This is the Linux Serial Programming HOWTO. All about how to program communications with other devices / computers over a serial line under Linux. Different techniques are explained: Canonical I/O (only complete lines are transmitted/received), asyncronous I/O, and waiting for input from multiple sources.
This is the first update to the initial release of the Linux Serial Programming HOWTO. The primary purpose of this update is to change the author information and convert the document to DocBook format. In terms of technical content, very little if anything has changed at this time. Sweeping changes to the technical content aren't going to happen overnight, but I'll work on it as much as time allows.
If you've been waiting in the wings for someone to take over this HOWTO, you've gotten your wish. Please send me any and all feedback you have, it'd be very much appreciated.
All examples were tested using a i386 Linux Kernel 2.0.29.
This document is copyrighted (c) 1997 Peter Baumann, (c) 2001 Gary Frerking and is distributed under the terms of the Linux Documentation Project (LDP) license, stated below.
Unless otherwise stated, Linux HOWTO documents are copyrighted by their respective authors. Linux HOWTO documents may be reproduced and distributed in whole or in part, in any medium physical or electronic, as long as this copyright notice is retained on all copies. Commercial redistribution is allowed and encouraged; however, the author would like to be notified of any such distributions.
All translations, derivative works, or aggregate works incorporating any Linux HOWTO documents must be covered under this copyright notice. That is, you may not produce a derivative work from a HOWTO and impose additional restrictions on its distribution. Exceptions to these rules may be granted under certain conditions; please contact the Linux HOWTO coordinator at the address given below.
In short, we wish to promote dissemination of this information through as many channels as possible. However, we do wish to retain copyright on the HOWTO documents, and would like to be notified of any plans to redistribute the HOWTOs.
If you have any questions, please contact <firstname.lastname@example.org>
No liability for the contents of this documents can be accepted. Use the concepts, examples and other content at your own risk. As this is a new edition of this document, there may be errors and inaccuracies, that may of course be damaging to your system. Proceed with caution, and although this is highly unlikely, the author(s) do not take any responsibility for that.
All copyrights are held by their by their respective owners, unless specifically noted otherwise. Use of a term in this document should not be regarded as affecting the validity of any trademark or service mark.
Naming of particular products or brands should not be seen as endorsements.
You are strongly recommended to take a backup of your system before major installation and backups at regular intervals.
As previously mentioned, not much is new in terms of technical content yet.
The original author thanked Mr. Strudthoff, Michael Carter, Peter Waltenberg, Antonino Ianella, Greg Hankins, Dave Pfaltzgraff, Sean Lincolne, Michael Wiedmann, and Adrey Bonar.
The best way to debug your code is to set up another Linux box, and connect the two computers via a null-modem cable. Use miniterm (available from the LDP programmers guide (ftp://sunsite.unc.edu/pub/Linux/docs/LDP/programmers-guide/lpg-0.4.tar.gz in the examples directory) to transmit characters to your Linux box. Miniterm can be compiled very easily and will transmit all keyboard input raw over the serial port. Only the define statement #define MODEMDEVICE "/dev/ttyS0" has to be checked. Set it to ttyS0 for COM1, ttyS1 for COM2, etc.. It is essential for testing, that all characters are transmitted raw (without output processing) over the line. To test your connection, start miniterm on both computers and just type away. The characters input on one computer should appear on the other computer and vice versa. The input will not be echoed to the attached screen.
To make a null-modem cable you have to cross the TxD (transmit) and RxD (receive) lines. For a description of a cable see sect. 7 of the Serial-HOWTO.
It is also possible to perform this testing with only one computer, if you have two unused serial ports. You can then run two miniterms off two virtual consoles. If you free a serial port by disconnecting the mouse, remember to redirect /dev/mouse if it exists. If you use a multiport serial card, be sure to configure it correctly. I had mine configured wrong and everything worked fine as long as I was testing only on my computer. When I connected to another computer, the port started loosing characters. Executing two programs on one computer just isn't fully asynchronous.
The devices /dev/ttyS* are intended to hook up terminals to your Linux box, and are configured for this use after startup. This has to be kept in mind when programming communication with a raw device. E.g. the ports are configured to echo characters sent from the device back to it, which normally has to be changed for data transmission.
