Merge tag 'nfs-for-6.6-4' of git://git.linux-nfs.org/projects/anna/linux-nfs

Pull NFS client fixes from Anna Schumaker:
 "Stable Fix:
   - Fix a pNFS hang in nfs4_evict_inode()

  Fixes:
   - Force update of suid/sgid bits after an NFS v4.2 ALLOCATE op
   - Fix a potential oops in nfs_inode_remove_request()
   - Check the validity of the layout pointer in ff_layout_mirror_prepare_stats()
   - Fix incorrectly marking the pNFS MDS with USE_PNFS_DS in some cases"

* tag 'nfs-for-6.6-4' of git://git.linux-nfs.org/projects/anna/linux-nfs:
  NFSv4.1: fixup use EXCHGID4_FLAG_USE_PNFS_DS for DS server
  pNFS/flexfiles: Check the layout validity in ff_layout_mirror_prepare_stats
  pNFS: Fix a hang in nfs4_evict_inode()
  NFS: Fix potential oops in nfs_inode_remove_request()
  nfs42: client needs to strip file mode's suid/sgid bit after ALLOCATE op

5 files changed, 38 insertions, 21 deletions

diff --git a/fs/nfs/flexfilelayout/flexfilelayout.c b/fs/nfs/flexfilelayout/flexfilelayout.c
index a1dc33864906..ef817a0475ff 100644
--- a/fs/nfs/flexfilelayout/flexfilelayout.c
+++ b/fs/nfs/flexfilelayout/flexfilelayout.c
@@ -2520,9 +2520,9 @@ ff_layout_mirror_prepare_stats(struct pnfs_layout_hdr *lo,
 	return i;
 }
 
-static int
-ff_layout_prepare_layoutstats(struct nfs42_layoutstat_args *args)
+static int ff_layout_prepare_layoutstats(struct nfs42_layoutstat_args *args)
 {
+	struct pnfs_layout_hdr *lo;
 	struct nfs4_flexfile_layout *ff_layout;
 	const int dev_count = PNFS_LAYOUTSTATS_MAXDEV;
 
@@ -2533,11 +2533,14 @@ ff_layout_prepare_layoutstats(struct nfs42_layoutstat_args *args)
 		return -ENOMEM;
 
 	spin_lock(&args->inode->i_lock);
-	ff_layout = FF_LAYOUT_FROM_HDR(NFS_I(args->inode)->layout);
-	args->num_dev = ff_layout_mirror_prepare_stats(&ff_layout->generic_hdr,
-						       &args->devinfo[0],
-						       dev_count,
-						       NFS4_FF_OP_LAYOUTSTATS);
+	lo = NFS_I(args->inode)->layout;
+	if (lo && pnfs_layout_is_valid(lo)) {
+		ff_layout = FF_LAYOUT_FROM_HDR(lo);
+		args->num_dev = ff_layout_mirror_prepare_stats(
+			&ff_layout->generic_hdr, &args->devinfo[0], dev_count,
+			NFS4_FF_OP_LAYOUTSTATS);
+	} else
+		args->num_dev = 0;
 	spin_unlock(&args->inode->i_lock);
 	if (!args->num_dev) {
 		kfree(args->devinfo);
diff --git a/fs/nfs/nfs42proc.c b/fs/nfs/nfs42proc.c
index 063e00aff87e..28704f924612 100644
--- a/fs/nfs/nfs42proc.c
+++ b/fs/nfs/nfs42proc.c
@@ -81,7 +81,8 @@ static int _nfs42_proc_fallocate(struct rpc_message *msg, struct file *filep,
 	if (status == 0) {
 		if (nfs_should_remove_suid(inode)) {
 			spin_lock(&inode->i_lock);
-			nfs_set_cache_invalid(inode, NFS_INO_INVALID_MODE);
+			nfs_set_cache_invalid(inode,
+				NFS_INO_REVAL_FORCED | NFS_INO_INVALID_MODE);
 			spin_unlock(&inode->i_lock);
 		}
 		status = nfs_post_op_update_inode_force_wcc(inode,
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 7016eaadf555..5ee283eb9660 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -8870,8 +8870,6 @@ static int _nfs4_proc_exchange_id(struct nfs_client *clp, const struct cred *cre
 	/* Save the EXCHANGE_ID verifier session trunk tests */
 	memcpy(clp->cl_confirm.data, argp->verifier.data,
 	       sizeof(clp->cl_confirm.data));
-	if (resp->flags & EXCHGID4_FLAG_USE_PNFS_DS)
-		set_bit(NFS_CS_DS, &clp->cl_flags);
 out:
 	trace_nfs4_exchange_id(clp, status);
 	rpc_put_task(task);
diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c
index 306cba0b9e69..84343aefbbd6 100644
--- a/fs/nfs/pnfs.c
+++ b/fs/nfs/pnfs.c
@@ -2634,31 +2634,44 @@ pnfs_should_return_unused_layout(struct pnfs_layout_hdr *lo,
 	return mode == 0;
 }
 
