1 files changed, 344 insertions, 214 deletions

diff --git a/fs/read_write.c b/fs/read_write.c
index 4771701c896b..befec0b5c537 100644
--- a/fs/read_write.c
+++ b/fs/read_write.c
@@ -36,22 +36,24 @@ EXPORT_SYMBOL(generic_ro_fops);
 
 static inline bool unsigned_offsets(struct file *file)
 {
-	return file->f_mode & FMODE_UNSIGNED_OFFSET;
+	return file->f_op->fop_flags & FOP_UNSIGNED_OFFSET;
 }
 
 /**
- * vfs_setpos - update the file offset for lseek
+ * vfs_setpos_cookie - update the file offset for lseek and reset cookie
  * @file:	file structure in question
  * @offset:	file offset to seek to
  * @maxsize:	maximum file size
+ * @cookie:	cookie to reset
  *
- * This is a low-level filesystem helper for updating the file offset to
- * the value specified by @offset if the given offset is valid and it is
- * not equal to the current file offset.
+ * Update the file offset to the value specified by @offset if the given
+ * offset is valid and it is not equal to the current file offset and
+ * reset the specified cookie to indicate that a seek happened.
  *
  * Return the specified offset on success and -EINVAL on invalid offset.
  */
-loff_t vfs_setpos(struct file *file, loff_t offset, loff_t maxsize)
+static loff_t vfs_setpos_cookie(struct file *file, loff_t offset,
+				loff_t maxsize, u64 *cookie)
 {
 	if (offset < 0 && !unsigned_offsets(file))
 		return -EINVAL;
@@ -60,35 +62,48 @@ loff_t vfs_setpos(struct file *file, loff_t offset, loff_t maxsize)
 
 	if (offset != file->f_pos) {
 		file->f_pos = offset;
-		file->f_version = 0;
+		if (cookie)
+			*cookie = 0;
 	}
 	return offset;
 }
-EXPORT_SYMBOL(vfs_setpos);
 
 /**
- * generic_file_llseek_size - generic llseek implementation for regular files
- * @file:	file structure to seek on
+ * vfs_setpos - update the file offset for lseek
+ * @file:	file structure in question
  * @offset:	file offset to seek to
- * @whence:	type of seek
- * @maxsize:	max size of this file in file system
- * @eof:	offset used for SEEK_END position
+ * @maxsize:	maximum file size
  *
- * This is a variant of generic_file_llseek that allows passing in a custom
- * maximum file size and a custom EOF position, for e.g. hashed directories
+ * This is a low-level filesystem helper for updating the file offset to
+ * the value specified by @offset if the given offset is valid and it is
+ * not equal to the current file offset.
  *
- * Synchronization:
- * SEEK_SET and SEEK_END are unsynchronized (but atomic on 64bit platforms)
- * SEEK_CUR is synchronized against other SEEK_CURs, but not read/writes.
- * read/writes behave like SEEK_SET against seeks.
+ * Return the specified offset on success and -EINVAL on invalid offset.
  */
-loff_t
-generic_file_llseek_size(struct file *file, loff_t offset, int whence,
-		loff_t maxsize, loff_t eof)
+loff_t vfs_setpos(struct file *file, loff_t offset, loff_t maxsize)
+{
+	return vfs_setpos_cookie(file, offset, maxsize, NULL);
+}
+EXPORT_SYMBOL(vfs_setpos);
+
+/**
+ * must_set_pos - check whether f_pos has to be updated
+ * @file: file to seek on
+ * @offset: offset to use
+ * @whence: type of seek operation
+ * @eof: end of file
+ *
+ * Check whether f_pos needs to be updated and update @offset according
+ * to @whence.
+ *
+ * Return: 0 if f_pos doesn't need to be updated, 1 if f_pos has to be
+ * updated, and negative error code on failure.
+ */
+static int must_set_pos(struct file *file, loff_t *offset, int whence, loff_t eof)
 {
 	switch (whence) {
 	case SEEK_END:
-		offset += eof;
+		*offset += eof;
 		break;
 	case SEEK_CUR:
 		/*
@@ -97,23 +112,17 @@ generic_file_llseek_size(struct file *file, loff_t offset, int whence,
 		 * f_pos value back to the file because a concurrent read(),
 		 * write() or lseek() might have altered it
 		 */
-		if (offset == 0)
-			return file->f_pos;
-		/*
-		 * f_lock protects against read/modify/write race with other
-		 * SEEK_CURs. Note that parallel writes and reads behave
-		 * like SEEK_SET.
-		 */
-		spin_lock(&file->f_lock);
-		offset = vfs_setpos(file, file->f_pos + offset, maxsize);
-		spin_unlock(&file->f_lock);
-		return offset;
+		if (*offset == 0) {
+			*offset = file->f_pos;
+			return 0;
+		}
+		break;
 	case SEEK_DATA:
 		/*
 		 * In the generic case the entire file is data, so as long as
 		 * offset isn't at the end of the file then the offset is data.
 		 */
-		if ((unsigned long long)offset >= eof)
+		if ((unsigned long long)*offset >= eof)
 			return -ENXIO;
 		break;
 	case SEEK_HOLE:
@@ -121,17 +130,103 @@ generic_file_llseek_size(struct file *file, loff_t offset, int whence,
 		 * There is a virtual hole at the end of the file, so as long as
 		 * offset isn't i_size or larger, return i_size.
 		 */
-		if ((unsigned long long)offset >= eof)
+		if ((unsigned long long)*offset >= eof)
 			return -ENXIO;
-		offset = eof;
+		*offset = eof;
 		break;
 	}
 
