nfs: disintegrate UAPI for nfs

This is to complete part of the Userspace API (UAPI) disintegration for which
the preparatory patches were pulled recently.  After these patches, userspace
headers will be segregated into:

        include/uapi/linux/.../foo.h

for the userspace interface stuff, and:

        include/linux/.../foo.h

for the strictly kernel internal stuff.

Signed-off-by: J. Bruce Fields <bfields@redhat.com>

7 files changed, 110 insertions, 40 deletions

diff --git a/net/ceph/ceph_common.c b/net/ceph/ceph_common.c
index 69e38db..a802029 100644
--- a/net/ceph/ceph_common.c
+++ b/net/ceph/ceph_common.c
@@ -84,7 +84,6 @@ int ceph_check_fsid(struct ceph_client *client, struct ceph_fsid *fsid)
 			return -1;
 		}
 	} else {
-		pr_info("client%lld fsid %pU\n", ceph_client_id(client), fsid);
 		memcpy(&client->fsid, fsid, sizeof(*fsid));
 	}
 	return 0;
diff --git a/net/ceph/debugfs.c b/net/ceph/debugfs.c
index 54b531a..38b5dc1 100644
--- a/net/ceph/debugfs.c
+++ b/net/ceph/debugfs.c
@@ -189,6 +189,9 @@ int ceph_debugfs_client_init(struct ceph_client *client)
 	snprintf(name, sizeof(name), "%pU.client%lld", &client->fsid,
 		 client->monc.auth->global_id);
 
+	dout("ceph_debugfs_client_init %p %s\n", client, name);
+
+	BUG_ON(client->debugfs_dir);
 	client->debugfs_dir = debugfs_create_dir(name, ceph_debugfs_dir);
 	if (!client->debugfs_dir)
 		goto out;
@@ -234,6 +237,7 @@ out:
 
 void ceph_debugfs_client_cleanup(struct ceph_client *client)
 {
+	dout("ceph_debugfs_client_cleanup %p\n", client);
 	debugfs_remove(client->debugfs_osdmap);
 	debugfs_remove(client->debugfs_monmap);
 	debugfs_remove(client->osdc.debugfs_file);
diff --git a/net/ceph/messenger.c b/net/ceph/messenger.c
index b979675..159aa8b 100644
--- a/net/ceph/messenger.c
+++ b/net/ceph/messenger.c
@@ -915,7 +915,6 @@ static int prepare_write_connect(struct ceph_connection *con)
 	con->out_connect.authorizer_len = auth ?
 		cpu_to_le32(auth->authorizer_buf_len) : 0;
 
-	con_out_kvec_reset(con);
 	con_out_kvec_add(con, sizeof (con->out_connect),
 					&con->out_connect);
 	if (auth && auth->authorizer_buf_len)
@@ -1074,16 +1073,13 @@ static int write_partial_msg_pages(struct ceph_connection *con)
 			BUG_ON(kaddr == NULL);
 			base = kaddr + con->out_msg_pos.page_pos + bio_offset;
 			crc = crc32c(crc, base, len);
+			kunmap(page);
 			msg->footer.data_crc = cpu_to_le32(crc);
 			con->out_msg_pos.did_page_crc = true;
 		}
 		ret = ceph_tcp_sendpage(con->sock, page,
 				      con->out_msg_pos.page_pos + bio_offset,
 				      len, 1);
-
-		if (do_datacrc)
-			kunmap(page);
-
 		if (ret <= 0)
 			goto out;
 
@@ -1557,6 +1553,7 @@ static int process_connect(struct ceph_connection *con)
 			return -1;
 		}
 		con->auth_retry = 1;
+		con_out_kvec_reset(con);
 		ret = prepare_write_connect(con);
 		if (ret < 0)
 			return ret;
@@ -1577,6 +1574,7 @@ static int process_connect(struct ceph_connection *con)
 		       ENTITY_NAME(con->peer_name),
 		       ceph_pr_addr(&con->peer_addr.in_addr));
 		reset_connection(con);
+		con_out_kvec_reset(con);
 		ret = prepare_write_connect(con);
 		if (ret < 0)
 			return ret;
@@ -1601,6 +1599,7 @@ static int process_connect(struct ceph_connection *con)
 		     le32_to_cpu(con->out_connect.connect_seq),
 		     le32_to_cpu(con->in_reply.connect_seq));
 		con->connect_seq = le32_to_cpu(con->in_reply.connect_seq);
+		con_out_kvec_reset(con);
 		ret = prepare_write_connect(con);
 		if (ret < 0)
 			return ret;
@@ -1617,6 +1616,7 @@ static int process_connect(struct ceph_connection *con)
 		     le32_to_cpu(con->in_reply.global_seq));
 		get_global_seq(con->msgr,
 			       le32_to_cpu(con->in_reply.global_seq));
+		con_out_kvec_reset(con);
 		ret = prepare_write_connect(con);
 		if (ret < 0)
 			return ret;
@@ -2135,7 +2135,11 @@ more:
 		BUG_ON(con->state != CON_STATE_CONNECTING);
 		con->state = CON_STATE_NEGOTIATING;
 
