Merge commit 'nfs/linux-next'

8 files changed, 106 insertions, 48 deletions

diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index c5c0175898f6..06f064d8fbbe 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -170,6 +170,8 @@ static void nfs4_shutdown_client(struct nfs_client *clp)
 	BUG_ON(!RB_EMPTY_ROOT(&clp->cl_state_owners));
 	if (__test_and_clear_bit(NFS_CS_IDMAP, &clp->cl_res_state))
 		nfs_idmap_delete(clp);
+
+	rpc_destroy_wait_queue(&clp->cl_rpcwaitq);
 #endif
 }
 
diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c
index 16844f98f50e..e44200579c8d 100644
--- a/fs/nfs/direct.c
+++ b/fs/nfs/direct.c
@@ -280,6 +280,7 @@ static ssize_t nfs_direct_read_schedule_segment(struct nfs_direct_req *dreq,
 		.rpc_client = NFS_CLIENT(inode),
 		.rpc_message = &msg,
 		.callback_ops = &nfs_read_direct_ops,
+		.workqueue = nfsiod_workqueue,
 		.flags = RPC_TASK_ASYNC,
 	};
 	unsigned int pgbase;
@@ -323,7 +324,7 @@ static ssize_t nfs_direct_read_schedule_segment(struct nfs_direct_req *dreq,
 		data->inode = inode;
 		data->cred = msg.rpc_cred;
 		data->args.fh = NFS_FH(inode);
-		data->args.context = ctx;
+		data->args.context = get_nfs_open_context(ctx);
 		data->args.offset = pos;
 		data->args.pgbase = pgbase;
 		data->args.pages = data->pagevec;
@@ -446,6 +447,7 @@ static void nfs_direct_write_reschedule(struct nfs_direct_req *dreq)
 	struct rpc_task_setup task_setup_data = {
 		.rpc_client = NFS_CLIENT(inode),
 		.callback_ops = &nfs_write_direct_ops,
+		.workqueue = nfsiod_workqueue,
 		.flags = RPC_TASK_ASYNC,
 	};
 
@@ -537,6 +539,7 @@ static void nfs_direct_commit_schedule(struct nfs_direct_req *dreq)
 		.rpc_message = &msg,
 		.callback_ops = &nfs_commit_direct_ops,
 		.callback_data = data,
+		.workqueue = nfsiod_workqueue,
 		.flags = RPC_TASK_ASYNC,
 	};
 
@@ -546,6 +549,7 @@ static void nfs_direct_commit_schedule(struct nfs_direct_req *dreq)
 	data->args.fh = NFS_FH(data->inode);
 	data->args.offset = 0;
 	data->args.count = 0;
+	data->args.context = get_nfs_open_context(dreq->ctx);
 	data->res.count = 0;
 	data->res.fattr = &data->fattr;
 	data->res.verf = &data->verf;
@@ -682,6 +686,7 @@ static ssize_t nfs_direct_write_schedule_segment(struct nfs_direct_req *dreq,
 		.rpc_client = NFS_CLIENT(inode),
 		.rpc_message = &msg,
 		.callback_ops = &nfs_write_direct_ops,
+		.workqueue = nfsiod_workqueue,
 		.flags = RPC_TASK_ASYNC,
 	};
 	size_t wsize = NFS_SERVER(inode)->wsize;
@@ -728,7 +733,7 @@ static ssize_t nfs_direct_write_schedule_segment(struct nfs_direct_req *dreq,
 		data->inode = inode;
 		data->cred = msg.rpc_cred;
 		data->args.fh = NFS_FH(inode);
-		data->args.context = ctx;
+		data->args.context = get_nfs_open_context(ctx);
 		data->args.offset = pos;
 		data->args.pgbase = pgbase;
 		data->args.pages = data->pagevec;
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index 966a8850aa30..36cb99985d22 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -521,8 +521,12 @@ struct nfs_open_context *get_nfs_open_context(struct nfs_open_context *ctx)
 
 static void __put_nfs_open_context(struct nfs_open_context *ctx, int wait)
 {
-	struct inode *inode = ctx->path.dentry->d_inode;
+	struct inode *inode;
 
