Merge tag 'nfs-for-5.11-2' of git://git.linux-nfs.org/projects/trondmy/linux-nfs

Pull NFS client fixes from Trond Myklebust:
 "Highlights include:

   - Fix parsing of link-local IPv6 addresses

   - Fix confusing logging of mount errors that was introduced by the
     fsopen() patchset.

   - Fix a tracing use after free in _nfs4_do_setlk()

   - Layout return-on-close fixes when called from nfs4_evict_inode()

   - Layout segments were being leaked in
     pnfs_generic_clear_request_commit()

   - Don't leak DS commits in pnfs_generic_retry_commit()

   - Fix an Oopsable use-after-free when nfs_delegation_find_inode_server()
     calls iput() on an inode after the super block has gone away"

* tag 'nfs-for-5.11-2' of git://git.linux-nfs.org/projects/trondmy/linux-nfs:
  NFS: nfs_igrab_and_active must first reference the superblock
  NFS: nfs_delegation_find_inode_server must first reference the superblock
  NFS/pNFS: Fix a leak of the layout 'plh_outstanding' counter
  NFS/pNFS: Don't leak DS commits in pnfs_generic_retry_commit()
  NFS/pNFS: Don't call pnfs_free_bucket_lseg() before removing the request
  pNFS: Stricter ordering of layoutget and layoutreturn
  pNFS: Clean up pnfs_layoutreturn_free_lsegs()
  pNFS: We want return-on-close to complete when evicting the inode
  pNFS: Mark layout for return if return-on-close was not sent
  net: sunrpc: interpret the return value of kstrtou32 correctly
  NFS: Adjust fs_context error logging
  NFS4: Fix use-after-free in trace_event_raw_event_nfs4_set_lock

8 files changed, 99 insertions, 82 deletions

diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c
index 816e1427f17e..04bf8066980c 100644
--- a/fs/nfs/delegation.c
+++ b/fs/nfs/delegation.c
@@ -1011,22 +1011,24 @@ nfs_delegation_find_inode_server(struct nfs_server *server,
 				 const struct nfs_fh *fhandle)
 {
 	struct nfs_delegation *delegation;
-	struct inode *freeme, *res = NULL;
+	struct super_block *freeme = NULL;
+	struct inode *res = NULL;
 
 	list_for_each_entry_rcu(delegation, &server->delegations, super_list) {
 		spin_lock(&delegation->lock);
 		if (delegation->inode != NULL &&
 		    !test_bit(NFS_DELEGATION_REVOKED, &delegation->flags) &&
 		    nfs_compare_fh(fhandle, &NFS_I(delegation->inode)->fh) == 0) {
-			freeme = igrab(delegation->inode);
-			if (freeme && nfs_sb_active(freeme->i_sb))
-				res = freeme;
+			if (nfs_sb_active(server->super)) {
+				freeme = server->super;
+				res = igrab(delegation->inode);
+			}
 			spin_unlock(&delegation->lock);
 			if (res != NULL)
 				return res;
 			if (freeme) {
 				rcu_read_unlock();
-				iput(freeme);
+				nfs_sb_deactive(freeme);
 				rcu_read_lock();
 			}
 			return ERR_PTR(-EAGAIN);
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index b840d0a91c9d..62d3189745cd 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -136,9 +136,29 @@ struct nfs_fs_context {
 	} clone_data;
 };
 
-#define nfs_errorf(fc, fmt, ...) errorf(fc, fmt, ## __VA_ARGS__)
-#define nfs_invalf(fc, fmt, ...) invalf(fc, fmt, ## __VA_ARGS__)
-#define nfs_warnf(fc, fmt, ...) warnf(fc, fmt, ## __VA_ARGS__)
+#define nfs_errorf(fc, fmt, ...) ((fc)->log.log ?		\
+	errorf(fc, fmt, ## __VA_ARGS__) :			\
+	({ dprintk(fmt "\n", ## __VA_ARGS__); }))
+
+#define nfs_ferrorf(fc, fac, fmt, ...) ((fc)->log.log ?		\
+	errorf(fc, fmt, ## __VA_ARGS__) :			\
+	({ dfprintk(fac, fmt "\n", ## __VA_ARGS__); }))
+
+#define nfs_invalf(fc, fmt, ...) ((fc)->log.log ?		\
+	invalf(fc, fmt, ## __VA_ARGS__) :			\
+	({ dprintk(fmt "\n", ## __VA_ARGS__);  -EINVAL; }))
+
+#define nfs_finvalf(fc, fac, fmt, ...) ((fc)->log.log ?		\
+	invalf(fc, fmt, ## __VA_ARGS__) :			\
+	({ dfprintk(fac, fmt "\n", ## __VA_ARGS__);  -EINVAL; }))
+
+#define nfs_warnf(fc, fmt, ...) ((fc)->log.log ?		\
+	warnf(fc, fmt, ## __VA_ARGS__) :			\
+	({ dprintk(fmt "\n", ## __VA_ARGS__); }))
+
+#define nfs_fwarnf(fc, fac, fmt, ...) ((fc)->log.log ?		\
+	warnf(fc, fmt, ## __VA_ARGS__) :			\
+	({ dfprintk(fac, fmt "\n", ## __VA_ARGS__); }))
 
