Linux block layer
 help / color / mirror / Atom feed
From: Jason Xing <kerneljasonxing@gmail.com>
To: axboe@kernel.dk, rostedt@goodmis.org, mhiramat@kernel.org,
	mathieu.desnoyers@efficios.com, akpm@linux-foundation.org
Cc: linux-kernel@vger.kernel.org, linux-block@vger.kernel.org,
	linux-trace-kernel@vger.kernel.org,
	Jason Xing <kernelxing@tencent.com>,
	Yushan Zhou <katrinzhou@tencent.com>
Subject: [PATCH v1 3/5] blktrace: use rbuf->stats.full as a drop indicator in relayfs
Date: Mon, 12 May 2025 10:49:33 +0800	[thread overview]
Message-ID: <20250512024935.64704-4-kerneljasonxing@gmail.com> (raw)
In-Reply-To: <20250512024935.64704-1-kerneljasonxing@gmail.com>

From: Jason Xing <kernelxing@tencent.com>

Replace internal subbuf_start with the default policy in relayfs.

Remove dropped field from struct blk_trace. Correspondingly, use per-cpu
buffer mechanism to replace atomic operation. Through incrementing
full counter to keep track of how many times we encountered a full
subbuffer, it aids the user space app in telling how many lost events
there were.

Sum up all the fields gathered from all cpus when application is reading.

Reviewed-by: Yushan Zhou <katrinzhou@tencent.com>
Signed-off-by: Jason Xing <kernelxing@tencent.com>
---
 kernel/trace/blktrace.c | 23 ++---------------------
 1 file changed, 2 insertions(+), 21 deletions(-)

diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c
index d1a89714e805..09d42c40ac9f 100644
--- a/kernel/trace/blktrace.c
+++ b/kernel/trace/blktrace.c
@@ -415,9 +415,9 @@ static ssize_t blk_dropped_read(struct file *filp, char __user *buffer,
 				size_t count, loff_t *ppos)
 {
 	struct blk_trace *bt = filp->private_data;
-	char buf[16];
+	char buf[RELAY_DUMP_BUF_MAX_LEN];
 
-	snprintf(buf, sizeof(buf), "%u\n", atomic_read(&bt->dropped));
+	relay_dump(bt->rchan, buf, RELAY_DUMP_BUF_MAX_LEN, RELAY_DUMP_BUF_FULL);
 
 	return simple_read_from_buffer(buffer, count, ppos, buf, strlen(buf));
 }
@@ -456,23 +456,6 @@ static const struct file_operations blk_msg_fops = {
 	.llseek =	noop_llseek,
 };
 
-/*
- * Keep track of how many times we encountered a full subbuffer, to aid
- * the user space app in telling how many lost events there were.
- */
-static int blk_subbuf_start_callback(struct rchan_buf *buf, void *subbuf,
-				     void *prev_subbuf)
-{
-	struct blk_trace *bt;
-
-	if (!relay_buf_full(buf))
-		return 1;
-
-	bt = buf->chan->private_data;
-	atomic_inc(&bt->dropped);
-	return 0;
-}
-
 static int blk_remove_buf_file_callback(struct dentry *dentry)
 {
 	debugfs_remove(dentry);
@@ -491,7 +474,6 @@ static struct dentry *blk_create_buf_file_callback(const char *filename,
 }
 
 static const struct rchan_callbacks blk_relay_callbacks = {
-	.subbuf_start		= blk_subbuf_start_callback,
 	.create_buf_file	= blk_create_buf_file_callback,
 	.remove_buf_file	= blk_remove_buf_file_callback,
 };
@@ -580,7 +562,6 @@ static int do_blk_trace_setup(struct request_queue *q, char *name, dev_t dev,
 	}
 
 	bt->dev = dev;
-	atomic_set(&bt->dropped, 0);
 	INIT_LIST_HEAD(&bt->running_list);
 
 	ret = -EIO;
-- 
2.43.5


  parent reply	other threads:[~2025-05-12  2:49 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-05-12  2:49 [PATCH v1 0/5] relayfs: misc changes Jason Xing
2025-05-12  2:49 ` [PATCH v1 1/5] relayfs: support a counter tracking if per-cpu buffers is full Jason Xing
2025-05-13  0:51   ` Andrew Morton
2025-05-13  1:37     ` Jason Xing
2025-05-12  2:49 ` [PATCH v1 2/5] relayfs: introduce dump of relayfs statistics function Jason Xing
2025-05-13  0:51   ` Andrew Morton
2025-05-13  1:48     ` Jason Xing
2025-05-13  2:04       ` Andrew Morton
2025-05-13  2:26         ` Jason Xing
2025-05-13  3:41           ` Andrew Morton
2025-05-13  3:48             ` Jason Xing
2025-05-13  9:58   ` Masami Hiramatsu
2025-05-13 10:32     ` Jason Xing
2025-05-13 13:22       ` Masami Hiramatsu
2025-05-13 13:46         ` Jason Xing
2025-05-14  1:29           ` Masami Hiramatsu
2025-05-14  2:06             ` Jason Xing
2025-05-12  2:49 ` Jason Xing [this message]
2025-05-12  2:49 ` [PATCH v1 4/5] relayfs: support a counter tracking if data is too big to write Jason Xing
2025-05-12  2:49 ` [PATCH v1 5/5] relayfs: uniformally use possible cpu iteration Jason Xing
2025-05-13  0:52   ` Andrew Morton
2025-05-13  2:03     ` Jason Xing
2025-05-13  3:21       ` Andrew Morton
2025-05-13  3:25         ` Jason Xing
2025-05-13  5:52     ` Jason Xing

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=20250512024935.64704-4-kerneljasonxing@gmail.com \
    --to=kerneljasonxing@gmail.com \
    --cc=akpm@linux-foundation.org \
    --cc=axboe@kernel.dk \
    --cc=katrinzhou@tencent.com \
    --cc=kernelxing@tencent.com \
    --cc=linux-block@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-trace-kernel@vger.kernel.org \
    --cc=mathieu.desnoyers@efficios.com \
    --cc=mhiramat@kernel.org \
    --cc=rostedt@goodmis.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox