recv(2) recv(2)
NAME [Toc] [Back]
recv, recvfrom, recvmsg - receive a message from a socket
SYNOPSIS [Toc] [Back]
#include <sys/socket.h>
int recv(int s, void *buf, int len, int flags);
int recvfrom(
int s,
void *buf,
int len,
int flags,
void *from,
int *fromlen
);
int recvmsg(int s, struct msghdr msg[], int flags);
_XOPEN_SOURCE_EXTENDED Only (UNIX 98)
ssize_t recv(int s, void *buf, size_t len, int flags);
ssize_t recvfrom(
int s,
void *buf,
size_t len,
int flags,
struct sockaddr *from,
socklen_t *fromlen
);
ssize_t recvmsg(int s, struct msghdr *msg, int flags);
Obsolescent _XOPEN_SOURCE_EXTENDED Only (UNIX 95) [Toc] [Back]
ssize_t recvfrom(
int s,
void *buf,
size_t len,
int flags,
struct sockaddr *from,
size_t *fromlen
);
DESCRIPTION [Toc] [Back]
The recv(), recvfrom(), and recvmsg() system calls are used to receive
messages from a socket.
s is a socket descriptor from which messages are received.
buf is a pointer to the buffer into which the messages are placed.
Hewlett-Packard Company - 1 - HP-UX 11i Version 2: August 2003
recv(2) recv(2)
len is the maximum number of bytes that can fit in the buffer
referenced by buf.
If the socket uses connection-based communications, such as a
SOCK_STREAM socket, these calls can only be used after the connection
has been established (see connect(2)). For connectionless sockets
such as SOCK_DGRAM, these calls can be used whether a connection has
been specified or not.
recvfrom() operates in the same manner as recv() except that it is
able to return the address of the socket from which the message was
sent. For connected datagram sockets, recvfrom() simply returns the
same address as getpeername() (see getpeername(2)). For stream
sockets, recvfrom() retrieves data in the same manner as recv(), but
does not return the socket address of the sender. If from is nonzero,
the source address of the message is placed in the socket address
structure pointed to by from. fromlen is a value-result parameter,
initialized to the size of the structure associated with from, and
modified on return to indicate the actual size of the address stored
there. If the memory pointed to by from is not large enough to
contain the entire address, only the first fromlen bytes of the
address are returned.
For message-based sockets such as SOCK_DGRAM, the entire message must
be read in a single operation. If a message is too long to fit in the
supplied buffer, the excess bytes are discarded. For stream-based
sockets such as SOCK_STREAM, there is no concept of message
boundaries. In this case, data is returned to the user as soon as it
becomes available, and no data is discarded. See the AF_CCITT Only
subsection below for a list of the exceptions to this behavior for
connections in the address family AF_CCITT.
recvmsg() performs the same action as recv(), but scatters the read
data into the buffers specified in the msghdr structure (see
_XOPEN_SOURCE_EXTENDED Only below). This structure is defined in
<sys/socket.h> and has the following form (HP-UX BSD Sockets Only):
struct msghdr {
caddr_t msg_name; /* optional address */
int msg_namelen; /* size of address */
struct iovec *msg_iov; /* scatter array for data */
int msg_iovlen; /* # of elements in msg_iov */
caddr_t msg_accrights; /* access rights */
int msg_accrightslen; /* size of msg_accrights */
}
msg_name points to a sockaddr structure in which the address of the
sending socket is to be stored, if the socket is connectionless;
msg_name may be a null pointer if no name is specified. msg_iov
specifies the locations of the character arrays for storing the
incoming data. msg_accrights specifies a buffer to receive any access
Hewlett-Packard Company - 2 - HP-UX 11i Version 2: August 2003
recv(2) recv(2)
rights sent along with the message. Access rights are limited to file
descriptors of size int. If access rights are not being transferred,
set the msg_accrights field to NULL. Access rights are supported only
for AF_UNIX.
By default, recv(), recvfrom() and recvmsg() are blocking system calls
and do not complete until there is data to read or the end of the
socket data stream is reached (the remote side of a connection-based
socket has performed an orderly shutdown and there is no more data to
read).
There are three ways to enable nonblocking mode:
+ With the FIOSNBIO ioctl() request
+ With the O_NONBLOCK fcntl() flag
+ With the O_NDELAY fcntl() flag
Although the use of FIONBIO is not recommended, if nonblocking I/O is
enabled using FIOSNBIO or the equivalent FIONBIO request (defined in
<sys/ioctl.h> and explained in ioctl(2), ioctl(5) and socket(7)), the
recv() request completes in one of six ways:
+ If there is enough data available to satisfy the entire request,
recv() completes successfully, having read all of the data, and
returns the number of bytes read.
