gnunet-service-transport.c File Reference

main for gnunet-service-transport More...

#include "platform.h"
#include "gnunet_util_lib.h"
#include "gnunet_statistics_service.h"
#include "gnunet_peerstore_service.h"
#include "gnunet_pils_service.h"
#include "gnunet_transport_communication_service.h"
#include "gnunet_nat_service.h"
#include "gnunet_hello_uri_lib.h"
#include "gnunet_signatures.h"
#include "transport.h"

Include dependency graph for gnunet-service-transport.c:

Go to the source code of this file.

Data Structures
struct	MessageUUIDP
	Unique identifier we attach to a message. More...
struct	AcknowledgementUUIDP
	Unique identifier to map an acknowledgement to a transmission. More...
struct	TransportBackchannelEncapsulationMessage
	Outer layer of an encapsulated backchannel message. More...
struct	EphemeralConfirmationPS
	Body by which a peer confirms that it is using an ephemeral key. More...
struct	TransportDVBoxPayloadP
	Plaintext of the variable-size payload that is encrypted within a struct TransportBackchannelEncapsulationMessage More...
struct	TransportReliabilityBoxMessage
	Outer layer of an encapsulated unfragmented application message sent over an unreliable channel. More...
struct	TransportCummulativeAckPayloadP
	Acknowledgement payload. More...
struct	TransportReliabilityAckMessage
	Confirmation that the receiver got a GNUNET_MESSAGE_TYPE_TRANSPORT_RELIABILITY_BOX. More...
struct	TransportFragmentBoxMessage
	Outer layer of an encapsulated fragmented application message. More...
struct	DvInitPS
	Content signed by the initiator during DV learning. More...
struct	DvHopPS
	Content signed by each peer during DV learning. More...
struct	DVPathEntryP
	An entry describing a peer on a path in a struct TransportDVLearnMessage message. More...
struct	TransportDVLearnMessage
	Internal message used by transport for distance vector learning. More...
struct	TransportDVBoxMessage
	Outer layer of an encapsulated message send over multiple hops. More...
struct	TransportValidationChallengeMessage
	Message send to another peer to validate that it can indeed receive messages at a particular address. More...
struct	TransportValidationPS
	Message signed by a peer to confirm that it can indeed receive messages at a particular address. More...
struct	TransportValidationResponseMessage
	Message send to a peer to respond to a #GNUNET_MESSAGE_TYPE_ADDRESS_VALIDATION_CHALLENGE. More...
struct	TransportGlobalNattedAddress
struct	TransportFlowControlMessage
	Message for Transport-to-Transport Flow control. More...
struct	LearnLaunchEntry
	When did we launch this DV learning activity? More...
struct	TransmissionHistoryEntry
	Information we keep per GOODPUT_AGING_SLOTS about historic (or current) transmission performance. More...
struct	PerformanceData
	Performance data for a transmission possibility. More...
struct	CommunicatorMessageContext
	Context from handle_incoming_msg(). More...
struct	RingBufferEntry
	Entry for the ring buffer caching messages send to core, when virtual link is available. More...
struct	CoreSentContext
	Closure for core_env_sent_cb. More...
struct	ReassemblyContext
	Information we keep for a message that we are reassembling. More...
struct	VirtualLink
	A virtual link is another reachable peer that is known to CORE. More...
struct	PendingAcknowledgement
	Data structure kept when we are waiting for an acknowledgement. More...
struct	DistanceVectorHop
	One possible hop towards a DV target. More...
struct	DistanceVector
	Entry in our dv_routes table, representing a (set of) distance vector routes to a particular peer. More...
struct	QueueEntry
	Entry identifying transmission in one of our struct Queue which still awaits an ACK. More...
struct	Queue
	Handle for a queue. More...
struct	Neighbour
	A neighbour that at least one communicator is connected to. More...
struct	IncomingRequest
	Another peer attempted to talk to us, we should try to establish a connection in the other direction. More...
struct	PeerRequest
	Information per peer and request. More...
struct	PendingMessage
	List containing all messages that are yet to be send. More...
struct	TransportCummulativeAckPayload
	Acknowledgement payload. More...
struct	AcknowledgementCummulator
	Data structure in which we track acknowledgements still to be sent to the. More...
struct	AddressListEntry
	One of the addresses of this peer. More...
struct	TransportClient
	Client connected to the transport service. More...
struct	ValidationState
	State we keep for validation activities. More...
struct	Backtalker
	A Backtalker is a peer sending us backchannel messages. More...
struct	PilsRequest
	DLL. More...
struct	MonitorEvent
	Details about what to notify monitors about. More...
struct	SignDvCls
struct	DVKeyState
	Structure of the key material used to encrypt backchannel messages. More...
struct	AddGlobalAddressesContext
struct	PilsAddressSignContext
	Helper context struct for HELLO update. More...
struct	SignedAddress
	Binary block we sign when we sign an address. More...
struct	FindByMessageUuidContext
	Closure for find_by_message_uuid. More...
struct	SignDhpCls
struct	NeighbourSelectionContext
	Closure for dv_neighbour_selection and dv_neighbour_transmission. More...
struct	DecapsDvBoxCls
struct	CheckKnownAddressContext
	Closure for check_known_address. More...
struct	SignTValidationCls
struct	CheckKnownChallengeContext
	Closure for check_known_challenge. More...
struct	PendingMessageScoreContext
	Context for select_best_pending_from_link(). More...
struct	QueueQualityContext
	Closure for check_connection_quality. More...
struct	SignDvInitCls
struct	TransportGlobalNattedAddressClosure
struct	UpdateHelloFromPidCtx

Macros
#define	RING_BUFFER_SIZE   16
	Size of ring buffer to cache CORE and forwarded DVBox messages.
#define	MAX_FC_RETRANSMIT_COUNT   1000
	Maximum number of FC retransmissions for a running retransmission task.
#define	MAX_CUMMULATIVE_ACKS   64
	Maximum number of messages we acknowledge together in one cumulative ACK.
#define	FC_NO_CHANGE_REPLY_PROBABILITY   8
	What is the 1:n chance that we send a Flow control response when receiving a flow control message that did not change anything for us? Basically, this is used in the case where both peers are stuck on flow control (no window changes), but one might continue sending flow control messages to the other peer as the first FC message when things stalled got lost, and then subsequently the other peer does usually not respond as nothing changed.
#define	IN_PACKET_SIZE_WITHOUT_MTU   128
	What is the size we assume for a read operation in the absence of an MTU for the purpose of flow control?
#define	GOODPUT_AGING_SLOTS   4
	Number of slots we keep of historic data for computation of goodput / message loss ratio.
#define	DEFAULT_WINDOW_SIZE   (128 * 1024)
	How big is the flow control window size by default; limits per-neighbour RAM utilization.
#define	MAX_INCOMING_REQUEST   16
	For how many incoming connections do we try to create a virtual link for (at the same time!).
#define	MAX_DV_DISCOVERY_SELECTION   16
	Maximum number of peers we select for forwarding DVInit messages at the same time (excluding initiator).
#define	RECV_WINDOW_SIZE   4
	Window size.
#define	MIN_DV_PATH_LENGTH_FOR_INITIATOR   3
	Minimum number of hops we should forward DV learn messages even if they are NOT useful for us in hope of looping back to the initiator?
#define	MAX_DV_HOPS_ALLOWED   16
	Maximum DV distance allowed ever.
#define	MAX_DV_LEARN_PENDING   64
	Maximum number of DV learning activities we may have pending at the same time.
#define	MAX_DV_PATHS_TO_TARGET   3
	Maximum number of DV paths we keep simultaneously to the same target.
#define	PILS_FEED_ADDRESSES_DELAY    GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_SECONDS, 3)
	Delay between added/removed addresses and PILS feed call.
#define	DELAY_WARN_THRESHOLD    GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_SECONDS, 5)
	If a queue delays the next message by more than this number of seconds we log a warning.
#define	DV_FORWARD_TIMEOUT    GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_SECONDS, 60)
	If a DVBox could not be forwarded after this number of seconds we drop it.
#define	DEFAULT_ACK_WAIT_DURATION    GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_SECONDS, 1)
	Default value for how long we wait for reliability ack.
#define	DV_QUALITY_RTT_THRESHOLD    GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_SECONDS, 1)
	We only consider queues as "quality" connections when suppressing the generation of DV initiation messages if the latency of the queue is below this threshold.
#define	DV_PATH_VALIDITY_TIMEOUT    GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_MINUTES, 5)
	How long do we consider a DV path valid if we see no further updates on it? Note: the value chosen here might be too low!
#define	BACKCHANNEL_INACTIVITY_TIMEOUT    GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_MINUTES, 5)
	How long do we cache backchannel (struct Backtalker) information after a backchannel goes inactive?
#define	DV_PATH_DISCOVERY_FREQUENCY    GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_MINUTES, 4)
	How long before paths expire would we like to (re)discover DV paths? Should be below DV_PATH_VALIDITY_TIMEOUT.
#define	EPHEMERAL_VALIDITY    GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_HOURS, 4)
	How long are ephemeral keys valid?
#define	REASSEMBLY_EXPIRATION    GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_MINUTES, 4)
	How long do we keep partially reassembled messages around before giving up?
#define	FAST_VALIDATION_CHALLENGE_FREQ    GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_MINUTES, 1)
	What is the fastest rate at which we send challenges if we keep learning an address (gossip, DHT, etc.)?
#define	MAX_VALIDATION_CHALLENGE_FREQ    GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_DAYS, 1)
	What is the slowest rate at which we send challenges?
#define	ACK_CUMMULATOR_TIMEOUT    GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_HOURS, 4)
	How long until we forget about historic accumulators and thus reset the ACK counter? Should exceed the maximum time an active connection experiences without an ACK.
#define	DV_LEARN_BASE_FREQUENCY   GNUNET_TIME_UNIT_MINUTES
	What is the non-randomized base frequency at which we would initiate DV learn messages?
#define	DV_LEARN_QUALITY_THRESHOLD   100
	How many good connections (confirmed, bi-directional, not DV) do we need to have to suppress initiating DV learn messages?
#define	MAX_ADDRESS_VALID_UNTIL    GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_MONTHS, 1)
	When do we forget an invalid address for sure?
#define	ADDRESS_VALIDATION_LIFETIME    GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_HOURS, 4)
	How long do we consider an address valid if we just checked?
#define	MIN_DELAY_ADDRESS_VALIDATION   GNUNET_TIME_UNIT_MILLISECONDS
	What is the maximum frequency at which we do address validation? A random value between 0 and this value is added when scheduling the validation_task (both to ensure we do not validate too often, and to randomize a bit).
#define	VALIDATION_RTT_BUFFER_FACTOR   3
	How many network RTTs before an address validation expires should we begin trying to revalidate? (Note that the RTT used here is the one that we experienced during the last validation, not necessarily the latest RTT observed).
#define	COMMUNICATOR_TOTAL_QUEUE_LIMIT   512
	How many messages can we have pending for a given communicator process before we start to throttle that communicator?
#define	QUEUE_LENGTH_LIMIT   32
	How many messages can we have pending for a given queue (queue to a particular peer via a communicator) process before we start to throttle that queue?
#define	QUEUE_ENTRY_TIMEOUT    GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_SECONDS, 5)
#define	RTT_DIFF    GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_SECONDS, 1)
	Difference of the average RTT for the DistanceVector calculate by us and the target we are willing to accept for starting the burst.

Typedefs
typedef void(*	DVMessageHandler) (void *cls, struct Neighbour *next_hop, const struct GNUNET_MessageHeader *hdr, enum RouteMessageOptions options)
	Function to call to further operate on the now DV encapsulated message hdr, forwarding it via next_hop under respect of options.

Enumerations
enum	ClientType {   CT_NONE = 0 , CT_CORE = 1 , CT_MONITOR = 2 , CT_COMMUNICATOR = 3 ,   CT_APPLICATION = 4 }
	What type of client is the struct TransportClient about? More...
enum	RouteMessageOptions {   RMO_NONE = 0 , RMO_DV_ALLOWED = 1 , RMO_UNCONFIRMED_ALLOWED = 2 , RMO_ANYTHING_GOES = (RMO_DV_ALLOWED | RMO_UNCONFIRMED_ALLOWED) ,   RMO_REDUNDANT = 4 }
	Which transmission options are allowable for transmission? Interpreted bit-wise! More...
enum	PendingMessageType { PMT_CORE = 0 , PMT_FRAGMENT_BOX = 1 , PMT_RELIABILITY_BOX = 2 , PMT_DV_BOX = 3 }
	Types of different pending messages. More...

Functions
static unsigned int	get_age ()
	Get an offset into the transmission history buffer for struct PerformanceData.
static void	free_incoming_request (struct IncomingRequest *ir)
	Release ir data structure.
static void	free_pending_acknowledgement (struct PendingAcknowledgement *pa)
	Release pa data structure.
static void	free_fragment_tree (struct PendingMessage *root)
	Free fragment tree below root, excluding root itself.
static void	free_pending_message (struct PendingMessage *pm)
	Release memory associated with pm and remove pm from associated data structures.
static void	free_reassembly_context (struct ReassemblyContext *rc)
	Free rc.
static void	reassembly_cleanup_task (void *cls)
	Task run to clean up reassembly context of a neighbour that have expired.
static int	free_reassembly_cb (void *cls, uint32_t key, void *value)
	function called to free_reassembly_context().
static void	free_virtual_link (struct VirtualLink *vl)
	Free virtual link.
static void	free_validation_state (struct ValidationState *vs)
	Free validation state.
static struct Neighbour *	lookup_neighbour (const struct GNUNET_PeerIdentity *pid)
	Lookup neighbour for peer pid.
static struct VirtualLink *	lookup_virtual_link (const struct GNUNET_PeerIdentity *pid)
	Lookup virtual link for peer pid.
static void	free_distance_vector_hop (struct DistanceVectorHop *dvh)
	Free a dvh.
static void	check_link_down (void *cls)
	Task run to check whether the hops of the cls still are validated, or if we need to core about disconnection.
static void	cores_send_disconnect_info (const struct GNUNET_PeerIdentity *pid)
	Send message to CORE clients that we lost a connection.
static void	free_dv_route (struct DistanceVector *dv)
	Free entry in dv_routes.
static void	notify_monitor (struct TransportClient *tc, const struct GNUNET_PeerIdentity *peer, const char *address, enum GNUNET_NetworkType nt, const struct MonitorEvent *me)
	Notify monitor tc about an event.
static void	notify_monitors (const struct GNUNET_PeerIdentity *peer, const char *address, enum GNUNET_NetworkType nt, const struct MonitorEvent *me)
	Send information in me about a peer's status with respect to some address to all monitors that care.
static void *	client_connect_cb (void *cls, struct GNUNET_SERVICE_Client *client, struct GNUNET_MQ_Handle *mq)
	Called whenever a client connects.
static enum GNUNET_GenericReturnValue	remove_global_addresses (void *cls, const struct GNUNET_PeerIdentity *pid, void *value)
static void	free_neighbour (struct Neighbour *neighbour, enum GNUNET_GenericReturnValue drop_link)
	Release memory used by neighbour.
static void	core_send_connect_info (struct TransportClient *tc, const struct GNUNET_PeerIdentity *pid)
	Send message to CORE clients that we lost a connection.
static void	cores_send_connect_info (const struct GNUNET_PeerIdentity *pid)
	Send message to CORE clients that we gained a connection.
static void	transmit_on_queue (void *cls)
	We believe we are ready to transmit a message on a queue.
static unsigned int	check_for_queue_with_higher_prio (struct Queue *queue, struct Queue *queue_head)
	Check if the communicator has another queue with higher prio ready for sending.
static void	schedule_transmit_on_queue (struct GNUNET_TIME_Relative delay, struct Queue *queue, enum GNUNET_SCHEDULER_Priority p)
	Called whenever something changed that might effect when we try to do the next transmission on queue using transmit_on_queue().
static void	free_queue (struct Queue *queue)
	Free queue.
static void	free_address_list_entry (struct AddressListEntry *ale)
	Free ale.
static int	stop_peer_request (void *cls, const struct GNUNET_PeerIdentity *pid, void *value)
	Stop the peer request in value.
static void	do_shutdown (void *cls)
	Function called when the service shuts down.
static void	client_disconnect_cb (void *cls, struct GNUNET_SERVICE_Client *client, void *app_ctx)
	Called whenever a client is disconnected.
static int	notify_client_connect_info (void *cls, const struct GNUNET_PeerIdentity *pid, void *value)
	Iterator telling new CORE client about all existing connections to peers.
static void	finish_cmc_handling_with_continue (struct CommunicatorMessageContext *cmc, unsigned int free_cmc)
	Send ACK to communicator (if requested) and free cmc.
static enum GNUNET_GenericReturnValue	resume_communicators (void *cls, const struct GNUNET_PeerIdentity *pid, void *value)
static void	handle_client_start (void *cls, const struct StartMessage *start)
	Initialize a "CORE" client.
static int	check_client_send (void *cls, const struct OutboundMessage *obm)
	Client asked for transmission to a peer.
static void	client_send_response (struct PendingMessage *pm)
	Send a response to the pm that we have processed a "send" request.
static unsigned int	pick_random_dv_hops (const struct DistanceVector *dv, enum RouteMessageOptions options, struct DistanceVectorHop **hops_array, unsigned int hops_array_length)
	Pick hops_array_length random DV paths satisfying options.
static int	check_communicator_available (void *cls, const struct GNUNET_TRANSPORT_CommunicatorAvailableMessage *cam)
	Communicator started.
static void	finish_cmc_handling (struct CommunicatorMessageContext *cmc)
static void	handle_client_recv_ok (void *cls, const struct RecvOkMessage *rom)
	Client confirms that it is done handling message(s) to a particular peer.
static void	handle_communicator_available (void *cls, const struct GNUNET_TRANSPORT_CommunicatorAvailableMessage *cam)
	Communicator started.
static int	check_communicator_backchannel (void *cls, const struct GNUNET_TRANSPORT_CommunicatorBackchannel *cb)
	Communicator requests backchannel transmission.
static void	sign_dv_cb (void *cls, const struct GNUNET_PeerIdentity *pid, const struct GNUNET_CRYPTO_EddsaSignature *sig)
static void	sign_ephemeral (struct DistanceVector *dv)
	Sign ephemeral keys in our dv are current.
static void	free_queue_entry (struct QueueEntry *qe, struct TransportClient *tc)
static void	free_timedout_queue_entry (void *cls)
static void	queue_send_msg (struct Queue *queue, struct PendingMessage *pm, const void *payload, size_t payload_size)
	Send the message payload on queue.
static struct GNUNET_TIME_Relative	route_via_neighbour (const struct Neighbour *n, const struct GNUNET_MessageHeader *hdr, enum RouteMessageOptions options)
	Pick a queue of n under constraints options and schedule transmission of hdr.
static void	dv_setup_key_state_from_km (const struct GNUNET_ShortHashCode *km, const struct GNUNET_ShortHashCode *iv, struct DVKeyState *key)
	Given the key material in km and the initialization vector iv, setup the key material for the backchannel in key.
static void	dv_hmac (const struct DVKeyState *key, struct GNUNET_HashCode *hmac, const void *data, size_t data_size)
	Do HMAC calculation for backchannel messages over data using key material from key.
static void	dv_encrypt (struct DVKeyState *key, const void *in, void *dst, size_t in_size)
	Perform backchannel encryption using symmetric secret in key to encrypt data from in to dst.
static enum GNUNET_GenericReturnValue	dv_decrypt (struct DVKeyState *key, void *out, const void *ciph, size_t out_size)
	Perform backchannel encryption using symmetric secret in key to encrypt data from in to dst.
static void	dv_key_clean (struct DVKeyState *key)
	Clean up key material in key.
static struct GNUNET_TIME_Relative	encapsulate_for_dv (struct DistanceVector *dv, unsigned int num_dvhs, struct DistanceVectorHop **dvhs, const struct GNUNET_MessageHeader *hdr, DVMessageHandler use, void *use_cls, enum RouteMessageOptions options, enum GNUNET_GenericReturnValue without_fc)
	Pick a path of dv under constraints options and schedule transmission of hdr.
static void	send_dv_to_neighbour (void *cls, struct Neighbour *next_hop, const struct GNUNET_MessageHeader *hdr, enum RouteMessageOptions options)
	Wrapper around route_via_neighbour() that matches the DVMessageHandler structure.
static struct GNUNET_TIME_Relative	route_control_message_without_fc (struct VirtualLink *vl, const struct GNUNET_MessageHeader *hdr, enum RouteMessageOptions options)
	We need to transmit hdr to target.
static void	consider_sending_fc (void *cls)
	Something changed on the virtual link with respect to flow control.
static void	task_consider_sending_fc (void *cls)
	Something changed on the virtual link with respect to flow control.
static char *	get_address_without_port (const char *address)
	Get the IP address without the port number.
static enum GNUNET_GenericReturnValue	add_global_addresses (void *cls, const struct GNUNET_PeerIdentity *pid, void *value)
static struct GNUNET_TIME_Relative	calculate_rtt (struct DistanceVector *dv)
static void	check_vl_transmission (struct VirtualLink *vl)
	There is a message at the head of the pending messages for vl which may be ready for transmission.
static void	handle_client_send (void *cls, const struct OutboundMessage *obm)
	Client asked for transmission to a peer.
static void	handle_communicator_backchannel (void *cls, const struct GNUNET_TRANSPORT_CommunicatorBackchannel *cb)
	Communicator requests backchannel transmission.
static int	check_add_address (void *cls, const struct GNUNET_TRANSPORT_AddAddressMessage *aam)
	Address of our peer added.
static void	store_pi (void *cls)
	Ask peerstore to store our address.
static void	shc_cont (void *cls, int success)
static void	pils_sign_hello_cb (void *cls, const struct GNUNET_PeerIdentity *pid, const struct GNUNET_CRYPTO_EddsaSignature *sig)
	Get HELLO signature and create message to store in PEERSTORE.
static void	peerstore_store_own_cb (void *cls, int success)
	Function called when peerstore is done storing our address.
static void	pils_sign_addr_cb (void *cls, const struct GNUNET_PeerIdentity *pid, const struct GNUNET_CRYPTO_EddsaSignature *sig)
void	pils_sign_address (struct AddressListEntry *ale, struct GNUNET_TIME_Absolute mono_time)
	Build address record by signing raw information with private key of the peer identity.
static struct AddressListEntry *	create_address_entry (struct TransportClient *tc, struct GNUNET_TIME_Relative expiration, enum GNUNET_NetworkType nt, const char *address, uint32_t aid, size_t slen)
static void	feed_addresses_to_pils (void *cls)
static void	handle_add_address (void *cls, const struct GNUNET_TRANSPORT_AddAddressMessage *aam)
	Address of our peer added.
static void	handle_del_address (void *cls, const struct GNUNET_TRANSPORT_DelAddressMessage *dam)
	Address of our peer deleted.
static void	demultiplex_with_cmc (struct CommunicatorMessageContext *cmc)
	Given an inbound message msg from a communicator cmc, demultiplex it based on the type calling the right handler.
static void	core_env_sent_cb (void *cls)
	Function called when we are done giving a message of a certain size to CORE and should thus decrement the number of bytes of RAM reserved for that peer's MQ.
static void	finish_handling_raw_message (struct VirtualLink *vl, const struct GNUNET_MessageHeader *mh, struct CommunicatorMessageContext *cmc, unsigned int free_cmc)
static void	handle_raw_message (void *cls, const struct GNUNET_MessageHeader *mh)
	Communicator gave us an unencapsulated message to pass as-is to CORE.
static int	check_fragment_box (void *cls, const struct TransportFragmentBoxMessage *fb)
	Communicator gave us a fragment box.
static void	destroy_ack_cummulator (void *cls)
	Clean up an idle cumulative acknowledgement data structure.
static void	transmit_cummulative_ack_cb (void *cls)
	Do the transmission of a cumulative acknowledgement now.
static void	cummulative_ack (const struct GNUNET_PeerIdentity *pid, const struct AcknowledgementUUIDP *ack_uuid, struct GNUNET_TIME_Absolute max_delay)
	Transmit an acknowledgement for ack_uuid to pid delaying transmission by at most ack_delay.
static int	find_by_message_uuid (void *cls, uint32_t key, void *value)
	Iterator called to find a reassembly context by the message UUID in the multihashmap32.
static void	handle_fragment_box (void *cls, const struct TransportFragmentBoxMessage *fb)
	Communicator gave us a fragment.
static int	check_reliability_box (void *cls, const struct TransportReliabilityBoxMessage *rb)
	Communicator gave us a reliability box.
static void	handle_reliability_box (void *cls, const struct TransportReliabilityBoxMessage *rb)
	Communicator gave us a reliability box.
static void	update_pd_age (struct PerformanceData *pd, unsigned int age)
	Check if we have advanced to another age since the last time.
static void	update_performance_data (struct PerformanceData *pd, struct GNUNET_TIME_Relative rtt, uint16_t bytes_transmitted_ok)
	Update pd based on the latest rtt and the number of bytes that were confirmed to be successfully transmitted.
static void	update_queue_performance (struct Queue *q, struct GNUNET_TIME_Relative rtt, uint16_t bytes_transmitted_ok)
	We have successfully transmitted data via q, update metrics.
static void	update_dvh_performance (struct DistanceVectorHop *dvh, struct GNUNET_TIME_Relative rtt, uint16_t bytes_transmitted_ok)
	We have successfully transmitted data via dvh, update metrics.
static void	completed_pending_message (struct PendingMessage *pm)
	We have completed transmission of pm, remove it from the transmission queues (and if it is a fragment, continue up the tree as necessary).
static void	handle_acknowledged (struct PendingAcknowledgement *pa, struct GNUNET_TIME_Relative ack_delay)
	The pa was acknowledged, process the acknowledgement.
static int	check_reliability_ack (void *cls, const struct TransportReliabilityAckMessage *ra)
	Communicator gave us a reliability ack.
static void	handle_reliability_ack (void *cls, const struct TransportReliabilityAckMessage *ra)
	Communicator gave us a reliability ack.
static int	check_backchannel_encapsulation (void *cls, const struct TransportBackchannelEncapsulationMessage *be)
	Communicator gave us a backchannel encapsulation.
static void	handle_backchannel_encapsulation (void *cls, const struct TransportBackchannelEncapsulationMessage *be)
	Communicator gave us a backchannel encapsulation.
static void	path_cleanup_cb (void *cls)
	Task called when we should check if any of the DV paths we have learned to a target are due for garbage collection.
static void	send_msg_from_cache (struct VirtualLink *vl)
static void	activate_core_visible_dv_path (struct DistanceVectorHop *hop)
	The hop is a validated path to the respective target peer and we should tell core about it – and schedule a job to revoke the state.
static int	learn_dv_path (const struct GNUNET_PeerIdentity *path, unsigned int path_len, struct GNUNET_TIME_Relative network_latency, struct GNUNET_TIME_Absolute path_valid_until)
	We have learned a path through the network to some other peer, add it to our DV data structure (returning GNUNET_YES on success).
static int	check_dv_learn (void *cls, const struct TransportDVLearnMessage *dvl)
	Communicator gave us a DV learn message.
static void	sign_dhp_cp (void *cls, const struct GNUNET_PeerIdentity *pid, const struct GNUNET_CRYPTO_EddsaSignature *sig)
static void	forward_dv_learn (const struct GNUNET_PeerIdentity *next_hop, const struct TransportDVLearnMessage *msg, uint16_t bi_history, uint16_t nhops, const struct DVPathEntryP *hops, struct GNUNET_TIME_Absolute in_time)
	Build and forward a DV learn message to next_hop.
static int	validate_dv_initiator_signature (struct GNUNET_TIME_AbsoluteNBO sender_monotonic_time, const struct GNUNET_PeerIdentity *init, const struct GNUNET_CRYPTO_ChallengeNonceP *challenge, const struct GNUNET_CRYPTO_EddsaSignature *init_sig)
	Check signature of type GNUNET_SIGNATURE_PURPOSE_TRANSPORT_DV_INITIATOR.
static int	dv_neighbour_selection (void *cls, const struct GNUNET_PeerIdentity *pid, void *value)
	Function called for each neighbour during handle_dv_learn.
static int	dv_neighbour_transmission (void *cls, const struct GNUNET_PeerIdentity *pid, void *value)
	Function called for each neighbour during handle_dv_learn.
static unsigned int	calculate_fork_degree (unsigned int hops_taken, unsigned int neighbour_count, unsigned int eligible_count)
	Computes the number of neighbours we should forward a DVInit message to given that it has so far taken hops_taken hops though the network and that the number of neighbours we have in total is neighbour_count, out of which eligible_count are not yet on the path.
static void	neighbour_store_dvmono_cb (void *cls, int success)
	Function called when peerstore is done storing a DV monotonic time.
static struct GNUNET_TIME_Relative	get_network_latency (const struct TransportDVLearnMessage *dvl)
static void	handle_dv_learn (void *cls, const struct TransportDVLearnMessage *dvl)
	Communicator gave us a DV learn message.
static int	check_dv_box (void *cls, const struct TransportDVBoxMessage *dvb)
	Communicator gave us a DV box.
static void	forward_dv_box (struct Neighbour *next_hop, struct TransportDVBoxMessage *hdr, uint16_t total_hops, uint16_t num_hops, const struct GNUNET_PeerIdentity *hops, const void *enc_payload, uint16_t enc_payload_size)
	Create a DV Box message and queue it for transmission to next_hop.
static void	free_backtalker (struct Backtalker *b)
	Free data structures associated with b.
static int	free_backtalker_cb (void *cls, const struct GNUNET_PeerIdentity *pid, void *value)
	Callback to free backtalker records.
static void	backtalker_timeout_cb (void *cls)
	Function called when it is time to clean up a backtalker.
static void	backtalker_monotime_cb (void *cls, const struct GNUNET_PEERSTORE_Record *record, const char *emsg)
	Function called with the monotonic time of a backtalker by PEERSTORE.
static void	backtalker_monotime_store_cb (void *cls, int success)
	Function called by PEERSTORE when the store operation of a backtalker's monotonic time is complete.
static void	update_backtalker_monotime (struct Backtalker *b)
	The backtalker b monotonic time changed.
static void	decaps_dv_box_cb (void *cls, const struct GNUNET_ShortHashCode *km)
static void	handle_dv_box (void *cls, const struct TransportDVBoxMessage *dvb)
	Communicator gave us a DV box.
static int	check_incoming_msg (void *cls, const struct GNUNET_TRANSPORT_IncomingMessage *im)
	Client notified us about transmission from a peer.
static int	check_known_address (void *cls, const struct GNUNET_PeerIdentity *pid, void *value)
	Test if the validation state in value matches the address from cls.
static void	validation_start_cb (void *cls)
	Task run periodically to validate some address based on validation_heap.
static void	update_next_challenge_time (struct ValidationState *vs, struct GNUNET_TIME_Absolute new_time)
	Set the time for next_challenge of vs to new_time.
static void	start_address_validation (const struct GNUNET_PeerIdentity *pid, const char *address)
	Start address validation.
static struct Queue *	find_queue (const struct GNUNET_PeerIdentity *pid, const char *address)
	Find the queue matching pid and address.
static void	suggest_to_connect (const struct GNUNET_PeerIdentity *pid, const char *address)
	Signature of a function called with a communicator address of a peer pid that an application wants us to connect to.
static void	hello_for_incoming_cb (void *cls, const struct GNUNET_PeerIdentity *pid, const char *uri)
static void	handle_hello_for_incoming (void *cls, const struct GNUNET_PEERSTORE_Record *record, const char *emsg)
	Function called by PEERSTORE for each matching record.
static void	hello_for_incoming_error_cb (void *cls)
static void	hello_for_incoming_sync_cb (void *cls)
static void	sign_t_validation_cb (void *cls, const struct GNUNET_PeerIdentity *pid, const struct GNUNET_CRYPTO_EddsaSignature *sig)
static void	handle_validation_challenge (void *cls, const struct TransportValidationChallengeMessage *tvc)
	Communicator gave us a transport address validation challenge.
static int	check_known_challenge (void *cls, const struct GNUNET_PeerIdentity *pid, void *value)
	Test if the validation state in value matches the challenge from cls.
static void	peerstore_store_validation_cb (void *cls, int success)
	Function called when peerstore is done storing a validated address.
static void	validation_transmit_on_queue (struct Queue *q, struct ValidationState *vs)
	The queue q (which matches the peer and address in vs) is ready for queueing.
static void	revalidation_start_cb (void *cls)
static enum GNUNET_GenericReturnValue	revalidate_map_it (void *cls, const struct GNUNET_HashCode *key, void *value)
static void	handle_validation_response (void *cls, const struct TransportValidationResponseMessage *tvr)
	Communicator gave us a transport address validation response.
static void	handle_incoming_msg (void *cls, const struct GNUNET_TRANSPORT_IncomingMessage *im)
	Incoming message.
static int	check_flow_control (void *cls, const struct TransportFlowControlMessage *fc)
	Communicator gave us a transport address validation response.
static void	iterate_address_start_burst (void *cls, const struct GNUNET_PeerIdentity *pid, const char *uri)
static void	check_for_burst_address (void *cls, const struct GNUNET_PEERSTORE_Record *record, const char *emsg)
static void	burst_timeout (void *cls)
static void	start_burst (void *cls)
static void	queue_burst (void *cls)
static void	handle_flow_control (void *cls, const struct TransportFlowControlMessage *fc)
	Communicator gave us a transport address validation response.
static int	check_add_queue_message (void *cls, const struct GNUNET_TRANSPORT_AddQueueMessage *aqm)
	New queue became available.
static void	set_pending_message_uuid (struct PendingMessage *pm)
	If necessary, generates the UUID for a pm.
static struct PendingAcknowledgement *	prepare_pending_acknowledgement (struct Queue *queue, struct DistanceVectorHop *dvh, struct PendingMessage *pm)
	Setup data structure waiting for acknowledgements.
static struct PendingMessage *	fragment_message (struct Queue *queue, struct DistanceVectorHop *dvh, struct PendingMessage *pm)
	Fragment the given pm to the given mtu.
static struct PendingMessage *	reliability_box_message (struct Queue *queue, struct DistanceVectorHop *dvh, struct PendingMessage *pm)
	Reliability-box the given pm.
static void	reorder_root_pm (struct PendingMessage *pm, struct GNUNET_TIME_Absolute next_attempt)
static unsigned int	check_next_attempt_tree (struct PendingMessage *pm, struct PendingMessage *root)
static void	harmonize_flight_round (struct PendingMessage *pm)
static void	update_pm_next_attempt (struct PendingMessage *pm, struct GNUNET_TIME_Absolute next_attempt)
	Change the value of the next_attempt field of pm to next_attempt and re-order pm in the transmission list as required by the new timestamp.
static void	select_best_pending_from_link (struct PendingMessageScoreContext *sc, struct Queue *queue, struct VirtualLink *vl, struct DistanceVectorHop *dvh, size_t overhead)
	Select the best pending message from vl for transmission via queue.
static void	extract_box_cb (void *cls, struct Neighbour *next_hop, const struct GNUNET_MessageHeader *hdr, enum RouteMessageOptions options)
	Function to call to further operate on the now DV encapsulated message hdr, forwarding it via next_hop under respect of options.
static void	handle_del_queue_message (void *cls, const struct GNUNET_TRANSPORT_DelQueueMessage *dqm)
	Queue to a peer went down.
static void	handle_send_message_ack (void *cls, const struct GNUNET_TRANSPORT_SendMessageToAck *sma)
	Message was transmitted.
static void	handle_burst_finished (void *cls, const struct GNUNET_TRANSPORT_BurstFinished *bf)
	The burst finished.
static int	notify_client_queues (void *cls, const struct GNUNET_PeerIdentity *pid, void *value)
	Iterator telling new MONITOR client about all existing queues to peers.
static void	handle_monitor_start (void *cls, const struct GNUNET_TRANSPORT_MonitorStart *start)
	Initialize a monitor client.
static struct TransportClient *	lookup_communicator (const char *prefix)
	Find transport client providing communication service for the protocol prefix.
static int	check_connection_quality (void *cls, const struct GNUNET_PeerIdentity *pid, void *value)
	Check whether any queue to the given neighbour is of a good "quality" and if so, increment the counter.
static void	start_dv_learn (void *cls)
	Task run when we CONSIDER initiating a DV learn process.
static void	sign_dv_init_cb (void *cls, const struct GNUNET_PeerIdentity *pid, const struct GNUNET_CRYPTO_EddsaSignature *sig)
static int	check_validation_request_pending (void *cls, const struct GNUNET_PeerIdentity *pid, void *value)
	A new queue has been created, check if any address validation requests have been waiting for it.
static void	neighbour_dv_monotime_cb (void *cls, const struct GNUNET_PEERSTORE_Record *record, const char *emsg)
	Function called with the monotonic time of a DV initiator by PEERSTORE.
static void	iterate_address_and_compare_cb (void *cls, const struct GNUNET_PeerIdentity *pid, const char *uri)
static enum GNUNET_GenericReturnValue	contains_address (void *cls, const struct GNUNET_PeerIdentity *pid, void *value)
static void	check_for_global_natted_error_cb (void *cls)
static void	check_for_global_natted_sync_cb (void *cls)
static void	check_for_global_natted (void *cls, const struct GNUNET_PEERSTORE_Record *record, const char *emsg)
static void	handle_add_queue_message (void *cls, const struct GNUNET_TRANSPORT_AddQueueMessage *aqm)
	New queue became available.
static void	handle_update_queue_message (void *cls, const struct GNUNET_TRANSPORT_UpdateQueueMessage *msg)
	Handle updates to queues.
static void	handle_queue_create_ok (void *cls, const struct GNUNET_TRANSPORT_CreateQueueResponse *cqr)
	Communicator tells us that our request to create a queue "worked", that is setting up the queue is now in process.
static void	handle_queue_create_fail (void *cls, const struct GNUNET_TRANSPORT_CreateQueueResponse *cqr)
	Communicator tells us that our request to create a queue failed.
static void	handle_suggest_cancel (void *cls, const struct ExpressPreferenceMessage *msg)
	We have received a struct ExpressPreferenceMessage from an application client.
static void	hello_for_client_cb (void *cls, const struct GNUNET_PeerIdentity *pid, const char *uri)
static void	handle_hello_for_client (void *cls, const struct GNUNET_PEERSTORE_Record *record, const char *emsg)
	Function called by PEERSTORE for each matching record.
static void	hello_for_client_error_cb (void *cls)
static void	hello_for_client_sync_cb (void *cls)
static void	handle_suggest (void *cls, const struct ExpressPreferenceMessage *msg)
	We have received a struct ExpressPreferenceMessage from an application client.
static int	check_request_hello_validation (void *cls, const struct RequestHelloValidationMessage *m)
	Check GNUNET_MESSAGE_TYPE_TRANSPORT_REQUEST_HELLO_VALIDATION messages.
static void	handle_request_hello_validation (void *cls, const struct RequestHelloValidationMessage *m)
	A client encountered an address of another peer.
static int	free_neighbour_cb (void *cls, const struct GNUNET_PeerIdentity *pid, void *value)
	Free neighbour entry.
static int	free_dv_routes_cb (void *cls, const struct GNUNET_PeerIdentity *pid, void *value)
	Free DV route entry.
static int	free_validation_state_cb (void *cls, const struct GNUNET_PeerIdentity *pid, void *value)
	Free validation state.
static int	free_pending_ack_cb (void *cls, const struct GNUNET_Uuid *key, void *value)
	Free pending acknowledgement.
static int	free_ack_cummulator_cb (void *cls, const struct GNUNET_PeerIdentity *pid, void *value)
	Free acknowledgement cummulator.
static void	shutdown_task (void *cls)
static void	update_hello_from_pid_change_cb (void *cls, int success)
void	print_address_list (void *cls, const struct GNUNET_PeerIdentity *pid, const char *uri)
static void	pils_pid_change_cb (void *cls, const struct GNUNET_HELLO_Parser *parser, const struct GNUNET_HashCode *hash)
	Callback called when pils service updates us with our new peer identity.
static void	run (void *cls, const struct GNUNET_CONFIGURATION_Handle *c, struct GNUNET_SERVICE_Handle *service)
	Initiate transport service.
	GNUNET_SERVICE_MAIN (GNUNET_OS_project_data_gnunet(), "transport", GNUNET_SERVICE_OPTION_SOFT_SHUTDOWN, &run, &client_connect_cb, &client_disconnect_cb, NULL, GNUNET_MQ_hd_fixed_size(suggest, GNUNET_MESSAGE_TYPE_TRANSPORT_SUGGEST, struct ExpressPreferenceMessage, NULL), GNUNET_MQ_hd_fixed_size(suggest_cancel, GNUNET_MESSAGE_TYPE_TRANSPORT_SUGGEST_CANCEL, struct ExpressPreferenceMessage, NULL), GNUNET_MQ_hd_var_size(request_hello_validation, GNUNET_MESSAGE_TYPE_TRANSPORT_REQUEST_HELLO_VALIDATION, struct RequestHelloValidationMessage, NULL), GNUNET_MQ_hd_fixed_size(client_start, GNUNET_MESSAGE_TYPE_TRANSPORT_START, struct StartMessage, NULL), GNUNET_MQ_hd_var_size(client_send, GNUNET_MESSAGE_TYPE_TRANSPORT_SEND, struct OutboundMessage, NULL), GNUNET_MQ_hd_fixed_size(client_recv_ok, GNUNET_MESSAGE_TYPE_TRANSPORT_RECV_OK, struct RecvOkMessage, NULL), GNUNET_MQ_hd_var_size(communicator_available, GNUNET_MESSAGE_TYPE_TRANSPORT_NEW_COMMUNICATOR, struct GNUNET_TRANSPORT_CommunicatorAvailableMessage, NULL), GNUNET_MQ_hd_var_size(communicator_backchannel, GNUNET_MESSAGE_TYPE_TRANSPORT_COMMUNICATOR_BACKCHANNEL, struct GNUNET_TRANSPORT_CommunicatorBackchannel, NULL), GNUNET_MQ_hd_var_size(add_address, GNUNET_MESSAGE_TYPE_TRANSPORT_ADD_ADDRESS, struct GNUNET_TRANSPORT_AddAddressMessage, NULL), GNUNET_MQ_hd_fixed_size(del_address, GNUNET_MESSAGE_TYPE_TRANSPORT_DEL_ADDRESS, struct GNUNET_TRANSPORT_DelAddressMessage, NULL), GNUNET_MQ_hd_var_size(incoming_msg, GNUNET_MESSAGE_TYPE_TRANSPORT_INCOMING_MSG, struct GNUNET_TRANSPORT_IncomingMessage, NULL), GNUNET_MQ_hd_fixed_size(queue_create_ok, GNUNET_MESSAGE_TYPE_TRANSPORT_QUEUE_CREATE_OK, struct GNUNET_TRANSPORT_CreateQueueResponse, NULL), GNUNET_MQ_hd_fixed_size(queue_create_fail, GNUNET_MESSAGE_TYPE_TRANSPORT_QUEUE_CREATE_FAIL, struct GNUNET_TRANSPORT_CreateQueueResponse, NULL), GNUNET_MQ_hd_var_size(add_queue_message, GNUNET_MESSAGE_TYPE_TRANSPORT_QUEUE_SETUP, struct GNUNET_TRANSPORT_AddQueueMessage, NULL), GNUNET_MQ_hd_fixed_size(update_queue_message, GNUNET_MESSAGE_TYPE_TRANSPORT_QUEUE_UPDATE, struct GNUNET_TRANSPORT_UpdateQueueMessage, NULL), GNUNET_MQ_hd_fixed_size(del_queue_message, GNUNET_MESSAGE_TYPE_TRANSPORT_QUEUE_TEARDOWN, struct GNUNET_TRANSPORT_DelQueueMessage, NULL), GNUNET_MQ_hd_fixed_size(send_message_ack, GNUNET_MESSAGE_TYPE_TRANSPORT_SEND_MSG_ACK, struct GNUNET_TRANSPORT_SendMessageToAck, NULL), GNUNET_MQ_hd_fixed_size(burst_finished, GNUNET_MESSAGE_TYPE_TRANSPORT_BURST_FINISHED, struct GNUNET_TRANSPORT_BurstFinished, NULL), GNUNET_MQ_hd_fixed_size(monitor_start, GNUNET_MESSAGE_TYPE_TRANSPORT_MONITOR_START, struct GNUNET_TRANSPORT_MonitorStart, NULL), GNUNET_MQ_handler_end())
	Define "main" method using service macro.

Variables
static struct RingBufferEntry *	ring_buffer [RING_BUFFER_SIZE]
	Ring buffer for a CORE message we did not deliver to CORE, because of missing virtual link to sender.
static unsigned int	ring_buffer_head
	Head of the ring buffer.
static unsigned int	is_ring_buffer_full
	Is the ring buffer filled up to RING_BUFFER_SIZE.
static struct PendingMessage *	ring_buffer_dv [RING_BUFFER_SIZE]
	Ring buffer for a forwarded DVBox message we did not deliver to the next hop, because of missing virtual link that hop.
static unsigned int	ring_buffer_dv_head
	Head of the ring buffer.
static unsigned int	is_ring_buffer_dv_full
	Is the ring buffer filled up to RING_BUFFER_SIZE.
static struct TransportClient *	clients_head
	Head of linked list of all clients to this service.
static struct TransportClient *	clients_tail
	Tail of linked list of all clients to this service.
static struct GNUNET_STATISTICS_Handle *	GST_stats
	Statistics handle.
static const struct GNUNET_CONFIGURATION_Handle *	GST_cfg
	Configuration handle.
static struct GNUNET_PeerIdentity *	GST_my_identity
	Our public key.
struct GNUNET_HELLO_Builder *	GST_my_hello
	Our HELLO.
static struct GNUNET_CONTAINER_MultiPeerMap *	neighbours
	Map from PIDs to struct Neighbour entries.
static struct GNUNET_CONTAINER_MultiPeerMap *	backtalkers
	Map from PIDs to struct Backtalker entries.
static struct GNUNET_CONTAINER_MultiPeerMap *	ack_cummulators
	Map from PIDs to struct AcknowledgementCummulators.
static struct GNUNET_CONTAINER_MultiUuidmap *	pending_acks
	Map of pending acknowledgements, mapping struct AcknowledgementUUID to a struct PendingAcknowledgement.
static struct GNUNET_CONTAINER_MultiPeerMap *	dv_routes
	Map from PIDs to struct DistanceVector entries describing known paths to the peer.
static struct GNUNET_CONTAINER_MultiPeerMap *	validation_map
	Map from PIDs to struct ValidationState entries describing addresses we are aware of and their validity state.
static struct GNUNET_CONTAINER_MultiHashMap *	revalidation_map
	Map from addresses to struct ValidationState entries describing addresses we are aware of and their validity state.
static struct GNUNET_CONTAINER_MultiPeerMap *	links
	Map from PIDs to struct VirtualLink entries describing links CORE knows to exist.
static struct GNUNET_CONTAINER_MultiShortmap *	dvlearn_map
	Map from challenges to struct LearnLaunchEntry values.
static struct LearnLaunchEntry *	lle_head = NULL
	Head of a DLL sorted by launch time.
static struct LearnLaunchEntry *	lle_tail = NULL
	Tail of a DLL sorted by launch time.
static struct GNUNET_CONTAINER_Heap *	validation_heap
	MIN Heap sorted by "next_challenge" to struct ValidationState entries sorting addresses we are aware of by when we should next try to (re)validate (or expire) them.
struct GNUNET_NAT_Handle *	nh
	Handle for connect to the NAT service.
static struct GNUNET_PEERSTORE_Handle *	peerstore
	Database for peer's HELLOs.
static struct GNUNET_PILS_Handle *	pils
	Service that manages our peer id.
static struct PilsRequest *	pils_requests_head
	PILS Operation DLL.
static struct PilsRequest *	pils_requests_tail
	PILS Operation DLL.
static struct GNUNET_SCHEDULER_Task *	dvlearn_task
	Task run to initiate DV learning.
static struct GNUNET_SCHEDULER_Task *	validation_task
	Task to run address validation.
static struct GNUNET_SCHEDULER_Task *	pils_feed_task
	Task to feed addresses to PILS.
static struct IncomingRequest *	ir_head
	List of incoming connections where we are trying to get a connection back established.
static struct IncomingRequest *	ir_tail
	Tail of DLL starting at ir_head.
static unsigned int	ir_total
	Length of the DLL starting at ir_head.
static unsigned long long	logging_uuid_gen
	Generator of logging_uuid in struct PendingMessage.
static enum GNUNET_GenericReturnValue	burst_running
	Is there a burst running?
static struct GNUNET_TIME_Absolute	hello_mono_time
	Monotonic time we use for HELLOs generated at this time.
static int	in_shutdown
	Indication if we have received a shutdown signal and are in the process of cleaning up.
static struct GNUNET_SCHEDULER_Task *	burst_task
	The task to start the burst.
struct GNUNET_SCHEDULER_Task *	burst_timeout_task
enum GNUNET_GenericReturnValue	use_burst

Detailed Description

main for gnunet-service-transport

Author

Christian Grothoff

TODO: Implement next:

• review retransmission logic, right now there is no smartness there! => congestion control, etc [PERFORMANCE-BASICS]

Optimizations-Statistics:

• Track ACK losses based on ACK-counter [ROUTING]
• Need to track total bandwidth per VirtualLink and adjust how frequently we send FC messages based on bandwidth-delay-product (and relation to the window size!). See OPTIMIZE-FC-BDP.
• Consider more statistics in check_connection_quality() [FIXME-CONQ-STATISTICS]
• Adapt available_fc_window_size, using larger values for high-bandwidth and high-latency links if we have the RAM [GOODPUT / utilization / stalls]
• Set last_window_consum_limit promise properly based on latency and bandwidth of the respective connection [GOODPUT / utilization / stalls]

Optimizations-DV:

• When forwarding DV learn messages, if a peer is reached that has a bidirectional link to the origin beyond 1st hop, do NOT forward it to peers other than the origin, as there is clearly a better path directly from the origin to whatever else we could reach.
• When we passively learned DV (with unconfirmed freshness), we right now add the path to our list but with a zero path_valid_until time and only use it for unconfirmed routes. However, we could consider triggering an explicit validation mechanism ourselves, specifically routing a challenge-response message over the path [ROUTING] = if available, try to confirm unconfirmed DV paths when trying to establish virtual link for a struct IncomingRequest. (i.e. if DVH is unconfirmed, incoming requests cause us to try to validate a passively learned path (requires new message type!))

Optimizations-Fragmentation:

• Fragments send over a reliable channel could do without the AcknowledgementUUIDP altogether, as they won't be acked! [BANDWIDTH] (-> have 2nd type of acknowledgment message; low priority, as we do not have an MTU-limited reliable communicator) [FIXME-FRAG-REL-UUID]
• if messages are below MTU, consider adding ACKs and other stuff to the same transmission to avoid tiny messages (requires planning at receiver, and additional MST-style demultiplex at receiver!) [PACKET COUNT]

Optimizations-internals:

• queue_send_msg by API design has to make a copy of the payload, and route_message on top of that requires a malloc/free. Change design to approximate "zero" copy better... [CPU]
• could avoid copying body of message into each fragment and keep fragments as just pointers into the original message and only fully build fragments just before transmission (optimization, should reduce CPU and memory use) [CPU, MEMORY]

Definition in file gnunet-service-transport.c.

Macro Definition Documentation

RING_BUFFER_SIZE

#define RING_BUFFER_SIZE   16

Size of ring buffer to cache CORE and forwarded DVBox messages.

Definition at line 89 of file gnunet-service-transport.c.

MAX_FC_RETRANSMIT_COUNT

#define MAX_FC_RETRANSMIT_COUNT   1000

Maximum number of FC retransmissions for a running retransmission task.

Definition at line 94 of file gnunet-service-transport.c.

MAX_CUMMULATIVE_ACKS

#define MAX_CUMMULATIVE_ACKS   64

Maximum number of messages we acknowledge together in one cumulative ACK.

Larger values may save a bit of bandwidth.

Definition at line 100 of file gnunet-service-transport.c.

FC_NO_CHANGE_REPLY_PROBABILITY

#define FC_NO_CHANGE_REPLY_PROBABILITY   8

What is the 1:n chance that we send a Flow control response when receiving a flow control message that did not change anything for us? Basically, this is used in the case where both peers are stuck on flow control (no window changes), but one might continue sending flow control messages to the other peer as the first FC message when things stalled got lost, and then subsequently the other peer does usually not respond as nothing changed.

So to ensure that eventually the FC messages stop, we do send with 1/8th probability an FC message even if nothing changed. That prevents one peer being stuck in sending (useless) FC messages "forever".

Definition at line 114 of file gnunet-service-transport.c.

IN_PACKET_SIZE_WITHOUT_MTU

#define IN_PACKET_SIZE_WITHOUT_MTU   128

What is the size we assume for a read operation in the absence of an MTU for the purpose of flow control?

Definition at line 120 of file gnunet-service-transport.c.

GOODPUT_AGING_SLOTS

#define GOODPUT_AGING_SLOTS   4

Number of slots we keep of historic data for computation of goodput / message loss ratio.

Definition at line 126 of file gnunet-service-transport.c.

DEFAULT_WINDOW_SIZE

#define DEFAULT_WINDOW_SIZE   (128 * 1024)

How big is the flow control window size by default; limits per-neighbour RAM utilization.

Definition at line 132 of file gnunet-service-transport.c.

MAX_INCOMING_REQUEST

#define MAX_INCOMING_REQUEST   16

For how many incoming connections do we try to create a virtual link for (at the same time!).

This does NOT limit the number of incoming connections, just the number for which we are actively trying to find working addresses in the absence (!) of our own applications wanting the link to go up.

Definition at line 142 of file gnunet-service-transport.c.

MAX_DV_DISCOVERY_SELECTION

#define MAX_DV_DISCOVERY_SELECTION   16

Maximum number of peers we select for forwarding DVInit messages at the same time (excluding initiator).

Definition at line 148 of file gnunet-service-transport.c.

RECV_WINDOW_SIZE

#define RECV_WINDOW_SIZE   4

Window size.

How many messages to the same target do we pass to CORE without a RECV_OK in between? Small values limit throughput, large values will increase latency.

FIXME-OPTIMIZE: find out what good values are experimentally, maybe set adaptively (i.e. to observed available bandwidth).

Definition at line 158 of file gnunet-service-transport.c.

MIN_DV_PATH_LENGTH_FOR_INITIATOR

#define MIN_DV_PATH_LENGTH_FOR_INITIATOR   3

Minimum number of hops we should forward DV learn messages even if they are NOT useful for us in hope of looping back to the initiator?

FIXME: allow initiator some control here instead?

Definition at line 167 of file gnunet-service-transport.c.

MAX_DV_HOPS_ALLOWED

#define MAX_DV_HOPS_ALLOWED   16

Maximum DV distance allowed ever.

Definition at line 172 of file gnunet-service-transport.c.

MAX_DV_LEARN_PENDING

#define MAX_DV_LEARN_PENDING   64

Maximum number of DV learning activities we may have pending at the same time.

Definition at line 178 of file gnunet-service-transport.c.

MAX_DV_PATHS_TO_TARGET

#define MAX_DV_PATHS_TO_TARGET   3

Maximum number of DV paths we keep simultaneously to the same target.

Definition at line 183 of file gnunet-service-transport.c.

PILS_FEED_ADDRESSES_DELAY

#define PILS_FEED_ADDRESSES_DELAY    GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_SECONDS, 3)

Delay between added/removed addresses and PILS feed call.

Introduced to handle cases with high address churn across communicators (startup, location change etc)

Definition at line 191 of file gnunet-service-transport.c.

{
  uint64_t uuid GNUNET_PACKED;
};
 
 
struct AcknowledgementUUIDP
{
  struct GNUNET_Uuid value;
};
 
struct TransportBackchannelEncapsulationMessage
{
  struct GNUNET_MessageHeader header;
 
  /* Followed by *another* message header which is the message to
     the communicator */
 
  /* Followed by a 0-terminated name of the communicator */
};
 
 
struct EphemeralConfirmationPS
{
  struct GNUNET_CRYPTO_EccSignaturePurpose purpose;
 
  struct GNUNET_TIME_AbsoluteNBO sender_monotonic_time;
 
  struct GNUNET_PeerIdentity target;
 
  struct GNUNET_CRYPTO_HpkeEncapsulation ephemeral_key;
};
 
 
struct TransportDVBoxPayloadP
{
  struct GNUNET_PeerIdentity sender;
 
  struct GNUNET_CRYPTO_EddsaSignature sender_sig;
 
  struct GNUNET_TIME_AbsoluteNBO monotonic_time;
 
  /* Followed by a `struct GNUNET_MessageHeader` with a message
     for the target peer */
};
 
 
struct TransportReliabilityBoxMessage
{
  struct GNUNET_MessageHeader header;
 
  uint32_t ack_countdown GNUNET_PACKED;
 
  struct AcknowledgementUUIDP ack_uuid;
};
 
 
struct TransportCummulativeAckPayloadP
{
  struct GNUNET_TIME_RelativeNBO ack_delay;
 
  struct AcknowledgementUUIDP ack_uuid;
};
 
 
struct TransportReliabilityAckMessage
{
  struct GNUNET_MessageHeader header;
 
  uint32_t ack_counter GNUNET_PACKED;
 
  /* followed by any number of `struct TransportCummulativeAckPayloadP`
     messages providing ACKs */
};
 
 
struct TransportFragmentBoxMessage
{
  struct GNUNET_MessageHeader header;
 
  uint16_t frag_off GNUNET_PACKED;
 
  uint16_t msg_size GNUNET_PACKED;
 
  struct AcknowledgementUUIDP ack_uuid;
 
  struct MessageUUIDP msg_uuid;
};
 
 
struct DvInitPS
{
  struct GNUNET_CRYPTO_EccSignaturePurpose purpose;
 
  struct GNUNET_TIME_AbsoluteNBO monotonic_time;
 
  struct GNUNET_CRYPTO_ChallengeNonceP challenge;
};
 
 
struct DvHopPS
{
  struct GNUNET_CRYPTO_EccSignaturePurpose purpose;
 
  struct GNUNET_PeerIdentity pred;
 
  struct GNUNET_PeerIdentity succ;
 
  struct GNUNET_CRYPTO_ChallengeNonceP challenge;
};
 
 
struct DVPathEntryP
{
  struct GNUNET_PeerIdentity hop;
 
  struct GNUNET_CRYPTO_EddsaSignature hop_sig;
};
 
 
struct TransportDVLearnMessage
{
  struct GNUNET_MessageHeader header;
 
  uint16_t num_hops GNUNET_PACKED;
 
  uint16_t bidirectional GNUNET_PACKED;
 
  struct GNUNET_TIME_RelativeNBO non_network_delay;
 
  struct GNUNET_TIME_AbsoluteNBO monotonic_time;
 
  struct GNUNET_CRYPTO_EddsaSignature init_sig;
 
  struct GNUNET_PeerIdentity initiator;
 
  struct GNUNET_CRYPTO_ChallengeNonceP challenge;
 
  /* Followed by @e num_hops `struct DVPathEntryP` values,
     excluding the initiator of the DV trace; the last entry is the
     current sender; the current peer must not be included. */
};
 
 
struct TransportDVBoxMessage
{
  struct GNUNET_MessageHeader header;
 
  unsigned int without_fc;
 
  uint16_t total_hops GNUNET_PACKED;
 
  uint16_t num_hops GNUNET_PACKED;
 
  struct GNUNET_CRYPTO_HpkeEncapsulation ephemeral_key;
 
  struct GNUNET_ShortHashCode iv;
 
  struct GNUNET_HashCode hmac;
 
  uint16_t orig_size GNUNET_PACKED;
 
  /* Followed by @e num_hops `struct GNUNET_PeerIdentity` values;
     excluding the @e origin and the current peer, the last must be
     the ultimate target; if @e num_hops is zero, the receiver of this
     message is the ultimate target. */
 
  /* Followed by encrypted, variable-size payload, which
     must begin with a `struct TransportDVBoxPayloadP` */
 
  /* Followed by the actual message, which itself must not be a
     a DV_LEARN or DV_BOX message! */
};
 
 
struct TransportValidationChallengeMessage
{
  struct GNUNET_MessageHeader header;
 
  uint32_t reserved GNUNET_PACKED;
 
  struct GNUNET_CRYPTO_ChallengeNonceP challenge;
 
  struct GNUNET_TIME_AbsoluteNBO sender_time;
};
 
 
struct TransportValidationPS
{
  struct GNUNET_CRYPTO_EccSignaturePurpose purpose;
 
  struct GNUNET_TIME_RelativeNBO validity_duration;
 
  struct GNUNET_CRYPTO_ChallengeNonceP challenge;
};
 
 
struct TransportValidationResponseMessage
{
  struct GNUNET_MessageHeader header;
 
  uint32_t reserved GNUNET_PACKED;
 
  struct GNUNET_CRYPTO_EddsaSignature signature;
 
  struct GNUNET_CRYPTO_ChallengeNonceP challenge;
 
  struct GNUNET_TIME_AbsoluteNBO origin_time;
 
  struct GNUNET_TIME_RelativeNBO validity_duration;
};
 
struct TransportGlobalNattedAddress
{
  unsigned int address_length;
 
  /* Followed by @e address_length bytes of the address. */
};
 
struct TransportFlowControlMessage
{
  struct GNUNET_MessageHeader header;
 
  uint32_t seq GNUNET_PACKED;
 
  uint64_t inbound_window_size GNUNET_PACKED;
 
  uint64_t outbound_sent GNUNET_PACKED;
 
  uint64_t outbound_window_size GNUNET_PACKED;
 
  struct GNUNET_TIME_AbsoluteNBO sender_time;
 
  struct GNUNET_TIME_RelativeNBO rtt;
 
  unsigned int sync_ready;
 
  unsigned int number_of_addresses;
 
  size_t size_of_addresses;
 
  /* Followed by @e number_of_addresses struct TransportGlobalNattedAddress. */
};
 
GNUNET_NETWORK_STRUCT_END
 
 
enum ClientType
{
  CT_NONE = 0,
 
  CT_CORE = 1,
 
  CT_MONITOR = 2,
 
  CT_COMMUNICATOR = 3,
 
  CT_APPLICATION = 4
};
 
 
enum RouteMessageOptions
{
  RMO_NONE = 0,
 
  RMO_DV_ALLOWED = 1,
 
  RMO_UNCONFIRMED_ALLOWED = 2,
 
  RMO_ANYTHING_GOES = (RMO_DV_ALLOWED | RMO_UNCONFIRMED_ALLOWED),
 
  RMO_REDUNDANT = 4
};
 
 
struct LearnLaunchEntry
{
  struct LearnLaunchEntry *prev;
 
  struct LearnLaunchEntry *next;
 
  struct GNUNET_CRYPTO_ChallengeNonceP challenge;
 
  struct GNUNET_TIME_Absolute launch_time;
};
 
 
struct TransmissionHistoryEntry
{
  uint64_t bytes_sent;
 
  uint64_t bytes_received;
};
 
 
struct PerformanceData
{
  struct GNUNET_TIME_Relative aged_rtt;
 
  struct TransmissionHistoryEntry the[GOODPUT_AGING_SLOTS];
 
  unsigned int last_age;
};
 
 
struct TransportClient;
 
struct Neighbour;
 
struct DistanceVector;
 
struct Queue;
 
struct PendingMessage;
 
struct DistanceVectorHop;
 
struct VirtualLink;
 
 
struct CommunicatorMessageContext
{
  struct CommunicatorMessageContext *next;
 
  struct CommunicatorMessageContext *prev;
 
  struct TransportClient *tc;
 
  struct GNUNET_TRANSPORT_IncomingMessage im;
 
  const struct GNUNET_MessageHeader *mh;
 
  uint16_t total_hops;
 
  unsigned int continue_send;
};
 
 
struct RingBufferEntry
{
  struct CommunicatorMessageContext *cmc;
 
  struct GNUNET_MessageHeader *mh;
};
 
 
struct CoreSentContext
{
  struct CoreSentContext *next;
 
  struct CoreSentContext *prev;
 
  struct VirtualLink *vl;
 
  uint16_t size;
 
  uint16_t isize;
};
 
 
struct ReassemblyContext
{
  struct MessageUUIDP msg_uuid;
 
  struct VirtualLink *virtual_link;
 
  struct GNUNET_CONTAINER_HeapNode *hn;
 
  uint8_t *bitfield;
 
  struct GNUNET_TIME_Absolute reassembly_timeout;
 
  struct GNUNET_TIME_Absolute last_frag;
 
  uint16_t msg_size;
 
  uint16_t msg_missing;
 
  /* Followed by @e msg_size bytes of the (partially) defragmented original
   * message */
 
  /* Followed by @e bitfield data */
};
 
 
struct VirtualLink
{
  struct GNUNET_PeerIdentity target;
 
  struct GNUNET_CONTAINER_MultiHashMap32 *reassembly_map;
 
  struct GNUNET_CONTAINER_Heap *reassembly_heap;
 
  struct GNUNET_SCHEDULER_Task *reassembly_timeout_task;
 
  struct CommunicatorMessageContext *cmc_head;
 
  struct CommunicatorMessageContext *cmc_tail;
 
  struct PendingMessage *pending_msg_head;
 
  struct PendingMessage *pending_msg_tail;
 
  struct CoreSentContext *csc_tail;
 
  struct CoreSentContext *csc_head;
 
  struct GNUNET_SCHEDULER_Task *visibility_task;
 
  struct GNUNET_SCHEDULER_Task *fc_retransmit_task;
 
  struct GNUNET_StartBurstCls *sb_cls;
 
  char *burst_addr;
 
  unsigned int fc_retransmit_count;
 
  unsigned int confirmed;
 
  struct Neighbour *n;
 
  struct DistanceVector *dv;
 
  struct GNUNET_TIME_Absolute n_challenge_time;
 
  struct GNUNET_TIME_Absolute last_fc_transmission;
 
  struct GNUNET_TIME_Absolute last_fc_timestamp;
 
  struct GNUNET_TIME_Relative last_fc_rtt;
 
  struct GNUNET_TIME_Relative other_rtt;
 
  struct GNUNET_PEERSTORE_IterateContext *ic;
 
  uint64_t message_uuid_ctr;
 
  uint64_t available_fc_window_size;
 
  uint64_t incoming_fc_window_size_ram;
 
  uint64_t incoming_fc_window_size;
 
  uint64_t incoming_fc_window_size_used;
 
  int64_t incoming_fc_window_size_loss;
 
  uint64_t outbound_fc_window_size;
 
  uint64_t outbound_fc_window_size_used;
 
  uint64_t last_outbound_window_size_received;
 
  uint32_t fc_seq_gen;
 
  uint32_t last_fc_seq;
 
  int core_recv_window;
 
  enum GNUNET_GenericReturnValue sync_ready;
};
 
 
struct PendingAcknowledgement
{
  struct PendingAcknowledgement *next_pm;
 
  struct PendingAcknowledgement *prev_pm;
 
  struct PendingAcknowledgement *next_queue;
 
  struct PendingAcknowledgement *prev_queue;
 
  struct PendingAcknowledgement *next_dvh;
 
  struct PendingAcknowledgement *prev_dvh;
 
  struct PendingAcknowledgement *next_pa;
 
  struct PendingAcknowledgement *prev_pa;
 
  struct AcknowledgementUUIDP ack_uuid;
 
  struct PendingMessage *pm;
 
  struct DistanceVectorHop *dvh;
 
  struct Queue *queue;
 
  struct GNUNET_TIME_Absolute transmission_time;
 
  uint16_t message_size;
 
  unsigned int num_send;
};
 
 
struct DistanceVectorHop
{
  struct DistanceVectorHop *next_dv;
 
  struct DistanceVectorHop *prev_dv;
 
  struct DistanceVectorHop *next_neighbour;
 
  struct DistanceVectorHop *prev_neighbour;
 
  struct PendingAcknowledgement *pa_head;
 
  struct PendingAcknowledgement *pa_tail;
 
  struct Neighbour *next_hop;
 
  struct DistanceVector *dv;
 
  const struct GNUNET_PeerIdentity *path;
 
  struct GNUNET_TIME_Absolute timeout;
 
  struct GNUNET_TIME_Absolute path_valid_until;
 
  struct PerformanceData pd;
 
  unsigned int distance;
};
 
 
struct DistanceVector
{
  struct GNUNET_PeerIdentity target;
 
  struct DistanceVectorHop *dv_head;
 
  struct DistanceVectorHop *dv_tail;
 
  struct GNUNET_SCHEDULER_Task *timeout_task;
 
  struct VirtualLink *vl;
 
  struct GNUNET_CRYPTO_EddsaSignature sender_sig;
 
  struct GNUNET_TIME_Absolute ephemeral_validity;
 
  struct GNUNET_TIME_Absolute monotime;
 
  struct GNUNET_CRYPTO_HpkeEncapsulation ephemeral_key;
 
  struct GNUNET_ShortHashCode *km;
};
 
 
struct QueueEntry
{
  struct QueueEntry *next;
 
  struct QueueEntry *prev;
 
  struct Queue *queue;
 
  struct PendingMessage *pm;
 
  uint64_t mid;
 
  struct GNUNET_TIME_Absolute creation_timestamp;
};
 
 
struct Queue
{
  struct Queue *next_neighbour;
 
  struct Queue *prev_neighbour;
 
  struct Queue *prev_client;
 
  struct Queue *next_client;
 
  struct PendingAcknowledgement *pa_head;
 
  struct PendingAcknowledgement *pa_tail;
 
  struct QueueEntry *queue_head;
 
  struct QueueEntry *queue_tail;
 
  struct Neighbour *neighbour;
 
  struct TransportClient *tc;
 
  const char *address;
 
  unsigned int unlimited_length;
 
  struct GNUNET_SCHEDULER_Task *transmit_task;
 
  struct GNUNET_TIME_Absolute validated_until;
 
  struct PerformanceData pd;
 
  struct GNUNET_PEERSTORE_Monitor *mo;
 
  uint64_t mid_gen;
 
  uint32_t qid;
 
  uint32_t mtu;
 
  uint32_t num_msg_pending;
 
  uint32_t num_bytes_pending;
 
  unsigned int queue_length;
 
  uint64_t q_capacity;
 
  uint32_t priority;
 
  enum GNUNET_NetworkType nt;
 
  enum GNUNET_TRANSPORT_ConnectionStatus cs;
 
  int idle;
 
  enum GNUNET_GenericReturnValue is_global_natted;
};
 
 
struct Neighbour
{
  struct GNUNET_PeerIdentity pid;
 
  struct DistanceVectorHop *dv_head;
 
  struct DistanceVectorHop *dv_tail;
 
  struct Queue *queue_head;
 
  struct Queue *queue_tail;
 
  struct GNUNET_PEERSTORE_IterateContext *get;
 
  struct GNUNET_PEERSTORE_StoreContext *sc;
 
  struct VirtualLink *vl;
 
  struct GNUNET_TIME_Absolute last_dv_learn_monotime;
 
  int dv_monotime_available;
 
  struct GNUNET_CONTAINER_MultiPeerMap *natted_addresses;
 
  unsigned int number_of_addresses;
 
  size_t size_of_global_addresses;
 
  enum GNUNET_GenericReturnValue is_global_natted;
};
 
 
struct IncomingRequest
{
  struct IncomingRequest *next;
 
  struct IncomingRequest *prev;
 
  struct GNUNET_PEERSTORE_Monitor *nc;
 
  struct GNUNET_PeerIdentity pid;
};
 
 
struct PeerRequest
{
  struct GNUNET_PeerIdentity pid;
 
  struct TransportClient *tc;
 
  struct GNUNET_PEERSTORE_Monitor *nc;
 
  enum GNUNET_MQ_PriorityPreferences pk;
 
  struct GNUNET_BANDWIDTH_Value32NBO bw;
};
 
 
enum PendingMessageType
{
  PMT_CORE = 0,
 
  PMT_FRAGMENT_BOX = 1,
 
  PMT_RELIABILITY_BOX = 2,
 
  PMT_DV_BOX = 3
};
 
 
struct PendingMessage
{
  struct PendingMessage *next_vl;
 
  struct PendingMessage *prev_vl;
 
  struct PendingMessage *next_client;
 
  struct PendingMessage *prev_client;
 
  struct PendingMessage *next_frag;
 
  struct PendingMessage *prev_frag;
 
  struct PendingAcknowledgement *pa_head;
 
  struct PendingAcknowledgement *pa_tail;
 
  struct PendingMessage *bpm;
 
  struct VirtualLink *vl;
 
  struct GNUNET_PeerIdentity target;
 
  struct QueueEntry *qe;
 
  struct TransportClient *client;
 
  struct PendingMessage *head_frag;
 
  struct PendingMessage *tail_frag;
 
  struct PendingMessage *frag_parent;
 
  struct GNUNET_TIME_Absolute timeout;
 
  struct GNUNET_TIME_Absolute next_attempt;
 
  struct MessageUUIDP msg_uuid;
 
  uint64_t logging_uuid;
 
  enum PendingMessageType pmt;
 
  enum GNUNET_MQ_PriorityPreferences prefs;
 
  struct DistanceVectorHop *used_dvh;
 
  uint16_t bytes_msg;
 
  uint16_t frag_off;
 
  uint32_t frags_in_flight;
 
  uint32_t frags_in_flight_round;
 
  uint16_t frag_count;
 
  int16_t msg_uuid_set;
 
  /* Followed by @e bytes_msg to transmit */
};
 
 
struct TransportCummulativeAckPayload
{
  struct GNUNET_TIME_Absolute receive_time;
 
  struct AcknowledgementUUIDP ack_uuid;
};
 
 
struct AcknowledgementCummulator
{
  struct GNUNET_PeerIdentity target;
 
  struct TransportCummulativeAckPayload ack_uuids[MAX_CUMMULATIVE_ACKS];
 
  struct GNUNET_SCHEDULER_Task *task;
 
  struct GNUNET_TIME_Absolute min_transmission_time;
 
  uint32_t ack_counter;
 
  unsigned int num_acks;
};
 
 
struct AddressListEntry
{
  struct AddressListEntry *next;
 
  struct AddressListEntry *prev;
 
  struct TransportClient *tc;
 
  struct GNUNET_PEERSTORE_StoreHelloContext *shc;
 
  const char *address;
 
  void *signed_address;
 
  size_t signed_address_len;
 
  struct GNUNET_PEERSTORE_StoreContext *sc;
 
  struct GNUNET_SCHEDULER_Task *st;
 
  struct GNUNET_TIME_Relative expiration;
 
  uint32_t aid;
 
  enum GNUNET_NetworkType nt;
};
 
 
struct TransportClient
{
  struct TransportClient *next;
 
  struct TransportClient *prev;
 
  struct GNUNET_SERVICE_Client *client;
 
  struct GNUNET_MQ_Handle *mq;
 
  enum ClientType type;
 
  union
  {
    struct
    {
      struct PendingMessage *pending_msg_head;
 
      struct PendingMessage *pending_msg_tail;
    } core;
 
    struct
    {
      struct GNUNET_PeerIdentity peer;
 
      int one_shot;
    } monitor;
 
 
    struct
    {
      char *address_prefix;
 
      struct Queue *queue_head;
 
      struct Queue *queue_tail;
 
      struct AddressListEntry *addr_head;
 
      struct AddressListEntry *addr_tail;
 
      unsigned int total_queue_length;
 
      struct GNUNET_SCHEDULER_Task *free_queue_entry_task;
 
      enum GNUNET_TRANSPORT_CommunicatorCharacteristics cc;
 
      enum GNUNET_GenericReturnValue can_burst;
 
    } communicator;
 
    struct
    {
      struct GNUNET_CONTAINER_MultiPeerMap *requests;
    } application;
  } details;
};
 
 
struct ValidationState
{
  struct GNUNET_PeerIdentity pid;
 
  struct GNUNET_TIME_Absolute valid_until;
 
  struct GNUNET_TIME_Absolute validated_until;
 
  struct GNUNET_TIME_Absolute first_challenge_use;
 
  struct GNUNET_TIME_Absolute last_challenge_use;
 
  struct GNUNET_TIME_Absolute next_challenge;
 
  struct GNUNET_TIME_Relative challenge_backoff;
 
  struct GNUNET_TIME_Relative validation_rtt;
 
  struct GNUNET_CRYPTO_ChallengeNonceP challenge;
 
  struct GNUNET_HashCode hc;
 
  struct GNUNET_SCHEDULER_Task *revalidation_task;
 
  char *address;
 
  struct GNUNET_CONTAINER_HeapNode *hn;
 
  struct GNUNET_PEERSTORE_StoreContext *sc;
 
  uint32_t last_window_consum_limit;
 
  int awaiting_queue;
};
 
 
struct Backtalker
{
  struct GNUNET_PeerIdentity pid;
 
  struct GNUNET_TIME_Absolute monotonic_time;
 
  struct GNUNET_TIME_Absolute timeout;
 
  struct GNUNET_CRYPTO_HpkeEncapsulation last_ephemeral;
 
  struct GNUNET_SCHEDULER_Task *task;
 
  struct CommunicatorMessageContext *cmc;
 
  struct GNUNET_PEERSTORE_IterateContext *get;
 
  struct GNUNET_PEERSTORE_StoreContext *sc;
 
  size_t body_size;
};
 
static struct RingBufferEntry *ring_buffer[RING_BUFFER_SIZE];
 
static unsigned int ring_buffer_head;
 
static unsigned int is_ring_buffer_full;
 
static struct PendingMessage *ring_buffer_dv[RING_BUFFER_SIZE];
 
static unsigned int ring_buffer_dv_head;
 
static unsigned int is_ring_buffer_dv_full;
 
static struct TransportClient *clients_head;
 
static struct TransportClient *clients_tail;
 
static struct GNUNET_STATISTICS_Handle *GST_stats;
 
static const struct GNUNET_CONFIGURATION_Handle *GST_cfg;
 
static struct GNUNET_PeerIdentity *GST_my_identity;
 
struct GNUNET_HELLO_Builder *GST_my_hello;
 
static struct GNUNET_CONTAINER_MultiPeerMap *neighbours;
 
static struct GNUNET_CONTAINER_MultiPeerMap *backtalkers;
 
static struct GNUNET_CONTAINER_MultiPeerMap *ack_cummulators;
 
static struct GNUNET_CONTAINER_MultiUuidmap *pending_acks;
 
static struct GNUNET_CONTAINER_MultiPeerMap *dv_routes;
 
static struct GNUNET_CONTAINER_MultiPeerMap *validation_map;
 
static struct GNUNET_CONTAINER_MultiHashMap *revalidation_map;
 
static struct GNUNET_CONTAINER_MultiPeerMap *links;
 
static struct GNUNET_CONTAINER_MultiShortmap *dvlearn_map;
 
static struct LearnLaunchEntry *lle_head = NULL;
 
static struct LearnLaunchEntry *lle_tail = NULL;
 
static struct GNUNET_CONTAINER_Heap *validation_heap;
 
struct GNUNET_NAT_Handle *nh;
 
static struct GNUNET_PEERSTORE_Handle *peerstore;
 
static struct GNUNET_PILS_Handle *pils;
 
struct PilsRequest
{
  struct PilsRequest *prev;
 
  struct PilsRequest *next;
 
  struct GNUNET_PILS_Operation *op;
};
 
static struct PilsRequest *pils_requests_head;
 
static struct PilsRequest *pils_requests_tail;
 
static struct GNUNET_SCHEDULER_Task *dvlearn_task;
 
static struct GNUNET_SCHEDULER_Task *validation_task;
 
static struct GNUNET_SCHEDULER_Task *pils_feed_task;
 
static struct IncomingRequest *ir_head;
 
static struct IncomingRequest *ir_tail;
 
static unsigned int ir_total;
 
static unsigned long long logging_uuid_gen;
 
static enum GNUNET_GenericReturnValue burst_running;
 
static struct GNUNET_TIME_Absolute hello_mono_time;
 
static int in_shutdown;
 
static struct GNUNET_SCHEDULER_Task *burst_task;
 
struct GNUNET_SCHEDULER_Task *burst_timeout_task;
 
enum GNUNET_GenericReturnValue use_burst;
 
static unsigned int
get_age ()
{
  struct GNUNET_TIME_Absolute now;
 
  now = GNUNET_TIME_absolute_get ();
  return now.abs_value_us / GNUNET_TIME_UNIT_MINUTES.rel_value_us / 15;
}
 
 
static void
free_incoming_request (struct IncomingRequest *ir)
{
  GNUNET_CONTAINER_DLL_remove (ir_head, ir_tail, ir);
  GNUNET_assert (ir_total > 0);
  ir_total--;
  if (NULL != ir->nc)
    GNUNET_PEERSTORE_monitor_stop (ir->nc);
  ir->nc = NULL;
  GNUNET_free (ir);
}
 
 
static void
free_pending_acknowledgement (struct PendingAcknowledgement *pa)
{
  struct Queue *q = pa->queue;
  struct PendingMessage *pm = pa->pm;
  struct DistanceVectorHop *dvh = pa->dvh;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "free_pending_acknowledgement\n");
  if (NULL != q)
  {
    GNUNET_CONTAINER_MDLL_remove (queue, q->pa_head, q->pa_tail, pa);
    pa->queue = NULL;
  }
  if (NULL != pm)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "remove pa from message\n");
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "remove pa from message %" PRIu64 "\n",
                pm->logging_uuid);
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "remove pa from message %u\n",
                pm->pmt);
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "remove pa from message %s\n",
                GNUNET_uuid2s (&pa->ack_uuid.value));
    GNUNET_CONTAINER_MDLL_remove (pm, pm->pa_head, pm->pa_tail, pa);
    pa->pm = NULL;
  }
  if (NULL != dvh)
  {
    GNUNET_CONTAINER_MDLL_remove (dvh, dvh->pa_head, dvh->pa_tail, pa);
    pa->queue = NULL;
  }
  GNUNET_assert (GNUNET_YES ==
                 GNUNET_CONTAINER_multiuuidmap_remove (pending_acks,
                                                       &pa->ack_uuid.value,
                                                       pa));
  GNUNET_free (pa);
}
 
 
static void
free_fragment_tree (struct PendingMessage *root)
{
  struct PendingMessage *frag;
 
  while (NULL != (frag = root->head_frag))
  {
    struct PendingAcknowledgement *pa;
 
    free_fragment_tree (frag);
    while (NULL != (pa = frag->pa_head))
    {
      GNUNET_CONTAINER_MDLL_remove (pm, frag->pa_head, frag->pa_tail, pa);
      pa->pm = NULL;
    }
    GNUNET_CONTAINER_MDLL_remove (frag, root->head_frag, root->tail_frag, frag);
    if (NULL != frag->qe)
    {
      GNUNET_assert (frag == frag->qe->pm);
      frag->qe->pm = NULL;
    }
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Free frag %p\n",
                frag);
    GNUNET_free (frag);
  }
}
 
 
static void
free_pending_message (struct PendingMessage *pm)
{
  struct TransportClient *tc = pm->client;
  struct VirtualLink *vl = pm->vl;
  struct PendingAcknowledgement *pa;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Freeing pm %p\n",
              pm);
  if (NULL != tc)
  {
    GNUNET_CONTAINER_MDLL_remove (client,
                                  tc->details.core.pending_msg_head,
                                  tc->details.core.pending_msg_tail,
                                  pm);
  }
  if ((NULL != vl) && (NULL == pm->frag_parent))
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Removing pm %" PRIu64 "\n",
                pm->logging_uuid);
    GNUNET_CONTAINER_MDLL_remove (vl,
                                  vl->pending_msg_head,
                                  vl->pending_msg_tail,
                                  pm);
  }
  else if (NULL != pm->frag_parent && PMT_DV_BOX != pm->pmt)
  {
    struct PendingMessage *root = pm->frag_parent;
 
    while (NULL != root->frag_parent && PMT_DV_BOX != root->pmt)
      root = root->frag_parent;
 
    root->frag_count--;
  }
  while (NULL != (pa = pm->pa_head))
  {
    if (NULL == pa)
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "free pending pa  null\n");
    if (NULL == pm->pa_tail)
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "free pending pa_tail null\n");
    if (NULL == pa->prev_pa)
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "free pending pa prev null\n");
    if (NULL == pa->next_pa)
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "free pending pa next null\n");
    GNUNET_CONTAINER_MDLL_remove (pm, pm->pa_head, pm->pa_tail, pa);
    pa->pm = NULL;
  }
 
  free_fragment_tree (pm);
  if (NULL != pm->qe)
  {
    GNUNET_assert (pm == pm->qe->pm);
    pm->qe->pm = NULL;
  }
  if (NULL != pm->bpm)
  {
    free_fragment_tree (pm->bpm);
    if (NULL != pm->bpm->qe)
    {
      struct QueueEntry *qe = pm->bpm->qe;
 
      qe->pm = NULL;
    }
    GNUNET_free (pm->bpm);
  }
 
  GNUNET_free (pm);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Freeing pm done\n");
}
 
 
static void
free_reassembly_context (struct ReassemblyContext *rc)
{
  struct VirtualLink *vl = rc->virtual_link;
 
  GNUNET_assert (rc == GNUNET_CONTAINER_heap_remove_node (rc->hn));
  GNUNET_assert (GNUNET_OK ==
                 GNUNET_CONTAINER_multihashmap32_remove (vl->reassembly_map,
                                                         rc->msg_uuid.uuid,
                                                         rc));
  GNUNET_free (rc);
}
 
 
static void
reassembly_cleanup_task (void *cls)
{
  struct VirtualLink *vl = cls;
  struct ReassemblyContext *rc;
 
  vl->reassembly_timeout_task = NULL;
  while (NULL != (rc = GNUNET_CONTAINER_heap_peek (vl->reassembly_heap)))
  {
    if (0 == GNUNET_TIME_absolute_get_remaining (rc->reassembly_timeout)
        .rel_value_us)
    {
      free_reassembly_context (rc);
      continue;
    }
    GNUNET_assert (NULL == vl->reassembly_timeout_task);
    vl->reassembly_timeout_task =
      GNUNET_SCHEDULER_add_at (rc->reassembly_timeout,
                               &reassembly_cleanup_task,
                               vl);
    return;
  }
}
 
 
static int
free_reassembly_cb (void *cls, uint32_t key, void *value)
{
  struct ReassemblyContext *rc = value;
 
  (void) cls;
  (void) key;
  free_reassembly_context (rc);
  return GNUNET_OK;
}
 
 
static void
free_virtual_link (struct VirtualLink *vl)
{
  struct PendingMessage *pm;
  struct CoreSentContext *csc;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "free virtual link %p\n",
              vl);
 
  if (NULL != vl->reassembly_map)
  {
    GNUNET_CONTAINER_multihashmap32_iterate (vl->reassembly_map,
                                             &free_reassembly_cb,
                                             NULL);
    GNUNET_CONTAINER_multihashmap32_destroy (vl->reassembly_map);
    vl->reassembly_map = NULL;
    GNUNET_CONTAINER_heap_destroy (vl->reassembly_heap);
    vl->reassembly_heap = NULL;
  }
  if (NULL != vl->reassembly_timeout_task)
  {
    GNUNET_SCHEDULER_cancel (vl->reassembly_timeout_task);
    vl->reassembly_timeout_task = NULL;
  }
  while (NULL != (pm = vl->pending_msg_head))
    free_pending_message (pm);
  GNUNET_assert (GNUNET_YES ==
                 GNUNET_CONTAINER_multipeermap_remove (links, &vl->target, vl));
  if (NULL != vl->visibility_task)
  {
    GNUNET_SCHEDULER_cancel (vl->visibility_task);
    vl->visibility_task = NULL;
  }
  if (NULL != vl->fc_retransmit_task)
  {
    GNUNET_SCHEDULER_cancel (vl->fc_retransmit_task);
    vl->fc_retransmit_task = NULL;
  }
  while (NULL != (csc = vl->csc_head))
  {
    GNUNET_CONTAINER_DLL_remove (vl->csc_head, vl->csc_tail, csc);
    GNUNET_assert (vl == csc->vl);
    csc->vl = NULL;
  }
  GNUNET_break (NULL == vl->n);
  GNUNET_break (NULL == vl->dv);
  GNUNET_free (vl);
}
 
 
static void
free_validation_state (struct ValidationState *vs)
{
  if (NULL != vs->revalidation_task)
  {
    GNUNET_SCHEDULER_cancel (vs->revalidation_task);
    vs->revalidation_task = NULL;
  }
  /*memcpy (&hkey,
          &hc,
          sizeof (hkey));*/
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Remove key %s for address %s map size %u contains %u during freeing state\n",
              GNUNET_h2s (&vs->hc),
              vs->address,
              GNUNET_CONTAINER_multihashmap_size (revalidation_map),
              GNUNET_CONTAINER_multihashmap_contains (revalidation_map,
                                                      &vs->hc));
  GNUNET_CONTAINER_multihashmap_remove (revalidation_map, &vs->hc, vs);
  GNUNET_assert (
    GNUNET_YES ==
    GNUNET_CONTAINER_multipeermap_remove (validation_map, &vs->pid, vs));
  GNUNET_CONTAINER_heap_remove_node (vs->hn);
  vs->hn = NULL;
  if (NULL != vs->sc)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "store cancel\n");
    GNUNET_PEERSTORE_store_cancel (vs->sc);
    vs->sc = NULL;
  }
  GNUNET_free (vs->address);
  GNUNET_free (vs);
}
 
 
static struct Neighbour *
lookup_neighbour (const struct GNUNET_PeerIdentity *pid)
{
  return GNUNET_CONTAINER_multipeermap_get (neighbours, pid);
}
 
 
static struct VirtualLink *
lookup_virtual_link (const struct GNUNET_PeerIdentity *pid)
{
  return GNUNET_CONTAINER_multipeermap_get (links, pid);
}
 
 
struct MonitorEvent
{
  struct GNUNET_TIME_Absolute last_validation;
  struct GNUNET_TIME_Absolute valid_until;
  struct GNUNET_TIME_Absolute next_validation;
 
  struct GNUNET_TIME_Relative rtt;
 
  enum GNUNET_TRANSPORT_ConnectionStatus cs;
 
  uint32_t num_msg_pending;
 
  uint32_t num_bytes_pending;
};
 
 
static void
free_distance_vector_hop (struct DistanceVectorHop *dvh)
{
  struct Neighbour *n = dvh->next_hop;
  struct DistanceVector *dv = dvh->dv;
  struct PendingAcknowledgement *pa;
 
  while (NULL != (pa = dvh->pa_head))
  {
    GNUNET_CONTAINER_MDLL_remove (dvh, dvh->pa_head, dvh->pa_tail, pa);
    pa->dvh = NULL;
  }
  GNUNET_CONTAINER_MDLL_remove (neighbour, n->dv_head, n->dv_tail, dvh);
  GNUNET_CONTAINER_MDLL_remove (dv, dv->dv_head, dv->dv_tail, dvh);
  GNUNET_free (dvh);
}
 
 
static void
check_link_down (void *cls);
 
 
static void
cores_send_disconnect_info (const struct GNUNET_PeerIdentity *pid)
{
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Informing CORE clients about disconnect from %s\n",
              GNUNET_i2s (pid));
  for (struct TransportClient *tc = clients_head; NULL != tc; tc = tc->next)
  {
    struct GNUNET_MQ_Envelope *env;
    struct DisconnectInfoMessage *dim;
 
    if (CT_CORE != tc->type)
      continue;
    env = GNUNET_MQ_msg (dim, GNUNET_MESSAGE_TYPE_TRANSPORT_DISCONNECT);
    dim->peer = *pid;
    GNUNET_MQ_send (tc->mq, env);
  }
}
 
 
static void
free_dv_route (struct DistanceVector *dv)
{
  struct DistanceVectorHop *dvh;
  struct VirtualLink *vl;
 
  while (NULL != (dvh = dv->dv_head))
    free_distance_vector_hop (dvh);
 
  GNUNET_assert (
    GNUNET_YES ==
    GNUNET_CONTAINER_multipeermap_remove (dv_routes, &dv->target, dv));
  if (NULL != (vl = dv->vl))
  {
    GNUNET_assert (dv == vl->dv);
    vl->dv = NULL;
    if (NULL == vl->n)
    {
      cores_send_disconnect_info (&dv->target);
      free_virtual_link (vl);
    }
    else
    {
      GNUNET_SCHEDULER_cancel (vl->visibility_task);
      vl->visibility_task = GNUNET_SCHEDULER_add_now (&check_link_down, vl);
    }
    dv->vl = NULL;
  }
 
  if (NULL != dv->timeout_task)
  {
    GNUNET_SCHEDULER_cancel (dv->timeout_task);
    dv->timeout_task = NULL;
  }
  GNUNET_free (dv->km);
  GNUNET_free (dv);
}
 
 
static void
notify_monitor (struct TransportClient *tc,
                const struct GNUNET_PeerIdentity *peer,
                const char *address,
                enum GNUNET_NetworkType nt,
                const struct MonitorEvent *me)
{
  struct GNUNET_MQ_Envelope *env;
  struct GNUNET_TRANSPORT_MonitorData *md;
  size_t addr_len = strlen (address) + 1;
 
  env = GNUNET_MQ_msg_extra (md,
                             addr_len,
                             GNUNET_MESSAGE_TYPE_TRANSPORT_MONITOR_DATA);
  md->nt = htonl ((uint32_t) nt);
  md->peer = *peer;
  md->last_validation = GNUNET_TIME_absolute_hton (me->last_validation);
  md->valid_until = GNUNET_TIME_absolute_hton (me->valid_until);
  md->next_validation = GNUNET_TIME_absolute_hton (me->next_validation);
  md->rtt = GNUNET_TIME_relative_hton (me->rtt);
  md->cs = htonl ((uint32_t) me->cs);
  md->num_msg_pending = htonl (me->num_msg_pending);
  md->num_bytes_pending = htonl (me->num_bytes_pending);
  memcpy (&md[1], address, addr_len);
  GNUNET_MQ_send (tc->mq, env);
}
 
 
static void
notify_monitors (const struct GNUNET_PeerIdentity *peer,
                 const char *address,
                 enum GNUNET_NetworkType nt,
                 const struct MonitorEvent *me)
{
  for (struct TransportClient *tc = clients_head; NULL != tc; tc = tc->next)
  {
    if (CT_MONITOR != tc->type)
      continue;
    if (tc->details.monitor.one_shot)
      continue;
    if ((GNUNET_NO == GNUNET_is_zero (&tc->details.monitor.peer)) &&
        (0 != GNUNET_memcmp (&tc->details.monitor.peer, peer)))
      continue;
    notify_monitor (tc, peer, address, nt, me);
  }
}
 
 
static void *
client_connect_cb (void *cls,
                   struct GNUNET_SERVICE_Client *client,
                   struct GNUNET_MQ_Handle *mq)
{
  struct TransportClient *tc;
 
  (void) cls;
  tc = GNUNET_new (struct TransportClient);
  tc->client = client;
  tc->mq = mq;
  GNUNET_CONTAINER_DLL_insert (clients_head, clients_tail, tc);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Client %p of type %u connected\n",
              tc,
              tc->type);
  return tc;
}
 
 
static enum GNUNET_GenericReturnValue
remove_global_addresses (void *cls,
                         const struct GNUNET_PeerIdentity *pid,
                         void *value)
{
  struct TransportGlobalNattedAddress *tgna = value;
  (void) cls;
 
  GNUNET_free (tgna);
 
  return GNUNET_OK;
}
 
 
static void
free_neighbour (struct Neighbour *neighbour,
                enum GNUNET_GenericReturnValue drop_link)
{
  struct DistanceVectorHop *dvh;
  struct VirtualLink *vl;
 
  GNUNET_assert (NULL == neighbour->queue_head);
  GNUNET_assert (GNUNET_YES ==
                 GNUNET_CONTAINER_multipeermap_remove (neighbours,
                                                       &neighbour->pid,
                                                       neighbour));
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Freeing neighbour\n");
  GNUNET_CONTAINER_multipeermap_iterate (neighbour->natted_addresses,
                                         &remove_global_addresses,
                                         NULL);
  GNUNET_CONTAINER_multipeermap_destroy (neighbour->natted_addresses);
  while (NULL != (dvh = neighbour->dv_head))
  {
    struct DistanceVector *dv = dvh->dv;
 
    free_distance_vector_hop (dvh);
    if (NULL == dv->dv_head)
      free_dv_route (dv);
  }
  if (NULL != neighbour->get)
  {
    GNUNET_PEERSTORE_iteration_stop (neighbour->get);
    neighbour->get = NULL;
  }
  if (NULL != neighbour->sc)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "store cancel\n");
    GNUNET_PEERSTORE_store_cancel (neighbour->sc);
    neighbour->sc = NULL;
  }
  if (NULL != (vl = neighbour->vl))
  {
    GNUNET_assert (neighbour == vl->n);
    vl->n = NULL;
    if ((GNUNET_YES == drop_link) || (NULL == vl->dv))
    {
      cores_send_disconnect_info (&vl->target);
      free_virtual_link (vl);
    }
    else
    {
      GNUNET_SCHEDULER_cancel (vl->visibility_task);
      vl->visibility_task = GNUNET_SCHEDULER_add_now (&check_link_down, vl);
    }
    neighbour->vl = NULL;
  }
  GNUNET_free (neighbour);
}
 
 
static void
core_send_connect_info (struct TransportClient *tc,
                        const struct GNUNET_PeerIdentity *pid)
{
  struct GNUNET_MQ_Envelope *env;
  struct ConnectInfoMessage *cim;
 
  GNUNET_assert (CT_CORE == tc->type);
  env = GNUNET_MQ_msg (cim, GNUNET_MESSAGE_TYPE_TRANSPORT_CONNECT);
  cim->id = *pid;
  GNUNET_MQ_send (tc->mq, env);
}
 
 
static void
cores_send_connect_info (const struct GNUNET_PeerIdentity *pid)
{
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Informing CORE clients about connection to %s\n",
              GNUNET_i2s (pid));
  for (struct TransportClient *tc = clients_head; NULL != tc; tc = tc->next)
  {
    if (CT_CORE != tc->type)
      continue;
    core_send_connect_info (tc, pid);
  }
}
 
 
static void
transmit_on_queue (void *cls);
 
 
static unsigned int
check_for_queue_with_higher_prio (struct Queue *queue, struct Queue *queue_head)
{
  for (struct Queue *s = queue_head; NULL != s;
       s = s->next_client)
  {
    if (s->tc->details.communicator.address_prefix !=
        queue->tc->details.communicator.address_prefix)
    {
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "queue address %s qid %u compare with queue: address %s qid %u\n",
                  queue->address,
                  queue->qid,
                  s->address,
                  s->qid);
      if ((s->priority > queue->priority) && (0 < s->q_capacity) &&
          (QUEUE_LENGTH_LIMIT > s->queue_length) )
        return GNUNET_YES;
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "Lower prio\n");
    }
  }
  return GNUNET_NO;
}
 
 
static void
schedule_transmit_on_queue (struct GNUNET_TIME_Relative delay,
                            struct Queue *queue,
                            enum GNUNET_SCHEDULER_Priority p)
{
  struct GNUNET_TIME_Absolute now = GNUNET_TIME_absolute_get ();
 
  if (queue->validated_until.abs_value_us < now.abs_value_us)
    return;
  if (check_for_queue_with_higher_prio (queue,
                                        queue->tc->details.communicator.
                                        queue_head))
    return;
 
  if (queue->tc->details.communicator.total_queue_length >=
      COMMUNICATOR_TOTAL_QUEUE_LIMIT)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Transmission on queue %s (QID %u) throttled due to communicator queue limit\n",
                queue->address,
                queue->qid);
    GNUNET_STATISTICS_update (
      GST_stats,
      "# Transmission throttled due to communicator queue limit",
      1,
      GNUNET_NO);
    queue->idle = GNUNET_NO;
    return;
  }
  if (queue->queue_length >= QUEUE_LENGTH_LIMIT)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Transmission on queue %s (QID %u) throttled due to communicator queue length limit\n",
                queue->address,
                queue->qid);
    GNUNET_STATISTICS_update (GST_stats,
                              "# Transmission throttled due to queue queue limit",
                              1,
                              GNUNET_NO);
    queue->idle = GNUNET_NO;
    return;
  }
  if (0 == queue->q_capacity)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Transmission on queue %s (QID %u) throttled due to communicator message  has capacity %"
                PRIu64 ".\n",
                queue->address,
                queue->qid,
                queue->q_capacity);
    GNUNET_STATISTICS_update (GST_stats,
                              "# Transmission throttled due to message queue capacity",
                              1,
                              GNUNET_NO);
    queue->idle = GNUNET_NO;
    return;
  }
  /* queue might indeed be ready, schedule it */
  if (NULL != queue->transmit_task)
    GNUNET_SCHEDULER_cancel (queue->transmit_task);
  queue->transmit_task =
    GNUNET_SCHEDULER_add_delayed_with_priority (delay, p, &transmit_on_queue,
                                                queue);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Considering transmission on queue `%s' QID %llu to %s\n",
              queue->address,
              (unsigned long long) queue->qid,
              GNUNET_i2s (&queue->neighbour->pid));
}
 
 
static void
check_link_down (void *cls)
{
  struct VirtualLink *vl = cls;
  struct DistanceVector *dv = vl->dv;
  struct Neighbour *n = vl->n;
  struct GNUNET_TIME_Absolute dvh_timeout;
  struct GNUNET_TIME_Absolute q_timeout;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Checking if link is down\n");
  vl->visibility_task = NULL;
  dvh_timeout = GNUNET_TIME_UNIT_ZERO_ABS;
  if (NULL != dv)
  {
    for (struct DistanceVectorHop *pos = dv->dv_head; NULL != pos;
         pos = pos->next_dv)
      dvh_timeout = GNUNET_TIME_absolute_max (dvh_timeout,
                                              pos->path_valid_until);
    if (0 == GNUNET_TIME_absolute_get_remaining (dvh_timeout).rel_value_us)
    {
      vl->dv->vl = NULL;
      vl->dv = NULL;
    }
  }
  q_timeout = GNUNET_TIME_UNIT_ZERO_ABS;
  for (struct Queue *q = n->queue_head; NULL != q; q = q->next_neighbour)
    q_timeout = GNUNET_TIME_absolute_max (q_timeout, q->validated_until);
  if (0 == GNUNET_TIME_absolute_get_remaining (q_timeout).rel_value_us)
  {
    vl->n->vl = NULL;
    vl->n = NULL;
  }
  if ((NULL == vl->n) && (NULL == vl->dv))
  {
    cores_send_disconnect_info (&vl->target);
    free_virtual_link (vl);
    return;
  }
  vl->visibility_task =
    GNUNET_SCHEDULER_add_at (GNUNET_TIME_absolute_max (q_timeout, dvh_timeout),
                             &check_link_down,
                             vl);
}
 
 
static void
free_queue (struct Queue *queue)
{
  struct Neighbour *neighbour = queue->neighbour;
  struct TransportClient *tc = queue->tc;
  struct MonitorEvent me = { .cs = GNUNET_TRANSPORT_CS_DOWN,
                             .rtt = GNUNET_TIME_UNIT_FOREVER_REL };
  struct QueueEntry *qe;
  int maxxed;
  struct PendingAcknowledgement *pa;
  struct VirtualLink *vl;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Cleaning up queue %u\n", queue->qid);
  if (NULL != queue->mo)
  {
    GNUNET_PEERSTORE_monitor_stop (queue->mo);
    queue->mo = NULL;
  }
  if (NULL != queue->transmit_task)
  {
    GNUNET_SCHEDULER_cancel (queue->transmit_task);
    queue->transmit_task = NULL;
  }
  while (NULL != (pa = queue->pa_head))
  {
    GNUNET_CONTAINER_MDLL_remove (queue, queue->pa_head, queue->pa_tail, pa);
    pa->queue = NULL;
  }
 
  GNUNET_CONTAINER_MDLL_remove (neighbour,
                                neighbour->queue_head,
                                neighbour->queue_tail,
                                queue);
  GNUNET_CONTAINER_MDLL_remove (client,
                                tc->details.communicator.queue_head,
                                tc->details.communicator.queue_tail,
                                queue);
  maxxed = (COMMUNICATOR_TOTAL_QUEUE_LIMIT <=
            tc->details.communicator.total_queue_length);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Cleaning up queue with length %u\n",
              queue->queue_length);
  while (NULL != (qe = queue->queue_head))
  {
    GNUNET_CONTAINER_DLL_remove (queue->queue_head, queue->queue_tail, qe);
    queue->queue_length--;
    tc->details.communicator.total_queue_length--;
    if (NULL != qe->pm)
    {
      GNUNET_assert (qe == qe->pm->qe);
      qe->pm->qe = NULL;
    }
    GNUNET_free (qe);
  }
  GNUNET_assert (0 == queue->queue_length);
  if ((maxxed) && (COMMUNICATOR_TOTAL_QUEUE_LIMIT >
                   tc->details.communicator.total_queue_length))
  {
    /* Communicator dropped below threshold, resume all _other_ queues */
    GNUNET_STATISTICS_update (
      GST_stats,
      "# Transmission throttled due to communicator queue limit",
      -1,
      GNUNET_NO);
    for (struct Queue *s = tc->details.communicator.queue_head; NULL != s;
         s = s->next_client)
      schedule_transmit_on_queue (GNUNET_TIME_UNIT_ZERO,
                                  s,
                                  GNUNET_SCHEDULER_PRIORITY_DEFAULT);
  }
  notify_monitors (&neighbour->pid, queue->address, queue->nt, &me);
  GNUNET_free (queue);
 
  vl = lookup_virtual_link (&neighbour->pid);
  if ((NULL != vl) && (neighbour == vl->n))
  {
    GNUNET_SCHEDULER_cancel (vl->visibility_task);
    check_link_down (vl);
  }
  if (NULL == neighbour->queue_head)
  {
    free_neighbour (neighbour, GNUNET_NO);
  }
}
 
 
static void
free_address_list_entry (struct AddressListEntry *ale)
{
  struct TransportClient *tc = ale->tc;
 
  GNUNET_CONTAINER_DLL_remove (tc->details.communicator.addr_head,
                               tc->details.communicator.addr_tail,
                               ale);
  if (NULL != ale->sc)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "store cancel\n");
    GNUNET_PEERSTORE_store_cancel (ale->sc);
    ale->sc = NULL;
  }
  if (NULL != ale->st)
  {
    GNUNET_SCHEDULER_cancel (ale->st);
    ale->st = NULL;
  }
  if (NULL != ale->signed_address)
    GNUNET_free (ale->signed_address);
  GNUNET_free (ale);
}
 
 
static int
stop_peer_request (void *cls,
                   const struct GNUNET_PeerIdentity *pid,
                   void *value)
{
  struct TransportClient *tc = cls;
  struct PeerRequest *pr = value;
 
  if (NULL != pr->nc)
    GNUNET_PEERSTORE_monitor_stop (pr->nc);
  pr->nc = NULL;
  GNUNET_assert (
    GNUNET_YES ==
    GNUNET_CONTAINER_multipeermap_remove (tc->details.application.requests,
                                          pid,
                                          pr));
  GNUNET_free (pr);
 
  return GNUNET_OK;
}
 
 
static void
do_shutdown (void *cls);
 
static void
client_disconnect_cb (void *cls,
                      struct GNUNET_SERVICE_Client *client,
                      void *app_ctx)
{
  struct TransportClient *tc = app_ctx;
 
  (void) cls;
  (void) client;
  GNUNET_CONTAINER_DLL_remove (clients_head, clients_tail, tc);
  switch (tc->type)
  {
  case CT_NONE:
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Unknown Client %p disconnected, cleaning up.\n",
                tc);
    break;
 
  case CT_CORE: {
      struct PendingMessage *pm;
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "CORE Client %p disconnected, cleaning up.\n",
                  tc);
 
 
      while (NULL != (pm = tc->details.core.pending_msg_head))
      {
        GNUNET_CONTAINER_MDLL_remove (client,
                                      tc->details.core.pending_msg_head,
                                      tc->details.core.pending_msg_tail,
                                      pm);
        pm->client = NULL;
      }
    }
    break;
 
  case CT_MONITOR:
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "MONITOR Client %p disconnected, cleaning up.\n",
                tc);
 
    break;
 
  case CT_COMMUNICATOR: {
      struct Queue *q;
      struct AddressListEntry *ale;
 
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "COMMUNICATOR Client %p disconnected, cleaning up.\n",
                  tc);
 
      if (NULL != tc->details.communicator.free_queue_entry_task)
        GNUNET_SCHEDULER_cancel (
          tc->details.communicator.free_queue_entry_task);
      while (NULL != (q = tc->details.communicator.queue_head))
        free_queue (q);
      while (NULL != (ale = tc->details.communicator.addr_head))
        free_address_list_entry (ale);
      GNUNET_free (tc->details.communicator.address_prefix);
    }
    break;
 
  case CT_APPLICATION:
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "APPLICATION Client %p disconnected, cleaning up.\n",
                tc);
 
    GNUNET_CONTAINER_multipeermap_iterate (tc->details.application.requests,
                                           &stop_peer_request,
                                           tc);
    GNUNET_CONTAINER_multipeermap_destroy (tc->details.application.requests);
    break;
  }
  GNUNET_free (tc);
  if ((GNUNET_YES == in_shutdown) && (NULL == clients_head))
  {
    GNUNET_log (GNUNET_ERROR_TYPE_ERROR,
                "Our last client disconnected\n");
    do_shutdown (cls);
  }
}
 
 
static int
notify_client_connect_info (void *cls,
                            const struct GNUNET_PeerIdentity *pid,
                            void *value)
{
  struct TransportClient *tc = cls;
  struct VirtualLink *vl = value;
 
  if ((NULL == vl) || (GNUNET_NO == vl->confirmed))
    return GNUNET_OK;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Telling new CORE client about existing connection to %s\n",
              GNUNET_i2s (pid));
  core_send_connect_info (tc, pid);
  return GNUNET_OK;
}
 
 
static void
finish_cmc_handling_with_continue (struct CommunicatorMessageContext *cmc,
                                   unsigned
                                   int free_cmc);
 
static enum GNUNET_GenericReturnValue
resume_communicators (void *cls,
                      const struct GNUNET_PeerIdentity *pid,
                      void *value)
{
  struct VirtualLink *vl = value;
  struct CommunicatorMessageContext *cmc;
 
