[PATCH] posix-timers: CPU clock support for POSIX timers

POSIX requires that when you claim _POSIX_CPUTIME and _POSIX_THREAD_CPUTIME,
not only the clock_* calls but also timer_* calls must support the thread and
process CPU time clocks.  This patch provides that support, building on my
recent additions to support these clocks in the POSIX clock_* interfaces.
This patch will not work without those changes, as well as the patch fixing
the timer lock-siglock deadlock problem.

The apparent pervasive changes to posix-timers.c are simply that some fields
of struct k_itimer have changed name and moved into a union.  This was
appropriate since the data structures required for the existing real-time
timer support and for the new thread/process CPU-time timers are quite
different.

The glibc patches to support CPU time clocks using the new kernel support is
in http://people.redhat.com/roland/glibc/kernel-cpuclocks.patch, and that
includes tests for the timer support (if you build glibc with NPTL).
From: Christoph Lameter <clameter@sgi.com>

  Your patch breaks the mmtimer driver because it used k_itimer values for
  its own purposes.  Here is a fix by defining an additional structure in
  k_itimer (same approach for mmtimer as the cpu timers):
From: Roland McGrath <roland@redhat.com>

Fix bug identified by Alexander Nyberg <alexn@dsv.su.se>

> The problem arises from code touching the union in alloc_posix_timer()
> which makes firing go non-zero. When firing is checked in
> posix_cpu_timer_set() it will be positive causing an infinite loop.
>
> So either the below fix or preferably move the INIT_LIST_HEAD(x) from
> alloc_posix_timer() to somewhere later where it doesn't disturb the other
> union members.

Thanks for finding this problem.  The latter is what I think is the right
solution.  This patch does that, and also removes some superfluous rezeroing.
Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

drivers/char/mmtimer.c

@@ -418,19 +418,19 @@ static int inline reschedule_periodic_timer(mmtimer_t *x)
 	int n;
 	struct k_itimer *t = x->timer;
 
-	t->it_timer.magic = x->i;
+	t->it.mmtimer.clock = x->i;
 	t->it_overrun--;
 
 	n = 0;
 	do {
 
-		t->it_timer.expires += t->it_incr << n;
+		t->it.mmtimer.expires += t->it.mmtimer.incr << n;
 		t->it_overrun += 1 << n;
 		n++;
 		if (n > 20)
 			return 1;
 
-	} while (mmtimer_setup(x->i, t->it_timer.expires));
+	} while (mmtimer_setup(x->i, t->it.mmtimer.expires));
 
 	return 0;
 }
@@ -466,7 +466,7 @@ mmtimer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 		spin_lock(&base[i].lock);
 		if (base[i].cpu == smp_processor_id()) {
 			if (base[i].timer)
-				expires = base[i].timer->it_timer.expires;
+				expires = base[i].timer->it.mmtimer.expires;
 			/* expires test won't work with shared irqs */
 			if ((mmtimer_int_pending(i) > 0) ||
 				(expires && (expires < rtc_time()))) {
@@ -503,7 +503,7 @@ void mmtimer_tasklet(unsigned long data) {
 
 		t->it_overrun++;
 	}
-	if(t->it_incr) {
+	if(t->it.mmtimer.incr) {
 		/* Periodic timer */
 		if (reschedule_periodic_timer(x)) {
 			printk(KERN_WARNING "mmtimer: unable to reschedule\n");
@@ -511,7 +511,7 @@ void mmtimer_tasklet(unsigned long data) {
 		}
 	} else {
 		/* Ensure we don't false trigger in mmtimer_interrupt */
-		t->it_timer.expires = 0;
+		t->it.mmtimer.expires = 0;
 	}
 	t->it_overrun_last = t->it_overrun;
 out:
@@ -522,7 +522,7 @@ void mmtimer_tasklet(unsigned long data) {
 static int sgi_timer_create(struct k_itimer *timer)
 {
 	/* Insure that a newly created timer is off */
-	timer->it_timer.magic = TIMER_OFF;
+	timer->it.mmtimer.clock = TIMER_OFF;
 	return 0;
 }
 
@@ -533,8 +533,8 @@ static int sgi_timer_create(struct k_itimer *timer)
  */
 static int sgi_timer_del(struct k_itimer *timr)
 {
-	int i = timr->it_timer.magic;
-	cnodeid_t nodeid = timr->it_timer.data;
+	int i = timr->it.mmtimer.clock;
+	cnodeid_t nodeid = timr->it.mmtimer.node;
 	mmtimer_t *t = timers + nodeid * NUM_COMPARATORS +i;
 	unsigned long irqflags;
 