+ If there is not enough data available to satisfy the entire
request, recv() completes successfully, having read as much data
as possible, and returns the number of bytes it was able to read.
+ If there is no data available, recv() returns -1 and errno is set
to [EWOULDBLOCK].
+ If the remote side of a connection-based socket has performed an
orderly shutdown and there is no more data to read (the socket
has reached the end of its data stream), recv() returns 0.
+ If the remote side of a connection-based socket has reset the
connection, recv() returns -1 and sets errno to [ECONNRESET].
+ If an error occurs, recv() returns -1 and sets errno to indicate
the error.
If nonblocking I/O is disabled using FIOSNBIO, recv() always executes
completely (blocking as necessary) and returns the number of bytes
read.
If the O_NONBLOCK flag is set using fcntl() (defined in <sys/fcntl.h>
and explained in fcntl(2) and fcntl(5)), POSIX-style nonblocking I/O
is enabled. In this case, the recv() request completes in one of six
ways:
Hewlett-Packard Company - 3 - HP-UX 11i Version 2: August 2003
recv(2) recv(2)
+ If there is enough data available to satisfy the entire request,
recv() completes successfully, having read all the data, and
returns the number of bytes read.
+ If there is not enough data available to satisfy the entire
request, recv() completes successfully, having read as much data
as possible, and returns the number of bytes it was able to read.
+ If there is no data available, recv() completes, having read no
data, and returns -1 and sets errno to [EAGAIN].
+ If the remote side of a connection-based socket has performed an
orderly shutdown and there is no more data to read (the socket
has reached the end of its data stream), recv() returns 0.
+ If the remote side of a connection-based socket has reset the
connection, recv() returns -1 and sets errno to [ECONNRESET].
+ If an error occurs, recv() returns -1 and errno is set to
indicate the error.
If the O_NDELAY flag is set using fcntl() (defined in <sys/fcntl.h>
and explained in fcntl(2) and fcntl(5)), nonblocking I/O is enabled.
In this case, the recv() request completes in one of five ways:
+ If there is enough data available to satisfy the entire request,
recv() completes successfully, having read all the data, and
returns the number of bytes read.
+ If there is not enough data to satisfy the entire request recv()
completes successfully, having read as much data as possible, and
returns the number of bytes it was able to read.
+ If the remote side of a connection-based socket has performed an
orderly shutdown and there is no more data to read (the socket
has reached the end of its data stream), recv() completes and
returns 0. The recv() call will also complete and return 0 when
there is no data to read. To distinguish this condition from the
end of the data stream, you can use a select() or poll() call to
check for data before calling recv(). If the select() or poll()
call indicates that an event has occurred on the socket, the
subsequent but recv() call will only return 0 if the socket has
reached the end of the data stream.
+ If the remote side of a connection-based socket has reset the
connection, recv() returns -1 and sets errno to [ECONNRESET].
+ If an error occurs, recv() returns -1 and errno is set to
indicate the error.
Hewlett-Packard Company - 4 - HP-UX 11i Version 2: August 2003
recv(2) recv(2)
If the O_NONBLOCK or O_NDELAY flag is cleared using fcntl(), the
corresponding style of nonblocking I/O, if previously enabled, is
disabled. In this case, recv() always executes completely (blocking
as necessary) and returns the number of bytes read.
Since both the fcntl() O_NONBLOCK and O_NDELAY flags and ioctl()
FIOSNBIO request are supported, some clarification on how these
features interact is necessary. If the O_NONBLOCK or O_NDELAY flag
has been set, recv() requests behave accordingly, regardless of any
FIOSNBIO requests. If neither the O_NONBLOCK flag nor the O_NDELAY
flag has been set, FIOSNBIO requests control the behavior of recv().
By default nonblocking I/O is disabled.
select() can be used to determine when more data arrives by selecting
the socket for reading.
The flags parameter can be set to MSG_PEEK, MSG_OOB, both, or zero.
If it is set to MSG_PEEK, any data returned to the user still is
treated as if it had not been read. The next recv() rereads the same
data. The MSG_OOB flag is used to receive out-of-band data. For TCP
SOCK_STREAM sockets, both the MSG_PEEK and MSG_OOB flags can be set at
the same time. The MSG_OOB flag value is supported for TCP
SOCK_STREAM sockets only. MSG_OOB is not supported for AF_UNIX or
AF_VME_LINK sockets.
A read() call made to a socket behaves in exactly the same way as a
recv() with flags set to zero.
AF_CCITT Only [Toc] [Back]
Connections in the address family AF_CCITT support message-based
sockets only. Although the user specifies connection-based
communications (SOCK_STREAM), the X.25 subsystem communicates via
messages. This address family does not support SOCK_DGRAM socket
types.
Normally, each recv() returns one complete X.25 message. If the
socket is in nonblocking mode, recv() behaves as described above.
