From:   Byungchul Park <max.byungchul.park@gmail.com>
To:     linux-kernel@vger.kernel.org
Cc:     torvalds@linux-foundation.org, damien.lemoal@opensource.wdc.com,
        linux-ide@vger.kernel.org, adilger.kernel@dilger.ca,
        linux-ext4@vger.kernel.org, mingo@redhat.com, peterz@infradead.org,
        will@kernel.org, tglx@linutronix.de, rostedt@goodmis.org,
        joel@joelfernandes.org, sashal@kernel.org, daniel.vetter@ffwll.ch,
        duyuyang@gmail.com, johannes.berg@intel.com, tj@kernel.org,
        tytso@mit.edu, willy@infradead.org, david@fromorbit.com,
        amir73il@gmail.com, gregkh@linuxfoundation.org,
        kernel-team@lge.com, linux-mm@kvack.org, akpm@linux-foundation.org,
        mhocko@kernel.org, minchan@kernel.org, hannes@cmpxchg.org,
        vdavydov.dev@gmail.com, sj@kernel.org, jglisse@redhat.com,
        dennis@kernel.org, cl@linux.com, penberg@kernel.org,
        rientjes@google.com, vbabka@suse.cz, ngupta@vflare.org,
        linux-block@vger.kernel.org, paolo.valente@linaro.org,
        josef@toxicpanda.com, linux-fsdevel@vger.kernel.org,
        viro@zeniv.linux.org.uk, jack@suse.cz, jlayton@kernel.org,
        dan.j.williams@intel.com, hch@infradead.org, djwong@kernel.org,
        dri-devel@lists.freedesktop.org, rodrigosiqueiramelo@gmail.com,
        melissa.srw@gmail.com, hamohammed.sa@gmail.com,
        42.hyeyoo@gmail.com, chris.p.wilson@intel.com,
        gwan-gyeong.mun@intel.com, max.byungchul.park@gmail.com,
        boqun.feng@gmail.com, longman@redhat.com, hdanton@sina.com
Subject: [PATCH v9 23/25] dept: Record the latest one out of consecutive waits of the same class
Date:   Tue, 31 Jan 2023 17:39:52 +0900
Message-Id: <1675154394-25598-24-git-send-email-max.byungchul.park@gmail.com>
In-Reply-To: <1675154394-25598-1-git-send-email-max.byungchul.park@gmail.com>
References: <1675154394-25598-1-git-send-email-max.byungchul.park@gmail.com>
Precedence: bulk

The current code records all the waits for later use to track relation
between waits and events in each context. However, since the same class
is handled the same way, it'd be okay to record only one on behalf of
the others if they all have the same class.

Even though it's the ideal to search the whole history buffer for that,
since it'd cost too high, alternatively, let's keep the latest one at
least when the same class'ed waits consecutively appear.

Signed-off-by: Byungchul Park <max.byungchul.park@gmail.com>
---
 kernel/dependency/dept.c | 21 ++++++++++++++++++++-
 1 file changed, 20 insertions(+), 1 deletion(-)

diff --git a/kernel/dependency/dept.c b/kernel/dependency/dept.c
index 3ab999f..2887858 100644
--- a/kernel/dependency/dept.c
+++ b/kernel/dependency/dept.c
@@ -1522,9 +1522,28 @@ static inline struct dept_wait_hist *new_hist(void)
 	return wh;
 }
 
+static inline struct dept_wait_hist *last_hist(void)
+{
+	int pos_n = hist_pos_next();
+	struct dept_wait_hist *wh_n = hist(pos_n);
+
+	/*
+	 * This is the first try.
+	 */
+	if (!pos_n && !wh_n->wait)
+		return NULL;
+
+	return hist(pos_n + DEPT_MAX_WAIT_HIST - 1);
+}
+
 static void add_hist(struct dept_wait *w, unsigned int wg, unsigned int ctxt_id)
 {
-	struct dept_wait_hist *wh = new_hist();
+	struct dept_wait_hist *wh;
+
+	wh = last_hist();
+
+	if (!wh || wh->wait->class != w->class || wh->ctxt_id != ctxt_id)
+		wh = new_hist();
 
 	if (likely(wh->wait))
 		put_wait(wh->wait);
-- 
1.9.1