+	return 1;
+}
+
+/**
+ * generic_file_llseek_size - generic llseek implementation for regular files
+ * @file:	file structure to seek on
+ * @offset:	file offset to seek to
+ * @whence:	type of seek
+ * @maxsize:	max size of this file in file system
+ * @eof:	offset used for SEEK_END position
+ *
+ * This is a variant of generic_file_llseek that allows passing in a custom
+ * maximum file size and a custom EOF position, for e.g. hashed directories
+ *
+ * Synchronization:
+ * SEEK_SET and SEEK_END are unsynchronized (but atomic on 64bit platforms)
+ * SEEK_CUR is synchronized against other SEEK_CURs, but not read/writes.
+ * read/writes behave like SEEK_SET against seeks.
+ */
+loff_t
+generic_file_llseek_size(struct file *file, loff_t offset, int whence,
+		loff_t maxsize, loff_t eof)
+{
+	int ret;
+
+	ret = must_set_pos(file, &offset, whence, eof);
+	if (ret < 0)
+		return ret;
+	if (ret == 0)
+		return offset;
+
+	if (whence == SEEK_CUR) {
+		/*
+		 * f_lock protects against read/modify/write race with
+		 * other SEEK_CURs. Note that parallel writes and reads
+		 * behave like SEEK_SET.
+		 */
+		guard(spinlock)(&file->f_lock);
+		return vfs_setpos(file, file->f_pos + offset, maxsize);
+	}
+
 	return vfs_setpos(file, offset, maxsize);
 }
 EXPORT_SYMBOL(generic_file_llseek_size);
 
 /**
+ * generic_llseek_cookie - versioned llseek implementation
+ * @file:	file structure to seek on
+ * @offset:	file offset to seek to
+ * @whence:	type of seek
+ * @cookie:	cookie to update
+ *
+ * See generic_file_llseek for a general description and locking assumptions.
+ *
+ * In contrast to generic_file_llseek, this function also resets a
+ * specified cookie to indicate a seek took place.
+ */
+loff_t generic_llseek_cookie(struct file *file, loff_t offset, int whence,
+			     u64 *cookie)
+{
+	struct inode *inode = file->f_mapping->host;
+	loff_t maxsize = inode->i_sb->s_maxbytes;
+	loff_t eof = i_size_read(inode);
+	int ret;
+
+	if (WARN_ON_ONCE(!cookie))
+		return -EINVAL;
+
+	/*
+	 * Require that this is only used for directories that guarantee
+	 * synchronization between readdir and seek so that an update to
+	 * @cookie is correctly synchronized with concurrent readdir.
+	 */
+	if (WARN_ON_ONCE(!(file->f_mode & FMODE_ATOMIC_POS)))
+		return -EINVAL;
+
+	ret = must_set_pos(file, &offset, whence, eof);
+	if (ret < 0)
+		return ret;
+	if (ret == 0)
+		return offset;
+
+	/* No need to hold f_lock because we know that f_pos_lock is held. */
+	if (whence == SEEK_CUR)
+		return vfs_setpos_cookie(file, file->f_pos + offset, maxsize, cookie);
+
+	return vfs_setpos_cookie(file, offset, maxsize, cookie);
+}
+EXPORT_SYMBOL(generic_llseek_cookie);
+
+/**
  * generic_file_llseek - generic llseek implementation for regular files
  * @file:	file structure to seek on
  * @offset:	file offset to seek to
@@ -270,10 +365,8 @@ loff_t default_llseek(struct file *file, loff_t offset, int whence)
 	}
 	retval = -EINVAL;
 	if (offset >= 0 || unsigned_offsets(file)) {
-		if (offset != file->f_pos) {
+		if (offset != file->f_pos)
 			file->f_pos = offset;
-			file->f_version = 0;
-		}
 		retval = offset;
 	}
 out:
@@ -294,12 +387,12 @@ static off_t ksys_lseek(unsigned int fd, off_t offset, unsigned int whence)
 {
 	off_t retval;
 	struct fd f = fdget_pos(fd);
-	if (!f.file)
+	if (!fd_file(f))
 		return -EBADF;
 
 	retval = -EINVAL;
 	if (whence <= SEEK_MAX) {
-		loff_t res = vfs_llseek(f.file, offset, whence);
+		loff_t res = vfs_llseek(fd_file(f), offset, whence);
 		retval = res;
 		if (res != (loff_t)retval)
 			retval = -EOVERFLOW;	/* LFS: should only happen on 32 bit platforms */
@@ -330,14 +423,14 @@ SYSCALL_DEFINE5(llseek, unsigned int, fd, unsigned long, offset_high,
 	struct fd f = fdget_pos(fd);
 	loff_t offset;
 
