fallocate should be a file operation

Currently all filesystems except XFS implement fallocate asynchronously,
while XFS forced a commit.  Both of these are suboptimal - in case of O_SYNC
I/O we really want our allocation on disk, especially for the !KEEP_SIZE
case where we actually grow the file with user-visible zeroes.  On the
other hand always commiting the transaction is a bad idea for fast-path
uses of fallocate like for example in recent Samba versions.   Given
that block allocation is a data plane operation anyway change it from
an inode operation to a file operation so that we have the file structure
available that lets us check for O_SYNC.

This also includes moving the code around for a few of the filesystems,
and remove the already unnedded S_ISDIR checks given that we only wire
up fallocate for regular files.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

3 files changed, 5 insertions, 8 deletions

diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h
index 1de65f572033..0c8d97b56f34 100644
--- a/fs/ext4/ext4.h
+++ b/fs/ext4/ext4.h
@@ -2065,7 +2065,7 @@ extern int ext4_ext_map_blocks(handle_t *handle, struct inode *inode,
 extern void ext4_ext_truncate(struct inode *);
 extern void ext4_ext_init(struct super_block *);
 extern void ext4_ext_release(struct super_block *);
-extern long ext4_fallocate(struct inode *inode, int mode, loff_t offset,
+extern long ext4_fallocate(struct file *file, int mode, loff_t offset,
                           loff_t len);
 extern int ext4_convert_unwritten_extents(struct inode *inode, loff_t offset,
                           ssize_t len);
diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index 4bdd160854eb..63a75810b7c3 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -3627,14 +3627,15 @@ static void ext4_falloc_update_inode(struct inode *inode,
 }
 
 /*
- * preallocate space for a file. This implements ext4's fallocate inode
+ * preallocate space for a file. This implements ext4's fallocate file
  * operation, which gets called from sys_fallocate system call.
  * For block-mapped files, posix_fallocate should fall back to the method
  * of writing zeroes to the required new blocks (the same behavior which is
  * expected for file systems which do not support fallocate() system call).
  */
-long ext4_fallocate(struct inode *inode, int mode, loff_t offset, loff_t len)
+long ext4_fallocate(struct file *file, int mode, loff_t offset, loff_t len)
 {
+        struct inode *inode = file->f_path.dentry->d_inode;
         handle_t *handle;
         loff_t new_size;
         unsigned int max_blocks;
@@ -3655,10 +3656,6 @@ long ext4_fallocate(struct inode *inode, int mode, loff_t offset, loff_t len)
         if (!(ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)))
                 return -EOPNOTSUPP;
 
-        /* preallocation to directories is currently not supported */
-        if (S_ISDIR(inode->i_mode))
-                return -ENODEV;
-
         map.m_lblk = offset >> blkbits;
         /*
          * We can't just convert len to max_blocks because
diff --git a/fs/ext4/file.c b/fs/ext4/file.c
index bb003dc9ffff..2e8322c8aa88 100644
--- a/fs/ext4/file.c
+++ b/fs/ext4/file.c
@@ -210,6 +210,7 @@ const struct file_operations ext4_file_operations = {
         .fsync          = ext4_sync_file,
         .splice_read    = generic_file_splice_read,
         .splice_write   = generic_file_splice_write,
+        .fallocate      = ext4_fallocate,
 };
 
 const struct inode_operations ext4_file_inode_operations = {
@@ -223,7 +224,6 @@ const struct inode_operations ext4_file_inode_operations = {
         .removexattr    = generic_removexattr,
 #endif
         .check_acl      = ext4_check_acl,
-        .fallocate      = ext4_fallocate,
         .fiemap         = ext4_fiemap,
 };