time: Avoid making adjustments if we haven't accumulated anything
authorJohn Stultz <john.stultz@linaro.org>
Wed, 22 Aug 2012 00:30:49 +0000 (20:30 -0400)
committerThomas Gleixner <tglx@linutronix.de>
Wed, 22 Aug 2012 08:42:13 +0000 (10:42 +0200)
If update_wall_time() is called and the current offset isn't large
enough to accumulate, avoid re-calling timekeeping_adjust which may
change the clock freq and can cause 1ns inconsistencies with
CLOCK_REALTIME_COARSE/CLOCK_MONOTONIC_COARSE.

Signed-off-by: John Stultz <john.stultz@linaro.org>
Cc: Prarit Bhargava <prarit@redhat.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: stable@vger.kernel.org
Link: http://lkml.kernel.org/r/1345595449-34965-5-git-send-email-john.stultz@linaro.org
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
kernel/time/timekeeping.c

index a5a9389c4c3090221d1b31bb89512cba5b36af0c..0c1485e42be6eec17d2525cbb4f2383daf8e2274 100644 (file)
@@ -1152,6 +1152,10 @@ static void update_wall_time(void)
        offset = (clock->read(clock) - clock->cycle_last) & clock->mask;
 #endif
 
+       /* Check if there's really nothing to do */
+       if (offset < tk->cycle_interval)
+               goto out;
+
        /*
         * With NO_HZ we may have to accumulate many cycle_intervals
         * (think "ticks") worth of time at once. To do this efficiently,