Merge commit 'v3.1-rc7' into perf/core

Merge reason: Pick up the latest upstream fixes.

Signed-off-by: Ingo Molnar <mingo@elte.hu>

kernel/events/core.c

@@ -400,14 +400,54 @@ void perf_cgroup_switch(struct task_struct *task, int mode)
 	local_irq_restore(flags);
 }
 
-static inline void perf_cgroup_sched_out(struct task_struct *task)
+static inline void perf_cgroup_sched_out(struct task_struct *task,
+					 struct task_struct *next)
 {
-	perf_cgroup_switch(task, PERF_CGROUP_SWOUT);
+	struct perf_cgroup *cgrp1;
+	struct perf_cgroup *cgrp2 = NULL;
+
+	/*
+	 * we come here when we know perf_cgroup_events > 0
+	 */
+	cgrp1 = perf_cgroup_from_task(task);
+
+	/*
+	 * next is NULL when called from perf_event_enable_on_exec()
+	 * that will systematically cause a cgroup_switch()
+	 */
+	if (next)
+		cgrp2 = perf_cgroup_from_task(next);
+
+	/*
+	 * only schedule out current cgroup events if we know
+	 * that we are switching to a different cgroup. Otherwise,
+	 * do no touch the cgroup events.
+	 */
+	if (cgrp1 != cgrp2)
+		perf_cgroup_switch(task, PERF_CGROUP_SWOUT);
 }
 
-static inline void perf_cgroup_sched_in(struct task_struct *task)
+static inline void perf_cgroup_sched_in(struct task_struct *prev,
+					struct task_struct *task)
 {
-	perf_cgroup_switch(task, PERF_CGROUP_SWIN);
+	struct perf_cgroup *cgrp1;
+	struct perf_cgroup *cgrp2 = NULL;
+
+	/*
+	 * we come here when we know perf_cgroup_events > 0
+	 */
+	cgrp1 = perf_cgroup_from_task(task);
+
+	/* prev can never be NULL */
+	cgrp2 = perf_cgroup_from_task(prev);
+
+	/*
+	 * only need to schedule in cgroup events if we are changing
+	 * cgroup during ctxsw. Cgroup events were not scheduled
+	 * out of ctxsw out if that was not the case.
+	 */
+	if (cgrp1 != cgrp2)
+		perf_cgroup_switch(task, PERF_CGROUP_SWIN);
 }
 
 static inline int perf_cgroup_connect(int fd, struct perf_event *event,
@@ -519,11 +559,13 @@ static inline void update_cgrp_time_from_cpuctx(struct perf_cpu_context *cpuctx)
 {
 }
 
-static inline void perf_cgroup_sched_out(struct task_struct *task)
+static inline void perf_cgroup_sched_out(struct task_struct *task,
+					 struct task_struct *next)
 {
 }
 
-static inline void perf_cgroup_sched_in(struct task_struct *task)
+static inline void perf_cgroup_sched_in(struct task_struct *prev,
+					struct task_struct *task)
 {
 }
 
@@ -1989,7 +2031,7 @@ void __perf_event_task_sched_out(struct task_struct *task,
 	 * cgroup event are system-wide mode only
 	 */
 	if (atomic_read(&__get_cpu_var(perf_cgroup_events)))
-		perf_cgroup_sched_out(task);
+		perf_cgroup_sched_out(task, next);
 }
 
 static void task_ctx_sched_out(struct perf_event_context *ctx)
@@ -2154,7 +2196,8 @@ static void perf_event_context_sched_in(struct perf_event_context *ctx,
  * accessing the event control register. If a NMI hits, then it will
  * keep the event running.
  */
-void __perf_event_task_sched_in(struct task_struct *task)
+void __perf_event_task_sched_in(struct task_struct *prev,
+				struct task_struct *task)
 {
 	struct perf_event_context *ctx;
 	int ctxn;
@@ -2172,7 +2215,7 @@ void __perf_event_task_sched_in(struct task_struct *task)
 	 * cgroup event are system-wide mode only
 	 */
 	if (atomic_read(&__get_cpu_var(perf_cgroup_events)))
-		perf_cgroup_sched_in(task);
+		perf_cgroup_sched_in(prev, task);
 }
 
 static u64 perf_calculate_period(struct perf_event *event, u64 nsec, u64 count)
@@ -2428,7 +2471,7 @@ static void perf_event_enable_on_exec(struct perf_event_context *ctx)
 	 * ctxswin cgroup events which are already scheduled
 	 * in.
 	 */
-	perf_cgroup_sched_out(current);
+	perf_cgroup_sched_out(current, NULL);
 
 	raw_spin_lock(&ctx->lock);
 	task_ctx_sched_out(ctx);
@@ -3354,8 +3397,8 @@ static int perf_event_index(struct perf_event *event)
 }
 
 static void calc_timer_values(struct perf_event *event,
-				u64 *running,
-				u64 *enabled)
+				u64 *enabled,
+				u64 *running)
 {
 	u64 now, ctx_time;
 

kernel/irq/chip.c

@@ -178,7 +178,7 @@ void irq_shutdown(struct irq_desc *desc)
 	desc->depth = 1;
 	if (desc->irq_data.chip->irq_shutdown)
 		desc->irq_data.chip->irq_shutdown(&desc->irq_data);
-	if (desc->irq_data.chip->irq_disable)
+	else if (desc->irq_data.chip->irq_disable)
 		desc->irq_data.chip->irq_disable(&desc->irq_data);
 	else
 		desc->irq_data.chip->irq_mask(&desc->irq_data);

kernel/irq/irqdesc.c

@@ -344,6 +344,7 @@ EXPORT_SYMBOL_GPL(irq_free_descs);
  * @from:	Start the search from this irq number
  * @cnt:	Number of consecutive irqs to allocate.
  * @node:	Preferred node on which the irq descriptor should be allocated
+ * @owner:	Owning module (can be NULL)
  *
  * Returns the first irq number or error code
  */

kernel/irq/manage.c

@@ -1331,7 +1331,6 @@ int request_threaded_irq(unsigned int irq, irq_handler_t handler,
 		if (!thread_fn)
 			return -EINVAL;
 		handler = irq_default_primary_handler;
-		irqflags |= IRQF_ONESHOT;
 	}
 
 	action = kzalloc(sizeof(struct irqaction), GFP_KERNEL);

kernel/power/Kconfig

@@ -231,3 +231,7 @@ config PM_CLK
 config PM_GENERIC_DOMAINS
 	bool
 	depends on PM
+
+config PM_GENERIC_DOMAINS_RUNTIME
+	def_bool y
+	depends on PM_RUNTIME && PM_GENERIC_DOMAINS

kernel/printk.c

@@ -1604,7 +1604,7 @@ static int __init printk_late_init(void)
 	struct console *con;
 
 	for_each_console(con) {
-		if (con->flags & CON_BOOT) {
+		if (!keep_bootcon && con->flags & CON_BOOT) {
 			printk(KERN_INFO "turn off boot console %s%d\n",
 				con->name, con->index);
 			unregister_console(con);

kernel/sched.c

@@ -3065,7 +3065,7 @@ static void finish_task_switch(struct rq *rq, struct task_struct *prev)
 #ifdef __ARCH_WANT_INTERRUPTS_ON_CTXSW
 	local_irq_disable();
 #endif /* __ARCH_WANT_INTERRUPTS_ON_CTXSW */
-	perf_event_task_sched_in(current);
+	perf_event_task_sched_in(prev, current);
 #ifdef __ARCH_WANT_INTERRUPTS_ON_CTXSW
 	local_irq_enable();
 #endif /* __ARCH_WANT_INTERRUPTS_ON_CTXSW */
@@ -4279,9 +4279,9 @@ pick_next_task(struct rq *rq)
 }
 
