7 files changed, 85 insertions, 132 deletions

diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c
index 6aa2e3f..90b3954 100644
--- a/fs/ceph/addr.c
+++ b/fs/ceph/addr.c
@@ -691,7 +691,7 @@ static int ceph_writepages_start(struct address_space *mapping,
 	     (wbc->sync_mode == WB_SYNC_ALL ? "ALL" : "HOLD"));
 
 	if (fsc->mount_state == CEPH_MOUNT_SHUTDOWN) {
-		pr_warning("writepage_start %p on forced umount\n", inode);
+		pr_warn("writepage_start %p on forced umount\n", inode);
 		return -EIO; /* we're in a forced umount, don't write! */
 	}
 	if (fsc->mount_options->wsize && fsc->mount_options->wsize < wsize)
@@ -1184,8 +1184,8 @@ static int ceph_write_end(struct file *file, struct address_space *mapping,
  * never get called.
  */
 static ssize_t ceph_direct_io(int rw, struct kiocb *iocb,
-			      const struct iovec *iov,
-			      loff_t pos, unsigned long nr_segs)
+			      struct iov_iter *iter,
+			      loff_t pos)
 {
 	WARN_ON(1);
 	return -EINVAL;
diff --git a/fs/ceph/debugfs.c b/fs/ceph/debugfs.c
index 16b54aa..5a743ac 100644
--- a/fs/ceph/debugfs.c
+++ b/fs/ceph/debugfs.c
@@ -71,9 +71,9 @@ static int mdsc_show(struct seq_file *s, void *p)
 		seq_printf(s, "%s", ceph_mds_op_name(req->r_op));
 
 		if (req->r_got_unsafe)
-			seq_printf(s, "\t(unsafe)");
+			seq_puts(s, "\t(unsafe)");
 		else
-			seq_printf(s, "\t");
+			seq_puts(s, "\t");
 
 		if (req->r_inode) {
 			seq_printf(s, " #%llx", ceph_ino(req->r_inode));
@@ -119,7 +119,7 @@ static int mdsc_show(struct seq_file *s, void *p)
 				seq_printf(s, " %s", req->r_path2);
 		}
 
-		seq_printf(s, "\n");
+		seq_puts(s, "\n");
 	}
 	mutex_unlock(&mdsc->mutex);
 
diff --git a/fs/ceph/file.c b/fs/ceph/file.c
index 88a6df4..3020851 100644
--- a/fs/ceph/file.c
+++ b/fs/ceph/file.c
@@ -418,7 +418,7 @@ static ssize_t ceph_sync_read(struct kiocb *iocb, struct iov_iter *i,
 	struct page **pages;
 	u64 off = iocb->ki_pos;
 	int num_pages, ret;
-	size_t len = i->count;
+	size_t len = iov_iter_count(i);
 
 	dout("sync_read on file %p %llu~%u %s\n", file, off,
 	     (unsigned)len,
@@ -436,25 +436,26 @@ static ssize_t ceph_sync_read(struct kiocb *iocb, struct iov_iter *i,
 
 	if (file->f_flags & O_DIRECT) {
 		while (iov_iter_count(i)) {
-			void __user *data = i->iov[0].iov_base + i->iov_offset;
-			size_t len = i->iov[0].iov_len - i->iov_offset;
+			size_t start;
+			ssize_t n;
 
-			num_pages = calc_pages_for((unsigned long)data, len);
-			pages = ceph_get_direct_page_vector(data,
-							    num_pages, true);
-			if (IS_ERR(pages))
-				return PTR_ERR(pages);
+			n = iov_iter_get_pages_alloc(i, &pages, INT_MAX, &start);
+			if (n < 0)
+				return n;
 
