Merge branch 'timers-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull timer changes for v3.4 from Ingo Molnar

* 'timers-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (32 commits)
  ntp: Fix integer overflow when setting time
  math: Introduce div64_long
  cs5535-clockevt: Allow the MFGPT IRQ to be shared
  cs5535-clockevt: Don't ignore MFGPT on SMP-capable kernels
  x86/time: Eliminate unused irq0_irqs counter
  clocksource: scx200_hrt: Fix the build
  x86/tsc: Reduce the TSC sync check time for core-siblings
  timer: Fix bad idle check on irq entry
  nohz: Remove ts->Einidle checks before restarting the tick
  nohz: Remove update_ts_time_stat from tick_nohz_start_idle
  clockevents: Leave the broadcast device in shutdown mode when not needed
  clocksource: Load the ACPI PM clocksource asynchronously
  clocksource: scx200_hrt: Convert scx200 to use clocksource_register_hz
  clocksource: Get rid of clocksource_calc_mult_shift()
  clocksource: dbx500: convert to clocksource_register_hz()
  clocksource: scx200_hrt:  use pr_<level> instead of printk
  time: Move common updates to a function
  time: Reorder so the hot data is together
  time: Remove most of xtime_lock usage in timekeeping.c
  ntp: Add ntp_lock to replace xtime_locking
  ...

arch/x86/kernel/time.c

@@ -57,9 +57,6 @@ EXPORT_SYMBOL(profile_pc);
  */
 static irqreturn_t timer_interrupt(int irq, void *dev_id)
 {
-	/* Keep nmi watchdog up to date */
-	inc_irq_stat(irq0_irqs);
-
 	global_clock_event->event_handler(global_clock_event);
 
 	/* MCA bus quirk: Acknowledge irq0 by setting bit 7 in port 0x61 */

arch/x86/kernel/tsc_sync.c

@@ -42,7 +42,7 @@ static __cpuinitdata int nr_warps;
 /*
  * TSC-warp measurement loop running on both CPUs:
  */
-static __cpuinit void check_tsc_warp(void)
+static __cpuinit void check_tsc_warp(unsigned int timeout)
 {
 	cycles_t start, now, prev, end;
 	int i;
@@ -51,9 +51,9 @@ static __cpuinit void check_tsc_warp(void)
 	start = get_cycles();
 	rdtsc_barrier();
 	/*
-	 * The measurement runs for 20 msecs:
+	 * The measurement runs for 'timeout' msecs:
 	 */
-	end = start + tsc_khz * 20ULL;
+	end = start + (cycles_t) tsc_khz * timeout;
 	now = start;
 
 	for (i = 0; ; i++) {
@@ -98,6 +98,25 @@ static __cpuinit void check_tsc_warp(void)
 			now-start, end-start);
 }
 
+/*
+ * If the target CPU coming online doesn't have any of its core-siblings
+ * online, a timeout of 20msec will be used for the TSC-warp measurement
+ * loop. Otherwise a smaller timeout of 2msec will be used, as we have some
+ * information about this socket already (and this information grows as we
+ * have more and more logical-siblings in that socket).
+ *
+ * Ideally we should be able to skip the TSC sync check on the other
+ * core-siblings, if the first logical CPU in a socket passed the sync test.
+ * But as the TSC is per-logical CPU and can potentially be modified wrongly
+ * by the bios, TSC sync test for smaller duration should be able
+ * to catch such errors. Also this will catch the condition where all the
+ * cores in the socket doesn't get reset at the same time.
+ */
+static inline unsigned int loop_timeout(int cpu)
+{
+	return (cpumask_weight(cpu_core_mask(cpu)) > 1) ? 2 : 20;
+}
+
 /*
  * Source CPU calls into this - it waits for the freshly booted
  * target CPU to arrive and then starts the measurement:
@@ -135,7 +154,7 @@ void __cpuinit check_tsc_sync_source(int cpu)
 	 */
 	atomic_inc(&start_count);
 
-	check_tsc_warp();
+	check_tsc_warp(loop_timeout(cpu));
 
 	while (atomic_read(&stop_count) != cpus-1)
 		cpu_relax();
@@ -183,7 +202,7 @@ void __cpuinit check_tsc_sync_target(void)
 	while (atomic_read(&start_count) != cpus)
 		cpu_relax();
 
-	check_tsc_warp();
+	check_tsc_warp(loop_timeout(smp_processor_id()));
 
 	/*
 	 * Ok, we are done: