1 files changed, 265 insertions, 185 deletions

diff --git a/fs/cifs/file.c b/fs/cifs/file.c
index 48b29d24c9f..e90a1e9aa62 100644
--- a/fs/cifs/file.c
+++ b/fs/cifs/file.c
@@ -183,6 +183,7 @@ cifs_nt_open(char *full_path, struct inode *inode, struct cifs_sb_info *cifs_sb,
 	int create_options = CREATE_NOT_DIR;
 	FILE_ALL_INFO *buf;
 	struct TCP_Server_Info *server = tcon->ses->server;
+	struct cifs_open_parms oparms;
 
 	if (!server->ops->open)
 		return -ENOSYS;
@@ -224,9 +225,16 @@ cifs_nt_open(char *full_path, struct inode *inode, struct cifs_sb_info *cifs_sb,
 	if (backup_cred(cifs_sb))
 		create_options |= CREATE_OPEN_BACKUP_INTENT;
 
-	rc = server->ops->open(xid, tcon, full_path, disposition,
-			       desired_access, create_options, fid, oplock, buf,
-			       cifs_sb);
+	oparms.tcon = tcon;
+	oparms.cifs_sb = cifs_sb;
+	oparms.desired_access = desired_access;
+	oparms.create_options = create_options;
+	oparms.disposition = disposition;
+	oparms.path = full_path;
+	oparms.fid = fid;
+	oparms.reconnect = false;
+
+	rc = server->ops->open(xid, &oparms, oplock, buf);
 
 	if (rc)
 		goto out;
@@ -236,7 +244,7 @@ cifs_nt_open(char *full_path, struct inode *inode, struct cifs_sb_info *cifs_sb,
 					      xid);
 	else
 		rc = cifs_get_inode_info(&inode, full_path, buf, inode->i_sb,
-					 xid, &fid->netfid);
+					 xid, fid);
 
 out:
 	kfree(buf);
@@ -305,8 +313,7 @@ cifs_new_fileinfo(struct cifs_fid *fid, struct file *file,
 	 * If the server returned a read oplock and we have mandatory brlocks,
 	 * set oplock level to None.
 	 */
-	if (oplock == server->vals->oplock_read &&
-						cifs_has_mand_locks(cinode)) {
+	if (server->ops->is_read_op(oplock) && cifs_has_mand_locks(cinode)) {
 		cifs_dbg(FYI, "Reset oplock val from read to None due to mand locks\n");
 		oplock = 0;
 	}
@@ -316,6 +323,7 @@ cifs_new_fileinfo(struct cifs_fid *fid, struct file *file,
 		oplock = fid->pending_open->oplock;
 	list_del(&fid->pending_open->olist);
 
+	fid->purge_cache = false;
 	server->ops->set_fid(cfile, fid, oplock);
 
 	list_add(&cfile->tlist, &tcon->openFileList);
@@ -326,6 +334,9 @@ cifs_new_fileinfo(struct cifs_fid *fid, struct file *file,
 		list_add_tail(&cfile->flist, &cinode->openFileList);
 	spin_unlock(&cifs_file_list_lock);
 
+	if (fid->purge_cache)
+		cifs_zap_mapping(inode);
+
 	file->private_data = cfile;
 	return cfile;
 }
@@ -381,7 +392,7 @@ void cifsFileInfo_put(struct cifsFileInfo *cifs_file)
 		 * again and get at least level II oplock.
 		 */
 		if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_STRICT_IO)
-			CIFS_I(inode)->invalid_mapping = true;
+			set_bit(CIFS_INO_INVALID_MAPPING, &cifsi->flags);
 		cifs_set_oplock_level(cifsi, 0);
 	}
 	spin_unlock(&cifs_file_list_lock);
@@ -553,11 +564,10 @@ cifs_relock_file(struct cifsFileInfo *cfile)
 	struct cifs_tcon *tcon = tlink_tcon(cfile->tlink);
 	int rc = 0;
 
-	/* we are going to update can_cache_brlcks here - need a write access */
-	down_write(&cinode->lock_sem);
+	down_read(&cinode->lock_sem);
 	if (cinode->can_cache_brlcks) {
-		/* can cache locks - no need to push them */
-		up_write(&cinode->lock_sem);
+		/* can cache locks - no need to relock */
+		up_read(&cinode->lock_sem);
 		return rc;
 	}
 
@@ -568,7 +578,7 @@ cifs_relock_file(struct cifsFileInfo *cfile)
 	else
 		rc = tcon->ses->server->ops->push_mand_locks(cfile);
 
-	up_write(&cinode->lock_sem);
+	up_read(&cinode->lock_sem);
 	return rc;
 }
 
@@ -587,7 +597,7 @@ cifs_reopen_file(struct cifsFileInfo *cfile, bool can_flush)
 	int desired_access;
 	int disposition = FILE_OPEN;
 	int create_options = CREATE_NOT_DIR;
-	struct cifs_fid fid;
+	struct cifs_open_parms oparms;
 
 	xid = get_xid();
 	mutex_lock(&cfile->fh_mutex);
@@ -637,9 +647,10 @@ cifs_reopen_file(struct cifsFileInfo *cfile, bool can_flush)
 
 		rc = cifs_posix_open(full_path, NULL, inode->i_sb,
 				     cifs_sb->mnt_file_mode /* ignored */,
-				     oflags, &oplock, &fid.netfid, xid);
+				     oflags, &oplock, &cfile->fid.netfid, xid);
 		if (rc == 0) {
 			cifs_dbg(FYI, "posix reopen succeeded\n");
+			oparms.reconnect = true;
 			goto reopen_success;
 		}
 		/*
@@ -654,18 +665,32 @@ cifs_reopen_file(struct cifsFileInfo *cfile, bool can_flush)
 		create_options |= CREATE_OPEN_BACKUP_INTENT;
 
 	if (server->ops->get_lease_key)
-		server->ops->get_lease_key(inode, &fid);
+		server->ops->get_lease_key(inode, &cfile->fid);
+
+	oparms.tcon = tcon;
+	oparms.cifs_sb = cifs_sb;
+	oparms.desired_access = desired_access;
+	oparms.create_options = create_options;
+	oparms.disposition = disposition;
+	oparms.path = full_path;
+	oparms.fid = &cfile->fid;
+	oparms.reconnect = true;
 
 	/*
 	 * Can not refresh inode by passing in file_info buf to be returned by
-	 * CIFSSMBOpen and then calling get_inode_info with returned buf since
+	 * ops->open and then calling get_inode_info with returned buf since
 	 * file might have write behind data that needs to be flushed and server
 	 * version of file size can be stale. If we knew for sure that inode was
 	 * not dirty locally we could do this.
 	 */
-	rc = server->ops->open(xid, tcon, full_path, disposition,
-			       desired_access, create_options, &fid, &oplock,
-			       NULL, cifs_sb);
+	rc = server->ops->open(xid, &oparms, &oplock, NULL);
+	if (rc == -ENOENT && oparms.reconnect == false) {
+		/* durable handle timeout is expired - open the file again */
+		rc = server->ops->open(xid, &oparms, &oplock, NULL);
+		/* indicate that we need to relock the file */
+		oparms.reconnect = true;
+	}
+
 	if (rc) {
 		mutex_unlock(&cfile->fh_mutex);
 		cifs_dbg(FYI, "cifs_reopen returned 0x%x\n", rc);
@@ -696,8 +721,9 @@ reopen_success:
 	 * to the server to get the new inode info.
 	 */
 
-	server->ops->set_fid(cfile, &fid, oplock);
-	cifs_relock_file(cfile);
+	server->ops->set_fid(cfile, &cfile->fid, oplock);
+	if (oparms.reconnect)
+		cifs_relock_file(cfile);
 
 reopen_error_exit:
 	kfree(full_path);
@@ -999,7 +1025,7 @@ try_again:
 		rc = wait_event_interruptible(flock->fl_wait, !flock->fl_next);
 		if (!rc)
 			goto try_again;
-		locks_delete_block(flock);
+		posix_unblock_lock(flock);
 	}
 	return rc;
 }
@@ -1092,6 +1118,7 @@ struct lock_to_push {
 static int
 cifs_push_posix_locks(struct cifsFileInfo *cfile)
 {
+	struct inode *inode = cfile->dentry->d_inode;
 	struct cifs_tcon *tcon = tlink_tcon(cfile->tlink);
 	struct file_lock *flock, **before;
 	unsigned int count = 0, i = 0;
@@ -1102,12 +1129,12 @@ cifs_push_posix_locks(struct cifsFileInfo *cfile)
 
 	xid = get_xid();
 
-	lock_flocks();
-	cifs_for_each_lock(cfile->dentry->d_inode, before) {
+	spin_lock(&inode->i_lock);
+	cifs_for_each_lock(inode, before) {
 		if ((*before)->fl_flags & FL_POSIX)
 			count++;
 	}
-	unlock_flocks();
+	spin_unlock(&inode->i_lock);
 
 	INIT_LIST_HEAD(&locks_to_send);
 
@@ -1126,8 +1153,8 @@ cifs_push_posix_locks(struct cifsFileInfo *cfile)
 	}
 
 	el = locks_to_send.next;
-	lock_flocks();
-	cifs_for_each_lock(cfile->dentry->d_inode, before) {
+	spin_lock(&inode->i_lock);
+	cifs_for_each_lock(inode, before) {
 		flock = *before;
 		if ((flock->fl_flags & FL_POSIX) == 0)
 			continue;
@@ -1152,7 +1179,7 @@ cifs_push_posix_locks(struct cifsFileInfo *cfile)
 		lck->offset = flock->fl_start;
 		el = el->next;
 	}
-	unlock_flocks();
+	spin_unlock(&inode->i_lock);
 
 	list_for_each_entry_safe(lck, tmp, &locks_to_send, llist) {
 		int stored_rc;
@@ -1500,12 +1527,12 @@ cifs_setlk(struct file *file, struct file_lock *flock, __u32 type,
 		 * read won't conflict with non-overlapted locks due to
 		 * pagereading.
 		 */
-		if (!CIFS_I(inode)->clientCanCacheAll &&
-					CIFS_I(inode)->clientCanCacheRead) {
-			cifs_invalidate_mapping(inode);
+		if (!CIFS_CACHE_WRITE(CIFS_I(inode)) &&
+					CIFS_CACHE_READ(CIFS_I(inode))) {
+			cifs_zap_mapping(inode);
 			cifs_dbg(FYI, "Set no oplock for inode=%p due to mand locks\n",
 				 inode);
-			CIFS_I(inode)->clientCanCacheRead = false;
+			CIFS_I(inode)->oplock = 0;
 		}
 
