Merge tag 'nfs-for-5.7-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs

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

  Stable fixes:
   - Fix a page leak in nfs_destroy_unlinked_subrequests()

   - Fix use-after-free issues in nfs_pageio_add_request()

   - Fix new mount code constant_table array definitions

   - finish_automount() requires us to hold 2 refs to the mount record

  Features:
   - Improve the accuracy of telldir/seekdir by using 64-bit cookies
     when possible.

   - Allow one RDMA active connection and several zombie connections to
     prevent blocking if the remote server is unresponsive.

   - Limit the size of the NFS access cache by default

   - Reduce the number of references to credentials that are taken by
     NFS

   - pNFS files and flexfiles drivers now support per-layout segment
     COMMIT lists.

   - Enable partial-file layout segments in the pNFS/flexfiles driver.

   - Add support for CB_RECALL_ANY to the pNFS flexfiles layout type

   - pNFS/flexfiles Report NFS4ERR_DELAY and NFS4ERR_GRACE errors from
     the DS using the layouterror mechanism.

  Bugfixes and cleanups:
   - SUNRPC: Fix krb5p regressions

   - Don't specify NFS version in "UDP not supported" error

   - nfsroot: set tcp as the default transport protocol

   - pnfs: Return valid stateids in nfs_layout_find_inode_by_stateid()

   - alloc_nfs_open_context() must use the file cred when available

   - Fix locking when dereferencing the delegation cred

   - Fix memory leaks in O_DIRECT when nfs_get_lock_context() fails

   - Various clean ups of the NFS O_DIRECT commit code

   - Clean up RDMA connect/disconnect

   - Replace zero-length arrays with C99-style flexible arrays"

* tag 'nfs-for-5.7-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs: (86 commits)
  NFS: Clean up process of marking inode stale.
  SUNRPC: Don't start a timer on an already queued rpc task
  NFS/pnfs: Reference the layout cred in pnfs_prepare_layoutreturn()
  NFS/pnfs: Fix dereference of layout cred in pnfs_layoutcommit_inode()
  NFS: Beware when dereferencing the delegation cred
  NFS: Add a module parameter to set nfs_mountpoint_expiry_timeout
  NFS: finish_automount() requires us to hold 2 refs to the mount record
  NFS: Fix a few constant_table array definitions
  NFS: Try to join page groups before an O_DIRECT retransmission
  NFS: Refactor nfs_lock_and_join_requests()
  NFS: Reverse the submission order of requests in __nfs_pageio_add_request()
  NFS: Clean up nfs_lock_and_join_requests()
  NFS: Remove the redundant function nfs_pgio_has_mirroring()
  NFS: Fix memory leaks in nfs_pageio_stop_mirroring()
  NFS: Fix a request reference leak in nfs_direct_write_clear_reqs()
  NFS: Fix use-after-free issues in nfs_pageio_add_request()
  NFS: Fix races nfs_page_group_destroy() vs nfs_destroy_unlinked_subrequests()
  NFS: Fix a page leak in nfs_destroy_unlinked_subrequests()
  NFS: Remove unused FLUSH_SYNC support in nfs_initiate_pgio()
  pNFS/flexfiles: Specify the layout segment range in LAYOUTGET
  ...

1 files changed, 111 insertions, 32 deletions

diff --git a/fs/nfs/pnfs.h b/fs/nfs/pnfs.h
index 0fafdadc9c8d..8e0ada581b92 100644
--- a/fs/nfs/pnfs.h
+++ b/fs/nfs/pnfs.h
@@ -66,6 +66,7 @@ struct nfs4_pnfs_ds {
 struct pnfs_layout_segment {
 	struct list_head pls_list;
 	struct list_head pls_lc_list;
+	struct list_head pls_commits;
 	struct pnfs_layout_range pls_range;
 	refcount_t pls_refcount;
 	u32 pls_seq;
@@ -105,6 +106,7 @@ enum {
 	NFS_LAYOUT_INVALID_STID,	/* layout stateid id is invalid */
 	NFS_LAYOUT_FIRST_LAYOUTGET,	/* Serialize first layoutget */
 	NFS_LAYOUT_INODE_FREEING,	/* The inode is being freed */
+	NFS_LAYOUT_HASHED,		/* The layout visible */
 };
 
 enum layoutdriver_policy_flags {
@@ -148,22 +150,6 @@ struct pnfs_layoutdriver_type {
 	const struct nfs_pageio_ops *pg_write_ops;
 
