From 860a0d9e511f278bedab62d555a457c18e0841d5 Mon Sep 17 00:00:00 2001
From: Jeff Layton <jlayton@primarydata.com>
Date: Tue, 28 Oct 2014 14:24:12 -0400
Subject: sunrpc: add some tracepoints in svc_rqst handling functions

...just around svc_send, svc_recv and svc_process for now.

Signed-off-by: Jeff Layton <jlayton@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
---
 net/sunrpc/svc.c      | 21 ++++++++++++---------
 net/sunrpc/svc_xprt.c | 31 +++++++++++++++++++++----------
 2 files changed, 33 insertions(+), 19 deletions(-)

(limited to 'net/sunrpc')

diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c
index ca8a7958f4e6..371a8bbb43d6 100644
--- a/net/sunrpc/svc.c
+++ b/net/sunrpc/svc.c
@@ -28,6 +28,8 @@
 #include <linux/sunrpc/clnt.h>
 #include <linux/sunrpc/bc_xprt.h>
 
+#include <trace/events/sunrpc.h>
+
 #define RPCDBG_FACILITY	RPCDBG_SVCDSP
 
 static void svc_unregister(const struct svc_serv *serv, struct net *net);
@@ -1314,24 +1316,25 @@ svc_process(struct svc_rqst *rqstp)
 	rqstp->rq_res.tail[0].iov_base = NULL;
 	rqstp->rq_res.tail[0].iov_len = 0;
 
-	rqstp->rq_xid = svc_getu32(argv);
-
 	dir  = svc_getnl(argv);
 	if (dir != 0) {
 		/* direction != CALL */
 		svc_printk(rqstp, "bad direction %d, dropping request\n", dir);
 		serv->sv_stats->rpcbadfmt++;
-		svc_drop(rqstp);
-		return 0;
+		goto out_drop;
 	}
 
 	/* Returns 1 for send, 0 for drop */
-	if (svc_process_common(rqstp, argv, resv))
-		return svc_send(rqstp);
-	else {
-		svc_drop(rqstp);
-		return 0;
+	if (likely(svc_process_common(rqstp, argv, resv))) {
+		int ret = svc_send(rqstp);
+
+		trace_svc_process(rqstp, ret);
+		return ret;
 	}
+out_drop:
+	trace_svc_process(rqstp, 0);
+	svc_drop(rqstp);
+	return 0;
 }
 
 #if defined(CONFIG_SUNRPC_BACKCHANNEL)
diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c
index c179ca2a5aa4..bbb3b044b877 100644
--- a/net/sunrpc/svc_xprt.c
+++ b/net/sunrpc/svc_xprt.c
@@ -15,6 +15,7 @@
 #include <linux/sunrpc/svcsock.h>
 #include <linux/sunrpc/xprt.h>
 #include <linux/module.h>
+#include <trace/events/sunrpc.h>
 
 #define RPCDBG_FACILITY	RPCDBG_SVCXPRT
 
@@ -773,35 +774,43 @@ int svc_recv(struct svc_rqst *rqstp, long timeout)
 
 	err = svc_alloc_arg(rqstp);
 	if (err)
-		return err;
+		goto out;
 
 	try_to_freeze();
 	cond_resched();
+	err = -EINTR;
 	if (signalled() || kthread_should_stop())
-		return -EINTR;
+		goto out;
 
 	xprt = svc_get_next_xprt(rqstp, timeout);
-	if (IS_ERR(xprt))
-		return PTR_ERR(xprt);
+	if (IS_ERR(xprt)) {
+		err = PTR_ERR(xprt);
+		goto out;
+	}
 
 	len = svc_handle_xprt(rqstp, xprt);
 
 	/* No data, incomplete (TCP) read, or accept() */
+	err = -EAGAIN;
 	if (len <= 0)
-		goto out;
+		goto out_release;
 
 	clear_bit(XPT_OLD, &xprt->xpt_flags);
 
 	rqstp->rq_secure = xprt->xpt_ops->xpo_secure_port(rqstp);
 	rqstp->rq_chandle.defer = svc_defer;
+	rqstp->rq_xid = svc_getu32(&rqstp->rq_arg.head[0]);
 
 	if (serv->sv_stats)
 		serv->sv_stats->netcnt++;
+	trace_svc_recv(rqstp, len);
 	return len;
-out:
+out_release:
 	rqstp->rq_res.len = 0;
 	svc_xprt_release(rqstp);
-	return -EAGAIN;
+out:
+	trace_svc_recv(rqstp, err);
+	return err;
 }
 EXPORT_SYMBOL_GPL(svc_recv);
 
@@ -821,12 +830,12 @@ EXPORT_SYMBOL_GPL(svc_drop);
 int svc_send(struct svc_rqst *rqstp)
 {
 	struct svc_xprt	*xprt;
-	int		len;
+	int		len = -EFAULT;
 	struct xdr_buf	*xb;
 
 	xprt = rqstp->rq_xprt;
 	if (!xprt)
-		return -EFAULT;
+		goto out;
 
 	/* release the receive skb before sending the reply */
 	rqstp->rq_xprt->xpt_ops->xpo_release_rqst(rqstp);
@@ -849,7 +858,9 @@ int svc_send(struct svc_rqst *rqstp)
 	svc_xprt_release(rqstp);
 
 	if (len == -ECONNREFUSED || len == -ENOTCONN || len == -EAGAIN)
-		return 0;
+		len = 0;
+out:
+	trace_svc_send(rqstp, len);
 	return len;
 }
 
-- 
cgit v1.2.3


From 3705ad64f123271b2b88dbff0c9891b7b90299d2 Mon Sep 17 00:00:00 2001
From: Jeff Layton <jlayton@primarydata.com>
Date: Tue, 28 Oct 2014 14:24:13 -0400
Subject: sunrpc: add new tracepoints in xprt handling code

...so we can keep track of when calls are sent and replies received.

Signed-off-by: Jeff Layton <jlayton@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
---
 net/sunrpc/xprt.c     | 9 ++++++++-
 net/sunrpc/xprtsock.c | 8 +++++++-
 2 files changed, 15 insertions(+), 2 deletions(-)

(limited to 'net/sunrpc')

diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c
index 56e4e150e80e..1b2e5e616cae 100644
--- a/net/sunrpc/xprt.c
+++ b/net/sunrpc/xprt.c
@@ -49,6 +49,8 @@
 #include <linux/sunrpc/metrics.h>
 #include <linux/sunrpc/bc_xprt.h>
 
+#include <trace/events/sunrpc.h>
+
 #include "sunrpc.h"
 
 /*
@@ -772,11 +774,14 @@ struct rpc_rqst *xprt_lookup_rqst(struct rpc_xprt *xprt, __be32 xid)
 	struct rpc_rqst *entry;
 
 	list_for_each_entry(entry, &xprt->recv, rq_list)
-		if (entry->rq_xid == xid)
+		if (entry->rq_xid == xid) {
+			trace_xprt_lookup_rqst(xprt, xid, 0);
 			return entry;
+		}
 
 	dprintk("RPC:       xprt_lookup_rqst did not find xid %08x\n",
 			ntohl(xid));
+	trace_xprt_lookup_rqst(xprt, xid, -ENOENT);
 	xprt->stat.bad_xids++;
 	return NULL;
 }
@@ -810,6 +815,7 @@ void xprt_complete_rqst(struct rpc_task *task, int copied)
 
 	dprintk("RPC: %5u xid %08x complete (%d bytes received)\n",
 			task->tk_pid, ntohl(req->rq_xid), copied);
+	trace_xprt_complete_rqst(xprt, req->rq_xid, copied);
 
 	xprt->stat.recvs++;
 	req->rq_rtt = ktime_sub(ktime_get(), req->rq_xtime);
@@ -926,6 +932,7 @@ void xprt_transmit(struct rpc_task *task)
 
 	req->rq_xtime = ktime_get();
 	status = xprt->ops->send_request(task);
+	trace_xprt_transmit(xprt, req->rq_xid, status);
 	if (status != 0) {
 		task->tk_status = status;
 		return;
diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c
index 3b305ab17afe..b63e26272dc2 100644
--- a/net/sunrpc/xprtsock.c
+++ b/net/sunrpc/xprtsock.c
@@ -1454,12 +1454,15 @@ static void xs_tcp_data_ready(struct sock *sk)
 	struct rpc_xprt *xprt;
 	read_descriptor_t rd_desc;
 	int read;
+	unsigned long total = 0;
 
 	dprintk("RPC:       xs_tcp_data_ready...\n");
 
 	read_lock_bh(&sk->sk_callback_lock);
-	if (!(xprt = xprt_from_sock(sk)))
+	if (!(xprt = xprt_from_sock(sk))) {
+		read = 0;
 		goto out;
+	}
 	/* Any data means we had a useful conversation, so
 	 * the we don't need to delay the next reconnect
 	 */
@@ -1471,8 +1474,11 @@ static void xs_tcp_data_ready(struct sock *sk)
 	do {
 		rd_desc.count = 65536;
 		read = tcp_read_sock(sk, &rd_desc, xs_tcp_data_recv);
+		if (read > 0)
+			total += read;
 	} while (read > 0);
 out:
+	trace_xs_tcp_data_ready(xprt, read, total);
 	read_unlock_bh(&sk->sk_callback_lock);
 }
 
-- 
cgit v1.2.3


From 1a867a0898b2e366a1eb5b7fe21413a2b2b1629f Mon Sep 17 00:00:00 2001
From: Jeff Layton <jlayton@primarydata.com>
Date: Tue, 28 Oct 2014 14:24:14 -0400
Subject: sunrpc: add tracepoints in xs_tcp_data_recv

Add tracepoints inside the main loop on xs_tcp_data_recv that allow
us to keep an eye on what's happening during each phase of it.

Signed-off-by: Jeff Layton <jlayton@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
---
 net/sunrpc/xprtsock.c | 61 ++-------------------------------------------------
 1 file changed, 2 insertions(+), 59 deletions(-)

(limited to 'net/sunrpc')

diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c
index b63e26272dc2..31c015196a29 100644
--- a/net/sunrpc/xprtsock.c
+++ b/net/sunrpc/xprtsock.c
@@ -216,65 +216,6 @@ static inline void xs_pktdump(char *msg, u32 *packet, unsigned int count)
 }
 #endif
 
-struct sock_xprt {
-	struct rpc_xprt		xprt;
-
-	/*
-	 * Network layer
-	 */
-	struct socket *		sock;
-	struct sock *		inet;
-
-	/*
-	 * State of TCP reply receive
-	 */
-	__be32			tcp_fraghdr,
-				tcp_xid,
-				tcp_calldir;
-
-	u32			tcp_offset,
-				tcp_reclen;
-
-	unsigned long		tcp_copied,
-				tcp_flags;
-
-	/*
-	 * Connection of transports
-	 */
-	struct delayed_work	connect_worker;
-	struct sockaddr_storage	srcaddr;
-	unsigned short		srcport;
-
-	/*
-	 * UDP socket buffer size parameters
-	 */
-	size_t			rcvsize,
-				sndsize;
-
-	/*
-	 * Saved socket callback addresses
-	 */
-	void			(*old_data_ready)(struct sock *);
-	void			(*old_state_change)(struct sock *);
-	void			(*old_write_space)(struct sock *);
-	void			(*old_error_report)(struct sock *);
-};
-
-/*
- * TCP receive state flags
- */
-#define TCP_RCV_LAST_FRAG	(1UL << 0)
-#define TCP_RCV_COPY_FRAGHDR	(1UL << 1)
-#define TCP_RCV_COPY_XID	(1UL << 2)
-#define TCP_RCV_COPY_DATA	(1UL << 3)
-#define TCP_RCV_READ_CALLDIR	(1UL << 4)
-#define TCP_RCV_COPY_CALLDIR	(1UL << 5)
-
-/*
- * TCP RPC flags
- */
-#define TCP_RPC_REPLY		(1UL << 6)
-
 static inline struct rpc_xprt *xprt_from_sock(struct sock *sk)
 {
 	return (struct rpc_xprt *) sk->sk_user_data;
@@ -1415,6 +1356,7 @@ static int xs_tcp_data_recv(read_descriptor_t *rd_desc, struct sk_buff *skb, uns
 
 	dprintk("RPC:       xs_tcp_data_recv started\n");
 	do {
+		trace_xs_tcp_data_recv(transport);
 		/* Read in a new fragment marker if necessary */
 		/* Can we ever really expect to get completely empty fragments? */
 		if (transport->tcp_flags & TCP_RCV_COPY_FRAGHDR) {
@@ -1439,6 +1381,7 @@ static int xs_tcp_data_recv(read_descriptor_t *rd_desc, struct sk_buff *skb, uns
 		/* Skip over any trailing bytes on short reads */
 		xs_tcp_read_discard(transport, &desc);
 	} while (desc.count);
+	trace_xs_tcp_data_recv(transport);
 	dprintk("RPC:       xs_tcp_data_recv done\n");
 	return len - desc.count;
 }
-- 
cgit v1.2.3


From f895b252d4edf66b2895fb5a7b17a638665f3e1f Mon Sep 17 00:00:00 2001
From: Jeff Layton <jlayton@primarydata.com>
Date: Mon, 17 Nov 2014 16:58:04 -0500
Subject: sunrpc: eliminate RPC_DEBUG

It's always set to whatever CONFIG_SUNRPC_DEBUG is, so just use that.

Signed-off-by: Jeff Layton <jlayton@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
---
 net/sunrpc/auth.c                       | 4 ++--
 net/sunrpc/auth_generic.c               | 2 +-
 net/sunrpc/auth_gss/auth_gss.c          | 2 +-
 net/sunrpc/auth_gss/gss_generic_token.c | 2 +-
 net/sunrpc/auth_gss/gss_krb5_crypto.c   | 2 +-
 net/sunrpc/auth_gss/gss_krb5_keys.c     | 2 +-
 net/sunrpc/auth_gss/gss_krb5_mech.c     | 2 +-
 net/sunrpc/auth_gss/gss_krb5_seal.c     | 2 +-
 net/sunrpc/auth_gss/gss_krb5_seqnum.c   | 2 +-
 net/sunrpc/auth_gss/gss_krb5_unseal.c   | 2 +-
 net/sunrpc/auth_gss/gss_krb5_wrap.c     | 2 +-
 net/sunrpc/auth_gss/gss_mech_switch.c   | 2 +-
 net/sunrpc/auth_gss/gss_rpc_xdr.h       | 2 +-
 net/sunrpc/auth_gss/svcauth_gss.c       | 2 +-
 net/sunrpc/auth_null.c                  | 4 ++--
 net/sunrpc/auth_unix.c                  | 2 +-
 net/sunrpc/backchannel_rqst.c           | 2 +-
 net/sunrpc/clnt.c                       | 6 +++---
 net/sunrpc/rpcb_clnt.c                  | 2 +-
 net/sunrpc/sched.c                      | 4 ++--
 net/sunrpc/sunrpc_syms.c                | 4 ++--
 net/sunrpc/svc.c                        | 2 +-
 net/sunrpc/sysctl.c                     | 2 +-
 net/sunrpc/xprt.c                       | 2 +-
 net/sunrpc/xprtrdma/rpc_rdma.c          | 4 ++--
 net/sunrpc/xprtrdma/transport.c         | 8 ++++----
 net/sunrpc/xprtrdma/verbs.c             | 8 ++++----
 net/sunrpc/xprtsock.c                   | 8 ++++----
 28 files changed, 44 insertions(+), 44 deletions(-)

(limited to 'net/sunrpc')

diff --git a/net/sunrpc/auth.c b/net/sunrpc/auth.c
index 383eb919ac0b..47f38be4155f 100644
--- a/net/sunrpc/auth.c
+++ b/net/sunrpc/auth.c
@@ -16,7 +16,7 @@
 #include <linux/sunrpc/gss_api.h>
 #include <linux/spinlock.h>
 
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 # define RPCDBG_FACILITY	RPCDBG_AUTH
 #endif
 
@@ -646,7 +646,7 @@ rpcauth_init_cred(struct rpc_cred *cred, const struct auth_cred *acred,
 	cred->cr_auth = auth;
 	cred->cr_ops = ops;
 	cred->cr_expire = jiffies;
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 	cred->cr_magic = RPCAUTH_CRED_MAGIC;
 #endif
 	cred->cr_uid = acred->uid;
diff --git a/net/sunrpc/auth_generic.c b/net/sunrpc/auth_generic.c
index 6f6b829c9e8e..41248b1820c7 100644
--- a/net/sunrpc/auth_generic.c
+++ b/net/sunrpc/auth_generic.c
@@ -14,7 +14,7 @@
 #include <linux/sunrpc/debug.h>
 #include <linux/sunrpc/sched.h>
 
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 # define RPCDBG_FACILITY	RPCDBG_AUTH
 #endif
 
diff --git a/net/sunrpc/auth_gss/auth_gss.c b/net/sunrpc/auth_gss/auth_gss.c
index 53ed8d3f8897..dace13d7638e 100644
--- a/net/sunrpc/auth_gss/auth_gss.c
+++ b/net/sunrpc/auth_gss/auth_gss.c
@@ -66,7 +66,7 @@ static unsigned int gss_expired_cred_retry_delay = GSS_RETRY_EXPIRED;
 #define GSS_KEY_EXPIRE_TIMEO 240
 static unsigned int gss_key_expire_timeo = GSS_KEY_EXPIRE_TIMEO;
 
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 # define RPCDBG_FACILITY	RPCDBG_AUTH
 #endif
 
diff --git a/net/sunrpc/auth_gss/gss_generic_token.c b/net/sunrpc/auth_gss/gss_generic_token.c
index c586e92bcf76..254defe446a7 100644
--- a/net/sunrpc/auth_gss/gss_generic_token.c
+++ b/net/sunrpc/auth_gss/gss_generic_token.c
@@ -38,7 +38,7 @@
 #include <linux/sunrpc/gss_asn1.h>
 
 
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 # define RPCDBG_FACILITY        RPCDBG_AUTH
 #endif
 
diff --git a/net/sunrpc/auth_gss/gss_krb5_crypto.c b/net/sunrpc/auth_gss/gss_krb5_crypto.c
index f5ed9f6ece06..b5408e8a37f2 100644
--- a/net/sunrpc/auth_gss/gss_krb5_crypto.c
+++ b/net/sunrpc/auth_gss/gss_krb5_crypto.c
@@ -45,7 +45,7 @@
 #include <linux/sunrpc/gss_krb5.h>
 #include <linux/sunrpc/xdr.h>
 
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 # define RPCDBG_FACILITY        RPCDBG_AUTH
 #endif
 
diff --git a/net/sunrpc/auth_gss/gss_krb5_keys.c b/net/sunrpc/auth_gss/gss_krb5_keys.c
index 24589bd2a4b6..234fa8d0fd9b 100644
--- a/net/sunrpc/auth_gss/gss_krb5_keys.c
+++ b/net/sunrpc/auth_gss/gss_krb5_keys.c
@@ -61,7 +61,7 @@
 #include <linux/sunrpc/xdr.h>
 #include <linux/lcm.h>
 
