to_kclock(const clockid_t id); + /* * we assume that the new SIGEV_THREAD_ID shares no bits with the other * SIGEV values. Here we put out an error if this assumption fails. @@ -124,20 +127,6 @@ static DEFINE_SPINLOCK(hash_lock); * have is CLOCK_REALTIME and its high res counter part, both of * which we beg off on and pass to do_sys_settimeofday(). */ - -/* - * These ones are defined below. - */ -static int common_nsleep(const clockid_t, int flags, struct timespec64 *t, - struct timespec __user *rmtp); -static int common_timer_create(struct k_itimer *new_timer); -static void common_timer_get(struct k_itimer *, struct itimerspec64 *); -static int common_timer_set(struct k_itimer *, int, - struct itimerspec64 *, struct itimerspec64 *); -static int common_timer_del(struct k_itimer *timer); - -static enum hrtimer_restart posix_timer_fn(struct hrtimer *data); - static struct k_itimer *__lock_timer(timer_t timer_id, unsigned long *flags); #define lock_timer(tid, flags) \ @@ -278,82 +267,6 @@ static int posix_get_hrtimer_res(clockid_t which_clock, struct timespec64 *tp) return 0; } - -static const struct k_clock clock_realtime = { - .clock_getres = posix_get_hrtimer_res, - .clock_get = posix_clock_realtime_get, - .clock_set = posix_clock_realtime_set, - .clock_adj = posix_clock_realtime_adj, - .nsleep = common_nsleep, - .nsleep_restart = hrtimer_nanosleep_restart, - .timer_create = common_timer_create, - .timer_set = common_timer_set, - .timer_get = common_timer_get, - .timer_del = common_timer_del, -}; - -static const struct k_clock clock_monotonic = { - .clock_getres = posix_get_hrtimer_res, - .clock_get = posix_ktime_get_ts, - .nsleep = common_nsleep, - .nsleep_restart = hrtimer_nanosleep_restart, - .timer_create = common_timer_create, - .timer_set = common_timer_set, - .timer_get = common_timer_get, - .timer_del = common_timer_del, -}; - -static const struct k_clock clock_monotonic_raw = { - .clock_getres = posix_get_hrtimer_res, - .clock_get = posix_get_monotonic_raw, -}; - -static const struct k_clock clock_realtime_coarse = { - .clock_getres = posix_get_coarse_res, - .clock_get = posix_get_realtime_coarse, -}; - -static const struct k_clock clock_monotonic_coarse = { - .clock_getres = posix_get_coarse_res, - .clock_get = posix_get_monotonic_coarse, -}; - -static const struct k_clock clock_tai = { - .clock_getres = posix_get_hrtimer_res, - .clock_get = posix_get_tai, - .nsleep = common_nsleep, - .nsleep_restart = hrtimer_nanosleep_restart, - .timer_create = common_timer_create, - .timer_set = common_timer_set, - .timer_get = common_timer_get, - .timer_del = common_timer_del, -}; - -static const struct k_clock clock_boottime = { - .clock_getres = posix_get_hrtimer_res, - .clock_get = posix_get_boottime, - .nsleep = common_nsleep, - .nsleep_restart = hrtimer_nanosleep_restart, - .timer_create = common_timer_create, - .timer_set = common_timer_set, - .timer_get = common_timer_get, - .timer_del = common_timer_del, -}; - -static const struct k_clock * const posix_clocks[] = { - [CLOCK_REALTIME] = &clock_realtime, - [CLOCK_MONOTONIC] = &clock_monotonic, - [CLOCK_PROCESS_CPUTIME_ID] = &clock_process, - [CLOCK_THREAD_CPUTIME_ID] = &clock_thread, - [CLOCK_MONOTONIC_RAW] = &clock_monotonic_raw, - [CLOCK_REALTIME_COARSE] = &clock_realtime_coarse, - [CLOCK_MONOTONIC_COARSE] = &clock_monotonic_coarse, - [CLOCK_BOOTTIME] = &clock_boottime, - [CLOCK_REALTIME_ALARM] = &alarm_clock, - [CLOCK_BOOTTIME_ALARM] = &alarm_clock, - [CLOCK_TAI] = &clock_tai, -}; - /* * Initialize everything, well, just everything in Posix clocks/timers ;) */ @@ -567,17 +480,6 @@ static void release_posix_timer(struct k_itimer *tmr, int it_id_set) call_rcu(&tmr->it.rcu, k_itimer_rcu_free); } -static const struct k_clock *clockid_to_kclock(const clockid_t id) -{ - if (id < 0) - return (id & CLOCKFD_MASK) == CLOCKFD ? - &clock_posix_dynamic : &clock_posix_cpu; - - if (id >= ARRAY_SIZE(posix_clocks) || !posix_clocks[id]) - return NULL; - return posix_clocks[id]; -} - static int common_timer_create(struct k_itimer *new_timer) { hrtimer_init(&new_timer->it.real.timer, new_timer->it_clock, 0); @@ -1129,3 +1031,89 @@ long clock_nanosleep_restart(struct restart_block *restart_block) return kc->nsleep_restart(restart_block); } + +static const struct k_clock clock_realtime = { + .clock_getres = posix_get_hrtimer_res, + .clock_get = posix_clock_realtime_get, + .clock_set = posix_clock_realtime_set, + .clock_adj = posix_clock_realtime_adj, + .nsleep = common_nsleep, + .nsleep_restart = hrtimer_nanosleep_restart, + .timer_create = common_timer_create, + .timer_set = common_timer_set, + .timer_get = common_timer_get, + .timer_del = common_timer_del, +}; + +static const struct k_clock clock_monotonic = { + .clock_getres = posix_get_hrtimer_res, + .clock_get = posix_ktime_get_ts, + .nsleep = common_nsleep, + .nsleep_restart = hrtimer_nanosleep_restart, + .timer_create = common_timer_create, + .timer_set = common_timer_set, + .timer_get = common_timer_get, + .timer_del = common_timer_del, +}; + +static const struct k_clock clock_monotonic_raw = { + .clock_getres = posix_get_hrtimer_res, + .clock_get = posix_get_monotonic_raw, +}; + +static const struct k_clock clock_realtime_coarse = { + .clock_getres = posix_get_coarse_res, + .clock_get = posix_get_realtime_coarse, +}; + +static const struct k_clock clock_monotonic_coarse = { + .clock_getres = posix_get_coarse_res, + .clock_get = posix_get_monotonic_coarse, +}; + +static const struct k_clock clock_tai = { + .clock_getres = posix_get_hrtimer_res, + .clock_get = posix_get_tai, + .nsleep = common_nsleep, + .nsleep_restart = hrtimer_nanosleep_restart, + .timer_create = common_timer_create, + .timer_set = common_timer_set, + .timer_get = common_timer_get, + .timer_del = common_timer_del, +}; + +static const struct k_clock clock_boottime = { + .clock_getres = posix_get_hrtimer_res, + .clock_get = posix_get_boottime, + .nsleep = common_nsleep, + .nsleep_restart = hrtimer_nanosleep_restart, + .timer_create = common_timer_create, + .timer_set = common_timer_set, + .timer_get = common_timer_get, + .timer_del = common_timer_del, +}; + +static const struct k_clock * const posix_clocks[] = { + [CLOCK_REALTIME] = &clock_realtime, + [CLOCK_MONOTONIC] = &clock_monotonic, + [CLOCK_PROCESS_CPUTIME_ID] = &clock_process, + [CLOCK_THREAD_CPUTIME_ID] = &clock_thread, + [CLOCK_MONOTONIC_RAW] = &clock_monotonic_raw, + [CLOCK_REALTIME_COARSE] = &clock_realtime_coarse, + [CLOCK_MONOTONIC_COARSE] = &clock_monotonic_coarse, + [CLOCK_BOOTTIME] = &clock_boottime, + [CLOCK_REALTIME_ALARM] = &alarm_clock, + [CLOCK_BOOTTIME_ALARM] = &alarm_clock, + [CLOCK_TAI] = &clock_tai, +}; + +static const struct k_clock *clockid_to_kclock(const clockid_t id) +{ + if (id < 0) + return (id & CLOCKFD_MASK) == CLOCKFD ? + &clock_posix_dynamic : &clock_posix_cpu; + + if (id >= ARRAY_SIZE(posix_clocks) || !posix_clocks[id]) + return NULL; + return posix_clocks[id]; +} -- cgit v1.2.3 From bab0aae9dcba9466dcc968b8bd21914f8f691631 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 30 May 2017 23:15:41 +0200 Subject: posix-timers: Move posix-timer internals to core None of these declarations is required outside of kernel/time. Move them to an internal header. Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: John Stultz Cc: Christoph Hellwig Link: http://lkml.kernel.org/r/20170530211656.394803853@linutronix.de --- kernel/time/alarmtimer.c | 2 ++ kernel/time/posix-clock.c | 2 ++ kernel/time/posix-cpu-timers.c | 2 ++ kernel/time/posix-timers.c | 1 + kernel/time/posix-timers.h | 29 +++++++++++++++++++++++++++++ 5 files changed, 36 insertions(+) create mode 100644 kernel/time/posix-timers.h (limited to 'kernel/time') diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index 2a8675f9aac5..36855d675da5 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -28,6 +28,8 @@ #include #include +#include "posix-timers.h" + #define CREATE_TRACE_POINTS #include diff --git a/kernel/time/posix-clock.c b/kernel/time/posix-clock.c index bd4fb785652f..17cdc554c9fe 100644 --- a/kernel/time/posix-clock.c +++ b/kernel/time/posix-clock.c @@ -25,6 +25,8 @@ #include #include +#include "posix-timers.h" + static void delete_clock(struct kref *kref); /* diff --git a/kernel/time/posix-cpu-timers.c b/kernel/time/posix-cpu-timers.c index c99434739fd5..a77a792f2570 100644 --- a/kernel/time/posix-cpu-timers.c +++ b/kernel/time/posix-cpu-timers.c @@ -13,6 +13,8 @@ #include #include +#include "posix-timers.h" + /* * Called after updating RLIMIT_CPU to run cpu timer and update * tsk->signal->cputime_expires expiration cache if necessary. Needs diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c index b60b655dfbcd..dee6a0d911d4 100644 --- a/kernel/time/posix-timers.c +++ b/kernel/time/posix-timers.c @@ -51,6 +51,7 @@ #include #include "timekeeping.h" +#include "posix-timers.h" /* * Management arrays for POSIX timers. Timers are now kept in static hash table diff --git a/kernel/time/posix-timers.h b/kernel/time/posix-timers.h new file mode 100644 index 000000000000..ad2dbd29b389 --- /dev/null +++ b/kernel/time/posix-timers.h @@ -0,0 +1,29 @@ +#define TIMER_RETRY 1 + +struct k_clock { + int (*clock_getres) (const clockid_t which_clock, struct timespec64 *tp); + int (*clock_set) (const clockid_t which_clock, + const struct timespec64 *tp); + int (*clock_get) (const clockid_t which_clock, struct timespec64 *tp); + int (*clock_adj) (const clockid_t which_clock, struct timex *tx); + int (*timer_create) (struct k_itimer *timer); + int (*nsleep) (const clockid_t which_clock, int flags, + struct timespec64 *, struct timespec __user *); + long (*nsleep_restart) (struct restart_block *restart_block); + int (*timer_set) (struct k_itimer *timr, int flags, + struct itimerspec64 *new_setting, + struct itimerspec64 *old_setting); + int (*timer_del) (struct k_itimer *timr); + void (*timer_get) (struct k_itimer *timr, + struct itimerspec64 *cur_setting); +}; + +extern const struct k_clock clock_posix_cpu; +extern const struct k_clock clock_posix_dynamic; +extern const struct k_clock clock_process; +extern const struct k_clock clock_thread; +extern const struct k_clock alarm_clock; + +int posix_timer_event(struct k_itimer *timr, int si_private); + +void posix_cpu_timer_schedule(struct k_itimer *timer); -- cgit v1.2.3 From af888d677a3f4473c198b4720319dd037f398b51 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 30 May 2017 23:15:42 +0200 Subject: posix-timers: Unify overrun/requeue_pending handling hrtimer based posix-timers and posix-cpu-timers handle the update of the rearming and overflow related status fields differently. Move that update to the common rearming code. Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: John Stultz Link: http://lkml.kernel.org/r/20170530211656.484936964@linutronix.de --- kernel/time/posix-cpu-timers.c | 18 +++++++----------- kernel/time/posix-timers.c | 15 ++++++++------- 2 files changed, 15 insertions(+), 18 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/posix-cpu-timers.c b/kernel/time/posix-cpu-timers.c index a77a792f2570..1683e503179e 100644 --- a/kernel/time/posix-cpu-timers.c +++ b/kernel/time/posix-cpu-timers.c @@ -527,6 +527,7 @@ static void cpu_timer_fire(struct k_itimer *timer) * ticking in case the signal is deliverable next time. */ posix_cpu_timer_schedule(timer); + ++timer->it_requeue_pending; } } @@ -997,12 +998,12 @@ void posix_cpu_timer_schedule(struct k_itimer *timer) cpu_clock_sample(timer->it_clock, p, &now); bump_cpu_timer(timer, now); if (unlikely(p->exit_state)) - goto out; + return; /* Protect timer list r/w in arm_timer() */ sighand = lock_task_sighand(p, &flags); if (!sighand) - goto out; + return; } else { /* * Protect arm_timer() and timer sampling in case of call to @@ -1015,11 +1016,10 @@ void posix_cpu_timer_schedule(struct k_itimer *timer) * We can't even collect a sample any more. */ timer->it.cpu.expires = 0; - goto out; + return; } else if (unlikely(p->exit_state) && thread_group_empty(p)) { - unlock_task_sighand(p, &flags); - /* Optimizations: if the process is dying, no need to rearm */ - goto out; + /* If the process is dying, no need to rearm */ + goto unlock; } cpu_timer_sample_group(timer->it_clock, p, &now); bump_cpu_timer(timer, now); @@ -1031,12 +1031,8 @@ void posix_cpu_timer_schedule(struct k_itimer *timer) */ WARN_ON_ONCE(!irqs_disabled()); arm_timer(timer); +unlock: unlock_task_sighand(p, &flags); - -out: - timer->it_overrun_last = timer->it_overrun; - timer->it_overrun = -1; - ++timer->it_requeue_pending; } /** diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c index dee6a0d911d4..79a00e0f1ef9 100644 --- a/kernel/time/posix-timers.c +++ b/kernel/time/posix-timers.c @@ -291,10 +291,6 @@ static void schedule_next_timer(struct k_itimer *timr) timr->it_overrun += (unsigned int) hrtimer_forward(timer, timer->base->get_time(), timr->it.real.interval); - - timr->it_overrun_last = timr->it_overrun; - timr->it_overrun = -1; - ++timr->it_requeue_pending; hrtimer_restart(timer); } @@ -315,18 +311,23 @@ void do_schedule_next_timer(struct siginfo *info) unsigned long flags; timr = lock_timer(info->si_tid, &flags); + if (!timr) + return; - if (timr && timr->it_requeue_pending == info->si_sys_private) { + if (timr->it_requeue_pending == info->si_sys_private) { if (timr->it_clock < 0) posix_cpu_timer_schedule(timr); else schedule_next_timer(timr); + timr->it_overrun_last = timr->it_overrun; + timr->it_overrun = -1; + ++timr->it_requeue_pending; + info->si_overrun += timr->it_overrun_last; } - if (timr) - unlock_timer(timr, flags); + unlock_timer(timr, flags); } int posix_timer_event(struct k_itimer *timr, int si_private) -- cgit v1.2.3 From 80105cd0e62ba8a2caf8eebd52f42952c7c04046 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 30 May 2017 23:15:43 +0200 Subject: posix-timers: Move interval out of the union Preparatory patch to unify the alarm timer and hrtimer based posix interval timer handling. The interval is used as a criteria for rearming decisions so moving it out of the clock specific data structures allows later unification. Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: John Stultz Link: http://lkml.kernel.org/r/20170530211656.563922908@linutronix.de --- kernel/time/alarmtimer.c | 13 ++++++------- kernel/time/posix-timers.c | 20 ++++++++++---------- 2 files changed, 16 insertions(+), 17 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index 36855d675da5..5b8cf4b61854 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -527,9 +527,8 @@ static enum alarmtimer_restart alarm_handle_timer(struct alarm *alarm, } /* Re-add periodic timers */ - if (ptr->it.alarm.interval) { - ptr->it_overrun += alarm_forward(alarm, now, - ptr->it.alarm.interval); + if (ptr->it_interval) { + ptr->it_overrun += alarm_forward(alarm, now, ptr->it_interval); result = ALARMTIMER_RESTART; } spin_unlock_irqrestore(&ptr->it_lock, flags); @@ -613,7 +612,7 @@ static void alarm_timer_get(struct k_itimer *timr, cur_setting->it_value.tv_nsec = 0; } - cur_setting->it_interval = ktime_to_timespec64(timr->it.alarm.interval); + cur_setting->it_interval = ktime_to_timespec64(timr->it_interval); } /** @@ -662,14 +661,14 @@ static int alarm_timer_set(struct k_itimer *timr, int flags, return TIMER_RETRY; /* start the timer */ - timr->it.alarm.interval = timespec64_to_ktime(new_setting->it_interval); + timr->it_interval = timespec64_to_ktime(new_setting->it_interval); /* * Rate limit to the tick as a hot fix to prevent DOS. Will be * mopped up later. */ - if (timr->it.alarm.interval < TICK_NSEC) - timr->it.alarm.interval = TICK_NSEC; + if (timr->it_interval < TICK_NSEC) + timr->it_interval = TICK_NSEC; exp = timespec64_to_ktime(new_setting->it_value); /* Convert (if necessary) to absolute time */ diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c index 79a00e0f1ef9..7dd992cc7105 100644 --- a/kernel/time/posix-timers.c +++ b/kernel/time/posix-timers.c @@ -285,12 +285,12 @@ static void schedule_next_timer(struct k_itimer *timr) { struct hrtimer *timer = &timr->it.real.timer; - if (timr->it.real.interval == 0) + if (!timr->it_interval) return; timr->it_overrun += (unsigned int) hrtimer_forward(timer, timer->base->get_time(), - timr->it.real.interval); + timr->it_interval); hrtimer_restart(timer); } @@ -375,7 +375,7 @@ static enum hrtimer_restart posix_timer_fn(struct hrtimer *timer) timr = container_of(timer, struct k_itimer, it.real.timer); spin_lock_irqsave(&timr->it_lock, flags); - if (timr->it.real.interval != 0) + if (timr->it_interval != 0) si_private = ++timr->it_requeue_pending; if (posix_timer_event(timr, si_private)) { @@ -384,7 +384,7 @@ static enum hrtimer_restart posix_timer_fn(struct hrtimer *timer) * we will not get a call back to restart it AND * it should be restarted. */ - if (timr->it.real.interval != 0) { + if (timr->it_interval != 0) { ktime_t now = hrtimer_cb_get_time(timer); /* @@ -413,13 +413,13 @@ static enum hrtimer_restart posix_timer_fn(struct hrtimer *timer) { ktime_t kj = NSEC_PER_SEC / HZ; - if (timr->it.real.interval < kj) + if (timr->it_interval < kj) now = ktime_add(now, kj); } #endif timr->it_overrun += (unsigned int) hrtimer_forward(timer, now, - timr->it.real.interval); + timr->it_interval); ret = HRTIMER_RESTART; ++timr->it_requeue_pending; } @@ -631,7 +631,7 @@ common_timer_get(struct k_itimer *timr, struct itimerspec64 *cur_setting) memset(cur_setting, 0, sizeof(*cur_setting)); - iv = timr->it.real.interval; + iv = timr->it_interval; /* interval timer ? */ if (iv) @@ -732,7 +732,7 @@ common_timer_set(struct k_itimer *timr, int flags, common_timer_get(timr, old_setting); /* disable the timer */ - timr->it.real.interval = 0; + timr->it_interval = 0; /* * careful here. If smp we could be in the "fire" routine which will * be spinning as we hold the lock. But this is ONLY an SMP issue. @@ -755,7 +755,7 @@ common_timer_set(struct k_itimer *timr, int flags, hrtimer_set_expires(timer, timespec64_to_ktime(new_setting->it_value)); /* Convert interval */ - timr->it.real.interval = timespec64_to_ktime(new_setting->it_interval); + timr->it_interval = timespec64_to_ktime(new_setting->it_interval); /* SIGEV_NONE timers are not queued ! See common_timer_get */ if (((timr->it_sigev_notify & ~SIGEV_THREAD_ID) == SIGEV_NONE)) { @@ -820,7 +820,7 @@ retry: static int common_timer_del(struct k_itimer *timer) { - timer->it.real.interval = 0; + timer->it_interval = 0; if (hrtimer_try_to_cancel(&timer->it.real.timer) < 0) return TIMER_RETRY; -- cgit v1.2.3 From d97bb75ddd2f38068df01da8abf26df78756253c Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 30 May 2017 23:15:44 +0200 Subject: posix-timers: Store k_clock pointer in k_itimer Having the k_clock pointer in the k_itimer struct avoids the lookup in several code pathes and makes the next steps of unification of the hrtimer and alarmtimer based posix timers simpler. Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: John Stultz Link: http://lkml.kernel.org/r/20170530211656.641222072@linutronix.de --- kernel/time/posix-cpu-timers.c | 2 ++ kernel/time/posix-timers.c | 7 ++++--- 2 files changed, 6 insertions(+), 3 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/posix-cpu-timers.c b/kernel/time/posix-cpu-timers.c index 1683e503179e..0123ece6851b 100644 --- a/kernel/time/posix-cpu-timers.c +++ b/kernel/time/posix-cpu-timers.c @@ -324,6 +324,8 @@ static int posix_cpu_timer_create(struct k_itimer *new_timer) if (CPUCLOCK_WHICH(new_timer->it_clock) >= CPUCLOCK_MAX) return -EINVAL; + new_timer->kclock = &clock_posix_cpu; + INIT_LIST_HEAD(&new_timer->it.cpu.entry); rcu_read_lock(); diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c index 7dd992cc7105..eb007e19811d 100644 --- a/kernel/time/posix-timers.c +++ b/kernel/time/posix-timers.c @@ -519,6 +519,7 @@ SYSCALL_DEFINE3(timer_create, const clockid_t, which_clock, it_id_set = IT_ID_SET; new_timer->it_id = (timer_t) new_timer_id; new_timer->it_clock = which_clock; + new_timer->kclock = kc; new_timer->it_overrun = -1; if (timer_event_spec) { @@ -679,7 +680,7 @@ SYSCALL_DEFINE2(timer_gettime, timer_t, timer_id, if (!timr) return -EINVAL; - kc = clockid_to_kclock(timr->it_clock); + kc = timr->kclock; if (WARN_ON_ONCE(!kc || !kc->timer_get)) ret = -EINVAL; else @@ -798,7 +799,7 @@ retry: if (!timr) return -EINVAL; - kc = clockid_to_kclock(timr->it_clock); + kc = timr->kclock; if (WARN_ON_ONCE(!kc || !kc->timer_set)) error = -EINVAL; else @@ -829,7 +830,7 @@ static int common_timer_del(struct k_itimer *timer) static inline int timer_delete_hook(struct k_itimer *timer) { - const struct k_clock *kc = clockid_to_kclock(timer->it_clock); + const struct k_clock *kc = timer->kclock; if (WARN_ON_ONCE(!kc || !kc->timer_del)) return -EINVAL; -- cgit v1.2.3 From 30802945893bc944b5971b408b37511a03b54e5c Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 30 May 2017 23:15:45 +0200 Subject: posix-timers: Add timer_rearm() callback Add a timer_rearm() callback which is used to make the rescheduling of posix interval timers independent of the underlying clock implementation. Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: John Stultz Link: http://lkml.kernel.org/r/20170530211656.732632167@linutronix.de --- kernel/time/posix-timers.h | 33 ++++++++++++++++++--------------- 1 file changed, 18 insertions(+), 15 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/posix-timers.h b/kernel/time/posix-timers.h index ad2dbd29b389..02ffd1b9d230 100644 --- a/kernel/time/posix-timers.h +++ b/kernel/time/posix-timers.h @@ -1,21 +1,24 @@ #define TIMER_RETRY 1 struct k_clock { - int (*clock_getres) (const clockid_t which_clock, struct timespec64 *tp); - int (*clock_set) (const clockid_t which_clock, - const struct timespec64 *tp); - int (*clock_get) (const clockid_t which_clock, struct timespec64 *tp); - int (*clock_adj) (const clockid_t which_clock, struct timex *tx); - int (*timer_create) (struct k_itimer *timer); - int (*nsleep) (const clockid_t which_clock, int flags, - struct timespec64 *, struct timespec __user *); - long (*nsleep_restart) (struct restart_block *restart_block); - int (*timer_set) (struct k_itimer *timr, int flags, - struct itimerspec64 *new_setting, - struct itimerspec64 *old_setting); - int (*timer_del) (struct k_itimer *timr); - void (*timer_get) (struct k_itimer *timr, - struct itimerspec64 *cur_setting); + int (*clock_getres)(const clockid_t which_clock, + struct timespec64 *tp); + int (*clock_set)(const clockid_t which_clock, + const struct timespec64 *tp); + int (*clock_get)(const clockid_t which_clock, + struct timespec64 *tp); + int (*clock_adj)(const clockid_t which_clock, struct timex *tx); + int (*timer_create)(struct k_itimer *timer); + int (*nsleep)(const clockid_t which_clock, int flags, + struct timespec64 *, struct timespec __user *); + long (*nsleep_restart)(struct restart_block *restart_block); + int (*timer_set)(struct k_itimer *timr, int flags, + struct itimerspec64 *new_setting, + struct itimerspec64 *old_setting); + int (*timer_del)(struct k_itimer *timr); + void (*timer_get)(struct k_itimer *timr, + struct itimerspec64 *cur_setting); + void (*timer_rearm)(struct k_itimer *timr); }; extern const struct k_clock clock_posix_cpu; -- cgit v1.2.3 From 96fe3b072f134e4993f829d599eaa1e0eb5a10e5 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 30 May 2017 23:15:46 +0200 Subject: posix-timers: Rename do_schedule_next_timer That function is a misnomer. Rename it with a proper prefix to posixtimer_rearm(). Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: John Stultz Link: http://lkml.kernel.org/r/20170530211656.811362578@linutronix.de --- kernel/time/posix-cpu-timers.c | 2 +- kernel/time/posix-timers.c | 10 +++++----- 2 files changed, 6 insertions(+), 6 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/posix-cpu-timers.c b/kernel/time/posix-cpu-timers.c index 0123ece6851b..1ba576d3151a 100644 --- a/kernel/time/posix-cpu-timers.c +++ b/kernel/time/posix-cpu-timers.c @@ -981,7 +981,7 @@ static void check_process_timers(struct task_struct *tsk, } /* - * This is called from the signal code (via do_schedule_next_timer) + * This is called from the signal code (via posixtimer_rearm) * when the last timer signal was delivered and we have to reload the timer. */ void posix_cpu_timer_schedule(struct k_itimer *timer) diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c index eb007e19811d..036b7e70c65c 100644 --- a/kernel/time/posix-timers.c +++ b/kernel/time/posix-timers.c @@ -305,7 +305,7 @@ static void schedule_next_timer(struct k_itimer *timr) * To protect against the timer going away while the interrupt is queued, * we require that the it_requeue_pending flag be set. */ -void do_schedule_next_timer(struct siginfo *info) +void posixtimer_rearm(struct siginfo *info) { struct k_itimer *timr; unsigned long flags; @@ -336,12 +336,12 @@ int posix_timer_event(struct k_itimer *timr, int si_private) int shared, ret = -1; /* * FIXME: if ->sigq is queued we can race with - * dequeue_signal()->do_schedule_next_timer(). + * dequeue_signal()->posixtimer_rearm(). * * If dequeue_signal() sees the "right" value of - * si_sys_private it calls do_schedule_next_timer(). + * si_sys_private it calls posixtimer_rearm(). * We re-queue ->sigq and drop ->it_lock(). - * do_schedule_next_timer() locks the timer + * posixtimer_rearm() locks the timer * and re-schedules it while ->sigq is pending. * Not really bad, but not that we want. */ @@ -701,7 +701,7 @@ SYSCALL_DEFINE2(timer_gettime, timer_t, timer_id, * accumulating overruns on the next timer. The overrun is frozen when * the signal is delivered, either at the notify time (if the info block * is not queued) or at the actual delivery time (as we are informed by - * the call back to do_schedule_next_timer(). So all we need to do is + * the call back to posixtimer_rearm(). So all we need to do is * to pick up the frozen overrun. */ SYSCALL_DEFINE1(timer_getoverrun, timer_t, timer_id) -- cgit v1.2.3 From f37fb0aa4f453c7c785bbcecc4991ac48c5c0e51 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 30 May 2017 23:15:47 +0200 Subject: posix-timers: Use timer_rearm() callback in posixtimer_rearm() Use the new timer_rearm() callback to replace the conditional hardcoded calls into the hrtimer and cpu timer code. This allows later to bring the same logic to alarmtimers. Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: John Stultz Link: http://lkml.kernel.org/r/20170530211656.889661919@linutronix.de --- kernel/time/posix-cpu-timers.c | 7 +++++-- kernel/time/posix-timers.c | 12 ++++++------ kernel/time/posix-timers.h | 2 -- 3 files changed, 11 insertions(+), 10 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/posix-cpu-timers.c b/kernel/time/posix-cpu-timers.c index 1ba576d3151a..96c833a61ade 100644 --- a/kernel/time/posix-cpu-timers.c +++ b/kernel/time/posix-cpu-timers.c @@ -15,6 +15,8 @@ #include "posix-timers.h" +static void posix_cpu_timer_rearm(struct k_itimer *timer); + /* * Called after updating RLIMIT_CPU to run cpu timer and update * tsk->signal->cputime_expires expiration cache if necessary. Needs @@ -528,7 +530,7 @@ static void cpu_timer_fire(struct k_itimer *timer) * reload the timer. But we need to keep it * ticking in case the signal is deliverable next time. */ - posix_cpu_timer_schedule(timer); + posix_cpu_timer_rearm(timer); ++timer->it_requeue_pending; } } @@ -984,7 +986,7 @@ static void check_process_timers(struct task_struct *tsk, * This is called from the signal code (via posixtimer_rearm) * when the last timer signal was delivered and we have to reload the timer. */ -void posix_cpu_timer_schedule(struct k_itimer *timer) +static void posix_cpu_timer_rearm(struct k_itimer *timer) { struct sighand_struct *sighand; unsigned long flags; @@ -1431,6 +1433,7 @@ const struct k_clock clock_posix_cpu = { .timer_set = posix_cpu_timer_set, .timer_del = posix_cpu_timer_del, .timer_get = posix_cpu_timer_get, + .timer_rearm = posix_cpu_timer_rearm, }; const struct k_clock clock_process = { diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c index 036b7e70c65c..b12582a4b122 100644 --- a/kernel/time/posix-timers.c +++ b/kernel/time/posix-timers.c @@ -278,10 +278,9 @@ static __init int init_posix_timers(void) NULL); return 0; } - __initcall(init_posix_timers); -static void schedule_next_timer(struct k_itimer *timr) +static void common_hrtimer_rearm(struct k_itimer *timr) { struct hrtimer *timer = &timr->it.real.timer; @@ -315,10 +314,7 @@ void posixtimer_rearm(struct siginfo *info) return; if (timr->it_requeue_pending == info->si_sys_private) { - if (timr->it_clock < 0) - posix_cpu_timer_schedule(timr); - else - schedule_next_timer(timr); + timr->kclock->timer_rearm(timr); timr->it_overrun_last = timr->it_overrun; timr->it_overrun = -1; @@ -1046,6 +1042,7 @@ static const struct k_clock clock_realtime = { .timer_set = common_timer_set, .timer_get = common_timer_get, .timer_del = common_timer_del, + .timer_rearm = common_hrtimer_rearm, }; static const struct k_clock clock_monotonic = { @@ -1057,6 +1054,7 @@ static const struct k_clock clock_monotonic = { .timer_set = common_timer_set, .timer_get = common_timer_get, .timer_del = common_timer_del, + .timer_rearm = common_hrtimer_rearm, }; static const struct k_clock clock_monotonic_raw = { @@ -1083,6 +1081,7 @@ static const struct k_clock clock_tai = { .timer_set = common_timer_set, .timer_get = common_timer_get, .timer_del = common_timer_del, + .timer_rearm = common_hrtimer_rearm, }; static const struct k_clock clock_boottime = { @@ -1094,6 +1093,7 @@ static const struct k_clock clock_boottime = { .timer_set = common_timer_set, .timer_get = common_timer_get, .timer_del = common_timer_del, + .timer_rearm = common_hrtimer_rearm, }; static const struct k_clock * const posix_clocks[] = { diff --git a/kernel/time/posix-timers.h b/kernel/time/posix-timers.h index 02ffd1b9d230..1f6f6f9a6a37 100644 --- a/kernel/time/posix-timers.h +++ b/kernel/time/posix-timers.h @@ -28,5 +28,3 @@ extern const struct k_clock clock_thread; extern const struct k_clock alarm_clock; int posix_timer_event(struct k_itimer *timr, int si_private); - -void posix_cpu_timer_schedule(struct k_itimer *timer); -- cgit v1.2.3 From 21e55c1f83880a56360287c00f2b5cd5e5a4a912 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 30 May 2017 23:15:48 +0200 Subject: posix-timers: Add active flag to k_itimer Keep track of the activation state of posix timers. This is a preparatory change for making common_timer_get() usable by both hrtimer and alarm timer implementations. Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: John Stultz Link: http://lkml.kernel.org/r/20170530211656.967783982@linutronix.de --- kernel/time/posix-timers.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) (limited to 'kernel/time') diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c index b12582a4b122..795215bba73d 100644 --- a/kernel/time/posix-timers.c +++ b/kernel/time/posix-timers.c @@ -316,6 +316,7 @@ void posixtimer_rearm(struct siginfo *info) if (timr->it_requeue_pending == info->si_sys_private) { timr->kclock->timer_rearm(timr); + timr->it_active = 1; timr->it_overrun_last = timr->it_overrun; timr->it_overrun = -1; ++timr->it_requeue_pending; @@ -371,6 +372,7 @@ static enum hrtimer_restart posix_timer_fn(struct hrtimer *timer) timr = container_of(timer, struct k_itimer, it.real.timer); spin_lock_irqsave(&timr->it_lock, flags); + timr->it_active = 0; if (timr->it_interval != 0) si_private = ++timr->it_requeue_pending; @@ -418,6 +420,7 @@ static enum hrtimer_restart posix_timer_fn(struct hrtimer *timer) timr->it_interval); ret = HRTIMER_RESTART; ++timr->it_requeue_pending; + timr->it_active = 1; } } @@ -737,7 +740,8 @@ common_timer_set(struct k_itimer *timr, int flags, if (hrtimer_try_to_cancel(timer) < 0) return TIMER_RETRY; - timr->it_requeue_pending = (timr->it_requeue_pending + 2) & + timr->it_active = 0; + timr->it_requeue_pending = (timr->it_requeue_pending + 2) & ~REQUEUE_PENDING; timr->it_overrun_last = 0; @@ -763,6 +767,7 @@ common_timer_set(struct k_itimer *timr, int flags, return 0; } + timr->it_active = 1; hrtimer_start_expires(timer, mode); return 0; } @@ -821,6 +826,7 @@ static int common_timer_del(struct k_itimer *timer) if (hrtimer_try_to_cancel(&timer->it.real.timer) < 0) return TIMER_RETRY; + timer->it_active = 0; return 0; } -- cgit v1.2.3 From 63841b2a6969501de183efafc14d20175e402804 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 30 May 2017 23:15:49 +0200 Subject: posix-timers: Add forward/remaining callbacks Add two callbacks to kclock which allow using common_)timer_get() for both hrtimer and alarm timer based clocks. Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: John Stultz Link: http://lkml.kernel.org/r/20170530211657.044915536@linutronix.de --- kernel/time/posix-timers.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'kernel/time') diff --git a/kernel/time/posix-timers.h b/kernel/time/posix-timers.h index 1f6f6f9a6a37..3bc5b74c342f 100644 --- a/kernel/time/posix-timers.h +++ b/kernel/time/posix-timers.h @@ -19,6 +19,8 @@ struct k_clock { void (*timer_get)(struct k_itimer *timr, struct itimerspec64 *cur_setting); void (*timer_rearm)(struct k_itimer *timr); + int (*timer_forward)(struct k_itimer *timr, ktime_t now); + ktime_t (*timer_remaining)(struct k_itimer *timr, ktime_t now); }; extern const struct k_clock clock_posix_cpu; -- cgit v1.2.3 From 91d57bae08689199c8acc77a8b3b41150cafab1c Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 30 May 2017 23:15:50 +0200 Subject: posix-timers: Make use of forward/remaining callbacks Replace the hrtimer calls by calls to the new forward/remaining kclock callbacks and move the hrtimer specific implementation into the corresponding callback functions. Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: John Stultz Link: http://lkml.kernel.org/r/20170530211657.121437232@linutronix.de --- kernel/time/posix-timers.c | 64 +++++++++++++++++++++++++++++++++++----------- 1 file changed, 49 insertions(+), 15 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c index 795215bba73d..48f6c37ae5df 100644 --- a/kernel/time/posix-timers.c +++ b/kernel/time/posix-timers.c @@ -607,6 +607,20 @@ static struct k_itimer *__lock_timer(timer_t timer_id, unsigned long *flags) return NULL; } +static ktime_t common_hrtimer_remaining(struct k_itimer *timr, ktime_t now) +{ + struct hrtimer *timer = &timr->it.real.timer; + + return __hrtimer_expires_remaining_adjusted(timer, now); +} + +static int common_hrtimer_forward(struct k_itimer *timr, ktime_t now) +{ + struct hrtimer *timer = &timr->it.real.timer; + + return (int)hrtimer_forward(timer, now, timr->it_interval); +} + /* * Get the time remaining on a POSIX.1b interval timer. This function * is ALWAYS called with spin_lock_irq on the timer, thus it must not @@ -626,42 +640,54 @@ static struct k_itimer *__lock_timer(timer_t timer_id, unsigned long *flags) static void common_timer_get(struct k_itimer *timr, struct itimerspec64 *cur_setting) { + const struct k_clock *kc = timr->kclock; ktime_t now, remaining, iv; - struct hrtimer *timer = &timr->it.real.timer; + struct timespec64 ts64; + bool sig_none; memset(cur_setting, 0, sizeof(*cur_setting)); + sig_none = (timr->it_sigev_notify & ~SIGEV_THREAD_ID) != SIGEV_NONE; iv = timr->it_interval; /* interval timer ? */ - if (iv) + if (iv) { cur_setting->it_interval = ktime_to_timespec64(iv); - else if (!hrtimer_active(timer) && - (timr->it_sigev_notify & ~SIGEV_THREAD_ID) != SIGEV_NONE) - return; + } else if (!timr->it_active) { + /* + * SIGEV_NONE oneshot timers are never queued. Check them + * below. + */ + if (!sig_none) + return; + } - now = timer->base->get_time(); + /* + * The timespec64 based conversion is suboptimal, but it's not + * worth to implement yet another callback. + */ + kc->clock_get(timr->it_clock, &ts64); + now = timespec64_to_ktime(ts64); /* - * When a requeue is pending or this is a SIGEV_NONE - * timer move the expiry time forward by intervals, so - * expiry is > now. + * When a requeue is pending or this is a SIGEV_NONE timer move the + * expiry time forward by intervals, so expiry is > now. */ - if (iv && (timr->it_requeue_pending & REQUEUE_PENDING || - (timr->it_sigev_notify & ~SIGEV_THREAD_ID) == SIGEV_NONE)) - timr->it_overrun += (unsigned int) hrtimer_forward(timer, now, iv); + if (iv && (timr->it_requeue_pending & REQUEUE_PENDING || sig_none)) + timr->it_overrun += kc->timer_forward(timr, now); - remaining = __hrtimer_expires_remaining_adjusted(timer, now); + remaining = kc->timer_remaining(timr, now); /* Return 0 only, when the timer is expired and not pending */ if (remaining <= 0) { /* * A single shot SIGEV_NONE timer must return 0, when * it is expired ! */ - if ((timr->it_sigev_notify & ~SIGEV_THREAD_ID) != SIGEV_NONE) + if (!sig_none) cur_setting->it_value.tv_nsec = 1; - } else + } else { cur_setting->it_value = ktime_to_timespec64(remaining); + } } /* Get the time remaining on a POSIX.1b interval timer. */ @@ -1049,6 +1075,8 @@ static const struct k_clock clock_realtime = { .timer_get = common_timer_get, .timer_del = common_timer_del, .timer_rearm = common_hrtimer_rearm, + .timer_forward = common_hrtimer_forward, + .timer_remaining= common_hrtimer_remaining, }; static const struct k_clock clock_monotonic = { @@ -1061,6 +1089,8 @@ static const struct k_clock clock_monotonic = { .timer_get = common_timer_get, .timer_del = common_timer_del, .timer_rearm = common_hrtimer_rearm, + .timer_forward = common_hrtimer_forward, + .timer_remaining= common_hrtimer_remaining, }; static const struct k_clock clock_monotonic_raw = { @@ -1088,6 +1118,8 @@ static const struct k_clock clock_tai = { .timer_get = common_timer_get, .timer_del = common_timer_del, .timer_rearm = common_hrtimer_rearm, + .timer_forward = common_hrtimer_forward, + .timer_remaining= common_hrtimer_remaining, }; static const struct k_clock clock_boottime = { @@ -1100,6 +1132,8 @@ static const struct k_clock clock_boottime = { .timer_get = common_timer_get, .timer_del = common_timer_del, .timer_rearm = common_hrtimer_rearm, + .timer_forward = common_hrtimer_forward, + .timer_remaining= common_hrtimer_remaining, }; static const struct k_clock * const posix_clocks[] = { -- cgit v1.2.3 From eabdec04385376d560078992710104cc7be2ce1b Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 30 May 2017 23:15:51 +0200 Subject: posix-timers: Zero settings value in common code Zero out the settings struct in the common code so the callbacks do not have to do it themself. Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: John Stultz Link: http://lkml.kernel.org/r/20170530211657.200870713@linutronix.de --- kernel/time/posix-cpu-timers.c | 5 +---- kernel/time/posix-timers.c | 3 +-- 2 files changed, 2 insertions(+), 6 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/posix-cpu-timers.c b/kernel/time/posix-cpu-timers.c index 96c833a61ade..cb4a4eb44279 100644 --- a/kernel/time/posix-cpu-timers.c +++ b/kernel/time/posix-cpu-timers.c @@ -719,10 +719,8 @@ static void posix_cpu_timer_get(struct k_itimer *timer, struct itimerspec64 *itp */ itp->it_interval = ns_to_timespec64(timer->it.cpu.incr); - if (timer->it.cpu.expires == 0) { /* Timer not armed at all. */ - itp->it_value.tv_sec = itp->it_value.tv_nsec = 0; + if (!timer->it.cpu.expires) return; - } /* * Sample the clock to take the difference with the expiry time. @@ -746,7 +744,6 @@ static void posix_cpu_timer_get(struct k_itimer *timer, struct itimerspec64 *itp * Call the timer disarmed, nothing else to do. */ timer->it.cpu.expires = 0; - itp->it_value = ns_to_timespec64(timer->it.cpu.expires); return; } else { cpu_timer_sample_group(timer->it_clock, p, &now); diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c index 48f6c37ae5df..0332f7a60fd6 100644 --- a/kernel/time/posix-timers.c +++ b/kernel/time/posix-timers.c @@ -645,8 +645,6 @@ common_timer_get(struct k_itimer *timr, struct itimerspec64 *cur_setting) struct timespec64 ts64; bool sig_none; - memset(cur_setting, 0, sizeof(*cur_setting)); - sig_none = (timr->it_sigev_notify & ~SIGEV_THREAD_ID) != SIGEV_NONE; iv = timr->it_interval; @@ -705,6 +703,7 @@ SYSCALL_DEFINE2(timer_gettime, timer_t, timer_id, if (!timr) return -EINVAL; + memset(&cur_setting64, 0, sizeof(cur_setting64)); kc = timr->kclock; if (WARN_ON_ONCE(!kc || !kc->timer_get)) ret = -EINVAL; -- cgit v1.2.3 From 525b8ed91671e29e187dfe02d408b11190ccf494 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 30 May 2017 23:15:52 +0200 Subject: posix-timers: Add cancel/arm callbacks Add timer_try_to_cancel() and timer_arm() callbacks to kclock which allow to make common_timer_set() usable by both hrtimer and alarmtimer based clocks. Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: John Stultz Link: http://lkml.kernel.org/r/20170530211657.278022962@linutronix.de --- kernel/time/posix-timers.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'kernel/time') diff --git a/kernel/time/posix-timers.h b/kernel/time/posix-timers.h index 3bc5b74c342f..b0ad77e18886 100644 --- a/kernel/time/posix-timers.h +++ b/kernel/time/posix-timers.h @@ -21,6 +21,9 @@ struct k_clock { void (*timer_rearm)(struct k_itimer *timr); int (*timer_forward)(struct k_itimer *timr, ktime_t now); ktime_t (*timer_remaining)(struct k_itimer *timr, ktime_t now); + int (*timer_try_to_cancel)(struct k_itimer *timr); + void (*timer_arm)(struct k_itimer *timr, ktime_t expires, + bool absolute, bool sigev_none); }; extern const struct k_clock clock_posix_cpu; -- cgit v1.2.3 From eae1c4ae275fe3e024454c012a548ee0d700f54c Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 30 May 2017 23:15:53 +0200 Subject: posix-timers: Make use of cancel/arm callbacks Replace the hrtimer calls by calls to the new try_to_cancel()/arm() kclock callbacks and move the hrtimer specific implementation into the corresponding callback functions. Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: John Stultz Link: http://lkml.kernel.org/r/20170530211657.355396667@linutronix.de --- kernel/time/posix-timers.c | 181 +++++++++++++++++++++++++-------------------- 1 file changed, 100 insertions(+), 81 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c index 0332f7a60fd6..8acc9ee2c2d6 100644 --- a/kernel/time/posix-timers.c +++ b/kernel/time/posix-timers.c @@ -744,25 +744,49 @@ SYSCALL_DEFINE1(timer_getoverrun, timer_t, timer_id) return overrun; } +static void common_hrtimer_arm(struct k_itimer *timr, ktime_t expires, + bool absolute, bool sigev_none) +{ + struct hrtimer *timer = &timr->it.real.timer; + enum hrtimer_mode mode; + + mode = absolute ? HRTIMER_MODE_ABS : HRTIMER_MODE_REL; + hrtimer_init(&timr->it.real.timer, timr->it_clock, mode); + timr->it.real.timer.function = posix_timer_fn; + + if (!absolute) + expires = ktime_add_safe(expires, timer->base->get_time()); + hrtimer_set_expires(timer, expires); + + if (!sigev_none) + hrtimer_start_expires(timer, HRTIMER_MODE_ABS); +} + +static int common_hrtimer_try_to_cancel(struct k_itimer *timr) +{ + return hrtimer_try_to_cancel(&timr->it.real.timer); +} + /* Set a POSIX.1b interval timer. */ -/* timr->it_lock is taken. */ static int common_timer_set(struct k_itimer *timr, int flags, - struct itimerspec64 *new_setting, struct itimerspec64 *old_setting) + struct itimerspec64 *new_setting, + struct itimerspec64 *old_setting) { - struct hrtimer *timer = &timr->it.real.timer; - enum hrtimer_mode mode; + const struct k_clock *kc = timr->kclock; + bool sigev_none; + ktime_t expires; if (old_setting) common_timer_get(timr, old_setting); - /* disable the timer */ + /* Prevent rearming by clearing the interval */ timr->it_interval = 0; /* - * careful here. If smp we could be in the "fire" routine which will - * be spinning as we hold the lock. But this is ONLY an SMP issue. + * Careful here. On SMP systems the timer expiry function could be + * active and spinning on timr->it_lock. */ - if (hrtimer_try_to_cancel(timer) < 0) + if (kc->timer_try_to_cancel(timr) < 0) return TIMER_RETRY; timr->it_active = 0; @@ -770,30 +794,16 @@ common_timer_set(struct k_itimer *timr, int flags, ~REQUEUE_PENDING; timr->it_overrun_last = 0; - /* switch off the timer when it_value is zero */ + /* Switch off the timer when it_value is zero */ if (!new_setting->it_value.tv_sec && !new_setting->it_value.tv_nsec) return 0; - mode = flags & TIMER_ABSTIME ? HRTIMER_MODE_ABS : HRTIMER_MODE_REL; - hrtimer_init(&timr->it.real.timer, timr->it_clock, mode); - timr->it.real.timer.function = posix_timer_fn; - - hrtimer_set_expires(timer, timespec64_to_ktime(new_setting->it_value)); - - /* Convert interval */ timr->it_interval = timespec64_to_ktime(new_setting->it_interval); + expires = timespec64_to_ktime(new_setting->it_value); + sigev_none = (timr->it_sigev_notify & ~SIGEV_THREAD_ID) == SIGEV_NONE; - /* SIGEV_NONE timers are not queued ! See common_timer_get */ - if (((timr->it_sigev_notify & ~SIGEV_THREAD_ID) == SIGEV_NONE)) { - /* Setup correct expiry time for relative timers */ - if (mode == HRTIMER_MODE_REL) { - hrtimer_add_expires(timer, timer->base->get_time()); - } - return 0; - } - - timr->it_active = 1; - hrtimer_start_expires(timer, mode); + kc->timer_arm(timr, expires, flags & TIMER_ABSTIME, sigev_none); + timr->it_active = !sigev_none; return 0; } @@ -847,9 +857,10 @@ retry: static int common_timer_del(struct k_itimer *timer) { - timer->it_interval = 0; + const struct k_clock *kc = timer->kclock; - if (hrtimer_try_to_cancel(&timer->it.real.timer) < 0) + timer->it_interval = 0; + if (kc->timer_try_to_cancel(timer) < 0) return TIMER_RETRY; timer->it_active = 0; return 0; @@ -1063,76 +1074,84 @@ long clock_nanosleep_restart(struct restart_block *restart_block) } static const struct k_clock clock_realtime = { - .clock_getres = posix_get_hrtimer_res, - .clock_get = posix_clock_realtime_get, - .clock_set = posix_clock_realtime_set, - .clock_adj = posix_clock_realtime_adj, - .nsleep = common_nsleep, - .nsleep_restart = hrtimer_nanosleep_restart, - .timer_create = common_timer_create, - .timer_set = common_timer_set, - .timer_get = common_timer_get, - .timer_del = common_timer_del, - .timer_rearm = common_hrtimer_rearm, - .timer_forward = common_hrtimer_forward, - .timer_remaining= common_hrtimer_remaining, + .clock_getres = posix_get_hrtimer_res, + .clock_get = posix_clock_realtime_get, + .clock_set = posix_clock_realtime_set, + .clock_adj = posix_clock_realtime_adj, + .nsleep = common_nsleep, + .nsleep_restart = hrtimer_nanosleep_restart, + .timer_create = common_timer_create, + .timer_set = common_timer_set, + .timer_get = common_timer_get, + .timer_del = common_timer_del, + .timer_rearm = common_hrtimer_rearm, + .timer_forward = common_hrtimer_forward, + .timer_remaining = common_hrtimer_remaining, + .timer_try_to_cancel = common_hrtimer_try_to_cancel, + .timer_arm = common_hrtimer_arm, }; static const struct k_clock clock_monotonic = { - .clock_getres = posix_get_hrtimer_res, - .clock_get = posix_ktime_get_ts, - .nsleep = common_nsleep, - .nsleep_restart = hrtimer_nanosleep_restart, - .timer_create = common_timer_create, - .timer_set = common_timer_set, - .timer_get = common_timer_get, - .timer_del = common_timer_del, - .timer_rearm = common_hrtimer_rearm, - .timer_forward = common_hrtimer_forward, - .timer_remaining= common_hrtimer_remaining, + .clock_getres = posix_get_hrtimer_res, + .clock_get = posix_ktime_get_ts, + .nsleep = common_nsleep, + .nsleep_restart = hrtimer_nanosleep_restart, + .timer_create = common_timer_create, + .timer_set = common_timer_set, + .timer_get = common_timer_get, + .timer_del = common_timer_del, + .timer_rearm = common_hrtimer_rearm, + .timer_forward = common_hrtimer_forward, + .timer_remaining = common_hrtimer_remaining, + .timer_try_to_cancel = common_hrtimer_try_to_cancel, + .timer_arm = common_hrtimer_arm, }; static const struct k_clock clock_monotonic_raw = { - .clock_getres = posix_get_hrtimer_res, - .clock_get = posix_get_monotonic_raw, + .clock_getres = posix_get_hrtimer_res, + .clock_get = posix_get_monotonic_raw, }; static const struct k_clock clock_realtime_coarse = { - .clock_getres = posix_get_coarse_res, - .clock_get = posix_get_realtime_coarse, + .clock_getres = posix_get_coarse_res, + .clock_get = posix_get_realtime_coarse, }; static const struct k_clock clock_monotonic_coarse = { - .clock_getres = posix_get_coarse_res, - .clock_get = posix_get_monotonic_coarse, + .clock_getres = posix_get_coarse_res, + .clock_get = posix_get_monotonic_coarse, }; static const struct k_clock clock_tai = { - .clock_getres = posix_get_hrtimer_res, - .clock_get = posix_get_tai, - .nsleep = common_nsleep, - .nsleep_restart = hrtimer_nanosleep_restart, - .timer_create = common_timer_create, - .timer_set = common_timer_set, - .timer_get = common_timer_get, - .timer_del = common_timer_del, - .timer_rearm = common_hrtimer_rearm, - .timer_forward = common_hrtimer_forward, - .timer_remaining= common_hrtimer_remaining, + .clock_getres = posix_get_hrtimer_res, + .clock_get = posix_get_tai, + .nsleep = common_nsleep, + .nsleep_restart = hrtimer_nanosleep_restart, + .timer_create = common_timer_create, + .timer_set = common_timer_set, + .timer_get = common_timer_get, + .timer_del = common_timer_del, + .timer_rearm = common_hrtimer_rearm, + .timer_forward = common_hrtimer_forward, + .timer_remaining = common_hrtimer_remaining, + .timer_try_to_cancel = common_hrtimer_try_to_cancel, + .timer_arm = common_hrtimer_arm, }; static const struct k_clock clock_boottime = { - .clock_getres = posix_get_hrtimer_res, - .clock_get = posix_get_boottime, - .nsleep = common_nsleep, - .nsleep_restart = hrtimer_nanosleep_restart, - .timer_create = common_timer_create, - .timer_set = common_timer_set, - .timer_get = common_timer_get, - .timer_del = common_timer_del, - .timer_rearm = common_hrtimer_rearm, - .timer_forward = common_hrtimer_forward, - .timer_remaining= common_hrtimer_remaining, + .clock_getres = posix_get_hrtimer_res, + .clock_get = posix_get_boottime, + .nsleep = common_nsleep, + .nsleep_restart = hrtimer_nanosleep_restart, + .timer_create = common_timer_create, + .timer_set = common_timer_set, + .timer_get = common_timer_get, + .timer_del = common_timer_del, + .timer_rearm = common_hrtimer_rearm, + .timer_forward = common_hrtimer_forward, + .timer_remaining = common_hrtimer_remaining, + .timer_try_to_cancel = common_hrtimer_try_to_cancel, + .timer_arm = common_hrtimer_arm, }; static const struct k_clock * const posix_clocks[] = { -- cgit v1.2.3 From b3db80f77a95a45dbb2136f7b2a364dc797ea914 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 30 May 2017 23:15:54 +0200 Subject: alarmtimer: Implement timer_rearm() callback Preparatory change to utilize the common posix timer mechanisms. Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: John Stultz Link: http://lkml.kernel.org/r/20170530211657.434598989@linutronix.de --- kernel/time/alarmtimer.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) (limited to 'kernel/time') diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index 5b8cf4b61854..be85e3cbfe1b 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -536,6 +536,18 @@ static enum alarmtimer_restart alarm_handle_timer(struct alarm *alarm, return result; } +/** + * alarm_timer_rearm - Posix timer callback for rearming timer + * @timr: Pointer to the posixtimer data struct + */ +static void alarm_timer_rearm(struct k_itimer *timr) +{ + struct alarm *alarm = &timr->it.alarm.alarmtimer; + + timr->it_overrun += alarm_forward_now(alarm, timr->it_interval); + alarm_start(alarm, alarm->node.expires); +} + /** * alarm_clock_getres - posix getres interface * @which_clock: clockid @@ -594,7 +606,7 @@ static int alarm_timer_create(struct k_itimer *new_timer) /** * alarm_timer_get - posix timer_get interface - * @new_timer: k_itimer pointer + * @timr: k_itimer pointer * @cur_setting: itimerspec data to fill * * Copies out the current itimerspec data @@ -863,6 +875,7 @@ const struct k_clock alarm_clock = { .timer_set = alarm_timer_set, .timer_del = alarm_timer_del, .timer_get = alarm_timer_get, + .timer_rearm = alarm_timer_rearm, .nsleep = alarm_timer_nsleep, }; #endif /* CONFIG_POSIX_TIMERS */ -- cgit v1.2.3 From e7561f1633ac735df48c55ad09a2530e9ab9fab1 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 30 May 2017 23:15:55 +0200 Subject: alarmtimer: Implement forward callback Preparatory change to utilize the common posix timer mechanisms. Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: John Stultz Link: http://lkml.kernel.org/r/20170530211657.513694229@linutronix.de --- kernel/time/alarmtimer.c | 13 +++++++++++++ 1 file changed, 13 insertions(+) (limited to 'kernel/time') diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index be85e3cbfe1b..6082cf1af876 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -548,6 +548,18 @@ static void alarm_timer_rearm(struct k_itimer *timr) alarm_start(alarm, alarm->node.expires); } +/** + * alarm_timer_forward - Posix timer callback for forwarding timer + * @timr: Pointer to the posixtimer data struct + * @now: Current time to forward the timer against + */ +static int alarm_timer_forward(struct k_itimer *timr, ktime_t now) +{ + struct alarm *alarm = &timr->it.alarm.alarmtimer; + + return (int) alarm_forward(alarm, timr->it_interval, now); +} + /** * alarm_clock_getres - posix getres interface * @which_clock: clockid @@ -876,6 +888,7 @@ const struct k_clock alarm_clock = { .timer_del = alarm_timer_del, .timer_get = alarm_timer_get, .timer_rearm = alarm_timer_rearm, + .timer_forward = alarm_timer_forward, .nsleep = alarm_timer_nsleep, }; #endif /* CONFIG_POSIX_TIMERS */ -- cgit v1.2.3 From d653d8457c76da11f047af1f66256ac9b8421b69 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 30 May 2017 23:15:56 +0200 Subject: alarmtimer: Implement remaining callback Preparatory change to utilize the common posix timer mechanisms. Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: John Stultz Link: http://lkml.kernel.org/r/20170530211657.592676753@linutronix.de --- kernel/time/alarmtimer.c | 31 ++++++++++++++++++++++--------- 1 file changed, 22 insertions(+), 9 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index 6082cf1af876..02ddc40f19fe 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -560,6 +560,18 @@ static int alarm_timer_forward(struct k_itimer *timr, ktime_t now) return (int) alarm_forward(alarm, timr->it_interval, now); } +/** + * alarm_timer_remaining - Posix timer callback to retrieve remaining time + * @timr: Pointer to the posixtimer data struct + * @now: Current time to calculate against + */ +static ktime_t alarm_timer_remaining(struct k_itimer *timr, ktime_t now) +{ + struct alarm *alarm = &timr->it.alarm.alarmtimer; + + return ktime_sub(now, alarm->node.expires); +} + /** * alarm_clock_getres - posix getres interface * @which_clock: clockid @@ -881,15 +893,16 @@ out: } const struct k_clock alarm_clock = { - .clock_getres = alarm_clock_getres, - .clock_get = alarm_clock_get, - .timer_create = alarm_timer_create, - .timer_set = alarm_timer_set, - .timer_del = alarm_timer_del, - .timer_get = alarm_timer_get, - .timer_rearm = alarm_timer_rearm, - .timer_forward = alarm_timer_forward, - .nsleep = alarm_timer_nsleep, + .clock_getres = alarm_clock_getres, + .clock_get = alarm_clock_get, + .timer_create = alarm_timer_create, + .timer_set = alarm_timer_set, + .timer_del = alarm_timer_del, + .timer_get = alarm_timer_get, + .timer_rearm = alarm_timer_rearm, + .timer_forward = alarm_timer_forward, + .timer_remaining = alarm_timer_remaining, + .nsleep = alarm_timer_nsleep, }; #endif /* CONFIG_POSIX_TIMERS */ -- cgit v1.2.3 From e344c9e76bc6af997926171bfd90d25bbae0a2c5 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 30 May 2017 23:15:57 +0200 Subject: alarmtimer: Implement try_to_cancel callback Preparatory change to utilize the common posix timer mechanisms. Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: John Stultz Link: http://lkml.kernel.org/r/20170530211657.670026824@linutronix.de --- kernel/time/alarmtimer.c | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'kernel/time') diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index 02ddc40f19fe..374bd855a488 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -572,6 +572,15 @@ static ktime_t alarm_timer_remaining(struct k_itimer *timr, ktime_t now) return ktime_sub(now, alarm->node.expires); } +/** + * alarm_timer_try_to_cancel - Posix timer callback to cancel a timer + * @timr: Pointer to the posixtimer data struct + */ +static int alarm_timer_try_to_cancel(struct k_itimer *timr) +{ + return alarm_try_to_cancel(&timr->it.alarm.alarmtimer); +} + /** * alarm_clock_getres - posix getres interface * @which_clock: clockid @@ -902,6 +911,7 @@ const struct k_clock alarm_clock = { .timer_rearm = alarm_timer_rearm, .timer_forward = alarm_timer_forward, .timer_remaining = alarm_timer_remaining, + .timer_try_to_cancel = alarm_timer_try_to_cancel, .nsleep = alarm_timer_nsleep, }; #endif /* CONFIG_POSIX_TIMERS */ -- cgit v1.2.3 From b3bf6f369d50ece9dec6338741648005d95c19e4 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 30 May 2017 23:15:58 +0200 Subject: alarmtimer: Implement arm callback Preparatory change to utilize the common posix timer mechanisms. Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: John Stultz Link: http://lkml.kernel.org/r/20170530211657.747567162@linutronix.de --- kernel/time/alarmtimer.c | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+) (limited to 'kernel/time') diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index 374bd855a488..c618a44bb054 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -581,6 +581,27 @@ static int alarm_timer_try_to_cancel(struct k_itimer *timr) return alarm_try_to_cancel(&timr->it.alarm.alarmtimer); } +/** + * alarm_timer_arm - Posix timer callback to arm a timer + * @timr: Pointer to the posixtimer data struct + * @expires: The new expiry time + * @absolute: Expiry value is absolute time + * @sigev_none: Posix timer does not deliver signals + */ +static void alarm_timer_arm(struct k_itimer *timr, ktime_t expires, + bool absolute, bool sigev_none) +{ + struct alarm *alarm = &timr->it.alarm.alarmtimer; + struct alarm_base *base = &alarm_bases[alarm->type]; + + if (!absolute) + expires = ktime_add_safe(expires, base->gettime()); + if (sigev_none) + alarm->node.expires = expires; + else + alarm_start(&timr->it.alarm.alarmtimer, expires); +} + /** * alarm_clock_getres - posix getres interface * @which_clock: clockid @@ -908,6 +929,7 @@ const struct k_clock alarm_clock = { .timer_set = alarm_timer_set, .timer_del = alarm_timer_del, .timer_get = alarm_timer_get, + .timer_arm = alarm_timer_arm, .timer_rearm = alarm_timer_rearm, .timer_forward = alarm_timer_forward, .timer_remaining = alarm_timer_remaining, -- cgit v1.2.3 From f2c45807d3992fe0f173f34af9c347d907c31686 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 30 May 2017 23:15:59 +0200 Subject: alarmtimer: Switch over to generic set/get/rearm routine All required callbacks are in place. Switch the alarm timer based posix interval timer callbacks to the common implementation and remove the incorrect private implementation. Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: John Stultz Link: http://lkml.kernel.org/r/20170530211657.825471962@linutronix.de --- kernel/time/alarmtimer.c | 121 +++++++-------------------------------------- kernel/time/posix-timers.c | 12 ++--- kernel/time/posix-timers.h | 6 +++ 3 files changed, 29 insertions(+), 110 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index c618a44bb054..d8a7a7e214de 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -515,20 +515,26 @@ static enum alarmtimer_type clock2alarm(clockid_t clockid) static enum alarmtimer_restart alarm_handle_timer(struct alarm *alarm, ktime_t now) { - unsigned long flags; struct k_itimer *ptr = container_of(alarm, struct k_itimer, - it.alarm.alarmtimer); + it.alarm.alarmtimer); enum alarmtimer_restart result = ALARMTIMER_NORESTART; + unsigned long flags; + int si_private = 0; spin_lock_irqsave(&ptr->it_lock, flags); - if ((ptr->it_sigev_notify & ~SIGEV_THREAD_ID) != SIGEV_NONE) { - if (posix_timer_event(ptr, 0)) - ptr->it_overrun++; - } - /* Re-add periodic timers */ - if (ptr->it_interval) { - ptr->it_overrun += alarm_forward(alarm, now, ptr->it_interval); + ptr->it_active = 0; + if (ptr->it_interval) + si_private = ++ptr->it_requeue_pending; + + if (posix_timer_event(ptr, si_private) && ptr->it_interval) { + /* + * Handle ignored signals and rearm the timer. This will go + * away once we handle ignored signals proper. + */ + ptr->it_overrun += alarm_forward_now(alarm, ptr->it_interval); + ++ptr->it_requeue_pending; + ptr->it_active = 1; result = ALARMTIMER_RESTART; } spin_unlock_irqrestore(&ptr->it_lock, flags); @@ -658,97 +664,6 @@ static int alarm_timer_create(struct k_itimer *new_timer) return 0; } -/** - * alarm_timer_get - posix timer_get interface - * @timr: k_itimer pointer - * @cur_setting: itimerspec data to fill - * - * Copies out the current itimerspec data - */ -static void alarm_timer_get(struct k_itimer *timr, - struct itimerspec64 *cur_setting) -{ - ktime_t relative_expiry_time = - alarm_expires_remaining(&(timr->it.alarm.alarmtimer)); - - if (ktime_to_ns(relative_expiry_time) > 0) { - cur_setting->it_value = ktime_to_timespec64(relative_expiry_time); - } else { - cur_setting->it_value.tv_sec = 0; - cur_setting->it_value.tv_nsec = 0; - } - - cur_setting->it_interval = ktime_to_timespec64(timr->it_interval); -} - -/** - * alarm_timer_del - posix timer_del interface - * @timr: k_itimer pointer to be deleted - * - * Cancels any programmed alarms for the given timer. - */ -static int alarm_timer_del(struct k_itimer *timr) -{ - if (!rtcdev) - return -ENOTSUPP; - - if (alarm_try_to_cancel(&timr->it.alarm.alarmtimer) < 0) - return TIMER_RETRY; - - return 0; -} - -/** - * alarm_timer_set - posix timer_set interface - * @timr: k_itimer pointer to be deleted - * @flags: timer flags - * @new_setting: itimerspec to be used - * @old_setting: itimerspec being replaced - * - * Sets the timer to new_setting, and starts the timer. - */ -static int alarm_timer_set(struct k_itimer *timr, int flags, - struct itimerspec64 *new_setting, - struct itimerspec64 *old_setting) -{ - ktime_t exp; - - if (!rtcdev) - return -ENOTSUPP; - - if (flags & ~TIMER_ABSTIME) - return -EINVAL; - - if (old_setting) - alarm_timer_get(timr, old_setting); - - /* If the timer was already set, cancel it */ - if (alarm_try_to_cancel(&timr->it.alarm.alarmtimer) < 0) - return TIMER_RETRY; - - /* start the timer */ - timr->it_interval = timespec64_to_ktime(new_setting->it_interval); - - /* - * Rate limit to the tick as a hot fix to prevent DOS. Will be - * mopped up later. - */ - if (timr->it_interval < TICK_NSEC) - timr->it_interval = TICK_NSEC; - - exp = timespec64_to_ktime(new_setting->it_value); - /* Convert (if necessary) to absolute time */ - if (flags != TIMER_ABSTIME) { - ktime_t now; - - now = alarm_bases[timr->it.alarm.alarmtimer.type].gettime(); - exp = ktime_add_safe(now, exp); - } - - alarm_start(&timr->it.alarm.alarmtimer, exp); - return 0; -} - /** * alarmtimer_nsleep_wakeup - Wakeup function for alarm_timer_nsleep * @alarm: ptr to alarm that fired @@ -926,9 +841,9 @@ const struct k_clock alarm_clock = { .clock_getres = alarm_clock_getres, .clock_get = alarm_clock_get, .timer_create = alarm_timer_create, - .timer_set = alarm_timer_set, - .timer_del = alarm_timer_del, - .timer_get = alarm_timer_get, + .timer_set = common_timer_set, + .timer_del = common_timer_del, + .timer_get = common_timer_get, .timer_arm = alarm_timer_arm, .timer_rearm = alarm_timer_rearm, .timer_forward = alarm_timer_forward, diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c index 8acc9ee2c2d6..6e7a70b1bf37 100644 --- a/kernel/time/posix-timers.c +++ b/kernel/time/posix-timers.c @@ -637,8 +637,7 @@ static int common_hrtimer_forward(struct k_itimer *timr, ktime_t now) * it is the same as a requeue pending timer WRT to what we should * report. */ -static void -common_timer_get(struct k_itimer *timr, struct itimerspec64 *cur_setting) +void common_timer_get(struct k_itimer *timr, struct itimerspec64 *cur_setting) { const struct k_clock *kc = timr->kclock; ktime_t now, remaining, iv; @@ -768,10 +767,9 @@ static int common_hrtimer_try_to_cancel(struct k_itimer *timr) } /* Set a POSIX.1b interval timer. */ -static int -common_timer_set(struct k_itimer *timr, int flags, - struct itimerspec64 *new_setting, - struct itimerspec64 *old_setting) +int common_timer_set(struct k_itimer *timr, int flags, + struct itimerspec64 *new_setting, + struct itimerspec64 *old_setting) { const struct k_clock *kc = timr->kclock; bool sigev_none; @@ -855,7 +853,7 @@ retry: return error; } -static int common_timer_del(struct k_itimer *timer) +int common_timer_del(struct k_itimer *timer) { const struct k_clock *kc = timer->kclock; diff --git a/kernel/time/posix-timers.h b/kernel/time/posix-timers.h index b0ad77e18886..b086f5ba2f5b 100644 --- a/kernel/time/posix-timers.h +++ b/kernel/time/posix-timers.h @@ -33,3 +33,9 @@ extern const struct k_clock clock_thread; extern const struct k_clock alarm_clock; int posix_timer_event(struct k_itimer *timr, int si_private); + +void common_timer_get(struct k_itimer *timr, struct itimerspec64 *cur_setting); +int common_timer_set(struct k_itimer *timr, int flags, + struct itimerspec64 *new_setting, + struct itimerspec64 *old_setting); +int common_timer_del(struct k_itimer *timer); -- cgit v1.2.3 From f99973e18b65ca1fff8c81532e3132b8f622aea8 Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Thu, 1 Jun 2017 16:47:09 +0200 Subject: nohz: Fix buggy tick delay on IRQ storms When the tick is stopped and we reach the dynticks evaluation code on IRQ exit, we perform a soft tick restart if we observe an expired timer from there. It means we program the nearest possible tick but we stay in dynticks mode (ts->tick_stopped = 1) because we may need to stop the tick again after that expired timer is handled. Now this solution works most of the time but if we suffer an IRQ storm and those interrupts trigger faster than the hardware clockevents min delay, our tick won't fire until that IRQ storm is finished. Here is the problem: on IRQ exit we reprog the timer to at least NOW() + min_clockevents_delay. Another IRQ fires before the tick so we reschedule again to NOW() + min_clockevents_delay, etc... The tick is eternally rescheduled min_clockevents_delay ahead. A solution is to simply remove this soft tick restart. After all the normal dynticks evaluation path can handle 0 delay just fine. And by doing that we benefit from the optimization branch which avoids clock reprogramming if the clockevents deadline hasn't changed since the last reprog. This fixes our issue because we don't do repetitive clock reprog that always add hardware min delay. As a side effect it should even optimize the 0 delay path in general. Reported-and-tested-by: Octavian Purdila Signed-off-by: Frederic Weisbecker Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Rik van Riel Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/1496328429-13317-1-git-send-email-fweisbec@gmail.com Signed-off-by: Ingo Molnar --- kernel/time/tick-sched.c | 21 ++------------------- 1 file changed, 2 insertions(+), 19 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c index e3043873fcdc..9d31f1e0067b 100644 --- a/kernel/time/tick-sched.c +++ b/kernel/time/tick-sched.c @@ -713,8 +713,6 @@ static ktime_t tick_nohz_stop_sched_tick(struct tick_sched *ts, */ delta = next_tick - basemono; if (delta <= (u64)TICK_NSEC) { - tick = 0; - /* * Tell the timer code that the base is not idle, i.e. undo * the effect of get_next_timer_interrupt(): @@ -724,23 +722,8 @@ static ktime_t tick_nohz_stop_sched_tick(struct tick_sched *ts, * We've not stopped the tick yet, and there's a timer in the * next period, so no point in stopping it either, bail. */ - if (!ts->tick_stopped) - goto out; - - /* - * If, OTOH, we did stop it, but there's a pending (expired) - * timer reprogram the timer hardware to fire now. - * - * We will not restart the tick proper, just prod the timer - * hardware into firing an interrupt to process the pending - * timers. Just like tick_irq_exit() will not restart the tick - * for 'normal' interrupts. - * - * Only once we exit the idle loop will we re-enable the tick, - * see tick_nohz_idle_exit(). - */ - if (delta == 0) { - tick_nohz_restart(ts, now); + if (!ts->tick_stopped) { + tick = 0; goto out; } } -- cgit v1.2.3 From fe5ac724d81a3c7803e60c2232718f212f3f38d4 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Thu, 11 May 2017 11:26:22 -0700 Subject: rcu: Remove nohz_full full-system-idle state machine The NO_HZ_FULL_SYSIDLE full-system-idle capability was added in 2013 by commit 0edd1b1784cb ("nohz_full: Add full-system-idle state machine"), but has not been used. This commit therefore removes it. If it turns out to be needed later, this commit can always be reverted. Signed-off-by: Paul E. McKenney Cc: Frederic Weisbecker Cc: Rik van Riel Cc: Ingo Molnar Acked-by: Linus Torvalds --- kernel/time/Kconfig | 50 -------------------------------------------------- 1 file changed, 50 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/Kconfig b/kernel/time/Kconfig index 4008d9f95dd7..ac09bc29eb08 100644 --- a/kernel/time/Kconfig +++ b/kernel/time/Kconfig @@ -126,56 +126,6 @@ config NO_HZ_FULL_ALL Note the boot CPU will still be kept outside the range to handle the timekeeping duty. -config NO_HZ_FULL_SYSIDLE - bool "Detect full-system idle state for full dynticks system" - depends on NO_HZ_FULL - default n - help - At least one CPU must keep the scheduling-clock tick running for - timekeeping purposes whenever there is a non-idle CPU, where - "non-idle" also includes dynticks CPUs as long as they are - running non-idle tasks. Because the underlying adaptive-tick - support cannot distinguish between all CPUs being idle and - all CPUs each running a single task in dynticks mode, the - underlying support simply ensures that there is always a CPU - handling the scheduling-clock tick, whether or not all CPUs - are idle. This Kconfig option enables scalable detection of - the all-CPUs-idle state, thus allowing the scheduling-clock - tick to be disabled when all CPUs are idle. Note that scalable - detection of the all-CPUs-idle state means that larger systems - will be slower to declare the all-CPUs-idle state. - - Say Y if you would like to help debug all-CPUs-idle detection. - - Say N if you are unsure. - -config NO_HZ_FULL_SYSIDLE_SMALL - int "Number of CPUs above which large-system approach is used" - depends on NO_HZ_FULL_SYSIDLE - range 1 NR_CPUS - default 8 - help - The full-system idle detection mechanism takes a lazy approach - on large systems, as is required to attain decent scalability. - However, on smaller systems, scalability is not anywhere near as - large a concern as is energy efficiency. The sysidle subsystem - therefore uses a fast but non-scalable algorithm for small - systems and a lazier but scalable algorithm for large systems. - This Kconfig parameter defines the number of CPUs in the largest - system that will be considered to be "small". - - The default value will be fine in most cases. Battery-powered - systems that (1) enable NO_HZ_FULL_SYSIDLE, (2) have larger - numbers of CPUs, and (3) are suffering from battery-lifetime - problems due to long sysidle latencies might wish to experiment - with larger values for this Kconfig parameter. On the other - hand, they might be even better served by disabling NO_HZ_FULL - entirely, given that NO_HZ_FULL is intended for HPC and - real-time workloads that at present do not tend to be run on - battery-powered systems. - - Take the default if you are unsure. - config NO_HZ bool "Old Idle dynticks config" depends on !ARCH_USES_GETTIMEOFFSET && GENERIC_CLOCKEVENTS -- cgit v1.2.3 From c6503be587e9c5c0aac4e2b45de982352f676a5b Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Mon, 12 Jun 2017 17:21:26 +0200 Subject: posix-timers: Fix inverted SIGEV_NONE logic in common_timer_get() The refactoring of the posix-timer core to allow better code sharing introduced inverted logic vs. SIGEV_NONE timers in common_timer_get(). That causes hrtimer_forward() to be called on active timers, which rightfully triggers the warning hrtimer_forward(). Make sig_none what it says: signal mode == SIGEV_NONE. Fixes: 91d57bae0868 ("posix-timers: Make use of forward/remaining callbacks") Reported-by: Ye Xiaolong Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: John Stultz Link: http://lkml.kernel.org/r/20170609104457.GA39907@inn.lkp.intel.com --- kernel/time/posix-timers.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/time') diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c index 6e7a70b1bf37..b53a0b562516 100644 --- a/kernel/time/posix-timers.c +++ b/kernel/time/posix-timers.c @@ -644,7 +644,7 @@ void common_timer_get(struct k_itimer *timr, struct itimerspec64 *cur_setting) struct timespec64 ts64; bool sig_none; - sig_none = (timr->it_sigev_notify & ~SIGEV_THREAD_ID) != SIGEV_NONE; + sig_none = (timr->it_sigev_notify & ~SIGEV_THREAD_ID) == SIGEV_NONE; iv = timr->it_interval; /* interval timer ? */ -- cgit v1.2.3 From 94114c367553f3301747e47f6947cabde947575f Mon Sep 17 00:00:00 2001 From: Stephen Boyd Date: Wed, 7 Jun 2017 23:36:03 -0700 Subject: tick/broadcast: Make tick_broadcast_setup_oneshot() static This function isn't used outside of tick-broadcast.c, so let's mark it static. Signed-off-by: Stephen Boyd Link: http://lkml.kernel.org/r/20170608063603.13276-1-sboyd@codeaurora.org Signed-off-by: Thomas Gleixner --- kernel/time/tick-broadcast.c | 4 +++- kernel/time/tick-internal.h | 2 -- 2 files changed, 3 insertions(+), 3 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c index 987e496bb51a..b398c2ea69b2 100644 --- a/kernel/time/tick-broadcast.c +++ b/kernel/time/tick-broadcast.c @@ -37,9 +37,11 @@ static int tick_broadcast_forced; static __cacheline_aligned_in_smp DEFINE_RAW_SPINLOCK(tick_broadcast_lock); #ifdef CONFIG_TICK_ONESHOT +static void tick_broadcast_setup_oneshot(struct clock_event_device *bc); static void tick_broadcast_clear_oneshot(int cpu); static void tick_resume_broadcast_oneshot(struct clock_event_device *bc); #else +static inline void tick_broadcast_setup_oneshot(struct clock_event_device *bc) { BUG(); } static inline void tick_broadcast_clear_oneshot(int cpu) { } static inline void tick_resume_broadcast_oneshot(struct clock_event_device *bc) { } #endif @@ -867,7 +869,7 @@ static void tick_broadcast_init_next_event(struct cpumask *mask, /** * tick_broadcast_setup_oneshot - setup the broadcast device */ -void tick_broadcast_setup_oneshot(struct clock_event_device *bc) +static void tick_broadcast_setup_oneshot(struct clock_event_device *bc) { int cpu = smp_processor_id(); diff --git a/kernel/time/tick-internal.h b/kernel/time/tick-internal.h index f738251000fe..be0ac01f2e12 100644 --- a/kernel/time/tick-internal.h +++ b/kernel/time/tick-internal.h @@ -126,7 +126,6 @@ static inline int tick_check_oneshot_change(int allow_nohz) { return 0; } /* Functions related to oneshot broadcasting */ #if defined(CONFIG_GENERIC_CLOCKEVENTS_BROADCAST) && defined(CONFIG_TICK_ONESHOT) -extern void tick_broadcast_setup_oneshot(struct clock_event_device *bc); extern void tick_broadcast_switch_to_oneshot(void); extern void tick_shutdown_broadcast_oneshot(unsigned int cpu); extern int tick_broadcast_oneshot_active(void); @@ -134,7 +133,6 @@ extern void tick_check_oneshot_broadcast_this_cpu(void); bool tick_broadcast_oneshot_available(void); extern struct cpumask *tick_get_broadcast_oneshot_mask(void); #else /* !(BROADCAST && ONESHOT): */ -static inline void tick_broadcast_setup_oneshot(struct clock_event_device *bc) { BUG(); } static inline void tick_broadcast_switch_to_oneshot(void) { } static inline void tick_shutdown_broadcast_oneshot(unsigned int cpu) { } static inline int tick_broadcast_oneshot_active(void) { return 0; } -- cgit v1.2.3 From 5c7a3a3d20a4e175304c0e23809e3d70be8fed8a Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Mon, 12 Jun 2017 19:44:09 +0200 Subject: posix-timers: Zero out oldval itimerspec The recent posix timer rework moved the clearing of the itimerspec to the real syscall implementation, but forgot that the kclock->timer_get() is used by timer_settime() as well. That results in an uninitialized variable and bogus values returned to user space. Add the missing memset to timer_settime(). Fixes: eabdec043853 ("posix-timers: Zero settings value in common code") Reported-by: Andrei Vagin Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: John Stultz Cc: Cyrill Gorcunov Link: http://lkml.kernel.org/r/20170609201156.GB21491@outlook.office365.com --- kernel/time/posix-timers.