 	struct pnfs_ds_commit_info *(*get_ds_info) (struct inode *inode);
-	void (*mark_request_commit) (struct nfs_page *req,
-				     struct pnfs_layout_segment *lseg,
-				     struct nfs_commit_info *cinfo,
-				     u32 ds_commit_idx);
-	void (*clear_request_commit) (struct nfs_page *req,
-				      struct nfs_commit_info *cinfo);
-	int (*scan_commit_lists) (struct nfs_commit_info *cinfo,
-				  int max);
-	void (*recover_commit_reqs) (struct list_head *list,
-				     struct nfs_commit_info *cinfo);
-	struct nfs_page * (*search_commit_reqs)(struct nfs_commit_info *cinfo,
-						struct page *page);
-	int (*commit_pagelist)(struct inode *inode,
-			       struct list_head *mds_pages,
-			       int how,
-			       struct nfs_commit_info *cinfo);
 
 	int (*sync)(struct inode *inode, bool datasync);
 
@@ -186,6 +172,29 @@ struct pnfs_layoutdriver_type {
 	int (*prepare_layoutstats) (struct nfs42_layoutstat_args *args);
 };
 
+struct pnfs_commit_ops {
+	void (*setup_ds_info)(struct pnfs_ds_commit_info *,
+			      struct pnfs_layout_segment *);
+	void (*release_ds_info)(struct pnfs_ds_commit_info *,
+				struct inode *inode);
+	int (*commit_pagelist)(struct inode *inode,
+			       struct list_head *mds_pages,
+			       int how,
+			       struct nfs_commit_info *cinfo);
+	void (*mark_request_commit) (struct nfs_page *req,
+				     struct pnfs_layout_segment *lseg,
+				     struct nfs_commit_info *cinfo,
+				     u32 ds_commit_idx);
+	void (*clear_request_commit) (struct nfs_page *req,
+				      struct nfs_commit_info *cinfo);
+	int (*scan_commit_lists) (struct nfs_commit_info *cinfo,
+				  int max);
+	void (*recover_commit_reqs) (struct list_head *list,
+				     struct nfs_commit_info *cinfo);
+	struct nfs_page * (*search_commit_reqs)(struct nfs_commit_info *cinfo,
+						struct page *page);
+};
+
 struct pnfs_layout_hdr {
 	refcount_t		plh_refcount;
 	atomic_t		plh_outstanding; /* number of RPCs out */
@@ -203,6 +212,7 @@ struct pnfs_layout_hdr {
 	loff_t			plh_lwb; /* last write byte for layoutcommit */
 	const struct cred	*plh_lc_cred; /* layoutcommit cred */
 	struct inode		*plh_inode;
+	struct rcu_head		plh_rcu;
 };
 
 struct pnfs_device {
@@ -242,6 +252,7 @@ void pnfs_put_lseg(struct pnfs_layout_segment *lseg);
 void set_pnfs_layoutdriver(struct nfs_server *, const struct nfs_fh *, struct nfs_fsinfo *);
 void unset_pnfs_layoutdriver(struct nfs_server *);
 void pnfs_generic_pg_check_layout(struct nfs_pageio_descriptor *pgio);
+void pnfs_generic_pg_check_range(struct nfs_pageio_descriptor *pgio, struct nfs_page *req);
 void pnfs_generic_pg_init_read(struct nfs_pageio_descriptor *, struct nfs_page *);
 int pnfs_generic_pg_readpages(struct nfs_pageio_descriptor *desc);
 void pnfs_generic_pg_init_write(struct nfs_pageio_descriptor *pgio,
@@ -267,6 +278,7 @@ bool nfs4_layout_refresh_old_stateid(nfs4_stateid *dst,
 void pnfs_put_layout_hdr(struct pnfs_layout_hdr *lo);
 void pnfs_set_layout_stateid(struct pnfs_layout_hdr *lo,
 			     const nfs4_stateid *new,
+			     const struct cred *cred,
 			     bool update_barrier);
 int pnfs_mark_matching_lsegs_invalid(struct pnfs_layout_hdr *lo,
 				struct list_head *tmp_list,
@@ -326,6 +338,9 @@ int pnfs_write_done_resend_to_mds(struct nfs_pgio_header *);
 struct nfs4_threshold *pnfs_mdsthreshold_alloc(void);
 void pnfs_error_mark_layout_for_return(struct inode *inode,
 				       struct pnfs_layout_segment *lseg);
+void pnfs_layout_return_unused_byclid(struct nfs_client *clp,
+				      enum pnfs_iomode iomode);
+
 /* nfs4_deviceid_flags */
 enum {
 	NFS_DEVICEID_INVALID = 0,       /* set when MDS clientid recalled */
@@ -360,6 +375,16 @@ bool nfs4_test_deviceid_unavailable(struct nfs4_deviceid_node *node);
 void nfs4_deviceid_purge_client(const struct nfs_client *);
 