-			ret = striped_read(inode, off, len,
+			num_pages = (n + start + PAGE_SIZE - 1) / PAGE_SIZE;
+
+			ret = striped_read(inode, off, n,
 					   pages, num_pages, checkeof,
-					   1, (unsigned long)data & ~PAGE_MASK);
+					   1, start);
+
 			ceph_put_page_vector(pages, num_pages, true);
 
 			if (ret <= 0)
 				break;
 			off += ret;
 			iov_iter_advance(i, ret);
-			if (ret < len)
+			if (ret < n)
 				break;
 		}
 	} else {
@@ -466,25 +467,14 @@ static ssize_t ceph_sync_read(struct kiocb *iocb, struct iov_iter *i,
 					num_pages, checkeof, 0, 0);
 		if (ret > 0) {
 			int l, k = 0;
-			size_t left = len = ret;
+			size_t left = ret;
 
 			while (left) {
-				void __user *data = i->iov[0].iov_base
-							+ i->iov_offset;
-				l = min(i->iov[0].iov_len - i->iov_offset,
-					left);
-
-				ret = ceph_copy_page_vector_to_user(&pages[k],
-								    data, off,
-								    l);
-				if (ret > 0) {
-					iov_iter_advance(i, ret);
-					left -= ret;
-					off += ret;
-					k = calc_pages_for(iocb->ki_pos,
-							   len - left + 1) - 1;
-					BUG_ON(k >= num_pages && left);
-				} else
+				int copy = min_t(size_t, PAGE_SIZE, left);
+				l = copy_page_to_iter(pages[k++], 0, copy, i);
+				off += l;
+				left -= l;
+				if (l < copy)
 					break;
 			}
 		}
@@ -541,8 +531,7 @@ static void ceph_sync_write_unsafe(struct ceph_osd_request *req, bool unsafe)
  * objects, rollback on failure, etc.)
  */
 static ssize_t
-ceph_sync_direct_write(struct kiocb *iocb, const struct iovec *iov,
-		       unsigned long nr_segs, size_t count)
+ceph_sync_direct_write(struct kiocb *iocb, struct iov_iter *from)
 {
 	struct file *file = iocb->ki_filp;
 	struct inode *inode = file_inode(file);
@@ -556,11 +545,10 @@ ceph_sync_direct_write(struct kiocb *iocb, const struct iovec *iov,
 	int written = 0;
 	int flags;
 	int check_caps = 0;
-	int page_align;
 	int ret;
 	struct timespec mtime = CURRENT_TIME;
 	loff_t pos = iocb->ki_pos;
-	struct iov_iter i;
+	size_t count = iov_iter_count(from);
 
 	if (ceph_snap(file_inode(file)) != CEPH_NOSNAP)
 		return -EROFS;
@@ -582,13 +570,10 @@ ceph_sync_direct_write(struct kiocb *iocb, const struct iovec *iov,
 		CEPH_OSD_FLAG_ONDISK |
 		CEPH_OSD_FLAG_WRITE;
 
-	iov_iter_init(&i, iov, nr_segs, count, 0);
-
-	while (iov_iter_count(&i) > 0) {
-		void __user *data = i.iov->iov_base + i.iov_offset;
-		u64 len = i.iov->iov_len - i.iov_offset;
-
-		page_align = (unsigned long)data & ~PAGE_MASK;
+	while (iov_iter_count(from) > 0) {
+		u64 len = iov_iter_single_seg_count(from);
+		size_t start;
+		ssize_t n;
 
 		snapc = ci->i_snap_realm->cached_context;
 		vino = ceph_vino(inode);
@@ -604,20 +589,21 @@ ceph_sync_direct_write(struct kiocb *iocb, const struct iovec *iov,
 			break;
 		}
 
-		num_pages = calc_pages_for(page_align, len);
-		pages = ceph_get_direct_page_vector(data, num_pages, false);
-		if (IS_ERR(pages)) {
-			ret = PTR_ERR(pages);
-			goto out;
+		n = iov_iter_get_pages_alloc(from, &pages, len, &start);
+		if (unlikely(n < 0)) {
+			ret = n;
+			ceph_osdc_put_request(req);
+			break;
 		}
 
