[PATCH 0/15 v2] BTRFS DIO inode locking/D_SYNC fix

[PATCH 0/15 v2] BTRFS DIO inode locking/D_SYNC fix

[Goldwyn Rodrigues]

I have merged two series here:
(1) Using inode_lock_shared for <EOF DIO writes
(2) Fix O_DSYNC | O_DIRECT

Both require inode locking to be pushed closer to I/O calls and thought
was relevant. (2) requires changes in iomap code. If there is a need to 
separate the two please let me know. I don't have any ulterior motives
here ;)

Testing: xfstests on both btrfs and xfs

Git: https://github.com/goldwynr/linux/tree/btrfs-inode-lock

Changes since v1:

 - Changed fix for deadlock due to O_DSYNC (iomap patches added)
 - btrfs_inode_lock() shifted to inode.c
 - Reinstated lockdep_assert_held() in iomap_dio_rw()

-- 
Goldwyn

 fs/btrfs/btrfs_inode.h |   28 --
 fs/btrfs/ctree.h       |   13 +
 fs/btrfs/file.c        |  497 +++++++++++++++++++++++++++----------------------
 fs/btrfs/inode.c       |  176 +++--------------
 fs/btrfs/transaction.h |    1
 fs/direct-io.c         |   19 -
 fs/iomap/direct-io.c   |   38 ++-
 include/linux/fs.h     |    2
 include/linux/iomap.h  |    5
 9 files changed, 355 insertions(+), 424 deletions(-)

[PATCH 01/15] fs: remove dio_end_io()

[Goldwyn Rodrigues]

From: Goldwyn Rodrigues <rgoldwyn@suse.com>

Since we removed the last user of dio_end_io(), remove the helper
function dio_end_io().

Reviewed-by: Nikolay Borisov <nborisov@suse.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
---
 fs/direct-io.c     | 19 -------------------
 include/linux/fs.h |  2 --
 2 files changed, 21 deletions(-)

diff --git a/fs/direct-io.c b/fs/direct-io.c
index 183299892465..abf535b036ab 100644
--- a/fs/direct-io.c
+++ b/fs/direct-io.c
@@ -386,25 +386,6 @@ static void dio_bio_end_io(struct bio *bio)
 	spin_unlock_irqrestore(&dio->bio_lock, flags);
 }
 
-/**
- * dio_end_io - handle the end io action for the given bio
- * @bio: The direct io bio thats being completed
- *
- * This is meant to be called by any filesystem that uses their own dio_submit_t
- * so that the DIO specific endio actions are dealt with after the filesystem
- * has done it's completion work.
- */
-void dio_end_io(struct bio *bio)
-{
-	struct dio *dio = bio->bi_private;
-
-	if (dio->is_async)
-		dio_bio_end_aio(bio);
-	else
-		dio_bio_end_io(bio);
-}
-EXPORT_SYMBOL_GPL(dio_end_io);
-
 static inline void
 dio_bio_alloc(struct dio *dio, struct dio_submit *sdio,
 	      struct block_device *bdev,
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 7519ae003a08..8e2842a9c0b3 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -3079,8 +3079,6 @@ enum {
 	DIO_SKIP_HOLES	= 0x02,
 };
 
-void dio_end_io(struct bio *bio);
-
 ssize_t __blockdev_direct_IO(struct kiocb *iocb, struct inode *inode,
 			     struct block_device *bdev, struct iov_iter *iter,
 			     get_block_t get_block,
-- 
2.26.2

[PATCH 02/15] btrfs: remove BTRFS_INODE_READDIO_NEED_LOCK

[Goldwyn Rodrigues]

From: Goldwyn Rodrigues <rgoldwyn@suse.com>

Since we now perform direct reads using i_rwsem, we can remove this
inode flag used to co-ordinate unlocked reads.

The truncate call takes i_rwsem. This means it is correctly synchronized
with concurrent direct reads.

Reviewed-by: Nikolay Borisov <nborisov@suse.com>
Reviewed-by: Johannes Thumshirn <jth@kernel.org>
Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
---
 fs/btrfs/btrfs_inode.h | 18 ------------------
 fs/btrfs/inode.c       |  3 ---
 2 files changed, 21 deletions(-)

diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h
index 6fdb46d58299..738009a22320 100644
--- a/fs/btrfs/btrfs_inode.h
+++ b/fs/btrfs/btrfs_inode.h
@@ -33,7 +33,6 @@ enum {
 	BTRFS_INODE_NEEDS_FULL_SYNC,
 	BTRFS_INODE_COPY_EVERYTHING,
 	BTRFS_INODE_IN_DELALLOC_LIST,
-	BTRFS_INODE_READDIO_NEED_LOCK,
 	BTRFS_INODE_HAS_PROPS,
 	BTRFS_INODE_SNAPSHOT_FLUSH,
 };
@@ -334,23 +333,6 @@ struct btrfs_dio_private {
 	u8 csums[];
 };
 
-/*
- * Disable DIO read nolock optimization, so new dio readers will be forced
- * to grab i_mutex. It is used to avoid the endless truncate due to
- * nonlocked dio read.
- */
-static inline void btrfs_inode_block_unlocked_dio(struct btrfs_inode *inode)
-{
-	set_bit(BTRFS_INODE_READDIO_NEED_LOCK, &inode->runtime_flags);
-	smp_mb();
-}
-
-static inline void btrfs_inode_resume_unlocked_dio(struct btrfs_inode *inode)
-{
-	smp_mb__before_atomic();
-	clear_bit(BTRFS_INODE_READDIO_NEED_LOCK, &inode->runtime_flags);
-}
-
 /* Array of bytes with variable length, hexadecimal format 0x1234 */
 #define CSUM_FMT				"0x%*phN"
 #define CSUM_FMT_VALUE(size, bytes)		size, bytes
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 18d171b2d544..3db4697ff1ba 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -4855,10 +4855,7 @@ static int btrfs_setsize(struct inode *inode, struct iattr *attr)
 
 		truncate_setsize(inode, newsize);
 
-		/* Disable nonlocked read DIO to avoid the endless truncate */
-		btrfs_inode_block_unlocked_dio(BTRFS_I(inode));
 		inode_dio_wait(inode);
-		btrfs_inode_resume_unlocked_dio(BTRFS_I(inode));
 
 		ret = btrfs_truncate(inode, newsize == oldsize);
 		if (ret && inode->i_nlink) {
-- 
2.26.2

[PATCH 03/15] iomap: Allow filesystem to call iomap_dio_complete without i_rwsem

[Goldwyn Rodrigues]

From: Christoph Hellwig <hch@lst.de>

This is to avoid the deadlock caused in btrfs because of O_DIRECT |
O_DSYNC.

Filesystems such as btrfs require i_rwsem while performing sync on a
file. iomap_dio_rw() is called under i_rw_sem. This leads to a
deadlock because of:

iomap_dio_complete()
  generic_write_sync()
    btrfs_sync_file()

Separate out iomap_dio_complete() from iomap_dio_rw(), so filesystems
can call iomap_dio_complete() after unlocking i_rwsem.

Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
---
 fs/iomap/direct-io.c  | 34 +++++++++++++++++++++++++---------
 include/linux/iomap.h |  5 +++++
 2 files changed, 30 insertions(+), 9 deletions(-)

diff --git a/fs/iomap/direct-io.c b/fs/iomap/direct-io.c
index c1aafb2ab990..d970c6bbbe11 100644
--- a/fs/iomap/direct-io.c
+++ b/fs/iomap/direct-io.c
@@ -76,7 +76,7 @@ static void iomap_dio_submit_bio(struct iomap_dio *dio, struct iomap *iomap,
 		dio->submit.cookie = submit_bio(bio);
 }
 
-static ssize_t iomap_dio_complete(struct iomap_dio *dio)
+ssize_t iomap_dio_complete(struct iomap_dio *dio)
 {
 	const struct iomap_dio_ops *dops = dio->dops;
 	struct kiocb *iocb = dio->iocb;
@@ -130,6 +130,7 @@ static ssize_t iomap_dio_complete(struct iomap_dio *dio)
 
 	return ret;
 }
+EXPORT_SYMBOL_GPL(iomap_dio_complete);
 
 static void iomap_dio_complete_work(struct work_struct *work)
 {
@@ -406,8 +407,8 @@ iomap_dio_actor(struct inode *inode, loff_t pos, loff_t length,
  * Returns -ENOTBLK In case of a page invalidation invalidation failure for
  * writes.  The callers needs to fall back to buffered I/O in this case.
  */
-ssize_t
-iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
+struct iomap_dio *
+__iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
 		const struct iomap_ops *ops, const struct iomap_dio_ops *dops,
 		bool wait_for_completion)
 {
@@ -421,14 +422,14 @@ iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
 	struct iomap_dio *dio;
 
 	if (!count)
-		return 0;
+		return NULL;
 
 	if (WARN_ON(is_sync_kiocb(iocb) && !wait_for_completion))
-		return -EIO;
+		return ERR_PTR(-EIO);
 
 	dio = kmalloc(sizeof(*dio), GFP_KERNEL);
 	if (!dio)
-		return -ENOMEM;
+		return ERR_PTR(-ENOMEM);
 
 	dio->iocb = iocb;
 	atomic_set(&dio->ref, 1);
@@ -558,7 +559,7 @@ iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
 	dio->wait_for_completion = wait_for_completion;
 	if (!atomic_dec_and_test(&dio->ref)) {
 		if (!wait_for_completion)
-			return -EIOCBQUEUED;
+			return ERR_PTR(-EIOCBQUEUED);
 
 		for (;;) {
 			set_current_state(TASK_UNINTERRUPTIBLE);
@@ -574,10 +575,25 @@ iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
 		__set_current_state(TASK_RUNNING);
 	}
 
-	return iomap_dio_complete(dio);
+	return dio;
 
 out_free_dio:
 	kfree(dio);
-	return ret;
+	return ERR_PTR(ret);
+}
+EXPORT_SYMBOL_GPL(__iomap_dio_rw);
+
+ssize_t
+iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
+		const struct iomap_ops *ops, const struct iomap_dio_ops *dops,
+		bool wait_for_completion)
+{
+	struct iomap_dio *dio;
+
+	dio = __iomap_dio_rw(iocb, iter, ops, dops, wait_for_completion);
+	if (IS_ERR_OR_NULL(dio))
+		return PTR_ERR_OR_ZERO(dio);
+	return iomap_dio_complete(dio);
 }
 EXPORT_SYMBOL_GPL(iomap_dio_rw);
+
diff --git a/include/linux/iomap.h b/include/linux/iomap.h
index 4d1d3c3469e9..172b3397a1a3 100644
--- a/include/linux/iomap.h
+++ b/include/linux/iomap.h
@@ -13,6 +13,7 @@
 struct address_space;
 struct fiemap_extent_info;
 struct inode;
+struct iomap_dio;
 struct iomap_writepage_ctx;
 struct iov_iter;
 struct kiocb;
@@ -258,6 +259,10 @@ struct iomap_dio_ops {
 ssize_t iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
 		const struct iomap_ops *ops, const struct iomap_dio_ops *dops,
 		bool wait_for_completion);
+struct iomap_dio *__iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
+		const struct iomap_ops *ops, const struct iomap_dio_ops *dops,
+		bool wait_for_completion);
+ssize_t iomap_dio_complete(struct iomap_dio *dio);
 int iomap_dio_iopoll(struct kiocb *kiocb, bool spin);
 
 #ifdef CONFIG_SWAP
-- 
2.26.2

[PATCH 04/15] iomap: Call inode_dio_end() before generic_write_sync()

[Goldwyn Rodrigues]

From: Goldwyn Rodrigues <rgoldwyn@suse.com>

iomap complete routine can deadlock with btrfs_fallocate because of the
call to generic_write_sync().

P0                      P1
inode_lock()            fallocate(FALLOC_FL_ZERO_RANGE)
__iomap_dio_rw()        inode_lock()
                        <block>
<submits IO>
<completes IO>
inode_unlock()
                        <gets inode_lock()>
                        inode_dio_wait()
iomap_dio_complete()
  generic_write_sync()
    btrfs_file_fsync()
      inode_lock()
      <deadlock>

inode_dio_end() is used to notify the end of DIO data in order
to synchronize with truncate. Call inode_dio_end() before calling
generic_write_sync(), so filesystems can lock i_rwsem during a sync.

---
 fs/iomap/direct-io.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/iomap/direct-io.c b/fs/iomap/direct-io.c
index d970c6bbbe11..e01f81e7b76f 100644
--- a/fs/iomap/direct-io.c
+++ b/fs/iomap/direct-io.c
@@ -118,6 +118,7 @@ ssize_t iomap_dio_complete(struct iomap_dio *dio)
 			dio_warn_stale_pagecache(iocb->ki_filp);
 	}
 
+	inode_dio_end(file_inode(iocb->ki_filp));
 	/*
 	 * If this is a DSYNC write, make sure we push it to stable storage now
 	 * that we've written data.
@@ -125,7 +126,6 @@ ssize_t iomap_dio_complete(struct iomap_dio *dio)
 	if (ret > 0 && (dio->flags & IOMAP_DIO_NEED_SYNC))
 		ret = generic_write_sync(iocb, ret);
 
-	inode_dio_end(file_inode(iocb->ki_filp));
 	kfree(dio);
 
 	return ret;
-- 
2.26.2

[PATCH 05/15] btrfs: split btrfs_direct_IO to read and write

[Goldwyn Rodrigues]

From: Goldwyn Rodrigues <rgoldwyn@suse.com>

The read and write DIO don't have anything in common except for the
call to iomap_dio_rw. Extract the write call into a new function to get
rid of conditional statements for direct write.