-		/* Banner is good, exchange connection info */
+		/*
+		 * Received banner is good, exchange connection info.
+		 * Do not reset out_kvec, as sending our banner raced
+		 * with receiving peer banner after connect completed.
+		 */
 		ret = prepare_write_connect(con);
 		if (ret < 0)
 			goto out;
diff --git a/net/ceph/mon_client.c b/net/ceph/mon_client.c
index 105d533..812eb3b 100644
--- a/net/ceph/mon_client.c
+++ b/net/ceph/mon_client.c
@@ -311,6 +311,17 @@ int ceph_monc_open_session(struct ceph_mon_client *monc)
 EXPORT_SYMBOL(ceph_monc_open_session);
 
 /*
+ * We require the fsid and global_id in order to initialize our
+ * debugfs dir.
+ */
+static bool have_debugfs_info(struct ceph_mon_client *monc)
+{
+	dout("have_debugfs_info fsid %d globalid %lld\n",
+	     (int)monc->client->have_fsid, monc->auth->global_id);
+	return monc->client->have_fsid && monc->auth->global_id > 0;
+}
+
+/*
  * The monitor responds with mount ack indicate mount success.  The
  * included client ticket allows the client to talk to MDSs and OSDs.
  */
@@ -320,9 +331,12 @@ static void ceph_monc_handle_map(struct ceph_mon_client *monc,
 	struct ceph_client *client = monc->client;
 	struct ceph_monmap *monmap = NULL, *old = monc->monmap;
 	void *p, *end;
+	int had_debugfs_info, init_debugfs = 0;
 
 	mutex_lock(&monc->mutex);
 
+	had_debugfs_info = have_debugfs_info(monc);
+
 	dout("handle_monmap\n");
 	p = msg->front.iov_base;
 	end = p + msg->front.iov_len;
@@ -344,12 +358,22 @@ static void ceph_monc_handle_map(struct ceph_mon_client *monc,
 
 	if (!client->have_fsid) {
 		client->have_fsid = true;
+		if (!had_debugfs_info && have_debugfs_info(monc)) {
+			pr_info("client%lld fsid %pU\n",
+				ceph_client_id(monc->client),
+				&monc->client->fsid);
+			init_debugfs = 1;
+		}
 		mutex_unlock(&monc->mutex);
-		/*
-		 * do debugfs initialization without mutex to avoid
-		 * creating a locking dependency
-		 */
-		ceph_debugfs_client_init(client);
+
+		if (init_debugfs) {
+			/*
+			 * do debugfs initialization without mutex to avoid
+			 * creating a locking dependency
+			 */
+			ceph_debugfs_client_init(monc->client);
+		}
+
 		goto out_unlocked;
 	}
 out:
@@ -613,7 +637,7 @@ bad:
 /*
  * Do a synchronous pool op.
  */
-int ceph_monc_do_poolop(struct ceph_mon_client *monc, u32 op,
+static int do_poolop(struct ceph_mon_client *monc, u32 op,
 			u32 pool, u64 snapid,
 			char *buf, int len)
 {
@@ -663,7 +687,7 @@ out:
 int ceph_monc_create_snapid(struct ceph_mon_client *monc,
 			    u32 pool, u64 *snapid)
 {
-	return ceph_monc_do_poolop(monc,  POOL_OP_CREATE_UNMANAGED_SNAP,
+	return do_poolop(monc,  POOL_OP_CREATE_UNMANAGED_SNAP,
 				   pool, 0, (char *)snapid, sizeof(*snapid));
 
 }
@@ -672,7 +696,7 @@ EXPORT_SYMBOL(ceph_monc_create_snapid);
 int ceph_monc_delete_snapid(struct ceph_mon_client *monc,
 			    u32 pool, u64 snapid)
 {
-	return ceph_monc_do_poolop(monc,  POOL_OP_CREATE_UNMANAGED_SNAP,
+	return do_poolop(monc,  POOL_OP_CREATE_UNMANAGED_SNAP,
 				   pool, snapid, 0, 0);
 
 }
@@ -745,7 +769,6 @@ static int build_initial_monmap(struct ceph_mon_client *monc)
 		monc->monmap->mon_inst[i].name.num = cpu_to_le64(i);
 	}
 	monc->monmap->num_mon = num_mon;
-	monc->have_fsid = false;
 	return 0;
 }
 