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 # define RPCDBG_FACILITY        RPCDBG_AUTH
 #endif
 
diff --git a/net/sunrpc/auth_gss/gss_krb5_mech.c b/net/sunrpc/auth_gss/gss_krb5_mech.c
index 0d3c158ef8fa..28db442a0034 100644
--- a/net/sunrpc/auth_gss/gss_krb5_mech.c
+++ b/net/sunrpc/auth_gss/gss_krb5_mech.c
@@ -45,7 +45,7 @@
 #include <linux/crypto.h>
 #include <linux/sunrpc/gss_krb5_enctypes.h>
 
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 # define RPCDBG_FACILITY	RPCDBG_AUTH
 #endif
 
diff --git a/net/sunrpc/auth_gss/gss_krb5_seal.c b/net/sunrpc/auth_gss/gss_krb5_seal.c
index 42768e5c3994..1d74d653e6c0 100644
--- a/net/sunrpc/auth_gss/gss_krb5_seal.c
+++ b/net/sunrpc/auth_gss/gss_krb5_seal.c
@@ -64,7 +64,7 @@
 #include <linux/random.h>
 #include <linux/crypto.h>
 
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 # define RPCDBG_FACILITY        RPCDBG_AUTH
 #endif
 
diff --git a/net/sunrpc/auth_gss/gss_krb5_seqnum.c b/net/sunrpc/auth_gss/gss_krb5_seqnum.c
index 62ac90c62cb1..20d55c793eb6 100644
--- a/net/sunrpc/auth_gss/gss_krb5_seqnum.c
+++ b/net/sunrpc/auth_gss/gss_krb5_seqnum.c
@@ -35,7 +35,7 @@
 #include <linux/sunrpc/gss_krb5.h>
 #include <linux/crypto.h>
 
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 # define RPCDBG_FACILITY        RPCDBG_AUTH
 #endif
 
diff --git a/net/sunrpc/auth_gss/gss_krb5_unseal.c b/net/sunrpc/auth_gss/gss_krb5_unseal.c
index 6c981ddc19f8..dcf9515d9aef 100644
--- a/net/sunrpc/auth_gss/gss_krb5_unseal.c
+++ b/net/sunrpc/auth_gss/gss_krb5_unseal.c
@@ -62,7 +62,7 @@
 #include <linux/sunrpc/gss_krb5.h>
 #include <linux/crypto.h>
 
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 # define RPCDBG_FACILITY        RPCDBG_AUTH
 #endif
 
diff --git a/net/sunrpc/auth_gss/gss_krb5_wrap.c b/net/sunrpc/auth_gss/gss_krb5_wrap.c
index 4b614c604fe0..ca7e92a32f84 100644
--- a/net/sunrpc/auth_gss/gss_krb5_wrap.c
+++ b/net/sunrpc/auth_gss/gss_krb5_wrap.c
@@ -35,7 +35,7 @@
 #include <linux/pagemap.h>
 #include <linux/crypto.h>
 
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 # define RPCDBG_FACILITY	RPCDBG_AUTH
 #endif
 
diff --git a/net/sunrpc/auth_gss/gss_mech_switch.c b/net/sunrpc/auth_gss/gss_mech_switch.c
index 92d5ab99fbf3..7063d856a598 100644
--- a/net/sunrpc/auth_gss/gss_mech_switch.c
+++ b/net/sunrpc/auth_gss/gss_mech_switch.c
@@ -46,7 +46,7 @@
 #include <linux/sunrpc/gss_api.h>
 #include <linux/sunrpc/clnt.h>
 
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 # define RPCDBG_FACILITY        RPCDBG_AUTH
 #endif
 
diff --git a/net/sunrpc/auth_gss/gss_rpc_xdr.h b/net/sunrpc/auth_gss/gss_rpc_xdr.h
index 685a688f3d8a..9d88c6239f01 100644
--- a/net/sunrpc/auth_gss/gss_rpc_xdr.h
+++ b/net/sunrpc/auth_gss/gss_rpc_xdr.h
@@ -25,7 +25,7 @@
 #include <linux/sunrpc/clnt.h>
 #include <linux/sunrpc/xprtsock.h>
 
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 # define RPCDBG_FACILITY	RPCDBG_AUTH
 #endif
 
diff --git a/net/sunrpc/auth_gss/svcauth_gss.c b/net/sunrpc/auth_gss/svcauth_gss.c
index c548ab213f76..de856ddf5fed 100644
--- a/net/sunrpc/auth_gss/svcauth_gss.c
+++ b/net/sunrpc/auth_gss/svcauth_gss.c
@@ -51,7 +51,7 @@
 #include "gss_rpc_upcall.h"
 
 
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 # define RPCDBG_FACILITY	RPCDBG_AUTH
 #endif
 
diff --git a/net/sunrpc/auth_null.c b/net/sunrpc/auth_null.c
index 712c123e04e9..c2a2b584a056 100644
--- a/net/sunrpc/auth_null.c
+++ b/net/sunrpc/auth_null.c
@@ -10,7 +10,7 @@
 #include <linux/module.h>
 #include <linux/sunrpc/clnt.h>
 
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 # define RPCDBG_FACILITY	RPCDBG_AUTH
 #endif
 
@@ -138,7 +138,7 @@ struct rpc_cred null_cred = {
 	.cr_ops		= &null_credops,
 	.cr_count	= ATOMIC_INIT(1),
 	.cr_flags	= 1UL << RPCAUTH_CRED_UPTODATE,
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 	.cr_magic	= RPCAUTH_CRED_MAGIC,
 #endif
 };
diff --git a/net/sunrpc/auth_unix.c b/net/sunrpc/auth_unix.c
index d5d692366294..4feda2d0a833 100644
--- a/net/sunrpc/auth_unix.c
+++ b/net/sunrpc/auth_unix.c
@@ -25,7 +25,7 @@ struct unx_cred {
 
 #define UNX_WRITESLACK		(21 + (UNX_MAXNODENAME >> 2))
 
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 # define RPCDBG_FACILITY	RPCDBG_AUTH
 #endif
 
diff --git a/net/sunrpc/backchannel_rqst.c b/net/sunrpc/backchannel_rqst.c
index 9761a0da964d..651f49ab601f 100644
--- a/net/sunrpc/backchannel_rqst.c
+++ b/net/sunrpc/backchannel_rqst.c
@@ -27,7 +27,7 @@ SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 #include <linux/export.h>
 #include <linux/sunrpc/bc_xprt.h>
 
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 #define RPCDBG_FACILITY	RPCDBG_TRANS
 #endif
 
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index 9acd6ce88db7..36c64ef460cf 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -42,7 +42,7 @@
 #include "sunrpc.h"
 #include "netns.h"
 
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 # define RPCDBG_FACILITY	RPCDBG_CALL
 #endif
 
@@ -1396,7 +1396,7 @@ rpc_restart_call(struct rpc_task *task)
 }
 EXPORT_SYMBOL_GPL(rpc_restart_call);
 
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 static const char *rpc_proc_name(const struct rpc_task *task)
 {
 	const struct rpc_procinfo *proc = task->tk_msg.rpc_proc;
@@ -2421,7 +2421,7 @@ struct rpc_task *rpc_call_null(struct rpc_clnt *clnt, struct rpc_cred *cred, int
 }
 EXPORT_SYMBOL_GPL(rpc_call_null);
 
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 static void rpc_show_header(void)
 {
 	printk(KERN_INFO "-pid- flgs status -client- --rqstp- "
diff --git a/net/sunrpc/rpcb_clnt.c b/net/sunrpc/rpcb_clnt.c
index 1891a1022c17..05202012bcfc 100644
--- a/net/sunrpc/rpcb_clnt.c
+++ b/net/sunrpc/rpcb_clnt.c
@@ -32,7 +32,7 @@
 
 #include "netns.h"
 
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 # define RPCDBG_FACILITY	RPCDBG_BIND
 #endif
 
diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c
index fe3441abdbe5..574b2977fc4b 100644
--- a/net/sunrpc/sched.c
+++ b/net/sunrpc/sched.c
@@ -24,7 +24,7 @@
 
 #include "sunrpc.h"
 
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 #define RPCDBG_FACILITY		RPCDBG_SCHED
 #endif
 
@@ -258,7 +258,7 @@ static int rpc_wait_bit_killable(struct wait_bit_key *key)
 	return 0;
 }
 
-#if defined(RPC_DEBUG) || defined(RPC_TRACEPOINTS)
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG) || defined(RPC_TRACEPOINTS)
 static void rpc_task_set_debuginfo(struct rpc_task *task)
 {
 	static atomic_t rpc_pid;
diff --git a/net/sunrpc/sunrpc_syms.c b/net/sunrpc/sunrpc_syms.c
index cd30120de9e4..f632e476ab6c 100644
--- a/net/sunrpc/sunrpc_syms.c
+++ b/net/sunrpc/sunrpc_syms.c
@@ -97,7 +97,7 @@ init_sunrpc(void)
 	err = register_rpc_pipefs();
 	if (err)
 		goto out4;
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 	rpc_register_sysctl();
 #endif
 	svc_init_xprt_sock();	/* svc sock transport */
@@ -123,7 +123,7 @@ cleanup_sunrpc(void)
 	unregister_rpc_pipefs();
 	rpc_destroy_mempool();
 	unregister_pernet_subsys(&sunrpc_net_ops);
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 	rpc_unregister_sysctl();
 #endif
 	rcu_barrier(); /* Wait for completion of call_rcu()'s */
diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c
index 371a8bbb43d6..2783fd80c229 100644
--- a/net/sunrpc/svc.c
+++ b/net/sunrpc/svc.c
@@ -1042,7 +1042,7 @@ static void svc_unregister(const struct svc_serv *serv, struct net *net)
 /*
  * dprintk the given error with the address of the client that caused it.
  */
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 static __printf(2, 3)
 void svc_printk(struct svc_rqst *rqstp, const char *fmt, ...)
 {
diff --git a/net/sunrpc/sysctl.c b/net/sunrpc/sysctl.c
index c99c58e2ee66..887f0183b4c6 100644
--- a/net/sunrpc/sysctl.c
+++ b/net/sunrpc/sysctl.c
@@ -37,7 +37,7 @@ EXPORT_SYMBOL_GPL(nfsd_debug);
 unsigned int	nlm_debug;
 EXPORT_SYMBOL_GPL(nlm_debug);
 