@@ -542,8 +542,8 @@ static int sgi_timer_del(struct k_itimer *timr)
 		spin_lock_irqsave(&t->lock, irqflags);
 		mmtimer_disable_int(cnodeid_to_nasid(nodeid),i);
 		t->timer = NULL;
-		timr->it_timer.magic = TIMER_OFF;
-		timr->it_timer.expires = 0;
+		timr->it.mmtimer.clock = TIMER_OFF;
+		timr->it.mmtimer.expires = 0;
 		spin_unlock_irqrestore(&t->lock, irqflags);
 	}
 	return 0;
@@ -556,7 +556,7 @@ static int sgi_timer_del(struct k_itimer *timr)
 static void sgi_timer_get(struct k_itimer *timr, struct itimerspec *cur_setting)
 {
 
-	if (timr->it_timer.magic == TIMER_OFF) {
+	if (timr->it.mmtimer.clock == TIMER_OFF) {
 		cur_setting->it_interval.tv_nsec = 0;
 		cur_setting->it_interval.tv_sec = 0;
 		cur_setting->it_value.tv_nsec = 0;
@@ -564,8 +564,8 @@ static void sgi_timer_get(struct k_itimer *timr, struct itimerspec *cur_setting)
 		return;
 	}
 
-	ns_to_timespec(cur_setting->it_interval, timr->it_incr * sgi_clock_period);
-	ns_to_timespec(cur_setting->it_value, (timr->it_timer.expires - rtc_time())* sgi_clock_period);
+	ns_to_timespec(cur_setting->it_interval, timr->it.mmtimer.incr * sgi_clock_period);
+	ns_to_timespec(cur_setting->it_value, (timr->it.mmtimer.expires - rtc_time())* sgi_clock_period);
 	return;
 }
 
@@ -638,19 +638,19 @@ static int sgi_timer_set(struct k_itimer *timr, int flags,
 	base[i].timer = timr;
 	base[i].cpu = smp_processor_id();
 
-	timr->it_timer.magic = i;
-	timr->it_timer.data = nodeid;
-	timr->it_incr = period;
-	timr->it_timer.expires = when;
+	timr->it.mmtimer.clock = i;
+	timr->it.mmtimer.node = nodeid;
+	timr->it.mmtimer.incr = period;
+	timr->it.mmtimer.expires = when;
 
 	if (period == 0) {
 		if (mmtimer_setup(i, when)) {
 			mmtimer_disable_int(-1, i);
 			posix_timer_event(timr, 0);
-			timr->it_timer.expires = 0;
+			timr->it.mmtimer.expires = 0;
 		}
 	} else {
-		timr->it_timer.expires -= period;
+		timr->it.mmtimer.expires -= period;
 		if (reschedule_periodic_timer(base+i))
 			err = -EINVAL;
 	}

include/linux/init_task.h

@@ -51,6 +51,7 @@
 		.list = LIST_HEAD_INIT(sig.shared_pending.list),	\
 		.signal =  {{0}}}, \
 	.posix_timers	 = LIST_HEAD_INIT(sig.posix_timers),		\
+	.cpu_timers	= INIT_CPU_TIMERS(sig.cpu_timers),		\
 	.rlim		= INIT_RLIMITS,					\
 }
 
@@ -112,8 +113,16 @@ extern struct group_info init_groups;
 	.proc_lock	= SPIN_LOCK_UNLOCKED,				\
 	.switch_lock	= SPIN_LOCK_UNLOCKED,				\
 	.journal_info	= NULL,						\
+	.cpu_timers	= INIT_CPU_TIMERS(tsk.cpu_timers),		\
 }
 
 
+#define INIT_CPU_TIMERS(cpu_timers)					\
+{									\
+	LIST_HEAD_INIT(cpu_timers[0]),					\
+	LIST_HEAD_INIT(cpu_timers[1]),					\
+	LIST_HEAD_INIT(cpu_timers[2]),					\
+}
+
 