  /* resume communicators */
  while (NULL != (cmc = vl->cmc_tail))
  {
    GNUNET_CONTAINER_DLL_remove (vl->cmc_head, vl->cmc_tail, cmc);
    if (GNUNET_NO == cmc->continue_send)
      finish_cmc_handling_with_continue (cmc, GNUNET_YES);
  }
  return GNUNET_OK;
}
 
 
static void
handle_client_start (void *cls, const struct StartMessage *start)
{
  struct TransportClient *tc = cls;
  // uint32_t options;
  //
  // FIXME ignore the check of the peer ids for now.
  //       (also deprecate the old way of obtaining our own peer ID)
  // options = ntohl (start->options);
  // if ((0 != (1 & options)) &&
  //    (0 != GNUNET_memcmp (&start->self, &GST_my_identity)))
  // {
  //  /* client thinks this is a different peer, reject */
  //  GNUNET_break (0);
  //  GNUNET_SERVICE_client_drop (tc->client);
  //  return;
  // }
  if (CT_NONE != tc->type)
  {
    GNUNET_break (0);
    GNUNET_SERVICE_client_drop (tc->client);
    return;
  }
  tc->type = CT_CORE;
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "New CORE client with PID %s registered\n",
              GNUNET_i2s (&start->self));
  GNUNET_CONTAINER_multipeermap_iterate (links,
                                         &notify_client_connect_info,
                                         tc);
  GNUNET_CONTAINER_multipeermap_iterate (links,
                                         &resume_communicators,
                                         NULL);
  GNUNET_SERVICE_client_continue (tc->client);
}
 
 
static int
check_client_send (void *cls, const struct OutboundMessage *obm)
{
  struct TransportClient *tc = cls;
  uint16_t size;
  const struct GNUNET_MessageHeader *obmm;
 
  if (CT_CORE != tc->type)
  {
    GNUNET_break (0);
    return GNUNET_SYSERR;
  }
  size = ntohs (obm->header.size) - sizeof(struct OutboundMessage);
  if (size < sizeof(struct GNUNET_MessageHeader))
  {
    GNUNET_break (0);
    return GNUNET_SYSERR;
  }
  obmm = (const struct GNUNET_MessageHeader *) &obm[1];
  if (size != ntohs (obmm->size))
  {
    GNUNET_break (0);
    return GNUNET_SYSERR;
  }
  return GNUNET_OK;
}
 
 
static void
client_send_response (struct PendingMessage *pm)
{
  struct TransportClient *tc = pm->client;
  struct VirtualLink *vl = pm->vl;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "client send response\n");
  if (NULL != tc)
  {
    struct GNUNET_MQ_Envelope *env;
    struct SendOkMessage *so_msg;
 
    env = GNUNET_MQ_msg (so_msg, GNUNET_MESSAGE_TYPE_TRANSPORT_SEND_OK);
    so_msg->peer = vl->target;
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Confirming transmission of <%" PRIu64 "> to %s\n",
                pm->logging_uuid,
                GNUNET_i2s (&vl->target));
    GNUNET_MQ_send (tc->mq, env);
  }
  free_pending_message (pm);
}
 
 
static unsigned int
pick_random_dv_hops (const struct DistanceVector *dv,
                     enum RouteMessageOptions options,
                     struct DistanceVectorHop **hops_array,
                     unsigned int hops_array_length)
{
  uint64_t choices[hops_array_length];
  uint64_t num_dv;
  unsigned int dv_count;
 
  /* Pick random vectors, but weighted by distance, giving more weight
     to shorter vectors */
  num_dv = 0;
  dv_count = 0;
  for (struct DistanceVectorHop *pos = dv->dv_head; NULL != pos;
       pos = pos->next_dv)
  {
    if ((0 == (options & RMO_UNCONFIRMED_ALLOWED)) &&
        (GNUNET_TIME_absolute_get_remaining (pos->path_valid_until)
         .rel_value_us == 0))
      continue;   /* pos unconfirmed and confirmed required */
    num_dv += MAX_DV_HOPS_ALLOWED - pos->distance;
    dv_count++;
  }
  if (0 == dv_count)
    return 0;
  if (dv_count <= hops_array_length)
  {
    dv_count = 0;
    for (struct DistanceVectorHop *pos = dv->dv_head; NULL != pos;
         pos = pos->next_dv)
      hops_array[dv_count++] = pos;
    return dv_count;
  }
  for (unsigned int i = 0; i < hops_array_length; i++)
  {
    int ok = GNUNET_NO;
    while (GNUNET_NO == ok)
    {
      choices[i] =
        GNUNET_CRYPTO_random_u64 (GNUNET_CRYPTO_QUALITY_WEAK, num_dv);
      ok = GNUNET_YES;
      for (unsigned int j = 0; j < i; j++)
        if (choices[i] == choices[j])
        {
          ok = GNUNET_NO;
          break;
        }
    }
  }
  dv_count = 0;
  num_dv = 0;
  for (struct DistanceVectorHop *pos = dv->dv_head; NULL != pos;
       pos = pos->next_dv)
  {
    uint32_t delta = MAX_DV_HOPS_ALLOWED - pos->distance;
 
    if ((0 == (options & RMO_UNCONFIRMED_ALLOWED)) &&
        (GNUNET_TIME_absolute_get_remaining (pos->path_valid_until)
         .rel_value_us == 0))
      continue;   /* pos unconfirmed and confirmed required */
    for (unsigned int i = 0; i < hops_array_length; i++)
      if ((num_dv <= choices[i]) && (num_dv + delta > choices[i]))
        hops_array[dv_count++] = pos;
    num_dv += delta;
  }
  return dv_count;
}
 
 
static int
check_communicator_available (
  void *cls,
  const struct GNUNET_TRANSPORT_CommunicatorAvailableMessage *cam)
{
  struct TransportClient *tc = cls;
  uint16_t size;
 
  if (CT_NONE != tc->type)
  {
    GNUNET_break (0);
    return GNUNET_SYSERR;
  }
  tc->type = CT_COMMUNICATOR;
  size = ntohs (cam->header.size) - sizeof(*cam);
  if (0 == size)
    return GNUNET_OK; /* receive-only communicator */
  GNUNET_MQ_check_zero_termination (cam);
  return GNUNET_OK;
}
 
 
static void
finish_cmc_handling_with_continue (struct CommunicatorMessageContext *cmc,
                                   unsigned
                                   int free_cmc)
{
  if (0 != ntohl (cmc->im.fc_on))
  {
    /* send ACK when done to communicator for flow control! */
    struct GNUNET_MQ_Envelope *env;
    struct GNUNET_TRANSPORT_IncomingMessageAck *ack;
 
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Acknowledge message with flow control id %" PRIu64 "\n",
                cmc->im.fc_id);
    env = GNUNET_MQ_msg (ack, GNUNET_MESSAGE_TYPE_TRANSPORT_INCOMING_MSG_ACK);
    ack->reserved = htonl (0);
    ack->fc_id = cmc->im.fc_id;
    ack->sender = cmc->im.neighbour_sender;
    GNUNET_MQ_send (cmc->tc->mq, env);
  }
 
  GNUNET_SERVICE_client_continue (cmc->tc->client);
 
  if (GNUNET_YES == free_cmc)
  {
    GNUNET_free (cmc);
  }
}
 
 
static void
finish_cmc_handling (struct CommunicatorMessageContext *cmc)
{
  finish_cmc_handling_with_continue (cmc, GNUNET_YES);
}
 
 
static void
handle_client_recv_ok (void *cls, const struct RecvOkMessage *rom)
{
  struct TransportClient *tc = cls;
  struct VirtualLink *vl;
  uint32_t delta;
  struct CommunicatorMessageContext *cmc;
 
  if (CT_CORE != tc->type)
  {
    GNUNET_break (0);
    GNUNET_SERVICE_client_drop (tc->client);
    return;
  }
  vl = lookup_virtual_link (&rom->peer);
  if ((NULL == vl) || (GNUNET_NO == vl->confirmed))
  {
    GNUNET_STATISTICS_update (GST_stats,
                              "# RECV_OK dropped: virtual link unknown",
                              1,
                              GNUNET_NO);
    GNUNET_SERVICE_client_continue (tc->client);
    return;
  }
  delta = ntohl (rom->increase_window_delta);
  vl->core_recv_window += delta;
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "CORE ack receiving message, increased CORE recv window to %d\n",
              vl->core_recv_window);
  GNUNET_SERVICE_client_continue (tc->client);
  if (vl->core_recv_window <= 0)
    return;
  /* resume communicators */
  while (NULL != (cmc = vl->cmc_tail))
  {
    GNUNET_CONTAINER_DLL_remove (vl->cmc_head, vl->cmc_tail, cmc);
    if (GNUNET_NO == cmc->continue_send)
      finish_cmc_handling_with_continue (cmc, GNUNET_YES);
  }
}
 
 
static void
handle_communicator_available (
  void *cls,
  const struct GNUNET_TRANSPORT_CommunicatorAvailableMessage *cam)
{
  struct TransportClient *tc = cls;
  uint16_t size;
 
  size = ntohs (cam->header.size) - sizeof(*cam);
  if (0 == size)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Receive-only communicator connected\n");
    return;   /* receive-only communicator */
  }
  tc->details.communicator.address_prefix =
    GNUNET_strdup ((const char *) &cam[1]);
  tc->details.communicator.cc = ntohl (cam->cc);
  tc->details.communicator.can_burst = ntohl (cam->can_burst);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Communicator for peer %s with prefix '%s' connected %s\n",
              GNUNET_i2s (GST_my_identity),
              tc->details.communicator.address_prefix,
              tc->details.communicator.can_burst ? "can burst" :
              "can not burst");
  GNUNET_SERVICE_client_continue (tc->client);
}
 
 
static int
check_communicator_backchannel (
  void *cls,
  const struct GNUNET_TRANSPORT_CommunicatorBackchannel *cb)
{
  const struct GNUNET_MessageHeader *inbox;
  const char *is;
  uint16_t msize;
  uint16_t isize;
 
  (void) cls;
  msize = ntohs (cb->header.size) - sizeof(*cb);
  inbox = (const struct GNUNET_MessageHeader *) &cb[1];
  isize = ntohs (inbox->size);
  if (isize >= msize)
  {
    GNUNET_break (0);
    return GNUNET_SYSERR;
  }
  is = (const char *) inbox;
  is += isize;
  msize -= isize;
  GNUNET_assert (0 < msize);
  if ('\0' != is[msize - 1])
  {
    GNUNET_break (0);
    return GNUNET_SYSERR;
  }
  return GNUNET_OK;
}
 
 
struct SignDvCls
{
  struct DistanceVector *dv;
  struct PilsRequest *req;
};
 
 
static void
sign_dv_cb (void *cls,
            const struct GNUNET_PeerIdentity *pid,
            const struct GNUNET_CRYPTO_EddsaSignature *sig)
{
  struct SignDvCls *sign_dv_cls = cls;
  struct DistanceVector *dv = sign_dv_cls->dv;
  struct PilsRequest *pr = sign_dv_cls->req;
 
  pr->op = NULL;
  GNUNET_CONTAINER_DLL_remove (pils_requests_head,
                               pils_requests_tail,
                               pr);
  GNUNET_free (pr);
 
  dv->sender_sig = *sig;
}
 
 
static void
sign_ephemeral (struct DistanceVector *dv)
{
  struct EphemeralConfirmationPS ec;
  struct SignDvCls *sign_dv_cls;
 
  dv->monotime = GNUNET_TIME_absolute_get_monotonic (GST_cfg);
  dv->ephemeral_validity =
    GNUNET_TIME_absolute_add (dv->monotime, EPHEMERAL_VALIDITY);
  ec.purpose.purpose = htonl (GNUNET_SIGNATURE_PURPOSE_TRANSPORT_EPHEMERAL);
  ec.target = dv->target;
  ec.ephemeral_key = dv->ephemeral_key;
  ec.sender_monotonic_time = GNUNET_TIME_absolute_hton (dv->monotime);
  ec.purpose.size = htonl (sizeof(ec));
  sign_dv_cls = GNUNET_new (struct SignDvCls);
  sign_dv_cls->req = GNUNET_new (struct PilsRequest);
  sign_dv_cls->dv = dv;
  GNUNET_CONTAINER_DLL_insert (pils_requests_head,
                               pils_requests_tail,
                               sign_dv_cls->req);
  sign_dv_cls->req->op = GNUNET_PILS_sign_by_peer_identity (pils,
                                                            &ec.purpose,
                                                            sign_dv_cb,
                                                            sign_dv_cls);
}
 
 
static void
free_queue_entry (struct QueueEntry *qe,
                  struct TransportClient *tc);
 
 
static void
free_timedout_queue_entry (void *cls)
{
  struct TransportClient *tc = cls;
  struct GNUNET_TIME_Absolute now = GNUNET_TIME_absolute_get ();
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "freeing timedout queue entries\n");
 
  tc->details.communicator.free_queue_entry_task = NULL;
  for (struct Queue *queue = tc->details.communicator.queue_head; NULL != queue;
       queue = queue->next_client)
  {
    struct QueueEntry *qep = queue->queue_head;
 
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "checking QID %u for timedout queue entries\n",
                queue->qid);
    while (NULL != qep)
    {
      struct QueueEntry *pos = qep;
      struct GNUNET_TIME_Relative diff = GNUNET_TIME_absolute_get_difference (
        pos->creation_timestamp, now);
      qep = qep->next;
 
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "diff to now %s \n",
                  GNUNET_TIME_relative2s (diff, GNUNET_NO));
      if (GNUNET_TIME_relative_cmp (QUEUE_ENTRY_TIMEOUT, <, diff))
      {
        GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                    "Freeing timed out QueueEntry with MID %" PRIu64
                    " and QID %u\n",
                    pos->mid,
                    queue->qid);
        free_queue_entry (pos, tc);
      }
    }
  }
}
 
 
static void
queue_send_msg (struct Queue *queue,
                struct PendingMessage *pm,
                const void *payload,
                size_t payload_size)
{
  struct Neighbour *n = queue->neighbour;
  struct GNUNET_TRANSPORT_SendMessageTo *smt;
  struct GNUNET_MQ_Envelope *env;
  struct PendingAcknowledgement *pa;
 
  GNUNET_log (
    GNUNET_ERROR_TYPE_DEBUG,
    "Queueing %u bytes of payload for transmission <%" PRIu64
    "> on queue %llu to %s\n",
    (unsigned int) payload_size,
    (NULL == pm) ? 0 : pm->logging_uuid,
    (unsigned long long) queue->qid,
    GNUNET_i2s (&queue->neighbour->pid));
  env = GNUNET_MQ_msg_extra (smt,
                             payload_size,
                             GNUNET_MESSAGE_TYPE_TRANSPORT_SEND_MSG);
  smt->qid = htonl (queue->qid);
  smt->mid = GNUNET_htonll (queue->mid_gen);
  smt->receiver = n->pid;
  memcpy (&smt[1], payload, payload_size);
  {
    /* Pass the env to the communicator of queue for transmission. */
    struct QueueEntry *qe;
 
    qe = GNUNET_new (struct QueueEntry);
    qe->creation_timestamp = GNUNET_TIME_absolute_get ();
    qe->mid = queue->mid_gen;
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Create QueueEntry with MID %" PRIu64
                " and QID %u and prefix %s\n",
                qe->mid,
                queue->qid,
                queue->tc->details.communicator.address_prefix);
    queue->mid_gen++;
    qe->queue = queue;
    if (NULL != pm)
    {
      qe->pm = pm;
      // TODO Why do we have a retransmission. When we know, make decision if we still want this.
      // GNUNET_assert (NULL == pm->qe);
      if (NULL != pm->qe)
      {
        GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                    "Retransmitting message <%" PRIu64
                    "> remove pm from qe with MID: %llu \n",
                    pm->logging_uuid,
                    (unsigned long long) pm->qe->mid);
        pm->qe->pm = NULL;
      }
      pm->qe = qe;
    }
    GNUNET_assert (CT_COMMUNICATOR == queue->tc->type);
    if (0 == queue->q_capacity)
    {
      // Messages without FC or fragments can get here.
      if (NULL != pm)
      {
        GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                    "Message %" PRIu64
                    " (pm type %u) was not send because queue has no capacity.\n",
                    pm->logging_uuid,
                    pm->pmt);
        pm->qe = NULL;
      }
      GNUNET_free (env);
      GNUNET_free (qe);
      return;
    }
    GNUNET_CONTAINER_DLL_insert (queue->queue_head, queue->queue_tail, qe);
    queue->queue_length++;
    queue->tc->details.communicator.total_queue_length++;
    if (GNUNET_NO == queue->unlimited_length)
      queue->q_capacity--;
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Queue %s with qid %u has capacity %" PRIu64 "\n",
                queue->address,
                queue->qid,
                queue->q_capacity);
    if (COMMUNICATOR_TOTAL_QUEUE_LIMIT ==
        queue->tc->details.communicator.total_queue_length)
      queue->idle = GNUNET_NO;
    if (QUEUE_LENGTH_LIMIT == queue->queue_length)
      queue->idle = GNUNET_NO;
    if (0 == queue->q_capacity)
      queue->idle = GNUNET_NO;
 
    if (GNUNET_NO == queue->idle)
    {
      struct TransportClient *tc = queue->tc;
 
      if (NULL == tc->details.communicator.free_queue_entry_task)
        tc->details.communicator.free_queue_entry_task =
          GNUNET_SCHEDULER_add_delayed (GNUNET_TIME_UNIT_SECONDS,
                                        &
                                        free_timedout_queue_entry,
                                        tc);
    }
    if (NULL != pm && NULL != (pa = pm->pa_head))
    {
      while (pm != pa->pm)
        pa = pa->next_pa;
      pa->num_send++;
    }
    // GNUNET_CONTAINER_multiuuidmap_get (pending_acks, &ack[i].ack_uuid.value);
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Sending message MID %" PRIu64
                " of type %u (%u) and size %lu with MQ %p queue %s (QID %u) pending %"
                PRIu64 "\n",
                GNUNET_ntohll (smt->mid),
                ntohs (((const struct GNUNET_MessageHeader *) payload)->type),
                ntohs (smt->header.size),
                (unsigned long) payload_size,
                queue->tc->mq,
                queue->address,
                queue->qid,
                (NULL == pm) ? 0 : pm->logging_uuid);
    GNUNET_MQ_send (queue->tc->mq, env);
  }
}
 
 
static struct GNUNET_TIME_Relative
route_via_neighbour (const struct Neighbour *n,
                     const struct GNUNET_MessageHeader *hdr,
                     enum RouteMessageOptions options)
{
  struct GNUNET_TIME_Absolute now;
  unsigned int candidates;
  unsigned int sel1;
  unsigned int sel2;
  struct GNUNET_TIME_Relative rtt;
 
  /* Pick one or two 'random' queues from n (under constraints of options) */
  now = GNUNET_TIME_absolute_get ();
  /* FIXME-OPTIMIZE: give queues 'weights' and pick proportional to
     weight in the future; weight could be assigned by observed
     bandwidth (note: not sure if we should do this for this type
     of control traffic though). */
  candidates = 0;
  for (struct Queue *pos = n->queue_head; NULL != pos;
       pos = pos->next_neighbour)
  {
    if ((0 != (options & RMO_UNCONFIRMED_ALLOWED)) ||
        (pos->validated_until.abs_value_us > now.abs_value_us))
      candidates++;
  }
  if (0 == candidates)
  {
    /* This can happen rarely if the last confirmed queue timed
       out just as we were beginning to process this message. */
    GNUNET_log (GNUNET_ERROR_TYPE_INFO,
                "Could not route message of type %u to %s: no valid queue\n",
                ntohs (hdr->type),
                GNUNET_i2s (&n->pid));
    GNUNET_STATISTICS_update (GST_stats,
                              "# route selection failed (all no valid queue)",
                              1,
                              GNUNET_NO);
    return GNUNET_TIME_UNIT_FOREVER_REL;
  }
 
  rtt = GNUNET_TIME_UNIT_FOREVER_REL;
  sel1 = GNUNET_CRYPTO_random_u32 (GNUNET_CRYPTO_QUALITY_WEAK, candidates);
  if (0 == (options & RMO_REDUNDANT))
    sel2 = candidates; /* picks none! */
  else
    sel2 = GNUNET_CRYPTO_random_u32 (GNUNET_CRYPTO_QUALITY_WEAK, candidates);
  candidates = 0;
  for (struct Queue *pos = n->queue_head; NULL != pos;
       pos = pos->next_neighbour)
  {
    if ((0 != (options & RMO_UNCONFIRMED_ALLOWED)) ||
        (pos->validated_until.abs_value_us > now.abs_value_us))
    {
      if ((sel1 == candidates) || (sel2 == candidates))
      {
        GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                    "Routing message of type %u to %s using %s (#%u)\n",
                    ntohs (hdr->type),
                    GNUNET_i2s (&n->pid),
                    pos->address,
                    (sel1 == candidates) ? 1 : 2);
        rtt = GNUNET_TIME_relative_min (rtt, pos->pd.aged_rtt);
        queue_send_msg (pos, NULL, hdr, ntohs (hdr->size));
      }
      candidates++;
    }
  }
  return rtt;
}
 
 
struct DVKeyState
{
  gcry_cipher_hd_t cipher;
 
  struct
  {
    struct GNUNET_CRYPTO_AuthKey hmac_key;
 
    char aes_key[256 / 8];
 
    char aes_ctr[128 / 8];
  } material;
};
 
 
static void
dv_setup_key_state_from_km (const struct GNUNET_ShortHashCode *km,
                            const struct GNUNET_ShortHashCode *iv,
                            struct DVKeyState *key)
{
  /* must match what we defive from decapsulated key */
  GNUNET_assert (GNUNET_YES ==
                 GNUNET_CRYPTO_hkdf_expand (&key->material,
                                            sizeof(key->material),
                                            km,
                                            "gnunet-transport-dv-key",
                                            strlen ("gnunet-transport-dv-key")
                                            ,
                                            km,
                                            sizeof(*km),
                                            iv,
                                            sizeof(*iv),
                                            NULL));
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Deriving backchannel key based on KM %s and IV %s\n",
              GNUNET_sh2s (km),
              GNUNET_sh2s (iv));
  GNUNET_assert (0 == gcry_cipher_open (&key->cipher,
                                        GCRY_CIPHER_AES256 /* low level: go for speed */
                                        ,
                                        GCRY_CIPHER_MODE_CTR,
                                        0 /* flags */));
  GNUNET_assert (0 == gcry_cipher_setkey (key->cipher,
                                          &key->material.aes_key,
                                          sizeof(key->material.aes_key)));
  gcry_cipher_setctr (key->cipher,
                      &key->material.aes_ctr,
                      sizeof(key->material.aes_ctr));
}
 
 
static void
dv_hmac (const struct DVKeyState *key,
         struct GNUNET_HashCode *hmac,
         const void *data,
         size_t data_size)
{
  GNUNET_CRYPTO_hmac (&key->material.hmac_key, data, data_size, hmac);
}
 
 
static void
dv_encrypt (struct DVKeyState *key, const void *in, void *dst, size_t in_size)
{
  GNUNET_assert (0 ==
                 gcry_cipher_encrypt (key->cipher, dst, in_size, in, in_size));
}
 
 
static enum GNUNET_GenericReturnValue
dv_decrypt (struct DVKeyState *key,
            void *out,
            const void *ciph,
            size_t out_size)
{
  return (0 ==
          gcry_cipher_decrypt (key->cipher,
                               out, out_size,
                               ciph, out_size)) ? GNUNET_OK : GNUNET_SYSERR;
}
 
 
static void
dv_key_clean (struct DVKeyState *key)
{
  gcry_cipher_close (key->cipher);
  GNUNET_CRYPTO_zero_keys (&key->material, sizeof(key->material));
}
 
 
typedef void (*DVMessageHandler) (void *cls,
                                  struct Neighbour *next_hop,
                                  const struct GNUNET_MessageHeader *hdr,
                                  enum RouteMessageOptions options);
 
static struct GNUNET_TIME_Relative
encapsulate_for_dv (struct DistanceVector *dv,
                    unsigned int num_dvhs,
                    struct DistanceVectorHop **dvhs,
                    const struct GNUNET_MessageHeader *hdr,
                    DVMessageHandler use,
                    void *use_cls,
                    enum RouteMessageOptions options,
                    enum GNUNET_GenericReturnValue without_fc)
{
  struct TransportDVBoxMessage box_hdr;
  struct TransportDVBoxPayloadP payload_hdr;
  uint16_t enc_body_size = ntohs (hdr->size);
  char enc[sizeof(struct TransportDVBoxPayloadP) + enc_body_size] GNUNET_ALIGN;
  struct DVKeyState *key;
  struct GNUNET_TIME_Relative rtt;
  struct GNUNET_ShortHashCode km;
 
  key = GNUNET_new (struct DVKeyState);
  /* Encrypt payload */
  box_hdr.header.type = htons (GNUNET_MESSAGE_TYPE_TRANSPORT_DV_BOX);
  box_hdr.total_hops = htons (0);
  box_hdr.without_fc = htons (without_fc);
  // update_ephemeral (dv);
  if (0 ==
      GNUNET_TIME_absolute_get_remaining (dv->ephemeral_validity).rel_value_us)
  {
    GNUNET_CRYPTO_eddsa_kem_encaps (&dv->target.public_key,
                                    &dv->ephemeral_key,
                                    &km);
    dv->km = GNUNET_new (struct GNUNET_ShortHashCode);
    GNUNET_memcpy (dv->km, &km, sizeof(struct GNUNET_ShortHashCode));
    sign_ephemeral (dv);
  }
  box_hdr.ephemeral_key = dv->ephemeral_key;
  payload_hdr.sender_sig = dv->sender_sig;
 
  GNUNET_CRYPTO_random_block (GNUNET_CRYPTO_QUALITY_NONCE,
                              &box_hdr.iv,
                              sizeof(box_hdr.iv));
  // We are creating this key, so this must work.
  // FIXME: Possibly also add return values here. We are processing
  // Input from other peers...
  dv_setup_key_state_from_km (dv->km, &box_hdr.iv, key);
  payload_hdr.sender = *GST_my_identity;
  payload_hdr.monotonic_time = GNUNET_TIME_absolute_hton (dv->monotime);
  dv_encrypt (key, &payload_hdr, enc, sizeof(payload_hdr));
  dv_encrypt (key,
              hdr,
              &enc[sizeof(struct TransportDVBoxPayloadP)],
              enc_body_size);
  dv_hmac (key, &box_hdr.hmac, enc, sizeof(enc));
  dv_key_clean (key);
  rtt = GNUNET_TIME_UNIT_FOREVER_REL;
  /* For each selected path, take the pre-computed header and body
     and add the path in the middle of the message; then send it. */
  for (unsigned int i = 0; i < num_dvhs; i++)
  {
    struct DistanceVectorHop *dvh = dvhs[i];
    unsigned int num_hops = dvh->distance + 1;
    char buf[sizeof(struct TransportDVBoxMessage)
             + sizeof(struct GNUNET_PeerIdentity) * num_hops
             + sizeof(struct TransportDVBoxPayloadP)
             + enc_body_size] GNUNET_ALIGN;
    struct GNUNET_PeerIdentity *dhops;
 
    box_hdr.header.size = htons (sizeof(buf));
    box_hdr.orig_size = htons (sizeof(buf));
    box_hdr.num_hops = htons (num_hops);
    memcpy (buf, &box_hdr, sizeof(box_hdr));
    dhops = (struct GNUNET_PeerIdentity *) &buf[sizeof(box_hdr)];
    memcpy (dhops,
            dvh->path,
            dvh->distance * sizeof(struct GNUNET_PeerIdentity));
    dhops[dvh->distance] = dv->target;
    if (GNUNET_EXTRA_LOGGING > 0)
    {
      char *path;
 
      path = GNUNET_strdup (GNUNET_i2s (GST_my_identity));
      for (unsigned int j = 0; j < num_hops; j++)
      {
        char *tmp;
 
        GNUNET_asprintf (&tmp, "%s-%s", path, GNUNET_i2s (&dhops[j]));
        GNUNET_free (path);
        path = tmp;
      }
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "Routing message of type %u to %s using DV (#%u/%u) via %s\n",
                  ntohs (hdr->type),
                  GNUNET_i2s (&dv->target),
                  i + 1,
                  num_dvhs,
                  path);
      GNUNET_free (path);
    }
    rtt = GNUNET_TIME_relative_min (rtt, dvh->pd.aged_rtt);
    memcpy (&dhops[num_hops], enc, sizeof(enc));
    use (use_cls,
         dvh->next_hop,
         (const struct GNUNET_MessageHeader *) buf,
         options);
    GNUNET_free (key);
  }
  return rtt;
}
 
 
static void
send_dv_to_neighbour (void *cls,
                      struct Neighbour *next_hop,
                      const struct GNUNET_MessageHeader *hdr,
                      enum RouteMessageOptions options)
{
  (void) cls;
  (void) route_via_neighbour (next_hop, hdr, RMO_UNCONFIRMED_ALLOWED);
}
 
 
static struct GNUNET_TIME_Relative
route_control_message_without_fc (struct VirtualLink *vl,
// route_control_message_without_fc (const struct GNUNET_PeerIdentity *target,
                                  const struct GNUNET_MessageHeader *hdr,
                                  enum RouteMessageOptions options)
{
  // struct VirtualLink *vl;
  struct Neighbour *n;
  struct DistanceVector *dv;
  struct GNUNET_TIME_Relative rtt1;
  struct GNUNET_TIME_Relative rtt2;
  const struct GNUNET_PeerIdentity *target = &vl->target;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Trying to route message of type %u to %s without fc\n",
              ntohs (hdr->type),
              GNUNET_i2s (target));
 
  // TODO Do this elsewhere. vl should be given as parameter to method.
  // vl = lookup_virtual_link (target);
  GNUNET_assert (NULL != vl && GNUNET_YES == vl->confirmed);
  if (NULL == vl)
    return GNUNET_TIME_UNIT_FOREVER_REL;
  n = vl->n;
  dv = (0 != (options & RMO_DV_ALLOWED)) ? vl->dv : NULL;
  if (0 == (options & RMO_UNCONFIRMED_ALLOWED))
  {
    /* if confirmed is required, and we do not have anything
       confirmed, drop respective options */
    if (NULL == n)
      n = lookup_neighbour (target);
    if ((NULL == dv) && (0 != (options & RMO_DV_ALLOWED)))
      dv = GNUNET_CONTAINER_multipeermap_get (dv_routes, target);
  }
  if ((NULL == n) && (NULL == dv))
  {
    GNUNET_log (GNUNET_ERROR_TYPE_INFO,
                "Cannot route message of type %u to %s: no route\n",
                ntohs (hdr->type),
                GNUNET_i2s (target));
    GNUNET_STATISTICS_update (GST_stats,
                              "# Messages dropped in routing: no acceptable method",
                              1,
                              GNUNET_NO);
    return GNUNET_TIME_UNIT_FOREVER_REL;
  }
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Routing message of type %u to %s with options %X\n",
              ntohs (hdr->type),
              GNUNET_i2s (target),
              (unsigned int) options);
  /* If both dv and n are possible and we must choose:
     flip a coin for the choice between the two; for now 50/50 */
  if ((NULL != n) && (NULL != dv) && (0 == (options & RMO_REDUNDANT)))
  {
    if (0 == GNUNET_CRYPTO_random_u32 (GNUNET_CRYPTO_QUALITY_WEAK, 2))
      n = NULL;
    else
      dv = NULL;
  }
  if ((NULL != n) && (NULL != dv))
    options &= ~RMO_REDUNDANT; /* We will do one DV and one direct, that's
                                  enough for redundancy, so clear the flag. */
  rtt1 = GNUNET_TIME_UNIT_FOREVER_REL;
  rtt2 = GNUNET_TIME_UNIT_FOREVER_REL;
  if (NULL != n)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Try to route message of type %u to %s without fc via neighbour\n",
                ntohs (hdr->type),
                GNUNET_i2s (target));
    rtt1 = route_via_neighbour (n, hdr, options);
  }
  if (NULL != dv)
  {
    struct DistanceVectorHop *hops[2];
    unsigned int res;
 
    res = pick_random_dv_hops (dv,
                               options,
                               hops,
                               (0 == (options & RMO_REDUNDANT)) ? 1 : 2);
    if (0 == res)
    {
      GNUNET_log (GNUNET_ERROR_TYPE_INFO,
                  "Failed to route message, could not determine DV path\n");
      return rtt1;
    }
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "encapsulate_for_dv 1\n");
    rtt2 = encapsulate_for_dv (dv,
                               res,
                               hops,
                               hdr,
                               &send_dv_to_neighbour,
                               NULL,
                               options & (~RMO_REDUNDANT),
                               GNUNET_YES);
  }
  return GNUNET_TIME_relative_min (rtt1, rtt2);
}
 
 
static void
consider_sending_fc (void *cls);
 
static void
task_consider_sending_fc (void *cls)
{
  struct VirtualLink *vl = cls;
  vl->fc_retransmit_task = NULL;
  consider_sending_fc (cls);
}
 
 
static char *
get_address_without_port (const char *address);
 
 
struct AddGlobalAddressesContext
{
  size_t off;
  char *tgnas;
};
 
 
static enum GNUNET_GenericReturnValue
add_global_addresses (void *cls,
                      const struct GNUNET_PeerIdentity *pid,
                      void *value)
{
  struct AddGlobalAddressesContext *ctx = cls;
  struct TransportGlobalNattedAddress *tgna = value;
  char *addr = (char *) &tgna[1];
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "sending address %s length %u\n",
              addr,
              ntohl (tgna->address_length));
  GNUNET_memcpy (&(ctx->tgnas[ctx->off]), tgna, sizeof (struct
                                                        TransportGlobalNattedAddress)
                 + ntohl (tgna->address_length));
  ctx->off += sizeof(struct TransportGlobalNattedAddress) + ntohl (tgna->
                                                                   address_length);
 
  return GNUNET_OK;
}
 
 
static struct GNUNET_TIME_Relative
calculate_rtt (struct DistanceVector *dv);
 
 
static void
consider_sending_fc (void *cls)
{
  struct VirtualLink *vl = cls;
  struct GNUNET_TIME_Absolute monotime;
  struct TransportFlowControlMessage *fc;
  struct GNUNET_TIME_Relative duration;
  struct GNUNET_TIME_Relative rtt;
  struct GNUNET_TIME_Relative rtt_average;
  struct Neighbour *n = vl->n;
 
  if (NULL != n && 0 < n->number_of_addresses)
  {
    size_t addresses_size =
      n->number_of_addresses * sizeof (struct TransportGlobalNattedAddress) + n
      ->size_of_global_addresses;
    char *tgnas = GNUNET_malloc (addresses_size);
    struct AddGlobalAddressesContext ctx;
    ctx.off = 0;
    ctx.tgnas = tgnas;
 
    fc = GNUNET_malloc (sizeof (struct TransportFlowControlMessage)
                        + addresses_size);
    fc->header.size = htons (sizeof(struct TransportFlowControlMessage)
                             + addresses_size);
    fc->size_of_addresses = htonl (n->size_of_global_addresses);
    fc->number_of_addresses = htonl (n->number_of_addresses);
    GNUNET_CONTAINER_multipeermap_iterate (n->natted_addresses,
                                           &add_global_addresses,
                                           &ctx);
    GNUNET_memcpy (&fc[1], tgnas, addresses_size);
    GNUNET_free (tgnas);
  }
  else
  {
    fc = GNUNET_malloc (sizeof (struct TransportFlowControlMessage));
    fc->header.size = htons (sizeof(struct TransportFlowControlMessage));
  }
 
  duration = GNUNET_TIME_absolute_get_duration (vl->last_fc_transmission);
  /* OPTIMIZE-FC-BDP: decide sane criteria on when to do this, instead of doing
     it always! */
  /* For example, we should probably ONLY do this if a bit more than
     an RTT has passed, or if the window changed "significantly" since
     then. See vl->last_fc_rtt! NOTE: to do this properly, we also
     need an estimate for the bandwidth-delay-product for the entire
     VL, as that determines "significantly". We have the delay, but
     the bandwidth statistics need to be added for the VL!*/(void) duration;
 
  if (NULL != vl->dv)
    rtt_average = calculate_rtt (vl->dv);
  else
    rtt_average = GNUNET_TIME_UNIT_FOREVER_REL;
  fc->rtt = GNUNET_TIME_relative_hton (rtt_average);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Sending FC seq %u to %s with new window %llu %lu %u\n",
              (unsigned int) vl->fc_seq_gen,
              GNUNET_i2s (&vl->target),
              (unsigned long long) vl->incoming_fc_window_size,
              (unsigned long) rtt_average.rel_value_us,
              vl->sync_ready);
  monotime = GNUNET_TIME_absolute_get_monotonic (GST_cfg);
  vl->last_fc_transmission = monotime;
  fc->sync_ready = vl->sync_ready;
  fc->header.type = htons (GNUNET_MESSAGE_TYPE_TRANSPORT_FLOW_CONTROL);
  fc->seq = htonl (vl->fc_seq_gen++);
  fc->inbound_window_size = GNUNET_htonll (vl->incoming_fc_window_size
                                           + vl->incoming_fc_window_size_used
                                           + vl->incoming_fc_window_size_loss);
  fc->outbound_sent = GNUNET_htonll (vl->outbound_fc_window_size_used);
  fc->outbound_window_size = GNUNET_htonll (vl->outbound_fc_window_size);
  fc->sender_time = GNUNET_TIME_absolute_hton (monotime);
  rtt = route_control_message_without_fc (vl, &fc->header, RMO_DV_ALLOWED);
  if (GNUNET_TIME_UNIT_FOREVER_REL.rel_value_us == rtt.rel_value_us)
  {
    rtt = GNUNET_TIME_UNIT_SECONDS;
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "FC retransmission to %s failed, will retry in %s\n",
                GNUNET_i2s (&vl->target),
                GNUNET_STRINGS_relative_time_to_string (rtt, GNUNET_YES));
    vl->last_fc_rtt = GNUNET_TIME_UNIT_ZERO;
  }
  else
  {
    /* OPTIMIZE-FC-BDP: rtt is not ideal, we can do better! */
    vl->last_fc_rtt = rtt;
  }
  if (NULL != vl->fc_retransmit_task)
    GNUNET_SCHEDULER_cancel (vl->fc_retransmit_task);
  if (MAX_FC_RETRANSMIT_COUNT == vl->fc_retransmit_count)
  {
    rtt = GNUNET_TIME_UNIT_MINUTES;
    vl->fc_retransmit_count = 0;
  }
  vl->fc_retransmit_task =
    GNUNET_SCHEDULER_add_delayed (rtt, &task_consider_sending_fc, vl);
  vl->fc_retransmit_count++;
  GNUNET_free (fc);
}
 
 
static void
check_vl_transmission (struct VirtualLink *vl)
{
  struct Neighbour *n = vl->n;
  struct DistanceVector *dv = vl->dv;
  struct GNUNET_TIME_Absolute now;
  struct VirtualLink *vl_next_hop;
  int elig;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "check_vl_transmission to target %s\n",
              GNUNET_i2s (&vl->target));
  /* Check that we have an eligible pending message!
     (cheaper than having #transmit_on_queue() find out!) */
  elig = GNUNET_NO;
  for (struct PendingMessage *pm = vl->pending_msg_head; NULL != pm;
       pm = pm->next_vl)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "check_vl_transmission loop\n");
    if (NULL != pm->qe)
      continue;   /* not eligible, is in a queue! */
    if (pm->bytes_msg + vl->outbound_fc_window_size_used >
        vl->outbound_fc_window_size)
    {
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "Stalled message %" PRIu64
                  " transmission on VL %s due to flow control: %llu < %llu\n",
                  pm->logging_uuid,
                  GNUNET_i2s (&vl->target),
                  (unsigned long long) vl->outbound_fc_window_size,
                  (unsigned long long) (pm->bytes_msg
                                        + vl->outbound_fc_window_size_used));
      consider_sending_fc (vl);
      return;     /* We have a message, but flow control says "nope" */
    }
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Target window on VL %s not stalled. Scheduling transmission on queue\n",
                GNUNET_i2s (&vl->target));
    /* Notify queues at direct neighbours that we are interested */
    now = GNUNET_TIME_absolute_get ();
    if (NULL != n)
    {
      for (struct Queue *queue = n->queue_head; NULL != queue;
           queue = queue->next_neighbour)
      {
        if ((GNUNET_YES == queue->idle) &&
            (queue->validated_until.abs_value_us > now.abs_value_us))
        {
          GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                      "Direct neighbour %s not stalled\n",
                      GNUNET_i2s (&n->pid));
          schedule_transmit_on_queue (GNUNET_TIME_UNIT_ZERO,
                                      queue,
                                      GNUNET_SCHEDULER_PRIORITY_DEFAULT);
          elig = GNUNET_YES;
        }
        else
          GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                      "Neighbour Queue QID: %u (%u) busy or invalid\n",
                      queue->qid,
                      queue->idle);
      }
    }
    /* Notify queues via DV that we are interested */
    if (NULL != dv)
    {
      /* Do DV with lower scheduler priority, which effectively means that
         IF a neighbour exists and is available, we prefer it. */
      for (struct DistanceVectorHop *pos = dv->dv_head; NULL != pos;
           pos = pos->next_dv)
      {
        struct Neighbour *nh_iter = pos->next_hop;
 
 
        if (pos->path_valid_until.abs_value_us <= now.abs_value_us)
          continue;   /* skip this one: path not validated */
        else
        {
          vl_next_hop = lookup_virtual_link (&nh_iter->pid);
          GNUNET_assert (NULL != vl_next_hop);
          if (pm->bytes_msg + vl_next_hop->outbound_fc_window_size_used >
              vl_next_hop->outbound_fc_window_size)
          {
            GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                        "Stalled message %" PRIu64
                        " transmission on next hop %s due to flow control: %llu < %llu\n",
                        pm->logging_uuid,
                        GNUNET_i2s (&vl_next_hop->target),
                        (unsigned long
                         long) vl_next_hop->outbound_fc_window_size,
                        (unsigned long long) (pm->bytes_msg
                                              + vl_next_hop->
                                              outbound_fc_window_size_used));
            consider_sending_fc (vl_next_hop);
            continue; /* We have a message, but flow control says "nope" for the first hop of this path */
          }
          for (struct Queue *queue = nh_iter->queue_head; NULL != queue;
               queue = queue->next_neighbour)
            if ((GNUNET_YES == queue->idle) &&
                (queue->validated_until.abs_value_us > now.abs_value_us))
            {
              GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                          "Next hop neighbour %s not stalled\n",
                          GNUNET_i2s (&nh_iter->pid));
              schedule_transmit_on_queue (GNUNET_TIME_UNIT_ZERO,
                                          queue,
                                          GNUNET_SCHEDULER_PRIORITY_BACKGROUND);
              elig = GNUNET_YES;
            }
            else
              GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                          "DV Queue QID: %u (%u) busy or invalid\n",
                          queue->qid,
                          queue->idle);
        }
      }
    }
    if (GNUNET_YES == elig)
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "Eligible message %" PRIu64 " of size %u to %s: %llu/%llu\n",
                  pm->logging_uuid,
                  pm->bytes_msg,
                  GNUNET_i2s (&vl->target),
                  (unsigned long long) vl->outbound_fc_window_size,
                  (unsigned long long) (pm->bytes_msg
                                        + vl->outbound_fc_window_size_used));
    break;
  }
}
 
 
static void
handle_client_send (void *cls, const struct OutboundMessage *obm)
{
  struct TransportClient *tc = cls;
  struct PendingMessage *pm;
  const struct GNUNET_MessageHeader *obmm;
  uint32_t bytes_msg;
  struct VirtualLink *vl;
  enum GNUNET_MQ_PriorityPreferences pp;
 
  GNUNET_assert (CT_CORE == tc->type);
  obmm = (const struct GNUNET_MessageHeader *) &obm[1];
  bytes_msg = ntohs (obmm->size);
  pp = ntohl (obm->priority);
  vl = lookup_virtual_link (&obm->peer);
  if ((NULL == vl) || (GNUNET_NO == vl->confirmed))
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Don't have %s as a neighbour (anymore).\n",
                GNUNET_i2s (&obm->peer));
    /* Failure: don't have this peer as a neighbour (anymore).
       Might have gone down asynchronously, so this is NOT
       a protocol violation by CORE. Still count the event,
       as this should be rare. */
    GNUNET_SERVICE_client_continue (tc->client);
    GNUNET_STATISTICS_update (GST_stats,
                              "# messages dropped (neighbour unknown)",
                              1,
                              GNUNET_NO);
    return;
  }
 
  pm = GNUNET_malloc (sizeof(struct PendingMessage) + bytes_msg);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "1 created pm %p storing vl %p\n",
              pm,
              vl);
  pm->logging_uuid = logging_uuid_gen++;
  pm->prefs = pp;
  pm->client = tc;
  pm->vl = vl;
  pm->bytes_msg = bytes_msg;
  memcpy (&pm[1], obmm, bytes_msg);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Sending message of type %u  with %u bytes as <%" PRIu64
              "> to %s\n",
              ntohs (obmm->type),
              bytes_msg,
              pm->logging_uuid,
              GNUNET_i2s (&obm->peer));
  GNUNET_CONTAINER_MDLL_insert (client,
                                tc->details.core.pending_msg_head,
                                tc->details.core.pending_msg_tail,
                                pm);
  GNUNET_CONTAINER_MDLL_insert (vl,
                                vl->pending_msg_head,
                                vl->pending_msg_tail,
                                pm);
  check_vl_transmission (vl);
  GNUNET_SERVICE_client_continue (tc->client);
}
 
 
static void
handle_communicator_backchannel (
  void *cls,
  const struct GNUNET_TRANSPORT_CommunicatorBackchannel *cb)
{
  struct Neighbour *n;
  struct VirtualLink *vl;
  struct TransportClient *tc = cls;
  const struct GNUNET_MessageHeader *inbox =
    (const struct GNUNET_MessageHeader *) &cb[1];
  uint16_t isize = ntohs (inbox->size);
  const char *is = ((const char *) &cb[1]) + isize;
  size_t slen = strlen (is) + 1;
  char
    mbuf[slen + isize
         + sizeof(struct
                  TransportBackchannelEncapsulationMessage)] GNUNET_ALIGN;
  struct TransportBackchannelEncapsulationMessage *be =
    (struct TransportBackchannelEncapsulationMessage *) mbuf;
 
  /* 0-termination of 'is' was checked already in
   #check_communicator_backchannel() */
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Preparing backchannel transmission to %s:%s of type %u and size %u\n",
              GNUNET_i2s (&cb->pid),
              is,
              ntohs (inbox->type),
              ntohs (inbox->size));
  /* encapsulate and encrypt message */
  be->header.type =
    htons (GNUNET_MESSAGE_TYPE_TRANSPORT_BACKCHANNEL_ENCAPSULATION);
  be->header.size = htons (sizeof(mbuf));
  memcpy (&be[1], inbox, isize);
  memcpy (&mbuf[sizeof(struct TransportBackchannelEncapsulationMessage)
                + isize],
          is,
          strlen (is) + 1);
  // route_control_message_without_fc (&cb->pid, &be->header, RMO_DV_ALLOWED);
  vl = lookup_virtual_link (&cb->pid);
  if ((NULL != vl) && (GNUNET_YES == vl->confirmed))
  {
    route_control_message_without_fc (vl, &be->header, RMO_DV_ALLOWED);
  }
  else
  {
    /* Use route via neighbour */
    n = lookup_neighbour (&cb->pid);
    if (NULL != n)
      route_via_neighbour (
        n,
        &be->header,
        RMO_NONE);
  }
  GNUNET_SERVICE_client_continue (tc->client);
}
 
 
static int
check_add_address (void *cls,
                   const struct GNUNET_TRANSPORT_AddAddressMessage *aam)
{
  struct TransportClient *tc = cls;
 
  if (CT_COMMUNICATOR != tc->type)
  {
    GNUNET_break (0);
    return GNUNET_SYSERR;
  }
  GNUNET_MQ_check_zero_termination (aam);
  return GNUNET_OK;
}
 
 
static void
store_pi (void *cls);
 
 
struct PilsAddressSignContext
{
 
  struct AddressListEntry *ale;
 
  struct PilsRequest *req;
 
 
  struct GNUNET_TIME_Absolute et;
};
 
 
static void
shc_cont (void *cls, int success)
{
  struct PilsAddressSignContext *pc = cls;
 
  GNUNET_assert (NULL == pc->req);
  if (GNUNET_OK != success)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
                "Failed to store our address `%s' with peerstore\n",
                pc->ale->address);
    if (NULL == pc->ale->st)
    {
      pc->ale->st = GNUNET_SCHEDULER_add_delayed (GNUNET_TIME_UNIT_SECONDS,
                                                  &store_pi,
                                                  pc->ale);
    }
  }
  GNUNET_free (pc);
}
 
 
static void
pils_sign_hello_cb (void *cls,
                    const struct GNUNET_PeerIdentity *pid,
                    const struct GNUNET_CRYPTO_EddsaSignature *sig)
{
  struct PilsAddressSignContext *pc = cls;
  struct GNUNET_MQ_Envelope *env;
  const struct GNUNET_MessageHeader *msg;
 
  pc->req->op = NULL;
  GNUNET_CONTAINER_DLL_remove (pils_requests_head,
                               pils_requests_tail,
                               pc->req);
  GNUNET_free (pc->req);
  pc->req = NULL;
  env = GNUNET_HELLO_builder_to_env (
    GST_my_hello,
    pid,
    sig,
    pc->et);
  msg = GNUNET_MQ_env_get_msg (env);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "store_pi 1\n");
  pc->ale->shc = GNUNET_PEERSTORE_hello_add (peerstore,
                                             msg,
                                             shc_cont,
                                             pc);
  GNUNET_free (env);
}
 
 
static void
peerstore_store_own_cb (void *cls, int success)
{
  struct PilsAddressSignContext *pc = cls;
 
  pc->ale->sc = NULL;
  if (GNUNET_YES != success)
    GNUNET_log (GNUNET_ERROR_TYPE_ERROR,
                "Failed to store our own address `%s' in peerstore!\n",
                pc->ale->address);
  else
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Successfully stored our own address `%s' in peerstore!\n",
                pc->ale->address);
  /* refresh period is 1/4 of expiration time, that should be plenty
     without being excessive. */
  if (NULL == pc->ale->st)
  {
    pc->ale->st =
      GNUNET_SCHEDULER_add_delayed (
        GNUNET_TIME_relative_divide (pc->ale->expiration,
                                     4ULL),
        &store_pi,
        pc->ale);
  }
 
  /* Now we have to update our HELLO! */
  pc->et = GNUNET_TIME_relative_to_absolute (GNUNET_HELLO_ADDRESS_EXPIRATION);
  pc->req = GNUNET_new (struct PilsRequest);
  GNUNET_CONTAINER_DLL_insert (pils_requests_head,
                               pils_requests_tail,
                               pc->req);
  pc->req->op = GNUNET_PILS_sign_hello (pils,
                                        GST_my_hello,
                                        pc->et,
                                        &pils_sign_hello_cb,
                                        pc);
}
 
 
// This function
static void
pils_sign_addr_cb (void *cls,
                   const struct GNUNET_PeerIdentity *pid,
                   const struct GNUNET_CRYPTO_EddsaSignature *sig)
{
  struct PilsAddressSignContext *pc = cls;
  char *sig_str;
  void *result;
  size_t result_size;
 
  pc->req->op = NULL;
  GNUNET_CONTAINER_DLL_remove (pils_requests_head,
                               pils_requests_tail,
                               pc->req);
  GNUNET_free (pc->req);
  sig_str = NULL;
  (void) GNUNET_STRINGS_base64_encode (sig, sizeof(*sig), &sig_str);
  result_size =
    1 + GNUNET_asprintf (
      (char **) &result,
      "%s;%llu;%u;%s",
      sig_str,
      (unsigned long long) pc->et.abs_value_us,
      (unsigned int) pc->ale->nt,
      pc->ale->address);
  GNUNET_free (sig_str);
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Build our HELLO URI `%s'\n",
              (char*) result);
 
  pc->ale->signed_address = result;
  pc->ale->signed_address_len = result_size;
  struct GNUNET_TIME_Absolute expiration;
 
  expiration = GNUNET_TIME_relative_to_absolute (pc->ale->expiration);
  pc->ale->sc = GNUNET_PEERSTORE_store (peerstore,
                                        "transport",
                                        GST_my_identity, // FIXME
                                        GNUNET_PEERSTORE_TRANSPORT_HELLO_KEY,
                                        result,
                                        result_size,
                                        expiration,
                                        GNUNET_PEERSTORE_STOREOPTION_MULTIPLE,
                                        &peerstore_store_own_cb,
                                        pc);
}
 
 
struct SignedAddress
{
  struct GNUNET_CRYPTO_EccSignaturePurpose purpose;
 
  struct GNUNET_TIME_AbsoluteNBO mono_time;
 
  struct GNUNET_HashCode addr_hash GNUNET_PACKED;
};
 
 
void
pils_sign_address (
  struct AddressListEntry *ale,
  struct GNUNET_TIME_Absolute mono_time)
{
  struct SignedAddress sa;
  struct PilsAddressSignContext *pc;
 
  sa.purpose.purpose = htonl (GNUNET_SIGNATURE_PURPOSE_TRANSPORT_ADDRESS);
  sa.purpose.size = htonl (sizeof(sa));
  sa.mono_time = GNUNET_TIME_absolute_hton (mono_time);
  GNUNET_CRYPTO_hash (ale->address, strlen (ale->address), &sa.addr_hash);
  pc = GNUNET_new (struct PilsAddressSignContext);
  pc->ale = ale;
  pc->et = mono_time;
  pc->req = GNUNET_new (struct PilsRequest);
  pc->req->op = GNUNET_PILS_sign_by_peer_identity (pils,
                                                   &sa.purpose,
                                                   pils_sign_addr_cb,
                                                   pc);
  GNUNET_CONTAINER_DLL_insert (pils_requests_head,
                               pils_requests_tail,
                               pc->req);
}
 
 
static void
store_pi (void *cls)
{
  struct AddressListEntry *ale = cls;
  const char *dash;
  char *address_uri;
  char *prefix = GNUNET_HELLO_address_to_prefix (ale->address);
  unsigned int add_success;
 
  if (NULL == GST_my_identity)
  {
    ale->st = GNUNET_SCHEDULER_add_delayed (GNUNET_TIME_UNIT_MILLISECONDS,
                                            &store_pi,
                                            ale);
    return;
  }
 
  dash = strchr (ale->address, '-');
  GNUNET_assert (NULL != dash);
  dash++;
  GNUNET_asprintf (&address_uri,
                   "%s://%s",
                   prefix,
                   dash);
  GNUNET_free (prefix);
  ale->st = NULL;
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Storing our address `%s' in peerstore until %s!\n",
              ale->address,
              GNUNET_STRINGS_absolute_time_to_string (hello_mono_time));
  add_success = GNUNET_HELLO_builder_add_address (GST_my_hello,
                                                  address_uri);
  if (GNUNET_OK != add_success)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
                "Storing our address `%s' %s\n",
                address_uri,
                GNUNET_NO == add_success ? "not done" : "failed");
    GNUNET_free (address_uri);
    return;
  }
  else
  {
 
    GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
                "Storing our address `%s'\n",
                address_uri);
  }
  // FIXME hello_mono_time used here?? What about expiration in ale?
  pils_sign_address (ale,
                     hello_mono_time);
  // TODO keep track of op and potentially cancle/clean
  GNUNET_free (address_uri);
}
 
 
static struct AddressListEntry *
create_address_entry (struct TransportClient *tc,
                      struct GNUNET_TIME_Relative expiration,
                      enum GNUNET_NetworkType nt,
                      const char *address,
                      uint32_t aid,
                      size_t slen)
{
  struct AddressListEntry *ale;
  char *address_without_port;
 
  ale = GNUNET_malloc (sizeof(struct AddressListEntry) + slen);
  ale->tc = tc;
  ale->address = (const char *) &ale[1];
  ale->expiration = expiration;
  ale->aid = aid;
  ale->nt = nt;
  memcpy (&ale[1], address, slen);
  address_without_port = get_address_without_port (ale->address);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Is this %s a local address (%s)\n",
              address_without_port,
              ale->address);
  if (0 != strcmp ("127.0.0.1", address_without_port))
  {
    if (NULL != ale->st)
    {
      GNUNET_SCHEDULER_cancel (ale->st);
    }
    ale->st = GNUNET_SCHEDULER_add_now (&store_pi, ale);
  }
  GNUNET_free (address_without_port);
 
  return ale;
}
 
 
static void
feed_addresses_to_pils (void *cls)
{
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Feeding addresses to PILS\n");
  pils_feed_task = NULL;
 
  GNUNET_PILS_feed_addresses (pils,
                              GST_my_hello);
}
 
 
static void
handle_add_address (void *cls,
                    const struct GNUNET_TRANSPORT_AddAddressMessage *aam)
{
  struct TransportClient *tc = cls;
  struct AddressListEntry *ale;
  size_t slen;
  char *address;
 
  /* 0-termination of &aam[1] was checked in #check_add_address */
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Communicator added address `%s'!\n",
              (const char *) &aam[1]);
  slen = ntohs (aam->header.size) - sizeof(*aam);
  address = GNUNET_malloc (slen);
  memcpy (address, &aam[1], slen);
  ale = create_address_entry (tc,
                              GNUNET_TIME_relative_ntoh (aam->expiration),
                              ntohl (aam->nt),
                              address,
                              aam->aid,
                              slen);
  GNUNET_CONTAINER_DLL_insert (tc->details.communicator.addr_head,
                               tc->details.communicator.addr_tail,
                               ale);
  {
    for (struct AddressListEntry *iter = tc->details.communicator.addr_head;
         (NULL != iter && NULL != iter->next);
         iter = iter->next)
    {
      char *address_uri;
      char *dash = strchr (ale->address, '-');
      char *prefix = GNUNET_HELLO_address_to_prefix (ale->address);
      GNUNET_assert (NULL != dash);
      dash++;
      GNUNET_asprintf (&address_uri,
                       "%s://%s",
                       prefix,
                       dash);
      GNUNET_free (prefix);
      GNUNET_HELLO_builder_add_address (GST_my_hello, address_uri);
      GNUNET_free (address_uri);
    }
    if (NULL != pils_feed_task)
      GNUNET_SCHEDULER_cancel (pils_feed_task);
    pils_feed_task = GNUNET_SCHEDULER_add_delayed (PILS_FEED_ADDRESSES_DELAY,
                                                   &feed_addresses_to_pils,
                                                   NULL);
  }
  GNUNET_SERVICE_client_continue (tc->client);
  GNUNET_free (address);
}
 
 
static void
handle_del_address (void *cls,
                    const struct GNUNET_TRANSPORT_DelAddressMessage *dam)
{
  struct TransportClient *tc = cls;
  struct AddressListEntry *alen;
 
  if (CT_COMMUNICATOR != tc->type)
  {
    GNUNET_break (0);
    GNUNET_SERVICE_client_drop (tc->client);
    return;
  }
  for (struct AddressListEntry *ale = tc->details.communicator.addr_head;
       NULL != ale;
       ale = alen)
  {
    alen = ale->next;
    if (dam->aid != ale->aid)
      continue;
    GNUNET_assert (ale->tc == tc);
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Communicator deleted address `%s'!\n",
                ale->address);
    GNUNET_HELLO_builder_del_address (GST_my_hello,
                                      ale->address);
    if (NULL != pils_feed_task)
      GNUNET_SCHEDULER_cancel (pils_feed_task);
    pils_feed_task = GNUNET_SCHEDULER_add_delayed (PILS_FEED_ADDRESSES_DELAY,
                                                   &feed_addresses_to_pils,
                                                   NULL);
    free_address_list_entry (ale);
    GNUNET_SERVICE_client_continue (tc->client);
    return;
  }
  GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
              "Communicator removed address we did not even have.\n");
  GNUNET_SERVICE_client_continue (tc->client);
  // GNUNET_SERVICE_client_drop (tc->client);
}
 
 
static void
demultiplex_with_cmc (struct CommunicatorMessageContext *cmc);
 
 
static void
core_env_sent_cb (void *cls)
{
  struct CoreSentContext *ctx = cls;
  struct VirtualLink *vl = ctx->vl;
 
  if (NULL == vl)
  {
    /* lost the link in the meantime, ignore */
    GNUNET_free (ctx);
    return;
  }
  GNUNET_CONTAINER_DLL_remove (vl->csc_head, vl->csc_tail, ctx);
  GNUNET_assert (vl->incoming_fc_window_size_ram >= ctx->size);
  vl->incoming_fc_window_size_ram -= ctx->size;
  vl->incoming_fc_window_size_used += ctx->isize;
  consider_sending_fc (vl);
  GNUNET_free (ctx);
}
 
 
static void
finish_handling_raw_message (struct VirtualLink *vl,
                             const struct GNUNET_MessageHeader *mh,
                             struct CommunicatorMessageContext *cmc,
                             unsigned int free_cmc)
{
  uint16_t size = ntohs (mh->size);
  int have_core;
 
  if (vl->incoming_fc_window_size_ram > UINT_MAX - size)
  {
    GNUNET_STATISTICS_update (GST_stats,
                              "# CORE messages dropped (FC arithmetic overflow)",
                              1,
                              GNUNET_NO);
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "CORE messages of type %u with %u bytes dropped (FC arithmetic overflow)\n",
                (unsigned int) ntohs (mh->type),
                (unsigned int) ntohs (mh->size));
    if (GNUNET_YES == free_cmc)
      finish_cmc_handling_with_continue (cmc, GNUNET_YES);
    return;
  }
  if (vl->incoming_fc_window_size_ram + size > vl->available_fc_window_size)
  {
    GNUNET_STATISTICS_update (GST_stats,
                              "# CORE messages dropped (FC window overflow)",
                              1,
                              GNUNET_NO);
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "CORE messages of type %u with %u bytes dropped (FC window overflow)\n",
                (unsigned int) ntohs (mh->type),
                (unsigned int) ntohs (mh->size));
    if (GNUNET_YES == free_cmc)
      finish_cmc_handling_with_continue (cmc, GNUNET_YES);
    return;
  }
 
  /* Forward to all CORE clients */
  have_core = GNUNET_NO;
  for (struct TransportClient *tc = clients_head; NULL != tc; tc = tc->next)
  {
    struct GNUNET_MQ_Envelope *env;
    struct InboundMessage *im;
    struct CoreSentContext *ctx;
 
    if (CT_CORE != tc->type)
      continue;
    vl->incoming_fc_window_size_ram += size;
    env = GNUNET_MQ_msg_extra (im, size, GNUNET_MESSAGE_TYPE_TRANSPORT_RECV);
    ctx = GNUNET_new (struct CoreSentContext);
    ctx->vl = vl;
    ctx->size = size;
    ctx->isize = (GNUNET_NO == have_core) ? size : 0;
    have_core = GNUNET_YES;
    GNUNET_CONTAINER_DLL_insert (vl->csc_head, vl->csc_tail, ctx);
    GNUNET_MQ_notify_sent (env, &core_env_sent_cb, ctx);
    im->peer = cmc->im.sender;
    memcpy (&im[1], mh, size);
    GNUNET_MQ_send (tc->mq, env);
    vl->core_recv_window--;
  }
  if (GNUNET_NO == have_core)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
                "Dropped message to CORE: no CORE client connected!\n");
    /* Nevertheless, count window as used, as it is from the
       perspective of the other peer! */
    vl->incoming_fc_window_size_used += size;
    /* TODO-M1 */
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Dropped message of type %u with %u bytes to CORE: no CORE client connected!\n",
                (unsigned int) ntohs (mh->type),
                (unsigned int) ntohs (mh->size));
    if (GNUNET_YES == free_cmc)
      finish_cmc_handling_with_continue (cmc, GNUNET_YES);
    return;
  }
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Delivered message from %s of type %u to CORE recv window %d\n",
              GNUNET_i2s (&cmc->im.sender),
              ntohs (mh->type),
              vl->core_recv_window);
  if (vl->core_recv_window > 0)
  {
    if (GNUNET_YES == free_cmc)
      finish_cmc_handling_with_continue (cmc, GNUNET_YES);
    return;
  }
  /* Wait with calling #finish_cmc_handling(cmc) until the message
     was processed by CORE MQs (for CORE flow control)! */
  if (GNUNET_YES == free_cmc)
    GNUNET_CONTAINER_DLL_insert (vl->cmc_head, vl->cmc_tail, cmc);
}
 
 
static void
handle_raw_message (void *cls, const struct GNUNET_MessageHeader *mh)
{
  struct CommunicatorMessageContext *cmc = cls;
  // struct CommunicatorMessageContext *cmc_copy =
  // GNUNET_new (struct CommunicatorMessageContext);
  struct GNUNET_MessageHeader *mh_copy;
  struct RingBufferEntry *rbe;
  struct VirtualLink *vl;
  uint16_t size = ntohs (mh->size);
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Handling raw message of type %u with %u bytes\n",
              (unsigned int) ntohs (mh->type),
              (unsigned int) ntohs (mh->size));
 
  if ((size > UINT16_MAX - sizeof(struct InboundMessage)) ||
      (size < sizeof(struct GNUNET_MessageHeader)))
  {
    struct GNUNET_SERVICE_Client *client = cmc->tc->client;
 
    GNUNET_break (0);
    finish_cmc_handling (cmc);
    GNUNET_SERVICE_client_drop (client);
    return;
  }
  vl = lookup_virtual_link (&cmc->im.sender);
  if ((NULL == vl) || (GNUNET_NO == vl->confirmed))
  {
    /* FIXME: sender is giving us messages for CORE but we don't have
       the link up yet! I *suspect* this can happen right now (i.e.
       sender has verified us, but we didn't verify sender), but if
       we pass this on, CORE would be confused (link down, messages
       arrive).  We should investigate more if this happens often,
       or in a persistent manner, and possibly do "something" about
       it. Thus logging as error for now. */
 
    mh_copy = GNUNET_malloc (size);
    rbe = GNUNET_new (struct RingBufferEntry);
    rbe->cmc = cmc;
    /*cmc_copy->tc = cmc->tc;
       cmc_copy->im = cmc->im;*/
    GNUNET_memcpy (mh_copy, mh, size);
 
    rbe->mh = mh_copy;
 
    if (GNUNET_YES == is_ring_buffer_full)
    {
      struct RingBufferEntry *rbe_old = ring_buffer[ring_buffer_head];
      GNUNET_free (rbe_old->cmc);
      GNUNET_free (rbe_old->mh);
      GNUNET_free (rbe_old);
    }
    ring_buffer[ring_buffer_head] = rbe;// cmc_copy;
    // cmc_copy->mh = (const struct GNUNET_MessageHeader *) mh_copy;
    cmc->mh = (const struct GNUNET_MessageHeader *) mh_copy;
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Storing message for %s and type %u (%u) in ring buffer head %u is full %u\n",
                GNUNET_i2s (&cmc->im.sender),
                (unsigned int) ntohs (mh->type),
                (unsigned int) ntohs (mh_copy->type),
                ring_buffer_head,
                is_ring_buffer_full);
    if (RING_BUFFER_SIZE - 1 == ring_buffer_head)
    {
      ring_buffer_head = 0;
      is_ring_buffer_full = GNUNET_YES;
    }
    else
      ring_buffer_head++;
 
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "%u items stored in ring buffer\n",
                GNUNET_YES == is_ring_buffer_full ? RING_BUFFER_SIZE :
                ring_buffer_head);
 
    /*GNUNET_break_op (0);
    GNUNET_STATISTICS_update (GST_stats,
                              "# CORE messages dropped (virtual link still down)",
                              1,
                              GNUNET_NO);
 
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "CORE messages of type %u with %u bytes dropped (virtual link still down)\n",
                (unsigned int) ntohs (mh->type),
                (unsigned int) ntohs (mh->size));
    finish_cmc_handling (cmc);*/
    finish_cmc_handling_with_continue (cmc, GNUNET_NO);
    cmc->continue_send = GNUNET_YES;
    // GNUNET_free (cmc);
    return;
  }
  finish_handling_raw_message (vl, mh, cmc, GNUNET_YES);
}
 
 
static int
check_fragment_box (void *cls, const struct TransportFragmentBoxMessage *fb)
{
  uint16_t size = ntohs (fb->header.size);
  uint16_t bsize = size - sizeof(*fb);
 
  (void) cls;
  if (0 == bsize)
  {
    GNUNET_break_op (0);
    return GNUNET_SYSERR;
  }
  if (bsize + ntohs (fb->frag_off) > ntohs (fb->msg_size))
  {
    GNUNET_break_op (0);
    return GNUNET_SYSERR;
  }
  if (ntohs (fb->frag_off) >= ntohs (fb->msg_size))
  {
    GNUNET_break_op (0);
    return GNUNET_SYSERR;
  }
  return GNUNET_YES;
}
 
 
static void
destroy_ack_cummulator (void *cls)
{
  struct AcknowledgementCummulator *ac = cls;
 
  ac->task = NULL;
  GNUNET_assert (0 == ac->num_acks);
  GNUNET_assert (
    GNUNET_YES ==
    GNUNET_CONTAINER_multipeermap_remove (ack_cummulators, &ac->target, ac));
  GNUNET_free (ac);
}
 
 
static void
transmit_cummulative_ack_cb (void *cls)
{
  struct Neighbour *n;
  struct VirtualLink *vl;
  struct AcknowledgementCummulator *ac = cls;
  char buf[sizeof(struct TransportReliabilityAckMessage)
           + ac->num_acks
           * sizeof(struct TransportCummulativeAckPayloadP)] GNUNET_ALIGN;
  struct TransportReliabilityAckMessage *ack =
    (struct TransportReliabilityAckMessage *) buf;
  struct TransportCummulativeAckPayloadP *ap;
 
  ac->task = NULL;
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Sending ACK with %u components to %s\n",
              ac->num_acks,
              GNUNET_i2s (&ac->target));
  GNUNET_assert (0 < ac->num_acks);
  ack->header.type = htons (GNUNET_MESSAGE_TYPE_TRANSPORT_RELIABILITY_ACK);
  ack->header.size =
    htons (sizeof(*ack)
           + ac->num_acks * sizeof(struct TransportCummulativeAckPayloadP));
  ack->ack_counter = htonl (ac->ack_counter += ac->num_acks);
  ap = (struct TransportCummulativeAckPayloadP *) &ack[1];
  for (unsigned int i = 0; i < ac->num_acks; i++)
  {
    ap[i].ack_uuid = ac->ack_uuids[i].ack_uuid;
    ap[i].ack_delay = GNUNET_TIME_relative_hton (
      GNUNET_TIME_absolute_get_duration (ac->ack_uuids[i].receive_time));
  }
  /*route_control_message_without_fc (
    &ac->target,
    &ack->header,
    RMO_DV_ALLOWED);*/
  vl = lookup_virtual_link (&ac->target);
  if ((NULL != vl) && (GNUNET_YES == vl->confirmed))
  {
    route_control_message_without_fc (
      vl,
      &ack->header,
      RMO_DV_ALLOWED);
  }
  else
  {
    /* Use route via neighbour */
    n = lookup_neighbour (&ac->target);
    if (NULL != n)
      route_via_neighbour (
        n,
        &ack->header,
        RMO_NONE);
  }
  ac->num_acks = 0;
  ac->task = GNUNET_SCHEDULER_add_delayed (ACK_CUMMULATOR_TIMEOUT,
                                           &destroy_ack_cummulator,
                                           ac);
}
 
 
static void
cummulative_ack (const struct GNUNET_PeerIdentity *pid,
                 const struct AcknowledgementUUIDP *ack_uuid,
                 struct GNUNET_TIME_Absolute max_delay)
{
  struct AcknowledgementCummulator *ac;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Scheduling ACK %s for transmission to %s\n",
              GNUNET_uuid2s (&ack_uuid->value),
              GNUNET_i2s (pid));
  ac = GNUNET_CONTAINER_multipeermap_get (ack_cummulators, pid);
  if (NULL == ac)
  {
    ac = GNUNET_new (struct AcknowledgementCummulator);
    ac->target = *pid;
    ac->min_transmission_time = max_delay;
    GNUNET_assert (GNUNET_YES ==
                   GNUNET_CONTAINER_multipeermap_put (
                     ack_cummulators,
                     &ac->target,
                     ac,
                     GNUNET_CONTAINER_MULTIHASHMAPOPTION_UNIQUE_ONLY));
  }
  else
  {
    if (MAX_CUMMULATIVE_ACKS == ac->num_acks)
    {
      /* must run immediately, ack buffer full! */
      transmit_cummulative_ack_cb (ac);
    }
    GNUNET_SCHEDULER_cancel (ac->task);
    ac->min_transmission_time =
      GNUNET_TIME_absolute_min (ac->min_transmission_time, max_delay);
  }
  GNUNET_assert (ac->num_acks < MAX_CUMMULATIVE_ACKS);
  ac->ack_uuids[ac->num_acks].receive_time = GNUNET_TIME_absolute_get ();
  ac->ack_uuids[ac->num_acks].ack_uuid = *ack_uuid;
  ac->num_acks++;
  ac->task = GNUNET_SCHEDULER_add_at (ac->min_transmission_time,
                                      &transmit_cummulative_ack_cb,
                                      ac);
}
 
 
struct FindByMessageUuidContext
{
  struct MessageUUIDP message_uuid;
 
  struct ReassemblyContext *rc;
};
 
 
static int
find_by_message_uuid (void *cls, uint32_t key, void *value)
{
  struct FindByMessageUuidContext *fc = cls;
  struct ReassemblyContext *rc = value;
 
  (void) key;
  if (0 == GNUNET_memcmp (&fc->message_uuid, &rc->msg_uuid))
  {
    fc->rc = rc;
    return GNUNET_NO;
  }
  return GNUNET_YES;
}
 
 
static void
handle_fragment_box (void *cls, const struct TransportFragmentBoxMessage *fb)
{
  struct CommunicatorMessageContext *cmc = cls;
  struct VirtualLink *vl;
  struct ReassemblyContext *rc;
  const struct GNUNET_MessageHeader *msg;
  uint16_t msize;
  uint16_t fsize;
  uint16_t frag_off;
  char *target;
  struct GNUNET_TIME_Relative cdelay;
  struct FindByMessageUuidContext fc;
 
  vl = lookup_virtual_link (&cmc->im.sender);
  if ((NULL == vl) || (GNUNET_NO == vl->confirmed))
  {
    struct GNUNET_SERVICE_Client *client = cmc->tc->client;
 
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "No virtual link for %s to handle fragment\n",
                GNUNET_i2s (&cmc->im.sender));
    GNUNET_break (0);
    finish_cmc_handling (cmc);
    GNUNET_SERVICE_client_drop (client);
    return;
  }
  if (NULL == vl->reassembly_map)
  {
    vl->reassembly_map = GNUNET_CONTAINER_multihashmap32_create (8);
    vl->reassembly_heap =
      GNUNET_CONTAINER_heap_create (GNUNET_CONTAINER_HEAP_ORDER_MIN);
    vl->reassembly_timeout_task =
      GNUNET_SCHEDULER_add_delayed (REASSEMBLY_EXPIRATION,
                                    &reassembly_cleanup_task,
                                    vl);
  }
  msize = ntohs (fb->msg_size);
  fc.message_uuid = fb->msg_uuid;
  fc.rc = NULL;
  (void) GNUNET_CONTAINER_multihashmap32_get_multiple (vl->reassembly_map,
                                                       fb->msg_uuid.uuid,
                                                       &find_by_message_uuid,
                                                       &fc);
  fsize = ntohs (fb->header.size) - sizeof(*fb);
  if (NULL == (rc = fc.rc))
  {
    rc = GNUNET_malloc (sizeof(*rc) + msize    /* reassembly payload buffer */
                        + (msize + 7) / 8 * sizeof(uint8_t) /* bitfield */);
    rc->msg_uuid = fb->msg_uuid;
    rc->virtual_link = vl;
    rc->msg_size = msize;
    rc->reassembly_timeout =
      GNUNET_TIME_relative_to_absolute (REASSEMBLY_EXPIRATION);
    rc->last_frag = GNUNET_TIME_absolute_get ();
    rc->hn = GNUNET_CONTAINER_heap_insert (vl->reassembly_heap,
                                           rc,
                                           rc->reassembly_timeout.abs_value_us);
    GNUNET_assert (GNUNET_OK ==
                   GNUNET_CONTAINER_multihashmap32_put (
                     vl->reassembly_map,
                     rc->msg_uuid.uuid,
                     rc,
                     GNUNET_CONTAINER_MULTIHASHMAPOPTION_MULTIPLE));
    target = (char *) &rc[1];
    rc->bitfield = (uint8_t *) (target + rc->msg_size);
    if (fsize != rc->msg_size)
      rc->msg_missing = rc->msg_size;
    else
      rc->msg_missing = 0;
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Received fragment with size %u at offset %u/%u %u bytes missing from %s for NEW message %"
                PRIu64 "\n",
                fsize,
                ntohs (fb->frag_off),
                msize,
                rc->msg_missing,
                GNUNET_i2s (&cmc->im.sender),
                fb->msg_uuid.uuid);
  }
  else
  {
    target = (char *) &rc[1];
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Received fragment at offset %u/%u from %s for message %u\n",
                ntohs (fb->frag_off),
                msize,
                GNUNET_i2s (&cmc->im.sender),
                (unsigned int) fb->msg_uuid.uuid);
  }
  if (msize != rc->msg_size)
  {
    GNUNET_break (0);
    finish_cmc_handling (cmc);
    return;
  }
 
  /* reassemble */
  if (0 == fsize)
  {
    GNUNET_break (0);
    finish_cmc_handling (cmc);
    return;
  }
  frag_off = ntohs (fb->frag_off);
  if (frag_off + fsize > msize)
  {
    /* Fragment (plus fragment size) exceeds message size! */
    GNUNET_break_op (0);
    finish_cmc_handling (cmc);
    return;
  }
  memcpy (&target[frag_off], &fb[1], fsize);
  /* update bitfield and msg_missing */
  for (unsigned int i = frag_off; i < frag_off + fsize; i++)
  {
    if (0 == (rc->bitfield[i / 8] & (1 << (i % 8))))
    {
      rc->bitfield[i / 8] |= (1 << (i % 8));
      rc->msg_missing--;
    }
  }
 
  /* Compute cumulative ACK */
  cdelay = GNUNET_TIME_absolute_get_duration (rc->last_frag);
  cdelay = GNUNET_TIME_relative_multiply (cdelay, rc->msg_missing / fsize);
  if (0 == rc->msg_missing)
    cdelay = GNUNET_TIME_UNIT_ZERO;
  cummulative_ack (&cmc->im.sender,
                   &fb->ack_uuid,
                   GNUNET_TIME_relative_to_absolute (cdelay));
  rc->last_frag = GNUNET_TIME_absolute_get ();
  /* is reassembly complete? */
  if (0 != rc->msg_missing)
  {
    finish_cmc_handling (cmc);
    return;
  }
  /* reassembly is complete, verify result */
  msg = (const struct GNUNET_MessageHeader *) &rc[1];
  if (ntohs (msg->size) != rc->msg_size)
  {
    GNUNET_break (0);
    free_reassembly_context (rc);
    finish_cmc_handling (cmc);
    return;
  }
  /* successful reassembly */
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Fragment reassembly complete for message %u\n",
              (unsigned int) fb->msg_uuid.uuid);
  /* FIXME: check that the resulting msg is NOT a
     DV Box or Reliability Box, as that is NOT allowed! */
  cmc->mh = msg;
  demultiplex_with_cmc (cmc);
  /* FIXME-OPTIMIZE: really free here? Might be bad if fragments are still
     en-route and we forget that we finished this reassembly immediately!
     -> keep around until timeout?
     -> shorten timeout based on ACK? */
  free_reassembly_context (rc);
}
 
 
static int
check_reliability_box (void *cls,
                       const struct TransportReliabilityBoxMessage *rb)
{
  const struct GNUNET_MessageHeader *box =  (const struct
                                             GNUNET_MessageHeader *) &rb[1];
  (void) cls;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "check_send_msg with size %u: inner msg type %u and size %u (%lu %lu)\n",
              ntohs (rb->header.size),
              ntohs (box->type),
              ntohs (box->size),
              sizeof (struct TransportReliabilityBoxMessage),
              sizeof (struct GNUNET_MessageHeader));
  GNUNET_MQ_check_boxed_message (rb);
  return GNUNET_YES;
}
 
 
static void
handle_reliability_box (void *cls,
                        const struct TransportReliabilityBoxMessage *rb)
{
  struct CommunicatorMessageContext *cmc = cls;
  const struct GNUNET_MessageHeader *inbox =
    (const struct GNUNET_MessageHeader *) &rb[1];
  struct GNUNET_TIME_Relative rtt;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Received reliability box from %s with UUID %s of type %u\n",
              GNUNET_i2s (&cmc->im.sender),
              GNUNET_uuid2s (&rb->ack_uuid.value),
              (unsigned int) ntohs (inbox->type));
  rtt = GNUNET_TIME_UNIT_SECONDS; /* FIXME: should base this on "RTT", but we
                                     do not really have an RTT for the
                                   * incoming* queue (should we have
                                     the sender add it to the rb message?) */
  cummulative_ack (
    &cmc->im.sender,
    &rb->ack_uuid,
    (0 == ntohl (rb->ack_countdown))
    ? GNUNET_TIME_UNIT_ZERO_ABS
    : GNUNET_TIME_relative_to_absolute (
      GNUNET_TIME_relative_divide (rtt, 8 /* FIXME: magic constant */)));
  /* continue with inner message */
  /* FIXME: check that inbox is NOT a DV Box, fragment or another
     reliability box (not allowed!) */
  cmc->mh = inbox;
  demultiplex_with_cmc (cmc);
}
 
 
static void
update_pd_age (struct PerformanceData *pd, unsigned int age)
{
  unsigned int sage;
 
  if (age == pd->last_age)
    return; /* nothing to do */
  sage = GNUNET_MAX (pd->last_age, age - 2 * GOODPUT_AGING_SLOTS);
  for (unsigned int i = sage; i <= age - GOODPUT_AGING_SLOTS; i++)
  {
    struct TransmissionHistoryEntry *the = &pd->the[i % GOODPUT_AGING_SLOTS];
 
    the->bytes_sent = 0;
    the->bytes_received = 0;
  }
  pd->last_age = age;
}
 
 
static void
update_performance_data (struct PerformanceData *pd,
                         struct GNUNET_TIME_Relative rtt,
                         uint16_t bytes_transmitted_ok)
{
  uint64_t nval = rtt.rel_value_us;
  uint64_t oval = pd->aged_rtt.rel_value_us;
  unsigned int age = get_age ();
  struct TransmissionHistoryEntry *the = &pd->the[age % GOODPUT_AGING_SLOTS];
 
  if (oval == GNUNET_TIME_UNIT_FOREVER_REL.rel_value_us)
    pd->aged_rtt = rtt;
  else
    pd->aged_rtt.rel_value_us = (nval + 7 * oval) / 8;
  update_pd_age (pd, age);
  the->bytes_received += bytes_transmitted_ok;
}
 
 
static void
update_queue_performance (struct Queue *q,
                          struct GNUNET_TIME_Relative rtt,
                          uint16_t bytes_transmitted_ok)
{
  update_performance_data (&q->pd, rtt, bytes_transmitted_ok);
}
 
 
static void
update_dvh_performance (struct DistanceVectorHop *dvh,
                        struct GNUNET_TIME_Relative rtt,
                        uint16_t bytes_transmitted_ok)
{
  update_performance_data (&dvh->pd, rtt, bytes_transmitted_ok);
}
 
 
static void
completed_pending_message (struct PendingMessage *pm)
{
  struct PendingMessage *pos;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Complete transmission of message %" PRIu64 " %u\n",
              pm->logging_uuid,
              pm->pmt);
  switch (pm->pmt)
  {
  case PMT_CORE:
  case PMT_RELIABILITY_BOX:
    /* Full message sent, we are done */
    client_send_response (pm);
    return;
 
  case PMT_FRAGMENT_BOX:
    /* Fragment sent over reliable channel */
    pos = pm->frag_parent;
    GNUNET_CONTAINER_MDLL_remove (frag, pos->head_frag, pos->tail_frag, pm);
    free_pending_message (pm);
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "pos frag_off %lu pos bytes_msg %lu pmt %u parent %u\n",
                (unsigned long) pos->frag_off,
                (unsigned long) pos->bytes_msg,
                pos->pmt,
                NULL == pos->frag_parent ? 1 : 0);
    /* check if subtree is done */
    while ((NULL == pos->head_frag) && (pos->frag_off == (pos->bytes_msg
                                                          - sizeof(struct
                                                                   TransportFragmentBoxMessage)))
           &&
           (NULL != pos->frag_parent))
    {
      pm = pos;
      pos = pm->frag_parent;
      if ((NULL == pos) && (PMT_DV_BOX == pm->pmt))
      {
        client_send_response (pm);
        return;
      }
      else if (PMT_DV_BOX == pm->pmt)
      {
        client_send_response (pos);
        return;
      }
      GNUNET_CONTAINER_MDLL_remove (frag, pos->head_frag, pos->tail_frag, pm);
      free_pending_message (pm);
    }
 
    /* Was this the last applicable fragment? */
    if ((NULL == pos->head_frag) && (NULL == pos->frag_parent || PMT_DV_BOX ==
                                     pos->pmt) &&
        (pos->frag_off == pos->bytes_msg))
      client_send_response (pos);
    return;
 
  case PMT_DV_BOX:
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Completed transmission of message %" PRIu64 " (DV Box)\n",
                pm->logging_uuid);
    if (NULL != pm->frag_parent)
    {
      pos = pm->frag_parent;
      free_pending_message (pm);
      pos->bpm = NULL;
      client_send_response (pos);
    }
    else
      client_send_response (pm);
    return;
  }
}
 
 
static void
handle_acknowledged (struct PendingAcknowledgement *pa,
                     struct GNUNET_TIME_Relative ack_delay)
{
  struct GNUNET_TIME_Relative delay;
 
  delay = GNUNET_TIME_absolute_get_duration (pa->transmission_time);
  delay = GNUNET_TIME_relative_subtract (delay, ack_delay);
  if (NULL != pa->queue && 1 == pa->num_send)
    update_queue_performance (pa->queue, delay, pa->message_size);
  if (NULL != pa->dvh && 1 == pa->num_send)
    update_dvh_performance (pa->dvh, delay, pa->message_size);
  if (NULL != pa->pm)
    completed_pending_message (pa->pm);
  free_pending_acknowledgement (pa);
}
 
 
static int
check_reliability_ack (void *cls,
                       const struct TransportReliabilityAckMessage *ra)
{
  unsigned int n_acks;
 
  (void) cls;
  n_acks = (ntohs (ra->header.size) - sizeof(*ra))
           / sizeof(struct TransportCummulativeAckPayloadP);
  if (0 == n_acks)
  {
    GNUNET_break_op (0);
    return GNUNET_SYSERR;
  }
  if ((ntohs (ra->header.size) - sizeof(*ra)) !=
      n_acks * sizeof(struct TransportCummulativeAckPayloadP))
  {
    GNUNET_break_op (0);
    return GNUNET_SYSERR;
  }
  return GNUNET_OK;
}
 
 
static void
handle_reliability_ack (void *cls,
                        const struct TransportReliabilityAckMessage *ra)
{
  struct CommunicatorMessageContext *cmc = cls;
  const struct TransportCummulativeAckPayloadP *ack;
  unsigned int n_acks;
  uint32_t ack_counter;
 
  n_acks = (ntohs (ra->header.size) - sizeof(*ra))
           / sizeof(struct TransportCummulativeAckPayloadP);
  ack = (const struct TransportCummulativeAckPayloadP *) &ra[1];
  for (unsigned int i = 0; i < n_acks; i++)
  {
    struct PendingAcknowledgement *pa =
      GNUNET_CONTAINER_multiuuidmap_get (pending_acks, &ack[i].ack_uuid.value);
    if (NULL == pa)
    {
      GNUNET_log (GNUNET_ERROR_TYPE_INFO,
                  "Received ACK from %s with UUID %s which is unknown to us!\n",
                  GNUNET_i2s (&cmc->im.sender),
                  GNUNET_uuid2s (&ack[i].ack_uuid.value));
      GNUNET_STATISTICS_update (
        GST_stats,
        "# FRAGMENT_ACKS dropped, no matching pending message",
        1,
        GNUNET_NO);
      continue;
    }
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Received ACK from %s with UUID %s\n",
                GNUNET_i2s (&cmc->im.sender),
                GNUNET_uuid2s (&ack[i].ack_uuid.value));
    handle_acknowledged (pa, GNUNET_TIME_relative_ntoh (ack[i].ack_delay));
  }
 
  ack_counter = htonl (ra->ack_counter);
  (void) ack_counter;  /* silence compiler warning for now */
  // FIXME-OPTIMIZE: track ACK losses based on ack_counter somewhere!
  // (DV and/or Neighbour?)
  finish_cmc_handling (cmc);
}
 
 
static int
check_backchannel_encapsulation (
  void *cls,
  const struct TransportBackchannelEncapsulationMessage *be)
{
  uint16_t size = ntohs (be->header.size) - sizeof(*be);
  const struct GNUNET_MessageHeader *inbox =
    (const struct GNUNET_MessageHeader *) &be[1];
  const char *is;
  uint16_t isize;
 
  (void) cls;
  if (ntohs (inbox->size) >= size)
  {
    GNUNET_break_op (0);
    return GNUNET_SYSERR;
  }
  isize = ntohs (inbox->size);
  is = ((const char *) inbox) + isize;
  size -= isize;
  if ('\0' != is[size - 1])
  {
    GNUNET_break_op (0);
    return GNUNET_SYSERR;
  }
  return GNUNET_YES;
}
 
 
static void
handle_backchannel_encapsulation (
  void *cls,
  const struct TransportBackchannelEncapsulationMessage *be)
{
  struct CommunicatorMessageContext *cmc = cls;
  struct GNUNET_TRANSPORT_CommunicatorBackchannelIncoming *cbi;
  struct GNUNET_MQ_Envelope *env;
  struct TransportClient *tc;
  const struct GNUNET_MessageHeader *inbox =
    (const struct GNUNET_MessageHeader *) &be[1];
  uint16_t isize = ntohs (inbox->size);
  const char *target_communicator = ((const char *) inbox) + isize;
  char *sender;
  char *self;
 
  GNUNET_asprintf (&sender,
                   "%s",
                   GNUNET_i2s (&cmc->im.sender));
  GNUNET_asprintf (&self,
                   "%s",
                   GNUNET_i2s (GST_my_identity));
 
  /* Find client providing this communicator */
  for (tc = clients_head; NULL != tc; tc = tc->next)
    if ((CT_COMMUNICATOR == tc->type) &&
        (0 ==
         strcmp (tc->details.communicator.address_prefix, target_communicator)))
      break;
  if (NULL == tc)
  {
    char *stastr;
 
    GNUNET_asprintf (
      &stastr,
      "# Backchannel message dropped: target communicator `%s' unknown",
      target_communicator);
    GNUNET_STATISTICS_update (GST_stats, stastr, 1, GNUNET_NO);
    GNUNET_free (stastr);
    finish_cmc_handling (cmc);
    return;
  }
  /* Finally, deliver backchannel message to communicator */
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Delivering backchannel message from %s to %s of type %u to %s\n",
              sender,
              self,
              ntohs (inbox->type),
              target_communicator);
  env = GNUNET_MQ_msg_extra (
    cbi,
    isize,
    GNUNET_MESSAGE_TYPE_TRANSPORT_COMMUNICATOR_BACKCHANNEL_INCOMING);
  cbi->pid = cmc->im.sender;
  memcpy (&cbi[1], inbox, isize);
  GNUNET_MQ_send (tc->mq, env);
  finish_cmc_handling (cmc);
}
 
 
static void
path_cleanup_cb (void *cls)
{
  struct DistanceVector *dv = cls;
  struct DistanceVectorHop *pos;
 
  dv->timeout_task = NULL;
  while (NULL != (pos = dv->dv_head))
  {
    GNUNET_assert (dv == pos->dv);
    if (GNUNET_TIME_absolute_get_remaining (pos->timeout).rel_value_us > 0)
      break;
    free_distance_vector_hop (pos);
  }
  if (NULL == pos)
  {
    free_dv_route (dv);
    return;
  }
  dv->timeout_task =
    GNUNET_SCHEDULER_add_at (pos->timeout, &path_cleanup_cb, dv);
}
 
 
static void
send_msg_from_cache (struct VirtualLink *vl)
{
 
  const struct GNUNET_PeerIdentity target = vl->target;
 
 
  if ((GNUNET_YES == is_ring_buffer_full) || (0 < ring_buffer_head))
  {
    struct RingBufferEntry *ring_buffer_copy[RING_BUFFER_SIZE];
    unsigned int tail = GNUNET_YES == is_ring_buffer_full ? ring_buffer_head :
                        0;
    unsigned int head = GNUNET_YES == is_ring_buffer_full ? RING_BUFFER_SIZE :
                        ring_buffer_head;
    struct GNUNET_TRANSPORT_IncomingMessage im;
    struct CommunicatorMessageContext *cmc;
    struct RingBufferEntry *rbe;
    struct GNUNET_MessageHeader *mh;
 
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Sending from ring buffer, which has %u items\n",
                head);
 
    ring_buffer_head = 0;
    for (unsigned int i = 0; i < head; i++)
    {
      rbe = ring_buffer[(i + tail) % RING_BUFFER_SIZE];
      cmc = rbe->cmc;
      mh = rbe->mh;
 
      im = cmc->im;
      // mh = cmc->mh;
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "Sending message of type %u to ring buffer target %s using vl target %s index %u\n",
                  mh->type,
                  GNUNET_i2s (&im.sender),
                  GNUNET_i2s2 (&target),
                  (i + tail) % RING_BUFFER_SIZE);
      if (0 == GNUNET_memcmp (&target, &im.sender))
      {
        GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                    "Finish handling message of type %u and size %u\n",
                    (unsigned int) ntohs (mh->type),
                    (unsigned int) ntohs (mh->size));
        finish_handling_raw_message (vl, mh, cmc, GNUNET_NO);
        GNUNET_free (mh);
        GNUNET_free (rbe->cmc);
        GNUNET_free (rbe);
      }
      else
      {
        ring_buffer_copy[ring_buffer_head] = rbe;
        ring_buffer_head++;
      }
    }
 
    if ((GNUNET_YES == is_ring_buffer_full) && (RING_BUFFER_SIZE - 1 >
                                                ring_buffer_head))
    {
      is_ring_buffer_full = GNUNET_NO;
    }
 
    for (unsigned int i = 0; i < ring_buffer_head; i++)
    {
      ring_buffer[i] = ring_buffer_copy[i];
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "ring_buffer_copy[i]->mh->type for i %u %u\n",
                  i,
                  ring_buffer_copy[i]->mh->type);
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "ring_buffer[i]->mh->type for i %u %u\n",
                  i,
                  ring_buffer[i]->mh->type);
    }
 
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "%u items still in ring buffer\n",
                ring_buffer_head);
  }
 
  if ((GNUNET_YES == is_ring_buffer_dv_full) || (0 < ring_buffer_dv_head))
  {
    struct PendingMessage *ring_buffer_dv_copy[RING_BUFFER_SIZE];
    struct PendingMessage *pm;
    unsigned int tail = GNUNET_YES == is_ring_buffer_dv_full ?
                        ring_buffer_dv_head :
                        0;
    unsigned int head = GNUNET_YES == is_ring_buffer_dv_full ?
                        RING_BUFFER_SIZE :
                        ring_buffer_dv_head;
 
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Sending from ring buffer dv, which has %u items\n",
                head);
 
    ring_buffer_dv_head = 0;
    for (unsigned int i = 0; i < head; i++)
    {
      pm = ring_buffer_dv[(i + tail) % RING_BUFFER_SIZE];
 
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "Sending to ring buffer target %s using vl target %s\n",
                  GNUNET_i2s (&pm->target),
                  GNUNET_i2s2 (&target));
      if (0 == GNUNET_memcmp (&target, &pm->target))
      {
        GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                    "Adding PendingMessage to vl, checking transmission.\n");
        pm->vl = vl;
        GNUNET_CONTAINER_MDLL_insert (vl,
                                      vl->pending_msg_head,
                                      vl->pending_msg_tail,
                                      pm);
 
        check_vl_transmission (vl);
      }
      else
      {
        ring_buffer_dv_copy[ring_buffer_dv_head] = pm;
        ring_buffer_dv_head++;
      }
    }
 
    if (is_ring_buffer_dv_full && (RING_BUFFER_SIZE - 1 > ring_buffer_dv_head))
    {
      is_ring_buffer_dv_full = GNUNET_NO;
    }
 
    for (unsigned int i = 0; i < ring_buffer_dv_head; i++)
      ring_buffer_dv[i] = ring_buffer_dv_copy[i];
 
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "%u items still in ring buffer dv.\n",
                ring_buffer_dv_head);
 
  }
}
 
 
static void
activate_core_visible_dv_path (struct DistanceVectorHop *hop)
{
  struct DistanceVector *dv = hop->dv;
  struct VirtualLink *vl;
 
  vl = lookup_virtual_link (&dv->target);
  if (NULL == vl)
  {
 
    vl = GNUNET_new (struct VirtualLink);
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Creating new virtual link %p to %s using DV!\n",
                vl,
                GNUNET_i2s (&dv->target));
    vl->burst_addr = NULL;
    vl->confirmed = GNUNET_YES;
    vl->message_uuid_ctr =
      GNUNET_CRYPTO_random_u64 (GNUNET_CRYPTO_QUALITY_WEAK, UINT64_MAX);
    vl->target = dv->target;
    vl->core_recv_window = RECV_WINDOW_SIZE;
    vl->available_fc_window_size = DEFAULT_WINDOW_SIZE;
    vl->incoming_fc_window_size = DEFAULT_WINDOW_SIZE;
    GNUNET_break (GNUNET_YES ==
                  GNUNET_CONTAINER_multipeermap_put (
                    links,
                    &vl->target,
                    vl,
                    GNUNET_CONTAINER_MULTIHASHMAPOPTION_UNIQUE_ONLY));
    vl->dv = dv;
    dv->vl = vl;
    vl->visibility_task =
      GNUNET_SCHEDULER_add_at (hop->path_valid_until, &check_link_down, vl);
    consider_sending_fc (vl);
    /* We lacked a confirmed connection to the target
       before, so tell CORE about it (finally!) */
    cores_send_connect_info (&dv->target);
    send_msg_from_cache (vl);
  }
  else
  {
    /* Link was already up, remember dv is also now available and we are done */
    vl->dv = dv;
    dv->vl = vl;
    if (GNUNET_NO == vl->confirmed)
    {
      vl->confirmed = GNUNET_YES;
      vl->visibility_task =
        GNUNET_SCHEDULER_add_at (hop->path_valid_until, &check_link_down, vl);
      consider_sending_fc (vl);
      /* We lacked a confirmed connection to the target
         before, so tell CORE about it (finally!) */
      cores_send_connect_info (&dv->target);
      send_msg_from_cache (vl);
    }
    else
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "Virtual link to %s could now also use DV!\n",
                  GNUNET_i2s (&dv->target));
  }
}
 
 
static int
learn_dv_path (const struct GNUNET_PeerIdentity *path,
               unsigned int path_len,
               struct GNUNET_TIME_Relative network_latency,
               struct GNUNET_TIME_Absolute path_valid_until)
{
  struct DistanceVectorHop *hop;
  struct DistanceVector *dv;
  struct Neighbour *next_hop;
  unsigned int shorter_distance;
 
  if (path_len < 3)
  {
    /* what a boring path! not allowed! */
    GNUNET_break (0);
    return GNUNET_SYSERR;
  }
  GNUNET_assert (0 == GNUNET_memcmp (GST_my_identity, &path[0]));
  next_hop = lookup_neighbour (&path[1]);
  if (NULL == next_hop)
  {
    /* next hop must be a neighbour, otherwise this whole thing is useless! */
    GNUNET_break (0);
    return GNUNET_SYSERR;
  }
  for (unsigned int i = 2; i < path_len; i++)
  {
    struct Neighbour *n = lookup_neighbour (&path[i]);
    struct GNUNET_TIME_Absolute q_timeout;
 
    if (NULL != n)
    {
      q_timeout = GNUNET_TIME_UNIT_ZERO_ABS;
      for (struct Queue *q = n->queue_head; NULL != q; q = q->next_neighbour)
        q_timeout = GNUNET_TIME_absolute_max (q_timeout, q->validated_until);
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "remaining %lu to %s\n",
                  (unsigned long) GNUNET_TIME_absolute_get_remaining (q_timeout)
                  .rel_value_us,
                  GNUNET_i2s (&n->pid));
      if (0 != GNUNET_TIME_absolute_get_remaining (q_timeout).rel_value_us)
      {
        /* Useless path: we have a direct active connection to some hop
           in the middle of the path, so this one is not even
           terribly useful for redundancy */
        GNUNET_log (GNUNET_ERROR_TYPE_INFO,
                    "Path of %u hops useless: directly link to hop %u (%s)\n",
                    path_len,
                    i,
                    GNUNET_i2s (&path[i]));
        GNUNET_STATISTICS_update (GST_stats,
                                  "# Useless DV path ignored: hop is neighbour",
                                  1,
                                  GNUNET_NO);
        return GNUNET_SYSERR;
      }
    }
  }
  dv = GNUNET_CONTAINER_multipeermap_get (dv_routes, &path[path_len - 1]);
  if (NULL == dv)
  {
    dv = GNUNET_new (struct DistanceVector);
    dv->target = path[path_len - 1];
    dv->timeout_task = GNUNET_SCHEDULER_add_delayed (DV_PATH_VALIDITY_TIMEOUT,
                                                     &path_cleanup_cb,
                                                     dv);
    GNUNET_assert (GNUNET_OK ==
                   GNUNET_CONTAINER_multipeermap_put (
                     dv_routes,
                     &dv->target,
                     dv,
                     GNUNET_CONTAINER_MULTIHASHMAPOPTION_UNIQUE_ONLY));
  }
  /* Check if we have this path already! */
  shorter_distance = 0;
  for (struct DistanceVectorHop *pos = dv->dv_head; NULL != pos;
       pos = pos->next_dv)
  {
    if (pos->distance < path_len - 3)
      shorter_distance++;
    /* Note that the distances in 'pos' excludes us (path[0]),
       the next_hop (path[1]) and the target so we need to subtract three
       and check next_hop explicitly */
    if ((pos->distance == path_len - 3) && (pos->next_hop == next_hop))
    {
      int match = GNUNET_YES;
 
      for (unsigned int i = 0; i < pos->distance; i++)
      {
        if (0 != GNUNET_memcmp (&pos->path[i], &path[i + 2]))
        {
          match = GNUNET_NO;
          break;
        }
      }
      if (GNUNET_YES == match)
      {
        struct GNUNET_TIME_Relative last_timeout;
 
        /* Re-discovered known path, update timeout */
        GNUNET_STATISTICS_update (GST_stats,
                                  "# Known DV path refreshed",
                                  1,
                                  GNUNET_NO);
        last_timeout = GNUNET_TIME_absolute_get_remaining (pos->timeout);
        pos->timeout =
          GNUNET_TIME_relative_to_absolute (DV_PATH_VALIDITY_TIMEOUT);
        pos->path_valid_until =
          GNUNET_TIME_absolute_max (pos->path_valid_until, path_valid_until);
        GNUNET_CONTAINER_MDLL_remove (dv, dv->dv_head, dv->dv_tail, pos);
        GNUNET_CONTAINER_MDLL_insert (dv, dv->dv_head, dv->dv_tail, pos);
        if (0 <
            GNUNET_TIME_absolute_get_remaining (path_valid_until).rel_value_us)
          activate_core_visible_dv_path (pos);
        if (last_timeout.rel_value_us <
            GNUNET_TIME_relative_subtract (DV_PATH_VALIDITY_TIMEOUT,
                                           DV_PATH_DISCOVERY_FREQUENCY)
            .rel_value_us)
        {
          /* Some peer send DV learn messages too often, we are learning
             the same path faster than it would be useful; do not forward! */
          GNUNET_log (GNUNET_ERROR_TYPE_INFO,
                      "Rediscovered path too quickly, not forwarding further\n")
          ;
          return GNUNET_NO;
        }
        GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                    "Refreshed known path to %s valid until %s, forwarding further\n",
                    GNUNET_i2s (&dv->target),
                    GNUNET_STRINGS_absolute_time_to_string (
                      pos->path_valid_until));
        return GNUNET_YES;
      }
    }
  }
  /* Count how many shorter paths we have (incl. direct
     neighbours) before simply giving up on this one! */
  if (shorter_distance >= MAX_DV_PATHS_TO_TARGET)
  {
    /* We have a shorter path already! */
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Have many shorter DV paths %s, not forwarding further\n",
                GNUNET_i2s (&dv->target));
    return GNUNET_NO;
  }
  /* create new DV path entry */
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Discovered new DV path to %s valid until %s\n",
              GNUNET_i2s (&dv->target),
              GNUNET_STRINGS_absolute_time_to_string (path_valid_until));
  hop = GNUNET_malloc (sizeof(struct DistanceVectorHop)
                       + sizeof(struct GNUNET_PeerIdentity) * (path_len - 3));
  hop->next_hop = next_hop;
  hop->dv = dv;
  hop->path = (const struct GNUNET_PeerIdentity *) &hop[1];
  memcpy (&hop[1],
          &path[2],
          sizeof(struct GNUNET_PeerIdentity) * (path_len - 3));
  hop->timeout = GNUNET_TIME_relative_to_absolute (DV_PATH_VALIDITY_TIMEOUT);
  hop->path_valid_until = path_valid_until;
  hop->distance = path_len - 3;
  hop->pd.aged_rtt = network_latency;
  GNUNET_CONTAINER_MDLL_insert (dv, dv->dv_head, dv->dv_tail, hop);
  GNUNET_CONTAINER_MDLL_insert (neighbour,
                                next_hop->dv_head,
                                next_hop->dv_tail,
                                hop);
  if (0 < GNUNET_TIME_absolute_get_remaining (path_valid_until).rel_value_us)
    activate_core_visible_dv_path (hop);
  return GNUNET_YES;
}
 
 
static int
check_dv_learn (void *cls, const struct TransportDVLearnMessage *dvl)
{
  uint16_t size = ntohs (dvl->header.size);
  uint16_t num_hops = ntohs (dvl->num_hops);
  const struct DVPathEntryP *hops = (const struct DVPathEntryP *) &dvl[1];
 
  (void) cls;
  if (size != sizeof(*dvl) + num_hops * sizeof(struct DVPathEntryP))
  {
    GNUNET_break_op (0);
    return GNUNET_SYSERR;
  }
  if (num_hops > MAX_DV_HOPS_ALLOWED)
  {
    GNUNET_break_op (0);
    return GNUNET_SYSERR;
  }
  for (unsigned int i = 0; i < num_hops; i++)
  {
    if (0 == GNUNET_memcmp (&dvl->initiator, &hops[i].hop))
    {
      GNUNET_break_op (0);
      return GNUNET_SYSERR;
    }
    if (0 == GNUNET_memcmp (GST_my_identity, &hops[i].hop))
    {
      GNUNET_break_op (0);
      return GNUNET_SYSERR;
    }
  }
  return GNUNET_YES;
}
 
 
struct SignDhpCls
{
  struct DVPathEntryP *dhops;
  uint16_t nhops;
  const struct GNUNET_PeerIdentity *next_hop;
  struct TransportDVLearnMessage *fwd;
  struct PilsRequest *pr;
};
 
 
static void
sign_dhp_cp (void *cls,
             const struct GNUNET_PeerIdentity *pid,
             const struct GNUNET_CRYPTO_EddsaSignature *sig)
{
  struct SignDhpCls *sign_dhp_cls = cls;
  struct VirtualLink *vl;
  struct DVPathEntryP *dhops = sign_dhp_cls->dhops;
  uint16_t nhops = sign_dhp_cls->nhops;
  const struct GNUNET_PeerIdentity *next_hop = sign_dhp_cls->next_hop;
  struct TransportDVLearnMessage *fwd = sign_dhp_cls->fwd;
  struct Neighbour *n;
 
  sign_dhp_cls->pr->op = NULL;
  GNUNET_CONTAINER_DLL_remove (pils_requests_head,
                               pils_requests_tail,
                               sign_dhp_cls->pr);
  GNUNET_free (sign_dhp_cls->pr);
  dhops[nhops].hop_sig = *sig;
 
  /*route_control_message_without_fc (next_hop,
                                    &fwd->header,
                                    RMO_UNCONFIRMED_ALLOWED);*/
  vl = lookup_virtual_link (next_hop);
  if ((NULL != vl) && (GNUNET_YES == vl->confirmed))
  {
    route_control_message_without_fc (vl,
                                      &fwd->header,
                                      RMO_UNCONFIRMED_ALLOWED);
  }
  else
  {
    /* Use route via neighbour */
    n = lookup_neighbour (next_hop);
    if (NULL != n)
      route_via_neighbour (
        n,
        &fwd->header,
        RMO_UNCONFIRMED_ALLOWED);
  }
  GNUNET_free (sign_dhp_cls);
}
 
 
static void
forward_dv_learn (const struct GNUNET_PeerIdentity *next_hop,
                  const struct TransportDVLearnMessage *msg,
                  uint16_t bi_history,
                  uint16_t nhops,
                  const struct DVPathEntryP *hops,
                  struct GNUNET_TIME_Absolute in_time)
{
  struct DVPathEntryP *dhops;
  char buf[sizeof(struct TransportDVLearnMessage)
           + (nhops + 1) * sizeof(struct DVPathEntryP)] GNUNET_ALIGN;
  struct TransportDVLearnMessage *fwd = (struct TransportDVLearnMessage *) buf;
  struct GNUNET_TIME_Relative nnd;
 
  /* compute message for forwarding */
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Forwarding DV learn message originating from %s to %s\n",
              GNUNET_i2s (&msg->initiator),
              GNUNET_i2s2 (next_hop));
  GNUNET_assert (nhops < MAX_DV_HOPS_ALLOWED);
  fwd->header.type = htons (GNUNET_MESSAGE_TYPE_TRANSPORT_DV_LEARN);
  fwd->header.size = htons (sizeof(struct TransportDVLearnMessage)
                            + (nhops + 1) * sizeof(struct DVPathEntryP));
  fwd->num_hops = htons (nhops + 1);
  fwd->bidirectional = htons (bi_history);
  nnd = GNUNET_TIME_relative_add (GNUNET_TIME_absolute_get_duration (in_time),
                                  GNUNET_TIME_relative_ntoh (
                                    msg->non_network_delay));
  fwd->non_network_delay = GNUNET_TIME_relative_hton (nnd);
  fwd->init_sig = msg->init_sig;
  fwd->initiator = msg->initiator;
  fwd->challenge = msg->challenge;
  fwd->monotonic_time = msg->monotonic_time;
  dhops = (struct DVPathEntryP *) &fwd[1];
  GNUNET_memcpy (dhops, hops, sizeof(struct DVPathEntryP) * nhops);
  dhops[nhops].hop = *GST_my_identity;
  {
    struct DvHopPS dhp = {
      .purpose.purpose = htonl (GNUNET_SIGNATURE_PURPOSE_TRANSPORT_DV_HOP),
      .purpose.size = htonl (sizeof(dhp)),
      .pred = (0 == nhops) ? msg->initiator : dhops[nhops - 1].hop,
      .succ = *next_hop,
      .challenge = msg->challenge
    };
    struct SignDhpCls *sign_dhp_cls = GNUNET_new (struct SignDhpCls);
    sign_dhp_cls->dhops = dhops;
    sign_dhp_cls->nhops = nhops;
    sign_dhp_cls->next_hop = next_hop;
    sign_dhp_cls->fwd = fwd;
    sign_dhp_cls->pr = GNUNET_new (struct PilsRequest);
    GNUNET_CONTAINER_DLL_insert (pils_requests_head,
                                 pils_requests_tail,
                                 sign_dhp_cls->pr);
    sign_dhp_cls->pr->op =
      GNUNET_PILS_sign_by_peer_identity (pils,
                                         &dhp.purpose,
                                         sign_dhp_cp,
                                         sign_dhp_cls);
  }
}
 
 
static int
validate_dv_initiator_signature (
  struct GNUNET_TIME_AbsoluteNBO sender_monotonic_time,
  const struct GNUNET_PeerIdentity *init,
  const struct GNUNET_CRYPTO_ChallengeNonceP *challenge,
  const struct GNUNET_CRYPTO_EddsaSignature *init_sig)
{
  struct DvInitPS ip = { .purpose.purpose = htonl (
                           GNUNET_SIGNATURE_PURPOSE_TRANSPORT_DV_INITIATOR),
                         .purpose.size = htonl (sizeof(ip)),
                         .monotonic_time = sender_monotonic_time,
                         .challenge = *challenge };
 
  if (
    GNUNET_OK !=
    GNUNET_CRYPTO_eddsa_verify (GNUNET_SIGNATURE_PURPOSE_TRANSPORT_DV_INITIATOR,
                                &ip,
                                init_sig,
                                &init->public_key))
  {
    GNUNET_break_op (0);
    return GNUNET_SYSERR;
  }
  return GNUNET_OK;
}
 
 
struct NeighbourSelectionContext
{
  const struct TransportDVLearnMessage *dvl;
 
  const struct DVPathEntryP *hops;
 
  struct GNUNET_TIME_Absolute in_time;
 
  uint32_t selections[MAX_DV_DISCOVERY_SELECTION];
 
  unsigned int num_eligible;
 
  unsigned int num_selections;
 
  uint16_t nhops;
 
  uint16_t bi_history;
};
 
 
static int
dv_neighbour_selection (void *cls,
                        const struct GNUNET_PeerIdentity *pid,
                        void *value)
{
  struct NeighbourSelectionContext *nsc = cls;
 