 /*
- * schedule() is the main scheduler function.
+ * __schedule() is the main scheduler function.
  */
-asmlinkage void __sched schedule(void)
+static void __sched __schedule(void)
 {
 	struct task_struct *prev, *next;
 	unsigned long *switch_count;
@@ -4322,16 +4322,6 @@ need_resched:
 				if (to_wakeup)
 					try_to_wake_up_local(to_wakeup);
 			}
-
-			/*
-			 * If we are going to sleep and we have plugged IO
-			 * queued, make sure to submit it to avoid deadlocks.
-			 */
-			if (blk_needs_flush_plug(prev)) {
-				raw_spin_unlock(&rq->lock);
-				blk_schedule_flush_plug(prev);
-				raw_spin_lock(&rq->lock);
-			}
 		}
 		switch_count = &prev->nvcsw;
 	}
@@ -4369,6 +4359,26 @@ need_resched:
 	if (need_resched())
 		goto need_resched;
 }
+
+static inline void sched_submit_work(struct task_struct *tsk)
+{
+	if (!tsk->state)
+		return;
+	/*
+	 * If we are going to sleep and we have plugged IO queued,
+	 * make sure to submit it to avoid deadlocks.
+	 */
+	if (blk_needs_flush_plug(tsk))
+		blk_schedule_flush_plug(tsk);
+}
+
+asmlinkage void schedule(void)
+{
+	struct task_struct *tsk = current;
+
+	sched_submit_work(tsk);
+	__schedule();
+}
 EXPORT_SYMBOL(schedule);
 
 #ifdef CONFIG_MUTEX_SPIN_ON_OWNER
@@ -4435,7 +4445,7 @@ asmlinkage void __sched notrace preempt_schedule(void)
 
 	do {
 		add_preempt_count_notrace(PREEMPT_ACTIVE);
-		schedule();
+		__schedule();
 		sub_preempt_count_notrace(PREEMPT_ACTIVE);
 
 		/*
@@ -4463,7 +4473,7 @@ asmlinkage void __sched preempt_schedule_irq(void)
 	do {
 		add_preempt_count(PREEMPT_ACTIVE);
 		local_irq_enable();
-		schedule();
+		__schedule();
 		local_irq_disable();
 		sub_preempt_count(PREEMPT_ACTIVE);
 
@@ -5588,7 +5598,7 @@ static inline int should_resched(void)
 static void __cond_resched(void)
 {
 	add_preempt_count(PREEMPT_ACTIVE);
-	schedule();
+	__schedule();
 	sub_preempt_count(PREEMPT_ACTIVE);
 }
 
@@ -7443,6 +7453,7 @@ static void __sdt_free(const struct cpumask *cpu_map)
 			struct sched_domain *sd = *per_cpu_ptr(sdd->sd, j);
 			if (sd && (sd->flags & SD_OVERLAP))
 				free_sched_groups(sd->groups, 0);
+			kfree(*per_cpu_ptr(sdd->sd, j));
 			kfree(*per_cpu_ptr(sdd->sg, j));
 			kfree(*per_cpu_ptr(sdd->sgp, j));
 		}

kernel/sys.c

@@ -37,6 +37,8 @@
 #include <linux/fs_struct.h>
 #include <linux/gfp.h>
 #include <linux/syscore_ops.h>
+#include <linux/version.h>
+#include <linux/ctype.h>
 
 #include <linux/compat.h>
 #include <linux/syscalls.h>
@@ -44,6 +46,8 @@
 #include <linux/user_namespace.h>
 
 #include <linux/kmsg_dump.h>
+/* Move somewhere else to avoid recompiling? */
+#include <generated/utsrelease.h>
 
 #include <asm/uaccess.h>
 #include <asm/io.h>
@@ -1161,6 +1165,34 @@ DECLARE_RWSEM(uts_sem);
 #define override_architecture(name)	0
 #endif
 
