vfs,mm: fix return value of read() at s_maxbytes

commit d05c5f7b upstream.

We truncated the possible read iterator to s_maxbytes in commit
c2a9737f ("vfs,mm: fix a dead loop in truncate_inode_pages_range()"),
but our end condition handling was wrong: it's not an error to try to
read at the end of the file.

Reading past the end should return EOF (0), not EINVAL.

See for example

  https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1649342
  http://lists.gnu.org/archive/html/bug-coreutils/2016-12/msg00008.html

where a md5sum of a maximally sized file fails because the final read is
exactly at s_maxbytes.

Fixes: c2a9737f ("vfs,mm: fix a dead loop in truncate_inode_pages_range()")
Reported-by: Joseph Salisbury <joseph.salisbury@canonical.com>
Cc: Wei Fang <fangwei1@huawei.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>

mm/filemap.c

@@ -1465,7 +1465,7 @@ static ssize_t do_generic_file_read(struct file *filp, loff_t *ppos,
 	int error = 0;
 
 	if (unlikely(*ppos >= inode->i_sb->s_maxbytes))
-		return -EINVAL;
+		return 0;
 	iov_iter_truncate(iter, inode->i_sb->s_maxbytes);
 
 	index = *ppos >> PAGE_CACHE_SHIFT;