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 
 static struct ctl_table_header *sunrpc_table_header;
 static struct ctl_table sunrpc_table[];
diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c
index 1b2e5e616cae..894d071426b2 100644
--- a/net/sunrpc/xprt.c
+++ b/net/sunrpc/xprt.c
@@ -57,7 +57,7 @@
  * Local variables
  */
 
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 # define RPCDBG_FACILITY	RPCDBG_XPRT
 #endif
 
diff --git a/net/sunrpc/xprtrdma/rpc_rdma.c b/net/sunrpc/xprtrdma/rpc_rdma.c
index 6166c985fe24..df01d124936c 100644
--- a/net/sunrpc/xprtrdma/rpc_rdma.c
+++ b/net/sunrpc/xprtrdma/rpc_rdma.c
@@ -49,11 +49,11 @@
 
 #include <linux/highmem.h>
 
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 # define RPCDBG_FACILITY	RPCDBG_TRANS
 #endif
 
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 static const char transfertypes[][12] = {
 	"pure inline",	/* no chunks */
 	" read chunk",	/* some argument via rdma read */
diff --git a/net/sunrpc/xprtrdma/transport.c b/net/sunrpc/xprtrdma/transport.c
index 6a4615dd0261..ef58ebadb3ae 100644
--- a/net/sunrpc/xprtrdma/transport.c
+++ b/net/sunrpc/xprtrdma/transport.c
@@ -55,7 +55,7 @@
 
 #include "xprt_rdma.h"
 
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 # define RPCDBG_FACILITY	RPCDBG_TRANS
 #endif
 
@@ -75,7 +75,7 @@ static unsigned int xprt_rdma_inline_write_padding;
 static unsigned int xprt_rdma_memreg_strategy = RPCRDMA_FRMR;
                 int xprt_rdma_pad_optimize = 0;
 
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 
 static unsigned int min_slot_table_size = RPCRDMA_MIN_SLOT_TABLE;
 static unsigned int max_slot_table_size = RPCRDMA_MAX_SLOT_TABLE;
@@ -705,7 +705,7 @@ static void __exit xprt_rdma_cleanup(void)
 	int rc;
 
 	dprintk("RPCRDMA Module Removed, deregister RPC RDMA transport\n");
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 	if (sunrpc_table_header) {
 		unregister_sysctl_table(sunrpc_table_header);
 		sunrpc_table_header = NULL;
@@ -736,7 +736,7 @@ static int __init xprt_rdma_init(void)
 	dprintk("\tPadding %d\n\tMemreg %d\n",
 		xprt_rdma_inline_write_padding, xprt_rdma_memreg_strategy);
 
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 	if (!sunrpc_table_header)
 		sunrpc_table_header = register_sysctl_table(sunrpc_table);
 #endif
diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c
index 61c41298b4ea..b92b04083e40 100644
--- a/net/sunrpc/xprtrdma/verbs.c
+++ b/net/sunrpc/xprtrdma/verbs.c
@@ -57,7 +57,7 @@
  * Globals/Macros
  */
 
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 # define RPCDBG_FACILITY	RPCDBG_TRANS
 #endif
 
@@ -313,7 +313,7 @@ rpcrdma_flush_cqs(struct rpcrdma_ep *ep)
 	rpcrdma_sendcq_upcall(ep->rep_attr.send_cq, ep);
 }
 
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 static const char * const conn[] = {
 	"address resolved",
 	"address error",
@@ -344,7 +344,7 @@ rpcrdma_conn_upcall(struct rdma_cm_id *id, struct rdma_cm_event *event)
 	struct rpcrdma_xprt *xprt = id->context;
 	struct rpcrdma_ia *ia = &xprt->rx_ia;
 	struct rpcrdma_ep *ep = &xprt->rx_ep;
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 	struct sockaddr_in *addr = (struct sockaddr_in *) &ep->rep_remote_addr;
 #endif
 	struct ib_qp_attr attr;
@@ -408,7 +408,7 @@ connected:
 		break;
 	}
 
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 	if (connstate == 1) {
 		int ird = attr.max_dest_rd_atomic;
 		int tird = ep->rep_remote_cma.responder_resources;
diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c
index 31c015196a29..87ce7e8bb8dc 100644
--- a/net/sunrpc/xprtsock.c
+++ b/net/sunrpc/xprtsock.c
@@ -75,7 +75,7 @@ static unsigned int xs_tcp_fin_timeout __read_mostly = XS_TCP_LINGER_TO;
  * someone else's file names!
  */
 
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 
 static unsigned int min_slot_table_size = RPC_MIN_SLOT_TABLE;
 static unsigned int max_slot_table_size = RPC_MAX_SLOT_TABLE;
@@ -186,7 +186,7 @@ static struct ctl_table sunrpc_table[] = {
  */
 #define XS_IDLE_DISC_TO		(5U * 60 * HZ)
 
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 # undef  RPC_DEBUG_DATA
 # define RPCDBG_FACILITY	RPCDBG_TRANS
 #endif
@@ -2991,7 +2991,7 @@ static struct xprt_class	xs_bc_tcp_transport = {
  */
 int init_socket_xprt(void)
 {
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 	if (!sunrpc_table_header)
 		sunrpc_table_header = register_sysctl_table(sunrpc_table);
 #endif
@@ -3010,7 +3010,7 @@ int init_socket_xprt(void)
  */
 void cleanup_socket_xprt(void)
 {
-#ifdef RPC_DEBUG
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 	if (sunrpc_table_header) {
 		unregister_sysctl_table(sunrpc_table_header);
 		sunrpc_table_header = NULL;
-- 
cgit v1.2.3


From 1306729b0d4f4a0bd0d098711ed3d938dc5a1a28 Mon Sep 17 00:00:00 2001
From: Jeff Layton <jlayton@primarydata.com>
Date: Mon, 17 Nov 2014 16:58:05 -0500
Subject: sunrpc: eliminate RPC_TRACEPOINTS

It's always set to the same value as CONFIG_TRACEPOINTS, so we can just
use that instead.

Signed-off-by: Jeff Layton <jlayton@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
---
 net/sunrpc/sched.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'net/sunrpc')

diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c
index 574b2977fc4b..d20f2329eea3 100644
--- a/net/sunrpc/sched.c
+++ b/net/sunrpc/sched.c
@@ -258,7 +258,7 @@ static int rpc_wait_bit_killable(struct wait_bit_key *key)
 	return 0;
 }
 
-#if IS_ENABLED(CONFIG_SUNRPC_DEBUG) || defined(RPC_TRACEPOINTS)
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG) || IS_ENABLED(CONFIG_TRACEPOINTS)
 static void rpc_task_set_debuginfo(struct rpc_task *task)
 {
 	static atomic_t rpc_pid;
-- 
cgit v1.2.3


From 92b98361f119a6919061d067c6584de9ae2de9f4 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Sat, 8 Nov 2014 20:14:12 -0500
Subject: xprtrdma: Return an errno from rpcrdma_register_external()

The RPC/RDMA send_request method and the chunk registration code
expects an errno from the registration function. This allows
the upper layers to distinguish between a recoverable failure
(for example, temporary memory exhaustion) and a hard failure
(for example, a bug in the registration logic).

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
---
 net/sunrpc/xprtrdma/verbs.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'net/sunrpc')

diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c
index 61c41298b4ea..6ea29420f0ac 100644
--- a/net/sunrpc/xprtrdma/verbs.c
+++ b/net/sunrpc/xprtrdma/verbs.c
@@ -1918,10 +1918,10 @@ rpcrdma_register_external(struct rpcrdma_mr_seg *seg,
 		break;
 
 	default:
-		return -1;
+		return -EIO;
 	}
 	if (rc)
-		return -1;
+		return rc;
 
 	return nsegs;
 }
-- 
cgit v1.2.3


From e7104a2a96069975d489c60a30564372c6273a85 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Sat, 8 Nov 2014 20:14:20 -0500
Subject: xprtrdma: Cap req_cqinit

Recent work made FRMR registration and invalidation completions
unsignaled. This greatly reduces the adapter interrupt rate.

Every so often, however, a posted send Work Request is allowed to
signal. Otherwise, the provider's Work Queue will wrap and the
workload will hang.

The number of Work Requests that are allowed to remain unsignaled is
determined by the value of req_cqinit. Currently, this is set to the
size of the send Work Queue divided by two, minus 1.

For FRMR, the send Work Queue is the maximum number of concurrent
RPCs (currently 32) times the maximum number of Work Requests an
RPC might use (currently 7, though some adapters may need more).

For mlx4, this is 224 entries. This leaves completion signaling
disabled for 111 send Work Requests.

Some providers hold back dispatching Work Requests until a CQE is
generated.  If completions are disabled, then no CQEs are generated
for quite some time, and that can stall the Work Queue.

I've seen this occur running xfstests generic/113 over NFSv4, where
eventually, posting a FAST_REG_MR Work Request fails with -ENOMEM
because the Work Queue has overflowed. The connection is dropped
and re-established.

Cap the rep_cqinit setting so completions are not left turned off
for too long.