 		rc = server->ops->mand_lock(xid, cfile, flock->fl_start, length,
@@ -2016,7 +2043,8 @@ retry:
 			}
 			wdata->pid = wdata->cfile->pid;
 			server = tlink_tcon(wdata->cfile->tlink)->ses->server;
-			rc = server->ops->async_writev(wdata);
+			rc = server->ops->async_writev(wdata,
+							cifs_writedata_release);
 		} while (wbc->sync_mode == WB_SYNC_ALL && rc == -EAGAIN);
 
 		for (i = 0; i < nr_pages; ++i)
@@ -2189,8 +2217,8 @@ int cifs_strict_fsync(struct file *file, loff_t start, loff_t end,
 	cifs_dbg(FYI, "Sync file - name: %s datasync: 0x%x\n",
 		 file->f_path.dentry->d_name.name, datasync);
 
-	if (!CIFS_I(inode)->clientCanCacheRead) {
-		rc = cifs_invalidate_mapping(inode);
+	if (!CIFS_CACHE_READ(CIFS_I(inode))) {
+		rc = cifs_zap_mapping(inode);
 		if (rc) {
 			cifs_dbg(FYI, "rc: %d during invalidate phase\n", rc);
 			rc = 0; /* don't care about it in fsync */
@@ -2304,9 +2332,20 @@ size_t get_numpages(const size_t wsize, const size_t len, size_t *cur_len)
 }
 
 static void
-cifs_uncached_writev_complete(struct work_struct *work)
+cifs_uncached_writedata_release(struct kref *refcount)
 {
 	int i;
+	struct cifs_writedata *wdata = container_of(refcount,
+					struct cifs_writedata, refcount);
+
+	for (i = 0; i < wdata->nr_pages; i++)
+		put_page(wdata->pages[i]);
+	cifs_writedata_release(refcount);
+}
+
+static void
+cifs_uncached_writev_complete(struct work_struct *work)
+{
 	struct cifs_writedata *wdata = container_of(work,
 					struct cifs_writedata, work);
 	struct inode *inode = wdata->cfile->dentry->d_inode;
@@ -2320,12 +2359,7 @@ cifs_uncached_writev_complete(struct work_struct *work)
 
 	complete(&wdata->done);
 
-	if (wdata->result != -EAGAIN) {
-		for (i = 0; i < wdata->nr_pages; i++)
-			put_page(wdata->pages[i]);
-	}
-
-	kref_put(&wdata->refcount, cifs_writedata_release);
+	kref_put(&wdata->refcount, cifs_uncached_writedata_release);
 }
 
 /* attempt to send write to server, retry on any -EAGAIN errors */
@@ -2343,21 +2377,20 @@ cifs_uncached_retry_writev(struct cifs_writedata *wdata)
 			if (rc != 0)
 				continue;
 		}
-		rc = server->ops->async_writev(wdata);
+		rc = server->ops->async_writev(wdata,
+					       cifs_uncached_writedata_release);
 	} while (rc == -EAGAIN);
 
 	return rc;
 }
 
 static ssize_t
-cifs_iovec_write(struct file *file, const struct iovec *iov,
-		 unsigned long nr_segs, loff_t *poffset)
+cifs_iovec_write(struct file *file, struct iov_iter *from, loff_t *poffset)
 {
 	unsigned long nr_pages, i;
-	size_t copied, len, cur_len;
+	size_t bytes, copied, len, cur_len;
 	ssize_t total_written = 0;
 	loff_t offset;
-	struct iov_iter it;
 	struct cifsFileInfo *open_file;
 	struct cifs_tcon *tcon;
 	struct cifs_sb_info *cifs_sb;
@@ -2366,14 +2399,16 @@ cifs_iovec_write(struct file *file, const struct iovec *iov,
 	int rc;
 	pid_t pid;
 
-	len = iov_length(iov, nr_segs);
-	if (!len)
-		return 0;
-
+	len = iov_iter_count(from);
 	rc = generic_write_checks(file, poffset, &len, 0);
 	if (rc)
 		return rc;
 
+	if (!len)
+		return 0;
+
+	iov_iter_truncate(from, len);
+
 	INIT_LIST_HEAD(&wdata_list);
 	cifs_sb = CIFS_SB(file->f_path.dentry->d_sb);
 	open_file = file->private_data;
@@ -2389,7 +2424,6 @@ cifs_iovec_write(struct file *file, const struct iovec *iov,
 	else
 		pid = current->tgid;
 
-	iov_iter_init(&it, iov, nr_segs, len, 0);
 	do {
 		size_t save_len;
 
@@ -2409,14 +2443,44 @@ cifs_iovec_write(struct file *file, const struct iovec *iov,
 
 		save_len = cur_len;
 		for (i = 0; i < nr_pages; i++) {
-			copied = min_t(const size_t, cur_len, PAGE_SIZE);
-			copied = iov_iter_copy_from_user(wdata->pages[i], &it,
-							 0, copied);
+			bytes = min_t(size_t, cur_len, PAGE_SIZE);
+			copied = copy_page_from_iter(wdata->pages[i], 0, bytes,
+						     from);
 			cur_len -= copied;
-			iov_iter_advance(&it, copied);
+			/*
+			 * If we didn't copy as much as we expected, then that
+			 * may mean we trod into an unmapped area. Stop copying
+			 * at that point. On the next pass through the big
+			 * loop, we'll likely end up getting a zero-length
+			 * write and bailing out of it.
+			 */
+			if (copied < bytes)
+				break;
 		}
 		cur_len = save_len - cur_len;
 
+		/*
+		 * If we have no data to send, then that probably means that
+		 * the copy above failed altogether. That's most likely because
+		 * the address in the iovec was bogus. Set the rc to -EFAULT,
+		 * free anything we allocated and bail out.
+		 */
+		if (!cur_len) {
+			for (i = 0; i < nr_pages; i++)
+				put_page(wdata->pages[i]);
+			kfree(wdata);
+			rc = -EFAULT;
+			break;
+		}
+
+		/*
+		 * i + 1 now represents the number of pages we actually used in
+		 * the copy phase above. Bring nr_pages down to that, and free
+		 * any pages that we didn't use.
+		 */
+		for ( ; nr_pages > i + 1; nr_pages--)
+			put_page(wdata->pages[nr_pages - 1]);
+
 		wdata->sync_mode = WB_SYNC_ALL;
 		wdata->nr_pages = nr_pages;
 		wdata->offset = (__u64)offset;
@@ -2427,7 +2491,8 @@ cifs_iovec_write(struct file *file, const struct iovec *iov,
 		wdata->tailsz = cur_len - ((nr_pages - 1) * PAGE_SIZE);
 		rc = cifs_uncached_retry_writev(wdata);
 		if (rc) {
-			kref_put(&wdata->refcount, cifs_writedata_release);
+			kref_put(&wdata->refcount,
+				 cifs_uncached_writedata_release);
 			break;
 		}
 
@@ -2469,7 +2534,7 @@ restart_loop:
 			}
 		}
 		list_del_init(&wdata->list);
-		kref_put(&wdata->refcount, cifs_writedata_release);
+		kref_put(&wdata->refcount, cifs_uncached_writedata_release);
 	}
 
 	if (total_written > 0)
@@ -2479,11 +2544,11 @@ restart_loop:
 	return total_written ? total_written : (ssize_t)rc;
 }
 
-ssize_t cifs_user_writev(struct kiocb *iocb, const struct iovec *iov,
-				unsigned long nr_segs, loff_t pos)
+ssize_t cifs_user_writev(struct kiocb *iocb, struct iov_iter *from)
 {
 	ssize_t written;
 	struct inode *inode;
+	loff_t pos = iocb->ki_pos;
 
 	inode = file_inode(iocb->ki_filp);
 
@@ -2493,9 +2558,9 @@ ssize_t cifs_user_writev(struct kiocb *iocb, const struct iovec *iov,
 	 * write request.
 	 */
 
-	written = cifs_iovec_write(iocb->ki_filp, iov, nr_segs, &pos);
+	written = cifs_iovec_write(iocb->ki_filp, from, &pos);
 	if (written > 0) {
-		CIFS_I(inode)->invalid_mapping = true;
+		set_bit(CIFS_INO_INVALID_MAPPING, &CIFS_I(inode)->flags);
 		iocb->ki_pos = pos;
 	}
 
@@ -2503,8 +2568,7 @@ ssize_t cifs_user_writev(struct kiocb *iocb, const struct iovec *iov,
 }
 
 static ssize_t
-cifs_writev(struct kiocb *iocb, const struct iovec *iov,
-	    unsigned long nr_segs, loff_t pos)
+cifs_writev(struct kiocb *iocb, struct iov_iter *from)
 {
 	struct file *file = iocb->ki_filp;
 	struct cifsFileInfo *cfile = (struct cifsFileInfo *)file->private_data;
@@ -2512,38 +2576,38 @@ cifs_writev(struct kiocb *iocb, const struct iovec *iov,
 	struct cifsInodeInfo *cinode = CIFS_I(inode);
 	struct TCP_Server_Info *server = tlink_tcon(cfile->tlink)->ses->server;
 	ssize_t rc = -EACCES;
-
-	BUG_ON(iocb->ki_pos != pos);
+	loff_t lock_pos = iocb->ki_pos;
 
 	/*
 	 * We need to hold the sem to be sure nobody modifies lock list
 	 * with a brlock that prevents writing.
 	 */
 	down_read(&cinode->lock_sem);
-	if (!cifs_find_lock_conflict(cfile, pos, iov_length(iov, nr_segs),
+	mutex_lock(&inode->i_mutex);
+	if (file->f_flags & O_APPEND)
+		lock_pos = i_size_read(inode);
+	if (!cifs_find_lock_conflict(cfile, lock_pos, iov_iter_count(from),
 				     server->vals->exclusive_lock_type, NULL,
 				     CIFS_WRITE_OP)) {
-		mutex_lock(&inode->i_mutex);
-		rc = __generic_file_aio_write(iocb, iov, nr_segs,
-					       &iocb->ki_pos);
+		rc = __generic_file_write_iter(iocb, from);
 		mutex_unlock(&inode->i_mutex);
-	}
 
