All of lore.kernel.org
 help / color / mirror / Atom feed
From: Sahitya Tummala <stummala@codeaurora.org>
To: Jaegeuk Kim <jaegeuk@kernel.org>, Chao Yu <yuchao0@huawei.com>,
	linux-f2fs-devel@lists.sourceforge.net
Cc: linux-kernel@vger.kernel.org
Subject: [f2fs-dev] [PATCH] f2fs: change to use rwsem for cp_mutex
Date: Mon, 23 Nov 2020 10:58:32 +0530	[thread overview]
Message-ID: <1606109312-1944-1-git-send-email-stummala@codeaurora.org> (raw)

Use rwsem to ensure serialization of the callers and to avoid
starvation of high priority tasks, when the system is under
heavy IO workload.

Signed-off-by: Sahitya Tummala <stummala@codeaurora.org>
---
 fs/f2fs/checkpoint.c | 8 ++++----
 fs/f2fs/f2fs.h       | 2 +-
 fs/f2fs/gc.c         | 4 ++--
 fs/f2fs/recovery.c   | 4 ++--
 fs/f2fs/super.c      | 2 +-
 5 files changed, 10 insertions(+), 10 deletions(-)

diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c
index 023462e..7fb8c80 100644
--- a/fs/f2fs/checkpoint.c
+++ b/fs/f2fs/checkpoint.c
@@ -348,13 +348,13 @@ static int f2fs_write_meta_pages(struct address_space *mapping,
 		goto skip_write;
 
 	/* if locked failed, cp will flush dirty pages instead */
-	if (!mutex_trylock(&sbi->cp_mutex))
+	if (!down_write_trylock(&sbi->cp_global_sem))
 		goto skip_write;
 
 	trace_f2fs_writepages(mapping->host, wbc, META);
 	diff = nr_pages_to_write(sbi, META, wbc);
 	written = f2fs_sync_meta_pages(sbi, META, wbc->nr_to_write, FS_META_IO);
-	mutex_unlock(&sbi->cp_mutex);
+	up_write(&sbi->cp_global_sem);
 	wbc->nr_to_write = max((long)0, wbc->nr_to_write - written - diff);
 	return 0;
 
@@ -1572,7 +1572,7 @@ int f2fs_write_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 		f2fs_warn(sbi, "Start checkpoint disabled!");
 	}
 	if (cpc->reason != CP_RESIZE)
-		mutex_lock(&sbi->cp_mutex);
+		down_write(&sbi->cp_global_sem);
 
 	if (!is_sbi_flag_set(sbi, SBI_IS_DIRTY) &&
 		((cpc->reason & CP_FASTBOOT) || (cpc->reason & CP_SYNC) ||
@@ -1647,7 +1647,7 @@ int f2fs_write_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 	trace_f2fs_write_checkpoint(sbi->sb, cpc->reason, "finish checkpoint");
 out:
 	if (cpc->reason != CP_RESIZE)
-		mutex_unlock(&sbi->cp_mutex);
+		up_write(&sbi->cp_global_sem);
 	return err;
 }
 
diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
index e082677..842c2ca 100644
--- a/fs/f2fs/f2fs.h
+++ b/fs/f2fs/f2fs.h
@@ -1369,7 +1369,7 @@ struct f2fs_sb_info {
 	int cur_cp_pack;			/* remain current cp pack */
 	spinlock_t cp_lock;			/* for flag in ckpt */
 	struct inode *meta_inode;		/* cache meta blocks */
-	struct mutex cp_mutex;			/* checkpoint procedure lock */
+	struct rw_semaphore cp_global_sem;	/* checkpoint procedure lock */
 	struct rw_semaphore cp_rwsem;		/* blocking FS operations */
 	struct rw_semaphore node_write;		/* locking node writes */
 	struct rw_semaphore node_change;	/* locking node change */
diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c
index 05641a1..3ef84e6 100644
--- a/fs/f2fs/gc.c
+++ b/fs/f2fs/gc.c
@@ -1986,7 +1986,7 @@ int f2fs_resize_fs(struct f2fs_sb_info *sbi, __u64 block_count)
 
 	freeze_super(sbi->sb);
 	down_write(&sbi->gc_lock);
-	mutex_lock(&sbi->cp_mutex);
+	down_write(&sbi->cp_global_sem);
 
 	spin_lock(&sbi->stat_lock);
 	if (shrunk_blocks + valid_user_blocks(sbi) +
@@ -2031,7 +2031,7 @@ int f2fs_resize_fs(struct f2fs_sb_info *sbi, __u64 block_count)
 		spin_unlock(&sbi->stat_lock);
 	}
 out_err:
-	mutex_unlock(&sbi->cp_mutex);
+	up_write(&sbi->cp_global_sem);
 	up_write(&sbi->gc_lock);
 	thaw_super(sbi->sb);
 	clear_sbi_flag(sbi, SBI_IS_RESIZEFS);
diff --git a/fs/f2fs/recovery.c b/fs/f2fs/recovery.c
index 0947d36..da75d5d 100644
--- a/fs/f2fs/recovery.c
+++ b/fs/f2fs/recovery.c
@@ -799,7 +799,7 @@ int f2fs_recover_fsync_data(struct f2fs_sb_info *sbi, bool check_only)
 	INIT_LIST_HEAD(&dir_list);
 