Originally proposed by Christoph Hellwig <hch@lst.de>

Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
---
 fs/btrfs/ctree.h |  5 ++-
 fs/btrfs/file.c  | 98 +++++++++++++++++++++++++++++++++++++++++++-----
 fs/btrfs/inode.c | 91 ++------------------------------------------
 3 files changed, 95 insertions(+), 99 deletions(-)

diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index cd644c755142..b47a8dcff028 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -28,6 +28,7 @@
 #include <linux/dynamic_debug.h>
 #include <linux/refcount.h>
 #include <linux/crc32c.h>
+#include <linux/iomap.h>
 #include "extent-io-tree.h"
 #include "extent_io.h"
 #include "extent_map.h"
@@ -3048,7 +3049,9 @@ int btrfs_writepage_cow_fixup(struct page *page, u64 start, u64 end);
 void btrfs_writepage_endio_finish_ordered(struct page *page, u64 start,
 					  u64 end, int uptodate);
 extern const struct dentry_operations btrfs_dentry_operations;
-ssize_t btrfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter);
+extern const struct iomap_ops btrfs_dio_iomap_ops;
+extern const struct iomap_dio_ops btrfs_dio_ops;
+extern const struct iomap_dio_ops btrfs_sync_dops;
 
 /* ioctl.c */
 long btrfs_ioctl(struct file *file, unsigned int cmd, unsigned long arg);
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index 038e0afaf3d0..910e2fd234a9 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -1855,21 +1855,68 @@ static noinline ssize_t btrfs_buffered_write(struct kiocb *iocb,
 	return num_written ? num_written : ret;
 }
 
-static ssize_t __btrfs_direct_write(struct kiocb *iocb, struct iov_iter *from)
+static ssize_t check_direct_IO(struct btrfs_fs_info *fs_info,
+		const struct iov_iter *iter, loff_t offset)
+{
+	const unsigned int blocksize_mask = fs_info->sectorsize - 1;
+
+	if (offset & blocksize_mask)
+		return -EINVAL;
+
+	if (iov_iter_alignment(iter) & blocksize_mask)
+		return -EINVAL;
+
+	return 0;
+}
+
+static ssize_t btrfs_direct_write(struct kiocb *iocb, struct iov_iter *from)
 {
 	struct file *file = iocb->ki_filp;
 	struct inode *inode = file_inode(file);
-	loff_t pos;
-	ssize_t written;
+	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
+	loff_t pos = iocb->ki_pos;
+	ssize_t written = 0;
+	bool relock = false;
 	ssize_t written_buffered;
 	loff_t endbyte;
 	int err;
 
-	written = btrfs_direct_IO(iocb, from);
+	if (check_direct_IO(fs_info, from, pos))
+		goto buffered;
+
+	/*
+	 * If the write DIO is beyond the EOF, we need update
+	 * the isize, but it is protected by i_mutex. So we can
+	 * not unlock the i_mutex at this case.
+	 */
+	if (pos + iov_iter_count(from) <= inode->i_size) {
+		inode_unlock(inode);
+		relock = true;
+	}
+	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.
+	 */
+	if (current->journal_info)
+		written = iomap_dio_rw(iocb, from, &btrfs_dio_iomap_ops,
+				&btrfs_sync_dops, is_sync_kiocb(iocb));
+	else
+		written = iomap_dio_rw(iocb, from, &btrfs_dio_iomap_ops,
+				&btrfs_dio_ops, is_sync_kiocb(iocb));
+
+	if (written == -ENOTBLK)
+		written = 0;
+
+	up_read(&BTRFS_I(inode)->dio_sem);
+	if (relock)
+		inode_lock(inode);
 
 	if (written < 0 || !iov_iter_count(from))
 		return written;
 
+buffered:
 	pos = iocb->ki_pos;
 	written_buffered = btrfs_buffered_write(iocb, from);
 	if (written_buffered < 0) {
@@ -2043,7 +2090,7 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb,
 			iocb->ki_flags &= ~IOCB_DSYNC;
 			current->journal_info = BTRFS_DIO_SYNC_STUB;
 		}
-		num_written = __btrfs_direct_write(iocb, from);
+		num_written = btrfs_direct_write(iocb, from);
 
 		/*
 		 * As stated above, we cleared journal_info, so we need to do
@@ -3618,16 +3665,47 @@ static int btrfs_file_open(struct inode *inode, struct file *filp)
 	return generic_file_open(inode, filp);
 }
 
+static int check_direct_read(struct btrfs_fs_info *fs_info,
+		const struct iov_iter *iter, loff_t offset)
+{
+	int ret;
+	int i, seg;
+
+	ret = check_direct_IO(fs_info, iter, offset);
+	if (ret < 0)
+		return ret;
+
+	if (!iter_is_iovec(iter))
+		return 0;
+
+	for (seg = 0; seg < iter->nr_segs; seg++)
+		for (i = seg + 1; i < iter->nr_segs; i++)
+			if (iter->iov[seg].iov_base == iter->iov[i].iov_base)
+				return -EINVAL;
+	return 0;
+}
+
+static ssize_t btrfs_direct_read(struct kiocb *iocb, struct iov_iter *to)
+{
+	struct inode *inode = file_inode(iocb->ki_filp);
+	ssize_t ret;
+
+	if (check_direct_read(btrfs_sb(inode->i_sb), to, iocb->ki_pos))
+		return 0;
+
+	inode_lock_shared(inode);
+	ret = iomap_dio_rw(iocb, to, &btrfs_dio_iomap_ops, &btrfs_dio_ops,
+			is_sync_kiocb(iocb));
+	inode_unlock_shared(inode);
+	return ret;
+}
+
 static ssize_t btrfs_file_read_iter(struct kiocb *iocb, struct iov_iter *to)
 {
 	ssize_t ret = 0;
 
 	if (iocb->ki_flags & IOCB_DIRECT) {
-		struct inode *inode = file_inode(iocb->ki_filp);
-
-		inode_lock_shared(inode);
-		ret = btrfs_direct_IO(iocb, to);
-		inode_unlock_shared(inode);
+		ret = btrfs_direct_read(iocb, to);
 		if (ret < 0)
 			return ret;
 	}
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 3db4697ff1ba..0730131b6590 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -7898,39 +7898,6 @@ static blk_qc_t btrfs_submit_direct(struct inode *inode, struct iomap *iomap,
 	return BLK_QC_T_NONE;
 }
 
-static ssize_t check_direct_IO(struct btrfs_fs_info *fs_info,
-			       const struct iov_iter *iter, loff_t offset)
-{
-	int seg;
-	int i;
-	unsigned int blocksize_mask = fs_info->sectorsize - 1;
-	ssize_t retval = -EINVAL;
-
-	if (offset & blocksize_mask)
-		goto out;
-
-	if (iov_iter_alignment(iter) & blocksize_mask)
-		goto out;
-
-	/* If this is a write we don't need to check anymore */
-	if (iov_iter_rw(iter) != READ || !iter_is_iovec(iter))
-		return 0;
-	/*
-	 * Check to make sure we don't have duplicate iov_base's in this
-	 * iovec, if so return EINVAL, otherwise we'll get csum errors
-	 * when reading back.
-	 */
-	for (seg = 0; seg < iter->nr_segs; seg++) {
-		for (i = seg + 1; i < iter->nr_segs; i++) {
-			if (iter->iov[seg].iov_base == iter->iov[i].iov_base)
-				goto out;
-		}
-	}
-	retval = 0;
-out:
-	return retval;
-}
-
 static inline int btrfs_maybe_fsync_end_io(struct kiocb *iocb, ssize_t size,
 					   int error, unsigned flags)
 {
@@ -7955,72 +7922,20 @@ static inline int btrfs_maybe_fsync_end_io(struct kiocb *iocb, ssize_t size,
 	return 0;
 }
 
-static const struct iomap_ops btrfs_dio_iomap_ops = {
+const struct iomap_ops btrfs_dio_iomap_ops = {
 	.iomap_begin            = btrfs_dio_iomap_begin,
 	.iomap_end              = btrfs_dio_iomap_end,
 };
 
-static const struct iomap_dio_ops btrfs_dio_ops = {
+const struct iomap_dio_ops btrfs_dio_ops = {
 	.submit_io		= btrfs_submit_direct,
 };
 
-static const struct iomap_dio_ops btrfs_sync_dops = {
+const struct iomap_dio_ops btrfs_sync_dops = {
 	.submit_io		= btrfs_submit_direct,
 	.end_io			= btrfs_maybe_fsync_end_io,
 };
 
-ssize_t btrfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter)
-{
-	struct file *file = iocb->ki_filp;
-	struct inode *inode = file->f_mapping->host;
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
-	struct extent_changeset *data_reserved = NULL;
-	loff_t offset = iocb->ki_pos;
-	size_t count = 0;
-	bool relock = false;
-	ssize_t ret;
-
-	if (check_direct_IO(fs_info, iter, offset))
-		return 0;
-
-	count = iov_iter_count(iter);
-	if (iov_iter_rw(iter) == WRITE) {
-		/*
-		 * If the write DIO is beyond the EOF, we need update
-		 * the isize, but it is protected by i_mutex. So we can
-		 * not unlock the i_mutex at this case.
-		 */
-		if (offset + count <= inode->i_size) {
-			inode_unlock(inode);
-			relock = true;
-		}
-		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.
-	 */
-	if (current->journal_info)
-		ret = iomap_dio_rw(iocb, iter, &btrfs_dio_iomap_ops,
-				   &btrfs_sync_dops, is_sync_kiocb(iocb));
-	else
-		ret = iomap_dio_rw(iocb, iter, &btrfs_dio_iomap_ops,
-				   &btrfs_dio_ops, is_sync_kiocb(iocb));
-
-	if (ret == -ENOTBLK)
-		ret = 0;
-
-	if (iov_iter_rw(iter) == WRITE)
-		up_read(&BTRFS_I(inode)->dio_sem);
-
-	if (relock)
-		inode_lock(inode);
-
-	extent_changeset_free(data_reserved);
-	return ret;
-}
-
 static int btrfs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 			u64 start, u64 len)
 {
-- 
2.26.2

[PATCH 06/15] btrfs: Move pos increment and pagecache extension to btrfs_buffered_write()

[Goldwyn Rodrigues]

From: Goldwyn Rodrigues <rgoldwyn@suse.com>

While we do this, correct the call to pagecache_isize_extended():
 - pagecache_isisze_extended needs to be called to the starting of the
   write as opposed to i_size
 - We don't need to check range before the call, this is done in the
   function

Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
---
 fs/btrfs/file.c | 13 +++++--------
 1 file changed, 5 insertions(+), 8 deletions(-)

diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index 910e2fd234a9..4c40a2742aab 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -1632,6 +1632,7 @@ static noinline ssize_t btrfs_buffered_write(struct kiocb *iocb,
 	int ret = 0;
 	bool only_release_metadata = false;
 	bool force_page_uptodate = false;
+	loff_t old_isize = i_size_read(inode);
 
 	nrptrs = min(DIV_ROUND_UP(iov_iter_count(i), PAGE_SIZE),
 			PAGE_SIZE / (sizeof(struct page *)));
@@ -1852,6 +1853,10 @@ static noinline ssize_t btrfs_buffered_write(struct kiocb *iocb,
 	}
 
 	extent_changeset_free(data_reserved);
+	if (num_written > 0) {
+		pagecache_isize_extended(inode, old_isize, iocb->ki_pos);
+		iocb->ki_pos += num_written;
+	}
 	return num_written ? num_written : ret;
 }
 
@@ -1975,7 +1980,6 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb,
 	loff_t pos;
 	size_t count;
 	loff_t oldsize;
-	int clean_page = 0;
 
 	if (!(iocb->ki_flags & IOCB_DIRECT) &&
 	    (iocb->ki_flags & IOCB_NOWAIT))
@@ -2057,8 +2061,6 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb,
 			inode_unlock(inode);
 			goto out;
 		}
-		if (start_pos > round_up(oldsize, fs_info->sectorsize))
-			clean_page = 1;
 	}
 
 	if (sync)
@@ -2101,11 +2103,6 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb,
 		current->journal_info = NULL;
 	} else {
 		num_written = btrfs_buffered_write(iocb, from);
-		if (num_written > 0)
-			iocb->ki_pos = pos + num_written;
-		if (clean_page)
-			pagecache_isize_extended(inode, oldsize,
-						i_size_read(inode));
 	}
 
 	inode_unlock(inode);
-- 
2.26.2

[PATCH 07/15] btrfs: Move FS error state bit early during write

[Goldwyn Rodrigues]

From: Goldwyn Rodrigues <rgoldwyn@suse.com>

fs_info->fs_state is a filesystem bit check as opposed to inode
and can be performed before we begin with write checks. This eliminates
inode lock/unlock in case of error bit is set.

Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
---
 fs/btrfs/file.c | 21 +++++++++------------
 1 file changed, 9 insertions(+), 12 deletions(-)

diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index 4c40a2742aab..ca374cb5ffc9 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -1981,6 +1981,15 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb,
 	size_t count;
 	loff_t oldsize;
 
+	/*
+	 * If BTRFS flips readonly due to some impossible error
+	 * (fs_info->fs_state now has BTRFS_SUPER_FLAG_ERROR),
+	 * although we have opened a file as writable, we have
+	 * to stop this write operation to ensure FS consistency.
+	 */
+	if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state))
+		return -EROFS;
+
 	if (!(iocb->ki_flags & IOCB_DIRECT) &&
 	    (iocb->ki_flags & IOCB_NOWAIT))
 		return -EOPNOTSUPP;
@@ -2030,18 +2039,6 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb,
 		goto out;
 	}
 
-	/*
-	 * If BTRFS flips readonly due to some impossible error
-	 * (fs_info->fs_state now has BTRFS_SUPER_FLAG_ERROR),
-	 * although we have opened a file as writable, we have
-	 * to stop this write operation to ensure FS consistency.
-	 */
-	if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state)) {
-		inode_unlock(inode);
-		err = -EROFS;
-		goto out;
-	}
-
 	/*
 	 * We reserve space for updating the inode when we reserve space for the
 	 * extent we are going to write, so we will enospc out there.  We don't
-- 
2.26.2

[PATCH 08/15] btrfs: Introduce btrfs_write_check()

[Goldwyn Rodrigues]

From: Goldwyn Rodrigues <rgoldwyn@suse.com>

btrfs_write_check() checks for all parameters in one place before
beginning a write. This does away with inode_unlock() after every check.
In the later patches, it will help push inode_lock/unlock() in buffered
and direct write functions respectively.

Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
---
 fs/btrfs/file.c | 159 ++++++++++++++++++++++++------------------------
 1 file changed, 81 insertions(+), 78 deletions(-)

diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index ca374cb5ffc9..0f961ce1fa98 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -1615,6 +1615,85 @@ void btrfs_check_nocow_unlock(struct btrfs_inode *inode)
 	btrfs_drew_write_unlock(&inode->root->snapshot_lock);
 }
 
+static void update_time_for_write(struct inode *inode)
+{
+	struct timespec64 now;
+
+	if (IS_NOCMTIME(inode))
+		return;
+
+	now = current_time(inode);
+	if (!timespec64_equal(&inode->i_mtime, &now))
+		inode->i_mtime = now;
+
+	if (!timespec64_equal(&inode->i_ctime, &now))
+		inode->i_ctime = now;
+
+	if (IS_I_VERSION(inode))
+		inode_inc_iversion(inode);
+}
+
+static size_t btrfs_write_check(struct kiocb *iocb, struct iov_iter *from)
+{
+	struct file *file = iocb->ki_filp;
+	struct inode *inode = file_inode(file);
+	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
+	loff_t pos = iocb->ki_pos;
+	size_t count = iov_iter_count(from);
+	int err;
+	loff_t oldsize;
+	loff_t start_pos;
+
+	err = generic_write_checks(iocb, from);
+	if (err <= 0)
+		return err;
+
+	if (iocb->ki_flags & IOCB_NOWAIT) {
+		size_t nocow_bytes = count;
+
+		/*
+		 * We will allocate space in case nodatacow is not set,
+		 * so bail
+		 */
+		if (check_nocow_nolock(BTRFS_I(inode), pos, &nocow_bytes)
+		    <= 0)
+			return -EAGAIN;
+		/*
+		 * There are holes in the range or parts of the range that must
+		 * be COWed (shared extents, RO block groups, etc), so just bail
+		 * out.
+		 */
+		if (nocow_bytes < count)
+			return -EAGAIN;
+	}
+
+	current->backing_dev_info = inode_to_bdi(inode);
+	err = file_remove_privs(file);
+	if (err)
+		return err;
+
+	/*
+	 * We reserve space for updating the inode when we reserve space for the
+	 * extent we are going to write, so we will enospc out there.  We don't
+	 * need to start yet another transaction to update the inode as we will
+	 * update the inode when we finish writing whatever data we write.
+	 */
+	update_time_for_write(inode);
+
+	start_pos = round_down(pos, fs_info->sectorsize);
+	oldsize = i_size_read(inode);
+	if (start_pos > oldsize) {
+		/* Expand hole size to cover write data, preventing empty gap */
+		loff_t end_pos = round_up(pos + count,
+					  fs_info->sectorsize);
+		err = btrfs_cont_expand(inode, oldsize, end_pos);
+		if (err)
+			return err;
+	}
+
+	return count;
+}
+
 static noinline ssize_t btrfs_buffered_write(struct kiocb *iocb,
 					       struct iov_iter *i)
 {
@@ -1947,24 +2026,6 @@ static ssize_t btrfs_direct_write(struct kiocb *iocb, struct iov_iter *from)
 	return written ? written : err;
 }
 
-static void update_time_for_write(struct inode *inode)
-{
-	struct timespec64 now;
-
-	if (IS_NOCMTIME(inode))
-		return;
-
-	now = current_time(inode);
-	if (!timespec64_equal(&inode->i_mtime, &now))
-		inode->i_mtime = now;
-
-	if (!timespec64_equal(&inode->i_ctime, &now))
-		inode->i_ctime = now;
-
-	if (IS_I_VERSION(inode))
-		inode_inc_iversion(inode);
-}
-
 static ssize_t btrfs_file_write_iter(struct kiocb *iocb,
 				    struct iov_iter *from)
 {
@@ -1972,14 +2033,9 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb,
 	struct inode *inode = file_inode(file);
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct btrfs_root *root = BTRFS_I(inode)->root;
-	u64 start_pos;
-	u64 end_pos;
 	ssize_t num_written = 0;
 	const bool sync = iocb->ki_flags & IOCB_DSYNC;
 	ssize_t err;
-	loff_t pos;
-	size_t count;
-	loff_t oldsize;
 
 	/*
 	 * If BTRFS flips readonly due to some impossible error
@@ -2001,65 +2057,12 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb,
 		inode_lock(inode);
 	}
 
-	err = generic_write_checks(iocb, from);
+	err = btrfs_write_check(iocb, from);
 	if (err <= 0) {
 		inode_unlock(inode);
 		return err;
 	}
 
-	pos = iocb->ki_pos;
-	count = iov_iter_count(from);
-	if (iocb->ki_flags & IOCB_NOWAIT) {
-		size_t nocow_bytes = count;
-
-		/*
-		 * We will allocate space in case nodatacow is not set,
-		 * so bail
-		 */
-		if (check_nocow_nolock(BTRFS_I(inode), pos, &nocow_bytes)
-		    <= 0) {
-			inode_unlock(inode);
-			return -EAGAIN;
-		}
-		/*
-		 * There are holes in the range or parts of the range that must
-		 * be COWed (shared extents, RO block groups, etc), so just bail
-		 * out.
-		 */
-		if (nocow_bytes < count) {
-			inode_unlock(inode);
-			return -EAGAIN;
-		}
-	}
-
-	current->backing_dev_info = inode_to_bdi(inode);
-	err = file_remove_privs(file);
-	if (err) {
-		inode_unlock(inode);
-		goto out;
-	}
-
-	/*
-	 * We reserve space for updating the inode when we reserve space for the
-	 * extent we are going to write, so we will enospc out there.  We don't
-	 * need to start yet another transaction to update the inode as we will
-	 * update the inode when we finish writing whatever data we write.
-	 */
-	update_time_for_write(inode);
-
-	start_pos = round_down(pos, fs_info->sectorsize);
-	oldsize = i_size_read(inode);
-	if (start_pos > oldsize) {
-		/* Expand hole size to cover write data, preventing empty gap */
-		end_pos = round_up(pos + count,
-				   fs_info->sectorsize);
-		err = btrfs_cont_expand(inode, oldsize, end_pos);
-		if (err) {
-			inode_unlock(inode);
-			goto out;
-		}
-	}
-
 	if (sync)
 		atomic_inc(&BTRFS_I(inode)->sync_writers);
 
@@ -2117,7 +2120,7 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb,
 
 	if (sync)
 		atomic_dec(&BTRFS_I(inode)->sync_writers);
-out:
+
 	current->backing_dev_info = NULL;
 	return num_written ? num_written : err;
 }
-- 
2.26.2

[PATCH 09/15] btrfs: Introduce btrfs_inode_lock()/unlock()

[Goldwyn Rodrigues]

From: Goldwyn Rodrigues <rgoldwyn@suse.com>

Helper functions for locking/unlocking i_rwsem.
btrfs_inode_lock/unlock() acquires the inode->i_rwsem depending on the
flags passed. ilock_flags determines the type of lock to be taken:

BTRFS_ILOCK_SHARED - for shared locks, for possible parallel DIO
BTRFS_ILOCK_TRY - for the RWF_NOWAIT sequence

Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
---
 fs/btrfs/ctree.h |  7 +++++++
 fs/btrfs/file.c  | 31 ++++++++++++++++---------------
 fs/btrfs/inode.c | 31 +++++++++++++++++++++++++++++++
 3 files changed, 54 insertions(+), 15 deletions(-)

diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index b47a8dcff028..ea15771bf3da 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -3053,6 +3053,13 @@ extern const struct iomap_ops btrfs_dio_iomap_ops;
 extern const struct iomap_dio_ops btrfs_dio_ops;
 extern const struct iomap_dio_ops btrfs_sync_dops;
 
+/* ilock flags definition */
+#define BTRFS_ILOCK_SHARED	(1 << 0)
+#define BTRFS_ILOCK_TRY 	(1 << 1)
+
+int btrfs_inode_lock(struct inode *inode, int ilock_flags);
+void btrfs_inode_unlock(struct inode *inode, int ilock_flags);
+
 /* ioctl.c */
 long btrfs_ioctl(struct file *file, unsigned int cmd, unsigned long arg);
 long btrfs_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg);
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index 0f961ce1fa98..7e18334e8121 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -1974,7 +1974,7 @@ static ssize_t btrfs_direct_write(struct kiocb *iocb, struct iov_iter *from)
 	 * not unlock the i_mutex at this case.
 	 */
 	if (pos + iov_iter_count(from) <= inode->i_size) {
-		inode_unlock(inode);
+		btrfs_inode_unlock(inode, 0);
 		relock = true;
 	}
 	down_read(&BTRFS_I(inode)->dio_sem);
@@ -1995,7 +1995,7 @@ static ssize_t btrfs_direct_write(struct kiocb *iocb, struct iov_iter *from)
 
 	up_read(&BTRFS_I(inode)->dio_sem);
 	if (relock)
-		inode_lock(inode);
+		btrfs_inode_lock(inode, 0);
 
 	if (written < 0 || !iov_iter_count(from))
 		return written;
@@ -2036,6 +2036,7 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb,
 	ssize_t num_written = 0;
 	const bool sync = iocb->ki_flags & IOCB_DSYNC;
 	ssize_t err;
+	int ilock_flags = 0;
 
 	/*
 	 * If BTRFS flips readonly due to some impossible error
@@ -2050,16 +2051,16 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb,
 	    (iocb->ki_flags & IOCB_NOWAIT))
 		return -EOPNOTSUPP;
 
-	if (iocb->ki_flags & IOCB_NOWAIT) {
-		if (!inode_trylock(inode))
-			return -EAGAIN;
-	} else {
-		inode_lock(inode);
-	}
+	if (iocb->ki_flags & IOCB_NOWAIT)
+		ilock_flags |= BTRFS_ILOCK_TRY;
+
+	err = btrfs_inode_lock(inode, ilock_flags);
+	if (err < 0)
+		return err;
 
 	err = btrfs_write_check(iocb, from);
 	if (err <= 0) {
-		inode_unlock(inode);
+		btrfs_inode_unlock(inode, ilock_flags);
 		return err;
 	}
 
@@ -2105,7 +2106,7 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb,
 		num_written = btrfs_buffered_write(iocb, from);
 	}
 
-	inode_unlock(inode);
+	btrfs_inode_unlock(inode, ilock_flags);
 
 	/*
 	 * We also have to set last_sub_trans to the current log transid,
@@ -3405,7 +3406,7 @@ static long btrfs_fallocate(struct file *file, int mode,
 			return ret;
 	}
 
-	inode_lock(inode);
+	btrfs_inode_lock(inode, 0);
 
 	if (!(mode & FALLOC_FL_KEEP_SIZE) && offset + len > inode->i_size) {
 		ret = inode_newsize_ok(inode, offset + len);
@@ -3644,9 +3645,9 @@ static loff_t btrfs_file_llseek(struct file *file, loff_t offset, int whence)
 		return generic_file_llseek(file, offset, whence);
 	case SEEK_DATA:
 	case SEEK_HOLE:
-		inode_lock_shared(inode);
+		btrfs_inode_lock(inode, BTRFS_ILOCK_SHARED);
 		offset = find_desired_extent(inode, offset, whence);
-		inode_unlock_shared(inode);
+		btrfs_inode_unlock(inode, BTRFS_ILOCK_SHARED);
 		break;
 	}
 
@@ -3690,10 +3691,10 @@ static ssize_t btrfs_direct_read(struct kiocb *iocb, struct iov_iter *to)
 	if (check_direct_read(btrfs_sb(inode->i_sb), to, iocb->ki_pos))
 		return 0;
 
-	inode_lock_shared(inode);
+	btrfs_inode_lock(inode, BTRFS_ILOCK_SHARED);
 	ret = iomap_dio_rw(iocb, to, &btrfs_dio_iomap_ops, &btrfs_dio_ops,
 			is_sync_kiocb(iocb));
-	inode_unlock_shared(inode);
+	btrfs_inode_unlock(inode, BTRFS_ILOCK_SHARED);
 	return ret;
 }
 
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 0730131b6590..f305efac75ae 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -96,6 +96,37 @@ static void __endio_write_update_ordered(struct btrfs_inode *inode,
 					 const u64 offset, const u64 bytes,
 					 const bool uptodate);
 
+int btrfs_inode_lock(struct inode *inode, int ilock_flags)
+{
+	if (ilock_flags & BTRFS_ILOCK_SHARED) {
+		if (ilock_flags & BTRFS_ILOCK_TRY) {
+			if (!inode_trylock_shared(inode))
+				return -EAGAIN;
+			else
+				return 0;
+		}
+		inode_lock_shared(inode);
+	} else {
+		if (ilock_flags & BTRFS_ILOCK_TRY) {
+			if (!inode_trylock(inode))
+				return -EAGAIN;
+			else
+				return 0;
+		}
+		inode_lock(inode);
+	}
+	return 0;
+}
+
+void btrfs_inode_unlock(struct inode *inode, int ilock_flags)
+{
+	if (ilock_flags & BTRFS_ILOCK_SHARED)
+		inode_unlock_shared(inode);
+	else
+		inode_unlock(inode);
+
+}
+
 /*
  * Cleanup all submitted ordered extents in specified range to handle errors
  * from the btrfs_run_delalloc_range() callback.
-- 
2.26.2

[PATCH 10/15] btrfs: Push inode locking and unlocking into buffered/direct write

[Goldwyn Rodrigues]

From: Goldwyn Rodrigues <rgoldwyn@suse.com>

Push inode locking and unlocking closer to where we perform the I/O. For
this we need to move the write checks inside the respective functions as
well.

pos is assigned after write checks because generic_write_check can
modify iocb->ki_pos.

Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
---
 fs/btrfs/file.c | 67 ++++++++++++++++++++++++++++++++-----------------
 1 file changed, 44 insertions(+), 23 deletions(-)

diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index 7e18334e8121..d9c3be19d7b3 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -1698,7 +1698,7 @@ static noinline ssize_t btrfs_buffered_write(struct kiocb *iocb,
 					       struct iov_iter *i)
 {
 	struct file *file = iocb->ki_filp;
-	loff_t pos = iocb->ki_pos;
+	loff_t pos;
 	struct inode *inode = file_inode(file);
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct page **pages = NULL;
@@ -1712,14 +1712,29 @@ static noinline ssize_t btrfs_buffered_write(struct kiocb *iocb,
 	bool only_release_metadata = false;
 	bool force_page_uptodate = false;
 	loff_t old_isize = i_size_read(inode);
+	int ilock_flags = 0;
+
+	if (iocb->ki_flags & IOCB_NOWAIT)
+		ilock_flags |= BTRFS_ILOCK_TRY;
+
+	ret = btrfs_inode_lock(inode, ilock_flags);
+	if (ret < 0)
+		return ret;
+
+	ret = btrfs_write_check(iocb, i);
+	if (ret <= 0)
+		goto out;
 
+	pos = iocb->ki_pos;
 	nrptrs = min(DIV_ROUND_UP(iov_iter_count(i), PAGE_SIZE),
 			PAGE_SIZE / (sizeof(struct page *)));
 	nrptrs = min(nrptrs, current->nr_dirtied_pause - current->nr_dirtied);
 	nrptrs = max(nrptrs, 8);
 	pages = kmalloc_array(nrptrs, sizeof(struct page *), GFP_KERNEL);
-	if (!pages)
-		return -ENOMEM;
+	if (!pages) {
+		ret = -ENOMEM;
+		goto out;
+	}
 
 	while (iov_iter_count(i) > 0) {
 		struct extent_state *cached_state = NULL;
@@ -1936,6 +1951,8 @@ static noinline ssize_t btrfs_buffered_write(struct kiocb *iocb,
 		pagecache_isize_extended(inode, old_isize, iocb->ki_pos);
 		iocb->ki_pos += num_written;
 	}
+out:
+	btrfs_inode_unlock(inode, ilock_flags);
 	return num_written ? num_written : ret;
 }
 
@@ -1958,15 +1975,33 @@ static ssize_t btrfs_direct_write(struct kiocb *iocb, struct iov_iter *from)
 	struct file *file = iocb->ki_filp;
 	struct inode *inode = file_inode(file);
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
-	loff_t pos = iocb->ki_pos;
+	loff_t pos;
 	ssize_t written = 0;
 	bool relock = false;
 	ssize_t written_buffered;
 	loff_t endbyte;
 	int err;
+	int ilock_flags = 0;
+
+	if (iocb->ki_flags & IOCB_NOWAIT)
+		ilock_flags |= BTRFS_ILOCK_TRY;
+
+	err = btrfs_inode_lock(inode, ilock_flags);
+	if (err < 0)
+		return err;
+
+	err = btrfs_write_check(iocb, from);
+	if (err <= 0) {
+		btrfs_inode_unlock(inode, ilock_flags);
+		goto out;
+	}
+
+	pos = iocb->ki_pos;
 
-	if (check_direct_IO(fs_info, from, pos))
+	if (check_direct_IO(fs_info, from, pos)) {
+		btrfs_inode_unlock(inode, ilock_flags);
 		goto buffered;
+	}
 
 	/*
 	 * If the write DIO is beyond the EOF, we need update
@@ -1997,8 +2032,10 @@ static ssize_t btrfs_direct_write(struct kiocb *iocb, struct iov_iter *from)
 	if (relock)
 		btrfs_inode_lock(inode, 0);
 
-	if (written < 0 || !iov_iter_count(from))
-		return written;
+	if (written < 0 || !iov_iter_count(from)) {
+		err = written;
+		goto out;
+	}
 
 buffered:
 	pos = iocb->ki_pos;
@@ -2036,7 +2073,6 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb,
 	ssize_t num_written = 0;
 	const bool sync = iocb->ki_flags & IOCB_DSYNC;
 	ssize_t err;
-	int ilock_flags = 0;
 
 	/*
 	 * If BTRFS flips readonly due to some impossible error
@@ -2051,19 +2087,6 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb,
 	    (iocb->ki_flags & IOCB_NOWAIT))
 		return -EOPNOTSUPP;
 
-	if (iocb->ki_flags & IOCB_NOWAIT)
-		ilock_flags |= BTRFS_ILOCK_TRY;
-
-	err = btrfs_inode_lock(inode, ilock_flags);
-	if (err < 0)
-		return err;
-
-	err = btrfs_write_check(iocb, from);
-	if (err <= 0) {
-		btrfs_inode_unlock(inode, ilock_flags);
-		return err;
-	}
-
 	if (sync)
 		atomic_inc(&BTRFS_I(inode)->sync_writers);
 
@@ -2106,8 +2129,6 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb,
 		num_written = btrfs_buffered_write(iocb, from);
 	}
 
-	btrfs_inode_unlock(inode, ilock_flags);
-
 	/*
 	 * We also have to set last_sub_trans to the current log transid,
 	 * otherwise subsequent syncs to a file that's been synced in this
-- 
2.26.2

[PATCH 11/15] btrfs: Use inode_lock_shared() for direct writes within EOF

[Goldwyn Rodrigues]

From: Goldwyn Rodrigues <rgoldwyn@suse.com>

Direct writes within EOF are safe to be performed with inode shared lock
to improve parallelization with other direct writes or reads because EOF
is not changed and there is no race with truncate().

Direct reads are already performed under shared inode lock.

This patch is precursor to removing btrfs_inode->dio_sem.

Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
---
 fs/btrfs/file.c | 33 +++++++++++++++++++++------------
 1 file changed, 21 insertions(+), 12 deletions(-)

diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index d9c3be19d7b3..50092d24eee2 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -1977,7 +1977,6 @@ static ssize_t btrfs_direct_write(struct kiocb *iocb, struct iov_iter *from)
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	loff_t pos;
 	ssize_t written = 0;
-	bool relock = false;
 	ssize_t written_buffered;
 	loff_t endbyte;
 	int err;
@@ -1986,6 +1985,15 @@ static ssize_t btrfs_direct_write(struct kiocb *iocb, struct iov_iter *from)
 	if (iocb->ki_flags & IOCB_NOWAIT)
 		ilock_flags |= BTRFS_ILOCK_TRY;
 
+	/*
+	 * If the write DIO within EOF,  use a shared lock
+	 */
+	if (iocb->ki_pos + iov_iter_count(from) <= i_size_read(inode))
+		ilock_flags |= BTRFS_ILOCK_SHARED;
+	else if (iocb->ki_flags & IOCB_NOWAIT)
+		return -EAGAIN;
+
+relock:
 	err = btrfs_inode_lock(inode, ilock_flags);
 	if (err < 0)
 		return err;
@@ -1997,21 +2005,23 @@ static ssize_t btrfs_direct_write(struct kiocb *iocb, struct iov_iter *from)
 	}
 
 	pos = iocb->ki_pos;
+	/*
+	 * Re-check since file size may have changed
+	 * just before taking the lock or pos may have changed
+	 * because of O_APPEND in generic_write_check()
+	 */
+	if ((ilock_flags & BTRFS_ILOCK_SHARED) &&
+	    pos + iov_iter_count(from) > i_size_read(inode)) {
+		btrfs_inode_unlock(inode, ilock_flags);
+		ilock_flags &= ~BTRFS_ILOCK_SHARED;
+		goto relock;
+	}
 
 	if (check_direct_IO(fs_info, from, pos)) {
 		btrfs_inode_unlock(inode, ilock_flags);
 		goto buffered;
 	}
 
-	/*
-	 * If the write DIO is beyond the EOF, we need update
-	 * the isize, but it is protected by i_mutex. So we can
-	 * not unlock the i_mutex at this case.
-	 */
-	if (pos + iov_iter_count(from) <= inode->i_size) {
-		btrfs_inode_unlock(inode, 0);
-		relock = true;
-	}
 	down_read(&BTRFS_I(inode)->dio_sem);
 
 	/*
@@ -2029,8 +2039,7 @@ static ssize_t btrfs_direct_write(struct kiocb *iocb, struct iov_iter *from)
 		written = 0;
 
 	up_read(&BTRFS_I(inode)->dio_sem);
-	if (relock)
-		btrfs_inode_lock(inode, 0);
+	btrfs_inode_unlock(inode, ilock_flags);
 
 	if (written < 0 || !iov_iter_count(from)) {
 		err = written;
-- 
2.26.2

[PATCH 12/15] btrfs: Remove dio_sem

[Goldwyn Rodrigues]

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>
---
 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 50092d24eee2..193af84f5405 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -2022,8 +2022,6 @@ 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.
@@ -2038,7 +2036,6 @@ 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)) {
@@ -2248,13 +2245,6 @@ 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);
 
 	/*
@@ -2285,7 +2275,6 @@ 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;
 	}
@@ -2382,7 +2371,6 @@ 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) {
@@ -2413,7 +2401,6 @@ 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 f305efac75ae..96ff8e4a3b45 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -8580,7 +8580,6 @@ 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

[PATCH 13/15] btrfs: Call iomap_dio_complete() without inode_lock

[Goldwyn Rodrigues]

From: Goldwyn Rodrigues <rgoldwyn@suse.com>

If direct writes are called with O_DIRECT | O_DSYNC, it will result in a
deadlock because iomap_dio_rw() is called under i_rwsem which calls
iomap_dio_complete()
  generic_write_sync()
    btrfs_sync_file().

btrfs_sync_file() requires i_rwsem, so call __iomap_dio_rw() with the
i_rwsem locked, and call iomap_dio_complete() after unlocking i_rwsem.

Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
---
 fs/btrfs/file.c | 24 +++++++++++-------------
 1 file changed, 11 insertions(+), 13 deletions(-)

diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index 193af84f5405..9c7a2d4b4148 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -1981,6 +1981,7 @@ static ssize_t btrfs_direct_write(struct kiocb *iocb, struct iov_iter *from)
 	loff_t endbyte;
 	int err;
 	int ilock_flags = 0;
+	struct iomap_dio *dio = NULL;
 
 	if (iocb->ki_flags & IOCB_NOWAIT)
 		ilock_flags |= BTRFS_ILOCK_TRY;
@@ -2022,22 +2023,19 @@ static ssize_t btrfs_direct_write(struct kiocb *iocb, struct iov_iter *from)
 		goto buffered;
 	}
 
-	/*
-	 * We have are actually a sync iocb, so we need our fancy endio to know
-	 * if we need to sync.
-	 */
-	if (current->journal_info)
-		written = iomap_dio_rw(iocb, from, &btrfs_dio_iomap_ops,
-				&btrfs_sync_dops, is_sync_kiocb(iocb));
-	else
-		written = iomap_dio_rw(iocb, from, &btrfs_dio_iomap_ops,
-				&btrfs_dio_ops, is_sync_kiocb(iocb));
-
-	if (written == -ENOTBLK)
-		written = 0;
+	dio = __iomap_dio_rw(iocb, from, &btrfs_dio_iomap_ops,
+			&btrfs_dio_ops, is_sync_kiocb(iocb));
 
 	btrfs_inode_unlock(inode, ilock_flags);
 
+	if (IS_ERR_OR_NULL(dio)) {
+		err = PTR_ERR_OR_ZERO(dio);
+		if (err < 0 && err != -ENOTBLK)
+			goto out;
+	} else {
+		written = iomap_dio_complete(dio);
+	}
+
 	if (written < 0 || !iov_iter_count(from)) {
 		err = written;
 		goto out;
-- 
2.26.2

[PATCH 14/15] btrfs: Revert 09745ff88d93 ("btrfs: dio iomap DSYNC workaround")

[Goldwyn Rodrigues]

From: Goldwyn Rodrigues <rgoldwyn@suse.com>

iomap_dio_complete()->generic_write_sync() is not called under i_rwsem
anymore, revert the workaround.

Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
---
 fs/btrfs/ctree.h       |  1 -
 fs/btrfs/file.c        | 38 ++------------------------------
 fs/btrfs/inode.c       | 50 ------------------------------------------
 fs/btrfs/transaction.h |  1 -
 4 files changed, 2 insertions(+), 88 deletions(-)

diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index ea15771bf3da..bc96c52021b2 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -3051,7 +3051,6 @@ void btrfs_writepage_endio_finish_ordered(struct page *page, u64 start,
 extern const struct dentry_operations btrfs_dentry_operations;
 extern const struct iomap_ops btrfs_dio_iomap_ops;
 extern const struct iomap_dio_ops btrfs_dio_ops;
-extern const struct iomap_dio_ops btrfs_sync_dops;
 
 /* ilock flags definition */
 #define BTRFS_ILOCK_SHARED	(1 << 0)
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index 9c7a2d4b4148..1d3363e7b09d 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -2094,44 +2094,10 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb,
 	if (sync)
 		atomic_inc(&BTRFS_I(inode)->sync_writers);
 
-	if (iocb->ki_flags & IOCB_DIRECT) {
-		/*
-		 * 1. We must always clear IOCB_DSYNC in order to not deadlock
-		 *    in iomap, as it calls generic_write_sync() in this case.
-		 * 2. If we are async, we can call iomap_dio_complete() either
-		 *    in
-		 *
-		 *    2.1. A worker thread from the last bio completed.  In this
-		 *	   case we need to mark the btrfs_dio_data that it is
-		 *	   async in order to call generic_write_sync() properly.
-		 *	   This is handled by setting BTRFS_DIO_SYNC_STUB in the
-		 *	   current->journal_info.
-		 *    2.2  The submitter context, because all IO completed
-		 *         before we exited iomap_dio_rw().  In this case we can
-		 *         just re-set the IOCB_DSYNC on the iocb and we'll do
-		 *         the sync below.  If our ->end_io() gets called and
-		 *         current->journal_info is set, then we know we're in
-		 *         our current context and we will clear
-		 *         current->journal_info to indicate that we need to
-		 *         sync below.
-		 */
-		if (sync) {
-			ASSERT(current->journal_info == NULL);
-			iocb->ki_flags &= ~IOCB_DSYNC;
-			current->journal_info = BTRFS_DIO_SYNC_STUB;
-		}
+	if (iocb->ki_flags & IOCB_DIRECT)
 		num_written = btrfs_direct_write(iocb, from);
-
-		/*
-		 * As stated above, we cleared journal_info, so we need to do
-		 * the sync ourselves.
-		 */
-		if (sync && current->journal_info == NULL)
-			iocb->ki_flags |= IOCB_DSYNC;
-		current->journal_info = NULL;
-	} else {
+	else
 		num_written = btrfs_buffered_write(iocb, from);
-	}
 
 	/*
 	 * We also have to set last_sub_trans to the current log transid,
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 96ff8e4a3b45..41e32ef6b6ef 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -62,7 +62,6 @@ struct btrfs_dio_data {
 	loff_t length;
 	ssize_t submitted;
 	struct extent_changeset *data_reserved;
-	bool sync;
 };
 
 static const struct inode_operations btrfs_dir_inode_operations;
@@ -7362,17 +7361,6 @@ static int btrfs_dio_iomap_begin(struct inode *inode, loff_t start,
 	int ret = 0;
 	u64 len = length;
 	bool unlock_extents = false;
-	bool sync = (current->journal_info == BTRFS_DIO_SYNC_STUB);
-
-	/*
-	 * We used current->journal_info here to see if we were sync, but
-	 * there's a lot of tests in the enospc machinery to not do flushing if
-	 * we have a journal_info set, so we need to clear this out and re-set
-	 * it in iomap_end.
-	 */
-	ASSERT(current->journal_info == NULL ||
-	       current->journal_info == BTRFS_DIO_SYNC_STUB);
-	current->journal_info = NULL;
 
 	if (!write)
 		len = min_t(u64, len, fs_info->sectorsize);
@@ -7398,7 +7386,6 @@ static int btrfs_dio_iomap_begin(struct inode *inode, loff_t start,
 	if (!dio_data)
 		return -ENOMEM;
 