+		num_pages = (n + start + PAGE_SIZE - 1) / PAGE_SIZE;
 		/*
 		 * throw out any page cache pages in this range. this
 		 * may block.
 		 */
 		truncate_inode_pages_range(inode->i_mapping, pos,
-				   (pos+len) | (PAGE_CACHE_SIZE-1));
-		osd_req_op_extent_osd_data_pages(req, 0, pages, len, page_align,
+				   (pos+n) | (PAGE_CACHE_SIZE-1));
+		osd_req_op_extent_osd_data_pages(req, 0, pages, n, start,
 						false, false);
 
 		/* BUG_ON(vino.snap != CEPH_NOSNAP); */
@@ -629,22 +615,20 @@ ceph_sync_direct_write(struct kiocb *iocb, const struct iovec *iov,
 
 		ceph_put_page_vector(pages, num_pages, false);
 
-out:
 		ceph_osdc_put_request(req);
-		if (ret == 0) {
-			pos += len;
-			written += len;
-			iov_iter_advance(&i, (size_t)len);
-
-			if (pos > i_size_read(inode)) {
-				check_caps = ceph_inode_set_size(inode, pos);
-				if (check_caps)
-					ceph_check_caps(ceph_inode(inode),
-							CHECK_CAPS_AUTHONLY,
-							NULL);
-			}
-		} else
+		if (ret)
 			break;
+		pos += n;
+		written += n;
+		iov_iter_advance(from, n);
+
+		if (pos > i_size_read(inode)) {
+			check_caps = ceph_inode_set_size(inode, pos);
+			if (check_caps)
+				ceph_check_caps(ceph_inode(inode),
+						CHECK_CAPS_AUTHONLY,
+						NULL);
+		}
 	}
 
 	if (ret != -EOLDSNAPC && written > 0) {
@@ -662,8 +646,7 @@ out:
  * correct atomic write, we should e.g. take write locks on all
  * objects, rollback on failure, etc.)
  */
-static ssize_t ceph_sync_write(struct kiocb *iocb, const struct iovec *iov,
-			       unsigned long nr_segs, size_t count)
+static ssize_t ceph_sync_write(struct kiocb *iocb, struct iov_iter *from)
 {
 	struct file *file = iocb->ki_filp;
 	struct inode *inode = file_inode(file);
@@ -681,7 +664,7 @@ static ssize_t ceph_sync_write(struct kiocb *iocb, const struct iovec *iov,
 	int ret;
 	struct timespec mtime = CURRENT_TIME;
 	loff_t pos = iocb->ki_pos;
-	struct iov_iter i;
+	size_t count = iov_iter_count(from);
 
 	if (ceph_snap(file_inode(file)) != CEPH_NOSNAP)
 		return -EROFS;
@@ -703,9 +686,7 @@ static ssize_t ceph_sync_write(struct kiocb *iocb, const struct iovec *iov,
 		CEPH_OSD_FLAG_WRITE |
 		CEPH_OSD_FLAG_ACK;
 
-	iov_iter_init(&i, iov, nr_segs, count, 0);
-
-	while ((len = iov_iter_count(&i)) > 0) {
+	while ((len = iov_iter_count(from)) > 0) {
 		size_t left;
 		int n;
 
@@ -737,13 +718,12 @@ static ssize_t ceph_sync_write(struct kiocb *iocb, const struct iovec *iov,
 		left = len;
 		for (n = 0; n < num_pages; n++) {
 			size_t plen = min_t(size_t, left, PAGE_SIZE);
-			ret = iov_iter_copy_from_user(pages[n], &i, 0, plen);
+			ret = copy_page_from_iter(pages[n], 0, plen, from);
 			if (ret != plen) {
 				ret = -EFAULT;
 				break;
 			}
 			left -= ret;
-			iov_iter_advance(&i, ret);
 		}
 
