All of lore.kernel.org
 help / color / mirror / Atom feed
From: John Stultz <john.stultz@linaro.org>
To: lkml <linux-kernel@vger.kernel.org>
Cc: John Stultz <john.stultz@linaro.org>
Subject: [PATCH 09/12] clocksource: Improve clocksource watchdog reporting
Date: Fri,  6 Mar 2015 18:50:00 -0800	[thread overview]
Message-ID: <1425696603-16878-10-git-send-email-john.stultz@linaro.org> (raw)
In-Reply-To: <1425696603-16878-1-git-send-email-john.stultz@linaro.org>

The clocksource watchdog reporting has been less helpful
then desired, as it just printed the delta between
the two clocksources. This prevents any useful analysis
of why the skew occurred.

Thus this patch tries to improve the output when we
mark a clocksource as unstable, printing out the cycle
last and now values for both the current clocksource
and the watchdog clocksource. This will allow us to see
if the result was due to a false positive caused by
a problematic watchdog.

Signed-off-by: John Stultz <john.stultz@linaro.org>
---
 kernel/time/clocksource.c | 21 ++++++++++++---------
 1 file changed, 12 insertions(+), 9 deletions(-)

diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c
index e6c752b..51c7b3a 100644
--- a/kernel/time/clocksource.c
+++ b/kernel/time/clocksource.c
@@ -142,13 +142,6 @@ static void __clocksource_unstable(struct clocksource *cs)
 		schedule_work(&watchdog_work);
 }
 
-static void clocksource_unstable(struct clocksource *cs, int64_t delta)
-{
-	printk(KERN_WARNING "Clocksource %s unstable (delta = %Ld ns)\n",
-	       cs->name, delta);
-	__clocksource_unstable(cs);
-}
-
 /**
  * clocksource_mark_unstable - mark clocksource unstable via watchdog
  * @cs:		clocksource to be marked unstable
@@ -174,7 +167,7 @@ void clocksource_mark_unstable(struct clocksource *cs)
 static void clocksource_watchdog(unsigned long data)
 {
 	struct clocksource *cs;
-	cycle_t csnow, wdnow, delta;
+	cycle_t csnow, wdnow, cslast, wdlast, delta;
 	int64_t wd_nsec, cs_nsec;
 	int next_cpu, reset_pending;
 
@@ -213,6 +206,8 @@ static void clocksource_watchdog(unsigned long data)
 
 		delta = clocksource_delta(csnow, cs->cs_last, cs->mask);
 		cs_nsec = clocksource_cyc2ns(delta, cs->mult, cs->shift);
+		wdlast = cs->wd_last; /* save these incase we print them */
+		cslast = cs->cs_last;
 		cs->cs_last = csnow;
 		cs->wd_last = wdnow;
 
@@ -221,7 +216,15 @@ static void clocksource_watchdog(unsigned long data)
 
 		/* Check the deviation from the watchdog clocksource. */
 		if ((abs(cs_nsec - wd_nsec) > WATCHDOG_THRESHOLD)) {
-			clocksource_unstable(cs, cs_nsec - wd_nsec);
+			pr_warn("Watchdog: clocksource %s unstable\n",
+				cs->name);
+			pr_warn("	"
+				"%s wd_now: %llx wd_last: %llx mask: %llx\n",
+				watchdog->name, wdnow, wdlast, watchdog->mask);
+			pr_warn("	"
+				"%s cs_now: %llx cs_last: %llx mask: %llx\n",
+				cs->name, csnow, cslast, cs->mask);
+			__clocksource_unstable(cs);
 			continue;
 		}
 
-- 
1.9.1


  parent reply	other threads:[~2015-03-07  2:52 UTC|newest]

Thread overview: 29+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-03-07  2:49 [PATCH 00/12] Increased clocksource validation and cleanups (v3) John Stultz
2015-03-07  2:49 ` [PATCH 01/12] clocksource: Simplify clocks_calc_max_nsecs logic John Stultz
2015-03-07  9:20   ` Ingo Molnar
2015-03-07  2:49 ` [PATCH 02/12] clocksource: Simplify logic around clocksource wrapping saftey margins John Stultz
2015-03-07  9:22   ` Ingo Molnar
2015-03-07  2:49 ` [PATCH 03/12] clocksource: Remove clocksource_max_deferment() John Stultz
2015-03-07  9:18   ` Ingo Molnar
2015-03-07  2:49 ` [PATCH 04/12] clocksource: Add max_cycles to clocksource structure John Stultz
2015-03-07  9:54   ` Ingo Molnar
2015-03-07  2:49 ` [PATCH 05/12] time: Add debugging checks to warn if we see delays John Stultz
2015-03-07  9:22   ` Paul Bolle
2015-03-07  9:29   ` Ingo Molnar
2015-03-07  2:49 ` [PATCH 06/12] time: Add infrastructure to cap clocksource reads to the max_cycles value John Stultz
2015-03-07  9:32   ` Ingo Molnar
2015-03-07  2:49 ` [PATCH 07/12] time: Try to catch clocksource delta underflows John Stultz
2015-03-07  9:34   ` Ingo Molnar
2015-03-07  2:49 ` [PATCH 08/12] time: Add warnings when overflows or underflows are observed John Stultz
2015-03-07  9:40   ` Ingo Molnar
2015-03-09 16:50     ` John Stultz
2015-03-10  5:05       ` Ingo Molnar
2015-03-07  2:50 ` John Stultz [this message]
2015-03-07  2:50 ` [PATCH 10/12] clocksource: Mostly kill clocksource_register() John Stultz
2015-03-07  9:45   ` Ingo Molnar
2015-03-07  2:50 ` [PATCH 11/12] sparc: Convert to using clocksource_register_hz() John Stultz
2015-03-07  9:46   ` Ingo Molnar
2015-03-07  2:50 ` [PATCH 12/12] clocksource: Add some debug info about clocksources being registered John Stultz
2015-03-07  9:50   ` Ingo Molnar
2015-03-12  3:16     ` John Stultz
  -- strict thread matches above, loose matches on Subject: below --
2015-01-23  0:09 [PATCH 00/12][RFC] Increased clocksource validation and cleanups (v2) John Stultz
2015-01-23  0:09 ` [PATCH 09/12] clocksource: Improve clocksource watchdog reporting John Stultz

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1425696603-16878-10-git-send-email-john.stultz@linaro.org \
    --to=john.stultz@linaro.org \
    --cc=linux-kernel@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.