clocksource: Fix abs() usage w/ 64bit values
authorJohn Stultz <john.stultz@linaro.org>
Tue, 15 Sep 2015 01:05:20 +0000 (18:05 -0700)
committerThomas Gleixner <tglx@linutronix.de>
Fri, 2 Oct 2015 20:53:01 +0000 (22:53 +0200)
This patch fixes one cases where abs() was being used with 64-bit
nanosecond values, where the result may be capped at 32-bits.

This potentially could cause watchdog false negatives on 32-bit
systems, so this patch addresses the issue by using abs64().

Signed-off-by: John Stultz <john.stultz@linaro.org>
Cc: Prarit Bhargava <prarit@redhat.com>
Cc: Richard Cochran <richardcochran@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: stable@vger.kernel.org
Link: http://lkml.kernel.org/r/1442279124-7309-2-git-send-email-john.stultz@linaro.org
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
kernel/time/clocksource.c

index 841b72f..3a38775 100644 (file)
@@ -217,7 +217,7 @@ static void clocksource_watchdog(unsigned long data)
                        continue;
 
                /* Check the deviation from the watchdog clocksource. */
-               if ((abs(cs_nsec - wd_nsec) > WATCHDOG_THRESHOLD)) {
+               if (abs64(cs_nsec - wd_nsec) > WATCHDOG_THRESHOLD) {
                        pr_warn("timekeeping watchdog: Marking clocksource '%s' as unstable because the skew is too large:\n",
                                cs->name);
                        pr_warn("                      '%s' wd_now: %llx wd_last: %llx mask: %llx\n",