  (void) value;
  if (0 == GNUNET_memcmp (pid, &nsc->dvl->initiator))
    return GNUNET_YES; /* skip initiator */
  for (unsigned int i = 0; i < nsc->nhops; i++)
    if (0 == GNUNET_memcmp (pid, &nsc->hops[i].hop))
      return GNUNET_YES;
  /* skip peers on path */
  nsc->num_eligible++;
  return GNUNET_YES;
}
 
 
static int
dv_neighbour_transmission (void *cls,
                           const struct GNUNET_PeerIdentity *pid,
                           void *value)
{
  struct NeighbourSelectionContext *nsc = cls;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "transmission %s\n",
              GNUNET_i2s (pid));
  (void) value;
  if (0 == GNUNET_memcmp (pid, &nsc->dvl->initiator))
    return GNUNET_YES; /* skip initiator */
  for (unsigned int i = 0; i < nsc->nhops; i++)
    if (0 == GNUNET_memcmp (pid, &nsc->hops[i].hop))
      return GNUNET_YES;
  /* skip peers on path */
  for (unsigned int i = 0; i < nsc->num_selections; i++)
  {
    if (nsc->selections[i] == nsc->num_eligible)
    {
      forward_dv_learn (pid,
                        nsc->dvl,
                        nsc->bi_history,
                        nsc->nhops,
                        nsc->hops,
                        nsc->in_time);
      break;
    }
  }
  nsc->num_eligible++;
  return GNUNET_YES;
}
 
 
static unsigned int
calculate_fork_degree (unsigned int hops_taken,
                       unsigned int neighbour_count,
                       unsigned int eligible_count)
{
  double target_total = 50.0; /* FIXME: use LOG(NSE)? */
  double eligible_ratio =
    ((double) eligible_count) / ((double) neighbour_count);
  double boost_factor = eligible_ratio * eligible_ratio;
  unsigned int rnd;
  double left;
 
  if (hops_taken >= 64)
  {
    GNUNET_break (0);
    return 0;   /* precaution given bitshift below */
  }
  for (unsigned int i = 1; i < hops_taken; i++)
  {
    /* For each hop, subtract the expected number of targets
       reached at distance d (so what remains divided by 2^d) */
    target_total -= (target_total * boost_factor / (1LLU << i));
  }
  rnd =
    (unsigned int) floor (target_total * boost_factor / (1LLU << hops_taken));
  /* round up or down probabilistically depending on how close we were
     when floor()ing to rnd */
  left = target_total - (double) rnd;
  if (UINT32_MAX * left >
      GNUNET_CRYPTO_random_u64 (GNUNET_CRYPTO_QUALITY_WEAK, UINT32_MAX))
    rnd++; /* round up */
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Forwarding DV learn message of %u hops %u(/%u/%u) times\n",
              hops_taken,
              rnd,
              eligible_count,
              neighbour_count);
  return rnd;
}
 
 
static void
neighbour_store_dvmono_cb (void *cls, int success)
{
  struct Neighbour *n = cls;
 
  n->sc = NULL;
  if (GNUNET_YES != success)
    GNUNET_log (GNUNET_ERROR_TYPE_ERROR,
                "Failed to store other peer's monotonic time in peerstore!\n");
}
 
 
static struct GNUNET_TIME_Relative
get_network_latency (const struct TransportDVLearnMessage *dvl)
{
  struct GNUNET_TIME_Relative host_latency_sum;
  struct GNUNET_TIME_Relative latency;
  struct GNUNET_TIME_Relative network_latency;
  uint16_t nhops = ntohs (dvl->num_hops);;
 
  /* We initiated this, learn the forward path! */
  host_latency_sum = GNUNET_TIME_relative_ntoh (dvl->non_network_delay);
 
  // Need also something to lookup initiation time
  // to compute RTT! -> add RTT argument here?
  latency = GNUNET_TIME_absolute_get_duration (GNUNET_TIME_absolute_ntoh (
                                                 dvl->monotonic_time));
  GNUNET_assert (latency.rel_value_us >= host_latency_sum.rel_value_us);
  // latency = GNUNET_TIME_UNIT_FOREVER_REL;   // FIXME: initialize properly
  // (based on dvl->challenge, we can identify time of origin!)
 
  network_latency = GNUNET_TIME_relative_subtract (latency, host_latency_sum);
  /* assumption: latency on all links is the same */
  network_latency = GNUNET_TIME_relative_divide (network_latency, nhops);
 
  return network_latency;
}
 
 
static void
handle_dv_learn (void *cls, const struct TransportDVLearnMessage *dvl)
{
  struct CommunicatorMessageContext *cmc = cls;
  enum GNUNET_TRANSPORT_CommunicatorCharacteristics cc;
  int bi_hop;
  uint16_t nhops;
  uint16_t bi_history;
  const struct DVPathEntryP *hops;
  int do_fwd;
  int did_initiator;
  struct GNUNET_TIME_Absolute in_time;
  struct Neighbour *n;
 
  nhops = ntohs (dvl->num_hops);  /* 0 = sender is initiator */
  bi_history = ntohs (dvl->bidirectional);
  hops = (const struct DVPathEntryP *) &dvl[1];
  if (0 == nhops)
  {
    /* sanity check */
    if (0 != GNUNET_memcmp (&dvl->initiator, &cmc->im.sender))
    {
      GNUNET_break (0);
      finish_cmc_handling (cmc);
      return;
    }
  }
  else
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "handle dv learn message last hop %s\n",
                GNUNET_i2s (&hops[nhops - 1].hop));
    /* sanity check */
    if (0 != GNUNET_memcmp (&hops[nhops - 1].hop, &cmc->im.sender))
    {
      GNUNET_break (0);
      finish_cmc_handling (cmc);
      return;
    }
  }
 
  GNUNET_assert (CT_COMMUNICATOR == cmc->tc->type);
  cc = cmc->tc->details.communicator.cc;
  bi_hop = (GNUNET_TRANSPORT_CC_RELIABLE ==
            cc); // FIXME: add bi-directional flag to cc?
  in_time = GNUNET_TIME_absolute_get ();
 
  /* continue communicator here, everything else can happen asynchronous! */
  finish_cmc_handling (cmc);
 
  n = lookup_neighbour (&dvl->initiator);
  if (NULL != n)
  {
    if ((n->dv_monotime_available == GNUNET_YES) &&
        (GNUNET_TIME_absolute_ntoh (dvl->monotonic_time).abs_value_us <
         n->last_dv_learn_monotime.abs_value_us))
    {
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "DV learn from %s discarded due to time travel",
                  GNUNET_i2s (&dvl->initiator));
      GNUNET_STATISTICS_update (GST_stats,
                                "# DV learn discarded due to time travel",
                                1,
                                GNUNET_NO);
      return;
    }
    if (GNUNET_OK != validate_dv_initiator_signature (dvl->monotonic_time,
                                                      &dvl->initiator,
                                                      &dvl->challenge,
                                                      &dvl->init_sig))
    {
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "DV learn signature from %s invalid\n",
                  GNUNET_i2s (&dvl->initiator));
      GNUNET_break_op (0);
      return;
    }
    n->last_dv_learn_monotime = GNUNET_TIME_absolute_ntoh (dvl->monotonic_time);
    if (GNUNET_YES == n->dv_monotime_available)
    {
      if (NULL != n->sc)
      {
        GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                    "store cancel\n");
        GNUNET_PEERSTORE_store_cancel (n->sc);
      }
      n->sc =
        GNUNET_PEERSTORE_store (peerstore,
                                "transport",
                                &dvl->initiator,
                                GNUNET_PEERSTORE_TRANSPORT_DVLEARN_MONOTIME,
                                &dvl->monotonic_time,
                                sizeof(dvl->monotonic_time),
                                GNUNET_TIME_UNIT_FOREVER_ABS,
                                GNUNET_PEERSTORE_STOREOPTION_REPLACE,
                                &neighbour_store_dvmono_cb,
                                n);
    }
  }
  /* OPTIMIZE-FIXME: asynchronously (!) verify signatures!,
     If signature verification load too high, implement random drop strategy */
  for (unsigned int i = 0; i < nhops; i++)
  {
    struct DvHopPS dhp = { .purpose.purpose =
                             htonl (GNUNET_SIGNATURE_PURPOSE_TRANSPORT_DV_HOP),
                           .purpose.size = htonl (sizeof(dhp)),
                           .pred = (0 == i) ? dvl->initiator : hops[i - 1].hop,
                           .succ = (nhops == i + 1) ? *GST_my_identity
                                   : hops[i + 1].hop,
                           .challenge = dvl->challenge };
 
    if (GNUNET_OK !=
        GNUNET_CRYPTO_eddsa_verify (GNUNET_SIGNATURE_PURPOSE_TRANSPORT_DV_HOP,
                                    &dhp,
                                    &hops[i].hop_sig,
                                    &hops[i].hop.public_key))
    {
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "DV learn from %s signature of hop %u invalid\n",
                  GNUNET_i2s (&dvl->initiator),
                  i);
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "signature of hop %s invalid\n",
                  GNUNET_i2s (&hops[i].hop));
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "pred %s\n",
                  GNUNET_i2s (&dhp.pred));
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "succ %s\n",
                  GNUNET_i2s (&dhp.succ));
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "hash %s\n",
                  GNUNET_sh2s (&dhp.challenge.value));
      GNUNET_break_op (0);
      return;
    }
  }
  if (GNUNET_EXTRA_LOGGING > 0)
  {
    char *path;
 
    path = GNUNET_strdup (GNUNET_i2s (&dvl->initiator));
    for (unsigned int i = 0; i < nhops; i++)
    {
      char *tmp;
 
      GNUNET_asprintf (&tmp,
                       "%s%s%s",
                       path,
                       (bi_history & (1 << (nhops - i))) ? "<->" : "-->",
                       GNUNET_i2s (&hops[i].hop));
      GNUNET_free (path);
      path = tmp;
    }
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Received DVInit via %s%s%s\n",
                path,
                bi_hop ? "<->" : "-->",
                GNUNET_i2s (GST_my_identity));
    GNUNET_free (path);
  }
  do_fwd = GNUNET_YES;
  if (0 == GNUNET_memcmp (GST_my_identity, &dvl->initiator))
  {
    struct GNUNET_PeerIdentity path[nhops + 1];
    struct GNUNET_TIME_Relative network_latency;
 
    /* We initiated this, learn the forward path! */
    path[0] = *GST_my_identity;
    path[1] = hops[0].hop;
 
    network_latency = get_network_latency (dvl);
 
    for (unsigned int i = 2; i <= nhops; i++)
    {
      struct GNUNET_TIME_Relative ilat;
 
      /* assumption: linear latency increase per hop */
      ilat = GNUNET_TIME_relative_multiply (network_latency, i);
      path[i] = hops[i - 1].hop;
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "Learned path with %u hops to %s with latency %s\n",
                  i,
                  GNUNET_i2s (&path[i]),
                  GNUNET_STRINGS_relative_time_to_string (ilat, GNUNET_YES));
      learn_dv_path (path,
                     i + 1,
                     ilat,
                     GNUNET_TIME_relative_to_absolute (
                       ADDRESS_VALIDATION_LIFETIME));
    }
    /* as we initiated, do not forward again (would be circular!) */
    do_fwd = GNUNET_NO;
    return;
  }
  if (bi_hop)
  {
    /* last hop was bi-directional, we could learn something here! */
    struct GNUNET_PeerIdentity path[nhops + 2];
    struct GNUNET_TIME_Relative ilat;
    struct GNUNET_TIME_Relative network_latency;
 
    path[0] = *GST_my_identity;
    path[1] = hops[nhops - 1].hop;   /* direct neighbour == predecessor! */
    for (unsigned int i = 0; i < nhops; i++)
    {
      int iret;
 
      if (0 == (bi_history & (1 << i)))
        break;     /* i-th hop not bi-directional, stop learning! */
      if (i == nhops - 1)
      {
        path[i + 2] = dvl->initiator;
      }
      else
      {
        path[i + 2] = hops[nhops - i - 2].hop;
      }
 
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "Learned inverse path with %u hops to %s\n",
                  i + 2,
                  GNUNET_i2s (&path[i + 2]));
      network_latency = get_network_latency (dvl);
      ilat = GNUNET_TIME_relative_multiply (network_latency, i + 2);
      iret = learn_dv_path (path,
                            i + 3,
                            ilat,
                            GNUNET_TIME_relative_to_absolute (
                              ADDRESS_VALIDATION_LIFETIME));
      if (GNUNET_SYSERR == iret)
      {
        /* path invalid or too long to be interesting for US, thus should also
           not be interesting to our neighbours, cut path when forwarding to
           'i' hops, except of course for the one that goes back to the
           initiator */
        GNUNET_STATISTICS_update (GST_stats,
                                  "# DV learn not forwarded due invalidity of path",
                                  1,
                                  GNUNET_NO);
        do_fwd = GNUNET_NO;
        break;
      }
      if ((GNUNET_NO == iret) && (nhops == i + 1))
      {
        /* we have better paths, and this is the longest target,
           so there cannot be anything interesting later */
        GNUNET_STATISTICS_update (GST_stats,
                                  "# DV learn not forwarded, got better paths",
                                  1,
                                  GNUNET_NO);
        do_fwd = GNUNET_NO;
        break;
      }
    }
  }
  if (MAX_DV_HOPS_ALLOWED == nhops)
  {
    /* At limit, we're out of here! */
    return;
  }
 
  /* Forward to initiator, if path non-trivial and possible */
  bi_history = (bi_history << 1) | (bi_hop ? 1 : 0);
  did_initiator = GNUNET_NO;
  if ((1 <= nhops) &&
      (GNUNET_YES ==
       GNUNET_CONTAINER_multipeermap_contains (neighbours, &dvl->initiator)))
  {
    /* send back to origin! */
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Sending DVL back to initiator %s\n",
                GNUNET_i2s (&dvl->initiator));
    forward_dv_learn (&dvl->initiator, dvl, bi_history, nhops, hops, in_time);
    did_initiator = GNUNET_YES;
  }
  /* We forward under two conditions: either we still learned something
     ourselves (do_fwd), or the path was darn short and thus the initiator is
     likely to still be very interested in this (and we did NOT already
     send it back to the initiator) */
  if ((do_fwd) || ((nhops < MIN_DV_PATH_LENGTH_FOR_INITIATOR) &&
                   (GNUNET_NO == did_initiator)))
  {
    /* Pick random neighbours that are not yet on the path */
    struct NeighbourSelectionContext nsc;
    unsigned int n_cnt;
 
    n_cnt = GNUNET_CONTAINER_multipeermap_size (neighbours);
    nsc.nhops = nhops;
    nsc.dvl = dvl;
    nsc.bi_history = bi_history;
    nsc.hops = hops;
    nsc.in_time = in_time;
    nsc.num_eligible = 0;
    GNUNET_CONTAINER_multipeermap_iterate (neighbours,
                                           &dv_neighbour_selection,
                                           &nsc);
    if (0 == nsc.num_eligible)
      return;   /* done here, cannot forward to anyone else */
    nsc.num_selections = calculate_fork_degree (nhops, n_cnt, nsc.num_eligible);
    nsc.num_selections =
      GNUNET_MIN (MAX_DV_DISCOVERY_SELECTION, nsc.num_selections);
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Forwarding DVL to %u other peers\n",
                nsc.num_selections);
    for (unsigned int i = 0; i < nsc.num_selections; i++)
      nsc.selections[i] =
        (nsc.num_selections == n_cnt)
        ? i   /* all were selected, avoid collisions by chance */
        : GNUNET_CRYPTO_random_u32 (GNUNET_CRYPTO_QUALITY_WEAK, n_cnt);
    nsc.num_eligible = 0;
    GNUNET_CONTAINER_multipeermap_iterate (neighbours,
                                           &dv_neighbour_transmission,
                                           &nsc);
  }
}
 
 
static int
check_dv_box (void *cls, const struct TransportDVBoxMessage *dvb)
{
  uint16_t size = ntohs (dvb->header.size);
  uint16_t num_hops = ntohs (dvb->num_hops);
  const struct GNUNET_PeerIdentity *hops =
    (const struct GNUNET_PeerIdentity *) &dvb[1];
 
  (void) cls;
  if (size < sizeof(*dvb) + num_hops * sizeof(struct GNUNET_PeerIdentity)
      + sizeof(struct GNUNET_MessageHeader))
  {
    GNUNET_break_op (0);
    return GNUNET_SYSERR;
  }
  /* This peer must not be on the path */
  for (unsigned int i = 0; i < num_hops; i++)
    if (0 == GNUNET_memcmp (&hops[i], GST_my_identity))
    {
      GNUNET_break_op (0);
      return GNUNET_SYSERR;
    }
  return GNUNET_YES;
}
 
 
static void
forward_dv_box (struct Neighbour *next_hop,
                struct TransportDVBoxMessage *hdr,
                uint16_t total_hops,
                uint16_t num_hops,
                const struct GNUNET_PeerIdentity *hops,
                const void *enc_payload,
                uint16_t enc_payload_size)
{
  struct VirtualLink *vl = next_hop->vl;
  struct PendingMessage *pm;
  size_t msg_size = sizeof(struct TransportDVBoxMessage)
                    + num_hops * sizeof(struct GNUNET_PeerIdentity)
                    + enc_payload_size;
  char *buf;
  char msg_buf[msg_size] GNUNET_ALIGN;
  struct GNUNET_PeerIdentity *dhops;
 
  hdr->num_hops = htons (num_hops);
  hdr->total_hops = htons (total_hops);
  hdr->header.size = htons (msg_size);
  memcpy (msg_buf, hdr, sizeof(*hdr));
  dhops = (struct GNUNET_PeerIdentity *) &msg_buf[sizeof(struct
                                                         TransportDVBoxMessage)]
  ;
  memcpy (dhops, hops, num_hops * sizeof(struct GNUNET_PeerIdentity));
  memcpy (&dhops[num_hops], enc_payload, enc_payload_size);
 
  if (GNUNET_YES == ntohs (hdr->without_fc))
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Forwarding control message (payload size %u) in DV Box to next hop %s (%u/%u) \n",
                enc_payload_size,
                GNUNET_i2s (&next_hop->pid),
                (unsigned int) num_hops,
                (unsigned int) total_hops);
    route_via_neighbour (next_hop, (const struct
                                    GNUNET_MessageHeader *) msg_buf,
                         RMO_ANYTHING_GOES);
  }
  else
  {
    pm = GNUNET_malloc (sizeof(struct PendingMessage) + msg_size);
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "2 created pm %p storing vl %p \n",
                pm,
                vl);
    pm->pmt = PMT_DV_BOX;
    pm->vl = vl;
    pm->target = next_hop->pid;
    pm->timeout = GNUNET_TIME_relative_to_absolute (DV_FORWARD_TIMEOUT);
    pm->logging_uuid = logging_uuid_gen++;
    pm->prefs = GNUNET_MQ_PRIO_BACKGROUND;
    pm->bytes_msg = msg_size;
    buf = (char *) &pm[1];
    memcpy (buf, msg_buf, msg_size);
 
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Created pending message %" PRIu64
                " for DV Box with next hop %s (%u/%u)\n",
                pm->logging_uuid,
                GNUNET_i2s (&next_hop->pid),
                (unsigned int) num_hops,
                (unsigned int) total_hops);
 
    if ((NULL != vl) && (GNUNET_YES == vl->confirmed))
    {
      GNUNET_CONTAINER_MDLL_insert (vl,
                                    vl->pending_msg_head,
                                    vl->pending_msg_tail,
                                    pm);
 
      check_vl_transmission (vl);
    }
    else
    {
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "The virtual link is not ready for forwarding a DV Box with payload, storing PendingMessage in ring buffer.\n");
 
      if (NULL != ring_buffer_dv[ring_buffer_dv_head])
      {
        struct PendingMessage *pm_old = ring_buffer_dv[ring_buffer_dv_head];
 
        GNUNET_free (pm_old);
      }
      ring_buffer_dv[ring_buffer_dv_head] = pm;
      if (RING_BUFFER_SIZE - 1 == ring_buffer_dv_head)
      {
        ring_buffer_dv_head = 0;
        is_ring_buffer_dv_full = GNUNET_YES;
      }
      else
        ring_buffer_dv_head++;
 
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "%u items stored in DV ring buffer\n",
                  GNUNET_YES == is_ring_buffer_dv_full ? RING_BUFFER_SIZE :
                  ring_buffer_dv_head);
    }
  }
}
 
 
static void
free_backtalker (struct Backtalker *b)
{
  if (NULL != b->get)
  {
    GNUNET_PEERSTORE_iteration_stop (b->get);
    b->get = NULL;
    GNUNET_assert (NULL != b->cmc);
    finish_cmc_handling (b->cmc);
    b->cmc = NULL;
  }
  if (NULL != b->task)
  {
    GNUNET_SCHEDULER_cancel (b->task);
    b->task = NULL;
  }
  if (NULL != b->sc)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "store cancel\n");
    GNUNET_PEERSTORE_store_cancel (b->sc);
    b->sc = NULL;
  }
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Removing backtalker for %s\n",
              GNUNET_i2s (&b->pid));
  GNUNET_assert (
    GNUNET_YES ==
    GNUNET_CONTAINER_multipeermap_remove (backtalkers, &b->pid, b));
  GNUNET_free (b);
}
 
 
static int
free_backtalker_cb (void *cls,
                    const struct GNUNET_PeerIdentity *pid,
                    void *value)
{
  struct Backtalker *b = value;
 
  (void) cls;
  (void) pid;
  free_backtalker (b);
  return GNUNET_OK;
}
 
 
static void
backtalker_timeout_cb (void *cls)
{
  struct Backtalker *b = cls;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "backtalker timeout.\n");
  b->task = NULL;
  if (0 != GNUNET_TIME_absolute_get_remaining (b->timeout).rel_value_us)
  {
    b->task = GNUNET_SCHEDULER_add_at (b->timeout, &backtalker_timeout_cb, b);
    return;
  }
  GNUNET_assert (NULL == b->sc);
  free_backtalker (b);
}
 
 
static void
backtalker_monotime_cb (void *cls,
                        const struct GNUNET_PEERSTORE_Record *record,
                        const char *emsg)
{
  struct Backtalker *b = cls;
  struct GNUNET_TIME_AbsoluteNBO *mtbe;
  struct GNUNET_TIME_Absolute mt;
 
  (void) emsg;
  if (NULL == record)
  {
    /* we're done with #backtalker_monotime_cb() invocations,
       continue normal processing */
    b->get = NULL;
    GNUNET_assert (NULL != b->cmc);
    b->cmc->mh = (const struct GNUNET_MessageHeader *) &b[1];
    if (0 != b->body_size)
      demultiplex_with_cmc (b->cmc);
    else
      finish_cmc_handling (b->cmc);
    b->cmc = NULL;
    return;
  }
  if (sizeof(*mtbe) != record->value_size)
  {
    GNUNET_PEERSTORE_iteration_next (b->get, 1);
    GNUNET_break (0);
    return;
  }
  mtbe = record->value;
  mt = GNUNET_TIME_absolute_ntoh (*mtbe);
  if (mt.abs_value_us > b->monotonic_time.abs_value_us)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Backtalker message from %s dropped, monotime in the past\n",
                GNUNET_i2s (&b->pid));
    GNUNET_STATISTICS_update (
      GST_stats,
      "# Backchannel messages dropped: monotonic time not increasing",
      1,
      GNUNET_NO);
    b->monotonic_time = mt;
    /* Setting body_size to 0 prevents call to #forward_backchannel_payload()
     */
    b->body_size = 0;
  }
  GNUNET_PEERSTORE_iteration_next (b->get, 1);
}
 
 
static void
backtalker_monotime_store_cb (void *cls, int success)
{
  struct Backtalker *b = cls;
 
  if (GNUNET_OK != success)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_ERROR,
                "Failed to store backtalker's monotonic time in PEERSTORE!\n");
  }
  b->sc = NULL;
  if (NULL != b->task)
  {
    GNUNET_SCHEDULER_cancel (b->task);
    b->task = NULL;
  }
  b->task = GNUNET_SCHEDULER_add_at (b->timeout, &backtalker_timeout_cb, b);
}
 
 
static void
update_backtalker_monotime (struct Backtalker *b)
{
  struct GNUNET_TIME_AbsoluteNBO mtbe;
 
  if (NULL != b->sc)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "store cancel before store with sc %p\n",
                b->sc);
    /*GNUNET_PEERSTORE_store_cancel (b->sc);
      b->sc = NULL;*/
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "store cancel before store with sc %p is null\n",
                b->sc);
  }
  else
  {
    GNUNET_SCHEDULER_cancel (b->task);
    b->task = NULL;
  }
  mtbe = GNUNET_TIME_absolute_hton (b->monotonic_time);
  b->sc =
    GNUNET_PEERSTORE_store (peerstore,
                            "transport",
                            &b->pid,
                            GNUNET_PEERSTORE_TRANSPORT_BACKCHANNEL_MONOTIME,
                            &mtbe,
                            sizeof(mtbe),
                            GNUNET_TIME_UNIT_FOREVER_ABS,
                            GNUNET_PEERSTORE_STOREOPTION_REPLACE,
                            &backtalker_monotime_store_cb,
                            b);
}
 
 
struct DecapsDvBoxCls
{
  struct CommunicatorMessageContext *cmc;
  const struct TransportDVBoxMessage *dvb;
  struct PilsRequest *pr;
};
 
 
static void
decaps_dv_box_cb (void *cls, const struct GNUNET_ShortHashCode *km)
{
  struct DecapsDvBoxCls *decaps_dv_box_cls = cls;
  struct CommunicatorMessageContext *cmc = decaps_dv_box_cls->cmc;
  const struct TransportDVBoxMessage *dvb = dvb;
  struct DVKeyState key;
  const char *hdr;
  size_t hdr_len;
  struct GNUNET_HashCode hmac;
 
  decaps_dv_box_cls->pr->op = NULL;
  GNUNET_CONTAINER_DLL_remove (pils_requests_head,
                               pils_requests_tail,
                               decaps_dv_box_cls->pr);
  GNUNET_free (decaps_dv_box_cls->pr);
  if (NULL == km)
  {
    GNUNET_break_op (0);
    finish_cmc_handling (cmc);
    return;
  }
  dv_setup_key_state_from_km (km, &dvb->iv, &key);
  hdr = (const char *) &dvb[1];
  hdr_len = ntohs (dvb->orig_size) - sizeof(*dvb) - sizeof(struct
                                                           GNUNET_PeerIdentity)
            * ntohs (dvb->total_hops);
 
  dv_hmac (&key, &hmac, hdr, hdr_len);
  if (0 != GNUNET_memcmp (&hmac, &dvb->hmac))
  {
    /* HMAC mismatch, discard! */
    GNUNET_break_op (0);
    finish_cmc_handling (cmc);
    return;
  }
  /* begin actual decryption */
  {
    struct Backtalker *b;
    struct GNUNET_TIME_Absolute monotime;
    struct TransportDVBoxPayloadP ppay;
    char body[hdr_len - sizeof(ppay)] GNUNET_ALIGN;
    const struct GNUNET_MessageHeader *mh;
 
    GNUNET_assert (hdr_len >=
                   sizeof(ppay) + sizeof(struct GNUNET_MessageHeader));
    if (GNUNET_OK != dv_decrypt (&key, &ppay, hdr, sizeof(ppay)))
    {
      GNUNET_log (GNUNET_ERROR_TYPE_ERROR,
                  "Error decrypting DV payload header\n");
      GNUNET_break_op (0);
      finish_cmc_handling (cmc);
      return;
    }
    if (GNUNET_OK != dv_decrypt (&key, body,
                                 &hdr[sizeof(ppay)], hdr_len - sizeof(ppay)))
    {
      GNUNET_log (GNUNET_ERROR_TYPE_ERROR,
                  "Error decrypting DV payload\n");
      GNUNET_break_op (0);
      finish_cmc_handling (cmc);
      return;
    }
    mh = (const struct GNUNET_MessageHeader *) body;
    dv_key_clean (&key);
    if (ntohs (mh->size) != sizeof(body))
    {
      GNUNET_break_op (0);
      finish_cmc_handling (cmc);
      return;
    }
    /* need to prevent box-in-a-box (and DV_LEARN) so check inbox type! */
    switch (ntohs (mh->type))
    {
    case GNUNET_MESSAGE_TYPE_TRANSPORT_DV_BOX:
      GNUNET_break_op (0);
      finish_cmc_handling (cmc);
      return;
 
    case GNUNET_MESSAGE_TYPE_TRANSPORT_DV_LEARN:
      GNUNET_break_op (0);
      finish_cmc_handling (cmc);
      return;
 
    default:
      /* permitted, continue */
      break;
    }
    monotime = GNUNET_TIME_absolute_ntoh (ppay.monotonic_time);
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Decrypted backtalk from %s\n",
                GNUNET_i2s (&ppay.sender));
    b = GNUNET_CONTAINER_multipeermap_get (backtalkers, &ppay.sender);
    if ((NULL != b) && (monotime.abs_value_us < b->monotonic_time.abs_value_us))
    {
      GNUNET_STATISTICS_update (
        GST_stats,
        "# Backchannel messages dropped: monotonic time not increasing",
        1,
        GNUNET_NO);
      finish_cmc_handling (cmc);
      return;
    }
    if ((NULL == b) ||
        (0 != GNUNET_memcmp (&b->last_ephemeral, &dvb->ephemeral_key)))
    {
      /* Check signature */
      struct EphemeralConfirmationPS ec;
 
      ec.purpose.purpose = htonl (GNUNET_SIGNATURE_PURPOSE_TRANSPORT_EPHEMERAL);
      ec.target = *GST_my_identity;
      ec.ephemeral_key = dvb->ephemeral_key;
      ec.purpose.size =  htonl (sizeof(ec));
      ec.sender_monotonic_time = ppay.monotonic_time;
      if (
        GNUNET_OK !=
        GNUNET_CRYPTO_eddsa_verify (
          GNUNET_SIGNATURE_PURPOSE_TRANSPORT_EPHEMERAL,
          &ec,
          &ppay.sender_sig,
          &ppay.sender.public_key))
      {
        /* Signature invalid, discard! */
        GNUNET_break_op (0);
        finish_cmc_handling (cmc);
        return;
      }
    }
    /* Update sender, we now know the real origin! */
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "DVBox received for me from %s via %s\n",
                GNUNET_i2s2 (&ppay.sender),
                GNUNET_i2s (&cmc->im.sender));
    cmc->im.sender = ppay.sender;
 
    if (NULL != b)
    {
      /* update key cache and mono time */
      b->last_ephemeral = dvb->ephemeral_key;
      b->monotonic_time = monotime;
      update_backtalker_monotime (b);
      b->timeout =
        GNUNET_TIME_relative_to_absolute (BACKCHANNEL_INACTIVITY_TIMEOUT);
      cmc->mh = mh;
      demultiplex_with_cmc (cmc);
      return;
    }
    /* setup data structure to cache signature AND check
       monotonic time with PEERSTORE before forwarding backchannel payload */
    b = GNUNET_malloc (sizeof(struct Backtalker) + sizeof(body));
    b->pid = ppay.sender;
    b->body_size = sizeof(body);
    memcpy (&b[1], body, sizeof(body));
    GNUNET_assert (GNUNET_YES ==
                   GNUNET_CONTAINER_multipeermap_put (
                     backtalkers,
                     &b->pid,
                     b,
                     GNUNET_CONTAINER_MULTIHASHMAPOPTION_UNIQUE_ONLY));
    b->monotonic_time = monotime; /* NOTE: to be checked still! */
    b->cmc = cmc;
    b->timeout =
      GNUNET_TIME_relative_to_absolute (BACKCHANNEL_INACTIVITY_TIMEOUT);
    b->task = GNUNET_SCHEDULER_add_at (b->timeout, &backtalker_timeout_cb, b);
    b->get =
      GNUNET_PEERSTORE_iteration_start (peerstore,
                                        "transport",
                                        &b->pid,
                                        GNUNET_PEERSTORE_TRANSPORT_BACKCHANNEL_MONOTIME,
                                        &backtalker_monotime_cb,
                                        b);
  } /* end actual decryption */
}
 
 
static void
handle_dv_box (void *cls, const struct TransportDVBoxMessage *dvb)
{
  struct CommunicatorMessageContext *cmc = cls;
  uint16_t size = ntohs (dvb->header.size) - sizeof(*dvb);
  uint16_t num_hops = ntohs (dvb->num_hops);
  const struct GNUNET_PeerIdentity *hops =
    (const struct GNUNET_PeerIdentity *) &dvb[1];
  const char *enc_payload = (const char *) &hops[num_hops];
  uint16_t enc_payload_size =
    size - (num_hops * sizeof(struct GNUNET_PeerIdentity));
  struct DecapsDvBoxCls *decaps_dv_box_cls;
 
  if (GNUNET_EXTRA_LOGGING > 0)
  {
    char *path;
 
    path = GNUNET_strdup (GNUNET_i2s (GST_my_identity));
    for (unsigned int i = 0; i < num_hops; i++)
    {
      char *tmp;
 
      GNUNET_asprintf (&tmp, "%s->%s", path, GNUNET_i2s (&hops[i]));
      GNUNET_free (path);
      path = tmp;
    }
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Received DVBox with remaining path %s\n",
                path);
    GNUNET_free (path);
  }
 
  if (num_hops > 0)
  {
    /* We're trying from the end of the hops array, as we may be
       able to find a shortcut unknown to the origin that way */
    for (int i = num_hops - 1; i >= 0; i--)
    {
      struct Neighbour *n;
 
      if (0 == GNUNET_memcmp (&hops[i], GST_my_identity))
      {
        GNUNET_break_op (0);
        finish_cmc_handling (cmc);
        return;
      }
      n = lookup_neighbour (&hops[i]);
      if (NULL == n)
        continue;
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "Skipping %u/%u hops ahead while routing DV Box\n",
                  i,
                  num_hops);
 
      forward_dv_box (n,
                      (struct TransportDVBoxMessage *) dvb,
                      ntohs (dvb->total_hops) + 1,
                      num_hops - i - 1,    /* number of hops left */
                      &hops[i + 1],    /* remaining hops */
                      enc_payload,
                      enc_payload_size);
      GNUNET_STATISTICS_update (GST_stats,
                                "# DV hops skipped routing boxes",
                                i,
                                GNUNET_NO);
      GNUNET_STATISTICS_update (GST_stats,
                                "# DV boxes routed (total)",
                                1,
                                GNUNET_NO);
      finish_cmc_handling (cmc);
      return;
    }
    /* Woopsie, next hop not in neighbours, drop! */
    GNUNET_STATISTICS_update (GST_stats,
                              "# DV Boxes dropped: next hop unknown",
                              1,
                              GNUNET_NO);
    finish_cmc_handling (cmc);
    return;
  }
  /* We are the target. Unbox and handle message. */
  GNUNET_STATISTICS_update (GST_stats,
                            "# DV boxes opened (ultimate target)",
                            1,
                            GNUNET_NO);
  cmc->total_hops = ntohs (dvb->total_hops);
 
  {
    // DH key derivation with received DV, could be garbage.
    decaps_dv_box_cls = GNUNET_new (struct DecapsDvBoxCls);
    decaps_dv_box_cls->cmc = cmc;
    decaps_dv_box_cls->dvb = dvb;
    decaps_dv_box_cls->pr = GNUNET_new (struct PilsRequest);
 
    GNUNET_CONTAINER_DLL_insert (pils_requests_head,
                                 pils_requests_tail,
                                 decaps_dv_box_cls->pr);
    decaps_dv_box_cls->pr->op = GNUNET_PILS_kem_decaps (pils,
                                                        &dvb->ephemeral_key,
                                                        decaps_dv_box_cb,
                                                        decaps_dv_box_cls);
  }
  // TODO keep track of cls and potentially clean
}
 
 
static int
check_incoming_msg (void *cls,
                    const struct GNUNET_TRANSPORT_IncomingMessage *im)
{
  struct TransportClient *tc = cls;
 
  if (CT_COMMUNICATOR != tc->type)
  {
    GNUNET_break (0);
    return GNUNET_SYSERR;
  }
  GNUNET_MQ_check_boxed_message (im);
  return GNUNET_OK;
}
 
 
struct CheckKnownAddressContext
{
  const char *address;
 
  struct ValidationState *vs;
};
 
 
static int
check_known_address (void *cls,
                     const struct GNUNET_PeerIdentity *pid,
                     void *value)
{
  struct CheckKnownAddressContext *ckac = cls;
  struct ValidationState *vs = value;
 
  (void) pid;
  if (0 != strcmp (vs->address, ckac->address))
    return GNUNET_OK;
  ckac->vs = vs;
  return GNUNET_NO;
}
 
 
static void
validation_start_cb (void *cls);
 
 
static void
update_next_challenge_time (struct ValidationState *vs,
                            struct GNUNET_TIME_Absolute new_time)
{
  struct GNUNET_TIME_Relative delta;
 
  if (new_time.abs_value_us == vs->next_challenge.abs_value_us)
    return; /* be lazy */
  vs->next_challenge = new_time;
  if (NULL == vs->hn)
    vs->hn =
      GNUNET_CONTAINER_heap_insert (validation_heap, vs, new_time.abs_value_us);
  else
    GNUNET_CONTAINER_heap_update_cost (vs->hn, new_time.abs_value_us);
  if ((vs != GNUNET_CONTAINER_heap_peek (validation_heap)) &&
      (NULL != validation_task))
    return;
  if (NULL != validation_task)
    GNUNET_SCHEDULER_cancel (validation_task);
  /* randomize a bit */
  delta.rel_value_us =
    GNUNET_CRYPTO_random_u64 (GNUNET_CRYPTO_QUALITY_WEAK,
                              MIN_DELAY_ADDRESS_VALIDATION.rel_value_us);
  new_time = GNUNET_TIME_absolute_add (new_time, delta);
  validation_task =
    GNUNET_SCHEDULER_add_at (new_time, &validation_start_cb, NULL);
}
 
 
static void
start_address_validation (const struct GNUNET_PeerIdentity *pid,
                          const char *address)
{
  struct GNUNET_TIME_Absolute now;
  struct ValidationState *vs;
  struct CheckKnownAddressContext ckac = { .address = address, .vs = NULL };
 
  (void) GNUNET_CONTAINER_multipeermap_get_multiple (validation_map,
                                                     pid,
                                                     &check_known_address,
                                                     &ckac);
  if (NULL != (vs = ckac.vs))
  {
    /* if 'vs' is not currently valid, we need to speed up retrying the
     * validation */
    if (vs->validated_until.abs_value_us < vs->next_challenge.abs_value_us)
    {
      /* reduce backoff as we got a fresh advertisement */
      vs->challenge_backoff =
        GNUNET_TIME_relative_min (FAST_VALIDATION_CHALLENGE_FREQ,
                                  GNUNET_TIME_relative_divide (
                                    vs->challenge_backoff,
                                    2));
      update_next_challenge_time (vs,
                                  GNUNET_TIME_relative_to_absolute (
                                    vs->challenge_backoff));
    }
    return;
  }
  now = GNUNET_TIME_absolute_get_monotonic (GST_cfg);
  vs = GNUNET_new (struct ValidationState);
  vs->pid = *pid;
  vs->valid_until =
    GNUNET_TIME_relative_to_absolute (ADDRESS_VALIDATION_LIFETIME);
  vs->first_challenge_use = now;
  vs->validation_rtt = GNUNET_TIME_UNIT_FOREVER_REL;
  GNUNET_CRYPTO_random_block (GNUNET_CRYPTO_QUALITY_NONCE,
                              &vs->challenge,
                              sizeof(vs->challenge));
  vs->address = GNUNET_strdup (address);
  GNUNET_CRYPTO_hash (vs->address, strlen (vs->address), &vs->hc);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Starting address validation `%s' of peer %s using challenge %s\n",
              address,
              GNUNET_i2s (pid),
              GNUNET_sh2s (&vs->challenge.value));
  GNUNET_assert (GNUNET_YES ==
                 GNUNET_CONTAINER_multipeermap_put (
                   validation_map,
                   &vs->pid,
                   vs,
                   GNUNET_CONTAINER_MULTIHASHMAPOPTION_MULTIPLE));
  update_next_challenge_time (vs, now);
}
 
 
static struct Queue *
find_queue (const struct GNUNET_PeerIdentity *pid, const char *address);
 
 
static void
suggest_to_connect (const struct GNUNET_PeerIdentity *pid, const char *address);
 
 
static void
hello_for_incoming_cb (void *cls,
                       const struct GNUNET_PeerIdentity *pid,
                       const char *uri)
{
  struct Queue *q;
  int pfx_len;
  const char *eou;
  char *address;
  (void) cls;
 
  eou = strstr (uri,
                "://");
  pfx_len = eou - uri;
  eou += 3;
  GNUNET_asprintf (&address,
                   "%.*s-%s",
                   pfx_len,
                   uri,
                   eou);
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "helo for client %s\n",
              address);
  q = find_queue (pid, address);
  if (NULL == q)
  {
    suggest_to_connect (pid, address);
  }
  else
    start_address_validation (pid, address);
  GNUNET_free (address);
}
 
 
static void
handle_hello_for_incoming (void *cls,
                           const struct GNUNET_PEERSTORE_Record *record,
                           const char *emsg)
{
  struct IncomingRequest *ir = cls;
  struct GNUNET_HELLO_Parser *parser;
  struct GNUNET_MessageHeader *hello;
 
  if (NULL != emsg)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
                "Got failure from PEERSTORE: %s\n",
                emsg);
    return;
  }
  hello = record->value;
  if (0 == GNUNET_memcmp (&record->peer, GST_my_identity))
  {
    GNUNET_PEERSTORE_monitor_next (ir->nc, 1);
    return;
  }
  parser = GNUNET_HELLO_parser_from_msg (hello);
  GNUNET_HELLO_parser_iterate (parser,
                               hello_for_incoming_cb,
                               NULL);
  GNUNET_HELLO_parser_free (parser);
}
 
 
static void
hello_for_incoming_error_cb (void *cls)
{
  GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
              "Error in PEERSTORE monitoring\n");
}
 
 
static void
hello_for_incoming_sync_cb (void *cls)
{
  GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
              "Done with initial PEERSTORE iteration during monitoring\n");
}
 
 
struct SignTValidationCls
{
  struct CommunicatorMessageContext *cmc;
  struct TransportValidationResponseMessage tvr;
  struct PilsRequest *pr;
};
 
 
static void
sign_t_validation_cb (void *cls,
                      const struct GNUNET_PeerIdentity *pid,
                      const struct GNUNET_CRYPTO_EddsaSignature *sig)
{
  struct SignTValidationCls *sign_t_validation_cls = cls;
  struct CommunicatorMessageContext *cmc = sign_t_validation_cls->cmc;
  struct TransportValidationResponseMessage tvr = sign_t_validation_cls->tvr;
  struct VirtualLink *vl;
  struct Neighbour *n;
  struct IncomingRequest *ir;
  struct GNUNET_PeerIdentity sender;
 
  sign_t_validation_cls->pr->op = NULL;
  GNUNET_CONTAINER_DLL_remove (pils_requests_head,
                               pils_requests_tail,
                               sign_t_validation_cls->pr);
  GNUNET_free (sign_t_validation_cls->pr);
  tvr.signature = *sig;
  sender = cmc->im.sender;
  vl = lookup_virtual_link (&sender);
  if ((NULL != vl) && (GNUNET_YES == vl->confirmed))
  {
    // route_control_message_without_fc (&cmc->im.sender,
    route_control_message_without_fc (vl,
                                      &tvr.header,
                                      RMO_ANYTHING_GOES | RMO_REDUNDANT);
  }
  else
  {
    /* Use route via neighbour */
    n = lookup_neighbour (&sender);
    if (NULL != n)
      route_via_neighbour (n, &tvr.header,
                           RMO_ANYTHING_GOES | RMO_REDUNDANT
                           | RMO_UNCONFIRMED_ALLOWED);
  }
 
  finish_cmc_handling (cmc);
  if (NULL != vl)
    return;
 
  /* For us, the link is still down, but we need bi-directional
     connections (for flow-control and for this to be useful for
     CORE), so we must try to bring the link up! */
 
  /* (1) Check existing queues, if any, we may be lucky! */
  n = lookup_neighbour (&sender);
  if (NULL != n)
    for (struct Queue *q = n->queue_head; NULL != q; q = q->next_neighbour)
      start_address_validation (&sender, q->address);
  /* (2) Also try to see if we have addresses in PEERSTORE for this peer
     we could use */
  for (ir = ir_head; NULL != ir; ir = ir->next)
    if (0 == GNUNET_memcmp (&ir->pid, &sender))
      return;
  /* we are already trying */
  ir = GNUNET_new (struct IncomingRequest);
  ir->pid = sender;
  GNUNET_CONTAINER_DLL_insert (ir_head, ir_tail, ir);
 
  ir->nc = GNUNET_PEERSTORE_monitor_start (GST_cfg,
                                           GNUNET_YES,
                                           "peerstore",
                                           NULL,
                                           GNUNET_PEERSTORE_HELLO_KEY,
                                           &hello_for_incoming_error_cb,
                                           NULL,
                                           &hello_for_incoming_sync_cb,
                                           NULL,
                                           &handle_hello_for_incoming,
                                           ir);
  ir_total++;
  /* Bound attempts we do in parallel here, might otherwise get excessive */
  while (ir_total > MAX_INCOMING_REQUEST)
    free_incoming_request (ir_head);
};
 
 
static void
handle_validation_challenge (
  void *cls,
  const struct TransportValidationChallengeMessage *tvc)
{
  struct CommunicatorMessageContext *cmc = cls;
  struct TransportValidationResponseMessage tvr;
  struct GNUNET_TIME_RelativeNBO validity_duration;
 
  /* DV-routed messages are not allowed for validation challenges */
  if (cmc->total_hops > 0)
  {
    GNUNET_break_op (0);
    finish_cmc_handling (cmc);
    return;
  }
  validity_duration = cmc->im.expected_address_validity;
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Received address validation challenge %s\n",
              GNUNET_sh2s (&tvc->challenge.value));
  /* If we have a virtual link, we use this mechanism to signal the
     size of the flow control window, and to allow the sender
     to ask for increases. If for us the virtual link is still down,
     we will always give a window size of zero. */
  tvr.header.type =
    htons (GNUNET_MESSAGE_TYPE_TRANSPORT_ADDRESS_VALIDATION_RESPONSE);
  tvr.header.size = htons (sizeof(tvr));
  tvr.reserved = htonl (0);
  tvr.challenge = tvc->challenge;
  tvr.origin_time = tvc->sender_time;
  tvr.validity_duration = validity_duration;
  {
    /* create signature */
    struct TransportValidationPS tvp = {
      .purpose.purpose = htonl (GNUNET_SIGNATURE_PURPOSE_TRANSPORT_CHALLENGE),
      .purpose.size = htonl (sizeof(tvp)),
      .validity_duration = validity_duration,
      .challenge = tvc->challenge
    };
    struct SignTValidationCls *sign_t_validation_cls;
 
    sign_t_validation_cls = GNUNET_new (struct SignTValidationCls);
    sign_t_validation_cls->cmc = cmc;
    sign_t_validation_cls->tvr = tvr;
    sign_t_validation_cls->pr = GNUNET_new (struct PilsRequest);
    GNUNET_CONTAINER_DLL_insert (pils_requests_head,
                                 pils_requests_tail,
                                 sign_t_validation_cls->pr);
    sign_t_validation_cls->pr->op =
      GNUNET_PILS_sign_by_peer_identity (pils,
                                         &tvp.purpose,
                                         &sign_t_validation_cb,
                                         sign_t_validation_cls);
  }
}
 
 
struct CheckKnownChallengeContext
{
  const struct GNUNET_CRYPTO_ChallengeNonceP *challenge;
 
  struct ValidationState *vs;
};
 
 
static int
check_known_challenge (void *cls,
                       const struct GNUNET_PeerIdentity *pid,
                       void *value)
{
  struct CheckKnownChallengeContext *ckac = cls;
  struct ValidationState *vs = value;
 
  (void) pid;
  if (0 != GNUNET_memcmp (&vs->challenge, ckac->challenge))
    return GNUNET_OK;
  ckac->vs = vs;
  return GNUNET_NO;
}
 
 
static void
peerstore_store_validation_cb (void *cls, int success)
{
  struct ValidationState *vs = cls;
 
  vs->sc = NULL;
  if (GNUNET_YES == success)
    return;
  GNUNET_STATISTICS_update (GST_stats,
                            "# Peerstore failed to store foreign address",
                            1,
                            GNUNET_NO);
}
 
 
static struct Queue *
find_queue (const struct GNUNET_PeerIdentity *pid, const char *address)
{
  struct Neighbour *n;
 
  n = lookup_neighbour (pid);
  if (NULL == n)
    return NULL;
  for (struct Queue *pos = n->queue_head; NULL != pos;
       pos = pos->next_neighbour)
  {
    if (0 == strcmp (pos->address, address))
      return pos;
  }
  return NULL;
}
 
 
static void
validation_transmit_on_queue (struct Queue *q, struct ValidationState *vs);
 
static void
revalidation_start_cb (void *cls)
{
  struct ValidationState *vs = cls;
  struct Queue *q;
  struct GNUNET_TIME_Absolute now;
 
  vs->revalidation_task = NULL;
  q = find_queue (&vs->pid, vs->address);
  if (NULL == q)
  {
    now = GNUNET_TIME_absolute_get ();
    vs->awaiting_queue = GNUNET_YES;
    suggest_to_connect (&vs->pid, vs->address);
    update_next_challenge_time (vs, now);
  }
  else
    validation_transmit_on_queue (q, vs);
}
 
 
static enum GNUNET_GenericReturnValue
revalidate_map_it (
  void *cls,
  const struct GNUNET_HashCode *key,
  void *value)
{
  (void) cls;
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Key in revalidate map  %s \n",
              GNUNET_h2s (key));
  return GNUNET_YES;
}
 
 
static void
handle_validation_response (
  void *cls,
  const struct TransportValidationResponseMessage *tvr)
{
  struct CommunicatorMessageContext *cmc = cls;
  struct ValidationState *vs;
  struct CheckKnownChallengeContext ckac = { .challenge = &tvr->challenge,
                                             .vs = NULL};
  struct GNUNET_TIME_Absolute origin_time;
  struct Queue *q;
  struct Neighbour *n;
  struct VirtualLink *vl;
  const struct GNUNET_TIME_Absolute now = GNUNET_TIME_absolute_get_monotonic (
    GST_cfg);
 
  /* check this is one of our challenges */
  (void) GNUNET_CONTAINER_multipeermap_get_multiple (validation_map,
                                                     &cmc->im.sender,
                                                     &check_known_challenge,
                                                     &ckac);
  if (NULL == (vs = ckac.vs))
  {
    /* This can happen simply if we 'forgot' the challenge by now,
       i.e. because we received the validation response twice */
    GNUNET_STATISTICS_update (GST_stats,
                              "# Validations dropped, challenge unknown",
                              1,
                              GNUNET_NO);
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Validation response %s dropped, challenge unknown\n",
                GNUNET_sh2s (&tvr->challenge.value));
    finish_cmc_handling (cmc);
    return;
  }
 
  /* sanity check on origin time */
  origin_time = GNUNET_TIME_absolute_ntoh (tvr->origin_time);
  if ((origin_time.abs_value_us < vs->first_challenge_use.abs_value_us) ||
      (origin_time.abs_value_us > vs->last_challenge_use.abs_value_us))
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Diff first use %" PRIu64 " and last use %" PRIu64 "\n",
                vs->first_challenge_use.abs_value_us - origin_time.abs_value_us,
                origin_time.abs_value_us - vs->last_challenge_use.abs_value_us);
    GNUNET_break_op (0);
    finish_cmc_handling (cmc);
    return;
  }
 
  {
    /* check signature */
    struct TransportValidationPS tvp = {
      .purpose.purpose = htonl (GNUNET_SIGNATURE_PURPOSE_TRANSPORT_CHALLENGE),
      .purpose.size = htonl (sizeof(tvp)),
      .validity_duration = tvr->validity_duration,
      .challenge = tvr->challenge
    };
 
    if (
      GNUNET_OK !=
      GNUNET_CRYPTO_eddsa_verify (GNUNET_SIGNATURE_PURPOSE_TRANSPORT_CHALLENGE,
                                  &tvp,
                                  &tvr->signature,
                                  &cmc->im.sender.public_key))
    {
      GNUNET_break_op (0);
      finish_cmc_handling (cmc);
      return;
    }
  }
 
  /* validity is capped by our willingness to keep track of the
     validation entry and the maximum the other peer allows */
  vs->valid_until = GNUNET_TIME_relative_to_absolute (
    GNUNET_TIME_relative_min (GNUNET_TIME_relative_ntoh (
                                tvr->validity_duration),
                              MAX_ADDRESS_VALID_UNTIL));
  vs->validated_until =
    GNUNET_TIME_absolute_min (vs->valid_until,
                              GNUNET_TIME_relative_to_absolute (
                                ADDRESS_VALIDATION_LIFETIME));
  vs->validation_rtt = GNUNET_TIME_absolute_get_duration (origin_time);
  vs->challenge_backoff = GNUNET_TIME_UNIT_ZERO;
  GNUNET_CRYPTO_random_block (GNUNET_CRYPTO_QUALITY_NONCE,
                              &vs->challenge,
                              sizeof(vs->challenge));
  vs->first_challenge_use = GNUNET_TIME_absolute_subtract (
    vs->validated_until,
    GNUNET_TIME_relative_multiply (vs->validation_rtt,
                                   VALIDATION_RTT_BUFFER_FACTOR));
  if (GNUNET_TIME_absolute_cmp (vs->first_challenge_use, <, now))
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "First challenge use is now %" PRIu64 " %s \n",
                vs->first_challenge_use.abs_value_us,
                GNUNET_sh2s (&vs->challenge.value));
    vs->first_challenge_use = now;
  }
  else
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "First challenge use is later %" PRIu64 " %s \n",
                vs->first_challenge_use.abs_value_us,
                GNUNET_sh2s (&vs->challenge.value));
  vs->last_challenge_use =
    GNUNET_TIME_UNIT_ZERO_ABS; /* challenge was not yet used */
  update_next_challenge_time (vs, vs->first_challenge_use);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Validation response %s from %s accepted, address valid until %s\n",
              GNUNET_sh2s (&tvr->challenge.value),
              GNUNET_i2s (&cmc->im.sender),
              GNUNET_STRINGS_absolute_time_to_string (vs->valid_until));
  /*memcpy (&hkey,
          &hc,
          sizeof (hkey));*/
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Key %s for address %s map size %u contains %u\n",
              GNUNET_h2s (&vs->hc),
              vs->address,
              GNUNET_CONTAINER_multihashmap_size (revalidation_map),
              GNUNET_CONTAINER_multihashmap_contains (revalidation_map,
                                                      &vs->hc));
  GNUNET_assert (GNUNET_YES ==
                 GNUNET_CONTAINER_multihashmap_put (
                   revalidation_map,
                   &vs->hc,
                   vs,
                   GNUNET_CONTAINER_MULTIHASHMAPOPTION_UNIQUE_ONLY));
  GNUNET_CONTAINER_multihashmap_iterate (revalidation_map,
                                         revalidate_map_it,
                                         NULL);
  vs->revalidation_task =
    GNUNET_SCHEDULER_add_at (GNUNET_TIME_absolute_subtract (vs->next_challenge,
                                                            GNUNET_TIME_UNIT_MINUTES),
                             &revalidation_start_cb, vs);
  vs->sc = GNUNET_PEERSTORE_store (peerstore,
                                   "transport",
                                   &cmc->im.sender,
                                   GNUNET_PEERSTORE_TRANSPORT_URLADDRESS_KEY,
                                   vs->address,
                                   strlen (vs->address) + 1,
                                   vs->valid_until,
                                   GNUNET_PEERSTORE_STOREOPTION_MULTIPLE,
                                   &peerstore_store_validation_cb,
                                   vs);
  finish_cmc_handling (cmc);
 
  /* Finally, we now possibly have a confirmed (!) working queue,
     update queue status (if queue still is around) */
  q = find_queue (&vs->pid, vs->address);
  if (NULL == q)
  {
    GNUNET_STATISTICS_update (GST_stats,
                              "# Queues lost at time of successful validation",
                              1,
                              GNUNET_NO);
    return;
  }
  q->validated_until = vs->validated_until;
  q->pd.aged_rtt = vs->validation_rtt;
  n = q->neighbour;
  vl = lookup_virtual_link (&vs->pid);
  if (NULL == vl)
  {
    vl = GNUNET_new (struct VirtualLink);
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Creating new virtual link %p to %s using direct neighbour!\n",
                vl,
                GNUNET_i2s (&vs->pid));
    vl->burst_addr = NULL;
    vl->confirmed = GNUNET_YES;
    vl->message_uuid_ctr =
      GNUNET_CRYPTO_random_u64 (GNUNET_CRYPTO_QUALITY_WEAK, UINT64_MAX);
    vl->target = n->pid;
    vl->core_recv_window = RECV_WINDOW_SIZE;
    vl->available_fc_window_size = DEFAULT_WINDOW_SIZE;
    vl->incoming_fc_window_size = DEFAULT_WINDOW_SIZE;
    GNUNET_break (GNUNET_YES ==
                  GNUNET_CONTAINER_multipeermap_put (
                    links,
                    &vl->target,
                    vl,
                    GNUNET_CONTAINER_MULTIHASHMAPOPTION_UNIQUE_ONLY));
    vl->n = n;
    n->vl = vl;
    q->idle = GNUNET_YES;
    vl->visibility_task =
      GNUNET_SCHEDULER_add_at (q->validated_until, &check_link_down, vl);
    consider_sending_fc (vl);
    /* We lacked a confirmed connection to the target
       before, so tell CORE about it (finally!) */
    cores_send_connect_info (&n->pid);
    send_msg_from_cache (vl);
  }
  else
  {
    /* Link was already up, remember n is also now available and we are done */
    if (NULL == vl->n)
    {
      vl->n = n;
      n->vl = vl;
      if (GNUNET_YES == vl->confirmed)
        GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                    "Virtual link to %s could now also use direct neighbour!\n",
                    GNUNET_i2s (&vs->pid));
    }
    else
    {
      GNUNET_assert (n == vl->n);
    }
    if (GNUNET_NO == vl->confirmed)
    {
      vl->confirmed = GNUNET_YES;
      q->idle = GNUNET_YES;
      vl->visibility_task =
        GNUNET_SCHEDULER_add_at (q->validated_until, &check_link_down, vl);
      consider_sending_fc (vl);
      /* We lacked a confirmed connection to the target
         before, so tell CORE about it (finally!) */
      cores_send_connect_info (&n->pid);
      send_msg_from_cache (vl);
    }
  }
}
 
 
static void
handle_incoming_msg (void *cls,
                     const struct GNUNET_TRANSPORT_IncomingMessage *im)
{
  struct TransportClient *tc = cls;
  struct CommunicatorMessageContext *cmc =
    GNUNET_new (struct CommunicatorMessageContext);
 
  cmc->tc = tc;
  cmc->im = *im;
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Received message with size %u and flow control id %" PRIu64
              " via communicator from peer %s\n",
              ntohs (im->header.size),
              im->fc_id,
              GNUNET_i2s (&im->sender));
  cmc->im.neighbour_sender = cmc->im.sender;
  cmc->mh = (const struct GNUNET_MessageHeader *) &im[1];
  demultiplex_with_cmc (cmc);
}
 
 
static int
check_flow_control (void *cls, const struct TransportFlowControlMessage *fc)
{
  unsigned int number_of_addresses = ntohl (fc->number_of_addresses);
  (void) cls;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Flow control header size %u size of addresses %u number of addresses %u size of message struct %lu second struct %lu\n",
              ntohs (fc->header.size),
              ntohl (fc->size_of_addresses),
              ntohl (fc->number_of_addresses),
              sizeof(struct TransportFlowControlMessage),
              sizeof (struct TransportGlobalNattedAddress));
 
  if (0 == number_of_addresses || ntohs (fc->header.size) == sizeof(struct
                                                                    TransportFlowControlMessage)
      + ntohl (fc->number_of_addresses) * sizeof (struct
                                                  TransportGlobalNattedAddress)
      + ntohl (fc->size_of_addresses))
    return GNUNET_OK;
  else
  {
    GNUNET_break_op (0);
    return GNUNET_SYSERR;
  }
}
 
 
static struct GNUNET_TIME_Relative
calculate_rtt (struct DistanceVector *dv)
{
  struct GNUNET_TIME_Relative ret = GNUNET_TIME_UNIT_ZERO;
  unsigned int n_hops = 0;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "calculate_rtt\n");
  for (struct DistanceVectorHop *pos = dv->dv_head; NULL != pos;
       pos = pos->next_dv)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "calculate_rtt %lu\n",
                (unsigned long) pos->pd.aged_rtt.rel_value_us);
    n_hops++;
    ret = GNUNET_TIME_relative_add (GNUNET_TIME_relative_multiply (pos->pd.
                                                                   aged_rtt, pos
                                                                   ->distance
                                                                   + 2), ret);
  }
 
  GNUNET_assert (0 != n_hops);
 
  return ret;
}
 
 
static void
iterate_address_start_burst (void *cls,
                             const struct GNUNET_PeerIdentity *pid,
                             const char *uri)
{
  struct VirtualLink *vl = cls;
  const char *slash;
  char *address_uri;
  char *prefix;
  char *uri_without_port;
 
  slash = strrchr (uri, '/');
  prefix = GNUNET_strndup (uri, (slash - uri) - 2);
  GNUNET_assert (NULL != slash);
  slash++;
  GNUNET_asprintf (&address_uri,
                   "%s-%s",
                   prefix,
                   slash);
 
  uri_without_port = get_address_without_port (address_uri);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "iterate_address_start_burst %s %s %s %s\n",
              uri_without_port,
              uri,
              address_uri,
              slash);
  if (0 == strcmp (uri_without_port, slash))
  {
    vl->burst_addr = GNUNET_strndup (uri_without_port, strlen (uri_without_port)
                                     );
  }
  else
    vl->burst_addr = NULL;
 
  GNUNET_free (prefix);
  GNUNET_free (uri_without_port);
}
 
 
static void
check_for_burst_address (void *cls,
                         const struct GNUNET_PEERSTORE_Record *record,
                         const char *emsg)
{
  struct GNUNET_StartBurstCls *sb_cls = cls;
  struct VirtualLink *vl = sb_cls->vl;
  struct GNUNET_MessageHeader *hello;
  struct GNUNET_HELLO_Parser *parser;
 
  if (NULL != emsg)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
                "Got failure from PEERSTORE: %s\n",
                emsg);
    return;
  }
  if (NULL == record)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Hello iteration end for %s\n",
                GNUNET_i2s (&vl->target));
    vl->ic = NULL;
    GNUNET_free (sb_cls);
    return;
  }
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "check_for_burst_address\n");
  hello = record->value;
  parser = GNUNET_HELLO_parser_from_msg (hello);
  GNUNET_HELLO_parser_iterate (parser,
                               &iterate_address_start_burst,
                               vl);
  GNUNET_HELLO_parser_free (parser);
 
  GNUNET_PEERSTORE_iteration_stop (vl->ic);
  GNUNET_free (sb_cls);
}
 
 
static void
burst_timeout (void *cls)
{
  burst_running = GNUNET_NO;
}
 
 
static void
start_burst (void *cls)
{
  struct GNUNET_StartBurstCls *sb_cls = cls;
  struct VirtualLink *vl = sb_cls->vl;
  struct GNUNET_TRANSPORT_StartBurst *sb;
  struct GNUNET_MQ_Envelope *env;
  char *uri_without_port = vl->burst_addr;
 
  burst_task = NULL;
  /*char buf[strlen (uri_without_port) + 1];
 
  GNUNET_memcpy (buf, uri_without_port,  strlen (uri_without_port));
  buf[strlen (uri_without_port)] = '\0';*/
  env =
    GNUNET_MQ_msg_extra (sb,
                         strlen (uri_without_port) + 1,
                         GNUNET_MESSAGE_TYPE_TRANSPORT_START_BURST);
  sb->rtt = GNUNET_TIME_relative_hton (sb_cls->rtt);
  sb->pid = vl->target;
  memcpy (&sb[1], uri_without_port, strlen (uri_without_port) + 1);
  for (struct TransportClient *tc = clients_head; NULL != tc; tc = tc->next)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "iterate_address_start_burst client tc prefix %s\n",
                tc->details.communicator.address_prefix);
    if (CT_COMMUNICATOR != tc->type)
      continue;
    if (GNUNET_YES == tc->details.communicator.can_burst)
    {
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "iterate_address_start_burst %s call %lu %u rtt %lu\n",
                  uri_without_port,
                  strlen (uri_without_port),
                  ntohs (sb->header.size),
                  (unsigned long) sb_cls->rtt.rel_value_us);
      GNUNET_MQ_send (tc->mq, env);
      burst_running = GNUNET_YES;
      burst_timeout_task = GNUNET_SCHEDULER_add_delayed (
        GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_SECONDS,
                                       60),
        &burst_timeout,
        NULL);
      // TODO We need some algo to choose from available communicators. Can we run two bursts at once? Atm we only implemented udp burst.
      break;
    }
  }
  GNUNET_free (env);
  GNUNET_free (sb_cls);
}
 
 
static void
queue_burst (void *cls)
{
  struct GNUNET_StartBurstCls *sb_cls = cls;
  struct VirtualLink *vl = sb_cls->vl;
 
  if (GNUNET_YES != use_burst)
    return;
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "burst_task %p ready %s burst addr %s (%p)\n",
              burst_task,
              sb_cls->sync_ready ? "yes" : "no",
              vl->burst_addr,
              vl->burst_addr);
  if (NULL != burst_task && GNUNET_NO == sb_cls->sync_ready)
  {
    GNUNET_SCHEDULER_cancel (burst_task);
    burst_task = NULL;
    GNUNET_free (sb_cls);
    return;
  }
  if (GNUNET_NO == burst_running && NULL != vl->burst_addr && NULL == burst_task
      )
  {
    burst_task = GNUNET_SCHEDULER_add_delayed (sb_cls->delay,
                                               &start_burst,
                                               sb_cls);
  }
  else if (NULL == vl->burst_addr)
  {
    vl->ic = GNUNET_PEERSTORE_iteration_start (peerstore,
                                               "peerstore",
                                               &vl->target,
                                               GNUNET_PEERSTORE_HELLO_KEY,
                                               check_for_burst_address,
                                               sb_cls);
  }
}
 
 
static void
handle_flow_control (void *cls, const struct TransportFlowControlMessage *fc)
{
  struct CommunicatorMessageContext *cmc = cls;
  struct VirtualLink *vl;
  struct GNUNET_TIME_Absolute q_timeout = GNUNET_TIME_UNIT_ZERO_ABS;
  uint32_t seq;
  struct GNUNET_TIME_Absolute st;
  uint64_t os;
  uint64_t wnd;
  uint32_t random;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Received FC from %s\n", GNUNET_i2s (&cmc->im.sender));
  vl = lookup_virtual_link (&cmc->im.sender);
  if (NULL == vl)
  {
    vl = GNUNET_new (struct VirtualLink);
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "No virtual link for %p FC creating new unconfirmed virtual link to %s!\n",
                vl,
                GNUNET_i2s (&cmc->im.sender));
    vl->burst_addr = NULL;
    vl->confirmed = GNUNET_NO;
    vl->message_uuid_ctr =
      GNUNET_CRYPTO_random_u64 (GNUNET_CRYPTO_QUALITY_WEAK, UINT64_MAX);
    vl->target = cmc->im.sender;
    vl->core_recv_window = RECV_WINDOW_SIZE;
    vl->available_fc_window_size = DEFAULT_WINDOW_SIZE;
    vl->incoming_fc_window_size = DEFAULT_WINDOW_SIZE;
    GNUNET_break (GNUNET_YES ==
                  GNUNET_CONTAINER_multipeermap_put (
                    links,
                    &vl->target,
                    vl,
                    GNUNET_CONTAINER_MULTIHASHMAPOPTION_UNIQUE_ONLY));
  }
  if (NULL != vl->n)
  {
    for (struct Queue *q = vl->n->queue_head; NULL != q; q = q->next_neighbour)
      q_timeout = GNUNET_TIME_absolute_max (q_timeout, q->validated_until);
  }
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "remaining %lu timeout for neighbour %p\n",
              (unsigned long) GNUNET_TIME_absolute_get_remaining (q_timeout).
              rel_value_us,
              vl->n);
  if (NULL == vl->n ||
      0 == GNUNET_TIME_absolute_get_remaining (q_timeout).rel_value_us)
  {
    struct GNUNET_TIME_Relative rtt;
    struct GNUNET_BurstSync burst_sync;
    struct GNUNET_StartBurstCls *bcls;
 
    bcls = GNUNET_new (struct GNUNET_StartBurstCls);
    bcls->vl = vl;
    vl->sb_cls = bcls;
    if (NULL != vl->dv)
      rtt = calculate_rtt (vl->dv);
    else
      rtt = GNUNET_TIME_UNIT_FOREVER_REL;
    burst_sync.rtt_average = fc->rtt;
    bcls->rtt = GNUNET_TIME_relative_ntoh (burst_sync.rtt_average);
    burst_sync.sync_ready = fc->sync_ready;
 
    GNUNET_is_burst_ready (rtt,
                           &burst_sync,
                           &queue_burst,
                           bcls);
  }
  if (0 != ntohl (fc->number_of_addresses))
  {
    unsigned int number_of_addresses = ntohl (fc->number_of_addresses);
    const char *tgnas;
    unsigned int off = 0;
 
    tgnas = (const char *) &fc[1];
 
    for (int i = 1; i <= number_of_addresses; i++)
    {
      struct TransportGlobalNattedAddress *tgna;
      char *addr;
      unsigned int address_length;
 
      tgna = (struct TransportGlobalNattedAddress*) &tgnas[off];
      addr = (char *) &tgna[1];
      address_length = ntohl (tgna->address_length);
      off += sizeof(struct TransportGlobalNattedAddress) + address_length;
 
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "received address %s length %u\n",
                  addr,
                  ntohl (tgna->address_length));
 
      GNUNET_NAT_add_global_address (nh, addr, ntohl (tgna->address_length));
    }
  }
  st = GNUNET_TIME_absolute_ntoh (fc->sender_time);
  if (st.abs_value_us < vl->last_fc_timestamp.abs_value_us)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "FC dropped: Message out of order\n");
    /* out of order, drop */
    GNUNET_STATISTICS_update (GST_stats,
                              "# FC dropped: message out of order",
                              1,
                              GNUNET_NO);
    finish_cmc_handling (cmc);
    return;
  }
  seq = ntohl (fc->seq);
  if (seq < vl->last_fc_seq)
  {
    /* Wrap-around/reset of other peer; start all counters from zero */
    vl->outbound_fc_window_size_used = 0;
  }
  vl->last_fc_seq = seq;
  vl->last_fc_timestamp = st;
  vl->outbound_fc_window_size = GNUNET_ntohll (fc->inbound_window_size);
  os = GNUNET_ntohll (fc->outbound_sent);
  vl->incoming_fc_window_size_loss =
    (int64_t) (os - vl->incoming_fc_window_size_used);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Received FC from %s, seq %u, new window %llu (loss at %lld)\n",
              GNUNET_i2s (&vl->target),
              (unsigned int) seq,
              (unsigned long long) vl->outbound_fc_window_size,
              (long long) vl->incoming_fc_window_size_loss);
  wnd = GNUNET_ntohll (fc->outbound_window_size);
  random = GNUNET_CRYPTO_random_u32 (GNUNET_CRYPTO_QUALITY_WEAK,
                                     UINT32_MAX);
  if ((GNUNET_YES == vl->confirmed) && ((wnd < vl->incoming_fc_window_size
                                         + vl->incoming_fc_window_size_used
                                         + vl->incoming_fc_window_size_loss) ||
                                        (vl->last_outbound_window_size_received
                                         != wnd) ||
                                        (0 == random
                                         % FC_NO_CHANGE_REPLY_PROBABILITY)))
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Consider re-sending our FC message, as clearly the other peer's idea of the window is not up-to-date (%llu vs %llu) or %llu last received differs, or random reply %u\n",
                (unsigned long long) wnd,
                (unsigned long long) vl->incoming_fc_window_size,
                (unsigned long long) vl->last_outbound_window_size_received,
                random % FC_NO_CHANGE_REPLY_PROBABILITY);
    consider_sending_fc (vl);
  }
  if ((wnd == vl->incoming_fc_window_size
       + vl->incoming_fc_window_size_used
       + vl->incoming_fc_window_size_loss) &&
      (vl->last_outbound_window_size_received == wnd) &&
      (NULL != vl->fc_retransmit_task))
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Stopping FC retransmission to %s: peer is current at window %llu\n",
                GNUNET_i2s (&vl->target),
                (unsigned long long) wnd);
    GNUNET_SCHEDULER_cancel (vl->fc_retransmit_task);
    vl->fc_retransmit_task = NULL;
    vl->fc_retransmit_count = 0;
  }
  vl->last_outbound_window_size_received = wnd;
  /* FC window likely increased, check transmission possibilities! */
  check_vl_transmission (vl);
  finish_cmc_handling (cmc);
}
 
 
static void
demultiplex_with_cmc (struct CommunicatorMessageContext *cmc)
{
  struct GNUNET_MQ_MessageHandler handlers[] =
  { GNUNET_MQ_hd_var_size (fragment_box,
                           GNUNET_MESSAGE_TYPE_TRANSPORT_FRAGMENT,
                           struct TransportFragmentBoxMessage,
                           cmc),
    GNUNET_MQ_hd_var_size (reliability_box,
                           GNUNET_MESSAGE_TYPE_TRANSPORT_RELIABILITY_BOX,
                           struct TransportReliabilityBoxMessage,
                           cmc),
    GNUNET_MQ_hd_var_size (reliability_ack,
                           GNUNET_MESSAGE_TYPE_TRANSPORT_RELIABILITY_ACK,
                           struct TransportReliabilityAckMessage,
                           cmc),
    GNUNET_MQ_hd_var_size (backchannel_encapsulation,
                           GNUNET_MESSAGE_TYPE_TRANSPORT_BACKCHANNEL_ENCAPSULATION,
                           struct TransportBackchannelEncapsulationMessage,
                           cmc),
    GNUNET_MQ_hd_var_size (dv_learn,
                           GNUNET_MESSAGE_TYPE_TRANSPORT_DV_LEARN,
                           struct TransportDVLearnMessage,
                           cmc),
    GNUNET_MQ_hd_var_size (dv_box,
                           GNUNET_MESSAGE_TYPE_TRANSPORT_DV_BOX,
                           struct TransportDVBoxMessage,
                           cmc),
    GNUNET_MQ_hd_var_size (flow_control,
                           GNUNET_MESSAGE_TYPE_TRANSPORT_FLOW_CONTROL,
                           struct TransportFlowControlMessage,
                           cmc),
    GNUNET_MQ_hd_fixed_size (
      validation_challenge,
      GNUNET_MESSAGE_TYPE_TRANSPORT_ADDRESS_VALIDATION_CHALLENGE,
      struct TransportValidationChallengeMessage,
      cmc),
    GNUNET_MQ_hd_fixed_size (
      validation_response,
      GNUNET_MESSAGE_TYPE_TRANSPORT_ADDRESS_VALIDATION_RESPONSE,
      struct TransportValidationResponseMessage,
      cmc),
    GNUNET_MQ_handler_end () };
  int ret;
  const struct GNUNET_MessageHeader *msg = cmc->mh;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Handling message of type %u with %u bytes\n",
              (unsigned int) ntohs (msg->type),
              (unsigned int) ntohs (msg->size));
  ret = GNUNET_MQ_handle_message (handlers, msg);
  if (GNUNET_SYSERR == ret)
  {
    GNUNET_break (0);
    GNUNET_SERVICE_client_drop (cmc->tc->client);
    GNUNET_free (cmc);
    return;
  }
  if (GNUNET_NO == ret)
  {
    /* unencapsulated 'raw' message */
    handle_raw_message (cmc, msg);
  }
}
 
 
static int
check_add_queue_message (void *cls,
                         const struct GNUNET_TRANSPORT_AddQueueMessage *aqm)
{
  struct TransportClient *tc = cls;
 
  if (CT_COMMUNICATOR != tc->type)
  {
    GNUNET_break (0);
    return GNUNET_SYSERR;
  }
  GNUNET_MQ_check_zero_termination (aqm);
  return GNUNET_OK;
}
 
 
static void
set_pending_message_uuid (struct PendingMessage *pm)
{
  if (pm->msg_uuid_set)
    return;
  pm->msg_uuid.uuid = pm->vl->message_uuid_ctr++;
  pm->msg_uuid_set = GNUNET_YES;
}
 
 
static struct PendingAcknowledgement *
prepare_pending_acknowledgement (struct Queue *queue,
                                 struct DistanceVectorHop *dvh,
                                 struct PendingMessage *pm)
{
  struct PendingAcknowledgement *pa;
 
  pa = GNUNET_new (struct PendingAcknowledgement);
  pa->queue = queue;
  pa->dvh = dvh;
  pa->pm = pm;
  do
  {
    GNUNET_CRYPTO_random_block (GNUNET_CRYPTO_QUALITY_NONCE,
                                &pa->ack_uuid,
                                sizeof(pa->ack_uuid));
  }
  while (GNUNET_YES != GNUNET_CONTAINER_multiuuidmap_put (
           pending_acks,
           &pa->ack_uuid.value,
           pa,
           GNUNET_CONTAINER_MULTIHASHMAPOPTION_UNIQUE_ONLY));
  GNUNET_CONTAINER_MDLL_insert (queue, queue->pa_head, queue->pa_tail, pa);
  GNUNET_CONTAINER_MDLL_insert (pm, pm->pa_head, pm->pa_tail, pa);
  if (NULL != dvh)
    GNUNET_CONTAINER_MDLL_insert (dvh, dvh->pa_head, dvh->pa_tail, pa);
  pa->transmission_time = GNUNET_TIME_absolute_get ();
  pa->message_size = pm->bytes_msg;
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Waiting for ACKnowledgment `%s' for <%" PRIu64 ">\n",
              GNUNET_uuid2s (&pa->ack_uuid.value),
              pm->logging_uuid);
  return pa;
}
 
 
static struct PendingMessage *
fragment_message (struct Queue *queue,
                  struct DistanceVectorHop *dvh,
                  struct PendingMessage *pm)
{
  struct PendingAcknowledgement *pa;
  struct PendingMessage *ff;
  uint16_t mtu;
  uint16_t msize;
 
  mtu = (UINT16_MAX == queue->mtu)
        ? UINT16_MAX - sizeof(struct GNUNET_TRANSPORT_SendMessageTo)
        : queue->mtu;
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Fragmenting message <%" PRIu64
              "> with size %u to %s for MTU %u\n",
              pm->logging_uuid,
              pm->bytes_msg,
              GNUNET_i2s (&pm->vl->target),
              (unsigned int) mtu);
  set_pending_message_uuid (pm);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Fragmenting message %" PRIu64 " <%" PRIu64
              "> with size %u to %s for MTU %u\n",
              pm->msg_uuid.uuid,
              pm->logging_uuid,
              pm->bytes_msg,
              GNUNET_i2s (&pm->vl->target),
              (unsigned int) mtu);
 
  /* This invariant is established in #handle_add_queue_message() */
  GNUNET_assert (mtu > sizeof(struct TransportFragmentBoxMessage));
 
  /* select fragment for transmission, descending the tree if it has
     been expanded until we are at a leaf or at a fragment that is small
     enough
   */
  ff = pm;
  msize = ff->bytes_msg;
 
  while (((ff->bytes_msg > mtu) || (pm == ff)) &&
         (ff->frag_off == msize) && (NULL != ff->head_frag))
  {
    ff = ff->head_frag;   /* descent into fragmented fragments */
    msize = ff->bytes_msg - sizeof(struct TransportFragmentBoxMessage);
  }
 
  if (((ff->bytes_msg > mtu) || (pm == ff)) && (ff->frag_off < msize))
  {
    /* Did not yet calculate all fragments, calculate next fragment */
    struct PendingMessage *frag;
    struct TransportFragmentBoxMessage tfb;
    const char *orig;
    char *msg;
    uint16_t fragmax;
    uint16_t fragsize;
    uint16_t msize_ff;
    uint16_t xoff = 0;
    pm->frag_count++;
 
    orig = (const char *) &ff[1];
    msize_ff = ff->bytes_msg;
    if (pm != ff)
    {
      const struct TransportFragmentBoxMessage *tfbo;
 
      tfbo = (const struct TransportFragmentBoxMessage *) orig;
      orig += sizeof(struct TransportFragmentBoxMessage);
      msize_ff -= sizeof(struct TransportFragmentBoxMessage);
      xoff = ntohs (tfbo->frag_off);
    }
    fragmax = mtu - sizeof(struct TransportFragmentBoxMessage);
    fragsize = GNUNET_MIN (msize_ff - ff->frag_off, fragmax);
    frag =
      GNUNET_malloc (sizeof(struct PendingMessage)
                     + sizeof(struct TransportFragmentBoxMessage) + fragsize);
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "3 created pm %p from pm %p storing vl %p from pm %p\n",
                frag,
                ff,
                pm->vl,
                pm);
    frag->logging_uuid = logging_uuid_gen++;
    frag->vl = pm->vl;
    frag->frag_parent = ff;
    frag->timeout = pm->timeout;
    frag->bytes_msg = sizeof(struct TransportFragmentBoxMessage) + fragsize;
    frag->pmt = PMT_FRAGMENT_BOX;
    msg = (char *) &frag[1];
    tfb.header.type = htons (GNUNET_MESSAGE_TYPE_TRANSPORT_FRAGMENT);
    tfb.header.size =
      htons (sizeof(struct TransportFragmentBoxMessage) + fragsize);
    pa = prepare_pending_acknowledgement (queue, dvh, frag);
    tfb.ack_uuid = pa->ack_uuid;
    tfb.msg_uuid = pm->msg_uuid;
    tfb.frag_off = htons (ff->frag_off + xoff);
    tfb.msg_size = htons (pm->bytes_msg);
    memcpy (msg, &tfb, sizeof(tfb));
    memcpy (&msg[sizeof(tfb)], &orig[ff->frag_off], fragsize);
    GNUNET_CONTAINER_MDLL_insert (frag, ff->head_frag,
                                  ff->tail_frag, frag);
    ff->frag_off += fragsize;
    ff = frag;
  }
 
  /* Move head to the tail and return it */
  GNUNET_CONTAINER_MDLL_remove (frag,
                                ff->frag_parent->head_frag,
                                ff->frag_parent->tail_frag,
                                ff);
  GNUNET_CONTAINER_MDLL_insert_tail (frag,
                                     ff->frag_parent->head_frag,
                                     ff->frag_parent->tail_frag,
                                     ff);
 
  return ff;
}
 
 
static struct PendingMessage *
reliability_box_message (struct Queue *queue,
                         struct DistanceVectorHop *dvh,
                         struct PendingMessage *pm)
{
  struct TransportReliabilityBoxMessage rbox;
  struct PendingAcknowledgement *pa;
  struct PendingMessage *bpm;
  char *msg;
 
  if ((PMT_CORE != pm->pmt) && (PMT_DV_BOX != pm->pmt))
    return pm; /* already fragmented or reliability boxed, or control message:
                  do nothing */
  if (NULL != pm->bpm)
    return pm->bpm; /* already computed earlier: do nothing */
  // TODO I guess we do not need this assertion. We might have a DLL with
  // fragments, because the MTU changed, and we do not need to fragment anymore.
  // But we should keep the fragments until message was completed, because
  // the MTU might change again.
  // GNUNET_assert (NULL == pm->head_frag);
  if (pm->bytes_msg + sizeof(rbox) > UINT16_MAX)
  {
    /* failed hard */
    GNUNET_break (0);
    client_send_response (pm);
    return NULL;
  }
 
  pa = prepare_pending_acknowledgement (queue, dvh, pm);
 
  bpm = GNUNET_malloc (sizeof(struct PendingMessage) + sizeof(rbox)
                       + pm->bytes_msg);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "4 created pm %p storing vl %p from pm %p\n",
              bpm,
              pm->vl,
              pm);
  bpm->logging_uuid = logging_uuid_gen++;
  bpm->vl = pm->vl;
  bpm->frag_parent = pm;
  // Why was this needed?
  // GNUNET_CONTAINER_MDLL_insert (frag, pm->head_frag, pm->tail_frag, bpm);
  bpm->timeout = pm->timeout;
  bpm->pmt = PMT_RELIABILITY_BOX;
  bpm->bytes_msg = pm->bytes_msg + sizeof(rbox);
  set_pending_message_uuid (bpm);
  rbox.header.type = htons (GNUNET_MESSAGE_TYPE_TRANSPORT_RELIABILITY_BOX);
  rbox.header.size = htons (sizeof(rbox) + pm->bytes_msg);
  rbox.ack_countdown = htonl (0);  // FIXME: implement ACK countdown support
 
  rbox.ack_uuid = pa->ack_uuid;
  msg = (char *) &bpm[1];
  memcpy (msg, &rbox, sizeof(rbox));
  memcpy (&msg[sizeof(rbox)], &pm[1], pm->bytes_msg);
  pm->bpm = bpm;
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Preparing reliability box for message <%" PRIu64
              "> of size %d (%d) to %s on queue %s\n",
              pm->logging_uuid,
              pm->bytes_msg,
              ntohs (((const struct GNUNET_MessageHeader *) &pm[1])->size),
              GNUNET_i2s (&pm->vl->target),
              queue->address);
  return bpm;
}
 
 
static void
reorder_root_pm (struct PendingMessage *pm,
                 struct GNUNET_TIME_Absolute next_attempt)
{
  struct VirtualLink *vl = pm->vl;
  struct PendingMessage *pos;
 
  /* re-insert sort in neighbour list */
  GNUNET_CONTAINER_MDLL_remove (vl,
                                vl->pending_msg_head,
                                vl->pending_msg_tail,
                                pm);
  pos = vl->pending_msg_tail;
  while ((NULL != pos) &&
         (next_attempt.abs_value_us > pos->next_attempt.abs_value_us))
    pos = pos->prev_vl;
  GNUNET_CONTAINER_MDLL_insert_after (vl,
                                      vl->pending_msg_head,
                                      vl->pending_msg_tail,
                                      pos,
                                      pm);
}
 
 
static unsigned int
check_next_attempt_tree (struct PendingMessage *pm, struct PendingMessage *root)
{
  struct PendingMessage *pos;
  enum GNUNET_GenericReturnValue frags_in_flight;
 
  pos = pm->head_frag;
  while (NULL != pos)
  {
    if (pos->frags_in_flight_round == pm->frags_in_flight_round ||
        GNUNET_NO == check_next_attempt_tree (pos, root))
      frags_in_flight = GNUNET_NO;
    else
    {
      frags_in_flight = GNUNET_YES;
      break;
    }
    pos = pos->next_frag;
  }
 
  return frags_in_flight;
}
 
 
static void
harmonize_flight_round (struct PendingMessage *pm)
{
  struct PendingMessage *pos;
 
  pos = pm->head_frag;
  while (NULL != pos)
  {
    pos->frags_in_flight_round = pm->frags_in_flight_round;
    harmonize_flight_round (pos);
    pos = pos->next_frag;
  }
}
 
 
static void
update_pm_next_attempt (struct PendingMessage *pm,
                        struct GNUNET_TIME_Absolute next_attempt)
{
  if (NULL == pm->frag_parent)
  {
    pm->next_attempt = next_attempt;
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Next attempt for message <%" PRIu64 "> set to %" PRIu64 "\n",
                pm->logging_uuid,
                next_attempt.abs_value_us);
    reorder_root_pm (pm, next_attempt);
  }
  else if ((PMT_RELIABILITY_BOX == pm->pmt) || (PMT_DV_BOX == pm->pmt))// || (PMT_FRAGMENT_BOX == pm->pmt))
  {
    struct PendingMessage *root = pm->frag_parent;
 
    while (NULL != root->frag_parent)
      root = root->frag_parent;
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Next attempt for root message <%" PRIu64 "> set to %s\n",
                root->logging_uuid,
                GNUNET_STRINGS_absolute_time_to_string (next_attempt));
    root->next_attempt = next_attempt;
    reorder_root_pm (root, next_attempt);
  }
  else
  {
    struct PendingMessage *root = pm->frag_parent;
 
    while (NULL != root->frag_parent && PMT_DV_BOX != root->pmt)
      root = root->frag_parent;
 
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "frag_count next attempt %u\n",
                root->frag_count);
 
    if (GNUNET_NO == root->frags_in_flight)
    {
      root->next_attempt = next_attempt;
      harmonize_flight_round (root);
      root->frags_in_flight_round++;
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "Next attempt for fragmented message <%" PRIu64 "> (<%" PRIu64
                  ">)set to %" PRIu64 "\n",
                  pm->logging_uuid,
                  root->logging_uuid,
                  next_attempt.abs_value_us);
    }
 
    pm->next_attempt = root->next_attempt;
    pm->frags_in_flight_round = root->frags_in_flight_round;
    harmonize_flight_round (pm);
 
    if (root->bytes_msg == root->frag_off)
      root->frags_in_flight = check_next_attempt_tree (root, root);
    else
      root->frags_in_flight = GNUNET_YES;
 
    if (GNUNET_NO == root->frags_in_flight)
    {
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "We have no fragments in flight for message %" PRIu64
                  ", reorder root! Next attempt is %" PRIu64 "\n",
                  root->logging_uuid,
                  root->next_attempt.abs_value_us);
      if (PMT_DV_BOX == root->pmt)
        root = root->frag_parent;
      reorder_root_pm (root, root->next_attempt);
      // root->next_attempt = GNUNET_TIME_UNIT_ZERO_ABS;
    }
    else
    {
      double factor = ((double) root->frag_count - 1)
                      / (double) root->frag_count;
      struct GNUNET_TIME_Relative s1;
      struct GNUNET_TIME_Relative s2;
      struct GNUNET_TIME_Relative plus_mean =
        GNUNET_TIME_absolute_get_remaining (root->next_attempt);
      struct GNUNET_TIME_Relative plus = GNUNET_TIME_absolute_get_remaining (
        next_attempt);
 
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "frag_count %u after factor\n",
                  root->frag_count);
      s1 = GNUNET_TIME_relative_multiply_double (plus_mean,
                                                 factor);
      s2 = GNUNET_TIME_relative_divide (plus,
                                        root->frag_count);
      plus_mean = GNUNET_TIME_relative_add (s1, s2);
      root->next_attempt = GNUNET_TIME_relative_to_absolute (plus_mean);
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "We have fragments in flight for message %" PRIu64
                  ", do not reorder root! Actual next attempt %" PRIu64 "\n",
                  root->logging_uuid,
                  root->next_attempt.abs_value_us);
    }
  }
}
 
 
struct PendingMessageScoreContext
{
  struct PendingMessage *best;
 
  struct DistanceVectorHop *dvh;
 
  size_t real_overhead;
 
  unsigned int consideration_counter;
 
  int frag;
 
  int relb;
 
  int to_early;
 
  unsigned int frags_in_flight;
 
  struct GNUNET_TIME_Relative to_early_retry_delay;
};
 
 
static void
select_best_pending_from_link (struct PendingMessageScoreContext *sc,
                               struct Queue *queue,
                               struct VirtualLink *vl,
                               struct DistanceVectorHop *dvh,
                               size_t overhead)
{
  struct GNUNET_TIME_Absolute now;
 
  now = GNUNET_TIME_absolute_get ();
  sc->to_early = GNUNET_NO;
  sc->frags_in_flight = GNUNET_NO;
  for (struct PendingMessage *pos = vl->pending_msg_head; NULL != pos;
       pos = pos->next_vl)
  {
    size_t real_overhead = overhead;
    int frag;
    int relb;
 
    if ((NULL != dvh) && (PMT_DV_BOX == pos->pmt))
    {
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "DV messages must not be DV-routed to next hop!\n");
      continue;   /* DV messages must not be DV-routed to next hop! */
    }
    if (pos->next_attempt.abs_value_us > now.abs_value_us)
    {
      if (GNUNET_YES == pos->frags_in_flight)
      {
        sc->frags_in_flight = GNUNET_YES;
        GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                    "Fragments in flight for message %" PRIu64 "\n",
                    pos->logging_uuid);
      }
      else
      {
        GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                    "Maybe too early, because message are sorted by next_attempt, if there are no fragments in flight.Checked message %"
                    PRIu64 "\n",
                    pos->logging_uuid);
        sc->to_early = GNUNET_YES;
        sc->to_early_retry_delay = GNUNET_TIME_absolute_get_remaining (
          pos->next_attempt);
        continue;
      }
      // break;   /* too early for all messages, they are sorted by next_attempt */
    }
    if (NULL != pos->qe)
    {
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "not eligible\n");
      continue;   /* not eligible */
    }
    sc->consideration_counter++;
    /* determine if we have to fragment, if so add fragmentation
       overhead! */
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "check %" PRIu64 " for sc->best\n",
                pos->logging_uuid);
    frag = GNUNET_NO;
    if (((0 != queue->mtu) &&
         (pos->bytes_msg + real_overhead > queue->mtu)) ||
        (pos->bytes_msg > UINT16_MAX - sizeof(struct
                                              GNUNET_TRANSPORT_SendMessageTo))
        ||
        (NULL != pos->head_frag /* fragments already exist, should
                                     respect that even if MTU is UINT16_MAX for
                                     this queue */))
    {
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "fragment msg with size %u, realoverhead is %lu\n",
                  pos->bytes_msg,
                  real_overhead);
      frag = GNUNET_YES;
      if (GNUNET_TRANSPORT_CC_RELIABLE == queue->tc->details.communicator.cc)
      {
        /* FIXME-FRAG-REL-UUID: we could use an optimized, shorter fragmentation
           header without the ACK UUID when using a *reliable* channel! */
      }
      real_overhead = overhead + sizeof(struct TransportFragmentBoxMessage);
    }
    /* determine if we have to reliability-box, if so add reliability box
       overhead */
    relb = GNUNET_NO;
    if ((GNUNET_NO == frag) &&
        (0 == (pos->prefs & GNUNET_MQ_PREF_UNRELIABLE)) &&
        (GNUNET_TRANSPORT_CC_RELIABLE != queue->tc->details.communicator.cc))
    {
      real_overhead += sizeof(struct TransportReliabilityBoxMessage);
 
      if ((0 != queue->mtu) && (pos->bytes_msg + real_overhead > queue->mtu))
      {
        frag = GNUNET_YES;
        real_overhead = overhead + sizeof(struct TransportFragmentBoxMessage);
      }
      else
      {
        relb = GNUNET_YES;
      }
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "Create reliability box of msg with size %u, realoverhead is %lu %u %u %u\n",
                  pos->bytes_msg,
                  real_overhead,
                  queue->mtu,
                  frag,
                  relb);
    }
 
    /* Finally, compare to existing 'best' in sc to see if this 'pos' pending
       message would beat it! */
    if (GNUNET_NO == sc->frags_in_flight && NULL != sc->best)
    {
      /* CHECK if pos fits queue BETTER (=smaller) than pm, if not: continue;
         OPTIMIZE-ME: This is a heuristic, which so far has NOT been
         experimentally validated. There may be some huge potential for
         improvement here. Also, we right now only compare how well the
         given message fits _this_ queue, and do not consider how well other
         queues might suit the message. Taking other queues into consideration
         may further improve the result, but could also be expensive
         in terms of CPU time.  */
      long long sc_score = sc->frag * 40 + sc->relb * 20 + sc->real_overhead;
      long long pm_score = frag * 40 + relb * 20 + real_overhead;
      long long time_delta =
        (sc->best->next_attempt.abs_value_us - pos->next_attempt.abs_value_us)
        / 1000LL;
 
      /* "time_delta" considers which message has been 'ready' for transmission
         for longer, if a message has a preference for low latency, increase
         the weight of the time_delta by 10x if it is favorable for that message */
      if ((0 != (pos->prefs & GNUNET_MQ_PREF_LOW_LATENCY)) &&
          (0 != (sc->best->prefs & GNUNET_MQ_PREF_LOW_LATENCY)))
        time_delta *= 10;     /* increase weight (always, both are low latency) */
      else if ((0 != (pos->prefs & GNUNET_MQ_PREF_LOW_LATENCY)) &&
               (time_delta > 0))
        time_delta *= 10;     /* increase weight, favors 'pos', which is low latency */
      else if ((0 != (sc->best->prefs & GNUNET_MQ_PREF_LOW_LATENCY)) &&
               (time_delta < 0))
        time_delta *= 10;     /* increase weight, favors 'sc->best', which is low latency */
      if (0 != queue->mtu)
      {
        /* Grant bonus if we are below MTU, larger bonus the closer we will
           be to the MTU */
        if (queue->mtu > sc->real_overhead + sc->best->bytes_msg)
          sc_score -= queue->mtu - (sc->real_overhead + sc->best->bytes_msg);
        if (queue->mtu > real_overhead + pos->bytes_msg)
          pm_score -= queue->mtu - (real_overhead + pos->bytes_msg);
      }
      if (sc_score + time_delta > pm_score)
      {
        GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                    "sc_score of %" PRIu64 " larger, keep sc->best %" PRIu64
                    "\n",
                    pos->logging_uuid,
                    sc->best->logging_uuid);
        continue;     /* sc_score larger, keep sc->best */
      }
    }
    sc->best = pos;
    sc->dvh = dvh;
    sc->frag = frag;
    sc->relb = relb;
    sc->real_overhead = real_overhead;
  }
}
 
 
static void
extract_box_cb (void *cls,
                struct Neighbour *next_hop,
                const struct GNUNET_MessageHeader *hdr,
                enum RouteMessageOptions options)
{
  struct PendingMessageScoreContext *sc = cls;
  struct PendingMessage *pm = sc->best;
  struct PendingMessage *bpm;
  uint16_t bsize = ntohs (hdr->size);
 
  GNUNET_assert (NULL == pm->bpm);
  bpm = GNUNET_malloc (sizeof(struct PendingMessage) + bsize);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "5 created pm %p storing vl %p from pm %p\n",
              bpm,
              pm->vl,
              pm);
  bpm->logging_uuid = logging_uuid_gen++;
  bpm->pmt = PMT_DV_BOX;
  bpm->vl = pm->vl;
  bpm->timeout = pm->timeout;
  bpm->bytes_msg = bsize;
  bpm->frag_parent = pm;
  set_pending_message_uuid (bpm);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Creating DV Box %" PRIu64 " for original message %" PRIu64
              " (next hop is %s)\n",
              bpm->logging_uuid,
              pm->logging_uuid,
              GNUNET_i2s (&next_hop->pid));
  memcpy (&bpm[1], hdr, bsize);
  pm->bpm = bpm;
}
 
 
static void
transmit_on_queue (void *cls)
{
  struct Queue *queue = cls;
  struct Neighbour *n = queue->neighbour;
  struct PendingMessageScoreContext sc;
  struct PendingMessage *pm;
 
  queue->transmit_task = NULL;
  if (NULL == n->vl)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Virtual link `%s' is down, cannot have PM for queue `%s'\n",
                GNUNET_i2s (&n->pid),
                queue->address);
    queue->idle = GNUNET_YES;
    return;
  }
  memset (&sc, 0, sizeof(sc));
  select_best_pending_from_link (&sc, queue, n->vl, NULL, 0);
  if (NULL == sc.best)
  {
    /* Also look at DVH that have the n as first hop! */
    for (struct DistanceVectorHop *dvh = n->dv_head; NULL != dvh;
         dvh = dvh->next_neighbour)
    {
      select_best_pending_from_link (&sc,
                                     queue,
                                     dvh->dv->vl,
                                     dvh,
                                     sizeof(struct GNUNET_PeerIdentity)
                                     * (1 + dvh->distance)
                                     + sizeof(struct TransportDVBoxMessage)
                                     + sizeof(struct TransportDVBoxPayloadP));
    }
  }
  if (NULL == sc.best)
  {
    /* no message pending, nothing to do here! */
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "No pending messages, queue `%s' to %s now idle\n",
                queue->address,
                GNUNET_i2s (&n->pid));
    if (GNUNET_YES == sc.to_early)
      schedule_transmit_on_queue (sc.to_early_retry_delay,
                                  queue,
                                  GNUNET_SCHEDULER_PRIORITY_DEFAULT);
    queue->idle = GNUNET_YES;
    return;
  }
  /* There is a message pending, we are certainly not idle */
  queue->idle = GNUNET_NO;
 
  /* Given selection in `sc`, do transmission */
  pm = sc.best;
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Selected message <%" PRIu64 ">\n",
              pm->logging_uuid);
  if (NULL != sc.dvh)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Is this %u a DV box?\n",
                pm->pmt);
    GNUNET_assert (PMT_DV_BOX != pm->pmt);
    if ((NULL != sc.best->bpm) && (sc.best->bpm->used_dvh != sc.dvh))
    {
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "Discard old box, because we have a new DV path.\n");
      free_pending_message (sc.best->bpm);
      sc.best->bpm = NULL;
    }
 
    if (NULL == sc.best->bpm)
    {
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "encapsulate_for_dv 2\n");
      encapsulate_for_dv (sc.dvh->dv,
                          1,
                          &sc.dvh,
                          (const struct GNUNET_MessageHeader *) &sc.best[1],
                          &extract_box_cb,
                          &sc,
                          RMO_NONE,
                          GNUNET_NO);
      GNUNET_assert (NULL != sc.best->bpm);
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "%lu %lu %lu %lu %u\n",
                  sizeof(struct GNUNET_PeerIdentity),
                  sizeof(struct TransportDVBoxMessage),
                  sizeof(struct TransportDVBoxPayloadP),
                  sizeof(struct TransportFragmentBoxMessage),
                  ((const struct GNUNET_MessageHeader *) &sc.best[1])->size);
      sc.best->bpm->used_dvh = sc.dvh;
    }
    pm = sc.best->bpm;
  }
  if (GNUNET_YES == sc.frag)
  {
    pm = fragment_message (queue, sc.dvh, pm);
    if (NULL == pm)
    {
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "Fragmentation failed queue %s to %s for <%" PRIu64
                  ">, trying again\n",
                  queue->address,
                  GNUNET_i2s (&n->pid),
                  sc.best->logging_uuid);
      schedule_transmit_on_queue (GNUNET_TIME_UNIT_ZERO,
                                  queue,
                                  GNUNET_SCHEDULER_PRIORITY_DEFAULT);
      return;
    }
  }
  else if (GNUNET_YES == sc.relb)
  {
    pm = reliability_box_message (queue, sc.dvh, pm);
    if (NULL == pm)
    {
      /* Reliability boxing failed, try next message... */
      GNUNET_log (
        GNUNET_ERROR_TYPE_DEBUG,
        "Reliability boxing failed queue %s to %s for <%" PRIu64
        ">, trying again\n",
        queue->address,
        GNUNET_i2s (&n->pid),
        sc.best->logging_uuid);
      schedule_transmit_on_queue (GNUNET_TIME_UNIT_ZERO,
                                  queue,
                                  GNUNET_SCHEDULER_PRIORITY_DEFAULT);
      return;
    }
  }
 
  /* Pass 'pm' for transission to the communicator */
  GNUNET_log (
    GNUNET_ERROR_TYPE_DEBUG,
    "Passing message <%" PRIu64
    "> to queue %s for peer %s (considered %u others)\n",
    pm->logging_uuid,
    queue->address,
    GNUNET_i2s (&n->pid),
    sc.consideration_counter);
 
  /* Flow control: increment amount of traffic sent; if we are routing
     via DV (and thus the ultimate target of the pending message is for
     a different virtual link than the one of the queue), then we need
     to use up not only the window of the direct link but also the
     flow control window for the DV link! */
  pm->vl->outbound_fc_window_size_used += pm->bytes_msg;
 
  if (pm->vl != queue->neighbour->vl)
  {
    /* If the virtual link of the queue differs, this better be distance
       vector routing! */
    GNUNET_assert (NULL != sc.dvh);
    /* If we do distance vector routing, we better not do this for a
       message that was itself DV-routed */
    GNUNET_assert (PMT_DV_BOX != sc.best->pmt);
    /* We use the size of the unboxed message here, to avoid counting
       the DV-Box header which is eaten up on the way by intermediaries */
    queue->neighbour->vl->outbound_fc_window_size_used += sc.best->bytes_msg;
  }
  else
  {
    GNUNET_assert (NULL == sc.dvh);
  }
 
  queue_send_msg (queue, pm, &pm[1], pm->bytes_msg);
 
  /* Check if this transmission somehow conclusively finished handing 'pm'
     even without any explicit ACKs */
  if ((PMT_CORE == pm->pmt) ||
      (GNUNET_TRANSPORT_CC_RELIABLE == queue->tc->details.communicator.cc))
  {
    completed_pending_message (pm);
  }
  else
  {
    struct GNUNET_TIME_Relative wait_duration;
    unsigned int wait_multiplier;
 
    if (PMT_FRAGMENT_BOX == pm->pmt)
    {
      struct PendingMessage *root;
 
      root = pm->frag_parent;
      while (NULL != root->frag_parent && PMT_DV_BOX != root->pmt)
        root = root->frag_parent;
 
      wait_multiplier =  (unsigned int) ceil ((double) root->bytes_msg
                                              / ((double) root->frag_off
                                                 / (double) root->frag_count))
                        * 4;
    }
    else
    {
      // No fragments, we use 4 RTT before retransmitting.
      wait_multiplier = 4;
    }
 
    // Depending on how much pending message the VirtualLink is queueing, we wait longer.
    // wait_multiplier = wait_multiplier * pm->vl->pending_msg_num;
 
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Wait multiplier %u\n",
                wait_multiplier);
 
    /* Message not finished, waiting for acknowledgement.
       Update time by which we might retransmit 's' based on queue
       characteristics (i.e. RTT); it takes one RTT for the message to
       arrive and the ACK to come back in the best case; but the other
       side is allowed to delay ACKs by 2 RTTs, so we use 4 RTT before
       retransmitting.
 
       OPTIMIZE: Note that in the future this heuristic should likely
       be improved further (measure RTT stability, consider message
       urgency and size when delaying ACKs, etc.) */
 
    if (GNUNET_TIME_UNIT_FOREVER_REL.rel_value_us !=
        queue->pd.aged_rtt.rel_value_us)
      wait_duration = queue->pd.aged_rtt;
    else
    {
      wait_duration = DEFAULT_ACK_WAIT_DURATION;
      wait_multiplier = 4;
    }
    {
      struct GNUNET_TIME_Absolute next = GNUNET_TIME_relative_to_absolute (
        GNUNET_TIME_relative_multiply (
          wait_duration, wait_multiplier));
      struct GNUNET_TIME_Relative plus = GNUNET_TIME_relative_multiply (
        wait_duration, wait_multiplier);
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "Waiting %s for ACK until %s\n",
                  GNUNET_STRINGS_relative_time_to_string (plus, GNUNET_NO),
                  GNUNET_STRINGS_absolute_time_to_string (next));
      update_pm_next_attempt (pm,
                              GNUNET_TIME_relative_to_absolute (
                                GNUNET_TIME_relative_multiply (wait_duration,
                                                               wait_multiplier))
                              );
    }
  }
  /* finally, re-schedule queue transmission task itself */
  schedule_transmit_on_queue (GNUNET_TIME_UNIT_ZERO,
                              queue,
                              GNUNET_SCHEDULER_PRIORITY_DEFAULT);
}
 
 
static void
handle_del_queue_message (void *cls,
                          const struct GNUNET_TRANSPORT_DelQueueMessage *dqm)
{
  struct TransportClient *tc = cls;
 
  if (CT_COMMUNICATOR != tc->type)
  {
    GNUNET_break (0);
    GNUNET_SERVICE_client_drop (tc->client);
    return;
  }
  for (struct Queue *queue = tc->details.communicator.queue_head; NULL != queue;
       queue = queue->next_client)
  {
    struct Neighbour *neighbour = queue->neighbour;
 
    if ((ntohl (dqm->qid) != queue->qid) ||
        (0 != GNUNET_memcmp (&dqm->receiver, &neighbour->pid)))
      continue;
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Dropped queue %s to peer %s\n",
                queue->address,
                GNUNET_i2s (&neighbour->pid));
    free_queue (queue);
    GNUNET_SERVICE_client_continue (tc->client);
    return;
  }
  GNUNET_break (0);
  GNUNET_SERVICE_client_drop (tc->client);
}
 
 
static void
free_queue_entry (struct QueueEntry *qe,
                  struct TransportClient *tc)
{
  struct PendingMessage *pm;
 
  GNUNET_CONTAINER_DLL_remove (qe->queue->queue_head,
                               qe->queue->queue_tail,
                               qe);
  qe->queue->queue_length--;
  tc->details.communicator.total_queue_length--;
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Received ACK on queue %s (QID %u) to peer %s (new length: %u/%u)\n",
              qe->queue->address,
              qe->queue->qid,
              GNUNET_i2s (&qe->queue->neighbour->pid),
              qe->queue->queue_length,
              tc->details.communicator.total_queue_length);
 
  /* if applicable, resume transmissions that waited on ACK */
  if (COMMUNICATOR_TOTAL_QUEUE_LIMIT - 1 ==
      tc->details.communicator.total_queue_length)
  {
    /* Communicator dropped below threshold, resume all queues
       incident with this client! */
    GNUNET_STATISTICS_update (
      GST_stats,
      "# Transmission throttled due to communicator queue limit",
      -1,
      GNUNET_NO);
    for (struct Queue *queue = tc->details.communicator.queue_head;
         NULL != queue;
         queue = queue->next_client)
    {
      schedule_transmit_on_queue (GNUNET_TIME_UNIT_ZERO,
                                  queue,
                                  GNUNET_SCHEDULER_PRIORITY_DEFAULT);
    }
  }
  else if (QUEUE_LENGTH_LIMIT - 1 == qe->queue->queue_length)
  {
    /* queue dropped below threshold; only resume this one queue */
    GNUNET_STATISTICS_update (GST_stats,
                              "# Transmission throttled due to queue queue limit",
                              -1,
                              GNUNET_NO);
    schedule_transmit_on_queue (GNUNET_TIME_UNIT_ZERO,
                                qe->queue,
                                GNUNET_SCHEDULER_PRIORITY_DEFAULT);
  }
  else if (1 == qe->queue->q_capacity)
  {
    // TODO I guess this will never happen, because the communicator triggers this by updating its queue length itself.
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Transmission rescheduled due to communicator message queue with qid %u has capacity %"
                PRIu64 ".\n",
                qe->queue->qid,
                qe->queue->q_capacity);
    /* message queue has capacity; only resume this one queue */
    /* queue dropped below threshold; only resume this one queue */
    GNUNET_STATISTICS_update (GST_stats,
                              "# Transmission throttled due to message queue capacity",
                              -1,
                              GNUNET_NO);
    schedule_transmit_on_queue (GNUNET_TIME_UNIT_ZERO,
                                qe->queue,
                                GNUNET_SCHEDULER_PRIORITY_DEFAULT);
  }
 
  if (NULL != (pm = qe->pm))
  {
    struct VirtualLink *vl;
 
    // GNUNET_assert (qe == pm->qe);
    pm->qe = NULL;
    /* If waiting for this communicator may have blocked transmission
       of pm on other queues for this neighbour, force schedule
       transmit on queue for queues of the neighbour */
    if (NULL == pm->frag_parent)
    {
      vl = pm->vl;
      if ((NULL != vl) &&
          (NULL != vl->pending_msg_head) &&
          (vl->pending_msg_head == pm))
        check_vl_transmission (vl);
    }
  }
  GNUNET_free (qe);
}
 
 
static void
handle_send_message_ack (void *cls,
                         const struct GNUNET_TRANSPORT_SendMessageToAck *sma)
{
  struct TransportClient *tc = cls;
  struct QueueEntry *qe;
 
  if (CT_COMMUNICATOR != tc->type)
  {
    GNUNET_break (0);
    GNUNET_SERVICE_client_drop (tc->client);
    return;
  }
 
  /* find our queue entry matching the ACK */
  qe = NULL;
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Looking for queue for PID %s\n",
              GNUNET_i2s (&sma->receiver));
  for (struct Queue *queue = tc->details.communicator.queue_head; NULL != queue;
       queue = queue->next_client)
  {
    if (0 != GNUNET_memcmp (&queue->neighbour->pid, &sma->receiver))
      continue;
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Found PID %s\n",
                GNUNET_i2s (&queue->neighbour->pid));
 
 
    for (struct QueueEntry *qep = queue->queue_head; NULL != qep;
         qep = qep->next)
    {
      if (qep->mid != GNUNET_ntohll (sma->mid) || queue->qid != ntohl (
            sma->qid))
        continue;
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "QueueEntry MID: %" PRIu64 " on queue QID: %u, Ack MID: %"
                  PRIu64 " Ack QID %u\n",
                  qep->mid,
                  queue->qid,
                  GNUNET_ntohll (sma->mid),
                  ntohl (sma->qid));
      qe = qep;
      if ((NULL != qe->pm) && (qe->pm->qe != qe))
        GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                    "For pending message %" PRIu64 " we had retransmissions.\n",
                    qe->pm->logging_uuid);
      break;
    }
  }
  if (NULL == qe)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "No QueueEntry found for Ack MID %" PRIu64 " QID: %u\n",
                GNUNET_ntohll (sma->mid),
                ntohl (sma->qid));
    // TODO I guess this can happen, if the Ack from the peer comes before the Ack from the queue.
    // Update: Maybe QueueEntry was accidentally freed during freeing PendingMessage.
    /* this should never happen */
    // GNUNET_break (0);
    // GNUNET_SERVICE_client_drop (tc->client);
    GNUNET_SERVICE_client_continue (tc->client);
    return;
  }
  free_queue_entry (qe, tc);
  GNUNET_SERVICE_client_continue (tc->client);
}
 
 
static void
handle_burst_finished (void *cls,
                       const struct GNUNET_TRANSPORT_BurstFinished *bf)
{
  burst_running = GNUNET_NO;
}
 
 
static int
notify_client_queues (void *cls,
                      const struct GNUNET_PeerIdentity *pid,
                      void *value)
{
  struct TransportClient *tc = cls;
  struct Neighbour *neighbour = value;
 
  GNUNET_assert (CT_MONITOR == tc->type);
  for (struct Queue *q = neighbour->queue_head; NULL != q;
       q = q->next_neighbour)
  {
    struct MonitorEvent me = { .rtt = q->pd.aged_rtt,
                               .cs = q->cs,
                               .num_msg_pending = q->num_msg_pending,
                               .num_bytes_pending = q->num_bytes_pending };
 
    notify_monitor (tc, pid, q->address, q->nt, &me);
  }
  return GNUNET_OK;
}
 
 
static void
handle_monitor_start (void *cls,
                      const struct GNUNET_TRANSPORT_MonitorStart *start)
{
  struct TransportClient *tc = cls;
 
  if (CT_NONE != tc->type)
  {
    GNUNET_break (0);
    GNUNET_SERVICE_client_drop (tc->client);
    return;
  }
  tc->type = CT_MONITOR;
  tc->details.monitor.peer = start->peer;
  tc->details.monitor.one_shot = ntohl (start->one_shot);
  GNUNET_CONTAINER_multipeermap_iterate (neighbours, &notify_client_queues, tc);
  GNUNET_SERVICE_client_mark_monitor (tc->client);
  GNUNET_SERVICE_client_continue (tc->client);
}
 
 
static struct TransportClient *
lookup_communicator (const char *prefix)
{
  for (struct TransportClient *tc = clients_head; NULL != tc; tc = tc->next)
  {
    if (CT_COMMUNICATOR != tc->type)
      continue;
    if (0 == strcmp (prefix, tc->details.communicator.address_prefix))
      return tc;
  }
  GNUNET_log (
    GNUNET_ERROR_TYPE_WARNING,
    "Someone suggested use of communicator for `%s', but we do not have such a communicator!\n",
    prefix);
  return NULL;
}
 
 
static void
suggest_to_connect (const struct GNUNET_PeerIdentity *pid, const char *address)
{
  static uint32_t idgen = 0;
  struct TransportClient *tc;
  char *prefix;
  struct GNUNET_TRANSPORT_CreateQueue *cqm;
  struct GNUNET_MQ_Envelope *env;
  size_t alen;
 
  prefix = GNUNET_HELLO_address_to_prefix (address);
  if (NULL == prefix)
  {
    GNUNET_break (0);  /* We got an invalid address!? */
    return;
  }
  tc = lookup_communicator (prefix);
  if (NULL == tc)
  {
    GNUNET_STATISTICS_update (GST_stats,
                              "# Suggestions ignored due to missing communicator",
                              1,
                              GNUNET_NO);
    GNUNET_log (GNUNET_ERROR_TYPE_INFO,
                "Cannot connect to %s at `%s', no matching communicator present\n",
                GNUNET_i2s (pid),
                address);
    GNUNET_free (prefix);
    return;
  }
  /* forward suggestion for queue creation to communicator */
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Request #%u for `%s' communicator to create queue to `%s' at `%s'\n",
              (unsigned int) idgen,
              prefix,
              GNUNET_i2s (pid),
              address);
  GNUNET_free (prefix);
  alen = strlen (address) + 1;
  env =
    GNUNET_MQ_msg_extra (cqm, alen, GNUNET_MESSAGE_TYPE_TRANSPORT_QUEUE_CREATE);
  cqm->request_id = htonl (idgen++);
  cqm->receiver = *pid;
  memcpy (&cqm[1], address, alen);
  GNUNET_MQ_send (tc->mq, env);
}
 
 
static void
validation_transmit_on_queue (struct Queue *q, struct ValidationState *vs)
{
  struct TransportValidationChallengeMessage tvc;
  struct GNUNET_TIME_Absolute monotonic_time;
 
  if (NULL != vs->revalidation_task)
  {
    GNUNET_SCHEDULER_cancel (vs->revalidation_task);
    vs->revalidation_task = NULL;
  }
  /*memcpy (&hkey,
          &hc,
          sizeof (hkey));*/
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Remove key %s for address %s map size %u contains %u\n",
              GNUNET_h2s (&vs->hc),
              vs->address,
              GNUNET_CONTAINER_multihashmap_size (revalidation_map),
              GNUNET_CONTAINER_multihashmap_contains (revalidation_map,
                                                      &vs->hc));
  GNUNET_CONTAINER_multihashmap_remove (revalidation_map, &vs->hc, vs);
 
  monotonic_time  = GNUNET_TIME_absolute_get_monotonic (GST_cfg);
  if (GNUNET_TIME_UNIT_ZERO_ABS.abs_value_us ==
      vs->last_challenge_use.abs_value_us)
  {
    vs->first_challenge_use = monotonic_time;
  }
  vs->last_challenge_use = monotonic_time;
  tvc.header.type =
    htons (GNUNET_MESSAGE_TYPE_TRANSPORT_ADDRESS_VALIDATION_CHALLENGE);
  tvc.header.size = htons (sizeof(tvc));
  tvc.reserved = htonl (0);
  tvc.challenge = vs->challenge;
  tvc.sender_time = GNUNET_TIME_absolute_hton (vs->last_challenge_use);
  GNUNET_log (GNUNET_ERROR_TYPE_INFO,
              "Sending address validation challenge %s to %s\n",
              GNUNET_sh2s (&tvc.challenge.value),
              GNUNET_i2s (&q->neighbour->pid));
  queue_send_msg (q, NULL, &tvc, sizeof(tvc));
}
 
 
static void
validation_start_cb (void *cls)
{
  struct ValidationState *vs;
  struct Queue *q;
  const struct GNUNET_TIME_Absolute now = GNUNET_TIME_absolute_get_monotonic (
    GST_cfg);
 
  (void) cls;
  validation_task = NULL;
  vs = GNUNET_CONTAINER_heap_peek (validation_heap);
  /* drop validations past their expiration */
  while (
    (NULL != vs) &&
    (0 == GNUNET_TIME_absolute_get_remaining (vs->valid_until).rel_value_us))
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Validation response %s cleaned up\n",
                GNUNET_sh2s (&vs->challenge.value));
    free_validation_state (vs);
    vs = GNUNET_CONTAINER_heap_peek (validation_heap);
  }
  if (NULL == vs)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_INFO,
                "Address validation task not scheduled anymore, nothing to do\n");
    return;   /* woopsie, no more addresses known, should only
                 happen if we're really a lonely peer */
  }
  q = find_queue (&vs->pid, vs->address);
  if (GNUNET_TIME_absolute_cmp (vs->first_challenge_use, >, now))
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "To early to start next address validation for challenge %s\n",
                GNUNET_sh2s (&vs->challenge.value));
    return;
  }
  if (NULL == q)
  {
    vs->awaiting_queue = GNUNET_YES;
    suggest_to_connect (&vs->pid, vs->address);
  }
  else
    validation_transmit_on_queue (q, vs);
  /* Finally, reschedule next attempt */
  vs->challenge_backoff =
    GNUNET_TIME_randomized_backoff (vs->challenge_backoff,
                                    MAX_VALIDATION_CHALLENGE_FREQ);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Address validation task will run again in %s\n",
              GNUNET_STRINGS_relative_time_to_string (vs->challenge_backoff,
                                                      GNUNET_YES));
  update_next_challenge_time (vs,
                              GNUNET_TIME_relative_to_absolute (
                                vs->challenge_backoff));
}
 
 
struct QueueQualityContext
{
  struct Queue *q;
 
  unsigned int quality_count;
 
  unsigned int num_queues;
 
  unsigned int k;
};
 
 
static int
check_connection_quality (void *cls,
                          const struct GNUNET_PeerIdentity *pid,
                          void *value)
{
  struct QueueQualityContext *ctx = cls;
  struct Neighbour *n = value;
  int do_inc;
 
  (void) pid;
  do_inc = GNUNET_NO;
  for (struct Queue *q = n->queue_head; NULL != q; q = q->next_neighbour)
  {
    ctx->num_queues++;
    if (0 == ctx->k--)
      ctx->q = q;
    /* FIXME-CONQ-STATISTICS: in the future, add reliability / goodput
       statistics and consider those as well here? */
    if (q->pd.aged_rtt.rel_value_us < DV_QUALITY_RTT_THRESHOLD.rel_value_us)
      do_inc = GNUNET_YES;
  }
  if (GNUNET_YES == do_inc)
    ctx->quality_count++;
  return GNUNET_OK;
}
 
 
static void
start_dv_learn (void *cls);
 
 
struct SignDvInitCls
{
  struct TransportDVLearnMessage dvl;
  struct LearnLaunchEntry *lle;
  struct QueueQualityContext qqc;
  struct PilsRequest *pr;
};
 
 
static void
sign_dv_init_cb (void *cls,
                 const struct GNUNET_PeerIdentity *pid,
                 const struct GNUNET_CRYPTO_EddsaSignature *sig)
{
  struct SignDvInitCls *sign_dv_init_cls = cls;
  struct TransportDVLearnMessage dvl = sign_dv_init_cls->dvl;
  struct LearnLaunchEntry *lle = sign_dv_init_cls->lle;
  struct QueueQualityContext qqc = sign_dv_init_cls->qqc;
 
  sign_dv_init_cls->pr->op = NULL;
  GNUNET_CONTAINER_DLL_remove (pils_requests_head,
                               pils_requests_tail,
                               sign_dv_init_cls->pr);
  GNUNET_free (sign_dv_init_cls->pr);
 
  dvl.init_sig = *sig;
  dvl.initiator = *GST_my_identity;
  dvl.challenge = lle->challenge;
 
  qqc.quality_count = 0;
  qqc.k = GNUNET_CRYPTO_random_u32 (GNUNET_CRYPTO_QUALITY_WEAK, qqc.num_queues);
  qqc.num_queues = 0;
  qqc.q = NULL;
  GNUNET_CONTAINER_multipeermap_iterate (neighbours,
                                         &check_connection_quality,
                                         &qqc);
  GNUNET_assert (NULL != qqc.q);
 
  /* Do this as close to transmission time as possible! */
  lle->launch_time = GNUNET_TIME_absolute_get ();
 
  queue_send_msg (qqc.q, NULL, &dvl, sizeof(dvl));
  /* reschedule this job, randomizing the time it runs (but no
     actual backoff!) */
  dvlearn_task = GNUNET_SCHEDULER_add_delayed (GNUNET_TIME_randomize (
                                                 DV_LEARN_BASE_FREQUENCY),
                                               &start_dv_learn,
                                               NULL);
}
 
 
static void
start_dv_learn (void *cls)
{
  struct LearnLaunchEntry *lle;
  struct QueueQualityContext qqc;
  struct TransportDVLearnMessage dvl;
 
  (void) cls;
  dvlearn_task = NULL;
  if (0 == GNUNET_CONTAINER_multipeermap_size (neighbours))
    return; /* lost all connectivity, cannot do learning */
  qqc.quality_count = 0;
  qqc.num_queues = 0;
  qqc.k = GNUNET_CONTAINER_multipeermap_size (neighbours);
  GNUNET_CONTAINER_multipeermap_iterate (neighbours,
                                         &check_connection_quality,
                                         &qqc);
  if (qqc.quality_count > DV_LEARN_QUALITY_THRESHOLD)
  {
    struct GNUNET_TIME_Relative delay;
    unsigned int factor;
 
    /* scale our retries by how far we are above the threshold */
    factor = qqc.quality_count / DV_LEARN_QUALITY_THRESHOLD;
    delay = GNUNET_TIME_relative_multiply (DV_LEARN_BASE_FREQUENCY, factor);
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "At connection quality %u, will launch DV learn in %s\n",
                qqc.quality_count,
                GNUNET_STRINGS_relative_time_to_string (delay, GNUNET_YES));
    dvlearn_task = GNUNET_SCHEDULER_add_delayed (delay, &start_dv_learn, NULL);
    return;
  }
  /* remove old entries in #dvlearn_map if it has grown too big */
  while (MAX_DV_LEARN_PENDING <=
         GNUNET_CONTAINER_multishortmap_size (dvlearn_map))
  {
    lle = lle_tail;
    GNUNET_assert (GNUNET_YES ==
                   GNUNET_CONTAINER_multishortmap_remove (dvlearn_map,
                                                          &lle->challenge.value,
                                                          lle));
    GNUNET_CONTAINER_DLL_remove (lle_head, lle_tail, lle);
    GNUNET_free (lle);
  }
  /* setup data structure for learning */
  lle = GNUNET_new (struct LearnLaunchEntry);
  GNUNET_CRYPTO_random_block (GNUNET_CRYPTO_QUALITY_NONCE,
                              &lle->challenge,
                              sizeof(lle->challenge));
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Starting launch DV learn with challenge %s\n",
              GNUNET_sh2s (&lle->challenge.value));
  GNUNET_CONTAINER_DLL_insert (lle_head, lle_tail, lle);
  GNUNET_break (GNUNET_YES ==
                GNUNET_CONTAINER_multishortmap_put (
                  dvlearn_map,
                  &lle->challenge.value,
                  lle,
                  GNUNET_CONTAINER_MULTIHASHMAPOPTION_UNIQUE_ONLY));
  dvl.header.type = htons (GNUNET_MESSAGE_TYPE_TRANSPORT_DV_LEARN);
  dvl.header.size = htons (sizeof(dvl));
  dvl.num_hops = htons (0);
  dvl.bidirectional = htons (0);
  dvl.non_network_delay = GNUNET_TIME_relative_hton (GNUNET_TIME_UNIT_ZERO);
  dvl.monotonic_time =
    GNUNET_TIME_absolute_hton (GNUNET_TIME_absolute_get_monotonic (GST_cfg));
  {
    struct DvInitPS dvip = {
      .purpose.purpose = htonl (
        GNUNET_SIGNATURE_PURPOSE_TRANSPORT_DV_INITIATOR),
      .purpose.size = htonl (sizeof(dvip)),
      .monotonic_time = dvl.monotonic_time,
      .challenge = lle->challenge
    };
    struct SignDvInitCls *sign_dv_init_cls;
 
    sign_dv_init_cls = GNUNET_new (struct SignDvInitCls);
    sign_dv_init_cls->dvl = dvl;
    sign_dv_init_cls->lle = lle;
    sign_dv_init_cls->qqc = qqc;
    sign_dv_init_cls->pr = GNUNET_new (struct PilsRequest);
    GNUNET_CONTAINER_DLL_insert (pils_requests_head,
                                 pils_requests_tail,
                                 sign_dv_init_cls->pr);
    sign_dv_init_cls->pr->op =
      GNUNET_PILS_sign_by_peer_identity (pils,
                                         &dvip.purpose,
                                         sign_dv_init_cb,
                                         sign_dv_init_cls);
  }
}
 
 
static char *
get_address_without_port (const char *address)
{
  const char *colon;
  char *colon_rest;
  size_t colon_rest_length;
  char *address_without_port;
 
  colon = strchr (address,':');
  colon_rest = GNUNET_strndup (address, colon - address);
  colon_rest_length = strlen (colon_rest);
  address_without_port = GNUNET_strndup (&colon_rest[4], colon_rest_length - 4);
  GNUNET_free (colon_rest);
 
  return address_without_port;
}
 
 
static int
check_validation_request_pending (void *cls,
                                  const struct GNUNET_PeerIdentity *pid,
                                  void *value)
{
  struct Queue *q = cls;
  struct ValidationState *vs = value;
  char *address_without_port_vs;
  char *address_without_port_q;
  int success = GNUNET_YES;
 
  // TODO Check if this is really necessary.
  address_without_port_vs = get_address_without_port (vs->address);
  address_without_port_q = get_address_without_port (q->address);
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Check validation request pending for `%s' at `%s'/`%s' (vs)/(q)\n",
              GNUNET_i2s (pid),
              address_without_port_vs,
              address_without_port_q);
  (void) pid;
  if ((GNUNET_YES == vs->awaiting_queue) &&
      (0 == strcmp (address_without_port_vs, address_without_port_q)))
  {
 
    vs->awaiting_queue = GNUNET_NO;
    validation_transmit_on_queue (q, vs);
    success = GNUNET_NO;
  }
 
  GNUNET_free (address_without_port_vs);
  GNUNET_free (address_without_port_q);
  return success;
}
 
 
static void
neighbour_dv_monotime_cb (void *cls,
                          const struct GNUNET_PEERSTORE_Record *record,
                          const char *emsg)
{
  struct Neighbour *n = cls;
  struct GNUNET_TIME_AbsoluteNBO *mtbe;
 
  (void) emsg;
  if (NULL == record)
  {
    /* we're done with #neighbour_dv_monotime_cb() invocations,
       continue normal processing */
    n->get = NULL;
    n->dv_monotime_available = GNUNET_YES;
    return;
  }
  if (0 == record->value_size)
  {
    GNUNET_PEERSTORE_iteration_next (n->get, 1);
    GNUNET_break (0);
    return;
  }
  mtbe = record->value;
  n->last_dv_learn_monotime =
    GNUNET_TIME_absolute_max (n->last_dv_learn_monotime,
                              GNUNET_TIME_absolute_ntoh (*mtbe));
  GNUNET_PEERSTORE_iteration_next (n->get, 1);
}
 
 
static void
iterate_address_and_compare_cb (void *cls,
                                const struct GNUNET_PeerIdentity *pid,
                                const char *uri)
{
  struct Queue *queue = cls;
  struct sockaddr_in v4;
  const char *slash;
  char *address_uri;
  char *prefix;
  char *uri_without_port;
  char *address_uri_without_port;
 
  slash = strrchr (uri, '/');
  prefix = GNUNET_strndup (uri, (slash - uri) - 2);
  GNUNET_assert (NULL != slash);
  slash++;
  GNUNET_asprintf (&address_uri,
                   "%s-%s",
                   prefix,
                   slash);
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "1 not global natted_address %u %s %s %s\n",
              queue->is_global_natted,
              uri,
              queue->address,
              slash);
 
  uri_without_port = get_address_without_port (address_uri);
  if (1 != inet_pton (AF_INET, uri_without_port, &v4.sin_addr))
  {
    GNUNET_free (prefix);
    GNUNET_free (address_uri);
    GNUNET_free (uri_without_port);
    return;
  }
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "2 not global natted_address %u %s %s\n",
              queue->is_global_natted,
              uri,
              queue->address);
 
  if (GNUNET_NO == queue->is_global_natted)
  {
    GNUNET_free (prefix);
    GNUNET_free (address_uri);
    GNUNET_free (uri_without_port);
    return;
  }
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "3 not global natted_address %u %s %s\n",
              queue->is_global_natted,
              uri,
              queue->address);
 
  if (0 == strcmp (uri_without_port, address_uri))
  {
    GNUNET_free (prefix);
    GNUNET_free (address_uri);
    GNUNET_free (uri_without_port);
    return;
  }
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "4 not global natted_address %u %s %s\n",
              queue->is_global_natted,
              uri,
              queue->address);
 
  address_uri_without_port = get_address_without_port (queue->address);
  if (0 == strcmp (uri_without_port, address_uri_without_port))
  {
    queue->is_global_natted = GNUNET_NO;
  }
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "not global natted_address %u %s %s %s %s %s %u\n",
              queue->is_global_natted,
              uri,
              queue->address,
              uri_without_port,
              address_uri_without_port,
              prefix,
              GNUNET_NO);
  GNUNET_free (prefix);
  GNUNET_free (address_uri);
  GNUNET_free (address_uri_without_port);
  GNUNET_free (uri_without_port);
}
 
 
struct TransportGlobalNattedAddressClosure
{
  char *addr;
 
  struct TransportGlobalNattedAddress *tgna;
};
 
 
static enum GNUNET_GenericReturnValue
contains_address (void *cls,
                  const struct GNUNET_PeerIdentity *pid,
                  void *value)
{
  struct TransportGlobalNattedAddressClosure *tgna_cls = cls;
  struct TransportGlobalNattedAddress *tgna = value;
  char *addr = (char *) &tgna[1];
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Checking tgna %p with addr %s and length %u compare length %lu\n",
              tgna,
              addr,
              ntohl (tgna->address_length),
              strlen (tgna_cls->addr));
  if (strlen (tgna_cls->addr) == ntohl (tgna->address_length)
      && 0 == strncmp (addr, tgna_cls->addr, ntohl (tgna->address_length)))
  {
    tgna_cls->tgna = tgna;
    return GNUNET_NO;
  }
  return GNUNET_YES;
}
 
 
static void
check_for_global_natted_error_cb (void *cls)
{
  GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
              "Error in PEERSTORE monitoring for checking global natted\n");
}
 
 
static void
check_for_global_natted_sync_cb (void *cls)
{
  GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
              "Done with initial PEERSTORE iteration during monitoring  for checking global natted\n");
}
 
 
static void
check_for_global_natted (void *cls,
                         const struct GNUNET_PEERSTORE_Record *record,
                         const char *emsg)
{
  struct Queue *queue = cls;
  struct Neighbour *neighbour = queue->neighbour;
  struct GNUNET_HELLO_Parser *parser;
  struct GNUNET_MessageHeader *hello;
  struct TransportGlobalNattedAddressClosure tgna_cls;
  size_t address_len_without_port;
 
  if (NULL != emsg)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
                "Got failure from PEERSTORE: %s\n",
                emsg);
    return;
  }
  if (0 == record->value_size)
  {
    GNUNET_PEERSTORE_monitor_next (queue->mo, 1);
    GNUNET_break (0);
    return;
  }
  queue->is_global_natted = GNUNET_YES;
  hello = record->value;
  parser = GNUNET_HELLO_parser_from_msg (hello);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "before not global natted %u\n",
              queue->is_global_natted);
  GNUNET_HELLO_parser_iterate (parser,
                               &iterate_address_and_compare_cb,
                               queue);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "after not global natted %u\n",
              queue->is_global_natted);
  GNUNET_HELLO_parser_free (parser);
 
  tgna_cls.addr = get_address_without_port (queue->address);
  address_len_without_port = strlen (tgna_cls.addr);
  /*{
    char buf[address_len_without_port + 1];
 
      GNUNET_memcpy (&buf, addr, address_len_without_port);
      buf[address_len_without_port] = '\0';
      GNUNET_free (addr);
      GNUNET_memcpy (tgna_cls.addr, buf, address_len_without_port + 1);
      }*/
  tgna_cls.tgna = NULL;
  GNUNET_CONTAINER_multipeermap_get_multiple (neighbour->natted_addresses,
                                              &neighbour->pid,
                                              &contains_address,
                                              &tgna_cls);
  if (NULL != tgna_cls.tgna)
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                " tgna_cls.tgna tgna %p %lu %u %u\n",
                tgna_cls.tgna,
                neighbour->size_of_global_addresses,
                ntohl (tgna_cls.tgna->address_length),
                neighbour->number_of_addresses);
  if (NULL == tgna_cls.tgna && GNUNET_YES == queue->is_global_natted)
  {
    struct TransportGlobalNattedAddress *tgna;
 
    tgna = GNUNET_malloc (sizeof (struct TransportGlobalNattedAddress)
                          + address_len_without_port);
    tgna->address_length = htonl (address_len_without_port);
    GNUNET_memcpy (&tgna[1], tgna_cls.addr, address_len_without_port);
    GNUNET_CONTAINER_multipeermap_put (neighbour->natted_addresses,
                                       &neighbour->pid,
                                       tgna,
                                       GNUNET_CONTAINER_MULTIHASHMAPOPTION_MULTIPLE);
    neighbour->number_of_addresses++;
    neighbour->size_of_global_addresses += address_len_without_port + 1;
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Created tgna %p with address %s and length %lu\n",
                tgna,
                tgna_cls.addr,
                address_len_without_port + 1);
  }
  else if (NULL != tgna_cls.tgna && GNUNET_NO == queue->is_global_natted)
  {
    GNUNET_CONTAINER_multipeermap_remove (neighbour->natted_addresses,
                                          &neighbour->pid,
                                          tgna_cls.tgna);
    GNUNET_assert (neighbour->size_of_global_addresses >= ntohl (tgna_cls.tgna->
                                                                 address_length)
                   );
    neighbour->size_of_global_addresses -= ntohl (tgna_cls.tgna->address_length)
    ;
    GNUNET_assert (0 < neighbour->number_of_addresses);
    neighbour->number_of_addresses--;
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "removed tgna %p\n",
                tgna_cls.tgna);
    GNUNET_free (tgna_cls.tgna);
  }
  GNUNET_PEERSTORE_monitor_next (queue->mo, 1);
  GNUNET_free (tgna_cls.addr);
}
 
 
static void
handle_add_queue_message (void *cls,
                          const struct GNUNET_TRANSPORT_AddQueueMessage *aqm)
{
  struct TransportClient *tc = cls;
  struct Queue *queue;
  struct Neighbour *neighbour;
  const char *addr;
  uint16_t addr_len;
 
  if (ntohl (aqm->mtu) <= sizeof(struct TransportFragmentBoxMessage))
  {
    /* MTU so small as to be useless for transmissions,
       required for #fragment_message()! */
    GNUNET_break_op (0);
    GNUNET_SERVICE_client_drop (tc->client);
    return;
  }
  /* This may simply be a queue update */
  for (queue = tc->details.communicator.queue_head;
       NULL != queue;
       queue = queue->next_client)
  {
    if (queue->qid != ntohl (aqm->qid))
      continue;
    break;
  }
 
  if (NULL != queue)
  {
    neighbour = queue->neighbour;
  }
  else
  {
    struct GNUNET_TIME_Absolute validated_until = GNUNET_TIME_UNIT_ZERO_ABS;
 
    neighbour = lookup_neighbour (&aqm->receiver);
    if (NULL == neighbour)
    {
      neighbour = GNUNET_new (struct Neighbour);
      neighbour->natted_addresses = GNUNET_CONTAINER_multipeermap_create (16,
                                                                          GNUNET_YES);
      neighbour->pid = aqm->receiver;
      GNUNET_assert (GNUNET_OK ==
                     GNUNET_CONTAINER_multipeermap_put (
                       neighbours,
                       &neighbour->pid,
                       neighbour,
                       GNUNET_CONTAINER_MULTIHASHMAPOPTION_UNIQUE_ONLY));
      neighbour->get =
        GNUNET_PEERSTORE_iteration_start (peerstore,
                                          "transport",
                                          &neighbour->pid,
                                          GNUNET_PEERSTORE_TRANSPORT_DVLEARN_MONOTIME,
                                          &neighbour_dv_monotime_cb,
                                          neighbour);
    }
    addr_len = ntohs (aqm->header.size) - sizeof(*aqm);
    addr = (const char *) &aqm[1];
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "New queue %s to %s available with QID %u and q_len %" PRIu64
                " and mtu %u\n",
                addr,
                GNUNET_i2s (&aqm->receiver),
                ntohl (aqm->qid),
                GNUNET_ntohll (aqm->q_len),
                ntohl (aqm->mtu));
    queue = GNUNET_malloc (sizeof(struct Queue) + addr_len);
    queue->tc = tc;
    for (struct Queue *q = neighbour->queue_head; NULL != q; q = q->
                                                                 next_neighbour)
      validated_until = GNUNET_TIME_absolute_max (validated_until, q->
                                                  validated_until);
    if (0 == GNUNET_TIME_absolute_get_remaining (validated_until).rel_value_us)
    {
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "New queue with QID %u inherit validated until\n",
                  ntohl (aqm->qid));
      queue->validated_until = validated_until;
    }
    queue->address = (const char *) &queue[1];
    queue->pd.aged_rtt = GNUNET_TIME_UNIT_FOREVER_REL;
    queue->qid = ntohl (aqm->qid);
    queue->neighbour = neighbour;
    if (GNUNET_TRANSPORT_QUEUE_LENGTH_UNLIMITED == GNUNET_ntohll (aqm->q_len))
      queue->unlimited_length = GNUNET_YES;
    queue->q_capacity = GNUNET_ntohll (aqm->q_len);
    memcpy (&queue[1], addr, addr_len);
    /* notify monitors about new queue */
    {
      struct MonitorEvent me = { .rtt = queue->pd.aged_rtt, .cs = queue->cs };
 
      notify_monitors (&neighbour->pid, queue->address, queue->nt, &me);
    }
    GNUNET_CONTAINER_MDLL_insert (neighbour,
                                  neighbour->queue_head,
                                  neighbour->queue_tail,
                                  queue);
    GNUNET_CONTAINER_MDLL_insert (client,
                                  tc->details.communicator.queue_head,
                                  tc->details.communicator.queue_tail,
                                  queue);
 
  }
  queue->mtu = ntohl (aqm->mtu);
  queue->nt = ntohl (aqm->nt);
  queue->cs = ntohl (aqm->cs);
  queue->idle = GNUNET_YES;
 
  {
    struct sockaddr_in v4;
    char *addr_without = get_address_without_port (queue->address);
    if (1 == inet_pton (AF_INET, addr_without, &v4.sin_addr))
    {
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "start not global natted\n");
      queue->mo = GNUNET_PEERSTORE_monitor_start (GST_cfg,
                                                  GNUNET_YES,
                                                  "peerstore",
                                                  &neighbour->pid,
                                                  GNUNET_PEERSTORE_HELLO_KEY,
                                                  &
                                                  check_for_global_natted_error_cb,
                                                  NULL,
                                                  &
                                                  check_for_global_natted_sync_cb,
                                                  NULL,
                                                  &check_for_global_natted,
                                                  queue);
    }
    GNUNET_free (addr_without);
  }
  /* check if valdiations are waiting for the queue */
  if (GNUNET_YES == GNUNET_CONTAINER_multipeermap_contains (validation_map,
                                                            &aqm->receiver))
  {
    if (GNUNET_SYSERR != GNUNET_CONTAINER_multipeermap_get_multiple (
          validation_map,
          &aqm->
          receiver,
          &
          check_validation_request_pending,
          queue))
      start_address_validation (&aqm->receiver, queue->address);
  }
  else
    start_address_validation (&aqm->receiver, queue->address);
  /* look for traffic for this queue */
  // TODO Check whether this makes any sense at all.
  /*schedule_transmit_on_queue (GNUNET_TIME_UNIT_ZERO,
    queue, GNUNET_SCHEDULER_PRIORITY_DEFAULT);*/
  /* might be our first queue, try launching DV learning */
  if (NULL == dvlearn_task)
    dvlearn_task = GNUNET_SCHEDULER_add_now (&start_dv_learn, NULL);
  GNUNET_SERVICE_client_continue (tc->client);
}
 
 
static void
handle_update_queue_message (void *cls,
                             const struct
                             GNUNET_TRANSPORT_UpdateQueueMessage *msg)
{
  struct TransportClient *tc = cls;
  struct Queue *target_queue = NULL;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Received queue update message for %u with q_len %llu and mtu %u\n",
              ntohl (msg->qid),
              (unsigned long long) GNUNET_ntohll (msg->q_len),
              ntohl (msg->mtu));
  for (target_queue = tc->details.communicator.queue_head;
       NULL != target_queue;
       target_queue = target_queue->next_client)
  {
    if (ntohl (msg->qid) == target_queue->qid)
      break;
  }
  if (NULL == target_queue)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
                "Queue to update no longer exists! Discarding update.\n");
    return;
  }
 
  target_queue->nt = msg->nt;
  target_queue->mtu = ntohl (msg->mtu);
  target_queue->cs = msg->cs;
  target_queue->priority = ntohl (msg->priority);
  /* The update message indicates how many messages
   * the queue should be able to handle.
   */
  if (GNUNET_TRANSPORT_QUEUE_LENGTH_UNLIMITED == GNUNET_ntohll (msg->q_len))
    target_queue->unlimited_length = GNUNET_YES;
  else
    target_queue->unlimited_length = GNUNET_NO;
  target_queue->q_capacity += GNUNET_ntohll (msg->q_len);
  if (0 < target_queue->q_capacity)
    schedule_transmit_on_queue (GNUNET_TIME_UNIT_ZERO,
                                target_queue,
                                GNUNET_SCHEDULER_PRIORITY_DEFAULT);
  GNUNET_SERVICE_client_continue (tc->client);
}
 
 
static void
handle_queue_create_ok (void *cls,
                        const struct GNUNET_TRANSPORT_CreateQueueResponse *cqr)
{
  struct TransportClient *tc = cls;
 
  if (CT_COMMUNICATOR != tc->type)
  {
    GNUNET_break (0);
    GNUNET_SERVICE_client_drop (tc->client);
    return;
  }
  GNUNET_STATISTICS_update (GST_stats,
                            "# Suggestions succeeded at communicator",
                            1,
                            GNUNET_NO);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Request #%u for communicator to create queue succeeded\n",
              (unsigned int) ntohs (cqr->request_id));
  GNUNET_SERVICE_client_continue (tc->client);
}
 
 
static void
handle_queue_create_fail (
  void *cls,
  const struct GNUNET_TRANSPORT_CreateQueueResponse *cqr)
{
  struct TransportClient *tc = cls;
 
  if (CT_COMMUNICATOR != tc->type)
  {
    GNUNET_break (0);
    GNUNET_SERVICE_client_drop (tc->client);
    return;
  }
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Request #%u for communicator to create queue failed\n",
              (unsigned int) ntohl (cqr->request_id));
  GNUNET_STATISTICS_update (GST_stats,
                            "# Suggestions failed in queue creation at communicator",
                            1,
                            GNUNET_NO);
  GNUNET_SERVICE_client_continue (tc->client);
}
 
 
static void
handle_suggest_cancel (void *cls, const struct ExpressPreferenceMessage *msg)
{
  struct TransportClient *tc = cls;
  struct PeerRequest *pr;
 
  if (CT_APPLICATION != tc->type)
  {
    GNUNET_break (0);
    GNUNET_SERVICE_client_drop (tc->client);
    return;
  }
  pr = GNUNET_CONTAINER_multipeermap_get (tc->details.application.requests,
                                          &msg->peer);
  if (NULL == pr)
  {
    GNUNET_break (0);
    GNUNET_SERVICE_client_drop (tc->client);
    return;
  }
  (void) stop_peer_request (tc, &pr->pid, pr);
  GNUNET_SERVICE_client_continue (tc->client);
}
 
 
static void
hello_for_client_cb (void *cls,
                     const struct GNUNET_PeerIdentity *pid,
                     const char *uri)
{
  struct Queue *q;
  int pfx_len;
  const char *eou;
  char *address;
  (void) cls;
 
  eou = strstr (uri,
                "://");
  pfx_len = eou - uri;
  eou += 3;
  GNUNET_asprintf (&address,
                   "%.*s-%s",
                   pfx_len,
                   uri,
                   eou);
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "hello for client %s\n",
              address);
 
  q = find_queue (pid, address);
  if (NULL == q)
  {
    suggest_to_connect (pid, address);
  }
  else
    start_address_validation (pid, address);
  GNUNET_free (address);
}
 
 
static void
handle_hello_for_client (void *cls,
                         const struct GNUNET_PEERSTORE_Record *record,
                         const char *emsg)
{
  struct PeerRequest *pr = cls;
  struct GNUNET_HELLO_Parser *parser;
  struct GNUNET_MessageHeader *hello;
 
  if (NULL != emsg)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
                "Got failure from PEERSTORE: %s\n",
                emsg);
    return;
  }
  if (NULL == GST_my_identity)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
                "No identity given yet!\n");
    return;
  }
  hello = record->value;
  if (0 == GNUNET_memcmp (&record->peer, GST_my_identity))
  {
    GNUNET_PEERSTORE_monitor_next (pr->nc, 1);
    return;
  }
  parser = GNUNET_HELLO_parser_from_msg (hello);
  if (NULL == parser)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
                "HELLO cannot be parsed!\n");
    return;
  }
  GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
              "HELLO for `%s' could be parsed, iterating addresses...!\n",
              GNUNET_i2s (GNUNET_HELLO_parser_get_id (parser)));
  GNUNET_HELLO_parser_iterate (parser,
                               hello_for_client_cb,
                               NULL);
  GNUNET_HELLO_parser_free (parser);
}
 
 
static void
hello_for_client_error_cb (void *cls)
{
  GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
              "Error in PEERSTORE monitoring\n");
}
 
 
static void
hello_for_client_sync_cb (void *cls)
{
  GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
              "Done with initial PEERSTORE iteration during monitoring\n");
}
 
 
static void
handle_suggest (void *cls, const struct ExpressPreferenceMessage *msg)
{
  struct TransportClient *tc = cls;
  struct PeerRequest *pr;
 
  if (CT_NONE == tc->type)
  {
    tc->type = CT_APPLICATION;
    tc->details.application.requests =
      GNUNET_CONTAINER_multipeermap_create (16, GNUNET_YES);
  }
  if (CT_APPLICATION != tc->type)
  {
    GNUNET_break (0);
    GNUNET_SERVICE_client_drop (tc->client);
    return;
  }
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Client suggested we talk to %s with preference %d at rate %u\n",
              GNUNET_i2s (&msg->peer),
              (int) ntohl (msg->pk),
              (int) ntohl (msg->bw.value__));
  if (0 == GNUNET_memcmp (GST_my_identity, &msg->peer))
  {
    GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
                "Client suggested connection to ourselves, ignoring...\n");
    return;
  }
  pr = GNUNET_new (struct PeerRequest);
  pr->tc = tc;
  pr->pid = msg->peer;
  pr->bw = msg->bw;
  pr->pk = ntohl (msg->pk);
  if (GNUNET_YES != GNUNET_CONTAINER_multipeermap_put (
        tc->details.application.requests,
        &pr->pid,
        pr,
        GNUNET_CONTAINER_MULTIHASHMAPOPTION_UNIQUE_ONLY))
  {
    GNUNET_break (0);
    GNUNET_free (pr);
    GNUNET_SERVICE_client_drop (tc->client);
    return;
  }
  pr->nc =
    GNUNET_PEERSTORE_monitor_start (GST_cfg,
                                    GNUNET_YES,
                                    "peerstore",
                                    NULL,
                                    GNUNET_PEERSTORE_HELLO_KEY,
                                    &hello_for_client_error_cb,
                                    NULL,
                                    &hello_for_client_sync_cb,
                                    NULL,
                                    &handle_hello_for_client,
                                    pr);
  GNUNET_SERVICE_client_continue (tc->client);
}
 
 
static int
check_request_hello_validation (void *cls,
                                const struct RequestHelloValidationMessage *m)
{
  (void) cls;
  GNUNET_MQ_check_zero_termination (m);
  return GNUNET_OK;
}
 
 
static void
handle_request_hello_validation (void *cls,
                                 const struct RequestHelloValidationMessage *m)
{
  struct TransportClient *tc = cls;
  struct Queue *q;
 
  q = find_queue (&m->peer, (const char *) &m[1]);
  if (NULL == q)
  {
    suggest_to_connect (&m->peer, (const char *) &m[1]);
  }
  else
    start_address_validation (&m->peer, (const char *) &m[1]);
  GNUNET_SERVICE_client_continue (tc->client);
}
 
 
static int
free_neighbour_cb (void *cls,
                   const struct GNUNET_PeerIdentity *pid,
                   void *value)
{
  struct Neighbour *neighbour = value;
 
  (void) cls;
  (void) pid;
  GNUNET_break (0);  // should this ever happen?
  free_neighbour (neighbour, GNUNET_YES);
 
  return GNUNET_OK;
}
 
 
static int
free_dv_routes_cb (void *cls,
                   const struct GNUNET_PeerIdentity *pid,
                   void *value)
{
  struct DistanceVector *dv = value;
 
  (void) cls;
  (void) pid;
  free_dv_route (dv);
 
  return GNUNET_OK;
}
 
 
static int
free_validation_state_cb (void *cls,
                          const struct GNUNET_PeerIdentity *pid,
                          void *value)
{
  struct ValidationState *vs = value;
 
  (void) cls;
  (void) pid;
  free_validation_state (vs);
  return GNUNET_OK;
}
 
 
static int
free_pending_ack_cb (void *cls, const struct GNUNET_Uuid *key, void *value)
{
  struct PendingAcknowledgement *pa = value;
 
  (void) cls;
  (void) key;
  free_pending_acknowledgement (pa);
  return GNUNET_OK;
}
 
 
static int
free_ack_cummulator_cb (void *cls,
                        const struct GNUNET_PeerIdentity *pid,
                        void *value)
{
  struct AcknowledgementCummulator *ac = value;
 
  (void) cls;
  (void) pid;
  GNUNET_SCHEDULER_cancel (ac->task);
  GNUNET_free (ac);
  return GNUNET_OK;
}
 
 
static void
do_shutdown (void *cls)
{
  struct LearnLaunchEntry *lle;
  struct PilsRequest *pr;
  (void) cls;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "shutdown logic\n");
  GNUNET_NAT_unregister (nh);
  GNUNET_CONTAINER_multipeermap_iterate (neighbours,
                                         &free_neighbour_cb, NULL);
  if (NULL != validation_task)
  {
    GNUNET_SCHEDULER_cancel (validation_task);
    validation_task = NULL;
  }
  if (NULL != dvlearn_task)
  {
    GNUNET_SCHEDULER_cancel (dvlearn_task);
    dvlearn_task = NULL;
  }
  GNUNET_CONTAINER_multishortmap_destroy (dvlearn_map);
  dvlearn_map = NULL;
  GNUNET_CONTAINER_multipeermap_iterate (dv_routes, &free_dv_routes_cb, NULL);
  GNUNET_CONTAINER_multipeermap_destroy (dv_routes);
  dv_routes = NULL;
  if (NULL != GST_stats)
  {
    GNUNET_STATISTICS_destroy (GST_stats, GNUNET_NO);
    GST_stats = NULL;
  }
  if (NULL != GST_my_hello)
  {
    GNUNET_HELLO_builder_free (GST_my_hello);
    GST_my_hello = NULL;
  }
  if (NULL != GST_my_identity)
  {
    GNUNET_free (GST_my_identity);
    GST_my_identity = NULL;
  }
  GNUNET_CONTAINER_multipeermap_iterate (ack_cummulators,
                                         &free_ack_cummulator_cb,
                                         NULL);
  GNUNET_CONTAINER_multipeermap_destroy (ack_cummulators);
  ack_cummulators = NULL;
  GNUNET_CONTAINER_multiuuidmap_iterate (pending_acks,
                                         &free_pending_ack_cb,
                                         NULL);
  GNUNET_CONTAINER_multiuuidmap_destroy (pending_acks);
  pending_acks = NULL;
  GNUNET_break (0 == GNUNET_CONTAINER_multipeermap_size (neighbours));
  GNUNET_CONTAINER_multipeermap_destroy (neighbours);
  neighbours = NULL;
  GNUNET_break (0 == GNUNET_CONTAINER_multipeermap_size (links));
  GNUNET_CONTAINER_multipeermap_destroy (links);
  links = NULL;
  GNUNET_CONTAINER_multipeermap_iterate (backtalkers,
                                         &free_backtalker_cb,
                                         NULL);
  GNUNET_CONTAINER_multipeermap_destroy (backtalkers);
  backtalkers = NULL;
  GNUNET_CONTAINER_multipeermap_iterate (validation_map,
                                         &free_validation_state_cb,
                                         NULL);
  GNUNET_CONTAINER_multipeermap_destroy (validation_map);
  validation_map = NULL;
  GNUNET_CONTAINER_heap_destroy (validation_heap);
  validation_heap = NULL;
  GNUNET_CONTAINER_multihashmap_destroy (revalidation_map);
  revalidation_map = NULL;
  while (NULL != ir_head)
    free_incoming_request (ir_head);
  GNUNET_assert (0 == ir_total);
  while (NULL != (lle = lle_head))
  {
    GNUNET_CONTAINER_DLL_remove (lle_head, lle_tail, lle);
    GNUNET_free (lle);
  }
  while (NULL != (pr = pils_requests_head))
  {
    GNUNET_CONTAINER_DLL_remove (pils_requests_head,
                                 pils_requests_tail,
                                 pr);
    if (NULL != pr->op)
      GNUNET_PILS_cancel (pr->op);
    GNUNET_free (pr);
  }
  if (NULL != pils_feed_task)
  {
    GNUNET_SCHEDULER_cancel (pils_feed_task);
    pils_feed_task = NULL;
  }
  if (NULL != pils)
  {
    GNUNET_PILS_disconnect (pils);
    pils = NULL;
  }
  if (NULL != peerstore)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Disconnecting from PEERSTORE service\n");
    GNUNET_PEERSTORE_disconnect (peerstore);
    peerstore = NULL;
  }
  GNUNET_SCHEDULER_shutdown ();
}
 
 
static void
shutdown_task (void *cls)
{
  in_shutdown = GNUNET_YES;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Shutdown task executed\n");
  if (NULL != clients_head)
  {
    for (struct TransportClient *tc = clients_head; NULL != tc; tc = tc->next)
    {
      GNUNET_log (GNUNET_ERROR_TYPE_ERROR,
                  "client still connected: %u\n",
                  tc->type);
    }
  }
  else
    do_shutdown (cls);
 
}
 
 
struct UpdateHelloFromPidCtx
{
  struct GNUNET_PEERSTORE_StoreHelloContext *sc;
};
 
static void
update_hello_from_pid_change_cb (void *cls, int success)
{
  struct UpdateHelloFromPidCtx *pc = cls;
 
  if (GNUNET_OK != success)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
                "Failed to store our new hello with peerstore\n");
  }
  GNUNET_free (pc);
  GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
              "Stored our new hello with peerstore\n");
}
 
 
void
print_address_list (void *cls,
                    const struct GNUNET_PeerIdentity *pid,
                    const char *uri)
{
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "%s\n", uri);
}
 
 
static void
pils_pid_change_cb (void *cls,
                    const struct GNUNET_HELLO_Parser *parser,
                    const struct GNUNET_HashCode *hash)
{
  struct GNUNET_MQ_Envelope *env;
  const struct GNUNET_MessageHeader *msg;
  struct UpdateHelloFromPidCtx *sc;
  struct GNUNET_HELLO_Builder *nbuilder;
  struct GNUNET_PeerIdentity npid;
 
  if (NULL == GST_my_identity)
    GST_my_identity = GNUNET_new (struct GNUNET_PeerIdentity);
  if (NULL == GST_my_hello)
    GST_my_hello = GNUNET_HELLO_builder_new ();
  GNUNET_log (GNUNET_ERROR_TYPE_INFO,
              "My current identity is `%s'\n",
              GNUNET_i2s_full (GST_my_identity));
  nbuilder = GNUNET_HELLO_builder_from_parser (parser,
                                               &npid);
  if (GNUNET_NO ==
      GNUNET_HELLO_builder_address_list_cmp (GST_my_hello, nbuilder))
  {
    GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
                "New PID from PILS is derived from address list inconsistend with ours. Ignoring...\n");
    GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
                "Proposed address list:\n");
    GNUNET_HELLO_builder_iterate (nbuilder, &print_address_list, NULL);
    GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
                "Current address list:\n");
    GNUNET_HELLO_builder_iterate (GST_my_hello, &print_address_list, NULL);
    GNUNET_HELLO_builder_free (nbuilder);
    return;
  }
  GST_my_hello = nbuilder;
  memcpy (GST_my_identity, &npid, sizeof npid);
  GNUNET_log (GNUNET_ERROR_TYPE_INFO,
              "My new identity is `%s'\n",
              GNUNET_i2s_full (GST_my_identity));
  env = GNUNET_HELLO_parser_to_env (parser);
  msg = GNUNET_MQ_env_get_msg (env);
  sc = GNUNET_new (struct UpdateHelloFromPidCtx);
  sc->sc = GNUNET_PEERSTORE_hello_add (peerstore,
                                       msg,
                                       update_hello_from_pid_change_cb,
                                       sc);
  GNUNET_free (env);
}
 
 
static void
run (void *cls,
     const struct GNUNET_CONFIGURATION_Handle *c,
     struct GNUNET_SERVICE_Handle *service)
{
  (void) cls;
  (void) service;
  /* setup globals */
  hello_mono_time = GNUNET_TIME_absolute_get_monotonic (c);
  in_shutdown = GNUNET_NO;
  GST_cfg = c;
  backtalkers = GNUNET_CONTAINER_multipeermap_create (16, GNUNET_YES);
  pending_acks = GNUNET_CONTAINER_multiuuidmap_create (32768, GNUNET_YES);
  ack_cummulators = GNUNET_CONTAINER_multipeermap_create (256, GNUNET_YES);
  neighbours = GNUNET_CONTAINER_multipeermap_create (1024, GNUNET_YES);
  links = GNUNET_CONTAINER_multipeermap_create (512, GNUNET_YES);
  dv_routes = GNUNET_CONTAINER_multipeermap_create (1024, GNUNET_YES);
  dvlearn_map = GNUNET_CONTAINER_multishortmap_create (2 * MAX_DV_LEARN_PENDING,
                                                       GNUNET_YES);
  validation_map = GNUNET_CONTAINER_multipeermap_create (1024, GNUNET_YES);
  revalidation_map = GNUNET_CONTAINER_multihashmap_create (1024, GNUNET_YES);
  validation_heap =
    GNUNET_CONTAINER_heap_create (GNUNET_CONTAINER_HEAP_ORDER_MIN);
  // TODO check for all uses of GST_my_hello that it is not used uninitialized
  use_burst = GNUNET_CONFIGURATION_get_value_yesno (GST_cfg,
                                                    "transport",
                                                    "USE_BURST_NAT");
  if (GNUNET_SYSERR == use_burst)
    GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
                "Could not configure burst nat use. Default to no.\n");
  GST_my_hello = GNUNET_HELLO_builder_new ();
  GST_stats = GNUNET_STATISTICS_create ("transport", GST_cfg);
  GNUNET_SCHEDULER_add_shutdown (&shutdown_task, NULL);
  peerstore = GNUNET_PEERSTORE_connect (GST_cfg);
  nh = GNUNET_NAT_register (GST_cfg,
                            "transport",
                            0,
                            0,
                            NULL,
                            0,
                            NULL,
                            NULL,
                            NULL);
  if (NULL == peerstore)
  {
    GNUNET_break (0);
    GNUNET_SCHEDULER_shutdown ();
    return;
  }
  GST_my_identity = NULL;
  pils = GNUNET_PILS_connect (GST_cfg,
                              pils_pid_change_cb,
                              NULL);          // FIXME we need to wait for
  // our first peer id before
  // we can start the service
  // completely - PILS in turn
  // waits for the first
  // addresses from the
  // communicators in order to
  // be able to generate a
  // peer id
  if (NULL == pils)
  {
    GNUNET_break (0);
    GNUNET_SCHEDULER_shutdown ();
    return;
  }
}
 
 
GNUNET_SERVICE_MAIN (
  GNUNET_OS_project_data_gnunet (),
  "transport",
  GNUNET_SERVICE_OPTION_SOFT_SHUTDOWN,
  &run,
  &client_connect_cb,
  &client_disconnect_cb,
  NULL,
  /* communication with applications */
  GNUNET_MQ_hd_fixed_size (suggest,
                           GNUNET_MESSAGE_TYPE_TRANSPORT_SUGGEST,
                           struct ExpressPreferenceMessage,
                           NULL),
  GNUNET_MQ_hd_fixed_size (suggest_cancel,
                           GNUNET_MESSAGE_TYPE_TRANSPORT_SUGGEST_CANCEL,
                           struct ExpressPreferenceMessage,
                           NULL),
  GNUNET_MQ_hd_var_size (request_hello_validation,
                         GNUNET_MESSAGE_TYPE_TRANSPORT_REQUEST_HELLO_VALIDATION,
                         struct RequestHelloValidationMessage,
                         NULL),
  /* communication with core */
  GNUNET_MQ_hd_fixed_size (client_start,
                           GNUNET_MESSAGE_TYPE_TRANSPORT_START,
                           struct StartMessage,
                           NULL),
  GNUNET_MQ_hd_var_size (client_send,
                         GNUNET_MESSAGE_TYPE_TRANSPORT_SEND,
                         struct OutboundMessage,
                         NULL),
  GNUNET_MQ_hd_fixed_size (client_recv_ok,
                           GNUNET_MESSAGE_TYPE_TRANSPORT_RECV_OK,
                           struct RecvOkMessage,
                           NULL),
  /* communication with communicators */
  GNUNET_MQ_hd_var_size (communicator_available,
                         GNUNET_MESSAGE_TYPE_TRANSPORT_NEW_COMMUNICATOR,
                         struct GNUNET_TRANSPORT_CommunicatorAvailableMessage,
                         NULL),
  GNUNET_MQ_hd_var_size (communicator_backchannel,
                         GNUNET_MESSAGE_TYPE_TRANSPORT_COMMUNICATOR_BACKCHANNEL,
                         struct GNUNET_TRANSPORT_CommunicatorBackchannel,
                         NULL),
  GNUNET_MQ_hd_var_size (add_address,
                         GNUNET_MESSAGE_TYPE_TRANSPORT_ADD_ADDRESS,
                         struct GNUNET_TRANSPORT_AddAddressMessage,
                         NULL),
  GNUNET_MQ_hd_fixed_size (del_address,
                           GNUNET_MESSAGE_TYPE_TRANSPORT_DEL_ADDRESS,
                           struct GNUNET_TRANSPORT_DelAddressMessage,
                           NULL),
  GNUNET_MQ_hd_var_size (incoming_msg,
                         GNUNET_MESSAGE_TYPE_TRANSPORT_INCOMING_MSG,
                         struct GNUNET_TRANSPORT_IncomingMessage,
                         NULL),
  GNUNET_MQ_hd_fixed_size (queue_create_ok,
                           GNUNET_MESSAGE_TYPE_TRANSPORT_QUEUE_CREATE_OK,
                           struct GNUNET_TRANSPORT_CreateQueueResponse,
                           NULL),
  GNUNET_MQ_hd_fixed_size (queue_create_fail,
                           GNUNET_MESSAGE_TYPE_TRANSPORT_QUEUE_CREATE_FAIL,
                           struct GNUNET_TRANSPORT_CreateQueueResponse,
                           NULL),
  GNUNET_MQ_hd_var_size (add_queue_message,
                         GNUNET_MESSAGE_TYPE_TRANSPORT_QUEUE_SETUP,
                         struct GNUNET_TRANSPORT_AddQueueMessage,
                         NULL),
  GNUNET_MQ_hd_fixed_size (update_queue_message,
                           GNUNET_MESSAGE_TYPE_TRANSPORT_QUEUE_UPDATE,
                           struct GNUNET_TRANSPORT_UpdateQueueMessage,
                           NULL),
  GNUNET_MQ_hd_fixed_size (del_queue_message,
                           GNUNET_MESSAGE_TYPE_TRANSPORT_QUEUE_TEARDOWN,
                           struct GNUNET_TRANSPORT_DelQueueMessage,
                           NULL),
  GNUNET_MQ_hd_fixed_size (send_message_ack,
                           GNUNET_MESSAGE_TYPE_TRANSPORT_SEND_MSG_ACK,
                           struct GNUNET_TRANSPORT_SendMessageToAck,
                           NULL),
  GNUNET_MQ_hd_fixed_size (burst_finished,
                           GNUNET_MESSAGE_TYPE_TRANSPORT_BURST_FINISHED,
                           struct GNUNET_TRANSPORT_BurstFinished,
                           NULL),
  /* communication with monitors */
  GNUNET_MQ_hd_fixed_size (monitor_start,
                           GNUNET_MESSAGE_TYPE_TRANSPORT_MONITOR_START,
                           struct GNUNET_TRANSPORT_MonitorStart,
                           NULL),
  GNUNET_MQ_handler_end ());
 
 
/* end of file gnunet-service-transport.c */

DELAY_WARN_THRESHOLD

#define DELAY_WARN_THRESHOLD    GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_SECONDS, 5)

If a queue delays the next message by more than this number of seconds we log a warning.

Note: this is for testing, the value chosen here might be too aggressively low!

Definition at line 199 of file gnunet-service-transport.c.

DV_FORWARD_TIMEOUT

#define DV_FORWARD_TIMEOUT    GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_SECONDS, 60)

If a DVBox could not be forwarded after this number of seconds we drop it.

Definition at line 206 of file gnunet-service-transport.c.

DEFAULT_ACK_WAIT_DURATION

#define DEFAULT_ACK_WAIT_DURATION    GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_SECONDS, 1)

Default value for how long we wait for reliability ack.

Definition at line 212 of file gnunet-service-transport.c.

DV_QUALITY_RTT_THRESHOLD

#define DV_QUALITY_RTT_THRESHOLD    GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_SECONDS, 1)

We only consider queues as "quality" connections when suppressing the generation of DV initiation messages if the latency of the queue is below this threshold.

Definition at line 220 of file gnunet-service-transport.c.

DV_PATH_VALIDITY_TIMEOUT

#define DV_PATH_VALIDITY_TIMEOUT    GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_MINUTES, 5)

How long do we consider a DV path valid if we see no further updates on it? Note: the value chosen here might be too low!

Definition at line 227 of file gnunet-service-transport.c.

BACKCHANNEL_INACTIVITY_TIMEOUT

#define BACKCHANNEL_INACTIVITY_TIMEOUT    GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_MINUTES, 5)

How long do we cache backchannel (struct Backtalker) information after a backchannel goes inactive?

Definition at line 234 of file gnunet-service-transport.c.

DV_PATH_DISCOVERY_FREQUENCY

#define DV_PATH_DISCOVERY_FREQUENCY    GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_MINUTES, 4)

How long before paths expire would we like to (re)discover DV paths? Should be below DV_PATH_VALIDITY_TIMEOUT.

Definition at line 241 of file gnunet-service-transport.c.

EPHEMERAL_VALIDITY

#define EPHEMERAL_VALIDITY    GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_HOURS, 4)

How long are ephemeral keys valid?

Definition at line 247 of file gnunet-service-transport.c.

REASSEMBLY_EXPIRATION

#define REASSEMBLY_EXPIRATION    GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_MINUTES, 4)

How long do we keep partially reassembled messages around before giving up?

Definition at line 253 of file gnunet-service-transport.c.

FAST_VALIDATION_CHALLENGE_FREQ

#define FAST_VALIDATION_CHALLENGE_FREQ    GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_MINUTES, 1)

What is the fastest rate at which we send challenges if we keep learning an address (gossip, DHT, etc.)?

Definition at line 260 of file gnunet-service-transport.c.

MAX_VALIDATION_CHALLENGE_FREQ

#define MAX_VALIDATION_CHALLENGE_FREQ    GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_DAYS, 1)

What is the slowest rate at which we send challenges?

Definition at line 266 of file gnunet-service-transport.c.

ACK_CUMMULATOR_TIMEOUT

#define ACK_CUMMULATOR_TIMEOUT    GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_HOURS, 4)

How long until we forget about historic accumulators and thus reset the ACK counter? Should exceed the maximum time an active connection experiences without an ACK.

Definition at line 274 of file gnunet-service-transport.c.

DV_LEARN_BASE_FREQUENCY

#define DV_LEARN_BASE_FREQUENCY   GNUNET_TIME_UNIT_MINUTES

What is the non-randomized base frequency at which we would initiate DV learn messages?

Definition at line 281 of file gnunet-service-transport.c.

DV_LEARN_QUALITY_THRESHOLD

#define DV_LEARN_QUALITY_THRESHOLD   100

How many good connections (confirmed, bi-directional, not DV) do we need to have to suppress initiating DV learn messages?

Definition at line 287 of file gnunet-service-transport.c.

MAX_ADDRESS_VALID_UNTIL

#define MAX_ADDRESS_VALID_UNTIL    GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_MONTHS, 1)

When do we forget an invalid address for sure?

Definition at line 292 of file gnunet-service-transport.c.

ADDRESS_VALIDATION_LIFETIME

#define ADDRESS_VALIDATION_LIFETIME    GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_HOURS, 4)

How long do we consider an address valid if we just checked?

Definition at line 298 of file gnunet-service-transport.c.

MIN_DELAY_ADDRESS_VALIDATION

#define MIN_DELAY_ADDRESS_VALIDATION   GNUNET_TIME_UNIT_MILLISECONDS

What is the maximum frequency at which we do address validation? A random value between 0 and this value is added when scheduling the validation_task (both to ensure we do not validate too often, and to randomize a bit).

Definition at line 307 of file gnunet-service-transport.c.

VALIDATION_RTT_BUFFER_FACTOR

#define VALIDATION_RTT_BUFFER_FACTOR   3

How many network RTTs before an address validation expires should we begin trying to revalidate? (Note that the RTT used here is the one that we experienced during the last validation, not necessarily the latest RTT observed).

Definition at line 315 of file gnunet-service-transport.c.

COMMUNICATOR_TOTAL_QUEUE_LIMIT

#define COMMUNICATOR_TOTAL_QUEUE_LIMIT   512

How many messages can we have pending for a given communicator process before we start to throttle that communicator?

Used if a communicator might be CPU-bound and cannot handle the traffic.

Definition at line 323 of file gnunet-service-transport.c.

QUEUE_LENGTH_LIMIT

#define QUEUE_LENGTH_LIMIT   32

How many messages can we have pending for a given queue (queue to a particular peer via a communicator) process before we start to throttle that queue?

Definition at line 330 of file gnunet-service-transport.c.

QUEUE_ENTRY_TIMEOUT

#define QUEUE_ENTRY_TIMEOUT    GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_SECONDS, 5)

Definition at line 335 of file gnunet-service-transport.c.

RTT_DIFF

#define RTT_DIFF    GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_SECONDS, 1)

Difference of the average RTT for the DistanceVector calculate by us and the target we are willing to accept for starting the burst.

Definition at line 342 of file gnunet-service-transport.c.

Typedef Documentation

DVMessageHandler

typedef void(* DVMessageHandler) (void *cls, struct Neighbour *next_hop, const struct GNUNET_MessageHeader *hdr, enum RouteMessageOptions options)

Function to call to further operate on the now DV encapsulated message hdr, forwarding it via next_hop under respect of options.

Parameters

cls	closure
next_hop	next hop of the DV path
hdr	encapsulated message, technically a struct TransportDFBoxMessage
options	options of the original message

Definition at line 5213 of file gnunet-service-transport.c.

Enumeration Type Documentation

ClientType

enum ClientType

What type of client is the struct TransportClient about?

Enumerator
CT_NONE	We do not know yet (client is fresh).
CT_CORE	Is the CORE service, we need to forward traffic to it.
CT_MONITOR	It is a monitor, forward monitor data.
CT_COMMUNICATOR	It is a communicator, use for communication.
CT_APPLICATION	"Application" telling us where to connect (i.e. TOPOLOGY, DHT or CADET).

Definition at line 1032 of file gnunet-service-transport.c.

{
  CT_NONE = 0,
 
  CT_CORE = 1,
 
  CT_MONITOR = 2,
 
  CT_COMMUNICATOR = 3,
 
  CT_APPLICATION = 4
};

RouteMessageOptions

enum RouteMessageOptions

Which transmission options are allowable for transmission? Interpreted bit-wise!

Enumerator
RMO_NONE	Only confirmed, non-DV direct neighbours.
RMO_DV_ALLOWED	We are allowed to use DV routing for this hdr.
RMO_UNCONFIRMED_ALLOWED	We are allowed to use unconfirmed queues or DV routes for this message.
RMO_ANYTHING_GOES	Reliable and unreliable, DV and non-DV are all acceptable.
RMO_REDUNDANT	If we have multiple choices, it is OK to send this message over multiple channels at the same time to improve loss tolerance. (We do at most 2 transmissions.)

Definition at line 1065 of file gnunet-service-transport.c.

{
  RMO_NONE = 0,
 
  RMO_DV_ALLOWED = 1,
 
  RMO_UNCONFIRMED_ALLOWED = 2,
 
  RMO_ANYTHING_GOES = (RMO_DV_ALLOWED | RMO_UNCONFIRMED_ALLOWED),
 
  RMO_REDUNDANT = 4
};

PendingMessageType

enum PendingMessageType

Types of different pending messages.

Enumerator
PMT_CORE	Ordinary message received from the CORE service.
PMT_FRAGMENT_BOX	Fragment box.
PMT_RELIABILITY_BOX	Reliability box.
PMT_DV_BOX	Pending message created during forward_dv_box().

Definition at line 2215 of file gnunet-service-transport.c.

{
  PMT_CORE = 0,
 
  PMT_FRAGMENT_BOX = 1,
 
  PMT_RELIABILITY_BOX = 2,
 
  PMT_DV_BOX = 3
};

Function Documentation

get_age()

static unsigned int get_age ( )

static

Get an offset into the transmission history buffer for struct PerformanceData.

Note that the caller must perform the required modulo GOODPUT_AGING_SLOTS operation before indexing into the array!

An 'age' lasts 15 minute slots.

Returns

current age of the world

Definition at line 3127 of file gnunet-service-transport.c.

{
  struct GNUNET_TIME_Absolute now;
 
  now = GNUNET_TIME_absolute_get ();
  return now.abs_value_us / GNUNET_TIME_UNIT_MINUTES.rel_value_us / 15;
}

References GNUNET_TIME_Absolute::abs_value_us, GNUNET_TIME_absolute_get(), and GNUNET_TIME_UNIT_MINUTES.

Referenced by update_performance_data().

Here is the call graph for this function:

Here is the caller graph for this function:

free_incoming_request()

static void free_incoming_request ( struct IncomingRequest *  ir )

static

Release ir data structure.

Parameters

ir	data structure to release

Definition at line 3142 of file gnunet-service-transport.c.

{
  GNUNET_CONTAINER_DLL_remove (ir_head, ir_tail, ir);
  GNUNET_assert (ir_total > 0);
  ir_total--;
  if (NULL != ir->nc)
    GNUNET_PEERSTORE_monitor_stop (ir->nc);
  ir->nc = NULL;
  GNUNET_free (ir);
}

References GNUNET_assert, GNUNET_CONTAINER_DLL_remove, GNUNET_free, GNUNET_PEERSTORE_monitor_stop(), ir_head, ir_tail, ir_total, and IncomingRequest::nc.

Referenced by do_shutdown(), and sign_t_validation_cb().

Here is the call graph for this function:

Here is the caller graph for this function:

free_pending_acknowledgement()

static void free_pending_acknowledgement ( struct PendingAcknowledgement *  pa )

static

Release pa data structure.

Parameters

pa	data structure to release

Definition at line 3160 of file gnunet-service-transport.c.

{
  struct Queue *q = pa->queue;
  struct PendingMessage *pm = pa->pm;
  struct DistanceVectorHop *dvh = pa->dvh;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "free_pending_acknowledgement\n");
  if (NULL != q)
  {
    GNUNET_CONTAINER_MDLL_remove (queue, q->pa_head, q->pa_tail, pa);
    pa->queue = NULL;
  }
  if (NULL != pm)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "remove pa from message\n");
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "remove pa from message %" PRIu64 "\n",
                pm->logging_uuid);
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "remove pa from message %u\n",
                pm->pmt);
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "remove pa from message %s\n",
                GNUNET_uuid2s (&pa->ack_uuid.value));
    GNUNET_CONTAINER_MDLL_remove (pm, pm->pa_head, pm->pa_tail, pa);
    pa->pm = NULL;
  }
  if (NULL != dvh)
  {
    GNUNET_CONTAINER_MDLL_remove (dvh, dvh->pa_head, dvh->pa_tail, pa);
    pa->queue = NULL;
  }
  GNUNET_assert (GNUNET_YES ==
                 GNUNET_CONTAINER_multiuuidmap_remove (pending_acks,
                                                       &pa->ack_uuid.value,
                                                       pa));
  GNUNET_free (pa);
}

References PendingAcknowledgement::ack_uuid, PendingAcknowledgement::dvh, GNUNET_assert, GNUNET_CONTAINER_MDLL_remove, GNUNET_CONTAINER_multiuuidmap_remove(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_free, GNUNET_log, GNUNET_uuid2s(), GNUNET_YES, DistanceVectorHop::pa_head, DistanceVectorHop::pa_tail, pending_acks, PendingAcknowledgement::pm, pm, q, queue(), PendingAcknowledgement::queue, and AcknowledgementUUIDP::value.

Referenced by free_pending_ack_cb(), and handle_acknowledged().

Here is the call graph for this function:

Here is the caller graph for this function:

free_fragment_tree()

static void free_fragment_tree ( struct PendingMessage *  root )

static

Free fragment tree below root, excluding root itself.

FIXME: this does NOT seem to have the intended semantics based on how this is called. Seems we generally DO expect root to be free'ed itself as well!

Parameters

root	root of the tree to free

Definition at line 3211 of file gnunet-service-transport.c.

{
  struct PendingMessage *frag;
 
  while (NULL != (frag = root->head_frag))
  {
    struct PendingAcknowledgement *pa;
 
    free_fragment_tree (frag);
    while (NULL != (pa = frag->pa_head))
    {
      GNUNET_CONTAINER_MDLL_remove (pm, frag->pa_head, frag->pa_tail, pa);
      pa->pm = NULL;
    }
    GNUNET_CONTAINER_MDLL_remove (frag, root->head_frag, root->tail_frag, frag);
    if (NULL != frag->qe)
    {
      GNUNET_assert (frag == frag->qe->pm);
      frag->qe->pm = NULL;
    }
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Free frag %p\n",
                frag);
    GNUNET_free (frag);
  }
}

References free_fragment_tree(), GNUNET_assert, GNUNET_CONTAINER_MDLL_remove, GNUNET_ERROR_TYPE_DEBUG, GNUNET_free, GNUNET_log, PendingMessage::head_frag, PendingMessage::pa_head, PendingMessage::pa_tail, PendingAcknowledgement::pm, QueueEntry::pm, pm, PendingMessage::qe, and PendingMessage::tail_frag.

Referenced by free_fragment_tree(), and free_pending_message().

Here is the call graph for this function:

Here is the caller graph for this function:

free_pending_message()

static void free_pending_message ( struct PendingMessage *  pm )

static

Release memory associated with pm and remove pm from associated data structures.

pm must be a top-level pending message and not a fragment in the tree. The entire tree is freed (if applicable).

Parameters

pm	the pending message to free

Definition at line 3247 of file gnunet-service-transport.c.

{
  struct TransportClient *tc = pm->client;
  struct VirtualLink *vl = pm->vl;
  struct PendingAcknowledgement *pa;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Freeing pm %p\n",
              pm);
  if (NULL != tc)
  {
    GNUNET_CONTAINER_MDLL_remove (client,
                                  tc->details.core.pending_msg_head,
                                  tc->details.core.pending_msg_tail,
                                  pm);
  }
  if ((NULL != vl) && (NULL == pm->frag_parent))
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Removing pm %" PRIu64 "\n",
                pm->logging_uuid);
    GNUNET_CONTAINER_MDLL_remove (vl,
                                  vl->pending_msg_head,
                                  vl->pending_msg_tail,
                                  pm);
  }
  else if (NULL != pm->frag_parent && PMT_DV_BOX != pm->pmt)
  {
    struct PendingMessage *root = pm->frag_parent;
 
    while (NULL != root->frag_parent && PMT_DV_BOX != root->pmt)
      root = root->frag_parent;
 
    root->frag_count--;
  }
  while (NULL != (pa = pm->pa_head))
  {
    if (NULL == pa)
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "free pending pa  null\n");
    if (NULL == pm->pa_tail)
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "free pending pa_tail null\n");
    if (NULL == pa->prev_pa)
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "free pending pa prev null\n");
    if (NULL == pa->next_pa)
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "free pending pa next null\n");
    GNUNET_CONTAINER_MDLL_remove (pm, pm->pa_head, pm->pa_tail, pa);
    pa->pm = NULL;
  }
 
  free_fragment_tree (pm);
  if (NULL != pm->qe)
  {
    GNUNET_assert (pm == pm->qe->pm);
    pm->qe->pm = NULL;
  }
  if (NULL != pm->bpm)
  {
    free_fragment_tree (pm->bpm);
    if (NULL != pm->bpm->qe)
    {
      struct QueueEntry *qe = pm->bpm->qe;
 
      qe->pm = NULL;
    }
    GNUNET_free (pm->bpm);
  }
 
  GNUNET_free (pm);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Freeing pm done\n");
}

References PendingMessage::frag_count, PendingMessage::frag_parent, free_fragment_tree(), GNUNET_assert, GNUNET_CONTAINER_MDLL_remove, GNUNET_ERROR_TYPE_DEBUG, GNUNET_free, GNUNET_log, PendingAcknowledgement::next_pa, VirtualLink::pending_msg_head, VirtualLink::pending_msg_tail, PendingAcknowledgement::pm, pm, PendingMessage::pmt, PMT_DV_BOX, PendingAcknowledgement::prev_pa, qe, and tc.

Referenced by client_send_response(), completed_pending_message(), free_virtual_link(), and transmit_on_queue().

Here is the call graph for this function:

Here is the caller graph for this function:

free_reassembly_context()

static void free_reassembly_context ( struct ReassemblyContext *  rc )

static

Free rc.

Parameters

rc	data structure to free

Definition at line 3330 of file gnunet-service-transport.c.

{
  struct VirtualLink *vl = rc->virtual_link;
 
  GNUNET_assert (rc == GNUNET_CONTAINER_heap_remove_node (rc->hn));
  GNUNET_assert (GNUNET_OK ==
                 GNUNET_CONTAINER_multihashmap32_remove (vl->reassembly_map,
                                                         rc->msg_uuid.uuid,
                                                         rc));
  GNUNET_free (rc);
}

References GNUNET_assert, GNUNET_CONTAINER_heap_remove_node(), GNUNET_CONTAINER_multihashmap32_remove(), GNUNET_free, GNUNET_OK, ReassemblyContext::hn, ReassemblyContext::msg_uuid, VirtualLink::reassembly_map, MessageUUIDP::uuid, and ReassemblyContext::virtual_link.

Referenced by free_reassembly_cb(), handle_fragment_box(), and reassembly_cleanup_task().

Here is the call graph for this function:

Here is the caller graph for this function:

reassembly_cleanup_task()

static void reassembly_cleanup_task ( void *  cls )

static

Task run to clean up reassembly context of a neighbour that have expired.

Parameters

cls	a struct Neighbour

Definition at line 3349 of file gnunet-service-transport.c.

{
  struct VirtualLink *vl = cls;
  struct ReassemblyContext *rc;
 
  vl->reassembly_timeout_task = NULL;
  while (NULL != (rc = GNUNET_CONTAINER_heap_peek (vl->reassembly_heap)))
  {
    if (0 == GNUNET_TIME_absolute_get_remaining (rc->reassembly_timeout)
        .rel_value_us)
    {
      free_reassembly_context (rc);
      continue;
    }
    GNUNET_assert (NULL == vl->reassembly_timeout_task);
    vl->reassembly_timeout_task =
      GNUNET_SCHEDULER_add_at (rc->reassembly_timeout,
                               &reassembly_cleanup_task,
                               vl);
    return;
  }
}

References free_reassembly_context(), GNUNET_assert, GNUNET_CONTAINER_heap_peek(), GNUNET_SCHEDULER_add_at(), GNUNET_TIME_absolute_get_remaining(), reassembly_cleanup_task(), VirtualLink::reassembly_heap, ReassemblyContext::reassembly_timeout, VirtualLink::reassembly_timeout_task, and GNUNET_TIME_Relative::rel_value_us.

Referenced by handle_fragment_box(), and reassembly_cleanup_task().

Here is the call graph for this function:

Here is the caller graph for this function:

free_reassembly_cb()

static int free_reassembly_cb ( void *  cls, uint32_t  key, void *  value  )

static

function called to free_reassembly_context().

Parameters

cls	NULL
key	unused
value	a struct ReassemblyContext to free

Returns

GNUNET_OK (continue iteration)

Definition at line 3382 of file gnunet-service-transport.c.

{
  struct ReassemblyContext *rc = value;
 
  (void) cls;
  (void) key;
  free_reassembly_context (rc);
  return GNUNET_OK;
}

References free_reassembly_context(), GNUNET_OK, key, and value.

Referenced by free_virtual_link().

Here is the call graph for this function:

Here is the caller graph for this function:

free_virtual_link()

static void free_virtual_link ( struct VirtualLink *  vl )

static

Free virtual link.

Parameters

vl	link data to free

Definition at line 3399 of file gnunet-service-transport.c.

{
  struct PendingMessage *pm;
  struct CoreSentContext *csc;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "free virtual link %p\n",
              vl);
 
  if (NULL != vl->reassembly_map)
  {
    GNUNET_CONTAINER_multihashmap32_iterate (vl->reassembly_map,
                                             &free_reassembly_cb,
                                             NULL);
    GNUNET_CONTAINER_multihashmap32_destroy (vl->reassembly_map);
    vl->reassembly_map = NULL;
    GNUNET_CONTAINER_heap_destroy (vl->reassembly_heap);
    vl->reassembly_heap = NULL;
  }
  if (NULL != vl->reassembly_timeout_task)
  {
    GNUNET_SCHEDULER_cancel (vl->reassembly_timeout_task);
    vl->reassembly_timeout_task = NULL;
  }
  while (NULL != (pm = vl->pending_msg_head))
    free_pending_message (pm);
  GNUNET_assert (GNUNET_YES ==
                 GNUNET_CONTAINER_multipeermap_remove (links, &vl->target, vl));
  if (NULL != vl->visibility_task)
  {
    GNUNET_SCHEDULER_cancel (vl->visibility_task);
    vl->visibility_task = NULL;
  }
  if (NULL != vl->fc_retransmit_task)
  {
    GNUNET_SCHEDULER_cancel (vl->fc_retransmit_task);
    vl->fc_retransmit_task = NULL;
  }
  while (NULL != (csc = vl->csc_head))
  {
    GNUNET_CONTAINER_DLL_remove (vl->csc_head, vl->csc_tail, csc);
    GNUNET_assert (vl == csc->vl);
    csc->vl = NULL;
  }
  GNUNET_break (NULL == vl->n);
  GNUNET_break (NULL == vl->dv);
  GNUNET_free (vl);
}

References VirtualLink::csc_head, VirtualLink::csc_tail, VirtualLink::dv, VirtualLink::fc_retransmit_task, free_pending_message(), free_reassembly_cb(), GNUNET_assert, GNUNET_break, GNUNET_CONTAINER_DLL_remove, GNUNET_CONTAINER_heap_destroy(), GNUNET_CONTAINER_multihashmap32_destroy(), GNUNET_CONTAINER_multihashmap32_iterate(), GNUNET_CONTAINER_multipeermap_remove(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_free, GNUNET_log, GNUNET_SCHEDULER_cancel(), GNUNET_YES, links, VirtualLink::n, VirtualLink::pending_msg_head, pm, VirtualLink::reassembly_heap, VirtualLink::reassembly_map, VirtualLink::reassembly_timeout_task, VirtualLink::target, VirtualLink::visibility_task, and CoreSentContext::vl.

Referenced by check_link_down(), free_dv_route(), and free_neighbour().

Here is the call graph for this function:

Here is the caller graph for this function:

free_validation_state()

static void free_validation_state ( struct ValidationState *  vs )

static

Free validation state.

Parameters

vs	validation state to free

Definition at line 3455 of file gnunet-service-transport.c.

{
  if (NULL != vs->revalidation_task)
  {
    GNUNET_SCHEDULER_cancel (vs->revalidation_task);
    vs->revalidation_task = NULL;
  }
  /*memcpy (&hkey,
          &hc,
          sizeof (hkey));*/
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Remove key %s for address %s map size %u contains %u during freeing state\n",
              GNUNET_h2s (&vs->hc),
              vs->address,
              GNUNET_CONTAINER_multihashmap_size (revalidation_map),
              GNUNET_CONTAINER_multihashmap_contains (revalidation_map,
                                                      &vs->hc));
  GNUNET_CONTAINER_multihashmap_remove (revalidation_map, &vs->hc, vs);
  GNUNET_assert (
    GNUNET_YES ==
    GNUNET_CONTAINER_multipeermap_remove (validation_map, &vs->pid, vs));
  GNUNET_CONTAINER_heap_remove_node (vs->hn);
  vs->hn = NULL;
  if (NULL != vs->sc)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "store cancel\n");
    GNUNET_PEERSTORE_store_cancel (vs->sc);
    vs->sc = NULL;
  }
  GNUNET_free (vs->address);
  GNUNET_free (vs);
}

References GNUNET_assert, GNUNET_CONTAINER_heap_remove_node(), GNUNET_CONTAINER_multihashmap_contains(), GNUNET_CONTAINER_multihashmap_remove(), GNUNET_CONTAINER_multihashmap_size(), GNUNET_CONTAINER_multipeermap_remove(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_free, GNUNET_h2s(), GNUNET_log, GNUNET_PEERSTORE_store_cancel(), GNUNET_SCHEDULER_cancel(), GNUNET_YES, ValidationState::pid, revalidation_map, and validation_map.

Referenced by free_validation_state_cb(), and validation_start_cb().

Here is the call graph for this function:

Here is the caller graph for this function:

lookup_neighbour()

static struct Neighbour * lookup_neighbour ( const struct GNUNET_PeerIdentity *  pid )

static

Lookup neighbour for peer pid.

Parameters

pid	neighbour to look for

Returns

NULL if we do not have this peer as a neighbour

Definition at line 3497 of file gnunet-service-transport.c.

{
  return GNUNET_CONTAINER_multipeermap_get (neighbours, pid);
}

References GNUNET_CONTAINER_multipeermap_get(), neighbours, and pid.

Referenced by find_queue(), handle_add_queue_message(), handle_communicator_backchannel(), handle_dv_box(), handle_dv_learn(), learn_dv_path(), route_control_message_without_fc(), sign_dhp_cp(), sign_t_validation_cb(), and transmit_cummulative_ack_cb().

Here is the call graph for this function:

Here is the caller graph for this function:

lookup_virtual_link()

static struct VirtualLink * lookup_virtual_link ( const struct GNUNET_PeerIdentity *  pid )

static

Lookup virtual link for peer pid.

Parameters

pid	virtual link to look for

Returns

NULL if we do not have this peer as a virtual link

Definition at line 3510 of file gnunet-service-transport.c.

{
  return GNUNET_CONTAINER_multipeermap_get (links, pid);
}

References GNUNET_CONTAINER_multipeermap_get(), links, and pid.

Referenced by activate_core_visible_dv_path(), check_vl_transmission(), free_queue(), handle_client_recv_ok(), handle_client_send(), handle_communicator_backchannel(), handle_flow_control(), handle_fragment_box(), handle_raw_message(), handle_validation_response(), sign_dhp_cp(), sign_t_validation_cb(), and transmit_cummulative_ack_cb().

Here is the call graph for this function:

Here is the caller graph for this function:

free_distance_vector_hop()

static void free_distance_vector_hop ( struct DistanceVectorHop *  dvh )

static

Free a dvh.

Callers MAY want to check if this was the last path to the target, and if so call free_dv_route to also free the associated DV entry in dv_routes (if not, the associated scheduler job should eventually take care of it).

Parameters

dvh	hop to free

Definition at line 3559 of file gnunet-service-transport.c.

{
  struct Neighbour *n = dvh->next_hop;
  struct DistanceVector *dv = dvh->dv;
  struct PendingAcknowledgement *pa;
 
  while (NULL != (pa = dvh->pa_head))
  {
    GNUNET_CONTAINER_MDLL_remove (dvh, dvh->pa_head, dvh->pa_tail, pa);
    pa->dvh = NULL;
  }
  GNUNET_CONTAINER_MDLL_remove (neighbour, n->dv_head, n->dv_tail, dvh);
  GNUNET_CONTAINER_MDLL_remove (dv, dv->dv_head, dv->dv_tail, dvh);
  GNUNET_free (dvh);
}

References DistanceVectorHop::dv, DistanceVector::dv_head, Neighbour::dv_head, DistanceVector::dv_tail, Neighbour::dv_tail, PendingAcknowledgement::dvh, GNUNET_CONTAINER_MDLL_remove, GNUNET_free, DistanceVectorHop::next_hop, DistanceVectorHop::pa_head, and DistanceVectorHop::pa_tail.

Referenced by free_dv_route(), free_neighbour(), and path_cleanup_cb().

Here is the caller graph for this function:

check_link_down()

static void check_link_down ( void *  cls )

static

Task run to check whether the hops of the cls still are validated, or if we need to core about disconnection.

Parameters

cls	a struct VirtualLink

Definition at line 3998 of file gnunet-service-transport.c.

{
  struct VirtualLink *vl = cls;
  struct DistanceVector *dv = vl->dv;
  struct Neighbour *n = vl->n;
  struct GNUNET_TIME_Absolute dvh_timeout;
  struct GNUNET_TIME_Absolute q_timeout;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Checking if link is down\n");
  vl->visibility_task = NULL;
  dvh_timeout = GNUNET_TIME_UNIT_ZERO_ABS;
  if (NULL != dv)
  {
    for (struct DistanceVectorHop *pos = dv->dv_head; NULL != pos;
         pos = pos->next_dv)
      dvh_timeout = GNUNET_TIME_absolute_max (dvh_timeout,
                                              pos->path_valid_until);
    if (0 == GNUNET_TIME_absolute_get_remaining (dvh_timeout).rel_value_us)
    {
      vl->dv->vl = NULL;
      vl->dv = NULL;
    }
  }
  q_timeout = GNUNET_TIME_UNIT_ZERO_ABS;
  for (struct Queue *q = n->queue_head; NULL != q; q = q->next_neighbour)
    q_timeout = GNUNET_TIME_absolute_max (q_timeout, q->validated_until);
  if (0 == GNUNET_TIME_absolute_get_remaining (q_timeout).rel_value_us)
  {
    vl->n->vl = NULL;
    vl->n = NULL;
  }
  if ((NULL == vl->n) && (NULL == vl->dv))
  {
    cores_send_disconnect_info (&vl->target);
    free_virtual_link (vl);
    return;
  }
  vl->visibility_task =
    GNUNET_SCHEDULER_add_at (GNUNET_TIME_absolute_max (q_timeout, dvh_timeout),
                             &check_link_down,
                             vl);
}

References check_link_down(), cores_send_disconnect_info(), VirtualLink::dv, DistanceVector::dv_head, free_virtual_link(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_log, GNUNET_SCHEDULER_add_at(), GNUNET_TIME_absolute_get_remaining(), GNUNET_TIME_absolute_max(), GNUNET_TIME_UNIT_ZERO_ABS, VirtualLink::n, q, Neighbour::queue_head, GNUNET_TIME_Relative::rel_value_us, VirtualLink::target, VirtualLink::visibility_task, DistanceVector::vl, and Neighbour::vl.

Referenced by activate_core_visible_dv_path(), check_link_down(), free_dv_route(), free_neighbour(), free_queue(), and handle_validation_response().

Here is the call graph for this function:

Here is the caller graph for this function:

cores_send_disconnect_info()

static void cores_send_disconnect_info ( const struct GNUNET_PeerIdentity *  pid )

static

Send message to CORE clients that we lost a connection.

Parameters

pid	peer the connection was for

Definition at line 3592 of file gnunet-service-transport.c.

{
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Informing CORE clients about disconnect from %s\n",
              GNUNET_i2s (pid));
  for (struct TransportClient *tc = clients_head; NULL != tc; tc = tc->next)
  {
    struct GNUNET_MQ_Envelope *env;
    struct DisconnectInfoMessage *dim;
 
    if (CT_CORE != tc->type)
      continue;
    env = GNUNET_MQ_msg (dim, GNUNET_MESSAGE_TYPE_TRANSPORT_DISCONNECT);
    dim->peer = *pid;
    GNUNET_MQ_send (tc->mq, env);
  }
}

References clients_head, CT_CORE, dim, env, GNUNET_ERROR_TYPE_DEBUG, GNUNET_i2s(), GNUNET_log, GNUNET_MESSAGE_TYPE_TRANSPORT_DISCONNECT, GNUNET_MQ_msg, GNUNET_MQ_send(), pid, and tc.

Referenced by check_link_down(), free_dv_route(), and free_neighbour().

Here is the call graph for this function:

Here is the caller graph for this function:

free_dv_route()

static void free_dv_route ( struct DistanceVector *  dv )

static

Free entry in dv_routes.

First frees all hops to the target, and if there are no entries left, frees dv as well.

Parameters

dv	route to free

Definition at line 3618 of file gnunet-service-transport.c.

{
  struct DistanceVectorHop *dvh;
  struct VirtualLink *vl;
 
  while (NULL != (dvh = dv->dv_head))
    free_distance_vector_hop (dvh);
 
  GNUNET_assert (
    GNUNET_YES ==
    GNUNET_CONTAINER_multipeermap_remove (dv_routes, &dv->target, dv));
  if (NULL != (vl = dv->vl))
  {
    GNUNET_assert (dv == vl->dv);
    vl->dv = NULL;
    if (NULL == vl->n)
    {
      cores_send_disconnect_info (&dv->target);
      free_virtual_link (vl);
    }
    else
    {
      GNUNET_SCHEDULER_cancel (vl->visibility_task);
      vl->visibility_task = GNUNET_SCHEDULER_add_now (&check_link_down, vl);
    }
    dv->vl = NULL;
  }
 
  if (NULL != dv->timeout_task)
  {
    GNUNET_SCHEDULER_cancel (dv->timeout_task);
    dv->timeout_task = NULL;
  }
  GNUNET_free (dv->km);
  GNUNET_free (dv);
}

References check_link_down(), cores_send_disconnect_info(), VirtualLink::dv, DistanceVector::dv_head, dv_routes, free_distance_vector_hop(), free_virtual_link(), GNUNET_assert, GNUNET_CONTAINER_multipeermap_remove(), GNUNET_free, GNUNET_SCHEDULER_add_now(), GNUNET_SCHEDULER_cancel(), GNUNET_YES, DistanceVector::km, VirtualLink::n, DistanceVector::target, DistanceVector::timeout_task, VirtualLink::visibility_task, and DistanceVector::vl.

Referenced by free_dv_routes_cb(), free_neighbour(), and path_cleanup_cb().

Here is the call graph for this function:

Here is the caller graph for this function:

notify_monitor()

static void notify_monitor ( struct TransportClient *  tc, const struct GNUNET_PeerIdentity *  peer, const char *  address, enum GNUNET_NetworkType  nt, const struct MonitorEvent *  me  )

static

Notify monitor tc about an event.

That tc cares about the event has already been checked.

Send tc information in me about a peer's status with respect to some address to all monitors that care.

Parameters

tc	monitor to inform
peer	peer the information is about
address	address the information is about
nt	network type associated with address
me	detailed information to transmit

Definition at line 3670 of file gnunet-service-transport.c.

{
  struct GNUNET_MQ_Envelope *env;
  struct GNUNET_TRANSPORT_MonitorData *md;
  size_t addr_len = strlen (address) + 1;
 
  env = GNUNET_MQ_msg_extra (md,
                             addr_len,
                             GNUNET_MESSAGE_TYPE_TRANSPORT_MONITOR_DATA);
  md->nt = htonl ((uint32_t) nt);
  md->peer = *peer;
  md->last_validation = GNUNET_TIME_absolute_hton (me->last_validation);
  md->valid_until = GNUNET_TIME_absolute_hton (me->valid_until);
  md->next_validation = GNUNET_TIME_absolute_hton (me->next_validation);
  md->rtt = GNUNET_TIME_relative_hton (me->rtt);
  md->cs = htonl ((uint32_t) me->cs);
  md->num_msg_pending = htonl (me->num_msg_pending);
  md->num_bytes_pending = htonl (me->num_bytes_pending);
  memcpy (&md[1], address, addr_len);
  GNUNET_MQ_send (tc->mq, env);
}

References address, GNUNET_TRANSPORT_MonitorData::cs, env, GNUNET_MESSAGE_TYPE_TRANSPORT_MONITOR_DATA, GNUNET_MQ_msg_extra, GNUNET_MQ_send(), GNUNET_TIME_absolute_hton(), GNUNET_TIME_relative_hton(), GNUNET_TRANSPORT_MonitorData::last_validation, me, GNUNET_TRANSPORT_MonitorData::next_validation, nt, GNUNET_TRANSPORT_MonitorData::nt, GNUNET_TRANSPORT_MonitorData::num_bytes_pending, GNUNET_TRANSPORT_MonitorData::num_msg_pending, GNUNET_TRANSPORT_MonitorData::peer, GNUNET_TRANSPORT_MonitorData::rtt, tc, and GNUNET_TRANSPORT_MonitorData::valid_until.

Referenced by notify_client_queues(), and notify_monitors().

Here is the call graph for this function:

Here is the caller graph for this function:

notify_monitors()

static void notify_monitors ( const struct GNUNET_PeerIdentity *  peer, const char *  address, enum GNUNET_NetworkType  nt, const struct MonitorEvent *  me  )

static

Send information in me about a peer's status with respect to some address to all monitors that care.

Parameters

peer	peer the information is about
address	address the information is about
nt	network type associated with address
me	detailed information to transmit

Definition at line 3707 of file gnunet-service-transport.c.

{
  for (struct TransportClient *tc = clients_head; NULL != tc; tc = tc->next)
  {
    if (CT_MONITOR != tc->type)
      continue;
    if (tc->details.monitor.one_shot)
      continue;
    if ((GNUNET_NO == GNUNET_is_zero (&tc->details.monitor.peer)) &&
        (0 != GNUNET_memcmp (&tc->details.monitor.peer, peer)))
      continue;
    notify_monitor (tc, peer, address, nt, me);
  }
}

References address, clients_head, CT_MONITOR, GNUNET_is_zero, GNUNET_memcmp, GNUNET_NO, me, notify_monitor(), nt, GNUNET_TRANSPORT_MonitorData::peer, and tc.

Referenced by free_queue(), and handle_add_queue_message().

Here is the call graph for this function:

Here is the caller graph for this function:

client_connect_cb()

static void * client_connect_cb ( void *  cls, struct GNUNET_SERVICE_Client *  client, struct GNUNET_MQ_Handle *  mq  )

static

Called whenever a client connects.

Allocates our data structures associated with that client.

Parameters

cls	closure, NULL
client	identification of the client
mq	message queue for the client

Returns

our struct TransportClient

Definition at line 3736 of file gnunet-service-transport.c.

{
  struct TransportClient *tc;
 
  (void) cls;
  tc = GNUNET_new (struct TransportClient);
  tc->client = client;
  tc->mq = mq;
  GNUNET_CONTAINER_DLL_insert (clients_head, clients_tail, tc);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Client %p of type %u connected\n",
              tc,
              tc->type);
  return tc;
}

References TransportClient::client, clients_head, clients_tail, GNUNET_CONTAINER_DLL_insert, GNUNET_ERROR_TYPE_DEBUG, GNUNET_log, GNUNET_new, mq, and tc.

remove_global_addresses()

static enum GNUNET_GenericReturnValue remove_global_addresses ( void *  cls, const struct GNUNET_PeerIdentity *  pid, void *  value  )

static

Definition at line 3756 of file gnunet-service-transport.c.

{
  struct TransportGlobalNattedAddress *tgna = value;
  (void) cls;
 
  GNUNET_free (tgna);
 
  return GNUNET_OK;
}

References GNUNET_free, GNUNET_OK, and value.

Referenced by free_neighbour().

Here is the caller graph for this function:

free_neighbour()

static void free_neighbour ( struct Neighbour *  neighbour, enum GNUNET_GenericReturnValue  drop_link  )

static

Release memory used by neighbour.

Parameters

neighbour	neighbour entry to free
drop_link	flag to decide whether to drop its virtual link

Definition at line 3776 of file gnunet-service-transport.c.

{
  struct DistanceVectorHop *dvh;
  struct VirtualLink *vl;
 
  GNUNET_assert (NULL == neighbour->queue_head);
  GNUNET_assert (GNUNET_YES ==
                 GNUNET_CONTAINER_multipeermap_remove (neighbours,
                                                       &neighbour->pid,
                                                       neighbour));
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Freeing neighbour\n");
  GNUNET_CONTAINER_multipeermap_iterate (neighbour->natted_addresses,
                                         &remove_global_addresses,
                                         NULL);
  GNUNET_CONTAINER_multipeermap_destroy (neighbour->natted_addresses);
  while (NULL != (dvh = neighbour->dv_head))
  {
    struct DistanceVector *dv = dvh->dv;
 
    free_distance_vector_hop (dvh);
    if (NULL == dv->dv_head)
      free_dv_route (dv);
  }
  if (NULL != neighbour->get)
  {
    GNUNET_PEERSTORE_iteration_stop (neighbour->get);
    neighbour->get = NULL;
  }
  if (NULL != neighbour->sc)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "store cancel\n");
    GNUNET_PEERSTORE_store_cancel (neighbour->sc);
    neighbour->sc = NULL;
  }
  if (NULL != (vl = neighbour->vl))
  {
    GNUNET_assert (neighbour == vl->n);
    vl->n = NULL;
    if ((GNUNET_YES == drop_link) || (NULL == vl->dv))
    {
      cores_send_disconnect_info (&vl->target);
      free_virtual_link (vl);
    }
    else
    {
      GNUNET_SCHEDULER_cancel (vl->visibility_task);
      vl->visibility_task = GNUNET_SCHEDULER_add_now (&check_link_down, vl);
    }
    neighbour->vl = NULL;
  }
  GNUNET_free (neighbour);
}

References check_link_down(), cores_send_disconnect_info(), VirtualLink::dv, DistanceVectorHop::dv, DistanceVector::dv_head, Neighbour::dv_head, free_distance_vector_hop(), free_dv_route(), free_virtual_link(), Neighbour::get, GNUNET_assert, GNUNET_CONTAINER_multipeermap_destroy(), GNUNET_CONTAINER_multipeermap_iterate(), GNUNET_CONTAINER_multipeermap_remove(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_free, GNUNET_log, GNUNET_PEERSTORE_iteration_stop(), GNUNET_PEERSTORE_store_cancel(), GNUNET_SCHEDULER_add_now(), GNUNET_SCHEDULER_cancel(), GNUNET_YES, VirtualLink::n, Neighbour::natted_addresses, neighbours, Neighbour::pid, Neighbour::queue_head, remove_global_addresses(), Neighbour::sc, VirtualLink::target, VirtualLink::visibility_task, DistanceVector::vl, and Neighbour::vl.

Referenced by free_neighbour_cb(), and free_queue().

Here is the call graph for this function:

Here is the caller graph for this function:

core_send_connect_info()

static void core_send_connect_info ( struct TransportClient *  tc, const struct GNUNET_PeerIdentity *  pid  )

static

Send message to CORE clients that we lost a connection.

Parameters

tc	client to inform (must be CORE client)
pid	peer the connection is for

Definition at line 3840 of file gnunet-service-transport.c.

{
  struct GNUNET_MQ_Envelope *env;
  struct ConnectInfoMessage *cim;
 
  GNUNET_assert (CT_CORE == tc->type);
  env = GNUNET_MQ_msg (cim, GNUNET_MESSAGE_TYPE_TRANSPORT_CONNECT);
  cim->id = *pid;
  GNUNET_MQ_send (tc->mq, env);
}

References CT_CORE, env, GNUNET_assert, GNUNET_MESSAGE_TYPE_TRANSPORT_CONNECT, GNUNET_MQ_msg, GNUNET_MQ_send(), ConnectInfoMessage::id, pid, and tc.

Referenced by cores_send_connect_info(), and notify_client_connect_info().

Here is the call graph for this function:

Here is the caller graph for this function:

cores_send_connect_info()

static void cores_send_connect_info ( const struct GNUNET_PeerIdentity *  pid )

static

Send message to CORE clients that we gained a connection.

Parameters

pid	peer the queue was for

Definition at line 3859 of file gnunet-service-transport.c.

{
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Informing CORE clients about connection to %s\n",
              GNUNET_i2s (pid));
  for (struct TransportClient *tc = clients_head; NULL != tc; tc = tc->next)
  {
    if (CT_CORE != tc->type)
      continue;
    core_send_connect_info (tc, pid);
  }
}

References clients_head, core_send_connect_info(), CT_CORE, GNUNET_ERROR_TYPE_DEBUG, GNUNET_i2s(), GNUNET_log, pid, and tc.

Referenced by activate_core_visible_dv_path(), and handle_validation_response().

Here is the call graph for this function:

Here is the caller graph for this function:

transmit_on_queue()

static void transmit_on_queue ( void *  cls )

static

We believe we are ready to transmit a message on a queue.

We believe we are ready to transmit a struct PendingMessage on a queue, the big question is which one! We need to see if there is one pending that is allowed by flow control and congestion control and (ideally) matches our queue's performance profile.

Gives the message to the communicator for transmission (updating the tracker, and re-scheduling itself if applicable).

Parameters

cls	the struct Queue to process transmissions for

If such a message is found, we give the message to the communicator for transmission (updating the tracker, and re-scheduling ourselves if applicable).

If no such message is found, the queue's idle field must be set to GNUNET_YES.

Parameters

cls	the struct Queue to process transmissions for

Definition at line 11404 of file gnunet-service-transport.c.

{
  struct Queue *queue = cls;
  struct Neighbour *n = queue->neighbour;
  struct PendingMessageScoreContext sc;
  struct PendingMessage *pm;
 
  queue->transmit_task = NULL;
  if (NULL == n->vl)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Virtual link `%s' is down, cannot have PM for queue `%s'\n",
                GNUNET_i2s (&n->pid),
                queue->address);
    queue->idle = GNUNET_YES;
    return;
  }
  memset (&sc, 0, sizeof(sc));
  select_best_pending_from_link (&sc, queue, n->vl, NULL, 0);
  if (NULL == sc.best)
  {
    /* Also look at DVH that have the n as first hop! */
    for (struct DistanceVectorHop *dvh = n->dv_head; NULL != dvh;
         dvh = dvh->next_neighbour)
    {
      select_best_pending_from_link (&sc,
                                     queue,
                                     dvh->dv->vl,
                                     dvh,
                                     sizeof(struct GNUNET_PeerIdentity)
                                     * (1 + dvh->distance)
                                     + sizeof(struct TransportDVBoxMessage)
                                     + sizeof(struct TransportDVBoxPayloadP));
    }
  }
  if (NULL == sc.best)
  {
    /* no message pending, nothing to do here! */
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "No pending messages, queue `%s' to %s now idle\n",
                queue->address,
                GNUNET_i2s (&n->pid));
    if (GNUNET_YES == sc.to_early)
      schedule_transmit_on_queue (sc.to_early_retry_delay,
                                  queue,
                                  GNUNET_SCHEDULER_PRIORITY_DEFAULT);
    queue->idle = GNUNET_YES;
    return;
  }
  /* There is a message pending, we are certainly not idle */
  queue->idle = GNUNET_NO;
 
  /* Given selection in `sc`, do transmission */
  pm = sc.best;
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Selected message <%" PRIu64 ">\n",
              pm->logging_uuid);
  if (NULL != sc.dvh)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Is this %u a DV box?\n",
                pm->pmt);
    GNUNET_assert (PMT_DV_BOX != pm->pmt);
    if ((NULL != sc.best->bpm) && (sc.best->bpm->used_dvh != sc.dvh))
    {
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "Discard old box, because we have a new DV path.\n");
      free_pending_message (sc.best->bpm);
      sc.best->bpm = NULL;
    }
 
    if (NULL == sc.best->bpm)
    {
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "encapsulate_for_dv 2\n");
      encapsulate_for_dv (sc.dvh->dv,
                          1,
                          &sc.dvh,
                          (const struct GNUNET_MessageHeader *) &sc.best[1],
                          &extract_box_cb,
                          &sc,
                          RMO_NONE,
                          GNUNET_NO);
      GNUNET_assert (NULL != sc.best->bpm);
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "%lu %lu %lu %lu %u\n",
                  sizeof(struct GNUNET_PeerIdentity),
                  sizeof(struct TransportDVBoxMessage),
                  sizeof(struct TransportDVBoxPayloadP),
                  sizeof(struct TransportFragmentBoxMessage),
                  ((const struct GNUNET_MessageHeader *) &sc.best[1])->size);
      sc.best->bpm->used_dvh = sc.dvh;
    }
    pm = sc.best->bpm;
  }
  if (GNUNET_YES == sc.frag)
  {
    pm = fragment_message (queue, sc.dvh, pm);
    if (NULL == pm)
    {
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "Fragmentation failed queue %s to %s for <%" PRIu64
                  ">, trying again\n",
                  queue->address,
                  GNUNET_i2s (&n->pid),
                  sc.best->logging_uuid);
      schedule_transmit_on_queue (GNUNET_TIME_UNIT_ZERO,
                                  queue,
                                  GNUNET_SCHEDULER_PRIORITY_DEFAULT);
      return;
    }
  }
  else if (GNUNET_YES == sc.relb)
  {
    pm = reliability_box_message (queue, sc.dvh, pm);
    if (NULL == pm)
    {
      /* Reliability boxing failed, try next message... */
      GNUNET_log (
        GNUNET_ERROR_TYPE_DEBUG,
        "Reliability boxing failed queue %s to %s for <%" PRIu64
        ">, trying again\n",
        queue->address,
        GNUNET_i2s (&n->pid),
        sc.best->logging_uuid);
      schedule_transmit_on_queue (GNUNET_TIME_UNIT_ZERO,
                                  queue,
                                  GNUNET_SCHEDULER_PRIORITY_DEFAULT);
      return;
    }
  }
 
  /* Pass 'pm' for transission to the communicator */
  GNUNET_log (
    GNUNET_ERROR_TYPE_DEBUG,
    "Passing message <%" PRIu64
    "> to queue %s for peer %s (considered %u others)\n",
    pm->logging_uuid,
    queue->address,
    GNUNET_i2s (&n->pid),
    sc.consideration_counter);
 
  /* Flow control: increment amount of traffic sent; if we are routing
     via DV (and thus the ultimate target of the pending message is for
     a different virtual link than the one of the queue), then we need
     to use up not only the window of the direct link but also the
     flow control window for the DV link! */
  pm->vl->outbound_fc_window_size_used += pm->bytes_msg;
 
  if (pm->vl != queue->neighbour->vl)
  {
    /* If the virtual link of the queue differs, this better be distance
       vector routing! */
    GNUNET_assert (NULL != sc.dvh);
    /* If we do distance vector routing, we better not do this for a
       message that was itself DV-routed */
    GNUNET_assert (PMT_DV_BOX != sc.best->pmt);
    /* We use the size of the unboxed message here, to avoid counting
       the DV-Box header which is eaten up on the way by intermediaries */
    queue->neighbour->vl->outbound_fc_window_size_used += sc.best->bytes_msg;
  }
  else
  {
    GNUNET_assert (NULL == sc.dvh);
  }
 
  queue_send_msg (queue, pm, &pm[1], pm->bytes_msg);
 
  /* Check if this transmission somehow conclusively finished handing 'pm'
     even without any explicit ACKs */
  if ((PMT_CORE == pm->pmt) ||
      (GNUNET_TRANSPORT_CC_RELIABLE == queue->tc->details.communicator.cc))
  {
    completed_pending_message (pm);
  }
  else
  {
    struct GNUNET_TIME_Relative wait_duration;
    unsigned int wait_multiplier;
 
    if (PMT_FRAGMENT_BOX == pm->pmt)
    {
      struct PendingMessage *root;
 
      root = pm->frag_parent;
      while (NULL != root->frag_parent && PMT_DV_BOX != root->pmt)
        root = root->frag_parent;
 
      wait_multiplier =  (unsigned int) ceil ((double) root->bytes_msg
                                              / ((double) root->frag_off
                                                 / (double) root->frag_count))
                        * 4;
    }
    else
    {
      // No fragments, we use 4 RTT before retransmitting.
      wait_multiplier = 4;
    }
 
    // Depending on how much pending message the VirtualLink is queueing, we wait longer.
    // wait_multiplier = wait_multiplier * pm->vl->pending_msg_num;
 
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Wait multiplier %u\n",
                wait_multiplier);
 
    /* Message not finished, waiting for acknowledgement.
       Update time by which we might retransmit 's' based on queue
       characteristics (i.e. RTT); it takes one RTT for the message to
       arrive and the ACK to come back in the best case; but the other
       side is allowed to delay ACKs by 2 RTTs, so we use 4 RTT before
       retransmitting.
 