BugLink: https://bugzilla.linux-nfs.org/show_bug.cgi?id=269
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
---
 net/sunrpc/xprtrdma/verbs.c     | 4 +++-
 net/sunrpc/xprtrdma/xprt_rdma.h | 6 ++++++
 2 files changed, 9 insertions(+), 1 deletion(-)

(limited to 'net/sunrpc')

diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c
index 6ea29420f0ac..af45cf390126 100644
--- a/net/sunrpc/xprtrdma/verbs.c
+++ b/net/sunrpc/xprtrdma/verbs.c
@@ -733,7 +733,9 @@ rpcrdma_ep_create(struct rpcrdma_ep *ep, struct rpcrdma_ia *ia,
 
 	/* set trigger for requesting send completion */
 	ep->rep_cqinit = ep->rep_attr.cap.max_send_wr/2 - 1;
-	if (ep->rep_cqinit <= 2)
+	if (ep->rep_cqinit > RPCRDMA_MAX_UNSIGNALED_SENDS)
+		ep->rep_cqinit = RPCRDMA_MAX_UNSIGNALED_SENDS;
+	else if (ep->rep_cqinit <= 2)
 		ep->rep_cqinit = 0;
 	INIT_CQCOUNT(ep);
 	ep->rep_ia = ia;
diff --git a/net/sunrpc/xprtrdma/xprt_rdma.h b/net/sunrpc/xprtrdma/xprt_rdma.h
index ac7fc9a31342..b799041b75bf 100644
--- a/net/sunrpc/xprtrdma/xprt_rdma.h
+++ b/net/sunrpc/xprtrdma/xprt_rdma.h
@@ -97,6 +97,12 @@ struct rpcrdma_ep {
 	struct ib_wc		rep_recv_wcs[RPCRDMA_POLLSIZE];
 };
 
+/*
+ * Force a signaled SEND Work Request every so often,
+ * in case the provider needs to do some housekeeping.
+ */
+#define RPCRDMA_MAX_UNSIGNALED_SENDS	(32)
+
 #define INIT_CQCOUNT(ep) atomic_set(&(ep)->rep_cqcount, (ep)->rep_cqinit)
 #define DECR_CQCOUNT(ep) atomic_sub_return(1, &(ep)->rep_cqcount)
 
-- 
cgit v1.2.3


From 467c9674bccc073684ee34f4bd205cf1b135d76e Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Sat, 8 Nov 2014 20:14:29 -0500
Subject: xprtrdma: unmap all FMRs during transport disconnect

When using RPCRDMA_MTHCAFMR memory registration, after a few
transport disconnect / reconnect cycles, ib_map_phys_fmr() starts to
return EINVAL because the provider has exhausted its map pool.

Make sure that all FMRs are unmapped during transport disconnect,
and that ->send_request remarshals them during an RPC retransmit.
This resets the transport's MRs to ensure that none are leaked
during a disconnect.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
---
 net/sunrpc/xprtrdma/transport.c |  2 +-
 net/sunrpc/xprtrdma/verbs.c     | 42 ++++++++++++++++++++++++++++++++++++++++-
 2 files changed, 42 insertions(+), 2 deletions(-)

(limited to 'net/sunrpc')

diff --git a/net/sunrpc/xprtrdma/transport.c b/net/sunrpc/xprtrdma/transport.c
index 6a4615dd0261..cfe9a810e6bc 100644
--- a/net/sunrpc/xprtrdma/transport.c
+++ b/net/sunrpc/xprtrdma/transport.c
@@ -599,7 +599,7 @@ xprt_rdma_send_request(struct rpc_task *task)
 
 	if (req->rl_niovs == 0)
 		rc = rpcrdma_marshal_req(rqst);
-	else if (r_xprt->rx_ia.ri_memreg_strategy == RPCRDMA_FRMR)
+	else if (r_xprt->rx_ia.ri_memreg_strategy != RPCRDMA_ALLPHYSICAL)
 		rc = rpcrdma_marshal_chunks(rqst, 0);
 	if (rc < 0)
 		goto failed_marshal;
diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c
index af45cf390126..3c88276090e3 100644
--- a/net/sunrpc/xprtrdma/verbs.c
+++ b/net/sunrpc/xprtrdma/verbs.c
@@ -62,6 +62,7 @@
 #endif
 
 static void rpcrdma_reset_frmrs(struct rpcrdma_ia *);
+static void rpcrdma_reset_fmrs(struct rpcrdma_ia *);
 
 /*
  * internal functions
@@ -868,8 +869,19 @@ retry:
 		rpcrdma_ep_disconnect(ep, ia);
 		rpcrdma_flush_cqs(ep);
 
-		if (ia->ri_memreg_strategy == RPCRDMA_FRMR)
+		switch (ia->ri_memreg_strategy) {
+		case RPCRDMA_FRMR:
 			rpcrdma_reset_frmrs(ia);
+			break;
+		case RPCRDMA_MTHCAFMR:
+			rpcrdma_reset_fmrs(ia);
+			break;
+		case RPCRDMA_ALLPHYSICAL:
+			break;
+		default:
+			rc = -EIO;
+			goto out;
+		}
 
 		xprt = container_of(ia, struct rpcrdma_xprt, rx_ia);
 		id = rpcrdma_create_id(xprt, ia,
@@ -1289,6 +1301,34 @@ rpcrdma_buffer_destroy(struct rpcrdma_buffer *buf)
 	kfree(buf->rb_pool);
 }
 
+/* After a disconnect, unmap all FMRs.
+ *
+ * This is invoked only in the transport connect worker in order
+ * to serialize with rpcrdma_register_fmr_external().
+ */
+static void
+rpcrdma_reset_fmrs(struct rpcrdma_ia *ia)
+{
+	struct rpcrdma_xprt *r_xprt =
+				container_of(ia, struct rpcrdma_xprt, rx_ia);
+	struct rpcrdma_buffer *buf = &r_xprt->rx_buf;
+	struct list_head *pos;
+	struct rpcrdma_mw *r;
+	LIST_HEAD(l);
+	int rc;
+
+	list_for_each(pos, &buf->rb_all) {
+		r = list_entry(pos, struct rpcrdma_mw, mw_all);
+
+		INIT_LIST_HEAD(&l);
+		list_add(&r->r.fmr->list, &l);
+		rc = ib_unmap_fmr(&l);
+		if (rc)
+			dprintk("RPC:       %s: ib_unmap_fmr failed %i\n",
+				__func__, rc);
+	}
+}
+
 /* After a disconnect, a flushed FAST_REG_MR can leave an FRMR in
  * an unusable state. Find FRMRs in this state and dereg / reg
  * each.  FRMRs that are VALID and attached to an rpcrdma_req are
-- 
cgit v1.2.3


From f1a03b76fecdb23983e2ad7e817e98d093ece9f7 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Sat, 8 Nov 2014 20:14:37 -0500
Subject: xprtrdma: Refactor tasklet scheduling

Restore the separate function that schedules the reply handling
tasklet. I need to call it from two different paths.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
---
 net/sunrpc/xprtrdma/verbs.c | 17 ++++++++++++-----
 1 file changed, 12 insertions(+), 5 deletions(-)

(limited to 'net/sunrpc')

diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c
index 3c88276090e3..478b2fda83a6 100644
--- a/net/sunrpc/xprtrdma/verbs.c
+++ b/net/sunrpc/xprtrdma/verbs.c
@@ -106,6 +106,17 @@ rpcrdma_run_tasklet(unsigned long data)
 
 static DECLARE_TASKLET(rpcrdma_tasklet_g, rpcrdma_run_tasklet, 0UL);
 
+static void
+rpcrdma_schedule_tasklet(struct list_head *sched_list)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&rpcrdma_tk_lock_g, flags);
+	list_splice_tail(sched_list, &rpcrdma_tasklets_g);
+	spin_unlock_irqrestore(&rpcrdma_tk_lock_g, flags);
+	tasklet_schedule(&rpcrdma_tasklet_g);
+}
+
 static void
 rpcrdma_qp_async_error_upcall(struct ib_event *event, void *context)
 {
@@ -244,7 +255,6 @@ rpcrdma_recvcq_poll(struct ib_cq *cq, struct rpcrdma_ep *ep)
 	struct list_head sched_list;
 	struct ib_wc *wcs;
 	int budget, count, rc;
-	unsigned long flags;
 
 	INIT_LIST_HEAD(&sched_list);
 	budget = RPCRDMA_WC_BUDGET / RPCRDMA_POLLSIZE;
@@ -262,10 +272,7 @@ rpcrdma_recvcq_poll(struct ib_cq *cq, struct rpcrdma_ep *ep)
 	rc = 0;
 
 out_schedule:
-	spin_lock_irqsave(&rpcrdma_tk_lock_g, flags);
-	list_splice_tail(&sched_list, &rpcrdma_tasklets_g);
-	spin_unlock_irqrestore(&rpcrdma_tk_lock_g, flags);
-	tasklet_schedule(&rpcrdma_tasklet_g);
+	rpcrdma_schedule_tasklet(&sched_list);
 	return rc;
 }
 
-- 
cgit v1.2.3


From 5c166bef4fa138da45d756dc77a8cb29fced1714 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Sat, 8 Nov 2014 20:14:45 -0500
Subject: xprtrdma: Re-write rpcrdma_flush_cqs()

Currently rpcrdma_flush_cqs() attempts to avoid code duplication,
and simply invokes rpcrdma_recvcq_upcall and rpcrdma_sendcq_upcall.

1. rpcrdma_flush_cqs() can run concurrently with provider upcalls.
   Both flush_cqs() and the upcalls were invoking ib_poll_cq() in
   different threads using the same wc buffers (ep->rep_recv_wcs
   and ep->rep_send_wcs), added by commit 1c00dd077654 ("xprtrmda:
   Reduce calls to ib_poll_cq() in completion handlers").