-	if (!f.file)
+	if (!fd_file(f))
 		return -EBADF;
 
 	retval = -EINVAL;
 	if (whence > SEEK_MAX)
 		goto out_putf;
 
-	offset = vfs_llseek(f.file, ((loff_t) offset_high << 32) | offset_low,
+	offset = vfs_llseek(fd_file(f), ((loff_t) offset_high << 32) | offset_low,
 			whence);
 
 	retval = (int)offset;
@@ -354,6 +447,9 @@ out_putf:
 
 int rw_verify_area(int read_write, struct file *file, const loff_t *ppos, size_t count)
 {
+	int mask = read_write == READ ? MAY_READ : MAY_WRITE;
+	int ret;
+
 	if (unlikely((ssize_t) count < 0))
 		return -EINVAL;
 
@@ -371,8 +467,11 @@ int rw_verify_area(int read_write, struct file *file, const loff_t *ppos, size_t
 		}
 	}
 
-	return security_file_permission(file,
-				read_write == READ ? MAY_READ : MAY_WRITE);
+	ret = security_file_permission(file, mask);
+	if (ret)
+		return ret;
+
+	return fsnotify_file_area_perm(file, mask, ppos, count);
 }
 EXPORT_SYMBOL(rw_verify_area);
 
@@ -386,7 +485,7 @@ static ssize_t new_sync_read(struct file *filp, char __user *buf, size_t len, lo
 	kiocb.ki_pos = (ppos ? *ppos : 0);
 	iov_iter_ubuf(&iter, ITER_DEST, buf, len);
 
-	ret = call_read_iter(filp, &kiocb, &iter);
+	ret = filp->f_op->read_iter(&kiocb, &iter);
 	BUG_ON(ret == -EIOCBQUEUED);
 	if (ppos)
 		*ppos = kiocb.ki_pos;
@@ -488,7 +587,7 @@ static ssize_t new_sync_write(struct file *filp, const char __user *buf, size_t
 	kiocb.ki_pos = (ppos ? *ppos : 0);
 	iov_iter_ubuf(&iter, ITER_SOURCE, (void __user *)buf, len);
 
-	ret = call_write_iter(filp, &kiocb, &iter);
+	ret = filp->f_op->write_iter(&kiocb, &iter);
 	BUG_ON(ret == -EIOCBQUEUED);
 	if (ret > 0 && ppos)
 		*ppos = kiocb.ki_pos;
@@ -604,15 +703,15 @@ ssize_t ksys_read(unsigned int fd, char __user *buf, size_t count)
 	struct fd f = fdget_pos(fd);
 	ssize_t ret = -EBADF;
 
-	if (f.file) {
-		loff_t pos, *ppos = file_ppos(f.file);
+	if (fd_file(f)) {
+		loff_t pos, *ppos = file_ppos(fd_file(f));
 		if (ppos) {
 			pos = *ppos;
 			ppos = &pos;
 		}
-		ret = vfs_read(f.file, buf, count, ppos);
+		ret = vfs_read(fd_file(f), buf, count, ppos);
 		if (ret >= 0 && ppos)
-			f.file->f_pos = pos;
+			fd_file(f)->f_pos = pos;
 		fdput_pos(f);
 	}
 	return ret;
@@ -628,15 +727,15 @@ ssize_t ksys_write(unsigned int fd, const char __user *buf, size_t count)
 	struct fd f = fdget_pos(fd);
 	ssize_t ret = -EBADF;
 
-	if (f.file) {
-		loff_t pos, *ppos = file_ppos(f.file);
+	if (fd_file(f)) {
+		loff_t pos, *ppos = file_ppos(fd_file(f));
 		if (ppos) {
 			pos = *ppos;
 			ppos = &pos;
 		}
-		ret = vfs_write(f.file, buf, count, ppos);
+		ret = vfs_write(fd_file(f), buf, count, ppos);
 		if (ret >= 0 && ppos)
-			f.file->f_pos = pos;
+			fd_file(f)->f_pos = pos;
 		fdput_pos(f);
 	}
 
@@ -659,10 +758,10 @@ ssize_t ksys_pread64(unsigned int fd, char __user *buf, size_t count,
 		return -EINVAL;
 
 	f = fdget(fd);
-	if (f.file) {
+	if (fd_file(f)) {
 		ret = -ESPIPE;
-		if (f.file->f_mode & FMODE_PREAD)
-			ret = vfs_read(f.file, buf, count, &pos);
+		if (fd_file(f)->f_mode & FMODE_PREAD)
+			ret = vfs_read(fd_file(f), buf, count, &pos);
 		fdput(f);
 	}
 
@@ -693,10 +792,10 @@ ssize_t ksys_pwrite64(unsigned int fd, const char __user *buf,
 		return -EINVAL;
 
 	f = fdget(fd);
-	if (f.file) {
+	if (fd_file(f)) {
 		ret = -ESPIPE;
-		if (f.file->f_mode & FMODE_PWRITE)  
-			ret = vfs_write(f.file, buf, count, &pos);
+		if (fd_file(f)->f_mode & FMODE_PWRITE)
+			ret = vfs_write(fd_file(f), buf, count, &pos);
 		fdput(f);
 	}
 