-	if (rc > 0 || rc == -EIOCBQUEUED) {
-		ssize_t err;
+		if (rc > 0) {
+			ssize_t err;
 
-		err = generic_write_sync(file, pos, rc);
-		if (err < 0 && rc > 0)
-			rc = err;
+			err = generic_write_sync(file, iocb->ki_pos - rc, rc);
+			if (err < 0)
+				rc = err;
+		}
+	} else {
+		mutex_unlock(&inode->i_mutex);
 	}
-
 	up_read(&cinode->lock_sem);
 	return rc;
 }
 
 ssize_t
-cifs_strict_writev(struct kiocb *iocb, const struct iovec *iov,
-		   unsigned long nr_segs, loff_t pos)
+cifs_strict_writev(struct kiocb *iocb, struct iov_iter *from)
 {
 	struct inode *inode = file_inode(iocb->ki_filp);
 	struct cifsInodeInfo *cinode = CIFS_I(inode);
@@ -2553,12 +2617,19 @@ cifs_strict_writev(struct kiocb *iocb, const struct iovec *iov,
 	struct cifs_tcon *tcon = tlink_tcon(cfile->tlink);
 	ssize_t written;
 
-	if (cinode->clientCanCacheAll) {
+	written = cifs_get_writer(cinode);
+	if (written)
+		return written;
+
+	if (CIFS_CACHE_WRITE(cinode)) {
 		if (cap_unix(tcon->ses) &&
 		(CIFS_UNIX_FCNTL_CAP & le64_to_cpu(tcon->fsUnixInfo.Capability))
-		    && ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NOPOSIXBRL) == 0))
-			return generic_file_aio_write(iocb, iov, nr_segs, pos);
-		return cifs_writev(iocb, iov, nr_segs, pos);
+		  && ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NOPOSIXBRL) == 0)) {
+			written = generic_file_write_iter(iocb, from);
+			goto out;
+		}
+		written = cifs_writev(iocb, from);
+		goto out;
 	}
 	/*
 	 * For non-oplocked files in strict cache mode we need to write the data
@@ -2566,18 +2637,20 @@ cifs_strict_writev(struct kiocb *iocb, const struct iovec *iov,
 	 * affected pages because it may cause a error with mandatory locks on
 	 * these pages but not on the region from pos to ppos+len-1.
 	 */
-	written = cifs_user_writev(iocb, iov, nr_segs, pos);
-	if (written > 0 && cinode->clientCanCacheRead) {
+	written = cifs_user_writev(iocb, from);
+	if (written > 0 && CIFS_CACHE_READ(cinode)) {
 		/*
 		 * Windows 7 server can delay breaking level2 oplock if a write
 		 * request comes - break it on the client to prevent reading
 		 * an old data.
 		 */
-		cifs_invalidate_mapping(inode);
+		cifs_zap_mapping(inode);
 		cifs_dbg(FYI, "Set no oplock for inode=%p after a write operation\n",
 			 inode);
-		cinode->clientCanCacheRead = false;
+		cinode->oplock = 0;
 	}
+out:
+	cifs_put_writer(cinode);
 	return written;
 }
 
@@ -2672,56 +2745,27 @@ cifs_retry_async_readv(struct cifs_readdata *rdata)
 /**
  * cifs_readdata_to_iov - copy data from pages in response to an iovec
  * @rdata:	the readdata response with list of pages holding data
- * @iov:	vector in which we should copy the data
- * @nr_segs:	number of segments in vector
- * @offset:	offset into file of the first iovec
- * @copied:	used to return the amount of data copied to the iov
+ * @iter:	destination for our data
  *
  * This function copies data from a list of pages in a readdata response into
  * an array of iovecs. It will first calculate where the data should go
  * based on the info in the readdata and then copy the data into that spot.
  */
-static ssize_t
-cifs_readdata_to_iov(struct cifs_readdata *rdata, const struct iovec *iov,
-			unsigned long nr_segs, loff_t offset, ssize_t *copied)
+static int
+cifs_readdata_to_iov(struct cifs_readdata *rdata, struct iov_iter *iter)
 {
-	int rc = 0;
-	struct iov_iter ii;
-	size_t pos = rdata->offset - offset;
-	ssize_t remaining = rdata->bytes;
-	unsigned char *pdata;
+	size_t remaining = rdata->bytes;
 	unsigned int i;
 