   During transport disconnect processing, this sometimes resulted
   in the same reply getting added to the rpcrdma_tasklets_g list
   more than once, which corrupted the list.

2. The upcall functions drain only a limited number of CQEs,
   thanks to the poll budget added by commit 8301a2c047cc
   ("xprtrdma: Limit work done by completion handler").

Fixes: a7bc211ac926 ("xprtrdma: On disconnect, don't ignore ... ")
BugLink: https://bugzilla.linux-nfs.org/show_bug.cgi?id=276
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
---
 net/sunrpc/xprtrdma/verbs.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

(limited to 'net/sunrpc')

diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c
index 478b2fda83a6..e6ac9643fe56 100644
--- a/net/sunrpc/xprtrdma/verbs.c
+++ b/net/sunrpc/xprtrdma/verbs.c
@@ -317,8 +317,15 @@ rpcrdma_recvcq_upcall(struct ib_cq *cq, void *cq_context)
 static void
 rpcrdma_flush_cqs(struct rpcrdma_ep *ep)
 {
-	rpcrdma_recvcq_upcall(ep->rep_attr.recv_cq, ep);
-	rpcrdma_sendcq_upcall(ep->rep_attr.send_cq, ep);
+	struct ib_wc wc;
+	LIST_HEAD(sched_list);
+
+	while (ib_poll_cq(ep->rep_attr.recv_cq, 1, &wc) > 0)
+		rpcrdma_recvcq_process_wc(&wc, &sched_list);
+	if (!list_empty(&sched_list))
+		rpcrdma_schedule_tasklet(&sched_list);
+	while (ib_poll_cq(ep->rep_attr.send_cq, 1, &wc) > 0)
+		rpcrdma_sendcq_process_wc(&wc);
 }
 
 #ifdef RPC_DEBUG
-- 
cgit v1.2.3


From d5440e27d3e572272db957d6b9661a902b7c339f Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Sat, 8 Nov 2014 20:14:53 -0500
Subject: xprtrdma: Enable pad optimization

The Linux NFS/RDMA server used to reject NFSv3 WRITE requests when
pad optimization was enabled. That bug was fixed by commit
e560e3b510d2 ("svcrdma: Add zero padding if the client doesn't send
it").

We can now enable pad optimization on the client, which helps
performance and is supported now by both Linux and Solaris servers.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
---
 net/sunrpc/xprtrdma/transport.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'net/sunrpc')

diff --git a/net/sunrpc/xprtrdma/transport.c b/net/sunrpc/xprtrdma/transport.c
index cfe9a810e6bc..8ed25760764b 100644
--- a/net/sunrpc/xprtrdma/transport.c
+++ b/net/sunrpc/xprtrdma/transport.c
@@ -73,7 +73,7 @@ static unsigned int xprt_rdma_max_inline_read = RPCRDMA_DEF_INLINE;
 static unsigned int xprt_rdma_max_inline_write = RPCRDMA_DEF_INLINE;
 static unsigned int xprt_rdma_inline_write_padding;
 static unsigned int xprt_rdma_memreg_strategy = RPCRDMA_FRMR;
-                int xprt_rdma_pad_optimize = 0;
+		int xprt_rdma_pad_optimize = 1;
 
 #ifdef RPC_DEBUG
 
-- 
cgit v1.2.3


From 7ff11de1bae02a41cac6503f858218ac1b9a3cbe Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Sat, 8 Nov 2014 20:15:01 -0500
Subject: xprtrdma: Display async errors

An async error upcall is a hard error, and should be reported in
the system log.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
---
 net/sunrpc/xprtrdma/verbs.c | 36 ++++++++++++++++++++++++++++++++----
 1 file changed, 32 insertions(+), 4 deletions(-)

(limited to 'net/sunrpc')

diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c
index e6ac9643fe56..5783c1a55b09 100644
--- a/net/sunrpc/xprtrdma/verbs.c
+++ b/net/sunrpc/xprtrdma/verbs.c
@@ -106,6 +106,32 @@ rpcrdma_run_tasklet(unsigned long data)
 
 static DECLARE_TASKLET(rpcrdma_tasklet_g, rpcrdma_run_tasklet, 0UL);
 
+static const char * const async_event[] = {
+	"CQ error",
+	"QP fatal error",
+	"QP request error",
+	"QP access error",
+	"communication established",
+	"send queue drained",
+	"path migration successful",
+	"path mig error",
+	"device fatal error",
+	"port active",
+	"port error",
+	"LID change",
+	"P_key change",
+	"SM change",
+	"SRQ error",
+	"SRQ limit reached",
+	"last WQE reached",
+	"client reregister",
+	"GID change",
+};
+
+#define ASYNC_MSG(status)					\
+	((status) < ARRAY_SIZE(async_event) ?			\
+		async_event[(status)] : "unknown async error")
+
 static void
 rpcrdma_schedule_tasklet(struct list_head *sched_list)
 {
@@ -122,8 +148,9 @@ rpcrdma_qp_async_error_upcall(struct ib_event *event, void *context)
 {
 	struct rpcrdma_ep *ep = context;
 
-	dprintk("RPC:       %s: QP error %X on device %s ep %p\n",
-		__func__, event->event, event->device->name, context);
+	pr_err("RPC:       %s: %s on device %s ep %p\n",
+	       __func__, ASYNC_MSG(event->event),
+		event->device->name, context);
 	if (ep->rep_connected == 1) {
 		ep->rep_connected = -EIO;
 		ep->rep_func(ep);
@@ -136,8 +163,9 @@ rpcrdma_cq_async_error_upcall(struct ib_event *event, void *context)
 {
 	struct rpcrdma_ep *ep = context;
 
-	dprintk("RPC:       %s: CQ error %X on device %s ep %p\n",
-		__func__, event->event, event->device->name, context);
+	pr_err("RPC:       %s: %s on device %s ep %p\n",
+	       __func__, ASYNC_MSG(event->event),
+		event->device->name, context);
 	if (ep->rep_connected == 1) {
 		ep->rep_connected = -EIO;
 		ep->rep_func(ep);
-- 
cgit v1.2.3


From edef1297f33a4546559d905457b435a5ea160bab Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Sat, 8 Nov 2014 20:15:09 -0500
Subject: SUNRPC: serialize iostats updates

Occasionally mountstats reports a negative retransmission rate.
Ensure that two RPCs completing concurrently don't confuse the sums
in the transport's op_metrics array.

Since pNFS filelayout can invoke rpc_count_iostats() on another
transport from xprt_release(), we can't rely on simply holding the
transport_lock in xprt_release(). There's nothing for it but hard
serialization. One spin lock per RPC operation should make this as
painless as it can be.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
---
 net/sunrpc/stats.c | 21 ++++++++++++++++-----
 1 file changed, 16 insertions(+), 5 deletions(-)

(limited to 'net/sunrpc')

diff --git a/net/sunrpc/stats.c b/net/sunrpc/stats.c
index 54530490944e..9711a155bc50 100644
--- a/net/sunrpc/stats.c
+++ b/net/sunrpc/stats.c
@@ -116,7 +116,15 @@ EXPORT_SYMBOL_GPL(svc_seq_show);
  */
 struct rpc_iostats *rpc_alloc_iostats(struct rpc_clnt *clnt)
 {
-	return kcalloc(clnt->cl_maxproc, sizeof(struct rpc_iostats), GFP_KERNEL);
+	struct rpc_iostats *stats;
+	int i;
+
+	stats = kcalloc(clnt->cl_maxproc, sizeof(*stats), GFP_KERNEL);
+	if (stats) {
+		for (i = 0; i < clnt->cl_maxproc; i++)
+			spin_lock_init(&stats[i].om_lock);
+	}
+	return stats;
 }
 EXPORT_SYMBOL_GPL(rpc_alloc_iostats);
 
@@ -135,20 +143,21 @@ EXPORT_SYMBOL_GPL(rpc_free_iostats);
  * rpc_count_iostats - tally up per-task stats
  * @task: completed rpc_task
  * @stats: array of stat structures
- *
- * Relies on the caller for serialization.
  */
 void rpc_count_iostats(const struct rpc_task *task, struct rpc_iostats *stats)
 {
 	struct rpc_rqst *req = task->tk_rqstp;
 	struct rpc_iostats *op_metrics;
-	ktime_t delta;
+	ktime_t delta, now;
 
 	if (!stats || !req)
 		return;
 
+	now = ktime_get();
 	op_metrics = &stats[task->tk_msg.rpc_proc->p_statidx];
 
+	spin_lock(&op_metrics->om_lock);
+
 	op_metrics->om_ops++;
 	op_metrics->om_ntrans += req->rq_ntrans;
 	op_metrics->om_timeouts += task->tk_timeouts;
@@ -161,8 +170,10 @@ void rpc_count_iostats(const struct rpc_task *task, struct rpc_iostats *stats)
 
 	op_metrics->om_rtt = ktime_add(op_metrics->om_rtt, req->rq_rtt);
 
-	delta = ktime_sub(ktime_get(), task->tk_start);
+	delta = ktime_sub(now, task->tk_start);
 	op_metrics->om_execute = ktime_add(op_metrics->om_execute, delta);
+
+	spin_unlock(&op_metrics->om_lock);
 }
 EXPORT_SYMBOL_GPL(rpc_count_iostats);
 
-- 
cgit v1.2.3


From b4b9d2ccf0be61c69213f6ae4e33377c05194ef4 Mon Sep 17 00:00:00 2001
From: Jeff Layton <jlayton@primarydata.com>
Date: Wed, 26 Nov 2014 14:44:43 -0500
Subject: sunrpc: add debugfs file for displaying client rpc_task queue

It's possible to get a dump of the RPC task queue by writing a value to
/proc/sys/sunrpc/rpc_debug. If you write any value to that file, you get
a dump of the RPC client task list into the log buffer. This is a rather
inconvenient interface however, and makes it hard to get immediate info
about the task queue.

