Merge tag 'nfs-for-5.5-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs

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

  Features:

   - NFSv4.2 now supports cross device offloaded copy (i.e. offloaded
     copy of a file from one source server to a different target
     server).

   - New RDMA tracepoints for debugging congestion control and Local
     Invalidate WRs.

  Bugfixes and cleanups

   - Drop the NFSv4.1 session slot if nfs4_delegreturn_prepare waits for
     layoutreturn

   - Handle bad/dead sessions correctly in nfs41_sequence_process()

   - Various bugfixes to the delegation return operation.

   - Various bugfixes pertaining to delegations that have been revoked.

   - Cleanups to the NFS timespec code to avoid unnecessary conversions
     between timespec and timespec64.

   - Fix unstable RDMA connections after a reconnect

   - Close race between waking an RDMA sender and posting a receive

   - Wake pending RDMA tasks if connection fails

   - Fix MR list corruption, and clean up MR usage

   - Fix another RPCSEC_GSS issue with MIC buffer space"

* tag 'nfs-for-5.5-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs: (79 commits)
  SUNRPC: Capture completion of all RPC tasks
  SUNRPC: Fix another issue with MIC buffer space
  NFS4: Trace lock reclaims
  NFS4: Trace state recovery operation
  NFSv4.2 fix memory leak in nfs42_ssc_open
  NFSv4.2 fix kfree in __nfs42_copy_file_range
  NFS: remove duplicated include from nfs4file.c
  NFSv4: Make _nfs42_proc_copy_notify() static
  NFS: Fallocate should use the nfs4_fattr_bitmap
  NFS: Return -ETXTBSY when attempting to write to a swapfile
  fs: nfs: sysfs: Remove NULL check before kfree
  NFS: remove unneeded semicolon
  NFSv4: add declaration of current_stateid
  NFSv4.x: Drop the slot if nfs4_delegreturn_prepare waits for layoutreturn
  NFSv4.x: Handle bad/dead sessions correctly in nfs41_sequence_process()
  nfsv4: Move NFSPROC4_CLNT_COPY_NOTIFY to end of list
  SUNRPC: Avoid RPC delays when exiting suspend
  NFS: Add a tracepoint in nfs_fh_to_dentry()
  NFSv4: Don't retry the GETATTR on old stateid in nfs4_delegreturn_done()
  NFSv4: Handle NFS4ERR_OLD_STATEID in delegreturn
  ...

2 files changed, 226 insertions, 66 deletions

diff --git a/include/trace/events/rpcrdma.h b/include/trace/events/rpcrdma.h
index a13830616107..69a8278e5cef 100644
--- a/include/trace/events/rpcrdma.h
+++ b/include/trace/events/rpcrdma.h
@@ -85,6 +85,44 @@ DECLARE_EVENT_CLASS(xprtrdma_rxprt,
 				),					\
 				TP_ARGS(r_xprt))
 
+DECLARE_EVENT_CLASS(xprtrdma_connect_class,
+	TP_PROTO(
+		const struct rpcrdma_xprt *r_xprt,
+		int rc
+	),
+
+	TP_ARGS(r_xprt, rc),
+
+	TP_STRUCT__entry(
+		__field(const void *, r_xprt)
+		__field(int, rc)
+		__field(int, connect_status)
+		__string(addr, rpcrdma_addrstr(r_xprt))
+		__string(port, rpcrdma_portstr(r_xprt))
+	),
+
+	TP_fast_assign(
+		__entry->r_xprt = r_xprt;
+		__entry->rc = rc;
+		__entry->connect_status = r_xprt->rx_ep.rep_connected;
+		__assign_str(addr, rpcrdma_addrstr(r_xprt));
+		__assign_str(port, rpcrdma_portstr(r_xprt));
+	),
+
+	TP_printk("peer=[%s]:%s r_xprt=%p: rc=%d connect status=%d",
+		__get_str(addr), __get_str(port), __entry->r_xprt,
+		__entry->rc, __entry->connect_status
+	)
+);
+
+#define DEFINE_CONN_EVENT(name)						\
+		DEFINE_EVENT(xprtrdma_connect_class, xprtrdma_##name,	\
+				TP_PROTO(				\
+					const struct rpcrdma_xprt *r_xprt, \
+					int rc				\
+				),					\
+				TP_ARGS(r_xprt, rc))
+
 DECLARE_EVENT_CLASS(xprtrdma_rdch_event,
 	TP_PROTO(
 		const struct rpc_task *task,
@@ -333,47 +371,81 @@ TRACE_EVENT(xprtrdma_cm_event,
 	)
 );
 
-TRACE_EVENT(xprtrdma_disconnect,
+TRACE_EVENT(xprtrdma_inline_thresh,
 	TP_PROTO(
-		const struct rpcrdma_xprt *r_xprt,
-		int status
+		const struct rpcrdma_xprt *r_xprt
 	),
 
-	TP_ARGS(r_xprt, status),
+	TP_ARGS(r_xprt),
 
 	TP_STRUCT__entry(
 		__field(const void *, r_xprt)
-		__field(int, status)
-		__field(int, connected)
+		__field(unsigned int, inline_send)
+		__field(unsigned int, inline_recv)
+		__field(unsigned int, max_send)
+		__field(unsigned int, max_recv)
 		__string(addr, rpcrdma_addrstr(r_xprt))
 		__string(port, rpcrdma_portstr(r_xprt))
 	),
 
 	TP_fast_assign(
+		const struct rpcrdma_ep *ep = &r_xprt->rx_ep;
+
 		__entry->r_xprt = r_xprt;
-		__entry->status = status;
-		__entry->connected = r_xprt->rx_ep.rep_connected;
+		__entry->inline_send = ep->rep_inline_send;
+		__entry->inline_recv = ep->rep_inline_recv;
+		__entry->max_send = ep->rep_max_inline_send;
+		__entry->max_recv = ep->rep_max_inline_recv;
 		__assign_str(addr, rpcrdma_addrstr(r_xprt));
 		__assign_str(port, rpcrdma_portstr(r_xprt));
 	),
 
-	TP_printk("peer=[%s]:%s r_xprt=%p: status=%d %sconnected",
-		__get_str(addr), __get_str(port),
-		__entry->r_xprt, __entry->status,
-		__entry->connected == 1 ? "still " : "dis"
+	TP_printk("peer=[%s]:%s r_xprt=%p neg send/recv=%u/%u, calc send/recv=%u/%u",
+		__get_str(addr), __get_str(port), __entry->r_xprt,
+		__entry->inline_send, __entry->inline_recv,
+		__entry->max_send, __entry->max_recv
 	)
 );
 
-DEFINE_RXPRT_EVENT(xprtrdma_conn_start);
-DEFINE_RXPRT_EVENT(xprtrdma_conn_tout);
+DEFINE_CONN_EVENT(connect);
+DEFINE_CONN_EVENT(disconnect);
+
 DEFINE_RXPRT_EVENT(xprtrdma_create);
 DEFINE_RXPRT_EVENT(xprtrdma_op_destroy);
 DEFINE_RXPRT_EVENT(xprtrdma_remove);
 DEFINE_RXPRT_EVENT(xprtrdma_reinsert);
-DEFINE_RXPRT_EVENT(xprtrdma_reconnect);
 DEFINE_RXPRT_EVENT(xprtrdma_op_inject_dsc);
 DEFINE_RXPRT_EVENT(xprtrdma_op_close);
-DEFINE_RXPRT_EVENT(xprtrdma_op_connect);
+DEFINE_RXPRT_EVENT(xprtrdma_op_setport);
+
+TRACE_EVENT(xprtrdma_op_connect,
+	TP_PROTO(
+		const struct rpcrdma_xprt *r_xprt,
+		unsigned long delay
+	),
+
+	TP_ARGS(r_xprt, delay),
+
+	TP_STRUCT__entry(
+		__field(const void *, r_xprt)
+		__field(unsigned long, delay)
+		__string(addr, rpcrdma_addrstr(r_xprt))
+		__string(port, rpcrdma_portstr(r_xprt))
+	),
+
+	TP_fast_assign(
+		__entry->r_xprt = r_xprt;
+		__entry->delay = delay;
+		__assign_str(addr, rpcrdma_addrstr(r_xprt));
+		__assign_str(port, rpcrdma_portstr(r_xprt));
+	),
+
+	TP_printk("peer=[%s]:%s r_xprt=%p delay=%lu",
+		__get_str(addr), __get_str(port), __entry->r_xprt,
+		__entry->delay
+	)
+);
+
 
 TRACE_EVENT(xprtrdma_op_set_cto,
 	TP_PROTO(
@@ -532,6 +604,8 @@ DEFINE_WRCH_EVENT(write);
 DEFINE_WRCH_EVENT(reply);
 
 TRACE_DEFINE_ENUM(rpcrdma_noch);
+TRACE_DEFINE_ENUM(rpcrdma_noch_pullup);
+TRACE_DEFINE_ENUM(rpcrdma_noch_mapped);
 TRACE_DEFINE_ENUM(rpcrdma_readch);
 TRACE_DEFINE_ENUM(rpcrdma_areadch);
 TRACE_DEFINE_ENUM(rpcrdma_writech);
@@ -540,6 +614,8 @@ TRACE_DEFINE_ENUM(rpcrdma_replych);
 #define xprtrdma_show_chunktype(x)					\
 		__print_symbolic(x,					\
 				{ rpcrdma_noch, "inline" },		\
+				{ rpcrdma_noch_pullup, "pullup" },	\
+				{ rpcrdma_noch_mapped, "mapped" },	\
 				{ rpcrdma_readch, "read list" },	\
 				{ rpcrdma_areadch, "*read list" },	\
 				{ rpcrdma_writech, "write list" },	\
@@ -667,9 +743,8 @@ TRACE_EVENT(xprtrdma_post_send,
 		__entry->client_id = rqst->rq_task->tk_client ?
 				     rqst->rq_task->tk_client->cl_clid : -1;
 		__entry->req = req;
-		__entry->num_sge = req->rl_sendctx->sc_wr.num_sge;
-		__entry->signaled = req->rl_sendctx->sc_wr.send_flags &
-				    IB_SEND_SIGNALED;
+		__entry->num_sge = req->rl_wr.num_sge;
+		__entry->signaled = req->rl_wr.send_flags & IB_SEND_SIGNALED;
 		__entry->status = status;
 	),
 
@@ -735,6 +810,31 @@ TRACE_EVENT(xprtrdma_post_recvs,
 	)
 );
 
+TRACE_EVENT(xprtrdma_post_linv,
+	TP_PROTO(
+		const struct rpcrdma_req *req,
+		int status
+	),
+
+	TP_ARGS(req, status),
+
+	TP_STRUCT__entry(
+		__field(const void *, req)
+		__field(int, status)
+		__field(u32, xid)
+	),
+
+	TP_fast_assign(
+		__entry->req = req;
+		__entry->status = status;
+		__entry->xid = be32_to_cpu(req->rl_slot.rq_xid);
+	),
+
+	TP_printk("req=%p xid=0x%08x status=%d",
+		__entry->req, __entry->xid, __entry->status
+	)
+);
+
 /**
  ** Completion events
  **/
@@ -1021,66 +1121,32 @@ DEFINE_REPLY_EVENT(xprtrdma_reply_hdr);
 TRACE_EVENT(xprtrdma_fixup,
 	TP_PROTO(
 		const struct rpc_rqst *rqst,
-		int len,
-		int hdrlen
+		unsigned long fixup
 	),
 
-	TP_ARGS(rqst, len, hdrlen),
+	TP_ARGS(rqst, fixup),
 
 	TP_STRUCT__entry(
 		__field(unsigned int, task_id)
 		__field(unsigned int, client_id)
-		__field(const void *, base)
-		__field(int, len)
-		__field(int, hdrlen)
-	),
-
-	TP_fast_assign(
-		__entry->task_id = rqst->rq_task->tk_pid;
-		__entry->client_id = rqst->rq_task->tk_client->cl_clid;
-		__entry->base = rqst->rq_rcv_buf.head[0].iov_base;
-		__entry->len = len;
-		__entry->hdrlen = hdrlen;
-	),
-
-	TP_printk("task:%u@%u base=%p len=%d hdrlen=%d",
-		__entry->task_id, __entry->client_id,
-		__entry->base, __entry->len, __entry->hdrlen
-	)
-);
-
-TRACE_EVENT(xprtrdma_fixup_pg,
-	TP_PROTO(
-		const struct rpc_rqst *rqst,
-		int pageno,
-		const void *pos,
-		int len,
-		int curlen
-	),
-
-	TP_ARGS(rqst, pageno, pos, len, curlen),
-
-	TP_STRUCT__entry(
-		__field(unsigned int, task_id)
-		__field(unsigned int, client_id)
-		__field(const void *, pos)
-		__field(int, pageno)
-		__field(int, len)
-		__field(int, curlen)
+		__field(unsigned long, fixup)
+		__field(size_t, headlen)
+		__field(unsigned int, pagelen)
+		__field(size_t, taillen)
 	),
 
 	TP_fast_assign(
 		__entry->task_id = rqst->rq_task->tk_pid;
 		__entry->client_id = rqst->rq_task->tk_client->cl_clid;
-		__entry->pos = pos;
-		__entry->pageno = pageno;
-		__entry->len = len;
-		__entry->curlen = curlen;
+		__entry->fixup = fixup;
+		__entry->headlen = rqst->rq_rcv_buf.head[0].iov_len;
+		__entry->pagelen = rqst->rq_rcv_buf.page_len;
+		__entry->taillen = rqst->rq_rcv_buf.tail[0].iov_len;
 	),
 
