From: paulmck@kernel.org
To: rcu@vger.kernel.org
Cc: linux-kernel@vger.kernel.org, kernel-team@fb.com,
mingo@kernel.org, jiangshanlai@gmail.com, dipankar@in.ibm.com,
akpm@linux-foundation.org, mathieu.desnoyers@efficios.com,
josh@joshtriplett.org, tglx@linutronix.de, peterz@infradead.org,
rostedt@goodmis.org, dhowells@redhat.com, edumazet@google.com,
fweisbec@gmail.com, oleg@redhat.com, joel@joelfernandes.org,
urezki@gmail.com, "Paul E . McKenney" <paulmck@kernel.org>
Subject: [PATCH tip/core/rcu 23/24] rcu/segcblist: Prevent useless GP start if no CBs to accelerate
Date: Mon, 31 Aug 2020 11:01:15 -0700 [thread overview]
Message-ID: <20200831180116.32690-23-paulmck@kernel.org> (raw)
In-Reply-To: <20200831180050.GA32590@paulmck-ThinkPad-P72>
From: "Joel Fernandes (Google)" <joel@joelfernandes.org>
The rcu_segcblist_accelerate() function returns true iff it is necessary
to request another grace period. A tracing session showed that this
function unnecessarily requests grace periods.
For exmaple, consider the following sequence of events:
1. Callbacks are queued only on the NEXT segment of CPU A's callback list.
2. CPU A runs RCU_SOFTIRQ, accelerating these callbacks from NEXT to WAIT.
3. Thus rcu_segcblist_accelerate() returns true, requesting grace period N.
4. RCU's grace-period kthread wakes up on CPU B and starts grace period N.
4. CPU A notices the new grace period and invokes RCU_SOFTIRQ.
5. CPU A's RCU_SOFTIRQ again invokes rcu_segcblist_accelerate(), but
there are no new callbacks. However, rcu_segcblist_accelerate()
nevertheless (uselessly) requests a new grace period N+1.
This extra grace period results in additional lock contention and also
additional wakeups, all for no good reason.
This commit therefore adds a check to rcu_segcblist_accelerate() that
prevents the return of true when there are no new callbacks.
This change reduces the number of grace periods (GPs) and wakeups in each
of eleven five-second rcutorture runs as follows:
+----+-------------------+-------------------+
| # | Number of GPs | Number of Wakeups |
+====+=========+=========+=========+=========+
| 1 | With | Without | With | Without |
+----+---------+---------+---------+---------+
| 2 | 75 | 89 | 113 | 119 |
+----+---------+---------+---------+---------+
| 3 | 62 | 91 | 105 | 123 |
+----+---------+---------+---------+---------+
| 4 | 60 | 79 | 98 | 110 |
+----+---------+---------+---------+---------+
| 5 | 63 | 79 | 99 | 112 |
+----+---------+---------+---------+---------+
| 6 | 57 | 89 | 96 | 123 |
+----+---------+---------+---------+---------+
| 7 | 64 | 85 | 97 | 118 |
+----+---------+---------+---------+---------+
| 8 | 58 | 83 | 98 | 113 |
+----+---------+---------+---------+---------+
| 9 | 57 | 77 | 89 | 104 |
+----+---------+---------+---------+---------+
| 10 | 66 | 82 | 98 | 119 |
+----+---------+---------+---------+---------+
| 11 | 52 | 82 | 83 | 117 |
+----+---------+---------+---------+---------+
The reduction in the number of wakeups ranges from 5% to 40%.
Cc: urezki@gmail.com
[ paulmck: Rework commit log and comment. ]
Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
---
kernel/rcu/rcu_segcblist.c | 10 +++++++++-
1 file changed, 9 insertions(+), 1 deletion(-)
diff --git a/kernel/rcu/rcu_segcblist.c b/kernel/rcu/rcu_segcblist.c
index 9a0f661..2d2a6b6b9 100644
--- a/kernel/rcu/rcu_segcblist.c
+++ b/kernel/rcu/rcu_segcblist.c
@@ -475,8 +475,16 @@ bool rcu_segcblist_accelerate(struct rcu_segcblist *rsclp, unsigned long seq)
* Also advance to the oldest segment of callbacks whose
* ->gp_seq[] completion is at or after that passed in via "seq",
* skipping any empty segments.
+ *
+ * Note that segment "i" (and any lower-numbered segments
+ * containing older callbacks) will be unaffected, and their
+ * grace-period numbers remain unchanged. For example, if i ==
+ * WAIT_TAIL, then neither WAIT_TAIL nor DONE_TAIL will be touched.
+ * Instead, the CBs in NEXT_TAIL will be merged with those in
+ * NEXT_READY_TAIL and the grace-period number of NEXT_READY_TAIL
+ * would be updated. NEXT_TAIL would then be empty.