@@ -724,15 +823,15 @@ static ssize_t do_iter_readv_writev(struct file *filp, struct iov_iter *iter,
 	ssize_t ret;
 
 	init_sync_kiocb(&kiocb, filp);
-	ret = kiocb_set_rw_flags(&kiocb, flags);
+	ret = kiocb_set_rw_flags(&kiocb, flags, type);
 	if (ret)
 		return ret;
 	kiocb.ki_pos = (ppos ? *ppos : 0);
 
 	if (type == READ)
-		ret = call_read_iter(filp, &kiocb, iter);
+		ret = filp->f_op->read_iter(&kiocb, iter);
 	else
-		ret = call_write_iter(filp, &kiocb, iter);
+		ret = filp->f_op->write_iter(&kiocb, iter);
 	BUG_ON(ret == -EIOCBQUEUED);
 	if (ppos)
 		*ppos = kiocb.ki_pos;
@@ -773,12 +872,14 @@ static ssize_t do_loop_readv_writev(struct file *filp, struct iov_iter *iter,
 	return ret;
 }
 
-static ssize_t do_iter_read(struct file *file, struct iov_iter *iter,
-		loff_t *pos, rwf_t flags)
+ssize_t vfs_iocb_iter_read(struct file *file, struct kiocb *iocb,
+			   struct iov_iter *iter)
 {
 	size_t tot_len;
 	ssize_t ret = 0;
 
+	if (!file->f_op->read_iter)
+		return -EINVAL;
 	if (!(file->f_mode & FMODE_READ))
 		return -EBADF;
 	if (!(file->f_mode & FMODE_CAN_READ))
@@ -787,22 +888,20 @@ static ssize_t do_iter_read(struct file *file, struct iov_iter *iter,
 	tot_len = iov_iter_count(iter);
 	if (!tot_len)
 		goto out;
-	ret = rw_verify_area(READ, file, pos, tot_len);
+	ret = rw_verify_area(READ, file, &iocb->ki_pos, tot_len);
 	if (ret < 0)
 		return ret;
 
-	if (file->f_op->read_iter)
-		ret = do_iter_readv_writev(file, iter, pos, READ, flags);
-	else
-		ret = do_loop_readv_writev(file, iter, pos, READ, flags);
+	ret = file->f_op->read_iter(iocb, iter);
 out:
 	if (ret >= 0)
 		fsnotify_access(file);
 	return ret;
 }
+EXPORT_SYMBOL(vfs_iocb_iter_read);
 
-ssize_t vfs_iocb_iter_read(struct file *file, struct kiocb *iocb,
-			   struct iov_iter *iter)
+ssize_t vfs_iter_read(struct file *file, struct iov_iter *iter, loff_t *ppos,
+		      rwf_t flags)
 {
 	size_t tot_len;
 	ssize_t ret = 0;
@@ -817,33 +916,30 @@ ssize_t vfs_iocb_iter_read(struct file *file, struct kiocb *iocb,
 	tot_len = iov_iter_count(iter);
 	if (!tot_len)
 		goto out;
-	ret = rw_verify_area(READ, file, &iocb->ki_pos, tot_len);
+	ret = rw_verify_area(READ, file, ppos, tot_len);
 	if (ret < 0)
 		return ret;
 
-	ret = call_read_iter(file, iocb, iter);
+	ret = do_iter_readv_writev(file, iter, ppos, READ, flags);
 out:
 	if (ret >= 0)
 		fsnotify_access(file);
 	return ret;
 }
-EXPORT_SYMBOL(vfs_iocb_iter_read);
-
-ssize_t vfs_iter_read(struct file *file, struct iov_iter *iter, loff_t *ppos,
-		rwf_t flags)
-{
-	if (!file->f_op->read_iter)
-		return -EINVAL;
-	return do_iter_read(file, iter, ppos, flags);
-}
 EXPORT_SYMBOL(vfs_iter_read);
 
-static ssize_t do_iter_write(struct file *file, struct iov_iter *iter,
-		loff_t *pos, rwf_t flags)
+/*
+ * Caller is responsible for calling kiocb_end_write() on completion
+ * if async iocb was queued.
+ */
+ssize_t vfs_iocb_iter_write(struct file *file, struct kiocb *iocb,
+			    struct iov_iter *iter)
 {
 	size_t tot_len;
 	ssize_t ret = 0;
 
+	if (!file->f_op->write_iter)
+		return -EINVAL;
 	if (!(file->f_mode & FMODE_WRITE))
 		return -EBADF;
 	if (!(file->f_mode & FMODE_CAN_WRITE))
@@ -852,88 +948,127 @@ static ssize_t do_iter_write(struct file *file, struct iov_iter *iter,
 	tot_len = iov_iter_count(iter);
 	if (!tot_len)
 		return 0;
-	ret = rw_verify_area(WRITE, file, pos, tot_len);
+	ret = rw_verify_area(WRITE, file, &iocb->ki_pos, tot_len);
 	if (ret < 0)
 		return ret;
 