@@ -865,8 +888,10 @@ static void handle_auth_reply(struct ceph_mon_client *monc,
 {
 	int ret;
 	int was_auth = 0;
+	int had_debugfs_info, init_debugfs = 0;
 
 	mutex_lock(&monc->mutex);
+	had_debugfs_info = have_debugfs_info(monc);
 	if (monc->auth->ops)
 		was_auth = monc->auth->ops->is_authenticated(monc->auth);
 	monc->pending_auth = 0;
@@ -889,7 +914,22 @@ static void handle_auth_reply(struct ceph_mon_client *monc,
 		__send_subscribe(monc);
 		__resend_generic_request(monc);
 	}
+
+	if (!had_debugfs_info && have_debugfs_info(monc)) {
+		pr_info("client%lld fsid %pU\n",
+			ceph_client_id(monc->client),
+			&monc->client->fsid);
+		init_debugfs = 1;
+	}
 	mutex_unlock(&monc->mutex);
+
+	if (init_debugfs) {
+		/*
+		 * do debugfs initialization without mutex to avoid
+		 * creating a locking dependency
+		 */
+		ceph_debugfs_client_init(monc->client);
+	}
 }
 
 static int __validate_auth(struct ceph_mon_client *monc)
diff --git a/net/ceph/osd_client.c b/net/ceph/osd_client.c
index 42119c0..c1d756c 100644
--- a/net/ceph/osd_client.c
+++ b/net/ceph/osd_client.c
@@ -52,7 +52,7 @@ static int op_has_extent(int op)
 		op == CEPH_OSD_OP_WRITE);
 }
 
-void ceph_calc_raw_layout(struct ceph_osd_client *osdc,
+int ceph_calc_raw_layout(struct ceph_osd_client *osdc,
 			struct ceph_file_layout *layout,
 			u64 snapid,
 			u64 off, u64 *plen, u64 *bno,
@@ -62,12 +62,15 @@ void ceph_calc_raw_layout(struct ceph_osd_client *osdc,
 	struct ceph_osd_request_head *reqhead = req->r_request->front.iov_base;
 	u64 orig_len = *plen;
 	u64 objoff, objlen;    /* extent in object */
+	int r;
 
 	reqhead->snapid = cpu_to_le64(snapid);
 
 	/* object extent? */
-	ceph_calc_file_object_mapping(layout, off, plen, bno,
-				      &objoff, &objlen);
+	r = ceph_calc_file_object_mapping(layout, off, plen, bno,
+					  &objoff, &objlen);
+	if (r < 0)
+		return r;
 	if (*plen < orig_len)
 		dout(" skipping last %llu, final file extent %llu~%llu\n",
 		     orig_len - *plen, off, *plen);
@@ -83,7 +86,7 @@ void ceph_calc_raw_layout(struct ceph_osd_client *osdc,
 
 	dout("calc_layout bno=%llx %llu~%llu (%d pages)\n",
 	     *bno, objoff, objlen, req->r_num_pages);
-
+	return 0;
 }
 EXPORT_SYMBOL(ceph_calc_raw_layout);
 
@@ -112,20 +115,25 @@ EXPORT_SYMBOL(ceph_calc_raw_layout);
  *
  * fill osd op in request message.
  */
-static void calc_layout(struct ceph_osd_client *osdc,
-			struct ceph_vino vino,
-			struct ceph_file_layout *layout,
-			u64 off, u64 *plen,
-			struct ceph_osd_request *req,
-			struct ceph_osd_req_op *op)
+static int calc_layout(struct ceph_osd_client *osdc,
+		       struct ceph_vino vino,
+		       struct ceph_file_layout *layout,
+		       u64 off, u64 *plen,
+		       struct ceph_osd_request *req,
+		       struct ceph_osd_req_op *op)
 {
 	u64 bno;
+	int r;
 
-	ceph_calc_raw_layout(osdc, layout, vino.snap, off,
-			     plen, &bno, req, op);
+	r = ceph_calc_raw_layout(osdc, layout, vino.snap, off,
+				 plen, &bno, req, op);
+	if (r < 0)
+		return r;
 
 	snprintf(req->r_oid, sizeof(req->r_oid), "%llx.%08llx", vino.ino, bno);
 	req->r_oid_len = strlen(req->r_oid);
+
+	return r;
 }
 
