Merge tag 'nfs-for-4.5-4' of git://git.linux-nfs.org/projects/trondmy/linux-nfs

Pull NFS client bugfixes from Trond Myklebust:
 "Stable bugfixes:
   - Fix nfs_size_to_loff_t
   - NFSv4: Fix a dentry leak on alias use

  Other bugfixes:
   - Don't schedule a layoutreturn if the layout segment can be freed
     immediately.
   - Always set NFS_LAYOUT_RETURN_REQUESTED with lo->plh_return_iomode
   - rpcrdma_bc_receive_call() should init rq_private_buf.len
   - fix stateid handling for the NFS v4.2 operations
   - pnfs/blocklayout: fix a memeory leak when using,vmalloc_to_page
   - fix panic in gss_pipe_downcall() in fips mode
   - Fix a race between layoutget and pnfs_destroy_layout
   - Fix a race between layoutget and bulk recalls"

* tag 'nfs-for-4.5-4' of git://git.linux-nfs.org/projects/trondmy/linux-nfs:
  NFSv4.x/pnfs: Fix a race between layoutget and bulk recalls
  NFSv4.x/pnfs: Fix a race between layoutget and pnfs_destroy_layout
  auth_gss: fix panic in gss_pipe_downcall() in fips mode
  pnfs/blocklayout: fix a memeory leak when using,vmalloc_to_page
  nfs4: fix stateid handling for the NFS v4.2 operations
  NFSv4: Fix a dentry leak on alias use
  xprtrdma: rpcrdma_bc_receive_call() should init rq_private_buf.len
  pNFS: Always set NFS_LAYOUT_RETURN_REQUESTED with lo->plh_return_iomode
  pNFS: Fix pnfs_mark_matching_lsegs_return()
  nfs: fix nfs_size_to_loff_t

4 files changed, 126 insertions, 62 deletions

diff --git a/fs/nfs/blocklayout/extent_tree.c b/fs/nfs/blocklayout/extent_tree.c
index c59a59c37f3d..35ab51c04814 100644
--- a/fs/nfs/blocklayout/extent_tree.c
+++ b/fs/nfs/blocklayout/extent_tree.c
@@ -476,6 +476,7 @@ static void ext_tree_free_commitdata(struct nfs4_layoutcommit_args *arg,
 
 		for (i = 0; i < nr_pages; i++)
 			put_page(arg->layoutupdate_pages[i]);
+		vfree(arg->start_p);
 		kfree(arg->layoutupdate_pages);
 	} else {
 		put_page(arg->layoutupdate_page);
@@ -559,10 +560,15 @@ retry:
 
 	if (unlikely(arg->layoutupdate_pages != &arg->layoutupdate_page)) {
 		void *p = start_p, *end = p + arg->layoutupdate_len;
+		struct page *page = NULL;
 		int i = 0;
 
-		for ( ; p < end; p += PAGE_SIZE)
-			arg->layoutupdate_pages[i++] = vmalloc_to_page(p);
+		arg->start_p = start_p;
+		for ( ; p < end; p += PAGE_SIZE) {
+			page = vmalloc_to_page(p);
+			arg->layoutupdate_pages[i++] = page;
+			get_page(page);
+		}
 	}
 
 	dprintk("%s found %zu ranges\n", __func__, count);
diff --git a/fs/nfs/nfs42proc.c b/fs/nfs/nfs42proc.c
index bd25dc7077f7..dff83460e5a6 100644
--- a/fs/nfs/nfs42proc.c
+++ b/fs/nfs/nfs42proc.c
@@ -16,29 +16,8 @@
 
 #define NFSDBG_FACILITY NFSDBG_PROC
 
-static int nfs42_set_rw_stateid(nfs4_stateid *dst, struct file *file,
-				fmode_t fmode)
-{
-	struct nfs_open_context *open;
-	struct nfs_lock_context *lock;
-	int ret;
-
-	open = get_nfs_open_context(nfs_file_open_context(file));
-	lock = nfs_get_lock_context(open);
-	if (IS_ERR(lock)) {
-		put_nfs_open_context(open);
-		return PTR_ERR(lock);
-	}
-
-	ret = nfs4_set_rw_stateid(dst, open, lock, fmode);
-
-	nfs_put_lock_context(lock);
-	put_nfs_open_context(open);
-	return ret;
-}
-
 static int _nfs42_proc_fallocate(struct rpc_message *msg, struct file *filep,
-				 loff_t offset, loff_t len)
+		struct nfs_lock_context *lock, loff_t offset, loff_t len)
 {
 	struct inode *inode = file_inode(filep);
 	struct nfs_server *server = NFS_SERVER(inode);
@@ -56,7 +35,8 @@ static int _nfs42_proc_fallocate(struct rpc_message *msg, struct file *filep,
 	msg->rpc_argp = &args;
 	msg->rpc_resp = &res;
 