-	/* set up iov_iter and advance to the correct offset */
-	iov_iter_init(&ii, iov, nr_segs, iov_length(iov, nr_segs), 0);
-	iov_iter_advance(&ii, pos);
-
-	*copied = 0;
 	for (i = 0; i < rdata->nr_pages; i++) {
-		ssize_t copy;
 		struct page *page = rdata->pages[i];
-
-		/* copy a whole page or whatever's left */
-		copy = min_t(ssize_t, remaining, PAGE_SIZE);
-
-		/* ...but limit it to whatever space is left in the iov */
-		copy = min_t(ssize_t, copy, iov_iter_count(&ii));
-
-		/* go while there's data to be copied and no errors */
-		if (copy && !rc) {
-			pdata = kmap(page);
-			rc = memcpy_toiovecend(ii.iov, pdata, ii.iov_offset,
-						(int)copy);
-			kunmap(page);
-			if (!rc) {
-				*copied += copy;
-				remaining -= copy;
-				iov_iter_advance(&ii, copy);
-			}
-		}
+		size_t copy = min_t(size_t, remaining, PAGE_SIZE);
+		size_t written = copy_page_to_iter(page, 0, copy, iter);
+		remaining -= written;
+		if (written < copy && iov_iter_count(iter) > 0)
+			break;
 	}
-
-	return rc;
+	return remaining ? -EFAULT : 0;
 }
 
 static void
@@ -2782,14 +2826,13 @@ cifs_uncached_read_into_pages(struct TCP_Server_Info *server,
 	return total_read > 0 ? total_read : result;
 }
 
-static ssize_t
-cifs_iovec_read(struct file *file, const struct iovec *iov,
-		 unsigned long nr_segs, loff_t *poffset)
+ssize_t cifs_user_readv(struct kiocb *iocb, struct iov_iter *to)
 {
+	struct file *file = iocb->ki_filp;
 	ssize_t rc;
 	size_t len, cur_len;
 	ssize_t total_read = 0;
-	loff_t offset = *poffset;
+	loff_t offset = iocb->ki_pos;
 	unsigned int npages;
 	struct cifs_sb_info *cifs_sb;
 	struct cifs_tcon *tcon;
@@ -2798,10 +2841,7 @@ cifs_iovec_read(struct file *file, const struct iovec *iov,
 	struct list_head rdata_list;
 	pid_t pid;
 
-	if (!nr_segs)
-		return 0;
-
-	len = iov_length(iov, nr_segs);
+	len = iov_iter_count(to);
 	if (!len)
 		return 0;
 
@@ -2830,7 +2870,7 @@ cifs_iovec_read(struct file *file, const struct iovec *iov,
 					    cifs_uncached_readv_complete);
 		if (!rdata) {
 			rc = -ENOMEM;
-			goto error;
+			break;
 		}
 
 		rc = cifs_read_allocate_pages(rdata, npages);
@@ -2862,60 +2902,48 @@ error:
 	if (!list_empty(&rdata_list))
 		rc = 0;
 
+	len = iov_iter_count(to);
 	/* the loop below should proceed in the order of increasing offsets */
-restart_loop:
 	list_for_each_entry_safe(rdata, tmp, &rdata_list, list) {
+	again:
 		if (!rc) {
-			ssize_t copied;
-
 			/* FIXME: freezable sleep too? */
 			rc = wait_for_completion_killable(&rdata->done);
 			if (rc)
 				rc = -EINTR;
-			else if (rdata->result)
+			else if (rdata->result) {
 				rc = rdata->result;
-			else {
-				rc = cifs_readdata_to_iov(rdata, iov,
-							nr_segs, *poffset,
-							&copied);
-				total_read += copied;
+				/* resend call if it's a retryable error */
+				if (rc == -EAGAIN) {
+					rc = cifs_retry_async_readv(rdata);
+					goto again;
+				}
+			} else {
+				rc = cifs_readdata_to_iov(rdata, to);
 			}
 
-			/* resend call if it's a retryable error */
-			if (rc == -EAGAIN) {
-				rc = cifs_retry_async_readv(rdata);
-				goto restart_loop;
-			}
 		}
 		list_del_init(&rdata->list);
 		kref_put(&rdata->refcount, cifs_uncached_readdata_release);
 	}
 
+	total_read = len - iov_iter_count(to);
+
 	cifs_stats_bytes_read(tcon, total_read);
-	*poffset += total_read;
 
 	/* mask nodata case */
 	if (rc == -ENODATA)
 		rc = 0;
 
-	return total_read ? total_read : rc;
-}
-
-ssize_t cifs_user_readv(struct kiocb *iocb, const struct iovec *iov,
-			       unsigned long nr_segs, loff_t pos)
-{
-	ssize_t read;
-
-	read = cifs_iovec_read(iocb->ki_filp, iov, nr_segs, &pos);
-	if (read > 0)
-		iocb->ki_pos = pos;
-
-	return read;
+	if (total_read) {
+		iocb->ki_pos += total_read;
+		return total_read;
+	}
+	return rc;
 }
 
 ssize_t
-cifs_strict_readv(struct kiocb *iocb, const struct iovec *iov,
-		  unsigned long nr_segs, loff_t pos)
+cifs_strict_readv(struct kiocb *iocb, struct iov_iter *to)
 {
 	struct inode *inode = file_inode(iocb->ki_filp);
 	struct cifsInodeInfo *cinode = CIFS_I(inode);
@@ -2933,23 +2961,23 @@ cifs_strict_readv(struct kiocb *iocb, const struct iovec *iov,
 	 * on pages affected by this read but not on the region from pos to
 	 * pos+len-1.
 	 */
-	if (!cinode->clientCanCacheRead)
-		return cifs_user_readv(iocb, iov, nr_segs, pos);
+	if (!CIFS_CACHE_READ(cinode))
+		return cifs_user_readv(iocb, to);
 
 	if (cap_unix(tcon->ses) &&
 	    (CIFS_UNIX_FCNTL_CAP & le64_to_cpu(tcon->fsUnixInfo.Capability)) &&
 	    ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NOPOSIXBRL) == 0))
-		return generic_file_aio_read(iocb, iov, nr_segs, pos);
+		return generic_file_read_iter(iocb, to);
 
