rxrpc: Add tracepoint for working out where aborts happen

Add a tracepoint for working out where local aborts happen. Each tracepoint call is labelled with a 3-letter code so that they can be distinguished - and the DATA sequence number is added too where available. rxrpc_kernel_abort_call() also takes a 3-letter code so that AFS can indicate the circumstances when it aborts a call. Signed-off-by: David Howells <[email protected]>
opendata26 · Sep 7, 2016 · 5a42976 · 5a42976
1 parent e8d6bbb
commit 5a42976
Show file tree

Hide file tree

Showing 11 changed files with 132 additions and 98 deletions.
diff --git a/fs/afs/rxrpc.c b/fs/afs/rxrpc.c
@@ -377,7 +377,7 @@ int afs_make_call(struct in_addr *addr, struct afs_call *call, gfp_t gfp,
 	return wait_mode->wait(call);
 
 error_do_abort:
-	rxrpc_kernel_abort_call(afs_socket, rxcall, RX_USER_ABORT);
+	rxrpc_kernel_abort_call(afs_socket, rxcall, RX_USER_ABORT, -ret, "KSD");
 error_kill_call:
 	afs_end_call(call);
 	_leave(" = %d", ret);
@@ -425,12 +425,12 @@ static void afs_deliver_to_call(struct afs_call *call)
 		case -ENOTCONN:
 			abort_code = RX_CALL_DEAD;
 			rxrpc_kernel_abort_call(afs_socket, call->rxcall,
-						abort_code);
+						abort_code, -ret, "KNC");
 			goto do_abort;
 		case -ENOTSUPP:
 			abort_code = RX_INVALID_OPERATION;
 			rxrpc_kernel_abort_call(afs_socket, call->rxcall,
-						abort_code);
+						abort_code, -ret, "KIV");
 			goto do_abort;
 		case -ENODATA:
 		case -EBADMSG:
@@ -440,7 +440,7 @@ static void afs_deliver_to_call(struct afs_call *call)
 			if (call->state != AFS_CALL_AWAIT_REPLY)
 				abort_code = RXGEN_SS_UNMARSHAL;
 			rxrpc_kernel_abort_call(afs_socket, call->rxcall,
-						abort_code);
+						abort_code, EBADMSG, "KUM");
 			goto do_abort;
 		}
 	}
@@ -463,6 +463,7 @@ static void afs_deliver_to_call(struct afs_call *call)
  */
 static int afs_wait_for_call_to_complete(struct afs_call *call)
 {
+	const char *abort_why;
 	int ret;
 
 	DECLARE_WAITQUEUE(myself, current);
@@ -481,9 +482,11 @@ static int afs_wait_for_call_to_complete(struct afs_call *call)
 			continue;
 		}
 
+		abort_why = "KWC";
 		ret = call->error;
 		if (call->state == AFS_CALL_COMPLETE)
 			break;
+		abort_why = "KWI";
 		ret = -EINTR;
 		if (signal_pending(current))
 			break;
@@ -497,7 +500,7 @@ static int afs_wait_for_call_to_complete(struct afs_call *call)
 	if (call->state < AFS_CALL_COMPLETE) {
 		_debug("call incomplete");
 		rxrpc_kernel_abort_call(afs_socket, call->rxcall,
-					RX_CALL_DEAD);
+					RX_CALL_DEAD, -ret, abort_why);
 	}
 
 	_debug("call complete");
@@ -695,7 +698,7 @@ void afs_send_empty_reply(struct afs_call *call)
 	case -ENOMEM:
 		_debug("oom");
 		rxrpc_kernel_abort_call(afs_socket, call->rxcall,
-					RX_USER_ABORT);
+					RX_USER_ABORT, ENOMEM, "KOO");
 	default:
 		afs_end_call(call);
 		_leave(" [error]");
@@ -734,7 +737,7 @@ void afs_send_simple_reply(struct afs_call *call, const void *buf, size_t len)
 	if (n == -ENOMEM) {
 		_debug("oom");
 		rxrpc_kernel_abort_call(afs_socket, call->rxcall,
-					RX_USER_ABORT);
+					RX_USER_ABORT, ENOMEM, "KOO");
 	}
 	afs_end_call(call);
 	_leave(" [error]");

diff --git a/include/net/af_rxrpc.h b/include/net/af_rxrpc.h
@@ -35,7 +35,8 @@ int rxrpc_kernel_send_data(struct socket *, struct rxrpc_call *,
 			   struct msghdr *, size_t);
 int rxrpc_kernel_recv_data(struct socket *, struct rxrpc_call *,
 			   void *, size_t, size_t *, bool, u32 *);
-void rxrpc_kernel_abort_call(struct socket *, struct rxrpc_call *, u32);
+void rxrpc_kernel_abort_call(struct socket *, struct rxrpc_call *,
+			     u32, int, const char *);
 void rxrpc_kernel_end_call(struct socket *, struct rxrpc_call *);
 struct rxrpc_call *rxrpc_kernel_accept_call(struct socket *, unsigned long,
 					    rxrpc_notify_rx_t);

diff --git a/include/trace/events/rxrpc.h b/include/trace/events/rxrpc.h
@@ -84,6 +84,35 @@ TRACE_EVENT(rxrpc_skb,
 		      __entry->where)
 	    );
 