-	status = nfs42_set_rw_stateid(&args.falloc_stateid, filep, FMODE_WRITE);
+	status = nfs4_set_rw_stateid(&args.falloc_stateid, lock->open_context,
+			lock, FMODE_WRITE);
 	if (status)
 		return status;
 
@@ -78,15 +58,26 @@ static int nfs42_proc_fallocate(struct rpc_message *msg, struct file *filep,
 {
 	struct nfs_server *server = NFS_SERVER(file_inode(filep));
 	struct nfs4_exception exception = { };
+	struct nfs_lock_context *lock;
 	int err;
 
+	lock = nfs_get_lock_context(nfs_file_open_context(filep));
+	if (IS_ERR(lock))
+		return PTR_ERR(lock);
+
+	exception.inode = file_inode(filep);
+	exception.state = lock->open_context->state;
+
 	do {
-		err = _nfs42_proc_fallocate(msg, filep, offset, len);
-		if (err == -ENOTSUPP)
-			return -EOPNOTSUPP;
+		err = _nfs42_proc_fallocate(msg, filep, lock, offset, len);
+		if (err == -ENOTSUPP) {
+			err = -EOPNOTSUPP;
+			break;
+		}
 		err = nfs4_handle_exception(server, err, &exception);
 	} while (exception.retry);
 
+	nfs_put_lock_context(lock);
 	return err;
 }
 
@@ -135,7 +126,8 @@ int nfs42_proc_deallocate(struct file *filep, loff_t offset, loff_t len)
 	return err;
 }
 
-static loff_t _nfs42_proc_llseek(struct file *filep, loff_t offset, int whence)
+static loff_t _nfs42_proc_llseek(struct file *filep,
+		struct nfs_lock_context *lock, loff_t offset, int whence)
 {
 	struct inode *inode = file_inode(filep);
 	struct nfs42_seek_args args = {
@@ -156,7 +148,8 @@ static loff_t _nfs42_proc_llseek(struct file *filep, loff_t offset, int whence)
 	if (!nfs_server_capable(inode, NFS_CAP_SEEK))
 		return -ENOTSUPP;
 
-	status = nfs42_set_rw_stateid(&args.sa_stateid, filep, FMODE_READ);
+	status = nfs4_set_rw_stateid(&args.sa_stateid, lock->open_context,
+			lock, FMODE_READ);
 	if (status)
 		return status;
 
@@ -175,17 +168,28 @@ loff_t nfs42_proc_llseek(struct file *filep, loff_t offset, int whence)
 {
 	struct nfs_server *server = NFS_SERVER(file_inode(filep));
 	struct nfs4_exception exception = { };
+	struct nfs_lock_context *lock;
 	loff_t err;
 
+	lock = nfs_get_lock_context(nfs_file_open_context(filep));
+	if (IS_ERR(lock))
+		return PTR_ERR(lock);
+
+	exception.inode = file_inode(filep);
+	exception.state = lock->open_context->state;
+
 	do {
-		err = _nfs42_proc_llseek(filep, offset, whence);
+		err = _nfs42_proc_llseek(filep, lock, offset, whence);
 		if (err >= 0)
 			break;
-		if (err == -ENOTSUPP)
-			return -EOPNOTSUPP;
+		if (err == -ENOTSUPP) {
+			err = -EOPNOTSUPP;
+			break;
+		}
 		err = nfs4_handle_exception(server, err, &exception);
 	} while (exception.retry);
 
+	nfs_put_lock_context(lock);
 	return err;
 }
 
@@ -298,8 +302,9 @@ int nfs42_proc_layoutstats_generic(struct nfs_server *server,
 }
 
 static int _nfs42_proc_clone(struct rpc_message *msg, struct file *src_f,
-			     struct file *dst_f, loff_t src_offset,
-			     loff_t dst_offset, loff_t count)
+		struct file *dst_f, struct nfs_lock_context *src_lock,
+		struct nfs_lock_context *dst_lock, loff_t src_offset,
+		loff_t dst_offset, loff_t count)
 {
 	struct inode *src_inode = file_inode(src_f);
 	struct inode *dst_inode = file_inode(dst_f);
@@ -320,11 +325,13 @@ static int _nfs42_proc_clone(struct rpc_message *msg, struct file *src_f,
 	msg->rpc_argp = &args;
 	msg->rpc_resp = &res;
 
