rdma_cm.7

.\" Licensed under the OpenIB.org BSD license (FreeBSD Variant) - See COPYING.md
.TH "RDMA_CM" 7 "2010-07-19" "librdmacm" "Librdmacm Programmer's Manual" librdmacm
.SH NAME
rdma_cm \- RDMA communication manager.
.SH SYNOPSIS
.B "#include <rdma/rdma_cma.h>"
.SH "DESCRIPTION"
Used to establish communication over RDMA transports.
.SH "NOTES"
The RDMA CM is a communication manager used to setup reliable, connected
and unreliable datagram data transfers.  It provides an RDMA transport
neutral interface for establishing connections.  The API concepts are
based on sockets, but adapted for queue pair (QP) based semantics:
communication must be over a specific RDMA device, and data transfers
are message based.
.P
The RDMA CM can control both the QP and communication management (connection setup /
teardown) portions of an RDMA API, or only the communication management
piece.  It works in conjunction with the verbs
API defined by the libibverbs library.  The libibverbs library provides the
underlying interfaces needed to send and receive data.
.P
The RDMA CM can operate asynchronously or synchronously.  The mode of
operation is controlled by the user through the use of the rdma_cm event channel
parameter in specific calls.  If an event channel is provided, an rdma_cm identifier
will report its event data (results of connecting, for example), on that channel.
If a channel is not provided, then all rdma_cm operations for the selected
rdma_cm identifier will block until they complete.
.P
The RDMA CM gives an option to different libibverbs providers to advertise and
use various specific to that provider QP configuration options. This functionality
is called ECE (enhanced connection establishment).
.SH "RDMA VERBS"
The rdma_cm supports the full range of verbs available through the libibverbs
library and interfaces.  However, it also provides wrapper functions for some
of the more commonly used verbs funcationality.  The full set of abstracted
verb calls are:
.P
rdma_reg_msgs  - register an array of buffers for sending and receiving
.P
rdma_reg_read  - registers a buffer for RDMA read operations
.P
rdma_reg_write - registers a buffer for RDMA write operations
.P
rdma_dereg_mr  - deregisters a memory region
.P
rdma_post_recv  - post a buffer to receive a message
.P
rdma_post_send  - post a buffer to send a message
.P
rdma_post_read  - post an RDMA to read data into a buffer
.P
rdma_post_write - post an RDMA to send data from a buffer
.P
rdma_post_recvv  - post a vector of buffers to receive a message
.P
rdma_post_sendv  - post a vector of buffers to send a message
.P
rdma_post_readv  - post a vector of buffers to receive an RDMA read
.P
rdma_post_writev - post a vector of buffers to send an RDMA write
.P
rdma_post_ud_send - post a buffer to send a message on a UD QP
.P
rdma_get_send_comp - get completion status for a send or RDMA operation
.P
rdma_get_recv_comp - get information about a completed receive
.SH "CLIENT OPERATION"
This section provides a general overview of the basic operation for the active,
or client, side of communication.  This flow assume asynchronous operation with
low level call details shown.  For
synchronous operation, calls to rdma_create_event_channel, rdma_get_cm_event,
rdma_ack_cm_event, and rdma_destroy_event_channel
would be eliminated.  Abstracted calls, such as rdma_create_ep encapsulate
several of these calls under a single API.
Users may also refer to the example applications for
code samples.  A general connection flow would be:
.IP rdma_getaddrinfo
retrieve address information of the destination
.IP rdma_create_event_channel
create channel to receive events
.IP rdma_create_id
allocate an rdma_cm_id, this is conceptually similar to a socket
.IP rdma_resolve_addr
obtain a local RDMA device to reach the remote address
.IP rdma_get_cm_event
wait for RDMA_CM_EVENT_ADDR_RESOLVED event
.IP rdma_ack_cm_event
ack event
.IP rdma_create_qp
allocate a QP for the communication
.IP rdma_resolve_route
determine the route to the remote address
.IP rdma_get_cm_event
wait for RDMA_CM_EVENT_ROUTE_RESOLVED event
.IP rdma_ack_cm_event
ack event
.IP rdma_connect
connect to the remote server
.IP rdma_get_cm_event
wait for RDMA_CM_EVENT_ESTABLISHED event
.IP rdma_ack_cm_event
ack event
.P
Perform data transfers over connection
.IP rdma_disconnect
tear-down connection
.IP rdma_get_cm_event
wait for RDMA_CM_EVENT_DISCONNECTED event
.IP rdma_ack_cm_event
ack event
.IP rdma_destroy_qp
destroy the QP
.IP rdma_destroy_id
release the rdma_cm_id
.IP rdma_destroy_event_channel
release the event channel
.IP rdma_set_local_ece
set desired ECE options
.P
An almost identical process is used to setup unreliable datagram (UD)
communication between nodes.  No actual connection is formed between QPs
however, so disconnection is not needed.
