1 files changed, 114 insertions, 85 deletions

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index ca6dda0f68bb..6e1c9b2d92c5 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -980,6 +980,7 @@ static struct nfs4_opendata *nfs4_opendata_alloc(struct dentry *dentry,
 	struct dentry *parent = dget_parent(dentry);
 	struct inode *dir = parent->d_inode;
 	struct nfs_server *server = NFS_SERVER(dir);
+	struct nfs_seqid *(*alloc_seqid)(struct nfs_seqid_counter *, gfp_t);
 	struct nfs4_opendata *p;
 
 	p = kzalloc(sizeof(*p), gfp_mask);
@@ -990,8 +991,9 @@ static struct nfs4_opendata *nfs4_opendata_alloc(struct dentry *dentry,
 	if (IS_ERR(p->f_label))
 		goto err_free_p;
 
-	p->o_arg.seqid = nfs_alloc_seqid(&sp->so_seqid, gfp_mask);
-	if (p->o_arg.seqid == NULL)
+	alloc_seqid = server->nfs_client->cl_mvops->alloc_seqid;
+	p->o_arg.seqid = alloc_seqid(&sp->so_seqid, gfp_mask);
+	if (IS_ERR(p->o_arg.seqid))
 		goto err_free_label;
 	nfs_sb_active(dentry->d_sb);
 	p->dentry = dget(dentry);
@@ -1170,6 +1172,16 @@ static bool nfs_need_update_open_stateid(struct nfs4_state *state,
 	return false;
 }
 
+static void nfs_resync_open_stateid_locked(struct nfs4_state *state)
+{
+	if (state->n_wronly)
+		set_bit(NFS_O_WRONLY_STATE, &state->flags);
+	if (state->n_rdonly)
+		set_bit(NFS_O_RDONLY_STATE, &state->flags);
+	if (state->n_rdwr)
+		set_bit(NFS_O_RDWR_STATE, &state->flags);
+}
+
 static void nfs_clear_open_stateid_locked(struct nfs4_state *state,
 		nfs4_stateid *stateid, fmode_t fmode)
 {
@@ -1188,8 +1200,12 @@ static void nfs_clear_open_stateid_locked(struct nfs4_state *state,
 	}
 	if (stateid == NULL)
 		return;
-	if (!nfs_need_update_open_stateid(state, stateid))
+	/* Handle races with OPEN */
+	if (!nfs4_stateid_match_other(stateid, &state->open_stateid) ||
+	    !nfs4_stateid_is_newer(stateid, &state->open_stateid)) {
+		nfs_resync_open_stateid_locked(state);
 		return;
+	}
 	if (test_bit(NFS_DELEGATED_STATE, &state->flags) == 0)
 		nfs4_stateid_copy(&state->stateid, stateid);
 	nfs4_stateid_copy(&state->open_stateid, stateid);
@@ -1284,6 +1300,23 @@ no_delegation:
 	return ret;
 }
 
+static bool nfs4_update_lock_stateid(struct nfs4_lock_state *lsp,
+		const nfs4_stateid *stateid)
+{
+	struct nfs4_state *state = lsp->ls_state;
+	bool ret = false;
+
+	spin_lock(&state->state_lock);
+	if (!nfs4_stateid_match_other(stateid, &lsp->ls_stateid))
+		goto out_noupdate;
+	if (!nfs4_stateid_is_newer(stateid, &lsp->ls_stateid))
+		goto out_noupdate;
+	nfs4_stateid_copy(&lsp->ls_stateid, stateid);
+	ret = true;
+out_noupdate:
+	spin_unlock(&state->state_lock);
+	return ret;
+}
 
 static void nfs4_return_incompatible_delegation(struct inode *inode, fmode_t fmode)
 {
@@ -2590,6 +2623,11 @@ static void nfs4_close_done(struct rpc_task *task, void *data)
 		case -NFS4ERR_OLD_STATEID:
 		case -NFS4ERR_BAD_STATEID:
 		case -NFS4ERR_EXPIRED:
+			if (!nfs4_stateid_match(&calldata->arg.stateid,
+						&state->stateid)) {
+				rpc_restart_call_prepare(task);
+				goto out_release;
+			}
 			if (calldata->arg.fmode == 0)
 				break;
 		default:
@@ -2622,6 +2660,7 @@ static void nfs4_close_prepare(struct rpc_task *task, void *data)
 	is_rdwr = test_bit(NFS_O_RDWR_STATE, &state->flags);
 	is_rdonly = test_bit(NFS_O_RDONLY_STATE, &state->flags);
 	is_wronly = test_bit(NFS_O_WRONLY_STATE, &state->flags);
+	nfs4_stateid_copy(&calldata->arg.stateid, &state->stateid);
 	/* Calculate the change in open mode */
 	calldata->arg.fmode = 0;
 	if (state->n_rdwr == 0) {
@@ -2678,45 +2717,10 @@ static const struct rpc_call_ops nfs4_close_ops = {
 	.rpc_release = nfs4_free_closedata,
 };
 