       OPTIMIZE: Note that in the future this heuristic should likely
       be improved further (measure RTT stability, consider message
       urgency and size when delaying ACKs, etc.) */
 
    if (GNUNET_TIME_UNIT_FOREVER_REL.rel_value_us !=
        queue->pd.aged_rtt.rel_value_us)
      wait_duration = queue->pd.aged_rtt;
    else
    {
      wait_duration = DEFAULT_ACK_WAIT_DURATION;
      wait_multiplier = 4;
    }
    {
      struct GNUNET_TIME_Absolute next = GNUNET_TIME_relative_to_absolute (
        GNUNET_TIME_relative_multiply (
          wait_duration, wait_multiplier));
      struct GNUNET_TIME_Relative plus = GNUNET_TIME_relative_multiply (
        wait_duration, wait_multiplier);
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "Waiting %s for ACK until %s\n",
                  GNUNET_STRINGS_relative_time_to_string (plus, GNUNET_NO),
                  GNUNET_STRINGS_absolute_time_to_string (next));
      update_pm_next_attempt (pm,
                              GNUNET_TIME_relative_to_absolute (
                                GNUNET_TIME_relative_multiply (wait_duration,
                                                               wait_multiplier))
                              );
    }
  }
  /* finally, re-schedule queue transmission task itself */
  schedule_transmit_on_queue (GNUNET_TIME_UNIT_ZERO,
                              queue,
                              GNUNET_SCHEDULER_PRIORITY_DEFAULT);
}

References PendingMessage::bytes_msg, completed_pending_message(), DEFAULT_ACK_WAIT_DURATION, Neighbour::dv_head, encapsulate_for_dv(), extract_box_cb(), PendingMessage::frag_count, PendingMessage::frag_off, PendingMessage::frag_parent, fragment_message(), free_pending_message(), GNUNET_assert, GNUNET_ERROR_TYPE_DEBUG, GNUNET_i2s(), GNUNET_log, GNUNET_NO, GNUNET_SCHEDULER_PRIORITY_DEFAULT, GNUNET_STRINGS_absolute_time_to_string(), GNUNET_STRINGS_relative_time_to_string(), GNUNET_TIME_relative_multiply(), GNUNET_TIME_relative_to_absolute(), GNUNET_TIME_UNIT_FOREVER_REL, GNUNET_TIME_UNIT_ZERO, GNUNET_TRANSPORT_CC_RELIABLE, GNUNET_YES, Neighbour::pid, pm, PendingMessage::pmt, PMT_CORE, PMT_DV_BOX, PMT_FRAGMENT_BOX, queue(), queue_send_msg(), reliability_box_message(), RMO_NONE, sc, schedule_transmit_on_queue(), select_best_pending_from_link(), update_pm_next_attempt(), and Neighbour::vl.

Referenced by schedule_transmit_on_queue().

Here is the call graph for this function:

Here is the caller graph for this function:

check_for_queue_with_higher_prio()

static unsigned int check_for_queue_with_higher_prio ( struct Queue *  queue, struct Queue *  queue_head  )

static

Check if the communicator has another queue with higher prio ready for sending.

Definition at line 3888 of file gnunet-service-transport.c.

{
  for (struct Queue *s = queue_head; NULL != s;
       s = s->next_client)
  {
    if (s->tc->details.communicator.address_prefix !=
        queue->tc->details.communicator.address_prefix)
    {
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "queue address %s qid %u compare with queue: address %s qid %u\n",
                  queue->address,
                  queue->qid,
                  s->address,
                  s->qid);
      if ((s->priority > queue->priority) && (0 < s->q_capacity) &&
          (QUEUE_LENGTH_LIMIT > s->queue_length) )
        return GNUNET_YES;
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "Lower prio\n");
    }
  }
  return GNUNET_NO;
}

References GNUNET_ERROR_TYPE_DEBUG, GNUNET_log, GNUNET_NO, GNUNET_YES, Queue::next_client, queue(), queue_head, and QUEUE_LENGTH_LIMIT.

Referenced by schedule_transmit_on_queue().

Here is the call graph for this function:

Here is the caller graph for this function:

schedule_transmit_on_queue()

static void schedule_transmit_on_queue ( struct GNUNET_TIME_Relative  delay, struct Queue *  queue, enum GNUNET_SCHEDULER_Priority  p  )

static

Called whenever something changed that might effect when we try to do the next transmission on queue using transmit_on_queue().

Parameters

queue	the queue to do scheduling for
p	task priority to use, if queue is scheduled

Definition at line 3921 of file gnunet-service-transport.c.

{
  struct GNUNET_TIME_Absolute now = GNUNET_TIME_absolute_get ();
 
  if (queue->validated_until.abs_value_us < now.abs_value_us)
    return;
  if (check_for_queue_with_higher_prio (queue,
                                        queue->tc->details.communicator.
                                        queue_head))
    return;
 
  if (queue->tc->details.communicator.total_queue_length >=
      COMMUNICATOR_TOTAL_QUEUE_LIMIT)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Transmission on queue %s (QID %u) throttled due to communicator queue limit\n",
                queue->address,
                queue->qid);
    GNUNET_STATISTICS_update (
      GST_stats,
      "# Transmission throttled due to communicator queue limit",
      1,
      GNUNET_NO);
    queue->idle = GNUNET_NO;
    return;
  }
  if (queue->queue_length >= QUEUE_LENGTH_LIMIT)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Transmission on queue %s (QID %u) throttled due to communicator queue length limit\n",
                queue->address,
                queue->qid);
    GNUNET_STATISTICS_update (GST_stats,
                              "# Transmission throttled due to queue queue limit",
                              1,
                              GNUNET_NO);
    queue->idle = GNUNET_NO;
    return;
  }
  if (0 == queue->q_capacity)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Transmission on queue %s (QID %u) throttled due to communicator message  has capacity %"
                PRIu64 ".\n",
                queue->address,
                queue->qid,
                queue->q_capacity);
    GNUNET_STATISTICS_update (GST_stats,
                              "# Transmission throttled due to message queue capacity",
                              1,
                              GNUNET_NO);
    queue->idle = GNUNET_NO;
    return;
  }
  /* queue might indeed be ready, schedule it */
  if (NULL != queue->transmit_task)
    GNUNET_SCHEDULER_cancel (queue->transmit_task);
  queue->transmit_task =
    GNUNET_SCHEDULER_add_delayed_with_priority (delay, p, &transmit_on_queue,
                                                queue);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Considering transmission on queue `%s' QID %llu to %s\n",
              queue->address,
              (unsigned long long) queue->qid,
              GNUNET_i2s (&queue->neighbour->pid));
}

References GNUNET_TIME_Absolute::abs_value_us, check_for_queue_with_higher_prio(), COMMUNICATOR_TOTAL_QUEUE_LIMIT, GNUNET_ERROR_TYPE_DEBUG, GNUNET_i2s(), GNUNET_log, GNUNET_NO, GNUNET_SCHEDULER_add_delayed_with_priority(), GNUNET_SCHEDULER_cancel(), GNUNET_STATISTICS_update(), GNUNET_TIME_absolute_get(), GST_stats, p, queue(), queue_head, QUEUE_LENGTH_LIMIT, and transmit_on_queue().

Referenced by check_vl_transmission(), free_queue(), free_queue_entry(), handle_update_queue_message(), and transmit_on_queue().

Here is the call graph for this function:

Here is the caller graph for this function:

free_queue()

static void free_queue ( struct Queue *  queue )

static

Free queue.

Parameters

queue

the queue to free

Definition at line 4049 of file gnunet-service-transport.c.

{
  struct Neighbour *neighbour = queue->neighbour;
  struct TransportClient *tc = queue->tc;
  struct MonitorEvent me = { .cs = GNUNET_TRANSPORT_CS_DOWN,
                             .rtt = GNUNET_TIME_UNIT_FOREVER_REL };
  struct QueueEntry *qe;
  int maxxed;
  struct PendingAcknowledgement *pa;
  struct VirtualLink *vl;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Cleaning up queue %u\n", queue->qid);
  if (NULL != queue->mo)
  {
    GNUNET_PEERSTORE_monitor_stop (queue->mo);
    queue->mo = NULL;
  }
  if (NULL != queue->transmit_task)
  {
    GNUNET_SCHEDULER_cancel (queue->transmit_task);
    queue->transmit_task = NULL;
  }
  while (NULL != (pa = queue->pa_head))
  {
    GNUNET_CONTAINER_MDLL_remove (queue, queue->pa_head, queue->pa_tail, pa);
    pa->queue = NULL;
  }
 
  GNUNET_CONTAINER_MDLL_remove (neighbour,
                                neighbour->queue_head,
                                neighbour->queue_tail,
                                queue);
  GNUNET_CONTAINER_MDLL_remove (client,
                                tc->details.communicator.queue_head,
                                tc->details.communicator.queue_tail,
                                queue);
  maxxed = (COMMUNICATOR_TOTAL_QUEUE_LIMIT <=
            tc->details.communicator.total_queue_length);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Cleaning up queue with length %u\n",
              queue->queue_length);
  while (NULL != (qe = queue->queue_head))
  {
    GNUNET_CONTAINER_DLL_remove (queue->queue_head, queue->queue_tail, qe);
    queue->queue_length--;
    tc->details.communicator.total_queue_length--;
    if (NULL != qe->pm)
    {
      GNUNET_assert (qe == qe->pm->qe);
      qe->pm->qe = NULL;
    }
    GNUNET_free (qe);
  }
  GNUNET_assert (0 == queue->queue_length);
  if ((maxxed) && (COMMUNICATOR_TOTAL_QUEUE_LIMIT >
                   tc->details.communicator.total_queue_length))
  {
    /* Communicator dropped below threshold, resume all _other_ queues */
    GNUNET_STATISTICS_update (
      GST_stats,
      "# Transmission throttled due to communicator queue limit",
      -1,
      GNUNET_NO);
    for (struct Queue *s = tc->details.communicator.queue_head; NULL != s;
         s = s->next_client)
      schedule_transmit_on_queue (GNUNET_TIME_UNIT_ZERO,
                                  s,
                                  GNUNET_SCHEDULER_PRIORITY_DEFAULT);
  }
  notify_monitors (&neighbour->pid, queue->address, queue->nt, &me);
  GNUNET_free (queue);
 
  vl = lookup_virtual_link (&neighbour->pid);
  if ((NULL != vl) && (neighbour == vl->n))
  {
    GNUNET_SCHEDULER_cancel (vl->visibility_task);
    check_link_down (vl);
  }
  if (NULL == neighbour->queue_head)
  {
    free_neighbour (neighbour, GNUNET_NO);
  }
}

References check_link_down(), COMMUNICATOR_TOTAL_QUEUE_LIMIT, free_neighbour(), GNUNET_assert, GNUNET_CONTAINER_DLL_remove, GNUNET_CONTAINER_MDLL_remove, GNUNET_ERROR_TYPE_DEBUG, GNUNET_free, GNUNET_log, GNUNET_NO, GNUNET_PEERSTORE_monitor_stop(), GNUNET_SCHEDULER_cancel(), GNUNET_SCHEDULER_PRIORITY_DEFAULT, GNUNET_STATISTICS_update(), GNUNET_TIME_UNIT_FOREVER_REL, GNUNET_TIME_UNIT_ZERO, GNUNET_TRANSPORT_CS_DOWN, GST_stats, lookup_virtual_link(), me, VirtualLink::n, notify_monitors(), Neighbour::pid, qe, queue(), PendingAcknowledgement::queue, Neighbour::queue_head, Neighbour::queue_tail, schedule_transmit_on_queue(), tc, and VirtualLink::visibility_task.

Referenced by client_disconnect_cb(), and handle_del_queue_message().

Here is the call graph for this function:

Here is the caller graph for this function:

free_address_list_entry()

static void free_address_list_entry ( struct AddressListEntry *  ale )

static

Free ale.

Parameters

ale	address list entry to free

Definition at line 4141 of file gnunet-service-transport.c.

{
  struct TransportClient *tc = ale->tc;
 
  GNUNET_CONTAINER_DLL_remove (tc->details.communicator.addr_head,
                               tc->details.communicator.addr_tail,
                               ale);
  if (NULL != ale->sc)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "store cancel\n");
    GNUNET_PEERSTORE_store_cancel (ale->sc);
    ale->sc = NULL;
  }
  if (NULL != ale->st)
  {
    GNUNET_SCHEDULER_cancel (ale->st);
    ale->st = NULL;
  }
  if (NULL != ale->signed_address)
    GNUNET_free (ale->signed_address);
  GNUNET_free (ale);
}

References GNUNET_CONTAINER_DLL_remove, GNUNET_ERROR_TYPE_DEBUG, GNUNET_free, GNUNET_log, GNUNET_PEERSTORE_store_cancel(), GNUNET_SCHEDULER_cancel(), AddressListEntry::sc, AddressListEntry::signed_address, AddressListEntry::st, tc, and AddressListEntry::tc.

Referenced by client_disconnect_cb(), and handle_del_address().

Here is the call graph for this function:

Here is the caller graph for this function:

stop_peer_request()

static int stop_peer_request ( void *  cls, const struct GNUNET_PeerIdentity *  pid, void *  value  )

static

Stop the peer request in value.

Parameters

cls	a struct TransportClient that no longer makes the request
pid	the peer's identity
value	a struct PeerRequest

Returns

GNUNET_YES (always)

Definition at line 4175 of file gnunet-service-transport.c.

{
  struct TransportClient *tc = cls;
  struct PeerRequest *pr = value;
 
  if (NULL != pr->nc)
    GNUNET_PEERSTORE_monitor_stop (pr->nc);
  pr->nc = NULL;
  GNUNET_assert (
    GNUNET_YES ==
    GNUNET_CONTAINER_multipeermap_remove (tc->details.application.requests,
                                          pid,
                                          pr));
  GNUNET_free (pr);
 
  return GNUNET_OK;
}

References GNUNET_assert, GNUNET_CONTAINER_multipeermap_remove(), GNUNET_free, GNUNET_OK, GNUNET_PEERSTORE_monitor_stop(), GNUNET_YES, pid, PeerRequest::pr, tc, and value.

Referenced by client_disconnect_cb(), and handle_suggest_cancel().

Here is the call graph for this function:

Here is the caller graph for this function:

do_shutdown()

static void do_shutdown ( void *  cls )

static

Function called when the service shuts down.

Unloads our plugins and cancels pending validations.

Parameters

cls	closure, unused

Definition at line 13366 of file gnunet-service-transport.c.

{
  struct LearnLaunchEntry *lle;
  struct PilsRequest *pr;
  (void) cls;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "shutdown logic\n");
  GNUNET_NAT_unregister (nh);
  GNUNET_CONTAINER_multipeermap_iterate (neighbours,
                                         &free_neighbour_cb, NULL);
  if (NULL != validation_task)
  {
    GNUNET_SCHEDULER_cancel (validation_task);
    validation_task = NULL;
  }
  if (NULL != dvlearn_task)
  {
    GNUNET_SCHEDULER_cancel (dvlearn_task);
    dvlearn_task = NULL;
  }
  GNUNET_CONTAINER_multishortmap_destroy (dvlearn_map);
  dvlearn_map = NULL;
  GNUNET_CONTAINER_multipeermap_iterate (dv_routes, &free_dv_routes_cb, NULL);
  GNUNET_CONTAINER_multipeermap_destroy (dv_routes);
  dv_routes = NULL;
  if (NULL != GST_stats)
  {
    GNUNET_STATISTICS_destroy (GST_stats, GNUNET_NO);
    GST_stats = NULL;
  }
  if (NULL != GST_my_hello)
  {
    GNUNET_HELLO_builder_free (GST_my_hello);
    GST_my_hello = NULL;
  }
  if (NULL != GST_my_identity)
  {
    GNUNET_free (GST_my_identity);
    GST_my_identity = NULL;
  }
  GNUNET_CONTAINER_multipeermap_iterate (ack_cummulators,
                                         &free_ack_cummulator_cb,
                                         NULL);
  GNUNET_CONTAINER_multipeermap_destroy (ack_cummulators);
  ack_cummulators = NULL;
  GNUNET_CONTAINER_multiuuidmap_iterate (pending_acks,
                                         &free_pending_ack_cb,
                                         NULL);
  GNUNET_CONTAINER_multiuuidmap_destroy (pending_acks);
  pending_acks = NULL;
  GNUNET_break (0 == GNUNET_CONTAINER_multipeermap_size (neighbours));
  GNUNET_CONTAINER_multipeermap_destroy (neighbours);
  neighbours = NULL;
  GNUNET_break (0 == GNUNET_CONTAINER_multipeermap_size (links));
  GNUNET_CONTAINER_multipeermap_destroy (links);
  links = NULL;
  GNUNET_CONTAINER_multipeermap_iterate (backtalkers,
                                         &free_backtalker_cb,
                                         NULL);
  GNUNET_CONTAINER_multipeermap_destroy (backtalkers);
  backtalkers = NULL;
  GNUNET_CONTAINER_multipeermap_iterate (validation_map,
                                         &free_validation_state_cb,
                                         NULL);
  GNUNET_CONTAINER_multipeermap_destroy (validation_map);
  validation_map = NULL;
  GNUNET_CONTAINER_heap_destroy (validation_heap);
  validation_heap = NULL;
  GNUNET_CONTAINER_multihashmap_destroy (revalidation_map);
  revalidation_map = NULL;
  while (NULL != ir_head)
    free_incoming_request (ir_head);
  GNUNET_assert (0 == ir_total);
  while (NULL != (lle = lle_head))
  {
    GNUNET_CONTAINER_DLL_remove (lle_head, lle_tail, lle);
    GNUNET_free (lle);
  }
  while (NULL != (pr = pils_requests_head))
  {
    GNUNET_CONTAINER_DLL_remove (pils_requests_head,
                                 pils_requests_tail,
                                 pr);
    if (NULL != pr->op)
      GNUNET_PILS_cancel (pr->op);
    GNUNET_free (pr);
  }
  if (NULL != pils_feed_task)
  {
    GNUNET_SCHEDULER_cancel (pils_feed_task);
    pils_feed_task = NULL;
  }
  if (NULL != pils)
  {
    GNUNET_PILS_disconnect (pils);
    pils = NULL;
  }
  if (NULL != peerstore)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Disconnecting from PEERSTORE service\n");
    GNUNET_PEERSTORE_disconnect (peerstore);
    peerstore = NULL;
  }
  GNUNET_SCHEDULER_shutdown ();
}

References ack_cummulators, backtalkers, dv_routes, dvlearn_map, dvlearn_task, free_ack_cummulator_cb(), free_backtalker_cb(), free_dv_routes_cb(), free_incoming_request(), free_neighbour_cb(), free_pending_ack_cb(), free_validation_state_cb(), GNUNET_assert, GNUNET_break, GNUNET_CONTAINER_DLL_remove, GNUNET_CONTAINER_heap_destroy(), GNUNET_CONTAINER_multihashmap_destroy(), GNUNET_CONTAINER_multipeermap_destroy(), GNUNET_CONTAINER_multipeermap_iterate(), GNUNET_CONTAINER_multipeermap_size(), GNUNET_CONTAINER_multishortmap_destroy(), GNUNET_CONTAINER_multiuuidmap_destroy(), GNUNET_CONTAINER_multiuuidmap_iterate(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_free, GNUNET_HELLO_builder_free(), GNUNET_log, GNUNET_NAT_unregister(), GNUNET_NO, GNUNET_PEERSTORE_disconnect(), GNUNET_PILS_cancel(), GNUNET_PILS_disconnect(), GNUNET_SCHEDULER_cancel(), GNUNET_SCHEDULER_shutdown(), GNUNET_STATISTICS_destroy(), GST_my_hello, GST_my_identity, GST_stats, ir_head, ir_total, links, lle_head, lle_tail, neighbours, nh, PilsRequest::op, peerstore, pending_acks, pils, pils_feed_task, pils_requests_head, pils_requests_tail, revalidation_map, validation_heap, validation_map, and validation_task.

Here is the call graph for this function:

client_disconnect_cb()

static void client_disconnect_cb ( void *  cls, struct GNUNET_SERVICE_Client *  client, void *  app_ctx  )

static

Called whenever a client is disconnected.

Frees our resources associated with that client.

Parameters

cls	closure, NULL
client	identification of the client
app_ctx	our struct TransportClient

Definition at line 4208 of file gnunet-service-transport.c.

{
  struct TransportClient *tc = app_ctx;
 
  (void) cls;
  (void) client;
  GNUNET_CONTAINER_DLL_remove (clients_head, clients_tail, tc);
  switch (tc->type)
  {
  case CT_NONE:
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Unknown Client %p disconnected, cleaning up.\n",
                tc);
    break;
 
  case CT_CORE: {
      struct PendingMessage *pm;
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "CORE Client %p disconnected, cleaning up.\n",
                  tc);
 
 
      while (NULL != (pm = tc->details.core.pending_msg_head))
      {
        GNUNET_CONTAINER_MDLL_remove (client,
                                      tc->details.core.pending_msg_head,
                                      tc->details.core.pending_msg_tail,
                                      pm);
        pm->client = NULL;
      }
    }
    break;
 
  case CT_MONITOR:
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "MONITOR Client %p disconnected, cleaning up.\n",
                tc);
 
    break;
 
  case CT_COMMUNICATOR: {
      struct Queue *q;
      struct AddressListEntry *ale;
 
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "COMMUNICATOR Client %p disconnected, cleaning up.\n",
                  tc);
 
      if (NULL != tc->details.communicator.free_queue_entry_task)
        GNUNET_SCHEDULER_cancel (
          tc->details.communicator.free_queue_entry_task);
      while (NULL != (q = tc->details.communicator.queue_head))
        free_queue (q);
      while (NULL != (ale = tc->details.communicator.addr_head))
        free_address_list_entry (ale);
      GNUNET_free (tc->details.communicator.address_prefix);
    }
    break;
 
  case CT_APPLICATION:
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "APPLICATION Client %p disconnected, cleaning up.\n",
                tc);
 
    GNUNET_CONTAINER_multipeermap_iterate (tc->details.application.requests,
                                           &stop_peer_request,
                                           tc);
    GNUNET_CONTAINER_multipeermap_destroy (tc->details.application.requests);
    break;
  }
  GNUNET_free (tc);
  if ((GNUNET_YES == in_shutdown) && (NULL == clients_head))
  {
    GNUNET_log (GNUNET_ERROR_TYPE_ERROR,
                "Our last client disconnected\n");
    do_shutdown (cls);
  }
}

References PendingMessage::client, TransportClient::client, clients_head, clients_tail, CT_APPLICATION, CT_COMMUNICATOR, CT_CORE, CT_MONITOR, CT_NONE, do_shutdown, free_address_list_entry(), free_queue(), GNUNET_CONTAINER_DLL_remove, GNUNET_CONTAINER_MDLL_remove, GNUNET_CONTAINER_multipeermap_destroy(), GNUNET_CONTAINER_multipeermap_iterate(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_ERROR_TYPE_ERROR, GNUNET_free, GNUNET_log, GNUNET_SCHEDULER_cancel(), GNUNET_YES, in_shutdown, pm, q, stop_peer_request(), and tc.

Here is the call graph for this function:

notify_client_connect_info()

static int notify_client_connect_info ( void *  cls, const struct GNUNET_PeerIdentity *  pid, void *  value  )

static

Iterator telling new CORE client about all existing connections to peers.

Parameters

cls	the new struct TransportClient
pid	a connected peer
value	the struct Neighbour with more information

Returns

GNUNET_OK (continue to iterate)

Definition at line 4300 of file gnunet-service-transport.c.

{
  struct TransportClient *tc = cls;
  struct VirtualLink *vl = value;
 
  if ((NULL == vl) || (GNUNET_NO == vl->confirmed))
    return GNUNET_OK;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Telling new CORE client about existing connection to %s\n",
              GNUNET_i2s (pid));
  core_send_connect_info (tc, pid);
  return GNUNET_OK;
}

References VirtualLink::confirmed, core_send_connect_info(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_i2s(), GNUNET_log, GNUNET_NO, GNUNET_OK, pid, tc, and value.

Referenced by handle_client_start().

Here is the call graph for this function:

Here is the caller graph for this function:

finish_cmc_handling_with_continue()

static void finish_cmc_handling_with_continue ( struct CommunicatorMessageContext *  cmc, unsigned int  free_cmc  )

static

Send ACK to communicator (if requested) and free cmc.

Parameters

cmc	context for which we are done handling the message

Definition at line 4571 of file gnunet-service-transport.c.

{
  if (0 != ntohl (cmc->im.fc_on))
  {
    /* send ACK when done to communicator for flow control! */
    struct GNUNET_MQ_Envelope *env;
    struct GNUNET_TRANSPORT_IncomingMessageAck *ack;
 
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Acknowledge message with flow control id %" PRIu64 "\n",
                cmc->im.fc_id);
    env = GNUNET_MQ_msg (ack, GNUNET_MESSAGE_TYPE_TRANSPORT_INCOMING_MSG_ACK);
    ack->reserved = htonl (0);
    ack->fc_id = cmc->im.fc_id;
    ack->sender = cmc->im.neighbour_sender;
    GNUNET_MQ_send (cmc->tc->mq, env);
  }
 
  GNUNET_SERVICE_client_continue (cmc->tc->client);
 
  if (GNUNET_YES == free_cmc)
  {
    GNUNET_free (cmc);
  }
}

References TransportClient::client, env, GNUNET_TRANSPORT_IncomingMessage::fc_id, GNUNET_TRANSPORT_IncomingMessageAck::fc_id, GNUNET_TRANSPORT_IncomingMessage::fc_on, GNUNET_ERROR_TYPE_DEBUG, GNUNET_free, GNUNET_log, GNUNET_MESSAGE_TYPE_TRANSPORT_INCOMING_MSG_ACK, GNUNET_MQ_msg, GNUNET_MQ_send(), GNUNET_SERVICE_client_continue(), GNUNET_YES, CommunicatorMessageContext::im, TransportClient::mq, GNUNET_TRANSPORT_IncomingMessage::neighbour_sender, GNUNET_TRANSPORT_IncomingMessageAck::reserved, GNUNET_TRANSPORT_IncomingMessageAck::sender, and CommunicatorMessageContext::tc.

Referenced by finish_cmc_handling(), finish_handling_raw_message(), handle_client_recv_ok(), handle_raw_message(), and resume_communicators().

Here is the call graph for this function:

Here is the caller graph for this function:

resume_communicators()

static enum GNUNET_GenericReturnValue resume_communicators ( void *  cls, const struct GNUNET_PeerIdentity *  pid, void *  value  )

static

Definition at line 4329 of file gnunet-service-transport.c.

{
  struct VirtualLink *vl = value;
  struct CommunicatorMessageContext *cmc;
 
  /* resume communicators */
  while (NULL != (cmc = vl->cmc_tail))
  {
    GNUNET_CONTAINER_DLL_remove (vl->cmc_head, vl->cmc_tail, cmc);
    if (GNUNET_NO == cmc->continue_send)
      finish_cmc_handling_with_continue (cmc, GNUNET_YES);
  }
  return GNUNET_OK;
}

References VirtualLink::cmc_head, VirtualLink::cmc_tail, CommunicatorMessageContext::continue_send, finish_cmc_handling_with_continue(), GNUNET_CONTAINER_DLL_remove, GNUNET_NO, GNUNET_OK, GNUNET_YES, and value.

Referenced by handle_client_start().

Here is the call graph for this function:

Here is the caller graph for this function:

handle_client_start()

static void handle_client_start ( void *  cls, const struct StartMessage *  start  )

static

Initialize a "CORE" client.

We got a start message from this client, so add it to the list of clients for broadcasting of inbound messages.

Parameters

cls	the client
start	the start message that was sent

Definition at line 4356 of file gnunet-service-transport.c.

{
  struct TransportClient *tc = cls;
  // uint32_t options;
  //
  // FIXME ignore the check of the peer ids for now.
  //       (also deprecate the old way of obtaining our own peer ID)
  // options = ntohl (start->options);
  // if ((0 != (1 & options)) &&
  //    (0 != GNUNET_memcmp (&start->self, &GST_my_identity)))
  // {
  //  /* client thinks this is a different peer, reject */
  //  GNUNET_break (0);
  //  GNUNET_SERVICE_client_drop (tc->client);
  //  return;
  // }
  if (CT_NONE != tc->type)
  {
    GNUNET_break (0);
    GNUNET_SERVICE_client_drop (tc->client);
    return;
  }
  tc->type = CT_CORE;
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "New CORE client with PID %s registered\n",
              GNUNET_i2s (&start->self));
  GNUNET_CONTAINER_multipeermap_iterate (links,
                                         &notify_client_connect_info,
                                         tc);
  GNUNET_CONTAINER_multipeermap_iterate (links,
                                         &resume_communicators,
                                         NULL);
  GNUNET_SERVICE_client_continue (tc->client);
}

References CT_CORE, CT_NONE, GNUNET_break, GNUNET_CONTAINER_multipeermap_iterate(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_i2s(), GNUNET_log, GNUNET_SERVICE_client_continue(), GNUNET_SERVICE_client_drop(), links, notify_client_connect_info(), resume_communicators(), start, and tc.

Here is the call graph for this function:

check_client_send()

static int check_client_send ( void *  cls, const struct OutboundMessage *  obm  )

static

Client asked for transmission to a peer.

Process the request.

Parameters

cls	the client
obm	the send message that was sent

Definition at line 4399 of file gnunet-service-transport.c.

{
  struct TransportClient *tc = cls;
  uint16_t size;
  const struct GNUNET_MessageHeader *obmm;
 
  if (CT_CORE != tc->type)
  {
    GNUNET_break (0);
    return GNUNET_SYSERR;
  }
  size = ntohs (obm->header.size) - sizeof(struct OutboundMessage);
  if (size < sizeof(struct GNUNET_MessageHeader))
  {
    GNUNET_break (0);
    return GNUNET_SYSERR;
  }
  obmm = (const struct GNUNET_MessageHeader *) &obm[1];
  if (size != ntohs (obmm->size))
  {
    GNUNET_break (0);
    return GNUNET_SYSERR;
  }
  return GNUNET_OK;
}

References CT_CORE, GNUNET_break, GNUNET_OK, GNUNET_SYSERR, OutboundMessage::header, GNUNET_MessageHeader::size, size, and tc.

client_send_response()

static void client_send_response ( struct PendingMessage *  pm )

static

Send a response to the pm that we have processed a "send" request.

Sends a confirmation to the "core" client responsible for the original request and free's pm.

Parameters

pm	handle to the original pending message

Definition at line 4434 of file gnunet-service-transport.c.

{
  struct TransportClient *tc = pm->client;
  struct VirtualLink *vl = pm->vl;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "client send response\n");
  if (NULL != tc)
  {
    struct GNUNET_MQ_Envelope *env;
    struct SendOkMessage *so_msg;
 
    env = GNUNET_MQ_msg (so_msg, GNUNET_MESSAGE_TYPE_TRANSPORT_SEND_OK);
    so_msg->peer = vl->target;
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Confirming transmission of <%" PRIu64 "> to %s\n",
                pm->logging_uuid,
                GNUNET_i2s (&vl->target));
    GNUNET_MQ_send (tc->mq, env);
  }
  free_pending_message (pm);
}

References env, free_pending_message(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_i2s(), GNUNET_log, GNUNET_MESSAGE_TYPE_TRANSPORT_SEND_OK, GNUNET_MQ_msg, GNUNET_MQ_send(), SendOkMessage::peer, pm, VirtualLink::target, and tc.

Referenced by completed_pending_message(), and reliability_box_message().

Here is the call graph for this function:

Here is the caller graph for this function:

pick_random_dv_hops()

static unsigned int pick_random_dv_hops ( const struct DistanceVector *  dv, enum RouteMessageOptions  options, struct DistanceVectorHop **  hops_array, unsigned int  hops_array_length  )

static

Pick hops_array_length random DV paths satisfying options.

Parameters

	dv	data structure to pick paths from
	options	constraints to satisfy
[out]	hops_array	set to the result
	hops_array_length	length of the hops_array

Returns

number of entries set in hops_array

Definition at line 4468 of file gnunet-service-transport.c.

{
  uint64_t choices[hops_array_length];
  uint64_t num_dv;
  unsigned int dv_count;
 
  /* Pick random vectors, but weighted by distance, giving more weight
     to shorter vectors */
  num_dv = 0;
  dv_count = 0;
  for (struct DistanceVectorHop *pos = dv->dv_head; NULL != pos;
       pos = pos->next_dv)
  {
    if ((0 == (options & RMO_UNCONFIRMED_ALLOWED)) &&
        (GNUNET_TIME_absolute_get_remaining (pos->path_valid_until)
         .rel_value_us == 0))
      continue;   /* pos unconfirmed and confirmed required */
    num_dv += MAX_DV_HOPS_ALLOWED - pos->distance;
    dv_count++;
  }
  if (0 == dv_count)
    return 0;
  if (dv_count <= hops_array_length)
  {
    dv_count = 0;
    for (struct DistanceVectorHop *pos = dv->dv_head; NULL != pos;
         pos = pos->next_dv)
      hops_array[dv_count++] = pos;
    return dv_count;
  }
  for (unsigned int i = 0; i < hops_array_length; i++)
  {
    int ok = GNUNET_NO;
    while (GNUNET_NO == ok)
    {
      choices[i] =
        GNUNET_CRYPTO_random_u64 (GNUNET_CRYPTO_QUALITY_WEAK, num_dv);
      ok = GNUNET_YES;
      for (unsigned int j = 0; j < i; j++)
        if (choices[i] == choices[j])
        {
          ok = GNUNET_NO;
          break;
        }
    }
  }
  dv_count = 0;
  num_dv = 0;
  for (struct DistanceVectorHop *pos = dv->dv_head; NULL != pos;
       pos = pos->next_dv)
  {
    uint32_t delta = MAX_DV_HOPS_ALLOWED - pos->distance;
 
    if ((0 == (options & RMO_UNCONFIRMED_ALLOWED)) &&
        (GNUNET_TIME_absolute_get_remaining (pos->path_valid_until)
         .rel_value_us == 0))
      continue;   /* pos unconfirmed and confirmed required */
    for (unsigned int i = 0; i < hops_array_length; i++)
      if ((num_dv <= choices[i]) && (num_dv + delta > choices[i]))
        hops_array[dv_count++] = pos;
    num_dv += delta;
  }
  return dv_count;
}

References delta, DistanceVector::dv_head, GNUNET_CRYPTO_QUALITY_WEAK, GNUNET_CRYPTO_random_u64(), GNUNET_NO, GNUNET_TIME_absolute_get_remaining(), GNUNET_YES, MAX_DV_HOPS_ALLOWED, options, GNUNET_TIME_Relative::rel_value_us, and RMO_UNCONFIRMED_ALLOWED.

Referenced by route_control_message_without_fc().

Here is the call graph for this function:

Here is the caller graph for this function:

check_communicator_available()

static int check_communicator_available ( void *  cls, const struct GNUNET_TRANSPORT_CommunicatorAvailableMessage *  cam  )

static

Communicator started.

Test message is well-formed.

Parameters

cls	the client
cam	the send message that was sent

Definition at line 4544 of file gnunet-service-transport.c.

{
  struct TransportClient *tc = cls;
  uint16_t size;
 
  if (CT_NONE != tc->type)
  {
    GNUNET_break (0);
    return GNUNET_SYSERR;
  }
  tc->type = CT_COMMUNICATOR;
  size = ntohs (cam->header.size) - sizeof(*cam);
  if (0 == size)
    return GNUNET_OK; /* receive-only communicator */
  GNUNET_MQ_check_zero_termination (cam);
  return GNUNET_OK;
}

References CT_COMMUNICATOR, CT_NONE, GNUNET_break, GNUNET_MQ_check_zero_termination, GNUNET_OK, GNUNET_SYSERR, GNUNET_TRANSPORT_CommunicatorAvailableMessage::header, GNUNET_MessageHeader::size, size, and tc.

finish_cmc_handling()

static void finish_cmc_handling ( struct CommunicatorMessageContext *  cmc )

static

Definition at line 4601 of file gnunet-service-transport.c.

{
  finish_cmc_handling_with_continue (cmc, GNUNET_YES);
}

References finish_cmc_handling_with_continue(), and GNUNET_YES.

Referenced by backtalker_monotime_cb(), decaps_dv_box_cb(), free_backtalker(), handle_backchannel_encapsulation(), handle_dv_box(), handle_dv_learn(), handle_flow_control(), handle_fragment_box(), handle_raw_message(), handle_reliability_ack(), handle_validation_challenge(), handle_validation_response(), and sign_t_validation_cb().

Here is the call graph for this function:

Here is the caller graph for this function:

handle_client_recv_ok()

static void handle_client_recv_ok ( void *  cls, const struct RecvOkMessage *  rom  )

static

Client confirms that it is done handling message(s) to a particular peer.

We may now provide more messages to CORE for this peer.

Notifies the respective queues that more messages can now be received.

Parameters

cls	the client
rom	the message that was sent

Definition at line 4617 of file gnunet-service-transport.c.

{
  struct TransportClient *tc = cls;
  struct VirtualLink *vl;
  uint32_t delta;
  struct CommunicatorMessageContext *cmc;
 
  if (CT_CORE != tc->type)
  {
    GNUNET_break (0);
    GNUNET_SERVICE_client_drop (tc->client);
    return;
  }
  vl = lookup_virtual_link (&rom->peer);
  if ((NULL == vl) || (GNUNET_NO == vl->confirmed))
  {
    GNUNET_STATISTICS_update (GST_stats,
                              "# RECV_OK dropped: virtual link unknown",
                              1,
                              GNUNET_NO);
    GNUNET_SERVICE_client_continue (tc->client);
    return;
  }
  delta = ntohl (rom->increase_window_delta);
  vl->core_recv_window += delta;
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "CORE ack receiving message, increased CORE recv window to %d\n",
              vl->core_recv_window);
  GNUNET_SERVICE_client_continue (tc->client);
  if (vl->core_recv_window <= 0)
    return;
  /* resume communicators */
  while (NULL != (cmc = vl->cmc_tail))
  {
    GNUNET_CONTAINER_DLL_remove (vl->cmc_head, vl->cmc_tail, cmc);
    if (GNUNET_NO == cmc->continue_send)
      finish_cmc_handling_with_continue (cmc, GNUNET_YES);
  }
}

References VirtualLink::cmc_head, VirtualLink::cmc_tail, VirtualLink::confirmed, CommunicatorMessageContext::continue_send, VirtualLink::core_recv_window, CT_CORE, delta, finish_cmc_handling_with_continue(), GNUNET_break, GNUNET_CONTAINER_DLL_remove, GNUNET_ERROR_TYPE_DEBUG, GNUNET_log, GNUNET_NO, GNUNET_SERVICE_client_continue(), GNUNET_SERVICE_client_drop(), GNUNET_STATISTICS_update(), GNUNET_YES, GST_stats, RecvOkMessage::increase_window_delta, lookup_virtual_link(), RecvOkMessage::peer, and tc.

Here is the call graph for this function:

handle_communicator_available()

static void handle_communicator_available ( void *  cls, const struct GNUNET_TRANSPORT_CommunicatorAvailableMessage *  cam  )

static

Communicator started.

Process the request.

Parameters

cls	the client
cam	the send message that was sent

Definition at line 4665 of file gnunet-service-transport.c.

{
  struct TransportClient *tc = cls;
  uint16_t size;
 
  size = ntohs (cam->header.size) - sizeof(*cam);
  if (0 == size)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Receive-only communicator connected\n");
    return;   /* receive-only communicator */
  }
  tc->details.communicator.address_prefix =
    GNUNET_strdup ((const char *) &cam[1]);
  tc->details.communicator.cc = ntohl (cam->cc);
  tc->details.communicator.can_burst = ntohl (cam->can_burst);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Communicator for peer %s with prefix '%s' connected %s\n",
              GNUNET_i2s (GST_my_identity),
              tc->details.communicator.address_prefix,
              tc->details.communicator.can_burst ? "can burst" :
              "can not burst");
  GNUNET_SERVICE_client_continue (tc->client);
}

References GNUNET_TRANSPORT_CommunicatorAvailableMessage::can_burst, GNUNET_TRANSPORT_CommunicatorAvailableMessage::cc, GNUNET_ERROR_TYPE_DEBUG, GNUNET_i2s(), GNUNET_log, GNUNET_SERVICE_client_continue(), GNUNET_strdup, GST_my_identity, GNUNET_TRANSPORT_CommunicatorAvailableMessage::header, GNUNET_MessageHeader::size, size, and tc.

Here is the call graph for this function:

check_communicator_backchannel()

static int check_communicator_backchannel ( void *  cls, const struct GNUNET_TRANSPORT_CommunicatorBackchannel *  cb  )

static

Communicator requests backchannel transmission.

Check the request.

Parameters

cls	the client
cb	the send message that was sent

Returns

GNUNET_OK if message is well-formed

Definition at line 4701 of file gnunet-service-transport.c.

{
  const struct GNUNET_MessageHeader *inbox;
  const char *is;
  uint16_t msize;
  uint16_t isize;
 
  (void) cls;
  msize = ntohs (cb->header.size) - sizeof(*cb);
  inbox = (const struct GNUNET_MessageHeader *) &cb[1];
  isize = ntohs (inbox->size);
  if (isize >= msize)
  {
    GNUNET_break (0);
    return GNUNET_SYSERR;
  }
  is = (const char *) inbox;
  is += isize;
  msize -= isize;
  GNUNET_assert (0 < msize);
  if ('\0' != is[msize - 1])
  {
    GNUNET_break (0);
    return GNUNET_SYSERR;
  }
  return GNUNET_OK;
}

References GNUNET_assert, GNUNET_break, GNUNET_OK, GNUNET_SYSERR, GNUNET_TRANSPORT_CommunicatorBackchannel::header, is, and GNUNET_MessageHeader::size.

sign_dv_cb()

static void sign_dv_cb ( void *  cls, const struct GNUNET_PeerIdentity *  pid, const struct GNUNET_CRYPTO_EddsaSignature *  sig  )

static

Definition at line 4740 of file gnunet-service-transport.c.

{
  struct SignDvCls *sign_dv_cls = cls;
  struct DistanceVector *dv = sign_dv_cls->dv;
  struct PilsRequest *pr = sign_dv_cls->req;
 
  pr->op = NULL;
  GNUNET_CONTAINER_DLL_remove (pils_requests_head,
                               pils_requests_tail,
                               pr);
  GNUNET_free (pr);
 
  dv->sender_sig = *sig;
}

References SignDvCls::dv, GNUNET_CONTAINER_DLL_remove, GNUNET_free, PilsRequest::op, pils_requests_head, pils_requests_tail, SignDvCls::req, and DistanceVector::sender_sig.

Referenced by sign_ephemeral().

Here is the caller graph for this function:

sign_ephemeral()

static void sign_ephemeral ( struct DistanceVector *  dv )

static

Sign ephemeral keys in our dv are current.

Parameters

[in,out]

dv

virtual link to update ephemeral for

Definition at line 4764 of file gnunet-service-transport.c.

{
  struct EphemeralConfirmationPS ec;
  struct SignDvCls *sign_dv_cls;
 
  dv->monotime = GNUNET_TIME_absolute_get_monotonic (GST_cfg);
  dv->ephemeral_validity =
    GNUNET_TIME_absolute_add (dv->monotime, EPHEMERAL_VALIDITY);
  ec.purpose.purpose = htonl (GNUNET_SIGNATURE_PURPOSE_TRANSPORT_EPHEMERAL);
  ec.target = dv->target;
  ec.ephemeral_key = dv->ephemeral_key;
  ec.sender_monotonic_time = GNUNET_TIME_absolute_hton (dv->monotime);
  ec.purpose.size = htonl (sizeof(ec));
  sign_dv_cls = GNUNET_new (struct SignDvCls);
  sign_dv_cls->req = GNUNET_new (struct PilsRequest);
  sign_dv_cls->dv = dv;
  GNUNET_CONTAINER_DLL_insert (pils_requests_head,
                               pils_requests_tail,
                               sign_dv_cls->req);
  sign_dv_cls->req->op = GNUNET_PILS_sign_by_peer_identity (pils,
                                                            &ec.purpose,
                                                            sign_dv_cb,
                                                            sign_dv_cls);
}

References SignDvCls::dv, EphemeralConfirmationPS::ephemeral_key, DistanceVector::ephemeral_key, EPHEMERAL_VALIDITY, DistanceVector::ephemeral_validity, GNUNET_CONTAINER_DLL_insert, GNUNET_new, GNUNET_PILS_sign_by_peer_identity(), GNUNET_SIGNATURE_PURPOSE_TRANSPORT_EPHEMERAL, GNUNET_TIME_absolute_add(), GNUNET_TIME_absolute_get_monotonic(), GNUNET_TIME_absolute_hton(), GST_cfg, DistanceVector::monotime, PilsRequest::op, pils, pils_requests_head, pils_requests_tail, GNUNET_CRYPTO_EccSignaturePurpose::purpose, EphemeralConfirmationPS::purpose, SignDvCls::req, EphemeralConfirmationPS::sender_monotonic_time, sign_dv_cb(), GNUNET_CRYPTO_EccSignaturePurpose::size, EphemeralConfirmationPS::target, and DistanceVector::target.

Referenced by encapsulate_for_dv().

Here is the call graph for this function:

Here is the caller graph for this function:

free_queue_entry()

static void free_queue_entry ( struct QueueEntry *  qe, struct TransportClient *  tc  )

static

Definition at line 11693 of file gnunet-service-transport.c.

{
  struct PendingMessage *pm;
 
  GNUNET_CONTAINER_DLL_remove (qe->queue->queue_head,
                               qe->queue->queue_tail,
                               qe);
  qe->queue->queue_length--;
  tc->details.communicator.total_queue_length--;
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Received ACK on queue %s (QID %u) to peer %s (new length: %u/%u)\n",
              qe->queue->address,
              qe->queue->qid,
              GNUNET_i2s (&qe->queue->neighbour->pid),
              qe->queue->queue_length,
              tc->details.communicator.total_queue_length);
 
  /* if applicable, resume transmissions that waited on ACK */
  if (COMMUNICATOR_TOTAL_QUEUE_LIMIT - 1 ==
      tc->details.communicator.total_queue_length)
  {
    /* Communicator dropped below threshold, resume all queues
       incident with this client! */
    GNUNET_STATISTICS_update (
      GST_stats,
      "# Transmission throttled due to communicator queue limit",
      -1,
      GNUNET_NO);
    for (struct Queue *queue = tc->details.communicator.queue_head;
         NULL != queue;
         queue = queue->next_client)
    {
      schedule_transmit_on_queue (GNUNET_TIME_UNIT_ZERO,
                                  queue,
                                  GNUNET_SCHEDULER_PRIORITY_DEFAULT);
    }
  }
  else if (QUEUE_LENGTH_LIMIT - 1 == qe->queue->queue_length)
  {
    /* queue dropped below threshold; only resume this one queue */
    GNUNET_STATISTICS_update (GST_stats,
                              "# Transmission throttled due to queue queue limit",
                              -1,
                              GNUNET_NO);
    schedule_transmit_on_queue (GNUNET_TIME_UNIT_ZERO,
                                qe->queue,
                                GNUNET_SCHEDULER_PRIORITY_DEFAULT);
  }
  else if (1 == qe->queue->q_capacity)
  {
    // TODO I guess this will never happen, because the communicator triggers this by updating its queue length itself.
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Transmission rescheduled due to communicator message queue with qid %u has capacity %"
                PRIu64 ".\n",
                qe->queue->qid,
                qe->queue->q_capacity);
    /* message queue has capacity; only resume this one queue */
    /* queue dropped below threshold; only resume this one queue */
    GNUNET_STATISTICS_update (GST_stats,
                              "# Transmission throttled due to message queue capacity",
                              -1,
                              GNUNET_NO);
    schedule_transmit_on_queue (GNUNET_TIME_UNIT_ZERO,
                                qe->queue,
                                GNUNET_SCHEDULER_PRIORITY_DEFAULT);
  }
 
  if (NULL != (pm = qe->pm))
  {
    struct VirtualLink *vl;
 
    // GNUNET_assert (qe == pm->qe);
    pm->qe = NULL;
    /* If waiting for this communicator may have blocked transmission
       of pm on other queues for this neighbour, force schedule
       transmit on queue for queues of the neighbour */
    if (NULL == pm->frag_parent)
    {
      vl = pm->vl;
      if ((NULL != vl) &&
          (NULL != vl->pending_msg_head) &&
          (vl->pending_msg_head == pm))
        check_vl_transmission (vl);
    }
  }
  GNUNET_free (qe);
}

References check_vl_transmission(), COMMUNICATOR_TOTAL_QUEUE_LIMIT, GNUNET_CONTAINER_DLL_remove, GNUNET_ERROR_TYPE_DEBUG, GNUNET_free, GNUNET_i2s(), GNUNET_log, GNUNET_NO, GNUNET_SCHEDULER_PRIORITY_DEFAULT, GNUNET_STATISTICS_update(), GNUNET_TIME_UNIT_ZERO, GST_stats, VirtualLink::pending_msg_head, pm, qe, queue(), QUEUE_LENGTH_LIMIT, schedule_transmit_on_queue(), and tc.

Referenced by free_timedout_queue_entry(), and handle_send_message_ack().

Here is the call graph for this function:

Here is the caller graph for this function:

free_timedout_queue_entry()

static void free_timedout_queue_entry ( void *  cls )

static

Definition at line 4796 of file gnunet-service-transport.c.

{
  struct TransportClient *tc = cls;
  struct GNUNET_TIME_Absolute now = GNUNET_TIME_absolute_get ();
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "freeing timedout queue entries\n");
 
  tc->details.communicator.free_queue_entry_task = NULL;
  for (struct Queue *queue = tc->details.communicator.queue_head; NULL != queue;
       queue = queue->next_client)
  {
    struct QueueEntry *qep = queue->queue_head;
 
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "checking QID %u for timedout queue entries\n",
                queue->qid);
    while (NULL != qep)
    {
      struct QueueEntry *pos = qep;
      struct GNUNET_TIME_Relative diff = GNUNET_TIME_absolute_get_difference (
        pos->creation_timestamp, now);
      qep = qep->next;
 
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "diff to now %s \n",
                  GNUNET_TIME_relative2s (diff, GNUNET_NO));
      if (GNUNET_TIME_relative_cmp (QUEUE_ENTRY_TIMEOUT, <, diff))
      {
        GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                    "Freeing timed out QueueEntry with MID %" PRIu64
                    " and QID %u\n",
                    pos->mid,
                    queue->qid);
        free_queue_entry (pos, tc);
      }
    }
  }
}

References QueueEntry::creation_timestamp, free_queue_entry(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_log, GNUNET_NO, GNUNET_TIME_absolute_get(), GNUNET_TIME_absolute_get_difference(), GNUNET_TIME_relative2s(), GNUNET_TIME_relative_cmp, QueueEntry::mid, QueueEntry::next, queue(), QUEUE_ENTRY_TIMEOUT, and tc.

Referenced by queue_send_msg().

Here is the call graph for this function:

Here is the caller graph for this function:

queue_send_msg()

static void queue_send_msg ( struct Queue *  queue, struct PendingMessage *  pm, const void *  payload, size_t  payload_size  )

static

Send the message payload on queue.

Parameters

queue	the queue to use for transmission
pm	pending message to update once transmission is done, may be NULL!
payload	the payload to send (encapsulated in a GNUNET_MESSAGE_TYPE_TRANSPORT_SEND_MSG).
payload_size	number of bytes in payload

Definition at line 4847 of file gnunet-service-transport.c.

{
  struct Neighbour *n = queue->neighbour;
  struct GNUNET_TRANSPORT_SendMessageTo *smt;
  struct GNUNET_MQ_Envelope *env;
  struct PendingAcknowledgement *pa;
 
  GNUNET_log (
    GNUNET_ERROR_TYPE_DEBUG,
    "Queueing %u bytes of payload for transmission <%" PRIu64
    "> on queue %llu to %s\n",
    (unsigned int) payload_size,
    (NULL == pm) ? 0 : pm->logging_uuid,
    (unsigned long long) queue->qid,
    GNUNET_i2s (&queue->neighbour->pid));
  env = GNUNET_MQ_msg_extra (smt,
                             payload_size,
                             GNUNET_MESSAGE_TYPE_TRANSPORT_SEND_MSG);
  smt->qid = htonl (queue->qid);
  smt->mid = GNUNET_htonll (queue->mid_gen);
  smt->receiver = n->pid;
  memcpy (&smt[1], payload, payload_size);
  {
    /* Pass the env to the communicator of queue for transmission. */
    struct QueueEntry *qe;
 
    qe = GNUNET_new (struct QueueEntry);
    qe->creation_timestamp = GNUNET_TIME_absolute_get ();
    qe->mid = queue->mid_gen;
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Create QueueEntry with MID %" PRIu64
                " and QID %u and prefix %s\n",
                qe->mid,
                queue->qid,
                queue->tc->details.communicator.address_prefix);
    queue->mid_gen++;
    qe->queue = queue;
    if (NULL != pm)
    {
      qe->pm = pm;
      // TODO Why do we have a retransmission. When we know, make decision if we still want this.
      // GNUNET_assert (NULL == pm->qe);
      if (NULL != pm->qe)
      {
        GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                    "Retransmitting message <%" PRIu64
                    "> remove pm from qe with MID: %llu \n",
                    pm->logging_uuid,
                    (unsigned long long) pm->qe->mid);
        pm->qe->pm = NULL;
      }
      pm->qe = qe;
    }
    GNUNET_assert (CT_COMMUNICATOR == queue->tc->type);
    if (0 == queue->q_capacity)
    {
      // Messages without FC or fragments can get here.
      if (NULL != pm)
      {
        GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                    "Message %" PRIu64
                    " (pm type %u) was not send because queue has no capacity.\n",
                    pm->logging_uuid,
                    pm->pmt);
        pm->qe = NULL;
      }
      GNUNET_free (env);
      GNUNET_free (qe);
      return;
    }
    GNUNET_CONTAINER_DLL_insert (queue->queue_head, queue->queue_tail, qe);
    queue->queue_length++;
    queue->tc->details.communicator.total_queue_length++;
    if (GNUNET_NO == queue->unlimited_length)
      queue->q_capacity--;
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Queue %s with qid %u has capacity %" PRIu64 "\n",
                queue->address,
                queue->qid,
                queue->q_capacity);
    if (COMMUNICATOR_TOTAL_QUEUE_LIMIT ==
        queue->tc->details.communicator.total_queue_length)
      queue->idle = GNUNET_NO;
    if (QUEUE_LENGTH_LIMIT == queue->queue_length)
      queue->idle = GNUNET_NO;
    if (0 == queue->q_capacity)
      queue->idle = GNUNET_NO;
 
    if (GNUNET_NO == queue->idle)
    {
      struct TransportClient *tc = queue->tc;
 
      if (NULL == tc->details.communicator.free_queue_entry_task)
        tc->details.communicator.free_queue_entry_task =
          GNUNET_SCHEDULER_add_delayed (GNUNET_TIME_UNIT_SECONDS,
                                        &
                                        free_timedout_queue_entry,
                                        tc);
    }
    if (NULL != pm && NULL != (pa = pm->pa_head))
    {
      while (pm != pa->pm)
        pa = pa->next_pa;
      pa->num_send++;
    }
    // GNUNET_CONTAINER_multiuuidmap_get (pending_acks, &ack[i].ack_uuid.value);
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Sending message MID %" PRIu64
                " of type %u (%u) and size %lu with MQ %p queue %s (QID %u) pending %"
                PRIu64 "\n",
                GNUNET_ntohll (smt->mid),
                ntohs (((const struct GNUNET_MessageHeader *) payload)->type),
                ntohs (smt->header.size),
                (unsigned long) payload_size,
                queue->tc->mq,
                queue->address,
                queue->qid,
                (NULL == pm) ? 0 : pm->logging_uuid);
    GNUNET_MQ_send (queue->tc->mq, env);
  }
}

References COMMUNICATOR_TOTAL_QUEUE_LIMIT, CT_COMMUNICATOR, env, free_timedout_queue_entry(), GNUNET_assert, GNUNET_CONTAINER_DLL_insert, GNUNET_ERROR_TYPE_DEBUG, GNUNET_free, GNUNET_htonll(), GNUNET_i2s(), GNUNET_log, GNUNET_MESSAGE_TYPE_TRANSPORT_SEND_MSG, GNUNET_MQ_msg_extra, GNUNET_MQ_send(), GNUNET_new, GNUNET_NO, GNUNET_ntohll(), GNUNET_SCHEDULER_add_delayed(), GNUNET_TIME_absolute_get(), GNUNET_TIME_UNIT_SECONDS, GNUNET_TRANSPORT_SendMessageTo::header, GNUNET_TRANSPORT_SendMessageTo::mid, PendingAcknowledgement::next_pa, PendingAcknowledgement::num_send, payload, Neighbour::pid, PendingAcknowledgement::pm, pm, qe, GNUNET_TRANSPORT_SendMessageTo::qid, queue(), QUEUE_LENGTH_LIMIT, GNUNET_TRANSPORT_SendMessageTo::receiver, GNUNET_MessageHeader::size, tc, and type.

Referenced by route_via_neighbour(), sign_dv_init_cb(), transmit_on_queue(), and validation_transmit_on_queue().

Here is the call graph for this function:

Here is the caller graph for this function:

route_via_neighbour()

static struct GNUNET_TIME_Relative route_via_neighbour ( const struct Neighbour *  n, const struct GNUNET_MessageHeader *  hdr, enum RouteMessageOptions  options  )

static

Pick a queue of n under constraints options and schedule transmission of hdr.

Parameters

n	neighbour to send to
hdr	message to send as payload
options	whether queues must be confirmed or not, and whether we may pick multiple (2) queues

Returns

expected RTT for transmission, GNUNET_TIME_UNIT_FOREVER_REL if sending failed

Definition at line 4984 of file gnunet-service-transport.c.

{
  struct GNUNET_TIME_Absolute now;
  unsigned int candidates;
  unsigned int sel1;
  unsigned int sel2;
  struct GNUNET_TIME_Relative rtt;
 
  /* Pick one or two 'random' queues from n (under constraints of options) */
  now = GNUNET_TIME_absolute_get ();
  /* FIXME-OPTIMIZE: give queues 'weights' and pick proportional to
     weight in the future; weight could be assigned by observed
     bandwidth (note: not sure if we should do this for this type
     of control traffic though). */
  candidates = 0;
  for (struct Queue *pos = n->queue_head; NULL != pos;
       pos = pos->next_neighbour)
  {
    if ((0 != (options & RMO_UNCONFIRMED_ALLOWED)) ||
        (pos->validated_until.abs_value_us > now.abs_value_us))
      candidates++;
  }
  if (0 == candidates)
  {
    /* This can happen rarely if the last confirmed queue timed
       out just as we were beginning to process this message. */
    GNUNET_log (GNUNET_ERROR_TYPE_INFO,
                "Could not route message of type %u to %s: no valid queue\n",
                ntohs (hdr->type),
                GNUNET_i2s (&n->pid));
    GNUNET_STATISTICS_update (GST_stats,
                              "# route selection failed (all no valid queue)",
                              1,
                              GNUNET_NO);
    return GNUNET_TIME_UNIT_FOREVER_REL;
  }
 
  rtt = GNUNET_TIME_UNIT_FOREVER_REL;
  sel1 = GNUNET_CRYPTO_random_u32 (GNUNET_CRYPTO_QUALITY_WEAK, candidates);
  if (0 == (options & RMO_REDUNDANT))
    sel2 = candidates; /* picks none! */
  else
    sel2 = GNUNET_CRYPTO_random_u32 (GNUNET_CRYPTO_QUALITY_WEAK, candidates);
  candidates = 0;
  for (struct Queue *pos = n->queue_head; NULL != pos;
       pos = pos->next_neighbour)
  {
    if ((0 != (options & RMO_UNCONFIRMED_ALLOWED)) ||
        (pos->validated_until.abs_value_us > now.abs_value_us))
    {
      if ((sel1 == candidates) || (sel2 == candidates))
      {
        GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                    "Routing message of type %u to %s using %s (#%u)\n",
                    ntohs (hdr->type),
                    GNUNET_i2s (&n->pid),
                    pos->address,
                    (sel1 == candidates) ? 1 : 2);
        rtt = GNUNET_TIME_relative_min (rtt, pos->pd.aged_rtt);
        queue_send_msg (pos, NULL, hdr, ntohs (hdr->size));
      }
      candidates++;
    }
  }
  return rtt;
}

References GNUNET_TIME_Absolute::abs_value_us, GNUNET_CRYPTO_QUALITY_WEAK, GNUNET_CRYPTO_random_u32(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_ERROR_TYPE_INFO, GNUNET_i2s(), GNUNET_log, GNUNET_NO, GNUNET_STATISTICS_update(), GNUNET_TIME_absolute_get(), GNUNET_TIME_relative_min(), GNUNET_TIME_UNIT_FOREVER_REL, GST_stats, options, queue_send_msg(), RMO_REDUNDANT, RMO_UNCONFIRMED_ALLOWED, GNUNET_MessageHeader::size, and GNUNET_MessageHeader::type.

Referenced by forward_dv_box(), handle_communicator_backchannel(), route_control_message_without_fc(), send_dv_to_neighbour(), sign_dhp_cp(), sign_t_validation_cb(), and transmit_cummulative_ack_cb().

Here is the call graph for this function:

Here is the caller graph for this function:

dv_setup_key_state_from_km()

static void dv_setup_key_state_from_km ( const struct GNUNET_ShortHashCode *  km, const struct GNUNET_ShortHashCode *  iv, struct DVKeyState *  key  )

static

Given the key material in km and the initialization vector iv, setup the key material for the backchannel in key.

Parameters

	km	raw master secret
	iv	initialization vector
[out]	key	symmetric cipher and HMAC state to generate

Definition at line 5096 of file gnunet-service-transport.c.

{
  /* must match what we defive from decapsulated key */
  GNUNET_assert (GNUNET_YES ==
                 GNUNET_CRYPTO_hkdf_expand (&key->material,
                                            sizeof(key->material),
                                            km,
                                            "gnunet-transport-dv-key",
                                            strlen ("gnunet-transport-dv-key")
                                            ,
                                            km,
                                            sizeof(*km),
                                            iv,
                                            sizeof(*iv),
                                            NULL));
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Deriving backchannel key based on KM %s and IV %s\n",
              GNUNET_sh2s (km),
              GNUNET_sh2s (iv));
  GNUNET_assert (0 == gcry_cipher_open (&key->cipher,
                                        GCRY_CIPHER_AES256 /* low level: go for speed */
                                        ,
                                        GCRY_CIPHER_MODE_CTR,
                                        0 /* flags */));
  GNUNET_assert (0 == gcry_cipher_setkey (key->cipher,
                                          &key->material.aes_key,
                                          sizeof(key->material.aes_key)));
  gcry_cipher_setctr (key->cipher,
                      &key->material.aes_ctr,
                      sizeof(key->material.aes_ctr));
}

References GNUNET_assert, GNUNET_CRYPTO_hkdf_expand(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_log, GNUNET_sh2s(), GNUNET_YES, and key.

Referenced by decaps_dv_box_cb(), and encapsulate_for_dv().

Here is the call graph for this function:

Here is the caller graph for this function:

dv_hmac()

static void dv_hmac ( const struct DVKeyState *  key, struct GNUNET_HashCode *  hmac, const void *  data, size_t  data_size  )

static

Do HMAC calculation for backchannel messages over data using key material from key.

Parameters

	key	key material (from DH)
[out]	hmac	set to the HMAC
	data	data to perform HMAC calculation over
	data_size	number of bytes in data

Definition at line 5141 of file gnunet-service-transport.c.

{
  GNUNET_CRYPTO_hmac (&key->material.hmac_key, data, data_size, hmac);
}

References data, data_size, GNUNET_CRYPTO_hmac(), and key.

Referenced by decaps_dv_box_cb(), and encapsulate_for_dv().

Here is the call graph for this function:

Here is the caller graph for this function:

dv_encrypt()

static void dv_encrypt ( struct DVKeyState *  key, const void *  in, void *  dst, size_t  in_size  )

static

Perform backchannel encryption using symmetric secret in key to encrypt data from in to dst.

Parameters

[in,out]	key	key material to use
	dst	where to write the result
	in	input data to encrypt (plaintext)
	in_size	number of bytes of input in in and available at dst

Definition at line 5160 of file gnunet-service-transport.c.

{
  GNUNET_assert (0 ==
                 gcry_cipher_encrypt (key->cipher, dst, in_size, in, in_size));
}

References GNUNET_assert, and key.

Referenced by encapsulate_for_dv().

Here is the caller graph for this function:

dv_decrypt()

static enum GNUNET_GenericReturnValue dv_decrypt ( struct DVKeyState *  key, void *  out, const void *  ciph, size_t  out_size  )

static

Perform backchannel encryption using symmetric secret in key to encrypt data from in to dst.

Parameters

[in,out]	key	key material to use
	ciph	cipher text to decrypt
[out]	out	output data to generate (plaintext)
	out_size	number of bytes of input in ciph and available in out

Returns

GNUNET_OK on success

Definition at line 5178 of file gnunet-service-transport.c.

{
  return (0 ==
          gcry_cipher_decrypt (key->cipher,
                               out, out_size,
                               ciph, out_size)) ? GNUNET_OK : GNUNET_SYSERR;
}

References GNUNET_OK, GNUNET_SYSERR, and key.

Referenced by decaps_dv_box_cb().

Here is the caller graph for this function:

dv_key_clean()

static void dv_key_clean ( struct DVKeyState *  key )

static

Clean up key material in key.

Parameters

key	key material to clean up (memory must not be free'd!)

Definition at line 5196 of file gnunet-service-transport.c.

{
  gcry_cipher_close (key->cipher);
  GNUNET_CRYPTO_zero_keys (&key->material, sizeof(key->material));
}

References GNUNET_CRYPTO_zero_keys(), and key.

Referenced by decaps_dv_box_cb(), and encapsulate_for_dv().

Here is the call graph for this function:

Here is the caller graph for this function:

encapsulate_for_dv()

static struct GNUNET_TIME_Relative encapsulate_for_dv ( struct DistanceVector *  dv, unsigned int  num_dvhs, struct DistanceVectorHop **  dvhs, const struct GNUNET_MessageHeader *  hdr, DVMessageHandler  use, void *  use_cls, enum RouteMessageOptions  options, enum GNUNET_GenericReturnValue  without_fc  )

static

Pick a path of dv under constraints options and schedule transmission of hdr.

Parameters

target	neighbour to ultimately send to
num_dvhs	length of the dvhs array
dvhs	array of hops to send the message to
hdr	message to send as payload
use	function to call with the encapsulated message
use_cls	closure for use
options	whether path must be confirmed or not, to be passed to use
without_fc	shall this TransportDVBoxMessage be forwarded without flow control.

Returns

expected RTT for transmission, GNUNET_TIME_UNIT_FOREVER_REL if sending failed

Definition at line 5233 of file gnunet-service-transport.c.

{
  struct TransportDVBoxMessage box_hdr;
  struct TransportDVBoxPayloadP payload_hdr;
  uint16_t enc_body_size = ntohs (hdr->size);
  char enc[sizeof(struct TransportDVBoxPayloadP) + enc_body_size] GNUNET_ALIGN;
  struct DVKeyState *key;
  struct GNUNET_TIME_Relative rtt;
  struct GNUNET_ShortHashCode km;
 
  key = GNUNET_new (struct DVKeyState);
  /* Encrypt payload */
  box_hdr.header.type = htons (GNUNET_MESSAGE_TYPE_TRANSPORT_DV_BOX);
  box_hdr.total_hops = htons (0);
  box_hdr.without_fc = htons (without_fc);
  // update_ephemeral (dv);
  if (0 ==
      GNUNET_TIME_absolute_get_remaining (dv->ephemeral_validity).rel_value_us)
  {
    GNUNET_CRYPTO_eddsa_kem_encaps (&dv->target.public_key,
                                    &dv->ephemeral_key,
                                    &km);
    dv->km = GNUNET_new (struct GNUNET_ShortHashCode);
    GNUNET_memcpy (dv->km, &km, sizeof(struct GNUNET_ShortHashCode));
    sign_ephemeral (dv);
  }
  box_hdr.ephemeral_key = dv->ephemeral_key;
  payload_hdr.sender_sig = dv->sender_sig;
 
  GNUNET_CRYPTO_random_block (GNUNET_CRYPTO_QUALITY_NONCE,
                              &box_hdr.iv,
                              sizeof(box_hdr.iv));
  // We are creating this key, so this must work.
  // FIXME: Possibly also add return values here. We are processing
  // Input from other peers...
  dv_setup_key_state_from_km (dv->km, &box_hdr.iv, key);
  payload_hdr.sender = *GST_my_identity;
  payload_hdr.monotonic_time = GNUNET_TIME_absolute_hton (dv->monotime);
  dv_encrypt (key, &payload_hdr, enc, sizeof(payload_hdr));
  dv_encrypt (key,
              hdr,
              &enc[sizeof(struct TransportDVBoxPayloadP)],
              enc_body_size);
  dv_hmac (key, &box_hdr.hmac, enc, sizeof(enc));
  dv_key_clean (key);
  rtt = GNUNET_TIME_UNIT_FOREVER_REL;
  /* For each selected path, take the pre-computed header and body
     and add the path in the middle of the message; then send it. */
  for (unsigned int i = 0; i < num_dvhs; i++)
  {
    struct DistanceVectorHop *dvh = dvhs[i];
    unsigned int num_hops = dvh->distance + 1;
    char buf[sizeof(struct TransportDVBoxMessage)
             + sizeof(struct GNUNET_PeerIdentity) * num_hops
             + sizeof(struct TransportDVBoxPayloadP)
             + enc_body_size] GNUNET_ALIGN;
    struct GNUNET_PeerIdentity *dhops;
 
    box_hdr.header.size = htons (sizeof(buf));
    box_hdr.orig_size = htons (sizeof(buf));
    box_hdr.num_hops = htons (num_hops);
    memcpy (buf, &box_hdr, sizeof(box_hdr));
    dhops = (struct GNUNET_PeerIdentity *) &buf[sizeof(box_hdr)];
    memcpy (dhops,
            dvh->path,
            dvh->distance * sizeof(struct GNUNET_PeerIdentity));
    dhops[dvh->distance] = dv->target;
    if (GNUNET_EXTRA_LOGGING > 0)
    {
      char *path;
 
      path = GNUNET_strdup (GNUNET_i2s (GST_my_identity));
      for (unsigned int j = 0; j < num_hops; j++)
      {
        char *tmp;
 
        GNUNET_asprintf (&tmp, "%s-%s", path, GNUNET_i2s (&dhops[j]));
        GNUNET_free (path);
        path = tmp;
      }
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "Routing message of type %u to %s using DV (#%u/%u) via %s\n",
                  ntohs (hdr->type),
                  GNUNET_i2s (&dv->target),
                  i + 1,
                  num_dvhs,
                  path);
      GNUNET_free (path);
    }
    rtt = GNUNET_TIME_relative_min (rtt, dvh->pd.aged_rtt);
    memcpy (&dhops[num_hops], enc, sizeof(enc));
    use (use_cls,
         dvh->next_hop,
         (const struct GNUNET_MessageHeader *) buf,
         options);
    GNUNET_free (key);
  }
  return rtt;
}

References PerformanceData::aged_rtt, DistanceVectorHop::distance, dv_encrypt(), dv_hmac(), dv_key_clean(), dv_setup_key_state_from_km(), enc, TransportDVBoxMessage::ephemeral_key, GNUNET_ALIGN, GNUNET_asprintf(), GNUNET_CRYPTO_eddsa_kem_encaps(), GNUNET_CRYPTO_QUALITY_NONCE, GNUNET_CRYPTO_random_block(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_EXTRA_LOGGING, GNUNET_free, GNUNET_i2s(), GNUNET_log, GNUNET_memcpy, GNUNET_MESSAGE_TYPE_TRANSPORT_DV_BOX, GNUNET_new, GNUNET_strdup, GNUNET_TIME_absolute_get_remaining(), GNUNET_TIME_absolute_hton(), GNUNET_TIME_relative_min(), GNUNET_TIME_UNIT_FOREVER_REL, GST_my_identity, TransportDVBoxMessage::header, TransportDVBoxMessage::hmac, TransportDVBoxMessage::iv, key, TransportDVBoxPayloadP::monotonic_time, DistanceVectorHop::next_hop, TransportDVBoxMessage::num_hops, options, TransportDVBoxMessage::orig_size, DistanceVectorHop::path, DistanceVectorHop::pd, GNUNET_TIME_Relative::rel_value_us, TransportDVBoxPayloadP::sender, TransportDVBoxPayloadP::sender_sig, sign_ephemeral(), GNUNET_MessageHeader::size, TransportDVBoxMessage::total_hops, GNUNET_MessageHeader::type, and TransportDVBoxMessage::without_fc.

Referenced by route_control_message_without_fc(), and transmit_on_queue().

Here is the call graph for this function:

Here is the caller graph for this function:

send_dv_to_neighbour()

static void send_dv_to_neighbour ( void *  cls, struct Neighbour *  next_hop, const struct GNUNET_MessageHeader *  hdr, enum RouteMessageOptions  options  )

static

Wrapper around route_via_neighbour() that matches the DVMessageHandler structure.

Parameters

cls	unused
next_hop	where to send next
hdr	header of the message to send
options	message options for queue selection

Definition at line 5351 of file gnunet-service-transport.c.

{
  (void) cls;
  (void) route_via_neighbour (next_hop, hdr, RMO_UNCONFIRMED_ALLOWED);
}

References RMO_UNCONFIRMED_ALLOWED, and route_via_neighbour().

Referenced by route_control_message_without_fc().

Here is the call graph for this function:

Here is the caller graph for this function:

route_control_message_without_fc()

static struct GNUNET_TIME_Relative route_control_message_without_fc ( struct VirtualLink *  vl, const struct GNUNET_MessageHeader *  hdr, enum RouteMessageOptions  options  )

static

We need to transmit hdr to target.

If necessary, this may involve DV routing. This function routes without applying flow control or congestion control and should only be used for control traffic.

Parameters

target	peer to receive hdr
hdr	header of the message to route and GNUNET_free()
options	which transmission channels are allowed

Returns

expected RTT for transmission, GNUNET_TIME_UNIT_FOREVER_REL if sending failed

Definition at line 5373 of file gnunet-service-transport.c.

{
  // struct VirtualLink *vl;
  struct Neighbour *n;
  struct DistanceVector *dv;
  struct GNUNET_TIME_Relative rtt1;
  struct GNUNET_TIME_Relative rtt2;
  const struct GNUNET_PeerIdentity *target = &vl->target;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Trying to route message of type %u to %s without fc\n",
              ntohs (hdr->type),
              GNUNET_i2s (target));
 
  // TODO Do this elsewhere. vl should be given as parameter to method.
  // vl = lookup_virtual_link (target);
  GNUNET_assert (NULL != vl && GNUNET_YES == vl->confirmed);
  if (NULL == vl)
    return GNUNET_TIME_UNIT_FOREVER_REL;
  n = vl->n;
  dv = (0 != (options & RMO_DV_ALLOWED)) ? vl->dv : NULL;
  if (0 == (options & RMO_UNCONFIRMED_ALLOWED))
  {
    /* if confirmed is required, and we do not have anything
       confirmed, drop respective options */
    if (NULL == n)
      n = lookup_neighbour (target);
    if ((NULL == dv) && (0 != (options & RMO_DV_ALLOWED)))
      dv = GNUNET_CONTAINER_multipeermap_get (dv_routes, target);
  }
  if ((NULL == n) && (NULL == dv))
  {
    GNUNET_log (GNUNET_ERROR_TYPE_INFO,
                "Cannot route message of type %u to %s: no route\n",
                ntohs (hdr->type),
                GNUNET_i2s (target));
    GNUNET_STATISTICS_update (GST_stats,
                              "# Messages dropped in routing: no acceptable method",
                              1,
                              GNUNET_NO);
    return GNUNET_TIME_UNIT_FOREVER_REL;
  }
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Routing message of type %u to %s with options %X\n",
              ntohs (hdr->type),
              GNUNET_i2s (target),
              (unsigned int) options);
  /* If both dv and n are possible and we must choose:
     flip a coin for the choice between the two; for now 50/50 */
  if ((NULL != n) && (NULL != dv) && (0 == (options & RMO_REDUNDANT)))
  {
    if (0 == GNUNET_CRYPTO_random_u32 (GNUNET_CRYPTO_QUALITY_WEAK, 2))
      n = NULL;
    else
      dv = NULL;
  }
  if ((NULL != n) && (NULL != dv))
    options &= ~RMO_REDUNDANT; /* We will do one DV and one direct, that's
                                  enough for redundancy, so clear the flag. */
  rtt1 = GNUNET_TIME_UNIT_FOREVER_REL;
  rtt2 = GNUNET_TIME_UNIT_FOREVER_REL;
  if (NULL != n)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Try to route message of type %u to %s without fc via neighbour\n",
                ntohs (hdr->type),
                GNUNET_i2s (target));
    rtt1 = route_via_neighbour (n, hdr, options);
  }
  if (NULL != dv)
  {
    struct DistanceVectorHop *hops[2];
    unsigned int res;
 
    res = pick_random_dv_hops (dv,
                               options,
                               hops,
                               (0 == (options & RMO_REDUNDANT)) ? 1 : 2);
    if (0 == res)
    {
      GNUNET_log (GNUNET_ERROR_TYPE_INFO,
                  "Failed to route message, could not determine DV path\n");
      return rtt1;
    }
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "encapsulate_for_dv 1\n");
    rtt2 = encapsulate_for_dv (dv,
                               res,
                               hops,
                               hdr,
                               &send_dv_to_neighbour,
                               NULL,
                               options & (~RMO_REDUNDANT),
                               GNUNET_YES);
  }
  return GNUNET_TIME_relative_min (rtt1, rtt2);
}

References DistanceVectorHop::dv, dv_routes, encapsulate_for_dv(), GNUNET_assert, GNUNET_CONTAINER_multipeermap_get(), GNUNET_CRYPTO_QUALITY_WEAK, GNUNET_CRYPTO_random_u32(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_ERROR_TYPE_INFO, GNUNET_i2s(), GNUNET_log, GNUNET_NO, GNUNET_STATISTICS_update(), GNUNET_TIME_relative_min(), GNUNET_TIME_UNIT_FOREVER_REL, GNUNET_YES, GST_stats, lookup_neighbour(), options, pick_random_dv_hops(), res, RMO_DV_ALLOWED, RMO_REDUNDANT, RMO_UNCONFIRMED_ALLOWED, route_via_neighbour(), send_dv_to_neighbour(), and GNUNET_MessageHeader::type.

Referenced by consider_sending_fc(), handle_communicator_backchannel(), sign_dhp_cp(), sign_t_validation_cb(), and transmit_cummulative_ack_cb().

Here is the call graph for this function:

Here is the caller graph for this function:

consider_sending_fc()

static void consider_sending_fc ( void *  cls )

static

Something changed on the virtual link with respect to flow control.

Consider retransmitting the FC window size.

Parameters

cls	a struct VirtualLink to work with

Definition at line 5538 of file gnunet-service-transport.c.

{
  struct VirtualLink *vl = cls;
  struct GNUNET_TIME_Absolute monotime;
  struct TransportFlowControlMessage *fc;
  struct GNUNET_TIME_Relative duration;
  struct GNUNET_TIME_Relative rtt;
  struct GNUNET_TIME_Relative rtt_average;
  struct Neighbour *n = vl->n;
 
  if (NULL != n && 0 < n->number_of_addresses)
  {
    size_t addresses_size =
      n->number_of_addresses * sizeof (struct TransportGlobalNattedAddress) + n
      ->size_of_global_addresses;
    char *tgnas = GNUNET_malloc (addresses_size);
    struct AddGlobalAddressesContext ctx;
    ctx.off = 0;
    ctx.tgnas = tgnas;
 
    fc = GNUNET_malloc (sizeof (struct TransportFlowControlMessage)
                        + addresses_size);
    fc->header.size = htons (sizeof(struct TransportFlowControlMessage)
                             + addresses_size);
    fc->size_of_addresses = htonl (n->size_of_global_addresses);
    fc->number_of_addresses = htonl (n->number_of_addresses);
    GNUNET_CONTAINER_multipeermap_iterate (n->natted_addresses,
                                           &add_global_addresses,
                                           &ctx);
    GNUNET_memcpy (&fc[1], tgnas, addresses_size);
    GNUNET_free (tgnas);
  }
  else
  {
    fc = GNUNET_malloc (sizeof (struct TransportFlowControlMessage));
    fc->header.size = htons (sizeof(struct TransportFlowControlMessage));
  }
 
  duration = GNUNET_TIME_absolute_get_duration (vl->last_fc_transmission);
  /* OPTIMIZE-FC-BDP: decide sane criteria on when to do this, instead of doing
     it always! */
  /* For example, we should probably ONLY do this if a bit more than
     an RTT has passed, or if the window changed "significantly" since
     then. See vl->last_fc_rtt! NOTE: to do this properly, we also
     need an estimate for the bandwidth-delay-product for the entire
     VL, as that determines "significantly". We have the delay, but
     the bandwidth statistics need to be added for the VL!*/(void) duration;
 
  if (NULL != vl->dv)
    rtt_average = calculate_rtt (vl->dv);
  else
    rtt_average = GNUNET_TIME_UNIT_FOREVER_REL;
  fc->rtt = GNUNET_TIME_relative_hton (rtt_average);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Sending FC seq %u to %s with new window %llu %lu %u\n",
              (unsigned int) vl->fc_seq_gen,
              GNUNET_i2s (&vl->target),
              (unsigned long long) vl->incoming_fc_window_size,
              (unsigned long) rtt_average.rel_value_us,
              vl->sync_ready);
  monotime = GNUNET_TIME_absolute_get_monotonic (GST_cfg);
  vl->last_fc_transmission = monotime;
  fc->sync_ready = vl->sync_ready;
  fc->header.type = htons (GNUNET_MESSAGE_TYPE_TRANSPORT_FLOW_CONTROL);
  fc->seq = htonl (vl->fc_seq_gen++);
  fc->inbound_window_size = GNUNET_htonll (vl->incoming_fc_window_size
                                           + vl->incoming_fc_window_size_used
                                           + vl->incoming_fc_window_size_loss);
  fc->outbound_sent = GNUNET_htonll (vl->outbound_fc_window_size_used);
  fc->outbound_window_size = GNUNET_htonll (vl->outbound_fc_window_size);
  fc->sender_time = GNUNET_TIME_absolute_hton (monotime);
  rtt = route_control_message_without_fc (vl, &fc->header, RMO_DV_ALLOWED);
  if (GNUNET_TIME_UNIT_FOREVER_REL.rel_value_us == rtt.rel_value_us)
  {
    rtt = GNUNET_TIME_UNIT_SECONDS;
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "FC retransmission to %s failed, will retry in %s\n",
                GNUNET_i2s (&vl->target),
                GNUNET_STRINGS_relative_time_to_string (rtt, GNUNET_YES));
    vl->last_fc_rtt = GNUNET_TIME_UNIT_ZERO;
  }
  else
  {
    /* OPTIMIZE-FC-BDP: rtt is not ideal, we can do better! */
    vl->last_fc_rtt = rtt;
  }
  if (NULL != vl->fc_retransmit_task)
    GNUNET_SCHEDULER_cancel (vl->fc_retransmit_task);
  if (MAX_FC_RETRANSMIT_COUNT == vl->fc_retransmit_count)
  {
    rtt = GNUNET_TIME_UNIT_MINUTES;
    vl->fc_retransmit_count = 0;
  }
  vl->fc_retransmit_task =
    GNUNET_SCHEDULER_add_delayed (rtt, &task_consider_sending_fc, vl);
  vl->fc_retransmit_count++;
  GNUNET_free (fc);
}

References add_global_addresses(), calculate_rtt(), ctx, duration, VirtualLink::dv, VirtualLink::fc_retransmit_count, VirtualLink::fc_retransmit_task, VirtualLink::fc_seq_gen, GNUNET_CONTAINER_multipeermap_iterate(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_free, GNUNET_htonll(), GNUNET_i2s(), GNUNET_log, GNUNET_malloc, GNUNET_memcpy, GNUNET_MESSAGE_TYPE_TRANSPORT_FLOW_CONTROL, GNUNET_SCHEDULER_add_delayed(), GNUNET_SCHEDULER_cancel(), GNUNET_STRINGS_relative_time_to_string(), GNUNET_TIME_absolute_get_duration(), GNUNET_TIME_absolute_get_monotonic(), GNUNET_TIME_absolute_hton(), GNUNET_TIME_relative_hton(), GNUNET_TIME_UNIT_FOREVER_REL, GNUNET_TIME_UNIT_MINUTES, GNUNET_TIME_UNIT_SECONDS, GNUNET_TIME_UNIT_ZERO, GNUNET_YES, GST_cfg, VirtualLink::incoming_fc_window_size, VirtualLink::incoming_fc_window_size_loss, VirtualLink::incoming_fc_window_size_used, VirtualLink::last_fc_rtt, VirtualLink::last_fc_transmission, MAX_FC_RETRANSMIT_COUNT, VirtualLink::n, Neighbour::number_of_addresses, AddGlobalAddressesContext::off, VirtualLink::outbound_fc_window_size, VirtualLink::outbound_fc_window_size_used, GNUNET_TIME_Relative::rel_value_us, RMO_DV_ALLOWED, route_control_message_without_fc(), VirtualLink::sync_ready, VirtualLink::target, task_consider_sending_fc(), AddGlobalAddressesContext::tgnas, and Neighbour::vl.

Referenced by activate_core_visible_dv_path(), check_vl_transmission(), core_env_sent_cb(), handle_flow_control(), handle_validation_response(), and task_consider_sending_fc().

Here is the call graph for this function:

Here is the caller graph for this function:

task_consider_sending_fc()

static void task_consider_sending_fc ( void *  cls )

static

Something changed on the virtual link with respect to flow control.

Consider retransmitting the FC window size.

Parameters

cls	a struct VirtualLink to work with

Definition at line 5485 of file gnunet-service-transport.c.

{
  struct VirtualLink *vl = cls;
  vl->fc_retransmit_task = NULL;
  consider_sending_fc (cls);
}

References consider_sending_fc(), and VirtualLink::fc_retransmit_task.

Referenced by consider_sending_fc().

Here is the call graph for this function:

Here is the caller graph for this function:

get_address_without_port()

static char * get_address_without_port ( const char *  address )

static

Get the IP address without the port number.

Parameters

address

The string contains a communicator prefix, IP address and port like this 'tcp-92.68.150.1:55452'.

Returns

String with IP address only.

Definition at line 12368 of file gnunet-service-transport.c.

{
  const char *colon;
  char *colon_rest;
  size_t colon_rest_length;
  char *address_without_port;
 
  colon = strchr (address,':');
  colon_rest = GNUNET_strndup (address, colon - address);
  colon_rest_length = strlen (colon_rest);
  address_without_port = GNUNET_strndup (&colon_rest[4], colon_rest_length - 4);
  GNUNET_free (colon_rest);
 
  return address_without_port;
}

References address, GNUNET_free, and GNUNET_strndup.

Referenced by check_for_global_natted(), check_validation_request_pending(), create_address_entry(), handle_add_queue_message(), iterate_address_and_compare_cb(), and iterate_address_start_burst().

Here is the caller graph for this function:

add_global_addresses()

static enum GNUNET_GenericReturnValue add_global_addresses ( void *  cls, const struct GNUNET_PeerIdentity *  pid, void *  value  )

static

Definition at line 5505 of file gnunet-service-transport.c.

{
  struct AddGlobalAddressesContext *ctx = cls;
  struct TransportGlobalNattedAddress *tgna = value;
  char *addr = (char *) &tgna[1];
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "sending address %s length %u\n",
              addr,
              ntohl (tgna->address_length));
  GNUNET_memcpy (&(ctx->tgnas[ctx->off]), tgna, sizeof (struct
                                                        TransportGlobalNattedAddress)
                 + ntohl (tgna->address_length));
  ctx->off += sizeof(struct TransportGlobalNattedAddress) + ntohl (tgna->
                                                                   address_length);
 
  return GNUNET_OK;
}

References TransportGlobalNattedAddress::address_length, ctx, GNUNET_ERROR_TYPE_DEBUG, GNUNET_log, GNUNET_memcpy, GNUNET_OK, and value.

Referenced by consider_sending_fc().

Here is the caller graph for this function:

calculate_rtt()

static struct GNUNET_TIME_Relative calculate_rtt ( struct DistanceVector *  dv )

static

Definition at line 10195 of file gnunet-service-transport.c.

{
  struct GNUNET_TIME_Relative ret = GNUNET_TIME_UNIT_ZERO;
  unsigned int n_hops = 0;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "calculate_rtt\n");
  for (struct DistanceVectorHop *pos = dv->dv_head; NULL != pos;
       pos = pos->next_dv)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "calculate_rtt %lu\n",
                (unsigned long) pos->pd.aged_rtt.rel_value_us);
    n_hops++;
    ret = GNUNET_TIME_relative_add (GNUNET_TIME_relative_multiply (pos->pd.
                                                                   aged_rtt, pos
                                                                   ->distance
                                                                   + 2), ret);
  }
 
  GNUNET_assert (0 != n_hops);
 
  return ret;
}

References GNUNET_assert, GNUNET_ERROR_TYPE_DEBUG, GNUNET_log, GNUNET_TIME_relative_add(), GNUNET_TIME_relative_multiply(), GNUNET_TIME_UNIT_ZERO, and ret.

Referenced by consider_sending_fc(), and handle_flow_control().

Here is the call graph for this function:

Here is the caller graph for this function:

check_vl_transmission()

static void check_vl_transmission ( struct VirtualLink *  vl )

static

There is a message at the head of the pending messages for vl which may be ready for transmission.

Check if a queue is ready to take it.

This function must (1) check for flow control to ensure that we can right now send to vl, (2) check that the pending message in the queue is actually eligible, (3) determine if any applicable queue (direct neighbour or DVH path) is ready to accept messages, and (4) prioritize based on the preferences associated with the pending message.

So yeah, easy.

Parameters

vl	virtual link where we should check for transmission

Definition at line 5655 of file gnunet-service-transport.c.

{
  struct Neighbour *n = vl->n;
  struct DistanceVector *dv = vl->dv;
  struct GNUNET_TIME_Absolute now;
  struct VirtualLink *vl_next_hop;
  int elig;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "check_vl_transmission to target %s\n",
              GNUNET_i2s (&vl->target));
  /* Check that we have an eligible pending message!
     (cheaper than having #transmit_on_queue() find out!) */
  elig = GNUNET_NO;
  for (struct PendingMessage *pm = vl->pending_msg_head; NULL != pm;
       pm = pm->next_vl)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "check_vl_transmission loop\n");
    if (NULL != pm->qe)
      continue;   /* not eligible, is in a queue! */
    if (pm->bytes_msg + vl->outbound_fc_window_size_used >
        vl->outbound_fc_window_size)
    {
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "Stalled message %" PRIu64
                  " transmission on VL %s due to flow control: %llu < %llu\n",
                  pm->logging_uuid,
                  GNUNET_i2s (&vl->target),
                  (unsigned long long) vl->outbound_fc_window_size,
                  (unsigned long long) (pm->bytes_msg
                                        + vl->outbound_fc_window_size_used));
      consider_sending_fc (vl);
      return;     /* We have a message, but flow control says "nope" */
    }
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Target window on VL %s not stalled. Scheduling transmission on queue\n",
                GNUNET_i2s (&vl->target));
    /* Notify queues at direct neighbours that we are interested */
    now = GNUNET_TIME_absolute_get ();
    if (NULL != n)
    {
      for (struct Queue *queue = n->queue_head; NULL != queue;
           queue = queue->next_neighbour)
      {
        if ((GNUNET_YES == queue->idle) &&
            (queue->validated_until.abs_value_us > now.abs_value_us))
        {
          GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                      "Direct neighbour %s not stalled\n",
                      GNUNET_i2s (&n->pid));
          schedule_transmit_on_queue (GNUNET_TIME_UNIT_ZERO,
                                      queue,
                                      GNUNET_SCHEDULER_PRIORITY_DEFAULT);
          elig = GNUNET_YES;
        }
        else
          GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                      "Neighbour Queue QID: %u (%u) busy or invalid\n",
                      queue->qid,
                      queue->idle);
      }
    }
    /* Notify queues via DV that we are interested */
    if (NULL != dv)
    {
      /* Do DV with lower scheduler priority, which effectively means that
         IF a neighbour exists and is available, we prefer it. */
      for (struct DistanceVectorHop *pos = dv->dv_head; NULL != pos;
           pos = pos->next_dv)
      {
        struct Neighbour *nh_iter = pos->next_hop;
 
 
        if (pos->path_valid_until.abs_value_us <= now.abs_value_us)
          continue;   /* skip this one: path not validated */
        else
        {
          vl_next_hop = lookup_virtual_link (&nh_iter->pid);
          GNUNET_assert (NULL != vl_next_hop);
          if (pm->bytes_msg + vl_next_hop->outbound_fc_window_size_used >
              vl_next_hop->outbound_fc_window_size)
          {
            GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                        "Stalled message %" PRIu64
                        " transmission on next hop %s due to flow control: %llu < %llu\n",
                        pm->logging_uuid,
                        GNUNET_i2s (&vl_next_hop->target),
                        (unsigned long
                         long) vl_next_hop->outbound_fc_window_size,
                        (unsigned long long) (pm->bytes_msg
                                              + vl_next_hop->
                                              outbound_fc_window_size_used));
            consider_sending_fc (vl_next_hop);
            continue; /* We have a message, but flow control says "nope" for the first hop of this path */
          }
          for (struct Queue *queue = nh_iter->queue_head; NULL != queue;
               queue = queue->next_neighbour)
            if ((GNUNET_YES == queue->idle) &&
                (queue->validated_until.abs_value_us > now.abs_value_us))
            {
              GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                          "Next hop neighbour %s not stalled\n",
                          GNUNET_i2s (&nh_iter->pid));
              schedule_transmit_on_queue (GNUNET_TIME_UNIT_ZERO,
                                          queue,
                                          GNUNET_SCHEDULER_PRIORITY_BACKGROUND);
              elig = GNUNET_YES;
            }
            else
              GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                          "DV Queue QID: %u (%u) busy or invalid\n",
                          queue->qid,
                          queue->idle);
        }
      }
    }
    if (GNUNET_YES == elig)
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "Eligible message %" PRIu64 " of size %u to %s: %llu/%llu\n",
                  pm->logging_uuid,
                  pm->bytes_msg,
                  GNUNET_i2s (&vl->target),
                  (unsigned long long) vl->outbound_fc_window_size,
                  (unsigned long long) (pm->bytes_msg
                                        + vl->outbound_fc_window_size_used));
    break;
  }
}

References GNUNET_TIME_Absolute::abs_value_us, consider_sending_fc(), VirtualLink::dv, DistanceVector::dv_head, GNUNET_assert, GNUNET_ERROR_TYPE_DEBUG, GNUNET_i2s(), GNUNET_log, GNUNET_NO, GNUNET_SCHEDULER_PRIORITY_BACKGROUND, GNUNET_SCHEDULER_PRIORITY_DEFAULT, GNUNET_TIME_absolute_get(), GNUNET_TIME_UNIT_ZERO, GNUNET_YES, lookup_virtual_link(), VirtualLink::n, VirtualLink::outbound_fc_window_size, VirtualLink::outbound_fc_window_size_used, VirtualLink::pending_msg_head, Neighbour::pid, pm, queue(), Neighbour::queue_head, schedule_transmit_on_queue(), VirtualLink::target, DistanceVector::vl, and Neighbour::vl.

Referenced by forward_dv_box(), free_queue_entry(), handle_client_send(), handle_flow_control(), and send_msg_from_cache().

Here is the call graph for this function:

Here is the caller graph for this function:

handle_client_send()

static void handle_client_send ( void *  cls, const struct OutboundMessage *  obm  )

static

Client asked for transmission to a peer.

Process the request.

Parameters

cls	the client
obm	the send message that was sent

Definition at line 5793 of file gnunet-service-transport.c.

{
  struct TransportClient *tc = cls;
  struct PendingMessage *pm;
  const struct GNUNET_MessageHeader *obmm;
  uint32_t bytes_msg;
  struct VirtualLink *vl;
  enum GNUNET_MQ_PriorityPreferences pp;
 
  GNUNET_assert (CT_CORE == tc->type);
  obmm = (const struct GNUNET_MessageHeader *) &obm[1];
  bytes_msg = ntohs (obmm->size);
  pp = ntohl (obm->priority);
  vl = lookup_virtual_link (&obm->peer);
  if ((NULL == vl) || (GNUNET_NO == vl->confirmed))
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Don't have %s as a neighbour (anymore).\n",
                GNUNET_i2s (&obm->peer));
    /* Failure: don't have this peer as a neighbour (anymore).
       Might have gone down asynchronously, so this is NOT
       a protocol violation by CORE. Still count the event,
       as this should be rare. */
    GNUNET_SERVICE_client_continue (tc->client);
    GNUNET_STATISTICS_update (GST_stats,
                              "# messages dropped (neighbour unknown)",
                              1,
                              GNUNET_NO);
    return;
  }
 
  pm = GNUNET_malloc (sizeof(struct PendingMessage) + bytes_msg);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "1 created pm %p storing vl %p\n",
              pm,
              vl);
  pm->logging_uuid = logging_uuid_gen++;
  pm->prefs = pp;
  pm->client = tc;
  pm->vl = vl;
  pm->bytes_msg = bytes_msg;
  memcpy (&pm[1], obmm, bytes_msg);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Sending message of type %u  with %u bytes as <%" PRIu64
              "> to %s\n",
              ntohs (obmm->type),
              bytes_msg,
              pm->logging_uuid,
              GNUNET_i2s (&obm->peer));
  GNUNET_CONTAINER_MDLL_insert (client,
                                tc->details.core.pending_msg_head,
                                tc->details.core.pending_msg_tail,
                                pm);
  GNUNET_CONTAINER_MDLL_insert (vl,
                                vl->pending_msg_head,
                                vl->pending_msg_tail,
                                pm);
  check_vl_transmission (vl);
  GNUNET_SERVICE_client_continue (tc->client);
}

References check_vl_transmission(), VirtualLink::confirmed, CT_CORE, GNUNET_assert, GNUNET_CONTAINER_MDLL_insert, GNUNET_ERROR_TYPE_DEBUG, GNUNET_i2s(), GNUNET_log, GNUNET_malloc, GNUNET_NO, GNUNET_SERVICE_client_continue(), GNUNET_STATISTICS_update(), GST_stats, logging_uuid_gen, lookup_virtual_link(), OutboundMessage::peer, VirtualLink::pending_msg_head, VirtualLink::pending_msg_tail, pm, OutboundMessage::priority, GNUNET_MessageHeader::size, tc, and GNUNET_MessageHeader::type.

Here is the call graph for this function:

handle_communicator_backchannel()

static void handle_communicator_backchannel ( void *  cls, const struct GNUNET_TRANSPORT_CommunicatorBackchannel *  cb  )

static

Communicator requests backchannel transmission.

Process the request. Just repacks it into our struct TransportBackchannelEncapsulationMessage * (which for now has exactly the same format, only a different message type) and passes it on for routing.

Parameters

cls	the client
cb	the send message that was sent

Definition at line 5865 of file gnunet-service-transport.c.

{
  struct Neighbour *n;
  struct VirtualLink *vl;
  struct TransportClient *tc = cls;
  const struct GNUNET_MessageHeader *inbox =
    (const struct GNUNET_MessageHeader *) &cb[1];
  uint16_t isize = ntohs (inbox->size);
  const char *is = ((const char *) &cb[1]) + isize;
  size_t slen = strlen (is) + 1;
  char
    mbuf[slen + isize
         + sizeof(struct
                  TransportBackchannelEncapsulationMessage)] GNUNET_ALIGN;
  struct TransportBackchannelEncapsulationMessage *be =
    (struct TransportBackchannelEncapsulationMessage *) mbuf;
 
  /* 0-termination of 'is' was checked already in
   #check_communicator_backchannel() */
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Preparing backchannel transmission to %s:%s of type %u and size %u\n",
              GNUNET_i2s (&cb->pid),
              is,
              ntohs (inbox->type),
              ntohs (inbox->size));
  /* encapsulate and encrypt message */
  be->header.type =
    htons (GNUNET_MESSAGE_TYPE_TRANSPORT_BACKCHANNEL_ENCAPSULATION);
  be->header.size = htons (sizeof(mbuf));
  memcpy (&be[1], inbox, isize);
  memcpy (&mbuf[sizeof(struct TransportBackchannelEncapsulationMessage)
                + isize],
          is,
          strlen (is) + 1);
  // route_control_message_without_fc (&cb->pid, &be->header, RMO_DV_ALLOWED);
  vl = lookup_virtual_link (&cb->pid);
  if ((NULL != vl) && (GNUNET_YES == vl->confirmed))
  {
    route_control_message_without_fc (vl, &be->header, RMO_DV_ALLOWED);
  }
  else
  {
    /* Use route via neighbour */
    n = lookup_neighbour (&cb->pid);
    if (NULL != n)
      route_via_neighbour (
        n,
        &be->header,
        RMO_NONE);
  }
  GNUNET_SERVICE_client_continue (tc->client);
}

References VirtualLink::confirmed, GNUNET_ALIGN, GNUNET_ERROR_TYPE_DEBUG, GNUNET_i2s(), GNUNET_log, GNUNET_MESSAGE_TYPE_TRANSPORT_BACKCHANNEL_ENCAPSULATION, GNUNET_SERVICE_client_continue(), GNUNET_YES, TransportBackchannelEncapsulationMessage::header, is, lookup_neighbour(), lookup_virtual_link(), GNUNET_TRANSPORT_CommunicatorBackchannel::pid, RMO_DV_ALLOWED, RMO_NONE, route_control_message_without_fc(), route_via_neighbour(), GNUNET_MessageHeader::size, tc, and GNUNET_MessageHeader::type.

Here is the call graph for this function:

check_add_address()

static int check_add_address ( void *  cls, const struct GNUNET_TRANSPORT_AddAddressMessage *  aam  )

static

Address of our peer added.

Test message is well-formed.

Parameters

cls	the client
aam	the send message that was sent

Returns

GNUNET_OK if message is well-formed

Definition at line 5929 of file gnunet-service-transport.c.

{
  struct TransportClient *tc = cls;
 
  if (CT_COMMUNICATOR != tc->type)
  {
    GNUNET_break (0);
    return GNUNET_SYSERR;
  }
  GNUNET_MQ_check_zero_termination (aam);
  return GNUNET_OK;
}

References CT_COMMUNICATOR, GNUNET_break, GNUNET_MQ_check_zero_termination, GNUNET_OK, GNUNET_SYSERR, and tc.

store_pi()

static void store_pi ( void *  cls )

static

Ask peerstore to store our address.

Parameters

cls	an struct AddressListEntry *

Definition at line 6196 of file gnunet-service-transport.c.

{
  struct AddressListEntry *ale = cls;
  const char *dash;
  char *address_uri;
  char *prefix = GNUNET_HELLO_address_to_prefix (ale->address);
  unsigned int add_success;
 
  if (NULL == GST_my_identity)
  {
    ale->st = GNUNET_SCHEDULER_add_delayed (GNUNET_TIME_UNIT_MILLISECONDS,
                                            &store_pi,
                                            ale);
    return;
  }
 
  dash = strchr (ale->address, '-');
  GNUNET_assert (NULL != dash);
  dash++;
  GNUNET_asprintf (&address_uri,
                   "%s://%s",
                   prefix,
                   dash);
  GNUNET_free (prefix);
  ale->st = NULL;
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Storing our address `%s' in peerstore until %s!\n",
              ale->address,
              GNUNET_STRINGS_absolute_time_to_string (hello_mono_time));
  add_success = GNUNET_HELLO_builder_add_address (GST_my_hello,
                                                  address_uri);
  if (GNUNET_OK != add_success)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
                "Storing our address `%s' %s\n",
                address_uri,
                GNUNET_NO == add_success ? "not done" : "failed");
    GNUNET_free (address_uri);
    return;
  }
  else
  {
 
    GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
                "Storing our address `%s'\n",
                address_uri);
  }
  // FIXME hello_mono_time used here?? What about expiration in ale?
  pils_sign_address (ale,
                     hello_mono_time);
  // TODO keep track of op and potentially cancle/clean
  GNUNET_free (address_uri);
}

References AddressListEntry::address, GNUNET_asprintf(), GNUNET_assert, GNUNET_ERROR_TYPE_DEBUG, GNUNET_ERROR_TYPE_WARNING, GNUNET_free, GNUNET_HELLO_address_to_prefix(), GNUNET_HELLO_builder_add_address(), GNUNET_log, GNUNET_NO, GNUNET_OK, GNUNET_SCHEDULER_add_delayed(), GNUNET_STRINGS_absolute_time_to_string(), GNUNET_TIME_UNIT_MILLISECONDS, GST_my_hello, GST_my_identity, hello_mono_time, pils_sign_address(), prefix, AddressListEntry::st, and store_pi().

Referenced by create_address_entry(), peerstore_store_own_cb(), shc_cont(), and store_pi().

Here is the call graph for this function:

Here is the caller graph for this function:

shc_cont()

static void shc_cont ( void *  cls, int  success  )

static

Definition at line 5978 of file gnunet-service-transport.c.

{
  struct PilsAddressSignContext *pc = cls;
 
  GNUNET_assert (NULL == pc->req);
  if (GNUNET_OK != success)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
                "Failed to store our address `%s' with peerstore\n",
                pc->ale->address);
    if (NULL == pc->ale->st)
    {
      pc->ale->st = GNUNET_SCHEDULER_add_delayed (GNUNET_TIME_UNIT_SECONDS,
                                                  &store_pi,
                                                  pc->ale);
    }
  }
  GNUNET_free (pc);
}

References GNUNET_assert, GNUNET_ERROR_TYPE_WARNING, GNUNET_free, GNUNET_log, GNUNET_OK, GNUNET_SCHEDULER_add_delayed(), GNUNET_TIME_UNIT_SECONDS, pc, and store_pi().

Referenced by pils_sign_hello_cb().

Here is the call graph for this function:

Here is the caller graph for this function:

pils_sign_hello_cb()

static void pils_sign_hello_cb ( void *  cls, const struct GNUNET_PeerIdentity *  pid, const struct GNUNET_CRYPTO_EddsaSignature *  sig  )

static

Get HELLO signature and create message to store in PEERSTORE.

Definition at line 6003 of file gnunet-service-transport.c.

{
  struct PilsAddressSignContext *pc = cls;
  struct GNUNET_MQ_Envelope *env;
  const struct GNUNET_MessageHeader *msg;
 
  pc->req->op = NULL;
  GNUNET_CONTAINER_DLL_remove (pils_requests_head,
                               pils_requests_tail,
                               pc->req);
  GNUNET_free (pc->req);
  pc->req = NULL;
  env = GNUNET_HELLO_builder_to_env (
    GST_my_hello,
    pid,
    sig,
    pc->et);
  msg = GNUNET_MQ_env_get_msg (env);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "store_pi 1\n");
  pc->ale->shc = GNUNET_PEERSTORE_hello_add (peerstore,
                                             msg,
                                             shc_cont,
                                             pc);
  GNUNET_free (env);
}

References env, GNUNET_CONTAINER_DLL_remove, GNUNET_ERROR_TYPE_DEBUG, GNUNET_free, GNUNET_HELLO_builder_to_env(), GNUNET_log, GNUNET_MQ_env_get_msg(), GNUNET_PEERSTORE_hello_add(), GST_my_hello, msg, pc, peerstore, pid, pils_requests_head, pils_requests_tail, and shc_cont().

Referenced by peerstore_store_own_cb().

Here is the call graph for this function:

Here is the caller graph for this function:

peerstore_store_own_cb()

static void peerstore_store_own_cb ( void *  cls, int  success  )

static

Function called when peerstore is done storing our address.

Parameters

cls	a struct AddressListEntry
success	GNUNET_YES if peerstore was successful

Definition at line 6040 of file gnunet-service-transport.c.

{
  struct PilsAddressSignContext *pc = cls;
 
  pc->ale->sc = NULL;
  if (GNUNET_YES != success)
    GNUNET_log (GNUNET_ERROR_TYPE_ERROR,
                "Failed to store our own address `%s' in peerstore!\n",
                pc->ale->address);
  else
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Successfully stored our own address `%s' in peerstore!\n",
                pc->ale->address);
  /* refresh period is 1/4 of expiration time, that should be plenty
     without being excessive. */
  if (NULL == pc->ale->st)
  {
    pc->ale->st =
      GNUNET_SCHEDULER_add_delayed (
        GNUNET_TIME_relative_divide (pc->ale->expiration,
                                     4ULL),
        &store_pi,
        pc->ale);
  }
 
  /* Now we have to update our HELLO! */
  pc->et = GNUNET_TIME_relative_to_absolute (GNUNET_HELLO_ADDRESS_EXPIRATION);
  pc->req = GNUNET_new (struct PilsRequest);
  GNUNET_CONTAINER_DLL_insert (pils_requests_head,
                               pils_requests_tail,
                               pc->req);
  pc->req->op = GNUNET_PILS_sign_hello (pils,
                                        GST_my_hello,
                                        pc->et,
                                        &pils_sign_hello_cb,
                                        pc);
}

References GNUNET_CONTAINER_DLL_insert, GNUNET_ERROR_TYPE_DEBUG, GNUNET_ERROR_TYPE_ERROR, GNUNET_HELLO_ADDRESS_EXPIRATION, GNUNET_log, GNUNET_new, GNUNET_PILS_sign_hello(), GNUNET_SCHEDULER_add_delayed(), GNUNET_TIME_relative_divide(), GNUNET_TIME_relative_to_absolute(), GNUNET_YES, GST_my_hello, pc, pils, pils_requests_head, pils_requests_tail, pils_sign_hello_cb(), and store_pi().

Referenced by pils_sign_addr_cb().

Here is the call graph for this function:

Here is the caller graph for this function:

pils_sign_addr_cb()

static void pils_sign_addr_cb ( void *  cls, const struct GNUNET_PeerIdentity *  pid, const struct GNUNET_CRYPTO_EddsaSignature *  sig  )

static

Definition at line 6081 of file gnunet-service-transport.c.

{
  struct PilsAddressSignContext *pc = cls;
  char *sig_str;
  void *result;
  size_t result_size;
 
  pc->req->op = NULL;
  GNUNET_CONTAINER_DLL_remove (pils_requests_head,
                               pils_requests_tail,
                               pc->req);
  GNUNET_free (pc->req);
  sig_str = NULL;
  (void) GNUNET_STRINGS_base64_encode (sig, sizeof(*sig), &sig_str);
  result_size =
    1 + GNUNET_asprintf (
      (char **) &result,
      "%s;%llu;%u;%s",
      sig_str,
      (unsigned long long) pc->et.abs_value_us,
      (unsigned int) pc->ale->nt,
      pc->ale->address);
  GNUNET_free (sig_str);
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Build our HELLO URI `%s'\n",
              (char*) result);
 
  pc->ale->signed_address = result;
  pc->ale->signed_address_len = result_size;
  struct GNUNET_TIME_Absolute expiration;
 
  expiration = GNUNET_TIME_relative_to_absolute (pc->ale->expiration);
  pc->ale->sc = GNUNET_PEERSTORE_store (peerstore,
                                        "transport",
                                        GST_my_identity, // FIXME
                                        GNUNET_PEERSTORE_TRANSPORT_HELLO_KEY,
                                        result,
                                        result_size,
                                        expiration,
                                        GNUNET_PEERSTORE_STOREOPTION_MULTIPLE,
                                        &peerstore_store_own_cb,
                                        pc);
}

References expiration, GNUNET_asprintf(), GNUNET_CONTAINER_DLL_remove, GNUNET_ERROR_TYPE_DEBUG, GNUNET_free, GNUNET_log, GNUNET_PEERSTORE_store(), GNUNET_PEERSTORE_STOREOPTION_MULTIPLE, GNUNET_PEERSTORE_TRANSPORT_HELLO_KEY, GNUNET_STRINGS_base64_encode(), GNUNET_TIME_relative_to_absolute(), GST_my_identity, pc, peerstore, peerstore_store_own_cb(), pils_requests_head, pils_requests_tail, and result.

Referenced by pils_sign_address().

Here is the call graph for this function:

Here is the caller graph for this function:

pils_sign_address()

void pils_sign_address	(	struct AddressListEntry *	ale,
		struct GNUNET_TIME_Absolute	mono_time
	)

Build address record by signing raw information with private key of the peer identity.