 /* pnfs_nfs.c */
+struct pnfs_commit_array *pnfs_alloc_commit_array(size_t n, gfp_t gfp_flags);
+void pnfs_free_commit_array(struct pnfs_commit_array *p);
+struct pnfs_commit_array *pnfs_add_commit_array(struct pnfs_ds_commit_info *,
+						struct pnfs_commit_array *,
+						struct pnfs_layout_segment *);
+
+void pnfs_generic_ds_cinfo_release_lseg(struct pnfs_ds_commit_info *fl_cinfo,
+		struct pnfs_layout_segment *lseg);
+void pnfs_generic_ds_cinfo_destroy(struct pnfs_ds_commit_info *fl_cinfo);
+
 void pnfs_generic_clear_request_commit(struct nfs_page *req,
 				       struct nfs_commit_info *cinfo);
 void pnfs_generic_commit_release(void *calldata);
@@ -367,6 +392,8 @@ void pnfs_generic_prepare_to_resend_writes(struct nfs_commit_data *data);
 void pnfs_generic_rw_release(void *data);
 void pnfs_generic_recover_commit_reqs(struct list_head *dst,
 				      struct nfs_commit_info *cinfo);
+struct nfs_page *pnfs_generic_search_commit_reqs(struct nfs_commit_info *cinfo,
+						 struct page *page);
 int pnfs_generic_commit_pagelist(struct inode *inode,
 				 struct list_head *mds_pages,
 				 int how,
@@ -438,9 +465,11 @@ static inline int
 pnfs_commit_list(struct inode *inode, struct list_head *mds_pages, int how,
 		 struct nfs_commit_info *cinfo)
 {
-	if (cinfo->ds == NULL || cinfo->ds->ncommitting == 0)
+	struct pnfs_ds_commit_info *fl_cinfo = cinfo->ds;
+
+	if (fl_cinfo == NULL || fl_cinfo->ncommitting == 0)
 		return PNFS_NOT_ATTEMPTED;
-	return NFS_SERVER(inode)->pnfs_curr_ld->commit_pagelist(inode, mds_pages, how, cinfo);
+	return fl_cinfo->ops->commit_pagelist(inode, mds_pages, how, cinfo);
 }
 
 static inline struct pnfs_ds_commit_info *
@@ -454,6 +483,28 @@ pnfs_get_ds_info(struct inode *inode)
 }
 
 static inline void
+pnfs_init_ds_commit_info_ops(struct pnfs_ds_commit_info *fl_cinfo, struct inode *inode)
+{
+	struct pnfs_ds_commit_info *inode_cinfo = pnfs_get_ds_info(inode);
+	if (inode_cinfo != NULL)
+		fl_cinfo->ops = inode_cinfo->ops;
+}
+
+static inline void
+pnfs_init_ds_commit_info(struct pnfs_ds_commit_info *fl_cinfo)
+{
+	INIT_LIST_HEAD(&fl_cinfo->commits);
+	fl_cinfo->ops = NULL;
+}
+
+static inline void
+pnfs_release_ds_info(struct pnfs_ds_commit_info *fl_cinfo, struct inode *inode)
+{
+	if (fl_cinfo->ops != NULL && fl_cinfo->ops->release_ds_info != NULL)
+		fl_cinfo->ops->release_ds_info(fl_cinfo, inode);
+}
+
+static inline void
 pnfs_generic_mark_devid_invalid(struct nfs4_deviceid_node *node)
 {
 	set_bit(NFS_DEVICEID_INVALID, &node->flags);
@@ -463,24 +514,22 @@ static inline bool
 pnfs_mark_request_commit(struct nfs_page *req, struct pnfs_layout_segment *lseg,
 			 struct nfs_commit_info *cinfo, u32 ds_commit_idx)
 {
-	struct inode *inode = d_inode(nfs_req_openctx(req)->dentry);
-	struct pnfs_layoutdriver_type *ld = NFS_SERVER(inode)->pnfs_curr_ld;
+	struct pnfs_ds_commit_info *fl_cinfo = cinfo->ds;
 