-	if (file->f_op->write_iter)
-		ret = do_iter_readv_writev(file, iter, pos, WRITE, flags);
-	else
-		ret = do_loop_readv_writev(file, iter, pos, WRITE, flags);
+	kiocb_start_write(iocb);
+	ret = file->f_op->write_iter(iocb, iter);
+	if (ret != -EIOCBQUEUED)
+		kiocb_end_write(iocb);
 	if (ret > 0)
 		fsnotify_modify(file);
+
 	return ret;
 }
+EXPORT_SYMBOL(vfs_iocb_iter_write);
 
-ssize_t vfs_iocb_iter_write(struct file *file, struct kiocb *iocb,
-			    struct iov_iter *iter)
+ssize_t vfs_iter_write(struct file *file, struct iov_iter *iter, loff_t *ppos,
+		       rwf_t flags)
 {
 	size_t tot_len;
-	ssize_t ret = 0;
+	ssize_t ret;
 
-	if (!file->f_op->write_iter)
-		return -EINVAL;
 	if (!(file->f_mode & FMODE_WRITE))
 		return -EBADF;
 	if (!(file->f_mode & FMODE_CAN_WRITE))
 		return -EINVAL;
+	if (!file->f_op->write_iter)
+		return -EINVAL;
 
 	tot_len = iov_iter_count(iter);
 	if (!tot_len)
 		return 0;
-	ret = rw_verify_area(WRITE, file, &iocb->ki_pos, tot_len);
+
+	ret = rw_verify_area(WRITE, file, ppos, tot_len);
 	if (ret < 0)
 		return ret;
 
-	ret = call_write_iter(file, iocb, iter);
+	file_start_write(file);
+	ret = do_iter_readv_writev(file, iter, ppos, WRITE, flags);
 	if (ret > 0)
 		fsnotify_modify(file);
+	file_end_write(file);
 
 	return ret;
 }
-EXPORT_SYMBOL(vfs_iocb_iter_write);
-
-ssize_t vfs_iter_write(struct file *file, struct iov_iter *iter, loff_t *ppos,
-		rwf_t flags)
-{
-	if (!file->f_op->write_iter)
-		return -EINVAL;
-	return do_iter_write(file, iter, ppos, flags);
-}
 EXPORT_SYMBOL(vfs_iter_write);
 
 static ssize_t vfs_readv(struct file *file, const struct iovec __user *vec,
-		  unsigned long vlen, loff_t *pos, rwf_t flags)
+			 unsigned long vlen, loff_t *pos, rwf_t flags)
 {
 	struct iovec iovstack[UIO_FASTIOV];
 	struct iovec *iov = iovstack;
 	struct iov_iter iter;
-	ssize_t ret;
+	size_t tot_len;
+	ssize_t ret = 0;
 
-	ret = import_iovec(ITER_DEST, vec, vlen, ARRAY_SIZE(iovstack), &iov, &iter);
-	if (ret >= 0) {
-		ret = do_iter_read(file, &iter, pos, flags);
-		kfree(iov);
-	}
+	if (!(file->f_mode & FMODE_READ))
+		return -EBADF;
+	if (!(file->f_mode & FMODE_CAN_READ))
+		return -EINVAL;
+
+	ret = import_iovec(ITER_DEST, vec, vlen, ARRAY_SIZE(iovstack), &iov,
+			   &iter);
+	if (ret < 0)
+		return ret;
+
+	tot_len = iov_iter_count(&iter);
+	if (!tot_len)
+		goto out;
 
+	ret = rw_verify_area(READ, file, pos, tot_len);
+	if (ret < 0)
+		goto out;
+
+	if (file->f_op->read_iter)
+		ret = do_iter_readv_writev(file, &iter, pos, READ, flags);
+	else
+		ret = do_loop_readv_writev(file, &iter, pos, READ, flags);
+out:
+	if (ret >= 0)
+		fsnotify_access(file);
+	kfree(iov);
 	return ret;
 }
 
 static ssize_t vfs_writev(struct file *file, const struct iovec __user *vec,
-		   unsigned long vlen, loff_t *pos, rwf_t flags)
+			  unsigned long vlen, loff_t *pos, rwf_t flags)
 {
 	struct iovec iovstack[UIO_FASTIOV];
 	struct iovec *iov = iovstack;
 	struct iov_iter iter;
-	ssize_t ret;
+	size_t tot_len;
+	ssize_t ret = 0;
 
-	ret = import_iovec(ITER_SOURCE, vec, vlen, ARRAY_SIZE(iovstack), &iov, &iter);
-	if (ret >= 0) {
-		file_start_write(file);
-		ret = do_iter_write(file, &iter, pos, flags);
-		file_end_write(file);
-		kfree(iov);
-	}
+	if (!(file->f_mode & FMODE_WRITE))
+		return -EBADF;
+	if (!(file->f_mode & FMODE_CAN_WRITE))
+		return -EINVAL;
+
+	ret = import_iovec(ITER_SOURCE, vec, vlen, ARRAY_SIZE(iovstack), &iov,
+			   &iter);
+	if (ret < 0)
+		return ret;
+
+	tot_len = iov_iter_count(&iter);
+	if (!tot_len)
+		goto out;
+
+	ret = rw_verify_area(WRITE, file, pos, tot_len);
+	if (ret < 0)
+		goto out;
+
+	file_start_write(file);
+	if (file->f_op->write_iter)
+		ret = do_iter_readv_writev(file, &iter, pos, WRITE, flags);
+	else
+		ret = do_loop_readv_writev(file, &iter, pos, WRITE, flags);
+	if (ret > 0)
+		fsnotify_modify(file);
+	file_end_write(file);
+out:
+	kfree(iov);
 	return ret;
 }
 