 	/*
 	 * We need to hold the sem to be sure nobody modifies lock list
 	 * with a brlock that prevents reading.
 	 */
 	down_read(&cinode->lock_sem);
-	if (!cifs_find_lock_conflict(cfile, pos, iov_length(iov, nr_segs),
+	if (!cifs_find_lock_conflict(cfile, iocb->ki_pos, iov_iter_count(to),
 				     tcon->ses->server->vals->shared_lock_type,
 				     NULL, CIFS_READ_OP))
-		rc = generic_file_aio_read(iocb, iov, nr_segs, pos);
+		rc = generic_file_read_iter(iocb, to);
 	up_read(&cinode->lock_sem);
 	return rc;
 }
@@ -3058,6 +3086,7 @@ cifs_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
 
 static struct vm_operations_struct cifs_file_vm_ops = {
 	.fault = filemap_fault,
+	.map_pages = filemap_map_pages,
 	.page_mkwrite = cifs_page_mkwrite,
 	.remap_pages = generic_file_remap_pages,
 };
@@ -3069,8 +3098,8 @@ int cifs_file_strict_mmap(struct file *file, struct vm_area_struct *vma)
 
 	xid = get_xid();
 
-	if (!CIFS_I(inode)->clientCanCacheRead) {
-		rc = cifs_invalidate_mapping(inode);
+	if (!CIFS_CACHE_READ(CIFS_I(inode))) {
+		rc = cifs_zap_mapping(inode);
 		if (rc)
 			return rc;
 	}
@@ -3227,6 +3256,9 @@ static int cifs_readpages(struct file *file, struct address_space *mapping,
 	/*
 	 * Reads as many pages as possible from fscache. Returns -ENOBUFS
 	 * immediately if the cookie is negative
+	 *
+	 * After this point, every page in the list might have PG_fscache set,
+	 * so we will need to clean that up off of every page we don't use.
 	 */
 	rc = cifs_readpages_from_fscache(mapping->host, mapping, page_list,
 					 &num_pages);
@@ -3349,9 +3381,17 @@ static int cifs_readpages(struct file *file, struct address_space *mapping,
 		kref_put(&rdata->refcount, cifs_readdata_release);
 	}
 
+	/* Any pages that have been shown to fscache but didn't get added to
+	 * the pagecache must be uncached before they get returned to the
+	 * allocator.
+	 */
+	cifs_fscache_readpages_cancel(mapping->host, page_list);
 	return rc;
 }
 
+/*
+ * cifs_readpage_worker must be called with the page pinned
+ */
 static int cifs_readpage_worker(struct file *file, struct page *page,
 	loff_t *poffset)
 {
@@ -3363,7 +3403,6 @@ static int cifs_readpage_worker(struct file *file, struct page *page,
 	if (rc == 0)
 		goto read_complete;
 
-	page_cache_get(page);
 	read_data = kmap(page);
 	/* for reads over a certain size could initiate async read ahead */
 
@@ -3390,7 +3429,7 @@ static int cifs_readpage_worker(struct file *file, struct page *page,
 
 io_error:
 	kunmap(page);
-	page_cache_release(page);
+	unlock_page(page);
 
 read_complete:
 	return rc;
@@ -3415,8 +3454,6 @@ static int cifs_readpage(struct file *file, struct page *page)
 
 	rc = cifs_readpage_worker(file, page, &offset);
 
-	unlock_page(page);
-
 	free_xid(xid);
 	return rc;
 }
@@ -3470,6 +3507,7 @@ static int cifs_write_begin(struct file *file, struct address_space *mapping,
 			loff_t pos, unsigned len, unsigned flags,
 			struct page **pagep, void **fsdata)
 {
+	int oncethru = 0;
 	pgoff_t index = pos >> PAGE_CACHE_SHIFT;
 	loff_t offset = pos & (PAGE_CACHE_SIZE - 1);
 	loff_t page_start = pos & PAGE_MASK;
@@ -3479,6 +3517,7 @@ static int cifs_write_begin(struct file *file, struct address_space *mapping,
 
 	cifs_dbg(FYI, "write_begin from %lld len %d\n", (long long)pos, len);
 
+start:
 	page = grab_cache_page_write_begin(mapping, index, flags);
 	if (!page) {
 		rc = -ENOMEM;
@@ -3502,7 +3541,7 @@ static int cifs_write_begin(struct file *file, struct address_space *mapping,
 	 * is, when the page lies beyond the EOF, or straddles the EOF
 	 * and the write will cover all of the existing data.
 	 */
-	if (CIFS_I(mapping->host)->clientCanCacheRead) {
+	if (CIFS_CACHE_READ(CIFS_I(mapping->host))) {
 		i_size = i_size_read(mapping->host);
 		if (page_start >= i_size ||
 		    (offset == 0 && (pos + len) >= i_size)) {
@@ -3520,13 +3559,16 @@ static int cifs_write_begin(struct file *file, struct address_space *mapping,
 		}
 	}
 