Parameters

handle	handle to the PILS service
address	text address at communicator to sign
nt	network type of address
mono_time	monotonic time at which address was valid
cb	callback called once the signature is ready
cb_cls	closure to the cb callback

Returns

handle to the operation, NULL on error

Definition at line 6165 of file gnunet-service-transport.c.

{
  struct SignedAddress sa;
  struct PilsAddressSignContext *pc;
 
  sa.purpose.purpose = htonl (GNUNET_SIGNATURE_PURPOSE_TRANSPORT_ADDRESS);
  sa.purpose.size = htonl (sizeof(sa));
  sa.mono_time = GNUNET_TIME_absolute_hton (mono_time);
  GNUNET_CRYPTO_hash (ale->address, strlen (ale->address), &sa.addr_hash);
  pc = GNUNET_new (struct PilsAddressSignContext);
  pc->ale = ale;
  pc->et = mono_time;
  pc->req = GNUNET_new (struct PilsRequest);
  pc->req->op = GNUNET_PILS_sign_by_peer_identity (pils,
                                                   &sa.purpose,
                                                   pils_sign_addr_cb,
                                                   pc);
  GNUNET_CONTAINER_DLL_insert (pils_requests_head,
                               pils_requests_tail,
                               pc->req);
}

References AddressListEntry::address, PilsAddressSignContext::ale, GNUNET_CONTAINER_DLL_insert, GNUNET_CRYPTO_hash(), GNUNET_new, GNUNET_PILS_sign_by_peer_identity(), GNUNET_SIGNATURE_PURPOSE_TRANSPORT_ADDRESS, GNUNET_TIME_absolute_hton(), SignedAddress::mono_time, pc, pils, pils_requests_head, pils_requests_tail, pils_sign_addr_cb(), GNUNET_CRYPTO_EccSignaturePurpose::purpose, SignedAddress::purpose, and GNUNET_CRYPTO_EccSignaturePurpose::size.

Referenced by store_pi().

Here is the call graph for this function:

Here is the caller graph for this function:

create_address_entry()

static struct AddressListEntry * create_address_entry ( struct TransportClient *  tc, struct GNUNET_TIME_Relative  expiration, enum GNUNET_NetworkType  nt, const char *  address, uint32_t  aid, size_t  slen  )

static

Definition at line 6252 of file gnunet-service-transport.c.

{
  struct AddressListEntry *ale;
  char *address_without_port;
 
  ale = GNUNET_malloc (sizeof(struct AddressListEntry) + slen);
  ale->tc = tc;
  ale->address = (const char *) &ale[1];
  ale->expiration = expiration;
  ale->aid = aid;
  ale->nt = nt;
  memcpy (&ale[1], address, slen);
  address_without_port = get_address_without_port (ale->address);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Is this %s a local address (%s)\n",
              address_without_port,
              ale->address);
  if (0 != strcmp ("127.0.0.1", address_without_port))
  {
    if (NULL != ale->st)
    {
      GNUNET_SCHEDULER_cancel (ale->st);
    }
    ale->st = GNUNET_SCHEDULER_add_now (&store_pi, ale);
  }
  GNUNET_free (address_without_port);
 
  return ale;
}

References address, AddressListEntry::address, AddressListEntry::aid, expiration, AddressListEntry::expiration, get_address_without_port(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_free, GNUNET_log, GNUNET_malloc, GNUNET_SCHEDULER_add_now(), GNUNET_SCHEDULER_cancel(), nt, AddressListEntry::nt, AddressListEntry::st, store_pi(), tc, and AddressListEntry::tc.

Referenced by handle_add_address().

Here is the call graph for this function:

Here is the caller graph for this function:

feed_addresses_to_pils()

static void feed_addresses_to_pils ( void *  cls )

static

Definition at line 6289 of file gnunet-service-transport.c.

{
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Feeding addresses to PILS\n");
  pils_feed_task = NULL;
 
  GNUNET_PILS_feed_addresses (pils,
                              GST_my_hello);
}

References GNUNET_ERROR_TYPE_DEBUG, GNUNET_log, GNUNET_PILS_feed_addresses(), GST_my_hello, pils, and pils_feed_task.

Referenced by handle_add_address(), and handle_del_address().

Here is the call graph for this function:

Here is the caller graph for this function:

handle_add_address()

static void handle_add_address ( void *  cls, const struct GNUNET_TRANSPORT_AddAddressMessage *  aam  )

static

Address of our peer added.

Process the request.

Parameters

cls	the client
aam	the send message that was sent

Definition at line 6308 of file gnunet-service-transport.c.

{
  struct TransportClient *tc = cls;
  struct AddressListEntry *ale;
  size_t slen;
  char *address;
 
  /* 0-termination of &aam[1] was checked in #check_add_address */
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Communicator added address `%s'!\n",
              (const char *) &aam[1]);
  slen = ntohs (aam->header.size) - sizeof(*aam);
  address = GNUNET_malloc (slen);
  memcpy (address, &aam[1], slen);
  ale = create_address_entry (tc,
                              GNUNET_TIME_relative_ntoh (aam->expiration),
                              ntohl (aam->nt),
                              address,
                              aam->aid,
                              slen);
  GNUNET_CONTAINER_DLL_insert (tc->details.communicator.addr_head,
                               tc->details.communicator.addr_tail,
                               ale);
  {
    for (struct AddressListEntry *iter = tc->details.communicator.addr_head;
         (NULL != iter && NULL != iter->next);
         iter = iter->next)
    {
      char *address_uri;
      char *dash = strchr (ale->address, '-');
      char *prefix = GNUNET_HELLO_address_to_prefix (ale->address);
      GNUNET_assert (NULL != dash);
      dash++;
      GNUNET_asprintf (&address_uri,
                       "%s://%s",
                       prefix,
                       dash);
      GNUNET_free (prefix);
      GNUNET_HELLO_builder_add_address (GST_my_hello, address_uri);
      GNUNET_free (address_uri);
    }
    if (NULL != pils_feed_task)
      GNUNET_SCHEDULER_cancel (pils_feed_task);
    pils_feed_task = GNUNET_SCHEDULER_add_delayed (PILS_FEED_ADDRESSES_DELAY,
                                                   &feed_addresses_to_pils,
                                                   NULL);
  }
  GNUNET_SERVICE_client_continue (tc->client);
  GNUNET_free (address);
}

References address, AddressListEntry::address, GNUNET_TRANSPORT_AddAddressMessage::aid, create_address_entry(), GNUNET_TRANSPORT_AddAddressMessage::expiration, feed_addresses_to_pils(), GNUNET_asprintf(), GNUNET_assert, GNUNET_CONTAINER_DLL_insert, GNUNET_ERROR_TYPE_DEBUG, GNUNET_free, GNUNET_HELLO_address_to_prefix(), GNUNET_HELLO_builder_add_address(), GNUNET_log, GNUNET_malloc, GNUNET_SCHEDULER_add_delayed(), GNUNET_SCHEDULER_cancel(), GNUNET_SERVICE_client_continue(), GNUNET_TIME_relative_ntoh(), GST_my_hello, GNUNET_TRANSPORT_AddAddressMessage::header, GNUNET_TRANSPORT_AddAddressMessage::nt, PILS_FEED_ADDRESSES_DELAY, pils_feed_task, prefix, GNUNET_MessageHeader::size, and tc.

Here is the call graph for this function:

handle_del_address()

static void handle_del_address ( void *  cls, const struct GNUNET_TRANSPORT_DelAddressMessage *  dam  )

static

Address of our peer deleted.

Process the request.

Parameters

cls	the client
dam	the send message that was sent

Definition at line 6368 of file gnunet-service-transport.c.

{
  struct TransportClient *tc = cls;
  struct AddressListEntry *alen;
 
  if (CT_COMMUNICATOR != tc->type)
  {
    GNUNET_break (0);
    GNUNET_SERVICE_client_drop (tc->client);
    return;
  }
  for (struct AddressListEntry *ale = tc->details.communicator.addr_head;
       NULL != ale;
       ale = alen)
  {
    alen = ale->next;
    if (dam->aid != ale->aid)
      continue;
    GNUNET_assert (ale->tc == tc);
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Communicator deleted address `%s'!\n",
                ale->address);
    GNUNET_HELLO_builder_del_address (GST_my_hello,
                                      ale->address);
    if (NULL != pils_feed_task)
      GNUNET_SCHEDULER_cancel (pils_feed_task);
    pils_feed_task = GNUNET_SCHEDULER_add_delayed (PILS_FEED_ADDRESSES_DELAY,
                                                   &feed_addresses_to_pils,
                                                   NULL);
    free_address_list_entry (ale);
    GNUNET_SERVICE_client_continue (tc->client);
    return;
  }
  GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
              "Communicator removed address we did not even have.\n");
  GNUNET_SERVICE_client_continue (tc->client);
  // GNUNET_SERVICE_client_drop (tc->client);
}

References GNUNET_TRANSPORT_DelAddressMessage::aid, CT_COMMUNICATOR, feed_addresses_to_pils(), free_address_list_entry(), GNUNET_assert, GNUNET_break, GNUNET_ERROR_TYPE_DEBUG, GNUNET_ERROR_TYPE_WARNING, GNUNET_HELLO_builder_del_address(), GNUNET_log, GNUNET_SCHEDULER_add_delayed(), GNUNET_SCHEDULER_cancel(), GNUNET_SERVICE_client_continue(), GNUNET_SERVICE_client_drop(), GST_my_hello, AddressListEntry::next, PILS_FEED_ADDRESSES_DELAY, pils_feed_task, and tc.

Here is the call graph for this function:

demultiplex_with_cmc()

static void demultiplex_with_cmc ( struct CommunicatorMessageContext *  cmc )

static

Given an inbound message msg from a communicator cmc, demultiplex it based on the type calling the right handler.

Parameters

cmc	context for demultiplexing
msg	message to demultiplex

Definition at line 10586 of file gnunet-service-transport.c.

{
  struct GNUNET_MQ_MessageHandler handlers[] =
  { GNUNET_MQ_hd_var_size (fragment_box,
                           GNUNET_MESSAGE_TYPE_TRANSPORT_FRAGMENT,
                           struct TransportFragmentBoxMessage,
                           cmc),
    GNUNET_MQ_hd_var_size (reliability_box,
                           GNUNET_MESSAGE_TYPE_TRANSPORT_RELIABILITY_BOX,
                           struct TransportReliabilityBoxMessage,
                           cmc),
    GNUNET_MQ_hd_var_size (reliability_ack,
                           GNUNET_MESSAGE_TYPE_TRANSPORT_RELIABILITY_ACK,
                           struct TransportReliabilityAckMessage,
                           cmc),
    GNUNET_MQ_hd_var_size (backchannel_encapsulation,
                           GNUNET_MESSAGE_TYPE_TRANSPORT_BACKCHANNEL_ENCAPSULATION,
                           struct TransportBackchannelEncapsulationMessage,
                           cmc),
    GNUNET_MQ_hd_var_size (dv_learn,
                           GNUNET_MESSAGE_TYPE_TRANSPORT_DV_LEARN,
                           struct TransportDVLearnMessage,
                           cmc),
    GNUNET_MQ_hd_var_size (dv_box,
                           GNUNET_MESSAGE_TYPE_TRANSPORT_DV_BOX,
                           struct TransportDVBoxMessage,
                           cmc),
    GNUNET_MQ_hd_var_size (flow_control,
                           GNUNET_MESSAGE_TYPE_TRANSPORT_FLOW_CONTROL,
                           struct TransportFlowControlMessage,
                           cmc),
    GNUNET_MQ_hd_fixed_size (
      validation_challenge,
      GNUNET_MESSAGE_TYPE_TRANSPORT_ADDRESS_VALIDATION_CHALLENGE,
      struct TransportValidationChallengeMessage,
      cmc),
    GNUNET_MQ_hd_fixed_size (
      validation_response,
      GNUNET_MESSAGE_TYPE_TRANSPORT_ADDRESS_VALIDATION_RESPONSE,
      struct TransportValidationResponseMessage,
      cmc),
    GNUNET_MQ_handler_end () };
  int ret;
  const struct GNUNET_MessageHeader *msg = cmc->mh;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Handling message of type %u with %u bytes\n",
              (unsigned int) ntohs (msg->type),
              (unsigned int) ntohs (msg->size));
  ret = GNUNET_MQ_handle_message (handlers, msg);
  if (GNUNET_SYSERR == ret)
  {
    GNUNET_break (0);
    GNUNET_SERVICE_client_drop (cmc->tc->client);
    GNUNET_free (cmc);
    return;
  }
  if (GNUNET_NO == ret)
  {
    /* unencapsulated 'raw' message */
    handle_raw_message (cmc, msg);
  }
}

References TransportClient::client, GNUNET_break, GNUNET_ERROR_TYPE_DEBUG, GNUNET_free, GNUNET_log, GNUNET_MESSAGE_TYPE_TRANSPORT_ADDRESS_VALIDATION_CHALLENGE, GNUNET_MESSAGE_TYPE_TRANSPORT_ADDRESS_VALIDATION_RESPONSE, GNUNET_MESSAGE_TYPE_TRANSPORT_BACKCHANNEL_ENCAPSULATION, GNUNET_MESSAGE_TYPE_TRANSPORT_DV_BOX, GNUNET_MESSAGE_TYPE_TRANSPORT_DV_LEARN, GNUNET_MESSAGE_TYPE_TRANSPORT_FLOW_CONTROL, GNUNET_MESSAGE_TYPE_TRANSPORT_FRAGMENT, GNUNET_MESSAGE_TYPE_TRANSPORT_RELIABILITY_ACK, GNUNET_MESSAGE_TYPE_TRANSPORT_RELIABILITY_BOX, GNUNET_MQ_handle_message(), GNUNET_MQ_handler_end, GNUNET_MQ_hd_fixed_size, GNUNET_MQ_hd_var_size, GNUNET_NO, GNUNET_SERVICE_client_drop(), GNUNET_SYSERR, handle_raw_message(), handlers, CommunicatorMessageContext::mh, msg, ret, GNUNET_MessageHeader::size, CommunicatorMessageContext::tc, and GNUNET_MessageHeader::type.

Referenced by backtalker_monotime_cb(), decaps_dv_box_cb(), handle_fragment_box(), handle_incoming_msg(), and handle_reliability_box().

Here is the call graph for this function:

Here is the caller graph for this function:

core_env_sent_cb()

static void core_env_sent_cb ( void *  cls )

static

Function called when we are done giving a message of a certain size to CORE and should thus decrement the number of bytes of RAM reserved for that peer's MQ.

Parameters

cls	a struct CoreSentContext

Definition at line 6428 of file gnunet-service-transport.c.

{
  struct CoreSentContext *ctx = cls;
  struct VirtualLink *vl = ctx->vl;
 
  if (NULL == vl)
  {
    /* lost the link in the meantime, ignore */
    GNUNET_free (ctx);
    return;
  }
  GNUNET_CONTAINER_DLL_remove (vl->csc_head, vl->csc_tail, ctx);
  GNUNET_assert (vl->incoming_fc_window_size_ram >= ctx->size);
  vl->incoming_fc_window_size_ram -= ctx->size;
  vl->incoming_fc_window_size_used += ctx->isize;
  consider_sending_fc (vl);
  GNUNET_free (ctx);
}

References consider_sending_fc(), VirtualLink::csc_head, VirtualLink::csc_tail, ctx, GNUNET_assert, GNUNET_CONTAINER_DLL_remove, GNUNET_free, VirtualLink::incoming_fc_window_size_ram, and VirtualLink::incoming_fc_window_size_used.

Referenced by finish_handling_raw_message().

Here is the call graph for this function:

Here is the caller graph for this function:

finish_handling_raw_message()

static void finish_handling_raw_message ( struct VirtualLink *  vl, const struct GNUNET_MessageHeader *  mh, struct CommunicatorMessageContext *  cmc, unsigned int  free_cmc  )

static

Definition at line 6449 of file gnunet-service-transport.c.

{
  uint16_t size = ntohs (mh->size);
  int have_core;
 
  if (vl->incoming_fc_window_size_ram > UINT_MAX - size)
  {
    GNUNET_STATISTICS_update (GST_stats,
                              "# CORE messages dropped (FC arithmetic overflow)",
                              1,
                              GNUNET_NO);
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "CORE messages of type %u with %u bytes dropped (FC arithmetic overflow)\n",
                (unsigned int) ntohs (mh->type),
                (unsigned int) ntohs (mh->size));
    if (GNUNET_YES == free_cmc)
      finish_cmc_handling_with_continue (cmc, GNUNET_YES);
    return;
  }
  if (vl->incoming_fc_window_size_ram + size > vl->available_fc_window_size)
  {
    GNUNET_STATISTICS_update (GST_stats,
                              "# CORE messages dropped (FC window overflow)",
                              1,
                              GNUNET_NO);
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "CORE messages of type %u with %u bytes dropped (FC window overflow)\n",
                (unsigned int) ntohs (mh->type),
                (unsigned int) ntohs (mh->size));
    if (GNUNET_YES == free_cmc)
      finish_cmc_handling_with_continue (cmc, GNUNET_YES);
    return;
  }
 
  /* Forward to all CORE clients */
  have_core = GNUNET_NO;
  for (struct TransportClient *tc = clients_head; NULL != tc; tc = tc->next)
  {
    struct GNUNET_MQ_Envelope *env;
    struct InboundMessage *im;
    struct CoreSentContext *ctx;
 
    if (CT_CORE != tc->type)
      continue;
    vl->incoming_fc_window_size_ram += size;
    env = GNUNET_MQ_msg_extra (im, size, GNUNET_MESSAGE_TYPE_TRANSPORT_RECV);
    ctx = GNUNET_new (struct CoreSentContext);
    ctx->vl = vl;
    ctx->size = size;
    ctx->isize = (GNUNET_NO == have_core) ? size : 0;
    have_core = GNUNET_YES;
    GNUNET_CONTAINER_DLL_insert (vl->csc_head, vl->csc_tail, ctx);
    GNUNET_MQ_notify_sent (env, &core_env_sent_cb, ctx);
    im->peer = cmc->im.sender;
    memcpy (&im[1], mh, size);
    GNUNET_MQ_send (tc->mq, env);
    vl->core_recv_window--;
  }
  if (GNUNET_NO == have_core)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
                "Dropped message to CORE: no CORE client connected!\n");
    /* Nevertheless, count window as used, as it is from the
       perspective of the other peer! */
    vl->incoming_fc_window_size_used += size;
    /* TODO-M1 */
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Dropped message of type %u with %u bytes to CORE: no CORE client connected!\n",
                (unsigned int) ntohs (mh->type),
                (unsigned int) ntohs (mh->size));
    if (GNUNET_YES == free_cmc)
      finish_cmc_handling_with_continue (cmc, GNUNET_YES);
    return;
  }
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Delivered message from %s of type %u to CORE recv window %d\n",
              GNUNET_i2s (&cmc->im.sender),
              ntohs (mh->type),
              vl->core_recv_window);
  if (vl->core_recv_window > 0)
  {
    if (GNUNET_YES == free_cmc)
      finish_cmc_handling_with_continue (cmc, GNUNET_YES);
    return;
  }
  /* Wait with calling #finish_cmc_handling(cmc) until the message
     was processed by CORE MQs (for CORE flow control)! */
  if (GNUNET_YES == free_cmc)
    GNUNET_CONTAINER_DLL_insert (vl->cmc_head, vl->cmc_tail, cmc);
}

References VirtualLink::available_fc_window_size, clients_head, VirtualLink::cmc_head, VirtualLink::cmc_tail, core_env_sent_cb(), VirtualLink::core_recv_window, VirtualLink::csc_head, VirtualLink::csc_tail, CT_CORE, ctx, env, finish_cmc_handling_with_continue(), GNUNET_CONTAINER_DLL_insert, GNUNET_ERROR_TYPE_DEBUG, GNUNET_ERROR_TYPE_WARNING, GNUNET_i2s(), GNUNET_log, GNUNET_MESSAGE_TYPE_TRANSPORT_RECV, GNUNET_MQ_msg_extra, GNUNET_MQ_notify_sent(), GNUNET_MQ_send(), GNUNET_new, GNUNET_NO, GNUNET_STATISTICS_update(), GNUNET_YES, GST_stats, CommunicatorMessageContext::im, VirtualLink::incoming_fc_window_size_ram, VirtualLink::incoming_fc_window_size_used, mh, InboundMessage::peer, GNUNET_TRANSPORT_IncomingMessage::sender, size, tc, and CoreSentContext::vl.

Referenced by handle_raw_message(), and send_msg_from_cache().

Here is the call graph for this function:

Here is the caller graph for this function:

handle_raw_message()

static void handle_raw_message ( void *  cls, const struct GNUNET_MessageHeader *  mh  )

static

Communicator gave us an unencapsulated message to pass as-is to CORE.

Process the request.

Parameters

cls	a struct CommunicatorMessageContext (must call finish_cmc_handling() when done)
mh	the message that was received

Definition at line 6553 of file gnunet-service-transport.c.

{
  struct CommunicatorMessageContext *cmc = cls;
  // struct CommunicatorMessageContext *cmc_copy =
  // GNUNET_new (struct CommunicatorMessageContext);
  struct GNUNET_MessageHeader *mh_copy;
  struct RingBufferEntry *rbe;
  struct VirtualLink *vl;
  uint16_t size = ntohs (mh->size);
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Handling raw message of type %u with %u bytes\n",
              (unsigned int) ntohs (mh->type),
              (unsigned int) ntohs (mh->size));
 
  if ((size > UINT16_MAX - sizeof(struct InboundMessage)) ||
      (size < sizeof(struct GNUNET_MessageHeader)))
  {
    struct GNUNET_SERVICE_Client *client = cmc->tc->client;
 
    GNUNET_break (0);
    finish_cmc_handling (cmc);
    GNUNET_SERVICE_client_drop (client);
    return;
  }
  vl = lookup_virtual_link (&cmc->im.sender);
  if ((NULL == vl) || (GNUNET_NO == vl->confirmed))
  {
    /* FIXME: sender is giving us messages for CORE but we don't have
       the link up yet! I *suspect* this can happen right now (i.e.
       sender has verified us, but we didn't verify sender), but if
       we pass this on, CORE would be confused (link down, messages
       arrive).  We should investigate more if this happens often,
       or in a persistent manner, and possibly do "something" about
       it. Thus logging as error for now. */
 
    mh_copy = GNUNET_malloc (size);
    rbe = GNUNET_new (struct RingBufferEntry);
    rbe->cmc = cmc;
    /*cmc_copy->tc = cmc->tc;
       cmc_copy->im = cmc->im;*/
    GNUNET_memcpy (mh_copy, mh, size);
 
    rbe->mh = mh_copy;
 
    if (GNUNET_YES == is_ring_buffer_full)
    {
      struct RingBufferEntry *rbe_old = ring_buffer[ring_buffer_head];
      GNUNET_free (rbe_old->cmc);
      GNUNET_free (rbe_old->mh);
      GNUNET_free (rbe_old);
    }
    ring_buffer[ring_buffer_head] = rbe;// cmc_copy;
    // cmc_copy->mh = (const struct GNUNET_MessageHeader *) mh_copy;
    cmc->mh = (const struct GNUNET_MessageHeader *) mh_copy;
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Storing message for %s and type %u (%u) in ring buffer head %u is full %u\n",
                GNUNET_i2s (&cmc->im.sender),
                (unsigned int) ntohs (mh->type),
                (unsigned int) ntohs (mh_copy->type),
                ring_buffer_head,
                is_ring_buffer_full);
    if (RING_BUFFER_SIZE - 1 == ring_buffer_head)
    {
      ring_buffer_head = 0;
      is_ring_buffer_full = GNUNET_YES;
    }
    else
      ring_buffer_head++;
 
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "%u items stored in ring buffer\n",
                GNUNET_YES == is_ring_buffer_full ? RING_BUFFER_SIZE :
                ring_buffer_head);
 
    /*GNUNET_break_op (0);
    GNUNET_STATISTICS_update (GST_stats,
                              "# CORE messages dropped (virtual link still down)",
                              1,
                              GNUNET_NO);
 
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "CORE messages of type %u with %u bytes dropped (virtual link still down)\n",
                (unsigned int) ntohs (mh->type),
                (unsigned int) ntohs (mh->size));
    finish_cmc_handling (cmc);*/
    finish_cmc_handling_with_continue (cmc, GNUNET_NO);
    cmc->continue_send = GNUNET_YES;
    // GNUNET_free (cmc);
    return;
  }
  finish_handling_raw_message (vl, mh, cmc, GNUNET_YES);
}

References TransportClient::client, RingBufferEntry::cmc, VirtualLink::confirmed, CommunicatorMessageContext::continue_send, finish_cmc_handling(), finish_cmc_handling_with_continue(), finish_handling_raw_message(), GNUNET_break, GNUNET_ERROR_TYPE_DEBUG, GNUNET_free, GNUNET_i2s(), GNUNET_log, GNUNET_malloc, GNUNET_memcpy, GNUNET_new, GNUNET_NO, GNUNET_SERVICE_client_drop(), GNUNET_YES, CommunicatorMessageContext::im, is_ring_buffer_full, lookup_virtual_link(), mh, CommunicatorMessageContext::mh, RingBufferEntry::mh, ring_buffer, ring_buffer_head, RING_BUFFER_SIZE, GNUNET_TRANSPORT_IncomingMessage::sender, size, CommunicatorMessageContext::tc, and GNUNET_MessageHeader::type.

Referenced by demultiplex_with_cmc().

Here is the call graph for this function:

Here is the caller graph for this function:

check_fragment_box()

static int check_fragment_box ( void *  cls, const struct TransportFragmentBoxMessage *  fb  )

static

Communicator gave us a fragment box.

Check the message.

Parameters

cls	a struct CommunicatorMessageContext
fb	the send message that was sent

Returns

GNUNET_YES if message is well-formed

Definition at line 6656 of file gnunet-service-transport.c.

{
  uint16_t size = ntohs (fb->header.size);
  uint16_t bsize = size - sizeof(*fb);
 
  (void) cls;
  if (0 == bsize)
  {
    GNUNET_break_op (0);
    return GNUNET_SYSERR;
  }
  if (bsize + ntohs (fb->frag_off) > ntohs (fb->msg_size))
  {
    GNUNET_break_op (0);
    return GNUNET_SYSERR;
  }
  if (ntohs (fb->frag_off) >= ntohs (fb->msg_size))
  {
    GNUNET_break_op (0);
    return GNUNET_SYSERR;
  }
  return GNUNET_YES;
}

References bsize, TransportFragmentBoxMessage::frag_off, GNUNET_break_op, GNUNET_SYSERR, GNUNET_YES, TransportFragmentBoxMessage::header, TransportFragmentBoxMessage::msg_size, GNUNET_MessageHeader::size, and size.

destroy_ack_cummulator()

static void destroy_ack_cummulator ( void *  cls )

static

Clean up an idle cumulative acknowledgement data structure.

Parameters

cls	a struct AcknowledgementCummulator *

Definition at line 6687 of file gnunet-service-transport.c.

{
  struct AcknowledgementCummulator *ac = cls;
 
  ac->task = NULL;
  GNUNET_assert (0 == ac->num_acks);
  GNUNET_assert (
    GNUNET_YES ==
    GNUNET_CONTAINER_multipeermap_remove (ack_cummulators, &ac->target, ac));
  GNUNET_free (ac);
}

References ack_cummulators, GNUNET_assert, GNUNET_CONTAINER_multipeermap_remove(), GNUNET_free, GNUNET_YES, AcknowledgementCummulator::num_acks, AcknowledgementCummulator::target, and AcknowledgementCummulator::task.

Referenced by transmit_cummulative_ack_cb().

Here is the call graph for this function:

Here is the caller graph for this function:

transmit_cummulative_ack_cb()

static void transmit_cummulative_ack_cb ( void *  cls )

static

Do the transmission of a cumulative acknowledgement now.

Parameters

cls	a struct AcknowledgementCummulator *

Definition at line 6706 of file gnunet-service-transport.c.

{
  struct Neighbour *n;
  struct VirtualLink *vl;
  struct AcknowledgementCummulator *ac = cls;
  char buf[sizeof(struct TransportReliabilityAckMessage)
           + ac->num_acks
           * sizeof(struct TransportCummulativeAckPayloadP)] GNUNET_ALIGN;
  struct TransportReliabilityAckMessage *ack =
    (struct TransportReliabilityAckMessage *) buf;
  struct TransportCummulativeAckPayloadP *ap;
 
  ac->task = NULL;
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Sending ACK with %u components to %s\n",
              ac->num_acks,
              GNUNET_i2s (&ac->target));
  GNUNET_assert (0 < ac->num_acks);
  ack->header.type = htons (GNUNET_MESSAGE_TYPE_TRANSPORT_RELIABILITY_ACK);
  ack->header.size =
    htons (sizeof(*ack)
           + ac->num_acks * sizeof(struct TransportCummulativeAckPayloadP));
  ack->ack_counter = htonl (ac->ack_counter += ac->num_acks);
  ap = (struct TransportCummulativeAckPayloadP *) &ack[1];
  for (unsigned int i = 0; i < ac->num_acks; i++)
  {
    ap[i].ack_uuid = ac->ack_uuids[i].ack_uuid;
    ap[i].ack_delay = GNUNET_TIME_relative_hton (
      GNUNET_TIME_absolute_get_duration (ac->ack_uuids[i].receive_time));
  }
  /*route_control_message_without_fc (
    &ac->target,
    &ack->header,
    RMO_DV_ALLOWED);*/
  vl = lookup_virtual_link (&ac->target);
  if ((NULL != vl) && (GNUNET_YES == vl->confirmed))
  {
    route_control_message_without_fc (
      vl,
      &ack->header,
      RMO_DV_ALLOWED);
  }
  else
  {
    /* Use route via neighbour */
    n = lookup_neighbour (&ac->target);
    if (NULL != n)
      route_via_neighbour (
        n,
        &ack->header,
        RMO_NONE);
  }
  ac->num_acks = 0;
  ac->task = GNUNET_SCHEDULER_add_delayed (ACK_CUMMULATOR_TIMEOUT,
                                           &destroy_ack_cummulator,
                                           ac);
}

References TransportReliabilityAckMessage::ack_counter, AcknowledgementCummulator::ack_counter, ACK_CUMMULATOR_TIMEOUT, TransportCummulativeAckPayloadP::ack_delay, TransportCummulativeAckPayloadP::ack_uuid, TransportCummulativeAckPayload::ack_uuid, AcknowledgementCummulator::ack_uuids, VirtualLink::confirmed, destroy_ack_cummulator(), GNUNET_ALIGN, GNUNET_assert, GNUNET_ERROR_TYPE_DEBUG, GNUNET_i2s(), GNUNET_log, GNUNET_MESSAGE_TYPE_TRANSPORT_RELIABILITY_ACK, GNUNET_SCHEDULER_add_delayed(), GNUNET_TIME_absolute_get_duration(), GNUNET_TIME_relative_hton(), GNUNET_YES, TransportReliabilityAckMessage::header, lookup_neighbour(), lookup_virtual_link(), AcknowledgementCummulator::num_acks, TransportCummulativeAckPayload::receive_time, RMO_DV_ALLOWED, RMO_NONE, route_control_message_without_fc(), route_via_neighbour(), GNUNET_MessageHeader::size, AcknowledgementCummulator::target, AcknowledgementCummulator::task, and GNUNET_MessageHeader::type.

Referenced by cummulative_ack().

Here is the call graph for this function:

Here is the caller graph for this function:

cummulative_ack()

static void cummulative_ack ( const struct GNUNET_PeerIdentity *  pid, const struct AcknowledgementUUIDP *  ack_uuid, struct GNUNET_TIME_Absolute  max_delay  )

static

Transmit an acknowledgement for ack_uuid to pid delaying transmission by at most ack_delay.

Parameters

pid	target peer
ack_uuid	UUID to ack
max_delay	how long can the ACK wait

Definition at line 6774 of file gnunet-service-transport.c.

{
  struct AcknowledgementCummulator *ac;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Scheduling ACK %s for transmission to %s\n",
              GNUNET_uuid2s (&ack_uuid->value),
              GNUNET_i2s (pid));
  ac = GNUNET_CONTAINER_multipeermap_get (ack_cummulators, pid);
  if (NULL == ac)
  {
    ac = GNUNET_new (struct AcknowledgementCummulator);
    ac->target = *pid;
    ac->min_transmission_time = max_delay;
    GNUNET_assert (GNUNET_YES ==
                   GNUNET_CONTAINER_multipeermap_put (
                     ack_cummulators,
                     &ac->target,
                     ac,
                     GNUNET_CONTAINER_MULTIHASHMAPOPTION_UNIQUE_ONLY));
  }
  else
  {
    if (MAX_CUMMULATIVE_ACKS == ac->num_acks)
    {
      /* must run immediately, ack buffer full! */
      transmit_cummulative_ack_cb (ac);
    }
    GNUNET_SCHEDULER_cancel (ac->task);
    ac->min_transmission_time =
      GNUNET_TIME_absolute_min (ac->min_transmission_time, max_delay);
  }
  GNUNET_assert (ac->num_acks < MAX_CUMMULATIVE_ACKS);
  ac->ack_uuids[ac->num_acks].receive_time = GNUNET_TIME_absolute_get ();
  ac->ack_uuids[ac->num_acks].ack_uuid = *ack_uuid;
  ac->num_acks++;
  ac->task = GNUNET_SCHEDULER_add_at (ac->min_transmission_time,
                                      &transmit_cummulative_ack_cb,
                                      ac);
}

References ack_cummulators, TransportCummulativeAckPayload::ack_uuid, AcknowledgementCummulator::ack_uuids, GNUNET_assert, GNUNET_CONTAINER_MULTIHASHMAPOPTION_UNIQUE_ONLY, GNUNET_CONTAINER_multipeermap_get(), GNUNET_CONTAINER_multipeermap_put(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_i2s(), GNUNET_log, GNUNET_new, GNUNET_SCHEDULER_add_at(), GNUNET_SCHEDULER_cancel(), GNUNET_TIME_absolute_get(), GNUNET_TIME_absolute_min(), GNUNET_uuid2s(), GNUNET_YES, MAX_CUMMULATIVE_ACKS, AcknowledgementCummulator::min_transmission_time, AcknowledgementCummulator::num_acks, pid, TransportCummulativeAckPayload::receive_time, AcknowledgementCummulator::target, AcknowledgementCummulator::task, transmit_cummulative_ack_cb(), and AcknowledgementUUIDP::value.

Referenced by handle_fragment_box(), and handle_reliability_box().

Here is the call graph for this function:

Here is the caller graph for this function:

find_by_message_uuid()

static int find_by_message_uuid ( void *  cls, uint32_t  key, void *  value  )

static

Iterator called to find a reassembly context by the message UUID in the multihashmap32.

Parameters

cls	a struct FindByMessageUuidContext
key	a key (unused)
value	a struct ReassemblyContext

Returns

GNUNET_YES if not found, GNUNET_NO if found

Definition at line 6845 of file gnunet-service-transport.c.

{
  struct FindByMessageUuidContext *fc = cls;
  struct ReassemblyContext *rc = value;
 
  (void) key;
  if (0 == GNUNET_memcmp (&fc->message_uuid, &rc->msg_uuid))
  {
    fc->rc = rc;
    return GNUNET_NO;
  }
  return GNUNET_YES;
}

References GNUNET_memcmp, GNUNET_NO, GNUNET_YES, key, FindByMessageUuidContext::message_uuid, ReassemblyContext::msg_uuid, FindByMessageUuidContext::rc, and value.

Referenced by handle_fragment_box().

Here is the caller graph for this function:

handle_fragment_box()

static void handle_fragment_box ( void *  cls, const struct TransportFragmentBoxMessage *  fb  )

static

Communicator gave us a fragment.

Process the request.

Parameters

cls	a struct CommunicatorMessageContext (must call finish_cmc_handling() when done)
fb	the message that was received

Definition at line 6868 of file gnunet-service-transport.c.

{
  struct CommunicatorMessageContext *cmc = cls;
  struct VirtualLink *vl;
  struct ReassemblyContext *rc;
  const struct GNUNET_MessageHeader *msg;
  uint16_t msize;
  uint16_t fsize;
  uint16_t frag_off;
  char *target;
  struct GNUNET_TIME_Relative cdelay;
  struct FindByMessageUuidContext fc;
 
  vl = lookup_virtual_link (&cmc->im.sender);
  if ((NULL == vl) || (GNUNET_NO == vl->confirmed))
  {
    struct GNUNET_SERVICE_Client *client = cmc->tc->client;
 
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "No virtual link for %s to handle fragment\n",
                GNUNET_i2s (&cmc->im.sender));
    GNUNET_break (0);
    finish_cmc_handling (cmc);
    GNUNET_SERVICE_client_drop (client);
    return;
  }
  if (NULL == vl->reassembly_map)
  {
    vl->reassembly_map = GNUNET_CONTAINER_multihashmap32_create (8);
    vl->reassembly_heap =
      GNUNET_CONTAINER_heap_create (GNUNET_CONTAINER_HEAP_ORDER_MIN);
    vl->reassembly_timeout_task =
      GNUNET_SCHEDULER_add_delayed (REASSEMBLY_EXPIRATION,
                                    &reassembly_cleanup_task,
                                    vl);
  }
  msize = ntohs (fb->msg_size);
  fc.message_uuid = fb->msg_uuid;
  fc.rc = NULL;
  (void) GNUNET_CONTAINER_multihashmap32_get_multiple (vl->reassembly_map,
                                                       fb->msg_uuid.uuid,
                                                       &find_by_message_uuid,
                                                       &fc);
  fsize = ntohs (fb->header.size) - sizeof(*fb);
  if (NULL == (rc = fc.rc))
  {
    rc = GNUNET_malloc (sizeof(*rc) + msize    /* reassembly payload buffer */
                        + (msize + 7) / 8 * sizeof(uint8_t) /* bitfield */);
    rc->msg_uuid = fb->msg_uuid;
    rc->virtual_link = vl;
    rc->msg_size = msize;
    rc->reassembly_timeout =
      GNUNET_TIME_relative_to_absolute (REASSEMBLY_EXPIRATION);
    rc->last_frag = GNUNET_TIME_absolute_get ();
    rc->hn = GNUNET_CONTAINER_heap_insert (vl->reassembly_heap,
                                           rc,
                                           rc->reassembly_timeout.abs_value_us);
    GNUNET_assert (GNUNET_OK ==
                   GNUNET_CONTAINER_multihashmap32_put (
                     vl->reassembly_map,
                     rc->msg_uuid.uuid,
                     rc,
                     GNUNET_CONTAINER_MULTIHASHMAPOPTION_MULTIPLE));
    target = (char *) &rc[1];
    rc->bitfield = (uint8_t *) (target + rc->msg_size);
    if (fsize != rc->msg_size)
      rc->msg_missing = rc->msg_size;
    else
      rc->msg_missing = 0;
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Received fragment with size %u at offset %u/%u %u bytes missing from %s for NEW message %"
                PRIu64 "\n",
                fsize,
                ntohs (fb->frag_off),
                msize,
                rc->msg_missing,
                GNUNET_i2s (&cmc->im.sender),
                fb->msg_uuid.uuid);
  }
  else
  {
    target = (char *) &rc[1];
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Received fragment at offset %u/%u from %s for message %u\n",
                ntohs (fb->frag_off),
                msize,
                GNUNET_i2s (&cmc->im.sender),
                (unsigned int) fb->msg_uuid.uuid);
  }
  if (msize != rc->msg_size)
  {
    GNUNET_break (0);
    finish_cmc_handling (cmc);
    return;
  }
 
  /* reassemble */
  if (0 == fsize)
  {
    GNUNET_break (0);
    finish_cmc_handling (cmc);
    return;
  }
  frag_off = ntohs (fb->frag_off);
  if (frag_off + fsize > msize)
  {
    /* Fragment (plus fragment size) exceeds message size! */
    GNUNET_break_op (0);
    finish_cmc_handling (cmc);
    return;
  }
  memcpy (&target[frag_off], &fb[1], fsize);
  /* update bitfield and msg_missing */
  for (unsigned int i = frag_off; i < frag_off + fsize; i++)
  {
    if (0 == (rc->bitfield[i / 8] & (1 << (i % 8))))
    {
      rc->bitfield[i / 8] |= (1 << (i % 8));
      rc->msg_missing--;
    }
  }
 
  /* Compute cumulative ACK */
  cdelay = GNUNET_TIME_absolute_get_duration (rc->last_frag);
  cdelay = GNUNET_TIME_relative_multiply (cdelay, rc->msg_missing / fsize);
  if (0 == rc->msg_missing)
    cdelay = GNUNET_TIME_UNIT_ZERO;
  cummulative_ack (&cmc->im.sender,
                   &fb->ack_uuid,
                   GNUNET_TIME_relative_to_absolute (cdelay));
  rc->last_frag = GNUNET_TIME_absolute_get ();
  /* is reassembly complete? */
  if (0 != rc->msg_missing)
  {
    finish_cmc_handling (cmc);
    return;
  }
  /* reassembly is complete, verify result */
  msg = (const struct GNUNET_MessageHeader *) &rc[1];
  if (ntohs (msg->size) != rc->msg_size)
  {
    GNUNET_break (0);
    free_reassembly_context (rc);
    finish_cmc_handling (cmc);
    return;
  }
  /* successful reassembly */
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Fragment reassembly complete for message %u\n",
              (unsigned int) fb->msg_uuid.uuid);
  /* FIXME: check that the resulting msg is NOT a
     DV Box or Reliability Box, as that is NOT allowed! */
  cmc->mh = msg;
  demultiplex_with_cmc (cmc);
  /* FIXME-OPTIMIZE: really free here? Might be bad if fragments are still
     en-route and we forget that we finished this reassembly immediately!
     -> keep around until timeout?
     -> shorten timeout based on ACK? */
  free_reassembly_context (rc);
}

References GNUNET_TIME_Absolute::abs_value_us, TransportFragmentBoxMessage::ack_uuid, ReassemblyContext::bitfield, TransportClient::client, VirtualLink::confirmed, cummulative_ack(), demultiplex_with_cmc(), find_by_message_uuid(), finish_cmc_handling(), TransportFragmentBoxMessage::frag_off, free_reassembly_context(), GNUNET_assert, GNUNET_break, GNUNET_break_op, GNUNET_CONTAINER_heap_create(), GNUNET_CONTAINER_heap_insert(), GNUNET_CONTAINER_HEAP_ORDER_MIN, GNUNET_CONTAINER_multihashmap32_create(), GNUNET_CONTAINER_multihashmap32_get_multiple(), GNUNET_CONTAINER_multihashmap32_put(), GNUNET_CONTAINER_MULTIHASHMAPOPTION_MULTIPLE, GNUNET_ERROR_TYPE_DEBUG, GNUNET_i2s(), GNUNET_log, GNUNET_malloc, GNUNET_NO, GNUNET_OK, GNUNET_SCHEDULER_add_delayed(), GNUNET_SERVICE_client_drop(), GNUNET_TIME_absolute_get(), GNUNET_TIME_absolute_get_duration(), GNUNET_TIME_relative_multiply(), GNUNET_TIME_relative_to_absolute(), GNUNET_TIME_UNIT_ZERO, TransportFragmentBoxMessage::header, ReassemblyContext::hn, CommunicatorMessageContext::im, ReassemblyContext::last_frag, lookup_virtual_link(), FindByMessageUuidContext::message_uuid, CommunicatorMessageContext::mh, msg, ReassemblyContext::msg_missing, TransportFragmentBoxMessage::msg_size, ReassemblyContext::msg_size, TransportFragmentBoxMessage::msg_uuid, ReassemblyContext::msg_uuid, FindByMessageUuidContext::rc, reassembly_cleanup_task(), REASSEMBLY_EXPIRATION, VirtualLink::reassembly_heap, VirtualLink::reassembly_map, ReassemblyContext::reassembly_timeout, VirtualLink::reassembly_timeout_task, GNUNET_TRANSPORT_IncomingMessage::sender, GNUNET_MessageHeader::size, CommunicatorMessageContext::tc, MessageUUIDP::uuid, and ReassemblyContext::virtual_link.

Here is the call graph for this function:

check_reliability_box()

static int check_reliability_box ( void *  cls, const struct TransportReliabilityBoxMessage *  rb  )

static

Communicator gave us a reliability box.

Check the message.

Parameters

cls	a struct CommunicatorMessageContext
rb	the send message that was sent

Returns

GNUNET_YES if message is well-formed

Definition at line 7038 of file gnunet-service-transport.c.

{
  const struct GNUNET_MessageHeader *box =  (const struct
                                             GNUNET_MessageHeader *) &rb[1];
  (void) cls;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "check_send_msg with size %u: inner msg type %u and size %u (%lu %lu)\n",
              ntohs (rb->header.size),
              ntohs (box->type),
              ntohs (box->size),
              sizeof (struct TransportReliabilityBoxMessage),
              sizeof (struct GNUNET_MessageHeader));
  GNUNET_MQ_check_boxed_message (rb);
  return GNUNET_YES;
}

References GNUNET_ERROR_TYPE_DEBUG, GNUNET_log, GNUNET_MQ_check_boxed_message, GNUNET_YES, TransportReliabilityBoxMessage::header, GNUNET_MessageHeader::size, and GNUNET_MessageHeader::type.

handle_reliability_box()

static void handle_reliability_box ( void *  cls, const struct TransportReliabilityBoxMessage *  rb  )

static

Communicator gave us a reliability box.

Process the request.

Parameters

cls	a struct CommunicatorMessageContext (must call finish_cmc_handling() when done)
rb	the message that was received

Definition at line 7065 of file gnunet-service-transport.c.

{
  struct CommunicatorMessageContext *cmc = cls;
  const struct GNUNET_MessageHeader *inbox =
    (const struct GNUNET_MessageHeader *) &rb[1];
  struct GNUNET_TIME_Relative rtt;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Received reliability box from %s with UUID %s of type %u\n",
              GNUNET_i2s (&cmc->im.sender),
              GNUNET_uuid2s (&rb->ack_uuid.value),
              (unsigned int) ntohs (inbox->type));
  rtt = GNUNET_TIME_UNIT_SECONDS; /* FIXME: should base this on "RTT", but we
                                     do not really have an RTT for the
                                   * incoming* queue (should we have
                                     the sender add it to the rb message?) */
  cummulative_ack (
    &cmc->im.sender,
    &rb->ack_uuid,
    (0 == ntohl (rb->ack_countdown))
    ? GNUNET_TIME_UNIT_ZERO_ABS
    : GNUNET_TIME_relative_to_absolute (
      GNUNET_TIME_relative_divide (rtt, 8 /* FIXME: magic constant */)));
  /* continue with inner message */
  /* FIXME: check that inbox is NOT a DV Box, fragment or another
     reliability box (not allowed!) */
  cmc->mh = inbox;
  demultiplex_with_cmc (cmc);
}

References TransportReliabilityBoxMessage::ack_countdown, TransportReliabilityBoxMessage::ack_uuid, cummulative_ack(), demultiplex_with_cmc(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_i2s(), GNUNET_log, GNUNET_TIME_relative_divide(), GNUNET_TIME_relative_to_absolute(), GNUNET_TIME_UNIT_SECONDS, GNUNET_TIME_UNIT_ZERO_ABS, GNUNET_uuid2s(), CommunicatorMessageContext::im, CommunicatorMessageContext::mh, GNUNET_TRANSPORT_IncomingMessage::sender, GNUNET_MessageHeader::type, and AcknowledgementUUIDP::value.

Here is the call graph for this function:

update_pd_age()

static void update_pd_age ( struct PerformanceData *  pd, unsigned int  age  )

static

Check if we have advanced to another age since the last time.

If so, purge ancient statistics (more than GOODPUT_AGING_SLOTS before the current age)

Parameters

[in,out]	pd	data to update
	age	current age

Definition at line 7106 of file gnunet-service-transport.c.

{
  unsigned int sage;
 
  if (age == pd->last_age)
    return; /* nothing to do */
  sage = GNUNET_MAX (pd->last_age, age - 2 * GOODPUT_AGING_SLOTS);
  for (unsigned int i = sage; i <= age - GOODPUT_AGING_SLOTS; i++)
  {
    struct TransmissionHistoryEntry *the = &pd->the[i % GOODPUT_AGING_SLOTS];
 
    the->bytes_sent = 0;
    the->bytes_received = 0;
  }
  pd->last_age = age;
}

References TransmissionHistoryEntry::bytes_received, TransmissionHistoryEntry::bytes_sent, GNUNET_MAX, GOODPUT_AGING_SLOTS, PerformanceData::last_age, and PerformanceData::the.

Referenced by update_performance_data().

Here is the caller graph for this function:

update_performance_data()

static void update_performance_data ( struct PerformanceData *  pd, struct GNUNET_TIME_Relative  rtt, uint16_t  bytes_transmitted_ok  )

static

Update pd based on the latest rtt and the number of bytes that were confirmed to be successfully transmitted.

Parameters

[in,out]	pd	data to update
	rtt	latest round-trip time
	bytes_transmitted_ok	number of bytes receiver confirmed as received

Definition at line 7133 of file gnunet-service-transport.c.

{
  uint64_t nval = rtt.rel_value_us;
  uint64_t oval = pd->aged_rtt.rel_value_us;
  unsigned int age = get_age ();
  struct TransmissionHistoryEntry *the = &pd->the[age % GOODPUT_AGING_SLOTS];
 
  if (oval == GNUNET_TIME_UNIT_FOREVER_REL.rel_value_us)
    pd->aged_rtt = rtt;
  else
    pd->aged_rtt.rel_value_us = (nval + 7 * oval) / 8;
  update_pd_age (pd, age);
  the->bytes_received += bytes_transmitted_ok;
}

References PerformanceData::aged_rtt, TransmissionHistoryEntry::bytes_received, get_age(), GNUNET_TIME_UNIT_FOREVER_REL, GOODPUT_AGING_SLOTS, GNUNET_TIME_Relative::rel_value_us, PerformanceData::the, and update_pd_age().

Referenced by update_dvh_performance(), and update_queue_performance().

Here is the call graph for this function:

Here is the caller graph for this function:

update_queue_performance()

static void update_queue_performance ( struct Queue *  q, struct GNUNET_TIME_Relative  rtt, uint16_t  bytes_transmitted_ok  )

static

We have successfully transmitted data via q, update metrics.

Parameters

q	queue to update
rtt	round trip time observed
bytes_transmitted_ok	number of bytes successfully transmitted

Definition at line 7159 of file gnunet-service-transport.c.

{
  update_performance_data (&q->pd, rtt, bytes_transmitted_ok);
}

References q, and update_performance_data().

Referenced by handle_acknowledged().

Here is the call graph for this function:

Here is the caller graph for this function:

update_dvh_performance()

static void update_dvh_performance ( struct DistanceVectorHop *  dvh, struct GNUNET_TIME_Relative  rtt, uint16_t  bytes_transmitted_ok  )

static

We have successfully transmitted data via dvh, update metrics.

Parameters

dvh	distance vector path data to update
rtt	round trip time observed
bytes_transmitted_ok	number of bytes successfully transmitted

Definition at line 7175 of file gnunet-service-transport.c.

{
  update_performance_data (&dvh->pd, rtt, bytes_transmitted_ok);
}

References DistanceVectorHop::pd, and update_performance_data().

Referenced by handle_acknowledged().

Here is the call graph for this function:

Here is the caller graph for this function:

completed_pending_message()

static void completed_pending_message ( struct PendingMessage *  pm )

static

We have completed transmission of pm, remove it from the transmission queues (and if it is a fragment, continue up the tree as necessary).

Parameters

pm	pending message that was transmitted

Definition at line 7191 of file gnunet-service-transport.c.

{
  struct PendingMessage *pos;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Complete transmission of message %" PRIu64 " %u\n",
              pm->logging_uuid,
              pm->pmt);
  switch (pm->pmt)
  {
  case PMT_CORE:
  case PMT_RELIABILITY_BOX:
    /* Full message sent, we are done */
    client_send_response (pm);
    return;
 
  case PMT_FRAGMENT_BOX:
    /* Fragment sent over reliable channel */
    pos = pm->frag_parent;
    GNUNET_CONTAINER_MDLL_remove (frag, pos->head_frag, pos->tail_frag, pm);
    free_pending_message (pm);
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "pos frag_off %lu pos bytes_msg %lu pmt %u parent %u\n",
                (unsigned long) pos->frag_off,
                (unsigned long) pos->bytes_msg,
                pos->pmt,
                NULL == pos->frag_parent ? 1 : 0);
    /* check if subtree is done */
    while ((NULL == pos->head_frag) && (pos->frag_off == (pos->bytes_msg
                                                          - sizeof(struct
                                                                   TransportFragmentBoxMessage)))
           &&
           (NULL != pos->frag_parent))
    {
      pm = pos;
      pos = pm->frag_parent;
      if ((NULL == pos) && (PMT_DV_BOX == pm->pmt))
      {
        client_send_response (pm);
        return;
      }
      else if (PMT_DV_BOX == pm->pmt)
      {
        client_send_response (pos);
        return;
      }
      GNUNET_CONTAINER_MDLL_remove (frag, pos->head_frag, pos->tail_frag, pm);
      free_pending_message (pm);
    }
 
    /* Was this the last applicable fragment? */
    if ((NULL == pos->head_frag) && (NULL == pos->frag_parent || PMT_DV_BOX ==
                                     pos->pmt) &&
        (pos->frag_off == pos->bytes_msg))
      client_send_response (pos);
    return;
 
  case PMT_DV_BOX:
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Completed transmission of message %" PRIu64 " (DV Box)\n",
                pm->logging_uuid);
    if (NULL != pm->frag_parent)
    {
      pos = pm->frag_parent;
      free_pending_message (pm);
      pos->bpm = NULL;
      client_send_response (pos);
    }
    else
      client_send_response (pm);
    return;
  }
}

References PendingMessage::bpm, PendingMessage::bytes_msg, client_send_response(), PendingMessage::frag_off, PendingMessage::frag_parent, free_pending_message(), GNUNET_CONTAINER_MDLL_remove, GNUNET_ERROR_TYPE_DEBUG, GNUNET_log, PendingMessage::head_frag, pm, PendingMessage::pmt, PMT_CORE, PMT_DV_BOX, PMT_FRAGMENT_BOX, PMT_RELIABILITY_BOX, and PendingMessage::tail_frag.

Referenced by handle_acknowledged(), and transmit_on_queue().

Here is the call graph for this function:

Here is the caller graph for this function:

handle_acknowledged()

static void handle_acknowledged ( struct PendingAcknowledgement *  pa, struct GNUNET_TIME_Relative  ack_delay  )

static

The pa was acknowledged, process the acknowledgement.

Parameters

pa	the pending acknowledgement that was satisfied
ack_delay	artificial delay from cumulative acks created by the other peer

Definition at line 7274 of file gnunet-service-transport.c.

{
  struct GNUNET_TIME_Relative delay;
 
  delay = GNUNET_TIME_absolute_get_duration (pa->transmission_time);
  delay = GNUNET_TIME_relative_subtract (delay, ack_delay);
  if (NULL != pa->queue && 1 == pa->num_send)
    update_queue_performance (pa->queue, delay, pa->message_size);
  if (NULL != pa->dvh && 1 == pa->num_send)
    update_dvh_performance (pa->dvh, delay, pa->message_size);
  if (NULL != pa->pm)
    completed_pending_message (pa->pm);
  free_pending_acknowledgement (pa);
}

References completed_pending_message(), PendingAcknowledgement::dvh, free_pending_acknowledgement(), GNUNET_TIME_absolute_get_duration(), GNUNET_TIME_relative_subtract(), PendingAcknowledgement::message_size, PendingAcknowledgement::num_send, PendingAcknowledgement::pm, PendingAcknowledgement::queue, PendingAcknowledgement::transmission_time, update_dvh_performance(), and update_queue_performance().

Referenced by handle_reliability_ack().

Here is the call graph for this function:

Here is the caller graph for this function:

check_reliability_ack()

static int check_reliability_ack ( void *  cls, const struct TransportReliabilityAckMessage *  ra  )

static

Communicator gave us a reliability ack.

Check it is well-formed.

Parameters

cls	a struct CommunicatorMessageContext (unused)
ra	the message that was received

Returns

#GNUNET_Ok if ra is well-formed

Definition at line 7299 of file gnunet-service-transport.c.

{
  unsigned int n_acks;
 
  (void) cls;
  n_acks = (ntohs (ra->header.size) - sizeof(*ra))
           / sizeof(struct TransportCummulativeAckPayloadP);
  if (0 == n_acks)
  {
    GNUNET_break_op (0);
    return GNUNET_SYSERR;
  }
  if ((ntohs (ra->header.size) - sizeof(*ra)) !=
      n_acks * sizeof(struct TransportCummulativeAckPayloadP))
  {
    GNUNET_break_op (0);
    return GNUNET_SYSERR;
  }
  return GNUNET_OK;
}

References GNUNET_break_op, GNUNET_OK, GNUNET_SYSERR, TransportReliabilityAckMessage::header, and GNUNET_MessageHeader::size.

handle_reliability_ack()

static void handle_reliability_ack ( void *  cls, const struct TransportReliabilityAckMessage *  ra  )

static

Communicator gave us a reliability ack.

Process the request.

Parameters

cls	a struct CommunicatorMessageContext (must call finish_cmc_handling() when done)
ra	the message that was received

Definition at line 7330 of file gnunet-service-transport.c.

{
  struct CommunicatorMessageContext *cmc = cls;
  const struct TransportCummulativeAckPayloadP *ack;
  unsigned int n_acks;
  uint32_t ack_counter;
 
  n_acks = (ntohs (ra->header.size) - sizeof(*ra))
           / sizeof(struct TransportCummulativeAckPayloadP);
  ack = (const struct TransportCummulativeAckPayloadP *) &ra[1];
  for (unsigned int i = 0; i < n_acks; i++)
  {
    struct PendingAcknowledgement *pa =
      GNUNET_CONTAINER_multiuuidmap_get (pending_acks, &ack[i].ack_uuid.value);
    if (NULL == pa)
    {
      GNUNET_log (GNUNET_ERROR_TYPE_INFO,
                  "Received ACK from %s with UUID %s which is unknown to us!\n",
                  GNUNET_i2s (&cmc->im.sender),
                  GNUNET_uuid2s (&ack[i].ack_uuid.value));
      GNUNET_STATISTICS_update (
        GST_stats,
        "# FRAGMENT_ACKS dropped, no matching pending message",
        1,
        GNUNET_NO);
      continue;
    }
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Received ACK from %s with UUID %s\n",
                GNUNET_i2s (&cmc->im.sender),
                GNUNET_uuid2s (&ack[i].ack_uuid.value));
    handle_acknowledged (pa, GNUNET_TIME_relative_ntoh (ack[i].ack_delay));
  }
 
  ack_counter = htonl (ra->ack_counter);
  (void) ack_counter;  /* silence compiler warning for now */
  // FIXME-OPTIMIZE: track ACK losses based on ack_counter somewhere!
  // (DV and/or Neighbour?)
  finish_cmc_handling (cmc);
}

References TransportReliabilityAckMessage::ack_counter, PendingAcknowledgement::ack_uuid, finish_cmc_handling(), GNUNET_CONTAINER_multiuuidmap_get(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_ERROR_TYPE_INFO, GNUNET_i2s(), GNUNET_log, GNUNET_NO, GNUNET_STATISTICS_update(), GNUNET_TIME_relative_ntoh(), GNUNET_uuid2s(), GST_stats, handle_acknowledged(), TransportReliabilityAckMessage::header, CommunicatorMessageContext::im, pending_acks, GNUNET_TRANSPORT_IncomingMessage::sender, GNUNET_MessageHeader::size, and AcknowledgementUUIDP::value.

Here is the call graph for this function:

check_backchannel_encapsulation()

static int check_backchannel_encapsulation ( void *  cls, const struct TransportBackchannelEncapsulationMessage *  be  )

static

Communicator gave us a backchannel encapsulation.

Check the message.

Parameters

cls	a struct CommunicatorMessageContext
be	the send message that was sent

Returns

GNUNET_YES if message is well-formed

Definition at line 7381 of file gnunet-service-transport.c.

{
  uint16_t size = ntohs (be->header.size) - sizeof(*be);
  const struct GNUNET_MessageHeader *inbox =
    (const struct GNUNET_MessageHeader *) &be[1];
  const char *is;
  uint16_t isize;
 
  (void) cls;
  if (ntohs (inbox->size) >= size)
  {
    GNUNET_break_op (0);
    return GNUNET_SYSERR;
  }
  isize = ntohs (inbox->size);
  is = ((const char *) inbox) + isize;
  size -= isize;
  if ('\0' != is[size - 1])
  {
    GNUNET_break_op (0);
    return GNUNET_SYSERR;
  }
  return GNUNET_YES;
}

References GNUNET_break_op, GNUNET_SYSERR, GNUNET_YES, TransportBackchannelEncapsulationMessage::header, is, GNUNET_MessageHeader::size, and size.

handle_backchannel_encapsulation()

static void handle_backchannel_encapsulation ( void *  cls, const struct TransportBackchannelEncapsulationMessage *  be  )

static

Communicator gave us a backchannel encapsulation.

Process the request. (We are the destination of the backchannel here.)

Parameters

cls	a struct CommunicatorMessageContext (must call finish_cmc_handling() when done)
be	the message that was received

Definition at line 7418 of file gnunet-service-transport.c.

{
  struct CommunicatorMessageContext *cmc = cls;
  struct GNUNET_TRANSPORT_CommunicatorBackchannelIncoming *cbi;
  struct GNUNET_MQ_Envelope *env;
  struct TransportClient *tc;
  const struct GNUNET_MessageHeader *inbox =
    (const struct GNUNET_MessageHeader *) &be[1];
  uint16_t isize = ntohs (inbox->size);
  const char *target_communicator = ((const char *) inbox) + isize;
  char *sender;
  char *self;
 
  GNUNET_asprintf (&sender,
                   "%s",
                   GNUNET_i2s (&cmc->im.sender));
  GNUNET_asprintf (&self,
                   "%s",
                   GNUNET_i2s (GST_my_identity));
 
  /* Find client providing this communicator */
  for (tc = clients_head; NULL != tc; tc = tc->next)
    if ((CT_COMMUNICATOR == tc->type) &&
        (0 ==
         strcmp (tc->details.communicator.address_prefix, target_communicator)))
      break;
  if (NULL == tc)
  {
    char *stastr;
 
    GNUNET_asprintf (
      &stastr,
      "# Backchannel message dropped: target communicator `%s' unknown",
      target_communicator);
    GNUNET_STATISTICS_update (GST_stats, stastr, 1, GNUNET_NO);
    GNUNET_free (stastr);
    finish_cmc_handling (cmc);
    return;
  }
  /* Finally, deliver backchannel message to communicator */
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Delivering backchannel message from %s to %s of type %u to %s\n",
              sender,
              self,
              ntohs (inbox->type),
              target_communicator);
  env = GNUNET_MQ_msg_extra (
    cbi,
    isize,
    GNUNET_MESSAGE_TYPE_TRANSPORT_COMMUNICATOR_BACKCHANNEL_INCOMING);
  cbi->pid = cmc->im.sender;
  memcpy (&cbi[1], inbox, isize);
  GNUNET_MQ_send (tc->mq, env);
  finish_cmc_handling (cmc);
}

References clients_head, CT_COMMUNICATOR, env, finish_cmc_handling(), GNUNET_asprintf(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_free, GNUNET_i2s(), GNUNET_log, GNUNET_MESSAGE_TYPE_TRANSPORT_COMMUNICATOR_BACKCHANNEL_INCOMING, GNUNET_MQ_msg_extra, GNUNET_MQ_send(), GNUNET_NO, GNUNET_STATISTICS_update(), GST_my_identity, GST_stats, CommunicatorMessageContext::im, GNUNET_TRANSPORT_CommunicatorBackchannelIncoming::pid, GNUNET_TRANSPORT_IncomingMessage::sender, GNUNET_MessageHeader::size, tc, and GNUNET_MessageHeader::type.

Here is the call graph for this function:

path_cleanup_cb()

static void path_cleanup_cb ( void *  cls )

static

Task called when we should check if any of the DV paths we have learned to a target are due for garbage collection.

Collects stale paths, and possibly frees the entire DV entry if no paths are left. Otherwise re-schedules itself.

Parameters

cls	a struct DistanceVector

Definition at line 7487 of file gnunet-service-transport.c.

{
  struct DistanceVector *dv = cls;
  struct DistanceVectorHop *pos;
 
  dv->timeout_task = NULL;
  while (NULL != (pos = dv->dv_head))
  {
    GNUNET_assert (dv == pos->dv);
    if (GNUNET_TIME_absolute_get_remaining (pos->timeout).rel_value_us > 0)
      break;
    free_distance_vector_hop (pos);
  }
  if (NULL == pos)
  {
    free_dv_route (dv);
    return;
  }
  dv->timeout_task =
    GNUNET_SCHEDULER_add_at (pos->timeout, &path_cleanup_cb, dv);
}

References DistanceVectorHop::dv, DistanceVector::dv_head, free_distance_vector_hop(), free_dv_route(), GNUNET_assert, GNUNET_SCHEDULER_add_at(), GNUNET_TIME_absolute_get_remaining(), path_cleanup_cb(), GNUNET_TIME_Relative::rel_value_us, DistanceVectorHop::timeout, and DistanceVector::timeout_task.

Referenced by learn_dv_path(), and path_cleanup_cb().

Here is the call graph for this function:

Here is the caller graph for this function:

send_msg_from_cache()

static void send_msg_from_cache ( struct VirtualLink *  vl )

static

Definition at line 7511 of file gnunet-service-transport.c.

{
 
  const struct GNUNET_PeerIdentity target = vl->target;
 
 
  if ((GNUNET_YES == is_ring_buffer_full) || (0 < ring_buffer_head))
  {
    struct RingBufferEntry *ring_buffer_copy[RING_BUFFER_SIZE];
    unsigned int tail = GNUNET_YES == is_ring_buffer_full ? ring_buffer_head :
                        0;
    unsigned int head = GNUNET_YES == is_ring_buffer_full ? RING_BUFFER_SIZE :
                        ring_buffer_head;
    struct GNUNET_TRANSPORT_IncomingMessage im;
    struct CommunicatorMessageContext *cmc;
    struct RingBufferEntry *rbe;
    struct GNUNET_MessageHeader *mh;
 
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Sending from ring buffer, which has %u items\n",
                head);
 
    ring_buffer_head = 0;
    for (unsigned int i = 0; i < head; i++)
    {
      rbe = ring_buffer[(i + tail) % RING_BUFFER_SIZE];
      cmc = rbe->cmc;
      mh = rbe->mh;
 
      im = cmc->im;
      // mh = cmc->mh;
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "Sending message of type %u to ring buffer target %s using vl target %s index %u\n",
                  mh->type,
                  GNUNET_i2s (&im.sender),
                  GNUNET_i2s2 (&target),
                  (i + tail) % RING_BUFFER_SIZE);
      if (0 == GNUNET_memcmp (&target, &im.sender))
      {
        GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                    "Finish handling message of type %u and size %u\n",
                    (unsigned int) ntohs (mh->type),
                    (unsigned int) ntohs (mh->size));
        finish_handling_raw_message (vl, mh, cmc, GNUNET_NO);
        GNUNET_free (mh);
        GNUNET_free (rbe->cmc);
        GNUNET_free (rbe);
      }
      else
      {
        ring_buffer_copy[ring_buffer_head] = rbe;
        ring_buffer_head++;
      }
    }
 
    if ((GNUNET_YES == is_ring_buffer_full) && (RING_BUFFER_SIZE - 1 >
                                                ring_buffer_head))
    {
      is_ring_buffer_full = GNUNET_NO;
    }
 
    for (unsigned int i = 0; i < ring_buffer_head; i++)
    {
      ring_buffer[i] = ring_buffer_copy[i];
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "ring_buffer_copy[i]->mh->type for i %u %u\n",
                  i,
                  ring_buffer_copy[i]->mh->type);
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "ring_buffer[i]->mh->type for i %u %u\n",
                  i,
                  ring_buffer[i]->mh->type);
    }
 
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "%u items still in ring buffer\n",
                ring_buffer_head);
  }
 
  if ((GNUNET_YES == is_ring_buffer_dv_full) || (0 < ring_buffer_dv_head))
  {
    struct PendingMessage *ring_buffer_dv_copy[RING_BUFFER_SIZE];
    struct PendingMessage *pm;
    unsigned int tail = GNUNET_YES == is_ring_buffer_dv_full ?
                        ring_buffer_dv_head :
                        0;
    unsigned int head = GNUNET_YES == is_ring_buffer_dv_full ?
                        RING_BUFFER_SIZE :
                        ring_buffer_dv_head;
 
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Sending from ring buffer dv, which has %u items\n",
                head);
 
    ring_buffer_dv_head = 0;
    for (unsigned int i = 0; i < head; i++)
    {
      pm = ring_buffer_dv[(i + tail) % RING_BUFFER_SIZE];
 
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "Sending to ring buffer target %s using vl target %s\n",
                  GNUNET_i2s (&pm->target),
                  GNUNET_i2s2 (&target));
      if (0 == GNUNET_memcmp (&target, &pm->target))
      {
        GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                    "Adding PendingMessage to vl, checking transmission.\n");
        pm->vl = vl;
        GNUNET_CONTAINER_MDLL_insert (vl,
                                      vl->pending_msg_head,
                                      vl->pending_msg_tail,
                                      pm);
 
        check_vl_transmission (vl);
      }
      else
      {
        ring_buffer_dv_copy[ring_buffer_dv_head] = pm;
        ring_buffer_dv_head++;
      }
    }
 
    if (is_ring_buffer_dv_full && (RING_BUFFER_SIZE - 1 > ring_buffer_dv_head))
    {
      is_ring_buffer_dv_full = GNUNET_NO;
    }
 
    for (unsigned int i = 0; i < ring_buffer_dv_head; i++)
      ring_buffer_dv[i] = ring_buffer_dv_copy[i];
 
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "%u items still in ring buffer dv.\n",
                ring_buffer_dv_head);
 
  }
}

References check_vl_transmission(), RingBufferEntry::cmc, finish_handling_raw_message(), GNUNET_CONTAINER_MDLL_insert, GNUNET_ERROR_TYPE_DEBUG, GNUNET_free, GNUNET_i2s(), GNUNET_i2s2(), GNUNET_log, GNUNET_memcmp, GNUNET_NO, GNUNET_YES, is_ring_buffer_dv_full, is_ring_buffer_full, mh, RingBufferEntry::mh, VirtualLink::pending_msg_head, VirtualLink::pending_msg_tail, pm, ring_buffer, ring_buffer_dv, ring_buffer_dv_head, ring_buffer_head, RING_BUFFER_SIZE, GNUNET_TRANSPORT_IncomingMessage::sender, VirtualLink::target, PendingMessage::target, and PendingMessage::vl.

Referenced by activate_core_visible_dv_path(), and handle_validation_response().

Here is the call graph for this function:

Here is the caller graph for this function:

activate_core_visible_dv_path()

static void activate_core_visible_dv_path ( struct DistanceVectorHop *  hop )

static

The hop is a validated path to the respective target peer and we should tell core about it – and schedule a job to revoke the state.

Parameters

hop	a path to some peer that is the reason for activation

Definition at line 7657 of file gnunet-service-transport.c.

{
  struct DistanceVector *dv = hop->dv;
  struct VirtualLink *vl;
 
  vl = lookup_virtual_link (&dv->target);
  if (NULL == vl)
  {
 
    vl = GNUNET_new (struct VirtualLink);
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Creating new virtual link %p to %s using DV!\n",
                vl,
                GNUNET_i2s (&dv->target));
    vl->burst_addr = NULL;
    vl->confirmed = GNUNET_YES;
    vl->message_uuid_ctr =
      GNUNET_CRYPTO_random_u64 (GNUNET_CRYPTO_QUALITY_WEAK, UINT64_MAX);
    vl->target = dv->target;
    vl->core_recv_window = RECV_WINDOW_SIZE;
    vl->available_fc_window_size = DEFAULT_WINDOW_SIZE;
    vl->incoming_fc_window_size = DEFAULT_WINDOW_SIZE;
    GNUNET_break (GNUNET_YES ==
                  GNUNET_CONTAINER_multipeermap_put (
                    links,
                    &vl->target,
                    vl,
                    GNUNET_CONTAINER_MULTIHASHMAPOPTION_UNIQUE_ONLY));
    vl->dv = dv;
    dv->vl = vl;
    vl->visibility_task =
      GNUNET_SCHEDULER_add_at (hop->path_valid_until, &check_link_down, vl);
    consider_sending_fc (vl);
    /* We lacked a confirmed connection to the target
       before, so tell CORE about it (finally!) */
    cores_send_connect_info (&dv->target);
    send_msg_from_cache (vl);
  }
  else
  {
    /* Link was already up, remember dv is also now available and we are done */
    vl->dv = dv;
    dv->vl = vl;
    if (GNUNET_NO == vl->confirmed)
    {
      vl->confirmed = GNUNET_YES;
      vl->visibility_task =
        GNUNET_SCHEDULER_add_at (hop->path_valid_until, &check_link_down, vl);
      consider_sending_fc (vl);
      /* We lacked a confirmed connection to the target
         before, so tell CORE about it (finally!) */
      cores_send_connect_info (&dv->target);
      send_msg_from_cache (vl);
    }
    else
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "Virtual link to %s could now also use DV!\n",
                  GNUNET_i2s (&dv->target));
  }
}

References VirtualLink::available_fc_window_size, VirtualLink::burst_addr, check_link_down(), VirtualLink::confirmed, consider_sending_fc(), VirtualLink::core_recv_window, cores_send_connect_info(), DEFAULT_WINDOW_SIZE, VirtualLink::dv, DistanceVectorHop::dv, GNUNET_break, GNUNET_CONTAINER_MULTIHASHMAPOPTION_UNIQUE_ONLY, GNUNET_CONTAINER_multipeermap_put(), GNUNET_CRYPTO_QUALITY_WEAK, GNUNET_CRYPTO_random_u64(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_i2s(), GNUNET_log, GNUNET_new, GNUNET_NO, GNUNET_SCHEDULER_add_at(), GNUNET_YES, VirtualLink::incoming_fc_window_size, links, lookup_virtual_link(), VirtualLink::message_uuid_ctr, DistanceVectorHop::path_valid_until, RECV_WINDOW_SIZE, send_msg_from_cache(), VirtualLink::target, DistanceVector::target, VirtualLink::visibility_task, and DistanceVector::vl.

Referenced by learn_dv_path().

Here is the call graph for this function:

Here is the caller graph for this function:

learn_dv_path()

static int learn_dv_path ( const struct GNUNET_PeerIdentity *  path, unsigned int  path_len, struct GNUNET_TIME_Relative  network_latency, struct GNUNET_TIME_Absolute  path_valid_until  )

static

We have learned a path through the network to some other peer, add it to our DV data structure (returning GNUNET_YES on success).

We do not add paths if we have a sufficient number of shorter paths to this target already (returning GNUNET_NO).

We also do not add problematic paths, like those where we lack the first hop in our neighbour list (i.e. due to a topology change) or where some non-first hop is in our neighbour list (returning GNUNET_SYSERR).

Parameters

path	the path we learned, path[0] should be us, and then path contains a valid path from us to path[path_len-1] path[1] should be a direct neighbour (we should check!)
path_len	number of entries on the path, at least three!
network_latency	how long does the message take from us to path[path_len-1]? set to "forever" if unknown
path_valid_until	how long is this path considered validated? Maybe be zero.

Returns

GNUNET_YES on success, GNUNET_NO if we have better path(s) to the target GNUNET_SYSERR if the path is useless and/or invalid (i.e. path[1] not a direct neighbour or path[i+1] is a direct neighbour for i>0)

Definition at line 7745 of file gnunet-service-transport.c.

{
  struct DistanceVectorHop *hop;
  struct DistanceVector *dv;
  struct Neighbour *next_hop;
  unsigned int shorter_distance;
 
  if (path_len < 3)
  {
    /* what a boring path! not allowed! */
    GNUNET_break (0);
    return GNUNET_SYSERR;
  }
  GNUNET_assert (0 == GNUNET_memcmp (GST_my_identity, &path[0]));
  next_hop = lookup_neighbour (&path[1]);
  if (NULL == next_hop)
  {
    /* next hop must be a neighbour, otherwise this whole thing is useless! */
    GNUNET_break (0);
    return GNUNET_SYSERR;
  }
  for (unsigned int i = 2; i < path_len; i++)
  {
    struct Neighbour *n = lookup_neighbour (&path[i]);
    struct GNUNET_TIME_Absolute q_timeout;
 
    if (NULL != n)
    {
      q_timeout = GNUNET_TIME_UNIT_ZERO_ABS;
      for (struct Queue *q = n->queue_head; NULL != q; q = q->next_neighbour)
        q_timeout = GNUNET_TIME_absolute_max (q_timeout, q->validated_until);
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "remaining %lu to %s\n",
                  (unsigned long) GNUNET_TIME_absolute_get_remaining (q_timeout)
                  .rel_value_us,
                  GNUNET_i2s (&n->pid));
      if (0 != GNUNET_TIME_absolute_get_remaining (q_timeout).rel_value_us)
      {
        /* Useless path: we have a direct active connection to some hop
           in the middle of the path, so this one is not even
           terribly useful for redundancy */
        GNUNET_log (GNUNET_ERROR_TYPE_INFO,
                    "Path of %u hops useless: directly link to hop %u (%s)\n",
                    path_len,
                    i,
                    GNUNET_i2s (&path[i]));
        GNUNET_STATISTICS_update (GST_stats,
                                  "# Useless DV path ignored: hop is neighbour",
                                  1,
                                  GNUNET_NO);
        return GNUNET_SYSERR;
      }
    }
  }
  dv = GNUNET_CONTAINER_multipeermap_get (dv_routes, &path[path_len - 1]);
  if (NULL == dv)
  {
    dv = GNUNET_new (struct DistanceVector);
    dv->target = path[path_len - 1];
    dv->timeout_task = GNUNET_SCHEDULER_add_delayed (DV_PATH_VALIDITY_TIMEOUT,
                                                     &path_cleanup_cb,
                                                     dv);
    GNUNET_assert (GNUNET_OK ==
                   GNUNET_CONTAINER_multipeermap_put (
                     dv_routes,
                     &dv->target,
                     dv,
                     GNUNET_CONTAINER_MULTIHASHMAPOPTION_UNIQUE_ONLY));
  }
  /* Check if we have this path already! */
  shorter_distance = 0;
  for (struct DistanceVectorHop *pos = dv->dv_head; NULL != pos;
       pos = pos->next_dv)
  {
    if (pos->distance < path_len - 3)
      shorter_distance++;
    /* Note that the distances in 'pos' excludes us (path[0]),
       the next_hop (path[1]) and the target so we need to subtract three
       and check next_hop explicitly */
    if ((pos->distance == path_len - 3) && (pos->next_hop == next_hop))
    {
      int match = GNUNET_YES;
 
      for (unsigned int i = 0; i < pos->distance; i++)
      {
        if (0 != GNUNET_memcmp (&pos->path[i], &path[i + 2]))
        {
          match = GNUNET_NO;
          break;
        }
      }
      if (GNUNET_YES == match)
      {
        struct GNUNET_TIME_Relative last_timeout;
 
        /* Re-discovered known path, update timeout */
        GNUNET_STATISTICS_update (GST_stats,
                                  "# Known DV path refreshed",
                                  1,
                                  GNUNET_NO);
        last_timeout = GNUNET_TIME_absolute_get_remaining (pos->timeout);
        pos->timeout =
          GNUNET_TIME_relative_to_absolute (DV_PATH_VALIDITY_TIMEOUT);
        pos->path_valid_until =
          GNUNET_TIME_absolute_max (pos->path_valid_until, path_valid_until);
        GNUNET_CONTAINER_MDLL_remove (dv, dv->dv_head, dv->dv_tail, pos);
        GNUNET_CONTAINER_MDLL_insert (dv, dv->dv_head, dv->dv_tail, pos);
        if (0 <
            GNUNET_TIME_absolute_get_remaining (path_valid_until).rel_value_us)
          activate_core_visible_dv_path (pos);
        if (last_timeout.rel_value_us <
            GNUNET_TIME_relative_subtract (DV_PATH_VALIDITY_TIMEOUT,
                                           DV_PATH_DISCOVERY_FREQUENCY)
            .rel_value_us)
        {
          /* Some peer send DV learn messages too often, we are learning
             the same path faster than it would be useful; do not forward! */
          GNUNET_log (GNUNET_ERROR_TYPE_INFO,
                      "Rediscovered path too quickly, not forwarding further\n")
          ;
          return GNUNET_NO;
        }
        GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                    "Refreshed known path to %s valid until %s, forwarding further\n",
                    GNUNET_i2s (&dv->target),
                    GNUNET_STRINGS_absolute_time_to_string (
                      pos->path_valid_until));
        return GNUNET_YES;
      }
    }
  }
  /* Count how many shorter paths we have (incl. direct
     neighbours) before simply giving up on this one! */
  if (shorter_distance >= MAX_DV_PATHS_TO_TARGET)
  {
    /* We have a shorter path already! */
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Have many shorter DV paths %s, not forwarding further\n",
                GNUNET_i2s (&dv->target));
    return GNUNET_NO;
  }
  /* create new DV path entry */
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Discovered new DV path to %s valid until %s\n",
              GNUNET_i2s (&dv->target),
              GNUNET_STRINGS_absolute_time_to_string (path_valid_until));
  hop = GNUNET_malloc (sizeof(struct DistanceVectorHop)
                       + sizeof(struct GNUNET_PeerIdentity) * (path_len - 3));
  hop->next_hop = next_hop;
  hop->dv = dv;
  hop->path = (const struct GNUNET_PeerIdentity *) &hop[1];
  memcpy (&hop[1],
          &path[2],
          sizeof(struct GNUNET_PeerIdentity) * (path_len - 3));
  hop->timeout = GNUNET_TIME_relative_to_absolute (DV_PATH_VALIDITY_TIMEOUT);
  hop->path_valid_until = path_valid_until;
  hop->distance = path_len - 3;
  hop->pd.aged_rtt = network_latency;
  GNUNET_CONTAINER_MDLL_insert (dv, dv->dv_head, dv->dv_tail, hop);
  GNUNET_CONTAINER_MDLL_insert (neighbour,
                                next_hop->dv_head,
                                next_hop->dv_tail,
                                hop);
  if (0 < GNUNET_TIME_absolute_get_remaining (path_valid_until).rel_value_us)
    activate_core_visible_dv_path (hop);
  return GNUNET_YES;
}

References activate_core_visible_dv_path(), PerformanceData::aged_rtt, DistanceVectorHop::distance, DistanceVectorHop::dv, DistanceVector::dv_head, Neighbour::dv_head, DV_PATH_DISCOVERY_FREQUENCY, DV_PATH_VALIDITY_TIMEOUT, dv_routes, DistanceVector::dv_tail, Neighbour::dv_tail, GNUNET_assert, GNUNET_break, GNUNET_CONTAINER_MDLL_insert, GNUNET_CONTAINER_MDLL_remove, GNUNET_CONTAINER_MULTIHASHMAPOPTION_UNIQUE_ONLY, GNUNET_CONTAINER_multipeermap_get(), GNUNET_CONTAINER_multipeermap_put(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_ERROR_TYPE_INFO, GNUNET_i2s(), GNUNET_log, GNUNET_malloc, GNUNET_memcmp, GNUNET_new, GNUNET_NO, GNUNET_OK, GNUNET_SCHEDULER_add_delayed(), GNUNET_STATISTICS_update(), GNUNET_STRINGS_absolute_time_to_string(), GNUNET_SYSERR, GNUNET_TIME_absolute_get_remaining(), GNUNET_TIME_absolute_max(), GNUNET_TIME_relative_subtract(), GNUNET_TIME_relative_to_absolute(), GNUNET_TIME_UNIT_ZERO_ABS, GNUNET_YES, GST_my_identity, GST_stats, lookup_neighbour(), MAX_DV_PATHS_TO_TARGET, DistanceVectorHop::next_hop, DistanceVectorHop::path, path_cleanup_cb(), DistanceVectorHop::path_valid_until, DistanceVectorHop::pd, Neighbour::pid, q, Neighbour::queue_head, GNUNET_TIME_Relative::rel_value_us, DistanceVector::target, DistanceVectorHop::timeout, and DistanceVector::timeout_task.

Referenced by handle_dv_learn().

Here is the call graph for this function:

Here is the caller graph for this function:

check_dv_learn()

static int check_dv_learn ( void *  cls, const struct TransportDVLearnMessage *  dvl  )

static

Communicator gave us a DV learn message.

Check the message.

Parameters

cls	a struct CommunicatorMessageContext
dvl	the send message that was sent

Returns

GNUNET_YES if message is well-formed

Definition at line 7925 of file gnunet-service-transport.c.

{
  uint16_t size = ntohs (dvl->header.size);
  uint16_t num_hops = ntohs (dvl->num_hops);
  const struct DVPathEntryP *hops = (const struct DVPathEntryP *) &dvl[1];
 
  (void) cls;
  if (size != sizeof(*dvl) + num_hops * sizeof(struct DVPathEntryP))
  {
    GNUNET_break_op (0);
    return GNUNET_SYSERR;
  }
  if (num_hops > MAX_DV_HOPS_ALLOWED)
  {
    GNUNET_break_op (0);
    return GNUNET_SYSERR;
  }
  for (unsigned int i = 0; i < num_hops; i++)
  {
    if (0 == GNUNET_memcmp (&dvl->initiator, &hops[i].hop))
    {
      GNUNET_break_op (0);
      return GNUNET_SYSERR;
    }
    if (0 == GNUNET_memcmp (GST_my_identity, &hops[i].hop))
    {
      GNUNET_break_op (0);
      return GNUNET_SYSERR;
    }
  }
  return GNUNET_YES;
}

References GNUNET_break_op, GNUNET_memcmp, GNUNET_SYSERR, GNUNET_YES, GST_my_identity, TransportDVLearnMessage::header, DVPathEntryP::hop, TransportDVLearnMessage::initiator, MAX_DV_HOPS_ALLOWED, TransportDVLearnMessage::num_hops, GNUNET_MessageHeader::size, and size.

sign_dhp_cp()

static void sign_dhp_cp ( void *  cls, const struct GNUNET_PeerIdentity *  pid, const struct GNUNET_CRYPTO_EddsaSignature *  sig  )

static

Definition at line 7970 of file gnunet-service-transport.c.

{
  struct SignDhpCls *sign_dhp_cls = cls;
  struct VirtualLink *vl;
  struct DVPathEntryP *dhops = sign_dhp_cls->dhops;
  uint16_t nhops = sign_dhp_cls->nhops;
  const struct GNUNET_PeerIdentity *next_hop = sign_dhp_cls->next_hop;
  struct TransportDVLearnMessage *fwd = sign_dhp_cls->fwd;
  struct Neighbour *n;
 
  sign_dhp_cls->pr->op = NULL;
  GNUNET_CONTAINER_DLL_remove (pils_requests_head,
                               pils_requests_tail,
                               sign_dhp_cls->pr);
  GNUNET_free (sign_dhp_cls->pr);
  dhops[nhops].hop_sig = *sig;
 
  /*route_control_message_without_fc (next_hop,
                                    &fwd->header,
                                    RMO_UNCONFIRMED_ALLOWED);*/
  vl = lookup_virtual_link (next_hop);
  if ((NULL != vl) && (GNUNET_YES == vl->confirmed))
  {
    route_control_message_without_fc (vl,
                                      &fwd->header,
                                      RMO_UNCONFIRMED_ALLOWED);
  }
  else
  {
    /* Use route via neighbour */
    n = lookup_neighbour (next_hop);
    if (NULL != n)
      route_via_neighbour (
        n,
        &fwd->header,
        RMO_UNCONFIRMED_ALLOWED);
  }
  GNUNET_free (sign_dhp_cls);
}

References VirtualLink::confirmed, SignDhpCls::dhops, SignDhpCls::fwd, GNUNET_CONTAINER_DLL_remove, GNUNET_free, GNUNET_YES, TransportDVLearnMessage::header, DVPathEntryP::hop_sig, lookup_neighbour(), lookup_virtual_link(), SignDhpCls::next_hop, SignDhpCls::nhops, PilsRequest::op, pils_requests_head, pils_requests_tail, SignDhpCls::pr, RMO_UNCONFIRMED_ALLOWED, route_control_message_without_fc(), route_via_neighbour(), and Neighbour::vl.

Referenced by forward_dv_learn().

Here is the call graph for this function:

Here is the caller graph for this function:

forward_dv_learn()

static void forward_dv_learn ( const struct GNUNET_PeerIdentity *  next_hop, const struct TransportDVLearnMessage *  msg, uint16_t  bi_history, uint16_t  nhops, const struct DVPathEntryP *  hops, struct GNUNET_TIME_Absolute  in_time  )

static

Build and forward a DV learn message to next_hop.

Parameters

next_hop	peer to send the message to
msg	message received
bi_history	bitmask specifying hops on path that were bidirectional
nhops	length of the hops array
hops	path the message traversed so far
in_time	when did we receive the message, used to calculate network delay

Definition at line 8025 of file gnunet-service-transport.c.

{
  struct DVPathEntryP *dhops;
  char buf[sizeof(struct TransportDVLearnMessage)
           + (nhops + 1) * sizeof(struct DVPathEntryP)] GNUNET_ALIGN;
  struct TransportDVLearnMessage *fwd = (struct TransportDVLearnMessage *) buf;
  struct GNUNET_TIME_Relative nnd;
 
  /* compute message for forwarding */
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Forwarding DV learn message originating from %s to %s\n",
              GNUNET_i2s (&msg->initiator),
              GNUNET_i2s2 (next_hop));
  GNUNET_assert (nhops < MAX_DV_HOPS_ALLOWED);
  fwd->header.type = htons (GNUNET_MESSAGE_TYPE_TRANSPORT_DV_LEARN);
  fwd->header.size = htons (sizeof(struct TransportDVLearnMessage)
                            + (nhops + 1) * sizeof(struct DVPathEntryP));
  fwd->num_hops = htons (nhops + 1);
  fwd->bidirectional = htons (bi_history);
  nnd = GNUNET_TIME_relative_add (GNUNET_TIME_absolute_get_duration (in_time),
                                  GNUNET_TIME_relative_ntoh (
                                    msg->non_network_delay));
  fwd->non_network_delay = GNUNET_TIME_relative_hton (nnd);
  fwd->init_sig = msg->init_sig;
  fwd->initiator = msg->initiator;
  fwd->challenge = msg->challenge;
  fwd->monotonic_time = msg->monotonic_time;
  dhops = (struct DVPathEntryP *) &fwd[1];
  GNUNET_memcpy (dhops, hops, sizeof(struct DVPathEntryP) * nhops);
  dhops[nhops].hop = *GST_my_identity;
  {
    struct DvHopPS dhp = {
      .purpose.purpose = htonl (GNUNET_SIGNATURE_PURPOSE_TRANSPORT_DV_HOP),
      .purpose.size = htonl (sizeof(dhp)),
      .pred = (0 == nhops) ? msg->initiator : dhops[nhops - 1].hop,
      .succ = *next_hop,
      .challenge = msg->challenge
    };
    struct SignDhpCls *sign_dhp_cls = GNUNET_new (struct SignDhpCls);
    sign_dhp_cls->dhops = dhops;
    sign_dhp_cls->nhops = nhops;
    sign_dhp_cls->next_hop = next_hop;
    sign_dhp_cls->fwd = fwd;
    sign_dhp_cls->pr = GNUNET_new (struct PilsRequest);
    GNUNET_CONTAINER_DLL_insert (pils_requests_head,
                                 pils_requests_tail,
                                 sign_dhp_cls->pr);
    sign_dhp_cls->pr->op =
      GNUNET_PILS_sign_by_peer_identity (pils,
                                         &dhp.purpose,
                                         sign_dhp_cp,
                                         sign_dhp_cls);
  }
}

References TransportDVLearnMessage::bidirectional, TransportDVLearnMessage::challenge, SignDhpCls::dhops, SignDhpCls::fwd, GNUNET_ALIGN, GNUNET_assert, GNUNET_CONTAINER_DLL_insert, GNUNET_ERROR_TYPE_DEBUG, GNUNET_i2s(), GNUNET_i2s2(), GNUNET_log, GNUNET_memcpy, GNUNET_MESSAGE_TYPE_TRANSPORT_DV_LEARN, GNUNET_new, GNUNET_PILS_sign_by_peer_identity(), GNUNET_SIGNATURE_PURPOSE_TRANSPORT_DV_HOP, GNUNET_TIME_absolute_get_duration(), GNUNET_TIME_relative_add(), GNUNET_TIME_relative_hton(), GNUNET_TIME_relative_ntoh(), GST_my_identity, TransportDVLearnMessage::header, DVPathEntryP::hop, TransportDVLearnMessage::init_sig, TransportDVLearnMessage::initiator, MAX_DV_HOPS_ALLOWED, TransportDVLearnMessage::monotonic_time, msg, SignDhpCls::next_hop, SignDhpCls::nhops, TransportDVLearnMessage::non_network_delay, TransportDVLearnMessage::num_hops, PilsRequest::op, pils, pils_requests_head, pils_requests_tail, SignDhpCls::pr, GNUNET_CRYPTO_EccSignaturePurpose::purpose, DvHopPS::purpose, sign_dhp_cp(), GNUNET_MessageHeader::size, and GNUNET_MessageHeader::type.

Referenced by dv_neighbour_transmission(), and handle_dv_learn().

Here is the call graph for this function:

Here is the caller graph for this function:

validate_dv_initiator_signature()

static int validate_dv_initiator_signature ( struct GNUNET_TIME_AbsoluteNBO  sender_monotonic_time, const struct GNUNET_PeerIdentity *  init, const struct GNUNET_CRYPTO_ChallengeNonceP *  challenge, const struct GNUNET_CRYPTO_EddsaSignature *  init_sig  )

static

Check signature of type GNUNET_SIGNATURE_PURPOSE_TRANSPORT_DV_INITIATOR.

Parameters

sender_monotonic_time	monotonic time of the initiator
init	the signer
challenge	the challenge that was signed
init_sig	signature presumably by init

Returns

GNUNET_OK if the signature is valid

Definition at line 8096 of file gnunet-service-transport.c.

{
  struct DvInitPS ip = { .purpose.purpose = htonl (
                           GNUNET_SIGNATURE_PURPOSE_TRANSPORT_DV_INITIATOR),
                         .purpose.size = htonl (sizeof(ip)),
                         .monotonic_time = sender_monotonic_time,
                         .challenge = *challenge };
 
  if (
    GNUNET_OK !=
    GNUNET_CRYPTO_eddsa_verify (GNUNET_SIGNATURE_PURPOSE_TRANSPORT_DV_INITIATOR,
                                &ip,
                                init_sig,
                                &init->public_key))
  {
    GNUNET_break_op (0);
    return GNUNET_SYSERR;
  }
  return GNUNET_OK;
}

References DvInitPS::challenge, GNUNET_break_op, GNUNET_CRYPTO_eddsa_verify, GNUNET_OK, GNUNET_SIGNATURE_PURPOSE_TRANSPORT_DV_INITIATOR, GNUNET_SYSERR, init, GNUNET_CRYPTO_EccSignaturePurpose::purpose, and DvInitPS::purpose.

Referenced by handle_dv_learn().

Here is the caller graph for this function:

dv_neighbour_selection()

static int dv_neighbour_selection ( void *  cls, const struct GNUNET_PeerIdentity *  pid, void *  value  )

static

Function called for each neighbour during handle_dv_learn.

Parameters

cls	a struct NeighbourSelectionContext *
pid	identity of the peer
value	a struct Neighbour

Returns

GNUNET_YES (always)

Definition at line 8178 of file gnunet-service-transport.c.

{
  struct NeighbourSelectionContext *nsc = cls;
 
  (void) value;
  if (0 == GNUNET_memcmp (pid, &nsc->dvl->initiator))
    return GNUNET_YES; /* skip initiator */
  for (unsigned int i = 0; i < nsc->nhops; i++)
    if (0 == GNUNET_memcmp (pid, &nsc->hops[i].hop))
      return GNUNET_YES;
  /* skip peers on path */
  nsc->num_eligible++;
  return GNUNET_YES;
}

References NeighbourSelectionContext::dvl, GNUNET_memcmp, GNUNET_YES, DVPathEntryP::hop, NeighbourSelectionContext::hops, TransportDVLearnMessage::initiator, NeighbourSelectionContext::nhops, NeighbourSelectionContext::num_eligible, pid, and value.

Referenced by handle_dv_learn().

Here is the caller graph for this function:

dv_neighbour_transmission()

static int dv_neighbour_transmission ( void *  cls, const struct GNUNET_PeerIdentity *  pid, void *  value  )

static

Function called for each neighbour during handle_dv_learn.

We call forward_dv_learn() on the neighbour(s) selected during dv_neighbour_selection().

Parameters

cls	a struct NeighbourSelectionContext *
pid	identity of the peer
value	a struct Neighbour

Returns

GNUNET_YES (always)

Definition at line 8207 of file gnunet-service-transport.c.

{
  struct NeighbourSelectionContext *nsc = cls;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "transmission %s\n",
              GNUNET_i2s (pid));
  (void) value;
  if (0 == GNUNET_memcmp (pid, &nsc->dvl->initiator))
    return GNUNET_YES; /* skip initiator */
  for (unsigned int i = 0; i < nsc->nhops; i++)
    if (0 == GNUNET_memcmp (pid, &nsc->hops[i].hop))
      return GNUNET_YES;
  /* skip peers on path */
  for (unsigned int i = 0; i < nsc->num_selections; i++)
  {
    if (nsc->selections[i] == nsc->num_eligible)
    {
      forward_dv_learn (pid,
                        nsc->dvl,
                        nsc->bi_history,
                        nsc->nhops,
                        nsc->hops,
                        nsc->in_time);
      break;
    }
  }
  nsc->num_eligible++;
  return GNUNET_YES;
}

References NeighbourSelectionContext::bi_history, NeighbourSelectionContext::dvl, forward_dv_learn(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_i2s(), GNUNET_log, GNUNET_memcmp, GNUNET_YES, DVPathEntryP::hop, NeighbourSelectionContext::hops, NeighbourSelectionContext::in_time, TransportDVLearnMessage::initiator, NeighbourSelectionContext::nhops, NeighbourSelectionContext::num_eligible, NeighbourSelectionContext::num_selections, pid, NeighbourSelectionContext::selections, and value.

Referenced by handle_dv_learn().

Here is the call graph for this function:

Here is the caller graph for this function:

calculate_fork_degree()

static unsigned int calculate_fork_degree ( unsigned int  hops_taken, unsigned int  neighbour_count, unsigned int  eligible_count  )

static

Computes the number of neighbours we should forward a DVInit message to given that it has so far taken hops_taken hops though the network and that the number of neighbours we have in total is neighbour_count, out of which eligible_count are not yet on the path.

NOTE: technically we might want to include NSE in the formula to get a better grip on the overall network size. However, for now using NSE here would create a dependency issue in the build system. => Left for later, hardcoded to 50 for now.

The goal of the formula is that we want to reach a total of LOG(NSE) peers via DV (target_total). We want the reach to be spread out over various distances to the origin, with a bias towards shorter distances.

We make the strong assumption that the network topology looks "similar" at other hops, in particular the neighbour_count should be comparable at other hops.

If the local neighbourhood is densely connected, we expect that eligible_count is close to neighbour_count minus hops_taken as a lot of the path is already known. In that case, we should forward to few(er) peers to try to find a path out of the neighbourhood. OTOH, if eligible_count is close to neighbour_count, we should forward to many peers as we are either still close to the origin (i.e. hops_taken is small) or because we managed to get beyond a local cluster. We express this as the boost_factor using the square of the fraction of eligible neighbours (so if only 50% are eligible, we boost by 1/4, but if 99% are eligible, the 'boost' will be almost 1).

Second, the more hops we have taken, the larger the problem of an exponential traffic explosion gets. So we take the target_total, and compute our degree such that at each distance d 2^{-d} peers are selected (corrected by the boost_factor).

Parameters

hops_taken	number of hops DVInit has travelled so far
neighbour_count	number of neighbours we have in total
eligible_count	number of neighbours we could in theory forward to

Definition at line 8285 of file gnunet-service-transport.c.

{
  double target_total = 50.0; /* FIXME: use LOG(NSE)? */
  double eligible_ratio =
    ((double) eligible_count) / ((double) neighbour_count);
  double boost_factor = eligible_ratio * eligible_ratio;
  unsigned int rnd;
  double left;
 
  if (hops_taken >= 64)
  {
    GNUNET_break (0);
    return 0;   /* precaution given bitshift below */
  }
  for (unsigned int i = 1; i < hops_taken; i++)
  {
    /* For each hop, subtract the expected number of targets
       reached at distance d (so what remains divided by 2^d) */
    target_total -= (target_total * boost_factor / (1LLU << i));
  }
  rnd =
    (unsigned int) floor (target_total * boost_factor / (1LLU << hops_taken));
  /* round up or down probabilistically depending on how close we were
     when floor()ing to rnd */
  left = target_total - (double) rnd;
  if (UINT32_MAX * left >
      GNUNET_CRYPTO_random_u64 (GNUNET_CRYPTO_QUALITY_WEAK, UINT32_MAX))
    rnd++; /* round up */
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Forwarding DV learn message of %u hops %u(/%u/%u) times\n",
              hops_taken,
              rnd,
              eligible_count,
              neighbour_count);
  return rnd;
}

References GNUNET_break, GNUNET_CRYPTO_QUALITY_WEAK, GNUNET_CRYPTO_random_u64(), GNUNET_ERROR_TYPE_DEBUG, and GNUNET_log.

Referenced by handle_dv_learn().

Here is the call graph for this function:

Here is the caller graph for this function:

neighbour_store_dvmono_cb()

static void neighbour_store_dvmono_cb ( void *  cls, int  success  )

static

Function called when peerstore is done storing a DV monotonic time.

Parameters

cls	a struct Neighbour
success	GNUNET_YES if peerstore was successful

Definition at line 8332 of file gnunet-service-transport.c.

{
  struct Neighbour *n = cls;
 
  n->sc = NULL;
  if (GNUNET_YES != success)
    GNUNET_log (GNUNET_ERROR_TYPE_ERROR,
                "Failed to store other peer's monotonic time in peerstore!\n");
}

References GNUNET_ERROR_TYPE_ERROR, GNUNET_log, GNUNET_YES, and Neighbour::sc.

Referenced by handle_dv_learn().

Here is the caller graph for this function:

get_network_latency()

static struct GNUNET_TIME_Relative get_network_latency ( const struct TransportDVLearnMessage *  dvl )

static

Definition at line 8344 of file gnunet-service-transport.c.

{
  struct GNUNET_TIME_Relative host_latency_sum;
  struct GNUNET_TIME_Relative latency;
  struct GNUNET_TIME_Relative network_latency;
  uint16_t nhops = ntohs (dvl->num_hops);;
 
  /* We initiated this, learn the forward path! */
  host_latency_sum = GNUNET_TIME_relative_ntoh (dvl->non_network_delay);
 
  // Need also something to lookup initiation time
  // to compute RTT! -> add RTT argument here?
  latency = GNUNET_TIME_absolute_get_duration (GNUNET_TIME_absolute_ntoh (
                                                 dvl->monotonic_time));
  GNUNET_assert (latency.rel_value_us >= host_latency_sum.rel_value_us);
  // latency = GNUNET_TIME_UNIT_FOREVER_REL;   // FIXME: initialize properly
  // (based on dvl->challenge, we can identify time of origin!)
 
  network_latency = GNUNET_TIME_relative_subtract (latency, host_latency_sum);
  /* assumption: latency on all links is the same */
  network_latency = GNUNET_TIME_relative_divide (network_latency, nhops);
 
  return network_latency;
}

References GNUNET_assert, GNUNET_TIME_absolute_get_duration(), GNUNET_TIME_absolute_ntoh(), GNUNET_TIME_relative_divide(), GNUNET_TIME_relative_ntoh(), GNUNET_TIME_relative_subtract(), and GNUNET_TIME_Relative::rel_value_us.

Referenced by handle_dv_learn().

Here is the call graph for this function:

Here is the caller graph for this function:

handle_dv_learn()

static void handle_dv_learn ( void *  cls, const struct TransportDVLearnMessage *  dvl  )

static

Communicator gave us a DV learn message.

Process the request.

Parameters

cls	a struct CommunicatorMessageContext (must call finish_cmc_handling() when done)
dvl	the message that was received

Definition at line 8378 of file gnunet-service-transport.c.

