Dear RT folks! I'm pleased to announce the v3.12.6-rt9 patch set.
Changes since v3.12.6-rt8 - ARM's mach-sti is now using rawlock as boot_lock (like the other mach-*) - There was a callpath to rcu_preempt_qs() with interrupts enabled. Tiejun Chen posted a patch to call it with interrupt disabled like we always do. - A patch from Paul E. McKenney to not activate RCU core on NO_HZ_FULL CPUs - A patch from Thomas Gleixner not to raise the timer softirq unconditionally (only if a timer is pending) There is also a patch in the queue from Paul E. McKenney to move RCU processing from softirq into its own thread. After Mike Galbraith reported a few RCU stalls I decided to keep it disabled for now until I have some time to look at it. Known issues: - bcache is disabled. - Brian Silverman reported a BUG (via Debian BTS) where gdb's record command does something nasty and causes a double fault on x86-64 kernel with 32bit userland (the debugged application). 32bit and 64bit setup are not kernels are not affected. The problem is limited is limited to x86. - Sami Pietikäinen reported a crash in __ip_make_skb(). Nicholas Mc Guire is preparing a patch for it. The delta patch against v3.12.6-rt8 is appended below and can be found here: https://www.kernel.org/pub/linux/kernel/projects/rt/3.12/incr/patch-3.12.6-rt8-rt9.patch.xz The RT patch against 3.12.6 can be found here: https://www.kernel.org/pub/linux/kernel/projects/rt/3.12/patch-3.12.6-rt9.patch.xz The split quilt queue is available at: https://www.kernel.org/pub/linux/kernel/projects/rt/3.12/patches-3.12.6-rt9.tar.xz Sebastian diff --git a/arch/arm/mach-sti/platsmp.c b/arch/arm/mach-sti/platsmp.c index dce50d9..c05b764 100644 --- a/arch/arm/mach-sti/platsmp.c +++ b/arch/arm/mach-sti/platsmp.c @@ -35,7 +35,7 @@ static void write_pen_release(int val) outer_clean_range(__pa(&pen_release), __pa(&pen_release + 1)); } -static DEFINE_SPINLOCK(boot_lock); +static DEFINE_RAW_SPINLOCK(boot_lock); void sti_secondary_init(unsigned int cpu) { @@ -50,8 +50,8 @@ void sti_secondary_init(unsigned int cpu) /* * Synchronise with the boot thread. */ - spin_lock(&boot_lock); - spin_unlock(&boot_lock); + raw_spin_lock(&boot_lock); + raw_spin_unlock(&boot_lock); } int sti_boot_secondary(unsigned int cpu, struct task_struct *idle) @@ -62,7 +62,7 @@ int sti_boot_secondary(unsigned int cpu, struct task_struct *idle) * set synchronisation state between this boot processor * and the secondary one */ - spin_lock(&boot_lock); + raw_spin_lock(&boot_lock); /* * The secondary processor is waiting to be released from @@ -93,7 +93,7 @@ int sti_boot_secondary(unsigned int cpu, struct task_struct *idle) * now the secondary core is starting up let it run its * calibrations, then wait for it to finish */ - spin_unlock(&boot_lock); + raw_spin_unlock(&boot_lock); return pen_release != -1 ? -ENOSYS : 0; } diff --git a/include/linux/hrtimer.h b/include/linux/hrtimer.h index 79a7a35..bdbf77db 100644 --- a/include/linux/hrtimer.h +++ b/include/linux/hrtimer.h @@ -461,9 +461,8 @@ extern int schedule_hrtimeout_range_clock(ktime_t *expires, unsigned long delta, const enum hrtimer_mode mode, int clock); extern int schedule_hrtimeout(ktime_t *expires, const enum hrtimer_mode mode); -/* Soft interrupt function to run the hrtimer queues: */ +/* Called from the periodic timer tick */ extern void hrtimer_run_queues(void); -extern void hrtimer_run_pending(void); /* Bootup initialization: */ extern void __init hrtimers_init(void); diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c index c383841..7aa442e 100644 --- a/kernel/hrtimer.c +++ b/kernel/hrtimer.c @@ -1694,30 +1694,6 @@ static void run_hrtimer_softirq(struct softirq_action *h) } /* - * Called from timer softirq every jiffy, expire hrtimers: - * - * For HRT its the fall back code to run the softirq in the timer - * softirq context in case the hrtimer initialization failed or has - * not been done yet. - */ -void hrtimer_run_pending(void) -{ - if (hrtimer_hres_active()) - return; - - /* - * This _is_ ugly: We have to check in the softirq context, - * whether we can switch to highres and / or nohz mode. The - * clocksource switch happens in the timer