 		if (ret < 0) {
@@ -796,8 +776,7 @@ out:
  *
  * Hmm, the sync read case isn't actually async... should it be?
  */
-static ssize_t ceph_aio_read(struct kiocb *iocb, const struct iovec *iov,
-			     unsigned long nr_segs, loff_t pos)
+static ssize_t ceph_read_iter(struct kiocb *iocb, struct iov_iter *to)
 {
 	struct file *filp = iocb->ki_filp;
 	struct ceph_file_info *fi = filp->private_data;
@@ -823,40 +802,20 @@ again:
 	if ((got & (CEPH_CAP_FILE_CACHE|CEPH_CAP_FILE_LAZYIO)) == 0 ||
 	    (iocb->ki_filp->f_flags & O_DIRECT) ||
 	    (fi->flags & CEPH_F_SYNC)) {
-		struct iov_iter i;
 
 		dout("aio_sync_read %p %llx.%llx %llu~%u got cap refs on %s\n",
 		     inode, ceph_vinop(inode), iocb->ki_pos, (unsigned)len,
 		     ceph_cap_string(got));
 
-		if (!read) {
-			ret = generic_segment_checks(iov, &nr_segs,
-							&len, VERIFY_WRITE);
-			if (ret)
-				goto out;
-		}
-
-		iov_iter_init(&i, iov, nr_segs, len, read);
-
 		/* hmm, this isn't really async... */
-		ret = ceph_sync_read(iocb, &i, &checkeof);
+		ret = ceph_sync_read(iocb, to, &checkeof);
 	} else {
-		/*
-		 * We can't modify the content of iov,
-		 * so we only read from beginning.
-		 */
-		if (read) {
-			iocb->ki_pos = pos;
-			len = iocb->ki_nbytes;
-			read = 0;
-		}
 		dout("aio_read %p %llx.%llx %llu~%u got cap refs on %s\n",
-		     inode, ceph_vinop(inode), pos, (unsigned)len,
+		     inode, ceph_vinop(inode), iocb->ki_pos, (unsigned)len,
 		     ceph_cap_string(got));
 
-		ret = generic_file_aio_read(iocb, iov, nr_segs, pos);
+		ret = generic_file_read_iter(iocb, to);
 	}
-out:
 	dout("aio_read %p %llx.%llx dropping cap refs on %s = %d\n",
 	     inode, ceph_vinop(inode), ceph_cap_string(got), (int)ret);
 	ceph_put_cap_refs(ci, got);
@@ -872,6 +831,7 @@ out:
 			     ", reading more\n", iocb->ki_pos,
 			     inode->i_size);
 
+			iov_iter_advance(to, ret);
 			read += ret;
 			len -= ret;
 			checkeof = 0;
@@ -895,8 +855,7 @@ out:
  *
  * If we are near ENOSPC, write synchronously.
  */
-static ssize_t ceph_aio_write(struct kiocb *iocb, const struct iovec *iov,
-		       unsigned long nr_segs, loff_t pos)
+static ssize_t ceph_write_iter(struct kiocb *iocb, struct iov_iter *from)
 {
 	struct file *file = iocb->ki_filp;
 	struct ceph_file_info *fi = file->private_data;
@@ -904,18 +863,15 @@ static ssize_t ceph_aio_write(struct kiocb *iocb, const struct iovec *iov,
 	struct ceph_inode_info *ci = ceph_inode(inode);
 	struct ceph_osd_client *osdc =
 		&ceph_sb_to_client(inode->i_sb)->client->osdc;
-	ssize_t count, written = 0;
+	ssize_t count = iov_iter_count(from), written = 0;
 	int err, want, got;
+	loff_t pos = iocb->ki_pos;
 
 	if (ceph_snap(inode) != CEPH_NOSNAP)
 		return -EROFS;
 
 	mutex_lock(&inode->i_mutex);
 
-	err = generic_segment_checks(iov, &nr_segs, &count, VERIFY_READ);
-	if (err)
-		goto out;
-
 	/* We can write back this queue in page reclaim */
 	current->backing_dev_info = file->f_mapping->backing_dev_info;
 