-static bool nfs4_state_has_opener(struct nfs4_state *state)
-{
-	/* first check existing openers */
-	if (test_bit(NFS_O_RDONLY_STATE, &state->flags) != 0 &&
-	    state->n_rdonly != 0)
-		return true;
-
-	if (test_bit(NFS_O_WRONLY_STATE, &state->flags) != 0 &&
-	    state->n_wronly != 0)
-		return true;
-
-	if (test_bit(NFS_O_RDWR_STATE, &state->flags) != 0 &&
-	    state->n_rdwr != 0)
-		return true;
-
-	return false;
-}
-
 static bool nfs4_roc(struct inode *inode)
 {
-	struct nfs_inode *nfsi = NFS_I(inode);
-	struct nfs_open_context *ctx;
-	struct nfs4_state *state;
-
-	spin_lock(&inode->i_lock);
-	list_for_each_entry(ctx, &nfsi->open_files, list) {
-		state = ctx->state;
-		if (state == NULL)
-			continue;
-		if (nfs4_state_has_opener(state)) {
-			spin_unlock(&inode->i_lock);
-			return false;
-		}
-	}
-	spin_unlock(&inode->i_lock);
-
-	if (nfs4_check_delegation(inode, FMODE_READ))
+	if (!nfs_have_layout(inode))
 		return false;
-
 	return pnfs_roc(inode);
 }
 
@@ -2734,6 +2738,7 @@ static bool nfs4_roc(struct inode *inode)
 int nfs4_do_close(struct nfs4_state *state, gfp_t gfp_mask, int wait)
 {
 	struct nfs_server *server = NFS_SERVER(state->inode);
+	struct nfs_seqid *(*alloc_seqid)(struct nfs_seqid_counter *, gfp_t);
 	struct nfs4_closedata *calldata;
 	struct nfs4_state_owner *sp = state->owner;
 	struct rpc_task *task;
@@ -2760,10 +2765,10 @@ int nfs4_do_close(struct nfs4_state *state, gfp_t gfp_mask, int wait)
 	calldata->inode = state->inode;
 	calldata->state = state;
 	calldata->arg.fh = NFS_FH(state->inode);
-	calldata->arg.stateid = &state->open_stateid;
 	/* Serialization for the sequence id */
-	calldata->arg.seqid = nfs_alloc_seqid(&state->owner->so_seqid, gfp_mask);
-	if (calldata->arg.seqid == NULL)
+	alloc_seqid = server->nfs_client->cl_mvops->alloc_seqid;
+	calldata->arg.seqid = alloc_seqid(&state->owner->so_seqid, gfp_mask);
+	if (IS_ERR(calldata->arg.seqid))
 		goto out_free_calldata;
 	calldata->arg.fmode = 0;
 	calldata->arg.bitmask = server->cache_consistency_bitmask;
@@ -5356,7 +5361,6 @@ static struct nfs4_unlockdata *nfs4_alloc_unlockdata(struct file_lock *fl,
 	p->arg.fl = &p->fl;
 	p->arg.seqid = seqid;
 	p->res.seqid = seqid;
-	p->arg.stateid = &lsp->ls_stateid;
 	p->lsp = lsp;
 	atomic_inc(&lsp->ls_count);
 	/* Ensure we don't close file until we're done freeing locks! */
@@ -5383,14 +5387,18 @@ static void nfs4_locku_done(struct rpc_task *task, void *data)
 		return;
 	switch (task->tk_status) {
 		case 0:
-			nfs4_stateid_copy(&calldata->lsp->ls_stateid,
-					&calldata->res.stateid);
 			renew_lease(calldata->server, calldata->timestamp);
-			break;
+			do_vfs_lock(calldata->fl.fl_file, &calldata->fl);
+			if (nfs4_update_lock_stateid(calldata->lsp,
+					&calldata->res.stateid))
+				break;
 		case -NFS4ERR_BAD_STATEID:
 		case -NFS4ERR_OLD_STATEID:
 		case -NFS4ERR_STALE_STATEID:
 		case -NFS4ERR_EXPIRED:
+			if (!nfs4_stateid_match(&calldata->arg.stateid,
+						&calldata->lsp->ls_stateid))
+				rpc_restart_call_prepare(task);
 			break;
 		default:
 			if (nfs4_async_handle_error(task, calldata->server,
@@ -5406,6 +5414,7 @@ static void nfs4_locku_prepare(struct rpc_task *task, void *data)
 
