Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs

Pull third hunk of vfs changes from Al Viro:
 "This contains the ->direct_IO() changes from Omar + saner
  generic_write_checks() + dealing with fcntl()/{read,write}() races
  (mirroring O_APPEND/O_DIRECT into iocb->ki_flags and instead of
  repeatedly looking at ->f_flags, which can be changed by fcntl(2),
  check ->ki_flags - which cannot) + infrastructure bits for dhowells'
  d_inode annotations + Christophs switch of /dev/loop to
  vfs_iter_write()"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (30 commits)
  block: loop: switch to VFS ITER_BVEC
  configfs: Fix inconsistent use of file_inode() vs file->f_path.dentry->d_inode
  VFS: Make pathwalk use d_is_reg() rather than S_ISREG()
  VFS: Fix up debugfs to use d_is_dir() in place of S_ISDIR()
  VFS: Combine inode checks with d_is_negative() and d_is_positive() in pathwalk
  NFS: Don't use d_inode as a variable name
  VFS: Impose ordering on accesses of d_inode and d_flags
  VFS: Add owner-filesystem positive/negative dentry checks
  nfs: generic_write_checks() shouldn't be done on swapout...
  ocfs2: use __generic_file_write_iter()
  mirror O_APPEND and O_DIRECT into iocb->ki_flags
  switch generic_write_checks() to iocb and iter
  ocfs2: move generic_write_checks() before the alignment checks
  ocfs2_file_write_iter: stop messing with ppos
  udf_file_write_iter: reorder and simplify
  fuse: ->direct_IO() doesn't need generic_write_checks()
  ext4_file_write_iter: move generic_write_checks() up
  xfs_file_aio_write_checks: switch to iocb/iov_iter
  generic_write_checks(): drop isblk argument
  blkdev_write_iter: expand generic_file_checks() call in there
  ...

2 files changed, 43 insertions, 103 deletions

diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c
index 8d2bc84..f906a25 100644
--- a/fs/ocfs2/aops.c
+++ b/fs/ocfs2/aops.c
@@ -855,10 +855,9 @@ static ssize_t ocfs2_direct_IO_write(struct kiocb *iocb,
 		ocfs2_inode_unlock(inode, 1);
 	}
 
-	written = __blockdev_direct_IO(WRITE, iocb, inode, inode->i_sb->s_bdev,
-			iter, offset,
-			ocfs2_direct_IO_get_blocks,
-			ocfs2_dio_end_io, NULL, 0);
+	written = __blockdev_direct_IO(iocb, inode, inode->i_sb->s_bdev, iter,
+				       offset, ocfs2_direct_IO_get_blocks,
+				       ocfs2_dio_end_io, NULL, 0);
 	if (unlikely(written < 0)) {
 		loff_t i_size = i_size_read(inode);
 
@@ -946,9 +945,7 @@ out:
 	return ret;
 }
 
-static ssize_t ocfs2_direct_IO(int rw,
-			       struct kiocb *iocb,
-			       struct iov_iter *iter,
+static ssize_t ocfs2_direct_IO(struct kiocb *iocb, struct iov_iter *iter,
 			       loff_t offset)
 {
 	struct file *file = iocb->ki_filp;
@@ -970,12 +967,11 @@ static ssize_t ocfs2_direct_IO(int rw,
 	if (i_size_read(inode) <= offset && !full_coherency)
 		return 0;
 
-	if (rw == READ)
-		return __blockdev_direct_IO(rw, iocb, inode,
-				    inode->i_sb->s_bdev,
-				    iter, offset,
-				    ocfs2_direct_IO_get_blocks,
-				    ocfs2_dio_end_io, NULL, 0);
+	if (iov_iter_rw(iter) == READ)
+		return __blockdev_direct_IO(iocb, inode, inode->i_sb->s_bdev,
+					    iter, offset,
+					    ocfs2_direct_IO_get_blocks,
+					    ocfs2_dio_end_io, NULL, 0);
 	else
 		return ocfs2_direct_IO_write(iocb, iter, offset);
 }
diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c
index 8c48e98..913fc250 100644
--- a/fs/ocfs2/file.c
+++ b/fs/ocfs2/file.c
@@ -2106,7 +2106,7 @@ out:
 }
 
 static int ocfs2_prepare_inode_for_write(struct file *file,
-					 loff_t *ppos,
+					 loff_t pos,
 					 size_t count,
 					 int appending,
 					 int *direct_io,
@@ -2115,7 +2115,7 @@ static int ocfs2_prepare_inode_for_write(struct file *file,
 	int ret = 0, meta_level = 0;
 	struct dentry *dentry = file->f_path.dentry;
 	struct inode *inode = dentry->d_inode;
-	loff_t saved_pos = 0, end;
+	loff_t end;
 	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
 	int full_coherency = !(osb->s_mount_opt &
 		OCFS2_MOUNT_COHERENCY_BUFFERED);
@@ -2155,23 +2155,16 @@ static int ocfs2_prepare_inode_for_write(struct file *file,
 			}
 		}
 