@@ -925,6 +881,7 @@ static ssize_t ceph_aio_write(struct kiocb *iocb, const struct iovec *iov,
 
 	if (count == 0)
 		goto out;
+	iov_iter_truncate(from, count);
 
 	err = file_remove_suid(file);
 	if (err)
@@ -956,23 +913,26 @@ retry_snap:
 
 	if ((got & (CEPH_CAP_FILE_BUFFER|CEPH_CAP_FILE_LAZYIO)) == 0 ||
 	    (file->f_flags & O_DIRECT) || (fi->flags & CEPH_F_SYNC)) {
+		struct iov_iter data;
 		mutex_unlock(&inode->i_mutex);
+		/* we might need to revert back to that point */
+		data = *from;
 		if (file->f_flags & O_DIRECT)
-			written = ceph_sync_direct_write(iocb, iov,
-							 nr_segs, count);
+			written = ceph_sync_direct_write(iocb, &data);
 		else
-			written = ceph_sync_write(iocb, iov, nr_segs, count);
+			written = ceph_sync_write(iocb, &data);
 		if (written == -EOLDSNAPC) {
 			dout("aio_write %p %llx.%llx %llu~%u"
 				"got EOLDSNAPC, retrying\n",
 				inode, ceph_vinop(inode),
-				pos, (unsigned)iov->iov_len);
+				pos, (unsigned)count);
 			mutex_lock(&inode->i_mutex);
 			goto retry_snap;
 		}
+		if (written > 0)
+			iov_iter_advance(from, written);
 	} else {
 		loff_t old_size = inode->i_size;
-		struct iov_iter from;
 		/*
 		 * No need to acquire the i_truncate_mutex. Because
 		 * the MDS revokes Fwb caps before sending truncate
@@ -980,8 +940,7 @@ retry_snap:
 		 * are pending vmtruncate. So write and vmtruncate
 		 * can not run at the same time
 		 */
-		iov_iter_init(&from, iov, nr_segs, count, 0);
-		written = generic_perform_write(file, &from, pos);
+		written = generic_perform_write(file, from, pos);
 		if (likely(written >= 0))
 			iocb->ki_pos = pos + written;
 		if (inode->i_size > old_size)
@@ -999,7 +958,7 @@ retry_snap:
 	}
 
 	dout("aio_write %p %llx.%llx %llu~%u  dropping cap refs on %s\n",
-	     inode, ceph_vinop(inode), pos, (unsigned)iov->iov_len,
+	     inode, ceph_vinop(inode), pos, (unsigned)count,
 	     ceph_cap_string(got));
 	ceph_put_cap_refs(ci, got);
 
@@ -1276,16 +1235,16 @@ const struct file_operations ceph_file_fops = {
 	.open = ceph_open,
 	.release = ceph_release,
 	.llseek = ceph_llseek,
-	.read = do_sync_read,
-	.write = do_sync_write,
-	.aio_read = ceph_aio_read,
-	.aio_write = ceph_aio_write,
+	.read = new_sync_read,
+	.write = new_sync_write,
+	.read_iter = ceph_read_iter,
+	.write_iter = ceph_write_iter,
 	.mmap = ceph_mmap,
 	.fsync = ceph_fsync,
 	.lock = ceph_lock,
 	.flock = ceph_flock,
 	.splice_read = generic_file_splice_read,
-	.splice_write = generic_file_splice_write,
+	.splice_write = iter_file_splice_write,
 	.unlocked_ioctl = ceph_ioctl,
 	.compat_ioctl	= ceph_ioctl,
 	.fallocate	= ceph_fallocate,
diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c
index 0901f38..04c89c2 100644
--- a/fs/ceph/inode.c
+++ b/fs/ceph/inode.c
@@ -869,7 +869,7 @@ static int fill_inode(struct inode *inode,
 				__ceph_get_fmode(ci, cap_fmode);
 		}
 	} else if (cap_fmode >= 0) {
-		pr_warning("mds issued no caps on %llx.%llx\n",
+		pr_warn("mds issued no caps on %llx.%llx\n",
 			   ceph_vinop(inode));
 		__ceph_get_fmode(ci, cap_fmode);
 	}
diff --git a/fs/ceph/locks.c b/fs/ceph/locks.c
index 1913988..fbc39c4 100644
--- a/fs/ceph/locks.c
+++ b/fs/ceph/locks.c
@@ -53,10 +53,7 @@ static int ceph_lock_message(u8 lock_type, u16 operation, struct file *file,
 	else
 		length = fl->fl_end - fl->fl_start + 1;
 