+	if (ctx == NULL)
+		return;
+
+	inode = ctx->path.dentry->d_inode;
 	if (!atomic_dec_and_lock(&ctx->count, &inode->i_lock))
 		return;
 	list_del(&ctx->list);
@@ -1215,6 +1219,36 @@ static void nfs_destroy_inodecache(void)
 	kmem_cache_destroy(nfs_inode_cachep);
 }
 
+struct workqueue_struct *nfsiod_workqueue;
+
+/*
+ * start up the nfsiod workqueue
+ */
+static int nfsiod_start(void)
+{
+	struct workqueue_struct *wq;
+	dprintk("RPC:       creating workqueue nfsiod\n");
+	wq = create_singlethread_workqueue("nfsiod");
+	if (wq == NULL)
+		return -ENOMEM;
+	nfsiod_workqueue = wq;
+	return 0;
+}
+
+/*
+ * Destroy the nfsiod workqueue
+ */
+static void nfsiod_stop(void)
+{
+	struct workqueue_struct *wq;
+
+	wq = nfsiod_workqueue;
+	if (wq == NULL)
+		return;
+	nfsiod_workqueue = NULL;
+	destroy_workqueue(wq);
+}
+
 /*
  * Initialize NFS
  */
@@ -1222,6 +1256,10 @@ static int __init init_nfs_fs(void)
 {
 	int err;
 
+	err = nfsiod_start();
+	if (err)
+		goto out6;
+
 	err = nfs_fs_proc_init();
 	if (err)
 		goto out5;
@@ -1268,6 +1306,8 @@ out3:
 out4:
 	nfs_fs_proc_exit();
 out5:
+	nfsiod_stop();
+out6:
 	return err;
 }
 
@@ -1283,6 +1323,7 @@ static void __exit exit_nfs_fs(void)
 #endif
 	unregister_nfs_fs();
 	nfs_fs_proc_exit();
+	nfsiod_stop();
 }
 
 /* Not quite true; I just maintain it */
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index 0f5619611b8d..985dc2931031 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -143,6 +143,7 @@ extern struct rpc_procinfo nfs4_procedures[];
 extern int nfs_access_cache_shrinker(int nr_to_scan, gfp_t gfp_mask);
 
 /* inode.c */
+extern struct workqueue_struct *nfsiod_workqueue;
 extern struct inode *nfs_alloc_inode(struct super_block *sb);
 extern void nfs_destroy_inode(struct inode *);
 extern int nfs_write_inode(struct inode *,int);
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 7ce07862c2fb..bbb0d58ee6ac 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -51,6 +51,7 @@
 
 #include "nfs4_fs.h"
 #include "delegation.h"
+#include "internal.h"
 #include "iostat.h"
 
 #define NFSDBG_FACILITY		NFSDBG_PROC
@@ -773,6 +774,7 @@ static int _nfs4_proc_open_confirm(struct nfs4_opendata *data)
 		.rpc_message = &msg,
 		.callback_ops = &nfs4_open_confirm_ops,
 		.callback_data = data,
+		.workqueue = nfsiod_workqueue,
 		.flags = RPC_TASK_ASYNC,
 	};
 	int status;
@@ -910,6 +912,7 @@ static int _nfs4_proc_open(struct nfs4_opendata *data)
 		.rpc_message = &msg,
 		.callback_ops = &nfs4_open_ops,
 		.callback_data = data,
+		.workqueue = nfsiod_workqueue,
 		.flags = RPC_TASK_ASYNC,
 	};
 	int status;
@@ -1315,6 +1318,7 @@ int nfs4_do_close(struct path *path, struct nfs4_state *state, int wait)
 		.rpc_client = server->client,
 		.rpc_message = &msg,
 		.callback_ops = &nfs4_close_ops,
+		.workqueue = nfsiod_workqueue,
 		.flags = RPC_TASK_ASYNC,
 	};
 	int status = -ENOMEM;
@@ -2761,10 +2765,10 @@ nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server)
 		case -NFS4ERR_STALE_CLIENTID:
 		case -NFS4ERR_STALE_STATEID:
 		case -NFS4ERR_EXPIRED:
-			rpc_sleep_on(&clp->cl_rpcwaitq, task, NULL, NULL);
+			rpc_sleep_on(&clp->cl_rpcwaitq, task, NULL);
 			nfs4_schedule_state_recovery(clp);
 			if (test_bit(NFS4CLNT_STATE_RECOVER, &clp->cl_state) == 0)
-				rpc_wake_up_task(task);
+				rpc_wake_up_queued_task(&clp->cl_rpcwaitq, task);
 			task->tk_status = 0;
 			return -EAGAIN;
 		case -NFS4ERR_DELAY:
@@ -3235,6 +3239,7 @@ static struct rpc_task *nfs4_do_unlck(struct file_lock *fl,
 		.rpc_client = NFS_CLIENT(lsp->ls_state->inode),
 		.rpc_message = &msg,
 		.callback_ops = &nfs4_locku_ops,
+		.workqueue = nfsiod_workqueue,
 		.flags = RPC_TASK_ASYNC,
 	};
 