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'kernel/time') diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c index b53a0b562516..88517dcfe0ca 100644 --- a/kernel/time/posix-timers.c +++ b/kernel/time/posix-timers.c @@ -828,6 +828,8 @@ SYSCALL_DEFINE4(timer_settime, timer_t, timer_id, int, flags, if (!timespec64_valid(&new_spec64.it_interval) || !timespec64_valid(&new_spec64.it_value)) return -EINVAL; + if (rtn) + memset(rtn, 0, sizeof(*rtn)); retry: timr = lock_timer(timer_id, &flag); if (!timr) -- cgit v1.2.3 From 67edab48caeb75d412706f4b9d3107afd1e07623 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Mon, 12 Jun 2017 19:39:49 +0200 Subject: posix-timers: Handle relative posix-timers correctly The recent rework of the posix timer internals broke the magic posix mechanism, which requires that relative timers are not affected by modifications of the underlying clock. That means relative CLOCK_REALTIME timers cannot use CLOCK_REALTIME, because that can be set and adjusted. The underlying hrtimer switches the clock for these timers to CLOCK_MONOTONIC. That still works, but reading the remaining time of such a timer has been broken in the rework. The old code used the hrtimer internals directly and avoided the posix clock callbacks. Now common_timer_get() uses the underlying kclock->timer_get() callback, which is still CLOCK_REALTIME based. So the remaining time of such a timer is calculated against the wrong time base. Handle it by switching the k_itimer->kclock pointer according to the resulting hrtimer mode. k_itimer->it_clock still contains CLOCK_REALTIME because the timer might be set with ABSTIME later and then it needs to switch back to the realtime posix clock implementation. Fixes: eae1c4ae275f ("posix-timers: Make use of cancel/arm callbacks") Reported-by: Andrei Vagin Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: John Stultz Cc: Cyrill Gorcunov Link: http://lkml.kernel.org/r/20170609201156.GB21491@outlook.office365.com --- kernel/time/posix-timers.c | 13 +++++++++++++ 1 file changed, 13 insertions(+) (limited to 'kernel/time') diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c index 88517dcfe0ca..58c0f60b132f 100644 --- a/kernel/time/posix-timers.c +++ b/kernel/time/posix-timers.c @@ -72,6 +72,7 @@ static DEFINE_SPINLOCK(hash_lock); static const struct k_clock * const posix_clocks[]; static const struct k_clock *clockid_to_kclock(const clockid_t id); +static const struct k_clock clock_realtime, clock_monotonic; /* * we assume that the new SIGEV_THREAD_ID shares no bits with the other @@ -750,6 +751,18 @@ static void common_hrtimer_arm(struct k_itimer *timr, ktime_t expires, enum hrtimer_mode mode; mode = absolute ? HRTIMER_MODE_ABS : HRTIMER_MODE_REL; + /* + * Posix magic: Relative CLOCK_REALTIME timers are not affected by + * clock modifications, so they become CLOCK_MONOTONIC based under the + * hood. See hrtimer_init(). Update timr->kclock, so the generic + * functions which use timr->kclock->clock_get() work. + * + * Note: it_clock stays unmodified, because the next timer_set() might + * use ABSTIME, so it needs to switch back. + */ + if (timr->it_clock == CLOCK_REALTIME) + timr->kclock = absolute ? &clock_realtime : &clock_monotonic; + hrtimer_init(&timr->it.real.timer, timr->it_clock, mode); timr->it.real.timer.function = posix_timer_fn; -- cgit v1.2.3 From d4af6d933ccffd24286528f04d5c39e702c8580f Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Tue, 13 Jun 2017 06:04:14 +0200 Subject: nohz: Fix spurious warning when hrtimer and clockevent get out of sync The sanity check ensuring that the tick expiry cache (ts->next_tick) is actually in sync with the hardware clock (dev->next_event) makes the wrong assumption that the clock can't be programmed later than the hrtimer deadline. In fact the clock hardware can be programmed later on some conditions such as: * The hrtimer deadline is already in the past. * The hrtimer deadline is earlier than the minimum delay supported by the hardware. Such conditions can be met when we program the tick, for example if the last jiffies update hasn't been seen by the current CPU yet, we may program the hrtimer to a deadline that is earlier than ktime_get() because last_jiffies_update is our timestamp base to compute the next tick. As a result, we can randomly observe such warning: WARNING: CPU: 5 PID: 0 at kernel/time/tick-sched.c:794 tick_nohz_stop_sched_tick kernel/time/tick-sched.c:791 [inline] Call Trace: tick_nohz_irq_exit tick_irq_exit irq_exit exiting_irq smp_call_function_interrupt smp_call_function_single_interrupt call_function_single_interrupt Therefore, let's rather make sure that the tick expiry cache is sync'ed with the tick hrtimer deadline, against which it is not supposed to drift away. The clock hardware instead has its own will and can't be used as a reliable comparison point. Reported-and-tested-by: Sasha Levin Reported-and-tested-by: Abdul Haleem Signed-off-by: Frederic Weisbecker Cc: James Hartsock Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Rik van Riel Cc: Thomas Gleixner Cc: Tim Wright Link: http://lkml.kernel.org/r/1497326654-14122-1-git-send-email-fweisbec@gmail.com [ Minor readability edit. ] Signed-off-by: Ingo Molnar --- kernel/time/tick-sched.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c index 9d31f1e0067b..204600986e0d 100644 --- a/kernel/time/tick-sched.c +++ b/kernel/time/tick-sched.c @@ -768,7 +768,7 @@ static ktime_t tick_nohz_stop_sched_tick(struct tick_sched *ts, /* Skip reprogram of event if its not changed */ if (ts->tick_stopped && (expires == ts->next_tick)) { /* Sanity check: make sure clockevent is actually programmed */ - if (likely(dev->next_event <= ts->next_tick)) + if (tick == KTIME_MAX || ts->next_tick == hrtimer_get_expires(&ts->sched_timer)) goto out; WARN_ON_ONCE(1); @@ -806,8 +806,10 @@ static ktime_t tick_nohz_stop_sched_tick(struct tick_sched *ts, goto out; } + hrtimer_set_expires(&ts->sched_timer, tick); + if (ts->nohz_mode == NOHZ_MODE_HIGHRES) - hrtimer_start(&ts->sched_timer, tick, HRTIMER_MODE_ABS_PINNED); + hrtimer_start_expires(&ts->sched_timer, HRTIMER_MODE_ABS_PINNED); else tick_program_event(tick, 1); out: -- cgit v1.2.3 From 86a9c446c13ecd8793ea8599761322aed125d542 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 7 Jun 2017 09:42:26 +0100 Subject: posix-cpu-timers: Move copyout of timespec into do_cpu_nanosleep() The posix-cpu-timer nanosleep() implementation can be simplified by moving the copy out of the remaining time to do_cpu_nanosleep() which is shared between the real nanosleep function and the restart function. The pointer to the timespec64 which is updated has to be stored in the restart block anyway. Instead of storing it only in the restart case, store it before calling do_cpu_nanosleep() and copy the remaining time in the signal exit path. [ tglx: Added changelog ] Signed-off-by: Al Viro Signed-off-by: Thomas Gleixner Cc: John Stultz Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20170607084241.28657-1-viro@ZenIV.linux.org.uk --- kernel/time/posix-cpu-timers.c | 63 +++++++++++++++++++----------------------- 1 file changed, 28 insertions(+), 35 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/posix-cpu-timers.c b/kernel/time/posix-cpu-timers.c index cb4a4eb44279..239fff980fd0 100644 --- a/kernel/time/posix-cpu-timers.c +++ b/kernel/time/posix-cpu-timers.c @@ -1226,9 +1226,10 @@ void set_process_cpu_timer(struct task_struct *tsk, unsigned int clock_idx, } static int do_cpu_nanosleep(const clockid_t which_clock, int flags, - struct timespec64 *rqtp, struct itimerspec64 *it) + struct timespec64 *rqtp) { struct k_itimer timer; + struct itimerspec64 it; int error; /* @@ -1242,12 +1243,14 @@ static int do_cpu_nanosleep(const clockid_t which_clock, int flags, timer.it_process = current; if (!error) { static struct itimerspec64 zero_it; + struct restart_block *restart = ¤t->restart_block; + struct timespec __user *rmtp; - memset(it, 0, sizeof *it); - it->it_value = *rqtp; + memset(&it, 0, sizeof it); + it.it_value = *rqtp; spin_lock_irq(&timer.it_lock); - error = posix_cpu_timer_set(&timer, flags, it, NULL); + error = posix_cpu_timer_set(&timer, flags, &it, NULL); if (error) { spin_unlock_irq(&timer.it_lock); return error; @@ -1277,7 +1280,7 @@ static int do_cpu_nanosleep(const clockid_t which_clock, int flags, * We were interrupted by a signal. */ *rqtp = ns_to_timespec64(timer.it.cpu.expires); - error = posix_cpu_timer_set(&timer, 0, &zero_it, it); + error = posix_cpu_timer_set(&timer, 0, &zero_it, &it); if (!error) { /* * Timer is now unarmed, deletion can not fail. @@ -1297,7 +1300,7 @@ static int do_cpu_nanosleep(const clockid_t which_clock, int flags, spin_unlock_irq(&timer.it_lock); } - if ((it->it_value.tv_sec | it->it_value.tv_nsec) == 0) { + if ((it.it_value.tv_sec | it.it_value.tv_nsec) == 0) { /* * It actually did fire already. */ @@ -1305,6 +1308,18 @@ static int do_cpu_nanosleep(const clockid_t which_clock, int flags, } error = -ERESTART_RESTARTBLOCK; + /* + * Report back to the user the time still remaining. + */ + rmtp = restart->nanosleep.rmtp; + if (rmtp) { + struct timespec ts; + + ts = timespec64_to_timespec(it.it_value); + if (copy_to_user(rmtp, &ts, sizeof(*rmtp))) + return -EFAULT; + } + restart->nanosleep.expires = timespec64_to_ns(rqtp); } return error; @@ -1316,10 +1331,13 @@ static int posix_cpu_nsleep(const clockid_t which_clock, int flags, struct timespec64 *rqtp, struct timespec __user *rmtp) { struct restart_block *restart_block = ¤t->restart_block; - struct itimerspec64 it; - struct timespec ts; int error; + if (flags & TIMER_ABSTIME) + rmtp = NULL; + + restart_block->nanosleep.rmtp = rmtp; + /* * Diagnose required errors first. */ @@ -1328,23 +1346,15 @@ static int posix_cpu_nsleep(const clockid_t which_clock, int flags, CPUCLOCK_PID(which_clock) == task_pid_vnr(current))) return -EINVAL; - error = do_cpu_nanosleep(which_clock, flags, rqtp, &it); + error = do_cpu_nanosleep(which_clock, flags, rqtp); if (error == -ERESTART_RESTARTBLOCK) { if (flags & TIMER_ABSTIME) return -ERESTARTNOHAND; - /* - * Report back to the user the time still remaining. - */ - ts = timespec64_to_timespec(it.it_value); - if (rmtp && copy_to_user(rmtp, &ts, sizeof(*rmtp))) - return -EFAULT; restart_block->fn = posix_cpu_nsleep_restart; restart_block->nanosleep.clockid = which_clock; - restart_block->nanosleep.rmtp = rmtp; - restart_block->nanosleep.expires = timespec64_to_ns(rqtp); } return error; } @@ -1352,28 +1362,11 @@ static int posix_cpu_nsleep(const clockid_t which_clock, int flags, static long posix_cpu_nsleep_restart(struct restart_block *restart_block) { clockid_t which_clock = restart_block->nanosleep.clockid; - struct itimerspec64 it; struct timespec64 t; - struct timespec tmp; - int error; t = ns_to_timespec64(restart_block->nanosleep.expires); - error = do_cpu_nanosleep(which_clock, TIMER_ABSTIME, &t, &it); - - if (error == -ERESTART_RESTARTBLOCK) { - struct timespec __user *rmtp = restart_block->nanosleep.rmtp; - /* - * Report back to the user the time still remaining. - */ - tmp = timespec64_to_timespec(it.it_value); - if (rmtp && copy_to_user(rmtp, &tmp, sizeof(*rmtp))) - return -EFAULT; - - restart_block->nanosleep.expires = timespec64_to_ns(&t); - } - return error; - + return do_cpu_nanosleep(which_clock, TIMER_ABSTIME, &t); } #define PROCESS_CLOCK MAKE_PROCESS_CPUCLOCK(0, CPUCLOCK_SCHED) -- cgit v1.2.3 From 15f27ce24cb613e6e01ce27c4094c55e55dde5d4 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 7 Jun 2017 09:42:27 +0100 Subject: alarmtimer: Move copyout and freeze handling into alarmtimer_do_nsleep() The alarmtimer nanosleep() implementation can be simplified by moving the copy out of the remaining time to alarmtimer_do_nsleep() which is shared between the real nanosleep function and the restart function. The pointer to the timespec64 which is updated has to be stored in the restart block anyway. Instead of storing it only in the restart case, store it before calling alarmtimer_do_nsleep() and copy the remaining time in the signal exit path. [ tglx: Added changelog ] Signed-off-by: Al Viro Signed-off-by: Thomas Gleixner Cc: John Stultz Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20170607084241.28657-2-viro@ZenIV.linux.org.uk --- kernel/time/alarmtimer.c | 102 +++++++++++++++-------------------------------- 1 file changed, 32 insertions(+), 70 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index d8a7a7e214de..ac6e9bc6cc59 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -688,8 +688,10 @@ static enum alarmtimer_restart alarmtimer_nsleep_wakeup(struct alarm *alarm, * * Sets the alarm timer and sleeps until it is fired or interrupted. */ -static int alarmtimer_do_nsleep(struct alarm *alarm, ktime_t absexp) +static int alarmtimer_do_nsleep(struct alarm *alarm, ktime_t absexp, + enum alarmtimer_type type) { + struct timespec __user *rmtp; alarm->data = (void *)current; do { set_current_state(TASK_INTERRUPTIBLE); @@ -702,36 +704,26 @@ static int alarmtimer_do_nsleep(struct alarm *alarm, ktime_t absexp) __set_current_state(TASK_RUNNING); - return (alarm->data == NULL); -} - - -/** - * update_rmtp - Update remaining timespec value - * @exp: expiration time - * @type: timer type - * @rmtp: user pointer to remaining timepsec value - * - * Helper function that fills in rmtp value with time between - * now and the exp value - */ -static int update_rmtp(ktime_t exp, enum alarmtimer_type type, - struct timespec __user *rmtp) -{ - struct timespec rmt; - ktime_t rem; - - rem = ktime_sub(exp, alarm_bases[type].gettime()); - - if (rem <= 0) + if (!alarm->data) return 0; - rmt = ktime_to_timespec(rem); - if (copy_to_user(rmtp, &rmt, sizeof(*rmtp))) - return -EFAULT; + if (freezing(current)) + alarmtimer_freezerset(absexp, type); + rmtp = current->restart_block.nanosleep.rmtp; + if (rmtp) { + struct timespec rmt; + ktime_t rem; - return 1; + rem = ktime_sub(absexp, alarm_bases[type].gettime()); + if (rem <= 0) + return 0; + rmt = ktime_to_timespec(rem); + + if (copy_to_user(rmtp, &rmt, sizeof(*rmtp))) + return -EFAULT; + } + return -ERESTART_RESTARTBLOCK; } /** @@ -743,32 +735,12 @@ static int update_rmtp(ktime_t exp, enum alarmtimer_type type, static long __sched alarm_timer_nsleep_restart(struct restart_block *restart) { enum alarmtimer_type type = restart->nanosleep.clockid; - ktime_t exp; - struct timespec __user *rmtp; + ktime_t exp = restart->nanosleep.expires; struct alarm alarm; - int ret = 0; - exp = restart->nanosleep.expires; alarm_init(&alarm, type, alarmtimer_nsleep_wakeup); - if (alarmtimer_do_nsleep(&alarm, exp)) - goto out; - - if (freezing(current)) - alarmtimer_freezerset(exp, type); - - rmtp = restart->nanosleep.rmtp; - if (rmtp) { - ret = update_rmtp(exp, type, rmtp); - if (ret <= 0) - goto out; - } - - - /* The other values in restart are already filled in */ - ret = -ERESTART_RESTARTBLOCK; -out: - return ret; + return alarmtimer_do_nsleep(&alarm, exp, type); } /** @@ -785,11 +757,16 @@ static int alarm_timer_nsleep(const clockid_t which_clock, int flags, struct timespec __user *rmtp) { enum alarmtimer_type type = clock2alarm(which_clock); - struct restart_block *restart; + struct restart_block *restart = ¤t->restart_block; struct alarm alarm; ktime_t exp; int ret = 0; + if (flags & TIMER_ABSTIME) + rmtp = NULL; + + restart->nanosleep.rmtp = rmtp; + if (!alarmtimer_get_rtcdev()) return -ENOTSUPP; @@ -808,32 +785,17 @@ static int alarm_timer_nsleep(const clockid_t which_clock, int flags, exp = ktime_add(now, exp); } - if (alarmtimer_do_nsleep(&alarm, exp)) - goto out; - - if (freezing(current)) - alarmtimer_freezerset(exp, type); + ret = alarmtimer_do_nsleep(&alarm, exp, type); + if (ret != -ERESTART_RESTARTBLOCK) + return ret; /* abs timers don't set remaining time or restart */ - if (flags == TIMER_ABSTIME) { - ret = -ERESTARTNOHAND; - goto out; - } - - if (rmtp) { - ret = update_rmtp(exp, type, rmtp); - if (ret <= 0) - goto out; - } + if (flags == TIMER_ABSTIME) + return -ERESTARTNOHAND; - restart = ¤t->restart_block; restart->fn = alarm_timer_nsleep_restart; restart->nanosleep.clockid = type; restart->nanosleep.expires = exp; - restart->nanosleep.rmtp = rmtp; - ret = -ERESTART_RESTARTBLOCK; - -out: return ret; } -- cgit v1.2.3 From 192a82f9003fe8fabd6088aa646e829225a94c55 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 7 Jun 2017 09:42:28 +0100 Subject: hrtimer_nanosleep(): Pass rmtp in restart_block Store the pointer to the timespec which gets updated with the remaining time in the restart block and remove the function argument. [ tglx: Added changelog ] Signed-off-by: Al Viro Signed-off-by: Thomas Gleixner Cc: John Stultz Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20170607084241.28657-3-viro@ZenIV.linux.org.uk --- kernel/time/hrtimer.c | 11 ++++++----- kernel/time/posix-stubs.c | 5 ++++- kernel/time/posix-timers.c | 5 ++++- 3 files changed, 14 insertions(+), 7 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c index ac053bb5296e..4ae777f159de 100644 --- a/kernel/time/hrtimer.c +++ b/kernel/time/hrtimer.c @@ -1503,10 +1503,11 @@ out: return ret; } -long hrtimer_nanosleep(struct timespec64 *rqtp, struct timespec __user *rmtp, +long hrtimer_nanosleep(struct timespec64 *rqtp, const enum hrtimer_mode mode, const clockid_t clockid) { - struct restart_block *restart; + struct restart_block *restart = ¤t->restart_block; + struct timespec __user *rmtp; struct hrtimer_sleeper t; int ret = 0; u64 slack; @@ -1526,16 +1527,15 @@ long hrtimer_nanosleep(struct timespec64 *rqtp, struct timespec __user *rmtp, goto out; } + rmtp = restart->nanosleep.rmtp; if (rmtp) { ret = update_rmtp(&t.timer, rmtp); if (ret <= 0) goto out; } - restart = ¤t->restart_block; restart->fn = hrtimer_nanosleep_restart; restart->nanosleep.clockid = t.timer.base->clockid; - restart->nanosleep.rmtp = rmtp; restart->nanosleep.expires = hrtimer_get_expires_tv64(&t.timer); ret = -ERESTART_RESTARTBLOCK; @@ -1557,7 +1557,8 @@ SYSCALL_DEFINE2(nanosleep, struct timespec __user *, rqtp, if (!timespec64_valid(&tu64)) return -EINVAL; - return hrtimer_nanosleep(&tu64, rmtp, HRTIMER_MODE_REL, CLOCK_MONOTONIC); + current->restart_block.nanosleep.rmtp = rmtp; + return hrtimer_nanosleep(&tu64, HRTIMER_MODE_REL, CLOCK_MONOTONIC); } /* diff --git a/kernel/time/posix-stubs.c b/kernel/time/posix-stubs.c index c0cd53eb018a..156a5e6f3bd2 100644 --- a/kernel/time/posix-stubs.c +++ b/kernel/time/posix-stubs.c @@ -115,7 +115,10 @@ SYSCALL_DEFINE4(clock_nanosleep, const clockid_t, which_clock, int, flags, t64 = timespec_to_timespec64(t); if (!timespec64_valid(&t64)) return -EINVAL; - return hrtimer_nanosleep(&t64, rmtp, flags & TIMER_ABSTIME ? + if (flags & TIMER_ABSTIME) + rmtp = NULL; + current->restart_block.nanosleep.rmtp = rmtp; + return hrtimer_nanosleep(&t64, flags & TIMER_ABSTIME ? HRTIMER_MODE_ABS : HRTIMER_MODE_REL, which_clock); default: diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c index 58c0f60b132f..1a9f59f8afc2 100644 --- a/kernel/time/posix-timers.c +++ b/kernel/time/posix-timers.c @@ -1043,7 +1043,10 @@ SYSCALL_DEFINE2(clock_getres, const clockid_t, which_clock, static int common_nsleep(const clockid_t which_clock, int flags, struct timespec64 *tsave, struct timespec __user *rmtp) { - return hrtimer_nanosleep(tsave, rmtp, flags & TIMER_ABSTIME ? + if (flags & TIMER_ABSTIME) + rmtp = NULL; + current->restart_block.nanosleep.rmtp = rmtp; + return hrtimer_nanosleep(tsave, flags & TIMER_ABSTIME ? HRTIMER_MODE_ABS : HRTIMER_MODE_REL, which_clock); } -- cgit v1.2.3 From a7602681fc63f1a3ddd3da336296c9634c2ff974 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 7 Jun 2017 09:42:29 +0100 Subject: hrtimer: Move copyout of remaining time to do_nanosleep() The hrtimer nanosleep() implementation can be simplified by moving the copy out of the remaining time to do_nanosleep() which is shared between the real nanosleep function and the restart function. The pointer to the timespec64 which is updated is already stored in the restart block at the call site, so the seperate handling of nanosleep and restart function can be avoided. [ tglx: Added changelog ] Signed-off-by: Al Viro Signed-off-by: Thomas Gleixner Cc: John Stultz Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20170607084241.28657-4-viro@ZenIV.linux.org.uk --- kernel/time/hrtimer.c | 62 +++++++++++++++++---------------------------------- 1 file changed, 20 insertions(+), 42 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c index 4ae777f159de..baa7b846b6e3 100644 --- a/kernel/time/hrtimer.c +++ b/kernel/time/hrtimer.c @@ -1441,6 +1441,7 @@ EXPORT_SYMBOL_GPL(hrtimer_init_sleeper); static int __sched do_nanosleep(struct hrtimer_sleeper *t, enum hrtimer_mode mode) { + struct timespec __user *rmtp; hrtimer_init_sleeper(t, current); do { @@ -1457,48 +1458,33 @@ static int __sched do_nanosleep(struct hrtimer_sleeper *t, enum hrtimer_mode mod __set_current_state(TASK_RUNNING); - return t->task == NULL; -} - -static int update_rmtp(struct hrtimer *timer, struct timespec __user *rmtp) -{ - struct timespec rmt; - ktime_t rem; - - rem = hrtimer_expires_remaining(timer); - if (rem <= 0) + if (!t->task) return 0; - rmt = ktime_to_timespec(rem); - if (copy_to_user(rmtp, &rmt, sizeof(*rmtp))) - return -EFAULT; - - return 1; + rmtp = current->restart_block.nanosleep.rmtp; + if (rmtp) { + struct timespec rmt; + ktime_t rem = hrtimer_expires_remaining(&t->timer); + if (rem <= 0) + return 0; + rmt = ktime_to_timespec(rem); + + if (copy_to_user(rmtp, &rmt, sizeof(*rmtp))) + return -EFAULT; + } + return -ERESTART_RESTARTBLOCK; } long __sched hrtimer_nanosleep_restart(struct restart_block *restart) { struct hrtimer_sleeper t; - struct timespec __user *rmtp; - int ret = 0; + int ret; hrtimer_init_on_stack(&t.timer, restart->nanosleep.clockid, HRTIMER_MODE_ABS); hrtimer_set_expires_tv64(&t.timer, restart->nanosleep.expires); - if (do_nanosleep(&t, HRTIMER_MODE_ABS)) - goto out; - - rmtp = restart->nanosleep.rmtp; - if (rmtp) { - ret = update_rmtp(&t.timer, rmtp); - if (ret <= 0) - goto out; - } - - /* The other values in restart are already filled in */ - ret = -ERESTART_RESTARTBLOCK; -out: + ret = do_nanosleep(&t, HRTIMER_MODE_ABS); destroy_hrtimer_on_stack(&t.timer); return ret; } @@ -1506,8 +1492,7 @@ out: long hrtimer_nanosleep(struct timespec64 *rqtp, const enum hrtimer_mode mode, const clockid_t clockid) { - struct restart_block *restart = ¤t->restart_block; - struct timespec __user *rmtp; + struct restart_block *restart; struct hrtimer_sleeper t; int ret = 0; u64 slack; @@ -1518,7 +1503,8 @@ long hrtimer_nanosleep(struct timespec64 *rqtp, hrtimer_init_on_stack(&t.timer, clockid, mode); hrtimer_set_expires_range_ns(&t.timer, timespec64_to_ktime(*rqtp), slack); - if (do_nanosleep(&t, mode)) + ret = do_nanosleep(&t, mode); + if (ret != -ERESTART_RESTARTBLOCK) goto out; /* Absolute timers do not update the rmtp value and restart: */ @@ -1527,18 +1513,10 @@ long hrtimer_nanosleep(struct timespec64 *rqtp, goto out; } - rmtp = restart->nanosleep.rmtp; - if (rmtp) { - ret = update_rmtp(&t.timer, rmtp); - if (ret <= 0) - goto out; - } - + restart = ¤t->restart_block; restart->fn = hrtimer_nanosleep_restart; restart->nanosleep.clockid = t.timer.base->clockid; restart->nanosleep.expires = hrtimer_get_expires_tv64(&t.timer); - - ret = -ERESTART_RESTARTBLOCK; out: destroy_hrtimer_on_stack(&t.timer); return ret; -- cgit v1.2.3 From 99e6c0e6ec349575886ca7daffc9cb7ec583176f Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 7 Jun 2017 09:42:30 +0100 Subject: posix-timers: Store rmtp into restart_block in sys_clock_nanosleep() ... instead of doing that in every ->nsleep() instance Signed-off-by: Al Viro Signed-off-by: Thomas Gleixner Cc: John Stultz Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20170607084241.28657-5-viro@ZenIV.linux.org.uk --- kernel/time/alarmtimer.c | 8 +------- kernel/time/posix-cpu-timers.c | 12 +++--------- kernel/time/posix-timers.c | 10 +++++----- kernel/time/posix-timers.h | 2 +- 4 files changed, 10 insertions(+), 22 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index ac6e9bc6cc59..d859a3601ddd 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -753,8 +753,7 @@ static long __sched alarm_timer_nsleep_restart(struct restart_block *restart) * Handles clock_nanosleep calls against _ALARM clockids */ static int alarm_timer_nsleep(const clockid_t which_clock, int flags, - struct timespec64 *tsreq, - struct timespec __user *rmtp) + struct timespec64 *tsreq) { enum alarmtimer_type type = clock2alarm(which_clock); struct restart_block *restart = ¤t->restart_block; @@ -762,11 +761,6 @@ static int alarm_timer_nsleep(const clockid_t which_clock, int flags, ktime_t exp; int ret = 0; - if (flags & TIMER_ABSTIME) - rmtp = NULL; - - restart->nanosleep.rmtp = rmtp; - if (!alarmtimer_get_rtcdev()) return -ENOTSUPP; diff --git a/kernel/time/posix-cpu-timers.c b/kernel/time/posix-cpu-timers.c index 239fff980fd0..ec6258c9cde5 100644 --- a/kernel/time/posix-cpu-timers.c +++ b/kernel/time/posix-cpu-timers.c @@ -1328,16 +1328,11 @@ static int do_cpu_nanosleep(const clockid_t which_clock, int flags, static long posix_cpu_nsleep_restart(struct restart_block *restart_block); static int posix_cpu_nsleep(const clockid_t which_clock, int flags, - struct timespec64 *rqtp, struct timespec __user *rmtp) + struct timespec64 *rqtp) { struct restart_block *restart_block = ¤t->restart_block; int error; - if (flags & TIMER_ABSTIME) - rmtp = NULL; - - restart_block->nanosleep.rmtp = rmtp; - /* * Diagnose required errors first. */ @@ -1388,10 +1383,9 @@ static int process_cpu_timer_create(struct k_itimer *timer) return posix_cpu_timer_create(timer); } static int process_cpu_nsleep(const clockid_t which_clock, int flags, - struct timespec64 *rqtp, - struct timespec __user *rmtp) + struct timespec64 *rqtp) { - return posix_cpu_nsleep(PROCESS_CLOCK, flags, rqtp, rmtp); + return posix_cpu_nsleep(PROCESS_CLOCK, flags, rqtp); } static long process_cpu_nsleep_restart(struct restart_block *restart_block) { diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c index 1a9f59f8afc2..a3e5c01b430e 100644 --- a/kernel/time/posix-timers.c +++ b/kernel/time/posix-timers.c @@ -1041,11 +1041,8 @@ SYSCALL_DEFINE2(clock_getres, const clockid_t, which_clock, * nanosleep for monotonic and realtime clocks */ static int common_nsleep(const clockid_t which_clock, int flags, - struct timespec64 *tsave, struct timespec __user *rmtp) + struct timespec64 *tsave) { - if (flags & TIMER_ABSTIME) - rmtp = NULL; - current->restart_block.nanosleep.rmtp = rmtp; return hrtimer_nanosleep(tsave, flags & TIMER_ABSTIME ? HRTIMER_MODE_ABS : HRTIMER_MODE_REL, which_clock); @@ -1070,8 +1067,11 @@ SYSCALL_DEFINE4(clock_nanosleep, const clockid_t, which_clock, int, flags, t64 = timespec_to_timespec64(t); if (!timespec64_valid(&t64)) return -EINVAL; + if (flags & TIMER_ABSTIME) + rmtp = NULL; + current->restart_block.nanosleep.rmtp = rmtp; - return kc->nsleep(which_clock, flags, &t64, rmtp); + return kc->nsleep(which_clock, flags, &t64); } /* diff --git a/kernel/time/posix-timers.h b/kernel/time/posix-timers.h index b086f5ba2f5b..bfd9e15c6ce0 100644 --- a/kernel/time/posix-timers.h +++ b/kernel/time/posix-timers.h @@ -10,7 +10,7 @@ struct k_clock { int (*clock_adj)(const clockid_t which_clock, struct timex *tx); int (*timer_create)(struct k_itimer *timer); int (*nsleep)(const clockid_t which_clock, int flags, - struct timespec64 *, struct timespec __user *); + struct timespec64 *); long (*nsleep_restart)(struct restart_block *restart_block); int (*timer_set)(struct k_itimer *timr, int flags, struct itimerspec64 *new_setting, -- cgit v1.2.3 From edbeda46322fbcb15af2d2d0f2daffb0cd349a5a Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 7 Jun 2017 09:42:31 +0100 Subject: time/posix-timers: Move the compat copyouts to the nanosleep implementations Turn restart_block.nanosleep.