[PATCH 5.10 2/2] vfs: fix copy_file_range() averts filesystem freeze protection

[Amir Goldstein <amir73il@gmail.com>]

commit 10bc8e4af65946b727728d7479c028742321b60a upstream.

[backport comments for pre v5.15:
- ksmbd mentions are irrelevant - ksmbd hunks were dropped
- sb_write_started() is missing - assert was dropped
]

Commit 868f9f2f8e00 ("vfs: fix copy_file_range() regression in cross-fs
copies") removed fallback to generic_copy_file_range() for cross-fs
cases inside vfs_copy_file_range().

To preserve behavior of nfsd and ksmbd server-side-copy, the fallback to
generic_copy_file_range() was added in nfsd and ksmbd code, but that
call is missing sb_start_write(), fsnotify hooks and more.

Ideally, nfsd and ksmbd would pass a flag to vfs_copy_file_range() that
will take care of the fallback, but that code would be subtle and we got
vfs_copy_file_range() logic wrong too many times already.

Instead, add a flag to explicitly request vfs_copy_file_range() to
perform only generic_copy_file_range() and let nfsd and ksmbd use this
flag only in the fallback path.

This choise keeps the logic changes to minimum in the non-nfsd/ksmbd code
paths to reduce the risk of further regressions.

Fixes: 868f9f2f8e00 ("vfs: fix copy_file_range() regression in cross-fs copies")
Tested-by: Namjae Jeon <linkinjeon@kernel.org>
Tested-by: Luis Henriques <lhenriques@suse.de>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
---
 fs/nfsd/vfs.c      |  4 ++--
 fs/read_write.c    | 17 +++++++++++++----
 include/linux/fs.h |  8 ++++++++
 3 files changed, 23 insertions(+), 6 deletions(-)

diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index 9ecaf9481a37..e6bdc16aaf05 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -582,8 +582,8 @@ ssize_t nfsd_copy_file_range(struct file *src, u64 src_pos, struct file *dst,
 	ret = vfs_copy_file_range(src, src_pos, dst, dst_pos, count, 0);
 
 	if (ret == -EOPNOTSUPP || ret == -EXDEV)
-		ret = generic_copy_file_range(src, src_pos, dst, dst_pos,
-					      count, 0);
+		ret = vfs_copy_file_range(src, src_pos, dst, dst_pos, count,
+					  COPY_FILE_SPLICE);
 	return ret;
 }
 
diff --git a/fs/read_write.c b/fs/read_write.c
index e1e83d6c21e6..ce74f3e0a346 100644
--- a/fs/read_write.c
+++ b/fs/read_write.c
@@ -1444,7 +1444,9 @@ static int generic_copy_file_checks(struct file *file_in, loff_t pos_in,
 	 * and several different sets of file_operations, but they all end up
 	 * using the same ->copy_file_range() function pointer.
 	 */
-	if (file_out->f_op->copy_file_range) {
+	if (flags & COPY_FILE_SPLICE) {
+		/* cross sb splice is allowed */
+	} else if (file_out->f_op->copy_file_range) {
 		if (file_in->f_op->copy_file_range !=
 		    file_out->f_op->copy_file_range)
 			return -EXDEV;
@@ -1494,8 +1496,9 @@ ssize_t vfs_copy_file_range(struct file *file_in, loff_t pos_in,
 			    size_t len, unsigned int flags)
 {
 	ssize_t ret;
+	bool splice = flags & COPY_FILE_SPLICE;
 
-	if (flags != 0)
+	if (flags & ~COPY_FILE_SPLICE)
 		return -EINVAL;
 
 	ret = generic_copy_file_checks(file_in, pos_in, file_out, pos_out, &len,
@@ -1521,14 +1524,14 @@ ssize_t vfs_copy_file_range(struct file *file_in, loff_t pos_in,
 	 * same sb using clone, but for filesystems where both clone and copy
 	 * are supported (e.g. nfs,cifs), we only call the copy method.
 	 */
-	if (file_out->f_op->copy_file_range) {
+	if (!splice && file_out->f_op->copy_file_range) {
 		ret = file_out->f_op->copy_file_range(file_in, pos_in,
 						      file_out, pos_out,
 						      len, flags);
 		goto done;
 	}
 
-	if (file_in->f_op->remap_file_range &&
+	if (!splice && file_in->f_op->remap_file_range &&
 	    file_inode(file_in)->i_sb == file_inode(file_out)->i_sb) {
 		ret = file_in->f_op->remap_file_range(file_in, pos_in,
 				file_out, pos_out,
@@ -1548,6 +1551,8 @@ ssize_t vfs_copy_file_range(struct file *file_in, loff_t pos_in,
 	 * consistent story about which filesystems support copy_file_range()
 	 * and which filesystems do not, that will allow userspace tools to
 	 * make consistent desicions w.r.t using copy_file_range().
+	 *
+	 * We also get here if caller (e.g. nfsd) requested COPY_FILE_SPLICE.
 	 */
 	ret = generic_copy_file_range(file_in, pos_in, file_out, pos_out, len,
 				      flags);
@@ -1602,6 +1607,10 @@ SYSCALL_DEFINE6(copy_file_range, int, fd_in, loff_t __user *, off_in,
 		pos_out = f_out.file->f_pos;
 	}
 
+	ret = -EINVAL;
+	if (flags != 0)
+		goto out;
+
 	ret = vfs_copy_file_range(f_in.file, pos_in, f_out.file, pos_out, len,
 				  flags);
 	if (ret > 0) {
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 8bde32cf9711..acf8dd6288f7 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1817,6 +1817,14 @@ struct dir_context {
  */
 #define REMAP_FILE_ADVISORY		(REMAP_FILE_CAN_SHORTEN)
 
+/*
+ * These flags control the behavior of vfs_copy_file_range().
+ * They are not available to the user via syscall.
+ *
+ * COPY_FILE_SPLICE: call splice direct instead of fs clone/copy ops
+ */
+#define COPY_FILE_SPLICE		(1 << 0)
+
 struct iov_iter;
 
 struct file_operations {
-- 
2.16.5