-	dio_data->sync = sync;
 	dio_data->length = length;
 	if (write) {
 		dio_data->reserve = round_up(length, fs_info->sectorsize);
@@ -7546,14 +7533,6 @@ static int btrfs_dio_iomap_end(struct inode *inode, loff_t pos, loff_t length,
 		extent_changeset_free(dio_data->data_reserved);
 	}
 out:
-	/*
-	 * We're all done, we can re-set the current->journal_info now safely
-	 * for our endio.
-	 */
-	if (dio_data->sync) {
-		ASSERT(current->journal_info == NULL);
-		current->journal_info = BTRFS_DIO_SYNC_STUB;
-	}
 	kfree(dio_data);
 	iomap->private = NULL;
 
@@ -7929,30 +7908,6 @@ static blk_qc_t btrfs_submit_direct(struct inode *inode, struct iomap *iomap,
 	return BLK_QC_T_NONE;
 }
 
-static inline int btrfs_maybe_fsync_end_io(struct kiocb *iocb, ssize_t size,
-					   int error, unsigned flags)
-{
-	/*
-	 * Now if we're still in the context of our submitter we know we can't
-	 * safely run generic_write_sync(), so clear our flag here so that the
-	 * caller knows to follow up with a sync.
-	 */
-	if (current->journal_info == BTRFS_DIO_SYNC_STUB) {
-		current->journal_info = NULL;
-		return error;
-	}
-
-	if (error)
-		return error;
-
-	if (size) {
-		iocb->ki_flags |= IOCB_DSYNC;
-		return generic_write_sync(iocb, size);
-	}
-
-	return 0;
-}
-
 const struct iomap_ops btrfs_dio_iomap_ops = {
 	.iomap_begin            = btrfs_dio_iomap_begin,
 	.iomap_end              = btrfs_dio_iomap_end,
@@ -7962,11 +7917,6 @@ const struct iomap_dio_ops btrfs_dio_ops = {
 	.submit_io		= btrfs_submit_direct,
 };
 
-const struct iomap_dio_ops btrfs_sync_dops = {
-	.submit_io		= btrfs_submit_direct,
-	.end_io			= btrfs_maybe_fsync_end_io,
-};
-
 static int btrfs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 			u64 start, u64 len)
 {
diff --git a/fs/btrfs/transaction.h b/fs/btrfs/transaction.h
index 858d9153a1cd..8241c050ba71 100644
--- a/fs/btrfs/transaction.h
+++ b/fs/btrfs/transaction.h
@@ -112,7 +112,6 @@ struct btrfs_transaction {
 #define TRANS_EXTWRITERS	(__TRANS_START | __TRANS_ATTACH)
 
 #define BTRFS_SEND_TRANS_STUB	((void *)1)
-#define BTRFS_DIO_SYNC_STUB	((void *)2)
 
 struct btrfs_trans_handle {
 	u64 transid;
-- 
2.26.2

[PATCH 15/15] iomap: Reinstate lockdep_assert_held in iomap_dio_rw()

[Goldwyn Rodrigues]

From: Goldwyn Rodrigues <rgoldwyn@suse.com>

btrfs holds inode_lock_shared() while performing DIO within EOF, so
lockdep_assert_held() check can be re-instated.

Revert 3ad99bec6e82 ("iomap: remove lockdep_assert_held()")

Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
---
 fs/iomap/direct-io.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/fs/iomap/direct-io.c b/fs/iomap/direct-io.c
index e01f81e7b76f..b5e030971001 100644
--- a/fs/iomap/direct-io.c
+++ b/fs/iomap/direct-io.c
@@ -421,6 +421,8 @@ __iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
 	struct blk_plug plug;
 	struct iomap_dio *dio;
 
+	lockdep_assert_held(&inode->i_rwsem);
+
 	if (!count)
 		return NULL;
 
-- 
2.26.2

Re: [PATCH 03/15] iomap: Allow filesystem to call iomap_dio_complete without i_rwsem

[Johannes Thumshirn]

On 21/09/2020 16:44, Goldwyn Rodrigues wrote:
> From: Christoph Hellwig <hch@lst.de>
> 
> This is to avoid the deadlock caused in btrfs because of O_DIRECT |
> O_DSYNC.
> 
> Filesystems such as btrfs require i_rwsem while performing sync on a
> file. iomap_dio_rw() is called under i_rw_sem. This leads to a
> deadlock because of:
> 
> iomap_dio_complete()
>   generic_write_sync()
>     btrfs_sync_file()
> 
> Separate out iomap_dio_complete() from iomap_dio_rw(), so filesystems
> can call iomap_dio_complete() after unlocking i_rwsem.
> 
> Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
>

This is missing Christoph's S-o-b

Re: [PATCH 04/15] iomap: Call inode_dio_end() before generic_write_sync()

[Johannes Thumshirn]

On 21/09/2020 16:44, Goldwyn Rodrigues wrote:
> From: Goldwyn Rodrigues <rgoldwyn@suse.com>
> 
> iomap complete routine can deadlock with btrfs_fallocate because of the
> call to generic_write_sync().
> 
> P0                      P1
> inode_lock()            fallocate(FALLOC_FL_ZERO_RANGE)
> __iomap_dio_rw()        inode_lock()
>                         <block>
> <submits IO>
> <completes IO>
> inode_unlock()
>                         <gets inode_lock()>
>                         inode_dio_wait()
> iomap_dio_complete()
>   generic_write_sync()
>     btrfs_file_fsync()
>       inode_lock()
>       <deadlock>
> 
> inode_dio_end() is used to notify the end of DIO data in order
> to synchronize with truncate. Call inode_dio_end() before calling
> generic_write_sync(), so filesystems can lock i_rwsem during a sync.
> 
> ---
>  fs/iomap/direct-io.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)


Missing S-o-b as well.

Re: [PATCH 02/15] btrfs: remove BTRFS_INODE_READDIO_NEED_LOCK

[Christoph Hellwig]

On Mon, Sep 21, 2020 at 09:43:40AM -0500, Goldwyn Rodrigues wrote:
> From: Goldwyn Rodrigues <rgoldwyn@suse.com>
> 
> Since we now perform direct reads using i_rwsem, we can remove this
> inode flag used to co-ordinate unlocked reads.
> 
> The truncate call takes i_rwsem. This means it is correctly synchronized
> with concurrent direct reads.
> 
> Reviewed-by: Nikolay Borisov <nborisov@suse.com>
> Reviewed-by: Johannes Thumshirn <jth@kernel.org>
> Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>

Looks good,

Reviewed-by: Christoph Hellwig <hch@lst.de>

Re: [PATCH 03/15] iomap: Allow filesystem to call iomap_dio_complete without i_rwsem

[hch]

On Mon, Sep 21, 2020 at 03:09:29PM +0000, Johannes Thumshirn wrote:
> On 21/09/2020 16:44, Goldwyn Rodrigues wrote:
> > From: Christoph Hellwig <hch@lst.de>
> > 
> > This is to avoid the deadlock caused in btrfs because of O_DIRECT |
> > O_DSYNC.
> > 
> > Filesystems such as btrfs require i_rwsem while performing sync on a
> > file. iomap_dio_rw() is called under i_rw_sem. This leads to a
> > deadlock because of:
> > 
> > iomap_dio_complete()
> >   generic_write_sync()
> >     btrfs_sync_file()
> > 
> > Separate out iomap_dio_complete() from iomap_dio_rw(), so filesystems
> > can call iomap_dio_complete() after unlocking i_rwsem.
> > 
> > Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
> >
> 
> This is missing Christoph's S-o-b

Goldwyn picked this up for a rfc.  But this looks like my patch with
a sane commit log, so:

Signed-off-by: Christoph Hellwig <hch@lst.de>

Re: [PATCH 04/15] iomap: Call inode_dio_end() before generic_write_sync()

[Christoph Hellwig]

On Mon, Sep 21, 2020 at 09:43:42AM -0500, Goldwyn Rodrigues wrote:
> From: Goldwyn Rodrigues <rgoldwyn@suse.com>
> 
> iomap complete routine can deadlock with btrfs_fallocate because of the
> call to generic_write_sync().
> 
> P0                      P1
> inode_lock()            fallocate(FALLOC_FL_ZERO_RANGE)
> __iomap_dio_rw()        inode_lock()
>                         <block>
> <submits IO>
> <completes IO>
> inode_unlock()
>                         <gets inode_lock()>
>                         inode_dio_wait()
> iomap_dio_complete()
>   generic_write_sync()
>     btrfs_file_fsync()
>       inode_lock()
>       <deadlock>
> 
> inode_dio_end() is used to notify the end of DIO data in order
> to synchronize with truncate. Call inode_dio_end() before calling
> generic_write_sync(), so filesystems can lock i_rwsem during a sync.

Looks good:

Reviewed-by: Christoph Hellwig <hch@lst.de>

You might want to mention in the commit log that this matches the
sequence in the old fs/direct-io.c implementation.

Re: [PATCH 05/15] btrfs: split btrfs_direct_IO to read and write

[Christoph Hellwig]

On Mon, Sep 21, 2020 at 09:43:43AM -0500, Goldwyn Rodrigues wrote:
> From: Goldwyn Rodrigues <rgoldwyn@suse.com>
> 
> The read and write DIO don't have anything in common except for the
> call to iomap_dio_rw. Extract the write call into a new function to get
> rid of conditional statements for direct write.
> 
> Originally proposed by Christoph Hellwig <hch@lst.de>
> 
> Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>

Looks good,

Reviewed-by: Christoph Hellwig <hch@lst.de>

Re: [PATCH 06/15] btrfs: Move pos increment and pagecache extension to btrfs_buffered_write()

[Christoph Hellwig]

On Mon, Sep 21, 2020 at 09:43:44AM -0500, Goldwyn Rodrigues wrote:
> From: Goldwyn Rodrigues <rgoldwyn@suse.com>
> 
> While we do this, correct the call to pagecache_isize_extended():
>  - pagecache_isisze_extended needs to be called to the starting of the
>    write as opposed to i_size
>  - We don't need to check range before the call, this is done in the
>    function
> 
> Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>

Looks good,

Reviewed-by: Christoph Hellwig <hch@lst.de>

Re: [PATCH 08/15] btrfs: Introduce btrfs_write_check()

[Christoph Hellwig]

> +	if (iocb->ki_flags & IOCB_NOWAIT) {
> +		size_t nocow_bytes = count;
> +
> +		/*
> +		 * We will allocate space in case nodatacow is not set,
> +		 * so bail
> +		 */
> +		if (check_nocow_nolock(BTRFS_I(inode), pos, &nocow_bytes)
> +		    <= 0)

This could easily be:

		if (check_nocow_nolock(BTRFS_I(inode), pos, &nocow_bytes) <= 0)

as it perfectly fits withing 80 characters.

Re: [PATCH 15/15] iomap: Reinstate lockdep_assert_held in iomap_dio_rw()

[Christoph Hellwig]

On Mon, Sep 21, 2020 at 09:43:53AM -0500, Goldwyn Rodrigues wrote:
> From: Goldwyn Rodrigues <rgoldwyn@suse.com>
> 
> btrfs holds inode_lock_shared() while performing DIO within EOF, so
> lockdep_assert_held() check can be re-instated.
> 
> Revert 3ad99bec6e82 ("iomap: remove lockdep_assert_held()")

It turns out gfs2 also calls without the lock, so while I'd love to see
the assert come back we'd regress gfs2.

Re: [PATCH 01/15] fs: remove dio_end_io()

[Josef Bacik]

On 9/21/20 10:43 AM, Goldwyn Rodrigues wrote:
> From: Goldwyn Rodrigues <rgoldwyn@suse.com>
> 
> Since we removed the last user of dio_end_io(), remove the helper
> function dio_end_io().
> 
> Reviewed-by: Nikolay Borisov <nborisov@suse.com>
> Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
> Reviewed-by: Christoph Hellwig <hch@lst.de>
> Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>

Reviewed-by: Josef Bacik <josef@toxicpanda.com>

Thanks,

Josef

Re: [PATCH 02/15] btrfs: remove BTRFS_INODE_READDIO_NEED_LOCK

[Josef Bacik]

On 9/21/20 10:43 AM, Goldwyn Rodrigues wrote:
> From: Goldwyn Rodrigues <rgoldwyn@suse.com>
> 
> Since we now perform direct reads using i_rwsem, we can remove this
> inode flag used to co-ordinate unlocked reads.
> 
> The truncate call takes i_rwsem. This means it is correctly synchronized
> with concurrent direct reads.
> 
> Reviewed-by: Nikolay Borisov <nborisov@suse.com>
> Reviewed-by: Johannes Thumshirn <jth@kernel.org>
> Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>

Reviewed-by: Josef Bacik <josef@toxicpanda.com>

Thanks,

Josef

Re: [PATCH 03/15] iomap: Allow filesystem to call iomap_dio_complete without i_rwsem

[Josef Bacik]