 static inline struct nfs_fs_context *nfs_fc2context(const struct fs_context *fc)
 {
@@ -579,12 +599,14 @@ extern void nfs4_test_session_trunk(struct rpc_clnt *clnt,
 
 static inline struct inode *nfs_igrab_and_active(struct inode *inode)
 {
-	inode = igrab(inode);
-	if (inode != NULL && !nfs_sb_active(inode->i_sb)) {
-		iput(inode);
-		inode = NULL;
+	struct super_block *sb = inode->i_sb;
+
+	if (sb && nfs_sb_active(sb)) {
+		if (igrab(inode))
+			return inode;
+		nfs_sb_deactive(sb);
 	}
-	return inode;
+	return NULL;
 }
 
 static inline void nfs_iput_and_deactive(struct inode *inode)
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 0ce04e0e5d82..2f4679a62712 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -3536,10 +3536,8 @@ static void nfs4_close_done(struct rpc_task *task, void *data)
 	trace_nfs4_close(state, &calldata->arg, &calldata->res, task->tk_status);
 
 	/* Handle Layoutreturn errors */
-	if (pnfs_roc_done(task, calldata->inode,
-				&calldata->arg.lr_args,
-				&calldata->res.lr_res,
-				&calldata->res.lr_ret) == -EAGAIN)
+	if (pnfs_roc_done(task, &calldata->arg.lr_args, &calldata->res.lr_res,
+			  &calldata->res.lr_ret) == -EAGAIN)
 		goto out_restart;
 
 	/* hmm. we are done with the inode, and in the process of freeing
@@ -6384,10 +6382,8 @@ static void nfs4_delegreturn_done(struct rpc_task *task, void *calldata)
 	trace_nfs4_delegreturn_exit(&data->args, &data->res, task->tk_status);
 
 	/* Handle Layoutreturn errors */
-	if (pnfs_roc_done(task, data->inode,
-				&data->args.lr_args,
-				&data->res.lr_res,
-				&data->res.lr_ret) == -EAGAIN)
+	if (pnfs_roc_done(task, &data->args.lr_args, &data->res.lr_res,
+			  &data->res.lr_ret) == -EAGAIN)
 		goto out_restart;
 
 	switch (task->tk_status) {
@@ -6441,10 +6437,10 @@ static void nfs4_delegreturn_release(void *calldata)
 	struct nfs4_delegreturndata *data = calldata;
 	struct inode *inode = data->inode;
 
+	if (data->lr.roc)
+		pnfs_roc_release(&data->lr.arg, &data->lr.res,
+				 data->res.lr_ret);
 	if (inode) {
-		if (data->lr.roc)
-			pnfs_roc_release(&data->lr.arg, &data->lr.res,
-					data->res.lr_ret);
 		nfs_post_op_update_inode_force_wcc(inode, &data->fattr);
 		nfs_iput_and_deactive(inode);
 	}
@@ -6520,16 +6516,14 @@ static int _nfs4_proc_delegreturn(struct inode *inode, const struct cred *cred,
 	nfs_fattr_init(data->res.fattr);
 	data->timestamp = jiffies;
 	data->rpc_status = 0;
-	data->lr.roc = pnfs_roc(inode, &data->lr.arg, &data->lr.res, cred);
 	data->inode = nfs_igrab_and_active(inode);
-	if (data->inode) {
+	if (data->inode || issync) {
+		data->lr.roc = pnfs_roc(inode, &data->lr.arg, &data->lr.res,
+					cred);
 		if (data->lr.roc) {
 			data->args.lr_args = &data->lr.arg;
 			data->res.lr_res = &data->lr.res;
 		}
-	} else if (data->lr.roc) {
-		pnfs_roc_release(&data->lr.arg, &data->lr.res, 0);
-		data->lr.roc = false;
 	}
 