{rmtp,compat_rmtp} into a tagged union (kind = 1 -> native, kind = 2 -> compat, kind = 0 -> nothing) and make the places doing actual copyout handle compat as well as native (that will become a helper in the next commit). Result: compat wrappers, messing with reassignments, etc. are gone. [ tglx: Folded in a variant of Peter Zijlstras enum patch ] Signed-off-by: Al Viro Signed-off-by: Thomas Gleixner Cc: John Stultz Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20170607084241.28657-6-viro@ZenIV.linux.org.uk --- kernel/time/alarmtimer.c | 16 +++++++++--- kernel/time/hrtimer.c | 42 ++++++++++++++++++++++++++++---- kernel/time/posix-cpu-timers.c | 20 ++++++++++----- kernel/time/posix-stubs.c | 55 ++++++++++++++++++++++++++++++++---------- kernel/time/posix-timers.c | 32 ++++++++++++++++++------ 5 files changed, 129 insertions(+), 36 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index d859a3601ddd..57bcf94ee132 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -27,6 +27,7 @@ #include #include #include +#include #include "posix-timers.h" @@ -691,7 +692,7 @@ static enum alarmtimer_restart alarmtimer_nsleep_wakeup(struct alarm *alarm, static int alarmtimer_do_nsleep(struct alarm *alarm, ktime_t absexp, enum alarmtimer_type type) { - struct timespec __user *rmtp; + struct restart_block *restart; alarm->data = (void *)current; do { set_current_state(TASK_INTERRUPTIBLE); @@ -709,8 +710,8 @@ static int alarmtimer_do_nsleep(struct alarm *alarm, ktime_t absexp, if (freezing(current)) alarmtimer_freezerset(absexp, type); - rmtp = current->restart_block.nanosleep.rmtp; - if (rmtp) { + restart = ¤t->restart_block; + if (restart->nanosleep.type != TT_NONE) { struct timespec rmt; ktime_t rem; @@ -720,7 +721,14 @@ static int alarmtimer_do_nsleep(struct alarm *alarm, ktime_t absexp, return 0; rmt = ktime_to_timespec(rem); - if (copy_to_user(rmtp, &rmt, sizeof(*rmtp))) +#ifdef CONFIG_COMPAT + if (restart->nanosleep.type == TT_COMPAT) { + if (compat_put_timespec(&rmt, + restart->nanosleep.compat_rmtp)) + return -EFAULT; + } else +#endif + if (copy_to_user(restart->nanosleep.rmtp, &rmt, sizeof(rmt))) return -EFAULT; } return -ERESTART_RESTARTBLOCK; diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c index baa7b846b6e3..5370da8fc0a4 100644 --- a/kernel/time/hrtimer.c +++ b/kernel/time/hrtimer.c @@ -51,6 +51,7 @@ #include #include #include +#include #include @@ -1441,7 +1442,8 @@ EXPORT_SYMBOL_GPL(hrtimer_init_sleeper); static int __sched do_nanosleep(struct hrtimer_sleeper *t, enum hrtimer_mode mode) { - struct timespec __user *rmtp; + struct restart_block *restart; + hrtimer_init_sleeper(t, current); do { @@ -1461,15 +1463,23 @@ static int __sched do_nanosleep(struct hrtimer_sleeper *t, enum hrtimer_mode mod if (!t->task) return 0; - rmtp = current->restart_block.nanosleep.rmtp; - if (rmtp) { - struct timespec rmt; + restart = ¤t->restart_block; + if (restart->nanosleep.type != TT_NONE) { ktime_t rem = hrtimer_expires_remaining(&t->timer); + struct timespec rmt; + if (rem <= 0) return 0; rmt = ktime_to_timespec(rem); - if (copy_to_user(rmtp, &rmt, sizeof(*rmtp))) +#ifdef CONFIG_COMPAT + if (restart->nanosleep.type == TT_COMPAT) { + if (compat_put_timespec(&rmt, + restart->nanosleep.compat_rmtp)) + return -EFAULT; + } else +#endif + if (copy_to_user(restart->nanosleep.rmtp, &rmt, sizeof(rmt))) return -EFAULT; } return -ERESTART_RESTARTBLOCK; @@ -1535,10 +1545,32 @@ SYSCALL_DEFINE2(nanosleep, struct timespec __user *, rqtp, if (!timespec64_valid(&tu64)) return -EINVAL; + current->restart_block.nanosleep.type = rmtp ? TT_NATIVE : TT_NONE; current->restart_block.nanosleep.rmtp = rmtp; return hrtimer_nanosleep(&tu64, HRTIMER_MODE_REL, CLOCK_MONOTONIC); } +#ifdef CONFIG_COMPAT + +COMPAT_SYSCALL_DEFINE2(nanosleep, struct compat_timespec __user *, rqtp, + struct compat_timespec __user *, rmtp) +{ + struct timespec64 tu64; + struct timespec tu; + + if (compat_get_timespec(&tu, rqtp)) + return -EFAULT; + + tu64 = timespec_to_timespec64(tu); + if (!timespec64_valid(&tu64)) + return -EINVAL; + + current->restart_block.nanosleep.type = rmtp ? TT_COMPAT : TT_NONE; + current->restart_block.nanosleep.compat_rmtp = rmtp; + return hrtimer_nanosleep(&tu64, HRTIMER_MODE_REL, CLOCK_MONOTONIC); +} +#endif + /* * Functions related to boot-time initialization: */ diff --git a/kernel/time/posix-cpu-timers.c b/kernel/time/posix-cpu-timers.c index ec6258c9cde5..1563ca22cf1f 100644 --- a/kernel/time/posix-cpu-timers.c +++ b/kernel/time/posix-cpu-timers.c @@ -12,6 +12,7 @@ #include #include #include +#include #include "posix-timers.h" @@ -1243,10 +1244,9 @@ static int do_cpu_nanosleep(const clockid_t which_clock, int flags, timer.it_process = current; if (!error) { static struct itimerspec64 zero_it; - struct restart_block *restart = ¤t->restart_block; - struct timespec __user *rmtp; + struct restart_block *restart; - memset(&it, 0, sizeof it); + memset(&it, 0, sizeof(it)); it.it_value = *rqtp; spin_lock_irq(&timer.it_lock); @@ -1311,12 +1311,20 @@ static int do_cpu_nanosleep(const clockid_t which_clock, int flags, /* * Report back to the user the time still remaining. */ - rmtp = restart->nanosleep.rmtp; - if (rmtp) { + restart = ¤t->restart_block; + if (restart->nanosleep.type != TT_NONE) { struct timespec ts; ts = timespec64_to_timespec(it.it_value); - if (copy_to_user(rmtp, &ts, sizeof(*rmtp))) +#ifdef CONFIG_COMPAT + if (restart->nanosleep.type == TT_COMPAT) { + if (compat_put_timespec(&ts, + restart->nanosleep.compat_rmtp)) + return -EFAULT; + } else +#endif + if (copy_to_user(restart->nanosleep.rmtp, &ts, + sizeof(ts))) return -EFAULT; } restart->nanosleep.expires = timespec64_to_ns(rqtp); diff --git a/kernel/time/posix-stubs.c b/kernel/time/posix-stubs.c index 156a5e6f3bd2..749b76f2d757 100644 --- a/kernel/time/posix-stubs.c +++ b/kernel/time/posix-stubs.c @@ -17,6 +17,7 @@ #include #include #include +#include asmlinkage long sys_ni_posix_timers(void) { @@ -110,25 +111,53 @@ SYSCALL_DEFINE4(clock_nanosleep, const clockid_t, which_clock, int, flags, case CLOCK_REALTIME: case CLOCK_MONOTONIC: case CLOCK_BOOTTIME: - if (copy_from_user(&t, rqtp, sizeof (struct timespec))) - return -EFAULT; - t64 = timespec_to_timespec64(t); - if (!timespec64_valid(&t64)) - return -EINVAL; - if (flags & TIMER_ABSTIME) - rmtp = NULL; - current->restart_block.nanosleep.rmtp = rmtp; - return hrtimer_nanosleep(&t64, flags & TIMER_ABSTIME ? - HRTIMER_MODE_ABS : HRTIMER_MODE_REL, - which_clock); + break; default: return -EINVAL; } + + if (copy_from_user(&t, rqtp, sizeof (struct timespec))) + return -EFAULT; + t64 = timespec_to_timespec64(t); + if (!timespec64_valid(&t64)) + return -EINVAL; + if (flags & TIMER_ABSTIME) + rmtp = NULL; + current->restart_block.nanosleep.type = rmtp ? TT_NATIVE : TT_NONE; + current->restart_block.nanosleep.rmtp = rmtp; + return hrtimer_nanosleep(&t64, flags & TIMER_ABSTIME ? + HRTIMER_MODE_ABS : HRTIMER_MODE_REL, + which_clock); } #ifdef CONFIG_COMPAT -long clock_nanosleep_restart(struct restart_block *restart_block) +COMPAT_SYSCALL_DEFINE4(clock_nanosleep, clockid_t, which_clock, int, flags, + struct compat_timespec __user *, rqtp, + struct compat_timespec __user *, rmtp) { - return hrtimer_nanosleep_restart(restart_block); + struct timespec64 t64; + struct timespec t; + + switch (which_clock) { + case CLOCK_REALTIME: + case CLOCK_MONOTONIC: + case CLOCK_BOOTTIME: + break; + default: + return -EINVAL; + } + + if (compat_get_timespec(&t, rqtp)) + return -EFAULT; + t64 = timespec_to_timespec64(t); + if (!timespec64_valid(&t64)) + return -EINVAL; + if (flags & TIMER_ABSTIME) + rmtp = NULL; + current->restart_block.nanosleep.type = rmtp ? TT_COMPAT : TT_NONE; + current->restart_block.nanosleep.compat_rmtp = rmtp; + return hrtimer_nanosleep(&t64, flags & TIMER_ABSTIME ? + HRTIMER_MODE_ABS : HRTIMER_MODE_REL, + which_clock); } #endif diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c index a3e5c01b430e..bec86b6b9814 100644 --- a/kernel/time/posix-timers.c +++ b/kernel/time/posix-timers.c @@ -49,6 +49,7 @@ #include #include #include +#include #include "timekeeping.h" #include "posix-timers.h" @@ -1069,25 +1070,40 @@ SYSCALL_DEFINE4(clock_nanosleep, const clockid_t, which_clock, int, flags, return -EINVAL; if (flags & TIMER_ABSTIME) rmtp = NULL; + current->restart_block.nanosleep.type = rmtp ? TT_NATIVE : TT_NONE; current->restart_block.nanosleep.rmtp = rmtp; return kc->nsleep(which_clock, flags, &t64); } -/* - * This will restart clock_nanosleep. This is required only by - * compat_clock_nanosleep_restart for now. - */ -long clock_nanosleep_restart(struct restart_block *restart_block) +#ifdef CONFIG_COMPAT +COMPAT_SYSCALL_DEFINE4(clock_nanosleep, clockid_t, which_clock, int, flags, + struct compat_timespec __user *, rqtp, + struct compat_timespec __user *, rmtp) { - clockid_t which_clock = restart_block->nanosleep.clockid; const struct k_clock *kc = clockid_to_kclock(which_clock); + struct timespec64 t64; + struct timespec t; - if (WARN_ON_ONCE(!kc || !kc->nsleep_restart)) + if (!kc) return -EINVAL; + if (!kc->nsleep) + return -ENANOSLEEP_NOTSUP; + + if (compat_get_timespec(&t, rqtp)) + return -EFAULT; - return kc->nsleep_restart(restart_block); + t64 = timespec_to_timespec64(t); + if (!timespec64_valid(&t64)) + return -EINVAL; + if (flags & TIMER_ABSTIME) + rmtp = NULL; + current->restart_block.nanosleep.type = rmtp ? TT_COMPAT : TT_NONE; + current->restart_block.nanosleep.compat_rmtp = rmtp; + + return kc->nsleep(which_clock, flags, &t64); } +#endif static const struct k_clock clock_realtime = { .clock_getres = posix_get_hrtimer_res, -- cgit v1.2.3 From ce41aaf47af3d28c4c958e07675a3e0a51f09bd3 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 7 Jun 2017 09:42:32 +0100 Subject: hrtimers/posix-timers: Merge nanosleep timespec copyout logics into a new helper Signed-off-by: Al Viro Signed-off-by: Thomas Gleixner Cc: John Stultz Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20170607084241.28657-7-viro@ZenIV.linux.org.uk --- kernel/time/alarmtimer.c | 10 +--------- kernel/time/hrtimer.c | 29 ++++++++++++++++++++--------- kernel/time/posix-cpu-timers.c | 13 ++----------- 3 files changed, 23 insertions(+), 29 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index 57bcf94ee132..7bed4e44f9bd 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -721,15 +721,7 @@ static int alarmtimer_do_nsleep(struct alarm *alarm, ktime_t absexp, return 0; rmt = ktime_to_timespec(rem); -#ifdef CONFIG_COMPAT - if (restart->nanosleep.type == TT_COMPAT) { - if (compat_put_timespec(&rmt, - restart->nanosleep.compat_rmtp)) - return -EFAULT; - } else -#endif - if (copy_to_user(restart->nanosleep.rmtp, &rmt, sizeof(rmt))) - return -EFAULT; + return nanosleep_copyout(restart, &rmt); } return -ERESTART_RESTARTBLOCK; } diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c index 5370da8fc0a4..db2f5f7b4ba5 100644 --- a/kernel/time/hrtimer.c +++ b/kernel/time/hrtimer.c @@ -1440,6 +1440,25 @@ void hrtimer_init_sleeper(struct hrtimer_sleeper *sl, struct task_struct *task) } EXPORT_SYMBOL_GPL(hrtimer_init_sleeper); +int nanosleep_copyout(struct restart_block *restart, struct timespec *ts) +{ + switch(restart->nanosleep.type) { +#ifdef CONFIG_COMPAT + case TT_COMPAT: + if (compat_put_timespec(ts, restart->nanosleep.compat_rmtp)) + return -EFAULT; + break; +#endif + case TT_NATIVE: + if (copy_to_user(restart->nanosleep.rmtp, ts, sizeof(struct timespec))) + return -EFAULT; + break; + default: + BUG(); + } + return -ERESTART_RESTARTBLOCK; +} + static int __sched do_nanosleep(struct hrtimer_sleeper *t, enum hrtimer_mode mode) { struct restart_block *restart; @@ -1472,15 +1491,7 @@ static int __sched do_nanosleep(struct hrtimer_sleeper *t, enum hrtimer_mode mod return 0; rmt = ktime_to_timespec(rem); -#ifdef CONFIG_COMPAT - if (restart->nanosleep.type == TT_COMPAT) { - if (compat_put_timespec(&rmt, - restart->nanosleep.compat_rmtp)) - return -EFAULT; - } else -#endif - if (copy_to_user(restart->nanosleep.rmtp, &rmt, sizeof(rmt))) - return -EFAULT; + return nanosleep_copyout(restart, &rmt); } return -ERESTART_RESTARTBLOCK; } diff --git a/kernel/time/posix-cpu-timers.c b/kernel/time/posix-cpu-timers.c index 1563ca22cf1f..993a924d1399 100644 --- a/kernel/time/posix-cpu-timers.c +++ b/kernel/time/posix-cpu-timers.c @@ -1312,22 +1312,13 @@ static int do_cpu_nanosleep(const clockid_t which_clock, int flags, * Report back to the user the time still remaining. */ restart = ¤t->restart_block; + restart->nanosleep.expires = timespec64_to_ns(rqtp); if (restart->nanosleep.type != TT_NONE) { struct timespec ts; ts = timespec64_to_timespec(it.it_value); -#ifdef CONFIG_COMPAT - if (restart->nanosleep.type == TT_COMPAT) { - if (compat_put_timespec(&ts, - restart->nanosleep.compat_rmtp)) - return -EFAULT; - } else -#endif - if (copy_to_user(restart->nanosleep.rmtp, &ts, - sizeof(ts))) - return -EFAULT; + error = nanosleep_copyout(restart, &ts); } - restart->nanosleep.expires = timespec64_to_ns(rqtp); } return error; -- cgit v1.2.3 From fb923c4a3c2ee735755d4a93522150fc35d0ecbd Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 7 Jun 2017 09:42:33 +0100 Subject: posix-timers: Kill ->nsleep_restart() No more users. Signed-off-by: Al Viro Signed-off-by: Thomas Gleixner Cc: John Stultz Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20170607084241.28657-8-viro@ZenIV.linux.org.uk --- kernel/time/hrtimer.c | 2 +- kernel/time/posix-cpu-timers.c | 6 ------ kernel/time/posix-timers.c | 4 ---- kernel/time/posix-timers.h | 1 - 4 files changed, 1 insertion(+), 12 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c index db2f5f7b4ba5..45f83cc7c0c7 100644 --- a/kernel/time/hrtimer.c +++ b/kernel/time/hrtimer.c @@ -1496,7 +1496,7 @@ static int __sched do_nanosleep(struct hrtimer_sleeper *t, enum hrtimer_mode mod return -ERESTART_RESTARTBLOCK; } -long __sched hrtimer_nanosleep_restart(struct restart_block *restart) +static long __sched hrtimer_nanosleep_restart(struct restart_block *restart) { struct hrtimer_sleeper t; int ret; diff --git a/kernel/time/posix-cpu-timers.c b/kernel/time/posix-cpu-timers.c index 993a924d1399..515148d4eeb1 100644 --- a/kernel/time/posix-cpu-timers.c +++ b/kernel/time/posix-cpu-timers.c @@ -1386,10 +1386,6 @@ static int process_cpu_nsleep(const clockid_t which_clock, int flags, { return posix_cpu_nsleep(PROCESS_CLOCK, flags, rqtp); } -static long process_cpu_nsleep_restart(struct restart_block *restart_block) -{ - return -EINVAL; -} static int thread_cpu_clock_getres(const clockid_t which_clock, struct timespec64 *tp) { @@ -1412,7 +1408,6 @@ const struct k_clock clock_posix_cpu = { .clock_get = posix_cpu_clock_get, .timer_create = posix_cpu_timer_create, .nsleep = posix_cpu_nsleep, - .nsleep_restart = posix_cpu_nsleep_restart, .timer_set = posix_cpu_timer_set, .timer_del = posix_cpu_timer_del, .timer_get = posix_cpu_timer_get, @@ -1424,7 +1419,6 @@ const struct k_clock clock_process = { .clock_get = process_cpu_clock_get, .timer_create = process_cpu_timer_create, .nsleep = process_cpu_nsleep, - .nsleep_restart = process_cpu_nsleep_restart, }; const struct k_clock clock_thread = { diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c index bec86b6b9814..ea4a463436bf 100644 --- a/kernel/time/posix-timers.c +++ b/kernel/time/posix-timers.c @@ -1111,7 +1111,6 @@ static const struct k_clock clock_realtime = { .clock_set = posix_clock_realtime_set, .clock_adj = posix_clock_realtime_adj, .nsleep = common_nsleep, - .nsleep_restart = hrtimer_nanosleep_restart, .timer_create = common_timer_create, .timer_set = common_timer_set, .timer_get = common_timer_get, @@ -1127,7 +1126,6 @@ static const struct k_clock clock_monotonic = { .clock_getres = posix_get_hrtimer_res, .clock_get = posix_ktime_get_ts, .nsleep = common_nsleep, - .nsleep_restart = hrtimer_nanosleep_restart, .timer_create = common_timer_create, .timer_set = common_timer_set, .timer_get = common_timer_get, @@ -1158,7 +1156,6 @@ static const struct k_clock clock_tai = { .clock_getres = posix_get_hrtimer_res, .clock_get = posix_get_tai, .nsleep = common_nsleep, - .nsleep_restart = hrtimer_nanosleep_restart, .timer_create = common_timer_create, .timer_set = common_timer_set, .timer_get = common_timer_get, @@ -1174,7 +1171,6 @@ static const struct k_clock clock_boottime = { .clock_getres = posix_get_hrtimer_res, .clock_get = posix_get_boottime, .nsleep = common_nsleep, - .nsleep_restart = hrtimer_nanosleep_restart, .timer_create = common_timer_create, .timer_set = common_timer_set, .timer_get = common_timer_get, diff --git a/kernel/time/posix-timers.h b/kernel/time/posix-timers.h index bfd9e15c6ce0..5e69bb85629f 100644 --- a/kernel/time/posix-timers.h +++ b/kernel/time/posix-timers.h @@ -11,7 +11,6 @@ struct k_clock { int (*timer_create)(struct k_itimer *timer); int (*nsleep)(const clockid_t which_clock, int flags, struct timespec64 *); - long (*nsleep_restart)(struct restart_block *restart_block); int (*timer_set)(struct k_itimer *timr, int flags, struct itimerspec64 *new_setting, struct itimerspec64 *old_setting); -- cgit v1.2.3 From 3a4d44b6162555070194e486ff6b3799a8d323a2 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 7 Jun 2017 09:42:34 +0100 Subject: ntp: Move adjtimex related compat syscalls to native counterparts Get rid of set_fs() mess and sanitize compat_{get,put}_timex(), while we are at it. Signed-off-by: Al Viro Signed-off-by: Thomas Gleixner Cc: John Stultz Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20170607084241.28657-9-viro@ZenIV.linux.org.uk --- kernel/time/posix-stubs.c | 2 ++ kernel/time/posix-timers.c | 27 +++++++++++++++++++++++++++ kernel/time/time.c | 24 +++++++++++++++++++++++- 3 files changed, 52 insertions(+), 1 deletion(-) (limited to 'kernel/time') diff --git a/kernel/time/posix-stubs.c b/kernel/time/posix-stubs.c index 749b76f2d757..954d1d8ff9e6 100644 --- a/kernel/time/posix-stubs.c +++ b/kernel/time/posix-stubs.c @@ -28,6 +28,7 @@ asmlinkage long sys_ni_posix_timers(void) } #define SYS_NI(name) SYSCALL_ALIAS(sys_##name, sys_ni_posix_timers) +#define COMPAT_SYS_NI(name) SYSCALL_ALIAS(compat_sys_##name, sys_ni_posix_timers) SYS_NI(timer_create); SYS_NI(timer_gettime); @@ -40,6 +41,7 @@ SYS_NI(setitimer); #ifdef __ARCH_WANT_SYS_ALARM SYS_NI(alarm); #endif +COMPAT_SYS_NI(clock_adjtime); /* * We preserve minimal support for CLOCK_REALTIME and CLOCK_MONOTONIC diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c index ea4a463436bf..b1b6d52d6425 100644 --- a/kernel/time/posix-timers.c +++ b/kernel/time/posix-timers.c @@ -1018,6 +1018,33 @@ SYSCALL_DEFINE2(clock_adjtime, const clockid_t, which_clock, return err; } +#ifdef CONFIG_COMPAT + +COMPAT_SYSCALL_DEFINE2(clock_adjtime, clockid_t, which_clock, + struct compat_timex __user *, utp) +{ + const struct k_clock *kc = clockid_to_kclock(which_clock); + struct timex ktx; + int err; + + if (!kc) + return -EINVAL; + if (!kc->clock_adj) + return -EOPNOTSUPP; + + err = compat_get_timex(&ktx, utp); + if (err) + return err; + + err = kc->clock_adj(which_clock, &ktx); + + if (err >= 0) + err = compat_put_timex(utp, &ktx); + + return err; +} +#endif + SYSCALL_DEFINE2(clock_getres, const clockid_t, which_clock, struct timespec __user *, tp) { diff --git a/kernel/time/time.c b/kernel/time/time.c index 49c73c6ed648..400662f16c5a 100644 --- a/kernel/time/time.c +++ b/kernel/time/time.c @@ -39,6 +39,7 @@ #include #include +#include #include #include @@ -224,12 +225,33 @@ SYSCALL_DEFINE1(adjtimex, struct timex __user *, txc_p) * structure. But bear in mind that the structures * may change */ - if(copy_from_user(&txc, txc_p, sizeof(struct timex))) + if (copy_from_user(&txc, txc_p, sizeof(struct timex))) return -EFAULT; ret = do_adjtimex(&txc); return copy_to_user(txc_p, &txc, sizeof(struct timex)) ? -EFAULT : ret; } +#ifdef CONFIG_COMPAT + +COMPAT_SYSCALL_DEFINE1(adjtimex, struct compat_timex __user *, utp) +{ + struct timex txc; + int err, ret; + + err = compat_get_timex(&txc, utp); + if (err) + return err; + + ret = do_adjtimex(&txc); + + err = compat_put_timex(utp, &txc); + if (err) + return err; + + return ret; +} +#endif + /* * Convert jiffies to milliseconds and back. * -- cgit v1.2.3 From 1acbe7708b0313b33287bb4ffcbf26462ea3c588 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 7 Jun 2017 09:42:35 +0100 Subject: posix-timers: Take compat timer_settime(2) to native one ... and get rid of set_fs() in there Signed-off-by: Al Viro Signed-off-by: Thomas Gleixner Cc: John Stultz Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20170607084241.28657-10-viro@ZenIV.linux.org.uk --- kernel/time/posix-stubs.c | 1 + kernel/time/posix-timers.c | 87 +++++++++++++++++++++++++++++++++------------- 2 files changed, 64 insertions(+), 24 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/posix-stubs.c b/kernel/time/posix-stubs.c index 954d1d8ff9e6..ad263df132d6 100644 --- a/kernel/time/posix-stubs.c +++ b/kernel/time/posix-stubs.c @@ -42,6 +42,7 @@ SYS_NI(setitimer); SYS_NI(alarm); #endif COMPAT_SYS_NI(clock_adjtime); +COMPAT_SYS_NI(timer_settime); /* * We preserve minimal support for CLOCK_REALTIME and CLOCK_MONOTONIC diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c index b1b6d52d6425..a73feac191f9 100644 --- a/kernel/time/posix-timers.c +++ b/kernel/time/posix-timers.c @@ -819,31 +819,21 @@ int common_timer_set(struct k_itimer *timr, int flags, return 0; } -/* Set a POSIX.1b interval timer */ -SYSCALL_DEFINE4(timer_settime, timer_t, timer_id, int, flags, - const struct itimerspec __user *, new_setting, - struct itimerspec __user *, old_setting) +static int do_timer_settime(timer_t timer_id, int flags, + struct itimerspec64 *new_spec64, + struct itimerspec64 *old_spec64) { - struct itimerspec64 new_spec64, old_spec64; - struct itimerspec64 *rtn = old_setting ? &old_spec64 : NULL; - struct itimerspec new_spec, old_spec; + const struct k_clock *kc; struct k_itimer *timr; unsigned long flag; - const struct k_clock *kc; int error = 0; - if (!new_setting) + if (!timespec64_valid(&new_spec64->it_interval) || + !timespec64_valid(&new_spec64->it_value)) return -EINVAL; - if (copy_from_user(&new_spec, new_setting, sizeof (new_spec))) - return -EFAULT; - new_spec64 = itimerspec_to_itimerspec64(&new_spec); - - if (!timespec64_valid(&new_spec64.it_interval) || - !timespec64_valid(&new_spec64.it_value)) - return -EINVAL; - if (rtn) - memset(rtn, 0, sizeof(*rtn)); + if (old_spec64) + memset(old_spec64, 0, sizeof(*old_spec64)); retry: timr = lock_timer(timer_id, &flag); if (!timr) @@ -853,22 +843,71 @@ retry: if (WARN_ON_ONCE(!kc || !kc->timer_set)) error = -EINVAL; else - error = kc->timer_set(timr, flags, &new_spec64, rtn); + error = kc->timer_set(timr, flags, new_spec64, old_spec64); unlock_timer(timr, flag); if (error == TIMER_RETRY) { - rtn = NULL; // We already got the old time... + old_spec64 = NULL; // We already got the old time... goto retry; } - old_spec = itimerspec64_to_itimerspec(&old_spec64); - if (old_setting && !error && - copy_to_user(old_setting, &old_spec, sizeof (old_spec))) - error = -EFAULT; + return error; +} + +/* Set a POSIX.1b interval timer */ +SYSCALL_DEFINE4(timer_settime, timer_t, timer_id, int, flags, + const struct itimerspec __user *, new_setting, + struct itimerspec __user *, old_setting) +{ + struct itimerspec64 new_spec64, old_spec64; + struct itimerspec64 *rtn = old_setting ? &old_spec64 : NULL; + struct itimerspec new_spec; + int error = 0; + + if (!new_setting) + return -EINVAL; + if (copy_from_user(&new_spec, new_setting, sizeof (new_spec))) + return -EFAULT; + new_spec64 = itimerspec_to_itimerspec64(&new_spec); + + error = do_timer_settime(timer_id, flags, &new_spec64, rtn); + if (!error && old_setting) { + struct itimerspec old_spec; + old_spec = itimerspec64_to_itimerspec(&old_spec64); + if (copy_to_user(old_setting, &old_spec, sizeof (old_spec))) + error = -EFAULT; + } return error; } +#ifdef CONFIG_COMPAT +COMPAT_SYSCALL_DEFINE4(timer_settime, timer_t, timer_id, int, flags, + struct compat_itimerspec __user *, new, + struct compat_itimerspec __user *, old) +{ + struct itimerspec64 new_spec64, old_spec64; + struct itimerspec64 *rtn = old ? &old_spec64 : NULL; + struct itimerspec new_spec; + int error = 0; + + if (!new) + return -EINVAL; + if (get_compat_itimerspec(&new_spec, new)) + return -EFAULT; + + new_spec64 = itimerspec_to_itimerspec64(&new_spec); + error = do_timer_settime(timer_id, flags, &new_spec64, rtn); + if (!error && old) { + struct itimerspec old_spec; + old_spec = itimerspec64_to_itimerspec(&old_spec64); + if (put_compat_itimerspec(old, &old_spec)) + error = -EFAULT; + } + return error; +} +#endif + int common_timer_del(struct k_itimer *timer) { const struct k_clock *kc = timer->kclock; -- cgit v1.2.3 From b0dc12426ec404de99d7e75a12a22d9201d90914 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 7 Jun 2017 09:42:36 +0100 Subject: posix-timers: Take compat timer_gettime(2) to native one ... and get rid of set_fs() in there Signed-off-by: Al Viro Signed-off-by: Thomas Gleixner Cc: John Stultz Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20170607084241.28657-11-viro@ZenIV.linux.org.uk --- kernel/time/posix-stubs.c | 1 + kernel/time/posix-timers.c | 43 ++++++++++++++++++++++++++++++++++--------- 2 files changed, 35 insertions(+), 9 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/posix-stubs.c b/kernel/time/posix-stubs.c index ad263df132d6..f4a1962d1729 100644 --- a/kernel/time/posix-stubs.c +++ b/kernel/time/posix-stubs.c @@ -43,6 +43,7 @@ SYS_NI(alarm); #endif COMPAT_SYS_NI(clock_adjtime); COMPAT_SYS_NI(timer_settime); +COMPAT_SYS_NI(timer_gettime); /* * We preserve minimal support for CLOCK_REALTIME and CLOCK_MONOTONIC diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c index a73feac191f9..e82bb1fd614e 100644 --- a/kernel/time/posix-timers.c +++ b/kernel/time/posix-timers.c @@ -690,11 +690,8 @@ void common_timer_get(struct k_itimer *timr, struct itimerspec64 *cur_setting) } /* Get the time remaining on a POSIX.1b interval timer. */ -SYSCALL_DEFINE2(timer_gettime, timer_t, timer_id, - struct itimerspec __user *, setting) +static int do_timer_gettime(timer_t timer_id, struct itimerspec64 *setting) { - struct itimerspec64 cur_setting64; - struct itimerspec cur_setting; struct k_itimer *timr; const struct k_clock *kc; unsigned long flags; @@ -704,21 +701,49 @@ SYSCALL_DEFINE2(timer_gettime, timer_t, timer_id, if (!timr) return -EINVAL; - memset(&cur_setting64, 0, sizeof(cur_setting64)); + memset(setting, 0, sizeof(*setting)); kc = timr->kclock; if (WARN_ON_ONCE(!kc || !kc->timer_get)) ret = -EINVAL; else - kc->timer_get(timr, &cur_setting64); + kc->timer_get(timr, setting); unlock_timer(timr, flags); + return ret; +} - cur_setting = itimerspec64_to_itimerspec(&cur_setting64); - if (!ret && copy_to_user(setting, &cur_setting, sizeof (cur_setting))) - return -EFAULT; +/* Get the time remaining on a POSIX.1b interval timer. */ +SYSCALL_DEFINE2(timer_gettime, timer_t, timer_id, + struct itimerspec __user *, setting) +{ + struct itimerspec64 cur_setting64; + int ret = do_timer_gettime(timer_id, &cur_setting64); + if (!ret) { + struct itimerspec cur_setting; + cur_setting = itimerspec64_to_itimerspec(&cur_setting64); + if (copy_to_user(setting, &cur_setting, sizeof (cur_setting))) + ret = -EFAULT; + } + return ret; +} + +#ifdef CONFIG_COMPAT +COMPAT_SYSCALL_DEFINE2(timer_gettime, timer_t, timer_id, + struct compat_itimerspec __user *, setting) +{ + struct itimerspec64 cur_setting64; + + int ret = do_timer_gettime(timer_id, &cur_setting64); + if (!ret) { + struct itimerspec cur_setting; + cur_setting = itimerspec64_to_itimerspec(&cur_setting64); + if (put_compat_itimerspec(setting, &cur_setting)) + ret = -EFAULT; + } return ret; } +#endif /* * Get the number of overruns of a POSIX.1b interval timer. This is to -- cgit v1.2.3 From 54ad9c46c262ce4a603dc7887e37956896a0211d Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 7 Jun 2017 09:42:37 +0100 Subject: itimers: Move compat itimer syscalls to native ones get rid of set_fs(), sanitize compat copyin/copyout. Signed-off-by: Al Viro Signed-off-by: Thomas Gleixner Cc: John Stultz Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20170607084241.28657-12-viro@ZenIV.linux.org.uk --- kernel/time/itimer.c | 38 ++++++++++++++++++++++++++++++++++++++ kernel/time/posix-stubs.c | 2 ++ 2 files changed, 40 insertions(+) (limited to 'kernel/time') diff --git a/kernel/time/itimer.c b/kernel/time/itimer.c index 087d6a1279b8..9dd7ff5e445a 100644 --- a/kernel/time/itimer.c +++ b/kernel/time/itimer.c @@ -15,6 +15,7 @@ #include #include #include +#include #include @@ -116,6 +117,19 @@ SYSCALL_DEFINE2(getitimer, int, which, struct itimerval __user *, value) return error; } +#ifdef CONFIG_COMPAT +COMPAT_SYSCALL_DEFINE2(getitimer, int, which, + struct compat_itimerval __user *, it) +{ + struct itimerval kit; + int error = do_getitimer(which, &kit); + + if (!error && put_compat_itimerval(it, &kit)) + error = -EFAULT; + return error; +} +#endif + /* * The timer is automagically restarted, when interval != 0 @@ -294,3 +308,27 @@ SYSCALL_DEFINE3(setitimer, int, which, struct itimerval __user *, value, return -EFAULT; return 0; } + +#ifdef CONFIG_COMPAT +COMPAT_SYSCALL_DEFINE3(setitimer, int, which, + struct compat_itimerval __user *, in, + struct compat_itimerval __user *, out) +{ + struct itimerval kin, kout; + int error; + + if (in) { + if (get_compat_itimerval(&kin, in)) + return -EFAULT; + } else { + memset(&kin, 0, sizeof(kin)); + } + + error = do_setitimer(which, &kin, out ? &kout : NULL); + if (error || !out) + return error; + if (put_compat_itimerval(out, &kout)) + return -EFAULT; + return 0; +} +#endif diff --git a/kernel/time/posix-stubs.c b/kernel/time/posix-stubs.c index f4a1962d1729..7f88517461e8 100644 --- a/kernel/time/posix-stubs.c +++ b/kernel/time/posix-stubs.c @@ -44,6 +44,8 @@ SYS_NI(alarm); COMPAT_SYS_NI(clock_adjtime); COMPAT_SYS_NI(timer_settime); COMPAT_SYS_NI(timer_gettime); +COMPAT_SYS_NI(getitimer); +COMPAT_SYS_NI(setitimer); /* * We preserve minimal support for CLOCK_REALTIME and CLOCK_MONOTONIC -- cgit v1.2.3 From d822cdcce43f9d4dcddbf9c68f9537d542ccc3c3 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 7 Jun 2017 09:42:38 +0100 Subject: posix-timers: Move compat versions of clock_gettime/settime/getres Move them to the native implementations and get rid of the set_fs() hackery. Signed-off-by: Al Viro Signed-off-by: Thomas Gleixner Cc: John Stultz Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20170607084241.28657-13-viro@ZenIV.linux.org.uk --- kernel/time/posix-stubs.c | 53 +++++++++++++++++++++++++++++++++++++ kernel/time/posix-timers.c | 66 +++++++++++++++++++++++++++++++++++++++++++--- 2 files changed, 115 insertions(+), 4 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/posix-stubs.c b/kernel/time/posix-stubs.c index 7f88517461e8..a375c31cb352 100644 --- a/kernel/time/posix-stubs.c +++ b/kernel/time/posix-stubs.c @@ -137,6 +137,59 @@ SYSCALL_DEFINE4(clock_nanosleep, const clockid_t, which_clock, int, flags, } #ifdef CONFIG_COMPAT +COMPAT_SYSCALL_DEFINE2(clock_settime, const clockid_t, which_clock, + struct compat_timespec __user *, tp) +{ + struct timespec64 new_tp64; + struct timespec new_tp; + + if (which_clock != CLOCK_REALTIME) + return -EINVAL; + if (compat_get_timespec(&new_tp, tp)) + return -EFAULT; + + new_tp64 = timespec_to_timespec64(new_tp); + return do_sys_settimeofday64(&new_tp64, NULL); +} + +COMPAT_SYSCALL_DEFINE2(clock_gettime, const clockid_t, which_clock, + struct compat_timespec __user *,tp) +{ + struct timespec64 kernel_tp64; + struct timespec kernel_tp; + + switch (which_clock) { + case CLOCK_REALTIME: ktime_get_real_ts64(&kernel_tp64); break; + case CLOCK_MONOTONIC: ktime_get_ts64(&kernel_tp64); break; + case CLOCK_BOOTTIME: get_monotonic_boottime64(&kernel_tp64); break; + default: return -EINVAL; + } + + kernel_tp = timespec64_to_timespec(kernel_tp64); + if (compat_put_timespec(&kernel_tp, tp)) + return -EFAULT; + return 0; +} + +COMPAT_SYSCALL_DEFINE2(clock_getres, const clockid_t, which_clock, + struct compat_timespec __user *, tp) +{ + struct timespec rtn_tp = { + .tv_sec = 0, + .tv_nsec = hrtimer_resolution, + }; + + switch (which_clock) { + case CLOCK_REALTIME: + case CLOCK_MONOTONIC: + case CLOCK_BOOTTIME: + if (compat_put_timespec(&rtn_tp, tp)) + return -EFAULT; + return 0; + default: + return -EINVAL; + } +} COMPAT_SYSCALL_DEFINE4(clock_nanosleep, clockid_t, which_clock, int, flags, struct compat_timespec __user *, rqtp, struct compat_timespec __user *, rmtp) diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c index e82bb1fd614e..61a5fb91a3c7 100644 --- a/kernel/time/posix-timers.c +++ b/kernel/time/posix-timers.c @@ -1082,8 +1082,66 @@ SYSCALL_DEFINE2(clock_adjtime, const clockid_t, which_clock, return err; } +SYSCALL_DEFINE2(clock_getres, const clockid_t, which_clock, + struct timespec __user *, tp) +{ + const struct k_clock *kc = clockid_to_kclock(which_clock); + struct timespec64 rtn_tp64; + struct timespec rtn_tp; + int error; + + if (!kc) + return -EINVAL; + + error = kc->clock_getres(which_clock, &rtn_tp64); + rtn_tp = timespec64_to_timespec(rtn_tp64); + + if (!error && tp && copy_to_user(tp, &rtn_tp, sizeof (rtn_tp))) + error = -EFAULT; + + return error; +} + #ifdef CONFIG_COMPAT +COMPAT_SYSCALL_DEFINE2(clock_settime, clockid_t, which_clock, + struct compat_timespec __user *, tp) +{ + const struct k_clock *kc = clockid_to_kclock(which_clock); + struct timespec64 new_tp64; + struct timespec new_tp; + + if (!kc || !kc->clock_set) + return -EINVAL; + + if (compat_get_timespec(&new_tp, tp)) + return -EFAULT; + + new_tp64 = timespec_to_timespec64(new_tp); + + return kc->clock_set(which_clock, &new_tp64); +} + +COMPAT_SYSCALL_DEFINE2(clock_gettime, clockid_t, which_clock, + struct compat_timespec __user *, tp) +{ + const struct k_clock *kc = clockid_to_kclock(which_clock); + struct timespec64 kernel_tp64; + struct timespec kernel_tp; + int error; + + if (!kc) + return -EINVAL; + + error = kc->clock_get(which_clock, &kernel_tp64); + kernel_tp = timespec64_to_timespec(kernel_tp64); + + if (!error && compat_put_timespec(&kernel_tp, tp)) + error = -EFAULT; + + return error; +} + COMPAT_SYSCALL_DEFINE2(clock_adjtime, clockid_t, which_clock, struct compat_timex __user *, utp) { @@ -1107,10 +1165,9 @@ COMPAT_SYSCALL_DEFINE2(clock_adjtime, clockid_t, which_clock, return err; } -#endif -SYSCALL_DEFINE2(clock_getres, const clockid_t, which_clock, - struct timespec __user *, tp) +COMPAT_SYSCALL_DEFINE2(clock_getres, clockid_t, which_clock, + struct compat_timespec __user *, tp) { const struct k_clock *kc = clockid_to_kclock(which_clock); struct timespec64 rtn_tp64; @@ -1123,11 +1180,12 @@ SYSCALL_DEFINE2(clock_getres, const clockid_t, which_clock, error = kc->clock_getres(which_clock, &rtn_tp64); rtn_tp = timespec64_to_timespec(rtn_tp64); - if (!error && tp && copy_to_user(tp, &rtn_tp, sizeof (rtn_tp))) + if (!error && tp && compat_put_timespec(&rtn_tp, tp)) error = -EFAULT; return error; } +#endif /* * nanosleep for monotonic and realtime clocks -- cgit v1.2.3 From 2482097c6c0f01ad74c9b2cff120a519ac59846e Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 7 Jun 2017 09:42:39 +0100 Subject: posix-timers: Move compat_timer_create() to native, get rid of set_fs() Signed-off-by: Al Viro Signed-off-by: Thomas Gleixner Cc: John Stultz Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20170607084241.28657-14-viro@ZenIV.linux.org.uk --- kernel/time/posix-stubs.c | 1 + kernel/time/posix-timers.c | 59 ++++++++++++++++++++++++++++++++-------------- 2 files changed, 42 insertions(+), 18 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/posix-stubs.c b/kernel/time/posix-stubs.c index a375c31cb352..38f3b20efa29 100644 --- a/kernel/time/posix-stubs.c +++ b/kernel/time/posix-stubs.c @@ -41,6 +41,7 @@ SYS_NI(setitimer); #ifdef __ARCH_WANT_SYS_ALARM SYS_NI(alarm); #endif +COMPAT_SYS_NI(timer_create); COMPAT_SYS_NI(clock_adjtime); COMPAT_SYS_NI(timer_settime); COMPAT_SYS_NI(timer_gettime); diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c index 61a5fb91a3c7..c9f45a84fb8b 100644 --- a/kernel/time/posix-timers.c +++ b/kernel/time/posix-timers.c @@ -490,15 +490,12 @@ static int common_timer_create(struct k_itimer *new_timer) } /* Create a POSIX.1b interval timer. */ - -SYSCALL_DEFINE3(timer_create, const clockid_t, which_clock, - struct sigevent __user *, timer_event_spec, - timer_t __user *, created_timer_id) +static int do_timer_create(clockid_t which_clock, struct sigevent *event, + timer_t __user *created_timer_id) { const struct k_clock *kc = clockid_to_kclock(which_clock); struct k_itimer *new_timer; int error, new_timer_id; - sigevent_t event; int it_id_set = IT_ID_NOT_SET; if (!kc) @@ -523,29 +520,25 @@ SYSCALL_DEFINE3(timer_create, const clockid_t, which_clock, new_timer->kclock = kc; new_timer->it_overrun = -1; - if (timer_event_spec) { - if (copy_from_user(&event, timer_event_spec, sizeof (event))) { - error = -EFAULT; - goto out; - } + if (event) { rcu_read_lock(); - new_timer->it_pid = get_pid(good_sigevent(&event)); + new_timer->it_pid = get_pid(good_sigevent(event)); rcu_read_unlock(); if (!new_timer->it_pid) { error = -EINVAL; goto out; } + new_timer->it_sigev_notify = event->sigev_notify; + new_timer->sigq->info.si_signo = event->sigev_signo; + new_timer->sigq->info.si_value = event->sigev_value; } else { - memset(&event.sigev_value, 0, sizeof(event.sigev_value)); - event.sigev_notify = SIGEV_SIGNAL; - event.sigev_signo = SIGALRM; - event.sigev_value.sival_int = new_timer->it_id; + new_timer->it_sigev_notify = SIGEV_SIGNAL; + new_timer->sigq->info.si_signo = SIGALRM; + memset(&new_timer->sigq->info.si_value, 0, sizeof(sigval_t)); + new_timer->sigq->info.si_value.sival_int = new_timer->it_id; new_timer->it_pid = get_pid(task_tgid(current)); } - new_timer->it_sigev_notify = event.sigev_notify; - new_timer->sigq->info.si_signo = event.sigev_signo; - new_timer->sigq->info.si_value = event.sigev_value; new_timer->sigq->info.si_tid = new_timer->it_id; new_timer->sigq->info.si_code = SI_TIMER; @@ -576,6 +569,36 @@ out: return error; } +SYSCALL_DEFINE3(timer_create, const clockid_t, which_clock, + struct sigevent __user *, timer_event_spec, + timer_t __user *, created_timer_id) +{ + if (timer_event_spec) { + sigevent_t event; + + if (copy_from_user(&event, timer_event_spec, sizeof (event))) + return -EFAULT; + return do_timer_create(which_clock, &event, created_timer_id); + } + return do_timer_create(which_clock, NULL, created_timer_id); +} + +#ifdef CONFIG_COMPAT +COMPAT_SYSCALL_DEFINE3(timer_create, clockid_t, which_clock, + struct compat_sigevent __user *, timer_event_spec, + timer_t __user *, created_timer_id) +{ + if (timer_event_spec) { + sigevent_t event; + + if (get_compat_sigevent(&event, timer_event_spec)) + return -EFAULT; + return do_timer_create(which_clock, &event, created_timer_id); + } + return do_timer_create(which_clock, NULL, created_timer_id); +} +#endif + /* * Locking issues: We need to protect the result of the id look up until * we get the timer locked down so it is not deleted under us. The -- cgit v1.2.3 From b180db2c8ca6692a10b79631cadc18d03303d494 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 7 Jun 2017 09:42:40 +0100 Subject: time: Move compat_time()/stime() to native Signed-off-by: Al Viro Signed-off-by: Thomas Gleixner Cc: John Stultz Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20170607084241.28657-15-viro@ZenIV.linux.org.uk --- kernel/time/time.c | 41 +++++++++++++++++++++++++++++++++++++++++ 1 file changed, 41 insertions(+) (limited to 'kernel/time') diff --git a/kernel/time/time.c b/kernel/time/time.c index 400662f16c5a..e5d44999ff78 100644 --- a/kernel/time/time.c +++ b/kernel/time/time.c @@ -100,6 +100,47 @@ SYSCALL_DEFINE1(stime, time_t __user *, tptr) #endif /* __ARCH_WANT_SYS_TIME */ +#ifdef CONFIG_COMPAT +#ifdef __ARCH_WANT_COMPAT_SYS_TIME + +/* compat_time_t is a 32 bit "long" and needs to get converted. */ +COMPAT_SYSCALL_DEFINE1(time, compat_time_t __user *, tloc) +{ + struct timeval tv; + compat_time_t i; + + do_gettimeofday(&tv); + i = tv.tv_sec; + + if (tloc) { + if (put_user(i,tloc)) + return -EFAULT; + } + force_successful_syscall_return(); + return i; +} + +COMPAT_SYSCALL_DEFINE1(stime, compat_time_t __user *, tptr) +{ + struct timespec tv; + int err; + + if (get_user(tv.tv_sec, tptr)) + return -EFAULT; + + tv.tv_nsec = 0; + + err = security_settime(&tv, NULL); + if (err) + return err; + + do_settimeofday(&tv); + return 0; +} + +#endif /* __ARCH_WANT_COMPAT_SYS_TIME */ +#endif + SYSCALL_DEFINE2(gettimeofday, struct timeval __user *, tv, struct timezone __user *, tz) { -- cgit v1.2.3 From 2b2d02856b3176701c91d723356f766d6ee27853 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 7 Jun 2017 09:42:41 +0100 Subject: time: Move compat_gettimeofday()/settimeofday() to native Signed-off-by: Al Viro Signed-off-by: Thomas Gleixner Cc: John Stultz Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20170607084241.28657-16-viro@ZenIV.linux.org.uk --- kernel/time/time.c | 41 +++++++++++++++++++++++++++++++++++++++++ 1 file changed, 41 insertions(+) (limited to 'kernel/time') diff --git a/kernel/time/time.c b/kernel/time/time.c index e5d44999ff78..7c89e437c4d7 100644 --- a/kernel/time/time.c +++ b/kernel/time/time.c @@ -257,6 +257,47 @@ SYSCALL_DEFINE2(settimeofday, struct timeval __user *, tv, return do_sys_settimeofday64(tv ? &new_ts : NULL, tz ? &new_tz : NULL); } +#ifdef CONFIG_COMPAT +COMPAT_SYSCALL_DEFINE2(gettimeofday, struct compat_timeval __user *, tv, + struct timezone __user *, tz) +{ + if (tv) { + struct timeval ktv; + + do_gettimeofday(&ktv); + if (compat_put_timeval(&ktv, tv)) + return -EFAULT; + } + if (tz) { + if (copy_to_user(tz, &sys_tz, sizeof(sys_tz))) + return -EFAULT; + } + + return 0; +} + +COMPAT_SYSCALL_DEFINE2(settimeofday, struct compat_timeval __user *, tv, + struct timezone __user *, tz) +{ + struct timespec64 new_ts; + struct timeval user_tv; + struct timezone new_tz; + + if (tv) { + if (compat_get_timeval(&user_tv, tv)) + return -EFAULT; + new_ts.tv_sec = user_tv.tv_sec; + new_ts.tv_nsec = user_tv.tv_usec * NSEC_PER_USEC; + } + if (tz) { + if (copy_from_user(&new_tz, tz, sizeof(*tz))) + return -EFAULT; + } + + return do_sys_settimeofday64(tv ? &new_ts : NULL, tz ? &new_tz : NULL); +} +#endif + SYSCALL_DEFINE1(adjtimex, struct timex __user *, txc_p) { struct timex txc; /* Local copy of parameter */ -- cgit v1.2.3 From 343d8fc208e43e50525ae6e0fc4845b9966b7318 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 13 Jun 2017 23:29:14 +0200 Subject: posix-cpu-timers: Avoid timespec conversion in do_cpu_nanosleep() No point in converting the expiry time back and forth. No point either to update the value in the caller supplied variable. mark the rqtp argument const. Signed-off-by: Thomas Gleixner Cc: Al Viro Cc: John Stultz Cc: Peter Zijlstra --- kernel/time/posix-cpu-timers.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/posix-cpu-timers.c b/kernel/time/posix-cpu-timers.c index 515148d4eeb1..3adfa42ca24c 100644 --- a/kernel/time/posix-cpu-timers.c +++ b/kernel/time/posix-cpu-timers.c @@ -1227,10 +1227,11 @@ void set_process_cpu_timer(struct task_struct *tsk, unsigned int clock_idx, } static int do_cpu_nanosleep(const clockid_t which_clock, int flags, - struct timespec64 *rqtp) + const struct timespec64 *rqtp) { - struct k_itimer timer; struct itimerspec64 it; + struct k_itimer timer; + u64 expires; int error; /* @@ -1279,7 +1280,7 @@ static int do_cpu_nanosleep(const clockid_t which_clock, int flags, /* * We were interrupted by a signal. */ - *rqtp = ns_to_timespec64(timer.it.cpu.expires); + expires = timer.it.cpu.expires; error = posix_cpu_timer_set(&timer, 0, &zero_it, &it); if (!error) { /* @@ -1312,7 +1313,7 @@ static int do_cpu_nanosleep(const clockid_t which_clock, int flags, * Report back to the user the time still remaining. */ restart = ¤t->restart_block; - restart->nanosleep.expires = timespec64_to_ns(rqtp); + restart->nanosleep.expires = expires; if (restart->nanosleep.type != TT_NONE) { struct timespec ts; -- cgit v1.2.3 From 938e7cf2d569833a5acf689a8926faf507826253 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 13 Jun 2017 23:34:33 +0200 Subject: posix-timers: Make nanosleep timespec argument const No nanosleep implementation modifies the rqtp argument. Mark is const. Signed-off-by: Thomas Gleixner Cc: Al Viro Cc: John Stultz Cc: Peter Zijlstra --- kernel/time/alarmtimer.c | 2 +- kernel/time/hrtimer.c | 2 +- kernel/time/posix-cpu-timers.c | 4 ++-- kernel/time/posix-timers.c | 4 ++-- kernel/time/posix-timers.h | 2 +- 5 files changed, 7 insertions(+), 7 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index 7bed4e44f9bd..c991cf212c6d 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -753,7 +753,7 @@ static long __sched alarm_timer_nsleep_restart(struct restart_block *restart) * Handles clock_nanosleep calls against _ALARM clockids */ static int alarm_timer_nsleep(const clockid_t which_clock, int flags, - struct timespec64 *tsreq) + const struct timespec64 *tsreq) { enum alarmtimer_type type = clock2alarm(which_clock); struct restart_block *restart = ¤t->restart_block; diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c index 45f83cc7c0c7..81da124f1115 100644 --- a/kernel/time/hrtimer.c +++ b/kernel/time/hrtimer.c @@ -1510,7 +1510,7 @@ static long __sched hrtimer_nanosleep_restart(struct restart_block *restart) return ret; } -long hrtimer_nanosleep(struct timespec64 *rqtp, +long hrtimer_nanosleep(const struct timespec64 *rqtp, const enum hrtimer_mode mode, const clockid_t clockid) { struct restart_block *restart; diff --git a/kernel/time/posix-cpu-timers.c b/kernel/time/posix-cpu-timers.c index 3adfa42ca24c..9df618ee64cf 100644 --- a/kernel/time/posix-cpu-timers.c +++ b/kernel/time/posix-cpu-timers.c @@ -1328,7 +1328,7 @@ static int do_cpu_nanosleep(const clockid_t which_clock, int flags, static long posix_cpu_nsleep_restart(struct restart_block *restart_block); static int posix_cpu_nsleep(const clockid_t which_clock, int flags, - struct timespec64 *rqtp) + const struct timespec64 *rqtp) { struct restart_block *restart_block = ¤t->restart_block; int error; @@ -1383,7 +1383,7 @@ static int process_cpu_timer_create(struct k_itimer *timer) return posix_cpu_timer_create(timer); } static int process_cpu_nsleep(const clockid_t which_clock, int flags, - struct timespec64 *rqtp) + const struct timespec64 *rqtp) { return posix_cpu_nsleep(PROCESS_CLOCK, flags, rqtp); } diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c index c9f45a84fb8b..82d67be7d9d1 100644 --- a/kernel/time/posix-timers.c +++ b/kernel/time/posix-timers.c @@ -1214,9 +1214,9 @@ COMPAT_SYSCALL_DEFINE2(clock_getres, clockid_t, which_clock, * nanosleep for monotonic and realtime clocks */ static int common_nsleep(const clockid_t which_clock, int flags, - struct timespec64 *tsave) + const struct timespec64 *rqtp) { - return hrtimer_nanosleep(tsave, flags & TIMER_ABSTIME ? + return hrtimer_nanosleep(rqtp, flags & TIMER_ABSTIME ? HRTIMER_MODE_ABS : HRTIMER_MODE_REL, which_clock); } diff --git a/kernel/time/posix-timers.h b/kernel/time/posix-timers.h index 5e69bb85629f..fb303c3be4d3 100644 --- a/kernel/time/posix-timers.h +++ b/kernel/time/posix-timers.h @@ -10,7 +10,7 @@ struct k_clock { int (*clock_adj)(const clockid_t which_clock, struct timex *tx); int (*timer_create)(struct k_itimer *timer); int (*nsleep)(const clockid_t which_clock, int flags, - struct timespec64 *); + const struct timespec64 *); int (*timer_set)(struct k_itimer *timr, int flags, struct itimerspec64 *new_setting, struct itimerspec64 *old_setting); -- cgit v1.2.3 From ceea5e3771ed2378668455fa21861bead7504df5 Mon Sep 17 00:00:00 2001 From: John Stultz Date: Thu, 8 Jun 2017 16:44:20 -0700 Subject: time: Fix clock->read(clock) race around clocksource changes In tests, which excercise switching of clocksources, a NULL pointer dereference can be observed on AMR64 platforms in the clocksource read() function: u64 clocksource_mmio_readl_down(struct clocksource *c) { return ~(u64)readl_relaxed(to_mmio_clksrc(c)->reg) & c->mask; } This is called from the core timekeeping code via: cycle_now = tkr->read(tkr->clock); tkr->read is the cached tkr->clock->read() function pointer. When the clocksource is changed then tkr->clock and tkr->read are updated sequentially. The code above results in a sequential load operation of tkr->read and tkr->clock as well. If the store to tkr->clock hits between the loads of tkr->read and tkr->clock, then the old read() function is called with the new clock pointer. As a consequence the read() function dereferences a different data structure and the resulting 'reg' pointer can point anywhere including NULL. This problem was introduced when the timekeeping code was switched over to use struct tk_read_base. Before that, it was theoretically possible as well when the compiler decided to reload clock in the code sequence: now = tk->clock->read(tk->clock); Add a helper function which avoids the issue by reading tk_read_base->clock once into a local variable clk and then issue the read function via clk->read(clk). This guarantees that the read() function always gets the proper clocksource pointer handed in. Since there is now no use for the tkr.read pointer, this patch also removes it, and to address stopping the fast timekeeper during suspend/resume, it introduces a dummy clocksource to use rather then just a dummy read function. Signed-off-by: John Stultz Acked-by: Ingo Molnar Cc: Prarit Bhargava Cc: Richard Cochran Cc: Stephen Boyd Cc: stable Cc: Miroslav Lichvar Cc: Daniel Mentz Link: http://lkml.kernel.org/r/1496965462-20003-2-git-send-email-john.stultz@linaro.org Signed-off-by: Thomas Gleixner --- kernel/time/timekeeping.c | 52 ++++++++++++++++++++++++++++++++--------------- 1 file changed, 36 insertions(+), 16 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index 9652bc57fd09..eff94cb8e89e 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -118,6 +118,26 @@ static inline void tk_update_sleep_time(struct timekeeper *tk, ktime_t delta) tk->offs_boot = ktime_add(tk->offs_boot, delta); } +/* + * tk_clock_read - atomic clocksource