Merge tag 'nfs-for-7.2-2' of git://git.linux-nfs.org/projects/anna/linux-nfs

Pull NFS client fixes from Anna Schumaker:

 - SUNRPC:
    - Release lower rpc_clnt if killed waiting for XPRT_LOCKED
    - Pin upper rpc_clnt across the TLS connect_worker

 - NFS:
    - Include MAY_WRITE in open permission mask for O_TRUNC
    - Charge unstable writes by request size, not folio size

* tag 'nfs-for-7.2-2' of git://git.linux-nfs.org/projects/anna/linux-nfs:
  NFS: Charge unstable writes by request size, not folio size
  NFSv4: include MAY_WRITE in open permission mask for O_TRUNC
  SUNRPC: pin upper rpc_clnt across the TLS connect_worker
  SUNRPC: release lower rpc_clnt if killed waiting for XPRT_LOCKED

fs/nfs/dir.c

@@ -3346,6 +3346,8 @@ static int nfs_open_permission_mask(int openflags)
 			mask |= MAY_READ;
 		if ((openflags & O_ACCMODE) != O_RDONLY)
 			mask |= MAY_WRITE;
+		if (openflags & O_TRUNC)
+			mask |= MAY_WRITE;
 	}
 
 	return mask;

fs/nfs/internal.h

@@ -851,17 +851,19 @@ void nfs_super_set_maxbytes(struct super_block *sb, __u64 maxfilesize)
 }
 
 /*
- * Record the page as unstable (an extra writeback period) and mark its
- * inode as dirty.
+ * Record the request's range as unstable (an extra writeback period) and
+ * mark its inode as dirty.
  */
-static inline void nfs_folio_mark_unstable(struct folio *folio,
+static inline void nfs_folio_mark_unstable(struct nfs_page *req,
 					   struct nfs_commit_info *cinfo)
 {
+	struct folio *folio = nfs_page_to_folio(req);
+
 	if (folio && !cinfo->dreq) {
 		struct inode *inode = folio->mapping->host;
-		long nr = folio_nr_pages(folio);
+		long nr = DIV_ROUND_UP(req->wb_bytes, PAGE_SIZE);
 
-		/* This page is really still in write-back - just that the
+		/* This range is really still in write-back - just that the
 		 * writeback is happening on the server now.
 		 */
 		node_stat_mod_folio(folio, NR_WRITEBACK, nr);

fs/nfs/pnfs_nfs.c

@@ -1199,7 +1199,7 @@ pnfs_layout_mark_request_commit(struct nfs_page *req,
 
 	nfs_request_add_commit_list_locked(req, list, cinfo);
 	mutex_unlock(&NFS_I(cinfo->inode)->commit_mutex);
-	nfs_folio_mark_unstable(nfs_page_to_folio(req), cinfo);
+	nfs_folio_mark_unstable(req, cinfo);
 	return;
 out_resched:
 	mutex_unlock(&NFS_I(cinfo->inode)->commit_mutex);

fs/nfs/write.c

@@ -807,7 +807,7 @@ nfs_request_add_commit_list(struct nfs_page *req, struct nfs_commit_info *cinfo)
 	mutex_lock(&NFS_I(cinfo->inode)->commit_mutex);
 	nfs_request_add_commit_list_locked(req, &cinfo->mds->list, cinfo);
 	mutex_unlock(&NFS_I(cinfo->inode)->commit_mutex);
-	nfs_folio_mark_unstable(nfs_page_to_folio(req), cinfo);
+	nfs_folio_mark_unstable(req, cinfo);
 }
 EXPORT_SYMBOL_GPL(nfs_request_add_commit_list);
 
@@ -866,10 +866,12 @@ nfs_mark_request_commit(struct nfs_page *req, struct pnfs_layout_segment *lseg,
 	nfs_request_add_commit_list(req, cinfo);
 }
 
-static void nfs_folio_clear_commit(struct folio *folio)
+static void nfs_folio_clear_commit(struct nfs_page *req)
 {
+	struct folio *folio = nfs_page_to_folio(req);
+
 	if (folio) {
-		long nr = folio_nr_pages(folio);
+		long nr = DIV_ROUND_UP(req->wb_bytes, PAGE_SIZE);
 
 		node_stat_mod_folio(folio, NR_WRITEBACK, -nr);
 		bdi_wb_stat_mod(folio->mapping->host, WB_WRITEBACK, -nr);
@@ -889,7 +891,7 @@ static void nfs_clear_request_commit(struct nfs_commit_info *cinfo,
 			nfs_request_remove_commit_list(req, cinfo);
 		}
 		mutex_unlock(&NFS_I(inode)->commit_mutex);
-		nfs_folio_clear_commit(nfs_page_to_folio(req));
+		nfs_folio_clear_commit(req);
 	}
 }
 