@@ -3419,6 +3424,7 @@ static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *f
 		.rpc_client = NFS_CLIENT(state->inode),
 		.rpc_message = &msg,
 		.callback_ops = &nfs4_lock_ops,
+		.workqueue = nfsiod_workqueue,
 		.flags = RPC_TASK_ASYNC,
 	};
 	int ret;
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index b962397004c1..7775435ea7a5 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -292,8 +292,10 @@ struct nfs4_state_owner *nfs4_get_state_owner(struct nfs_server *server, struct
 	spin_unlock(&clp->cl_lock);
 	if (sp == new)
 		get_rpccred(cred);
-	else
+	else {
+		rpc_destroy_wait_queue(&new->so_sequence.wait);
 		kfree(new);
+	}
 	return sp;
 }
 
@@ -310,6 +312,7 @@ void nfs4_put_state_owner(struct nfs4_state_owner *sp)
 		return;
 	nfs4_remove_state_owner(clp, sp);
 	spin_unlock(&clp->cl_lock);
+	rpc_destroy_wait_queue(&sp->so_sequence.wait);
 	put_rpccred(cred);
 	kfree(sp);
 }
@@ -529,6 +532,7 @@ static void nfs4_free_lock_state(struct nfs4_lock_state *lsp)
 	spin_lock(&clp->cl_lock);
 	nfs_free_unique_id(&clp->cl_lockowner_id, &lsp->ls_id);
 	spin_unlock(&clp->cl_lock);
+	rpc_destroy_wait_queue(&lsp->ls_sequence.wait);
 	kfree(lsp);
 }
 
@@ -731,7 +735,7 @@ int nfs_wait_on_sequence(struct nfs_seqid *seqid, struct rpc_task *task)
 		list_add_tail(&seqid->list, &sequence->list);
 	if (list_first_entry(&sequence->list, struct nfs_seqid, list) == seqid)
 		goto unlock;
-	rpc_sleep_on(&sequence->wait, task, NULL, NULL);
+	rpc_sleep_on(&sequence->wait, task, NULL);
 	status = -EAGAIN;
 unlock:
 	spin_unlock(&sequence->lock);
diff --git a/fs/nfs/read.c b/fs/nfs/read.c
index 3d7d9631e125..be9e8270f4d7 100644
--- a/fs/nfs/read.c
+++ b/fs/nfs/read.c
@@ -58,22 +58,19 @@ struct nfs_read_data *nfs_readdata_alloc(unsigned int pagecount)
 	return p;
 }
 
-static void nfs_readdata_rcu_free(struct rcu_head *head)
+static void nfs_readdata_free(struct nfs_read_data *p)
 {
-	struct nfs_read_data *p = container_of(head, struct nfs_read_data, task.u.tk_rcu);
 	if (p && (p->pagevec != &p->page_array[0]))
 		kfree(p->pagevec);
 	mempool_free(p, nfs_rdata_mempool);
 }
 