 /*
@@ -213,7 +221,6 @@ struct ceph_osd_request *ceph_osdc_alloc_request(struct ceph_osd_client *osdc,
 	kref_init(&req->r_kref);
 	init_completion(&req->r_completion);
 	init_completion(&req->r_safe_completion);
-	rb_init_node(&req->r_node);
 	INIT_LIST_HEAD(&req->r_unsafe_item);
 	INIT_LIST_HEAD(&req->r_linger_item);
 	INIT_LIST_HEAD(&req->r_linger_osd);
@@ -456,6 +463,7 @@ struct ceph_osd_request *ceph_osdc_new_request(struct ceph_osd_client *osdc,
 {
 	struct ceph_osd_req_op ops[3];
 	struct ceph_osd_request *req;
+	int r;
 
 	ops[0].op = opcode;
 	ops[0].extent.truncate_seq = truncate_seq;
@@ -474,10 +482,12 @@ struct ceph_osd_request *ceph_osdc_new_request(struct ceph_osd_client *osdc,
 					 use_mempool,
 					 GFP_NOFS, NULL, NULL);
 	if (!req)
-		return NULL;
+		return ERR_PTR(-ENOMEM);
 
 	/* calculate max write size */
-	calc_layout(osdc, vino, layout, off, plen, req, ops);
+	r = calc_layout(osdc, vino, layout, off, plen, req, ops);
+	if (r < 0)
+		return ERR_PTR(r);
 	req->r_file_layout = *layout;  /* keep a copy */
 
 	/* in case it differs from natural (file) alignment that
@@ -1920,8 +1930,8 @@ int ceph_osdc_readpages(struct ceph_osd_client *osdc,
 				    CEPH_OSD_OP_READ, CEPH_OSD_FLAG_READ,
 				    NULL, 0, truncate_seq, truncate_size, NULL,
 				    false, 1, page_align);
-	if (!req)
-		return -ENOMEM;
+	if (IS_ERR(req))
+		return PTR_ERR(req);
 
 	/* it may be a short read due to an object boundary */
 	req->r_pages = pages;
@@ -1963,8 +1973,8 @@ int ceph_osdc_writepages(struct ceph_osd_client *osdc, struct ceph_vino vino,
 				    snapc, do_sync,
 				    truncate_seq, truncate_size, mtime,
 				    nofail, 1, page_align);
-	if (!req)
-		return -ENOMEM;
+	if (IS_ERR(req))
+		return PTR_ERR(req);
 
 	/* it may be a short write due to an object boundary */
 	req->r_pages = pages;
diff --git a/net/ceph/osdmap.c b/net/ceph/osdmap.c
index 3124b71..5433fb0 100644
--- a/net/ceph/osdmap.c
+++ b/net/ceph/osdmap.c
@@ -984,7 +984,7 @@ bad:
  * for now, we write only a single su, until we can
  * pass a stride back to the caller.
  */
-void ceph_calc_file_object_mapping(struct ceph_file_layout *layout,
+int ceph_calc_file_object_mapping(struct ceph_file_layout *layout,
 				   u64 off, u64 *plen,
 				   u64 *ono,
 				   u64 *oxoff, u64 *oxlen)
@@ -998,11 +998,17 @@ void ceph_calc_file_object_mapping(struct ceph_file_layout *layout,
 
 	dout("mapping %llu~%llu  osize %u fl_su %u\n", off, *plen,
 	     osize, su);
+	if (su == 0 || sc == 0)
+		goto invalid;
 	su_per_object = osize / su;
+	if (su_per_object == 0)
+		goto invalid;
 	dout("osize %u / su %u = su_per_object %u\n", osize, su,
 	     su_per_object);
 
-	BUG_ON((su & ~PAGE_MASK) != 0);
+	if ((su & ~PAGE_MASK) != 0)
+		goto invalid;
+
 	/* bl = *off / su; */
 	t = off;
 	do_div(t, su);
@@ -1030,6 +1036,14 @@ void ceph_calc_file_object_mapping(struct ceph_file_layout *layout,
 	*plen = *oxlen;
 
 	dout(" obj extent %llu~%llu\n", *oxoff, *oxlen);
+	return 0;
+
+invalid:
+	dout(" invalid layout\n");
+	*ono = 0;
+	*oxoff = 0;
+	*oxlen = 0;
+	return -EINVAL;
 }
 EXPORT_SYMBOL(ceph_calc_file_object_mapping);
 
diff --git a/net/ceph/pagelist.c b/net/ceph/pagelist.c
index 665cd23..92866be 100644
--- a/net/ceph/pagelist.c
+++ b/net/ceph/pagelist.c
@@ -1,4 +1,3 @@
-
 #include <linux/module.h>
 #include <linux/gfp.h>
 #include <linux/pagemap.h>
@@ -134,8 +133,8 @@ int ceph_pagelist_truncate(struct ceph_pagelist *pl,
 	ceph_pagelist_unmap_tail(pl);
 	while (pl->head.prev != c->page_lru) {
 		page = list_entry(pl->head.prev, struct page, lru);
-		list_del(&page->lru);                /* remove from pagelist */
-		list_add_tail(&page->lru, &pl->free_list); /* add to reserve */
+		/* move from pagelist to reserve */
+		list_move_tail(&page->lru, &pl->free_list);
 		++pl->num_pages_free;
 	}
 	pl->room = c->room;