-	TP_printk("task:%u@%u pageno=%d pos=%p len=%d curlen=%d",
-		__entry->task_id, __entry->client_id,
-		__entry->pageno, __entry->pos, __entry->len, __entry->curlen
+	TP_printk("task:%u@%u fixup=%lu xdr=%zu/%u/%zu",
+		__entry->task_id, __entry->client_id, __entry->fixup,
+		__entry->headlen, __entry->pagelen, __entry->taillen
 	)
 );
 
diff --git a/include/trace/events/sunrpc.h b/include/trace/events/sunrpc.h
index ffa3c51dbb1a..205bf28bcada 100644
--- a/include/trace/events/sunrpc.h
+++ b/include/trace/events/sunrpc.h
@@ -165,6 +165,7 @@ DECLARE_EVENT_CLASS(rpc_task_running,
 DEFINE_RPC_RUNNING_EVENT(begin);
 DEFINE_RPC_RUNNING_EVENT(run_action);
 DEFINE_RPC_RUNNING_EVENT(complete);
+DEFINE_RPC_RUNNING_EVENT(end);
 
 DECLARE_EVENT_CLASS(rpc_task_queued,
 
@@ -777,6 +778,99 @@ TRACE_EVENT(xprt_ping,
 			__get_str(addr), __get_str(port), __entry->status)
 );
 
+DECLARE_EVENT_CLASS(xprt_writelock_event,
+	TP_PROTO(
+		const struct rpc_xprt *xprt, const struct rpc_task *task
+	),
+
+	TP_ARGS(xprt, task),
+
+	TP_STRUCT__entry(
+		__field(unsigned int, task_id)
+		__field(unsigned int, client_id)
+		__field(unsigned int, snd_task_id)
+	),
+
+	TP_fast_assign(
+		if (task) {
+			__entry->task_id = task->tk_pid;
+			__entry->client_id = task->tk_client ?
+					     task->tk_client->cl_clid : -1;
+		} else {
+			__entry->task_id = -1;
+			__entry->client_id = -1;
+		}
+		__entry->snd_task_id = xprt->snd_task ?
+					xprt->snd_task->tk_pid : -1;
+	),
+
+	TP_printk("task:%u@%u snd_task:%u",
+			__entry->task_id, __entry->client_id,
+			__entry->snd_task_id)
+);
+
+#define DEFINE_WRITELOCK_EVENT(name) \
+	DEFINE_EVENT(xprt_writelock_event, xprt_##name, \
+			TP_PROTO( \
+				const struct rpc_xprt *xprt, \
+				const struct rpc_task *task \
+			), \
+			TP_ARGS(xprt, task))
+
+DEFINE_WRITELOCK_EVENT(reserve_xprt);
+DEFINE_WRITELOCK_EVENT(release_xprt);
+
+DECLARE_EVENT_CLASS(xprt_cong_event,
+	TP_PROTO(
+		const struct rpc_xprt *xprt, const struct rpc_task *task
+	),
+
+	TP_ARGS(xprt, task),
+
+	TP_STRUCT__entry(
+		__field(unsigned int, task_id)
+		__field(unsigned int, client_id)
+		__field(unsigned int, snd_task_id)
+		__field(unsigned long, cong)
+		__field(unsigned long, cwnd)
+		__field(bool, wait)
+	),
+
+	TP_fast_assign(
+		if (task) {
+			__entry->task_id = task->tk_pid;
+			__entry->client_id = task->tk_client ?
+					     task->tk_client->cl_clid : -1;
+		} else {
+			__entry->task_id = -1;
+			__entry->client_id = -1;
+		}
+		__entry->snd_task_id = xprt->snd_task ?
+					xprt->snd_task->tk_pid : -1;
+		__entry->cong = xprt->cong;
+		__entry->cwnd = xprt->cwnd;
+		__entry->wait = test_bit(XPRT_CWND_WAIT, &xprt->state);
+	),
+
+	TP_printk("task:%u@%u snd_task:%u cong=%lu cwnd=%lu%s",
+			__entry->task_id, __entry->client_id,
+			__entry->snd_task_id, __entry->cong, __entry->cwnd,
+			__entry->wait ? " (wait)" : "")
+);
+
+#define DEFINE_CONG_EVENT(name) \
+	DEFINE_EVENT(xprt_cong_event, xprt_##name, \
+			TP_PROTO( \
+				const struct rpc_xprt *xprt, \
+				const struct rpc_task *task \
+			), \
+			TP_ARGS(xprt, task))
+
+DEFINE_CONG_EVENT(reserve_cong);
+DEFINE_CONG_EVENT(release_cong);
+DEFINE_CONG_EVENT(get_cong);
+DEFINE_CONG_EVENT(put_cong);
+
 TRACE_EVENT(xs_stream_read_data,
 	TP_PROTO(struct rpc_xprt *xprt, ssize_t err, size_t total),