-static void nfs_readdata_free(struct nfs_read_data *rdata)
-{
-	call_rcu_bh(&rdata->task.u.tk_rcu, nfs_readdata_rcu_free);
-}
-
 void nfs_readdata_release(void *data)
 {
-        nfs_readdata_free(data);
+	struct nfs_read_data *rdata = data;
+
+	put_nfs_open_context(rdata->args.context);
+	nfs_readdata_free(rdata);
 }
 
 static
@@ -174,6 +171,7 @@ static void nfs_read_rpcsetup(struct nfs_page *req, struct nfs_read_data *data,
 		.rpc_message = &msg,
 		.callback_ops = call_ops,
 		.callback_data = data,
+		.workqueue = nfsiod_workqueue,
 		.flags = RPC_TASK_ASYNC | swap_flags,
 	};
 
@@ -186,7 +184,7 @@ static void nfs_read_rpcsetup(struct nfs_page *req, struct nfs_read_data *data,
 	data->args.pgbase = req->wb_pgbase + offset;
 	data->args.pages  = data->pagevec;
 	data->args.count  = count;
-	data->args.context = req->wb_context;
+	data->args.context = get_nfs_open_context(req->wb_context);
 
 	data->res.fattr   = &data->fattr;
 	data->res.count   = count;
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index f55c437124a2..1667e3984418 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -58,19 +58,13 @@ struct nfs_write_data *nfs_commit_alloc(void)
 	return p;
 }
 
-static void nfs_commit_rcu_free(struct rcu_head *head)
+void nfs_commit_free(struct nfs_write_data *p)
 {
-	struct nfs_write_data *p = container_of(head, struct nfs_write_data, task.u.tk_rcu);
 	if (p && (p->pagevec != &p->page_array[0]))
 		kfree(p->pagevec);
 	mempool_free(p, nfs_commit_mempool);
 }
 
-void nfs_commit_free(struct nfs_write_data *wdata)
-{
-	call_rcu_bh(&wdata->task.u.tk_rcu, nfs_commit_rcu_free);
-}
-
 struct nfs_write_data *nfs_writedata_alloc(unsigned int pagecount)
 {
 	struct nfs_write_data *p = mempool_alloc(nfs_wdata_mempool, GFP_NOFS);
@@ -92,21 +86,18 @@ struct nfs_write_data *nfs_writedata_alloc(unsigned int pagecount)
 	return p;
 }
 
-static void nfs_writedata_rcu_free(struct rcu_head *head)
+static void nfs_writedata_free(struct nfs_write_data *p)
 {
-	struct nfs_write_data *p = container_of(head, struct nfs_write_data, task.u.tk_rcu);
 	if (p && (p->pagevec != &p->page_array[0]))
 		kfree(p->pagevec);
 	mempool_free(p, nfs_wdata_mempool);
 }
 
-static void nfs_writedata_free(struct nfs_write_data *wdata)
+void nfs_writedata_release(void *data)
 {
-	call_rcu_bh(&wdata->task.u.tk_rcu, nfs_writedata_rcu_free);
-}
+	struct nfs_write_data *wdata = data;
 
-void nfs_writedata_release(void *wdata)
-{
+	put_nfs_open_context(wdata->args.context);
 	nfs_writedata_free(wdata);
 }
 
@@ -360,15 +351,13 @@ int nfs_writepages(struct address_space *mapping, struct writeback_control *wbc)
 /*
  * Insert a write request into an inode
  */
-static int nfs_inode_add_request(struct inode *inode, struct nfs_page *req)
+static void nfs_inode_add_request(struct inode *inode, struct nfs_page *req)
 {
 	struct nfs_inode *nfsi = NFS_I(inode);
 	int error;
 
 	error = radix_tree_insert(&nfsi->nfs_page_tree, req->wb_index, req);
-	BUG_ON(error == -EEXIST);
-	if (error)
-		return error;
+	BUG_ON(error);
 	if (!nfsi->npages) {
 		igrab(inode);
 		if (nfs_have_delegation(inode, FMODE_WRITE))
@@ -378,8 +367,8 @@ static int nfs_inode_add_request(struct inode *inode, struct nfs_page *req)
 	set_page_private(req->wb_page, (unsigned long)req);
 	nfsi->npages++;
 	kref_get(&req->wb_kref);
-	radix_tree_tag_set(&nfsi->nfs_page_tree, req->wb_index, NFS_PAGE_TAG_LOCKED);
-	return 0;
+	radix_tree_tag_set(&nfsi->nfs_page_tree, req->wb_index,
+				NFS_PAGE_TAG_LOCKED);
 }
 
