NFSv4: Fix a pNFS layout related use-after-free race when freeing the inode

When returning the layout in nfs4_evict_inode(), we need to ensure that
the layout is actually done being freed before we can proceed to free the
inode itself.

Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>

3 files changed, 37 insertions, 3 deletions

diff --git a/fs/nfs/nfs4super.c b/fs/nfs/nfs4super.c
index 93f5c1678ec2..984cc42ee54d 100644
--- a/fs/nfs/nfs4super.c
+++ b/fs/nfs/nfs4super.c
@@ -67,7 +67,7 @@ static void nfs4_evict_inode(struct inode *inode)
 	nfs_inode_evict_delegation(inode);
 	/* Note that above delegreturn would trigger pnfs return-on-close */
 	pnfs_return_layout(inode);
-	pnfs_destroy_layout(NFS_I(inode));
+	pnfs_destroy_layout_final(NFS_I(inode));
 	/* First call standard NFS clear_inode() code */
 	nfs_clear_inode(inode);
 	nfs4_xattr_cache_zap(inode);
diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c
index 0e50b9d45c32..07f59dc8cb2e 100644
--- a/fs/nfs/pnfs.c
+++ b/fs/nfs/pnfs.c
@@ -294,6 +294,7 @@ void
 pnfs_put_layout_hdr(struct pnfs_layout_hdr *lo)
 {
 	struct inode *inode;
+	unsigned long i_state;
 
 	if (!lo)
 		return;
@@ -304,8 +305,12 @@ pnfs_put_layout_hdr(struct pnfs_layout_hdr *lo)
 		if (!list_empty(&lo->plh_segs))
 			WARN_ONCE(1, "NFS: BUG unfreed layout segments.\n");
 		pnfs_detach_layout_hdr(lo);
+		i_state = inode->i_state;
 		spin_unlock(&inode->i_lock);
 		pnfs_free_layout_hdr(lo);
+		/* Notify pnfs_destroy_layout_final() that we're done */
+		if (i_state & (I_FREEING | I_CLEAR))
+			wake_up_var(lo);
 	}
 }
 
@@ -734,8 +739,7 @@ pnfs_free_lseg_list(struct list_head *free_me)
 	}
 }
 
-void
-pnfs_destroy_layout(struct nfs_inode *nfsi)
+static struct pnfs_layout_hdr *__pnfs_destroy_layout(struct nfs_inode *nfsi)
 {
 	struct pnfs_layout_hdr *lo;
 	LIST_HEAD(tmp_list);
@@ -753,9 +757,34 @@ pnfs_destroy_layout(struct nfs_inode *nfsi)
 		pnfs_put_layout_hdr(lo);
 	} else
 		spin_unlock(&nfsi->vfs_inode.i_lock);
+	return lo;
+}
+
+void pnfs_destroy_layout(struct nfs_inode *nfsi)
+{
+	__pnfs_destroy_layout(nfsi);
 }
 EXPORT_SYMBOL_GPL(pnfs_destroy_layout);
 
+static bool pnfs_layout_removed(struct nfs_inode *nfsi,
+				struct pnfs_layout_hdr *lo)
+{
+	bool ret;
+
+	spin_lock(&nfsi->vfs_inode.i_lock);
+	ret = nfsi->layout != lo;
+	spin_unlock(&nfsi->vfs_inode.i_lock);
+	return ret;
+}
+
+void pnfs_destroy_layout_final(struct nfs_inode *nfsi)
+{
+	struct pnfs_layout_hdr *lo = __pnfs_destroy_layout(nfsi);
+
+	if (lo)
+		wait_var_event(lo, pnfs_layout_removed(nfsi, lo));
+}
+
 static bool
 pnfs_layout_add_bulk_destroy_list(struct inode *inode,
 		struct list_head *layout_list)
diff --git a/fs/nfs/pnfs.h b/fs/nfs/pnfs.h
index f618c49697bb..bbd3de1025f2 100644
--- a/fs/nfs/pnfs.h
+++ b/fs/nfs/pnfs.h
@@ -268,6 +268,7 @@ struct pnfs_layout_segment *pnfs_layout_process(struct nfs4_layoutget *lgp);
 void pnfs_layoutget_free(struct nfs4_layoutget *lgp);
 void pnfs_free_lseg_list(struct list_head *tmp_list);
 void pnfs_destroy_layout(struct nfs_inode *);
+void pnfs_destroy_layout_final(struct nfs_inode *);
 void pnfs_destroy_all_layouts(struct nfs_client *);
 int pnfs_destroy_layouts_byfsid(struct nfs_client *clp,
 		struct nfs_fsid *fsid,
@@ -712,6 +713,10 @@ static inline void pnfs_destroy_layout(struct nfs_inode *nfsi)
 {
 }
 
+static inline void pnfs_destroy_layout_final(struct nfs_inode *nfsi)
+{
+}
+
 static inline struct pnfs_layout_segment *
 pnfs_get_lseg(struct pnfs_layout_segment *lseg)
 {