read() helper + * + * This helper is necessary to use in the read paths because, while the + * seqlock ensures we don't return a bad value while structures are updated, + * it doesn't protect from potential crashes. There is the possibility that + * the tkr's clocksource may change between the read reference, and the + * clock reference passed to the read function. This can cause crashes if + * the wrong clocksource is passed to the wrong read function. + * This isn't necessary to use when holding the timekeeper_lock or doing + * a read of the fast-timekeeper tkrs (which is protected by its own locking + * and update logic). + */ +static inline u64 tk_clock_read(struct tk_read_base *tkr) +{ + struct clocksource *clock = READ_ONCE(tkr->clock); + + return clock->read(clock); +} + #ifdef CONFIG_DEBUG_TIMEKEEPING #define WARNING_FREQ (HZ*300) /* 5 minute rate-limiting */ @@ -175,7 +195,7 @@ static inline u64 timekeeping_get_delta(struct tk_read_base *tkr) */ do { seq = read_seqcount_begin(&tk_core.seq); - now = tkr->read(tkr->clock); + now = tk_clock_read(tkr); last = tkr->cycle_last; mask = tkr->mask; max = tkr->clock->max_cycles; @@ -209,7 +229,7 @@ static inline u64 timekeeping_get_delta(struct tk_read_base *tkr) u64 cycle_now, delta; /* read clocksource */ - cycle_now = tkr->read(tkr->clock); + cycle_now = tk_clock_read(tkr); /* calculate the delta since the last update_wall_time */ delta = clocksource_delta(cycle_now, tkr->cycle_last, tkr->mask); @@ -238,12 +258,10 @@ static void tk_setup_internals(struct timekeeper *tk, struct clocksource *clock) ++tk->cs_was_changed_seq; old_clock = tk->tkr_mono.clock; tk->tkr_mono.clock = clock; - tk->tkr_mono.read = clock->read; tk->tkr_mono.mask = clock->mask; - tk->tkr_mono.cycle_last = tk->tkr_mono.read(clock); + tk->tkr_mono.cycle_last = tk_clock_read(&tk->tkr_mono); tk->tkr_raw.clock = clock; - tk->tkr_raw.read = clock->read; tk->tkr_raw.mask = clock->mask; tk->tkr_raw.cycle_last = tk->tkr_mono.cycle_last; @@ -404,7 +422,7 @@ static __always_inline u64 __ktime_get_fast_ns(struct tk_fast *tkf) now += timekeeping_delta_to_ns(tkr, clocksource_delta( - tkr->read(tkr->clock), + tk_clock_read(tkr), tkr->cycle_last, tkr->mask)); } while (read_seqcount_retry(&tkf->seq, seq)); @@ -461,6 +479,10 @@ static u64 dummy_clock_read(struct clocksource *cs) return cycles_at_suspend; } +static struct clocksource dummy_clock = { + .read = dummy_clock_read, +}; + /** * halt_fast_timekeeper - Prevent fast timekeeper from accessing clocksource. * @tk: Timekeeper to snapshot. @@ -477,13 +499,13 @@ static void halt_fast_timekeeper(struct timekeeper *tk) struct tk_read_base *tkr = &tk->tkr_mono; memcpy(&tkr_dummy, tkr, sizeof(tkr_dummy)); - cycles_at_suspend = tkr->read(tkr->clock); - tkr_dummy.read = dummy_clock_read; + cycles_at_suspend = tk_clock_read(tkr); + tkr_dummy.clock = &dummy_clock; update_fast_timekeeper(&tkr_dummy, &tk_fast_mono); tkr = &tk->tkr_raw; memcpy(&tkr_dummy, tkr, sizeof(tkr_dummy)); - tkr_dummy.read = dummy_clock_read; + tkr_dummy.clock = &dummy_clock; update_fast_timekeeper(&tkr_dummy, &tk_fast_raw); } @@ -649,11 +671,10 @@ static void timekeeping_update(struct timekeeper *tk, unsigned int action) */ static void timekeeping_forward_now(struct timekeeper *tk) { - struct clocksource *clock = tk->tkr_mono.clock; u64 cycle_now, delta; u64 nsec; - cycle_now = tk->tkr_mono.read(clock); + cycle_now = tk_clock_read(&tk->tkr_mono); delta = clocksource_delta(cycle_now, tk->tkr_mono.cycle_last, tk->tkr_mono.mask); tk->tkr_mono.cycle_last = cycle_now; tk->tkr_raw.cycle_last = cycle_now; @@ -929,8 +950,7 @@ void ktime_get_snapshot(struct system_time_snapshot *systime_snapshot) do { seq = read_seqcount_begin(&tk_core.seq); - - now = tk->tkr_mono.read(tk->tkr_mono.clock); + now = tk_clock_read(&tk->tkr_mono); systime_snapshot->cs_was_changed_seq = tk->cs_was_changed_seq; systime_snapshot->clock_was_set_seq = tk->clock_was_set_seq; base_real = ktime_add(tk->tkr_mono.base, @@ -1108,7 +1128,7 @@ int get_device_system_crosststamp(int (*get_time_fn) * Check whether the system counter value provided by the * device driver is on the current timekeeping interval. */ - now = tk->tkr_mono.read(tk->tkr_mono.clock); + now = tk_clock_read(&tk->tkr_mono); interval_start = tk->tkr_mono.cycle_last; if (!cycle_between(interval_start, cycles, now)) { clock_was_set_seq = tk->clock_was_set_seq; @@ -1629,7 +1649,7 @@ void timekeeping_resume(void) * The less preferred source will only be tried if there is no better * usable source. The rtc part is handled separately in rtc core code. */ - cycle_now = tk->tkr_mono.read(clock); + cycle_now = tk_clock_read(&tk->tkr_mono); if ((clock->flags & CLOCK_SOURCE_SUSPEND_NONSTOP) && cycle_now > tk->tkr_mono.cycle_last) { u64 nsec, cyc_delta; @@ -2030,7 +2050,7 @@ void update_wall_time(void) #ifdef CONFIG_ARCH_USES_GETTIMEOFFSET offset = real_tk->cycle_interval; #else - offset = clocksource_delta(tk->tkr_mono.read(tk->tkr_mono.clock), + offset = clocksource_delta(tk_clock_read(&tk->tkr_mono), tk->tkr_mono.cycle_last, tk->tkr_mono.mask); #endif -- cgit v1.2.3 From 3d88d56c5873f6eebe23e05c3da701960146b801 Mon Sep 17 00:00:00 2001 From: John Stultz Date: Thu, 8 Jun 2017 16:44:21 -0700 Subject: time: Fix CLOCK_MONOTONIC_RAW sub-nanosecond accounting Due to how the MONOTONIC_RAW accumulation logic was handled, there is the potential for a 1ns discontinuity when we do accumulations. This small discontinuity has for the most part gone un-noticed, but since ARM64 enabled CLOCK_MONOTONIC_RAW in their vDSO clock_gettime implementation, we've seen failures with the inconsistency-check test in kselftest. This patch addresses the issue by using the same sub-ns accumulation handling that CLOCK_MONOTONIC uses, which avoids the issue for in-kernel users. Since the ARM64 vDSO implementation has its own clock_gettime calculation logic, this patch reduces the frequency of errors, but failures are still seen. The ARM64 vDSO will need to be updated to include the sub-nanosecond xtime_nsec values in its calculation for this issue to be completely fixed. Signed-off-by: John Stultz Tested-by: Daniel Mentz Cc: Prarit Bhargava Cc: Kevin Brodsky Cc: Richard Cochran Cc: Stephen Boyd Cc: Will Deacon Cc: "stable #4 . 8+" Cc: Miroslav Lichvar Link: http://lkml.kernel.org/r/1496965462-20003-3-git-send-email-john.stultz@linaro.org Signed-off-by: Thomas Gleixner --- kernel/time/timekeeping.c | 19 ++++++++++--------- 1 file changed, 10 insertions(+), 9 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index eff94cb8e89e..b602c48cb841 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -280,7 +280,7 @@ static void tk_setup_internals(struct timekeeper *tk, struct clocksource *clock) /* Go back from cycles -> shifted ns */ tk->xtime_interval = interval * clock->mult; tk->xtime_remainder = ntpinterval - tk->xtime_interval; - tk->raw_interval = (interval * clock->mult) >> clock->shift; + tk->raw_interval = interval * clock->mult; /* if changing clocks, convert xtime_nsec shift units */ if (old_clock) { @@ -1996,7 +1996,7 @@ static u64 logarithmic_accumulation(struct timekeeper *tk, u64 offset, u32 shift, unsigned int *clock_set) { u64 interval = tk->cycle_interval << shift; - u64 raw_nsecs; + u64 snsec_per_sec; /* If the offset is smaller than a shifted interval, do nothing */ if (offset < interval) @@ -2011,14 +2011,15 @@ static u64 logarithmic_accumulation(struct timekeeper *tk, u64 offset, *clock_set |= accumulate_nsecs_to_secs(tk); /* Accumulate raw time */ - raw_nsecs = (u64)tk->raw_interval << shift; - raw_nsecs += tk->raw_time.tv_nsec; - if (raw_nsecs >= NSEC_PER_SEC) { - u64 raw_secs = raw_nsecs; - raw_nsecs = do_div(raw_secs, NSEC_PER_SEC); - tk->raw_time.tv_sec += raw_secs; + tk->tkr_raw.xtime_nsec += (u64)tk->raw_time.tv_nsec << tk->tkr_raw.shift; + tk->tkr_raw.xtime_nsec += tk->raw_interval << shift; + snsec_per_sec = (u64)NSEC_PER_SEC << tk->tkr_raw.shift; + while (tk->tkr_raw.xtime_nsec >= snsec_per_sec) { + tk->tkr_raw.xtime_nsec -= snsec_per_sec; + tk->raw_time.tv_sec++; } - tk->raw_time.tv_nsec = raw_nsecs; + tk->raw_time.tv_nsec = tk->tkr_raw.xtime_nsec >> tk->tkr_raw.shift; + tk->tkr_raw.xtime_nsec -= (u64)tk->raw_time.tv_nsec << tk->tkr_raw.shift; /* Accumulate error between NTP and clock interval */ tk->ntp_error += tk->ntp_tick << shift; -- cgit v1.2.3 From d15bc69affc57d7985a01745ca28eafa0772325b Mon Sep 17 00:00:00 2001 From: Peter Meerwald-Stadler Date: Tue, 30 May 2017 21:41:03 +0200 Subject: timers: Fix parameter description of try_to_del_timer_sync() Signed-off-by: Peter Meerwald-Stadler Link: http://lkml.kernel.org/r/20170530194103.7454-1-pmeerw@pmeerw.net Cc: John Stultz Cc: trivial@rustcorp.com.au Signed-off-by: Thomas Gleixner --- kernel/time/timer.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/time') diff --git a/kernel/time/timer.c b/kernel/time/timer.c index 152a706ef8b8..709a404bd133 100644 --- a/kernel/time/timer.c +++ b/kernel/time/timer.c @@ -1150,7 +1150,7 @@ EXPORT_SYMBOL(del_timer); /** * try_to_del_timer_sync - Try to deactivate a timer - * @timer: timer do del + * @timer: timer to delete * * This function tries to deactivate a timer. Upon successful (ret >= 0) * exit the timer is not queued and the handler is not running on any CPU. -- cgit v1.2.3 From 35eb7258c009dc478338e674a5a84d25d0929c56 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 20 Jun 2017 17:37:35 +0200 Subject: itimer: Make timeval to nsec conversion range limited The expiry time of a itimer is supplied through sys_setitimer() via a struct timeval. The timeval is validated for correctness. In the actual set timer implementation the timeval is converted to a scalar nanoseconds value. If the tv_sec part of the time spec is large enough the conversion to nanoseconds (sec * NSEC_PER_SEC) overflows 64bit. Mitigate that by using the timeval_to_ktime() conversion function, which checks the tv_sec part for a potential mult overflow and clamps the result to KTIME_MAX, which is about 292 years. Reported-by: Xishi Qiu Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: John Stultz Link: http://lkml.kernel.org/r/20170620154113.505981643@linutronix.de --- kernel/time/itimer.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/itimer.c b/kernel/time/itimer.c index 9dd7ff5e445a..2ef98a02376a 100644 --- a/kernel/time/itimer.c +++ b/kernel/time/itimer.c @@ -152,8 +152,12 @@ static void set_cpu_itimer(struct task_struct *tsk, unsigned int clock_id, u64 oval, nval, ointerval, ninterval; struct cpu_itimer *it = &tsk->signal->it[clock_id]; - nval = timeval_to_ns(&value->it_value); - ninterval = timeval_to_ns(&value->it_interval); + /* + * Use the to_ktime conversion because that clamps the maximum + * value to KTIME_MAX and avoid multiplication overflows. + */ + nval = ktime_to_ns(timeval_to_ktime(value->it_value)); + ninterval = ktime_to_ns(timeval_to_ktime(value->it_interval)); spin_lock_irq(&tsk->sighand->siglock); -- cgit v1.2.3 From 098b0e01a91c42aaaf0425605cd126b03fcb0bcf Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 20 Jun 2017 17:37:36 +0200 Subject: posix-cpu-timers: Make timespec to nsec conversion safe The expiry time of a posix cpu timer is supplied through sys_timer_set() via a struct timespec. The timespec is validated for correctness. In the actual set timer implementation the timespec is converted to a scalar nanoseconds value. If the tv_sec part of the time spec is large enough the conversion to nanoseconds (sec * NSEC_PER_SEC) overflows 64bit. Mitigate that by using the timespec_to_ktime() conversion function, which checks the tv_sec part for a potential mult overflow and clamps the result to KTIME_MAX, which is about 292 years. Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: Xishi Qiu Cc: John Stultz Link: http://lkml.kernel.org/r/20170620154113.588276707@linutronix.de --- kernel/time/posix-cpu-timers.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'kernel/time') diff --git a/kernel/time/posix-cpu-timers.c b/kernel/time/posix-cpu-timers.c index 9df618ee64cf..60cb24ac9ebc 100644 --- a/kernel/time/posix-cpu-timers.c +++ b/kernel/time/posix-cpu-timers.c @@ -580,7 +580,11 @@ static int posix_cpu_timer_set(struct k_itimer *timer, int timer_flags, WARN_ON_ONCE(p == NULL); - new_expires = timespec64_to_ns(&new->it_value); + /* + * Use the to_ktime conversion because that clamps the maximum + * value to KTIME_MAX and avoid multiplication overflows. + */ + new_expires = ktime_to_ns(timespec64_to_ktime(new->it_value)); /* * Protect against sighand release/switch in exit/exec and p->cpu_timers -- cgit v1.2.3 From fc6eead7c1e2e5376c25d2795d4539fdacbc0648 Mon Sep 17 00:00:00 2001 From: John Stultz Date: Mon, 22 May 2017 17:20:20 -0700 Subject: time: Clean up CLOCK_MONOTONIC_RAW time handling Now that we fixed the sub-ns handling for CLOCK_MONOTONIC_RAW, remove the duplicitive tk->raw_time.tv_nsec, which can be stored in tk->tkr_raw.xtime_nsec (similarly to how its handled for monotonic time). Cc: Thomas Gleixner Cc: Ingo Molnar Cc: Miroslav Lichvar Cc: Richard Cochran Cc: Prarit Bhargava Cc: Stephen Boyd Cc: Kevin Brodsky Cc: Will Deacon Cc: Daniel Mentz Tested-by: Daniel Mentz Signed-off-by: John Stultz --- kernel/time/timekeeping.c | 45 +++++++++++++++++++++++++-------------------- 1 file changed, 25 insertions(+), 20 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index b602c48cb841..0454bfa24353 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -72,6 +72,10 @@ static inline void tk_normalize_xtime(struct timekeeper *tk) tk->tkr_mono.xtime_nsec -= (u64)NSEC_PER_SEC << tk->tkr_mono.shift; tk->xtime_sec++; } + while (tk->tkr_raw.xtime_nsec >= ((u64)NSEC_PER_SEC << tk->tkr_raw.shift)) { + tk->tkr_raw.xtime_nsec -= (u64)NSEC_PER_SEC << tk->tkr_raw.shift; + tk->raw_sec++; + } } static inline struct timespec64 tk_xtime(struct timekeeper *tk) @@ -285,12 +289,14 @@ static void tk_setup_internals(struct timekeeper *tk, struct clocksource *clock) /* if changing clocks, convert xtime_nsec shift units */ if (old_clock) { int shift_change = clock->shift - old_clock->shift; - if (shift_change < 0) + if (shift_change < 0) { tk->tkr_mono.xtime_nsec >>= -shift_change; - else + tk->tkr_raw.xtime_nsec >>= -shift_change; + } else { tk->tkr_mono.xtime_nsec <<= shift_change; + tk->tkr_raw.xtime_nsec <<= shift_change; + } } - tk->tkr_raw.xtime_nsec = 0; tk->tkr_mono.shift = clock->shift; tk->tkr_raw.shift = clock->shift; @@ -619,9 +625,6 @@ static inline void tk_update_ktime_data(struct timekeeper *tk) nsec = (u32) tk->wall_to_monotonic.tv_nsec; tk->tkr_mono.base = ns_to_ktime(seconds * NSEC_PER_SEC + nsec); - /* Update the monotonic raw base */ - tk->tkr_raw.base = timespec64_to_ktime(tk->raw_time); - /* * The sum of the nanoseconds portions of xtime and * wall_to_monotonic can be greater/equal one second. Take @@ -631,6 +634,11 @@ static inline void tk_update_ktime_data(struct timekeeper *tk) if (nsec >= NSEC_PER_SEC) seconds++; tk->ktime_sec = seconds; + + /* Update the monotonic raw base */ + seconds = tk->raw_sec; + nsec = (u32)(tk->tkr_raw.xtime_nsec >> tk->tkr_raw.shift); + tk->tkr_raw.base = ns_to_ktime(seconds * NSEC_PER_SEC + nsec); } /* must hold timekeeper_lock */ @@ -672,7 +680,6 @@ static void timekeeping_update(struct timekeeper *tk, unsigned int action) static void timekeeping_forward_now(struct timekeeper *tk) { u64 cycle_now, delta; - u64 nsec; cycle_now = tk_clock_read(&tk->tkr_mono); delta = clocksource_delta(cycle_now, tk->tkr_mono.cycle_last, tk->tkr_mono.mask); @@ -684,10 +691,13 @@ static void timekeeping_forward_now(struct timekeeper *tk) /* If arch requires, add in get_arch_timeoffset() */ tk->tkr_mono.xtime_nsec += (u64)arch_gettimeoffset() << tk->tkr_mono.shift; - tk_normalize_xtime(tk); - nsec = clocksource_cyc2ns(delta, tk->tkr_raw.mult, tk->tkr_raw.shift); - timespec64_add_ns(&tk->raw_time, nsec); + tk->tkr_raw.xtime_nsec += delta * tk->tkr_raw.mult; + + /* If arch requires, add in get_arch_timeoffset() */ + tk->tkr_raw.xtime_nsec += (u64)arch_gettimeoffset() << tk->tkr_raw.shift; + + tk_normalize_xtime(tk); } /** @@ -1373,19 +1383,18 @@ int timekeeping_notify(struct clocksource *clock) void getrawmonotonic64(struct timespec64 *ts) { struct timekeeper *tk = &tk_core.timekeeper; - struct timespec64 ts64; unsigned long seq; u64 nsecs; do { seq = read_seqcount_begin(&tk_core.seq); + ts->tv_sec = tk->raw_sec; nsecs = timekeeping_get_ns(&tk->tkr_raw); - ts64 = tk->raw_time; } while (read_seqcount_retry(&tk_core.seq, seq)); - timespec64_add_ns(&ts64, nsecs); - *ts = ts64; + ts->tv_nsec = 0; + timespec64_add_ns(ts, nsecs); } EXPORT_SYMBOL(getrawmonotonic64); @@ -1509,8 +1518,7 @@ void __init timekeeping_init(void) tk_setup_internals(tk, clock); tk_set_xtime(tk, &now); - tk->raw_time.tv_sec = 0; - tk->raw_time.tv_nsec = 0; + tk->raw_sec = 0; if (boot.tv_sec == 0 && boot.tv_nsec == 0) boot = tk_xtime(tk); @@ -2011,15 +2019,12 @@ static u64 logarithmic_accumulation(struct timekeeper *tk, u64 offset, *clock_set |= accumulate_nsecs_to_secs(tk); /* Accumulate raw time */ - tk->tkr_raw.xtime_nsec += (u64)tk->raw_time.tv_nsec << tk->tkr_raw.shift; tk->tkr_raw.xtime_nsec += tk->raw_interval << shift; snsec_per_sec = (u64)NSEC_PER_SEC << tk->tkr_raw.shift; while (tk->tkr_raw.xtime_nsec >= snsec_per_sec) { tk->tkr_raw.xtime_nsec -= snsec_per_sec; - tk->raw_time.tv_sec++; + tk->raw_sec++; } - tk->raw_time.tv_nsec = tk->tkr_raw.xtime_nsec >> tk->tkr_raw.shift; - tk->tkr_raw.xtime_nsec -= (u64)tk->raw_time.tv_nsec << tk->tkr_raw.shift; /* Accumulate error between NTP and clock interval */ tk->ntp_error += tk->ntp_tick << shift; -- cgit v1.2.3 From 369adf04d80a7e179b9ea6d74cc01c233f142f47 Mon Sep 17 00:00:00 2001 From: John Stultz Date: Fri, 12 May 2017 10:58:18 -0700 Subject: time: Add warning about imminent deprecation of CONFIG_GENERIC_TIME_VSYSCALL_OLD CONFIG_GENERIC_TIME_VSYSCALL_OLD was introduced five years ago to allow a transition from the old vsyscall implementations to the new method (which simplified internal accounting and made timekeeping more precise). However, PPC and IA64 have yet to make the transition, despite in some cases me sending test patches to try to help it along. http://patches.linaro.org/patch/30501/ http://patches.linaro.org/patch/35412/ If its helpful, my last pass at the patches can be found here: https://git.linaro.org/people/john.stultz/linux.git dev/oldvsyscall-cleanup So I think its time to set a deadline and make it clear this is going away. So this patch adds warnings about this functionality being dropped. Likely to be in v4.15. Cc: Thomas Gleixner Cc: Ingo Molnar Cc: Miroslav Lichvar Cc: Richard Cochran Cc: Prarit Bhargava Cc: Marcelo Tosatti Cc: Paul Mackerras Cc: Anton Blanchard Cc: Benjamin Herrenschmidt Cc: Tony Luck Cc: Michael Ellerman Cc: Fenghua Yu Signed-off-by: John Stultz --- kernel/time/timekeeping.c | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel/time') diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index 0454bfa24353..cedafa008de5 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -516,6 +516,7 @@ static void halt_fast_timekeeper(struct timekeeper *tk) } #ifdef CONFIG_GENERIC_TIME_VSYSCALL_OLD +#warning Please contact your maintainers, as GENERIC_TIME_VSYSCALL_OLD compatibity will disappear soon. static inline void update_vsyscall(struct timekeeper *tk) { -- cgit v1.2.3 From 3c85d6db5e5f05ae6c3d7f5a0ceceb43746a5ca7 Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Mon, 19 Jun 2017 04:12:00 +0200 Subject: sched/loadavg: Generalize "_idle" naming to "_nohz" The loadavg naming code still assumes that nohz == idle whereas its code is actually handling well both nohz idle and nohz full. So lets fix the naming according to what the code actually does, to unconfuse the reader. Signed-off-by: Frederic Weisbecker Acked-by: Rik van Riel Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/1497838322-10913-2-git-send-email-fweisbec@gmail.com Signed-off-by: Ingo Molnar --- kernel/time/tick-sched.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c index 9c2dc64e31d8..b1b58a07e042 100644 --- a/kernel/time/tick-sched.c +++ b/kernel/time/tick-sched.c @@ -783,7 +783,7 @@ static ktime_t tick_nohz_stop_sched_tick(struct tick_sched *ts, */ if (!ts->tick_stopped) { nohz_balance_enter_idle(cpu); - calc_load_enter_idle(); + calc_load_nohz_start(); cpu_load_update_nohz_start(); ts->last_tick = hrtimer_get_expires(&ts->sched_timer); @@ -823,7 +823,7 @@ static void tick_nohz_restart_sched_tick(struct tick_sched *ts, ktime_t now) */ timer_clear_idle(); - calc_load_exit_idle(); + calc_load_nohz_stop(); touch_softlockup_watchdog_sched(); /* * Cancel the scheduled timer and restore the tick -- cgit v1.2.3 From a0db971e4eb69fc84eb3d7ef94f718b483550b4a Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Mon, 19 Jun 2017 04:12:01 +0200 Subject: nohz: Move idle balancer registration to the idle path The idle load balancing registration path assumes that we only stop the tick when the CPU is idle, ignoring the nohz full case. As a result, a nohz full CPU that is running a task may be chosen to perform idle load balancing. Lets make sure that only CPUs in dynticks idle mode can be picked as idle load balancers. Signed-off-by: Frederic Weisbecker Acked-by: Rik van Riel Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/1497838322-10913-3-git-send-email-fweisbec@gmail.com Signed-off-by: Ingo Molnar --- kernel/time/tick-sched.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c index b1b58a07e042..db023e9cbb25 100644 --- a/kernel/time/tick-sched.c +++ b/kernel/time/tick-sched.c @@ -782,7 +782,6 @@ static ktime_t tick_nohz_stop_sched_tick(struct tick_sched *ts, * the scheduler tick in nohz_restart_sched_tick. */ if (!ts->tick_stopped) { - nohz_balance_enter_idle(cpu); calc_load_nohz_start(); cpu_load_update_nohz_start(); @@ -923,8 +922,10 @@ static void __tick_nohz_idle_enter(struct tick_sched *ts) ts->idle_expires = expires; } - if (!was_stopped && ts->tick_stopped) + if (!was_stopped && ts->tick_stopped) { ts->idle_jiffies = ts->last_jiffies; + nohz_balance_enter_idle(cpu); + } } } -- cgit v1.2.3 From f59dd9c886acb3abb188e8e94a99436560976835 Mon Sep 17 00:00:00 2001 From: Deepa Dinamani Date: Sat, 24 Jun 2017 11:45:02 -0700 Subject: time: add get_timespec64 and put_timespec64 Add helper functions to convert between struct timespec64 and struct timespec at userspace boundaries. This is a preparatory patch to use timespec64 as the basic type internally in the kernel as timespec is not y2038 safe on 32 bit systems. The patch helps the cause by containing all data conversions at the userspace boundaries within these functions. Suggested-by: Arnd Bergmann Signed-off-by: Deepa Dinamani Signed-off-by: Al Viro --- kernel/time/time.c | 28 ++++++++++++++++++++++++++++ 1 file changed, 28 insertions(+) (limited to 'kernel/time') diff --git a/kernel/time/time.c b/kernel/time/time.c index 7c89e437c4d7..adb9853ca6b0 100644 --- a/kernel/time/time.c +++ b/kernel/time/time.c @@ -890,3 +890,31 @@ struct timespec64 timespec64_add_safe(const struct timespec64 lhs, return res; } + +int get_timespec64(struct timespec64 *ts, + const struct timespec __user *uts) +{ + struct timespec kts; + int ret; + + ret = copy_from_user(&kts, uts, sizeof(kts)); + if (ret) + return -EFAULT; + + ts->tv_sec = kts.tv_sec; + ts->tv_nsec = kts.tv_nsec; + + return 0; +} +EXPORT_SYMBOL_GPL(get_timespec64); + +int put_timespec64(const struct timespec64 *ts, + struct timespec __user *uts) +{ + struct timespec kts = { + .tv_sec = ts->tv_sec, + .tv_nsec = ts->tv_nsec + }; + return copy_to_user(uts, &kts, sizeof(kts)) ? -EFAULT : 0; +} +EXPORT_SYMBOL_GPL(put_timespec64); -- cgit v1.2.3 From d5b7ffbfbdacc29e4db035f90665951668fa9c58 Mon Sep 17 00:00:00 2001 From: Deepa Dinamani Date: Sat, 24 Jun 2017 11:45:03 -0700 Subject: time: introduce {get,put}_itimerspec64 As we change the user space type for the timerfd and posix timer functions to newer data types, we need some form of conversion helpers to avoid duplicating that logic. Suggested-by: Arnd Bergmann Signed-off-by: Deepa Dinamani Signed-off-by: Al Viro --- kernel/time/time.c | 30 ++++++++++++++++++++++++++++++ 1 file changed, 30 insertions(+) (limited to 'kernel/time') diff --git a/kernel/time/time.c b/kernel/time/time.c index adb9853ca6b0..44a8c1402133 100644 --- a/kernel/time/time.c +++ b/kernel/time/time.c @@ -918,3 +918,33 @@ int put_timespec64(const struct timespec64 *ts, return copy_to_user(uts, &kts, sizeof(kts)) ? -EFAULT : 0; } EXPORT_SYMBOL_GPL(put_timespec64); + +int get_itimerspec64(struct itimerspec64 *it, + const struct itimerspec __user *uit) +{ + int ret; + + ret = get_timespec64(&it->it_interval, &uit->it_interval); + if (ret) + return ret; + + ret = get_timespec64(&it->it_value, &uit->it_value); + + return ret; +} +EXPORT_SYMBOL_GPL(get_itimerspec64); + +int put_itimerspec64(const struct itimerspec64 *it, + struct itimerspec __user *uit) +{ + int ret; + + ret = put_timespec64(&it->it_interval, &uit->it_interval); + if (ret) + return ret; + + ret = put_timespec64(&it->it_value, &uit->it_value); + + return ret; +} +EXPORT_SYMBOL_GPL(put_itimerspec64); -- cgit v1.2.3 From 63a766a1780f9581e8885bdb64270a594a84f81a Mon Sep 17 00:00:00 2001 From: Deepa Dinamani Date: Sat, 24 Jun 2017 11:45:04 -0700 Subject: posix-stubs: Conditionally include COMPAT_SYS_NI defines These apis only need to be defined if CONFIG_COMPAT is enabled. Signed-off-by: Deepa Dinamani Signed-off-by: Al Viro --- kernel/time/posix-stubs.