 /*
@@ -591,6 +580,13 @@ static struct nfs_page * nfs_update_request(struct nfs_open_context* ctx,
 		/* Loop over all inode entries and see if we find
 		 * A request for the page we wish to update
 		 */
+		if (new) {
+			if (radix_tree_preload(GFP_NOFS)) {
+				nfs_release_request(new);
+				return ERR_PTR(-ENOMEM);
+			}
+		}
+
 		spin_lock(&inode->i_lock);
 		req = nfs_page_find_request_locked(page);
 		if (req) {
@@ -601,28 +597,27 @@ static struct nfs_page * nfs_update_request(struct nfs_open_context* ctx,
 				error = nfs_wait_on_request(req);
 				nfs_release_request(req);
 				if (error < 0) {
-					if (new)
+					if (new) {
+						radix_tree_preload_end();
 						nfs_release_request(new);
+					}
 					return ERR_PTR(error);
 				}
 				continue;
 			}
 			spin_unlock(&inode->i_lock);
-			if (new)
+			if (new) {
+				radix_tree_preload_end();
 				nfs_release_request(new);
+			}
 			break;
 		}
 
 		if (new) {
-			int error;
 			nfs_lock_request_dontget(new);
-			error = nfs_inode_add_request(inode, new);
-			if (error) {
-				spin_unlock(&inode->i_lock);
-				nfs_unlock_request(new);
-				return ERR_PTR(error);
-			}
+			nfs_inode_add_request(inode, new);
 			spin_unlock(&inode->i_lock);
+			radix_tree_preload_end();
 			req = new;
 			goto zero_page;
 		}
@@ -734,7 +729,7 @@ int nfs_updatepage(struct file *file, struct page *page,
 	 */
 	if (nfs_write_pageuptodate(page, inode) &&
 			inode->i_flock == NULL &&
-			!(file->f_mode & O_SYNC)) {
+			!(file->f_flags & O_SYNC)) {
 		count = max(count + offset, nfs_page_length(page));
 		offset = 0;
 	}
@@ -800,6 +795,7 @@ static void nfs_write_rpcsetup(struct nfs_page *req,
 		.rpc_message = &msg,
 		.callback_ops = call_ops,
 		.callback_data = data,
+		.workqueue = nfsiod_workqueue,
 		.flags = flags,
 		.priority = priority,
 	};
@@ -816,7 +812,7 @@ static void nfs_write_rpcsetup(struct nfs_page *req,
 	data->args.pgbase = req->wb_pgbase + offset;
 	data->args.pages  = data->pagevec;
 	data->args.count  = count;
-	data->args.context = req->wb_context;
+	data->args.context = get_nfs_open_context(req->wb_context);
 	data->args.stable  = NFS_UNSTABLE;
 	if (how & FLUSH_STABLE) {
 		data->args.stable = NFS_DATA_SYNC;
@@ -1153,8 +1149,11 @@ int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data)
 
 
 #if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4)
-void nfs_commit_release(void *wdata)
+void nfs_commit_release(void *data)
 {
+	struct nfs_write_data *wdata = data;
+
+	put_nfs_open_context(wdata->args.context);
 	nfs_commit_free(wdata);
 }
 
@@ -1181,6 +1180,7 @@ static void nfs_commit_rpcsetup(struct list_head *head,
 		.rpc_message = &msg,
 		.callback_ops = &nfs_commit_ops,
 		.callback_data = data,
+		.workqueue = nfsiod_workqueue,
 		.flags = flags,
 		.priority = priority,
 	};
@@ -1197,6 +1197,7 @@ static void nfs_commit_rpcsetup(struct list_head *head,
 	/* Note: we always request a commit of the entire inode */
 	data->args.offset = 0;
 	data->args.count  = 0;
+	data->args.context = get_nfs_open_context(first->wb_context);
 	data->res.count   = 0;
 	data->res.fattr   = &data->fattr;
 	data->res.verf    = &data->verf;