Merge tag 'nfs-for-5.20-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs

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

  Stable fixes:
   - pNFS/flexfiles: Fix infinite looping when the RDMA connection
     errors out

  Bugfixes:
   - NFS: fix port value parsing
   - SUNRPC: Reinitialise the backchannel request buffers before reuse
   - SUNRPC: fix expiry of auth creds
   - NFSv4: Fix races in the legacy idmapper upcall
   - NFS: O_DIRECT fixes from Jeff Layton
   - NFSv4.1: Fix OP_SEQUENCE error handling
   - SUNRPC: Fix an RPC/RDMA performance regression
   - NFS: Fix case insensitive renames
   - NFSv4/pnfs: Fix a use-after-free bug in open
   - NFSv4.1: RECLAIM_COMPLETE must handle EACCES

  Features:
   - NFSv4.1: session trunking enhancements
   - NFSv4.2: READ_PLUS performance optimisations
   - NFS: relax the rules for rsize/wsize mount options
   - NFS: don't unhash dentry during unlink/rename
   - SUNRPC: Fail faster on bad verifier
   - NFS/SUNRPC: Various tracing improvements"

* tag 'nfs-for-5.20-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs: (46 commits)
  NFS: Improve readpage/writepage tracing
  NFS: Improve O_DIRECT tracing
  NFS: Improve write error tracing
  NFS: don't unhash dentry during unlink/rename
  NFSv4/pnfs: Fix a use-after-free bug in open
  NFS: nfs_async_write_reschedule_io must not recurse into the writeback code
  SUNRPC: Don't reuse bvec on retransmission of the request
  SUNRPC: Reinitialise the backchannel request buffers before reuse
  NFSv4.1: RECLAIM_COMPLETE must handle EACCES
  NFSv4.1 probe offline transports for trunking on session creation
  SUNRPC create a function that probes only offline transports
  SUNRPC export xprt_iter_rewind function
  SUNRPC restructure rpc_clnt_setup_test_and_add_xprt
  NFSv4.1 remove xprt from xprt_switch if session trunking test fails
  SUNRPC create an rpc function that allows xprt removal from rpc_clnt
  SUNRPC enable back offline transports in trunking discovery
  SUNRPC create an iterator to list only OFFLINE xprts
  NFSv4.1 offline trunkable transports on DESTROY_SESSION
  SUNRPC add function to offline remove trunkable transports
  SUNRPC expose functions for offline remote xprt functionality
  ...

9 files changed, 54 insertions, 8 deletions

diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index a17c337dbdf1..b32ed68e7dc4 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -617,6 +617,15 @@ nfs_fileid_to_ino_t(u64 fileid)
 
 #define NFS_JUKEBOX_RETRY_TIME (5 * HZ)
 
+/* We need to block new opens while a file is being unlinked.
+ * If it is opened *before* we decide to unlink, we will silly-rename
+ * instead. If it is opened *after*, then we need to create or will fail.
+ * If we allow the two to race, we could end up with a file that is open
+ * but deleted on the server resulting in ESTALE.
+ * So use ->d_fsdata to record when the unlink is happening
+ * and block dentry revalidation while it is set.
+ */
+#define NFS_FSDATA_BLOCKED ((void*)1)
 
 # undef ifdebug
 # ifdef NFS_DEBUG
diff --git a/include/linux/nfs_page.h b/include/linux/nfs_page.h
index f0373a6cb5fb..ba7e2e4b0926 100644
--- a/include/linux/nfs_page.h
+++ b/include/linux/nfs_page.h
@@ -202,8 +202,7 @@ nfs_list_entry(struct list_head *head)
 	return list_entry(head, struct nfs_page, wb_list);
 }
 
-static inline
-loff_t req_offset(struct nfs_page *req)
+static inline loff_t req_offset(const struct nfs_page *req)
 {
 	return (((loff_t)req->wb_index) << PAGE_SHIFT) + req->wb_offset;
 }
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index 0e3aa0f5f324..e86cf6642d21 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -1600,6 +1600,7 @@ enum {
 	NFS_IOHDR_STAT,
 	NFS_IOHDR_RESEND_PNFS,
 	NFS_IOHDR_RESEND_MDS,
+	NFS_IOHDR_UNSTABLE_WRITES,
 };
 
 struct nfs_io_completion;
diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h
index 90501404fa49..75eea5ebb179 100644
--- a/include/linux/sunrpc/clnt.h
+++ b/include/linux/sunrpc/clnt.h
@@ -234,13 +234,18 @@ int		rpc_clnt_setup_test_and_add_xprt(struct rpc_clnt *,
 			struct rpc_xprt_switch *,
 			struct rpc_xprt *,
 			void *);
+void		rpc_clnt_manage_trunked_xprts(struct rpc_clnt *);
+void		rpc_clnt_probe_trunked_xprts(struct rpc_clnt *,
+			struct rpc_add_xprt_test *);
 
 const char *rpc_proc_name(const struct rpc_task *task);
 
 void rpc_clnt_xprt_switch_put(struct rpc_clnt *);
 void rpc_clnt_xprt_switch_add_xprt(struct rpc_clnt *, struct rpc_xprt *);
+void rpc_clnt_xprt_switch_remove_xprt(struct rpc_clnt *, struct rpc_xprt *);
 bool rpc_clnt_xprt_switch_has_addr(struct rpc_clnt *clnt,
 			const struct sockaddr *sap);
+void rpc_clnt_xprt_set_online(struct rpc_clnt *clnt, struct rpc_xprt *xprt);
 void rpc_cleanup_clids(void);
 
 static inline int rpc_reply_expected(struct rpc_task *task)
diff --git a/include/linux/sunrpc/sched.h b/include/linux/sunrpc/sched.h
index 1d7a3e51b795..acc62647317c 100644
--- a/include/linux/sunrpc/sched.h
+++ b/include/linux/sunrpc/sched.h
@@ -61,8 +61,6 @@ struct rpc_task {
 		struct rpc_wait		tk_wait;	/* RPC wait */
 	} u;
 
-	int			tk_rpc_status;	/* Result of last RPC operation */
-
 	/*
 	 * RPC call state
 	 */
@@ -82,6 +80,8 @@ struct rpc_task {
 	ktime_t			tk_start;	/* RPC task init timestamp */
 
 	pid_t			tk_owner;	/* Process id for batching tasks */
+
+	int			tk_rpc_status;	/* Result of last RPC operation */
 	unsigned short		tk_flags;	/* misc flags */
 	unsigned short		tk_timeouts;	/* maj timeouts */
 
diff --git a/include/linux/sunrpc/xdr.h b/include/linux/sunrpc/xdr.h
index 986c8a17ca5e..69175029abbb 100644
--- a/include/linux/sunrpc/xdr.h
+++ b/include/linux/sunrpc/xdr.h
@@ -258,10 +258,13 @@ extern __be32 *xdr_inline_decode(struct xdr_stream *xdr, size_t nbytes);
 extern unsigned int xdr_read_pages(struct xdr_stream *xdr, unsigned int len);
 extern void xdr_enter_page(struct xdr_stream *xdr, unsigned int len);
 extern int xdr_process_buf(const struct xdr_buf *buf, unsigned int offset, unsigned int len, int (*actor)(struct scatterlist *, void *), void *data);
-extern unsigned int xdr_align_data(struct xdr_stream *, unsigned int offset, unsigned int length);
-extern unsigned int xdr_expand_hole(struct xdr_stream *, unsigned int offset, unsigned int length);
+extern void xdr_set_pagelen(struct xdr_stream *, unsigned int len);
 extern bool xdr_stream_subsegment(struct xdr_stream *xdr, struct xdr_buf *subbuf,
 				  unsigned int len);
+extern unsigned int xdr_stream_move_subsegment(struct xdr_stream *xdr, unsigned int offset,
+					       unsigned int target, unsigned int length);
+extern unsigned int xdr_stream_zero(struct xdr_stream *xdr, unsigned int offset,
+				    unsigned int length);
 