.P
Although this example shows the client initiating the disconnect, either side
of a connection may initiate the disconnect.
.SH "SERVER OPERATION"
This section provides a general overview of the basic operation for the passive,
or server, side of communication.  A general connection flow would be:
.IP rdma_create_event_channel
create channel to receive events
.IP rdma_create_id
allocate an rdma_cm_id, this is conceptually similar to a socket
.IP rdma_bind_addr
set the local port number to listen on
.IP rdma_listen
begin listening for connection requests
.IP rdma_get_cm_event
wait for RDMA_CM_EVENT_CONNECT_REQUEST event with a new rdma_cm_id
.IP rdma_create_qp
allocate a QP for the communication on the new rdma_cm_id
.IP rdma_accept
accept the connection request
.IP rdma_ack_cm_event
ack event
.IP rdma_get_cm_event
wait for RDMA_CM_EVENT_ESTABLISHED event
.IP rdma_ack_cm_event
ack event
.P
Perform data transfers over connection
.IP rdma_get_cm_event
wait for RDMA_CM_EVENT_DISCONNECTED event
.IP rdma_ack_cm_event
ack event
.IP rdma_disconnect
tear-down connection
.IP rdma_destroy_qp
destroy the QP
.IP rdma_destroy_id
release the connected rdma_cm_id
.IP rdma_destroy_id
release the listening rdma_cm_id
.IP rdma_destroy_event_channel
release the event channel
.IP rdma_get_remote_ece
get ECe options sent by the client
.IP rdma_set_local_ece
set desired ECE options
.SH "RETURN CODES"
.IP "=  0"
success
.IP "= -1"
error - see errno for more details
.P
Most librdmacm functions return 0 to indicate success, and a -1 return value
to indicate failure.  If a function operates asynchronously, a return value of 0
means that the operation was successfully started.  The operation could still
complete in error; users should check the status of the related event.  If the
return value is -1, then errno will contain additional information
regarding the reason for the failure.
.P
Prior versions of the library would return -errno and not set errno for some cases
related to ENOMEM, ENODEV, ENODATA, EINVAL, and EADDRNOTAVAIL codes. Applications
that want to check these codes and have compatibility with prior library versions
must manually set errno to the negative of the return code if it is < -1.
.SH "SEE ALSO"
rdma_accept(3),
rdma_ack_cm_event(3),
rdma_bind_addr(3),
rdma_connect(3),
rdma_create_ep(3),
rdma_create_event_channel(3),
rdma_create_id(3),
rdma_create_qp(3),
rdma_dereg_mr(3),
rdma_destroy_ep(3),
rdma_destroy_event_channel(3),
rdma_destroy_id(3),
rdma_destroy_qp(3),
rdma_disconnect(3),
rdma_event_str(3),
rdma_free_devices(3),
rdma_getaddrinfo(3),
rdma_get_cm_event(3),
rdma_get_devices(3),
rdma_get_dst_port(3),
rdma_get_local_addr(3),
rdma_get_peer_addr(3),
rdma_get_recv_comp(3),
rdma_get_remote_ece(3),
rdma_get_request(3),
rdma_get_send_comp(3),
rdma_get_src_port(3),
rdma_join_multicast(3),
rdma_leave_multicast(3),
rdma_listen(3),
rdma_migrate_id(3),
rdma_notify(3),
rdma_post_read(3)
rdma_post_readv(3),
rdma_post_recv(3),
rdma_post_recvv(3),
rdma_post_send(3),
rdma_post_sendv(3),
rdma_post_ud_send(3),
rdma_post_write(3),
rdma_post_writev(3),
rdma_reg_msgs(3),
rdma_reg_read(3),
rdma_reg_write(3),
rdma_reject(3),
rdma_resolve_addr(3),
rdma_resolve_route(3),
rdma_get_remote_ece(3),
rdma_set_option(3),
mckey(1),
rdma_client(1),
rdma_server(1),
rping(1),
ucmatose(1),
udaddy(1)