linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Goldwyn Rodrigues <rgoldwyn@suse.de>
To: linux-fsdevel@vger.kernel.org
Cc: linux-btrfs@vger.kernel.org, david@fromorbit.com, hch@lst.de,
	johannes.thumshirn@wdc.com, dsterba@suse.com,
	darrick.wong@oracle.com, josef@toxicpanda.com,
	Goldwyn Rodrigues <rgoldwyn@suse.com>
Subject: [PATCH 12/14] btrfs: Remove dio_sem
Date: Thu, 24 Sep 2020 11:39:19 -0500	[thread overview]
Message-ID: <20200924163922.2547-13-rgoldwyn@suse.de> (raw)
In-Reply-To: <20200924163922.2547-1-rgoldwyn@suse.de>

From: Goldwyn Rodrigues <rgoldwyn@suse.com>

dio_sem can be eliminated because all DIO synchronization is performed
through inode->i_rwsem now.

Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
---
 fs/btrfs/btrfs_inode.h | 10 ----------
 fs/btrfs/file.c        | 13 -------------
 fs/btrfs/inode.c       |  1 -
 3 files changed, 24 deletions(-)

diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h
index 738009a22320..176abce59467 100644
--- a/fs/btrfs/btrfs_inode.h
+++ b/fs/btrfs/btrfs_inode.h
@@ -203,16 +203,6 @@ struct btrfs_inode {
 	/* Hook into fs_info->delayed_iputs */
 	struct list_head delayed_iput;
 
-	/*
-	 * To avoid races between lockless (i_mutex not held) direct IO writes
-	 * and concurrent fsync requests. Direct IO writes must acquire read
-	 * access on this semaphore for creating an extent map and its
-	 * corresponding ordered extent. The fast fsync path must acquire write
-	 * access on this semaphore before it collects ordered extents and
-	 * extent maps.
-	 */
-	struct rw_semaphore dio_sem;
-
 	struct inode vfs_inode;
 };
 
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index 83012d1e6f29..6854bf78d677 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -2021,8 +2021,6 @@ static ssize_t btrfs_direct_write(struct kiocb *iocb, struct iov_iter *from)
 		goto buffered;
 	}
 
-	down_read(&BTRFS_I(inode)->dio_sem);
-
 	/*
 	 * We have are actually a sync iocb, so we need our fancy endio to know
 	 * if we need to sync.
@@ -2037,7 +2035,6 @@ static ssize_t btrfs_direct_write(struct kiocb *iocb, struct iov_iter *from)
 	if (written == -ENOTBLK)
 		written = 0;
 
-	up_read(&BTRFS_I(inode)->dio_sem);
 	btrfs_inode_unlock(inode, ilock_flags);
 
 	if (written < 0 || !iov_iter_count(from)) {
@@ -2245,13 +2242,6 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 
 	inode_lock(inode);
 
-	/*
-	 * We take the dio_sem here because the tree log stuff can race with
-	 * lockless dio writes and get an extent map logged for an extent we
-	 * never waited on.  We need it this high up for lockdep reasons.
-	 */
-	down_write(&BTRFS_I(inode)->dio_sem);
-
 	atomic_inc(&root->log_batch);
 
 	/*
@@ -2282,7 +2272,6 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 	 */
 	ret = start_ordered_ops(inode, start, end);
 	if (ret) {
-		up_write(&BTRFS_I(inode)->dio_sem);
 		inode_unlock(inode);
 		goto out;
 	}
@@ -2379,7 +2368,6 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 	 * file again, but that will end up using the synchronization
 	 * inside btrfs_sync_log to keep things safe.
 	 */
-	up_write(&BTRFS_I(inode)->dio_sem);
 	inode_unlock(inode);
 
 	if (ret != BTRFS_NO_LOG_SYNC) {
@@ -2410,7 +2398,6 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 
 out_release_extents:
 	btrfs_release_log_ctx_extents(&ctx);
-	up_write(&BTRFS_I(inode)->dio_sem);
 	inode_unlock(inode);
 	goto out;
 }
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index cdde4422a6fa..17c97f30459c 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -8594,7 +8594,6 @@ struct inode *btrfs_alloc_inode(struct super_block *sb)
 	INIT_LIST_HEAD(&ei->delalloc_inodes);
 	INIT_LIST_HEAD(&ei->delayed_iput);
 	RB_CLEAR_NODE(&ei->rb_node);
-	init_rwsem(&ei->dio_sem);
 
 	return inode;
 }
-- 
2.26.2


  parent reply	other threads:[~2020-09-24 16:40 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-09-24 16:39 [PATCH 0/14 v3] BTRFS DIO inode locking/D_SYNC fix Goldwyn Rodrigues
2020-09-24 16:39 ` [PATCH 01/14] fs: remove dio_end_io() Goldwyn Rodrigues
2020-09-29 12:16   ` David Sterba
2020-09-24 16:39 ` [PATCH 02/14] btrfs: remove BTRFS_INODE_READDIO_NEED_LOCK Goldwyn Rodrigues
2020-09-24 16:39 ` [PATCH 03/14] iomap: Allow filesystem to call iomap_dio_complete without i_rwsem Goldwyn Rodrigues
2020-09-26  1:49   ` Darrick J. Wong
2020-09-24 16:39 ` [PATCH 04/14] iomap: Call inode_dio_end() before generic_write_sync() Goldwyn Rodrigues
2020-09-26  1:51   ` Darrick J. Wong
2020-09-28 15:04     ` David Sterba
2020-09-28 16:12       ` Darrick J. Wong
2020-09-24 16:39 ` [PATCH 05/14] btrfs: split btrfs_direct_IO to read and write Goldwyn Rodrigues
2020-09-24 16:39 ` [PATCH 06/14] btrfs: Move pos increment and pagecache extension to btrfs_buffered_write() Goldwyn Rodrigues
2020-09-24 16:39 ` [PATCH 07/14] btrfs: Move FS error state bit early during write Goldwyn Rodrigues
2020-09-24 16:39 ` [PATCH 08/14] btrfs: Introduce btrfs_write_check() Goldwyn Rodrigues
2020-10-09 14:21   ` Josef Bacik
2020-09-24 16:39 ` [PATCH 09/14] btrfs: Introduce btrfs_inode_lock()/unlock() Goldwyn Rodrigues
2020-09-24 16:39 ` [PATCH 10/14] btrfs: Push inode locking and unlocking into buffered/direct write Goldwyn Rodrigues
2020-09-24 16:39 ` [PATCH 11/14] btrfs: Use inode_lock_shared() for direct writes within EOF Goldwyn Rodrigues
2020-10-09 14:25   ` Josef Bacik
2020-09-24 16:39 ` Goldwyn Rodrigues [this message]
2020-09-24 16:39 ` [PATCH 13/14] btrfs: Call iomap_dio_complete() without inode_lock Goldwyn Rodrigues
2020-09-24 16:39 ` [PATCH 14/14] btrfs: Revert 09745ff88d93 ("btrfs: dio iomap DSYNC workaround") Goldwyn Rodrigues

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=20200924163922.2547-13-rgoldwyn@suse.de \
    --to=rgoldwyn@suse.de \
    --cc=darrick.wong@oracle.com \
    --cc=david@fromorbit.com \
    --cc=dsterba@suse.com \
    --cc=hch@lst.de \
    --cc=johannes.thumshirn@wdc.com \
    --cc=josef@toxicpanda.com \
    --cc=linux-btrfs@vger.kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=rgoldwyn@suse.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;
as well as URLs for NNTP newsgroup(s).