 	if (nfs_wait_on_sequence(calldata->arg.seqid, task) != 0)
 		goto out_wait;
+	nfs4_stateid_copy(&calldata->arg.stateid, &calldata->lsp->ls_stateid);
 	if (test_bit(NFS_LOCK_INITIALIZED, &calldata->lsp->ls_flags) == 0) {
 		/* Note: exit _without_ running nfs4_locku_done */
 		goto out_no_action;
@@ -5476,6 +5485,7 @@ static int nfs4_proc_unlck(struct nfs4_state *state, int cmd, struct file_lock *
 	struct nfs_seqid *seqid;
 	struct nfs4_lock_state *lsp;
 	struct rpc_task *task;
+	struct nfs_seqid *(*alloc_seqid)(struct nfs_seqid_counter *, gfp_t);
 	int status = 0;
 	unsigned char fl_flags = request->fl_flags;
 
@@ -5499,9 +5509,10 @@ static int nfs4_proc_unlck(struct nfs4_state *state, int cmd, struct file_lock *
 	lsp = request->fl_u.nfs4_fl.owner;
 	if (test_bit(NFS_LOCK_INITIALIZED, &lsp->ls_flags) == 0)
 		goto out;
-	seqid = nfs_alloc_seqid(&lsp->ls_seqid, GFP_KERNEL);
+	alloc_seqid = NFS_SERVER(inode)->nfs_client->cl_mvops->alloc_seqid;
+	seqid = alloc_seqid(&lsp->ls_seqid, GFP_KERNEL);
 	status = -ENOMEM;
-	if (seqid == NULL)
+	if (IS_ERR(seqid))
 		goto out;
 	task = nfs4_do_unlck(request, nfs_file_open_context(request->fl_file), lsp, seqid);
 	status = PTR_ERR(task);
@@ -5534,6 +5545,7 @@ static struct nfs4_lockdata *nfs4_alloc_lockdata(struct file_lock *fl,
 	struct nfs4_lockdata *p;
 	struct inode *inode = lsp->ls_state->inode;
 	struct nfs_server *server = NFS_SERVER(inode);
+	struct nfs_seqid *(*alloc_seqid)(struct nfs_seqid_counter *, gfp_t);
 
