public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
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,
	"Paul E. McKenney" <paulmck@kernel.org>
Subject: [PATCH tip/core/rcu 16/30] refperf: Add warmup and cooldown processing phases
Date: Mon, 22 Jun 2020 17:29:59 -0700	[thread overview]
Message-ID: <20200623003013.26252-16-paulmck@kernel.org> (raw)
In-Reply-To: <20200623002941.GA26089@paulmck-ThinkPad-P72>

From: "Paul E. McKenney" <paulmck@kernel.org>

This commit causes all the readers to start running unmeasured load
until all readers have done at least one such run (thus having warmed
up), then run the measured load, and then run unmeasured load until all
readers have completed their measured load.  This approach avoids any
thread running measured load while other readers are idle.

Cc: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
---
 kernel/rcu/refperf.c | 20 ++++++++++++++++++--
 1 file changed, 18 insertions(+), 2 deletions(-)

diff --git a/kernel/rcu/refperf.c b/kernel/rcu/refperf.c
index 234bb0e..445190b 100644
--- a/kernel/rcu/refperf.c
+++ b/kernel/rcu/refperf.c
@@ -100,6 +100,8 @@ static atomic_t nreaders_exp;
 // Use to wait for all threads to start.
 static atomic_t n_init;
 static atomic_t n_started;
+static atomic_t n_warmedup;
+static atomic_t n_cooleddown;
 
 // Track which experiment is currently running.
 static int exp_idx;
@@ -260,8 +262,15 @@ ref_perf_reader(void *arg)
 
 	VERBOSE_PERFOUT("ref_perf_reader %ld: experiment %d started", me, exp_idx);
 
-	// To prevent noise, keep interrupts disabled. This also has the
-	// effect of preventing entries into slow path for rcu_read_unlock().
+
+	// To reduce noise, do an initial cache-warming invocation, check
+	// in, and then keep warming until everyone has checked in.
+	cur_ops->readsection(loops);
+	if (!atomic_dec_return(&n_warmedup))
+		while (atomic_read_acquire(&n_warmedup))
+			cur_ops->readsection(loops);
+	// Also keep interrupts disabled.  This also has the effect
+	// of preventing entries into slow path for rcu_read_unlock().
 	local_irq_save(flags);
 	start = ktime_get_mono_fast_ns();
 
@@ -271,6 +280,11 @@ ref_perf_reader(void *arg)
 	local_irq_restore(flags);
 
 	rt->last_duration_ns = WARN_ON_ONCE(duration < 0) ? 0 : duration;
+	// To reduce runtime-skew noise, do maintain-load invocations until
+	// everyone is done.
+	if (!atomic_dec_return(&n_cooleddown))
+		while (atomic_read_acquire(&n_cooleddown))
+			cur_ops->readsection(loops);
 
 	if (atomic_dec_and_test(&nreaders_exp))
 		wake_up(&main_wq);
@@ -372,6 +386,8 @@ static int main_func(void *arg)
 		reset_readers();
 		atomic_set(&nreaders_exp, nreaders);
 		atomic_set(&n_started, nreaders);
+		atomic_set(&n_warmedup, nreaders);
+		atomic_set(&n_cooleddown, nreaders);
 
 		exp_idx = exp;
 
-- 
2.9.5


  parent reply	other threads:[~2020-06-23  0:31 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-06-23  0:29 [PATCH tip/core/rcu 0/30] Add read-side scalability tests for v5.9 Paul E. McKenney
2020-06-23  0:29 ` [PATCH tip/core/rcu 01/30] rcuperf: Remove useless while loops around wait_event paulmck
2020-06-23  0:29 ` [PATCH tip/core/rcu 02/30] refperf: Add a test to measure performance of read-side synchronization paulmck
2020-06-23  0:29 ` [PATCH tip/core/rcu 03/30] rcuperf: Add comments explaining the high reader overhead paulmck
2020-06-23  0:29 ` [PATCH tip/core/rcu 04/30] torture: Add refperf to the rcutorture scripting paulmck
2020-06-23  0:29 ` [PATCH tip/core/rcu 05/30] refperf: Add holdoff parameter to allow CPUs to come online paulmck
2020-06-23  0:29 ` [PATCH tip/core/rcu 06/30] refperf: Hoist function-pointer calls out of the loop paulmck
2020-06-23  0:29 ` [PATCH tip/core/rcu 07/30] refperf: Allow decimal nanoseconds paulmck
2020-06-23  0:29 ` [PATCH tip/core/rcu 08/30] refperf: Convert nreaders to a module parameter paulmck
2020-06-23  0:29 ` [PATCH tip/core/rcu 09/30] refperf: Provide module parameter to specify number of experiments paulmck
2020-06-23  0:29 ` [PATCH tip/core/rcu 10/30] refperf: Dynamically allocate experiment-summary output buffer paulmck
2020-06-23  0:29 ` [PATCH tip/core/rcu 11/30] refperf: Dynamically allocate thread-summary " paulmck
2020-06-23  0:29 ` [PATCH tip/core/rcu 12/30] refperf: Make functions static paulmck
2020-06-23  0:29 ` [PATCH tip/core/rcu 13/30] refperf: Tune reader measurement interval paulmck
2020-06-23  0:29 ` [PATCH tip/core/rcu 14/30] refperf: Convert reader_task structure's "start" field to int paulmck
2020-06-23  0:29 ` [PATCH tip/core/rcu 15/30] refperf: More closely synchronize reader start times paulmck
2020-06-23  0:29 ` paulmck [this message]
2020-06-23  0:30 ` [PATCH tip/core/rcu 17/30] refperf: Label experiment-number column "Runs" paulmck
2020-06-23  0:30 ` [PATCH tip/core/rcu 18/30] refperf: Output per-experiment data points paulmck
2020-06-23  0:30 ` [PATCH tip/core/rcu 19/30] refperf: Simplify initialization-time wakeup protocol paulmck
2020-06-23  0:30 ` [PATCH tip/core/rcu 20/30] refperf: Add read-side delay module parameter paulmck
2020-06-23  0:30 ` [PATCH tip/core/rcu 21/30] refperf: Adjust refperf.loop default value paulmck
2020-06-23  0:30 ` [PATCH tip/core/rcu 22/30] doc: Document rcuperf's module parameters paulmck
2020-06-23  0:30 ` [PATCH tip/core/rcu 23/30] refperf: Work around 64-bit division paulmck
2020-06-23  0:30 ` [PATCH tip/core/rcu 24/30] refperf: Change readdelay module parameter to nanoseconds paulmck
2020-06-23  0:30 ` [PATCH tip/core/rcu 25/30] refperf: Add test for RCU Tasks Trace readers paulmck
2020-06-23  0:30 ` [PATCH tip/core/rcu 26/30] refperf: Add test for RCU Tasks readers paulmck
2020-06-23  0:30 ` [PATCH tip/core/rcu 27/30] rcu-tasks: Fix synchronize_rcu_tasks_trace() header comment paulmck
2020-06-23  0:30 ` [PATCH tip/core/rcu 28/30] refperf: Rename RCU_REF_PERF_TEST to RCU_REF_SCALE_TEST paulmck
2020-06-23  0:30 ` [PATCH tip/core/rcu 29/30] refperf: Rename refperf.c to refscale.c and change internal names paulmck
2020-06-23  0:30 ` [PATCH tip/core/rcu 30/30] refscale: Change --torture type from refperf to refscale 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=20200623003013.26252-16-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 \
    /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