 	task_setup_data.callback_data = data;
@@ -7111,9 +7105,9 @@ static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *f
 					data->arg.new_lock_owner, ret);
 	} else
 		data->cancelled = true;
+	trace_nfs4_set_lock(fl, state, &data->res.stateid, cmd, ret);
 	rpc_put_task(task);
 	dprintk("%s: done, ret = %d!\n", __func__, ret);
-	trace_nfs4_set_lock(fl, state, &data->res.stateid, cmd, ret);
 	return ret;
 }
 
diff --git a/fs/nfs/nfs4super.c b/fs/nfs/nfs4super.c
index 984cc42ee54d..d09bcfd7db89 100644
--- a/fs/nfs/nfs4super.c
+++ b/fs/nfs/nfs4super.c
@@ -227,7 +227,7 @@ int nfs4_try_get_tree(struct fs_context *fc)
 			   fc, ctx->nfs_server.hostname,
 			   ctx->nfs_server.export_path);
 	if (err) {
-		nfs_errorf(fc, "NFS4: Couldn't follow remote path");
+		nfs_ferrorf(fc, MOUNT, "NFS4: Couldn't follow remote path");
 		dfprintk(MOUNT, "<-- nfs4_try_get_tree() = %d [error]\n", err);
 	} else {
 		dfprintk(MOUNT, "<-- nfs4_try_get_tree() = 0\n");
@@ -250,7 +250,7 @@ int nfs4_get_referral_tree(struct fs_context *fc)
 			    fc, ctx->nfs_server.hostname,
 			    ctx->nfs_server.export_path);
 	if (err) {
-		nfs_errorf(fc, "NFS4: Couldn't follow remote path");
+		nfs_ferrorf(fc, MOUNT, "NFS4: Couldn't follow remote path");
 		dfprintk(MOUNT, "<-- nfs4_get_referral_tree() = %d [error]\n", err);
 	} else {
 		dfprintk(MOUNT, "<-- nfs4_get_referral_tree() = 0\n");
diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c
index 07f59dc8cb2e..4f274f21c4ab 100644
--- a/fs/nfs/pnfs.c
+++ b/fs/nfs/pnfs.c
@@ -1152,7 +1152,7 @@ void pnfs_layoutreturn_free_lsegs(struct pnfs_layout_hdr *lo,
 	LIST_HEAD(freeme);
 
 	spin_lock(&inode->i_lock);
-	if (!pnfs_layout_is_valid(lo) || !arg_stateid ||
+	if (!pnfs_layout_is_valid(lo) ||
 	    !nfs4_stateid_match_other(&lo->plh_stateid, arg_stateid))
 		goto out_unlock;
 	if (stateid) {
@@ -1509,10 +1509,8 @@ out_noroc:
 	return false;
 }
 
-int pnfs_roc_done(struct rpc_task *task, struct inode *inode,
-		struct nfs4_layoutreturn_args **argpp,
-		struct nfs4_layoutreturn_res **respp,
-		int *ret)
+int pnfs_roc_done(struct rpc_task *task, struct nfs4_layoutreturn_args **argpp,
+		  struct nfs4_layoutreturn_res **respp, int *ret)
 {
 	struct nfs4_layoutreturn_args *arg = *argpp;
 	int retval = -EAGAIN;
@@ -1545,7 +1543,7 @@ int pnfs_roc_done(struct rpc_task *task, struct inode *inode,
 		return 0;
 	case -NFS4ERR_OLD_STATEID:
 		if (!nfs4_layout_refresh_old_stateid(&arg->stateid,
-					&arg->range, inode))
+						     &arg->range, arg->inode))
 			break;
 		*ret = -NFS4ERR_NOMATCHING_LAYOUT;
 		return -EAGAIN;
@@ -1560,23 +1558,28 @@ void pnfs_roc_release(struct nfs4_layoutreturn_args *args,
 		int ret)
 {
 	struct pnfs_layout_hdr *lo = args->layout;
-	const nfs4_stateid *arg_stateid = NULL;
+	struct inode *inode = args->inode;
 	const nfs4_stateid *res_stateid = NULL;
 	struct nfs4_xdr_opaque_data *ld_private = args->ld_private;
 
