From: "Paul E. McKenney" <paulmck@kernel.org>
To: Zhaolong Zhang <zhangzl2013@126.com>
Cc: Zhaolong Zhang <zhangzl2013@126.com.com>,
Josh Triplett <josh@joshtriplett.org>,
Steven Rostedt <rostedt@goodmis.org>,
Mathieu Desnoyers <mathieu.desnoyers@efficios.com>,
Lai Jiangshan <jiangshanlai@gmail.com>,
Joel Fernandes <joel@joelfernandes.org>,
rcu@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] rcu: Fix the (t=0 jiffies) false positive
Date: Thu, 5 Mar 2020 15:05:15 -0800 [thread overview]
Message-ID: <20200305230515.GN2935@paulmck-ThinkPad-P72> (raw)
In-Reply-To: <1583394357-11767-1-git-send-email-zhangzl2013@126.com>
On Thu, Mar 05, 2020 at 03:45:57PM +0800, Zhaolong Zhang wrote:
> Calculate 't' with the previously recorded 'gps' instead of 'gp_start'.
>
> Signed-off-by: Zhaolong Zhang <zhangzl2013@126.com>
Good catch, thank you!
I had to apply this by hand. My guess is that you developed against
mainline rather than the "dev" branch of the -rcu tree:
git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu.git
Or perhaps your email client corrupted the patch?
Regardless, please check the version below and let me know if I messed
anything up.
Thanx, Paul
------------------------------------------------------------------------
commit 7dd581bbbcc00246fec35eda60add4b32200b0a0
Author: Zhaolong Zhang <zhangzl2013@126.com>
Date: Thu Mar 5 14:56:11 2020 -0800
rcu: Fix the (t=0 jiffies) false positive
It is possible that an over-long grace period will end while the RCU
CPU stall warning message is printing. In this case, the estimate of
the offending grace period's duration can be erroneous due to refetching
of rcu_state.gp_start, which will now be the time of the newly started
grace period. Computation of this duration clearly needs to use the
start time for the old over-long grace period, not the fresh new one.
This commit avoids such errors by causing both print_other_cpu_stall() and
print_cpu_stall() to reuse the value previously fetched by their caller.
Signed-off-by: Zhaolong Zhang <zhangzl2013@126.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
diff --git a/kernel/rcu/tree_stall.h b/kernel/rcu/tree_stall.h
index b17cd9b..502b4dd 100644
--- a/kernel/rcu/tree_stall.h
+++ b/kernel/rcu/tree_stall.h
@@ -371,7 +371,7 @@ static void rcu_check_gp_kthread_starvation(void)
}
}
-static void print_other_cpu_stall(unsigned long gp_seq)
+static void print_other_cpu_stall(unsigned long gp_seq, unsigned long gps)
{
int cpu;
unsigned long flags;
@@ -408,7 +408,7 @@ static void print_other_cpu_stall(unsigned long gp_seq)
for_each_possible_cpu(cpu)
totqlen += rcu_get_n_cbs_cpu(cpu);
pr_cont("\t(detected by %d, t=%ld jiffies, g=%ld, q=%lu)\n",
- smp_processor_id(), (long)(jiffies - rcu_state.gp_start),
+ smp_processor_id(), (long)(jiffies - gps),
(long)rcu_seq_current(&rcu_state.gp_seq), totqlen);
if (ndetected) {
rcu_dump_cpu_stacks();
@@ -442,7 +442,7 @@ static void print_other_cpu_stall(unsigned long gp_seq)
rcu_force_quiescent_state(); /* Kick them all. */
}
-static void print_cpu_stall(void)
+static void print_cpu_stall(unsigned long gps)
{
int cpu;
unsigned long flags;
@@ -467,7 +467,7 @@ static void print_cpu_stall(void)
for_each_possible_cpu(cpu)
totqlen += rcu_get_n_cbs_cpu(cpu);
pr_cont("\t(t=%lu jiffies g=%ld q=%lu)\n",
- jiffies - rcu_state.gp_start,
+ jiffies - gps,
(long)rcu_seq_current(&rcu_state.gp_seq), totqlen);
rcu_check_gp_kthread_starvation();
@@ -546,7 +546,7 @@ static void check_cpu_stall(struct rcu_data *rdp)
cmpxchg(&rcu_state.jiffies_stall, js, jn) == js) {
/* We haven't checked in, so go dump stack. */
- print_cpu_stall();
+ print_cpu_stall(gps);
if (rcu_cpu_stall_ftrace_dump)
rcu_ftrace_dump(DUMP_ALL);
@@ -555,7 +555,7 @@ static void check_cpu_stall(struct rcu_data *rdp)
cmpxchg(&rcu_state.jiffies_stall, js, jn) == js) {
/* They had a few time units to dump stack, so complain. */
- print_other_cpu_stall(gs2);
+ print_other_cpu_stall(gs2, gps);
if (rcu_cpu_stall_ftrace_dump)
rcu_ftrace_dump(DUMP_ALL);
}
next prev parent reply other threads:[~2020-03-05 23:05 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-03-05 7:45 [PATCH] rcu: Fix the (t=0 jiffies) false positive Zhaolong Zhang
2020-03-05 23:05 ` Paul E. McKenney [this message]
2020-03-06 5:06 ` Zhaolong Zhang
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=20200305230515.GN2935@paulmck-ThinkPad-P72 \
--to=paulmck@kernel.org \
--cc=jiangshanlai@gmail.com \
--cc=joel@joelfernandes.org \
--cc=josh@joshtriplett.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mathieu.desnoyers@efficios.com \
--cc=rcu@vger.kernel.org \
--cc=rostedt@goodmis.org \
--cc=zhangzl2013@126.com \
--cc=zhangzl2013@126.com.com \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).