@@ -943,15 +1078,15 @@ static ssize_t do_readv(unsigned long fd, const struct iovec __user *vec,
 	struct fd f = fdget_pos(fd);
 	ssize_t ret = -EBADF;
 
-	if (f.file) {
-		loff_t pos, *ppos = file_ppos(f.file);
+	if (fd_file(f)) {
+		loff_t pos, *ppos = file_ppos(fd_file(f));
 		if (ppos) {
 			pos = *ppos;
 			ppos = &pos;
 		}
-		ret = vfs_readv(f.file, vec, vlen, ppos, flags);
+		ret = vfs_readv(fd_file(f), vec, vlen, ppos, flags);
 		if (ret >= 0 && ppos)
-			f.file->f_pos = pos;
+			fd_file(f)->f_pos = pos;
 		fdput_pos(f);
 	}
 
@@ -967,15 +1102,15 @@ static ssize_t do_writev(unsigned long fd, const struct iovec __user *vec,
 	struct fd f = fdget_pos(fd);
 	ssize_t ret = -EBADF;
 
-	if (f.file) {
-		loff_t pos, *ppos = file_ppos(f.file);
+	if (fd_file(f)) {
+		loff_t pos, *ppos = file_ppos(fd_file(f));
 		if (ppos) {
 			pos = *ppos;
 			ppos = &pos;
 		}
-		ret = vfs_writev(f.file, vec, vlen, ppos, flags);
+		ret = vfs_writev(fd_file(f), vec, vlen, ppos, flags);
 		if (ret >= 0 && ppos)
-			f.file->f_pos = pos;
+			fd_file(f)->f_pos = pos;
 		fdput_pos(f);
 	}
 
@@ -1001,10 +1136,10 @@ static ssize_t do_preadv(unsigned long fd, const struct iovec __user *vec,
 		return -EINVAL;
 
 	f = fdget(fd);
-	if (f.file) {
+	if (fd_file(f)) {
 		ret = -ESPIPE;
-		if (f.file->f_mode & FMODE_PREAD)
-			ret = vfs_readv(f.file, vec, vlen, &pos, flags);
+		if (fd_file(f)->f_mode & FMODE_PREAD)
+			ret = vfs_readv(fd_file(f), vec, vlen, &pos, flags);
 		fdput(f);
 	}
 
@@ -1024,10 +1159,10 @@ static ssize_t do_pwritev(unsigned long fd, const struct iovec __user *vec,
 		return -EINVAL;
 
 	f = fdget(fd);
-	if (f.file) {
+	if (fd_file(f)) {
 		ret = -ESPIPE;
-		if (f.file->f_mode & FMODE_PWRITE)
-			ret = vfs_writev(f.file, vec, vlen, &pos, flags);
+		if (fd_file(f)->f_mode & FMODE_PWRITE)
+			ret = vfs_writev(fd_file(f), vec, vlen, &pos, flags);
 		fdput(f);
 	}
 
@@ -1178,7 +1313,7 @@ COMPAT_SYSCALL_DEFINE6(pwritev2, compat_ulong_t, fd,
 #endif /* CONFIG_COMPAT */
 
 static ssize_t do_sendfile(int out_fd, int in_fd, loff_t *ppos,
-		  	   size_t count, loff_t max)
+			   size_t count, loff_t max)
 {
 	struct fd in, out;
 	struct inode *in_inode, *out_inode;
@@ -1193,19 +1328,19 @@ static ssize_t do_sendfile(int out_fd, int in_fd, loff_t *ppos,
 	 */
 	retval = -EBADF;
 	in = fdget(in_fd);
-	if (!in.file)
+	if (!fd_file(in))
 		goto out;
-	if (!(in.file->f_mode & FMODE_READ))
+	if (!(fd_file(in)->f_mode & FMODE_READ))
 		goto fput_in;
 	retval = -ESPIPE;
 	if (!ppos) {
-		pos = in.file->f_pos;
+		pos = fd_file(in)->f_pos;
 	} else {
 		pos = *ppos;
-		if (!(in.file->f_mode & FMODE_PREAD))
+		if (!(fd_file(in)->f_mode & FMODE_PREAD))
 			goto fput_in;
 	}
-	retval = rw_verify_area(READ, in.file, &pos, count);
+	retval = rw_verify_area(READ, fd_file(in), &pos, count);
 	if (retval < 0)
 		goto fput_in;
 	if (count > MAX_RW_COUNT)
@@ -1216,13 +1351,13 @@ static ssize_t do_sendfile(int out_fd, int in_fd, loff_t *ppos,
 	 */
 	retval = -EBADF;
 	out = fdget(out_fd);
-	if (!out.file)
+	if (!fd_file(out))
 		goto fput_in;
-	if (!(out.file->f_mode & FMODE_WRITE))
+	if (!(fd_file(out)->f_mode & FMODE_WRITE))
 		goto fput_out;
-	in_inode = file_inode(in.file);
-	out_inode = file_inode(out.file);
-	out_pos = out.file->f_pos;
+	in_inode = file_inode(fd_file(in));
+	out_inode = file_inode(fd_file(out));
+	out_pos = fd_file(out)->f_pos;
 
