timekeeping: Add warnings when overflows or underflows are observed

It was suggested that the underflow/overflow protection
should probably throw some sort of warning out, rather
than just silently fixing the issue.

So this patch adds some warnings here. The flag variables
used are not protected by locks, but since we can't print
from the reading functions, just being able to say we
saw an issue in the update interval is useful enough,
and can be slightly racy without real consequence.

The big complication is that we're only under a read
seqlock, so the data could shift under us during
our calculation to see if there was a problem. This
patch avoids this issue by nesting another seqlock
which allows us to snapshot the just required values
atomically. So we shouldn't see false positives.

I also added some basic rate-limiting here, since
on one build machine w/ skewed TSCs it was fairly
noisy at bootup.

Signed-off-by: John Stultz <john.stultz@linaro.org>
Cc: Dave Jones <davej@codemonkey.org.uk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Prarit Bhargava <prarit@redhat.com>
Cc: Richard Cochran <richardcochran@gmail.com>
Cc: Stephen Boyd <sboyd@codeaurora.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1426133800-29329-8-git-send-email-john.stultz@linaro.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
This commit is contained in:
John Stultz 2015-03-11 21:16:35 -07:00 committed by Ingo Molnar
parent 057b87e316
commit 4ca22c2648

View File

@ -119,6 +119,20 @@ static inline void tk_update_sleep_time(struct timekeeper *tk, ktime_t delta)
}
#ifdef CONFIG_DEBUG_TIMEKEEPING
#define WARNING_FREQ (HZ*300) /* 5 minute rate-limiting */
/*
* These simple flag variables are managed
* without locks, which is racy, but ok since
* we don't really care about being super
* precise about how many events were seen,
* just that a problem was observed.
*/
static int timekeeping_underflow_seen;
static int timekeeping_overflow_seen;
/* last_warning is only modified under the timekeeping lock */
static long timekeeping_last_warning;
static void timekeeping_check_update(struct timekeeper *tk, cycle_t offset)
{
@ -136,28 +150,64 @@ static void timekeeping_check_update(struct timekeeper *tk, cycle_t offset)
printk_deferred(" timekeeping: Your kernel is still fine, but is feeling a bit nervous\n");
}
}
if (timekeeping_underflow_seen) {
if (jiffies - timekeeping_last_warning > WARNING_FREQ) {
printk_deferred("WARNING: Underflow in clocksource '%s' observed, time update ignored.\n", name);
printk_deferred(" Please report this, consider using a different clocksource, if possible.\n");
printk_deferred(" Your kernel is probably still fine.\n");
timekeeping_last_warning = jiffies;
}
timekeeping_underflow_seen = 0;
}
if (timekeeping_overflow_seen) {
if (jiffies - timekeeping_last_warning > WARNING_FREQ) {
printk_deferred("WARNING: Overflow in clocksource '%s' observed, time update capped.\n", name);
printk_deferred(" Please report this, consider using a different clocksource, if possible.\n");
printk_deferred(" Your kernel is probably still fine.\n");
timekeeping_last_warning = jiffies;
}
timekeeping_overflow_seen = 0;
}
}
static inline cycle_t timekeeping_get_delta(struct tk_read_base *tkr)
{
cycle_t cycle_now, delta;
cycle_t now, last, mask, max, delta;
unsigned int seq;
/* read clocksource */
cycle_now = tkr->read(tkr->clock);
/*
* Since we're called holding a seqlock, the data may shift
* under us while we're doing the calculation. This can cause
* false positives, since we'd note a problem but throw the
* results away. So nest another seqlock here to atomically
* grab the points we are checking with.
*/
do {
seq = read_seqcount_begin(&tk_core.seq);
now = tkr->read(tkr->clock);
last = tkr->cycle_last;
mask = tkr->mask;
max = tkr->clock->max_cycles;
} while (read_seqcount_retry(&tk_core.seq, seq));
/* calculate the delta since the last update_wall_time */
delta = clocksource_delta(cycle_now, tkr->cycle_last, tkr->mask);
delta = clocksource_delta(now, last, mask);
/*
* Try to catch underflows by checking if we are seeing small
* mask-relative negative values.
*/
if (unlikely((~delta & tkr->mask) < (tkr->mask >> 3)))
if (unlikely((~delta & mask) < (mask >> 3))) {
timekeeping_underflow_seen = 1;
delta = 0;
}
/* Cap delta value to the max_cycles values to avoid mult overflows */
if (unlikely(delta > tkr->clock->max_cycles))
if (unlikely(delta > max)) {
timekeeping_overflow_seen = 1;
delta = tkr->clock->max_cycles;
}
return delta;
}