@@ -1741,7 +1743,7 @@ void nfs_retry_commit(struct list_head *page_list,
 		req = nfs_list_entry(page_list->next);
 		nfs_list_remove_request(req);
 		nfs_mark_request_commit(req, lseg, cinfo, ds_commit_idx);
-		nfs_folio_clear_commit(nfs_page_to_folio(req));
+		nfs_folio_clear_commit(req);
 		nfs_unlock_and_release_request(req);
 	}
 }
@@ -1813,7 +1815,7 @@ static void nfs_commit_release_pages(struct nfs_commit_data *data)
 		req = nfs_list_entry(data->pages.next);
 		nfs_list_remove_request(req);
 		folio = nfs_page_to_folio(req);
-		nfs_folio_clear_commit(folio);
+		nfs_folio_clear_commit(req);
 
 		dprintk("NFS:       commit (%s/%llu %d@%lld)",
 			nfs_req_openctx(req)->dentry->d_sb->s_id,

include/linux/sunrpc/clnt.h

@@ -190,6 +190,7 @@ int		rpc_switch_client_transport(struct rpc_clnt *,
 				const struct rpc_timeout *);
 
 void		rpc_shutdown_client(struct rpc_clnt *);
+void		rpc_hold_client(struct rpc_clnt *);
 void		rpc_release_client(struct rpc_clnt *);
 void		rpc_task_release_transport(struct rpc_task *);
 void		rpc_task_release_client(struct rpc_task *);

net/sunrpc/clnt.c

@@ -1026,8 +1026,23 @@ rpc_free_auth(struct rpc_clnt *clnt)
 	return NULL;
 }
 
-/*
- * Release reference to the RPC client
+/**
+ * rpc_hold_client - acquire a reference on an rpc_clnt
+ * @clnt: rpc_clnt to pin
+ *
+ * Pairs with rpc_release_client().
+ */
+void rpc_hold_client(struct rpc_clnt *clnt)
+{
+	refcount_inc(&clnt->cl_count);
+}
+
+/**
+ * rpc_release_client - release a reference on an rpc_clnt
+ * @clnt: rpc_clnt to release
+ *
+ * Pairs with rpc_hold_client(). The rpc_clnt's resources are
+ * freed once its reference count drops to zero.
  */
 void
 rpc_release_client(struct rpc_clnt *clnt)

net/sunrpc/xprtsock.c

@@ -2734,8 +2734,11 @@ static void xs_tcp_tls_setup_socket(struct work_struct *work)
 	lower_xprt = rcu_dereference(lower_clnt->cl_xprt);
 	rcu_read_unlock();
 
-	if (wait_on_bit_lock(&lower_xprt->state, XPRT_LOCKED, TASK_KILLABLE))
+	if (wait_on_bit_lock(&lower_xprt->state, XPRT_LOCKED, TASK_KILLABLE)) {
+		/* XPRT_LOCKED was never acquired. */
+		rpc_shutdown_client(lower_clnt);
 		goto out_unlock;
+	}
 
 	status = xs_tls_handshake_sync(lower_xprt, &upper_xprt->xprtsec);
 	if (status) {
@@ -2758,6 +2761,7 @@ static void xs_tcp_tls_setup_socket(struct work_struct *work)
 out_unlock:
 	current_restore_flags(pflags, PF_MEMALLOC);
 	upper_transport->clnt = NULL;
+	rpc_release_client(upper_clnt);
 	xprt_unlock_connect(upper_xprt, upper_transport);
 	return;
 
@@ -2805,7 +2809,15 @@ static void xs_connect(struct rpc_xprt *xprt, struct rpc_task *task)
 	} else
 		dprintk("RPC:       xs_connect scheduled xprt %p\n", xprt);
 
-	transport->clnt = task->tk_client;
+	/*
+	 * Only the TLS connect_worker reads transport->clnt; pinning
+	 * the upper rpc_clnt unconditionally would form a cycle with
+	 * cl_xprt and prevent xprt destruction.
+	 */
+	if (xprt->xprtsec.policy != RPC_XPRTSEC_NONE) {
+		rpc_hold_client(task->tk_client);
+		transport->clnt = task->tk_client;
+	}
 	queue_delayed_work(xprtiod_workqueue,
 			&transport->connect_worker,
 			delay);