Note that if the user specifies len less than the actual X.25 message
size, the excess data is discarded and no error indication is
returned. The size of the next available message as well as the state
of MDTF, D, and Q bits can be obtained with ioctl(X25_NEXT_MSG_STAT).
Connections of the address family AF_CCITT receive data in the same
way as message-based connections described above, with the following
additions and exceptions:
+ recvfrom() is supported; however, the from and fromlen parameters
are ignored (that is, it works in the same manner as recv()).
Hewlett-Packard Company - 5 - HP-UX 11i Version 2: August 2003
recv(2) recv(2)
+ To receive a message in fragments of the complete X.25 message,
use ioctl(X25_SET_FRAGMENT_SIZE). The state of the MDTF bit is 1
for all except the last fragment of the message.
+ The MSG_OOB flag is supported.
+ The MSG_PEEK flag is supported; the two flags can be combined.
+ If a message is received that is larger than the user-controlled
maximum message size, the X.25 subsystem RESETs the circuit,
discards the data, and sends the out-of-band event
OOB_VC_MESSAGE_TOO_BIG to the socket.
_XOPEN_SOURCE_EXTENDED Only
For X/Open Sockets, the msghdr structure has the following form:
(UNIX 98)
struct msghdr {
void *msg_name; /* optional address */
socklen_t msg_namelen; /* size of address */
struct iovec *msg_iov; /* scatter array for data */
int msg_iovlen; /* # of elements in msg_iov */
void *msg_control; /* ancillary data, see below */
socklen_t msg_controllen; /* ancillary data buffer len */
int msg_flags; /* flags on received message */
}
Obsolescent (UNIX 95) [Toc] [Back]
struct msghdr {
void *msg_name; /* optional address */
size_t msg_namelen; /* size of address */
struct iovec *msg_iov; /* scatter array for data */
int msg_iovlen; /* # of elements in msg_iov */
void *msg_control; /* ancillary data, see below */
size_t msg_controllen; /* ancillary data buffer len */
int msg_flags; /* flags on received message */
}
msg_control specifies a buffer to receive any ancillary data sent
along with the message. Ancillary data consists of a sequence of
pairs, each consisting of a cmsghdr structure followed by a data
array. The data array contains the ancillary data message, and the
cmsghdr structure contains descriptive information that allows an
application to correctly parse the data. cmsghdr has the following
structure:
(UNIX 98)
struct cmsghdr {
Hewlett-Packard Company - 6 - HP-UX 11i Version 2: August 2003
recv(2) recv(2)
socklen_t cmsg_len; /* data byte count, including hdr*/
int cmsg_level; /* originating protocol */
int cmsg_type; /* protocol-specific type */
}
Obsolescent (UNIX 95) [Toc] [Back]
struct cmsghdr {
size_t cmsg_len; /* data byte count, including hdr*/
int cmsg_level; /* originating protocol */
int cmsg_type; /* protocol-specific type */
}
If the cmsg_level is SOL_SOCKET, and cmsg_type is SCM_RIGHTS, then it
indicates that the data array contains the access rights to be
received. Access rights are supported only for AF_UNIX. Access
rights are limited to file descriptors of size int.
If the cmsg_level is IPPROTO_IPV6, then cmsg_type must be one of the
supported types: IPV6_PKTINFO, IPV6_HOPLIMIT, IPV6_NEXTHOP,
IPV6_RTHDR, IPV6_HOPOPTS, IPV6_DSTOPTS or IPV6_RTHDRDSTOPTS. (See
description in ip6(7P)).
If ancillary data are not being transferred, set the msg_control field
to NULL, and set the msg_controllen field to 0.
The flags parameter accepts a new value, MSG_WAITALL, which requests
that the function block until the full amount of data requested can be
returned. The function may return a smaller amount of data if a
signal is caught, the connection is terminated, or an error is pending
for the socket.
On successful completion of recvmsg(), the msg_flags member of the
message header is the bitwise-inclusive OR of all of the following
flags that indicate conditions detected for the received message.
MSG_EOR End of record was received (if supported by
the protocol).
MSG_OOB Out-of-band data was received.
MSG_TRUNC Normal data was truncated.
MSG_CTRUNC Control data was truncated.
DEPENDENCIES [Toc] [Back]
AF_CCITT
recvfrom() is supported; however, the from and fromlen parameters are
ignored (i.e., it works in the same manner as recv()).
Hewlett-Packard Company - 7 - HP-UX 11i Version 2: August 2003
recv(2) recv(2)
The O_NDELAY fcntl() call is not supported over X.25 links. Use the
FIOSNBIO ioctl() call instead to enable nonblocking I/0.
RETURN VALUE [Toc] [Back]
recv(), recvfrom(), and recvmsg() return the following values:
n Successful completion. n is the number of bytes received.