 	switch (ret) {
 	case -NFS4ERR_NOMATCHING_LAYOUT:
+		spin_lock(&inode->i_lock);
+		if (pnfs_layout_is_valid(lo) &&
+		    nfs4_stateid_match_other(&args->stateid, &lo->plh_stateid))
+			pnfs_set_plh_return_info(lo, args->range.iomode, 0);
+		pnfs_clear_layoutreturn_waitbit(lo);
+		spin_unlock(&inode->i_lock);
 		break;
 	case 0:
 		if (res->lrs_present)
 			res_stateid = &res->stateid;
 		fallthrough;
 	default:
-		arg_stateid = &args->stateid;
+		pnfs_layoutreturn_free_lsegs(lo, &args->stateid, &args->range,
+					     res_stateid);
 	}
 	trace_nfs4_layoutreturn_on_close(args->inode, &args->stateid, ret);
-	pnfs_layoutreturn_free_lsegs(lo, arg_stateid, &args->range,
-			res_stateid);
 	if (ld_private && ld_private->ops && ld_private->ops->free)
 		ld_private->ops->free(ld_private);
 	pnfs_put_layout_hdr(lo);
@@ -2015,6 +2018,27 @@ lookup_again:
 		goto lookup_again;
 	}
 
+	/*
+	 * Because we free lsegs when sending LAYOUTRETURN, we need to wait
+	 * for LAYOUTRETURN.
+	 */
+	if (test_bit(NFS_LAYOUT_RETURN, &lo->plh_flags)) {
+		spin_unlock(&ino->i_lock);
+		dprintk("%s wait for layoutreturn\n", __func__);
+		lseg = ERR_PTR(pnfs_prepare_to_retry_layoutget(lo));
+		if (!IS_ERR(lseg)) {
+			pnfs_put_layout_hdr(lo);
+			dprintk("%s retrying\n", __func__);
+			trace_pnfs_update_layout(ino, pos, count, iomode, lo,
+						 lseg,
+						 PNFS_UPDATE_LAYOUT_RETRY);
+			goto lookup_again;
+		}
+		trace_pnfs_update_layout(ino, pos, count, iomode, lo, lseg,
+					 PNFS_UPDATE_LAYOUT_RETURN);
+		goto out_put_layout_hdr;
+	}
+
 	lseg = pnfs_find_lseg(lo, &arg, strict_iomode);
 	if (lseg) {
 		trace_pnfs_update_layout(ino, pos, count, iomode, lo, lseg,
@@ -2067,28 +2091,6 @@ lookup_again:
 		nfs4_stateid_copy(&stateid, &lo->plh_stateid);
 	}
 
-	/*
-	 * Because we free lsegs before sending LAYOUTRETURN, we need to wait
-	 * for LAYOUTRETURN even if first is true.
-	 */
-	if (test_bit(NFS_LAYOUT_RETURN, &lo->plh_flags)) {
-		spin_unlock(&ino->i_lock);
-		dprintk("%s wait for layoutreturn\n", __func__);
-		lseg = ERR_PTR(pnfs_prepare_to_retry_layoutget(lo));
-		if (!IS_ERR(lseg)) {
-			if (first)
-				pnfs_clear_first_layoutget(lo);
-			pnfs_put_layout_hdr(lo);
-			dprintk("%s retrying\n", __func__);
-			trace_pnfs_update_layout(ino, pos, count, iomode, lo,
-					lseg, PNFS_UPDATE_LAYOUT_RETRY);
-			goto lookup_again;
-		}
-		trace_pnfs_update_layout(ino, pos, count, iomode, lo, lseg,
-				PNFS_UPDATE_LAYOUT_RETURN);
-		goto out_put_layout_hdr;
-	}
-
 	if (pnfs_layoutgets_blocked(lo)) {
 		trace_pnfs_update_layout(ino, pos, count, iomode, lo, lseg,
 				PNFS_UPDATE_LAYOUT_BLOCKED);
@@ -2242,6 +2244,7 @@ static void _lgopen_prepare_attached(struct nfs4_opendata *data,
 					     &rng, GFP_KERNEL);
 	if (!lgp) {
 		pnfs_clear_first_layoutget(lo);
+		nfs_layoutget_end(lo);
 		pnfs_put_layout_hdr(lo);
 		return;
 	}
diff --git a/fs/nfs/pnfs.h b/fs/nfs/pnfs.h
index bbd3de1025f2..d810ae674f4e 100644
--- a/fs/nfs/pnfs.h
+++ b/fs/nfs/pnfs.h
@@ -297,10 +297,8 @@ bool pnfs_roc(struct inode *ino,
 		struct nfs4_layoutreturn_args *args,
 		struct nfs4_layoutreturn_res *res,
 		const struct cred *cred);
-int pnfs_roc_done(struct rpc_task *task, struct inode *inode,
-		struct nfs4_layoutreturn_args **argpp,
-		struct nfs4_layoutreturn_res **respp,
-		int *ret);
+int pnfs_roc_done(struct rpc_task *task, struct nfs4_layoutreturn_args **argpp,
+		  struct nfs4_layoutreturn_res **respp, int *ret);
 void pnfs_roc_release(struct nfs4_layoutreturn_args *args,
 		struct nfs4_layoutreturn_res *res,
 		int ret);
@@ -772,7 +770,7 @@ pnfs_roc(struct inode *ino,
 }
 