+TRACE_EVENT(rxrpc_abort,
+	    TP_PROTO(const char *why, u32 cid, u32 call_id, rxrpc_seq_t seq,
+		     int abort_code, int error),
+
+	    TP_ARGS(why, cid, call_id, seq, abort_code, error),
+
+	    TP_STRUCT__entry(
+		    __array(char,			why, 4		)
+		    __field(u32,			cid		)
+		    __field(u32,			call_id		)
+		    __field(rxrpc_seq_t,		seq		)
+		    __field(int,			abort_code	)
+		    __field(int,			error		)
+			     ),
+
+	    TP_fast_assign(
+		    memcpy(__entry->why, why, 4);
+		    __entry->cid = cid;
+		    __entry->call_id = call_id;
+		    __entry->abort_code = abort_code;
+		    __entry->error = error;
+		    __entry->seq = seq;
+			   ),
+
+	    TP_printk("%08x:%08x s=%u a=%d e=%d %s",
+		      __entry->cid, __entry->call_id, __entry->seq,
+		      __entry->abort_code, __entry->error, __entry->why)
+	    );
+
 #endif /* _TRACE_RXRPC_H */
 
 /* This part must be outside protection */

diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h
@@ -155,7 +155,8 @@ struct rxrpc_security {
 			     void *);
 
 	/* verify the security on a received packet */
-	int (*verify_packet)(struct rxrpc_call *, struct sk_buff *, u32 *);
+	int (*verify_packet)(struct rxrpc_call *, struct sk_buff *,
+			     rxrpc_seq_t, u16);
 
 	/* issue a challenge */
 	int (*issue_challenge)(struct rxrpc_connection *);
@@ -637,9 +638,12 @@ static inline bool rxrpc_call_completed(struct rxrpc_call *call)
 /*
  * Record that a call is locally aborted.
  */
-static inline bool __rxrpc_abort_call(struct rxrpc_call *call,
+static inline bool __rxrpc_abort_call(const char *why, struct rxrpc_call *call,
+				      rxrpc_seq_t seq,
 				      u32 abort_code, int error)
 {
+	trace_rxrpc_abort(why, call->cid, call->call_id, seq,
+			  abort_code, error);
 	if (__rxrpc_set_call_completion(call,
 					RXRPC_CALL_LOCALLY_ABORTED,
 					abort_code, error)) {
@@ -649,13 +653,13 @@ static inline bool __rxrpc_abort_call(struct rxrpc_call *call,
 	return false;
 }
 
-static inline bool rxrpc_abort_call(struct rxrpc_call *call,
-				    u32 abort_code, int error)
+static inline bool rxrpc_abort_call(const char *why, struct rxrpc_call *call,
+				    rxrpc_seq_t seq, u32 abort_code, int error)
 {
 	bool ret;
 
 	write_lock_bh(&call->state_lock);
-	ret = __rxrpc_abort_call(call, abort_code, error);
+	ret = __rxrpc_abort_call(why, call, seq, abort_code, error);
 	write_unlock_bh(&call->state_lock);
 	return ret;
 }

diff --git a/net/rxrpc/call_event.c b/net/rxrpc/call_event.c
@@ -598,7 +598,8 @@ static int rxrpc_process_rx_queue(struct rxrpc_call *call,
 
 		/* secured packets must be verified and possibly decrypted */
 		if (call->conn->security->verify_packet(call, skb,
-							_abort_code) < 0)
+							sp->hdr.seq,
+							sp->hdr.cksum) < 0)
 			goto protocol_error;
 
 		rxrpc_insert_oos_packet(call, skb);
@@ -982,7 +983,7 @@ void rxrpc_process_call(struct work_struct *work)
 	}
 
 	if (test_bit(RXRPC_CALL_EV_LIFE_TIMER, &call->events)) {
-		rxrpc_abort_call(call, RX_CALL_TIMEOUT, ETIME);
+		rxrpc_abort_call("EXP", call, 0, RX_CALL_TIMEOUT, ETIME);
 