+/*
+ * Work around broken programs that cannot handle "Linux 3.0".
+ * Instead we map 3.x to 2.6.40+x, so e.g. 3.0 would be 2.6.40
+ */
+static int override_release(char __user *release, int len)
+{
+	int ret = 0;
+	char buf[len];
+
+	if (current->personality & UNAME26) {
+		char *rest = UTS_RELEASE;
+		int ndots = 0;
+		unsigned v;
+
+		while (*rest) {
+			if (*rest == '.' && ++ndots >= 3)
+				break;
+			if (!isdigit(*rest) && *rest != '.')
+				break;
+			rest++;
+		}
+		v = ((LINUX_VERSION_CODE >> 8) & 0xff) + 40;
+		snprintf(buf, len, "2.6.%u%s", v, rest);
+		ret = copy_to_user(release, buf, len);
+	}
+	return ret;
+}
+
 SYSCALL_DEFINE1(newuname, struct new_utsname __user *, name)
 {
 	int errno = 0;
@@ -1170,6 +1202,8 @@ SYSCALL_DEFINE1(newuname, struct new_utsname __user *, name)
 		errno = -EFAULT;
 	up_read(&uts_sem);
 
+	if (!errno && override_release(name->release, sizeof(name->release)))
+		errno = -EFAULT;
 	if (!errno && override_architecture(name))
 		errno = -EFAULT;
 	return errno;
@@ -1191,6 +1225,8 @@ SYSCALL_DEFINE1(uname, struct old_utsname __user *, name)
 		error = -EFAULT;
 	up_read(&uts_sem);
 
+	if (!error && override_release(name->release, sizeof(name->release)))
+		error = -EFAULT;
 	if (!error && override_architecture(name))
 		error = -EFAULT;
 	return error;
@@ -1225,6 +1261,8 @@ SYSCALL_DEFINE1(olduname, struct oldold_utsname __user *, name)
 
 	if (!error && override_architecture(name))
 		error = -EFAULT;
+	if (!error && override_release(name->release, sizeof(name->release)))
+		error = -EFAULT;
 	return error ? -EFAULT : 0;
 }
 #endif

kernel/sys_ni.c

@@ -16,7 +16,6 @@ asmlinkage long sys_ni_syscall(void)
 	return -ENOSYS;
 }
 
-cond_syscall(sys_nfsservctl);
 cond_syscall(sys_quotactl);
 cond_syscall(sys32_quotactl);
 cond_syscall(sys_acct);

kernel/sysctl_binary.c

@@ -1,6 +1,6 @@
 #include <linux/stat.h>
 #include <linux/sysctl.h>
-#include "../fs/xfs/linux-2.6/xfs_sysctl.h"
+#include "../fs/xfs/xfs_sysctl.h"
 #include <linux/sunrpc/debug.h>
 #include <linux/string.h>
 #include <net/ip_vs.h>

kernel/sysctl_check.c

@@ -1,6 +1,6 @@
 #include <linux/stat.h>
 #include <linux/sysctl.h>
-#include "../fs/xfs/linux-2.6/xfs_sysctl.h"
+#include "../fs/xfs/xfs_sysctl.h"
 #include <linux/sunrpc/debug.h>
 #include <linux/string.h>
 #include <net/ip_vs.h>

kernel/taskstats.c

@@ -655,6 +655,7 @@ static struct genl_ops taskstats_ops = {
 	.cmd		= TASKSTATS_CMD_GET,
 	.doit		= taskstats_user_cmd,
 	.policy		= taskstats_cmd_get_policy,
+	.flags		= GENL_ADMIN_PERM,
 };
 
 static struct genl_ops cgroupstats_ops = {

kernel/time/alarmtimer.c

@@ -441,6 +441,8 @@ static int alarm_timer_create(struct k_itimer *new_timer)
 static void alarm_timer_get(struct k_itimer *timr,
 				struct itimerspec *cur_setting)
 {
+	memset(cur_setting, 0, sizeof(struct itimerspec));
+
 	cur_setting->it_interval =
 			ktime_to_timespec(timr->it.alarmtimer.period);
 	cur_setting->it_value =
@@ -479,11 +481,17 @@ static int alarm_timer_set(struct k_itimer *timr, int flags,
 	if (!rtcdev)
 		return -ENOTSUPP;
 