Add a new directory hierarchy under debugfs:

    sunrpc/
        rpc_clnt/
            <clientid>/

Within each clientid directory we create a new "tasks" file that will
dump info similar to what shows up in the log buffer, but with a few
small differences -- we avoid printing raw kernel addresses in favor of
symbolic names and the XID is also displayed.

Signed-off-by: Jeff Layton <jlayton@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
---
 net/sunrpc/Kconfig       |   1 +
 net/sunrpc/Makefile      |   1 +
 net/sunrpc/clnt.c        |  10 ++-
 net/sunrpc/debugfs.c     | 191 +++++++++++++++++++++++++++++++++++++++++++++++
 net/sunrpc/sunrpc_syms.c |   8 ++
 5 files changed, 210 insertions(+), 1 deletion(-)
 create mode 100644 net/sunrpc/debugfs.c

(limited to 'net/sunrpc')

diff --git a/net/sunrpc/Kconfig b/net/sunrpc/Kconfig
index 0754d0f466d2..fb78117b896c 100644
--- a/net/sunrpc/Kconfig
+++ b/net/sunrpc/Kconfig
@@ -35,6 +35,7 @@ config RPCSEC_GSS_KRB5
 config SUNRPC_DEBUG
 	bool "RPC: Enable dprintk debugging"
 	depends on SUNRPC && SYSCTL
+	select DEBUG_FS
 	help
 	  This option enables a sysctl-based debugging interface
 	  that is be used by the 'rpcdebug' utility to turn on or off
diff --git a/net/sunrpc/Makefile b/net/sunrpc/Makefile
index e5a7a1cac8f3..15e6f6c23c5d 100644
--- a/net/sunrpc/Makefile
+++ b/net/sunrpc/Makefile
@@ -14,6 +14,7 @@ sunrpc-y := clnt.o xprt.o socklib.o xprtsock.o sched.o \
 	    addr.o rpcb_clnt.o timer.o xdr.o \
 	    sunrpc_syms.o cache.o rpc_pipe.o \
 	    svc_xprt.o
+sunrpc-$(CONFIG_SUNRPC_DEBUG) += debugfs.o
 sunrpc-$(CONFIG_SUNRPC_BACKCHANNEL) += backchannel_rqst.o bc_svc.o
 sunrpc-$(CONFIG_PROC_FS) += stats.o
 sunrpc-$(CONFIG_SYSCTL) += sysctl.o
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index 36c64ef460cf..05da12a33945 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -305,6 +305,10 @@ static int rpc_client_register(struct rpc_clnt *clnt,
 	struct super_block *pipefs_sb;
 	int err;
 
+	err = rpc_clnt_debugfs_register(clnt);
+	if (err)
+		return err;
+
 	pipefs_sb = rpc_get_sb_net(net);
 	if (pipefs_sb) {
 		err = rpc_setup_pipedir(pipefs_sb, clnt);
@@ -331,6 +335,7 @@ err_auth:
 out:
 	if (pipefs_sb)
 		rpc_put_sb_net(net);
+	rpc_clnt_debugfs_unregister(clnt);
 	return err;
 }
 
@@ -670,6 +675,7 @@ int rpc_switch_client_transport(struct rpc_clnt *clnt,
 
 	rpc_unregister_client(clnt);
 	__rpc_clnt_remove_pipedir(clnt);
+	rpc_clnt_debugfs_unregister(clnt);
 
 	/*
 	 * A new transport was created.  "clnt" therefore
@@ -771,6 +777,7 @@ rpc_free_client(struct rpc_clnt *clnt)
 			rcu_dereference(clnt->cl_xprt)->servername);
 	if (clnt->cl_parent != clnt)
 		parent = clnt->cl_parent;
+	rpc_clnt_debugfs_unregister(clnt);
 	rpc_clnt_remove_pipedir(clnt);
 	rpc_unregister_client(clnt);
 	rpc_free_iostats(clnt->cl_metrics);
@@ -1397,7 +1404,8 @@ rpc_restart_call(struct rpc_task *task)
 EXPORT_SYMBOL_GPL(rpc_restart_call);
 
 #if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
-static const char *rpc_proc_name(const struct rpc_task *task)
+const char
+*rpc_proc_name(const struct rpc_task *task)
 {
 	const struct rpc_procinfo *proc = task->tk_msg.rpc_proc;
 
diff --git a/net/sunrpc/debugfs.c b/net/sunrpc/debugfs.c
new file mode 100644
index 000000000000..3d7745683ca3
--- /dev/null
+++ b/net/sunrpc/debugfs.c
@@ -0,0 +1,191 @@
+/**
+ * debugfs interface for sunrpc
+ *
+ * (c) 2014 Jeff Layton <jlayton@primarydata.com>
+ */
+
+#include <linux/debugfs.h>
+#include <linux/sunrpc/sched.h>
+#include <linux/sunrpc/clnt.h>
+#include "netns.h"
+
+static struct dentry *topdir;
+static struct dentry *rpc_clnt_dir;
+
+struct rpc_clnt_iter {
+	struct rpc_clnt	*clnt;
+	loff_t		pos;
+};
+
+static int
+tasks_show(struct seq_file *f, void *v)
+{
+	u32 xid = 0;
+	struct rpc_task *task = v;
+	struct rpc_clnt *clnt = task->tk_client;
+	const char *rpc_waitq = "none";
+
+	if (RPC_IS_QUEUED(task))
+		rpc_waitq = rpc_qname(task->tk_waitqueue);
+
+	if (task->tk_rqstp)
+		xid = be32_to_cpu(task->tk_rqstp->rq_xid);
+
+	seq_printf(f, "%5u %04x %6d 0x%x 0x%x %8ld %ps %sv%u %s a:%ps q:%s\n",
+		task->tk_pid, task->tk_flags, task->tk_status,
+		clnt->cl_clid, xid, task->tk_timeout, task->tk_ops,
+		clnt->cl_program->name, clnt->cl_vers, rpc_proc_name(task),
+		task->tk_action, rpc_waitq);
+	return 0;
+}
+
+static void *
+tasks_start(struct seq_file *f, loff_t *ppos)
+	__acquires(&clnt->cl_lock)
+{
+	struct rpc_clnt_iter *iter = f->private;
+	loff_t pos = *ppos;
+	struct rpc_clnt *clnt = iter->clnt;
+	struct rpc_task *task;
+
+	iter->pos = pos + 1;
+	spin_lock(&clnt->cl_lock);
+	list_for_each_entry(task, &clnt->cl_tasks, tk_task)
+		if (pos-- == 0)
+			return task;
+	return NULL;
+}
+
+static void *
+tasks_next(struct seq_file *f, void *v, loff_t *pos)
+{
+	struct rpc_clnt_iter *iter = f->private;
+	struct rpc_clnt *clnt = iter->clnt;
+	struct rpc_task *task = v;
+	struct list_head *next = task->tk_task.next;
+
+	++iter->pos;
+	++*pos;
+
+	/* If there's another task on list, return it */
+	if (next == &clnt->cl_tasks)
+		return NULL;
+	return list_entry(next, struct rpc_task, tk_task);
+}
+
+static void
+tasks_stop(struct seq_file *f, void *v)
+	__releases(&clnt->cl_lock)
+{
+	struct rpc_clnt_iter *iter = f->private;
+	struct rpc_clnt *clnt = iter->clnt;
+
+	spin_unlock(&clnt->cl_lock);
+}
+
+static const struct seq_operations tasks_seq_operations = {
+	.start	= tasks_start,
+	.next	= tasks_next,
+	.stop	= tasks_stop,
+	.show	= tasks_show,
+};
+
+static int tasks_open(struct inode *inode, struct file *filp)
+{
+	int ret = seq_open_private(filp, &tasks_seq_operations,
+					sizeof(struct rpc_clnt_iter));
+
+	if (!ret) {
+		struct seq_file *seq = filp->private_data;
+		struct rpc_clnt_iter *iter = seq->private;
+
+		iter->clnt = inode->i_private;
+
+		if (!atomic_inc_not_zero(&iter->clnt->cl_count)) {
+			seq_release_private(inode, filp);
+			ret = -EINVAL;
+		}
+	}
+
+	return ret;
+}
+
+static int
+tasks_release(struct inode *inode, struct file *filp)
+{
+	struct seq_file *seq = filp->private_data;
+	struct rpc_clnt_iter *iter = seq->private;
+
+	rpc_release_client(iter->clnt);
+	return seq_release_private(inode, filp);
+}
+
+static const struct file_operations tasks_fops = {
+	.owner		= THIS_MODULE,
+	.open		= tasks_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= tasks_release,
+};
+
+int
+rpc_clnt_debugfs_register(struct rpc_clnt *clnt)
+{
+	int len;
+	char name[9]; /* 8 for hex digits + NULL terminator */
+
+	/* Already registered? */
+	if (clnt->cl_debugfs)
+		return 0;
+
+	len = snprintf(name, sizeof(name), "%x", clnt->cl_clid);
+	if (len >= sizeof(name))
+		return -EINVAL;
+
+	/* make the per-client dir */
+	clnt->cl_debugfs = debugfs_create_dir(name, rpc_clnt_dir);
+	if (!clnt->cl_debugfs)
+		return -ENOMEM;
+
+	/* make tasks file */
+	if (!debugfs_create_file("tasks", S_IFREG | S_IRUSR, clnt->cl_debugfs,
+				 clnt, &tasks_fops)) {
+		debugfs_remove_recursive(clnt->cl_debugfs);
+		clnt->cl_debugfs = NULL;
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+void
+rpc_clnt_debugfs_unregister(struct rpc_clnt *clnt)
+{
+	debugfs_remove_recursive(clnt->cl_debugfs);
+	clnt->cl_debugfs = NULL;
+}
+
+void __exit
+sunrpc_debugfs_exit(void)
+{
+	debugfs_remove_recursive(topdir);
+}
+
+int __init
+sunrpc_debugfs_init(void)
+{
+	topdir = debugfs_create_dir("sunrpc", NULL);
+	if (!topdir)
+		goto out;
+
+	rpc_clnt_dir = debugfs_create_dir("rpc_clnt", topdir);
+	if (!rpc_clnt_dir)
+		goto out_remove;
+
+	return 0;
+out_remove:
+	debugfs_remove_recursive(topdir);
+	topdir = NULL;
+out:
+	return -ENOMEM;
+}
diff --git a/net/sunrpc/sunrpc_syms.c b/net/sunrpc/sunrpc_syms.c
index f632e476ab6c..e37fbed87956 100644
--- a/net/sunrpc/sunrpc_syms.c
+++ b/net/sunrpc/sunrpc_syms.c
@@ -97,6 +97,11 @@ init_sunrpc(void)
 	err = register_rpc_pipefs();
 	if (err)
 		goto out4;
+
+	err = sunrpc_debugfs_init();
+	if (err)
+		goto out5;
+
 #if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 	rpc_register_sysctl();
 #endif
@@ -104,6 +109,8 @@ init_sunrpc(void)
 	init_socket_xprt();	/* clnt sock transport */
 	return 0;
 