 		_debug("post timeout");
 		if (rxrpc_post_message(call, RXRPC_SKB_MARK_LOCAL_ERROR,
@@ -1005,7 +1006,7 @@ void rxrpc_process_call(struct work_struct *work)
 		case -EKEYEXPIRED:
 		case -EKEYREJECTED:
 		case -EPROTO:
-			rxrpc_abort_call(call, abort_code, -ret);
+			rxrpc_abort_call("PRO", call, 0, abort_code, -ret);
 			goto kill_ACKs;
 		}
 	}

diff --git a/net/rxrpc/call_object.c b/net/rxrpc/call_object.c
@@ -563,7 +563,7 @@ void rxrpc_release_call(struct rxrpc_sock *rx, struct rxrpc_call *call)
 
 		if (call->state < RXRPC_CALL_COMPLETE) {
 			_debug("+++ ABORTING STATE %d +++\n", call->state);
-			__rxrpc_abort_call(call, RX_CALL_DEAD, ECONNRESET);
+			__rxrpc_abort_call("SKT", call, 0, RX_CALL_DEAD, ECONNRESET);
 			clear_bit(RXRPC_CALL_EV_ACK_FINAL, &call->events);
 			rxrpc_send_call_packet(call, RXRPC_PACKET_TYPE_ABORT);
 		}

diff --git a/net/rxrpc/conn_event.c b/net/rxrpc/conn_event.c
@@ -158,6 +158,11 @@ static void rxrpc_abort_calls(struct rxrpc_connection *conn,
 			lockdep_is_held(&conn->channel_lock));
 		if (call) {
 			rxrpc_see_call(call);
+			if (compl == RXRPC_CALL_LOCALLY_ABORTED)
+				trace_rxrpc_abort("CON", call->cid,
+						  call->call_id, 0,
+						  abort_code, error);
+
 			write_lock_bh(&call->state_lock);
 			if (rxrpc_set_call_completion(call, compl, abort_code,
 						      error)) {
@@ -167,6 +172,7 @@ static void rxrpc_abort_calls(struct rxrpc_connection *conn,
 			write_unlock_bh(&call->state_lock);
 			if (queue)
 				rxrpc_queue_call(call);
+
 		}
 	}
 

diff --git a/net/rxrpc/input.c b/net/rxrpc/input.c
@@ -429,7 +429,7 @@ void rxrpc_fast_process_packet(struct rxrpc_call *call, struct sk_buff *skb)
 	_debug("protocol error");
 	write_lock_bh(&call->state_lock);
 protocol_error_locked:
-	if (__rxrpc_abort_call(call, RX_PROTOCOL_ERROR, EPROTO))
+	if (__rxrpc_abort_call("FPR", call, 0, RX_PROTOCOL_ERROR, EPROTO))
 		rxrpc_queue_call(call);
 free_packet_unlock:
 	write_unlock_bh(&call->state_lock);
@@ -495,9 +495,10 @@ static void rxrpc_process_jumbo_packet(struct rxrpc_call *call,
 protocol_error:
 	_debug("protocol error");
 	rxrpc_free_skb(part);
-	rxrpc_free_skb(jumbo);
-	if (rxrpc_abort_call(call, RX_PROTOCOL_ERROR, EPROTO))
+	if (rxrpc_abort_call("PJP", call, sp->hdr.seq,
+			     RX_PROTOCOL_ERROR, EPROTO))
 		rxrpc_queue_call(call);
+	rxrpc_free_skb(jumbo);
 	_leave("");
 }
 

diff --git a/net/rxrpc/insecure.c b/net/rxrpc/insecure.c
@@ -23,31 +23,32 @@ static int none_prime_packet_security(struct rxrpc_connection *conn)
 }
 
 static int none_secure_packet(struct rxrpc_call *call,
-			       struct sk_buff *skb,
-			       size_t data_size,
-			       void *sechdr)
+			      struct sk_buff *skb,
+			      size_t data_size,
+			      void *sechdr)
 {
 	return 0;
 }
 
 static int none_verify_packet(struct rxrpc_call *call,
-			       struct sk_buff *skb,
-			       u32 *_abort_code)
+			      struct sk_buff *skb,
+			      rxrpc_seq_t seq,
+			      u16 expected_cksum)
 {
 	return 0;
 }
 
 static int none_respond_to_challenge(struct rxrpc_connection *conn,
-				      struct sk_buff *skb,
-				      u32 *_abort_code)
+				     struct sk_buff *skb,
+				     u32 *_abort_code)
 {
 	*_abort_code = RX_PROTOCOL_ERROR;
 	return -EPROTO;
 }
 
 static int none_verify_response(struct rxrpc_connection *conn,
-				 struct sk_buff *skb,
-				 u32 *_abort_code)
+				struct sk_buff *skb,
+				u32 *_abort_code)
 {
 	*_abort_code = RX_PROTOCOL_ERROR;
 	return -EPROTO;