-	status = nfs42_set_rw_stateid(&args.src_stateid, src_f, FMODE_READ);
+	status = nfs4_set_rw_stateid(&args.src_stateid, src_lock->open_context,
+			src_lock, FMODE_READ);
 	if (status)
 		return status;
 
-	status = nfs42_set_rw_stateid(&args.dst_stateid, dst_f, FMODE_WRITE);
+	status = nfs4_set_rw_stateid(&args.dst_stateid, dst_lock->open_context,
+			dst_lock, FMODE_WRITE);
 	if (status)
 		return status;
 
@@ -349,22 +356,48 @@ int nfs42_proc_clone(struct file *src_f, struct file *dst_f,
 	};
 	struct inode *inode = file_inode(src_f);
 	struct nfs_server *server = NFS_SERVER(file_inode(src_f));
-	struct nfs4_exception exception = { };
-	int err;
+	struct nfs_lock_context *src_lock;
+	struct nfs_lock_context *dst_lock;
+	struct nfs4_exception src_exception = { };
+	struct nfs4_exception dst_exception = { };
+	int err, err2;
 
 	if (!nfs_server_capable(inode, NFS_CAP_CLONE))
 		return -EOPNOTSUPP;
 
+	src_lock = nfs_get_lock_context(nfs_file_open_context(src_f));
+	if (IS_ERR(src_lock))
+		return PTR_ERR(src_lock);
+
+	src_exception.inode = file_inode(src_f);
+	src_exception.state = src_lock->open_context->state;
+
+	dst_lock = nfs_get_lock_context(nfs_file_open_context(dst_f));
+	if (IS_ERR(dst_lock)) {
+		err = PTR_ERR(dst_lock);
+		goto out_put_src_lock;
+	}
+
+	dst_exception.inode = file_inode(dst_f);
+	dst_exception.state = dst_lock->open_context->state;
+
 	do {
-		err = _nfs42_proc_clone(&msg, src_f, dst_f, src_offset,
-					dst_offset, count);
+		err = _nfs42_proc_clone(&msg, src_f, dst_f, src_lock, dst_lock,
+					src_offset, dst_offset, count);
 		if (err == -ENOTSUPP || err == -EOPNOTSUPP) {
 			NFS_SERVER(inode)->caps &= ~NFS_CAP_CLONE;
-			return -EOPNOTSUPP;
+			err = -EOPNOTSUPP;
+			break;
 		}
-		err = nfs4_handle_exception(server, err, &exception);
-	} while (exception.retry);
 
-	return err;
+		err2 = nfs4_handle_exception(server, err, &src_exception);
+		err = nfs4_handle_exception(server, err, &dst_exception);
+		if (!err)
+			err = err2;
+	} while (src_exception.retry || dst_exception.retry);
 
+	nfs_put_lock_context(dst_lock);
+out_put_src_lock:
+	nfs_put_lock_context(src_lock);
+	return err;
 }
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 4bfc33ad0563..14881594dd07 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -2466,9 +2466,9 @@ static int _nfs4_open_and_get_state(struct nfs4_opendata *opendata,
 		dentry = d_add_unique(dentry, igrab(state->inode));
 		if (dentry == NULL) {
 			dentry = opendata->dentry;
-		} else if (dentry != ctx->dentry) {
+		} else {
 			dput(ctx->dentry);
-			ctx->dentry = dget(dentry);
+			ctx->dentry = dentry;
 		}
 		nfs_set_verifier(dentry,
 				nfs_save_change_attribute(d_inode(opendata->dir)));
diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c
index 482b6e94bb37..2fa483e6dbe2 100644
--- a/fs/nfs/pnfs.c
+++ b/fs/nfs/pnfs.c
@@ -252,6 +252,27 @@ pnfs_put_layout_hdr(struct pnfs_layout_hdr *lo)
 	}
 }
 