On 9/21/20 10:43 AM, Goldwyn Rodrigues wrote:
> From: Christoph Hellwig <hch@lst.de>
> 
> This is to avoid the deadlock caused in btrfs because of O_DIRECT |
> O_DSYNC.
> 
> Filesystems such as btrfs require i_rwsem while performing sync on a
> file. iomap_dio_rw() is called under i_rw_sem. This leads to a
> deadlock because of:
> 
> iomap_dio_complete()
>    generic_write_sync()
>      btrfs_sync_file()
> 
> Separate out iomap_dio_complete() from iomap_dio_rw(), so filesystems
> can call iomap_dio_complete() after unlocking i_rwsem.
> 
> Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
> ---
>   fs/iomap/direct-io.c  | 34 +++++++++++++++++++++++++---------
>   include/linux/iomap.h |  5 +++++
>   2 files changed, 30 insertions(+), 9 deletions(-)
> 
> diff --git a/fs/iomap/direct-io.c b/fs/iomap/direct-io.c
> index c1aafb2ab990..d970c6bbbe11 100644
> --- a/fs/iomap/direct-io.c
> +++ b/fs/iomap/direct-io.c
> @@ -76,7 +76,7 @@ static void iomap_dio_submit_bio(struct iomap_dio *dio, struct iomap *iomap,
>   		dio->submit.cookie = submit_bio(bio);
>   }
>   
> -static ssize_t iomap_dio_complete(struct iomap_dio *dio)
> +ssize_t iomap_dio_complete(struct iomap_dio *dio)
>   {
>   	const struct iomap_dio_ops *dops = dio->dops;
>   	struct kiocb *iocb = dio->iocb;
> @@ -130,6 +130,7 @@ static ssize_t iomap_dio_complete(struct iomap_dio *dio)
>   
>   	return ret;
>   }
> +EXPORT_SYMBOL_GPL(iomap_dio_complete);
>   
>   static void iomap_dio_complete_work(struct work_struct *work)
>   {
> @@ -406,8 +407,8 @@ iomap_dio_actor(struct inode *inode, loff_t pos, loff_t length,
>    * Returns -ENOTBLK In case of a page invalidation invalidation failure for
>    * writes.  The callers needs to fall back to buffered I/O in this case.
>    */
> -ssize_t
> -iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
> +struct iomap_dio *
> +__iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
>   		const struct iomap_ops *ops, const struct iomap_dio_ops *dops,
>   		bool wait_for_completion)
>   {
> @@ -421,14 +422,14 @@ iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
>   	struct iomap_dio *dio;
>   
>   	if (!count)
> -		return 0;
> +		return NULL;
>   
>   	if (WARN_ON(is_sync_kiocb(iocb) && !wait_for_completion))
> -		return -EIO;
> +		return ERR_PTR(-EIO);
>   
>   	dio = kmalloc(sizeof(*dio), GFP_KERNEL);
>   	if (!dio)
> -		return -ENOMEM;
> +		return ERR_PTR(-ENOMEM);
>   
>   	dio->iocb = iocb;
>   	atomic_set(&dio->ref, 1);
> @@ -558,7 +559,7 @@ iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
>   	dio->wait_for_completion = wait_for_completion;
>   	if (!atomic_dec_and_test(&dio->ref)) {
>   		if (!wait_for_completion)
> -			return -EIOCBQUEUED;
> +			return ERR_PTR(-EIOCBQUEUED);
>   
>   		for (;;) {
>   			set_current_state(TASK_UNINTERRUPTIBLE);
> @@ -574,10 +575,25 @@ iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
>   		__set_current_state(TASK_RUNNING);
>   	}
>   
> -	return iomap_dio_complete(dio);
> +	return dio;
>   
>   out_free_dio:
>   	kfree(dio);
> -	return ret;
> +	return ERR_PTR(ret);
> +}
> +EXPORT_SYMBOL_GPL(__iomap_dio_rw);
> +
> +ssize_t
> +iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
> +		const struct iomap_ops *ops, const struct iomap_dio_ops *dops,
> +		bool wait_for_completion)
> +{
> +	struct iomap_dio *dio;
> +
> +	dio = __iomap_dio_rw(iocb, iter, ops, dops, wait_for_completion);
> +	if (IS_ERR_OR_NULL(dio))
> +		return PTR_ERR_OR_ZERO(dio);
> +	return iomap_dio_complete(dio);
>   }
>   EXPORT_SYMBOL_GPL(iomap_dio_rw);
> +

Got an extra + here for some reason.  Otherwise

Reviewed-by: Josef Bacik <josef@toxicpanda.com>

Thanks,

Josef

Re: [PATCH 04/15] iomap: Call inode_dio_end() before generic_write_sync()

[Josef Bacik]

On 9/21/20 10:43 AM, Goldwyn Rodrigues wrote:
> From: Goldwyn Rodrigues <rgoldwyn@suse.com>
> 
> iomap complete routine can deadlock with btrfs_fallocate because of the
> call to generic_write_sync().
> 
> P0                      P1
> inode_lock()            fallocate(FALLOC_FL_ZERO_RANGE)
> __iomap_dio_rw()        inode_lock()
>                          <block>
> <submits IO>
> <completes IO>
> inode_unlock()
>                          <gets inode_lock()>
>                          inode_dio_wait()
> iomap_dio_complete()
>    generic_write_sync()
>      btrfs_file_fsync()
>        inode_lock()
>        <deadlock>
> 
> inode_dio_end() is used to notify the end of DIO data in order
> to synchronize with truncate. Call inode_dio_end() before calling
> generic_write_sync(), so filesystems can lock i_rwsem during a sync.
> 
> ---
>   fs/iomap/direct-io.c | 2 +-
>   1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/fs/iomap/direct-io.c b/fs/iomap/direct-io.c
> index d970c6bbbe11..e01f81e7b76f 100644
> --- a/fs/iomap/direct-io.c
> +++ b/fs/iomap/direct-io.c
> @@ -118,6 +118,7 @@ ssize_t iomap_dio_complete(struct iomap_dio *dio)
>   			dio_warn_stale_pagecache(iocb->ki_filp);
>   	}
>   
> +	inode_dio_end(file_inode(iocb->ki_filp));
>   	/*
>   	 * If this is a DSYNC write, make sure we push it to stable storage now
>   	 * that we've written data.
> @@ -125,7 +126,6 @@ ssize_t iomap_dio_complete(struct iomap_dio *dio)
>   	if (ret > 0 && (dio->flags & IOMAP_DIO_NEED_SYNC))
>   		ret = generic_write_sync(iocb, ret);
>   
> -	inode_dio_end(file_inode(iocb->ki_filp));
>   	kfree(dio);
>   
>   	return ret;
> 

Did you verify that xfs or ext4 don't rely on the inode_dio_end() happening 
before the generic_write_sync()?  I wouldn't expect that they would, but we've 
already run into problems making those kind of assumptions.  If it's fine you 
can add

Reviewed-by: Josef Bacik <josef@toxicpanda.com>

Thanks,

Josef

Re: [PATCH 05/15] btrfs: split btrfs_direct_IO to read and write

[Josef Bacik]

On 9/21/20 10:43 AM, Goldwyn Rodrigues wrote:
> From: Goldwyn Rodrigues <rgoldwyn@suse.com>
> 
> The read and write DIO don't have anything in common except for the
> call to iomap_dio_rw. Extract the write call into a new function to get
> rid of conditional statements for direct write.
> 
> Originally proposed by Christoph Hellwig <hch@lst.de>
> 
> Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>

Reviewed-by: Josef Bacik <josef@toxicpanda.com>

Thanks,

Josef

Re: [PATCH 06/15] btrfs: Move pos increment and pagecache extension to btrfs_buffered_write()

[Josef Bacik]

On 9/21/20 10:43 AM, Goldwyn Rodrigues wrote:
> From: Goldwyn Rodrigues <rgoldwyn@suse.com>
> 
> While we do this, correct the call to pagecache_isize_extended():
>   - pagecache_isisze_extended needs to be called to the starting of the
>     write as opposed to i_size
>   - We don't need to check range before the call, this is done in the
>     function
> 
> Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>

Reviewed-by: Josef Bacik <josef@toxicpanda.com>

Thanks,

Josef

Re: [PATCH 07/15] btrfs: Move FS error state bit early during write

[Josef Bacik]

On 9/21/20 10:43 AM, Goldwyn Rodrigues wrote:
> From: Goldwyn Rodrigues <rgoldwyn@suse.com>
> 
> fs_info->fs_state is a filesystem bit check as opposed to inode
> and can be performed before we begin with write checks. This eliminates
> inode lock/unlock in case of error bit is set.
> 
> Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>

The check was there because remove_file_privs may have tripped the error 
starting a trans handle.  That being said we could easily have been set before 
now, so this is fine

Reviewed-by: Josef Bacik <josef@toxicpanda.com>

Thanks,

Josef

Re: [PATCH 08/15] btrfs: Introduce btrfs_write_check()

[Josef Bacik]

On 9/21/20 10:43 AM, Goldwyn Rodrigues wrote:
> From: Goldwyn Rodrigues <rgoldwyn@suse.com>
> 
> btrfs_write_check() checks for all parameters in one place before
> beginning a write. This does away with inode_unlock() after every check.
> In the later patches, it will help push inode_lock/unlock() in buffered
> and direct write functions respectively.
> 
> Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
> ---
>   fs/btrfs/file.c | 159 ++++++++++++++++++++++++------------------------
>   1 file changed, 81 insertions(+), 78 deletions(-)
> 
> diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
> index ca374cb5ffc9..0f961ce1fa98 100644
> --- a/fs/btrfs/file.c
> +++ b/fs/btrfs/file.c
> @@ -1615,6 +1615,85 @@ void btrfs_check_nocow_unlock(struct btrfs_inode *inode)
>   	btrfs_drew_write_unlock(&inode->root->snapshot_lock);
>   }
>   
> +static void update_time_for_write(struct inode *inode)
> +{
> +	struct timespec64 now;
> +
> +	if (IS_NOCMTIME(inode))
> +		return;
> +
> +	now = current_time(inode);
> +	if (!timespec64_equal(&inode->i_mtime, &now))
> +		inode->i_mtime = now;
> +
> +	if (!timespec64_equal(&inode->i_ctime, &now))
> +		inode->i_ctime = now;
> +
> +	if (IS_I_VERSION(inode))
> +		inode_inc_iversion(inode);
> +}
> +
> +static size_t btrfs_write_check(struct kiocb *iocb, struct iov_iter *from)
> +{
> +	struct file *file = iocb->ki_filp;
> +	struct inode *inode = file_inode(file);
> +	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
> +	loff_t pos = iocb->ki_pos;
> +	size_t count = iov_iter_count(from);
> +	int err;
> +	loff_t oldsize;
> +	loff_t start_pos;
> +
> +	err = generic_write_checks(iocb, from);
> +	if (err <= 0)
> +		return err;
> +
> +	if (iocb->ki_flags & IOCB_NOWAIT) {
> +		size_t nocow_bytes = count;
> +
> +		/*
> +		 * We will allocate space in case nodatacow is not set,
> +		 * so bail
> +		 */
> +		if (check_nocow_nolock(BTRFS_I(inode), pos, &nocow_bytes)
> +		    <= 0)
> +			return -EAGAIN;
> +		/*
> +		 * There are holes in the range or parts of the range that must
> +		 * be COWed (shared extents, RO block groups, etc), so just bail
> +		 * out.
> +		 */
> +		if (nocow_bytes < count)
> +			return -EAGAIN;
> +	}
> +
> +	current->backing_dev_info = inode_to_bdi(inode);
> +	err = file_remove_privs(file);
> +	if (err)
> +		return err;
> +
> +	/*
> +	 * We reserve space for updating the inode when we reserve space for the
> +	 * extent we are going to write, so we will enospc out there.  We don't
> +	 * need to start yet another transaction to update the inode as we will
> +	 * update the inode when we finish writing whatever data we write.
> +	 */
> +	update_time_for_write(inode);
> +
> +	start_pos = round_down(pos, fs_info->sectorsize);
> +	oldsize = i_size_read(inode);
> +	if (start_pos > oldsize) {
> +		/* Expand hole size to cover write data, preventing empty gap */
> +		loff_t end_pos = round_up(pos + count,
> +					  fs_info->sectorsize);
> +		err = btrfs_cont_expand(inode, oldsize, end_pos);
> +		if (err)
> +			return err;

In this case we're not re-setting current->backing_dev_info to NULL.  Thanks,

Josef

Re: [PATCH 09/15] btrfs: Introduce btrfs_inode_lock()/unlock()

[Josef Bacik]

On 9/21/20 10:43 AM, Goldwyn Rodrigues wrote:
> From: Goldwyn Rodrigues <rgoldwyn@suse.com>
> 
> Helper functions for locking/unlocking i_rwsem.
> btrfs_inode_lock/unlock() acquires the inode->i_rwsem depending on the
> flags passed. ilock_flags determines the type of lock to be taken:
> 
> BTRFS_ILOCK_SHARED - for shared locks, for possible parallel DIO
> BTRFS_ILOCK_TRY - for the RWF_NOWAIT sequence
> 
> Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
> ---
>   fs/btrfs/ctree.h |  7 +++++++
>   fs/btrfs/file.c  | 31 ++++++++++++++++---------------
>   fs/btrfs/inode.c | 31 +++++++++++++++++++++++++++++++
>   3 files changed, 54 insertions(+), 15 deletions(-)
> 
> diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
> index b47a8dcff028..ea15771bf3da 100644
> --- a/fs/btrfs/ctree.h
> +++ b/fs/btrfs/ctree.h
> @@ -3053,6 +3053,13 @@ extern const struct iomap_ops btrfs_dio_iomap_ops;
>   extern const struct iomap_dio_ops btrfs_dio_ops;
>   extern const struct iomap_dio_ops btrfs_sync_dops;
>   
> +/* ilock flags definition */
> +#define BTRFS_ILOCK_SHARED	(1 << 0)
> +#define BTRFS_ILOCK_TRY 	(1 << 1)
> +
> +int btrfs_inode_lock(struct inode *inode, int ilock_flags);
> +void btrfs_inode_unlock(struct inode *inode, int ilock_flags);
> +
>   /* ioctl.c */
>   long btrfs_ioctl(struct file *file, unsigned int cmd, unsigned long arg);
>   long btrfs_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg);
> diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
> index 0f961ce1fa98..7e18334e8121 100644
> --- a/fs/btrfs/file.c
> +++ b/fs/btrfs/file.c
> @@ -1974,7 +1974,7 @@ static ssize_t btrfs_direct_write(struct kiocb *iocb, struct iov_iter *from)
>   	 * not unlock the i_mutex at this case.
>   	 */
>   	if (pos + iov_iter_count(from) <= inode->i_size) {
> -		inode_unlock(inode);
> +		btrfs_inode_unlock(inode, 0);
>   		relock = true;
>   	}
>   	down_read(&BTRFS_I(inode)->dio_sem);
> @@ -1995,7 +1995,7 @@ static ssize_t btrfs_direct_write(struct kiocb *iocb, struct iov_iter *from)
>   
>   	up_read(&BTRFS_I(inode)->dio_sem);
>   	if (relock)
> -		inode_lock(inode);
> +		btrfs_inode_lock(inode, 0);
>   
>   	if (written < 0 || !iov_iter_count(from))
>   		return written;
> @@ -2036,6 +2036,7 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb,
>   	ssize_t num_written = 0;
>   	const bool sync = iocb->ki_flags & IOCB_DSYNC;
>   	ssize_t err;
> +	int ilock_flags = 0;
>   
>   	/*
>   	 * If BTRFS flips readonly due to some impossible error
> @@ -2050,16 +2051,16 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb,
>   	    (iocb->ki_flags & IOCB_NOWAIT))
>   		return -EOPNOTSUPP;
>   
> -	if (iocb->ki_flags & IOCB_NOWAIT) {
> -		if (!inode_trylock(inode))
> -			return -EAGAIN;
> -	} else {
> -		inode_lock(inode);
> -	}
> +	if (iocb->ki_flags & IOCB_NOWAIT)
> +		ilock_flags |= BTRFS_ILOCK_TRY;
> +
> +	err = btrfs_inode_lock(inode, ilock_flags);
> +	if (err < 0)
> +		return err;
>   
>   	err = btrfs_write_check(iocb, from);
>   	if (err <= 0) {
> -		inode_unlock(inode);
> +		btrfs_inode_unlock(inode, ilock_flags);
>   		return err;
>   	}
>   
> @@ -2105,7 +2106,7 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb,
>   		num_written = btrfs_buffered_write(iocb, from);
>   	}
>   
> -	inode_unlock(inode);
> +	btrfs_inode_unlock(inode, ilock_flags);
>   
>   	/*
>   	 * We also have to set last_sub_trans to the current log transid,
> @@ -3405,7 +3406,7 @@ static long btrfs_fallocate(struct file *file, int mode,
>   			return ret;
>   	}
>   
> -	inode_lock(inode);
> +	btrfs_inode_lock(inode, 0);
>   
>   	if (!(mode & FALLOC_FL_KEEP_SIZE) && offset + len > inode->i_size) {
>   		ret = inode_newsize_ok(inode, offset + len);
> @@ -3644,9 +3645,9 @@ static loff_t btrfs_file_llseek(struct file *file, loff_t offset, int whence)
>   		return generic_file_llseek(file, offset, whence);
>   	case SEEK_DATA:
>   	case SEEK_HOLE:
> -		inode_lock_shared(inode);
> +		btrfs_inode_lock(inode, BTRFS_ILOCK_SHARED);
>   		offset = find_desired_extent(inode, offset, whence);
> -		inode_unlock_shared(inode);
> +		btrfs_inode_unlock(inode, BTRFS_ILOCK_SHARED);
>   		break;
>   	}
>   
> @@ -3690,10 +3691,10 @@ static ssize_t btrfs_direct_read(struct kiocb *iocb, struct iov_iter *to)
>   	if (check_direct_read(btrfs_sb(inode->i_sb), to, iocb->ki_pos))
>   		return 0;
>   
> -	inode_lock_shared(inode);
> +	btrfs_inode_lock(inode, BTRFS_ILOCK_SHARED);
>   	ret = iomap_dio_rw(iocb, to, &btrfs_dio_iomap_ops, &btrfs_dio_ops,
>   			is_sync_kiocb(iocb));
> -	inode_unlock_shared(inode);
> +	btrfs_inode_unlock(inode, BTRFS_ILOCK_SHARED);
>   	return ret;
>   }
>   
> diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
> index 0730131b6590..f305efac75ae 100644
> --- a/fs/btrfs/inode.c
> +++ b/fs/btrfs/inode.c
> @@ -96,6 +96,37 @@ static void __endio_write_update_ordered(struct btrfs_inode *inode,
>   					 const u64 offset, const u64 bytes,
>   					 const bool uptodate);
>   
> +int btrfs_inode_lock(struct inode *inode, int ilock_flags)
> +{
> +	if (ilock_flags & BTRFS_ILOCK_SHARED) {
> +		if (ilock_flags & BTRFS_ILOCK_TRY) {
> +			if (!inode_trylock_shared(inode))
> +				return -EAGAIN;
> +			else
> +				return 0;
> +		}
> +		inode_lock_shared(inode);
> +	} else {
> +		if (ilock_flags & BTRFS_ILOCK_TRY) {
> +			if (!inode_trylock(inode))
> +				return -EAGAIN;
> +			else
> +				return 0;
> +		}
> +		inode_lock(inode);
> +	}
> +	return 0;
> +}
> +
> +void btrfs_inode_unlock(struct inode *inode, int ilock_flags)
> +{
> +	if (ilock_flags & BTRFS_ILOCK_SHARED)
> +		inode_unlock_shared(inode);
> +	else
> +		inode_unlock(inode);
> +
> +}
> +