 	p = kzalloc(sizeof(*p), gfp_mask);
 	if (p == NULL)
@@ -5542,12 +5554,12 @@ static struct nfs4_lockdata *nfs4_alloc_lockdata(struct file_lock *fl,
 	p->arg.fh = NFS_FH(inode);
 	p->arg.fl = &p->fl;
 	p->arg.open_seqid = nfs_alloc_seqid(&lsp->ls_state->owner->so_seqid, gfp_mask);
-	if (p->arg.open_seqid == NULL)
+	if (IS_ERR(p->arg.open_seqid))
 		goto out_free;
-	p->arg.lock_seqid = nfs_alloc_seqid(&lsp->ls_seqid, gfp_mask);
-	if (p->arg.lock_seqid == NULL)
+	alloc_seqid = server->nfs_client->cl_mvops->alloc_seqid;
+	p->arg.lock_seqid = alloc_seqid(&lsp->ls_seqid, gfp_mask);
+	if (IS_ERR(p->arg.lock_seqid))
 		goto out_free_seqid;
-	p->arg.lock_stateid = &lsp->ls_stateid;
 	p->arg.lock_owner.clientid = server->nfs_client->cl_clientid;
 	p->arg.lock_owner.id = lsp->ls_seqid.owner_id;
 	p->arg.lock_owner.s_dev = server->s_dev;
@@ -5574,15 +5586,19 @@ static void nfs4_lock_prepare(struct rpc_task *task, void *calldata)
 	if (nfs_wait_on_sequence(data->arg.lock_seqid, task) != 0)
 		goto out_wait;
 	/* Do we need to do an open_to_lock_owner? */
-	if (!(data->arg.lock_seqid->sequence->flags & NFS_SEQID_CONFIRMED)) {
+	if (!test_bit(NFS_LOCK_INITIALIZED, &data->lsp->ls_flags)) {
 		if (nfs_wait_on_sequence(data->arg.open_seqid, task) != 0) {
 			goto out_release_lock_seqid;
 		}
-		data->arg.open_stateid = &state->open_stateid;
+		nfs4_stateid_copy(&data->arg.open_stateid,
+				&state->open_stateid);
 		data->arg.new_lock_owner = 1;
 		data->res.open_seqid = data->arg.open_seqid;
-	} else
+	} else {
 		data->arg.new_lock_owner = 0;
+		nfs4_stateid_copy(&data->arg.lock_stateid,
+				&data->lsp->ls_stateid);
+	}
 	if (!nfs4_valid_open_stateid(state)) {
 		data->rpc_status = -EBADF;
 		task->tk_action = NULL;
@@ -5606,6 +5622,7 @@ out_wait:
 static void nfs4_lock_done(struct rpc_task *task, void *calldata)
 {
 	struct nfs4_lockdata *data = calldata;
+	struct nfs4_lock_state *lsp = data->lsp;
 
 	dprintk("%s: begin!\n", __func__);
 
@@ -5613,18 +5630,36 @@ static void nfs4_lock_done(struct rpc_task *task, void *calldata)
 		return;
 
 	data->rpc_status = task->tk_status;
-	if (data->arg.new_lock_owner != 0) {
-		if (data->rpc_status == 0)
-			nfs_confirm_seqid(&data->lsp->ls_seqid, 0);
-		else
-			goto out;
-	}
-	if (data->rpc_status == 0) {
-		nfs4_stateid_copy(&data->lsp->ls_stateid, &data->res.stateid);
-		set_bit(NFS_LOCK_INITIALIZED, &data->lsp->ls_flags);
-		renew_lease(NFS_SERVER(data->ctx->dentry->d_inode), data->timestamp);
+	switch (task->tk_status) {
+	case 0:
+		renew_lease(NFS_SERVER(data->ctx->dentry->d_inode),
+				data->timestamp);
+		if (data->arg.new_lock) {
+			data->fl.fl_flags &= ~(FL_SLEEP | FL_ACCESS);
+			if (do_vfs_lock(data->fl.fl_file, &data->fl) < 0) {
+				rpc_restart_call_prepare(task);
+				break;
+			}
+		}
+		if (data->arg.new_lock_owner != 0) {
+			nfs_confirm_seqid(&lsp->ls_seqid, 0);
+			nfs4_stateid_copy(&lsp->ls_stateid, &data->res.stateid);
+			set_bit(NFS_LOCK_INITIALIZED, &lsp->ls_flags);
+		} else if (!nfs4_update_lock_stateid(lsp, &data->res.stateid))
+			rpc_restart_call_prepare(task);
+		break;
+	case -NFS4ERR_BAD_STATEID:
+	case -NFS4ERR_OLD_STATEID:
+	case -NFS4ERR_STALE_STATEID:
+	case -NFS4ERR_EXPIRED:
+		if (data->arg.new_lock_owner != 0) {
+			if (!nfs4_stateid_match(&data->arg.open_stateid,
+						&lsp->ls_state->open_stateid))
+				rpc_restart_call_prepare(task);
+		} else if (!nfs4_stateid_match(&data->arg.lock_stateid,
+						&lsp->ls_stateid))
+				rpc_restart_call_prepare(task);
 	}
-out:
 	dprintk("%s: done, ret = %d!\n", __func__, data->rpc_status);
 }
 