interrupt with - * xtime_lock held. Notification from there only sets the - * check bit in the tick_oneshot code, otherwise we might - * deadlock vs. xtime_lock. - */ - if (tick_check_oneshot_change(!hrtimer_is_hres_enabled())) - hrtimer_switch_to_hres(); -} - -/* * Called from hardirq context every jiffy */ void hrtimer_run_queues(void) @@ -1730,6 +1706,13 @@ void hrtimer_run_queues(void) if (hrtimer_hres_active()) return; + /* + * Check whether we can switch to highres mode. + */ + if (tick_check_oneshot_change(!hrtimer_is_hres_enabled()) + && hrtimer_switch_to_hres()) + return; + for (index = 0; index < HRTIMER_MAX_CLOCK_BASES; index++) { base = &cpu_base->clock_base[index]; if (!timerqueue_getnext(&base->active)) diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 10365be..f4f61bb 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -204,7 +204,12 @@ static void rcu_preempt_qs(int cpu); void rcu_bh_qs(int cpu) { + unsigned long flags; + + /* Callers to this function, rcu_preempt_qs(), must disable irqs. */ + local_irq_save(flags); rcu_preempt_qs(cpu); + local_irq_restore(flags); } #else void rcu_bh_qs(int cpu) @@ -2674,6 +2679,10 @@ static int __rcu_pending(struct rcu_state *rsp, struct rcu_data *rdp) /* Check for CPU stalls, if enabled. */ check_cpu_stall(rsp, rdp); + /* Is this CPU a NO_HZ_FULL CPU that should ignore RCU? */ + if (rcu_nohz_full_cpu(rsp)) + return 0; + /* Is the RCU core waiting for a quiescent state from this CPU? */ if (rcu_scheduler_fully_active && rdp->qs_pending && !rdp->passed_quiesce) { diff --git a/kernel/rcutree.h b/kernel/rcutree.h index c36d59a..eb4fe67 100644 --- a/kernel/rcutree.h +++ b/kernel/rcutree.h @@ -563,6 +563,7 @@ static void rcu_sysidle_report_gp(struct rcu_state *rsp, int isidle, unsigned long maxj); static void rcu_bind_gp_kthread(void); static void rcu_sysidle_init_percpu_data(struct rcu_dynticks *rdtp); +static bool rcu_nohz_full_cpu(struct rcu_state *rsp); #endif /* #ifndef RCU_TREE_NONCORE */ diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h index 05bcc6f..c1735a1 100644 --- a/kernel/rcutree_plugin.h +++ b/kernel/rcutree_plugin.h @@ -2801,3 +2801,23 @@ static void rcu_sysidle_init_percpu_data(struct rcu_dynticks *rdtp) } #endif /* #else #ifdef CONFIG_NO_HZ_FULL_SYSIDLE */ + +/* + * Is this CPU a NO_HZ_FULL CPU that should ignore RCU so that the + * grace-period kthread will do force_quiescent_state() processing? + * The idea is to avoid waking up RCU core processing on such a + * CPU unless the grace period has extended for too long. + * + * This code relies on the fact that all NO_HZ_FULL CPUs are also + * CONFIG_RCU_NOCB_CPUs. + */ +static bool rcu_nohz_full_cpu(struct rcu_state *rsp) +{ +#ifdef CONFIG_NO_HZ_FULL + if (tick_nohz_full_cpu(smp_processor_id()) && + (!rcu_gp_in_progress(rsp) || + ULONG_CMP_LT(jiffies, ACCESS_ONCE(rsp->gp_start) + HZ))) + return 1; +#endif /* #ifdef CONFIG_NO_HZ_FULL */ + return 0; +} diff --git a/kernel/timer.c b/kernel/timer.c index b06c647..46467be 100644 --- a/kernel/timer.c +++ b/kernel/timer.c @@ -1443,8 +1443,6 @@ static void run_timer_softirq(struct softirq_action *h) irq_work_run(); #endif - hrtimer_run_pending(); - if (time_after_eq(jiffies, base->timer_jiffies)) __run_timers(base); } @@ -1454,8 +1452,27 @@ static void run_timer_softirq(struct softirq_action *h) */ void run_local_timers(void) { + struct tvec_base *base = __this_cpu_read(tvec_bases); + hrtimer_run_queues(); - raise_softirq(TIMER_SOFTIRQ); + /* + * We can access this lockless as we are in the timer + * interrupt. If there are no timers queued, nothing to do in + * the timer softirq. + */ + if (!spin_do_trylock(&base->lock)) { + raise_softirq(TIMER_SOFTIRQ); + return; + } + if (!base->active_timers) + goto out; + + /* Check whether the next pending timer has expired */ + if (time_before_eq(base->next_timer, jiffies)) + raise_softirq(TIMER_SOFTIRQ); +out: + rt_spin_unlock_after_trylock_in_irq(&base->lock); + } #ifdef __ARCH_WANT_SYS_ALARM diff --git a/localversion-rt b/localversion-rt index 700c857..22746d6 100644 --- a/localversion-rt +++ b/localversion-rt @@ -1 +1 @@ --rt8 +-rt9 -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/