 static inline int
-pnfs_roc_done(struct rpc_task *task, struct inode *inode,
+pnfs_roc_done(struct rpc_task *task,
 		struct nfs4_layoutreturn_args **argpp,
 		struct nfs4_layoutreturn_res **respp,
 		int *ret)
diff --git a/fs/nfs/pnfs_nfs.c b/fs/nfs/pnfs_nfs.c
index 2efcfdd348a1..49d3389bd813 100644
--- a/fs/nfs/pnfs_nfs.c
+++ b/fs/nfs/pnfs_nfs.c
@@ -78,22 +78,18 @@ void
 pnfs_generic_clear_request_commit(struct nfs_page *req,
 				  struct nfs_commit_info *cinfo)
 {
-	struct pnfs_layout_segment *freeme = NULL;
+	struct pnfs_commit_bucket *bucket = NULL;
 
 	if (!test_and_clear_bit(PG_COMMIT_TO_DS, &req->wb_flags))
 		goto out;
 	cinfo->ds->nwritten--;
-	if (list_is_singular(&req->wb_list)) {
-		struct pnfs_commit_bucket *bucket;
-
+	if (list_is_singular(&req->wb_list))
 		bucket = list_first_entry(&req->wb_list,
-					  struct pnfs_commit_bucket,
-					  written);
-		freeme = pnfs_free_bucket_lseg(bucket);
-	}
+					  struct pnfs_commit_bucket, written);
 out:
 	nfs_request_remove_commit_list(req, cinfo);
-	pnfs_put_lseg(freeme);
+	if (bucket)
+		pnfs_put_lseg(pnfs_free_bucket_lseg(bucket));
 }
 EXPORT_SYMBOL_GPL(pnfs_generic_clear_request_commit);
 
@@ -407,12 +403,16 @@ pnfs_bucket_get_committing(struct list_head *head,
 			   struct pnfs_commit_bucket *bucket,
 			   struct nfs_commit_info *cinfo)
 {
+	struct pnfs_layout_segment *lseg;
 	struct list_head *pos;
 
 	list_for_each(pos, &bucket->committing)
 		cinfo->ds->ncommitting--;
 	list_splice_init(&bucket->committing, head);
-	return pnfs_free_bucket_lseg(bucket);
+	lseg = pnfs_free_bucket_lseg(bucket);
+	if (!lseg)
+		lseg = pnfs_get_lseg(bucket->lseg);
+	return lseg;
 }
 
 static struct nfs_commit_data *
@@ -424,8 +424,6 @@ pnfs_bucket_fetch_commitdata(struct pnfs_commit_bucket *bucket,
 	if (!data)
 		return NULL;
 	data->lseg = pnfs_bucket_get_committing(&data->pages, bucket, cinfo);
-	if (!data->lseg)
-		data->lseg = pnfs_get_lseg(bucket->lseg);
 	return data;
 }
 
diff --git a/net/sunrpc/addr.c b/net/sunrpc/addr.c
index 010dcb876f9d..6e4dbd577a39 100644
--- a/net/sunrpc/addr.c
+++ b/net/sunrpc/addr.c
@@ -185,7 +185,7 @@ static int rpc_parse_scope_id(struct net *net, const char *buf,
 			scope_id = dev->ifindex;
 			dev_put(dev);
 		} else {
-			if (kstrtou32(p, 10, &scope_id) == 0) {
+			if (kstrtou32(p, 10, &scope_id) != 0) {
 				kfree(p);
 				return 0;
 			}