*/
- if (++i >= RCU_NEXT_TAIL)
+ if (rcu_segcblist_restempty(rsclp, i) || ++i >= RCU_NEXT_TAIL)
return false;
/*
--
2.9.5
next prev parent reply other threads:[~2020-08-31 18:01 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-08-31 18:00 [PATCH tip/core/rcu 0/24] Miscellaneous fixes for v5.10 Paul E. McKenney
2020-08-31 18:00 ` [PATCH tip/core/rcu 01/24] rcu: Remove KCSAN stubs paulmck
2020-08-31 18:00 ` [PATCH tip/core/rcu 02/24] rcu: Remove KCSAN stubs from update.c paulmck
2020-08-31 18:00 ` [PATCH tip/core/rcu 03/24] srcu: Remove KCSAN stubs paulmck
2020-08-31 18:00 ` [PATCH tip/core/rcu 04/24] rcu: Initialize at declaration time in rcu_exp_handler() paulmck
2020-08-31 18:00 ` [PATCH tip/core/rcu 05/24] rcu/trace: Print negative GP numbers correctly paulmck
2020-08-31 18:00 ` [PATCH tip/core/rcu 06/24] rcu/trace: Use gp_seq_req in acceleration's rcu_grace_period tracepoint paulmck
2020-08-31 18:00 ` [PATCH tip/core/rcu 07/24] nocb: Clarify RCU nocb CPU error message paulmck
2020-08-31 18:01 ` [PATCH tip/core/rcu 08/24] rcu/tree: Force quiescent state on callback overload paulmck
2020-08-31 18:01 ` [PATCH tip/core/rcu 09/24] rcu/tree: Remove CONFIG_PREMPT_RCU check in force_qs_rnp() paulmck
2020-08-31 18:01 ` [PATCH tip/core/rcu 10/24] nocb: Remove show_rcu_nocb_state() false positive printout paulmck
2020-08-31 18:01 ` [PATCH tip/core/rcu 11/24] rcu: Add READ_ONCE() to rcu_do_batch() access to rcu_divisor paulmck
2020-08-31 18:01 ` [PATCH tip/core/rcu 12/24] rcu: Add READ_ONCE() to rcu_do_batch() access to rcu_resched_ns paulmck
2020-08-31 18:01 ` [PATCH tip/core/rcu 13/24] rcu: Add READ_ONCE() to rcu_do_batch() access to rcu_kick_kthreads paulmck
2020-08-31 18:01 ` [PATCH tip/core/rcu 14/24] rcu: Add READ_ONCE() to rcu_do_batch() access to rcu_cpu_stall_ftrace_dump paulmck
2020-08-31 18:01 ` [PATCH tip/core/rcu 15/24] rcu: Fix kerneldoc comments in rcupdate.h paulmck
2020-08-31 18:01 ` [PATCH tip/core/rcu 16/24] rculist: Introduce list/hlist_for_each_entry_srcu() macros paulmck
2020-08-31 18:01 ` [PATCH tip/core/rcu 17/24] kvm: mmu: page_track: Fix RCU list API usage paulmck
2020-08-31 18:01 ` [PATCH tip/core/rcu 18/24] rcu: Move rcu_cpu_started per-CPU variable to rcu_data paulmck
2020-08-31 18:01 ` [PATCH tip/core/rcu 19/24] rcu/nocb: Add a warning for non-GP kthread running GP code paulmck
2020-08-31 18:01 ` [PATCH tip/core/rcu 20/24] rcu: Clarify comments about FQS loop reporting quiescent states paulmck
2020-08-31 18:01 ` [PATCH tip/core/rcu 21/24] rcu: Make FQS more aggressive in complaining about offline CPUs paulmck
2020-08-31 18:01 ` [PATCH tip/core/rcu 22/24] rcu: Remove unused __rcu_is_watching() function paulmck
2020-08-31 18:01 ` paulmck [this message]
2020-08-31 18:01 ` [PATCH tip/core/rcu 24/24] rcu: Shrink each possible cpu krcp paulmck
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=20200831180116.32690-23-paulmck@kernel.org \
--to=paulmck@kernel.org \
--cc=akpm@linux-foundation.org \
--cc=dhowells@redhat.com \
--cc=dipankar@in.ibm.com \
--cc=edumazet@google.com \
--cc=fweisbec@gmail.com \
--cc=jiangshanlai@gmail.com \
--cc=joel@joelfernandes.org \
--cc=josh@joshtriplett.org \
--cc=kernel-team@fb.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mathieu.desnoyers@efficios.com \
--cc=mingo@kernel.org \
--cc=oleg@redhat.com \
--cc=peterz@infradead.org \
--cc=rcu@vger.kernel.org \
--cc=rostedt@goodmis.org \
--cc=tglx@linutronix.de \
--cc=urezki@gmail.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).