 /**
  * xdr_set_scratch_buffer - Attach a scratch buffer for decoding data.
diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h
index 522bbf937957..b9f59aabee53 100644
--- a/include/linux/sunrpc/xprt.h
+++ b/include/linux/sunrpc/xprt.h
@@ -144,7 +144,8 @@ struct rpc_xprt_ops {
 	unsigned short	(*get_srcport)(struct rpc_xprt *xprt);
 	int		(*buf_alloc)(struct rpc_task *task);
 	void		(*buf_free)(struct rpc_task *task);
-	int		(*prepare_request)(struct rpc_rqst *req);
+	int		(*prepare_request)(struct rpc_rqst *req,
+					   struct xdr_buf *buf);
 	int		(*send_request)(struct rpc_rqst *req);
 	void		(*wait_for_reply_request)(struct rpc_task *task);
 	void		(*timer)(struct rpc_xprt *xprt, struct rpc_task *task);
@@ -505,4 +506,7 @@ static inline int xprt_test_and_set_binding(struct rpc_xprt *xprt)
 	return test_and_set_bit(XPRT_BINDING, &xprt->state);
 }
 
+void xprt_set_offline_locked(struct rpc_xprt *xprt, struct rpc_xprt_switch *xps);
+void xprt_set_online_locked(struct rpc_xprt *xprt, struct rpc_xprt_switch *xps);
+void xprt_delete_locked(struct rpc_xprt *xprt, struct rpc_xprt_switch *xps);
 #endif /* _LINUX_SUNRPC_XPRT_H */
diff --git a/include/linux/sunrpc/xprtmultipath.h b/include/linux/sunrpc/xprtmultipath.h
index bbb8a5fa0816..c0514c684b2c 100644
--- a/include/linux/sunrpc/xprtmultipath.h
+++ b/include/linux/sunrpc/xprtmultipath.h
@@ -55,7 +55,7 @@ extern void rpc_xprt_switch_set_roundrobin(struct rpc_xprt_switch *xps);
 extern void rpc_xprt_switch_add_xprt(struct rpc_xprt_switch *xps,
 		struct rpc_xprt *xprt);
 extern void rpc_xprt_switch_remove_xprt(struct rpc_xprt_switch *xps,
-		struct rpc_xprt *xprt);
+		struct rpc_xprt *xprt, bool offline);
 
 extern void xprt_iter_init(struct rpc_xprt_iter *xpi,
 		struct rpc_xprt_switch *xps);
@@ -63,8 +63,13 @@ extern void xprt_iter_init(struct rpc_xprt_iter *xpi,
 extern void xprt_iter_init_listall(struct rpc_xprt_iter *xpi,
 		struct rpc_xprt_switch *xps);
 
+extern void xprt_iter_init_listoffline(struct rpc_xprt_iter *xpi,
+		struct rpc_xprt_switch *xps);
+
 extern void xprt_iter_destroy(struct rpc_xprt_iter *xpi);
 
+extern void xprt_iter_rewind(struct rpc_xprt_iter *xpi);
+
 extern struct rpc_xprt_switch *xprt_iter_xchg_switch(
 		struct rpc_xprt_iter *xpi,
 		struct rpc_xprt_switch *newswitch);
diff --git a/include/trace/events/sunrpc.h b/include/trace/events/sunrpc.h
index 5c48be033cc7..f48f2ab9d238 100644
--- a/include/trace/events/sunrpc.h
+++ b/include/trace/events/sunrpc.h
@@ -1266,6 +1266,26 @@ TRACE_EVENT(xprt_reserve,
 	)
 );
 
+TRACE_EVENT(xs_data_ready,
+	TP_PROTO(
+		const struct rpc_xprt *xprt
+	),
+
+	TP_ARGS(xprt),
+
+	TP_STRUCT__entry(
+		__string(addr, xprt->address_strings[RPC_DISPLAY_ADDR])
+		__string(port, xprt->address_strings[RPC_DISPLAY_PORT])
+	),
+
+	TP_fast_assign(
+		__assign_str(addr, xprt->address_strings[RPC_DISPLAY_ADDR]);
+		__assign_str(port, xprt->address_strings[RPC_DISPLAY_PORT]);
+	),
+
+	TP_printk("peer=[%s]:%s", __get_str(addr), __get_str(port))
+);
+
 TRACE_EVENT(xs_stream_read_data,
 	TP_PROTO(struct rpc_xprt *xprt, ssize_t err, size_t total),