sendfile.2freebsd

Langue: en

Autres versions - même langue

Version: 338604 (ubuntu - 24/10/10)

Section: 2 (Appels système)


BSD mandoc

NAME

sendfile - send a file to a socket

LIBRARY

Lb libc

SYNOPSIS

In sys/types.h In sys/socket.h In sys/uio.h Ft int Fo sendfile Fa int fd int s off_t offset size_t nbytes Fa struct sf_hdtr *hdtr off_t *sbytes int flags Fc  

DESCRIPTION

The Fn sendfile system call sends a regular file specified by descriptor Fa fd out a stream socket specified by descriptor Fa s .

The Fa offset argument specifies where to begin in the file. Should Fa offset fall beyond the end of file, the system will return success and report 0 bytes sent as described below. The Fa nbytes argument specifies how many bytes of the file should be sent, with 0 having the special meaning of send until the end of file has been reached.

An optional header and/or trailer can be sent before and after the file data by specifying a pointer to a Vt struct sf_hdtr , which has the following structure:

 struct sf_hdtr {
         struct iovec *headers;  /* pointer to header iovecs */
         int hdr_cnt;            /* number of header iovecs */
         struct iovec *trailers; /* pointer to trailer iovecs */
         int trl_cnt;            /* number of trailer iovecs */
 };
 

The Fa headers and Fa trailers pointers, if non- NULL point to arrays of Vt struct iovec structures. See the Fn writev system call for information on the iovec structure. The number of iovecs in these arrays is specified by Fa hdr_cnt and Fa trl_cnt .

If non- NULL the system will write the total number of bytes sent on the socket to the variable pointed to by Fa sbytes .

The Fa flags argument is a bitmap of these values:

SF_NODISKIO
This flag causes any Fn sendfile call which would block on disk I/O to instead return Er EBUSY . Busy servers may benefit by transferring requests that would block to a separate I/O worker thread.
SF_MNOWAIT
(description missing)
SF_SYNC
sleeps until the network stack no longer references the VM pages of the file, making subsequent modifications to it safe. Please note that this is not a guarantee that the data has actually been sent.

When using a socket marked for non-blocking I/O, Fn sendfile may send fewer bytes than requested. In this case, the number of bytes successfully written is returned in Fa *sbytes (if specified), and the error Er EAGAIN is returned.

IMPLEMENTATION NOTES

The Fx implementation of Fn sendfile is "zero-copy", meaning that it has been optimized so that copying of the file data is avoided.

TUNING

On some architectures, this system call internally uses a special Fn sendfile buffer (Vt struct sf_buf ) to handle sending file data to the client. If the sending socket is blocking, and there are not enough Fn sendfile buffers available, Fn sendfile will block and report a state of ``sfbufa '' If the sending socket is non-blocking and there are not enough Fn sendfile buffers available, the call will block and wait for the necessary buffers to become available before finishing the call.

The number of Vt sf_buf Ns 's allocated should be proportional to the number of nmbclusters used to send data to a client via Fn sendfile . Tune accordingly to avoid blocking! Busy installations that make extensive use of Fn sendfile may want to increase these values to be inline with their kern.ipc.nmbclusters (see tuning(7) for details).

The number of Fn sendfile buffers available is determined at boot time by either the kern.ipc.nsfbufs loader.conf5 variable or the NSFBUFS kernel configuration tunable. The number of Fn sendfile buffers scales with kern.maxusers The kern.ipc.nsfbufsused and kern.ipc.nsfbufspeak read-only sysctl(8) variables show current and peak Fn sendfile buffers usage respectively. These values may also be viewed through netstat -m

If a value of zero is reported for kern.ipc.nsfbufs your architecture does not need to use Fn sendfile buffers because their task can be efficiently performed by the generic virtual memory structures.

RETURN VALUES

Rv -std sendfile

ERRORS

Bq Er EAGAIN
The socket is marked for non-blocking I/O and not all data was sent due to the socket buffer being filled. If specified, the number of bytes successfully sent will be returned in Fa *sbytes .
Bq Er EBADF
The Fa fd argument is not a valid file descriptor.
Bq Er EBADF
The Fa s argument is not a valid socket descriptor.
Bq Er EBUSY
Completing the entire transfer would have required disk I/O, so it was aborted. Partial data may have been sent. (This error can only occur when SF_NODISKIO is specified.)
Bq Er EFAULT
An invalid address was specified for an argument.
Bq Er EINTR
A signal interrupted Fn sendfile before it could be completed. If specified, the number of bytes successfully sent will be returned in Fa *sbytes .
Bq Er EINVAL
The Fa fd argument is not a regular file.
Bq Er EINVAL
The Fa s argument is not a SOCK_STREAM type socket.
Bq Er EINVAL
The Fa offset argument is negative.
Bq Er EIO
An error occurred while reading from Fa fd .
Bq Er ENOTCONN
The Fa s argument points to an unconnected socket.
Bq Er ENOTSOCK
The Fa s argument is not a socket.
Bq Er EOPNOTSUPP
The file system for descriptor Fa fd does not support Fn sendfile .
Bq Er EPIPE
The socket peer has closed the connection.

SEE ALSO

netstat(1), open(2), send(2), socket(2), writev(2), tuning(7)
K. Elmeleegy A. Chanda A. L. Cox W. Zwaenepoel A Portable Kernel Abstraction for Low-Overhead Ephemeral Mapping Management The Proceedings of the 2005 USENIX Annual Technical Conference pp 223-236 2005

HISTORY

The Fn sendfile system call first appeared in Fx 3.0 . This manual page first appeared in Fx 3.1 .

AUTHORS

The Fn sendfile system call and this manual page were written by An David G. Lawrence Aq dg@dglawrence.com .