-	/* Save old values */
-	old_setting->it_interval =
-			ktime_to_timespec(timr->it.alarmtimer.period);
-	old_setting->it_value =
-			ktime_to_timespec(timr->it.alarmtimer.node.expires);
+	/*
+	 * XXX HACK! Currently we can DOS a system if the interval
+	 * period on alarmtimers is too small. Cap the interval here
+	 * to 100us and solve this properly in a future patch! -jstultz
+	 */
+	if ((new_setting->it_interval.tv_sec == 0) &&
+			(new_setting->it_interval.tv_nsec < 100000))
+		new_setting->it_interval.tv_nsec = 100000;
+
+	if (old_setting)
+		alarm_timer_get(timr, old_setting);
 
 	/* If the timer was already set, cancel it */
 	alarm_cancel(&timr->it.alarmtimer);

kernel/trace/blktrace.c

@@ -206,6 +206,8 @@ static void __blk_add_trace(struct blk_trace *bt, sector_t sector, int bytes,
 	what |= MASK_TC_BIT(rw, RAHEAD);
 	what |= MASK_TC_BIT(rw, META);
 	what |= MASK_TC_BIT(rw, DISCARD);
+	what |= MASK_TC_BIT(rw, FLUSH);
+	what |= MASK_TC_BIT(rw, FUA);
 
 	pid = tsk->pid;
 	if (act_log_check(bt, what, sector, pid))
@@ -1054,6 +1056,9 @@ static void fill_rwbs(char *rwbs, const struct blk_io_trace *t)
 		goto out;
 	}
 
+	if (tc & BLK_TC_FLUSH)
+		rwbs[i++] = 'F';
+
 	if (tc & BLK_TC_DISCARD)
 		rwbs[i++] = 'D';
 	else if (tc & BLK_TC_WRITE)
@@ -1063,10 +1068,10 @@ static void fill_rwbs(char *rwbs, const struct blk_io_trace *t)
 	else
 		rwbs[i++] = 'N';
 
+	if (tc & BLK_TC_FUA)
+		rwbs[i++] = 'F';
 	if (tc & BLK_TC_AHEAD)
 		rwbs[i++] = 'A';
-	if (tc & BLK_TC_BARRIER)
-		rwbs[i++] = 'B';
 	if (tc & BLK_TC_SYNC)
 		rwbs[i++] = 'S';
 	if (tc & BLK_TC_META)
@@ -1132,7 +1137,7 @@ typedef int (blk_log_action_t) (struct trace_iterator *iter, const char *act);
 
 static int blk_log_action_classic(struct trace_iterator *iter, const char *act)
 {
-	char rwbs[6];
+	char rwbs[RWBS_LEN];
 	unsigned long long ts  = iter->ts;
 	unsigned long nsec_rem = do_div(ts, NSEC_PER_SEC);
 	unsigned secs	       = (unsigned long)ts;
@@ -1148,7 +1153,7 @@ static int blk_log_action_classic(struct trace_iterator *iter, const char *act)
 
 static int blk_log_action(struct trace_iterator *iter, const char *act)
 {
-	char rwbs[6];
+	char rwbs[RWBS_LEN];
 	const struct blk_io_trace *t = te_blk_io_trace(iter->ent);
 