-		/* work on a copy of ppos until we're sure that we won't have
-		 * to recalculate it due to relocking. */
-		if (appending)
-			saved_pos = i_size_read(inode);
-		else
-			saved_pos = *ppos;
-
-		end = saved_pos + count;
+		end = pos + count;
 
-		ret = ocfs2_check_range_for_refcount(inode, saved_pos, count);
+		ret = ocfs2_check_range_for_refcount(inode, pos, count);
 		if (ret == 1) {
 			ocfs2_inode_unlock(inode, meta_level);
 			meta_level = -1;
 
 			ret = ocfs2_prepare_inode_for_refcount(inode,
 							       file,
-							       saved_pos,
+							       pos,
 							       count,
 							       &meta_level);
 			if (has_refcount)
@@ -2227,7 +2220,7 @@ static int ocfs2_prepare_inode_for_write(struct file *file,
 		 * caller will have to retake some cluster
 		 * locks and initiate the io as buffered.
 		 */
-		ret = ocfs2_check_range_for_holes(inode, saved_pos, count);
+		ret = ocfs2_check_range_for_holes(inode, pos, count);
 		if (ret == 1) {
 			/*
 			 * Fallback to old way if the feature bit is not set.
@@ -2242,12 +2235,9 @@ static int ocfs2_prepare_inode_for_write(struct file *file,
 		break;
 	}
 
-	if (appending)
-		*ppos = saved_pos;
-
 out_unlock:
 	trace_ocfs2_prepare_inode_for_write(OCFS2_I(inode)->ip_blkno,
-					    saved_pos, appending, count,
+					    pos, appending, count,
 					    direct_io, has_refcount);
 
 	if (meta_level >= 0)
@@ -2260,19 +2250,20 @@ out:
 static ssize_t ocfs2_file_write_iter(struct kiocb *iocb,
 				    struct iov_iter *from)
 {
-	int ret, direct_io, appending, rw_level, have_alloc_sem  = 0;
+	int direct_io, appending, rw_level, have_alloc_sem  = 0;
 	int can_do_direct, has_refcount = 0;
 	ssize_t written = 0;
-	size_t count = iov_iter_count(from);
-	loff_t old_size, *ppos = &iocb->ki_pos;
+	ssize_t ret;
+	size_t count = iov_iter_count(from), orig_count;
+	loff_t old_size;
 	u32 old_clusters;
 	struct file *file = iocb->ki_filp;
 	struct inode *inode = file_inode(file);
-	struct address_space *mapping = file->f_mapping;
 	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
 	int full_coherency = !(osb->s_mount_opt &
 			       OCFS2_MOUNT_COHERENCY_BUFFERED);
 	int unaligned_dio = 0;
+	int dropped_dio = 0;
 
 	trace_ocfs2_file_aio_write(inode, file, file->f_path.dentry,
 		(unsigned long long)OCFS2_I(inode)->ip_blkno,
@@ -2283,8 +2274,8 @@ static ssize_t ocfs2_file_write_iter(struct kiocb *iocb,
 	if (count == 0)
 		return 0;
 
-	appending = file->f_flags & O_APPEND ? 1 : 0;
-	direct_io = file->f_flags & O_DIRECT ? 1 : 0;
+	appending = iocb->ki_flags & IOCB_APPEND ? 1 : 0;
+	direct_io = iocb->ki_flags & IOCB_DIRECT ? 1 : 0;
 
 	mutex_lock(&inode->i_mutex);
 
@@ -2329,8 +2320,17 @@ relock:
 		ocfs2_inode_unlock(inode, 1);
 	}
 
+	orig_count = iov_iter_count(from);
+	ret = generic_write_checks(iocb, from);
+	if (ret <= 0) {
+		if (ret)
+			mlog_errno(ret);
+		goto out;
+	}
+	count = ret;
+
 	can_do_direct = direct_io;
-	ret = ocfs2_prepare_inode_for_write(file, ppos, count, appending,
+	ret = ocfs2_prepare_inode_for_write(file, iocb->ki_pos, count, appending,
 					    &can_do_direct, &has_refcount);
 	if (ret < 0) {
 		mlog_errno(ret);
@@ -2338,7 +2338,7 @@ relock:
 	}
 
 	if (direct_io && !is_sync_kiocb(iocb))
-		unaligned_dio = ocfs2_is_io_unaligned(inode, count, *ppos);
+		unaligned_dio = ocfs2_is_io_unaligned(inode, count, iocb->ki_pos);
 