 	if (!max)
 		max = min(in_inode->i_sb->s_maxbytes, out_inode->i_sb->s_maxbytes);
@@ -1242,35 +1377,33 @@ static ssize_t do_sendfile(int out_fd, int in_fd, loff_t *ppos,
 	 * and the application is arguably buggy if it doesn't expect
 	 * EAGAIN on a non-blocking file descriptor.
 	 */
-	if (in.file->f_flags & O_NONBLOCK)
+	if (fd_file(in)->f_flags & O_NONBLOCK)
 		fl = SPLICE_F_NONBLOCK;
 #endif
-	opipe = get_pipe_info(out.file, true);
+	opipe = get_pipe_info(fd_file(out), true);
 	if (!opipe) {
-		retval = rw_verify_area(WRITE, out.file, &out_pos, count);
+		retval = rw_verify_area(WRITE, fd_file(out), &out_pos, count);
 		if (retval < 0)
 			goto fput_out;
-		file_start_write(out.file);
-		retval = do_splice_direct(in.file, &pos, out.file, &out_pos,
+		retval = do_splice_direct(fd_file(in), &pos, fd_file(out), &out_pos,
 					  count, fl);
-		file_end_write(out.file);
 	} else {
-		if (out.file->f_flags & O_NONBLOCK)
+		if (fd_file(out)->f_flags & O_NONBLOCK)
 			fl |= SPLICE_F_NONBLOCK;
 
-		retval = splice_file_to_pipe(in.file, opipe, &pos, count, fl);
+		retval = splice_file_to_pipe(fd_file(in), opipe, &pos, count, fl);
 	}
 
 	if (retval > 0) {
 		add_rchar(current, retval);
 		add_wchar(current, retval);
-		fsnotify_access(in.file);
-		fsnotify_modify(out.file);
-		out.file->f_pos = out_pos;
+		fsnotify_access(fd_file(in));
+		fsnotify_modify(fd_file(out));
+		fd_file(out)->f_pos = out_pos;
 		if (ppos)
 			*ppos = pos;
 		else
-			in.file->f_pos = pos;
+			fd_file(in)->f_pos = pos;
 	}
 
 	inc_syscr(current);
@@ -1362,38 +1495,6 @@ COMPAT_SYSCALL_DEFINE4(sendfile64, int, out_fd, int, in_fd,
 }
 #endif
 
-/**
- * generic_copy_file_range - copy data between two files
- * @file_in:	file structure to read from
- * @pos_in:	file offset to read from
- * @file_out:	file structure to write data to
- * @pos_out:	file offset to write data to
- * @len:	amount of data to copy
- * @flags:	copy flags
- *
- * This is a generic filesystem helper to copy data from one file to another.
- * It has no constraints on the source or destination file owners - the files
- * can belong to different superblocks and different filesystem types. Short
- * copies are allowed.
- *
- * This should be called from the @file_out filesystem, as per the
- * ->copy_file_range() method.
- *
- * Returns the number of bytes copied or a negative error indicating the
- * failure.
- */
-
-ssize_t generic_copy_file_range(struct file *file_in, loff_t pos_in,
-				struct file *file_out, loff_t pos_out,
-				size_t len, unsigned int flags)
-{
-	lockdep_assert(sb_write_started(file_inode(file_out)->i_sb));
-
-	return do_splice_direct(file_in, &pos_in, file_out, &pos_out,
-				len > MAX_RW_COUNT ? MAX_RW_COUNT : len, 0);
-}
-EXPORT_SYMBOL(generic_copy_file_range);
-
 /*
  * Performs necessary checks before doing a file copy
  *
@@ -1478,6 +1579,7 @@ ssize_t vfs_copy_file_range(struct file *file_in, loff_t pos_in,
 {
 	ssize_t ret;
 	bool splice = flags & COPY_FILE_SPLICE;
+	bool samesb = file_inode(file_in)->i_sb == file_inode(file_out)->i_sb;
 
 	if (flags & ~COPY_FILE_SPLICE)
 		return -EINVAL;
@@ -1509,19 +1611,24 @@ ssize_t vfs_copy_file_range(struct file *file_in, loff_t pos_in,
 		ret = file_out->f_op->copy_file_range(file_in, pos_in,
 						      file_out, pos_out,
 						      len, flags);
-		goto done;
-	}
-
-	if (!splice && file_in->f_op->remap_file_range &&
-	    file_inode(file_in)->i_sb == file_inode(file_out)->i_sb) {
+	} else if (!splice && file_in->f_op->remap_file_range && samesb) {
 		ret = file_in->f_op->remap_file_range(file_in, pos_in,
 				file_out, pos_out,
 				min_t(loff_t, MAX_RW_COUNT, len),
 				REMAP_FILE_CAN_SHORTEN);
-		if (ret > 0)
-			goto done;
+		/* fallback to splice */
+		if (ret <= 0)
+			splice = true;
+	} else if (samesb) {
+		/* Fallback to splice for same sb copy for backward compat */
+		splice = true;
 	}
 