 	fill_rwbs(rwbs, t);
@@ -1561,7 +1566,7 @@ static const struct {
 } mask_maps[] = {
 	{ BLK_TC_READ,		"read"		},
 	{ BLK_TC_WRITE,		"write"		},
-	{ BLK_TC_BARRIER,	"barrier"	},
+	{ BLK_TC_FLUSH,		"flush"		},
 	{ BLK_TC_SYNC,		"sync"		},
 	{ BLK_TC_QUEUE,		"queue"		},
 	{ BLK_TC_REQUEUE,	"requeue"	},
@@ -1573,6 +1578,7 @@ static const struct {
 	{ BLK_TC_META,		"meta"		},
 	{ BLK_TC_DISCARD,	"discard"	},
 	{ BLK_TC_DRV_DATA,	"drv_data"	},
+	{ BLK_TC_FUA,		"fua"		},
 };
 
 static int blk_trace_str2mask(const char *str)
@@ -1788,6 +1794,9 @@ void blk_fill_rwbs(char *rwbs, u32 rw, int bytes)
 {
 	int i = 0;
 
+	if (rw & REQ_FLUSH)
+		rwbs[i++] = 'F';
+
 	if (rw & WRITE)
 		rwbs[i++] = 'W';
 	else if (rw & REQ_DISCARD)
@@ -1797,6 +1806,8 @@ void blk_fill_rwbs(char *rwbs, u32 rw, int bytes)
 	else
 		rwbs[i++] = 'N';
 
+	if (rw & REQ_FUA)
+		rwbs[i++] = 'F';
 	if (rw & REQ_RAHEAD)
 		rwbs[i++] = 'A';
 	if (rw & REQ_SYNC)

kernel/tsacct.c

@@ -78,6 +78,7 @@ void bacct_add_tsk(struct taskstats *stats, struct task_struct *tsk)
 
 #define KB 1024
 #define MB (1024*KB)
+#define KB_MASK (~(KB-1))
 /*
  * fill in extended accounting fields
  */
@@ -95,14 +96,14 @@ void xacct_add_tsk(struct taskstats *stats, struct task_struct *p)
 		stats->hiwater_vm    = get_mm_hiwater_vm(mm)  * PAGE_SIZE / KB;
 		mmput(mm);
 	}
-	stats->read_char	= p->ioac.rchar;
-	stats->write_char	= p->ioac.wchar;
-	stats->read_syscalls	= p->ioac.syscr;
-	stats->write_syscalls	= p->ioac.syscw;
+	stats->read_char	= p->ioac.rchar & KB_MASK;
+	stats->write_char	= p->ioac.wchar & KB_MASK;
+	stats->read_syscalls	= p->ioac.syscr & KB_MASK;
+	stats->write_syscalls	= p->ioac.syscw & KB_MASK;
 #ifdef CONFIG_TASK_IO_ACCOUNTING
-	stats->read_bytes	= p->ioac.read_bytes;
-	stats->write_bytes	= p->ioac.write_bytes;
-	stats->cancelled_write_bytes = p->ioac.cancelled_write_bytes;
+	stats->read_bytes	= p->ioac.read_bytes & KB_MASK;
+	stats->write_bytes	= p->ioac.write_bytes & KB_MASK;
+	stats->cancelled_write_bytes = p->ioac.cancelled_write_bytes & KB_MASK;
 #else
 	stats->read_bytes	= 0;
 	stats->write_bytes	= 0;

kernel/workqueue.c

@@ -2412,8 +2412,13 @@ reflush:
 
 	for_each_cwq_cpu(cpu, wq) {
 		struct cpu_workqueue_struct *cwq = get_cwq(cpu, wq);
+		bool drained;
 
-		if (!cwq->nr_active && list_empty(&cwq->delayed_works))
+		spin_lock_irq(&cwq->gcwq->lock);
+		drained = !cwq->nr_active && list_empty(&cwq->delayed_works);
+		spin_unlock_irq(&cwq->gcwq->lock);
+
+		if (drained)
 			continue;
 
 		if (++flush_cnt == 10 ||