CC(9) FreeBSD Kernel Developer's Manual CC(9)
NAME
cc, DECLARE_CC_MODULE, CC_VAR -- Modular Congestion Control
SYNOPSIS
#include <netinet/cc.h>
#include <netinet/cc/cc_module.h>
DECLARE_CC_MODULE(ccname, ccalgo);
CC_VAR(ccv, what);
DESCRIPTION
The cc framework allows congestion control algorithms to be implemented
as dynamically loadable kernel modules via the kld(4) facility. Trans-
port protocols can select from the list of available algorithms on a con-
nection-by-connection basis, or use the system default (see cc(4) for
more details).
cc modules are identified by an ascii(7) name and set of hook functions
encapsulated in a struct cc_algo, which has the following members:
struct cc_algo {
char name[TCP_CA_NAME_MAX];
int (*mod_init) (void);
int (*mod_destroy) (void);
int (*cb_init) (struct cc_var *ccv);
void (*cb_destroy) (struct cc_var *ccv);
void (*conn_init) (struct cc_var *ccv);
void (*ack_received) (struct cc_var *ccv, uint16_t type);
void (*cong_signal) (struct cc_var *ccv, uint32_t type);
void (*post_recovery) (struct cc_var *ccv);
void (*after_idle) (struct cc_var *ccv);
};
The name field identifies the unique name of the algorithm, and should be
no longer than TCP_CA_NAME_MAX-1 characters in length (the
TCP_CA_NAME_MAX define lives in <netinet/tcp.h> for compatibility rea-
sons).
The mod_init function is called when a new module is loaded into the sys-
tem but before the registration process is complete. It should be imple-
mented if a module needs to set up some global state prior to being
available for use by new connections. Returning a non-zero value from
mod_init will cause the loading of the module to fail.
The mod_destroy function is called prior to unloading an existing module
from the kernel. It should be implemented if a module needs to clean up
any global state before being removed from the kernel. The return value
is currently ignored.
The cb_init function is called when a TCP control block struct tcpcb is
created. It should be implemented if a module needs to allocate memory
for storing private per-connection state. Returning a non-zero value
from cb_init will cause the connection set up to be aborted, terminating
the connection as a result.
The cb_destroy function is called when a TCP control block struct tcpcb
is destroyed. It should be implemented if a module needs to free memory
allocated in cb_init.
The conn_init function is called when a new connection has been estab-
lished and variables are being initialised. It should be implemented to
initialise congestion control algorithm variables for the newly estab-
lished connection.
The ack_received function is called when a TCP acknowledgement (ACK)
packet is received. Modules use the type argument as an input to their
congestion management algorithms. The ACK types currently reported by
the stack are CC_ACK and CC_DUPACK. CC_ACK indicates the received ACK
acknowledges previously unacknowledged data. CC_DUPACK indicates the
received ACK acknowledges data we have already received an ACK for.
The cong_signal function is called when a congestion event is detected by
the TCP stack. Modules use the type argument as an input to their con-
gestion management algorithms. The congestion event types currently
reported by the stack are CC_ECN, CC_RTO, CC_RTO_ERR and CC_NDUPACK.
CC_ECN is reported when the TCP stack receives an explicit congestion
notification (RFC3168). CC_RTO is reported when the retransmission time
out timer fires. CC_RTO_ERR is reported if the retransmission time out
timer fired in error. CC_NDUPACK is reported if N duplicate ACKs have
been received back-to-back, where N is the fast retransmit duplicate ack
threshold (N=3 currently as per RFC5681).
The post_recovery function is called after the TCP connection has recov-
ered from a congestion event. It should be implemented to adjust state
as required.
The after_idle function is called when data transfer resumes after an
idle period. It should be implemented to adjust state as required.
The DECLARE_CC_MODULE() macro provides a convenient wrapper around the
DECLARE_MODULE(9) macro, and is used to register a cc module with the cc
framework. The ccname argument specifies the module's name. The ccalgo
argument points to the module's struct cc_algo.
cc modules must instantiate a struct cc_algo, but are only required to
set the name field, and optionally any of the function pointers. The
stack will skip calling any function pointer which is NULL, so there is
no requirement to implement any of the function pointers. Using the C99
designated initialiser feature to set fields is encouraged.
Each function pointer which deals with congestion control state is passed
a pointer to a struct cc_var, which has the following members:
struct cc_var {
void *cc_data;
int bytes_this_ack;
tcp_seq curack;
uint32_t flags;
int type;
union ccv_container {
struct tcpcb *tcp;
struct sctp_nets *sctp;
} ccvc;
};
struct cc_var groups congestion control related variables into a single,
embeddable structure and adds a layer of indirection to accessing trans-
port protocol control blocks. The eventual goal is to allow a single set
of cc modules to be shared between all congestion aware transport proto-
cols, though currently only tcp(4) is supported.
To aid the eventual transition towards this goal, direct use of variables
from the transport protocol's data structures is strongly discouraged.
However, it is inevitable at the current time to require access to some
of these variables, and so the CC_VAR() macro exists as a convenience
accessor. The ccv argument points to the struct cc_var passed into the
function by the cc framework. The what argument specifies the name of
the variable to access.
Apart from the type and ccv_container fields, the remaining fields in
struct cc_var are for use by cc modules.
The cc_data field is available for algorithms requiring additional per-
connection state to attach a dynamic memory pointer to. The memory
should be allocated and attached in the module's cb_init hook function.
The bytes_this_ack field specifies the number of new bytes acknowledged
by the most recently received ACK packet. It is only valid in the
ack_received hook function.
The curack field specifies the sequence number of the most recently
received ACK packet. It is only valid in the ack_received, cong_signal
and post_recovery hook functions.
The flags field is used to pass useful information from the stack to a cc
module. The CCF_ABC_SENTAWND flag is relevant in ack_received and is set
when appropriate byte counting (RFC3465) has counted a window's worth of
bytes has been sent. It is the module's responsibility to clear the flag
after it has processed the signal. The CCF_CWND_LIMITED flag is relevant
in ack_received and is set when the connection's ability to send data is
currently constrained by the value of the congestion window. Algorithms
should use the abscence of this flag being set to avoid accumulating a
large difference between the congestion window and send window.
SEE ALSO
cc(4), cc_chd(4), cc_cubic(4), cc_hd(4), cc_htcp(4), cc_newreno(4),
cc_vegas(4), tcp(4)
ACKNOWLEDGEMENTS
Development and testing of this software were made possible in part by
grants from the FreeBSD Foundation and Cisco University Research Program
Fund at Community Foundation Silicon Valley.
FUTURE WORK
Integrate with sctp(4).
HISTORY
The modular Congestion Control (CC) framework first appeared in
FreeBSD 9.0.
The framework was first released in 2007 by James Healy and Lawrence
Stewart whilst working on the NewTCP research project at Swinburne Uni-
versity of Technology's Centre for Advanced Internet Architectures, Mel-
bourne, Australia, which was made possible in part by a grant from the
Cisco University Research Program Fund at Community Foundation Silicon
Valley. More details are available at:
http://caia.swin.edu.au/urp/newtcp/
AUTHORS
The cc framework was written by Lawrence Stewart <lstewart@FreeBSD.org>,
James Healy <jimmy@deefa.com> and David Hayes <david.hayes@ieee.org>.
This manual page was written by David Hayes <david.hayes@ieee.org> and
Lawrence Stewart <lstewart@FreeBSD.org>.
FreeBSD 9.0 February 15, 2011 FreeBSD 9.0
Man(1) output converted with
man2html