Cache xtime every call to update_wall_time

This avoids xtime lag seen with dynticks, because while 'xtime' itself
is still not updated often, we keep a 'xtime_cache' variable around that
contains the approximate real-time that _is_ updated each time we do a
'update_wall_time()', and is thus never off by more than one tick.

IOW, this restores the original semantics for 'xtime' users, as long as
you use the proper abstraction functions (ie 'current_kernel_time()' or
'get_seconds()' depending on whether you want a timespec or just the
seconds field).

[ Updated Patch.  As penance for my sins I've also yanked another #ifdef
  that was added to avoid the xtime lag w/ hrtimers.  ]

Signed-off-by: John Stultz <johnstul@us.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

3 files changed, 24 insertions, 12 deletions

diff --git a/include/linux/time.h b/include/linux/time.h
index 71181df8b744..6a5f503b4f1d 100644
--- a/include/linux/time.h
+++ b/include/linux/time.h
@@ -99,11 +99,7 @@ extern int update_persistent_clock(struct timespec now);
 extern int no_sync_cmos_clock __read_mostly;
 void timekeeping_init(void);
 
-static inline unsigned long get_seconds(void)
-{
-        return xtime.tv_sec;
-}
-
+unsigned long get_seconds(void);
 struct timespec current_kernel_time(void);
 
 #define CURRENT_TIME            (current_kernel_time())
diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c
index a7bb05e6cb63..c21ca6bfaa66 100644
--- a/kernel/hrtimer.c
+++ b/kernel/hrtimer.c
@@ -141,11 +141,7 @@ static void hrtimer_get_softirq_time(struct hrtimer_cpu_base *base)
 
         do {
                 seq = read_seqbegin(&xtime_lock);
-#ifdef CONFIG_NO_HZ
-                getnstimeofday(&xts);
-#else
                 xts = current_kernel_time();
-#endif
                 tom = wall_to_monotonic;
         } while (read_seqretry(&xtime_lock, seq));
 
diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c
index 07a3f1420c27..acc417b5a9b7 100644
--- a/kernel/time/timekeeping.c
+++ b/kernel/time/timekeeping.c
@@ -47,10 +47,22 @@ EXPORT_SYMBOL(xtime_lock);
 struct timespec xtime __attribute__ ((aligned (16)));
 struct timespec wall_to_monotonic __attribute__ ((aligned (16)));
 static unsigned long total_sleep_time;          /* seconds */
-
 EXPORT_SYMBOL(xtime);
 
 
+#ifdef CONFIG_NO_HZ
+static struct timespec xtime_cache __attribute__ ((aligned (16)));
+static inline void update_xtime_cache(u64 nsec)
+{
+        xtime_cache = xtime;
+        timespec_add_ns(&xtime_cache, nsec);
+}
+#else
+#define xtime_cache xtime
+/* We do *not* want to evaluate the argument for this case */
+#define update_xtime_cache(n) do { } while (0)
+#endif
+
 static struct clocksource *clock; /* pointer to current clocksource */
 
 
@@ -478,6 +490,8 @@ void update_wall_time(void)
         xtime.tv_nsec = (s64)clock->xtime_nsec >> clock->shift;
         clock->xtime_nsec -= (s64)xtime.tv_nsec << clock->shift;
 
+        update_xtime_cache(cyc2ns(clock, offset));
+
         /* check to see if there is a new clocksource to use */
         change_clocksource();
         update_vsyscall(&xtime, clock);
@@ -510,6 +524,13 @@ void monotonic_to_bootbased(struct timespec *ts)
         ts->tv_sec += total_sleep_time;
 }
 
+unsigned long get_seconds(void)
+{
+        return xtime_cache.tv_sec;
+}
+EXPORT_SYMBOL(get_seconds);
+
+
 struct timespec current_kernel_time(void)
 {
         struct timespec now;
@@ -518,10 +539,9 @@ struct timespec current_kernel_time(void)
         do {
                 seq = read_seqbegin(&xtime_lock);
 
-                now = xtime;
+                now = xtime_cache;
         } while (read_seqretry(&xtime_lock, seq));
 
         return now;
 }
-
 EXPORT_SYMBOL(current_kernel_time);