All parameters can be easily configured from within a program. The configuration is stored in a structure struct termios, which is defined in <asm/termbits.h>:
This file also includes all flag definitions. The input mode flags in c_iflag handle all input processing, which means that the characters sent from the device can be processed before they are read with read. Similarly c_oflag handles the output processing. c_cflag contains the settings for the port, as the baudrate, bits per character, stop bits, etc.. The local mode flags stored in c_lflag determine if characters are echoed, signals are sent to your program, etc.. Finally the array c_cc defines the control characters for end of file, stop, etc.. Default values for the control characters are defined in <asm/termios.h>. The flags are described in the manual page termios(3). The structure termios contains the c_line (line discipline) element, which is not used in POSIX compliant systems.
Here three different input concepts will be presented. The appropriate concept has to be chosen for the intended application. Whenever possible, do not loop reading single characters to get a complete string. When I did this, I lost characters, whereas a read for the whole string did not show any errors.
This is the normal processing mode for terminals, but can also be useful for communicating with other dl input is processed in units of lines, which means that a read will only return a full line of input. A line is by default terminated by a NL (ASCII LF), an end of file, or an end of line character. A CR (the DOS/Windows default end-of-line) will not terminate a line with the default settings.
Canonical input processing can also handle the erase, delete word, and reprint characters, translate CR to NL, etc..
Non-Canonical Input Processing will handle a fixed amount of characters per read, and allows for a character timer. This mode should be used if your application will always read a fixed number of characters, or if the connected device sends bursts of characters.
The two modes described above can be used in synchronous and asynchronous mode. Synchronous is the default, where a read statement will block, until the read is satisfied. In asynchronous mode the read statement will return immediatly and send a signal to the calling program upon completion. This signal can be received by a signal handler.
This is not a different input mode, but might be useful, if you are handling multiple devices. In my application I was handling input over a TCP/IP socket and input over a serial connection from another computer quasi-simultaneously. The program example given below will wait for input from two different input sources. If input from one source becomes available, it will be processed, and the program will then wait for new input.
The approach presented below seems rather complex, but it is important to keep in mind that Linux is a multi-processing operating system. The select system call will not load the CPU while waiting for input, whereas looping until input becomes available would slow down other processes executing at the same time.
All examples have been derived from miniterm.c. The type ahead buffer is limited to 255 characters, just like the maximum string length for canonical input processing (<linux/limits.h> or <posix1_lim.h>).
See the comments in the code for explanation of the use of the different input modes. I hope that the code is understandable. The example for canonical input is commented best, the other examples are commented only where they differ from the example for canonical input to emphasize the differences.
The descriptions are not complete, but you are encouraged to experiment with the examples to derive the best solution for your application.
Don't forget to give the appropriate serial ports the right permissions (e. g.: chmod a+rw /dev/ttyS1)!
In non-canonical input processing mode, input is not assembled into lines and input processing (erase, kill, delete, etc.) does not occur. Two parameters control the behavior of this mode: c_cc[VTIME] sets the character timer, and c_cc[VMIN] sets the minimum number of characters to receive before satisfying the read.
If MIN > 0 and TIME = 0, MIN sets the number of characters to receive before the read is satisfied. As TIME is zero, the timer is not used.
If MIN = 0 and TIME > 0, TIME serves as a timeout value. The read will be satisfied if a single character is read, or TIME is exceeded (t = TIME *0.1 s). If TIME is exceeded, no character will be returned.
If MIN > 0 and TIME > 0, TIME serves as an inter-character timer. The read will be satisfied if MIN characters are received, or the time between two characters exceeds TIME. The timer is restarted every time a character is received and only becomes active after the first character has been received.
If MIN = 0 and TIME = 0, read will be satisfied immediately. The number of characters currently available, or the number of characters requested will be returned. According to Antonino (see contributions), you could issue a fcntl(fd, F_SETFL, FNDELAY); before reading to get the same result.
By modifying newtio.c_cc[VTIME] and newtio.c_cc[VMIN] all modes described above can be tested.
This section is kept to a minimum. It is just intended to be a hint, and therefore the example code is kept short. This will not only work with serial ports, but with any set of file descriptors.
The select call and accompanying macros use a fd_set. This is a bit array, which has a bit entry for every valid file descriptor number. select will accept a fd_set with the bits set for the relevant file descriptors and returns a fd_set, in which the bits for the file descriptors are set where input, output, or an exception occurred. All handling of fd_set is done with the provided macros. See also the manual page select(2).
The given example blocks indefinitely, until input from one of the sources becomes available. If you need to timeout on input, just replace the select call by:
This example will timeout after 1 second. If a timeout occurs, select will return 0, but beware that Timeout is decremented by the time actually waited for input by select. If the timeout value is zero, select will return immediatly.