-static int
-pnfs_layout_return_unused_byserver(struct nfs_server *server, void *data)
+static int pnfs_layout_return_unused_byserver(struct nfs_server *server,
+					      void *data)
 {
 	const struct pnfs_layout_range *range = data;
+	const struct cred *cred;
 	struct pnfs_layout_hdr *lo;
 	struct inode *inode;
+	nfs4_stateid stateid;
+	enum pnfs_iomode iomode;
+
 restart:
 	rcu_read_lock();
 	list_for_each_entry_rcu(lo, &server->layouts, plh_layouts) {
-		if (!pnfs_layout_can_be_returned(lo) ||
+		inode = lo->plh_inode;
+		if (!inode || !pnfs_layout_can_be_returned(lo) ||
 		    test_bit(NFS_LAYOUT_RETURN_REQUESTED, &lo->plh_flags))
 			continue;
-		inode = lo->plh_inode;
 		spin_lock(&inode->i_lock);
-		if (!pnfs_should_return_unused_layout(lo, range)) {
+		if (!lo->plh_inode ||
+		    !pnfs_should_return_unused_layout(lo, range)) {
 			spin_unlock(&inode->i_lock);
 			continue;
 		}
+		pnfs_get_layout_hdr(lo);
+		pnfs_set_plh_return_info(lo, range->iomode, 0);
+		if (pnfs_mark_matching_lsegs_return(lo, &lo->plh_return_segs,
+						    range, 0) != 0 ||
+		    !pnfs_prepare_layoutreturn(lo, &stateid, &cred, &iomode)) {
+			spin_unlock(&inode->i_lock);
+			rcu_read_unlock();
+			pnfs_put_layout_hdr(lo);
+			cond_resched();
+			goto restart;
+		}
 		spin_unlock(&inode->i_lock);
-		inode = pnfs_grab_inode_layout_hdr(lo);
-		if (!inode)
-			continue;
 		rcu_read_unlock();
-		pnfs_mark_layout_for_return(inode, range);
-		iput(inode);
+		pnfs_send_layoutreturn(lo, &stateid, &cred, iomode, false);
+		pnfs_put_layout_hdr(lo);
 		cond_resched();
 		goto restart;
 	}
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index 7720b5e43014..9d82d50ce0b1 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -788,6 +788,8 @@ static void nfs_inode_add_request(struct nfs_page *req)
  */
 static void nfs_inode_remove_request(struct nfs_page *req)
 {
+	struct nfs_inode *nfsi = NFS_I(nfs_page_to_inode(req));
+
 	if (nfs_page_group_sync_on_bit(req, PG_REMOVE)) {
 		struct folio *folio = nfs_page_to_folio(req->wb_head);
 		struct address_space *mapping = folio_file_mapping(folio);
@@ -802,7 +804,7 @@ static void nfs_inode_remove_request(struct nfs_page *req)
 	}
 
 	if (test_and_clear_bit(PG_INODE_REF, &req->wb_flags)) {
-		atomic_long_dec(&NFS_I(nfs_page_to_inode(req))->nrequests);
+		atomic_long_dec(&nfsi->nrequests);
 		nfs_release_request(req);
 	}
 }