Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
From:   Shiyang Ruan <ruansy.fnst@cn.fujitsu.com>
To:     <linux-xfs@vger.kernel.org>, <linux-nvdimm@lists.01.org>,
        <darrick.wong@oracle.com>, <rgoldwyn@suse.de>, <hch@infradead.org>,
        <david@fromorbit.com>
CC:     <linux-kernel@vger.kernel.org>, <gujx@cn.fujitsu.com>,
        <qi.fuli@fujitsu.com>, <caoj.fnst@cn.fujitsu.com>,
        <ruansy.fnst@cn.fujitsu.com>
Subject: [RFC PATCH v2 7/7] xfs: support dedupe for fsdax.
Date:   Wed, 30 Oct 2019 12:13:58 +0800
Message-ID: <20191030041358.14450-8-ruansy.fnst@cn.fujitsu.com>
In-Reply-To: <20191030041358.14450-1-ruansy.fnst@cn.fujitsu.com>
References: <20191030041358.14450-1-ruansy.fnst@cn.fujitsu.com>
MIME-Version: 1.0
Content-Type: text/plain
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

Use xfs_break_layouts() to break files' layouts when locking them.  And
call dax_file_range_compare() function to compare range for DAX files.

Signed-off-by: Shiyang Ruan <ruansy.fnst@cn.fujitsu.com>
---
 fs/xfs/xfs_reflink.c | 77 ++++++++++++++++++++++++++------------------
 1 file changed, 45 insertions(+), 32 deletions(-)

diff --git a/fs/xfs/xfs_reflink.c b/fs/xfs/xfs_reflink.c
index e3620bc794a2..3d8d1d2f0ac0 100644
--- a/fs/xfs/xfs_reflink.c
+++ b/fs/xfs/xfs_reflink.c
@@ -1186,47 +1186,41 @@ xfs_reflink_remap_blocks(
  * back out both locks.
  */
 static int
-xfs_iolock_two_inodes_and_break_layout(
-	struct inode		*src,
-	struct inode		*dest)
+xfs_reflink_remap_lock_and_break_layouts(
+	struct file		*file_in,
+	struct file		*file_out)
 {
 	int			error;
+	struct inode		*inode_in = file_inode(file_in);
+	struct xfs_inode	*src = XFS_I(inode_in);
+	struct inode		*inode_out = file_inode(file_out);
+	struct xfs_inode	*dest = XFS_I(inode_out);
+	uint			iolock = XFS_IOLOCK_EXCL | XFS_MMAPLOCK_EXCL;
 
-	if (src > dest)
+	if (inode_in > inode_out) {
+		swap(inode_in, inode_out);
 		swap(src, dest);
-
-retry:
-	/* Wait to break both inodes' layouts before we start locking. */
-	error = break_layout(src, true);
-	if (error)
-		return error;
-	if (src != dest) {
-		error = break_layout(dest, true);
-		if (error)
-			return error;
 	}
 
-	/* Lock one inode and make sure nobody got in and leased it. */
-	inode_lock(src);
-	error = break_layout(src, false);
+	inode_lock(inode_in);
+	xfs_ilock(src, XFS_MMAPLOCK_EXCL);
+	error = xfs_break_layouts(inode_in, &iolock, BREAK_UNMAP);
+	xfs_iunlock(src, XFS_MMAPLOCK_EXCL);
 	if (error) {
-		inode_unlock(src);
-		if (error == -EWOULDBLOCK)
-			goto retry;
+		inode_unlock(inode_in);
 		return error;
 	}
 
-	if (src == dest)
+	if (inode_in == inode_out)
 		return 0;
 
-	/* Lock the other inode and make sure nobody got in and leased it. */
-	inode_lock_nested(dest, I_MUTEX_NONDIR2);
-	error = break_layout(dest, false);
+	inode_lock_nested(inode_out, I_MUTEX_NONDIR2);
+	xfs_ilock(dest, XFS_MMAPLOCK_EXCL);
+	error = xfs_break_layouts(inode_out, &iolock, BREAK_UNMAP);
+	xfs_iunlock(dest, XFS_MMAPLOCK_EXCL);
 	if (error) {
-		inode_unlock(src);
-		inode_unlock(dest);
-		if (error == -EWOULDBLOCK)
-			goto retry;
+		inode_unlock(inode_in);
+		inode_unlock(inode_out);
 		return error;
 	}
 
@@ -1245,6 +1239,11 @@ xfs_reflink_remap_unlock(
 	struct xfs_inode	*dest = XFS_I(inode_out);
 	bool			same_inode = (inode_in == inode_out);
 
+	if (inode_in > inode_out) {
+		swap(inode_in, inode_out);
+		swap(src, dest);
+	}
+
 	xfs_iunlock(dest, XFS_MMAPLOCK_EXCL);
 	if (!same_inode)
 		xfs_iunlock(src, XFS_MMAPLOCK_EXCL);
@@ -1275,6 +1274,14 @@ xfs_reflink_zero_posteof(
 			&xfs_buffered_write_iomap_ops);
 }
 
+int xfs_reflink_dedupe_file_range_compare(struct inode *src, loff_t srcoff,
+					  struct inode *dest, loff_t destoff,
+					  loff_t len, bool *is_same)
+{
+	return dax_file_range_compare(src, srcoff, dest, destoff, len, is_same,
+				      &xfs_read_iomap_ops);
+}
+
 /*
  * Prepare two files for range cloning.  Upon a successful return both inodes
  * will have the iolock and mmaplock held, the page cache of the out file will
@@ -1319,9 +1326,10 @@ xfs_reflink_remap_prep(
 	struct xfs_inode	*dest = XFS_I(inode_out);
 	bool			same_inode = (inode_in == inode_out);
 	ssize_t			ret;
+	compare_range_t		cmp;
 
 	/* Lock both files against IO */
-	ret = xfs_iolock_two_inodes_and_break_layout(inode_in, inode_out);
+	ret = xfs_reflink_remap_lock_and_break_layouts(file_in, file_out);
 	if (ret)
 		return ret;
 	if (same_inode)
@@ -1336,12 +1344,17 @@ xfs_reflink_remap_prep(
 	if (XFS_IS_REALTIME_INODE(src) || XFS_IS_REALTIME_INODE(dest))
 		goto out_unlock;
 
-	/* Don't share DAX file data for now. */
-	if (IS_DAX(inode_in) || IS_DAX(inode_out))
+	/* Don't share DAX file data with non-DAX file. */
+	if (IS_DAX(inode_in) != IS_DAX(inode_out))
 		goto out_unlock;
 
+	if (IS_DAX(inode_in))
+		cmp = xfs_reflink_dedupe_file_range_compare;
+	else
+		cmp = vfs_dedupe_file_range_compare;
+
 	ret = generic_remap_file_range_prep(file_in, pos_in, file_out, pos_out,
-			len, remap_flags, vfs_dedupe_file_range_compare);
+			len, remap_flags, cmp);
 	if (ret < 0 || *len == 0)
 		goto out_unlock;
 
-- 
2.23.0