0 The socket is blocking and the transport connection to the
remote node failed, or the remote side of a connection-based
socket has performed an orderly shutdown and there is no
more data to read (the socket has reached the end of its
data stream). Sockets with the O_NDELAY flag set may also
return 0 at any time when there is no data available.
-1 Failure. errno is set to indicate the error.
ERRORS [Toc] [Back]
If recv(), recvfrom(), or recvmsg() fails, errno is set to one of the
following values.
[EAGAIN] Non-blocking I/O is enabled using O_NONBLOCK
flag with fcntl() and the receive operation
would block, or the socket has an error that
was set asynchronously. An asynchronous
error can be caused by a gateway failing to
forward a datagram because the datagram
exceeds the MTU of the next-hop network and
the "Don't Fragment" (DF) bit in the datagram
is set. (See SO_PMTU in getsockopt(2).)
[EBADF] The argument s is an invalid descriptor.
[ECONNRESET] A connection was forcibly closed by a peer.
[EFAULT] An invalid pointer was specified in the buf,
from, or fromlen parameter, or in the msghdr
structure.
[EINTR] The receive was interrupted by delivery of a
signal before any data was available for the
receive.
[EINVAL] The len parameter or a length in the msghdr
structure is invalid; or no data is available
on receive of out of band data.
[EMFILE] The number of file descriptors available to
the calling process has been exceeded.
[EMSGSIZE] A length in the msghdr structure is invalid.
Hewlett-Packard Company - 8 - HP-UX 11i Version 2: August 2003
recv(2) recv(2)
[ENOBUFS] Insufficient resources were available in the
system to perform the operation.
[ENOPROTOOPT] The remote system or an intermediate system
in the communications path does not support a
protocol option sent by the local system.
This option may have been set using a
getsockopt() or setsockopt() call, or set as
a system parameter.
[ENOSR] Buffers could not be allocated for the
message that was to be created due to
insufficient STREAMS memory resources.
[ENOTCONN] Receive on a SOCK_STREAM socket that is not
yet connected.
[ENOTSOCK] The argument s is a valid file descriptor,
but it is not a socket.
[EOPNOTSUPP] The MSG_OOB flag was set for a UDP SOCK_DGRAM
message-based socket, or MSG_OOB or MSG_PEEK
was set for any AF_UNIX socket. The MSG_OOB
flag is supported only for stream-based TCP
SOCK_STREAM sockets. Neither MSG_PEEK nor
MSG_OOB is supported for AF_UNIX sockets.
AF_CCITT only: recv() was issued on a
listen() socket.
[ETIMEDOUT] The connection timed out during connection
establishment, or due to a transmission
timeout on active connection.
[EWOULDBLOCK] Non-blocking I/O is enabled using ioctl()
FIOSNBIO request, and the requested operation
would block.
WARNINGS [Toc] [Back]
IPv6 is supported on HP-UX 11i Version 1.0, with the optional IPv6
software installed. Currently, IPv6 is not supported on systems
running HP-UX 11i Version 1.6.
OBSOLESCENCE [Toc] [Back]
Currently, the socklen_t and size_t types are the same size. This is
compatible with both the UNIX 95 and UNIX 98 profiles. However, in a
future release, socklen_t might be a different size. In that case,
passing a size_t pointer will evoke compile-time warnings, which must
be corrected in order for the application to behave correctly. Also,
the size of the msghdr and cmsghdr structures and the relative
position of their members will be different, which might affect
Hewlett-Packard Company - 9 - HP-UX 11i Version 2: August 2003
recv(2) recv(2)
application behavior. Applications that use socklen_t now, where
appropriate, will avoid such migration problems. On the other hand,
applications that need to be portable to the UNIX 95 profile should
follow the X/Open specification (see xopen_networking(7)).
FUTURE DIRECTION [Toc] [Back]
Currently, the default behavior is the HP-UX BSD Sockets; however, it
might be changed to X/Open Sockets in a future release. At that time,
any HP-UX BSD Sockets behavior that is incompatible with X/Open
Sockets might be obsoleted. Applications that conform to the X/Open
specification now will avoid migration problems (see
xopen_networking(7)).
AUTHOR [Toc] [Back]
recv(), recvmsg(), and recvfrom() were developed by HP and the
University of California, Berkeley.
SEE ALSO [Toc] [Back]
getsockopt(2), read(2), select(2), send(2), socket(2),
thread_safety(5), inet(7F), socket(7), ip6(7P), tcp(7P), udp(7P),
unix(7P), xopen_networking(7).
STANDARDS CONFORMANCE [Toc] [Back]
recv(): XPG4
Hewlett-Packard Company - 10 - HP-UX 11i Version 2: August 2003 [ Back ] |