{
  struct CommunicatorMessageContext *cmc = cls;
  enum GNUNET_TRANSPORT_CommunicatorCharacteristics cc;
  int bi_hop;
  uint16_t nhops;
  uint16_t bi_history;
  const struct DVPathEntryP *hops;
  int do_fwd;
  int did_initiator;
  struct GNUNET_TIME_Absolute in_time;
  struct Neighbour *n;
 
  nhops = ntohs (dvl->num_hops);  /* 0 = sender is initiator */
  bi_history = ntohs (dvl->bidirectional);
  hops = (const struct DVPathEntryP *) &dvl[1];
  if (0 == nhops)
  {
    /* sanity check */
    if (0 != GNUNET_memcmp (&dvl->initiator, &cmc->im.sender))
    {
      GNUNET_break (0);
      finish_cmc_handling (cmc);
      return;
    }
  }
  else
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "handle dv learn message last hop %s\n",
                GNUNET_i2s (&hops[nhops - 1].hop));
    /* sanity check */
    if (0 != GNUNET_memcmp (&hops[nhops - 1].hop, &cmc->im.sender))
    {
      GNUNET_break (0);
      finish_cmc_handling (cmc);
      return;
    }
  }
 
  GNUNET_assert (CT_COMMUNICATOR == cmc->tc->type);
  cc = cmc->tc->details.communicator.cc;
  bi_hop = (GNUNET_TRANSPORT_CC_RELIABLE ==
            cc); // FIXME: add bi-directional flag to cc?
  in_time = GNUNET_TIME_absolute_get ();
 
  /* continue communicator here, everything else can happen asynchronous! */
  finish_cmc_handling (cmc);
 
  n = lookup_neighbour (&dvl->initiator);
  if (NULL != n)
  {
    if ((n->dv_monotime_available == GNUNET_YES) &&
        (GNUNET_TIME_absolute_ntoh (dvl->monotonic_time).abs_value_us <
         n->last_dv_learn_monotime.abs_value_us))
    {
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "DV learn from %s discarded due to time travel",
                  GNUNET_i2s (&dvl->initiator));
      GNUNET_STATISTICS_update (GST_stats,
                                "# DV learn discarded due to time travel",
                                1,
                                GNUNET_NO);
      return;
    }
    if (GNUNET_OK != validate_dv_initiator_signature (dvl->monotonic_time,
                                                      &dvl->initiator,
                                                      &dvl->challenge,
                                                      &dvl->init_sig))
    {
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "DV learn signature from %s invalid\n",
                  GNUNET_i2s (&dvl->initiator));
      GNUNET_break_op (0);
      return;
    }
    n->last_dv_learn_monotime = GNUNET_TIME_absolute_ntoh (dvl->monotonic_time);
    if (GNUNET_YES == n->dv_monotime_available)
    {
      if (NULL != n->sc)
      {
        GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                    "store cancel\n");
        GNUNET_PEERSTORE_store_cancel (n->sc);
      }
      n->sc =
        GNUNET_PEERSTORE_store (peerstore,
                                "transport",
                                &dvl->initiator,
                                GNUNET_PEERSTORE_TRANSPORT_DVLEARN_MONOTIME,
                                &dvl->monotonic_time,
                                sizeof(dvl->monotonic_time),
                                GNUNET_TIME_UNIT_FOREVER_ABS,
                                GNUNET_PEERSTORE_STOREOPTION_REPLACE,
                                &neighbour_store_dvmono_cb,
                                n);
    }
  }
  /* OPTIMIZE-FIXME: asynchronously (!) verify signatures!,
     If signature verification load too high, implement random drop strategy */
  for (unsigned int i = 0; i < nhops; i++)
  {
    struct DvHopPS dhp = { .purpose.purpose =
                             htonl (GNUNET_SIGNATURE_PURPOSE_TRANSPORT_DV_HOP),
                           .purpose.size = htonl (sizeof(dhp)),
                           .pred = (0 == i) ? dvl->initiator : hops[i - 1].hop,
                           .succ = (nhops == i + 1) ? *GST_my_identity
                                   : hops[i + 1].hop,
                           .challenge = dvl->challenge };
 
    if (GNUNET_OK !=
        GNUNET_CRYPTO_eddsa_verify (GNUNET_SIGNATURE_PURPOSE_TRANSPORT_DV_HOP,
                                    &dhp,
                                    &hops[i].hop_sig,
                                    &hops[i].hop.public_key))
    {
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "DV learn from %s signature of hop %u invalid\n",
                  GNUNET_i2s (&dvl->initiator),
                  i);
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "signature of hop %s invalid\n",
                  GNUNET_i2s (&hops[i].hop));
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "pred %s\n",
                  GNUNET_i2s (&dhp.pred));
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "succ %s\n",
                  GNUNET_i2s (&dhp.succ));
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "hash %s\n",
                  GNUNET_sh2s (&dhp.challenge.value));
      GNUNET_break_op (0);
      return;
    }
  }
  if (GNUNET_EXTRA_LOGGING > 0)
  {
    char *path;
 
    path = GNUNET_strdup (GNUNET_i2s (&dvl->initiator));
    for (unsigned int i = 0; i < nhops; i++)
    {
      char *tmp;
 
      GNUNET_asprintf (&tmp,
                       "%s%s%s",
                       path,
                       (bi_history & (1 << (nhops - i))) ? "<->" : "-->",
                       GNUNET_i2s (&hops[i].hop));
      GNUNET_free (path);
      path = tmp;
    }
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Received DVInit via %s%s%s\n",
                path,
                bi_hop ? "<->" : "-->",
                GNUNET_i2s (GST_my_identity));
    GNUNET_free (path);
  }
  do_fwd = GNUNET_YES;
  if (0 == GNUNET_memcmp (GST_my_identity, &dvl->initiator))
  {
    struct GNUNET_PeerIdentity path[nhops + 1];
    struct GNUNET_TIME_Relative network_latency;
 
    /* We initiated this, learn the forward path! */
    path[0] = *GST_my_identity;
    path[1] = hops[0].hop;
 
    network_latency = get_network_latency (dvl);
 
    for (unsigned int i = 2; i <= nhops; i++)
    {
      struct GNUNET_TIME_Relative ilat;
 
      /* assumption: linear latency increase per hop */
      ilat = GNUNET_TIME_relative_multiply (network_latency, i);
      path[i] = hops[i - 1].hop;
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "Learned path with %u hops to %s with latency %s\n",
                  i,
                  GNUNET_i2s (&path[i]),
                  GNUNET_STRINGS_relative_time_to_string (ilat, GNUNET_YES));
      learn_dv_path (path,
                     i + 1,
                     ilat,
                     GNUNET_TIME_relative_to_absolute (
                       ADDRESS_VALIDATION_LIFETIME));
    }
    /* as we initiated, do not forward again (would be circular!) */
    do_fwd = GNUNET_NO;
    return;
  }
  if (bi_hop)
  {
    /* last hop was bi-directional, we could learn something here! */
    struct GNUNET_PeerIdentity path[nhops + 2];
    struct GNUNET_TIME_Relative ilat;
    struct GNUNET_TIME_Relative network_latency;
 
    path[0] = *GST_my_identity;
    path[1] = hops[nhops - 1].hop;   /* direct neighbour == predecessor! */
    for (unsigned int i = 0; i < nhops; i++)
    {
      int iret;
 
      if (0 == (bi_history & (1 << i)))
        break;     /* i-th hop not bi-directional, stop learning! */
      if (i == nhops - 1)
      {
        path[i + 2] = dvl->initiator;
      }
      else
      {
        path[i + 2] = hops[nhops - i - 2].hop;
      }
 
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "Learned inverse path with %u hops to %s\n",
                  i + 2,
                  GNUNET_i2s (&path[i + 2]));
      network_latency = get_network_latency (dvl);
      ilat = GNUNET_TIME_relative_multiply (network_latency, i + 2);
      iret = learn_dv_path (path,
                            i + 3,
                            ilat,
                            GNUNET_TIME_relative_to_absolute (
                              ADDRESS_VALIDATION_LIFETIME));
      if (GNUNET_SYSERR == iret)
      {
        /* path invalid or too long to be interesting for US, thus should also
           not be interesting to our neighbours, cut path when forwarding to
           'i' hops, except of course for the one that goes back to the
           initiator */
        GNUNET_STATISTICS_update (GST_stats,
                                  "# DV learn not forwarded due invalidity of path",
                                  1,
                                  GNUNET_NO);
        do_fwd = GNUNET_NO;
        break;
      }
      if ((GNUNET_NO == iret) && (nhops == i + 1))
      {
        /* we have better paths, and this is the longest target,
           so there cannot be anything interesting later */
        GNUNET_STATISTICS_update (GST_stats,
                                  "# DV learn not forwarded, got better paths",
                                  1,
                                  GNUNET_NO);
        do_fwd = GNUNET_NO;
        break;
      }
    }
  }
  if (MAX_DV_HOPS_ALLOWED == nhops)
  {
    /* At limit, we're out of here! */
    return;
  }
 
  /* Forward to initiator, if path non-trivial and possible */
  bi_history = (bi_history << 1) | (bi_hop ? 1 : 0);
  did_initiator = GNUNET_NO;
  if ((1 <= nhops) &&
      (GNUNET_YES ==
       GNUNET_CONTAINER_multipeermap_contains (neighbours, &dvl->initiator)))
  {
    /* send back to origin! */
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Sending DVL back to initiator %s\n",
                GNUNET_i2s (&dvl->initiator));
    forward_dv_learn (&dvl->initiator, dvl, bi_history, nhops, hops, in_time);
    did_initiator = GNUNET_YES;
  }
  /* We forward under two conditions: either we still learned something
     ourselves (do_fwd), or the path was darn short and thus the initiator is
     likely to still be very interested in this (and we did NOT already
     send it back to the initiator) */
  if ((do_fwd) || ((nhops < MIN_DV_PATH_LENGTH_FOR_INITIATOR) &&
                   (GNUNET_NO == did_initiator)))
  {
    /* Pick random neighbours that are not yet on the path */
    struct NeighbourSelectionContext nsc;
    unsigned int n_cnt;
 
    n_cnt = GNUNET_CONTAINER_multipeermap_size (neighbours);
    nsc.nhops = nhops;
    nsc.dvl = dvl;
    nsc.bi_history = bi_history;
    nsc.hops = hops;
    nsc.in_time = in_time;
    nsc.num_eligible = 0;
    GNUNET_CONTAINER_multipeermap_iterate (neighbours,
                                           &dv_neighbour_selection,
                                           &nsc);
    if (0 == nsc.num_eligible)
      return;   /* done here, cannot forward to anyone else */
    nsc.num_selections = calculate_fork_degree (nhops, n_cnt, nsc.num_eligible);
    nsc.num_selections =
      GNUNET_MIN (MAX_DV_DISCOVERY_SELECTION, nsc.num_selections);
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Forwarding DVL to %u other peers\n",
                nsc.num_selections);
    for (unsigned int i = 0; i < nsc.num_selections; i++)
      nsc.selections[i] =
        (nsc.num_selections == n_cnt)
        ? i   /* all were selected, avoid collisions by chance */
        : GNUNET_CRYPTO_random_u32 (GNUNET_CRYPTO_QUALITY_WEAK, n_cnt);
    nsc.num_eligible = 0;
    GNUNET_CONTAINER_multipeermap_iterate (neighbours,
                                           &dv_neighbour_transmission,
                                           &nsc);
  }
}

References GNUNET_TIME_Absolute::abs_value_us, ADDRESS_VALIDATION_LIFETIME, NeighbourSelectionContext::bi_history, TransportDVLearnMessage::bidirectional, calculate_fork_degree(), TransportClient::cc, DvHopPS::challenge, TransportDVLearnMessage::challenge, TransportClient::communicator, CT_COMMUNICATOR, TransportClient::details, Neighbour::dv_monotime_available, dv_neighbour_selection(), dv_neighbour_transmission(), NeighbourSelectionContext::dvl, finish_cmc_handling(), forward_dv_learn(), get_network_latency(), GNUNET_asprintf(), GNUNET_assert, GNUNET_break, GNUNET_break_op, GNUNET_CONTAINER_multipeermap_contains(), GNUNET_CONTAINER_multipeermap_iterate(), GNUNET_CONTAINER_multipeermap_size(), GNUNET_CRYPTO_eddsa_verify, GNUNET_CRYPTO_QUALITY_WEAK, GNUNET_CRYPTO_random_u32(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_EXTRA_LOGGING, GNUNET_free, GNUNET_i2s(), GNUNET_log, GNUNET_memcmp, GNUNET_MIN, GNUNET_NO, GNUNET_OK, GNUNET_PEERSTORE_store(), GNUNET_PEERSTORE_store_cancel(), GNUNET_PEERSTORE_STOREOPTION_REPLACE, GNUNET_PEERSTORE_TRANSPORT_DVLEARN_MONOTIME, GNUNET_sh2s(), GNUNET_SIGNATURE_PURPOSE_TRANSPORT_DV_HOP, GNUNET_STATISTICS_update(), GNUNET_strdup, GNUNET_STRINGS_relative_time_to_string(), GNUNET_SYSERR, GNUNET_TIME_absolute_get(), GNUNET_TIME_absolute_ntoh(), GNUNET_TIME_relative_multiply(), GNUNET_TIME_relative_to_absolute(), GNUNET_TIME_UNIT_FOREVER_ABS, GNUNET_TRANSPORT_CC_RELIABLE, GNUNET_YES, GST_my_identity, GST_stats, DVPathEntryP::hop, NeighbourSelectionContext::hops, CommunicatorMessageContext::im, NeighbourSelectionContext::in_time, TransportDVLearnMessage::init_sig, TransportDVLearnMessage::initiator, Neighbour::last_dv_learn_monotime, learn_dv_path(), lookup_neighbour(), MAX_DV_DISCOVERY_SELECTION, MAX_DV_HOPS_ALLOWED, MIN_DV_PATH_LENGTH_FOR_INITIATOR, TransportDVLearnMessage::monotonic_time, neighbour_store_dvmono_cb(), neighbours, NeighbourSelectionContext::nhops, NeighbourSelectionContext::num_eligible, TransportDVLearnMessage::num_hops, NeighbourSelectionContext::num_selections, peerstore, DvHopPS::pred, GNUNET_CRYPTO_EccSignaturePurpose::purpose, DvHopPS::purpose, Neighbour::sc, NeighbourSelectionContext::selections, GNUNET_TRANSPORT_IncomingMessage::sender, DvHopPS::succ, CommunicatorMessageContext::tc, TransportClient::type, validate_dv_initiator_signature(), and GNUNET_CRYPTO_ChallengeNonceP::value.

Here is the call graph for this function:

check_dv_box()

static int check_dv_box ( void *  cls, const struct TransportDVBoxMessage *  dvb  )

static

Communicator gave us a DV box.

Check the message.

Parameters

cls	a struct CommunicatorMessageContext
dvb	the send message that was sent

Returns

GNUNET_YES if message is well-formed

Definition at line 8703 of file gnunet-service-transport.c.

{
  uint16_t size = ntohs (dvb->header.size);
  uint16_t num_hops = ntohs (dvb->num_hops);
  const struct GNUNET_PeerIdentity *hops =
    (const struct GNUNET_PeerIdentity *) &dvb[1];
 
  (void) cls;
  if (size < sizeof(*dvb) + num_hops * sizeof(struct GNUNET_PeerIdentity)
      + sizeof(struct GNUNET_MessageHeader))
  {
    GNUNET_break_op (0);
    return GNUNET_SYSERR;
  }
  /* This peer must not be on the path */
  for (unsigned int i = 0; i < num_hops; i++)
    if (0 == GNUNET_memcmp (&hops[i], GST_my_identity))
    {
      GNUNET_break_op (0);
      return GNUNET_SYSERR;
    }
  return GNUNET_YES;
}

References GNUNET_break_op, GNUNET_memcmp, GNUNET_SYSERR, GNUNET_YES, GST_my_identity, TransportDVBoxMessage::header, TransportDVBoxMessage::num_hops, GNUNET_MessageHeader::size, and size.

forward_dv_box()

static void forward_dv_box ( struct Neighbour *  next_hop, struct TransportDVBoxMessage *  hdr, uint16_t  total_hops, uint16_t  num_hops, const struct GNUNET_PeerIdentity *  hops, const void *  enc_payload, uint16_t  enc_payload_size  )

static

Create a DV Box message and queue it for transmission to next_hop.

Parameters

next_hop	peer to receive the message next
total_hops	how many hops did the message take so far
num_hops	length of the hops array
origin	origin of the message
hops	next peer(s) to the destination, including destination
payload	payload of the box
payload_size	number of bytes in payload

Definition at line 8741 of file gnunet-service-transport.c.

{
  struct VirtualLink *vl = next_hop->vl;
  struct PendingMessage *pm;
  size_t msg_size = sizeof(struct TransportDVBoxMessage)
                    + num_hops * sizeof(struct GNUNET_PeerIdentity)
                    + enc_payload_size;
  char *buf;
  char msg_buf[msg_size] GNUNET_ALIGN;
  struct GNUNET_PeerIdentity *dhops;
 
  hdr->num_hops = htons (num_hops);
  hdr->total_hops = htons (total_hops);
  hdr->header.size = htons (msg_size);
  memcpy (msg_buf, hdr, sizeof(*hdr));
  dhops = (struct GNUNET_PeerIdentity *) &msg_buf[sizeof(struct
                                                         TransportDVBoxMessage)]
  ;
  memcpy (dhops, hops, num_hops * sizeof(struct GNUNET_PeerIdentity));
  memcpy (&dhops[num_hops], enc_payload, enc_payload_size);
 
  if (GNUNET_YES == ntohs (hdr->without_fc))
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Forwarding control message (payload size %u) in DV Box to next hop %s (%u/%u) \n",
                enc_payload_size,
                GNUNET_i2s (&next_hop->pid),
                (unsigned int) num_hops,
                (unsigned int) total_hops);
    route_via_neighbour (next_hop, (const struct
                                    GNUNET_MessageHeader *) msg_buf,
                         RMO_ANYTHING_GOES);
  }
  else
  {
    pm = GNUNET_malloc (sizeof(struct PendingMessage) + msg_size);
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "2 created pm %p storing vl %p \n",
                pm,
                vl);
    pm->pmt = PMT_DV_BOX;
    pm->vl = vl;
    pm->target = next_hop->pid;
    pm->timeout = GNUNET_TIME_relative_to_absolute (DV_FORWARD_TIMEOUT);
    pm->logging_uuid = logging_uuid_gen++;
    pm->prefs = GNUNET_MQ_PRIO_BACKGROUND;
    pm->bytes_msg = msg_size;
    buf = (char *) &pm[1];
    memcpy (buf, msg_buf, msg_size);
 
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Created pending message %" PRIu64
                " for DV Box with next hop %s (%u/%u)\n",
                pm->logging_uuid,
                GNUNET_i2s (&next_hop->pid),
                (unsigned int) num_hops,
                (unsigned int) total_hops);
 
    if ((NULL != vl) && (GNUNET_YES == vl->confirmed))
    {
      GNUNET_CONTAINER_MDLL_insert (vl,
                                    vl->pending_msg_head,
                                    vl->pending_msg_tail,
                                    pm);
 
      check_vl_transmission (vl);
    }
    else
    {
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "The virtual link is not ready for forwarding a DV Box with payload, storing PendingMessage in ring buffer.\n");
 
      if (NULL != ring_buffer_dv[ring_buffer_dv_head])
      {
        struct PendingMessage *pm_old = ring_buffer_dv[ring_buffer_dv_head];
 
        GNUNET_free (pm_old);
      }
      ring_buffer_dv[ring_buffer_dv_head] = pm;
      if (RING_BUFFER_SIZE - 1 == ring_buffer_dv_head)
      {
        ring_buffer_dv_head = 0;
        is_ring_buffer_dv_full = GNUNET_YES;
      }
      else
        ring_buffer_dv_head++;
 
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "%u items stored in DV ring buffer\n",
                  GNUNET_YES == is_ring_buffer_dv_full ? RING_BUFFER_SIZE :
                  ring_buffer_dv_head);
    }
  }
}

References check_vl_transmission(), VirtualLink::confirmed, DV_FORWARD_TIMEOUT, GNUNET_ALIGN, GNUNET_CONTAINER_MDLL_insert, GNUNET_ERROR_TYPE_DEBUG, GNUNET_free, GNUNET_i2s(), GNUNET_log, GNUNET_malloc, GNUNET_MQ_PRIO_BACKGROUND, GNUNET_TIME_relative_to_absolute(), GNUNET_YES, TransportDVBoxMessage::header, is_ring_buffer_dv_full, logging_uuid_gen, TransportDVBoxMessage::num_hops, VirtualLink::pending_msg_head, VirtualLink::pending_msg_tail, Neighbour::pid, pm, PMT_DV_BOX, ring_buffer_dv, ring_buffer_dv_head, RING_BUFFER_SIZE, RMO_ANYTHING_GOES, route_via_neighbour(), GNUNET_MessageHeader::size, VirtualLink::target, TransportDVBoxMessage::total_hops, Neighbour::vl, and TransportDVBoxMessage::without_fc.

Referenced by handle_dv_box().

Here is the call graph for this function:

Here is the caller graph for this function:

free_backtalker()

static void free_backtalker ( struct Backtalker *  b )

static

Free data structures associated with b.

Parameters

b	data structure to release

Definition at line 8849 of file gnunet-service-transport.c.

{
  if (NULL != b->get)
  {
    GNUNET_PEERSTORE_iteration_stop (b->get);
    b->get = NULL;
    GNUNET_assert (NULL != b->cmc);
    finish_cmc_handling (b->cmc);
    b->cmc = NULL;
  }
  if (NULL != b->task)
  {
    GNUNET_SCHEDULER_cancel (b->task);
    b->task = NULL;
  }
  if (NULL != b->sc)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "store cancel\n");
    GNUNET_PEERSTORE_store_cancel (b->sc);
    b->sc = NULL;
  }
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Removing backtalker for %s\n",
              GNUNET_i2s (&b->pid));
  GNUNET_assert (
    GNUNET_YES ==
    GNUNET_CONTAINER_multipeermap_remove (backtalkers, &b->pid, b));
  GNUNET_free (b);
}

References backtalkers, Backtalker::cmc, finish_cmc_handling(), Backtalker::get, GNUNET_assert, GNUNET_CONTAINER_multipeermap_remove(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_free, GNUNET_i2s(), GNUNET_log, GNUNET_PEERSTORE_iteration_stop(), GNUNET_PEERSTORE_store_cancel(), GNUNET_SCHEDULER_cancel(), GNUNET_YES, Backtalker::pid, Backtalker::sc, and Backtalker::task.

Referenced by backtalker_timeout_cb(), and free_backtalker_cb().

Here is the call graph for this function:

Here is the caller graph for this function:

free_backtalker_cb()

static int free_backtalker_cb ( void *  cls, const struct GNUNET_PeerIdentity *  pid, void *  value  )

static

Callback to free backtalker records.

Parameters

cls	NULL
pid	unused
value	a struct Backtalker

Returns

GNUNET_OK (always)

Definition at line 8890 of file gnunet-service-transport.c.

{
  struct Backtalker *b = value;
 
  (void) cls;
  (void) pid;
  free_backtalker (b);
  return GNUNET_OK;
}

References free_backtalker(), GNUNET_OK, pid, and value.

Referenced by do_shutdown().

Here is the call graph for this function:

Here is the caller graph for this function:

backtalker_timeout_cb()

static void backtalker_timeout_cb ( void *  cls )

static

Function called when it is time to clean up a backtalker.

Parameters

cls	a struct Backtalker

Definition at line 8909 of file gnunet-service-transport.c.

{
  struct Backtalker *b = cls;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "backtalker timeout.\n");
  b->task = NULL;
  if (0 != GNUNET_TIME_absolute_get_remaining (b->timeout).rel_value_us)
  {
    b->task = GNUNET_SCHEDULER_add_at (b->timeout, &backtalker_timeout_cb, b);
    return;
  }
  GNUNET_assert (NULL == b->sc);
  free_backtalker (b);
}

References backtalker_timeout_cb(), free_backtalker(), GNUNET_assert, GNUNET_ERROR_TYPE_DEBUG, GNUNET_log, GNUNET_SCHEDULER_add_at(), GNUNET_TIME_absolute_get_remaining(), GNUNET_TIME_Relative::rel_value_us, Backtalker::sc, Backtalker::task, and Backtalker::timeout.

Referenced by backtalker_monotime_store_cb(), backtalker_timeout_cb(), and decaps_dv_box_cb().

Here is the call graph for this function:

Here is the caller graph for this function:

backtalker_monotime_cb()

static void backtalker_monotime_cb ( void *  cls, const struct GNUNET_PEERSTORE_Record *  record, const char *  emsg  )

static

Function called with the monotonic time of a backtalker by PEERSTORE.

Updates the time and continues processing.

Parameters

cls	a struct Backtalker
record	the information found, NULL for the last call
emsg	error message

Definition at line 8935 of file gnunet-service-transport.c.

{
  struct Backtalker *b = cls;
  struct GNUNET_TIME_AbsoluteNBO *mtbe;
  struct GNUNET_TIME_Absolute mt;
 
  (void) emsg;
  if (NULL == record)
  {
    /* we're done with #backtalker_monotime_cb() invocations,
       continue normal processing */
    b->get = NULL;
    GNUNET_assert (NULL != b->cmc);
    b->cmc->mh = (const struct GNUNET_MessageHeader *) &b[1];
    if (0 != b->body_size)
      demultiplex_with_cmc (b->cmc);
    else
      finish_cmc_handling (b->cmc);
    b->cmc = NULL;
    return;
  }
  if (sizeof(*mtbe) != record->value_size)
  {
    GNUNET_PEERSTORE_iteration_next (b->get, 1);
    GNUNET_break (0);
    return;
  }
  mtbe = record->value;
  mt = GNUNET_TIME_absolute_ntoh (*mtbe);
  if (mt.abs_value_us > b->monotonic_time.abs_value_us)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Backtalker message from %s dropped, monotime in the past\n",
                GNUNET_i2s (&b->pid));
    GNUNET_STATISTICS_update (
      GST_stats,
      "# Backchannel messages dropped: monotonic time not increasing",
      1,
      GNUNET_NO);
    b->monotonic_time = mt;
    /* Setting body_size to 0 prevents call to #forward_backchannel_payload()
     */
    b->body_size = 0;
  }
  GNUNET_PEERSTORE_iteration_next (b->get, 1);
}

References GNUNET_TIME_Absolute::abs_value_us, Backtalker::body_size, Backtalker::cmc, demultiplex_with_cmc(), finish_cmc_handling(), Backtalker::get, GNUNET_assert, GNUNET_break, GNUNET_ERROR_TYPE_DEBUG, GNUNET_i2s(), GNUNET_log, GNUNET_NO, GNUNET_PEERSTORE_iteration_next(), GNUNET_STATISTICS_update(), GNUNET_TIME_absolute_ntoh(), GST_stats, CommunicatorMessageContext::mh, Backtalker::monotonic_time, Backtalker::pid, and record().

Referenced by decaps_dv_box_cb().

Here is the call graph for this function:

Here is the caller graph for this function:

backtalker_monotime_store_cb()

static void backtalker_monotime_store_cb ( void *  cls, int  success  )

static

Function called by PEERSTORE when the store operation of a backtalker's monotonic time is complete.

Parameters

cls	the struct Backtalker
success	GNUNET_OK on success

Definition at line 8993 of file gnunet-service-transport.c.

{
  struct Backtalker *b = cls;
 
  if (GNUNET_OK != success)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_ERROR,
                "Failed to store backtalker's monotonic time in PEERSTORE!\n");
  }
  b->sc = NULL;
  if (NULL != b->task)
  {
    GNUNET_SCHEDULER_cancel (b->task);
    b->task = NULL;
  }
  b->task = GNUNET_SCHEDULER_add_at (b->timeout, &backtalker_timeout_cb, b);
}

References backtalker_timeout_cb(), GNUNET_ERROR_TYPE_ERROR, GNUNET_log, GNUNET_OK, GNUNET_SCHEDULER_add_at(), GNUNET_SCHEDULER_cancel(), Backtalker::sc, Backtalker::task, and Backtalker::timeout.

Referenced by update_backtalker_monotime().

Here is the call graph for this function:

Here is the caller graph for this function:

update_backtalker_monotime()

static void update_backtalker_monotime ( struct Backtalker *  b )

static

The backtalker b monotonic time changed.

Update PEERSTORE.

Parameters

b	a backtalker with updated monotonic time

Definition at line 9018 of file gnunet-service-transport.c.

{
  struct GNUNET_TIME_AbsoluteNBO mtbe;
 
  if (NULL != b->sc)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "store cancel before store with sc %p\n",
                b->sc);
    /*GNUNET_PEERSTORE_store_cancel (b->sc);
      b->sc = NULL;*/
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "store cancel before store with sc %p is null\n",
                b->sc);
  }
  else
  {
    GNUNET_SCHEDULER_cancel (b->task);
    b->task = NULL;
  }
  mtbe = GNUNET_TIME_absolute_hton (b->monotonic_time);
  b->sc =
    GNUNET_PEERSTORE_store (peerstore,
                            "transport",
                            &b->pid,
                            GNUNET_PEERSTORE_TRANSPORT_BACKCHANNEL_MONOTIME,
                            &mtbe,
                            sizeof(mtbe),
                            GNUNET_TIME_UNIT_FOREVER_ABS,
                            GNUNET_PEERSTORE_STOREOPTION_REPLACE,
                            &backtalker_monotime_store_cb,
                            b);
}

References backtalker_monotime_store_cb(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_log, GNUNET_PEERSTORE_store(), GNUNET_PEERSTORE_STOREOPTION_REPLACE, GNUNET_PEERSTORE_TRANSPORT_BACKCHANNEL_MONOTIME, GNUNET_SCHEDULER_cancel(), GNUNET_TIME_absolute_hton(), GNUNET_TIME_UNIT_FOREVER_ABS, Backtalker::monotonic_time, peerstore, Backtalker::pid, Backtalker::sc, and Backtalker::task.

Referenced by decaps_dv_box_cb().

Here is the call graph for this function:

Here is the caller graph for this function:

decaps_dv_box_cb()

static void decaps_dv_box_cb ( void *  cls, const struct GNUNET_ShortHashCode *  km  )

static

Definition at line 9062 of file gnunet-service-transport.c.

{
  struct DecapsDvBoxCls *decaps_dv_box_cls = cls;
  struct CommunicatorMessageContext *cmc = decaps_dv_box_cls->cmc;
  const struct TransportDVBoxMessage *dvb = dvb;
  struct DVKeyState key;
  const char *hdr;
  size_t hdr_len;
  struct GNUNET_HashCode hmac;
 
  decaps_dv_box_cls->pr->op = NULL;
  GNUNET_CONTAINER_DLL_remove (pils_requests_head,
                               pils_requests_tail,
                               decaps_dv_box_cls->pr);
  GNUNET_free (decaps_dv_box_cls->pr);
  if (NULL == km)
  {
    GNUNET_break_op (0);
    finish_cmc_handling (cmc);
    return;
  }
  dv_setup_key_state_from_km (km, &dvb->iv, &key);
  hdr = (const char *) &dvb[1];
  hdr_len = ntohs (dvb->orig_size) - sizeof(*dvb) - sizeof(struct
                                                           GNUNET_PeerIdentity)
            * ntohs (dvb->total_hops);
 
  dv_hmac (&key, &hmac, hdr, hdr_len);
  if (0 != GNUNET_memcmp (&hmac, &dvb->hmac))
  {
    /* HMAC mismatch, discard! */
    GNUNET_break_op (0);
    finish_cmc_handling (cmc);
    return;
  }
  /* begin actual decryption */
  {
    struct Backtalker *b;
    struct GNUNET_TIME_Absolute monotime;
    struct TransportDVBoxPayloadP ppay;
    char body[hdr_len - sizeof(ppay)] GNUNET_ALIGN;
    const struct GNUNET_MessageHeader *mh;
 
    GNUNET_assert (hdr_len >=
                   sizeof(ppay) + sizeof(struct GNUNET_MessageHeader));
    if (GNUNET_OK != dv_decrypt (&key, &ppay, hdr, sizeof(ppay)))
    {
      GNUNET_log (GNUNET_ERROR_TYPE_ERROR,
                  "Error decrypting DV payload header\n");
      GNUNET_break_op (0);
      finish_cmc_handling (cmc);
      return;
    }
    if (GNUNET_OK != dv_decrypt (&key, body,
                                 &hdr[sizeof(ppay)], hdr_len - sizeof(ppay)))
    {
      GNUNET_log (GNUNET_ERROR_TYPE_ERROR,
                  "Error decrypting DV payload\n");
      GNUNET_break_op (0);
      finish_cmc_handling (cmc);
      return;
    }
    mh = (const struct GNUNET_MessageHeader *) body;
    dv_key_clean (&key);
    if (ntohs (mh->size) != sizeof(body))
    {
      GNUNET_break_op (0);
      finish_cmc_handling (cmc);
      return;
    }
    /* need to prevent box-in-a-box (and DV_LEARN) so check inbox type! */
    switch (ntohs (mh->type))
    {
    case GNUNET_MESSAGE_TYPE_TRANSPORT_DV_BOX:
      GNUNET_break_op (0);
      finish_cmc_handling (cmc);
      return;
 
    case GNUNET_MESSAGE_TYPE_TRANSPORT_DV_LEARN:
      GNUNET_break_op (0);
      finish_cmc_handling (cmc);
      return;
 
    default:
      /* permitted, continue */
      break;
    }
    monotime = GNUNET_TIME_absolute_ntoh (ppay.monotonic_time);
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Decrypted backtalk from %s\n",
                GNUNET_i2s (&ppay.sender));
    b = GNUNET_CONTAINER_multipeermap_get (backtalkers, &ppay.sender);
    if ((NULL != b) && (monotime.abs_value_us < b->monotonic_time.abs_value_us))
    {
      GNUNET_STATISTICS_update (
        GST_stats,
        "# Backchannel messages dropped: monotonic time not increasing",
        1,
        GNUNET_NO);
      finish_cmc_handling (cmc);
      return;
    }
    if ((NULL == b) ||
        (0 != GNUNET_memcmp (&b->last_ephemeral, &dvb->ephemeral_key)))
    {
      /* Check signature */
      struct EphemeralConfirmationPS ec;
 
      ec.purpose.purpose = htonl (GNUNET_SIGNATURE_PURPOSE_TRANSPORT_EPHEMERAL);
      ec.target = *GST_my_identity;
      ec.ephemeral_key = dvb->ephemeral_key;
      ec.purpose.size =  htonl (sizeof(ec));
      ec.sender_monotonic_time = ppay.monotonic_time;
      if (
        GNUNET_OK !=
        GNUNET_CRYPTO_eddsa_verify (
          GNUNET_SIGNATURE_PURPOSE_TRANSPORT_EPHEMERAL,
          &ec,
          &ppay.sender_sig,
          &ppay.sender.public_key))
      {
        /* Signature invalid, discard! */
        GNUNET_break_op (0);
        finish_cmc_handling (cmc);
        return;
      }
    }
    /* Update sender, we now know the real origin! */
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "DVBox received for me from %s via %s\n",
                GNUNET_i2s2 (&ppay.sender),
                GNUNET_i2s (&cmc->im.sender));
    cmc->im.sender = ppay.sender;
 
    if (NULL != b)
    {
      /* update key cache and mono time */
      b->last_ephemeral = dvb->ephemeral_key;
      b->monotonic_time = monotime;
      update_backtalker_monotime (b);
      b->timeout =
        GNUNET_TIME_relative_to_absolute (BACKCHANNEL_INACTIVITY_TIMEOUT);
      cmc->mh = mh;
      demultiplex_with_cmc (cmc);
      return;
    }
    /* setup data structure to cache signature AND check
       monotonic time with PEERSTORE before forwarding backchannel payload */
    b = GNUNET_malloc (sizeof(struct Backtalker) + sizeof(body));
    b->pid = ppay.sender;
    b->body_size = sizeof(body);
    memcpy (&b[1], body, sizeof(body));
    GNUNET_assert (GNUNET_YES ==
                   GNUNET_CONTAINER_multipeermap_put (
                     backtalkers,
                     &b->pid,
                     b,
                     GNUNET_CONTAINER_MULTIHASHMAPOPTION_UNIQUE_ONLY));
    b->monotonic_time = monotime; /* NOTE: to be checked still! */
    b->cmc = cmc;
    b->timeout =
      GNUNET_TIME_relative_to_absolute (BACKCHANNEL_INACTIVITY_TIMEOUT);
    b->task = GNUNET_SCHEDULER_add_at (b->timeout, &backtalker_timeout_cb, b);
    b->get =
      GNUNET_PEERSTORE_iteration_start (peerstore,
                                        "transport",
                                        &b->pid,
                                        GNUNET_PEERSTORE_TRANSPORT_BACKCHANNEL_MONOTIME,
                                        &backtalker_monotime_cb,
                                        b);
  } /* end actual decryption */
}

References GNUNET_TIME_Absolute::abs_value_us, BACKCHANNEL_INACTIVITY_TIMEOUT, backtalker_monotime_cb(), backtalker_timeout_cb(), backtalkers, Backtalker::body_size, Backtalker::cmc, DecapsDvBoxCls::cmc, demultiplex_with_cmc(), dv_decrypt(), dv_hmac(), dv_key_clean(), dv_setup_key_state_from_km(), EphemeralConfirmationPS::ephemeral_key, TransportDVBoxMessage::ephemeral_key, finish_cmc_handling(), Backtalker::get, GNUNET_ALIGN, GNUNET_assert, GNUNET_break_op, GNUNET_CONTAINER_DLL_remove, GNUNET_CONTAINER_MULTIHASHMAPOPTION_UNIQUE_ONLY, GNUNET_CONTAINER_multipeermap_get(), GNUNET_CONTAINER_multipeermap_put(), GNUNET_CRYPTO_eddsa_verify, GNUNET_ERROR_TYPE_DEBUG, GNUNET_ERROR_TYPE_ERROR, GNUNET_free, GNUNET_i2s(), GNUNET_i2s2(), GNUNET_log, GNUNET_malloc, GNUNET_memcmp, GNUNET_MESSAGE_TYPE_TRANSPORT_DV_BOX, GNUNET_MESSAGE_TYPE_TRANSPORT_DV_LEARN, GNUNET_NO, GNUNET_OK, GNUNET_PEERSTORE_iteration_start(), GNUNET_PEERSTORE_TRANSPORT_BACKCHANNEL_MONOTIME, GNUNET_SCHEDULER_add_at(), GNUNET_SIGNATURE_PURPOSE_TRANSPORT_EPHEMERAL, GNUNET_STATISTICS_update(), GNUNET_TIME_absolute_ntoh(), GNUNET_TIME_relative_to_absolute(), GNUNET_YES, GST_my_identity, GST_stats, TransportDVBoxMessage::hmac, CommunicatorMessageContext::im, TransportDVBoxMessage::iv, key, Backtalker::last_ephemeral, mh, CommunicatorMessageContext::mh, TransportDVBoxPayloadP::monotonic_time, Backtalker::monotonic_time, PilsRequest::op, TransportDVBoxMessage::orig_size, peerstore, Backtalker::pid, pils_requests_head, pils_requests_tail, DecapsDvBoxCls::pr, GNUNET_PeerIdentity::public_key, GNUNET_CRYPTO_EccSignaturePurpose::purpose, EphemeralConfirmationPS::purpose, TransportDVBoxPayloadP::sender, GNUNET_TRANSPORT_IncomingMessage::sender, EphemeralConfirmationPS::sender_monotonic_time, TransportDVBoxPayloadP::sender_sig, GNUNET_CRYPTO_EccSignaturePurpose::size, EphemeralConfirmationPS::target, Backtalker::task, Backtalker::timeout, TransportDVBoxMessage::total_hops, and update_backtalker_monotime().

Referenced by handle_dv_box().

Here is the call graph for this function:

Here is the caller graph for this function:

handle_dv_box()

static void handle_dv_box ( void *  cls, const struct TransportDVBoxMessage *  dvb  )

static

Communicator gave us a DV box.

Process the request.

Parameters

cls	a struct CommunicatorMessageContext (must call finish_cmc_handling() when done)
dvb	the message that was received

Definition at line 9244 of file gnunet-service-transport.c.

{
  struct CommunicatorMessageContext *cmc = cls;
  uint16_t size = ntohs (dvb->header.size) - sizeof(*dvb);
  uint16_t num_hops = ntohs (dvb->num_hops);
  const struct GNUNET_PeerIdentity *hops =
    (const struct GNUNET_PeerIdentity *) &dvb[1];
  const char *enc_payload = (const char *) &hops[num_hops];
  uint16_t enc_payload_size =
    size - (num_hops * sizeof(struct GNUNET_PeerIdentity));
  struct DecapsDvBoxCls *decaps_dv_box_cls;
 
  if (GNUNET_EXTRA_LOGGING > 0)
  {
    char *path;
 
    path = GNUNET_strdup (GNUNET_i2s (GST_my_identity));
    for (unsigned int i = 0; i < num_hops; i++)
    {
      char *tmp;
 
      GNUNET_asprintf (&tmp, "%s->%s", path, GNUNET_i2s (&hops[i]));
      GNUNET_free (path);
      path = tmp;
    }
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Received DVBox with remaining path %s\n",
                path);
    GNUNET_free (path);
  }
 
  if (num_hops > 0)
  {
    /* We're trying from the end of the hops array, as we may be
       able to find a shortcut unknown to the origin that way */
    for (int i = num_hops - 1; i >= 0; i--)
    {
      struct Neighbour *n;
 
      if (0 == GNUNET_memcmp (&hops[i], GST_my_identity))
      {
        GNUNET_break_op (0);
        finish_cmc_handling (cmc);
        return;
      }
      n = lookup_neighbour (&hops[i]);
      if (NULL == n)
        continue;
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "Skipping %u/%u hops ahead while routing DV Box\n",
                  i,
                  num_hops);
 
      forward_dv_box (n,
                      (struct TransportDVBoxMessage *) dvb,
                      ntohs (dvb->total_hops) + 1,
                      num_hops - i - 1,    /* number of hops left */
                      &hops[i + 1],    /* remaining hops */
                      enc_payload,
                      enc_payload_size);
      GNUNET_STATISTICS_update (GST_stats,
                                "# DV hops skipped routing boxes",
                                i,
                                GNUNET_NO);
      GNUNET_STATISTICS_update (GST_stats,
                                "# DV boxes routed (total)",
                                1,
                                GNUNET_NO);
      finish_cmc_handling (cmc);
      return;
    }
    /* Woopsie, next hop not in neighbours, drop! */
    GNUNET_STATISTICS_update (GST_stats,
                              "# DV Boxes dropped: next hop unknown",
                              1,
                              GNUNET_NO);
    finish_cmc_handling (cmc);
    return;
  }
  /* We are the target. Unbox and handle message. */
  GNUNET_STATISTICS_update (GST_stats,
                            "# DV boxes opened (ultimate target)",
                            1,
                            GNUNET_NO);
  cmc->total_hops = ntohs (dvb->total_hops);
 
  {
    // DH key derivation with received DV, could be garbage.
    decaps_dv_box_cls = GNUNET_new (struct DecapsDvBoxCls);
    decaps_dv_box_cls->cmc = cmc;
    decaps_dv_box_cls->dvb = dvb;
    decaps_dv_box_cls->pr = GNUNET_new (struct PilsRequest);
 
    GNUNET_CONTAINER_DLL_insert (pils_requests_head,
                                 pils_requests_tail,
                                 decaps_dv_box_cls->pr);
    decaps_dv_box_cls->pr->op = GNUNET_PILS_kem_decaps (pils,
                                                        &dvb->ephemeral_key,
                                                        decaps_dv_box_cb,
                                                        decaps_dv_box_cls);
  }
  // TODO keep track of cls and potentially clean
}

References DecapsDvBoxCls::cmc, decaps_dv_box_cb(), DecapsDvBoxCls::dvb, TransportDVBoxMessage::ephemeral_key, finish_cmc_handling(), forward_dv_box(), GNUNET_asprintf(), GNUNET_break_op, GNUNET_CONTAINER_DLL_insert, GNUNET_ERROR_TYPE_DEBUG, GNUNET_EXTRA_LOGGING, GNUNET_free, GNUNET_i2s(), GNUNET_log, GNUNET_memcmp, GNUNET_new, GNUNET_NO, GNUNET_PILS_kem_decaps(), GNUNET_STATISTICS_update(), GNUNET_strdup, GST_my_identity, GST_stats, TransportDVBoxMessage::header, lookup_neighbour(), TransportDVBoxMessage::num_hops, PilsRequest::op, pils, pils_requests_head, pils_requests_tail, DecapsDvBoxCls::pr, GNUNET_MessageHeader::size, size, TransportDVBoxMessage::total_hops, and CommunicatorMessageContext::total_hops.

Here is the call graph for this function:

check_incoming_msg()

static int check_incoming_msg ( void *  cls, const struct GNUNET_TRANSPORT_IncomingMessage *  im  )

static

Client notified us about transmission from a peer.

Process the request.

Parameters

cls	a struct TransportClient which sent us the message
im	the send message that was sent

Returns

GNUNET_YES if message is well-formed

Definition at line 9357 of file gnunet-service-transport.c.

{
  struct TransportClient *tc = cls;
 
  if (CT_COMMUNICATOR != tc->type)
  {
    GNUNET_break (0);
    return GNUNET_SYSERR;
  }
  GNUNET_MQ_check_boxed_message (im);
  return GNUNET_OK;
}

References CT_COMMUNICATOR, GNUNET_break, GNUNET_MQ_check_boxed_message, GNUNET_OK, GNUNET_SYSERR, and tc.

check_known_address()

static int check_known_address ( void *  cls, const struct GNUNET_PeerIdentity *  pid, void *  value  )

static

Test if the validation state in value matches the address from cls.

Parameters

cls	a struct CheckKnownAddressContext
pid	unused (must match though)
value	a struct ValidationState

Returns

GNUNET_OK if not matching, GNUNET_NO if match found

Definition at line 9399 of file gnunet-service-transport.c.

{
  struct CheckKnownAddressContext *ckac = cls;
  struct ValidationState *vs = value;
 
  (void) pid;
  if (0 != strcmp (vs->address, ckac->address))
    return GNUNET_OK;
  ckac->vs = vs;
  return GNUNET_NO;
}

References CheckKnownAddressContext::address, GNUNET_NO, GNUNET_OK, pid, value, and CheckKnownAddressContext::vs.

Referenced by start_address_validation().

Here is the caller graph for this function:

validation_start_cb()

static void validation_start_cb ( void *  cls )

static

Task run periodically to validate some address based on validation_heap.

Parameters

cls

NULL

Definition at line 12067 of file gnunet-service-transport.c.

{
  struct ValidationState *vs;
  struct Queue *q;
  const struct GNUNET_TIME_Absolute now = GNUNET_TIME_absolute_get_monotonic (
    GST_cfg);
 
  (void) cls;
  validation_task = NULL;
  vs = GNUNET_CONTAINER_heap_peek (validation_heap);
  /* drop validations past their expiration */
  while (
    (NULL != vs) &&
    (0 == GNUNET_TIME_absolute_get_remaining (vs->valid_until).rel_value_us))
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Validation response %s cleaned up\n",
                GNUNET_sh2s (&vs->challenge.value));
    free_validation_state (vs);
    vs = GNUNET_CONTAINER_heap_peek (validation_heap);
  }
  if (NULL == vs)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_INFO,
                "Address validation task not scheduled anymore, nothing to do\n");
    return;   /* woopsie, no more addresses known, should only
                 happen if we're really a lonely peer */
  }
  q = find_queue (&vs->pid, vs->address);
  if (GNUNET_TIME_absolute_cmp (vs->first_challenge_use, >, now))
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "To early to start next address validation for challenge %s\n",
                GNUNET_sh2s (&vs->challenge.value));
    return;
  }
  if (NULL == q)
  {
    vs->awaiting_queue = GNUNET_YES;
    suggest_to_connect (&vs->pid, vs->address);
  }
  else
    validation_transmit_on_queue (q, vs);
  /* Finally, reschedule next attempt */
  vs->challenge_backoff =
    GNUNET_TIME_randomized_backoff (vs->challenge_backoff,
                                    MAX_VALIDATION_CHALLENGE_FREQ);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Address validation task will run again in %s\n",
              GNUNET_STRINGS_relative_time_to_string (vs->challenge_backoff,
                                                      GNUNET_YES));
  update_next_challenge_time (vs,
                              GNUNET_TIME_relative_to_absolute (
                                vs->challenge_backoff));
}

References find_queue(), free_validation_state(), GNUNET_CONTAINER_heap_peek(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_ERROR_TYPE_INFO, GNUNET_log, GNUNET_sh2s(), GNUNET_STRINGS_relative_time_to_string(), GNUNET_TIME_absolute_cmp, GNUNET_TIME_absolute_get_monotonic(), GNUNET_TIME_absolute_get_remaining(), GNUNET_TIME_randomized_backoff(), GNUNET_TIME_relative_to_absolute(), GNUNET_YES, GST_cfg, MAX_VALIDATION_CHALLENGE_FREQ, ValidationState::pid, q, GNUNET_TIME_Relative::rel_value_us, suggest_to_connect(), update_next_challenge_time(), ValidationState::valid_until, validation_heap, validation_task, and validation_transmit_on_queue().

Referenced by update_next_challenge_time().

Here is the call graph for this function:

Here is the caller graph for this function:

update_next_challenge_time()

static void update_next_challenge_time ( struct ValidationState *  vs, struct GNUNET_TIME_Absolute  new_time  )

static

Set the time for next_challenge of vs to new_time.

Updates the heap and if necessary reschedules the job.

Parameters

vs	validation state to update
new_time	new time for revalidation

Definition at line 9431 of file gnunet-service-transport.c.

{
  struct GNUNET_TIME_Relative delta;
 
  if (new_time.abs_value_us == vs->next_challenge.abs_value_us)
    return; /* be lazy */
  vs->next_challenge = new_time;
  if (NULL == vs->hn)
    vs->hn =
      GNUNET_CONTAINER_heap_insert (validation_heap, vs, new_time.abs_value_us);
  else
    GNUNET_CONTAINER_heap_update_cost (vs->hn, new_time.abs_value_us);
  if ((vs != GNUNET_CONTAINER_heap_peek (validation_heap)) &&
      (NULL != validation_task))
    return;
  if (NULL != validation_task)
    GNUNET_SCHEDULER_cancel (validation_task);
  /* randomize a bit */
  delta.rel_value_us =
    GNUNET_CRYPTO_random_u64 (GNUNET_CRYPTO_QUALITY_WEAK,
                              MIN_DELAY_ADDRESS_VALIDATION.rel_value_us);
  new_time = GNUNET_TIME_absolute_add (new_time, delta);
  validation_task =
    GNUNET_SCHEDULER_add_at (new_time, &validation_start_cb, NULL);
}

References GNUNET_TIME_Absolute::abs_value_us, delta, GNUNET_CONTAINER_heap_insert(), GNUNET_CONTAINER_heap_peek(), GNUNET_CONTAINER_heap_update_cost(), GNUNET_CRYPTO_QUALITY_WEAK, GNUNET_CRYPTO_random_u64(), GNUNET_SCHEDULER_add_at(), GNUNET_SCHEDULER_cancel(), GNUNET_TIME_absolute_add(), MIN_DELAY_ADDRESS_VALIDATION, GNUNET_TIME_Relative::rel_value_us, validation_heap, validation_start_cb(), and validation_task.

Referenced by handle_validation_response(), revalidation_start_cb(), start_address_validation(), and validation_start_cb().

Here is the call graph for this function:

Here is the caller graph for this function:

start_address_validation()

static void start_address_validation ( const struct GNUNET_PeerIdentity *  pid, const char *  address  )

static

Start address validation.

Parameters

pid	peer the address is for
address	an address to reach pid (presumably)

Definition at line 9466 of file gnunet-service-transport.c.

{
  struct GNUNET_TIME_Absolute now;
  struct ValidationState *vs;
  struct CheckKnownAddressContext ckac = { .address = address, .vs = NULL };
 
  (void) GNUNET_CONTAINER_multipeermap_get_multiple (validation_map,
                                                     pid,
                                                     &check_known_address,
                                                     &ckac);
  if (NULL != (vs = ckac.vs))
  {
    /* if 'vs' is not currently valid, we need to speed up retrying the
     * validation */
    if (vs->validated_until.abs_value_us < vs->next_challenge.abs_value_us)
    {
      /* reduce backoff as we got a fresh advertisement */
      vs->challenge_backoff =
        GNUNET_TIME_relative_min (FAST_VALIDATION_CHALLENGE_FREQ,
                                  GNUNET_TIME_relative_divide (
                                    vs->challenge_backoff,
                                    2));
      update_next_challenge_time (vs,
                                  GNUNET_TIME_relative_to_absolute (
                                    vs->challenge_backoff));
    }
    return;
  }
  now = GNUNET_TIME_absolute_get_monotonic (GST_cfg);
  vs = GNUNET_new (struct ValidationState);
  vs->pid = *pid;
  vs->valid_until =
    GNUNET_TIME_relative_to_absolute (ADDRESS_VALIDATION_LIFETIME);
  vs->first_challenge_use = now;
  vs->validation_rtt = GNUNET_TIME_UNIT_FOREVER_REL;
  GNUNET_CRYPTO_random_block (GNUNET_CRYPTO_QUALITY_NONCE,
                              &vs->challenge,
                              sizeof(vs->challenge));
  vs->address = GNUNET_strdup (address);
  GNUNET_CRYPTO_hash (vs->address, strlen (vs->address), &vs->hc);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Starting address validation `%s' of peer %s using challenge %s\n",
              address,
              GNUNET_i2s (pid),
              GNUNET_sh2s (&vs->challenge.value));
  GNUNET_assert (GNUNET_YES ==
                 GNUNET_CONTAINER_multipeermap_put (
                   validation_map,
                   &vs->pid,
                   vs,
                   GNUNET_CONTAINER_MULTIHASHMAPOPTION_MULTIPLE));
  update_next_challenge_time (vs, now);
}

References GNUNET_TIME_Absolute::abs_value_us, address, CheckKnownAddressContext::address, ADDRESS_VALIDATION_LIFETIME, check_known_address(), FAST_VALIDATION_CHALLENGE_FREQ, GNUNET_assert, GNUNET_CONTAINER_MULTIHASHMAPOPTION_MULTIPLE, GNUNET_CONTAINER_multipeermap_get_multiple(), GNUNET_CONTAINER_multipeermap_put(), GNUNET_CRYPTO_hash(), GNUNET_CRYPTO_QUALITY_NONCE, GNUNET_CRYPTO_random_block(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_i2s(), GNUNET_log, GNUNET_new, GNUNET_sh2s(), GNUNET_strdup, GNUNET_TIME_absolute_get_monotonic(), GNUNET_TIME_relative_divide(), GNUNET_TIME_relative_min(), GNUNET_TIME_relative_to_absolute(), GNUNET_TIME_UNIT_FOREVER_REL, GNUNET_YES, GST_cfg, ValidationState::pid, pid, update_next_challenge_time(), ValidationState::valid_until, ValidationState::validated_until, validation_map, and CheckKnownAddressContext::vs.

Referenced by handle_add_queue_message(), handle_request_hello_validation(), hello_for_client_cb(), hello_for_incoming_cb(), and sign_t_validation_cb().

Here is the call graph for this function:

Here is the caller graph for this function:

find_queue()

static struct Queue * find_queue ( const struct GNUNET_PeerIdentity *  pid, const char *  address  )

static

Find the queue matching pid and address.

Parameters

pid	peer the queue must go to
address	address the queue must use

Returns

NULL if no such queue exists

Definition at line 9842 of file gnunet-service-transport.c.

{
  struct Neighbour *n;
 
  n = lookup_neighbour (pid);
  if (NULL == n)
    return NULL;
  for (struct Queue *pos = n->queue_head; NULL != pos;
       pos = pos->next_neighbour)
  {
    if (0 == strcmp (pos->address, address))
      return pos;
  }
  return NULL;
}

References address, lookup_neighbour(), pid, and Neighbour::queue_head.

Referenced by handle_request_hello_validation(), handle_validation_response(), hello_for_client_cb(), hello_for_incoming_cb(), revalidation_start_cb(), and validation_start_cb().

Here is the call graph for this function:

Here is the caller graph for this function:

suggest_to_connect()

static void suggest_to_connect ( const struct GNUNET_PeerIdentity *  pid, const char *  address  )

static

Signature of a function called with a communicator address of a peer pid that an application wants us to connect to.

Parameters

pid	target peer
address	the address to try

Definition at line 11963 of file gnunet-service-transport.c.

{
  static uint32_t idgen = 0;
  struct TransportClient *tc;
  char *prefix;
  struct GNUNET_TRANSPORT_CreateQueue *cqm;
  struct GNUNET_MQ_Envelope *env;
  size_t alen;
 
  prefix = GNUNET_HELLO_address_to_prefix (address);
  if (NULL == prefix)
  {
    GNUNET_break (0);  /* We got an invalid address!? */
    return;
  }
  tc = lookup_communicator (prefix);
  if (NULL == tc)
  {
    GNUNET_STATISTICS_update (GST_stats,
                              "# Suggestions ignored due to missing communicator",
                              1,
                              GNUNET_NO);
    GNUNET_log (GNUNET_ERROR_TYPE_INFO,
                "Cannot connect to %s at `%s', no matching communicator present\n",
                GNUNET_i2s (pid),
                address);
    GNUNET_free (prefix);
    return;
  }
  /* forward suggestion for queue creation to communicator */
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Request #%u for `%s' communicator to create queue to `%s' at `%s'\n",
              (unsigned int) idgen,
              prefix,
              GNUNET_i2s (pid),
              address);
  GNUNET_free (prefix);
  alen = strlen (address) + 1;
  env =
    GNUNET_MQ_msg_extra (cqm, alen, GNUNET_MESSAGE_TYPE_TRANSPORT_QUEUE_CREATE);
  cqm->request_id = htonl (idgen++);
  cqm->receiver = *pid;
  memcpy (&cqm[1], address, alen);
  GNUNET_MQ_send (tc->mq, env);
}

References address, env, GNUNET_break, GNUNET_ERROR_TYPE_DEBUG, GNUNET_ERROR_TYPE_INFO, GNUNET_free, GNUNET_HELLO_address_to_prefix(), GNUNET_i2s(), GNUNET_log, GNUNET_MESSAGE_TYPE_TRANSPORT_QUEUE_CREATE, GNUNET_MQ_msg_extra, GNUNET_MQ_send(), GNUNET_NO, GNUNET_STATISTICS_update(), GST_stats, lookup_communicator(), pid, prefix, GNUNET_TRANSPORT_CreateQueue::receiver, GNUNET_TRANSPORT_CreateQueue::request_id, and tc.

Referenced by handle_request_hello_validation(), hello_for_client_cb(), hello_for_incoming_cb(), revalidation_start_cb(), and validation_start_cb().

Here is the call graph for this function:

Here is the caller graph for this function:

hello_for_incoming_cb()

static void hello_for_incoming_cb ( void *  cls, const struct GNUNET_PeerIdentity *  pid, const char *  uri  )

static

Definition at line 9531 of file gnunet-service-transport.c.

{
  struct Queue *q;
  int pfx_len;
  const char *eou;
  char *address;
  (void) cls;
 
  eou = strstr (uri,
                "://");
  pfx_len = eou - uri;
  eou += 3;
  GNUNET_asprintf (&address,
                   "%.*s-%s",
                   pfx_len,
                   uri,
                   eou);
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "helo for client %s\n",
              address);
  q = find_queue (pid, address);
  if (NULL == q)
  {
    suggest_to_connect (pid, address);
  }
  else
    start_address_validation (pid, address);
  GNUNET_free (address);
}

References address, find_queue(), GNUNET_asprintf(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_free, GNUNET_log, pid, q, start_address_validation(), suggest_to_connect(), and uri.

Referenced by handle_hello_for_incoming().

Here is the call graph for this function:

Here is the caller graph for this function:

handle_hello_for_incoming()

static void handle_hello_for_incoming ( void *  cls, const struct GNUNET_PEERSTORE_Record *  record, const char *  emsg  )

static

Function called by PEERSTORE for each matching record.

Parameters

cls	closure, a struct IncomingRequest
record	peerstore record information
emsg	error message, or NULL if no errors

Definition at line 9573 of file gnunet-service-transport.c.

{
  struct IncomingRequest *ir = cls;
  struct GNUNET_HELLO_Parser *parser;
  struct GNUNET_MessageHeader *hello;
 
  if (NULL != emsg)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
                "Got failure from PEERSTORE: %s\n",
                emsg);
    return;
  }
  hello = record->value;
  if (0 == GNUNET_memcmp (&record->peer, GST_my_identity))
  {
    GNUNET_PEERSTORE_monitor_next (ir->nc, 1);
    return;
  }
  parser = GNUNET_HELLO_parser_from_msg (hello);
  GNUNET_HELLO_parser_iterate (parser,
                               hello_for_incoming_cb,
                               NULL);
  GNUNET_HELLO_parser_free (parser);
}

References GNUNET_ERROR_TYPE_WARNING, GNUNET_HELLO_parser_free(), GNUNET_HELLO_parser_from_msg(), GNUNET_HELLO_parser_iterate(), GNUNET_log, GNUNET_memcmp, GNUNET_PEERSTORE_monitor_next(), GST_my_identity, hello_for_incoming_cb(), IncomingRequest::nc, and record().

Referenced by sign_t_validation_cb().

Here is the call graph for this function:

Here is the caller graph for this function:

hello_for_incoming_error_cb()

static void hello_for_incoming_error_cb ( void *  cls )

static

Definition at line 9603 of file gnunet-service-transport.c.

{
  GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
              "Error in PEERSTORE monitoring\n");
}

References GNUNET_ERROR_TYPE_WARNING, and GNUNET_log.

Referenced by sign_t_validation_cb().

Here is the caller graph for this function:

hello_for_incoming_sync_cb()

static void hello_for_incoming_sync_cb ( void *  cls )

static

Definition at line 9611 of file gnunet-service-transport.c.

{
  GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
              "Done with initial PEERSTORE iteration during monitoring\n");
}

References GNUNET_ERROR_TYPE_WARNING, and GNUNET_log.

Referenced by sign_t_validation_cb().

Here is the caller graph for this function:

sign_t_validation_cb()

static void sign_t_validation_cb ( void *  cls, const struct GNUNET_PeerIdentity *  pid, const struct GNUNET_CRYPTO_EddsaSignature *  sig  )

static

Definition at line 9627 of file gnunet-service-transport.c.

{
  struct SignTValidationCls *sign_t_validation_cls = cls;
  struct CommunicatorMessageContext *cmc = sign_t_validation_cls->cmc;
  struct TransportValidationResponseMessage tvr = sign_t_validation_cls->tvr;
  struct VirtualLink *vl;
  struct Neighbour *n;
  struct IncomingRequest *ir;
  struct GNUNET_PeerIdentity sender;
 
  sign_t_validation_cls->pr->op = NULL;
  GNUNET_CONTAINER_DLL_remove (pils_requests_head,
                               pils_requests_tail,
                               sign_t_validation_cls->pr);
  GNUNET_free (sign_t_validation_cls->pr);
  tvr.signature = *sig;
  sender = cmc->im.sender;
  vl = lookup_virtual_link (&sender);
  if ((NULL != vl) && (GNUNET_YES == vl->confirmed))
  {
    // route_control_message_without_fc (&cmc->im.sender,
    route_control_message_without_fc (vl,
                                      &tvr.header,
                                      RMO_ANYTHING_GOES | RMO_REDUNDANT);
  }
  else
  {
    /* Use route via neighbour */
    n = lookup_neighbour (&sender);
    if (NULL != n)
      route_via_neighbour (n, &tvr.header,
                           RMO_ANYTHING_GOES | RMO_REDUNDANT
                           | RMO_UNCONFIRMED_ALLOWED);
  }
 
  finish_cmc_handling (cmc);
  if (NULL != vl)
    return;
 
  /* For us, the link is still down, but we need bi-directional
     connections (for flow-control and for this to be useful for
     CORE), so we must try to bring the link up! */
 
  /* (1) Check existing queues, if any, we may be lucky! */
  n = lookup_neighbour (&sender);
  if (NULL != n)
    for (struct Queue *q = n->queue_head; NULL != q; q = q->next_neighbour)
      start_address_validation (&sender, q->address);
  /* (2) Also try to see if we have addresses in PEERSTORE for this peer
     we could use */
  for (ir = ir_head; NULL != ir; ir = ir->next)
    if (0 == GNUNET_memcmp (&ir->pid, &sender))
      return;
  /* we are already trying */
  ir = GNUNET_new (struct IncomingRequest);
  ir->pid = sender;
  GNUNET_CONTAINER_DLL_insert (ir_head, ir_tail, ir);
 
  ir->nc = GNUNET_PEERSTORE_monitor_start (GST_cfg,
                                           GNUNET_YES,
                                           "peerstore",
                                           NULL,
                                           GNUNET_PEERSTORE_HELLO_KEY,
                                           &hello_for_incoming_error_cb,
                                           NULL,
                                           &hello_for_incoming_sync_cb,
                                           NULL,
                                           &handle_hello_for_incoming,
                                           ir);
  ir_total++;
  /* Bound attempts we do in parallel here, might otherwise get excessive */
  while (ir_total > MAX_INCOMING_REQUEST)
    free_incoming_request (ir_head);
};

References SignTValidationCls::cmc, VirtualLink::confirmed, finish_cmc_handling(), free_incoming_request(), GNUNET_CONTAINER_DLL_insert, GNUNET_CONTAINER_DLL_remove, GNUNET_free, GNUNET_memcmp, GNUNET_new, GNUNET_PEERSTORE_HELLO_KEY, GNUNET_PEERSTORE_monitor_start(), GNUNET_YES, GST_cfg, handle_hello_for_incoming(), TransportValidationResponseMessage::header, hello_for_incoming_error_cb(), hello_for_incoming_sync_cb(), CommunicatorMessageContext::im, ir_head, ir_tail, ir_total, lookup_neighbour(), lookup_virtual_link(), MAX_INCOMING_REQUEST, IncomingRequest::nc, IncomingRequest::next, PilsRequest::op, IncomingRequest::pid, pils_requests_head, pils_requests_tail, SignTValidationCls::pr, q, Neighbour::queue_head, RMO_ANYTHING_GOES, RMO_REDUNDANT, RMO_UNCONFIRMED_ALLOWED, route_control_message_without_fc(), route_via_neighbour(), GNUNET_TRANSPORT_IncomingMessage::sender, TransportValidationResponseMessage::signature, start_address_validation(), and SignTValidationCls::tvr.

Referenced by handle_validation_challenge().

Here is the call graph for this function:

Here is the caller graph for this function:

handle_validation_challenge()

static void handle_validation_challenge ( void *  cls, const struct TransportValidationChallengeMessage *  tvc  )

static

Communicator gave us a transport address validation challenge.

Process the request.

Parameters

cls	a struct CommunicatorMessageContext (must call finish_cmc_handling() when done)
tvc	the message that was received

Definition at line 9714 of file gnunet-service-transport.c.

{
  struct CommunicatorMessageContext *cmc = cls;
  struct TransportValidationResponseMessage tvr;
  struct GNUNET_TIME_RelativeNBO validity_duration;
 
  /* DV-routed messages are not allowed for validation challenges */
  if (cmc->total_hops > 0)
  {
    GNUNET_break_op (0);
    finish_cmc_handling (cmc);
    return;
  }
  validity_duration = cmc->im.expected_address_validity;
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Received address validation challenge %s\n",
              GNUNET_sh2s (&tvc->challenge.value));
  /* If we have a virtual link, we use this mechanism to signal the
     size of the flow control window, and to allow the sender
     to ask for increases. If for us the virtual link is still down,
     we will always give a window size of zero. */
  tvr.header.type =
    htons (GNUNET_MESSAGE_TYPE_TRANSPORT_ADDRESS_VALIDATION_RESPONSE);
  tvr.header.size = htons (sizeof(tvr));
  tvr.reserved = htonl (0);
  tvr.challenge = tvc->challenge;
  tvr.origin_time = tvc->sender_time;
  tvr.validity_duration = validity_duration;
  {
    /* create signature */
    struct TransportValidationPS tvp = {
      .purpose.purpose = htonl (GNUNET_SIGNATURE_PURPOSE_TRANSPORT_CHALLENGE),
      .purpose.size = htonl (sizeof(tvp)),
      .validity_duration = validity_duration,
      .challenge = tvc->challenge
    };
    struct SignTValidationCls *sign_t_validation_cls;
 
    sign_t_validation_cls = GNUNET_new (struct SignTValidationCls);
    sign_t_validation_cls->cmc = cmc;
    sign_t_validation_cls->tvr = tvr;
    sign_t_validation_cls->pr = GNUNET_new (struct PilsRequest);
    GNUNET_CONTAINER_DLL_insert (pils_requests_head,
                                 pils_requests_tail,
                                 sign_t_validation_cls->pr);
    sign_t_validation_cls->pr->op =
      GNUNET_PILS_sign_by_peer_identity (pils,
                                         &tvp.purpose,
                                         &sign_t_validation_cb,
                                         sign_t_validation_cls);
  }
}

References TransportValidationChallengeMessage::challenge, TransportValidationResponseMessage::challenge, SignTValidationCls::cmc, GNUNET_TRANSPORT_IncomingMessage::expected_address_validity, finish_cmc_handling(), GNUNET_break_op, GNUNET_CONTAINER_DLL_insert, GNUNET_ERROR_TYPE_DEBUG, GNUNET_log, GNUNET_MESSAGE_TYPE_TRANSPORT_ADDRESS_VALIDATION_RESPONSE, GNUNET_new, GNUNET_PILS_sign_by_peer_identity(), GNUNET_sh2s(), GNUNET_SIGNATURE_PURPOSE_TRANSPORT_CHALLENGE, TransportValidationResponseMessage::header, CommunicatorMessageContext::im, PilsRequest::op, pils, pils_requests_head, pils_requests_tail, SignTValidationCls::pr, GNUNET_CRYPTO_EccSignaturePurpose::purpose, TransportValidationPS::purpose, TransportValidationResponseMessage::reserved, TransportValidationChallengeMessage::sender_time, sign_t_validation_cb(), GNUNET_MessageHeader::size, CommunicatorMessageContext::total_hops, SignTValidationCls::tvr, GNUNET_MessageHeader::type, TransportValidationPS::validity_duration, and GNUNET_CRYPTO_ChallengeNonceP::value.

Here is the call graph for this function:

check_known_challenge()

static int check_known_challenge ( void *  cls, const struct GNUNET_PeerIdentity *  pid, void *  value  )

static

Test if the validation state in value matches the challenge from cls.

Parameters

cls	a struct CheckKnownChallengeContext
pid	unused (must match though)
value	a struct ValidationState

Returns

GNUNET_OK if not matching, GNUNET_NO if match found

Definition at line 9797 of file gnunet-service-transport.c.

{
  struct CheckKnownChallengeContext *ckac = cls;
  struct ValidationState *vs = value;
 
  (void) pid;
  if (0 != GNUNET_memcmp (&vs->challenge, ckac->challenge))
    return GNUNET_OK;
  ckac->vs = vs;
  return GNUNET_NO;
}

References CheckKnownChallengeContext::challenge, GNUNET_memcmp, GNUNET_NO, GNUNET_OK, pid, value, and CheckKnownChallengeContext::vs.

Referenced by handle_validation_response().

Here is the caller graph for this function:

peerstore_store_validation_cb()

static void peerstore_store_validation_cb ( void *  cls, int  success  )

static

Function called when peerstore is done storing a validated address.

Parameters

cls	a struct ValidationState
success	GNUNET_YES on success

Definition at line 9820 of file gnunet-service-transport.c.

{
  struct ValidationState *vs = cls;
 
  vs->sc = NULL;
  if (GNUNET_YES == success)
    return;
  GNUNET_STATISTICS_update (GST_stats,
                            "# Peerstore failed to store foreign address",
                            1,
                            GNUNET_NO);
}

References GNUNET_NO, GNUNET_STATISTICS_update(), GNUNET_YES, and GST_stats.

Referenced by handle_validation_response().

Here is the call graph for this function:

Here is the caller graph for this function:

validation_transmit_on_queue()

static void validation_transmit_on_queue ( struct Queue *  q, struct ValidationState *  vs  )

static

The queue q (which matches the peer and address in vs) is ready for queueing.

We should now queue the validation request.

Parameters

q	queue to send on
vs	state to derive validation challenge from

Definition at line 12018 of file gnunet-service-transport.c.

{
  struct TransportValidationChallengeMessage tvc;
  struct GNUNET_TIME_Absolute monotonic_time;
 
  if (NULL != vs->revalidation_task)
  {
    GNUNET_SCHEDULER_cancel (vs->revalidation_task);
    vs->revalidation_task = NULL;
  }
  /*memcpy (&hkey,
          &hc,
          sizeof (hkey));*/
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Remove key %s for address %s map size %u contains %u\n",
              GNUNET_h2s (&vs->hc),
              vs->address,
              GNUNET_CONTAINER_multihashmap_size (revalidation_map),
              GNUNET_CONTAINER_multihashmap_contains (revalidation_map,
                                                      &vs->hc));
  GNUNET_CONTAINER_multihashmap_remove (revalidation_map, &vs->hc, vs);
 
  monotonic_time  = GNUNET_TIME_absolute_get_monotonic (GST_cfg);
  if (GNUNET_TIME_UNIT_ZERO_ABS.abs_value_us ==
      vs->last_challenge_use.abs_value_us)
  {
    vs->first_challenge_use = monotonic_time;
  }
  vs->last_challenge_use = monotonic_time;
  tvc.header.type =
    htons (GNUNET_MESSAGE_TYPE_TRANSPORT_ADDRESS_VALIDATION_CHALLENGE);
  tvc.header.size = htons (sizeof(tvc));
  tvc.reserved = htonl (0);
  tvc.challenge = vs->challenge;
  tvc.sender_time = GNUNET_TIME_absolute_hton (vs->last_challenge_use);
  GNUNET_log (GNUNET_ERROR_TYPE_INFO,
              "Sending address validation challenge %s to %s\n",
              GNUNET_sh2s (&tvc.challenge.value),
              GNUNET_i2s (&q->neighbour->pid));
  queue_send_msg (q, NULL, &tvc, sizeof(tvc));
}

References TransportValidationChallengeMessage::challenge, GNUNET_CONTAINER_multihashmap_contains(), GNUNET_CONTAINER_multihashmap_remove(), GNUNET_CONTAINER_multihashmap_size(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_ERROR_TYPE_INFO, GNUNET_h2s(), GNUNET_i2s(), GNUNET_log, GNUNET_MESSAGE_TYPE_TRANSPORT_ADDRESS_VALIDATION_CHALLENGE, GNUNET_SCHEDULER_cancel(), GNUNET_sh2s(), GNUNET_TIME_absolute_get_monotonic(), GNUNET_TIME_absolute_hton(), GNUNET_TIME_UNIT_ZERO_ABS, GST_cfg, TransportValidationChallengeMessage::header, q, queue_send_msg(), TransportValidationChallengeMessage::reserved, revalidation_map, TransportValidationChallengeMessage::sender_time, GNUNET_MessageHeader::size, GNUNET_MessageHeader::type, and GNUNET_CRYPTO_ChallengeNonceP::value.

Referenced by check_validation_request_pending(), revalidation_start_cb(), and validation_start_cb().

Here is the call graph for this function:

Here is the caller graph for this function:

revalidation_start_cb()

static void revalidation_start_cb ( void *  cls )

static

Definition at line 9863 of file gnunet-service-transport.c.

{
  struct ValidationState *vs = cls;
  struct Queue *q;
  struct GNUNET_TIME_Absolute now;
 
  vs->revalidation_task = NULL;
  q = find_queue (&vs->pid, vs->address);
  if (NULL == q)
  {
    now = GNUNET_TIME_absolute_get ();
    vs->awaiting_queue = GNUNET_YES;
    suggest_to_connect (&vs->pid, vs->address);
    update_next_challenge_time (vs, now);
  }
  else
    validation_transmit_on_queue (q, vs);
}

References find_queue(), GNUNET_TIME_absolute_get(), GNUNET_YES, ValidationState::pid, q, suggest_to_connect(), update_next_challenge_time(), and validation_transmit_on_queue().

Referenced by handle_validation_response().

Here is the call graph for this function:

Here is the caller graph for this function:

revalidate_map_it()

static enum GNUNET_GenericReturnValue revalidate_map_it ( void *  cls, const struct GNUNET_HashCode *  key, void *  value  )

static

Definition at line 9884 of file gnunet-service-transport.c.

{
  (void) cls;
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Key in revalidate map  %s \n",
              GNUNET_h2s (key));
  return GNUNET_YES;
}

References GNUNET_ERROR_TYPE_DEBUG, GNUNET_h2s(), GNUNET_log, GNUNET_YES, and key.

Referenced by handle_validation_response().

Here is the call graph for this function:

Here is the caller graph for this function:

handle_validation_response()

static void handle_validation_response ( void *  cls, const struct TransportValidationResponseMessage *  tvr  )

static

Communicator gave us a transport address validation response.

Process the request.

Parameters

cls	a struct CommunicatorMessageContext (must call finish_cmc_handling() when done)
tvr	the message that was received

Definition at line 9906 of file gnunet-service-transport.c.

{
  struct CommunicatorMessageContext *cmc = cls;
  struct ValidationState *vs;
  struct CheckKnownChallengeContext ckac = { .challenge = &tvr->challenge,
                                             .vs = NULL};
  struct GNUNET_TIME_Absolute origin_time;
  struct Queue *q;
  struct Neighbour *n;
  struct VirtualLink *vl;
  const struct GNUNET_TIME_Absolute now = GNUNET_TIME_absolute_get_monotonic (
    GST_cfg);
 
  /* check this is one of our challenges */
  (void) GNUNET_CONTAINER_multipeermap_get_multiple (validation_map,
                                                     &cmc->im.sender,
                                                     &check_known_challenge,
                                                     &ckac);
  if (NULL == (vs = ckac.vs))
  {
    /* This can happen simply if we 'forgot' the challenge by now,
       i.e. because we received the validation response twice */
    GNUNET_STATISTICS_update (GST_stats,
                              "# Validations dropped, challenge unknown",
                              1,
                              GNUNET_NO);
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Validation response %s dropped, challenge unknown\n",
                GNUNET_sh2s (&tvr->challenge.value));
    finish_cmc_handling (cmc);
    return;
  }
 
  /* sanity check on origin time */
  origin_time = GNUNET_TIME_absolute_ntoh (tvr->origin_time);
  if ((origin_time.abs_value_us < vs->first_challenge_use.abs_value_us) ||
      (origin_time.abs_value_us > vs->last_challenge_use.abs_value_us))
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Diff first use %" PRIu64 " and last use %" PRIu64 "\n",
                vs->first_challenge_use.abs_value_us - origin_time.abs_value_us,
                origin_time.abs_value_us - vs->last_challenge_use.abs_value_us);
    GNUNET_break_op (0);
    finish_cmc_handling (cmc);
    return;
  }
 
  {
    /* check signature */
    struct TransportValidationPS tvp = {
      .purpose.purpose = htonl (GNUNET_SIGNATURE_PURPOSE_TRANSPORT_CHALLENGE),
      .purpose.size = htonl (sizeof(tvp)),
      .validity_duration = tvr->validity_duration,
      .challenge = tvr->challenge
    };
 
    if (
      GNUNET_OK !=
      GNUNET_CRYPTO_eddsa_verify (GNUNET_SIGNATURE_PURPOSE_TRANSPORT_CHALLENGE,
                                  &tvp,
                                  &tvr->signature,
                                  &cmc->im.sender.public_key))
    {
      GNUNET_break_op (0);
      finish_cmc_handling (cmc);
      return;
    }
  }
 
  /* validity is capped by our willingness to keep track of the
     validation entry and the maximum the other peer allows */
  vs->valid_until = GNUNET_TIME_relative_to_absolute (
    GNUNET_TIME_relative_min (GNUNET_TIME_relative_ntoh (
                                tvr->validity_duration),
                              MAX_ADDRESS_VALID_UNTIL));
  vs->validated_until =
    GNUNET_TIME_absolute_min (vs->valid_until,
                              GNUNET_TIME_relative_to_absolute (
                                ADDRESS_VALIDATION_LIFETIME));
  vs->validation_rtt = GNUNET_TIME_absolute_get_duration (origin_time);
  vs->challenge_backoff = GNUNET_TIME_UNIT_ZERO;
  GNUNET_CRYPTO_random_block (GNUNET_CRYPTO_QUALITY_NONCE,
                              &vs->challenge,
                              sizeof(vs->challenge));
  vs->first_challenge_use = GNUNET_TIME_absolute_subtract (
    vs->validated_until,
    GNUNET_TIME_relative_multiply (vs->validation_rtt,
                                   VALIDATION_RTT_BUFFER_FACTOR));
  if (GNUNET_TIME_absolute_cmp (vs->first_challenge_use, <, now))
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "First challenge use is now %" PRIu64 " %s \n",
                vs->first_challenge_use.abs_value_us,
                GNUNET_sh2s (&vs->challenge.value));
    vs->first_challenge_use = now;
  }
  else
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "First challenge use is later %" PRIu64 " %s \n",
                vs->first_challenge_use.abs_value_us,
                GNUNET_sh2s (&vs->challenge.value));
  vs->last_challenge_use =
    GNUNET_TIME_UNIT_ZERO_ABS; /* challenge was not yet used */
  update_next_challenge_time (vs, vs->first_challenge_use);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Validation response %s from %s accepted, address valid until %s\n",
              GNUNET_sh2s (&tvr->challenge.value),
              GNUNET_i2s (&cmc->im.sender),
              GNUNET_STRINGS_absolute_time_to_string (vs->valid_until));
  /*memcpy (&hkey,
          &hc,
          sizeof (hkey));*/
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Key %s for address %s map size %u contains %u\n",
              GNUNET_h2s (&vs->hc),
              vs->address,
              GNUNET_CONTAINER_multihashmap_size (revalidation_map),
              GNUNET_CONTAINER_multihashmap_contains (revalidation_map,
                                                      &vs->hc));
  GNUNET_assert (GNUNET_YES ==
                 GNUNET_CONTAINER_multihashmap_put (
                   revalidation_map,
                   &vs->hc,
                   vs,
                   GNUNET_CONTAINER_MULTIHASHMAPOPTION_UNIQUE_ONLY));
  GNUNET_CONTAINER_multihashmap_iterate (revalidation_map,
                                         revalidate_map_it,
                                         NULL);
  vs->revalidation_task =
    GNUNET_SCHEDULER_add_at (GNUNET_TIME_absolute_subtract (vs->next_challenge,
                                                            GNUNET_TIME_UNIT_MINUTES),
                             &revalidation_start_cb, vs);
  vs->sc = GNUNET_PEERSTORE_store (peerstore,
                                   "transport",
                                   &cmc->im.sender,
                                   GNUNET_PEERSTORE_TRANSPORT_URLADDRESS_KEY,
                                   vs->address,
                                   strlen (vs->address) + 1,
                                   vs->valid_until,
                                   GNUNET_PEERSTORE_STOREOPTION_MULTIPLE,
                                   &peerstore_store_validation_cb,
                                   vs);
  finish_cmc_handling (cmc);
 
  /* Finally, we now possibly have a confirmed (!) working queue,
     update queue status (if queue still is around) */
  q = find_queue (&vs->pid, vs->address);
  if (NULL == q)
  {
    GNUNET_STATISTICS_update (GST_stats,
                              "# Queues lost at time of successful validation",
                              1,
                              GNUNET_NO);
    return;
  }
  q->validated_until = vs->validated_until;
  q->pd.aged_rtt = vs->validation_rtt;
  n = q->neighbour;
  vl = lookup_virtual_link (&vs->pid);
  if (NULL == vl)
  {
    vl = GNUNET_new (struct VirtualLink);
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Creating new virtual link %p to %s using direct neighbour!\n",
                vl,
                GNUNET_i2s (&vs->pid));
    vl->burst_addr = NULL;
    vl->confirmed = GNUNET_YES;
    vl->message_uuid_ctr =
      GNUNET_CRYPTO_random_u64 (GNUNET_CRYPTO_QUALITY_WEAK, UINT64_MAX);
    vl->target = n->pid;
    vl->core_recv_window = RECV_WINDOW_SIZE;
    vl->available_fc_window_size = DEFAULT_WINDOW_SIZE;
    vl->incoming_fc_window_size = DEFAULT_WINDOW_SIZE;
    GNUNET_break (GNUNET_YES ==
                  GNUNET_CONTAINER_multipeermap_put (
                    links,
                    &vl->target,
                    vl,
                    GNUNET_CONTAINER_MULTIHASHMAPOPTION_UNIQUE_ONLY));
    vl->n = n;
    n->vl = vl;
    q->idle = GNUNET_YES;
    vl->visibility_task =
      GNUNET_SCHEDULER_add_at (q->validated_until, &check_link_down, vl);
    consider_sending_fc (vl);
    /* We lacked a confirmed connection to the target
       before, so tell CORE about it (finally!) */
    cores_send_connect_info (&n->pid);
    send_msg_from_cache (vl);
  }
  else
  {
    /* Link was already up, remember n is also now available and we are done */
    if (NULL == vl->n)
    {
      vl->n = n;
      n->vl = vl;
      if (GNUNET_YES == vl->confirmed)
        GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                    "Virtual link to %s could now also use direct neighbour!\n",
                    GNUNET_i2s (&vs->pid));
    }
    else
    {
      GNUNET_assert (n == vl->n);
    }
    if (GNUNET_NO == vl->confirmed)
    {
      vl->confirmed = GNUNET_YES;
      q->idle = GNUNET_YES;
      vl->visibility_task =
        GNUNET_SCHEDULER_add_at (q->validated_until, &check_link_down, vl);
      consider_sending_fc (vl);
      /* We lacked a confirmed connection to the target
         before, so tell CORE about it (finally!) */
      cores_send_connect_info (&n->pid);
      send_msg_from_cache (vl);
    }
  }
}

References GNUNET_TIME_Absolute::abs_value_us, ADDRESS_VALIDATION_LIFETIME, VirtualLink::available_fc_window_size, VirtualLink::burst_addr, TransportValidationResponseMessage::challenge, CheckKnownChallengeContext::challenge, check_known_challenge(), check_link_down(), VirtualLink::confirmed, consider_sending_fc(), VirtualLink::core_recv_window, cores_send_connect_info(), DEFAULT_WINDOW_SIZE, find_queue(), finish_cmc_handling(), GNUNET_assert, GNUNET_break, GNUNET_break_op, GNUNET_CONTAINER_multihashmap_contains(), GNUNET_CONTAINER_multihashmap_iterate(), GNUNET_CONTAINER_multihashmap_put(), GNUNET_CONTAINER_multihashmap_size(), GNUNET_CONTAINER_MULTIHASHMAPOPTION_UNIQUE_ONLY, GNUNET_CONTAINER_multipeermap_get_multiple(), GNUNET_CONTAINER_multipeermap_put(), GNUNET_CRYPTO_eddsa_verify, GNUNET_CRYPTO_QUALITY_NONCE, GNUNET_CRYPTO_QUALITY_WEAK, GNUNET_CRYPTO_random_block(), GNUNET_CRYPTO_random_u64(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_h2s(), GNUNET_i2s(), GNUNET_log, GNUNET_new, GNUNET_NO, GNUNET_OK, GNUNET_PEERSTORE_store(), GNUNET_PEERSTORE_STOREOPTION_MULTIPLE, GNUNET_PEERSTORE_TRANSPORT_URLADDRESS_KEY, GNUNET_SCHEDULER_add_at(), GNUNET_sh2s(), GNUNET_SIGNATURE_PURPOSE_TRANSPORT_CHALLENGE, GNUNET_STATISTICS_update(), GNUNET_STRINGS_absolute_time_to_string(), GNUNET_TIME_absolute_cmp, GNUNET_TIME_absolute_get_duration(), GNUNET_TIME_absolute_get_monotonic(), GNUNET_TIME_absolute_min(), GNUNET_TIME_absolute_ntoh(), GNUNET_TIME_absolute_subtract(), GNUNET_TIME_relative_min(), GNUNET_TIME_relative_multiply(), GNUNET_TIME_relative_ntoh(), GNUNET_TIME_relative_to_absolute(), GNUNET_TIME_UNIT_MINUTES, GNUNET_TIME_UNIT_ZERO, GNUNET_TIME_UNIT_ZERO_ABS, GNUNET_YES, GST_cfg, GST_stats, CommunicatorMessageContext::im, VirtualLink::incoming_fc_window_size, links, lookup_virtual_link(), MAX_ADDRESS_VALID_UNTIL, VirtualLink::message_uuid_ctr, VirtualLink::n, peerstore, peerstore_store_validation_cb(), Neighbour::pid, ValidationState::pid, GNUNET_PeerIdentity::public_key, GNUNET_CRYPTO_EccSignaturePurpose::purpose, TransportValidationPS::purpose, q, RECV_WINDOW_SIZE, revalidate_map_it(), revalidation_map, revalidation_start_cb(), send_msg_from_cache(), GNUNET_TRANSPORT_IncomingMessage::sender, TransportValidationResponseMessage::signature, VirtualLink::target, update_next_challenge_time(), ValidationState::valid_until, ValidationState::validated_until, validation_map, VALIDATION_RTT_BUFFER_FACTOR, GNUNET_CRYPTO_ChallengeNonceP::value, VirtualLink::visibility_task, Neighbour::vl, and CheckKnownChallengeContext::vs.

Here is the call graph for this function:

handle_incoming_msg()

static void handle_incoming_msg ( void *  cls, const struct GNUNET_TRANSPORT_IncomingMessage *  im  )

static

Incoming message.

Process the request.

Parameters

im	the send message that was received

Definition at line 10137 of file gnunet-service-transport.c.

{
  struct TransportClient *tc = cls;
  struct CommunicatorMessageContext *cmc =
    GNUNET_new (struct CommunicatorMessageContext);
 
  cmc->tc = tc;
  cmc->im = *im;
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Received message with size %u and flow control id %" PRIu64
              " via communicator from peer %s\n",
              ntohs (im->header.size),
              im->fc_id,
              GNUNET_i2s (&im->sender));
  cmc->im.neighbour_sender = cmc->im.sender;
  cmc->mh = (const struct GNUNET_MessageHeader *) &im[1];
  demultiplex_with_cmc (cmc);
}

References demultiplex_with_cmc(), GNUNET_TRANSPORT_IncomingMessage::fc_id, GNUNET_ERROR_TYPE_DEBUG, GNUNET_i2s(), GNUNET_log, GNUNET_new, GNUNET_TRANSPORT_IncomingMessage::header, CommunicatorMessageContext::im, CommunicatorMessageContext::mh, GNUNET_TRANSPORT_IncomingMessage::neighbour_sender, GNUNET_TRANSPORT_IncomingMessage::sender, GNUNET_MessageHeader::size, tc, and CommunicatorMessageContext::tc.

Here is the call graph for this function:

check_flow_control()

static int check_flow_control ( void *  cls, const struct TransportFlowControlMessage *  fc  )

static

Communicator gave us a transport address validation response.

Check the request.

Parameters

cls	a struct CommunicatorMessageContext
fc	the message that was received

Returns

GNUNET_YES if message is well-formed

Definition at line 10167 of file gnunet-service-transport.c.

{
  unsigned int number_of_addresses = ntohl (fc->number_of_addresses);
  (void) cls;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Flow control header size %u size of addresses %u number of addresses %u size of message struct %lu second struct %lu\n",
              ntohs (fc->header.size),
              ntohl (fc->size_of_addresses),
              ntohl (fc->number_of_addresses),
              sizeof(struct TransportFlowControlMessage),
              sizeof (struct TransportGlobalNattedAddress));
 
  if (0 == number_of_addresses || ntohs (fc->header.size) == sizeof(struct
                                                                    TransportFlowControlMessage)
      + ntohl (fc->number_of_addresses) * sizeof (struct
                                                  TransportGlobalNattedAddress)
      + ntohl (fc->size_of_addresses))
    return GNUNET_OK;
  else
  {
    GNUNET_break_op (0);
    return GNUNET_SYSERR;
  }
}

References GNUNET_break_op, GNUNET_ERROR_TYPE_DEBUG, GNUNET_log, GNUNET_OK, GNUNET_SYSERR, TransportFlowControlMessage::header, TransportFlowControlMessage::number_of_addresses, GNUNET_MessageHeader::size, and TransportFlowControlMessage::size_of_addresses.

iterate_address_start_burst()

static void iterate_address_start_burst ( void *  cls, const struct GNUNET_PeerIdentity *  pid, const char *  uri  )

static

Definition at line 10222 of file gnunet-service-transport.c.

{
  struct VirtualLink *vl = cls;
  const char *slash;
  char *address_uri;
  char *prefix;
  char *uri_without_port;
 
  slash = strrchr (uri, '/');
  prefix = GNUNET_strndup (uri, (slash - uri) - 2);
  GNUNET_assert (NULL != slash);
  slash++;
  GNUNET_asprintf (&address_uri,
                   "%s-%s",
                   prefix,
                   slash);
 
  uri_without_port = get_address_without_port (address_uri);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "iterate_address_start_burst %s %s %s %s\n",
              uri_without_port,
              uri,
              address_uri,
              slash);
  if (0 == strcmp (uri_without_port, slash))
  {
    vl->burst_addr = GNUNET_strndup (uri_without_port, strlen (uri_without_port)
                                     );
  }
  else
    vl->burst_addr = NULL;
 
  GNUNET_free (prefix);
  GNUNET_free (uri_without_port);
}

References VirtualLink::burst_addr, get_address_without_port(), GNUNET_asprintf(), GNUNET_assert, GNUNET_ERROR_TYPE_DEBUG, GNUNET_free, GNUNET_log, GNUNET_strndup, prefix, and uri.

Referenced by check_for_burst_address().

Here is the call graph for this function:

Here is the caller graph for this function:

check_for_burst_address()

static void check_for_burst_address ( void *  cls, const struct GNUNET_PEERSTORE_Record *  record, const char *  emsg  )

static

Definition at line 10262 of file gnunet-service-transport.c.

{
  struct GNUNET_StartBurstCls *sb_cls = cls;
  struct VirtualLink *vl = sb_cls->vl;
  struct GNUNET_MessageHeader *hello;
  struct GNUNET_HELLO_Parser *parser;
 
  if (NULL != emsg)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
                "Got failure from PEERSTORE: %s\n",
                emsg);
    return;
  }
  if (NULL == record)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Hello iteration end for %s\n",
                GNUNET_i2s (&vl->target));
    vl->ic = NULL;
    GNUNET_free (sb_cls);
    return;
  }
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "check_for_burst_address\n");
  hello = record->value;
  parser = GNUNET_HELLO_parser_from_msg (hello);
  GNUNET_HELLO_parser_iterate (parser,
                               &iterate_address_start_burst,
                               vl);
  GNUNET_HELLO_parser_free (parser);
 
  GNUNET_PEERSTORE_iteration_stop (vl->ic);
  GNUNET_free (sb_cls);
}

References GNUNET_ERROR_TYPE_DEBUG, GNUNET_ERROR_TYPE_WARNING, GNUNET_free, GNUNET_HELLO_parser_free(), GNUNET_HELLO_parser_from_msg(), GNUNET_HELLO_parser_iterate(), GNUNET_i2s(), GNUNET_log, GNUNET_PEERSTORE_iteration_stop(), VirtualLink::ic, iterate_address_start_burst(), record(), VirtualLink::sb_cls, VirtualLink::target, and GNUNET_StartBurstCls::vl.

Referenced by queue_burst().

Here is the call graph for this function:

Here is the caller graph for this function:

burst_timeout()

static void burst_timeout ( void *  cls )

static

Definition at line 10303 of file gnunet-service-transport.c.

{
  burst_running = GNUNET_NO;
}

References burst_running, and GNUNET_NO.

Referenced by start_burst().

Here is the caller graph for this function:

start_burst()

static void start_burst ( void *  cls )

static

Definition at line 10310 of file gnunet-service-transport.c.

{
  struct GNUNET_StartBurstCls *sb_cls = cls;
  struct VirtualLink *vl = sb_cls->vl;
  struct GNUNET_TRANSPORT_StartBurst *sb;
  struct GNUNET_MQ_Envelope *env;
  char *uri_without_port = vl->burst_addr;
 
  burst_task = NULL;
  /*char buf[strlen (uri_without_port) + 1];
 
  GNUNET_memcpy (buf, uri_without_port,  strlen (uri_without_port));
  buf[strlen (uri_without_port)] = '\0';*/
  env =
    GNUNET_MQ_msg_extra (sb,
                         strlen (uri_without_port) + 1,
                         GNUNET_MESSAGE_TYPE_TRANSPORT_START_BURST);
  sb->rtt = GNUNET_TIME_relative_hton (sb_cls->rtt);
  sb->pid = vl->target;
  memcpy (&sb[1], uri_without_port, strlen (uri_without_port) + 1);
  for (struct TransportClient *tc = clients_head; NULL != tc; tc = tc->next)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "iterate_address_start_burst client tc prefix %s\n",
                tc->details.communicator.address_prefix);
    if (CT_COMMUNICATOR != tc->type)
      continue;
    if (GNUNET_YES == tc->details.communicator.can_burst)
    {
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "iterate_address_start_burst %s call %lu %u rtt %lu\n",
                  uri_without_port,
                  strlen (uri_without_port),
                  ntohs (sb->header.size),
                  (unsigned long) sb_cls->rtt.rel_value_us);
      GNUNET_MQ_send (tc->mq, env);
      burst_running = GNUNET_YES;
      burst_timeout_task = GNUNET_SCHEDULER_add_delayed (
        GNUNET_TIME_relative_multiply (GNUNET_TIME_UNIT_SECONDS,
                                       60),
        &burst_timeout,
        NULL);
      // TODO We need some algo to choose from available communicators. Can we run two bursts at once? Atm we only implemented udp burst.
      break;
    }
  }
  GNUNET_free (env);
  GNUNET_free (sb_cls);
}

References VirtualLink::burst_addr, burst_running, burst_task, burst_timeout(), burst_timeout_task, clients_head, CT_COMMUNICATOR, env, GNUNET_ERROR_TYPE_DEBUG, GNUNET_free, GNUNET_log, GNUNET_MESSAGE_TYPE_TRANSPORT_START_BURST, GNUNET_MQ_msg_extra, GNUNET_MQ_send(), GNUNET_SCHEDULER_add_delayed(), GNUNET_TIME_relative_hton(), GNUNET_TIME_relative_multiply(), GNUNET_TIME_UNIT_SECONDS, GNUNET_YES, GNUNET_TRANSPORT_StartBurst::header, GNUNET_TRANSPORT_StartBurst::pid, GNUNET_TIME_Relative::rel_value_us, GNUNET_StartBurstCls::rtt, GNUNET_TRANSPORT_StartBurst::rtt, VirtualLink::sb_cls, GNUNET_MessageHeader::size, VirtualLink::target, tc, and GNUNET_StartBurstCls::vl.

Referenced by queue_burst().

Here is the call graph for this function:

Here is the caller graph for this function:

queue_burst()

static void queue_burst ( void *  cls )

static

Definition at line 10362 of file gnunet-service-transport.c.

{
  struct GNUNET_StartBurstCls *sb_cls = cls;
  struct VirtualLink *vl = sb_cls->vl;
 
  if (GNUNET_YES != use_burst)
    return;
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "burst_task %p ready %s burst addr %s (%p)\n",
              burst_task,
              sb_cls->sync_ready ? "yes" : "no",
              vl->burst_addr,
              vl->burst_addr);
  if (NULL != burst_task && GNUNET_NO == sb_cls->sync_ready)
  {
    GNUNET_SCHEDULER_cancel (burst_task);
    burst_task = NULL;
    GNUNET_free (sb_cls);
    return;
  }
  if (GNUNET_NO == burst_running && NULL != vl->burst_addr && NULL == burst_task
      )
  {
    burst_task = GNUNET_SCHEDULER_add_delayed (sb_cls->delay,
                                               &start_burst,
                                               sb_cls);
  }
  else if (NULL == vl->burst_addr)
  {
    vl->ic = GNUNET_PEERSTORE_iteration_start (peerstore,
                                               "peerstore",
                                               &vl->target,
                                               GNUNET_PEERSTORE_HELLO_KEY,
                                               check_for_burst_address,
                                               sb_cls);
  }
}

References VirtualLink::burst_addr, burst_running, burst_task, check_for_burst_address(), GNUNET_StartBurstCls::delay, GNUNET_ERROR_TYPE_DEBUG, GNUNET_free, GNUNET_log, GNUNET_NO, GNUNET_PEERSTORE_HELLO_KEY, GNUNET_PEERSTORE_iteration_start(), GNUNET_SCHEDULER_add_delayed(), GNUNET_SCHEDULER_cancel(), GNUNET_YES, VirtualLink::ic, peerstore, VirtualLink::sb_cls, start_burst(), GNUNET_StartBurstCls::sync_ready, VirtualLink::target, use_burst, and GNUNET_StartBurstCls::vl.

Referenced by handle_flow_control().

Here is the call graph for this function:

Here is the caller graph for this function:

handle_flow_control()

static void handle_flow_control ( void *  cls, const struct TransportFlowControlMessage *  fc  )

static

Communicator gave us a transport address validation response.

Process the request.

Parameters

cls	a struct CommunicatorMessageContext (must call finish_cmc_handling() when done)
fc	the message that was received

Definition at line 10410 of file gnunet-service-transport.c.

{
  struct CommunicatorMessageContext *cmc = cls;
  struct VirtualLink *vl;
  struct GNUNET_TIME_Absolute q_timeout = GNUNET_TIME_UNIT_ZERO_ABS;
  uint32_t seq;
  struct GNUNET_TIME_Absolute st;
  uint64_t os;
  uint64_t wnd;
  uint32_t random;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Received FC from %s\n", GNUNET_i2s (&cmc->im.sender));
  vl = lookup_virtual_link (&cmc->im.sender);
  if (NULL == vl)
  {
    vl = GNUNET_new (struct VirtualLink);
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "No virtual link for %p FC creating new unconfirmed virtual link to %s!\n",
                vl,
                GNUNET_i2s (&cmc->im.sender));
    vl->burst_addr = NULL;
    vl->confirmed = GNUNET_NO;
    vl->message_uuid_ctr =
      GNUNET_CRYPTO_random_u64 (GNUNET_CRYPTO_QUALITY_WEAK, UINT64_MAX);
    vl->target = cmc->im.sender;
    vl->core_recv_window = RECV_WINDOW_SIZE;
    vl->available_fc_window_size = DEFAULT_WINDOW_SIZE;
    vl->incoming_fc_window_size = DEFAULT_WINDOW_SIZE;
    GNUNET_break (GNUNET_YES ==
                  GNUNET_CONTAINER_multipeermap_put (
                    links,
                    &vl->target,
                    vl,
                    GNUNET_CONTAINER_MULTIHASHMAPOPTION_UNIQUE_ONLY));
  }
  if (NULL != vl->n)
  {
    for (struct Queue *q = vl->n->queue_head; NULL != q; q = q->next_neighbour)
      q_timeout = GNUNET_TIME_absolute_max (q_timeout, q->validated_until);
  }
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "remaining %lu timeout for neighbour %p\n",
              (unsigned long) GNUNET_TIME_absolute_get_remaining (q_timeout).
              rel_value_us,
              vl->n);
  if (NULL == vl->n ||
      0 == GNUNET_TIME_absolute_get_remaining (q_timeout).rel_value_us)
  {
    struct GNUNET_TIME_Relative rtt;
    struct GNUNET_BurstSync burst_sync;
    struct GNUNET_StartBurstCls *bcls;
 
    bcls = GNUNET_new (struct GNUNET_StartBurstCls);
    bcls->vl = vl;
    vl->sb_cls = bcls;
    if (NULL != vl->dv)
      rtt = calculate_rtt (vl->dv);
    else
      rtt = GNUNET_TIME_UNIT_FOREVER_REL;
    burst_sync.rtt_average = fc->rtt;
    bcls->rtt = GNUNET_TIME_relative_ntoh (burst_sync.rtt_average);
    burst_sync.sync_ready = fc->sync_ready;
 
    GNUNET_is_burst_ready (rtt,
                           &burst_sync,
                           &queue_burst,
                           bcls);
  }
  if (0 != ntohl (fc->number_of_addresses))
  {
    unsigned int number_of_addresses = ntohl (fc->number_of_addresses);
    const char *tgnas;
    unsigned int off = 0;
 
    tgnas = (const char *) &fc[1];
 
    for (int i = 1; i <= number_of_addresses; i++)
    {
      struct TransportGlobalNattedAddress *tgna;
      char *addr;
      unsigned int address_length;
 
      tgna = (struct TransportGlobalNattedAddress*) &tgnas[off];
      addr = (char *) &tgna[1];
      address_length = ntohl (tgna->address_length);
      off += sizeof(struct TransportGlobalNattedAddress) + address_length;
 
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "received address %s length %u\n",
                  addr,
                  ntohl (tgna->address_length));
 
      GNUNET_NAT_add_global_address (nh, addr, ntohl (tgna->address_length));
    }
  }
  st = GNUNET_TIME_absolute_ntoh (fc->sender_time);
  if (st.abs_value_us < vl->last_fc_timestamp.abs_value_us)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "FC dropped: Message out of order\n");
    /* out of order, drop */
    GNUNET_STATISTICS_update (GST_stats,
                              "# FC dropped: message out of order",
                              1,
                              GNUNET_NO);
    finish_cmc_handling (cmc);
    return;
  }
  seq = ntohl (fc->seq);
  if (seq < vl->last_fc_seq)
  {
    /* Wrap-around/reset of other peer; start all counters from zero */
    vl->outbound_fc_window_size_used = 0;
  }
  vl->last_fc_seq = seq;
  vl->last_fc_timestamp = st;
  vl->outbound_fc_window_size = GNUNET_ntohll (fc->inbound_window_size);
  os = GNUNET_ntohll (fc->outbound_sent);
  vl->incoming_fc_window_size_loss =
    (int64_t) (os - vl->incoming_fc_window_size_used);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Received FC from %s, seq %u, new window %llu (loss at %lld)\n",
              GNUNET_i2s (&vl->target),
              (unsigned int) seq,
              (unsigned long long) vl->outbound_fc_window_size,
              (long long) vl->incoming_fc_window_size_loss);
  wnd = GNUNET_ntohll (fc->outbound_window_size);
  random = GNUNET_CRYPTO_random_u32 (GNUNET_CRYPTO_QUALITY_WEAK,
                                     UINT32_MAX);
  if ((GNUNET_YES == vl->confirmed) && ((wnd < vl->incoming_fc_window_size
                                         + vl->incoming_fc_window_size_used
                                         + vl->incoming_fc_window_size_loss) ||
                                        (vl->last_outbound_window_size_received
                                         != wnd) ||
                                        (0 == random
                                         % FC_NO_CHANGE_REPLY_PROBABILITY)))
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Consider re-sending our FC message, as clearly the other peer's idea of the window is not up-to-date (%llu vs %llu) or %llu last received differs, or random reply %u\n",
                (unsigned long long) wnd,
                (unsigned long long) vl->incoming_fc_window_size,
                (unsigned long long) vl->last_outbound_window_size_received,
                random % FC_NO_CHANGE_REPLY_PROBABILITY);
    consider_sending_fc (vl);
  }
  if ((wnd == vl->incoming_fc_window_size
       + vl->incoming_fc_window_size_used
       + vl->incoming_fc_window_size_loss) &&
      (vl->last_outbound_window_size_received == wnd) &&
      (NULL != vl->fc_retransmit_task))
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Stopping FC retransmission to %s: peer is current at window %llu\n",
                GNUNET_i2s (&vl->target),
                (unsigned long long) wnd);
    GNUNET_SCHEDULER_cancel (vl->fc_retransmit_task);
    vl->fc_retransmit_task = NULL;
    vl->fc_retransmit_count = 0;
  }
  vl->last_outbound_window_size_received = wnd;
  /* FC window likely increased, check transmission possibilities! */
  check_vl_transmission (vl);
  finish_cmc_handling (cmc);
}

References GNUNET_TIME_Absolute::abs_value_us, TransportGlobalNattedAddress::address_length, VirtualLink::available_fc_window_size, VirtualLink::burst_addr, calculate_rtt(), check_vl_transmission(), VirtualLink::confirmed, consider_sending_fc(), VirtualLink::core_recv_window, DEFAULT_WINDOW_SIZE, VirtualLink::dv, FC_NO_CHANGE_REPLY_PROBABILITY, VirtualLink::fc_retransmit_count, VirtualLink::fc_retransmit_task, finish_cmc_handling(), GNUNET_break, GNUNET_CONTAINER_MULTIHASHMAPOPTION_UNIQUE_ONLY, GNUNET_CONTAINER_multipeermap_put(), GNUNET_CRYPTO_QUALITY_WEAK, GNUNET_CRYPTO_random_u32(), GNUNET_CRYPTO_random_u64(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_i2s(), GNUNET_is_burst_ready(), GNUNET_log, GNUNET_NAT_add_global_address(), GNUNET_new, GNUNET_NO, GNUNET_ntohll(), GNUNET_SCHEDULER_cancel(), GNUNET_STATISTICS_update(), GNUNET_TIME_absolute_get_remaining(), GNUNET_TIME_absolute_max(), GNUNET_TIME_absolute_ntoh(), GNUNET_TIME_relative_ntoh(), GNUNET_TIME_UNIT_FOREVER_REL, GNUNET_TIME_UNIT_ZERO_ABS, GNUNET_YES, GST_stats, CommunicatorMessageContext::im, TransportFlowControlMessage::inbound_window_size, VirtualLink::incoming_fc_window_size, VirtualLink::incoming_fc_window_size_loss, VirtualLink::incoming_fc_window_size_used, VirtualLink::last_fc_seq, VirtualLink::last_fc_timestamp, VirtualLink::last_outbound_window_size_received, links, lookup_virtual_link(), VirtualLink::message_uuid_ctr, VirtualLink::n, nh, TransportFlowControlMessage::number_of_addresses, VirtualLink::outbound_fc_window_size, VirtualLink::outbound_fc_window_size_used, TransportFlowControlMessage::outbound_sent, TransportFlowControlMessage::outbound_window_size, q, queue_burst(), Neighbour::queue_head, RECV_WINDOW_SIZE, GNUNET_StartBurstCls::rtt, TransportFlowControlMessage::rtt, GNUNET_BurstSync::rtt_average, VirtualLink::sb_cls, GNUNET_TRANSPORT_IncomingMessage::sender, TransportFlowControlMessage::sender_time, TransportFlowControlMessage::seq, st, GNUNET_BurstSync::sync_ready, TransportFlowControlMessage::sync_ready, VirtualLink::target, and GNUNET_StartBurstCls::vl.

Here is the call graph for this function:

check_add_queue_message()

static int check_add_queue_message ( void *  cls, const struct GNUNET_TRANSPORT_AddQueueMessage *  aqm  )

static

New queue became available.

Check message.

Parameters

cls	the client
aqm	the send message that was sent

Definition at line 10658 of file gnunet-service-transport.c.

{
  struct TransportClient *tc = cls;
 
  if (CT_COMMUNICATOR != tc->type)
  {
    GNUNET_break (0);
    return GNUNET_SYSERR;
  }
  GNUNET_MQ_check_zero_termination (aqm);
  return GNUNET_OK;
}

References CT_COMMUNICATOR, GNUNET_break, GNUNET_MQ_check_zero_termination, GNUNET_OK, GNUNET_SYSERR, and tc.

set_pending_message_uuid()

static void set_pending_message_uuid ( struct PendingMessage *  pm )

static

If necessary, generates the UUID for a pm.

Parameters

pm	pending message to generate UUID for.

Definition at line 10679 of file gnunet-service-transport.c.

{
  if (pm->msg_uuid_set)
    return;
  pm->msg_uuid.uuid = pm->vl->message_uuid_ctr++;
  pm->msg_uuid_set = GNUNET_YES;
}

References GNUNET_YES, and pm.

Referenced by extract_box_cb(), fragment_message(), and reliability_box_message().

Here is the caller graph for this function:

prepare_pending_acknowledgement()

static struct PendingAcknowledgement * prepare_pending_acknowledgement ( struct Queue *  queue, struct DistanceVectorHop *  dvh, struct PendingMessage *  pm  )

static

Setup data structure waiting for acknowledgements.

Parameters

queue	queue the pm will be sent over
dvh	path the message will take, may be NULL
pm	the pending message for transmission

Returns

corresponding fresh pending acknowledgement

Definition at line 10697 of file gnunet-service-transport.c.

{
  struct PendingAcknowledgement *pa;
 
  pa = GNUNET_new (struct PendingAcknowledgement);
  pa->queue = queue;
  pa->dvh = dvh;
  pa->pm = pm;
  do
  {
    GNUNET_CRYPTO_random_block (GNUNET_CRYPTO_QUALITY_NONCE,
                                &pa->ack_uuid,
                                sizeof(pa->ack_uuid));
  }
  while (GNUNET_YES != GNUNET_CONTAINER_multiuuidmap_put (
           pending_acks,
           &pa->ack_uuid.value,
           pa,
           GNUNET_CONTAINER_MULTIHASHMAPOPTION_UNIQUE_ONLY));
  GNUNET_CONTAINER_MDLL_insert (queue, queue->pa_head, queue->pa_tail, pa);
  GNUNET_CONTAINER_MDLL_insert (pm, pm->pa_head, pm->pa_tail, pa);
  if (NULL != dvh)
    GNUNET_CONTAINER_MDLL_insert (dvh, dvh->pa_head, dvh->pa_tail, pa);
  pa->transmission_time = GNUNET_TIME_absolute_get ();
  pa->message_size = pm->bytes_msg;
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Waiting for ACKnowledgment `%s' for <%" PRIu64 ">\n",
              GNUNET_uuid2s (&pa->ack_uuid.value),
              pm->logging_uuid);
  return pa;
}

References PendingAcknowledgement::ack_uuid, PendingAcknowledgement::dvh, GNUNET_CONTAINER_MDLL_insert, GNUNET_CONTAINER_MULTIHASHMAPOPTION_UNIQUE_ONLY, GNUNET_CONTAINER_multiuuidmap_put(), GNUNET_CRYPTO_QUALITY_NONCE, GNUNET_CRYPTO_random_block(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_log, GNUNET_new, GNUNET_TIME_absolute_get(), GNUNET_uuid2s(), GNUNET_YES, PendingAcknowledgement::message_size, DistanceVectorHop::pa_head, DistanceVectorHop::pa_tail, pending_acks, PendingAcknowledgement::pm, pm, queue(), PendingAcknowledgement::queue, PendingAcknowledgement::transmission_time, and AcknowledgementUUIDP::value.

Referenced by fragment_message(), and reliability_box_message().

Here is the call graph for this function:

Here is the caller graph for this function:

fragment_message()

static struct PendingMessage * fragment_message ( struct Queue *  queue, struct DistanceVectorHop *  dvh, struct PendingMessage *  pm  )

static

Fragment the given pm to the given mtu.

Adds additional fragments to the neighbour as well. If the mtu is too small, generates and error for the pm and returns NULL.

Parameters

queue	which queue to fragment for
dvh	path the message will take, or NULL
pm	pending message to fragment for transmission

Returns

new message to transmit

Definition at line 10744 of file gnunet-service-transport.c.

{
  struct PendingAcknowledgement *pa;
  struct PendingMessage *ff;
  uint16_t mtu;
  uint16_t msize;
 
  mtu = (UINT16_MAX == queue->mtu)
        ? UINT16_MAX - sizeof(struct GNUNET_TRANSPORT_SendMessageTo)
        : queue->mtu;
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Fragmenting message <%" PRIu64
              "> with size %u to %s for MTU %u\n",
              pm->logging_uuid,
              pm->bytes_msg,
              GNUNET_i2s (&pm->vl->target),
              (unsigned int) mtu);
  set_pending_message_uuid (pm);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Fragmenting message %" PRIu64 " <%" PRIu64
              "> with size %u to %s for MTU %u\n",
              pm->msg_uuid.uuid,
              pm->logging_uuid,
              pm->bytes_msg,
              GNUNET_i2s (&pm->vl->target),
              (unsigned int) mtu);
 
  /* This invariant is established in #handle_add_queue_message() */
  GNUNET_assert (mtu > sizeof(struct TransportFragmentBoxMessage));
 
  /* select fragment for transmission, descending the tree if it has
     been expanded until we are at a leaf or at a fragment that is small
     enough
   */
  ff = pm;
  msize = ff->bytes_msg;
 
  while (((ff->bytes_msg > mtu) || (pm == ff)) &&
         (ff->frag_off == msize) && (NULL != ff->head_frag))
  {
    ff = ff->head_frag;   /* descent into fragmented fragments */
    msize = ff->bytes_msg - sizeof(struct TransportFragmentBoxMessage);
  }
 
  if (((ff->bytes_msg > mtu) || (pm == ff)) && (ff->frag_off < msize))
  {
    /* Did not yet calculate all fragments, calculate next fragment */
    struct PendingMessage *frag;
    struct TransportFragmentBoxMessage tfb;
    const char *orig;
    char *msg;
    uint16_t fragmax;
    uint16_t fragsize;
    uint16_t msize_ff;
    uint16_t xoff = 0;
    pm->frag_count++;
 
    orig = (const char *) &ff[1];
    msize_ff = ff->bytes_msg;
    if (pm != ff)
    {
      const struct TransportFragmentBoxMessage *tfbo;
 
      tfbo = (const struct TransportFragmentBoxMessage *) orig;
      orig += sizeof(struct TransportFragmentBoxMessage);
      msize_ff -= sizeof(struct TransportFragmentBoxMessage);
      xoff = ntohs (tfbo->frag_off);
    }
    fragmax = mtu - sizeof(struct TransportFragmentBoxMessage);
    fragsize = GNUNET_MIN (msize_ff - ff->frag_off, fragmax);
    frag =
      GNUNET_malloc (sizeof(struct PendingMessage)
                     + sizeof(struct TransportFragmentBoxMessage) + fragsize);
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "3 created pm %p from pm %p storing vl %p from pm %p\n",
                frag,
                ff,
                pm->vl,
                pm);
    frag->logging_uuid = logging_uuid_gen++;
    frag->vl = pm->vl;
    frag->frag_parent = ff;
    frag->timeout = pm->timeout;
    frag->bytes_msg = sizeof(struct TransportFragmentBoxMessage) + fragsize;
    frag->pmt = PMT_FRAGMENT_BOX;
    msg = (char *) &frag[1];
    tfb.header.type = htons (GNUNET_MESSAGE_TYPE_TRANSPORT_FRAGMENT);
    tfb.header.size =
      htons (sizeof(struct TransportFragmentBoxMessage) + fragsize);
    pa = prepare_pending_acknowledgement (queue, dvh, frag);
    tfb.ack_uuid = pa->ack_uuid;
    tfb.msg_uuid = pm->msg_uuid;
    tfb.frag_off = htons (ff->frag_off + xoff);
    tfb.msg_size = htons (pm->bytes_msg);
    memcpy (msg, &tfb, sizeof(tfb));
    memcpy (&msg[sizeof(tfb)], &orig[ff->frag_off], fragsize);
    GNUNET_CONTAINER_MDLL_insert (frag, ff->head_frag,
                                  ff->tail_frag, frag);
    ff->frag_off += fragsize;
    ff = frag;
  }
 
  /* Move head to the tail and return it */
  GNUNET_CONTAINER_MDLL_remove (frag,
                                ff->frag_parent->head_frag,
                                ff->frag_parent->tail_frag,
                                ff);
  GNUNET_CONTAINER_MDLL_insert_tail (frag,
                                     ff->frag_parent->head_frag,
                                     ff->frag_parent->tail_frag,
                                     ff);
 
  return ff;
}

References TransportFragmentBoxMessage::ack_uuid, PendingAcknowledgement::ack_uuid, PendingMessage::bytes_msg, TransportFragmentBoxMessage::frag_off, PendingMessage::frag_off, PendingMessage::frag_parent, GNUNET_assert, GNUNET_CONTAINER_MDLL_insert, GNUNET_CONTAINER_MDLL_insert_tail, GNUNET_CONTAINER_MDLL_remove, GNUNET_ERROR_TYPE_DEBUG, GNUNET_i2s(), GNUNET_log, GNUNET_malloc, GNUNET_MESSAGE_TYPE_TRANSPORT_FRAGMENT, GNUNET_MIN, PendingMessage::head_frag, TransportFragmentBoxMessage::header, PendingMessage::logging_uuid, logging_uuid_gen, msg, TransportFragmentBoxMessage::msg_size, TransportFragmentBoxMessage::msg_uuid, pm, PendingMessage::pmt, PMT_FRAGMENT_BOX, prepare_pending_acknowledgement(), queue(), set_pending_message_uuid(), GNUNET_MessageHeader::size, PendingMessage::tail_frag, PendingMessage::timeout, GNUNET_MessageHeader::type, and PendingMessage::vl.