 	/* prevent checkpoint */
-	mutex_lock(&sbi->cp_mutex);
+	down_write(&sbi->cp_global_sem);
 
 	/* step #1: find fsynced inode numbers */
 	err = find_fsync_dnodes(sbi, &inode_list, check_only);
@@ -850,7 +850,7 @@ int f2fs_recover_fsync_data(struct f2fs_sb_info *sbi, bool check_only)
 	if (!err)
 		clear_sbi_flag(sbi, SBI_POR_DOING);
 
-	mutex_unlock(&sbi->cp_mutex);
+	up_write(&sbi->cp_global_sem);
 
 	/* let's drop all the directory inodes for clean checkpoint */
 	destroy_fsync_dnodes(&dir_list, err);
diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
index 87f7a6e..e33b0da 100644
--- a/fs/f2fs/super.c
+++ b/fs/f2fs/super.c
@@ -3552,7 +3552,7 @@ static int f2fs_fill_super(struct super_block *sb, void *data, int silent)
 	sbi->valid_super_block = valid_super_block;
 	init_rwsem(&sbi->gc_lock);
 	mutex_init(&sbi->writepages);
-	mutex_init(&sbi->cp_mutex);
+	init_rwsem(&sbi->cp_global_sem);
 	init_rwsem(&sbi->node_write);
 	init_rwsem(&sbi->node_change);
 
-- 
Qualcomm India Private Limited, on behalf of Qualcomm Innovation Center, Inc.
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, a Linux Foundation Collaborative Project.



_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel

WARNING: multiple messages have this Message-ID (diff)
From: Sahitya Tummala <stummala@codeaurora.org>
To: Jaegeuk Kim <jaegeuk@kernel.org>, Chao Yu <yuchao0@huawei.com>,
	linux-f2fs-devel@lists.sourceforge.net
Cc: Sahitya Tummala <stummala@codeaurora.org>, linux-kernel@vger.kernel.org
Subject: [PATCH] f2fs: change to use rwsem for cp_mutex
Date: Mon, 23 Nov 2020 10:58:32 +0530	[thread overview]
Message-ID: <1606109312-1944-1-git-send-email-stummala@codeaurora.org> (raw)

Use rwsem to ensure serialization of the callers and to avoid
starvation of high priority tasks, when the system is under
heavy IO workload.

Signed-off-by: Sahitya Tummala <stummala@codeaurora.org>
---
 fs/f2fs/checkpoint.c | 8 ++++----
 fs/f2fs/f2fs.h       | 2 +-
 fs/f2fs/gc.c         | 4 ++--
 fs/f2fs/recovery.c   | 4 ++--
 fs/f2fs/super.c      | 2 +-
 5 files changed, 10 insertions(+), 10 deletions(-)

diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c
index 023462e..7fb8c80 100644
--- a/fs/f2fs/checkpoint.c
+++ b/fs/f2fs/checkpoint.c
@@ -348,13 +348,13 @@ static int f2fs_write_meta_pages(struct address_space *mapping,
 		goto skip_write;
 
 	/* if locked failed, cp will flush dirty pages instead */
-	if (!mutex_trylock(&sbi->cp_mutex))
+	if (!down_write_trylock(&sbi->cp_global_sem))
 		goto skip_write;
 
 	trace_f2fs_writepages(mapping->host, wbc, META);
 	diff = nr_pages_to_write(sbi, META, wbc);
 	written = f2fs_sync_meta_pages(sbi, META, wbc->nr_to_write, FS_META_IO);
-	mutex_unlock(&sbi->cp_mutex);
+	up_write(&sbi->cp_global_sem);
 	wbc->nr_to_write = max((long)0, wbc->nr_to_write - written - diff);
 	return 0;
 
@@ -1572,7 +1572,7 @@ int f2fs_write_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 		f2fs_warn(sbi, "Start checkpoint disabled!");
 	}
 	if (cpc->reason != CP_RESIZE)
-		mutex_lock(&sbi->cp_mutex);
+		down_write(&sbi->cp_global_sem);
 
 	if (!is_sbi_flag_set(sbi, SBI_IS_DIRTY) &&
 		((cpc->reason & CP_FASTBOOT) || (cpc->reason & CP_SYNC) ||
@@ -1647,7 +1647,7 @@ int f2fs_write_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 	trace_f2fs_write_checkpoint(sbi->sb, cpc->reason, "finish checkpoint");
 out:
 	if (cpc->reason != CP_RESIZE)
-		mutex_unlock(&sbi->cp_mutex);
+		up_write(&sbi->cp_global_sem);
 	return err;
 }
 
diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
index e082677..842c2ca 100644
--- a/fs/f2fs/f2fs.h
+++ b/fs/f2fs/f2fs.h
@@ -1369,7 +1369,7 @@ struct f2fs_sb_info {
 	int cur_cp_pack;			/* remain current cp pack */
 	spinlock_t cp_lock;			/* for flag in ckpt */
 	struct inode *meta_inode;		/* cache meta blocks */
-	struct mutex cp_mutex;			/* checkpoint procedure lock */
+	struct rw_semaphore cp_global_sem;	/* checkpoint procedure lock */
 	struct rw_semaphore cp_rwsem;		/* blocking FS operations */
 	struct rw_semaphore node_write;		/* locking node writes */
 	struct rw_semaphore node_change;	/* locking node change */
diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c
index 05641a1..3ef84e6 100644
--- a/fs/f2fs/gc.c
+++ b/fs/f2fs/gc.c
@@ -1986,7 +1986,7 @@ int f2fs_resize_fs(struct f2fs_sb_info *sbi, __u64 block_count)
 
 	freeze_super(sbi->sb);
 	down_write(&sbi->gc_lock);
-	mutex_lock(&sbi->cp_mutex);
+	down_write(&sbi->cp_global_sem);
 
 	spin_lock(&sbi->stat_lock);
 	if (shrunk_blocks + valid_user_blocks(sbi) +
@@ -2031,7 +2031,7 @@ int f2fs_resize_fs(struct f2fs_sb_info *sbi, __u64 block_count)
 		spin_unlock(&sbi->stat_lock);
 	}
 out_err:
-	mutex_unlock(&sbi->cp_mutex);
+	up_write(&sbi->cp_global_sem);
 	up_write(&sbi->gc_lock);
 	thaw_super(sbi->sb);
 	clear_sbi_flag(sbi, SBI_IS_RESIZEFS);
diff --git a/fs/f2fs/recovery.c b/fs/f2fs/recovery.c
index 0947d36..da75d5d 100644
--- a/fs/f2fs/recovery.c
+++ b/fs/f2fs/recovery.c
@@ -799,7 +799,7 @@ int f2fs_recover_fsync_data(struct f2fs_sb_info *sbi, bool check_only)
 	INIT_LIST_HEAD(&dir_list);
 
 	/* prevent checkpoint */
-	mutex_lock(&sbi->cp_mutex);
+	down_write(&sbi->cp_global_sem);
 
 	/* step #1: find fsynced inode numbers */
 	err = find_fsync_dnodes(sbi, &inode_list, check_only);
@@ -850,7 +850,7 @@ int f2fs_recover_fsync_data(struct f2fs_sb_info *sbi, bool check_only)
 	if (!err)
 		clear_sbi_flag(sbi, SBI_POR_DOING);
 
-	mutex_unlock(&sbi->cp_mutex);
+	up_write(&sbi->cp_global_sem);
 
 	/* let's drop all the directory inodes for clean checkpoint */
 	destroy_fsync_dnodes(&dir_list, err);
diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
index 87f7a6e..e33b0da 100644
--- a/fs/f2fs/super.c
+++ b/fs/f2fs/super.c
@@ -3552,7 +3552,7 @@ static int f2fs_fill_super(struct super_block *sb, void *data, int silent)
 	sbi->valid_super_block = valid_super_block;
 	init_rwsem(&sbi->gc_lock);
 	mutex_init(&sbi->writepages);
-	mutex_init(&sbi->cp_mutex);
+	init_rwsem(&sbi->cp_global_sem);
 	init_rwsem(&sbi->node_write);
 	init_rwsem(&sbi->node_change);
 
-- 
Qualcomm India Private Limited, on behalf of Qualcomm Innovation Center, Inc.
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, a Linux Foundation Collaborative Project.


             reply	other threads:[~2020-11-23  5:29 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-11-23  5:28 Sahitya Tummala [this message]
2020-11-23  5:28 ` [PATCH] f2fs: change to use rwsem for cp_mutex Sahitya Tummala
2020-11-23 17:05 ` [f2fs-dev] " David Laight
2020-11-23 17:05   ` David Laight
2020-11-24  3:12   ` [f2fs-dev] " Chao Yu
2020-11-24  3:12     ` Chao Yu
2020-11-24  9:12     ` [f2fs-dev] " David Laight
2020-11-24  9:12       ` David Laight
2020-11-25 13:08       ` [f2fs-dev] " Sahitya Tummala
2020-11-25 13:08         ` Sahitya Tummala
2020-11-24  3:06 ` [f2fs-dev] " Chao Yu
2020-11-24  3:06   ` Chao Yu

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=1606109312-1944-1-git-send-email-stummala@codeaurora.org \
    --to=stummala@codeaurora.org \
    --cc=jaegeuk@kernel.org \
    --cc=linux-f2fs-devel@lists.sourceforge.net \
    --cc=linux-kernel@vger.kernel.org \
    --cc=yuchao0@huawei.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.