+/*
+ * Mark a pnfs_layout_hdr and all associated layout segments as invalid
+ *
+ * In order to continue using the pnfs_layout_hdr, a full recovery
+ * is required.
+ * Note that caller must hold inode->i_lock.
+ */
+static int
+pnfs_mark_layout_stateid_invalid(struct pnfs_layout_hdr *lo,
+		struct list_head *lseg_list)
+{
+	struct pnfs_layout_range range = {
+		.iomode = IOMODE_ANY,
+		.offset = 0,
+		.length = NFS4_MAX_UINT64,
+	};
+
+	set_bit(NFS_LAYOUT_INVALID_STID, &lo->plh_flags);
+	return pnfs_mark_matching_lsegs_invalid(lo, lseg_list, &range);
+}
+
 static int
 pnfs_iomode_to_fail_bit(u32 iomode)
 {
@@ -554,9 +575,8 @@ pnfs_destroy_layout(struct nfs_inode *nfsi)
 	spin_lock(&nfsi->vfs_inode.i_lock);
 	lo = nfsi->layout;
 	if (lo) {
-		lo->plh_block_lgets++; /* permanently block new LAYOUTGETs */
-		pnfs_mark_matching_lsegs_invalid(lo, &tmp_list, NULL);
 		pnfs_get_layout_hdr(lo);
+		pnfs_mark_layout_stateid_invalid(lo, &tmp_list);
 		pnfs_layout_clear_fail_bit(lo, NFS_LAYOUT_RO_FAILED);
 		pnfs_layout_clear_fail_bit(lo, NFS_LAYOUT_RW_FAILED);
 		spin_unlock(&nfsi->vfs_inode.i_lock);
@@ -617,11 +637,6 @@ pnfs_layout_free_bulk_destroy_list(struct list_head *layout_list,
 {
 	struct pnfs_layout_hdr *lo;
 	struct inode *inode;
-	struct pnfs_layout_range range = {
-		.iomode = IOMODE_ANY,
-		.offset = 0,
-		.length = NFS4_MAX_UINT64,
-	};
 	LIST_HEAD(lseg_list);
 	int ret = 0;
 
@@ -636,11 +651,11 @@ pnfs_layout_free_bulk_destroy_list(struct list_head *layout_list,
 
 		spin_lock(&inode->i_lock);
 		list_del_init(&lo->plh_bulk_destroy);
-		lo->plh_block_lgets++; /* permanently block new LAYOUTGETs */
-		if (is_bulk_recall)
-			set_bit(NFS_LAYOUT_BULK_RECALL, &lo->plh_flags);
-		if (pnfs_mark_matching_lsegs_invalid(lo, &lseg_list, &range))
+		if (pnfs_mark_layout_stateid_invalid(lo, &lseg_list)) {
+			if (is_bulk_recall)
+				set_bit(NFS_LAYOUT_BULK_RECALL, &lo->plh_flags);
 			ret = -EAGAIN;
+		}
 		spin_unlock(&inode->i_lock);
 		pnfs_free_lseg_list(&lseg_list);
 		/* Free all lsegs that are attached to commit buckets */
@@ -1738,8 +1753,19 @@ pnfs_set_plh_return_iomode(struct pnfs_layout_hdr *lo, enum pnfs_iomode iomode)
 	if (lo->plh_return_iomode != 0)
 		iomode = IOMODE_ANY;
 	lo->plh_return_iomode = iomode;
+	set_bit(NFS_LAYOUT_RETURN_REQUESTED, &lo->plh_flags);
 }
 
+/**
+ * pnfs_mark_matching_lsegs_return - Free or return matching layout segments
+ * @lo: pointer to layout header
+ * @tmp_list: list header to be used with pnfs_free_lseg_list()
+ * @return_range: describe layout segment ranges to be returned
+ *
+ * This function is mainly intended for use by layoutrecall. It attempts
+ * to free the layout segment immediately, or else to mark it for return
+ * as soon as its reference count drops to zero.
+ */
 int
 pnfs_mark_matching_lsegs_return(struct pnfs_layout_hdr *lo,
 				struct list_head *tmp_list,
@@ -1762,12 +1788,11 @@ pnfs_mark_matching_lsegs_return(struct pnfs_layout_hdr *lo,
 				lseg, lseg->pls_range.iomode,
 				lseg->pls_range.offset,
 				lseg->pls_range.length);
+			if (mark_lseg_invalid(lseg, tmp_list))
+				continue;
+			remaining++;
 			set_bit(NFS_LSEG_LAYOUTRETURN, &lseg->pls_flags);
 			pnfs_set_plh_return_iomode(lo, return_range->iomode);
-			if (!mark_lseg_invalid(lseg, tmp_list))
-				remaining++;
-			set_bit(NFS_LAYOUT_RETURN_REQUESTED,
-					&lo->plh_flags);
 		}
 	return remaining;
 }