+	file_end_write(file_out);
+
+	if (!splice)
+		goto done;
+
 	/*
 	 * We can get here for same sb copy of filesystems that do not implement
 	 * ->copy_file_range() in case filesystem does not support clone or in
@@ -1533,11 +1640,16 @@ ssize_t vfs_copy_file_range(struct file *file_in, loff_t pos_in,
 	 * 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.
+	 * We also get here if caller (e.g. nfsd) requested COPY_FILE_SPLICE
+	 * for server-side-copy between any two sb.
+	 *
+	 * In any case, we call do_splice_direct() and not splice_file_range(),
+	 * without file_start_write() held, to avoid possible deadlocks related
+	 * to splicing from input file, while file_start_write() is held on
+	 * the output file on a different sb.
 	 */
-	ret = generic_copy_file_range(file_in, pos_in, file_out, pos_out, len,
-				      flags);
-
+	ret = do_splice_direct(file_in, &pos_in, file_out, &pos_out,
+			       min_t(size_t, len, MAX_RW_COUNT), 0);
 done:
 	if (ret > 0) {
 		fsnotify_access(file_in);
@@ -1549,8 +1661,6 @@ done:
 	inc_syscr(current);
 	inc_syscw(current);
 
-	file_end_write(file_out);
-
 	return ret;
 }
 EXPORT_SYMBOL(vfs_copy_file_range);
@@ -1566,11 +1676,11 @@ SYSCALL_DEFINE6(copy_file_range, int, fd_in, loff_t __user *, off_in,
 	ssize_t ret = -EBADF;
 
 	f_in = fdget(fd_in);
-	if (!f_in.file)
+	if (!fd_file(f_in))
 		goto out2;
 
 	f_out = fdget(fd_out);
-	if (!f_out.file)
+	if (!fd_file(f_out))
 		goto out1;
 
 	ret = -EFAULT;
@@ -1578,21 +1688,21 @@ SYSCALL_DEFINE6(copy_file_range, int, fd_in, loff_t __user *, off_in,
 		if (copy_from_user(&pos_in, off_in, sizeof(loff_t)))
 			goto out;
 	} else {
-		pos_in = f_in.file->f_pos;
+		pos_in = fd_file(f_in)->f_pos;
 	}
 
 	if (off_out) {
 		if (copy_from_user(&pos_out, off_out, sizeof(loff_t)))
 			goto out;
 	} else {
-		pos_out = f_out.file->f_pos;
+		pos_out = fd_file(f_out)->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,
+	ret = vfs_copy_file_range(fd_file(f_in), pos_in, fd_file(f_out), pos_out, len,
 				  flags);
 	if (ret > 0) {
 		pos_in += ret;
@@ -1602,14 +1712,14 @@ SYSCALL_DEFINE6(copy_file_range, int, fd_in, loff_t __user *, off_in,
 			if (copy_to_user(off_in, &pos_in, sizeof(loff_t)))
 				ret = -EFAULT;
 		} else {
-			f_in.file->f_pos = pos_in;
+			fd_file(f_in)->f_pos = pos_in;
 		}
 
 		if (off_out) {
 			if (copy_to_user(off_out, &pos_out, sizeof(loff_t)))
 				ret = -EFAULT;
 		} else {
-			f_out.file->f_pos = pos_out;
+			fd_file(f_out)->f_pos = pos_out;
 		}
 	}
 
@@ -1650,6 +1760,7 @@ int generic_write_check_limits(struct file *file, loff_t pos, loff_t *count)
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(generic_write_check_limits);
 
 /* Like generic_write_checks(), but takes size of write instead of iter. */
 int generic_write_checks_count(struct kiocb *iocb, loff_t *count)
@@ -1668,7 +1779,7 @@ int generic_write_checks_count(struct kiocb *iocb, loff_t *count)
 
 	if ((iocb->ki_flags & IOCB_NOWAIT) &&
 	    !((iocb->ki_flags & IOCB_DIRECT) ||
-	      (file->f_mode & FMODE_BUF_WASYNC)))
+	      (file->f_op->fop_flags & FOP_BUFFER_WASYNC)))
 		return -EINVAL;
 
 	return generic_write_check_limits(iocb->ki_filp, iocb->ki_pos, count);
@@ -1718,3 +1829,22 @@ int generic_file_rw_checks(struct file *file_in, struct file *file_out)
 
 	return 0;
 }
+
+int generic_atomic_write_valid(struct kiocb *iocb, struct iov_iter *iter)
+{
+	size_t len = iov_iter_count(iter);
+
+	if (!iter_is_ubuf(iter))
+		return -EINVAL;
+
+	if (!is_power_of_2(len))
+		return -EINVAL;
+
+	if (!IS_ALIGNED(iocb->ki_pos, len))
+		return -EINVAL;
+
+	if (!(iocb->ki_flags & IOCB_DIRECT))
+		return -EOPNOTSUPP;
+
+	return 0;
+}