-	if (lseg == NULL || ld->mark_request_commit == NULL)
+	if (!lseg || !fl_cinfo->ops->mark_request_commit)
 		return false;
-	ld->mark_request_commit(req, lseg, cinfo, ds_commit_idx);
+	fl_cinfo->ops->mark_request_commit(req, lseg, cinfo, ds_commit_idx);
 	return true;
 }
 
 static inline bool
 pnfs_clear_request_commit(struct nfs_page *req, struct nfs_commit_info *cinfo)
 {
-	struct inode *inode = d_inode(nfs_req_openctx(req)->dentry);
-	struct pnfs_layoutdriver_type *ld = NFS_SERVER(inode)->pnfs_curr_ld;
+	struct pnfs_ds_commit_info *fl_cinfo = cinfo->ds;
 
-	if (ld == NULL || ld->clear_request_commit == NULL)
+	if (!fl_cinfo || !fl_cinfo->ops || !fl_cinfo->ops->clear_request_commit)
 		return false;
-	ld->clear_request_commit(req, cinfo);
+	fl_cinfo->ops->clear_request_commit(req, cinfo);
 	return true;
 }
 
@@ -488,21 +537,31 @@ static inline int
 pnfs_scan_commit_lists(struct inode *inode, struct nfs_commit_info *cinfo,
 		       int max)
 {
-	if (cinfo->ds == NULL || cinfo->ds->nwritten == 0)
+	struct pnfs_ds_commit_info *fl_cinfo = cinfo->ds;
+
+	if (!fl_cinfo || fl_cinfo->nwritten == 0)
 		return 0;
-	else
-		return NFS_SERVER(inode)->pnfs_curr_ld->scan_commit_lists(cinfo, max);
+	return fl_cinfo->ops->scan_commit_lists(cinfo, max);
+}
+
+static inline void
+pnfs_recover_commit_reqs(struct list_head *head, struct nfs_commit_info *cinfo)
+{
+	struct pnfs_ds_commit_info *fl_cinfo = cinfo->ds;
+
+	if (fl_cinfo && fl_cinfo->nwritten != 0)
+		fl_cinfo->ops->recover_commit_reqs(head, cinfo);
 }
 
 static inline struct nfs_page *
 pnfs_search_commit_reqs(struct inode *inode, struct nfs_commit_info *cinfo,
 			struct page *page)
 {
-	struct pnfs_layoutdriver_type *ld = NFS_SERVER(inode)->pnfs_curr_ld;
+	struct pnfs_ds_commit_info *fl_cinfo = cinfo->ds;
 
-	if (ld == NULL || ld->search_commit_reqs == NULL)
+	if (!fl_cinfo->ops || !fl_cinfo->ops->search_commit_reqs)
 		return NULL;
-	return ld->search_commit_reqs(cinfo, page);
+	return fl_cinfo->ops->search_commit_reqs(cinfo, page);
 }
 
 /* Should the pNFS client commit and return the layout upon a setattr */
@@ -750,6 +809,21 @@ pnfs_get_ds_info(struct inode *inode)
 	return NULL;
 }
 
+static inline void
+pnfs_init_ds_commit_info_ops(struct pnfs_ds_commit_info *fl_cinfo, struct inode *inode)
+{
+}
+
+static inline void
+pnfs_init_ds_commit_info(struct pnfs_ds_commit_info *fl_cinfo)
+{
+}
+
+static inline void
+pnfs_release_ds_info(struct pnfs_ds_commit_info *fl_cinfo, struct inode *inode)
+{
+}
+
 static inline bool
 pnfs_mark_request_commit(struct nfs_page *req, struct pnfs_layout_segment *lseg,
 			 struct nfs_commit_info *cinfo, u32 ds_commit_idx)
@@ -770,6 +844,11 @@ pnfs_scan_commit_lists(struct inode *inode, struct nfs_commit_info *cinfo,
 	return 0;
 }
 
+static inline void
+pnfs_recover_commit_reqs(struct list_head *head, struct nfs_commit_info *cinfo)
+{
+}
+
 static inline struct nfs_page *
 pnfs_search_commit_reqs(struct inode *inode, struct nfs_commit_info *cinfo,
 			struct page *page)