-	if (lock_type == CEPH_LOCK_FCNTL)
-		owner = secure_addr(fl->fl_owner);
-	else
-		owner = secure_addr(fl->fl_file);
+	owner = secure_addr(fl->fl_owner);
 
 	dout("ceph_lock_message: rule: %d, op: %d, owner: %llx, pid: %llu, "
 	     "start: %llu, length: %llu, wait: %d, type: %d", (int)lock_type,
@@ -314,10 +311,7 @@ int lock_to_ceph_filelock(struct file_lock *lock,
 	cephlock->length = cpu_to_le64(lock->fl_end - lock->fl_start + 1);
 	cephlock->client = cpu_to_le64(0);
 	cephlock->pid = cpu_to_le64((u64)lock->fl_pid);
-	if (lock->fl_flags & FL_POSIX)
-		cephlock->owner = cpu_to_le64(secure_addr(lock->fl_owner));
-	else
-		cephlock->owner = cpu_to_le64(secure_addr(lock->fl_file));
+	cephlock->owner = cpu_to_le64(secure_addr(lock->fl_owner));
 
 	switch (lock->fl_type) {
 	case F_RDLCK:
diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c
index 7b19510..92a2548 100644
--- a/fs/ceph/mds_client.c
+++ b/fs/ceph/mds_client.c
@@ -2225,13 +2225,13 @@ static void handle_reply(struct ceph_mds_session *session, struct ceph_msg *msg)
 	/* dup? */
 	if ((req->r_got_unsafe && !head->safe) ||
 	    (req->r_got_safe && head->safe)) {
-		pr_warning("got a dup %s reply on %llu from mds%d\n",
+		pr_warn("got a dup %s reply on %llu from mds%d\n",
 			   head->safe ? "safe" : "unsafe", tid, mds);
 		mutex_unlock(&mdsc->mutex);
 		goto out;
 	}
 	if (req->r_got_safe && !head->safe) {
-		pr_warning("got unsafe after safe on %llu from mds%d\n",
+		pr_warn("got unsafe after safe on %llu from mds%d\n",
 			   tid, mds);
 		mutex_unlock(&mdsc->mutex);
 		goto out;
@@ -3532,7 +3532,7 @@ static void peer_reset(struct ceph_connection *con)
 	struct ceph_mds_session *s = con->private;
 	struct ceph_mds_client *mdsc = s->s_mdsc;
 
-	pr_warning("mds%d closed our session\n", s->s_mds);
+	pr_warn("mds%d closed our session\n", s->s_mds);
 	send_mds_reconnect(mdsc, s);
 }
 
diff --git a/fs/ceph/mdsmap.c b/fs/ceph/mdsmap.c
index 132b64e..261531e 100644
--- a/fs/ceph/mdsmap.c
+++ b/fs/ceph/mdsmap.c
@@ -62,7 +62,7 @@ struct ceph_mdsmap *ceph_mdsmap_decode(void **p, void *end)
 
 	ceph_decode_16_safe(p, end, version, bad);
 	if (version > 3) {
-		pr_warning("got mdsmap version %d > 3, failing", version);
+		pr_warn("got mdsmap version %d > 3, failing", version);
 		goto bad;
 	}