+out5:
+	unregister_rpc_pipefs();
 out4:
 	unregister_pernet_subsys(&sunrpc_net_ops);
 out3:
@@ -120,6 +127,7 @@ cleanup_sunrpc(void)
 	rpcauth_remove_module();
 	cleanup_socket_xprt();
 	svc_cleanup_xprt_sock();
+	sunrpc_debugfs_exit();
 	unregister_rpc_pipefs();
 	rpc_destroy_mempool();
 	unregister_pernet_subsys(&sunrpc_net_ops);
-- 
cgit v1.2.3


From 388f0c776781fe64ce951701bfe712b2182a31f2 Mon Sep 17 00:00:00 2001
From: Jeff Layton <jlayton@primarydata.com>
Date: Wed, 26 Nov 2014 14:44:44 -0500
Subject: sunrpc: add a debugfs rpc_xprt directory with an info file in it

Add a new directory heirarchy under the debugfs sunrpc/ directory:

    sunrpc/
        rpc_xprt/
            <xprt id>/

Within that directory, we can put files that give info about the
xprts. We do have the (minor) problem that there is no succinct,
unique identifier for rpc_xprts. So we generate them synthetically
with a static atomic_t counter.

For now, this directory just holds an "info" file, but we may add
other files to it in the future.

Signed-off-by: Jeff Layton <jlayton@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
---
 net/sunrpc/debugfs.c | 115 +++++++++++++++++++++++++++++++++++++++++++++++----
 net/sunrpc/xprt.c    |   8 ++++
 2 files changed, 116 insertions(+), 7 deletions(-)

(limited to 'net/sunrpc')

diff --git a/net/sunrpc/debugfs.c b/net/sunrpc/debugfs.c
index 3d7745683ca3..e811f390f9f6 100644
--- a/net/sunrpc/debugfs.c
+++ b/net/sunrpc/debugfs.c
@@ -11,6 +11,7 @@
 
 static struct dentry *topdir;
 static struct dentry *rpc_clnt_dir;
+static struct dentry *rpc_xprt_dir;
 
 struct rpc_clnt_iter {
 	struct rpc_clnt	*clnt;
@@ -131,8 +132,8 @@ static const struct file_operations tasks_fops = {
 int
 rpc_clnt_debugfs_register(struct rpc_clnt *clnt)
 {
-	int len;
-	char name[9]; /* 8 for hex digits + NULL terminator */
+	int len, err;
+	char name[24]; /* enough for "../../rpc_xprt/ + 8 hex digits + NULL */
 
 	/* Already registered? */
 	if (clnt->cl_debugfs)
@@ -148,14 +149,28 @@ rpc_clnt_debugfs_register(struct rpc_clnt *clnt)
 		return -ENOMEM;
 
 	/* make tasks file */
+	err = -ENOMEM;
 	if (!debugfs_create_file("tasks", S_IFREG | S_IRUSR, clnt->cl_debugfs,
-				 clnt, &tasks_fops)) {
-		debugfs_remove_recursive(clnt->cl_debugfs);
-		clnt->cl_debugfs = NULL;
-		return -ENOMEM;
-	}
+				 clnt, &tasks_fops))
+		goto out_err;
+
+	err = -EINVAL;
+	rcu_read_lock();
+	len = snprintf(name, sizeof(name), "../../rpc_xprt/%s",
+			rcu_dereference(clnt->cl_xprt)->debugfs->d_name.name);
+	rcu_read_unlock();
+	if (len >= sizeof(name))
+		goto out_err;
+
+	err = -ENOMEM;
+	if (!debugfs_create_symlink("xprt", clnt->cl_debugfs, name))
+		goto out_err;
 
 	return 0;
+out_err:
+	debugfs_remove_recursive(clnt->cl_debugfs);
+	clnt->cl_debugfs = NULL;
+	return err;
 }
 
 void
@@ -165,6 +180,88 @@ rpc_clnt_debugfs_unregister(struct rpc_clnt *clnt)
 	clnt->cl_debugfs = NULL;
 }
 
+static int
+xprt_info_show(struct seq_file *f, void *v)
+{
+	struct rpc_xprt *xprt = f->private;
+
+	seq_printf(f, "netid: %s\n", xprt->address_strings[RPC_DISPLAY_NETID]);
+	seq_printf(f, "addr:  %s\n", xprt->address_strings[RPC_DISPLAY_ADDR]);
+	seq_printf(f, "port:  %s\n", xprt->address_strings[RPC_DISPLAY_PORT]);
+	seq_printf(f, "state: 0x%lx\n", xprt->state);
+	return 0;
+}
+
+static int
+xprt_info_open(struct inode *inode, struct file *filp)
+{
+	int ret;
+	struct rpc_xprt *xprt = inode->i_private;
+
+	ret = single_open(filp, xprt_info_show, xprt);
+
+	if (!ret) {
+		if (!xprt_get(xprt)) {
+			single_release(inode, filp);
+			ret = -EINVAL;
+		}
+	}
+	return ret;
+}
+
+static int
+xprt_info_release(struct inode *inode, struct file *filp)
+{
+	struct rpc_xprt *xprt = inode->i_private;
+
+	xprt_put(xprt);
+	return single_release(inode, filp);
+}
+
+static const struct file_operations xprt_info_fops = {
+	.owner		= THIS_MODULE,
+	.open		= xprt_info_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= xprt_info_release,
+};
+
+int
+rpc_xprt_debugfs_register(struct rpc_xprt *xprt)
+{
+	int len, id;
+	static atomic_t	cur_id;
+	char		name[9]; /* 8 hex digits + NULL term */
+
+	id = (unsigned int)atomic_inc_return(&cur_id);
+
+	len = snprintf(name, sizeof(name), "%x", id);
+	if (len >= sizeof(name))
+		return -EINVAL;
+
+	/* make the per-client dir */
+	xprt->debugfs = debugfs_create_dir(name, rpc_xprt_dir);
+	if (!xprt->debugfs)
+		return -ENOMEM;
+
+	/* make tasks file */
+	if (!debugfs_create_file("info", S_IFREG | S_IRUSR, xprt->debugfs,
+				 xprt, &xprt_info_fops)) {
+		debugfs_remove_recursive(xprt->debugfs);
+		xprt->debugfs = NULL;
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+void
+rpc_xprt_debugfs_unregister(struct rpc_xprt *xprt)
+{
+	debugfs_remove_recursive(xprt->debugfs);
+	xprt->debugfs = NULL;
+}
+
 void __exit
 sunrpc_debugfs_exit(void)
 {
@@ -182,6 +279,10 @@ sunrpc_debugfs_init(void)
 	if (!rpc_clnt_dir)
 		goto out_remove;
 
+	rpc_xprt_dir = debugfs_create_dir("rpc_xprt", topdir);
+	if (!rpc_xprt_dir)
+		goto out_remove;
+
 	return 0;
 out_remove:
 	debugfs_remove_recursive(topdir);
diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c
index 894d071426b2..ebbefad21a37 100644
--- a/net/sunrpc/xprt.c
+++ b/net/sunrpc/xprt.c
@@ -1303,6 +1303,7 @@ static void xprt_init(struct rpc_xprt *xprt, struct net *net)
  */
 struct rpc_xprt *xprt_create_transport(struct xprt_create *args)
 {
+	int err;
 	struct rpc_xprt	*xprt;
 	struct xprt_class *t;
 
@@ -1343,6 +1344,12 @@ found:
 		return ERR_PTR(-ENOMEM);
 	}
 
+	err = rpc_xprt_debugfs_register(xprt);
+	if (err) {
+		xprt_destroy(xprt);
+		return ERR_PTR(err);
+	}
+
 	dprintk("RPC:       created transport %p with %u slots\n", xprt,
 			xprt->max_reqs);
 out:
@@ -1359,6 +1366,7 @@ static void xprt_destroy(struct rpc_xprt *xprt)
 	dprintk("RPC:       destroying transport %p\n", xprt);
 	del_timer_sync(&xprt->timer);
 
+	rpc_xprt_debugfs_unregister(xprt);
 	rpc_destroy_wait_queue(&xprt->binding);
 	rpc_destroy_wait_queue(&xprt->pending);
 	rpc_destroy_wait_queue(&xprt->sending);
-- 
cgit v1.2.3