 	/*
 	 * We can't complete the direct I/O as requested, fall back to
@@ -2351,6 +2351,9 @@ relock:
 		rw_level = -1;
 
 		direct_io = 0;
+		iocb->ki_flags &= ~IOCB_DIRECT;
+		iov_iter_reexpand(from, orig_count);
+		dropped_dio = 1;
 		goto relock;
 	}
 
@@ -2374,74 +2377,15 @@ relock:
 	/* communicate with ocfs2_dio_end_io */
 	ocfs2_iocb_set_rw_locked(iocb, rw_level);
 
-	ret = generic_write_checks(file, ppos, &count,
-				   S_ISBLK(inode->i_mode));
-	if (ret)
-		goto out_dio;
-
-	iov_iter_truncate(from, count);
-	if (direct_io) {
-		loff_t endbyte;
-		ssize_t written_buffered;
-		written = generic_file_direct_write(iocb, from, *ppos);
-		if (written < 0 || written == count) {
-			ret = written;
-			goto out_dio;
-		}
-
-		/*
-		 * for completing the rest of the request.
-		 */
-		count -= written;
-		written_buffered = generic_perform_write(file, from, *ppos);
-		/*
-		 * If generic_file_buffered_write() returned a synchronous error
-		 * then we want to return the number of bytes which were
-		 * direct-written, or the error code if that was zero. Note
-		 * that this differs from normal direct-io semantics, which
-		 * will return -EFOO even if some bytes were written.
-		 */
-		if (written_buffered < 0) {
-			ret = written_buffered;
-			goto out_dio;
-		}
-
-		/* We need to ensure that the page cache pages are written to
-		 * disk and invalidated to preserve the expected O_DIRECT
-		 * semantics.
-		 */
-		endbyte = *ppos + written_buffered - 1;
-		ret = filemap_write_and_wait_range(file->f_mapping, *ppos,
-				endbyte);
-		if (ret == 0) {
-			iocb->ki_pos = *ppos + written_buffered;
-			written += written_buffered;
-			invalidate_mapping_pages(mapping,
-					*ppos >> PAGE_CACHE_SHIFT,
-					endbyte >> PAGE_CACHE_SHIFT);
-		} else {
-			/*
-			 * We don't know how much we wrote, so just return
-			 * the number of bytes which were direct-written
-			 */
-		}
-	} else {
-		current->backing_dev_info = inode_to_bdi(inode);
-		written = generic_perform_write(file, from, *ppos);
-		if (likely(written >= 0))
-			iocb->ki_pos = *ppos + written;
-		current->backing_dev_info = NULL;
-	}
-
-out_dio:
+	written = __generic_file_write_iter(iocb, from);
 	/* buffered aio wouldn't have proper lock coverage today */
-	BUG_ON(ret == -EIOCBQUEUED && !(file->f_flags & O_DIRECT));
+	BUG_ON(written == -EIOCBQUEUED && !(iocb->ki_flags & IOCB_DIRECT));
 
 	if (unlikely(written <= 0))
 		goto no_sync;
 
-	if (((file->f_flags & O_DSYNC) && !direct_io) || IS_SYNC(inode) ||
-	    ((file->f_flags & O_DIRECT) && !direct_io)) {
+	if (((file->f_flags & O_DSYNC) && !direct_io) ||
+	    IS_SYNC(inode) || dropped_dio) {
 		ret = filemap_fdatawrite_range(file->f_mapping,
 					       iocb->ki_pos - written,
 					       iocb->ki_pos - 1);
@@ -2552,7 +2496,7 @@ static ssize_t ocfs2_file_read_iter(struct kiocb *iocb,
 	 * buffered reads protect themselves in ->readpage().  O_DIRECT reads
 	 * need locks to protect pending reads from racing with truncate.
 	 */
-	if (filp->f_flags & O_DIRECT) {
+	if (iocb->ki_flags & IOCB_DIRECT) {
 		have_alloc_sem = 1;
 		ocfs2_iocb_set_sem_locked(iocb);
 
@@ -2586,7 +2530,7 @@ static ssize_t ocfs2_file_read_iter(struct kiocb *iocb,
 	trace_generic_file_aio_read_ret(ret);
 
 	/* buffered aio wouldn't have proper lock coverage today */
-	BUG_ON(ret == -EIOCBQUEUED && !(filp->f_flags & O_DIRECT));
+	BUG_ON(ret == -EIOCBQUEUED && !(iocb->ki_flags & IOCB_DIRECT));
 
 	/* see ocfs2_file_write_iter */
 	if (ret == -EIOCBQUEUED || !ocfs2_iocb_is_rw_locked(iocb)) {