mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Ziyu Zhang <ziyuzhang201@gmail.com>
To: Jaegeuk Kim <jaegeuk@kernel.org>, Chao Yu <chao@kernel.org>
Cc: linux-f2fs-devel@lists.sourceforge.net,
	linux-kernel@vger.kernel.org, baijiaju1990@gmail.com,
	r33s3n6@gmail.com, gality369@gmail.com, zhenghaoran154@gmail.com,
	hanguidong02@gmail.com, zzzccc427@gmail.com,
	Ziyu Zhang <ziyuzhang201@gmail.com>
Subject: [PATCH] f2fs: make gc_wake test-and-clear atomic
Date: Sat, 16 May 2026 11:50:01 +0800	[thread overview]
Message-ID: <20260516035001.272335-1-ziyuzhang201@gmail.com> (raw)

gc_thread_func() tests gc_th->gc_wake and then clears it with
separate plain accesses. sysfs gc_urgent writes set the same flag and
wake the GC thread. If a sysfs writer stores true between the GC
thread's load and store, the later store false can clear the new wake
request.

Store gc_wake as an atomic_t. Use atomic_read() for the wait
condition, atomic_xchg(..., 0) in the GC thread, and atomic_set(..., 1)
in the sysfs trigger paths. This makes the consume-and-clear operation
atomic with respect to new wake requests: a set before the exchange is
consumed by the current iteration, while a set after the exchange stays
pending for the next wait.

Fixes: d9872a698c39 ("f2fs: introduce gc_urgent mode for background GC")
Signed-off-by: Ziyu Zhang <ziyuzhang201@gmail.com>
---
 fs/f2fs/gc.c    | 7 +++----
 fs/f2fs/gc.h    | 2 +-
 fs/f2fs/sysfs.c | 4 ++--
 3 files changed, 6 insertions(+), 7 deletions(-)

diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c
index 098e9f71421e..71e40e4083ad 100644
--- a/fs/f2fs/gc.c
+++ b/fs/f2fs/gc.c
@@ -49,15 +49,14 @@ static int gc_thread_func(void *data)
 		wait_event_freezable_timeout(*wq,
 				kthread_should_stop() ||
 				waitqueue_active(fggc_wq) ||
-				gc_th->gc_wake,
+				atomic_read(&gc_th->gc_wake),
 				msecs_to_jiffies(wait_ms));
 
 		if (test_opt(sbi, GC_MERGE) && waitqueue_active(fggc_wq))
 			foreground = true;
 
 		/* give it a try one time */
-		if (gc_th->gc_wake)
-			gc_th->gc_wake = false;
+		atomic_xchg(&gc_th->gc_wake, 0);
 
 		if (f2fs_readonly(sbi->sb)) {
 			stat_other_skip_bggc_count(sbi);
@@ -214,7 +213,7 @@ int f2fs_start_gc_thread(struct f2fs_sb_info *sbi)
 		gc_th->boost_zoned_gc_percent = 0;
 	}
 
-	gc_th->gc_wake = false;
+	atomic_set(&gc_th->gc_wake, 0);
 
 	sbi->gc_thread = gc_th;
 	init_waitqueue_head(&sbi->gc_thread->gc_wait_queue_head);
diff --git a/fs/f2fs/gc.h b/fs/f2fs/gc.h
index 24e8b1c27acc..65e5b062a0d3 100644
--- a/fs/f2fs/gc.h
+++ b/fs/f2fs/gc.h
@@ -56,7 +56,7 @@ struct f2fs_gc_kthread {
 	unsigned int no_gc_sleep_time;
 
 	/* for changing gc mode */
-	bool gc_wake;
+	atomic_t gc_wake;
 
 	/* for GC_MERGE mount option */
 	wait_queue_head_t fggc_wq;		/*
diff --git a/fs/f2fs/sysfs.c b/fs/f2fs/sysfs.c
index f736052dea50..6ca5943450d5 100644
--- a/fs/f2fs/sysfs.c
+++ b/fs/f2fs/sysfs.c
@@ -586,7 +586,7 @@ static ssize_t __sbi_store(struct f2fs_attr *a,
 		} else if (t == 1) {
 			sbi->gc_mode = GC_URGENT_HIGH;
 			if (sbi->gc_thread) {
-				sbi->gc_thread->gc_wake = true;
+				atomic_set(&sbi->gc_thread->gc_wake, 1);
 				wake_up_interruptible_all(
 					&sbi->gc_thread->gc_wait_queue_head);
 				wake_up_discard_thread(sbi, true);
@@ -596,7 +596,7 @@ static ssize_t __sbi_store(struct f2fs_attr *a,
 		} else if (t == 3) {
 			sbi->gc_mode = GC_URGENT_MID;
 			if (sbi->gc_thread) {
-				sbi->gc_thread->gc_wake = true;
+				atomic_set(&sbi->gc_thread->gc_wake, 1);
 				wake_up_interruptible_all(
 					&sbi->gc_thread->gc_wait_queue_head);
 			}
-- 
2.43.0


             reply	other threads:[~2026-05-16  3:50 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-05-16  3:50 Ziyu Zhang [this message]
2026-05-17  9:22 ` Chao Yu
2026-05-18  3:15   ` ziyu 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=20260516035001.272335-1-ziyuzhang201@gmail.com \
    --to=ziyuzhang201@gmail.com \
    --cc=baijiaju1990@gmail.com \
    --cc=chao@kernel.org \
    --cc=gality369@gmail.com \
    --cc=hanguidong02@gmail.com \
    --cc=jaegeuk@kernel.org \
    --cc=linux-f2fs-devel@lists.sourceforge.net \
    --cc=linux-kernel@vger.kernel.org \
    --cc=r33s3n6@gmail.com \
    --cc=zhenghaoran154@gmail.com \
    --cc=zzzccc427@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

all inboxes | Powered by JetHome®