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/posix-stubs.c b/kernel/time/posix-stubs.c index 38f3b20efa29..65878221cbfb 100644 --- a/kernel/time/posix-stubs.c +++ b/kernel/time/posix-stubs.c @@ -41,12 +41,6 @@ SYS_NI(setitimer); #ifdef __ARCH_WANT_SYS_ALARM SYS_NI(alarm); #endif -COMPAT_SYS_NI(timer_create); -COMPAT_SYS_NI(clock_adjtime); -COMPAT_SYS_NI(timer_settime); -COMPAT_SYS_NI(timer_gettime); -COMPAT_SYS_NI(getitimer); -COMPAT_SYS_NI(setitimer); /* * We preserve minimal support for CLOCK_REALTIME and CLOCK_MONOTONIC @@ -138,6 +132,13 @@ SYSCALL_DEFINE4(clock_nanosleep, const clockid_t, which_clock, int, flags, } #ifdef CONFIG_COMPAT +COMPAT_SYS_NI(timer_create); +COMPAT_SYS_NI(clock_adjtime); +COMPAT_SYS_NI(timer_settime); +COMPAT_SYS_NI(timer_gettime); +COMPAT_SYS_NI(getitimer); +COMPAT_SYS_NI(setitimer); + COMPAT_SYSCALL_DEFINE2(clock_settime, const clockid_t, which_clock, struct compat_timespec __user *, tp) { -- cgit v1.2.3 From 2287d8664fe7345ead891017eccd879fc605305e Mon Sep 17 00:00:00 2001 From: Sebastian Andrzej Siewior Date: Tue, 27 Jun 2017 18:15:38 +0200 Subject: timers: Make the cpu base lock raw The timers cpu base lock could not be converted to a raw spinlock becaue the lock held time was non-deterministic due to cascading and long lasting timer wheel traversals. The rework of the timer wheel to the new non-cascading model removed also the wheel traversals and the lock held times are deterministic now. This allows to make the lock raw and thereby unbreaks NOHz* on preempt-RT. Signed-off-by: Sebastian Andrzej Siewior Link: http://lkml.kernel.org/r/20170627161538.30257-1-bigeasy@linutronix.de Signed-off-by: Thomas Gleixner --- kernel/time/timer.c | 48 ++++++++++++++++++++++++------------------------ 1 file changed, 24 insertions(+), 24 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/timer.c b/kernel/time/timer.c index 709a404bd133..71ce3f4eead3 100644 --- a/kernel/time/timer.c +++ b/kernel/time/timer.c @@ -195,7 +195,7 @@ EXPORT_SYMBOL(jiffies_64); #endif struct timer_base { - spinlock_t lock; + raw_spinlock_t lock; struct timer_list *running_timer; unsigned long clk; unsigned long next_expiry; @@ -913,10 +913,10 @@ static struct timer_base *lock_timer_base(struct timer_list *timer, if (!(tf & TIMER_MIGRATING)) { base = get_timer_base(tf); - spin_lock_irqsave(&base->lock, *flags); + raw_spin_lock_irqsave(&base->lock, *flags); if (timer->flags == tf) return base; - spin_unlock_irqrestore(&base->lock, *flags); + raw_spin_unlock_irqrestore(&base->lock, *flags); } cpu_relax(); } @@ -986,9 +986,9 @@ __mod_timer(struct timer_list *timer, unsigned long expires, bool pending_only) /* See the comment in lock_timer_base() */ timer->flags |= TIMER_MIGRATING; - spin_unlock(&base->lock); + raw_spin_unlock(&base->lock); base = new_base; - spin_lock(&base->lock); + raw_spin_lock(&base->lock); WRITE_ONCE(timer->flags, (timer->flags & ~TIMER_BASEMASK) | base->cpu); } @@ -1013,7 +1013,7 @@ __mod_timer(struct timer_list *timer, unsigned long expires, bool pending_only) } out_unlock: - spin_unlock_irqrestore(&base->lock, flags); + raw_spin_unlock_irqrestore(&base->lock, flags); return ret; } @@ -1106,16 +1106,16 @@ void add_timer_on(struct timer_list *timer, int cpu) if (base != new_base) { timer->flags |= TIMER_MIGRATING; - spin_unlock(&base->lock); + raw_spin_unlock(&base->lock); base = new_base; - spin_lock(&base->lock); + raw_spin_lock(&base->lock); WRITE_ONCE(timer->flags, (timer->flags & ~TIMER_BASEMASK) | cpu); } debug_activate(timer, timer->expires); internal_add_timer(base, timer); - spin_unlock_irqrestore(&base->lock, flags); + raw_spin_unlock_irqrestore(&base->lock, flags); } EXPORT_SYMBOL_GPL(add_timer_on); @@ -1141,7 +1141,7 @@ int del_timer(struct timer_list *timer) if (timer_pending(timer)) { base = lock_timer_base(timer, &flags); ret = detach_if_pending(timer, base, true); - spin_unlock_irqrestore(&base->lock, flags); + raw_spin_unlock_irqrestore(&base->lock, flags); } return ret; @@ -1168,7 +1168,7 @@ int try_to_del_timer_sync(struct timer_list *timer) if (base->running_timer != timer) ret = detach_if_pending(timer, base, true); - spin_unlock_irqrestore(&base->lock, flags); + raw_spin_unlock_irqrestore(&base->lock, flags); return ret; } @@ -1299,13 +1299,13 @@ static void expire_timers(struct timer_base *base, struct hlist_head *head) data = timer->data; if (timer->flags & TIMER_IRQSAFE) { - spin_unlock(&base->lock); + raw_spin_unlock(&base->lock); call_timer_fn(timer, fn, data); - spin_lock(&base->lock); + raw_spin_lock(&base->lock); } else { - spin_unlock_irq(&base->lock); + raw_spin_unlock_irq(&base->lock); call_timer_fn(timer, fn, data); - spin_lock_irq(&base->lock); + raw_spin_lock_irq(&base->lock); } } } @@ -1474,7 +1474,7 @@ u64 get_next_timer_interrupt(unsigned long basej, u64 basem) if (cpu_is_offline(smp_processor_id())) return expires; - spin_lock(&base->lock); + raw_spin_lock(&base->lock); nextevt = __next_timer_interrupt(base); is_max_delta = (nextevt == base->clk + NEXT_TIMER_MAX_DELTA); base->next_expiry = nextevt; @@ -1502,7 +1502,7 @@ u64 get_next_timer_interrupt(unsigned long basej, u64 basem) if ((expires - basem) > TICK_NSEC) base->is_idle = true; } - spin_unlock(&base->lock); + raw_spin_unlock(&base->lock); return cmp_next_hrtimer_event(basem, expires); } @@ -1590,7 +1590,7 @@ static inline void __run_timers(struct timer_base *base) if (!time_after_eq(jiffies, base->clk)) return; - spin_lock_irq(&base->lock); + raw_spin_lock_irq(&base->lock); while (time_after_eq(jiffies, base->clk)) { @@ -1601,7 +1601,7 @@ static inline void __run_timers(struct timer_base *base) expire_timers(base, heads + levels); } base->running_timer = NULL; - spin_unlock_irq(&base->lock); + raw_spin_unlock_irq(&base->lock); } /* @@ -1786,16 +1786,16 @@ int timers_dead_cpu(unsigned int cpu) * The caller is globally serialized and nobody else * takes two locks at once, deadlock is not possible. */ - spin_lock_irq(&new_base->lock); - spin_lock_nested(&old_base->lock, SINGLE_DEPTH_NESTING); + raw_spin_lock_irq(&new_base->lock); + raw_spin_lock_nested(&old_base->lock, SINGLE_DEPTH_NESTING); BUG_ON(old_base->running_timer); for (i = 0; i < WHEEL_SIZE; i++) migrate_timer_list(new_base, old_base->vectors + i); - spin_unlock(&old_base->lock); - spin_unlock_irq(&new_base->lock); + raw_spin_unlock(&old_base->lock); + raw_spin_unlock_irq(&new_base->lock); put_cpu_ptr(&timer_bases); } return 0; @@ -1811,7 +1811,7 @@ static void __init init_timer_cpu(int cpu) for (i = 0; i < NR_BASES; i++) { base = per_cpu_ptr(&timer_bases[i], cpu); base->cpu = cpu; - spin_lock_init(&base->lock); + raw_spin_lock_init(&base->lock); base->clk = jiffies; } } -- cgit v1.2.3 From 5c4994102fb508d4a0f7a8afa46560c314c1ebd4 Mon Sep 17 00:00:00 2001 From: Deepa Dinamani Date: Sat, 24 Jun 2017 11:45:05 -0700 Subject: posix-timers: Use get_timespec64() and put_timespec64() Usage of these apis and their compat versions makes the syscalls: clock_gettime, clock_settime, clock_getres and their compat implementations simpler. This is a preparatory patch to isolate data conversions to struct timespec64 at userspace boundaries. This helps contain the changes needed to transition to new y2038 safe types. Signed-off-by: Deepa Dinamani Signed-off-by: Al Viro --- kernel/time/posix-stubs.c | 83 +++++++++++++++++++++++++--------------------- kernel/time/posix-timers.c | 63 ++++++++++++++--------------------- 2 files changed, 70 insertions(+), 76 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/posix-stubs.c b/kernel/time/posix-stubs.c index 65878221cbfb..06f34feb635e 100644 --- a/kernel/time/posix-stubs.c +++ b/kernel/time/posix-stubs.c @@ -51,40 +51,52 @@ SYS_NI(alarm); SYSCALL_DEFINE2(clock_settime, const clockid_t, which_clock, const struct timespec __user *, tp) { - struct timespec64 new_tp64; - struct timespec new_tp; + struct timespec64 new_tp; if (which_clock != CLOCK_REALTIME) return -EINVAL; - if (copy_from_user(&new_tp, tp, sizeof (*tp))) + if (get_timespec64(&new_tp, tp)) return -EFAULT; - new_tp64 = timespec_to_timespec64(new_tp); - return do_sys_settimeofday64(&new_tp64, NULL); + return do_sys_settimeofday64(&new_tp, NULL); } -SYSCALL_DEFINE2(clock_gettime, const clockid_t, which_clock, - struct timespec __user *,tp) +int do_clock_gettime(clockid_t which_clock, struct timespec64 *tp) { - struct timespec64 kernel_tp64; - struct timespec kernel_tp; - switch (which_clock) { - case CLOCK_REALTIME: ktime_get_real_ts64(&kernel_tp64); break; - case CLOCK_MONOTONIC: ktime_get_ts64(&kernel_tp64); break; - case CLOCK_BOOTTIME: get_monotonic_boottime64(&kernel_tp64); break; - default: return -EINVAL; + case CLOCK_REALTIME: + ktime_get_real_ts64(tp); + break; + case CLOCK_MONOTONIC: + ktime_get_ts64(tp); + break; + case CLOCK_BOOTTIME: + get_monotonic_boottime64(tp); + break; + default: + return -EINVAL; } - kernel_tp = timespec64_to_timespec(kernel_tp64); - if (copy_to_user(tp, &kernel_tp, sizeof (kernel_tp))) + return 0; +} +SYSCALL_DEFINE2(clock_gettime, const clockid_t, which_clock, + struct timespec __user *, tp) +{ + int ret; + struct timespec64 kernel_tp; + + ret = do_clock_gettime(which_clock, &kernel_tp); + if (ret) + return ret; + + if (put_timespec64(&kernel_tp, tp)) return -EFAULT; return 0; } SYSCALL_DEFINE2(clock_getres, const clockid_t, which_clock, struct timespec __user *, tp) { - struct timespec rtn_tp = { + struct timespec64 rtn_tp = { .tv_sec = 0, .tv_nsec = hrtimer_resolution, }; @@ -93,7 +105,7 @@ SYSCALL_DEFINE2(clock_getres, const clockid_t, which_clock, struct timespec __us case CLOCK_REALTIME: case CLOCK_MONOTONIC: case CLOCK_BOOTTIME: - if (copy_to_user(tp, &rtn_tp, sizeof(rtn_tp))) + if (put_timespec64(&rtn_tp, tp)) return -EFAULT; return 0; default: @@ -142,41 +154,35 @@ COMPAT_SYS_NI(setitimer); COMPAT_SYSCALL_DEFINE2(clock_settime, const clockid_t, which_clock, struct compat_timespec __user *, tp) { - struct timespec64 new_tp64; - struct timespec new_tp; + struct timespec64 new_tp; if (which_clock != CLOCK_REALTIME) return -EINVAL; - if (compat_get_timespec(&new_tp, tp)) + if (compat_get_timespec64(&new_tp, tp)) return -EFAULT; - new_tp64 = timespec_to_timespec64(new_tp); - return do_sys_settimeofday64(&new_tp64, NULL); + return do_sys_settimeofday64(&new_tp, NULL); } -COMPAT_SYSCALL_DEFINE2(clock_gettime, const clockid_t, which_clock, - struct compat_timespec __user *,tp) +COMPAT_SYSCALL_DEFINE2(clock_gettime, clockid_t, which_clock, + struct compat_timespec __user *, tp) { - struct timespec64 kernel_tp64; - struct timespec kernel_tp; + int ret; + struct timespec64 kernel_tp; - switch (which_clock) { - case CLOCK_REALTIME: ktime_get_real_ts64(&kernel_tp64); break; - case CLOCK_MONOTONIC: ktime_get_ts64(&kernel_tp64); break; - case CLOCK_BOOTTIME: get_monotonic_boottime64(&kernel_tp64); break; - default: return -EINVAL; - } + ret = do_clock_gettime(which_clock, &kernel_tp); + if (ret) + return ret; - kernel_tp = timespec64_to_timespec(kernel_tp64); - if (compat_put_timespec(&kernel_tp, tp)) + if (compat_put_timespec64(&kernel_tp, tp)) return -EFAULT; return 0; } -COMPAT_SYSCALL_DEFINE2(clock_getres, const clockid_t, which_clock, +COMPAT_SYSCALL_DEFINE2(clock_getres, clockid_t, which_clock, struct compat_timespec __user *, tp) { - struct timespec rtn_tp = { + struct timespec64 rtn_tp = { .tv_sec = 0, .tv_nsec = hrtimer_resolution, }; @@ -185,13 +191,14 @@ COMPAT_SYSCALL_DEFINE2(clock_getres, const clockid_t, which_clock, case CLOCK_REALTIME: case CLOCK_MONOTONIC: case CLOCK_BOOTTIME: - if (compat_put_timespec(&rtn_tp, tp)) + if (compat_put_timespec64(&rtn_tp, tp)) return -EFAULT; return 0; default: return -EINVAL; } } + COMPAT_SYSCALL_DEFINE4(clock_nanosleep, clockid_t, which_clock, int, flags, struct compat_timespec __user *, rqtp, struct compat_timespec __user *, rmtp) diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c index 82d67be7d9d1..39322ae5dd87 100644 --- a/kernel/time/posix-timers.c +++ b/kernel/time/posix-timers.c @@ -1049,34 +1049,30 @@ SYSCALL_DEFINE2(clock_settime, const clockid_t, which_clock, const struct timespec __user *, tp) { const struct k_clock *kc = clockid_to_kclock(which_clock); - struct timespec64 new_tp64; - struct timespec new_tp; + struct timespec64 new_tp; if (!kc || !kc->clock_set) return -EINVAL; - if (copy_from_user(&new_tp, tp, sizeof (*tp))) + if (get_timespec64(&new_tp, tp)) return -EFAULT; - new_tp64 = timespec_to_timespec64(new_tp); - return kc->clock_set(which_clock, &new_tp64); + return kc->clock_set(which_clock, &new_tp); } SYSCALL_DEFINE2(clock_gettime, const clockid_t, which_clock, struct timespec __user *,tp) { const struct k_clock *kc = clockid_to_kclock(which_clock); - struct timespec64 kernel_tp64; - struct timespec kernel_tp; + struct timespec64 kernel_tp; int error; if (!kc) return -EINVAL; - error = kc->clock_get(which_clock, &kernel_tp64); - kernel_tp = timespec64_to_timespec(kernel_tp64); + error = kc->clock_get(which_clock, &kernel_tp); - if (!error && copy_to_user(tp, &kernel_tp, sizeof (kernel_tp))) + if (!error && put_timespec64(&kernel_tp, tp)) error = -EFAULT; return error; @@ -1109,17 +1105,15 @@ SYSCALL_DEFINE2(clock_getres, const clockid_t, which_clock, struct timespec __user *, tp) { const struct k_clock *kc = clockid_to_kclock(which_clock); - struct timespec64 rtn_tp64; - struct timespec rtn_tp; + struct timespec64 rtn_tp; int error; if (!kc) return -EINVAL; - error = kc->clock_getres(which_clock, &rtn_tp64); - rtn_tp = timespec64_to_timespec(rtn_tp64); + error = kc->clock_getres(which_clock, &rtn_tp); - if (!error && tp && copy_to_user(tp, &rtn_tp, sizeof (rtn_tp))) + if (!error && tp && put_timespec64(&rtn_tp, tp)) error = -EFAULT; return error; @@ -1131,38 +1125,33 @@ COMPAT_SYSCALL_DEFINE2(clock_settime, clockid_t, which_clock, struct compat_timespec __user *, tp) { const struct k_clock *kc = clockid_to_kclock(which_clock); - struct timespec64 new_tp64; - struct timespec new_tp; + struct timespec64 ts; if (!kc || !kc->clock_set) return -EINVAL; - if (compat_get_timespec(&new_tp, tp)) + if (compat_get_timespec64(&ts, tp)) return -EFAULT; - new_tp64 = timespec_to_timespec64(new_tp); - - return kc->clock_set(which_clock, &new_tp64); + return kc->clock_set(which_clock, &ts); } COMPAT_SYSCALL_DEFINE2(clock_gettime, clockid_t, which_clock, struct compat_timespec __user *, tp) { const struct k_clock *kc = clockid_to_kclock(which_clock); - struct timespec64 kernel_tp64; - struct timespec kernel_tp; - int error; + struct timespec64 ts; + int err; if (!kc) return -EINVAL; - error = kc->clock_get(which_clock, &kernel_tp64); - kernel_tp = timespec64_to_timespec(kernel_tp64); + err = kc->clock_get(which_clock, &ts); - if (!error && compat_put_timespec(&kernel_tp, tp)) - error = -EFAULT; + if (!err && compat_put_timespec64(&ts, tp)) + err = -EFAULT; - return error; + return err; } COMPAT_SYSCALL_DEFINE2(clock_adjtime, clockid_t, which_clock, @@ -1193,21 +1182,19 @@ COMPAT_SYSCALL_DEFINE2(clock_getres, clockid_t, which_clock, struct compat_timespec __user *, tp) { const struct k_clock *kc = clockid_to_kclock(which_clock); - struct timespec64 rtn_tp64; - struct timespec rtn_tp; - int error; + struct timespec64 ts; + int err; if (!kc) return -EINVAL; - error = kc->clock_getres(which_clock, &rtn_tp64); - rtn_tp = timespec64_to_timespec(rtn_tp64); - - if (!error && tp && compat_put_timespec(&rtn_tp, tp)) - error = -EFAULT; + err = kc->clock_getres(which_clock, &ts); + if (!err && tp && compat_put_timespec64(&ts, tp)) + return -EFAULT; - return error; + return err; } + #endif /* -- cgit v1.2.3 From c0edd7c9acd0eaee149ab6cb4441cc71a1af87f0 Mon Sep 17 00:00:00 2001 From: Deepa Dinamani Date: Sat, 24 Jun 2017 11:45:06 -0700 Subject: nanosleep: Use get_timespec64() and put_timespec64() Usage of these apis and their compat versions makes the syscalls: clock_nanosleep and nanosleep and their compat implementations simpler. This is a preparatory patch to isolate data conversions to struct timespec64 at userspace boundaries. This helps contain the changes needed to transition to new y2038 safe types. Signed-off-by: Deepa Dinamani Signed-off-by: Al Viro --- kernel/time/alarmtimer.c | 4 ++-- kernel/time/hrtimer.c | 30 +++++++++++++----------------- kernel/time/posix-cpu-timers.c | 8 ++------ kernel/time/posix-timers.c | 20 ++++++++------------ 4 files changed, 25 insertions(+), 37 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index c991cf212c6d..0b8ff7d257ea 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -712,14 +712,14 @@ static int alarmtimer_do_nsleep(struct alarm *alarm, ktime_t absexp, alarmtimer_freezerset(absexp, type); restart = ¤t->restart_block; if (restart->nanosleep.type != TT_NONE) { - struct timespec rmt; + struct timespec64 rmt; ktime_t rem; rem = ktime_sub(absexp, alarm_bases[type].gettime()); if (rem <= 0) return 0; - rmt = ktime_to_timespec(rem); + rmt = ktime_to_timespec64(rem); return nanosleep_copyout(restart, &rmt); } diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c index 81da124f1115..88f75f92ef36 100644 --- a/kernel/time/hrtimer.c +++ b/kernel/time/hrtimer.c @@ -1440,17 +1440,17 @@ void hrtimer_init_sleeper(struct hrtimer_sleeper *sl, struct task_struct *task) } EXPORT_SYMBOL_GPL(hrtimer_init_sleeper); -int nanosleep_copyout(struct restart_block *restart, struct timespec *ts) +int nanosleep_copyout(struct restart_block *restart, struct timespec64 *ts) { switch(restart->nanosleep.type) { #ifdef CONFIG_COMPAT case TT_COMPAT: - if (compat_put_timespec(ts, restart->nanosleep.compat_rmtp)) + if (compat_put_timespec64(ts, restart->nanosleep.compat_rmtp)) return -EFAULT; break; #endif case TT_NATIVE: - if (copy_to_user(restart->nanosleep.rmtp, ts, sizeof(struct timespec))) + if (put_timespec64(ts, restart->nanosleep.rmtp)) return -EFAULT; break; default: @@ -1485,11 +1485,11 @@ static int __sched do_nanosleep(struct hrtimer_sleeper *t, enum hrtimer_mode mod restart = ¤t->restart_block; if (restart->nanosleep.type != TT_NONE) { ktime_t rem = hrtimer_expires_remaining(&t->timer); - struct timespec rmt; + struct timespec64 rmt; if (rem <= 0) return 0; - rmt = ktime_to_timespec(rem); + rmt = ktime_to_timespec64(rem); return nanosleep_copyout(restart, &rmt); } @@ -1546,19 +1546,17 @@ out: SYSCALL_DEFINE2(nanosleep, struct timespec __user *, rqtp, struct timespec __user *, rmtp) { - struct timespec64 tu64; - struct timespec tu; + struct timespec64 tu; - if (copy_from_user(&tu, rqtp, sizeof(tu))) + if (get_timespec64(&tu, rqtp)) return -EFAULT; - tu64 = timespec_to_timespec64(tu); - if (!timespec64_valid(&tu64)) + if (!timespec64_valid(&tu)) return -EINVAL; current->restart_block.nanosleep.type = rmtp ? TT_NATIVE : TT_NONE; current->restart_block.nanosleep.rmtp = rmtp; - return hrtimer_nanosleep(&tu64, HRTIMER_MODE_REL, CLOCK_MONOTONIC); + return hrtimer_nanosleep(&tu, HRTIMER_MODE_REL, CLOCK_MONOTONIC); } #ifdef CONFIG_COMPAT @@ -1566,19 +1564,17 @@ SYSCALL_DEFINE2(nanosleep, struct timespec __user *, rqtp, COMPAT_SYSCALL_DEFINE2(nanosleep, struct compat_timespec __user *, rqtp, struct compat_timespec __user *, rmtp) { - struct timespec64 tu64; - struct timespec tu; + struct timespec64 tu; - if (compat_get_timespec(&tu, rqtp)) + if (compat_get_timespec64(&tu, rqtp)) return -EFAULT; - tu64 = timespec_to_timespec64(tu); - if (!timespec64_valid(&tu64)) + if (!timespec64_valid(&tu)) return -EINVAL; current->restart_block.nanosleep.type = rmtp ? TT_COMPAT : TT_NONE; current->restart_block.nanosleep.compat_rmtp = rmtp; - return hrtimer_nanosleep(&tu64, HRTIMER_MODE_REL, CLOCK_MONOTONIC); + return hrtimer_nanosleep(&tu, HRTIMER_MODE_REL, CLOCK_MONOTONIC); } #endif diff --git a/kernel/time/posix-cpu-timers.c b/kernel/time/posix-cpu-timers.c index 9df618ee64cf..7323da5950cc 100644 --- a/kernel/time/posix-cpu-timers.c +++ b/kernel/time/posix-cpu-timers.c @@ -1314,12 +1314,8 @@ static int do_cpu_nanosleep(const clockid_t which_clock, int flags, */ restart = ¤t->restart_block; restart->nanosleep.expires = expires; - if (restart->nanosleep.type != TT_NONE) { - struct timespec ts; - - ts = timespec64_to_timespec(it.it_value); - error = nanosleep_copyout(restart, &ts); - } + if (restart->nanosleep.type != TT_NONE) + error = nanosleep_copyout(restart, &it.it_value); } return error; diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c index 39322ae5dd87..4b0fc3b0a1c4 100644 --- a/kernel/time/posix-timers.c +++ b/kernel/time/posix-timers.c @@ -1213,26 +1213,24 @@ SYSCALL_DEFINE4(clock_nanosleep, const clockid_t, which_clock, int, flags, struct timespec __user *, rmtp) { const struct k_clock *kc = clockid_to_kclock(which_clock); - struct timespec64 t64; - struct timespec t; + struct timespec64 t; if (!kc) return -EINVAL; if (!kc->nsleep) return -ENANOSLEEP_NOTSUP; - if (copy_from_user(&t, rqtp, sizeof (struct timespec))) + if (get_timespec64(&t, rqtp)) return -EFAULT; - t64 = timespec_to_timespec64(t); - if (!timespec64_valid(&t64)) + if (!timespec64_valid(&t)) return -EINVAL; if (flags & TIMER_ABSTIME) rmtp = NULL; current->restart_block.nanosleep.type = rmtp ? TT_NATIVE : TT_NONE; current->restart_block.nanosleep.rmtp = rmtp; - return kc->nsleep(which_clock, flags, &t64); + return kc->nsleep(which_clock, flags, &t); } #ifdef CONFIG_COMPAT @@ -1241,26 +1239,24 @@ COMPAT_SYSCALL_DEFINE4(clock_nanosleep, clockid_t, which_clock, int, flags, struct compat_timespec __user *, rmtp) { const struct k_clock *kc = clockid_to_kclock(which_clock); - struct timespec64 t64; - struct timespec t; + struct timespec64 t; if (!kc) return -EINVAL; if (!kc->nsleep) return -ENANOSLEEP_NOTSUP; - if (compat_get_timespec(&t, rqtp)) + if (compat_get_timespec64(&t, rqtp)) return -EFAULT; - t64 = timespec_to_timespec64(t); - if (!timespec64_valid(&t64)) + if (!timespec64_valid(&t)) return -EINVAL; if (flags & TIMER_ABSTIME) rmtp = NULL; current->restart_block.nanosleep.type = rmtp ? TT_COMPAT : TT_NONE; current->restart_block.nanosleep.compat_rmtp = rmtp; - return kc->nsleep(which_clock, flags, &t64); + return kc->nsleep(which_clock, flags, &t); } #endif -- cgit v1.2.3 From 725816e8aabb1c183baa2bc9572ab9a0d26b9ea1 Mon Sep 17 00:00:00 2001 From: Deepa Dinamani Date: Sat, 24 Jun 2017 11:45:08 -0700 Subject: posix_clocks: Use get_itimerspec64() and put_itimerspec64() Usage of these apis and their compat versions makes the syscalls: timer_settime and timer_gettime and their compat implementations simpler. This patch also serves as a preparatory patch for changing syscalls to use new time_t data types to support the y2038 effort by isolating the processing of user pointers through these apis. Signed-off-by: Deepa Dinamani Signed-off-by: Al Viro --- kernel/time/posix-timers.c | 44 ++++++++++++++++---------------------------- 1 file changed, 16 insertions(+), 28 deletions(-) (limited to 'kernel/time') diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c index 4b0fc3b0a1c4..13d6881f908b 100644 --- a/kernel/time/posix-timers.c +++ b/kernel/time/posix-timers.c @@ -739,13 +739,11 @@ static int do_timer_gettime(timer_t timer_id, struct itimerspec64 *setting) SYSCALL_DEFINE2(timer_gettime, timer_t, timer_id, struct itimerspec __user *, setting) { - struct itimerspec64 cur_setting64; + struct itimerspec64 cur_setting; - int ret = do_timer_gettime(timer_id, &cur_setting64); + int ret = do_timer_gettime(timer_id, &cur_setting); if (!ret) { - struct itimerspec cur_setting; - cur_setting = itimerspec64_to_itimerspec(&cur_setting64); - if (copy_to_user(setting, &cur_setting, sizeof (cur_setting))) + if (put_itimerspec64(&cur_setting, setting)) ret = -EFAULT; } return ret; @@ -755,13 +753,11 @@ SYSCALL_DEFINE2(timer_gettime, timer_t, timer_id, COMPAT_SYSCALL_DEFINE2(timer_gettime, timer_t, timer_id, struct compat_itimerspec __user *, setting) { - struct itimerspec64 cur_setting64; + struct itimerspec64 cur_setting; - int ret = do_timer_gettime(timer_id, &cur_setting64); + int ret = do_timer_gettime(timer_id, &cur_setting); if (!ret) { - struct itimerspec cur_setting; - cur_setting = itimerspec64_to_itimerspec(&cur_setting64); - if (put_compat_itimerspec(setting, &cur_setting)) + if (put_compat_itimerspec64(&cur_setting, setting)) ret = -EFAULT; } return ret; @@ -907,23 +903,19 @@ SYSCALL_DEFINE4(timer_settime, timer_t, timer_id, int, flags, const struct itimerspec __user *, new_setting, struct itimerspec __user *, old_setting) { - struct itimerspec64 new_spec64, old_spec64; - struct itimerspec64 *rtn = old_setting ? &old_spec64 : NULL; - struct itimerspec new_spec; + struct itimerspec64 new_spec, old_spec; + struct itimerspec64 *rtn = old_setting ? &old_spec : NULL; int error = 0; if (!new_setting) return -EINVAL; - if (copy_from_user(&new_spec, new_setting, sizeof (new_spec))) + if (get_itimerspec64(&new_spec, new_setting)) return -EFAULT; - new_spec64 = itimerspec_to_itimerspec64(&new_spec); - error = do_timer_settime(timer_id, flags, &new_spec64, rtn); + error = do_timer_settime(timer_id, flags, &new_spec, rtn); if (!error && old_setting) { - struct itimerspec old_spec; - old_spec = itimerspec64_to_itimerspec(&old_spec64); - if (copy_to_user(old_setting, &old_spec, sizeof (old_spec))) + if (put_itimerspec64(&old_spec, old_setting)) error = -EFAULT; } return error; @@ -934,22 +926,18 @@ COMPAT_SYSCALL_DEFINE4(timer_settime, timer_t, timer_id, int, flags, struct compat_itimerspec __user *, new, struct compat_itimerspec __user *, old) { - struct itimerspec64 new_spec64, old_spec64; - struct itimerspec64 *rtn = old ? &old_spec64 : NULL; - struct itimerspec new_spec; + struct itimerspec64 new_spec, old_spec; + struct itimerspec64 *rtn = old ? &old_spec : NULL; int error = 0; if (!new) return -EINVAL; - if (get_compat_itimerspec(&new_spec, new)) + if (get_compat_itimerspec64(&new_spec, new)) return -EFAULT; - new_spec64 = itimerspec_to_itimerspec64(&new_spec); - error = do_timer_settime(timer_id, flags, &new_spec64, rtn); + error = do_timer_settime(timer_id, flags, &new_spec, rtn); if (!error && old) { - struct itimerspec old_spec; - old_spec = itimerspec64_to_itimerspec(&old_spec64); - if (put_compat_itimerspec(old, &old_spec)) + if (put_compat_itimerspec64(&old_spec, old)) error = -EFAULT; } return error; -- cgit v1.2.3