Since you are going to have to respin anyway, function comments for these 
please.  Then you can add

Reviewed-by: Josef Bacik <josef@toxicpanda.com>

Thanks,

Josef

Re: [PATCH 10/15] btrfs: Push inode locking and unlocking into buffered/direct write

[Josef Bacik]

On 9/21/20 10:43 AM, Goldwyn Rodrigues wrote:
> From: Goldwyn Rodrigues <rgoldwyn@suse.com>
> 
> Push inode locking and unlocking closer to where we perform the I/O. For
> this we need to move the write checks inside the respective functions as
> well.
> 
> pos is assigned after write checks because generic_write_check can
> modify iocb->ki_pos.
> 
> Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>

Reviewed-by: Josef Bacik <josef@toxicpanda.com>

Thanks,

Josef

Re: [PATCH 11/15] btrfs: Use inode_lock_shared() for direct writes within EOF

[Josef Bacik]

On 9/21/20 10:43 AM, Goldwyn Rodrigues wrote:
> From: Goldwyn Rodrigues <rgoldwyn@suse.com>
> 
> Direct writes within EOF are safe to be performed with inode shared lock
> to improve parallelization with other direct writes or reads because EOF
> is not changed and there is no race with truncate().
> 
> Direct reads are already performed under shared inode lock.
> 
> This patch is precursor to removing btrfs_inode->dio_sem.
> 
> Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
> ---
>   fs/btrfs/file.c | 33 +++++++++++++++++++++------------
>   1 file changed, 21 insertions(+), 12 deletions(-)
> 
> diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
> index d9c3be19d7b3..50092d24eee2 100644
> --- a/fs/btrfs/file.c
> +++ b/fs/btrfs/file.c
> @@ -1977,7 +1977,6 @@ static ssize_t btrfs_direct_write(struct kiocb *iocb, struct iov_iter *from)
>   	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
>   	loff_t pos;
>   	ssize_t written = 0;
> -	bool relock = false;
>   	ssize_t written_buffered;
>   	loff_t endbyte;
>   	int err;
> @@ -1986,6 +1985,15 @@ static ssize_t btrfs_direct_write(struct kiocb *iocb, struct iov_iter *from)
>   	if (iocb->ki_flags & IOCB_NOWAIT)
>   		ilock_flags |= BTRFS_ILOCK_TRY;
>   
> +	/*
> +	 * If the write DIO within EOF,  use a shared lock
> +	 */
> +	if (iocb->ki_pos + iov_iter_count(from) <= i_size_read(inode))
> +		ilock_flags |= BTRFS_ILOCK_SHARED;
> +	else if (iocb->ki_flags & IOCB_NOWAIT)
> +		return -EAGAIN;
> +
> +relock:

Huh?  Why are you making it so EOF extending NOWAIT writes now fail?  We are 
still using ILOCK_TRY here, so we may still not block, am I missing something? 
Thanks,

Josef

Re: [PATCH 12/15] btrfs: Remove dio_sem

[Josef Bacik]

On 9/21/20 10:43 AM, Goldwyn Rodrigues wrote:
> 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>

Thanks,

Josef

Re: [PATCH 13/15] btrfs: Call iomap_dio_complete() without inode_lock

[Josef Bacik]

On 9/21/20 10:43 AM, Goldwyn Rodrigues wrote:
> From: Goldwyn Rodrigues <rgoldwyn@suse.com>
> 
> If direct writes are called with O_DIRECT | O_DSYNC, it will result in a
> deadlock because iomap_dio_rw() is called under i_rwsem which calls
> iomap_dio_complete()
>    generic_write_sync()
>      btrfs_sync_file().
> 
> btrfs_sync_file() requires i_rwsem, so call __iomap_dio_rw() with the
> i_rwsem locked, and call iomap_dio_complete() after unlocking i_rwsem.
> 
> Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>

Reviewed-by: Josef Bacik <josef@toxicpanda.com>

Thanks,

Josef

Re: [PATCH 14/15] btrfs: Revert 09745ff88d93 ("btrfs: dio iomap DSYNC workaround")

[Josef Bacik]

On 9/21/20 10:43 AM, Goldwyn Rodrigues wrote:
> From: Goldwyn Rodrigues <rgoldwyn@suse.com>
> 
> iomap_dio_complete()->generic_write_sync() is not called under i_rwsem
> anymore, revert the workaround.
> 
> Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>

Reviewed-by: Josef Bacik <josef@toxicpanda.com>

Thanks,

Josef

Re: [PATCH 04/15] iomap: Call inode_dio_end() before generic_write_sync()

[Darrick J. Wong]

On Tue, Sep 22, 2020 at 10:20:11AM -0400, Josef Bacik wrote:
> On 9/21/20 10:43 AM, Goldwyn Rodrigues wrote:
> > From: Goldwyn Rodrigues <rgoldwyn@suse.com>
> > 
> > iomap complete routine can deadlock with btrfs_fallocate because of the
> > call to generic_write_sync().
> > 
> > P0                      P1
> > inode_lock()            fallocate(FALLOC_FL_ZERO_RANGE)
> > __iomap_dio_rw()        inode_lock()
> >                          <block>
> > <submits IO>
> > <completes IO>
> > inode_unlock()
> >                          <gets inode_lock()>
> >                          inode_dio_wait()
> > iomap_dio_complete()
> >    generic_write_sync()
> >      btrfs_file_fsync()
> >        inode_lock()
> >        <deadlock>
> > 
> > inode_dio_end() is used to notify the end of DIO data in order
> > to synchronize with truncate. Call inode_dio_end() before calling
> > generic_write_sync(), so filesystems can lock i_rwsem during a sync.
> > 
> > ---
> >   fs/iomap/direct-io.c | 2 +-
> >   1 file changed, 1 insertion(+), 1 deletion(-)
> > 
> > diff --git a/fs/iomap/direct-io.c b/fs/iomap/direct-io.c
> > index d970c6bbbe11..e01f81e7b76f 100644
> > --- a/fs/iomap/direct-io.c
> > +++ b/fs/iomap/direct-io.c
> > @@ -118,6 +118,7 @@ ssize_t iomap_dio_complete(struct iomap_dio *dio)
> >   			dio_warn_stale_pagecache(iocb->ki_filp);
> >   	}
> > +	inode_dio_end(file_inode(iocb->ki_filp));
> >   	/*
> >   	 * If this is a DSYNC write, make sure we push it to stable storage now
> >   	 * that we've written data.
> > @@ -125,7 +126,6 @@ ssize_t iomap_dio_complete(struct iomap_dio *dio)
> >   	if (ret > 0 && (dio->flags & IOMAP_DIO_NEED_SYNC))
> >   		ret = generic_write_sync(iocb, ret);
> > -	inode_dio_end(file_inode(iocb->ki_filp));
> >   	kfree(dio);
> >   	return ret;
> > 
> 
> Did you verify that xfs or ext4 don't rely on the inode_dio_end() happening
> before the generic_write_sync()?  I wouldn't expect that they would, but
> we've already run into problems making those kind of assumptions.  If it's
> fine you can add

I was gonna ask the same question, but as there's no SoB on this patch I
hadn't really looked at it yet. ;)

Operations that rely on inode_dio_wait to have blocked until all the
directios are complete could get tripped up by iomap not having done the
generic_write_sync to stabilise the metadata, but I /think/ most
operations that do that also themselves flush the file.  But I don't
really know if there's a subtlety there if the inode_dio_wait thread
manages to grab the ILOCK before the generic_write_sync thread does.

--D

> Reviewed-by: Josef Bacik <josef@toxicpanda.com>
> 
> Thanks,
> 
> Josef

Re: [PATCH 04/15] iomap: Call inode_dio_end() before generic_write_sync()

[Goldwyn Rodrigues]

On  9:31 22/09, Darrick J. Wong wrote:
> On Tue, Sep 22, 2020 at 10:20:11AM -0400, Josef Bacik wrote:
> > On 9/21/20 10:43 AM, Goldwyn Rodrigues wrote:
> > > From: Goldwyn Rodrigues <rgoldwyn@suse.com>
> > > 
> > > iomap complete routine can deadlock with btrfs_fallocate because of the
> > > call to generic_write_sync().
> > > 
> > > P0                      P1
> > > inode_lock()            fallocate(FALLOC_FL_ZERO_RANGE)
> > > __iomap_dio_rw()        inode_lock()
> > >                          <block>
> > > <submits IO>
> > > <completes IO>
> > > inode_unlock()
> > >                          <gets inode_lock()>
> > >                          inode_dio_wait()
> > > iomap_dio_complete()
> > >    generic_write_sync()
> > >      btrfs_file_fsync()
> > >        inode_lock()
> > >        <deadlock>
> > > 
> > > inode_dio_end() is used to notify the end of DIO data in order
> > > to synchronize with truncate. Call inode_dio_end() before calling
> > > generic_write_sync(), so filesystems can lock i_rwsem during a sync.
> > > 
> > > ---
> > >   fs/iomap/direct-io.c | 2 +-
> > >   1 file changed, 1 insertion(+), 1 deletion(-)
> > > 
> > > diff --git a/fs/iomap/direct-io.c b/fs/iomap/direct-io.c
> > > index d970c6bbbe11..e01f81e7b76f 100644
> > > --- a/fs/iomap/direct-io.c
> > > +++ b/fs/iomap/direct-io.c
> > > @@ -118,6 +118,7 @@ ssize_t iomap_dio_complete(struct iomap_dio *dio)
> > >   			dio_warn_stale_pagecache(iocb->ki_filp);
> > >   	}
> > > +	inode_dio_end(file_inode(iocb->ki_filp));
> > >   	/*
> > >   	 * If this is a DSYNC write, make sure we push it to stable storage now
> > >   	 * that we've written data.
> > > @@ -125,7 +126,6 @@ ssize_t iomap_dio_complete(struct iomap_dio *dio)
> > >   	if (ret > 0 && (dio->flags & IOMAP_DIO_NEED_SYNC))
> > >   		ret = generic_write_sync(iocb, ret);
> > > -	inode_dio_end(file_inode(iocb->ki_filp));
> > >   	kfree(dio);
> > >   	return ret;
> > > 
> > 
> > Did you verify that xfs or ext4 don't rely on the inode_dio_end() happening
> > before the generic_write_sync()?  I wouldn't expect that they would, but
> > we've already run into problems making those kind of assumptions.  If it's
> > fine you can add
> 
> I was gonna ask the same question, but as there's no SoB on this patch I
> hadn't really looked at it yet. ;)
> 
> Operations that rely on inode_dio_wait to have blocked until all the
> directios are complete could get tripped up by iomap not having done the
> generic_write_sync to stabilise the metadata, but I /think/ most
> operations that do that also themselves flush the file.  But I don't
> really know if there's a subtlety there if the inode_dio_wait thread
> manages to grab the ILOCK before the generic_write_sync thread does.
> 

I ran xfstests and it was successful. I am testing ext4 now.