@@ -5705,7 +5740,8 @@ static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *f
 		if (recovery_type == NFS_LOCK_RECLAIM)
 			data->arg.reclaim = NFS_LOCK_RECLAIM;
 		nfs4_set_sequence_privileged(&data->arg.seq_args);
-	}
+	} else
+		data->arg.new_lock = 1;
 	task = rpc_run_task(&task_setup_data);
 	if (IS_ERR(task))
 		return PTR_ERR(task);
@@ -5829,10 +5865,8 @@ static int nfs41_lock_expired(struct nfs4_state *state, struct file_lock *reques
 
 static int _nfs4_proc_setlk(struct nfs4_state *state, int cmd, struct file_lock *request)
 {
-	struct nfs4_state_owner *sp = state->owner;
 	struct nfs_inode *nfsi = NFS_I(state->inode);
 	unsigned char fl_flags = request->fl_flags;
-	unsigned int seq;
 	int status = -ENOLCK;
 
 	if ((fl_flags & FL_POSIX) &&
@@ -5852,25 +5886,11 @@ static int _nfs4_proc_setlk(struct nfs4_state *state, int cmd, struct file_lock
 		/* ...but avoid races with delegation recall... */
 		request->fl_flags = fl_flags & ~FL_SLEEP;
 		status = do_vfs_lock(request->fl_file, request);
-		goto out_unlock;
-	}
-	seq = raw_seqcount_begin(&sp->so_reclaim_seqcount);
-	up_read(&nfsi->rwsem);
-	status = _nfs4_do_setlk(state, cmd, request, NFS_LOCK_NEW);
-	if (status != 0)
+		up_read(&nfsi->rwsem);
 		goto out;
-	down_read(&nfsi->rwsem);
-	if (read_seqcount_retry(&sp->so_reclaim_seqcount, seq)) {
-		status = -NFS4ERR_DELAY;
-		goto out_unlock;
 	}
-	/* Note: we always want to sleep here! */
-	request->fl_flags = fl_flags | FL_SLEEP;
-	if (do_vfs_lock(request->fl_file, request) < 0)
-		printk(KERN_WARNING "NFS: %s: VFS is out of sync with lock "
-			"manager!\n", __func__);
-out_unlock:
 	up_read(&nfsi->rwsem);
+	status = _nfs4_do_setlk(state, cmd, request, NFS_LOCK_NEW);
 out:
 	request->fl_flags = fl_flags;
 	return status;
@@ -8409,6 +8429,7 @@ static const struct nfs4_minor_version_ops nfs_v4_0_minor_ops = {
 	.match_stateid = nfs4_match_stateid,
 	.find_root_sec = nfs4_find_root_sec,
 	.free_lock_state = nfs4_release_lockowner,
+	.alloc_seqid = nfs_alloc_seqid,
 	.call_sync_ops = &nfs40_call_sync_ops,
 	.reboot_recovery_ops = &nfs40_reboot_recovery_ops,
 	.nograce_recovery_ops = &nfs40_nograce_recovery_ops,
@@ -8417,6 +8438,12 @@ static const struct nfs4_minor_version_ops nfs_v4_0_minor_ops = {
 };
 
 #if defined(CONFIG_NFS_V4_1)
+static struct nfs_seqid *
+nfs_alloc_no_seqid(struct nfs_seqid_counter *arg1, gfp_t arg2)
+{
+	return NULL;
+}
+
 static const struct nfs4_minor_version_ops nfs_v4_1_minor_ops = {
 	.minor_version = 1,
 	.init_caps = NFS_CAP_READDIRPLUS
@@ -8430,6 +8457,7 @@ static const struct nfs4_minor_version_ops nfs_v4_1_minor_ops = {
 	.match_stateid = nfs41_match_stateid,
 	.find_root_sec = nfs41_find_root_sec,
 	.free_lock_state = nfs41_free_lock_state,
+	.alloc_seqid = nfs_alloc_no_seqid,
 	.call_sync_ops = &nfs41_call_sync_ops,
 	.reboot_recovery_ops = &nfs41_reboot_recovery_ops,
 	.nograce_recovery_ops = &nfs41_nograce_recovery_ops,
@@ -8456,6 +8484,7 @@ static const struct nfs4_minor_version_ops nfs_v4_2_minor_ops = {
 	.find_root_sec = nfs41_find_root_sec,
 	.free_lock_state = nfs41_free_lock_state,
 	.call_sync_ops = &nfs41_call_sync_ops,
+	.alloc_seqid = nfs_alloc_no_seqid,
 	.reboot_recovery_ops = &nfs41_reboot_recovery_ops,
 	.nograce_recovery_ops = &nfs41_nograce_recovery_ops,
 	.state_renewal_ops = &nfs41_state_renewal_ops,