-	if ((file->f_flags & O_ACCMODE) != O_WRONLY) {
+	if ((file->f_flags & O_ACCMODE) != O_WRONLY && !oncethru) {
 		/*
 		 * might as well read a page, it is fast enough. If we get
 		 * an error, we don't need to return it. cifs_write_end will
 		 * do a sync write instead since PG_uptodate isn't set.
 		 */
 		cifs_readpage_worker(file, page, &page_start);
+		page_cache_release(page);
+		oncethru = 1;
+		goto start;
 	} else {
 		/* we could try using another file handle if there is one -
 		   but how would we lock it to prevent close of that handle
@@ -3546,11 +3588,12 @@ static int cifs_release_page(struct page *page, gfp_t gfp)
 	return cifs_fscache_release_page(page, gfp);
 }
 
-static void cifs_invalidate_page(struct page *page, unsigned long offset)
+static void cifs_invalidate_page(struct page *page, unsigned int offset,
+				 unsigned int length)
 {
 	struct cifsInodeInfo *cifsi = CIFS_I(page->mapping->host);
 
-	if (offset == 0)
+	if (offset == 0 && length == PAGE_CACHE_SIZE)
 		cifs_fscache_invalidate_page(page, &cifsi->vfs_inode);
 }
 
@@ -3575,6 +3618,13 @@ static int cifs_launder_page(struct page *page)
 	return rc;
 }
 
+static int
+cifs_pending_writers_wait(void *unused)
+{
+	schedule();
+	return 0;
+}
+
 void cifs_oplock_break(struct work_struct *work)
 {
 	struct cifsFileInfo *cfile = container_of(work, struct cifsFileInfo,
@@ -3582,25 +3632,32 @@ void cifs_oplock_break(struct work_struct *work)
 	struct inode *inode = cfile->dentry->d_inode;
 	struct cifsInodeInfo *cinode = CIFS_I(inode);
 	struct cifs_tcon *tcon = tlink_tcon(cfile->tlink);
+	struct TCP_Server_Info *server = tcon->ses->server;
 	int rc = 0;
 
-	if (!cinode->clientCanCacheAll && cinode->clientCanCacheRead &&
+	wait_on_bit(&cinode->flags, CIFS_INODE_PENDING_WRITERS,
+			cifs_pending_writers_wait, TASK_UNINTERRUPTIBLE);
+
+	server->ops->downgrade_oplock(server, cinode,
+		test_bit(CIFS_INODE_DOWNGRADE_OPLOCK_TO_L2, &cinode->flags));
+
+	if (!CIFS_CACHE_WRITE(cinode) && CIFS_CACHE_READ(cinode) &&
 						cifs_has_mand_locks(cinode)) {
 		cifs_dbg(FYI, "Reset oplock to None for inode=%p due to mand locks\n",
 			 inode);
-		cinode->clientCanCacheRead = false;
+		cinode->oplock = 0;
 	}
 
 	if (inode && S_ISREG(inode->i_mode)) {
-		if (cinode->clientCanCacheRead)
+		if (CIFS_CACHE_READ(cinode))
 			break_lease(inode, O_RDONLY);
 		else
 			break_lease(inode, O_WRONLY);
 		rc = filemap_fdatawrite(inode->i_mapping);
-		if (cinode->clientCanCacheRead == 0) {
+		if (!CIFS_CACHE_READ(cinode)) {
 			rc = filemap_fdatawait(inode->i_mapping);
 			mapping_set_error(inode->i_mapping, rc);
-			cifs_invalidate_mapping(inode);
+			cifs_zap_mapping(inode);
 		}
 		cifs_dbg(FYI, "Oplock flush inode %p rc %d\n", inode, rc);
 	}
@@ -3620,8 +3677,30 @@ void cifs_oplock_break(struct work_struct *work)
 							     cinode);
 		cifs_dbg(FYI, "Oplock release rc = %d\n", rc);
 	}
+	cifs_done_oplock_break(cinode);
 }
 
+/*
+ * The presence of cifs_direct_io() in the address space ops vector
+ * allowes open() O_DIRECT flags which would have failed otherwise.
+ *
+ * In the non-cached mode (mount with cache=none), we shunt off direct read and write requests
+ * so this method should never be called.
+ *
+ * Direct IO is not yet supported in the cached mode. 
+ */
+static ssize_t
+cifs_direct_io(int rw, struct kiocb *iocb, struct iov_iter *iter,
+               loff_t pos)
+{
+        /*
+         * FIXME
+         * Eventually need to support direct IO for non forcedirectio mounts
+         */
+        return -EINVAL;
+}
+
+
 const struct address_space_operations cifs_addr_ops = {
 	.readpage = cifs_readpage,
 	.readpages = cifs_readpages,
@@ -3631,6 +3710,7 @@ const struct address_space_operations cifs_addr_ops = {
 	.write_end = cifs_write_end,
 	.set_page_dirty = __set_page_dirty_nobuffers,
 	.releasepage = cifs_release_page,
+	.direct_IO = cifs_direct_io,
 	.invalidatepage = cifs_invalidate_page,
 	.launder_page = cifs_launder_page,
 };