-- 
Goldwyn

Re: [PATCH 11/15] btrfs: Use inode_lock_shared() for direct writes within EOF

[Goldwyn Rodrigues]

On 10:52 22/09, Josef Bacik wrote:
> On 9/21/20 10:43 AM, Goldwyn Rodrigues wrote:
> > From: Goldwyn Rodrigues <rgoldwyn@suse.com>
> > 
> > Direct writes within EOF are safe to be performed with inode shared lock
> > to improve parallelization with other direct writes or reads because EOF
> > is not changed and there is no race with truncate().
> > 
> > Direct reads are already performed under shared inode lock.
> > 
> > This patch is precursor to removing btrfs_inode->dio_sem.
> > 
> > Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
> > ---
> >   fs/btrfs/file.c | 33 +++++++++++++++++++++------------
> >   1 file changed, 21 insertions(+), 12 deletions(-)
> > 
> > diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
> > index d9c3be19d7b3..50092d24eee2 100644
> > --- a/fs/btrfs/file.c
> > +++ b/fs/btrfs/file.c
> > @@ -1977,7 +1977,6 @@ static ssize_t btrfs_direct_write(struct kiocb *iocb, struct iov_iter *from)
> >   	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
> >   	loff_t pos;
> >   	ssize_t written = 0;
> > -	bool relock = false;
> >   	ssize_t written_buffered;
> >   	loff_t endbyte;
> >   	int err;
> > @@ -1986,6 +1985,15 @@ static ssize_t btrfs_direct_write(struct kiocb *iocb, struct iov_iter *from)
> >   	if (iocb->ki_flags & IOCB_NOWAIT)
> >   		ilock_flags |= BTRFS_ILOCK_TRY;
> > +	/*
> > +	 * If the write DIO within EOF,  use a shared lock
> > +	 */
> > +	if (iocb->ki_pos + iov_iter_count(from) <= i_size_read(inode))
> > +		ilock_flags |= BTRFS_ILOCK_SHARED;
> > +	else if (iocb->ki_flags & IOCB_NOWAIT)
> > +		return -EAGAIN;
> > +
> > +relock:
> 
> Huh?  Why are you making it so EOF extending NOWAIT writes now fail?  We are
> still using ILOCK_TRY here, so we may still not block, am I missing
> something? Thanks,
> 

Yes, this is incorrect. I had thought of this would block on disk space
allocations. But did not consider the prealloc case.

I am removing this check to match the previous behavior.

-- 
Goldwyn

Re: [PATCH 04/15] iomap: Call inode_dio_end() before generic_write_sync()

[Dave Chinner]

On Tue, Sep 22, 2020 at 09:31:56AM -0700, Darrick J. Wong wrote:
> On Tue, Sep 22, 2020 at 10:20:11AM -0400, Josef Bacik wrote:
> > On 9/21/20 10:43 AM, Goldwyn Rodrigues wrote:
> > > From: Goldwyn Rodrigues <rgoldwyn@suse.com>
> > > 
> > > iomap complete routine can deadlock with btrfs_fallocate because of the
> > > call to generic_write_sync().
> > > 
> > > P0                      P1
> > > inode_lock()            fallocate(FALLOC_FL_ZERO_RANGE)
> > > __iomap_dio_rw()        inode_lock()
> > >                          <block>
> > > <submits IO>
> > > <completes IO>
> > > inode_unlock()
> > >                          <gets inode_lock()>
> > >                          inode_dio_wait()
> > > iomap_dio_complete()
> > >    generic_write_sync()
> > >      btrfs_file_fsync()
> > >        inode_lock()
> > >        <deadlock>
> > > 
> > > inode_dio_end() is used to notify the end of DIO data in order
> > > to synchronize with truncate. Call inode_dio_end() before calling
> > > generic_write_sync(), so filesystems can lock i_rwsem during a sync.
> > > 
> > > ---
> > >   fs/iomap/direct-io.c | 2 +-
> > >   1 file changed, 1 insertion(+), 1 deletion(-)
> > > 
> > > diff --git a/fs/iomap/direct-io.c b/fs/iomap/direct-io.c
> > > index d970c6bbbe11..e01f81e7b76f 100644
> > > --- a/fs/iomap/direct-io.c
> > > +++ b/fs/iomap/direct-io.c
> > > @@ -118,6 +118,7 @@ ssize_t iomap_dio_complete(struct iomap_dio *dio)
> > >   			dio_warn_stale_pagecache(iocb->ki_filp);
> > >   	}
> > > +	inode_dio_end(file_inode(iocb->ki_filp));
> > >   	/*
> > >   	 * If this is a DSYNC write, make sure we push it to stable storage now
> > >   	 * that we've written data.
> > > @@ -125,7 +126,6 @@ ssize_t iomap_dio_complete(struct iomap_dio *dio)
> > >   	if (ret > 0 && (dio->flags & IOMAP_DIO_NEED_SYNC))
> > >   		ret = generic_write_sync(iocb, ret);
> > > -	inode_dio_end(file_inode(iocb->ki_filp));
> > >   	kfree(dio);
> > >   	return ret;
> > > 
> > 
> > Did you verify that xfs or ext4 don't rely on the inode_dio_end() happening
> > before the generic_write_sync()?  I wouldn't expect that they would, but
> > we've already run into problems making those kind of assumptions.  If it's
> > fine you can add
> 
> I was gonna ask the same question, but as there's no SoB on this patch I
> hadn't really looked at it yet. ;)
> 
> Operations that rely on inode_dio_wait to have blocked until all the
> directios are complete could get tripped up by iomap not having done the
> generic_write_sync to stabilise the metadata, but I /think/ most
> operations that do that also themselves flush the file.  But I don't
> really know if there's a subtlety there if the inode_dio_wait thread
> manages to grab the ILOCK before the generic_write_sync thread does.

I did point out in the previous thread that this actually means that
inode_dio_wait() now has inconsistent wait semantics for O_DSYNC
writes. If it's a pure overwrite and we hit the FUA path, the
O_DSYNC write will be complete and guaranteed to be on stable storage
before the IO completes. If the inode is metadata dirty, then the IO
will now be signalled complete *before* the data and metadata are
flushed to stable storage.

Hence, from the perspective of writes to *stable* storage, this
makes the ordering of O_DSYNC DIO against anything waiting for it to
complete to be potentially inconsistent at the stable storage level.

That's an extremely subtle change of behaviour, and something that
would be largely impossible to test or reproduce. And, really, I
don't like having this sort of "oh, it should be fine" handwavy
justification when we are talking about data integrity operations...

Cheers,

Dave.
-- 
Dave Chinner
david@fromorbit.com

Re: [PATCH 04/15] iomap: Call inode_dio_end() before generic_write_sync()

[Christoph Hellwig]

On Wed, Sep 23, 2020 at 07:49:34AM +1000, Dave Chinner wrote:
> I did point out in the previous thread that this actually means that
> inode_dio_wait() now has inconsistent wait semantics for O_DSYNC
> writes. If it's a pure overwrite and we hit the FUA path, the
> O_DSYNC write will be complete and guaranteed to be on stable storage
> before the IO completes. If the inode is metadata dirty, then the IO
> will now be signalled complete *before* the data and metadata are
> flushed to stable storage.
> 
> Hence, from the perspective of writes to *stable* storage, this
> makes the ordering of O_DSYNC DIO against anything waiting for it to
> complete to be potentially inconsistent at the stable storage level.
> 
> That's an extremely subtle change of behaviour, and something that
> would be largely impossible to test or reproduce. And, really, I
> don't like having this sort of "oh, it should be fine" handwavy
> justification when we are talking about data integrity operations...

... and I replied with a detailed analysis of what it is fine, and
how this just restores the behavior we historically had before
switching to the iomap direct I/O code.  Although if we want to go
into the fine details we did not have the REQ_FUA path back then,
but that does not change the analysis.

Re: [PATCH 04/15] iomap: Call inode_dio_end() before generic_write_sync()

[Darrick J. Wong]

On Wed, Sep 23, 2020 at 07:16:58AM +0200, Christoph Hellwig wrote:
> On Wed, Sep 23, 2020 at 07:49:34AM +1000, Dave Chinner wrote:
> > I did point out in the previous thread that this actually means that
> > inode_dio_wait() now has inconsistent wait semantics for O_DSYNC
> > writes. If it's a pure overwrite and we hit the FUA path, the
> > O_DSYNC write will be complete and guaranteed to be on stable storage
> > before the IO completes. If the inode is metadata dirty, then the IO
> > will now be signalled complete *before* the data and metadata are
> > flushed to stable storage.
> > 
> > Hence, from the perspective of writes to *stable* storage, this
> > makes the ordering of O_DSYNC DIO against anything waiting for it to
> > complete to be potentially inconsistent at the stable storage level.
> > 
> > That's an extremely subtle change of behaviour, and something that
> > would be largely impossible to test or reproduce. And, really, I
> > don't like having this sort of "oh, it should be fine" handwavy
> > justification when we are talking about data integrity operations...
> 
> ... and I replied with a detailed analysis of what it is fine, and
> how this just restores the behavior we historically had before
> switching to the iomap direct I/O code.  Although if we want to go
> into the fine details we did not have the REQ_FUA path back then,
> but that does not change the analysis.

You did?  Got a link?  Not sure if vger/oraclemail are still delaying
messages for me.... :/

--D

Re: [PATCH 04/15] iomap: Call inode_dio_end() before generic_write_sync()

[Christoph Hellwig]

On Tue, Sep 22, 2020 at 10:31:49PM -0700, Darrick J. Wong wrote:
> > ... and I replied with a detailed analysis of what it is fine, and
> > how this just restores the behavior we historically had before
> > switching to the iomap direct I/O code.  Although if we want to go
> > into the fine details we did not have the REQ_FUA path back then,
> > but that does not change the analysis.
> 
> You did?  Got a link?  Not sure if vger/oraclemail are still delaying
> messages for me.... :/

Two replies from September 17 to the
"Re: [RFC PATCH] btrfs: don't call btrfs_sync_file from iomap context"

thread.

Msg IDs:

20200917055232.GA31646@lst.de

and

20200917064238.GA32441@lst.de

Re: [PATCH 04/15] iomap: Call inode_dio_end() before generic_write_sync()

[Dave Chinner]

On Wed, Sep 23, 2020 at 07:49:25AM +0200, Christoph Hellwig wrote:
> On Tue, Sep 22, 2020 at 10:31:49PM -0700, Darrick J. Wong wrote:
> > > ... and I replied with a detailed analysis of what it is fine, and
> > > how this just restores the behavior we historically had before
> > > switching to the iomap direct I/O code.  Although if we want to go
> > > into the fine details we did not have the REQ_FUA path back then,
> > > but that does not change the analysis.
> > 
> > You did?  Got a link?  Not sure if vger/oraclemail are still delaying
> > messages for me.... :/
> 
> Two replies from September 17 to the
> "Re: [RFC PATCH] btrfs: don't call btrfs_sync_file from iomap context"
> 
> thread.
> 
> Msg IDs:
> 
> 20200917055232.GA31646@lst.de
> 
> and
> 
> 20200917064238.GA32441@lst.de

<sigh>

That last one is not in my local archive - vger has been on the
blink lately, so I guess I'm not really that surprised that mail has
gone missing and not just delayed for a day or two....

Cheers,

Dave.
-- 
Dave Chinner
david@fromorbit.com

Re: [PATCH 07/15] btrfs: Move FS error state bit early during write

[Nikolay Borisov]

On 21.09.20 г. 17:43 ч., Goldwyn Rodrigues wrote:
> From: Goldwyn Rodrigues <rgoldwyn@suse.com>
> 
> fs_info->fs_state is a filesystem bit check as opposed to inode
> and can be performed before we begin with write checks. This eliminates
> inode lock/unlock in case of error bit is set.
> 
> Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
> ---
>  fs/btrfs/file.c | 21 +++++++++------------
>  1 file changed, 9 insertions(+), 12 deletions(-)
> 
> diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
> index 4c40a2742aab..ca374cb5ffc9 100644
> --- a/fs/btrfs/file.c
> +++ b/fs/btrfs/file.c
> @@ -1981,6 +1981,15 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb,
>  	size_t count;
>  	loff_t oldsize;
>  
> +	/*
> +	 * If BTRFS flips readonly due to some impossible error
> +	 * (fs_info->fs_state now has BTRFS_SUPER_FLAG_ERROR),
> +	 * although we have opened a file as writable, we have
> +	 * to stop this write operation to ensure FS consistency.
> +	 */
> +	if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state))
> +		return -EROFS;
> +

nit: Actually can't this check be eliminated altogether or the comment
vastly simplified because BTRFS_SUPER_FLAG_ERROR check is performed only
during mount so the description in the parantheses is invalid i.e the fs
won't flip to RO because BTRFS_SUPER_FLAG_ERROR is now set in the super
block. As a matter of fact how is this flag set - because I don't see it
set in the kernel code nor in btrfs-progs ?

<snip>

Re: [PATCH 07/15] btrfs: Move FS error state bit early during write

[Goldwyn Rodrigues]

On 12:10 23/09, Nikolay Borisov wrote:
> 
> 
> On 21.09.20 г. 17:43 ч., Goldwyn Rodrigues wrote:
> > From: Goldwyn Rodrigues <rgoldwyn@suse.com>
> > 
> > fs_info->fs_state is a filesystem bit check as opposed to inode
> > and can be performed before we begin with write checks. This eliminates
> > inode lock/unlock in case of error bit is set.
> > 
> > Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
> > ---
> >  fs/btrfs/file.c | 21 +++++++++------------
> >  1 file changed, 9 insertions(+), 12 deletions(-)
> > 
> > diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
> > index 4c40a2742aab..ca374cb5ffc9 100644
> > --- a/fs/btrfs/file.c
> > +++ b/fs/btrfs/file.c
> > @@ -1981,6 +1981,15 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb,
> >  	size_t count;
> >  	loff_t oldsize;
> >  
> > +	/*
> > +	 * If BTRFS flips readonly due to some impossible error
> > +	 * (fs_info->fs_state now has BTRFS_SUPER_FLAG_ERROR),
> > +	 * although we have opened a file as writable, we have
> > +	 * to stop this write operation to ensure FS consistency.
> > +	 */
> > +	if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state))
> > +		return -EROFS;
> > +
> 
> nit: Actually can't this check be eliminated altogether or the comment
> vastly simplified because BTRFS_SUPER_FLAG_ERROR check is performed only
> during mount so the description in the parantheses is invalid i.e the fs
> won't flip to RO because BTRFS_SUPER_FLAG_ERROR is now set in the super
> block. As a matter of fact how is this flag set - because I don't see it
> set in the kernel code nor in btrfs-progs ?

You are right. This flag originated from 
acce952b0263 ("Btrfs: forced readonly mounts on errors")

However, the following commit removed writing the super in case of the
error:
68ce9682a4bb ("Btrfs: remove superblock writing after fatal error")

So, it does not land in the super flags anyways.
The flag does not make sense if we use BTRFS_FS_STATE_ERROR.

I will remove BTRFS_SUPER_FLAG_ERROR comment for now.

-- 
Goldwyn