Referenced by transmit_on_queue().

Here is the call graph for this function:

Here is the caller graph for this function:

reliability_box_message()

static struct PendingMessage * reliability_box_message ( struct Queue *  queue, struct DistanceVectorHop *  dvh, struct PendingMessage *  pm  )

static

Reliability-box the given pm.

On error (can there be any), NULL may be returned, otherwise the "replacement" for pm (which should then be added to the respective neighbour's queue instead of pm). If the pm is already fragmented or reliability boxed, or itself an ACK, this function simply returns pm.

Parameters

queue	which queue to prepare transmission for
dvh	path the message will take, or NULL
pm	pending message to box for transmission over unreliabile queue

Returns

new message to transmit

Definition at line 10875 of file gnunet-service-transport.c.

{
  struct TransportReliabilityBoxMessage rbox;
  struct PendingAcknowledgement *pa;
  struct PendingMessage *bpm;
  char *msg;
 
  if ((PMT_CORE != pm->pmt) && (PMT_DV_BOX != pm->pmt))
    return pm; /* already fragmented or reliability boxed, or control message:
                  do nothing */
  if (NULL != pm->bpm)
    return pm->bpm; /* already computed earlier: do nothing */
  // TODO I guess we do not need this assertion. We might have a DLL with
  // fragments, because the MTU changed, and we do not need to fragment anymore.
  // But we should keep the fragments until message was completed, because
  // the MTU might change again.
  // GNUNET_assert (NULL == pm->head_frag);
  if (pm->bytes_msg + sizeof(rbox) > UINT16_MAX)
  {
    /* failed hard */
    GNUNET_break (0);
    client_send_response (pm);
    return NULL;
  }
 
  pa = prepare_pending_acknowledgement (queue, dvh, pm);
 
  bpm = GNUNET_malloc (sizeof(struct PendingMessage) + sizeof(rbox)
                       + pm->bytes_msg);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "4 created pm %p storing vl %p from pm %p\n",
              bpm,
              pm->vl,
              pm);
  bpm->logging_uuid = logging_uuid_gen++;
  bpm->vl = pm->vl;
  bpm->frag_parent = pm;
  // Why was this needed?
  // GNUNET_CONTAINER_MDLL_insert (frag, pm->head_frag, pm->tail_frag, bpm);
  bpm->timeout = pm->timeout;
  bpm->pmt = PMT_RELIABILITY_BOX;
  bpm->bytes_msg = pm->bytes_msg + sizeof(rbox);
  set_pending_message_uuid (bpm);
  rbox.header.type = htons (GNUNET_MESSAGE_TYPE_TRANSPORT_RELIABILITY_BOX);
  rbox.header.size = htons (sizeof(rbox) + pm->bytes_msg);
  rbox.ack_countdown = htonl (0);  // FIXME: implement ACK countdown support
 
  rbox.ack_uuid = pa->ack_uuid;
  msg = (char *) &bpm[1];
  memcpy (msg, &rbox, sizeof(rbox));
  memcpy (&msg[sizeof(rbox)], &pm[1], pm->bytes_msg);
  pm->bpm = bpm;
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Preparing reliability box for message <%" PRIu64
              "> of size %d (%d) to %s on queue %s\n",
              pm->logging_uuid,
              pm->bytes_msg,
              ntohs (((const struct GNUNET_MessageHeader *) &pm[1])->size),
              GNUNET_i2s (&pm->vl->target),
              queue->address);
  return bpm;
}

References TransportReliabilityBoxMessage::ack_countdown, TransportReliabilityBoxMessage::ack_uuid, PendingAcknowledgement::ack_uuid, PendingMessage::bpm, PendingMessage::bytes_msg, client_send_response(), PendingMessage::frag_parent, GNUNET_break, GNUNET_ERROR_TYPE_DEBUG, GNUNET_i2s(), GNUNET_log, GNUNET_malloc, GNUNET_MESSAGE_TYPE_TRANSPORT_RELIABILITY_BOX, TransportReliabilityBoxMessage::header, PendingMessage::logging_uuid, logging_uuid_gen, msg, pm, PendingMessage::pmt, PMT_CORE, PMT_DV_BOX, PMT_RELIABILITY_BOX, prepare_pending_acknowledgement(), queue(), set_pending_message_uuid(), GNUNET_MessageHeader::size, size, PendingMessage::timeout, GNUNET_MessageHeader::type, and PendingMessage::vl.

Referenced by transmit_on_queue().

Here is the call graph for this function:

Here is the caller graph for this function:

reorder_root_pm()

static void reorder_root_pm ( struct PendingMessage *  pm, struct GNUNET_TIME_Absolute  next_attempt  )

static

Definition at line 10942 of file gnunet-service-transport.c.

{
  struct VirtualLink *vl = pm->vl;
  struct PendingMessage *pos;
 
  /* re-insert sort in neighbour list */
  GNUNET_CONTAINER_MDLL_remove (vl,
                                vl->pending_msg_head,
                                vl->pending_msg_tail,
                                pm);
  pos = vl->pending_msg_tail;
  while ((NULL != pos) &&
         (next_attempt.abs_value_us > pos->next_attempt.abs_value_us))
    pos = pos->prev_vl;
  GNUNET_CONTAINER_MDLL_insert_after (vl,
                                      vl->pending_msg_head,
                                      vl->pending_msg_tail,
                                      pos,
                                      pm);
}

References GNUNET_TIME_Absolute::abs_value_us, GNUNET_CONTAINER_MDLL_insert_after, GNUNET_CONTAINER_MDLL_remove, PendingMessage::next_attempt, VirtualLink::pending_msg_head, VirtualLink::pending_msg_tail, pm, PendingMessage::prev_vl, and PendingMessage::vl.

Referenced by update_pm_next_attempt().

Here is the caller graph for this function:

check_next_attempt_tree()

static unsigned int check_next_attempt_tree ( struct PendingMessage *  pm, struct PendingMessage *  root  )

static

Definition at line 10966 of file gnunet-service-transport.c.

{
  struct PendingMessage *pos;
  enum GNUNET_GenericReturnValue frags_in_flight;
 
  pos = pm->head_frag;
  while (NULL != pos)
  {
    if (pos->frags_in_flight_round == pm->frags_in_flight_round ||
        GNUNET_NO == check_next_attempt_tree (pos, root))
      frags_in_flight = GNUNET_NO;
    else
    {
      frags_in_flight = GNUNET_YES;
      break;
    }
    pos = pos->next_frag;
  }
 
  return frags_in_flight;
}

References check_next_attempt_tree(), PendingMessage::frags_in_flight, PendingMessage::frags_in_flight_round, GNUNET_NO, GNUNET_YES, PendingMessage::head_frag, PendingMessage::next_frag, and pm.

Referenced by check_next_attempt_tree(), and update_pm_next_attempt().

Here is the call graph for this function:

Here is the caller graph for this function:

harmonize_flight_round()

static void harmonize_flight_round ( struct PendingMessage *  pm )

static

Definition at line 10990 of file gnunet-service-transport.c.

{
  struct PendingMessage *pos;
 
  pos = pm->head_frag;
  while (NULL != pos)
  {
    pos->frags_in_flight_round = pm->frags_in_flight_round;
    harmonize_flight_round (pos);
    pos = pos->next_frag;
  }
}

References PendingMessage::frags_in_flight_round, harmonize_flight_round(), PendingMessage::head_frag, PendingMessage::next_frag, and pm.

Referenced by harmonize_flight_round(), and update_pm_next_attempt().

Here is the call graph for this function:

Here is the caller graph for this function:

update_pm_next_attempt()

static void update_pm_next_attempt ( struct PendingMessage *  pm, struct GNUNET_TIME_Absolute  next_attempt  )

static

Change the value of the next_attempt field of pm to next_attempt and re-order pm in the transmission list as required by the new timestamp.

Parameters

pm	a pending message to update
next_attempt	timestamp to use

Definition at line 11013 of file gnunet-service-transport.c.

{
  if (NULL == pm->frag_parent)
  {
    pm->next_attempt = next_attempt;
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Next attempt for message <%" PRIu64 "> set to %" PRIu64 "\n",
                pm->logging_uuid,
                next_attempt.abs_value_us);
    reorder_root_pm (pm, next_attempt);
  }
  else if ((PMT_RELIABILITY_BOX == pm->pmt) || (PMT_DV_BOX == pm->pmt))// || (PMT_FRAGMENT_BOX == pm->pmt))
  {
    struct PendingMessage *root = pm->frag_parent;
 
    while (NULL != root->frag_parent)
      root = root->frag_parent;
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Next attempt for root message <%" PRIu64 "> set to %s\n",
                root->logging_uuid,
                GNUNET_STRINGS_absolute_time_to_string (next_attempt));
    root->next_attempt = next_attempt;
    reorder_root_pm (root, next_attempt);
  }
  else
  {
    struct PendingMessage *root = pm->frag_parent;
 
    while (NULL != root->frag_parent && PMT_DV_BOX != root->pmt)
      root = root->frag_parent;
 
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "frag_count next attempt %u\n",
                root->frag_count);
 
    if (GNUNET_NO == root->frags_in_flight)
    {
      root->next_attempt = next_attempt;
      harmonize_flight_round (root);
      root->frags_in_flight_round++;
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "Next attempt for fragmented message <%" PRIu64 "> (<%" PRIu64
                  ">)set to %" PRIu64 "\n",
                  pm->logging_uuid,
                  root->logging_uuid,
                  next_attempt.abs_value_us);
    }
 
    pm->next_attempt = root->next_attempt;
    pm->frags_in_flight_round = root->frags_in_flight_round;
    harmonize_flight_round (pm);
 
    if (root->bytes_msg == root->frag_off)
      root->frags_in_flight = check_next_attempt_tree (root, root);
    else
      root->frags_in_flight = GNUNET_YES;
 
    if (GNUNET_NO == root->frags_in_flight)
    {
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "We have no fragments in flight for message %" PRIu64
                  ", reorder root! Next attempt is %" PRIu64 "\n",
                  root->logging_uuid,
                  root->next_attempt.abs_value_us);
      if (PMT_DV_BOX == root->pmt)
        root = root->frag_parent;
      reorder_root_pm (root, root->next_attempt);
      // root->next_attempt = GNUNET_TIME_UNIT_ZERO_ABS;
    }
    else
    {
      double factor = ((double) root->frag_count - 1)
                      / (double) root->frag_count;
      struct GNUNET_TIME_Relative s1;
      struct GNUNET_TIME_Relative s2;
      struct GNUNET_TIME_Relative plus_mean =
        GNUNET_TIME_absolute_get_remaining (root->next_attempt);
      struct GNUNET_TIME_Relative plus = GNUNET_TIME_absolute_get_remaining (
        next_attempt);
 
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "frag_count %u after factor\n",
                  root->frag_count);
      s1 = GNUNET_TIME_relative_multiply_double (plus_mean,
                                                 factor);
      s2 = GNUNET_TIME_relative_divide (plus,
                                        root->frag_count);
      plus_mean = GNUNET_TIME_relative_add (s1, s2);
      root->next_attempt = GNUNET_TIME_relative_to_absolute (plus_mean);
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "We have fragments in flight for message %" PRIu64
                  ", do not reorder root! Actual next attempt %" PRIu64 "\n",
                  root->logging_uuid,
                  root->next_attempt.abs_value_us);
    }
  }
}

References GNUNET_TIME_Absolute::abs_value_us, PendingMessage::bytes_msg, check_next_attempt_tree(), PendingMessage::frag_count, PendingMessage::frag_off, PendingMessage::frag_parent, PendingMessage::frags_in_flight, PendingMessage::frags_in_flight_round, GNUNET_ERROR_TYPE_DEBUG, GNUNET_log, GNUNET_NO, GNUNET_STRINGS_absolute_time_to_string(), GNUNET_TIME_absolute_get_remaining(), GNUNET_TIME_relative_add(), GNUNET_TIME_relative_divide(), GNUNET_TIME_relative_multiply_double(), GNUNET_TIME_relative_to_absolute(), GNUNET_YES, harmonize_flight_round(), PendingMessage::logging_uuid, PendingMessage::next_attempt, pm, PendingMessage::pmt, PMT_DV_BOX, PMT_RELIABILITY_BOX, and reorder_root_pm().

Referenced by transmit_on_queue().

Here is the call graph for this function:

Here is the caller graph for this function:

select_best_pending_from_link()

static void select_best_pending_from_link ( struct PendingMessageScoreContext *  sc, struct Queue *  queue, struct VirtualLink *  vl, struct DistanceVectorHop *  dvh, size_t  overhead  )

static

Select the best pending message from vl for transmission via queue.

Parameters

[in,out]	sc	best message so far (NULL for none), plus scoring data
	queue	the queue that will be used for transmission
	vl	the virtual link providing the messages
	dvh	path we are currently considering, or NULL for none
	overhead	number of bytes of overhead to be expected from DV encapsulation (0 for without DV)

Definition at line 11177 of file gnunet-service-transport.c.

{
  struct GNUNET_TIME_Absolute now;
 
  now = GNUNET_TIME_absolute_get ();
  sc->to_early = GNUNET_NO;
  sc->frags_in_flight = GNUNET_NO;
  for (struct PendingMessage *pos = vl->pending_msg_head; NULL != pos;
       pos = pos->next_vl)
  {
    size_t real_overhead = overhead;
    int frag;
    int relb;
 
    if ((NULL != dvh) && (PMT_DV_BOX == pos->pmt))
    {
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "DV messages must not be DV-routed to next hop!\n");
      continue;   /* DV messages must not be DV-routed to next hop! */
    }
    if (pos->next_attempt.abs_value_us > now.abs_value_us)
    {
      if (GNUNET_YES == pos->frags_in_flight)
      {
        sc->frags_in_flight = GNUNET_YES;
        GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                    "Fragments in flight for message %" PRIu64 "\n",
                    pos->logging_uuid);
      }
      else
      {
        GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                    "Maybe too early, because message are sorted by next_attempt, if there are no fragments in flight.Checked message %"
                    PRIu64 "\n",
                    pos->logging_uuid);
        sc->to_early = GNUNET_YES;
        sc->to_early_retry_delay = GNUNET_TIME_absolute_get_remaining (
          pos->next_attempt);
        continue;
      }
      // break;   /* too early for all messages, they are sorted by next_attempt */
    }
    if (NULL != pos->qe)
    {
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "not eligible\n");
      continue;   /* not eligible */
    }
    sc->consideration_counter++;
    /* determine if we have to fragment, if so add fragmentation
       overhead! */
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "check %" PRIu64 " for sc->best\n",
                pos->logging_uuid);
    frag = GNUNET_NO;
    if (((0 != queue->mtu) &&
         (pos->bytes_msg + real_overhead > queue->mtu)) ||
        (pos->bytes_msg > UINT16_MAX - sizeof(struct
                                              GNUNET_TRANSPORT_SendMessageTo))
        ||
        (NULL != pos->head_frag /* fragments already exist, should
                                     respect that even if MTU is UINT16_MAX for
                                     this queue */))
    {
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "fragment msg with size %u, realoverhead is %lu\n",
                  pos->bytes_msg,
                  real_overhead);
      frag = GNUNET_YES;
      if (GNUNET_TRANSPORT_CC_RELIABLE == queue->tc->details.communicator.cc)
      {
        /* FIXME-FRAG-REL-UUID: we could use an optimized, shorter fragmentation
           header without the ACK UUID when using a *reliable* channel! */
      }
      real_overhead = overhead + sizeof(struct TransportFragmentBoxMessage);
    }
    /* determine if we have to reliability-box, if so add reliability box
       overhead */
    relb = GNUNET_NO;
    if ((GNUNET_NO == frag) &&
        (0 == (pos->prefs & GNUNET_MQ_PREF_UNRELIABLE)) &&
        (GNUNET_TRANSPORT_CC_RELIABLE != queue->tc->details.communicator.cc))
    {
      real_overhead += sizeof(struct TransportReliabilityBoxMessage);
 
      if ((0 != queue->mtu) && (pos->bytes_msg + real_overhead > queue->mtu))
      {
        frag = GNUNET_YES;
        real_overhead = overhead + sizeof(struct TransportFragmentBoxMessage);
      }
      else
      {
        relb = GNUNET_YES;
      }
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "Create reliability box of msg with size %u, realoverhead is %lu %u %u %u\n",
                  pos->bytes_msg,
                  real_overhead,
                  queue->mtu,
                  frag,
                  relb);
    }
 
    /* Finally, compare to existing 'best' in sc to see if this 'pos' pending
       message would beat it! */
    if (GNUNET_NO == sc->frags_in_flight && NULL != sc->best)
    {
      /* CHECK if pos fits queue BETTER (=smaller) than pm, if not: continue;
         OPTIMIZE-ME: This is a heuristic, which so far has NOT been
         experimentally validated. There may be some huge potential for
         improvement here. Also, we right now only compare how well the
         given message fits _this_ queue, and do not consider how well other
         queues might suit the message. Taking other queues into consideration
         may further improve the result, but could also be expensive
         in terms of CPU time.  */
      long long sc_score = sc->frag * 40 + sc->relb * 20 + sc->real_overhead;
      long long pm_score = frag * 40 + relb * 20 + real_overhead;
      long long time_delta =
        (sc->best->next_attempt.abs_value_us - pos->next_attempt.abs_value_us)
        / 1000LL;
 
      /* "time_delta" considers which message has been 'ready' for transmission
         for longer, if a message has a preference for low latency, increase
         the weight of the time_delta by 10x if it is favorable for that message */
      if ((0 != (pos->prefs & GNUNET_MQ_PREF_LOW_LATENCY)) &&
          (0 != (sc->best->prefs & GNUNET_MQ_PREF_LOW_LATENCY)))
        time_delta *= 10;     /* increase weight (always, both are low latency) */
      else if ((0 != (pos->prefs & GNUNET_MQ_PREF_LOW_LATENCY)) &&
               (time_delta > 0))
        time_delta *= 10;     /* increase weight, favors 'pos', which is low latency */
      else if ((0 != (sc->best->prefs & GNUNET_MQ_PREF_LOW_LATENCY)) &&
               (time_delta < 0))
        time_delta *= 10;     /* increase weight, favors 'sc->best', which is low latency */
      if (0 != queue->mtu)
      {
        /* Grant bonus if we are below MTU, larger bonus the closer we will
           be to the MTU */
        if (queue->mtu > sc->real_overhead + sc->best->bytes_msg)
          sc_score -= queue->mtu - (sc->real_overhead + sc->best->bytes_msg);
        if (queue->mtu > real_overhead + pos->bytes_msg)
          pm_score -= queue->mtu - (real_overhead + pos->bytes_msg);
      }
      if (sc_score + time_delta > pm_score)
      {
        GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                    "sc_score of %" PRIu64 " larger, keep sc->best %" PRIu64
                    "\n",
                    pos->logging_uuid,
                    sc->best->logging_uuid);
        continue;     /* sc_score larger, keep sc->best */
      }
    }
    sc->best = pos;
    sc->dvh = dvh;
    sc->frag = frag;
    sc->relb = relb;
    sc->real_overhead = real_overhead;
  }
}

References GNUNET_TIME_Absolute::abs_value_us, GNUNET_ERROR_TYPE_DEBUG, GNUNET_log, GNUNET_MQ_PREF_LOW_LATENCY, GNUNET_MQ_PREF_UNRELIABLE, GNUNET_NO, GNUNET_TIME_absolute_get(), GNUNET_TIME_absolute_get_remaining(), GNUNET_TRANSPORT_CC_RELIABLE, GNUNET_YES, VirtualLink::pending_msg_head, PMT_DV_BOX, queue(), and sc.

Referenced by transmit_on_queue().

Here is the call graph for this function:

Here is the caller graph for this function:

extract_box_cb()

static void extract_box_cb ( void *  cls, struct Neighbour *  next_hop, const struct GNUNET_MessageHeader *  hdr, enum RouteMessageOptions  options  )

static

Function to call to further operate on the now DV encapsulated message hdr, forwarding it via next_hop under respect of options.

Parameters

cls	a struct PendingMessageScoreContext
next_hop	next hop of the DV path
hdr	encapsulated message, technically a struct TransportDVBoxMessage
options	options of the original message

Definition at line 11353 of file gnunet-service-transport.c.

{
  struct PendingMessageScoreContext *sc = cls;
  struct PendingMessage *pm = sc->best;
  struct PendingMessage *bpm;
  uint16_t bsize = ntohs (hdr->size);
 
  GNUNET_assert (NULL == pm->bpm);
  bpm = GNUNET_malloc (sizeof(struct PendingMessage) + bsize);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "5 created pm %p storing vl %p from pm %p\n",
              bpm,
              pm->vl,
              pm);
  bpm->logging_uuid = logging_uuid_gen++;
  bpm->pmt = PMT_DV_BOX;
  bpm->vl = pm->vl;
  bpm->timeout = pm->timeout;
  bpm->bytes_msg = bsize;
  bpm->frag_parent = pm;
  set_pending_message_uuid (bpm);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Creating DV Box %" PRIu64 " for original message %" PRIu64
              " (next hop is %s)\n",
              bpm->logging_uuid,
              pm->logging_uuid,
              GNUNET_i2s (&next_hop->pid));
  memcpy (&bpm[1], hdr, bsize);
  pm->bpm = bpm;
}

References PendingMessage::bpm, bsize, PendingMessage::bytes_msg, PendingMessage::frag_parent, GNUNET_assert, GNUNET_ERROR_TYPE_DEBUG, GNUNET_i2s(), GNUNET_log, GNUNET_malloc, PendingMessage::logging_uuid, logging_uuid_gen, Neighbour::pid, pm, PendingMessage::pmt, PMT_DV_BOX, sc, set_pending_message_uuid(), GNUNET_MessageHeader::size, PendingMessage::timeout, and PendingMessage::vl.

Referenced by transmit_on_queue().

Here is the call graph for this function:

Here is the caller graph for this function:

handle_del_queue_message()

static void handle_del_queue_message ( void *  cls, const struct GNUNET_TRANSPORT_DelQueueMessage *  dqm  )

static

Queue to a peer went down.

Process the request.

Parameters

cls	the client
dqm	the send message that was sent

Definition at line 11660 of file gnunet-service-transport.c.

{
  struct TransportClient *tc = cls;
 
  if (CT_COMMUNICATOR != tc->type)
  {
    GNUNET_break (0);
    GNUNET_SERVICE_client_drop (tc->client);
    return;
  }
  for (struct Queue *queue = tc->details.communicator.queue_head; NULL != queue;
       queue = queue->next_client)
  {
    struct Neighbour *neighbour = queue->neighbour;
 
    if ((ntohl (dqm->qid) != queue->qid) ||
        (0 != GNUNET_memcmp (&dqm->receiver, &neighbour->pid)))
      continue;
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Dropped queue %s to peer %s\n",
                queue->address,
                GNUNET_i2s (&neighbour->pid));
    free_queue (queue);
    GNUNET_SERVICE_client_continue (tc->client);
    return;
  }
  GNUNET_break (0);
  GNUNET_SERVICE_client_drop (tc->client);
}

References CT_COMMUNICATOR, free_queue(), GNUNET_break, GNUNET_ERROR_TYPE_DEBUG, GNUNET_i2s(), GNUNET_log, GNUNET_memcmp, GNUNET_SERVICE_client_continue(), GNUNET_SERVICE_client_drop(), Neighbour::pid, GNUNET_TRANSPORT_DelQueueMessage::qid, queue(), GNUNET_TRANSPORT_DelQueueMessage::receiver, and tc.

Here is the call graph for this function:

handle_send_message_ack()

static void handle_send_message_ack ( void *  cls, const struct GNUNET_TRANSPORT_SendMessageToAck *  sma  )

static

Message was transmitted.

Process the request.

Parameters

cls	the client
sma	the send message that was sent

Definition at line 11790 of file gnunet-service-transport.c.

{
  struct TransportClient *tc = cls;
  struct QueueEntry *qe;
 
  if (CT_COMMUNICATOR != tc->type)
  {
    GNUNET_break (0);
    GNUNET_SERVICE_client_drop (tc->client);
    return;
  }
 
  /* find our queue entry matching the ACK */
  qe = NULL;
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Looking for queue for PID %s\n",
              GNUNET_i2s (&sma->receiver));
  for (struct Queue *queue = tc->details.communicator.queue_head; NULL != queue;
       queue = queue->next_client)
  {
    if (0 != GNUNET_memcmp (&queue->neighbour->pid, &sma->receiver))
      continue;
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Found PID %s\n",
                GNUNET_i2s (&queue->neighbour->pid));
 
 
    for (struct QueueEntry *qep = queue->queue_head; NULL != qep;
         qep = qep->next)
    {
      if (qep->mid != GNUNET_ntohll (sma->mid) || queue->qid != ntohl (
            sma->qid))
        continue;
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "QueueEntry MID: %" PRIu64 " on queue QID: %u, Ack MID: %"
                  PRIu64 " Ack QID %u\n",
                  qep->mid,
                  queue->qid,
                  GNUNET_ntohll (sma->mid),
                  ntohl (sma->qid));
      qe = qep;
      if ((NULL != qe->pm) && (qe->pm->qe != qe))
        GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                    "For pending message %" PRIu64 " we had retransmissions.\n",
                    qe->pm->logging_uuid);
      break;
    }
  }
  if (NULL == qe)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "No QueueEntry found for Ack MID %" PRIu64 " QID: %u\n",
                GNUNET_ntohll (sma->mid),
                ntohl (sma->qid));
    // TODO I guess this can happen, if the Ack from the peer comes before the Ack from the queue.
    // Update: Maybe QueueEntry was accidentally freed during freeing PendingMessage.
    /* this should never happen */
    // GNUNET_break (0);
    // GNUNET_SERVICE_client_drop (tc->client);
    GNUNET_SERVICE_client_continue (tc->client);
    return;
  }
  free_queue_entry (qe, tc);
  GNUNET_SERVICE_client_continue (tc->client);
}

References CT_COMMUNICATOR, free_queue_entry(), GNUNET_break, GNUNET_ERROR_TYPE_DEBUG, GNUNET_i2s(), GNUNET_log, GNUNET_memcmp, GNUNET_ntohll(), GNUNET_SERVICE_client_continue(), GNUNET_SERVICE_client_drop(), GNUNET_TRANSPORT_SendMessageToAck::mid, qe, GNUNET_TRANSPORT_SendMessageToAck::qid, queue(), GNUNET_TRANSPORT_SendMessageToAck::receiver, and tc.

Here is the call graph for this function:

handle_burst_finished()

static void handle_burst_finished ( void *  cls, const struct GNUNET_TRANSPORT_BurstFinished *  bf  )

static

The burst finished.

Parameters

cls	the client

Definition at line 11864 of file gnunet-service-transport.c.

{
  burst_running = GNUNET_NO;
}

References burst_running, and GNUNET_NO.

notify_client_queues()

static int notify_client_queues ( void *  cls, const struct GNUNET_PeerIdentity *  pid, void *  value  )

static

Iterator telling new MONITOR client about all existing queues to peers.

Parameters

cls	the new struct TransportClient
pid	a connected peer
value	the struct Neighbour with more information

Returns

GNUNET_OK (continue to iterate)

Definition at line 11881 of file gnunet-service-transport.c.

{
  struct TransportClient *tc = cls;
  struct Neighbour *neighbour = value;
 
  GNUNET_assert (CT_MONITOR == tc->type);
  for (struct Queue *q = neighbour->queue_head; NULL != q;
       q = q->next_neighbour)
  {
    struct MonitorEvent me = { .rtt = q->pd.aged_rtt,
                               .cs = q->cs,
                               .num_msg_pending = q->num_msg_pending,
                               .num_bytes_pending = q->num_bytes_pending };
 
    notify_monitor (tc, pid, q->address, q->nt, &me);
  }
  return GNUNET_OK;
}

References CT_MONITOR, GNUNET_assert, GNUNET_OK, me, notify_monitor(), pid, q, Neighbour::queue_head, tc, and value.

Referenced by handle_monitor_start().

Here is the call graph for this function:

Here is the caller graph for this function:

handle_monitor_start()

static void handle_monitor_start ( void *  cls, const struct GNUNET_TRANSPORT_MonitorStart *  start  )

static

Initialize a monitor client.

Parameters

cls	the client
start	the start message that was sent

Definition at line 11910 of file gnunet-service-transport.c.

{
  struct TransportClient *tc = cls;
 
  if (CT_NONE != tc->type)
  {
    GNUNET_break (0);
    GNUNET_SERVICE_client_drop (tc->client);
    return;
  }
  tc->type = CT_MONITOR;
  tc->details.monitor.peer = start->peer;
  tc->details.monitor.one_shot = ntohl (start->one_shot);
  GNUNET_CONTAINER_multipeermap_iterate (neighbours, &notify_client_queues, tc);
  GNUNET_SERVICE_client_mark_monitor (tc->client);
  GNUNET_SERVICE_client_continue (tc->client);
}

References CT_MONITOR, CT_NONE, GNUNET_break, GNUNET_CONTAINER_multipeermap_iterate(), GNUNET_SERVICE_client_continue(), GNUNET_SERVICE_client_drop(), GNUNET_SERVICE_client_mark_monitor(), neighbours, notify_client_queues(), start, and tc.

Here is the call graph for this function:

lookup_communicator()

static struct TransportClient * lookup_communicator ( const char *  prefix )

static

Find transport client providing communication service for the protocol prefix.

Parameters

prefix

communicator name

Returns

NULL if no such transport client is available

Definition at line 11938 of file gnunet-service-transport.c.

{
  for (struct TransportClient *tc = clients_head; NULL != tc; tc = tc->next)
  {
    if (CT_COMMUNICATOR != tc->type)
      continue;
    if (0 == strcmp (prefix, tc->details.communicator.address_prefix))
      return tc;
  }
  GNUNET_log (
    GNUNET_ERROR_TYPE_WARNING,
    "Someone suggested use of communicator for `%s', but we do not have such a communicator!\n",
    prefix);
  return NULL;
}

References clients_head, CT_COMMUNICATOR, GNUNET_ERROR_TYPE_WARNING, GNUNET_log, prefix, and tc.

Referenced by suggest_to_connect().

Here is the caller graph for this function:

check_connection_quality()

static int check_connection_quality ( void *  cls, const struct GNUNET_PeerIdentity *  pid, void *  value  )

static

Check whether any queue to the given neighbour is of a good "quality" and if so, increment the counter.

Also counts the total number of queues, and returns the k-th queue found.

Parameters

cls	a struct QueueQualityContext * with counters
pid	peer this is about
value	a struct Neighbour

Returns

GNUNET_OK (continue to iterate)

Definition at line 12164 of file gnunet-service-transport.c.

{
  struct QueueQualityContext *ctx = cls;
  struct Neighbour *n = value;
  int do_inc;
 
  (void) pid;
  do_inc = GNUNET_NO;
  for (struct Queue *q = n->queue_head; NULL != q; q = q->next_neighbour)
  {
    ctx->num_queues++;
    if (0 == ctx->k--)
      ctx->q = q;
    /* FIXME-CONQ-STATISTICS: in the future, add reliability / goodput
       statistics and consider those as well here? */
    if (q->pd.aged_rtt.rel_value_us < DV_QUALITY_RTT_THRESHOLD.rel_value_us)
      do_inc = GNUNET_YES;
  }
  if (GNUNET_YES == do_inc)
    ctx->quality_count++;
  return GNUNET_OK;
}

References ctx, DV_QUALITY_RTT_THRESHOLD, GNUNET_NO, GNUNET_OK, GNUNET_YES, pid, q, Neighbour::queue_head, and value.

Referenced by sign_dv_init_cb(), and start_dv_learn().

Here is the caller graph for this function:

start_dv_learn()

static void start_dv_learn ( void *  cls )

static

Task run when we CONSIDER initiating a DV learn process.

We first check that sending out a message is even possible (queues exist), then that it is desirable (if not, reschedule the task for later), and finally we may then begin the job. If there are too many entries in the dvlearn_map, we purge the oldest entry using lle_tail.

Parameters

cls

NULL

Definition at line 12268 of file gnunet-service-transport.c.

{
  struct LearnLaunchEntry *lle;
  struct QueueQualityContext qqc;
  struct TransportDVLearnMessage dvl;
 
  (void) cls;
  dvlearn_task = NULL;
  if (0 == GNUNET_CONTAINER_multipeermap_size (neighbours))
    return; /* lost all connectivity, cannot do learning */
  qqc.quality_count = 0;
  qqc.num_queues = 0;
  qqc.k = GNUNET_CONTAINER_multipeermap_size (neighbours);
  GNUNET_CONTAINER_multipeermap_iterate (neighbours,
                                         &check_connection_quality,
                                         &qqc);
  if (qqc.quality_count > DV_LEARN_QUALITY_THRESHOLD)
  {
    struct GNUNET_TIME_Relative delay;
    unsigned int factor;
 
    /* scale our retries by how far we are above the threshold */
    factor = qqc.quality_count / DV_LEARN_QUALITY_THRESHOLD;
    delay = GNUNET_TIME_relative_multiply (DV_LEARN_BASE_FREQUENCY, factor);
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "At connection quality %u, will launch DV learn in %s\n",
                qqc.quality_count,
                GNUNET_STRINGS_relative_time_to_string (delay, GNUNET_YES));
    dvlearn_task = GNUNET_SCHEDULER_add_delayed (delay, &start_dv_learn, NULL);
    return;
  }
  /* remove old entries in #dvlearn_map if it has grown too big */
  while (MAX_DV_LEARN_PENDING <=
         GNUNET_CONTAINER_multishortmap_size (dvlearn_map))
  {
    lle = lle_tail;
    GNUNET_assert (GNUNET_YES ==
                   GNUNET_CONTAINER_multishortmap_remove (dvlearn_map,
                                                          &lle->challenge.value,
                                                          lle));
    GNUNET_CONTAINER_DLL_remove (lle_head, lle_tail, lle);
    GNUNET_free (lle);
  }
  /* setup data structure for learning */
  lle = GNUNET_new (struct LearnLaunchEntry);
  GNUNET_CRYPTO_random_block (GNUNET_CRYPTO_QUALITY_NONCE,
                              &lle->challenge,
                              sizeof(lle->challenge));
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Starting launch DV learn with challenge %s\n",
              GNUNET_sh2s (&lle->challenge.value));
  GNUNET_CONTAINER_DLL_insert (lle_head, lle_tail, lle);
  GNUNET_break (GNUNET_YES ==
                GNUNET_CONTAINER_multishortmap_put (
                  dvlearn_map,
                  &lle->challenge.value,
                  lle,
                  GNUNET_CONTAINER_MULTIHASHMAPOPTION_UNIQUE_ONLY));
  dvl.header.type = htons (GNUNET_MESSAGE_TYPE_TRANSPORT_DV_LEARN);
  dvl.header.size = htons (sizeof(dvl));
  dvl.num_hops = htons (0);
  dvl.bidirectional = htons (0);
  dvl.non_network_delay = GNUNET_TIME_relative_hton (GNUNET_TIME_UNIT_ZERO);
  dvl.monotonic_time =
    GNUNET_TIME_absolute_hton (GNUNET_TIME_absolute_get_monotonic (GST_cfg));
  {
    struct DvInitPS dvip = {
      .purpose.purpose = htonl (
        GNUNET_SIGNATURE_PURPOSE_TRANSPORT_DV_INITIATOR),
      .purpose.size = htonl (sizeof(dvip)),
      .monotonic_time = dvl.monotonic_time,
      .challenge = lle->challenge
    };
    struct SignDvInitCls *sign_dv_init_cls;
 
    sign_dv_init_cls = GNUNET_new (struct SignDvInitCls);
    sign_dv_init_cls->dvl = dvl;
    sign_dv_init_cls->lle = lle;
    sign_dv_init_cls->qqc = qqc;
    sign_dv_init_cls->pr = GNUNET_new (struct PilsRequest);
    GNUNET_CONTAINER_DLL_insert (pils_requests_head,
                                 pils_requests_tail,
                                 sign_dv_init_cls->pr);
    sign_dv_init_cls->pr->op =
      GNUNET_PILS_sign_by_peer_identity (pils,
                                         &dvip.purpose,
                                         sign_dv_init_cb,
                                         sign_dv_init_cls);
  }
}

References TransportDVLearnMessage::bidirectional, LearnLaunchEntry::challenge, check_connection_quality(), DV_LEARN_BASE_FREQUENCY, DV_LEARN_QUALITY_THRESHOLD, SignDvInitCls::dvl, dvlearn_map, dvlearn_task, GNUNET_assert, GNUNET_break, GNUNET_CONTAINER_DLL_insert, GNUNET_CONTAINER_DLL_remove, GNUNET_CONTAINER_MULTIHASHMAPOPTION_UNIQUE_ONLY, GNUNET_CONTAINER_multipeermap_iterate(), GNUNET_CONTAINER_multipeermap_size(), GNUNET_CONTAINER_multishortmap_put(), GNUNET_CONTAINER_multishortmap_remove(), GNUNET_CONTAINER_multishortmap_size(), GNUNET_CRYPTO_QUALITY_NONCE, GNUNET_CRYPTO_random_block(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_free, GNUNET_log, GNUNET_MESSAGE_TYPE_TRANSPORT_DV_LEARN, GNUNET_new, GNUNET_PILS_sign_by_peer_identity(), GNUNET_SCHEDULER_add_delayed(), GNUNET_sh2s(), GNUNET_SIGNATURE_PURPOSE_TRANSPORT_DV_INITIATOR, GNUNET_STRINGS_relative_time_to_string(), GNUNET_TIME_absolute_get_monotonic(), GNUNET_TIME_absolute_hton(), GNUNET_TIME_relative_hton(), GNUNET_TIME_relative_multiply(), GNUNET_TIME_UNIT_ZERO, GNUNET_YES, GST_cfg, TransportDVLearnMessage::header, QueueQualityContext::k, SignDvInitCls::lle, lle_head, lle_tail, MAX_DV_LEARN_PENDING, TransportDVLearnMessage::monotonic_time, neighbours, TransportDVLearnMessage::non_network_delay, TransportDVLearnMessage::num_hops, QueueQualityContext::num_queues, PilsRequest::op, pils, pils_requests_head, pils_requests_tail, SignDvInitCls::pr, GNUNET_CRYPTO_EccSignaturePurpose::purpose, DvInitPS::purpose, SignDvInitCls::qqc, QueueQualityContext::quality_count, sign_dv_init_cb(), GNUNET_MessageHeader::size, start_dv_learn(), GNUNET_MessageHeader::type, and GNUNET_CRYPTO_ChallengeNonceP::value.

Referenced by handle_add_queue_message(), sign_dv_init_cb(), and start_dv_learn().

Here is the call graph for this function:

Here is the caller graph for this function:

sign_dv_init_cb()

static void sign_dv_init_cb ( void *  cls, const struct GNUNET_PeerIdentity *  pid, const struct GNUNET_CRYPTO_EddsaSignature *  sig  )

static

Definition at line 12215 of file gnunet-service-transport.c.

{
  struct SignDvInitCls *sign_dv_init_cls = cls;
  struct TransportDVLearnMessage dvl = sign_dv_init_cls->dvl;
  struct LearnLaunchEntry *lle = sign_dv_init_cls->lle;
  struct QueueQualityContext qqc = sign_dv_init_cls->qqc;
 
  sign_dv_init_cls->pr->op = NULL;
  GNUNET_CONTAINER_DLL_remove (pils_requests_head,
                               pils_requests_tail,
                               sign_dv_init_cls->pr);
  GNUNET_free (sign_dv_init_cls->pr);
 
  dvl.init_sig = *sig;
  dvl.initiator = *GST_my_identity;
  dvl.challenge = lle->challenge;
 
  qqc.quality_count = 0;
  qqc.k = GNUNET_CRYPTO_random_u32 (GNUNET_CRYPTO_QUALITY_WEAK, qqc.num_queues);
  qqc.num_queues = 0;
  qqc.q = NULL;
  GNUNET_CONTAINER_multipeermap_iterate (neighbours,
                                         &check_connection_quality,
                                         &qqc);
  GNUNET_assert (NULL != qqc.q);
 
  /* Do this as close to transmission time as possible! */
  lle->launch_time = GNUNET_TIME_absolute_get ();
 
  queue_send_msg (qqc.q, NULL, &dvl, sizeof(dvl));
  /* reschedule this job, randomizing the time it runs (but no
     actual backoff!) */
  dvlearn_task = GNUNET_SCHEDULER_add_delayed (GNUNET_TIME_randomize (
                                                 DV_LEARN_BASE_FREQUENCY),
                                               &start_dv_learn,
                                               NULL);
}

References TransportDVLearnMessage::challenge, LearnLaunchEntry::challenge, check_connection_quality(), DV_LEARN_BASE_FREQUENCY, SignDvInitCls::dvl, dvlearn_task, GNUNET_assert, GNUNET_CONTAINER_DLL_remove, GNUNET_CONTAINER_multipeermap_iterate(), GNUNET_CRYPTO_QUALITY_WEAK, GNUNET_CRYPTO_random_u32(), GNUNET_free, GNUNET_SCHEDULER_add_delayed(), GNUNET_TIME_absolute_get(), GNUNET_TIME_randomize(), GST_my_identity, TransportDVLearnMessage::init_sig, TransportDVLearnMessage::initiator, QueueQualityContext::k, LearnLaunchEntry::launch_time, SignDvInitCls::lle, neighbours, QueueQualityContext::num_queues, PilsRequest::op, pils_requests_head, pils_requests_tail, SignDvInitCls::pr, QueueQualityContext::q, SignDvInitCls::qqc, QueueQualityContext::quality_count, queue_send_msg(), and start_dv_learn().

Referenced by start_dv_learn().

Here is the call graph for this function:

Here is the caller graph for this function:

check_validation_request_pending()

static int check_validation_request_pending ( void *  cls, const struct GNUNET_PeerIdentity *  pid, void *  value  )

static

A new queue has been created, check if any address validation requests have been waiting for it.

Parameters

cls	a struct Queue
pid	peer concerned (unused)
value	a struct ValidationState

Returns

GNUNET_NO if a match was found and we can stop looking

Definition at line 12395 of file gnunet-service-transport.c.

{
  struct Queue *q = cls;
  struct ValidationState *vs = value;
  char *address_without_port_vs;
  char *address_without_port_q;
  int success = GNUNET_YES;
 
  // TODO Check if this is really necessary.
  address_without_port_vs = get_address_without_port (vs->address);
  address_without_port_q = get_address_without_port (q->address);
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Check validation request pending for `%s' at `%s'/`%s' (vs)/(q)\n",
              GNUNET_i2s (pid),
              address_without_port_vs,
              address_without_port_q);
  (void) pid;
  if ((GNUNET_YES == vs->awaiting_queue) &&
      (0 == strcmp (address_without_port_vs, address_without_port_q)))
  {
 
    vs->awaiting_queue = GNUNET_NO;
    validation_transmit_on_queue (q, vs);
    success = GNUNET_NO;
  }
 
  GNUNET_free (address_without_port_vs);
  GNUNET_free (address_without_port_q);
  return success;
}

References get_address_without_port(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_free, GNUNET_i2s(), GNUNET_log, GNUNET_NO, GNUNET_YES, pid, q, validation_transmit_on_queue(), and value.

Referenced by handle_add_queue_message().

Here is the call graph for this function:

Here is the caller graph for this function:

neighbour_dv_monotime_cb()

static void neighbour_dv_monotime_cb ( void *  cls, const struct GNUNET_PEERSTORE_Record *  record, const char *  emsg  )

static

Function called with the monotonic time of a DV initiator by PEERSTORE.

Updates the time.

Parameters

cls	a struct Neighbour
record	the information found, NULL for the last call
emsg	error message

Definition at line 12439 of file gnunet-service-transport.c.

{
  struct Neighbour *n = cls;
  struct GNUNET_TIME_AbsoluteNBO *mtbe;
 
  (void) emsg;
  if (NULL == record)
  {
    /* we're done with #neighbour_dv_monotime_cb() invocations,
       continue normal processing */
    n->get = NULL;
    n->dv_monotime_available = GNUNET_YES;
    return;
  }
  if (0 == record->value_size)
  {
    GNUNET_PEERSTORE_iteration_next (n->get, 1);
    GNUNET_break (0);
    return;
  }
  mtbe = record->value;
  n->last_dv_learn_monotime =
    GNUNET_TIME_absolute_max (n->last_dv_learn_monotime,
                              GNUNET_TIME_absolute_ntoh (*mtbe));
  GNUNET_PEERSTORE_iteration_next (n->get, 1);
}

References Neighbour::dv_monotime_available, Neighbour::get, GNUNET_break, GNUNET_PEERSTORE_iteration_next(), GNUNET_TIME_absolute_max(), GNUNET_TIME_absolute_ntoh(), GNUNET_YES, Neighbour::last_dv_learn_monotime, and record().

Referenced by handle_add_queue_message().

Here is the call graph for this function:

Here is the caller graph for this function:

iterate_address_and_compare_cb()

static void iterate_address_and_compare_cb ( void *  cls, const struct GNUNET_PeerIdentity *  pid, const char *  uri  )

static

Definition at line 12470 of file gnunet-service-transport.c.

{
  struct Queue *queue = cls;
  struct sockaddr_in v4;
  const char *slash;
  char *address_uri;
  char *prefix;
  char *uri_without_port;
  char *address_uri_without_port;
 
  slash = strrchr (uri, '/');
  prefix = GNUNET_strndup (uri, (slash - uri) - 2);
  GNUNET_assert (NULL != slash);
  slash++;
  GNUNET_asprintf (&address_uri,
                   "%s-%s",
                   prefix,
                   slash);
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "1 not global natted_address %u %s %s %s\n",
              queue->is_global_natted,
              uri,
              queue->address,
              slash);
 
  uri_without_port = get_address_without_port (address_uri);
  if (1 != inet_pton (AF_INET, uri_without_port, &v4.sin_addr))
  {
    GNUNET_free (prefix);
    GNUNET_free (address_uri);
    GNUNET_free (uri_without_port);
    return;
  }
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "2 not global natted_address %u %s %s\n",
              queue->is_global_natted,
              uri,
              queue->address);
 
  if (GNUNET_NO == queue->is_global_natted)
  {
    GNUNET_free (prefix);
    GNUNET_free (address_uri);
    GNUNET_free (uri_without_port);
    return;
  }
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "3 not global natted_address %u %s %s\n",
              queue->is_global_natted,
              uri,
              queue->address);
 
  if (0 == strcmp (uri_without_port, address_uri))
  {
    GNUNET_free (prefix);
    GNUNET_free (address_uri);
    GNUNET_free (uri_without_port);
    return;
  }
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "4 not global natted_address %u %s %s\n",
              queue->is_global_natted,
              uri,
              queue->address);
 
  address_uri_without_port = get_address_without_port (queue->address);
  if (0 == strcmp (uri_without_port, address_uri_without_port))
  {
    queue->is_global_natted = GNUNET_NO;
  }
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "not global natted_address %u %s %s %s %s %s %u\n",
              queue->is_global_natted,
              uri,
              queue->address,
              uri_without_port,
              address_uri_without_port,
              prefix,
              GNUNET_NO);
  GNUNET_free (prefix);
  GNUNET_free (address_uri);
  GNUNET_free (address_uri_without_port);
  GNUNET_free (uri_without_port);
}

References get_address_without_port(), GNUNET_asprintf(), GNUNET_assert, GNUNET_ERROR_TYPE_DEBUG, GNUNET_free, GNUNET_log, GNUNET_NO, GNUNET_strndup, prefix, queue(), and uri.

Referenced by check_for_global_natted().

Here is the call graph for this function:

Here is the caller graph for this function:

contains_address()

static enum GNUNET_GenericReturnValue contains_address ( void *  cls, const struct GNUNET_PeerIdentity *  pid, void *  value  )

static

Definition at line 12578 of file gnunet-service-transport.c.

{
  struct TransportGlobalNattedAddressClosure *tgna_cls = cls;
  struct TransportGlobalNattedAddress *tgna = value;
  char *addr = (char *) &tgna[1];
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Checking tgna %p with addr %s and length %u compare length %lu\n",
              tgna,
              addr,
              ntohl (tgna->address_length),
              strlen (tgna_cls->addr));
  if (strlen (tgna_cls->addr) == ntohl (tgna->address_length)
      && 0 == strncmp (addr, tgna_cls->addr, ntohl (tgna->address_length)))
  {
    tgna_cls->tgna = tgna;
    return GNUNET_NO;
  }
  return GNUNET_YES;
}

References TransportGlobalNattedAddressClosure::addr, TransportGlobalNattedAddress::address_length, GNUNET_ERROR_TYPE_DEBUG, GNUNET_log, GNUNET_NO, GNUNET_YES, TransportGlobalNattedAddressClosure::tgna, and value.

Referenced by check_for_global_natted().

Here is the caller graph for this function:

check_for_global_natted_error_cb()

static void check_for_global_natted_error_cb ( void *  cls )

static

Definition at line 12603 of file gnunet-service-transport.c.

{
  GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
              "Error in PEERSTORE monitoring for checking global natted\n");
}

References GNUNET_ERROR_TYPE_WARNING, and GNUNET_log.

Referenced by handle_add_queue_message().

Here is the caller graph for this function:

check_for_global_natted_sync_cb()

static void check_for_global_natted_sync_cb ( void *  cls )

static

Definition at line 12611 of file gnunet-service-transport.c.

{
  GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
              "Done with initial PEERSTORE iteration during monitoring  for checking global natted\n");
}

References GNUNET_ERROR_TYPE_WARNING, and GNUNET_log.

Referenced by handle_add_queue_message().

Here is the caller graph for this function:

check_for_global_natted()

static void check_for_global_natted ( void *  cls, const struct GNUNET_PEERSTORE_Record *  record, const char *  emsg  )

static

Definition at line 12619 of file gnunet-service-transport.c.

{
  struct Queue *queue = cls;
  struct Neighbour *neighbour = queue->neighbour;
  struct GNUNET_HELLO_Parser *parser;
  struct GNUNET_MessageHeader *hello;
  struct TransportGlobalNattedAddressClosure tgna_cls;
  size_t address_len_without_port;
 
  if (NULL != emsg)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
                "Got failure from PEERSTORE: %s\n",
                emsg);
    return;
  }
  if (0 == record->value_size)
  {
    GNUNET_PEERSTORE_monitor_next (queue->mo, 1);
    GNUNET_break (0);
    return;
  }
  queue->is_global_natted = GNUNET_YES;
  hello = record->value;
  parser = GNUNET_HELLO_parser_from_msg (hello);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "before not global natted %u\n",
              queue->is_global_natted);
  GNUNET_HELLO_parser_iterate (parser,
                               &iterate_address_and_compare_cb,
                               queue);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "after not global natted %u\n",
              queue->is_global_natted);
  GNUNET_HELLO_parser_free (parser);
 
  tgna_cls.addr = get_address_without_port (queue->address);
  address_len_without_port = strlen (tgna_cls.addr);
  /*{
    char buf[address_len_without_port + 1];
 
      GNUNET_memcpy (&buf, addr, address_len_without_port);
      buf[address_len_without_port] = '\0';
      GNUNET_free (addr);
      GNUNET_memcpy (tgna_cls.addr, buf, address_len_without_port + 1);
      }*/
  tgna_cls.tgna = NULL;
  GNUNET_CONTAINER_multipeermap_get_multiple (neighbour->natted_addresses,
                                              &neighbour->pid,
                                              &contains_address,
                                              &tgna_cls);
  if (NULL != tgna_cls.tgna)
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                " tgna_cls.tgna tgna %p %lu %u %u\n",
                tgna_cls.tgna,
                neighbour->size_of_global_addresses,
                ntohl (tgna_cls.tgna->address_length),
                neighbour->number_of_addresses);
  if (NULL == tgna_cls.tgna && GNUNET_YES == queue->is_global_natted)
  {
    struct TransportGlobalNattedAddress *tgna;
 
    tgna = GNUNET_malloc (sizeof (struct TransportGlobalNattedAddress)
                          + address_len_without_port);
    tgna->address_length = htonl (address_len_without_port);
    GNUNET_memcpy (&tgna[1], tgna_cls.addr, address_len_without_port);
    GNUNET_CONTAINER_multipeermap_put (neighbour->natted_addresses,
                                       &neighbour->pid,
                                       tgna,
                                       GNUNET_CONTAINER_MULTIHASHMAPOPTION_MULTIPLE);
    neighbour->number_of_addresses++;
    neighbour->size_of_global_addresses += address_len_without_port + 1;
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "Created tgna %p with address %s and length %lu\n",
                tgna,
                tgna_cls.addr,
                address_len_without_port + 1);
  }
  else if (NULL != tgna_cls.tgna && GNUNET_NO == queue->is_global_natted)
  {
    GNUNET_CONTAINER_multipeermap_remove (neighbour->natted_addresses,
                                          &neighbour->pid,
                                          tgna_cls.tgna);
    GNUNET_assert (neighbour->size_of_global_addresses >= ntohl (tgna_cls.tgna->
                                                                 address_length)
                   );
    neighbour->size_of_global_addresses -= ntohl (tgna_cls.tgna->address_length)
    ;
    GNUNET_assert (0 < neighbour->number_of_addresses);
    neighbour->number_of_addresses--;
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "removed tgna %p\n",
                tgna_cls.tgna);
    GNUNET_free (tgna_cls.tgna);
  }
  GNUNET_PEERSTORE_monitor_next (queue->mo, 1);
  GNUNET_free (tgna_cls.addr);
}

References TransportGlobalNattedAddressClosure::addr, TransportGlobalNattedAddress::address_length, contains_address(), get_address_without_port(), GNUNET_assert, GNUNET_break, GNUNET_CONTAINER_MULTIHASHMAPOPTION_MULTIPLE, GNUNET_CONTAINER_multipeermap_get_multiple(), GNUNET_CONTAINER_multipeermap_put(), GNUNET_CONTAINER_multipeermap_remove(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_ERROR_TYPE_WARNING, GNUNET_free, GNUNET_HELLO_parser_free(), GNUNET_HELLO_parser_from_msg(), GNUNET_HELLO_parser_iterate(), GNUNET_log, GNUNET_malloc, GNUNET_memcpy, GNUNET_NO, GNUNET_PEERSTORE_monitor_next(), GNUNET_YES, iterate_address_and_compare_cb(), Neighbour::natted_addresses, Neighbour::number_of_addresses, Neighbour::pid, queue(), record(), Neighbour::size_of_global_addresses, and TransportGlobalNattedAddressClosure::tgna.

Referenced by handle_add_queue_message().

Here is the call graph for this function:

Here is the caller graph for this function:

handle_add_queue_message()

static void handle_add_queue_message ( void *  cls, const struct GNUNET_TRANSPORT_AddQueueMessage *  aqm  )

static

New queue became available.

Process the request.

Parameters

cls	the client
aqm	the send message that was sent

Definition at line 12728 of file gnunet-service-transport.c.

{
  struct TransportClient *tc = cls;
  struct Queue *queue;
  struct Neighbour *neighbour;
  const char *addr;
  uint16_t addr_len;
 
  if (ntohl (aqm->mtu) <= sizeof(struct TransportFragmentBoxMessage))
  {
    /* MTU so small as to be useless for transmissions,
       required for #fragment_message()! */
    GNUNET_break_op (0);
    GNUNET_SERVICE_client_drop (tc->client);
    return;
  }
  /* This may simply be a queue update */
  for (queue = tc->details.communicator.queue_head;
       NULL != queue;
       queue = queue->next_client)
  {
    if (queue->qid != ntohl (aqm->qid))
      continue;
    break;
  }
 
  if (NULL != queue)
  {
    neighbour = queue->neighbour;
  }
  else
  {
    struct GNUNET_TIME_Absolute validated_until = GNUNET_TIME_UNIT_ZERO_ABS;
 
    neighbour = lookup_neighbour (&aqm->receiver);
    if (NULL == neighbour)
    {
      neighbour = GNUNET_new (struct Neighbour);
      neighbour->natted_addresses = GNUNET_CONTAINER_multipeermap_create (16,
                                                                          GNUNET_YES);
      neighbour->pid = aqm->receiver;
      GNUNET_assert (GNUNET_OK ==
                     GNUNET_CONTAINER_multipeermap_put (
                       neighbours,
                       &neighbour->pid,
                       neighbour,
                       GNUNET_CONTAINER_MULTIHASHMAPOPTION_UNIQUE_ONLY));
      neighbour->get =
        GNUNET_PEERSTORE_iteration_start (peerstore,
                                          "transport",
                                          &neighbour->pid,
                                          GNUNET_PEERSTORE_TRANSPORT_DVLEARN_MONOTIME,
                                          &neighbour_dv_monotime_cb,
                                          neighbour);
    }
    addr_len = ntohs (aqm->header.size) - sizeof(*aqm);
    addr = (const char *) &aqm[1];
    GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                "New queue %s to %s available with QID %u and q_len %" PRIu64
                " and mtu %u\n",
                addr,
                GNUNET_i2s (&aqm->receiver),
                ntohl (aqm->qid),
                GNUNET_ntohll (aqm->q_len),
                ntohl (aqm->mtu));
    queue = GNUNET_malloc (sizeof(struct Queue) + addr_len);
    queue->tc = tc;
    for (struct Queue *q = neighbour->queue_head; NULL != q; q = q->
                                                                 next_neighbour)
      validated_until = GNUNET_TIME_absolute_max (validated_until, q->
                                                  validated_until);
    if (0 == GNUNET_TIME_absolute_get_remaining (validated_until).rel_value_us)
    {
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "New queue with QID %u inherit validated until\n",
                  ntohl (aqm->qid));
      queue->validated_until = validated_until;
    }
    queue->address = (const char *) &queue[1];
    queue->pd.aged_rtt = GNUNET_TIME_UNIT_FOREVER_REL;
    queue->qid = ntohl (aqm->qid);
    queue->neighbour = neighbour;
    if (GNUNET_TRANSPORT_QUEUE_LENGTH_UNLIMITED == GNUNET_ntohll (aqm->q_len))
      queue->unlimited_length = GNUNET_YES;
    queue->q_capacity = GNUNET_ntohll (aqm->q_len);
    memcpy (&queue[1], addr, addr_len);
    /* notify monitors about new queue */
    {
      struct MonitorEvent me = { .rtt = queue->pd.aged_rtt, .cs = queue->cs };
 
      notify_monitors (&neighbour->pid, queue->address, queue->nt, &me);
    }
    GNUNET_CONTAINER_MDLL_insert (neighbour,
                                  neighbour->queue_head,
                                  neighbour->queue_tail,
                                  queue);
    GNUNET_CONTAINER_MDLL_insert (client,
                                  tc->details.communicator.queue_head,
                                  tc->details.communicator.queue_tail,
                                  queue);
 
  }
  queue->mtu = ntohl (aqm->mtu);
  queue->nt = ntohl (aqm->nt);
  queue->cs = ntohl (aqm->cs);
  queue->idle = GNUNET_YES;
 
  {
    struct sockaddr_in v4;
    char *addr_without = get_address_without_port (queue->address);
    if (1 == inet_pton (AF_INET, addr_without, &v4.sin_addr))
    {
      GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
                  "start not global natted\n");
      queue->mo = GNUNET_PEERSTORE_monitor_start (GST_cfg,
                                                  GNUNET_YES,
                                                  "peerstore",
                                                  &neighbour->pid,
                                                  GNUNET_PEERSTORE_HELLO_KEY,
                                                  &
                                                  check_for_global_natted_error_cb,
                                                  NULL,
                                                  &
                                                  check_for_global_natted_sync_cb,
                                                  NULL,
                                                  &check_for_global_natted,
                                                  queue);
    }
    GNUNET_free (addr_without);
  }
  /* check if valdiations are waiting for the queue */
  if (GNUNET_YES == GNUNET_CONTAINER_multipeermap_contains (validation_map,
                                                            &aqm->receiver))
  {
    if (GNUNET_SYSERR != GNUNET_CONTAINER_multipeermap_get_multiple (
          validation_map,
          &aqm->
          receiver,
          &
          check_validation_request_pending,
          queue))
      start_address_validation (&aqm->receiver, queue->address);
  }
  else
    start_address_validation (&aqm->receiver, queue->address);
  /* look for traffic for this queue */
  // TODO Check whether this makes any sense at all.
  /*schedule_transmit_on_queue (GNUNET_TIME_UNIT_ZERO,
    queue, GNUNET_SCHEDULER_PRIORITY_DEFAULT);*/
  /* might be our first queue, try launching DV learning */
  if (NULL == dvlearn_task)
    dvlearn_task = GNUNET_SCHEDULER_add_now (&start_dv_learn, NULL);
  GNUNET_SERVICE_client_continue (tc->client);
}

References check_for_global_natted(), check_for_global_natted_error_cb(), check_for_global_natted_sync_cb(), check_validation_request_pending(), GNUNET_TRANSPORT_AddQueueMessage::cs, dvlearn_task, Neighbour::get, get_address_without_port(), GNUNET_assert, GNUNET_break_op, GNUNET_CONTAINER_MDLL_insert, GNUNET_CONTAINER_MULTIHASHMAPOPTION_UNIQUE_ONLY, GNUNET_CONTAINER_multipeermap_contains(), GNUNET_CONTAINER_multipeermap_create(), GNUNET_CONTAINER_multipeermap_get_multiple(), GNUNET_CONTAINER_multipeermap_put(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_free, GNUNET_i2s(), GNUNET_log, GNUNET_malloc, GNUNET_new, GNUNET_ntohll(), GNUNET_OK, GNUNET_PEERSTORE_HELLO_KEY, GNUNET_PEERSTORE_iteration_start(), GNUNET_PEERSTORE_monitor_start(), GNUNET_PEERSTORE_TRANSPORT_DVLEARN_MONOTIME, GNUNET_SCHEDULER_add_now(), GNUNET_SERVICE_client_continue(), GNUNET_SERVICE_client_drop(), GNUNET_SYSERR, GNUNET_TIME_absolute_get_remaining(), GNUNET_TIME_absolute_max(), GNUNET_TIME_UNIT_FOREVER_REL, GNUNET_TIME_UNIT_ZERO_ABS, GNUNET_TRANSPORT_QUEUE_LENGTH_UNLIMITED, GNUNET_YES, GST_cfg, GNUNET_TRANSPORT_AddQueueMessage::header, lookup_neighbour(), me, GNUNET_TRANSPORT_AddQueueMessage::mtu, Neighbour::natted_addresses, neighbour_dv_monotime_cb(), neighbours, notify_monitors(), GNUNET_TRANSPORT_AddQueueMessage::nt, peerstore, Neighbour::pid, q, GNUNET_TRANSPORT_AddQueueMessage::q_len, GNUNET_TRANSPORT_AddQueueMessage::qid, queue(), Neighbour::queue_head, Neighbour::queue_tail, receiver(), GNUNET_TRANSPORT_AddQueueMessage::receiver, GNUNET_TIME_Relative::rel_value_us, GNUNET_MessageHeader::size, start_address_validation(), start_dv_learn(), tc, and validation_map.

Here is the call graph for this function:

handle_update_queue_message()

static void handle_update_queue_message ( void *  cls, const struct GNUNET_TRANSPORT_UpdateQueueMessage *  msg  )

static

Handle updates to queues.

Parameters

cls	the transport client.
msg	Message struct.

Definition at line 12892 of file gnunet-service-transport.c.

{
  struct TransportClient *tc = cls;
  struct Queue *target_queue = NULL;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Received queue update message for %u with q_len %llu and mtu %u\n",
              ntohl (msg->qid),
              (unsigned long long) GNUNET_ntohll (msg->q_len),
              ntohl (msg->mtu));
  for (target_queue = tc->details.communicator.queue_head;
       NULL != target_queue;
       target_queue = target_queue->next_client)
  {
    if (ntohl (msg->qid) == target_queue->qid)
      break;
  }
  if (NULL == target_queue)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
                "Queue to update no longer exists! Discarding update.\n");
    return;
  }
 
  target_queue->nt = msg->nt;
  target_queue->mtu = ntohl (msg->mtu);
  target_queue->cs = msg->cs;
  target_queue->priority = ntohl (msg->priority);
  /* The update message indicates how many messages
   * the queue should be able to handle.
   */
  if (GNUNET_TRANSPORT_QUEUE_LENGTH_UNLIMITED == GNUNET_ntohll (msg->q_len))
    target_queue->unlimited_length = GNUNET_YES;
  else
    target_queue->unlimited_length = GNUNET_NO;
  target_queue->q_capacity += GNUNET_ntohll (msg->q_len);
  if (0 < target_queue->q_capacity)
    schedule_transmit_on_queue (GNUNET_TIME_UNIT_ZERO,
                                target_queue,
                                GNUNET_SCHEDULER_PRIORITY_DEFAULT);
  GNUNET_SERVICE_client_continue (tc->client);
}

References Queue::cs, GNUNET_ERROR_TYPE_DEBUG, GNUNET_ERROR_TYPE_WARNING, GNUNET_log, GNUNET_NO, GNUNET_ntohll(), GNUNET_SCHEDULER_PRIORITY_DEFAULT, GNUNET_SERVICE_client_continue(), GNUNET_TIME_UNIT_ZERO, GNUNET_TRANSPORT_QUEUE_LENGTH_UNLIMITED, GNUNET_YES, msg, Queue::mtu, Queue::next_client, Queue::nt, Queue::priority, Queue::q_capacity, Queue::qid, schedule_transmit_on_queue(), tc, and Queue::unlimited_length.

Here is the call graph for this function:

handle_queue_create_ok()

static void handle_queue_create_ok ( void *  cls, const struct GNUNET_TRANSPORT_CreateQueueResponse *  cqr  )

static

Communicator tells us that our request to create a queue "worked", that is setting up the queue is now in process.

Parameters

cls	the struct TransportClient
cqr	confirmation message

Definition at line 12946 of file gnunet-service-transport.c.

{
  struct TransportClient *tc = cls;
 
  if (CT_COMMUNICATOR != tc->type)
  {
    GNUNET_break (0);
    GNUNET_SERVICE_client_drop (tc->client);
    return;
  }
  GNUNET_STATISTICS_update (GST_stats,
                            "# Suggestions succeeded at communicator",
                            1,
                            GNUNET_NO);
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Request #%u for communicator to create queue succeeded\n",
              (unsigned int) ntohs (cqr->request_id));
  GNUNET_SERVICE_client_continue (tc->client);
}

References CT_COMMUNICATOR, GNUNET_break, GNUNET_ERROR_TYPE_DEBUG, GNUNET_log, GNUNET_NO, GNUNET_SERVICE_client_continue(), GNUNET_SERVICE_client_drop(), GNUNET_STATISTICS_update(), GST_stats, GNUNET_TRANSPORT_CreateQueueResponse::request_id, and tc.

Here is the call graph for this function:

handle_queue_create_fail()

static void handle_queue_create_fail ( void *  cls, const struct GNUNET_TRANSPORT_CreateQueueResponse *  cqr  )

static

Communicator tells us that our request to create a queue failed.

This usually indicates that the provided address is simply invalid or that the communicator's resources are exhausted.

Parameters

cls	the struct TransportClient
cqr	failure message

Definition at line 12977 of file gnunet-service-transport.c.

{
  struct TransportClient *tc = cls;
 
  if (CT_COMMUNICATOR != tc->type)
  {
    GNUNET_break (0);
    GNUNET_SERVICE_client_drop (tc->client);
    return;
  }
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Request #%u for communicator to create queue failed\n",
              (unsigned int) ntohl (cqr->request_id));
  GNUNET_STATISTICS_update (GST_stats,
                            "# Suggestions failed in queue creation at communicator",
                            1,
                            GNUNET_NO);
  GNUNET_SERVICE_client_continue (tc->client);
}

References CT_COMMUNICATOR, GNUNET_break, GNUNET_ERROR_TYPE_DEBUG, GNUNET_log, GNUNET_NO, GNUNET_SERVICE_client_continue(), GNUNET_SERVICE_client_drop(), GNUNET_STATISTICS_update(), GST_stats, GNUNET_TRANSPORT_CreateQueueResponse::request_id, and tc.

Here is the call graph for this function:

handle_suggest_cancel()

static void handle_suggest_cancel ( void *  cls, const struct ExpressPreferenceMessage *  msg  )

static

We have received a struct ExpressPreferenceMessage from an application client.

Parameters

cls	handle to the client
msg	the start message

Definition at line 13008 of file gnunet-service-transport.c.

{
  struct TransportClient *tc = cls;
  struct PeerRequest *pr;
 
  if (CT_APPLICATION != tc->type)
  {
    GNUNET_break (0);
    GNUNET_SERVICE_client_drop (tc->client);
    return;
  }
  pr = GNUNET_CONTAINER_multipeermap_get (tc->details.application.requests,
                                          &msg->peer);
  if (NULL == pr)
  {
    GNUNET_break (0);
    GNUNET_SERVICE_client_drop (tc->client);
    return;
  }
  (void) stop_peer_request (tc, &pr->pid, pr);
  GNUNET_SERVICE_client_continue (tc->client);
}

References CT_APPLICATION, GNUNET_break, GNUNET_CONTAINER_multipeermap_get(), GNUNET_SERVICE_client_continue(), GNUNET_SERVICE_client_drop(), msg, PeerRequest::pr, stop_peer_request(), and tc.

Here is the call graph for this function:

hello_for_client_cb()

static void hello_for_client_cb ( void *  cls, const struct GNUNET_PeerIdentity *  pid, const char *  uri  )

static

Definition at line 13033 of file gnunet-service-transport.c.

{
  struct Queue *q;
  int pfx_len;
  const char *eou;
  char *address;
  (void) cls;
 
  eou = strstr (uri,
                "://");
  pfx_len = eou - uri;
  eou += 3;
  GNUNET_asprintf (&address,
                   "%.*s-%s",
                   pfx_len,
                   uri,
                   eou);
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "hello for client %s\n",
              address);
 
  q = find_queue (pid, address);
  if (NULL == q)
  {
    suggest_to_connect (pid, address);
  }
  else
    start_address_validation (pid, address);
  GNUNET_free (address);
}

References address, find_queue(), GNUNET_asprintf(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_free, GNUNET_log, pid, q, start_address_validation(), suggest_to_connect(), and uri.

Referenced by handle_hello_for_client().

Here is the call graph for this function:

Here is the caller graph for this function:

handle_hello_for_client()

static void handle_hello_for_client ( void *  cls, const struct GNUNET_PEERSTORE_Record *  record, const char *  emsg  )

static

Function called by PEERSTORE for each matching record.

Parameters

cls	closure, a struct PeerRequest
record	peerstore record information
emsg	error message, or NULL if no errors

Definition at line 13076 of file gnunet-service-transport.c.

{
  struct PeerRequest *pr = cls;
  struct GNUNET_HELLO_Parser *parser;
  struct GNUNET_MessageHeader *hello;
 
  if (NULL != emsg)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
                "Got failure from PEERSTORE: %s\n",
                emsg);
    return;
  }
  if (NULL == GST_my_identity)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
                "No identity given yet!\n");
    return;
  }
  hello = record->value;
  if (0 == GNUNET_memcmp (&record->peer, GST_my_identity))
  {
    GNUNET_PEERSTORE_monitor_next (pr->nc, 1);
    return;
  }
  parser = GNUNET_HELLO_parser_from_msg (hello);
  if (NULL == parser)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
                "HELLO cannot be parsed!\n");
    return;
  }
  GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
              "HELLO for `%s' could be parsed, iterating addresses...!\n",
              GNUNET_i2s (GNUNET_HELLO_parser_get_id (parser)));
  GNUNET_HELLO_parser_iterate (parser,
                               hello_for_client_cb,
                               NULL);
  GNUNET_HELLO_parser_free (parser);
}

References GNUNET_ERROR_TYPE_WARNING, GNUNET_HELLO_parser_free(), GNUNET_HELLO_parser_from_msg(), GNUNET_HELLO_parser_get_id(), GNUNET_HELLO_parser_iterate(), GNUNET_i2s(), GNUNET_log, GNUNET_memcmp, GNUNET_PEERSTORE_monitor_next(), GST_my_identity, hello_for_client_cb(), PeerRequest::nc, PeerRequest::pr, and record().

Referenced by handle_suggest().

Here is the call graph for this function:

Here is the caller graph for this function:

hello_for_client_error_cb()

static void hello_for_client_error_cb ( void *  cls )

static

Definition at line 13121 of file gnunet-service-transport.c.

{
  GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
              "Error in PEERSTORE monitoring\n");
}

References GNUNET_ERROR_TYPE_WARNING, and GNUNET_log.

Referenced by handle_suggest().

Here is the caller graph for this function:

hello_for_client_sync_cb()

static void hello_for_client_sync_cb ( void *  cls )

static

Definition at line 13129 of file gnunet-service-transport.c.

{
  GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
              "Done with initial PEERSTORE iteration during monitoring\n");
}

References GNUNET_ERROR_TYPE_WARNING, and GNUNET_log.

Referenced by handle_suggest().

Here is the caller graph for this function:

handle_suggest()

static void handle_suggest ( void *  cls, const struct ExpressPreferenceMessage *  msg  )

static

We have received a struct ExpressPreferenceMessage from an application client.

Parameters

cls	handle to the client
msg	the start message

Definition at line 13144 of file gnunet-service-transport.c.

{
  struct TransportClient *tc = cls;
  struct PeerRequest *pr;
 
  if (CT_NONE == tc->type)
  {
    tc->type = CT_APPLICATION;
    tc->details.application.requests =
      GNUNET_CONTAINER_multipeermap_create (16, GNUNET_YES);
  }
  if (CT_APPLICATION != tc->type)
  {
    GNUNET_break (0);
    GNUNET_SERVICE_client_drop (tc->client);
    return;
  }
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Client suggested we talk to %s with preference %d at rate %u\n",
              GNUNET_i2s (&msg->peer),
              (int) ntohl (msg->pk),
              (int) ntohl (msg->bw.value__));
  if (0 == GNUNET_memcmp (GST_my_identity, &msg->peer))
  {
    GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
                "Client suggested connection to ourselves, ignoring...\n");
    return;
  }
  pr = GNUNET_new (struct PeerRequest);
  pr->tc = tc;
  pr->pid = msg->peer;
  pr->bw = msg->bw;
  pr->pk = ntohl (msg->pk);
  if (GNUNET_YES != GNUNET_CONTAINER_multipeermap_put (
        tc->details.application.requests,
        &pr->pid,
        pr,
        GNUNET_CONTAINER_MULTIHASHMAPOPTION_UNIQUE_ONLY))
  {
    GNUNET_break (0);
    GNUNET_free (pr);
    GNUNET_SERVICE_client_drop (tc->client);
    return;
  }
  pr->nc =
    GNUNET_PEERSTORE_monitor_start (GST_cfg,
                                    GNUNET_YES,
                                    "peerstore",
                                    NULL,
                                    GNUNET_PEERSTORE_HELLO_KEY,
                                    &hello_for_client_error_cb,
                                    NULL,
                                    &hello_for_client_sync_cb,
                                    NULL,
                                    &handle_hello_for_client,
                                    pr);
  GNUNET_SERVICE_client_continue (tc->client);
}

References CT_APPLICATION, CT_NONE, GNUNET_break, GNUNET_CONTAINER_MULTIHASHMAPOPTION_UNIQUE_ONLY, GNUNET_CONTAINER_multipeermap_create(), GNUNET_CONTAINER_multipeermap_put(), GNUNET_ERROR_TYPE_DEBUG, GNUNET_ERROR_TYPE_WARNING, GNUNET_free, GNUNET_i2s(), GNUNET_log, GNUNET_memcmp, GNUNET_new, GNUNET_PEERSTORE_HELLO_KEY, GNUNET_PEERSTORE_monitor_start(), GNUNET_SERVICE_client_continue(), GNUNET_SERVICE_client_drop(), GNUNET_YES, GST_cfg, GST_my_identity, handle_hello_for_client(), hello_for_client_error_cb(), hello_for_client_sync_cb(), msg, PeerRequest::pr, and tc.

Here is the call graph for this function:

check_request_hello_validation()

static int check_request_hello_validation ( void *  cls, const struct RequestHelloValidationMessage *  m  )

static

Check GNUNET_MESSAGE_TYPE_TRANSPORT_REQUEST_HELLO_VALIDATION messages.

Parameters

cls	a struct TransportClient *
m	message to verify

Returns

GNUNET_OK on success

Definition at line 13213 of file gnunet-service-transport.c.

{
  (void) cls;
  GNUNET_MQ_check_zero_termination (m);
  return GNUNET_OK;
}

References GNUNET_MQ_check_zero_termination, GNUNET_OK, and m.

handle_request_hello_validation()

static void handle_request_hello_validation ( void *  cls, const struct RequestHelloValidationMessage *  m  )

static

A client encountered an address of another peer.

Consider validating it, and if validation succeeds, persist it to PEERSTORE.

Parameters

cls	a struct TransportClient *
m	message to verify

Definition at line 13230 of file gnunet-service-transport.c.

{
  struct TransportClient *tc = cls;
  struct Queue *q;
 
  q = find_queue (&m->peer, (const char *) &m[1]);
  if (NULL == q)
  {
    suggest_to_connect (&m->peer, (const char *) &m[1]);
  }
  else
    start_address_validation (&m->peer, (const char *) &m[1]);
  GNUNET_SERVICE_client_continue (tc->client);
}

References find_queue(), GNUNET_SERVICE_client_continue(), m, q, start_address_validation(), suggest_to_connect(), and tc.

Here is the call graph for this function:

free_neighbour_cb()

static int free_neighbour_cb ( void *  cls, const struct GNUNET_PeerIdentity *  pid, void *  value  )

static

Free neighbour entry.

Parameters

cls	NULL
pid	unused
value	a struct Neighbour

Returns

GNUNET_OK (always)

Definition at line 13256 of file gnunet-service-transport.c.

{
  struct Neighbour *neighbour = value;
 
  (void) cls;
  (void) pid;
  GNUNET_break (0);  // should this ever happen?
  free_neighbour (neighbour, GNUNET_YES);
 
  return GNUNET_OK;
}

References free_neighbour(), GNUNET_break, GNUNET_OK, GNUNET_YES, pid, and value.

Referenced by do_shutdown().

Here is the call graph for this function:

Here is the caller graph for this function:

free_dv_routes_cb()

static int free_dv_routes_cb ( void *  cls, const struct GNUNET_PeerIdentity *  pid, void *  value  )

static

Free DV route entry.

Parameters

cls	NULL
pid	unused
value	a struct DistanceVector

Returns

GNUNET_OK (always)

Definition at line 13280 of file gnunet-service-transport.c.

{
  struct DistanceVector *dv = value;
 
  (void) cls;
  (void) pid;
  free_dv_route (dv);
 
  return GNUNET_OK;
}

References free_dv_route(), GNUNET_OK, pid, and value.

Referenced by do_shutdown().

Here is the call graph for this function:

Here is the caller graph for this function:

free_validation_state_cb()

static int free_validation_state_cb ( void *  cls, const struct GNUNET_PeerIdentity *  pid, void *  value  )

static

Free validation state.

Parameters

cls	NULL
pid	unused
value	a struct ValidationState

Returns

GNUNET_OK (always)

Definition at line 13303 of file gnunet-service-transport.c.

{
  struct ValidationState *vs = value;
 
  (void) cls;
  (void) pid;
  free_validation_state (vs);
  return GNUNET_OK;
}

References free_validation_state(), GNUNET_OK, pid, and value.

Referenced by do_shutdown().

Here is the call graph for this function:

Here is the caller graph for this function:

free_pending_ack_cb()

static int free_pending_ack_cb ( void *  cls, const struct GNUNET_Uuid *  key, void *  value  )

static

Free pending acknowledgement.

Parameters

cls	NULL
key	unused
value	a struct PendingAcknowledgement

Returns

GNUNET_OK (always)

Definition at line 13325 of file gnunet-service-transport.c.

{
  struct PendingAcknowledgement *pa = value;
 
  (void) cls;
  (void) key;
  free_pending_acknowledgement (pa);
  return GNUNET_OK;
}

References free_pending_acknowledgement(), GNUNET_OK, key, and value.

Referenced by do_shutdown().

Here is the call graph for this function:

Here is the caller graph for this function:

free_ack_cummulator_cb()

static int free_ack_cummulator_cb ( void *  cls, const struct GNUNET_PeerIdentity *  pid, void *  value  )

static

Free acknowledgement cummulator.

Parameters

cls	NULL
pid	unused
value	a struct AcknowledgementCummulator

Returns

GNUNET_OK (always)

Definition at line 13345 of file gnunet-service-transport.c.

{
  struct AcknowledgementCummulator *ac = value;
 
  (void) cls;
  (void) pid;
  GNUNET_SCHEDULER_cancel (ac->task);
  GNUNET_free (ac);
  return GNUNET_OK;
}

References GNUNET_free, GNUNET_OK, GNUNET_SCHEDULER_cancel(), pid, AcknowledgementCummulator::task, and value.

Referenced by do_shutdown().

Here is the call graph for this function:

Here is the caller graph for this function:

shutdown_task()

static void shutdown_task ( void *  cls )

static

Definition at line 13476 of file gnunet-service-transport.c.

{
  in_shutdown = GNUNET_YES;
 
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "Shutdown task executed\n");
  if (NULL != clients_head)
  {
    for (struct TransportClient *tc = clients_head; NULL != tc; tc = tc->next)
    {
      GNUNET_log (GNUNET_ERROR_TYPE_ERROR,
                  "client still connected: %u\n",
                  tc->type);
    }
  }
  else
    do_shutdown (cls);
 
}

References clients_head, do_shutdown, GNUNET_ERROR_TYPE_DEBUG, GNUNET_ERROR_TYPE_ERROR, GNUNET_log, GNUNET_YES, in_shutdown, and tc.

update_hello_from_pid_change_cb()

static void update_hello_from_pid_change_cb ( void *  cls, int  success  )

static

Definition at line 13503 of file gnunet-service-transport.c.

{
  struct UpdateHelloFromPidCtx *pc = cls;
 
  if (GNUNET_OK != success)
  {
    GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
                "Failed to store our new hello with peerstore\n");
  }
  GNUNET_free (pc);
  GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
              "Stored our new hello with peerstore\n");
}

References GNUNET_ERROR_TYPE_WARNING, GNUNET_free, GNUNET_log, GNUNET_OK, and pc.

Referenced by pils_pid_change_cb().

Here is the caller graph for this function:

print_address_list()

void print_address_list	(	void *	cls,
		const struct GNUNET_PeerIdentity *	pid,
		const char *	uri
	)

Definition at line 13519 of file gnunet-service-transport.c.

{
  GNUNET_log (GNUNET_ERROR_TYPE_DEBUG,
              "%s\n", uri);
}

References GNUNET_ERROR_TYPE_DEBUG, GNUNET_log, and uri.

Referenced by pils_pid_change_cb().

Here is the caller graph for this function:

pils_pid_change_cb()

static void pils_pid_change_cb ( void *  cls, const struct GNUNET_HELLO_Parser *  parser, const struct GNUNET_HashCode *  hash  )

static

Callback called when pils service updates us with our new peer identity.

Parameters

cls	closure given to GNUNET_PILS_connect
peer_id	the new peer id
hash	The hash of addresses the peer id is based on. This hash is also returned by #GNUNET_PILS_feed_address.

FIXME we may want to have a sanity check here that verifies that our address list in the builder is the same as the one in the parser (and hence derived from the correct set). If it is NOT, then it is very likely that PILS is behind and will send another update shortly. In this case, we may want to hold off and not generate the new HELLO.

Definition at line 13538 of file gnunet-service-transport.c.

{
  struct GNUNET_MQ_Envelope *env;
  const struct GNUNET_MessageHeader *msg;
  struct UpdateHelloFromPidCtx *sc;
  struct GNUNET_HELLO_Builder *nbuilder;
  struct GNUNET_PeerIdentity npid;
 
  if (NULL == GST_my_identity)
    GST_my_identity = GNUNET_new (struct GNUNET_PeerIdentity);
  if (NULL == GST_my_hello)
    GST_my_hello = GNUNET_HELLO_builder_new ();
  GNUNET_log (GNUNET_ERROR_TYPE_INFO,
              "My current identity is `%s'\n",
              GNUNET_i2s_full (GST_my_identity));
  nbuilder = GNUNET_HELLO_builder_from_parser (parser,
                                               &npid);
  if (GNUNET_NO ==
      GNUNET_HELLO_builder_address_list_cmp (GST_my_hello, nbuilder))
  {
    GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
                "New PID from PILS is derived from address list inconsistend with ours. Ignoring...\n");
    GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
                "Proposed address list:\n");
    GNUNET_HELLO_builder_iterate (nbuilder, &print_address_list, NULL);
    GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
                "Current address list:\n");
    GNUNET_HELLO_builder_iterate (GST_my_hello, &print_address_list, NULL);
    GNUNET_HELLO_builder_free (nbuilder);
    return;
  }
  GST_my_hello = nbuilder;
  memcpy (GST_my_identity, &npid, sizeof npid);
  GNUNET_log (GNUNET_ERROR_TYPE_INFO,
              "My new identity is `%s'\n",
              GNUNET_i2s_full (GST_my_identity));
  env = GNUNET_HELLO_parser_to_env (parser);
  msg = GNUNET_MQ_env_get_msg (env);
  sc = GNUNET_new (struct UpdateHelloFromPidCtx);
  sc->sc = GNUNET_PEERSTORE_hello_add (peerstore,
                                       msg,
                                       update_hello_from_pid_change_cb,
                                       sc);
  GNUNET_free (env);
}

References env, GNUNET_ERROR_TYPE_INFO, GNUNET_ERROR_TYPE_WARNING, GNUNET_free, GNUNET_HELLO_builder_address_list_cmp(), GNUNET_HELLO_builder_free(), GNUNET_HELLO_builder_from_parser(), GNUNET_HELLO_builder_iterate(), GNUNET_HELLO_builder_new(), GNUNET_HELLO_parser_to_env(), GNUNET_i2s_full(), GNUNET_log, GNUNET_MQ_env_get_msg(), GNUNET_new, GNUNET_NO, GNUNET_PEERSTORE_hello_add(), GST_my_hello, GST_my_identity, msg, peerstore, print_address_list(), sc, and update_hello_from_pid_change_cb().

Referenced by run().

Here is the call graph for this function:

Here is the caller graph for this function:

run()

static void run ( void *  cls, const struct GNUNET_CONFIGURATION_Handle *  c, struct GNUNET_SERVICE_Handle *  service  )

static

Initiate transport service.

Parameters

cls	closure
c	configuration to use
service	the initialized service

Definition at line 13605 of file gnunet-service-transport.c.

{
  (void) cls;
  (void) service;
  /* setup globals */
  hello_mono_time = GNUNET_TIME_absolute_get_monotonic (c);
  in_shutdown = GNUNET_NO;
  GST_cfg = c;
  backtalkers = GNUNET_CONTAINER_multipeermap_create (16, GNUNET_YES);
  pending_acks = GNUNET_CONTAINER_multiuuidmap_create (32768, GNUNET_YES);
  ack_cummulators = GNUNET_CONTAINER_multipeermap_create (256, GNUNET_YES);
  neighbours = GNUNET_CONTAINER_multipeermap_create (1024, GNUNET_YES);
  links = GNUNET_CONTAINER_multipeermap_create (512, GNUNET_YES);
  dv_routes = GNUNET_CONTAINER_multipeermap_create (1024, GNUNET_YES);
  dvlearn_map = GNUNET_CONTAINER_multishortmap_create (2 * MAX_DV_LEARN_PENDING,
                                                       GNUNET_YES);
  validation_map = GNUNET_CONTAINER_multipeermap_create (1024, GNUNET_YES);
  revalidation_map = GNUNET_CONTAINER_multihashmap_create (1024, GNUNET_YES);
  validation_heap =
    GNUNET_CONTAINER_heap_create (GNUNET_CONTAINER_HEAP_ORDER_MIN);
  // TODO check for all uses of GST_my_hello that it is not used uninitialized
  use_burst = GNUNET_CONFIGURATION_get_value_yesno (GST_cfg,
                                                    "transport",
                                                    "USE_BURST_NAT");
  if (GNUNET_SYSERR == use_burst)
    GNUNET_log (GNUNET_ERROR_TYPE_WARNING,
                "Could not configure burst nat use. Default to no.\n");
  GST_my_hello = GNUNET_HELLO_builder_new ();
  GST_stats = GNUNET_STATISTICS_create ("transport", GST_cfg);
  GNUNET_SCHEDULER_add_shutdown (&shutdown_task, NULL);
  peerstore = GNUNET_PEERSTORE_connect (GST_cfg);
  nh = GNUNET_NAT_register (GST_cfg,
                            "transport",
                            0,
                            0,
                            NULL,
                            0,
                            NULL,
                            NULL,
                            NULL);
  if (NULL == peerstore)
  {
    GNUNET_break (0);
    GNUNET_SCHEDULER_shutdown ();
    return;
  }
  GST_my_identity = NULL;
  pils = GNUNET_PILS_connect (GST_cfg,
                              pils_pid_change_cb,
                              NULL);          // FIXME we need to wait for
  // our first peer id before
  // we can start the service
  // completely - PILS in turn
  // waits for the first
  // addresses from the
  // communicators in order to
  // be able to generate a
  // peer id
  if (NULL == pils)
  {
    GNUNET_break (0);
    GNUNET_SCHEDULER_shutdown ();
    return;
  }
}

References ack_cummulators, backtalkers, dv_routes, dvlearn_map, GNUNET_break, GNUNET_CONFIGURATION_get_value_yesno(), GNUNET_CONTAINER_heap_create(), GNUNET_CONTAINER_HEAP_ORDER_MIN, GNUNET_CONTAINER_multihashmap_create(), GNUNET_CONTAINER_multipeermap_create(), GNUNET_CONTAINER_multishortmap_create(), GNUNET_CONTAINER_multiuuidmap_create(), GNUNET_ERROR_TYPE_WARNING, GNUNET_HELLO_builder_new(), GNUNET_log, GNUNET_NAT_register(), GNUNET_NO, GNUNET_PEERSTORE_connect(), GNUNET_PILS_connect(), GNUNET_SCHEDULER_add_shutdown(), GNUNET_SCHEDULER_shutdown(), GNUNET_STATISTICS_create(), GNUNET_SYSERR, GNUNET_TIME_absolute_get_monotonic(), GNUNET_YES, GST_cfg, GST_my_hello, GST_my_identity, GST_stats, hello_mono_time, in_shutdown, links, MAX_DV_LEARN_PENDING, neighbours, nh, peerstore, pending_acks, pils, pils_pid_change_cb(), revalidation_map, service, shutdown_task, use_burst, validation_heap, and validation_map.

Here is the call graph for this function:

GNUNET_SERVICE_MAIN()

GNUNET_SERVICE_MAIN	(	GNUNET_OS_project_data_gnunet()	,
		"transport"	,
		GNUNET_SERVICE_OPTION_SOFT_SHUTDOWN	,
		&	run,
		&	client_connect_cb,
		&	client_disconnect_cb,
		NULL	,
		GNUNET_MQ_hd_fixed_size(suggest, GNUNET_MESSAGE_TYPE_TRANSPORT_SUGGEST, struct ExpressPreferenceMessage, NULL)	,
		GNUNET_MQ_hd_fixed_size(suggest_cancel, GNUNET_MESSAGE_TYPE_TRANSPORT_SUGGEST_CANCEL, struct ExpressPreferenceMessage, NULL)	,
		GNUNET_MQ_hd_var_size(request_hello_validation, GNUNET_MESSAGE_TYPE_TRANSPORT_REQUEST_HELLO_VALIDATION, struct RequestHelloValidationMessage, NULL)	,
		GNUNET_MQ_hd_fixed_size(client_start, GNUNET_MESSAGE_TYPE_TRANSPORT_START, struct StartMessage, NULL)	,
		GNUNET_MQ_hd_var_size(client_send, GNUNET_MESSAGE_TYPE_TRANSPORT_SEND, struct OutboundMessage, NULL)	,
		GNUNET_MQ_hd_fixed_size(client_recv_ok, GNUNET_MESSAGE_TYPE_TRANSPORT_RECV_OK, struct RecvOkMessage, NULL)	,
		GNUNET_MQ_hd_var_size(communicator_available, GNUNET_MESSAGE_TYPE_TRANSPORT_NEW_COMMUNICATOR, struct GNUNET_TRANSPORT_CommunicatorAvailableMessage, NULL)	,
		GNUNET_MQ_hd_var_size(communicator_backchannel, GNUNET_MESSAGE_TYPE_TRANSPORT_COMMUNICATOR_BACKCHANNEL, struct GNUNET_TRANSPORT_CommunicatorBackchannel, NULL)	,
		GNUNET_MQ_hd_var_size(add_address, GNUNET_MESSAGE_TYPE_TRANSPORT_ADD_ADDRESS, struct GNUNET_TRANSPORT_AddAddressMessage, NULL)	,
		GNUNET_MQ_hd_fixed_size(del_address, GNUNET_MESSAGE_TYPE_TRANSPORT_DEL_ADDRESS, struct GNUNET_TRANSPORT_DelAddressMessage, NULL)	,
		GNUNET_MQ_hd_var_size(incoming_msg, GNUNET_MESSAGE_TYPE_TRANSPORT_INCOMING_MSG, struct GNUNET_TRANSPORT_IncomingMessage, NULL)	,
		GNUNET_MQ_hd_fixed_size(queue_create_ok, GNUNET_MESSAGE_TYPE_TRANSPORT_QUEUE_CREATE_OK, struct GNUNET_TRANSPORT_CreateQueueResponse, NULL)	,
		GNUNET_MQ_hd_fixed_size(queue_create_fail, GNUNET_MESSAGE_TYPE_TRANSPORT_QUEUE_CREATE_FAIL, struct GNUNET_TRANSPORT_CreateQueueResponse, NULL)	,
		GNUNET_MQ_hd_var_size(add_queue_message, GNUNET_MESSAGE_TYPE_TRANSPORT_QUEUE_SETUP, struct GNUNET_TRANSPORT_AddQueueMessage, NULL)	,
		GNUNET_MQ_hd_fixed_size(update_queue_message, GNUNET_MESSAGE_TYPE_TRANSPORT_QUEUE_UPDATE, struct GNUNET_TRANSPORT_UpdateQueueMessage, NULL)	,
		GNUNET_MQ_hd_fixed_size(del_queue_message, GNUNET_MESSAGE_TYPE_TRANSPORT_QUEUE_TEARDOWN, struct GNUNET_TRANSPORT_DelQueueMessage, NULL)	,
		GNUNET_MQ_hd_fixed_size(send_message_ack, GNUNET_MESSAGE_TYPE_TRANSPORT_SEND_MSG_ACK, struct GNUNET_TRANSPORT_SendMessageToAck, NULL)	,
		GNUNET_MQ_hd_fixed_size(burst_finished, GNUNET_MESSAGE_TYPE_TRANSPORT_BURST_FINISHED, struct GNUNET_TRANSPORT_BurstFinished, NULL)	,
		GNUNET_MQ_hd_fixed_size(monitor_start, GNUNET_MESSAGE_TYPE_TRANSPORT_MONITOR_START, struct GNUNET_TRANSPORT_MonitorStart, NULL)	,
		GNUNET_MQ_handler_end()
	)

Define "main" method using service macro.

Variable Documentation

ring_buffer

struct RingBufferEntry* ring_buffer[RING_BUFFER_SIZE]

static

Ring buffer for a CORE message we did not deliver to CORE, because of missing virtual link to sender.

Definition at line 2876 of file gnunet-service-transport.c.

Referenced by handle_raw_message(), and send_msg_from_cache().

ring_buffer_head

unsigned int ring_buffer_head

static

Head of the ring buffer.

Definition at line 2881 of file gnunet-service-transport.c.

Referenced by handle_raw_message(), and send_msg_from_cache().

is_ring_buffer_full

unsigned int is_ring_buffer_full

static

Is the ring buffer filled up to RING_BUFFER_SIZE.

Definition at line 2886 of file gnunet-service-transport.c.

Referenced by handle_raw_message(), and send_msg_from_cache().

ring_buffer_dv

struct PendingMessage* ring_buffer_dv[RING_BUFFER_SIZE]

static

Ring buffer for a forwarded DVBox message we did not deliver to the next hop, because of missing virtual link that hop.

Definition at line 2891 of file gnunet-service-transport.c.

Referenced by forward_dv_box(), and send_msg_from_cache().

ring_buffer_dv_head

unsigned int ring_buffer_dv_head

static

Head of the ring buffer.

Definition at line 2896 of file gnunet-service-transport.c.

Referenced by forward_dv_box(), and send_msg_from_cache().

is_ring_buffer_dv_full

unsigned int is_ring_buffer_dv_full

static

Is the ring buffer filled up to RING_BUFFER_SIZE.

Definition at line 2901 of file gnunet-service-transport.c.

Referenced by forward_dv_box(), and send_msg_from_cache().

clients_head

struct TransportClient* clients_head

static

Head of linked list of all clients to this service.

Definition at line 2906 of file gnunet-service-transport.c.

Referenced by client_connect_cb(), client_disconnect_cb(), cores_send_connect_info(), cores_send_disconnect_info(), finish_handling_raw_message(), handle_backchannel_encapsulation(), lookup_communicator(), notify_monitors(), shutdown_task(), and start_burst().

clients_tail

struct TransportClient* clients_tail

static

Tail of linked list of all clients to this service.

Definition at line 2911 of file gnunet-service-transport.c.

Referenced by client_connect_cb(), and client_disconnect_cb().

GST_stats

struct GNUNET_STATISTICS_Handle* GST_stats

static

Statistics handle.

Definition at line 2916 of file gnunet-service-transport.c.

Referenced by backtalker_monotime_cb(), decaps_dv_box_cb(), do_shutdown(), finish_handling_raw_message(), free_queue(), free_queue_entry(), handle_backchannel_encapsulation(), handle_client_recv_ok(), handle_client_send(), handle_dv_box(), handle_dv_learn(), handle_flow_control(), handle_queue_create_fail(), handle_queue_create_ok(), handle_reliability_ack(), handle_validation_response(), learn_dv_path(), peerstore_store_validation_cb(), route_control_message_without_fc(), route_via_neighbour(), run(), schedule_transmit_on_queue(), and suggest_to_connect().

GST_cfg

const struct GNUNET_CONFIGURATION_Handle* GST_cfg

static

Configuration handle.

Definition at line 2921 of file gnunet-service-transport.c.

Referenced by consider_sending_fc(), handle_add_queue_message(), handle_suggest(), handle_validation_response(), run(), sign_ephemeral(), sign_t_validation_cb(), start_address_validation(), start_dv_learn(), validation_start_cb(), and validation_transmit_on_queue().

GST_my_identity

struct GNUNET_PeerIdentity* GST_my_identity

static

Our public key.

Definition at line 2926 of file gnunet-service-transport.c.

Referenced by check_dv_box(), check_dv_learn(), decaps_dv_box_cb(), do_shutdown(), encapsulate_for_dv(), forward_dv_learn(), handle_backchannel_encapsulation(), handle_communicator_available(), handle_dv_box(), handle_dv_learn(), handle_hello_for_client(), handle_hello_for_incoming(), handle_suggest(), learn_dv_path(), pils_pid_change_cb(), pils_sign_addr_cb(), run(), sign_dv_init_cb(), and store_pi().

GST_my_hello

struct GNUNET_HELLO_Builder* GST_my_hello

Our HELLO.

Definition at line 2931 of file gnunet-service-transport.c.

Referenced by do_shutdown(), feed_addresses_to_pils(), handle_add_address(), handle_del_address(), peerstore_store_own_cb(), pils_pid_change_cb(), pils_sign_hello_cb(), run(), and store_pi().

neighbours

struct GNUNET_CONTAINER_MultiPeerMap* neighbours

static

Map from PIDs to struct Neighbour entries.

A peer is a neighbour if we have an MQ to it from some communicator.

Definition at line 2937 of file gnunet-service-transport.c.

Referenced by do_shutdown(), free_neighbour(), handle_add_queue_message(), handle_dv_learn(), handle_monitor_start(), lookup_neighbour(), run(), sign_dv_init_cb(), and start_dv_learn().

backtalkers

struct GNUNET_CONTAINER_MultiPeerMap* backtalkers

static

Map from PIDs to struct Backtalker entries.

A peer is a backtalker if it recently send us backchannel messages.

Definition at line 2943 of file gnunet-service-transport.c.

Referenced by decaps_dv_box_cb(), do_shutdown(), free_backtalker(), and run().

ack_cummulators

struct GNUNET_CONTAINER_MultiPeerMap* ack_cummulators

static

Map from PIDs to struct AcknowledgementCummulators.

Here we track the cumulative ACKs for transmission.

Definition at line 2949 of file gnunet-service-transport.c.

Referenced by cummulative_ack(), destroy_ack_cummulator(), do_shutdown(), and run().

pending_acks

struct GNUNET_CONTAINER_MultiUuidmap* pending_acks

static

Map of pending acknowledgements, mapping struct AcknowledgementUUID to a struct PendingAcknowledgement.

Definition at line 2955 of file gnunet-service-transport.c.

Referenced by do_shutdown(), free_pending_acknowledgement(), handle_reliability_ack(), prepare_pending_acknowledgement(), and run().

dv_routes

struct GNUNET_CONTAINER_MultiPeerMap* dv_routes

static

Map from PIDs to struct DistanceVector entries describing known paths to the peer.

Definition at line 2961 of file gnunet-service-transport.c.

Referenced by do_shutdown(), free_dv_route(), learn_dv_path(), route_control_message_without_fc(), and run().

validation_map

struct GNUNET_CONTAINER_MultiPeerMap* validation_map

static

Map from PIDs to struct ValidationState entries describing addresses we are aware of and their validity state.

Definition at line 2967 of file gnunet-service-transport.c.

Referenced by do_shutdown(), free_validation_state(), handle_add_queue_message(), handle_validation_response(), run(), and start_address_validation().

revalidation_map

struct GNUNET_CONTAINER_MultiHashMap* revalidation_map

static

Map from addresses to struct ValidationState entries describing addresses we are aware of and their validity state.

Definition at line 2973 of file gnunet-service-transport.c.

Referenced by do_shutdown(), free_validation_state(), handle_validation_response(), run(), and validation_transmit_on_queue().

links

struct GNUNET_CONTAINER_MultiPeerMap* links

static

Map from PIDs to struct VirtualLink entries describing links CORE knows to exist.

Definition at line 2979 of file gnunet-service-transport.c.

Referenced by activate_core_visible_dv_path(), do_shutdown(), free_virtual_link(), handle_client_start(), handle_flow_control(), handle_validation_response(), load_message_store_links(), lookup_virtual_link(), and run().

dvlearn_map

struct GNUNET_CONTAINER_MultiShortmap* dvlearn_map

static

Map from challenges to struct LearnLaunchEntry values.

Definition at line 2984 of file gnunet-service-transport.c.

Referenced by do_shutdown(), run(), and start_dv_learn().

lle_head

struct LearnLaunchEntry* lle_head = NULL

static

Head of a DLL sorted by launch time.

Definition at line 2989 of file gnunet-service-transport.c.

Referenced by do_shutdown(), and start_dv_learn().

lle_tail

struct LearnLaunchEntry* lle_tail = NULL

static

Tail of a DLL sorted by launch time.

Definition at line 2994 of file gnunet-service-transport.c.

Referenced by do_shutdown(), and start_dv_learn().

validation_heap

struct GNUNET_CONTAINER_Heap* validation_heap

static

MIN Heap sorted by "next_challenge" to struct ValidationState entries sorting addresses we are aware of by when we should next try to (re)validate (or expire) them.

Definition at line 3001 of file gnunet-service-transport.c.

Referenced by do_shutdown(), run(), update_next_challenge_time(), and validation_start_cb().

nh

struct GNUNET_NAT_Handle* nh

Handle for connect to the NAT service.

Definition at line 3006 of file gnunet-service-transport.c.

Referenced by do_shutdown(), handle_flow_control(), and run().

peerstore

struct GNUNET_PEERSTORE_Handle* peerstore

static

Database for peer's HELLOs.

Definition at line 3011 of file gnunet-service-transport.c.

Referenced by decaps_dv_box_cb(), do_shutdown(), handle_add_queue_message(), handle_dv_learn(), handle_validation_response(), pils_pid_change_cb(), pils_sign_addr_cb(), pils_sign_hello_cb(), queue_burst(), run(), and update_backtalker_monotime().

pils

struct GNUNET_PILS_Handle* pils

static

Service that manages our peer id.

Definition at line 3016 of file gnunet-service-transport.c.

Referenced by do_shutdown(), feed_addresses_to_pils(), forward_dv_learn(), handle_dv_box(), handle_validation_challenge(), peerstore_store_own_cb(), pils_sign_address(), run(), sign_ephemeral(), and start_dv_learn().

pils_requests_head

struct PilsRequest* pils_requests_head

static

PILS Operation DLL.

Definition at line 3042 of file gnunet-service-transport.c.

Referenced by decaps_dv_box_cb(), do_shutdown(), forward_dv_learn(), handle_dv_box(), handle_validation_challenge(), peerstore_store_own_cb(), pils_sign_addr_cb(), pils_sign_address(), pils_sign_hello_cb(), sign_dhp_cp(), sign_dv_cb(), sign_dv_init_cb(), sign_ephemeral(), sign_t_validation_cb(), and start_dv_learn().

pils_requests_tail

struct PilsRequest* pils_requests_tail

static

PILS Operation DLL.

Definition at line 3047 of file gnunet-service-transport.c.

Referenced by decaps_dv_box_cb(), do_shutdown(), forward_dv_learn(), handle_dv_box(), handle_validation_challenge(), peerstore_store_own_cb(), pils_sign_addr_cb(), pils_sign_address(), pils_sign_hello_cb(), sign_dhp_cp(), sign_dv_cb(), sign_dv_init_cb(), sign_ephemeral(), sign_t_validation_cb(), and start_dv_learn().

dvlearn_task

struct GNUNET_SCHEDULER_Task* dvlearn_task

static

Task run to initiate DV learning.

Definition at line 3052 of file gnunet-service-transport.c.

Referenced by do_shutdown(), handle_add_queue_message(), sign_dv_init_cb(), and start_dv_learn().

validation_task

struct GNUNET_SCHEDULER_Task* validation_task

static

Task to run address validation.

Definition at line 3057 of file gnunet-service-transport.c.

Referenced by do_shutdown(), update_next_challenge_time(), and validation_start_cb().

pils_feed_task

struct GNUNET_SCHEDULER_Task* pils_feed_task

static

Task to feed addresses to PILS.

Definition at line 3062 of file gnunet-service-transport.c.

Referenced by do_shutdown(), feed_addresses_to_pils(), handle_add_address(), and handle_del_address().

ir_head

struct IncomingRequest* ir_head

static

List of incoming connections where we are trying to get a connection back established.

Length kept in ir_total.

Definition at line 3069 of file gnunet-service-transport.c.

Referenced by do_shutdown(), free_incoming_request(), and sign_t_validation_cb().

ir_tail

struct IncomingRequest* ir_tail

static

Tail of DLL starting at ir_head.

Definition at line 3074 of file gnunet-service-transport.c.

Referenced by free_incoming_request(), and sign_t_validation_cb().

ir_total

unsigned int ir_total

static

Length of the DLL starting at ir_head.

Definition at line 3079 of file gnunet-service-transport.c.

Referenced by do_shutdown(), free_incoming_request(), and sign_t_validation_cb().

logging_uuid_gen

unsigned long long logging_uuid_gen

static

Generator of logging_uuid in struct PendingMessage.

Definition at line 3084 of file gnunet-service-transport.c.

Referenced by extract_box_cb(), forward_dv_box(), fragment_message(), handle_client_send(), and reliability_box_message().

burst_running

enum GNUNET_GenericReturnValue burst_running

static

Is there a burst running?

Definition at line 3089 of file gnunet-service-transport.c.

Referenced by burst_timeout(), handle_burst_finished(), queue_burst(), and start_burst().

hello_mono_time

struct GNUNET_TIME_Absolute hello_mono_time

static

Monotonic time we use for HELLOs generated at this time.

TODO: we should increase this value from time to time (i.e. whenever a struct AddressListEntry actually expires), but IF we do this, we must also update all (remaining) addresses in the PEERSTORE at that time! (So for now only increased when the peer is restarted, which hopefully roughly matches whenever our addresses change.)

Definition at line 3099 of file gnunet-service-transport.c.

Referenced by run(), and store_pi().

in_shutdown

int in_shutdown

static

Indication if we have received a shutdown signal and are in the process of cleaning up.

Definition at line 3105 of file gnunet-service-transport.c.

Referenced by client_disconnect_cb(), run(), and shutdown_task().

burst_task

struct GNUNET_SCHEDULER_Task* burst_task

static

The task to start the burst.

Definition at line 3110 of file gnunet-service-transport.c.

Referenced by queue_burst(), and start_burst().

burst_timeout_task

struct GNUNET_SCHEDULER_Task* burst_timeout_task

Definition at line 3112 of file gnunet-service-transport.c.

Referenced by start_burst().

use_burst

enum GNUNET_GenericReturnValue use_burst

Definition at line 3114 of file gnunet-service-transport.c.

Referenced by queue_burst(), and run().