 #endif

include/linux/posix-timers.h

@@ -3,8 +3,21 @@
 
 #include <linux/spinlock.h>
 #include <linux/list.h>
+#include <linux/sched.h>
 
-#define CPUCLOCK_PID(clock)	((pid_t) ~((clock) >> 3))
+union cpu_time_count {
+	cputime_t cpu;
+	unsigned long long sched;
+};
+
+struct cpu_timer_list {
+	struct list_head entry;
+	union cpu_time_count expires, incr;
+	struct task_struct *task;
+	int firing;
+};
+
+#define CPUCLOCK_PID(clock)		((pid_t) ~((clock) >> 3))
 #define CPUCLOCK_PERTHREAD(clock) \
 	(((clock) & (clockid_t) CPUCLOCK_PERTHREAD_MASK) != 0)
 #define CPUCLOCK_PID_MASK	7
@@ -30,15 +43,27 @@ struct k_itimer {
 	int it_overrun;			/* overrun on pending signal  */
 	int it_overrun_last;		/* overrun on last delivered signal */
 	int it_requeue_pending;         /* waiting to requeue this timer */
+#define REQUEUE_PENDING 1
 	int it_sigev_notify;		/* notify word of sigevent struct */
 	int it_sigev_signo;		/* signo word of sigevent struct */
 	sigval_t it_sigev_value;	/* value word of sigevent struct */
-	unsigned long it_incr;		/* interval specified in jiffies */
 	struct task_struct *it_process;	/* process to send signal to */
-	struct timer_list it_timer;
 	struct sigqueue *sigq;		/* signal queue entry. */
-	struct list_head abs_timer_entry; /* clock abs_timer_list */
-	struct timespec wall_to_prev;   /* wall_to_monotonic used when set */
+	union {
+		struct {
+			struct timer_list timer;
+			struct list_head abs_timer_entry; /* clock abs_timer_list */
+			struct timespec wall_to_prev;   /* wall_to_monotonic used when set */
+			unsigned long incr; /* interval in jiffies */
+		} real;
+		struct cpu_timer_list cpu;
+		struct {
+			unsigned int clock;
+			unsigned int node;
+			unsigned long incr;
+			unsigned long expires;
+		} mmtimer;
+	} it;
 };
 
 struct k_clock_abs {
@@ -57,6 +82,7 @@ struct k_clock {
 			  struct itimerspec * new_setting,
 			  struct itimerspec * old_setting);
 	int (*timer_del) (struct k_itimer * timr);
+#define TIMER_RETRY 1
 	void (*timer_get) (struct k_itimer * timr,
 			   struct itimerspec * cur_setting);
 };
@@ -82,10 +108,11 @@ struct now_struct {
 #define posix_bump_timer(timr, now)					\
          do {								\
               long delta, orun;						\
-	      delta = now.jiffies - (timr)->it_timer.expires;		\
+	      delta = now.jiffies - (timr)->it.real.timer.expires;	\
               if (delta >= 0) {						\
-	           orun = 1 + (delta / (timr)->it_incr);		\
-	          (timr)->it_timer.expires += orun * (timr)->it_incr;	\
+	           orun = 1 + (delta / (timr)->it.real.incr);		\
+	          (timr)->it.real.timer.expires +=			\
+			 orun * (timr)->it.real.incr;			\
                   (timr)->it_overrun += orun;				\
               }								\
             }while (0)
@@ -95,12 +122,16 @@ int posix_cpu_clock_get(clockid_t which_clock, struct timespec *);
 int posix_cpu_clock_set(clockid_t which_clock, const struct timespec *tp);
 int posix_cpu_timer_create(struct k_itimer *);
 int posix_cpu_nsleep(clockid_t, int, struct timespec *);
-#define posix_cpu_timer_create do_posix_clock_notimer_create
-#define posix_cpu_nsleep do_posix_clock_nonanosleep
 int posix_cpu_timer_set(struct k_itimer *, int,
 			struct itimerspec *, struct itimerspec *);
 int posix_cpu_timer_del(struct k_itimer *);
 void posix_cpu_timer_get(struct k_itimer *, struct itimerspec *);
 
+void posix_cpu_timer_schedule(struct k_itimer *);
+
+void run_posix_cpu_timers(struct task_struct *);
+void posix_cpu_timers_exit(struct task_struct *);
+void posix_cpu_timers_exit_group(struct task_struct *);
+
 
 #endif

include/linux/sched.h

@@ -338,6 +338,8 @@ struct signal_struct {
 	 * have no need to disable irqs.
 	 */
 	struct rlimit rlim[RLIM_NLIMITS];
+
+	struct list_head cpu_timers[3];
 };
 
 /*
@@ -612,6 +614,11 @@ struct task_struct {
 	struct timespec start_time;
 /* mm fault and swap info: this can arguably be seen as either mm-specific or thread-specific */
 	unsigned long min_flt, maj_flt;
+
+  	cputime_t it_prof_expires, it_virt_expires;
+	unsigned long long it_sched_expires;
+	struct list_head cpu_timers[3];
+
 /* process credentials */
 	uid_t uid,euid,suid,fsuid;
 	gid_t gid,egid,sgid,fsgid;

kernel/exit.c

@@ -759,6 +759,9 @@ static void exit_notify(struct task_struct *tsk)
 	 */
 	tsk->it_virt_value = cputime_zero;
 	tsk->it_prof_value = cputime_zero;
+ 	tsk->it_virt_expires = cputime_zero;
+ 	tsk->it_prof_expires = cputime_zero;
+	tsk->it_sched_expires = 0;
 
 	write_unlock_irq(&tasklist_lock);
 

kernel/fork.c

@@ -750,6 +750,9 @@ static inline int copy_signal(unsigned long clone_flags, struct task_struct * ts
 	sig->nvcsw = sig->nivcsw = sig->cnvcsw = sig->cnivcsw = 0;
 	sig->min_flt = sig->maj_flt = sig->cmin_flt = sig->cmaj_flt = 0;
 	sig->sched_time = 0;
+	INIT_LIST_HEAD(&sig->cpu_timers[0]);
+	INIT_LIST_HEAD(&sig->cpu_timers[1]);
+	INIT_LIST_HEAD(&sig->cpu_timers[2]);
 
 	task_lock(current->group_leader);
 	memcpy(sig->rlim, current->signal->rlim, sizeof sig->rlim);
@@ -885,6 +888,13 @@ static task_t *copy_process(unsigned long clone_flags,
 	p->syscw = 0;		/* I/O counter: write syscalls */
 	acct_clear_integrals(p);
 
+ 	p->it_virt_expires = cputime_zero;
+	p->it_prof_expires = cputime_zero;
+ 	p->it_sched_expires = 0;
+ 	INIT_LIST_HEAD(&p->cpu_timers[0]);
+ 	INIT_LIST_HEAD(&p->cpu_timers[1]);
+ 	INIT_LIST_HEAD(&p->cpu_timers[2]);
+
 	p->lock_depth = -1;		/* -1 = no lock */
 	do_posix_clock_monotonic_gettime(&p->start_time);
 	p->security = NULL;
@@ -1017,6 +1027,16 @@ static task_t *copy_process(unsigned long clone_flags,
 			set_tsk_thread_flag(p, TIF_SIGPENDING);
 		}
 
+		if (!list_empty(&current->signal->cpu_timers[0]) ||
+		    !list_empty(&current->signal->cpu_timers[1]) ||
+		    !list_empty(&current->signal->cpu_timers[2])) {
+			/*
+			 * Have child wake up on its first tick to check
+			 * for process CPU timers.
+			 */
+			p->it_prof_expires = jiffies_to_cputime(1);
+		}
+
 		spin_unlock(&current->sighand->siglock);
 	}
 

kernel/signal.c

@@ -22,6 +22,7 @@
 #include <linux/security.h>
 #include <linux/syscalls.h>
 #include <linux/ptrace.h>
+#include <linux/posix-timers.h>
 #include <asm/param.h>
 #include <asm/uaccess.h>
 #include <asm/unistd.h>
@@ -347,7 +348,9 @@ void __exit_signal(struct task_struct *tsk)
 	if (!atomic_read(&sig->count))
 		BUG();
 	spin_lock(&sighand->siglock);
+	posix_cpu_timers_exit(tsk);
 	if (atomic_dec_and_test(&sig->count)) {
+		posix_cpu_timers_exit_group(tsk);
 		if (tsk == sig->curr_target)
 			sig->curr_target = next_thread(tsk);
 		tsk->signal = NULL;

kernel/timer.c

@@ -30,6 +30,7 @@
 #include <linux/thread_info.h>
 #include <linux/time.h>
 #include <linux/jiffies.h>
+#include <linux/posix-timers.h>
 #include <linux/cpu.h>
 #include <linux/syscalls.h>
 
@@ -825,6 +826,7 @@ void update_process_times(int user_tick)
 	if (rcu_pending(cpu))
 		rcu_check_callbacks(cpu, user_tick);
 	scheduler_tick();
+ 	run_posix_cpu_timers(p);
 }
 
 /*